Abstract
Background
Convalescent plasma may reduce mortality in patients with viral respiratory diseases, and is being investigated as a potential therapy for coronavirus disease 2019 (COVID‐19). A thorough understanding of the current body of evidence regarding benefits and risks of this intervention is required.
Objectives
To assess the effectiveness and safety of convalescent plasma transfusion in the treatment of people with COVID‐19; and to maintain the currency of the evidence using a living systematic review approach.
Search methods
To identify completed and ongoing studies, we searched the World Health Organization (WHO) COVID‐19 Global literature on coronavirus disease Research Database, MEDLINE, Embase, Cochrane COVID‐19 Study Register, and the Epistemonikos COVID‐19 L*OVE Platform. We searched monthly until 03 March 2022.
Selection criteria
We included randomised controlled trials (RCTs) evaluating convalescent plasma for COVID‐19, irrespective of disease severity, age, gender or ethnicity.
We excluded studies that included populations with other coronavirus diseases (severe acute respiratory syndrome (SARS) or Middle East respiratory syndrome (MERS)), as well as studies evaluating standard immunoglobulin.
Data collection and analysis
We followed standard Cochrane methodology.
To assess bias in included studies we used RoB 2. We used the GRADE approach to rate the certainty of evidence for the following outcomes: all‐cause mortality at up to day 28, worsening and improvement of clinical status (for individuals with moderate to severe disease), hospital admission or death, COVID‐19 symptoms resolution (for individuals with mild disease), quality of life, grade 3 or 4 adverse events, and serious adverse events.
Main results
In this fourth review update version, we included 33 RCTs with 24,861 participants, of whom 11,432 received convalescent plasma. Of these, nine studies are single‐centre studies and 24 are multi‐centre studies. Fourteen studies took place in America, eight in Europe, three in South‐East Asia, two in Africa, two in western Pacific and three in eastern Mediterranean regions and one in multiple regions. We identified a further 49 ongoing studies evaluating convalescent plasma, and 33 studies reporting as being completed.
Individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease
29 RCTs investigated the use of convalescent plasma for 22,728 participants with moderate to severe disease. 23 RCTs with 22,020 participants compared convalescent plasma to placebo or standard care alone, five compared to standard plasma and one compared to human immunoglobulin. We evaluate subgroups on detection of antibodies detection, symptom onset, country income groups and several co‐morbidities in the full text.
Convalescent plasma versus placebo or standard care alone
Convalescent plasma does not reduce all‐cause mortality at up to day 28 (risk ratio (RR) 0.98, 95% confidence interval (CI) 0.92 to 1.03; 220 per 1000; 21 RCTs, 19,021 participants; high‐certainty evidence). It has little to no impact on need for invasive mechanical ventilation, or death (RR 1.03, 95% CI 0.97 to 1.11; 296 per 1000; 6 RCTs, 14,477 participants; high‐certainty evidence) and has no impact on whether participants are discharged from hospital (RR 1.00, 95% CI 0.97 to 1.02; 665 per 1000; 6 RCTs, 12,721 participants; high‐certainty evidence). Convalescent plasma may have little to no impact on quality of life (MD 1.00, 95% CI −2.14 to 4.14; 1 RCT, 483 participants; low‐certainty evidence). Convalescent plasma may have little to no impact on the risk of grades 3 and 4 adverse events (RR 1.17, 95% CI 0.96 to 1.42; 212 per 1000; 6 RCTs, 2392 participants; low‐certainty evidence). It has probably little to no effect on the risk of serious adverse events (RR 1.14, 95% CI 0.91 to 1.44; 135 per 1000; 6 RCTs, 3901 participants; moderate‐certainty evidence).
Convalescent plasma versus standard plasma
We are uncertain whether convalescent plasma reduces or increases all‐cause mortality at up to day 28 (RR 0.73, 95% CI 0.45 to 1.19; 129 per 1000; 4 RCTs, 484 participants; very low‐certainty evidence). We are uncertain whether convalescent plasma reduces or increases the need for invasive mechanical ventilation, or death (RR 5.59, 95% CI 0.29 to 108.38; 311 per 1000; 1 study, 34 participants; very low‐certainty evidence) and whether it reduces or increases the risk of serious adverse events (RR 0.80, 95% CI 0.55 to 1.15; 236 per 1000; 3 RCTs, 327 participants; very low‐certainty evidence). We did not identify any study reporting other key outcomes.
Convalescent plasma versus human immunoglobulin
Convalescent plasma may have little to no effect on all‐cause mortality at up to day 28 (RR 1.07, 95% CI 0.76 to 1.50; 464 per 1000; 1 study, 190 participants; low‐certainty evidence). We did not identify any study reporting other key outcomes.
Individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and mild disease
We identified two RCTs reporting on 536 participants, comparing convalescent plasma to placebo or standard care alone, and two RCTs reporting on 1597 participants with mild disease, comparing convalescent plasma to standard plasma.
Convalescent plasma versus placebo or standard care alone
We are uncertain whether convalescent plasma reduces all‐cause mortality at up to day 28 (odds ratio (OR) 0.36, 95% CI 0.09 to 1.46; 8 per 1000; 2 RCTs, 536 participants; very low‐certainty evidence). It may have little to no effect on admission to hospital or death within 28 days (RR 1.05, 95% CI 0.60 to 1.84; 117 per 1000; 1 RCT, 376 participants; low‐certainty evidence), on time to COVID‐19 symptom resolution (hazard ratio (HR) 1.05, 95% CI 0.85 to 1.30; 483 per 1000; 1 RCT, 376 participants; low‐certainty evidence), on the risk of grades 3 and 4 adverse events (RR 1.29, 95% CI 0.75 to 2.19; 144 per 1000; 1 RCT, 376 participants; low‐certainty evidence) and the risk of serious adverse events (RR 1.14, 95% CI 0.66 to 1.94; 133 per 1000; 1 RCT, 376 participants; low‐certainty evidence). We did not identify any study reporting other key outcomes.
Convalescent plasma versus standard plasma
We are uncertain whether convalescent plasma reduces all‐cause mortality at up to day 28 (OR 0.30, 95% CI 0.05 to 1.75; 2 per 1000; 2 RCTs, 1597 participants; very low‐certainty evidence). It probably reduces admission to hospital or death within 28 days (RR 0.49, 95% CI 0.31 to 0.75; 36 per 1000; 2 RCTs, 1595 participants; moderate‐certainty evidence). Convalescent plasma may have little to no effect on initial symptom resolution at up to day 28 (RR 1.12, 95% CI 0.98 to 1.27; 1 RCT, 416 participants; low‐certainty evidence). We did not identify any study reporting other key outcomes.
This is a living systematic review. We search monthly for new evidence and update the review when we identify relevant new evidence.
Authors' conclusions
For the comparison of convalescent plasma versus placebo or standard care alone, our certainty in the evidence that convalescent plasma for individuals with moderate to severe disease does not reduce mortality and has little to no impact on clinical improvement or worsening is high. It probably has little to no effect on SAEs. For individuals with mild disease, we have low certainty evidence for our primary outcomes. There are 49 ongoing studies, and 33 studies reported as complete in a trials registry. Publication of ongoing studies might resolve some of the uncertainties around convalescent plasma therapy for people with asymptomatic or mild disease.
Plain language summary
Is plasma from the blood of people who have recovered from COVID‐19 an effective treatment for other people with COVID‐19?
Key messages
• We are very confident that plasma from the blood of people who have recovered from COVID‐19 (convalescent plasma) has no benefits for the treatment of people with moderate to severe COVID‐19.
• Convalescent plasma may have little to no benefit for treating people with mild COVID‐19.
• We found 49 ongoing studies and 33 finished studies with unpublished results. We will update our review with evidence from these studies as soon as possible. New evidence may answer our remaining questions, especially for people with mild COVID‐19 or who have no symptoms.
What is convalescent plasma?
The body produces antibodies as one of its defences against infection. Antibodies are found in part of the blood called plasma. Plasma from people who have recovered from the COVID‐19 virus contains COVID‐19 antibodies, and it can be used to make convalescent plasma, which is plasma that contains these antibodies.
Convalescent plasma has been used successfully to treat some viruses. This treatment (given by a drip or injection) is generally well‐tolerated, but can cause unwanted effects.
What did we want to find out?
We wanted to find out whether convalescent plasma is an effective treatment for people with confirmed COVID‐19. We looked at:
• deaths from any cause after treatment with convalescent plasma;
• worsening of patients’ condition, measured by the number of people who needed support from a ventilator (a machine that helps people breathe if they cannot breathe on their own) or died; and improvement of patients’ condition, measured by participants discharged from hospital;
• quality of life; and
• unwanted effects.
What did we do?
We searched for studies that investigated convalescent plasma to treat people with COVID‐19. Studies could take place anywhere in the world and include people of any age, gender or ethnicity, with mild, moderate or severe COVID‐19.
Where possible we pooled (added up) the studies’ results to analyse them. We rated our confidence in the evidence, based on factors such as study methods and sizes.
What did we find?
We found 33 studies with 24,861 participants that investigated convalescent plasma. Among these, 29 studies included people with moderate to severe COVID‐19 and four studies included people with mild COVID‐19. Studies mainly took place in hospitals, in countries all over the world. The following findings apply to convalescent plasma compared with placebo (the same treatment but with no active ingredients) or standard care.
People with moderate to severe COVID‐19
• Convalescent plasma makes no difference to deaths from any cause at up to 28 days after treatment, about 225 in 1000 people died, compared to 220 in 1000 people who had been given convalescent plasma (21 studies, 19,021 people).
• Convalescent plasma makes little to no difference to needing invasive mechanical ventilation or dying. About 287 in 1000 people needed invasive mechanical ventilation support or died, compared to 296 in 1000 people given convalescent plasma (6 studies, 14,477 people). It makes no difference toparticipants being discharged from hospital. About 665 in 1000 people were discharged from hospital, compared to 665 in 1000 people given convalescent plasma (6 studies, 12,721 people).
• Convalescent plasma probably makes no difference to serious unwanted effects, about 118 in 1000 people may be at risk to have serious unwanted effects compared to 133 in 1000 people given convalescent plasma (6 studies, 4901 people).
• Convalescent plasma may result in no difference in quality of life (1 study, 483 people).
People with mild COVID‐19 • Convalescent plasma may result in no difference to deaths from any cause up to 28 days after treatment. About 22 in 1000 people given placebo or standard care died, compared to 9 in 1000 people given convalescent plasma (2 studies, 536 people).
• Convalescent plasma may result in no difference to admission to hospital or death within 28 days after treatment. About 112 in 1000 people given placebo or standard care were admitted to hospital or died, compared to 117 in 1000 people given convalescent plasma (1 study, 376 people).
• Convalescent plasma may result in no difference in the time until COVID‐19 symptoms resolved (1 study, 376 people).
• Convalescent plasma may result in no difference to serious unwanted effects.
What are the limitations of the evidence?
• We are very confident in the evidence for deaths from any cause, and worsening and improvement of patients’ condition in people with moderate to severe COVID‐19, as the results are consistent and are from many high‐quality studies.
• Our confidence in the other evidence for people with moderate and severe, and mild COVID‐19 is still limited, as we could not identify enough consistent results from a lot of studies.
• We still have little evidence on quality of life and for people with mild disease, and none for people without COVID‐19 symptoms.
How up to date is this evidence?
This is the fifth version of our review. The evidence is up to date to 03 March 2022.
Summary of findings
Summary of findings 1. Summary of findings table ‐ Convalescent plasma compared to placebo or standard care alone for individuals with moderate to severe disease.
| Convalescent plasma compared to placebo or standard care alone for individuals with moderate to severe disease | ||||||
| Patient or population: individualswith moderate to severe disease Setting: inpatient Intervention: convalescent plasma Comparison: placebo or standard care alone | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | № of participants (studies) | Certainty of the evidence (GRADE) | Comments | |
| Risk with placebo or standard care alone | Risk with convalescent plasma | |||||
| All‐cause mortality at up to day 28 ‐ total | 225 per 1000 | 220 per 1000 (207 to 232) | RR 0.98 (0.92 to 1.03) | 19021 (21 RCTs) | ⊕⊕⊕⊕ High | Convalescent plasma does not reduce all‐cause mortality at up to day 28. We did not identify any significant subgroup differences for the effect of convalescent plasma. |
| Clinical worsening: need for invasive mechanical ventilation, or death at up to day 28 | 287 per 1000 | 296 per 1000 (278 to 319) | RR 1.03 (0.97 to 1.11) | 14477 (6 RCTs) | ⊕⊕⊕⊕ High | Convalescent plasma has little to no impact on the need for invasive mechanical ventilation, or death at day 28. We identified significant subgroup differences for the effect of convalescent plasma with regard to antibodies in recipients detected at baseline (P = 0.02; 2 studies), participants' age (P = 0.03; 3 studies) and sex (P = 0.04; 3 studies). |
| Clinical improvement: participants discharged from hospital | 665 per 1000 | 665 per 1000 (645 to 678) | RR 1.00 (0.97 to 1.02) | 12721 (6 RCTs) | ⊕⊕⊕⊕ High | Convalescent plasma has no impact on whether participants are discharged from hospital. We identified significant subgroup differences for the effect of convalescent plasma with regard to antibodies in recipients detected at baseline (P = 0.04; 2 studies) and time since symptom onset (P = 0.09; 2 studies). |
| Quality of life, assessed with EQ‐5D‐5L‐questionnaire, at day 28 (0 indicates worst heath and 100 best health) | The mean quality of life, assessed with EQ‐5D‐5L‐questionnaire, at day 28 was 72 | MD 1 higher (2.14 lower to 4.14 higher) | ‐ | 483 (1 RCT) | ⊕⊕⊝⊝ Lowa | Convalescent plasma may have little to no impact on quality of life at up to day 28. We did not identify any significant subgroup differences. |
| Grades 3 and 4 adverse events follow‐up: 28 days | 181 per 1000 | 212 per 1000 (174 to 257) | RR 1.17 (0.96 to 1.42) | 2392 (6 RCTs) | ⊕⊕⊝⊝ Lowb,c | Convalescent plasma may have little to no impact on the risk of grades 3 and 4 adverse events. We identified significant subgroup differences for the effect of convalescent plasma with regard to country income groups (P = 0.07; 5 studies). |
| Serious adverse events ‐ total follow‐up: 28 days | 118 per 1000 | 135 per 1000 (108 to 170) | RR 1.14 (0.91 to 1.44) | 3901 (6 RCTs) | ⊕⊕⊕⊝ Moderatec | Convalescent plasma probably has little to no effect on the risk of serious adverse events. We did not identify any significant subgroup differences. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; RR: risk ratio | ||||||
| GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. | ||||||
| See interactive version of this table: https://gdt.gradepro.org/presentations/#/isof/isof_question_revman_web_423533352409190077. | ||||||
a Downgraded two levels for very serious imprecision, because of few participants and wide confidence interval. b Downgraded one level for serious indirectness, because some studies did not provide clear definitions of how the adverse events were measured and the number of events are inconsistent. c Downgraded one level for publication bias, because safety outcomes were assessed and reported in most studies for convalescent plasma group only.
Summary of findings 2. Summary of findings table ‐ Convalescent plasma compared to standard plasma for individuals with moderate to severe disease.
| Convalescent plasma compared to standard plasma for individuals with moderate to severe disease | ||||||
| Patient or population: individuals with moderate to severe disease Setting: inpatient Intervention: convalescent plasma Comparison: standard plasma | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | № of participants (studies) | Certainty of the evidence (GRADE) | Comments | |
| Risk with standard plasma | Risk with convalescent plasma | |||||
| All‐cause mortality at up to day 28 | 177 per 1000 | 129 per 1000 (80 to 210) | RR 0.73 (0.45 to 1.19) | 484 (4 RCTs) | ⊕⊝⊝⊝ Very lowa,b | We are uncertain whether convalescent plasma reduces or increases all‐cause mortality at up to day 28. We did not identify any significant subgroup differences. |
| Clinical worsening: need for invasive mechanical ventilation, or death at up to day 28 | 56 per 1000 | 311 per 1000 (16 to 1000) | RR 5.59 (0.29 to 108.38) | 34 (1 RCT) | ⊕⊝⊝⊝ Very lowc | We are uncertain whether convalescent plasma reduces or increases the risk of need for invasive mechanical ventilation, or death at day 28. We did not identify any significant subgroup differences. |
| Clinical improvement: participants discharged from hospital ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| Quality of life, assessed with standardised scales at longest follow‐up ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| Grades 3 and 4 adverse events follow‐up: 28 days | The identified study reported the number of participants experiencing any event of grade 3 (27/147 in CP group versus 17/72 in SP group), or grade 4 (26/147 in CP group versus 15/72 in SP group). | 248 (1 RCT) | ⊕⊝⊝⊝ Very lowd,e | We are uncertain whether convalescent plasma reduced or increases the risk for grades 3 and 4 adverse events. | ||
| Serious adverse events follow‐up: 28 days | 295 per 1000 | 236 per 1000 (162 to 340) | RR 0.80 (0.55 to 1.15) | 327 (3 RCTs) | ⊕⊝⊝⊝ Very lowf,g | We are uncertain whether convalescent plasma reduces or increases the risk of serious adverse events. We did not identify any significant subgroup differences. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RR: risk ratio; SMD: standardised mean difference | ||||||
| GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. | ||||||
| See interactive version of this table: https://gdt.gradepro.org/presentations/#/isof/isof_question_revman_web_423575562054934976. | ||||||
a Downgraded one level for serious inconsistency, because direction of effect is not consistent in all the studies. b Downgraded two levels for very serious imprecision, because of few participants and because optimal information size is not met at a power of 0.80. c Downgraded three levels due to extreme imprecision, because of extremely few participants, extremely few events, very wide confidence interval and optimal information size is not met at a power of 0.80. d Downgraded one level for serious indirectness, because definition of outcomes was different to the definition used in our review. e Downgraded two levels for very serious imprecision, because of few participants, and few events. f Downgraded two levels for very serious imprecision, because of few participants, and few events and optimal information size is not met at a power of 0.80. g Downgraded one level for publication bias, because safety outcomes were assessed and reported in most studies for convalescent plasma group only.
Summary of findings 3. Summary of findings table ‐ Convalescent plasma compared to human immunoglobulin for individuals with moderate to severe disease.
| Convalescent plasma compared to human immunoglobulin for individuals with moderate to severe disease | ||||||
| Patient or population: individuals with moderate to severe disease Setting: inpatient Intervention: convalescent plasma Comparison: human immunoglobulin | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | № of participants (studies) | Certainty of the evidence (GRADE) | Comments | |
| Risk with human immunoglobulin | Risk with convalescent plasma | |||||
| All‐cause mortality at up to day 28 | 433 per 1000 | 464 per 1000 (329 to 650) | RR 1.07 (0.76 to 1.50) | 190 (1 RCT) | ⊕⊕⊝⊝ Lowa | Convalescent plasma may have little to no effect on all‐cause mortality at up to day 28. |
| Clinical worsening: need for invasive mechanical ventilation, or death at up to day 28 ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| Clinical improvement: participants discharged from hospital ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| Quality of life ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| Grades 3 and 4 adverse events ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| Serious adverse events ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; OR: odds ratio; RR: risk ratio | ||||||
| GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. | ||||||
| See interactive version of this table: https://gdt.gradepro.org/presentations/#/isof/isof_question_revman_web_432005298021526745. | ||||||
a Downgraded by two levels for very serious imprecision, because of few events, few participants and because optimal information size is not met at a power of 0.90.
Summary of findings 4. Summary of findings table ‐ Convalescent plasma compared to placebo or standard care alone for individuals with mild disease.
| Convalescent plasma compared to placebo or standard care alone for individuals with mild disease | ||||||
| Patient or population: outpatients withmild disease Setting: outpatient Intervention: convalescent plasma Comparison: placebo or standard care alone | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | № of participants (studies) | Certainty of the evidence (GRADE) | Comments | |
| Risk with placebo or standard care alone | Risk with convalescent plasma | |||||
| All‐cause mortality at up to day 28 | 22 per 1000 | 8 per 1000 (2 to 32) | OR 0.36 (0.09 to 1.46) | 536 (2 RCTs) | ⊕⊝⊝⊝ Very lowa,b | We are uncertain whether or not convalescent plasma reduces all‐cause mortality at up to day 28. We did not identify any significant subgroup differences. |
| Admission to hospital or death within 28 days | 112 per 1000 | 117 per 1000 (67 to 206) | RR 1.05 (0.60 to 1.84) | 376 (1 RCT) | ⊕⊕⊝⊝ Lowb | Convalescent plasma may have little to no impact on admission to hospital or death within 28 days. |
| Symptom resolution‐ all initial symptoms resolved ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| Time to symptom resolution (absolute effect calculated for day 12) follow‐up: 60 days | Low | HR 1.05 (0.85 to 1.30) [symptoms resolution ] | 376 (1 RCT) | ⊕⊕⊝⊝ Lowb | Convalescent plasma may have little to no impact on time to symptom resolution. | |
| 500 per 1000 | 483 per 1000 (406 to 555) | |||||
| Quality of life, assessed with standardised scales at longest follow‐up ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| Grades 3 and 4 adverse events follow‐up: 28 days | 112 per 1000 | 144 per 1000 (84 to 245) | RR 1.29 (0.75 to 2.19) | 376 (1 RCT) | ⊕⊕⊝⊝ Lowb | Convalescent plasma may have little to no impact on the risk of grades 3 and 4 adverse events. |
| Serious adverse events follow‐up: 28 days | 117 per 1000 | 133 per 1000 (77 to 227) | RR 1.14 (0.66 to 1.94) | 376 (1 RCT) | ⊕⊕⊝⊝ Lowb | Convalescent plasma may have little to no impact on the risk of serious adverse events. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; HR: hazard Ratio; OR: odds ratio; RR: risk ratio; SMD: standardised mean difference | ||||||
| GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. | ||||||
| See interactive version of this table: https://gdt.gradepro.org/presentations/#/isof/isof_question_revman_web_423575913930302502. | ||||||
a Downgraded one level for serious indirectness, because outcome definition did not exactly match our definition (defined as death associated with COVID‐19). b Downgraded two levels for very serious imprecision, because of few participants, few events, wide confidence intervals and because optimal information size is not met for a power of 0.90.
Summary of findings 5. Summary of findings table ‐ Convalescent plasma compared to standard plasma for outpatients with mild disease.
| Convalescent plasma compared to standard plasma for outpatients withmild disease | ||||||
| Patient or population: outpatients withmild disease Setting: outpatient Intervention: convalescent plasma Comparison: standard plasma | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | № of participants (studies) | Certainty of the evidence (GRADE) | Comments | |
| Risk with standard plasma | Risk with convalescent plasma | |||||
| All‐cause mortality at up to day 28 | 5 per 1000 | 2 per 1000 (0 to 9) | OR 0.30 (0.05 to 1.75) | 1597 (2 RCTs) | ⊕⊝⊝⊝ Very lowa | We are uncertain whether or not convalescent plasma reduces all‐cause mortality at up to day 28. We did not identify any significant subgroup differences. |
| Admission to hospital or death within 28 days | 73 per 1000 | 36 per 1000 (23 to 55) | RR 0.49 (0.31 to 0.75) | 1595 (2 RCTs) | ⊕⊕⊕⊝ Moderateb | Convalescent plasma probably reduces admission to hospital or death at up to day 28. We identified significant subgroup differences for the effect of convalescent plasma with regard to the age of participants (P = 0.02; 2 studies). |
| Symptom resolution‐ all initial symptoms resolved follow‐up: 28 days | 657 per 1000 | 736 per 1000 (644 to 835) | RR 1.12 (0.98 to 1.27) | 416 (1 RCT) | ⊕⊕⊝⊝ Lowc | Convalescent plasma may have little to no effect on symptom resolution at up to day 28. We did not identify any significant subgroup difference. |
| Time to symptom resolution ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| Quality of life, assessed with standardised scales ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| Grades 3 and 4 adverse events ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| Serious adverse events ‐ not reported | ‐ | ‐ | ‐ | ‐ | ‐ | We did not identify any study reporting this outcome. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; OR: odds ratio; RR: risk ratio; SMD: standardised mean difference | ||||||
| GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. | ||||||
| See interactive version of this table: https://gdt.gradepro.org/presentations/#/isof/isof_question_revman_web_432020765287988004. | ||||||
a Downgraded by three levels for extremely serious imprecision, because few events, extremely wide confidence intervals and because optimal information size is not met for a power of 0.90. b Downgraded one level for serious imprecision, because optimal information size is not met for a power of 0.90. c Downgraded by two levels for very serious imprecision, because few events, very wide confidence intervals and because optimal information size is not met for a power of 0.90.
Background
‐
Description of the condition
The clinical syndrome coronavirus disease 2019 (COVID‐19) is a new, rapidly emerging zoonotic infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2; WHO 2020a). On 22 March 2020, the World Health Organization (WHO) declared the current COVID‐19 outbreak to be a pandemic, with the outbreak resulting in more than 535 million confirmed cases and over 6.3 million deaths worldwide as of June 2022 (WHO 2020b; WHO 2021a). Although there are similarities with historic coronavirus epidemics, with severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS) responsible for 813 and 858 deaths respectively, the scale and impact of the COVID‐19 pandemic present unprecedented challenges to health facilities and healthcare workers all over the world (WHO 2007; WHO 2019). Concurrently, new SARS‐CoV‐2 variants emerged, potentially having an effect on the transmission and characteristics of the disease, the effectiveness of vaccines and treatments, or on public health and social measures (WHO 2022b).
Early reports suggested case fatality rates between 0.7% and 4% (WHO 2020a; WHO 2020c). More recent reports estimate wide‐ranging case fatality rates, as low as 0. 1% in Singapore and up to 5.2% in Peru (Johns Hopkins 2022). However, these numbers should be interpreted with great care due to testing availability, underreporting of cases and delays from confirmation of a case to time of death (Kim 2020), ethnicity, underlying health conditions, access to health care, and socioeconomic status (Williamson 2020).
The median incubation period and time to symptom onset of SARS‐CoV‐2 was reported to be five days, with 97.5% of cases developing symptoms within 11.5 days of infection (Brandal 2021; Lauer 2020). However, the median incubation time depends on the virus variant and is estimated to be only three days (range: 0 to 8 days) in case of the Omicron variant, which is shorter compared to the estimate for the Delta variant, for instance, or other previous variants (Brandal 2021). Common signs and symptoms can include fever, dry cough, fatigue and sputum production (WHO 2020a). Postviral olfactory dysfunction is reported in 5% to 85% of cases, with loss of both smell and taste reported (Izquierdo‐Dominguez 2020). Other less commonly reported signs and symptoms are shortness of breath, sore throat, headache, myalgia or arthralgia, chills, nausea or vomiting, nasal congestion, diarrhoea, haemoptysis and conjunctival congestion (WHO 2020a). Of the reported cases, 80% are estimated to have a mild or asymptomatic course of infection, and an estimated 5% of cases are admitted to the intensive care unit (ICU) with acute respiratory distress syndrome (ARDS), septic shock, multiple organ failure, or all three conditions (Team 2020; WHO 2020a). A risk factor for developing infection and progressing to severe disease is old age, with people aged over 80 years at highest risk of mortality. Other risk factors are cardiovascular disease, obesity, hypertension, diabetes, chronic respiratory disease, cancer and compromised immune status (Chen 2020a; Huang 2020; Liang 2020; WHO 2020a; Wu 2020a). Individuals who are immune‐compromised are at higher risk of adverse outcomes from COVID‐19, including people who have received recent chemotherapy and chemoimmunotherapy, recipients of solid organ and allogeneic transplants and individuals with haematological malignancies and solid organ cancer (Chavez‐MacGregor 2022; Fung 2020).
SARS‐CoV‐2 is a positive‐sense, single‐stranded ribonucleic acid (RNA) virus with a large genome. There are indications that the virus is capable of inducing an excessive immune reaction in the host, with highly activated but decreased numbers of CD4+ and CD8+ T cells detected in the peripheral blood of people with COVID‐19 (Xu 2020a). Early reports also showed that people critically ill with COVID‐19 frequently exhibit a hypercoagulable state and endothelial inflammation, which is hypothesised to lead to the high burden of thromboembolic events seen in this population (Driggin 2020). SARS‐CoV‐2 binds to the angiotensin‐converting enzyme 2 (ACE2). ACE2 is a protein that functions as the receptor, facilitating entry of SARS‐CoV‐2 into the host cell, and is most abundant on type II alveolar cells in the lungs (Tolouian 2020; Van de Veerdonk 2020).
Widespread availability of vaccines against SARS‐CoV‐2 has reduced infections and the risk of severe disease. However, individuals who are immunocompromised have reduced rates of seroconversion following vaccination and therefore remain susceptible to severe COVID‐19 and in need of effective therapies (Lee 2022; Teh 2022).
Description of the intervention
Convalescent plasma and convalescent serum prepared from convalescent plasma are interventions that have been used in the past to treat conditions when no vaccine or pharmacological interventions were available. Diphtheria, pneumococcal pneumonia, hepatitis A and B, mumps, polio, measles, and rabies are conditions where convalescent plasma has been shown to be effective (Eibl 2008).
A systematic review has shown that convalescent plasma may have clinical benefit for people with influenza and SARS (Mair‐Jenkins 2015). This systematic review included observational studies and randomised controlled trials (RCTs) that investigated the use of convalescent plasma or serum for treating severe acute respiratory infections of laboratory‐confirmed or suspected viral aetiology, and included investigations with patients of any age and sex. Control interventions consisted of sham or placebo therapy and no therapy. Although the included studies were generally small and of low quality, with a moderate to high risk of bias, the review authors concluded that the use of convalescent plasma may reduce mortality, and appears safe (Mair‐Jenkins 2015). The authors also suggested that the effectiveness of convalescent plasma in reducing hospital length of stay is dependent on early administration of the therapy, and use as prophylaxis is more likely to be beneficial than treating severe disease. However, the optimal timing and dosage of convalescent plasma therapy are unknown.
There is conflicting evidence about the effect of convalescent plasma for treating severe acute respiratory infections. For instance, studies that investigated the effectiveness of immune plasma for influenza showed no benefit (Beigel 2017; Beigel 2019).
Although convalescent plasma is generally thought to be a well‐tolerated therapy, adverse events can occur. Limited information is available about specific adverse events related to convalescent plasma therapy, but symptoms that have been reported are similar to those for other types of plasma blood components, including fever or chills, allergic reactions, and transfusion‐related acute lung injury (TRALI; Beigel 2019; Chun 2016; Luke 2006). Furthermore, the transfer of coagulation factors present in plasma products is potentially harmful for people with COVID‐19, who are already at an increased risk of thromboembolic events (Driggin 2020). Plasma transfusions are also known to cause transfusion‐associated circulatory overload (TACO). TACO and TRALI are especially important to consider, because COVID‐19 patients with comorbidities, who might be eligible for experimental treatment with convalescent plasma therapy, are at an increased risk of these adverse events. Risk‐mitigation strategies can be implemented to prevent TRALI. These include limiting donations from female donors, especially those with a history of pregnancy, and screening donors for antibodies that are implicated in TRALI (Otrock 2017). In addition to the aforementioned adverse events, transfusion‐transmitted infections, red blood cell allo‐immunisation and haemolytic transfusion reactions have also been described following plasma transfusion, although they are less common (Pandey 2012). Pathogen inactivation can be implemented to decrease the risk of transmitting infections by transfusion (Rock 2011), with likely preservation of neutralising antibodies (Focosi 2021; Kostin 2021; Larrea 2022).
A theoretical risk related to virus‐specific antibodies, which are transferred with convalescent plasma administration, is antibody‐dependent enhancement of infection (Morens 1994). Here, virus‐binding antibodies facilitate the entry and replication of virus particles into monocytes, macrophages and granulocytic cells and thereby increase the risk of more severe disease in the infected host. Although antibody‐dependent enhancement has not been demonstrated in COVID‐19, it has been seen with previous coronavirus infections when the antibodies given targeted a different serotype of the virus (Wan 2020; Wang 2014). A mechanism for antibody‐dependent enhancement in COVID‐19 has recently been proposed, with non‐neutralising antibodies to variable S domains potentially enabling an alternative infection pathway via Fc receptor‐mediated uptake (Ricke 2020). Antibody‐dependent enhancement is therefore a potentially harmful consequence of convalescent plasma therapy for COVID‐19.
In summary, the benefits of the intervention for convalescent plasma should be carefully considered in view of the risks of adverse events.
How the intervention might work
Convalescent plasma contains pathogen‐specific neutralising antibodies, which can neutralise viral particles and may confer passive immunity to recipients. The duration of conferred protection can differ depending on the timing of administration, ranging from weeks to months after treatment (Casadevall 2020).
By neutralising SARS‐CoV‐2 particles, early treatment with convalescent plasma is postulated to increase the patient’s own capacity to clear the initial inoculum (Casadevall 2020; Robbins 1995). This could lead to a reduction in mortality and fewer hospitalised patients progressing to the ICU. Furthermore, convalescent plasma may reduce the length of ICU stay in critically ill patients (Mair‐Jenkins 2015), thus helping to lift pressure from global healthcare systems and increasing ICU capacity.
Although, initially re‐infection was not thought to be likely (Bao 2020a; Wu 2020b), this has changed since the end of 2021 when Omicron variants of SARS‐CoV‐2 became dominant (ONS 2022).
This implies that convalescent plasma from people who have recovered from SARS‐CoV‐2 infection may be capable of conferring passive immunity. Retrospective studies also observed a potential correlation between the level of antibody titres in convalescent plasma and recovery after treatment (Joyner 2021; Shen 2020). It is important to note, however, that research in other coronavirus species has shown that immunity may not be long‐lasting, with two to three years of protection estimated from work with SARS and MERS (Mo 2006; Payne 2016). Furthermore, there are indications that the severity of infection has an impact on antibody titres, with less‐severe disease leading to lower neutralising antibody response in people with SARS and COVID‐19 (Ho 2005; Zhao 2020). It is unclear exactly how often reinfection occurs, with the burden of reinfection likely to be underestimated, while at the same time a number of case reports of severe reinfection have been published (Iwasaki 2021).
Why it is important to do this review
There is an ongoing need for information to guide clinical decision making for COVID‐19 patients. Pharmacological treatment options have been and are continuing to be investigated in many ongoing trials, with treatment guidelines updated accordingly (WHO 2022a). Current treatments that have been shown to be effective in non‐severe COVID‐19 at risk of hospitalisation include nirmatrelvir‐ritonavir, remdesivir and molnupiravir (WHO 2022a), and in severe or critical COVID‐19 systemic corticosteroids, interleukin‐6 receptor blockers (tocilizumab and sarilumab), and baracitinib (Janus kinase inhibitor) (Beigel 2020; Horby 2020; Horby 2021a; WHO 2022a). Current treatment also consists of supportive care with extracorporeal membrane oxygenation in severe cases and oxygen supply in less severe cases (CDC 2020; WHO 2020d). Although the risk of hospitalisation and death has decreased during the course of the pandemic with emergence of different variants including omicron (Adjei 2022; Nyberg 2022; Ulloa 2022), even with current therapies, patients with COVID‐19 remain at increased risk of mortality, particularly older individuals and those with comorbidities (Adjei 2022).
Currently approved and available COVID‐19 vaccines have been shown to be well‐tolerated and effective (CDC 2022). They can prevent transmission to those who are at risk for becoming seriously ill and reduce the risk for developing severe disease (CDC 2022). However, even with the available effective vaccines, not everyone can be effectively vaccinated; for example, people who are temporarily or permanently immune‐compromised. Convalescent plasma can be prepared and made rapidly available by blood banks and hospitals when enough potential donors have recovered from the infection, using readily available materials and methods (Bloch 2020). However, its safety and efficacy are not well‐characterised, and there are costs associated with pursuing the use of convalescent plasma for treatment of COVID‐19.
A multitude of clinical trials investigating the safety and effectiveness of convalescent plasma have been announced, and their results will need to be interpreted with care. Thus, there needs to be a thorough understanding of the current body of evidence regarding the use of convalescent plasma for people with COVID‐19, and an extensive review of the available literature is required.
Objectives
To assess the effectiveness and safety of convalescent plasma transfusion in the treatment of people with COVID‐19; and to maintain the currency of the evidence using a living systematic review approach.
Methods
Criteria for considering studies for this review
Types of studies
The main description of methods is based on a template from Cochrane Haematology and is in line with a series of Cochrane Reviews investigating treatments and therapies for COVID‐19. The protocol for this review was registered with the Center for Open Science on 17 April 2020 (Piechotta 2020a). Amendments that have been made since are summarised in Differences between protocol and review and Table 6.
1. Summary of PICO development from protocol stage to current review version.
| Publication date | Participants | Interventions | Comparators | Outcomes | Study designs | |
|
Protocol Piechotta 2020a |
17 April 2020 |
Inclusion
Exclusion
|
Inclusion
|
Inclusion
|
All criteria based on COMET Initiative for COVID‐19 patients (COMET 2020) Primary outcomes
Secondary outcomes
|
Planned inclusion priority, determined by availability of sufficient evidence
|
|
Version 1 Valk 2020 |
14 May 2020 | See above |
Inclusion
Exclusion
|
See above | All criteria based onCOMET Initiative for COVID‐19 patients (COMET 2020) Primary outcomes
Secondary outcomes
|
Inclusion
No evidence available for
|
| Changesb | None | Added exclusion criteria
|
None | Revised secondary outcome "Improvement of clinical symptoms, assessed through need for respiratory support":
|
None | |
|
Version 2 Piechotta 2020b |
10 July 2020 | See above | See above | Inclusion
|
All criteria based on COMET Initiative for COVID‐19 patients (COMET 2020) Primary outcomes
Secondary outcomes
|
Inclusion
Further inclusion
According to originally planned inclusion priorities |
| Changesb | None | None | Added eligible control treatment
|
Added a secondary outcome:
|
Added inclusion criteria for safety data
|
|
|
Version 3 Chai 2020 |
12 October 2020 | See above | See above | See above | All criteria based on COMET Initiative for COVID‐19 patients (COMET 2020) Primary outcomes
Secondary outcomes
|
Inclusion
Further inclusion
According to originally planned inclusion priorities Exclusion
|
| Changesb | None | None | None | Revised and renamed secondary outcome “Improvement of clinical symptoms”
Added secondary outcome:
|
Added exclusion criteria:
|
|
| Piechotta 2021 |
Inclusion
Exclusion
|
See above |
Inclusion
|
All criteria based on COMET Initiative for COVID‐19 patients (COMET 2020), and outcomes prioritised by consumer representatives, referees of previous versions of this review, and the German guideline panel for inpatient therapy of people with COVID‐19. Individuals with a confirmed diagnosis of COVID‐19 and moderate to severe diseaseEffectiveness of convalescent plasma Prioritised outcomes
Additional outcomes
Safety of convalescent plasma
Individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease Effectiveness of convalescent plasma Prioritised outcomes
Additional outcomes
Safety of convalescent plasma
|
Inclusion
|
|
| Changesb | Introduced separate populationsc
|
None | Added eligible control treatment
Added specifications on placebo treatment
|
Changed primary and secondary outcomes to prioritised (included in 'Summary of findings' table) and additional outcomes (not included in 'Summary of findings' table). Revised and specified outcomes per population. Individuals with moderate to severe disease
Added outcomes for individuals with asymptomatic or mild disease |
Added inclusion criteria
Added exclusion criteria
|
|
| Version 5 | (Current version) | See above |
Inclusion
Exclusion
|
See above |
Individuals with a confirmed diagnosis of COVID‐19 and moderate to severe diseaseEffectiveness of convalescent plasma Primary outcomes
Secondary outcomes
Individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease Effectiveness of convalescent plasma Primary outcomes
Secondary outcomes
|
Inclusion
|
| Changesb | None | Removed inclusion criteria:
|
None | We renamed 'Prioritised outcomes' to 'Primary outcomes' and 'Additional outcomes' to 'Secondary outcomes'. We revised and redefined outcomes for individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease:
We revised and redefined outcomes for individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease.
|
Added exclusion criteria
|
|
| AE: adverse event; CBA: controlled before‐and‐after; COMET: Core Outcome Measures in Effectiveness Trials; ECMO: extracorporeal membrane oxygenation; ICU: intensive care unit;IMV: invasive mechanical ventilation; ITS: interrupted time series; MV: mechanical ventilation; NIV: non‐invasive ventilation; NRSI: non‐randomised studies of interventions; RCT: randomised controlled trial; RT‐PCR: reverse‐transcription polymerase chain reaction; SAE: serious adverse event; TACO: transfusion‐associated circulatory overload;TAD: transfusion‐associated dyspnoea; TRALI: transfusion‐related acute lung injury; WHO: World Health Organization; WHOQOL‐100: World Health Organization quality‐of‐life scale | ||||||
aIncluding changes in study designs and methodology. bChanges in PICO compared to the previously published version. cAccording to the latest WHO clinical progression score (WHO 2020e).
To assess the benefits and safety of convalescent plasma therapy for the treatment of COVID‐19, we included RCTs, as such studies, if performed appropriately, give the best evidence for experimental therapies in highly controlled therapeutic settings. We used the methods recommended by the Cochrane Handbook for Systematic Reviews of Interventions (Lefebvre 2022a). If we had identified non‐standard RCT designs, such as cluster‐randomised trials and cross‐over trials, we would have considered only the results from the first cycle of cross‐over RCTs.
We included full‐text publications, preprint articles, abstract publications, and results published in trials registries, if sufficient information was available on study design, characteristics of participants, interventions and outcomes. We did not apply any limitation with respect to the length of follow‐up.
Types of participants
We included individuals with a confirmed diagnosis of COVID‐19, with no age, gender or ethnicity restrictions.
We included trials that included participants with any disease severity. We performed separate analyses for populations with ambulatory mild disease and for hospitalised participants with moderate to severe disease, according to the latest WHO clinical progression score (see Table 7; WHO 2020e).
2. World Health Organization clinical progression scale a.
| Patient state | Descriptor | Score |
| Uninfected | Uninfected; no viral RNA detected | 0 |
| Ambulatory mild disease | Asymptomatic; viral RNA detected | 1 |
| Symptomatic; independent | 2 | |
| Symptomatic; assistance needed | 3 | |
| Hospitalised: moderate disease | Hospitalised; no oxygen therapyb | 4 |
| Hospitalised; oxygen by mask or nasal prongs | 5 | |
| Hospitalised: severe disease | Hospitalised; oxygen by non‐invasive mechanical ventilation or high flow | 6 |
| Intubation and mechanical ventilation; pO2/FiO2 ≥ 150 or SpO2/FiO2 ≥ 200 | 7 | |
| Invasive mechanical ventilation; pO2/FiO2 < 150 (SpO2/FiO2 < 200) or vasopressors | 8 | |
| Invasive mechanical ventilation; pO2/FiO2 < 150 and vasopressors, dialysis or ECMO | 9 | |
| Dead | Dead | 10 |
| ECMO: extracorporeal membrane oxygenation; FiO2:fraction of inspired oxygen; pO2: partial pressure of oxygen; SpO2: oxygen saturation | ||
aWorld Health Organization (WHO) clinical progression scale (WHO 2020e). bIf hospitalised for isolation only, record status as for ambulatory patient.
We excluded studies that included populations with other coronavirus diseases (SARS or MERS). We also excluded studies that included populations with mixed viral diseases (e.g. influenza), unless the trial authors provided subgroup data for people with COVID‐19.
Types of interventions
We included the following intervention.
Convalescent plasma from people who had recovered from SARS‐CoV‐2 infection
We did not include studies on standard immunoglobulin as intervention.
We included the following comparisons for the control arm.
Convalescent plasma therapy versus control treatment, for example, drug treatments (including but not limited to hydroxychloroquine, remdesivir), standard immunoglobulin. Co‐interventions were allowed, but must have been comparable between intervention groups.
Convalescent plasma versus standard care or placebo (i.e. saline solution)
Convalescent plasma versus standard plasma (i.e. fresh frozen plasma)
Types of outcome measures
We evaluated core outcomes as predefined by the Core Outcome Measures in Effectiveness Trials (COMET) Initiative for COVID‐19 patients (COMET 2020), and additional outcomes that have been prioritised by consumer representatives, referees of previous versions of this review, and the German guideline panel for inpatient therapy of people with COVID‐19.
We defined outcome sets for two populations: individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease, and individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease, according to the WHO clinical progression scale (WHO 2020e).
We assessed disease severity with need for respiratory support according to the WHO clinical progression scale (WHO 2020e).
Timing of outcome measurement
For time‐to‐event outcomes, such as mortality, discharge from hospital, and improvement of clinical symptoms, we included outcome measures representing the longest follow‐up time available.
We included all other outcome categories for the observational periods that the study publications reported. We included those adverse events occurring during active treatment and had planned to include long‐term adverse events as well. If sufficient data had been available, we planned to group the measurement time points of eligible outcomes, for example, adverse events and serious adverse events, into those measured directly after treatment (up to seven days after treatment), medium‐term outcomes (15 days after treatment) and longer‐term outcomes (over 30 days after treatment).
Primary outcomes
These critical outcomes will be included in the summary of findings tables.
Individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease
All‐cause mortality at day 28, day 60, time to death and during hospital stay
-
Clinical status, at day 28, day 60, and up to the longest follow‐up, including the following:
worsening of clinical status: participants with clinical deterioration (new need for invasive mechanical ventilation) or death;
improvement of clinical status: participants discharged from hospital. Participants should be discharged without clinical deterioration
Quality of life, including fatigue and neurological status, assessed with standardised scales (e.g. WHOQOL‐100, a standardised scale for assessing quality of life) at up to 7 days, up to 28 days, and longest follow‐up available
Adverse events (any grade, grades 1‐2, grades 3‐4), defined as the number of participants with any event and including potential relationship between intervention and adverse reaction (e.g. TRALI, transfusion‐transmitted infection, TACO, transfusion‐associated dyspnoea (TAD), acute transfusion reactions, headache, thromboembolic events)
Serious adverse events, defined as the number of participants with any serious adverse event (serious as defined according to CTCAE (Common Terminology Criteria for Adverse Events))
Individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease
All‐cause mortality at day 28, day 60, time to event, and at the longest follow‐up
Admission to hospital or death within 28 days
-
Symptom resolution:
all initial symptoms resolved (asymptomatic) at day 14, day 28, and up to the longest follow‐up;
time to symptom resolution
Quality of life, including fatigue and neurological status, assessed with standardised scales (e.g. WHOQOL‐100) at up to 7 days, up to 28 days, and longest follow‐up available;
Adverse events (any grade, grades 1‐2, grades 3‐4)
Serious adverse events, defined as the number of participants with any serious adverse event (serious as defined according to CTCAE (Common Terminology Criteria for Adverse Events))
Secondary outcomes
These outcomes will not be included in the summary of findings tables.
Individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease
-
Improvement of clinical status, at day 28 and up to the longest follow‐up, including:
weaning or liberation from invasive mechanical ventilation in surviving participants;
ventilator‐free days (defined as days alive and free from mechanical ventilation);
liberation from supplemental oxygen in surviving participants
Need for dialysis at up to 28 days
Admission to the ICU on day 28
Duration of hospitalisation
Viral clearance, assessed with reverse transcription polymerase chain reaction (RT‐PCR) test for SARS‐CoV‐2 at baseline, and up to 3, 7, and 14 days.
Individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease
-
Worsening of clinical status, at day 28 and up to the longest follow‐up, including (moderate to severe COVID‐19 symptoms):
need for hospitalisation with oxygen by mask or nasal prongs, or death;
need for invasive mechanical ventilation, or death
Viral clearance, assessed with RT‐PCR for SARS‐CoV‐2 at baseline, and up to 3, 7, and 14 days
Search methods for identification of studies
We carried out weekly searches until 11 August 2021 and from August 2021 onwards, we carried out monthly searches for completed and ongoing studies. In order to limit language bias, studies reported in all languages were eligible. We checked review search methods and strategies approximately monthly to ensure they reflected any terminology changes in the topic area, or in the databases. We adapted the strategy where necessary.
Electronic searches
We designed and tested search strategies for electronic databases according to methods suggested in the Cochrane Handbook for Systematic Reviews of Interventions (Lefebvre 2022a). One review author (CD) developed the original strategies and Cochrane Haematology's Information Specialist (IM) peer‐reviewed and revised them at various times, to reflect the current state of knowledge. In this emerging field, we expected that at least study abstracts would be in English. If studies were published in other languages than those our review team could accommodate (English, Dutch, German, French, Italian, Malay and Spanish), we involved Cochrane TaskExchange to identify people within Cochrane to translate these studies.
As publication bias might influence all subsequent analyses and conclusions, we searched all potentially relevant trials registries in detail to detect ongoing studies as well as studies that had been completed but not yet published. It is mandatory to provide trial results in the trials registry, so we planned to extract and analyse these data, in case results were not published elsewhere. However, no outcome data have yet been added to the trials registries.
We searched the following databases and sources from 1 January 2019 to 03 March 2022.
-
Databases of medical literature (Appendix 1):
MEDLINE Ovid (1 January 2020 to 02 March 2022);
Embase Ovid (1 January 2020 to 02 March 2022);
Cochrane COVID‐19 Study Register (covid-19.cochrane.org; inception to 03 March 2022)*
PubMed (for epublications ahead of print only; 1 January 2020 to 03 March 2022)
World Health Organization COVID‐19 Global literature on coronavirus disease (bvsalud.org/global-literature-on-novel-coronavirus-2019-ncov; inception to 03 March 2022) without references of MEDLINE and PubMed)
Epistemonikos, L*OVE List Coronavirus disease (COVID‐19) (app.iloveevidence.com; inception to 03 March 2022)
*The Cochrane COVID‐19 Study Register is a specialised register built within the Cochrane Register of Studies (CRS) and is maintained by Cochrane Information Specialists. Complete data sources and search methods for the register are available at community.cochrane.org/about-covid-19-study-register. The register contains study reports from several sources, including:
weekly searches of PubMed;
daily searches of ClinicalTrials.gov;
weekly searches of Embase.com;
weekly searches of the WHO International Clinical Trials Registry Platform (ICTRP);
monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL).
Searching other resources
We handsearched the reference lists of all identified studies, relevant review articles and current treatment guidelines for further literature. We also contacted experts in the field, drug manufacturers and regulatory agencies in order to retrieve information on unpublished studies.
Data collection and analysis
Selection of studies
Using Covidence software, two review authors (from among SJV, KLC, VP, CK, CI and NS) independently screened the results of the search strategies for eligibility, by reading the abstracts. We coded the abstracts as either 'retrieve' or 'do not retrieve'. In the case of disagreement, or if it was unclear whether we should retrieve the abstract or not, we obtained the full‐text publication for further discussion. Two review authors assessed the full‐text articles of selected studies. If the two review authors were unable to reach a consensus, they consulted a third review author to reach a final decision.
We documented the study selection process in a flow chart, as recommended in the PRISMA statement (Moher 2009), and show the total numbers of retrieved references and the numbers of included and excluded studies. We list all studies that we excluded after full‐text assessment and the reasons for their exclusion in the Characteristics of excluded studies table.
Data extraction and management
Two review authors (from among CI, NK, and EA) independently assessed eligible studies obtained in the process of study selection (as described above) for methodological quality and risk of bias. If the review authors were unable to reach a consensus, we consulted a third review author.
Two review authors (from among CI, NK, NC, and EA) extracted data using a customised data extraction form, developed in Microsoft Excel (Microsoft Corporation 2018). Another review author (CI, NK, or NS) verified the accuracy and (where applicable) the plausibility of extractions and assessment. We conducted data extraction according to the guidelines proposed by Cochrane (Li 2022). If the review authors were unable to reach a consensus, we consulted a third review author.
We collated multiple reports of one study so that the study, and not the report, is the unit of analysis.
We extracted the following information.
General information: author, title, source, publication date, country, language, duplicate publications
Quality assessment: study design, bias arising from the randomisation process, due to deviations from the intended interventions, due to missing outcome data, in measurement of the outcome, and in selection of the reported results
Study characteristics: trial design, setting and dates, source of participants, inclusion/exclusion criteria, comparability of groups, treatment cross‐overs, compliance with assigned treatment, length of follow‐up
Participant characteristics: age, sex, ethnicity, number of participants recruited/allocated/evaluated, disease, severity of disease, additional diagnoses, previous treatments (e.g. experimental drug therapies, oxygen therapy, ventilation), whether the donors were tested by nasal swabs or whether the plasma was tested
-
Interventions: convalescent plasma therapy, concomitant therapy, duration of follow‐up, donors' disease severity, how donations were tested for neutralising antibody
For studies that included a control group: comparator (type)
Outcomes: as specified in Types of outcome measures
Assessment of risk of bias in included studies
We used the RoB 2 to analyse the risk of bias in the underlying study results (Sterne 2019). Of interest for this review was the effect of the assignment to the intervention (the intention‐to‐treat (ITT) effect) and we performed all assessments with RoB 2 on this effect. The outcomes that we addressed are those specified for inclusion in Summary of findings table 1. Accordingly, the outcomes had been prioritised according to the COMET Initiative for COVID‐19 patients (COMET 2020).
Two review authors (from among CI, VP and EA) independently assessed the risk of bias for each study result. In case of discrepancies among their judgements or inability to reach consensus, we consulted a third review author to reach a final decision. We assessed the following types of bias as outlined in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022a).
Bias arising from the randomisation process
Bias due to deviations from the intended interventions
Bias due to missing outcome data
Bias in measurement of the outcome
Bias in selection of the reported result
For cluster‐RCTs, we had planned to add an additional domain to assess bias arising from the timing of identification and recruitment of participants in relation to timing of randomisation, as recommended in the archived RoB 2 guidance for cluster‐randomised trials (Eldridge 2016), and in Chapter 23 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022b).
To address these types of bias we used the signalling questions recommended in RoB 2 and made a judgement using the following options:
'yes': if there is firm evidence that the question is fulfilled in the study (i.e. the study is at low or high risk of bias given the direction of the question);
'probably yes': a judgement has been made that the question is fulfilled in the study (i.e. the study is at low or high risk of bias given the direction of the question);
'no': if there is firm evidence that the question is unfulfilled in the study (i.e. the study is at low or high risk of bias for the given the direction of the question);
'probably no': a judgement has been made that the question is unfulfilled in the study (i.e. the study is at low or high risk of bias given the direction of the question);
'no information': if the study report does not provide sufficient information to allow any judgement.
We used the algorithms proposed by RoB 2 to assign each domain one of the following levels of bias:
low risk of bias;
some concerns;
high risk of bias.
Subsequently, we derived a risk of bias rating for each prespecified outcome in each study in accordance with the following suggestions.
'Low risk of bias': we judged the trial to be at low risk of bias for all domains for this result.
'Some concerns': we judged the trial to raise some concerns in at least one domain for this result, but not to be at high risk of bias for any domain.
'High risk of bias': we judged the trial to be at high risk of bias in at least one domain for the result or we judged the trial to have some concerns for multiple domains in a way that substantially lowers confidence in the results.
We used the RoB 2 Excel tool to implement RoB 2 (available on the riskofbiasinfo.org website), added our judgements to the analysis for each assessed study and outcome, and stored our detailed RoB 2 assessments as supplementary online material. We used the overall risk of bias judgement, derived from the RoB 2 Excel tool, to inform our GRADE decision on downgrading for risk of bias.
Measures of treatment effect
For continuous outcomes, we recorded the mean, standard deviation and total number of participants in both the treatment and control groups. For dichotomous outcomes, we recorded the number of events and total number of participants in both the treatment and control groups.
For continuous outcomes using the same scale, we performed analyses using the mean difference (MD) with 95% confidence intervals (CIs). For continuous outcomes measured with different scales, we performed analyses using the standardised mean difference (SMD). For interpreting SMDs, we re‐expressed SMDs in the original units of a particular scale with the most clinical relevance and impact.
If available, we extracted and reported hazard ratios (HRs) for time‐to‐event outcomes (e.g. time to symptom resolution). If HRs were not available, we made every effort to estimate the HR as accurately as possible using the available data and a purpose‐built method based on the Parmar and Tierney approach (Parmar 1998; Tierney 2007). If sufficient studies provided HRs, we used HRs rather than risk ratios (RRs) or MDs in a meta‐analysis.
For dichotomous outcomes, we planned to report the pooled RR with a 95% CI (Deeks 2022). If the number of observed events had been small (less than 5% of sample per group), and if studies had balanced treatment groups, we planned to report the Peto odds ratio (OR) with 95% CI (Deeks 2022).
Unit of analysis issues
As recommended in Chapter 6 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022c), for studies with multiple treatment groups, we planned to combine arms if they could be regarded as subtypes of the same intervention.
When arms could not be pooled this way, we planned to compare each arm with the common comparator separately. For pair‐wise meta‐analysis, we planned to split the ‘shared’ group into two or more groups with smaller sample sizes, and include two or more (reasonably independent) comparisons. For this purpose, for dichotomous outcomes, both the number of events and the total number of participants would be divided up, and for continuous outcomes, the total number of participants would be divided up with unchanged means and standard deviations (SDs).
Dealing with missing data
Chapter 6 of the Cochrane Handbook for Systematic Reviews of Interventions suggests a number of potential sources for missing data, which we needed to take into account: at study level, at outcome level and at summary data level (Higgins 2022c). In the first instance, it is of the utmost importance to differentiate between data 'missing at random' and 'not missing at random'.
We handled missing data by conducting an available‐case analysis and extracted all the available data. Further, we requested missing data from the study authors. For the previous update version, we contacted 11 principal investigators from included studies (Agarwal 2020; AlQahtani 2021; Avendano‐Sola 2021; Bajpai 2020; Gharbharan 2021; Hamdy Salman 2020; Horby 2021b; Li 2020; Libster 2020; Ray 2022; Simonovich 2020). We received six responses: one each from Agarwal 2020; AlQahtani 2021; Avendano‐Sola 2021; Gharbharan 2021; Horby 2021b and Li 2020, providing all requested information. For this current update version, we contacted another 11 principal investigators from included studies (Baldeon 2022; Bar 2021; Bennett‐Guerrero 2021; De Santis 2022; Devos 2021; Holm 2021; Koerper 2021; Menichetti 2021; NCT04421404; Ortigoza 2022; Sekine 2021). We received three responses (De Santis 2022; Baldeon 2022; Bennett‐Guerrero 2021).
We further contacted all principal investigators (where contact information was available) from ongoing studies, asking for their prospective completion dates, as well as completed or terminated studies without published results, and invited them to share their data with us for this update. We received a response from one study (NCT04374526), informing us that their trial was completed, but that the results will not be published soon and that they are willing to share their data for this update (no data were received from the investigators before submission of our review).
Assessment of heterogeneity
We assessed heterogeneity of treatment effects between trials using a Chi2 test with a significance level at P < 0.1, and visual examination. We used the I2 statistic (Higgins 2003), to quantify possible heterogeneity (I2 > 30% to signify moderate heterogeneity, I2 > 75% to signify considerable heterogeneity; Deeks 2022). If heterogeneity had been above 80%, we would have explored potential causes through sensitivity and subgroup analyses. If we had not found a reason for heterogeneity, we would not have performed a meta‐analysis, but would have only commented on results from all studies and presented these in tables.
Assessment of reporting biases
As mentioned above, we searched trials registries to identify completed studies that have not been published elsewhere, to minimise or determine publication bias. We included studies irrespective of their publication status, as recommended in the Cochrane Handbook for Systematic Reviews of Interventions (McKenzie 2022).
For meta‐analyses involving at least 10 studies, we intended to explore potential publication bias by generating a funnel plot and statistically testing this by conducting a linear regression test (Sterne 2019). We considered P < 0.1 as significant for this test.
Data synthesis
If the clinical and methodological characteristics of individual studies were sufficiently homogeneous, we pooled the data in meta‐analysis. We performed separate analyses for populations with ambulatory mild disease and for hospitalised participants with moderate to severe disease, according to the latest WHO clinical progression score (WHO 2020e). We performed analyses according to the recommendations of the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2022). We did not conduct meta‐analyses that included different study designs. We conducted separate meta‐analyses for each comparison.
We used RevMan Web 2022 software for analyses. One review author entered the data into the software, and a second review author checked the data for accuracy.
We used the random‐effects model for all analyses, as we anticipated that true effects in included studies would be related but would not be the same. For binary outcomes, we based the estimation of the between‐study variance using the Mantel‐Haenszel method. We used the inverse variance method for continuous outcomes, outcomes that include data from cluster‐RCTs, or outcomes where HRs are available. We planned to explore heterogeneity above 80% with subgroup analyses. If we could not find a cause for the heterogeneity or if study outcomes were too clinically heterogeneous to be combined, we did not perform a meta‐analysis, but commented on the results in a narrative analysis, with the results from all studies presented in tables.
Subgroup analysis and investigation of heterogeneity
We performed subgroup analyses of the following characteristics for our prioritised outcomes, as specified in the Summary of findings section.
-
Severity of condition for inpatients only (assessed with need for respiratory support according to WHO clinical progression scale (WHO 2020e) are divided into:
moderate, when at least 90% of participants are WHO level 4 or higher, and below WHO level 6,
severe disease, when at least 90% of participants are WHO level 6 or higher, and
moderate to severe, when 90% of participants are in both the 'moderate' and severe' categories
Length of time since symptom onset (divided into up to and including 7 days and more than 7 days)
Antibodies in recipients detected at baseline (divided into 'detected in a maximum of 20% of recipients' versus 'detected in at least 80% of recipients')
Age of participants (divided into age groups: children, 18 to 64 years, 65 years and older)
Pre‐existing conditions (diabetes, respiratory disease, hypertension, immunosuppression)
Level of antibody titre in donors (divided into high and low titres, using the US Food and Drug Administration (FDA) definition for 'low' and 'high' titre using the definition provided by the study)
Equity impact: sex (divided into female and male)
Equity impact: country income groups, according to the World Bank definitions (divided into high‐ and low‐ or middle‐income countries; The World Bank 2022)
We used the tests for interaction to test for differences between subgroup results.
We had further planned to perform additional subgroup analyses of the following characteristics, but we did not find outcome data for:
SARS‐CoV‐2 variants (e.g. B1.1.7, B.1.351, P.1, and other variants that may occur in the future)
Equity impact: ethnicity
Sensitivity analysis
We performed sensitivity analyses for the following.
Risk of bias assessment components (studies with a low risk of bias or some concerns versus studies with a high risk of bias)
Influence of completed, but not published studies (preprints)
Influence of premature termination of studies
Summary of findings and assessment of the certainty of the evidence
We used the GRADE approach to assess the certainty of the evidence for the following outcomes. We prepared three summary of findings tables for the population of individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease, and two summary of findings tables for individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease.
Individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease
All‐cause mortality at day 28 and if not reported, all‐cause mortality at day 60, time‐to‐event estimate, or during hospital stay
-
Clinical status, at day 28 and if not reported, clinical status at day 60, or up to the longest follow‐up, including the following:
worsening of clinical status: participants with clinical deterioration (new need for invasive mechanical ventilation) or death;
improvement of clinical status: participants discharged from hospital. Participants should be discharged without clinical deterioration.
Quality of life, including fatigue and neurological status, assessed with standardised scales (e.g. WHOQOL‐100, a standardised scale for assessing quality of life) at up to 7 days, up to 28 days, or longest follow‐up available
Grade 3 or 4 adverse events, defined as the number of participants with any event and including potential relationship between intervention and adverse reaction (e.g. TRALI, transfusion‐transmitted infection, TACO, TAD, acute transfusion reactions, headache, thromboembolic events)
Serious adverse events, defined as the number of participants with any serious adverse event (serious as defined according to CTCAE (Common Terminology Criteria for Adverse Events)).
Individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease
All‐cause mortality at day 28 and if not reported, all‐cause mortality at day 60, or time‐to‐event estimate
-
Symptom resolution:
all initial symptoms resolved (asymptomatic) at day 28 and if not reported, at day 14 or up to the longest follow‐up;
length of time to symptom resolution
Quality of life, including fatigue and functional independence; assessed with standardised scales (e.g. WHOQOL‐100) at longest follow‐up available
Grade 3 or 4 adverse events
Serious adverse events, defined as the number of participants with any serious adverse event (serious as defined according to CTCAE (Common Terminology Criteria for Adverse Events))
We followed the current GRADE guidance for these assessments in its entirety, as recommended in Chapter 14 of the Cochrane Handbook for Systematic Reviews of Interventions (Schünemann 2022). We used GRADEpro GDT software to create a summary of findings table (Schünemann 2022). For RCTs, we used the overall risk of bias judgement, derived from the RoB 2 Excel tool, to inform our decision on downgrading for risk of bias. For time‐to‐event outcomes, we calculated absolute effects at specific time points, as recommended in the GRADE guidance 27 (Skoetz 2020). We phrased the findings and certainty of the evidence as suggested in the informative statement guidance (Santesso 2020).
Results
Description of studies
Results of the search
As of 3 March 2022, we identified for this update 7625 new records, in addition to the 22,570 potentially relevant records from the previous versions (altogether 30,195 references). After removing duplicates, we screened 6455 new records for this update (altogether 22,267 records) based on their titles and abstracts, and we excluded 21,929 records that did not meet the prespecified inclusion criteria. We evaluated the remaining 338 records and screened the full texts, or, if these were not available, abstract publications or trial registry entries. See Figure 1 for the study flow diagram (Moher 2009).
1.
Study flow diagram
We identified 82 eligible studies within 108 citations: 33 included studies (60 records) (Agarwal 2020; Alemany 2022; AlQahtani 2021; Avendano‐Sola 2021; Bajpai 2020; Baldeon 2022; Bar 2021; Begin 2021; Beltran Gonzalez 2021; Bennett‐Guerrero 2021; CoV‐Early; De Santis 2022; Devos 2021; Estcourt 2021; Gharbharan 2021; Hamdy Salman 2020; Holm 2021; Horby 2021b; Kirenga 2021; Koerper 2021; Korley 2021; Li 2020; Libster 2020; Menichetti 2021; NCT04421404; Ortigoza 2022; O’Donnell 2021; Pouladzadeh 2021; Ray 2022; Sekine 2021; Simonovich 2020; Sullivan 2022; Van den Berg 2022) and 49 ongoing studies (see 'Ongoing studies' below).
Included studies
We included 33 studies reporting on 24,861 participants, of whom 11,432 received convalescent plasma (Agarwal 2020; Alemany 2022; AlQahtani 2021; Avendano‐Sola 2021; Bajpai 2020; Baldeon 2022; Bar 2021; Begin 2021; Beltran Gonzalez 2021; Bennett‐Guerrero 2021; CoV‐Early; De Santis 2022; Devos 2021; Estcourt 2021; Gharbharan 2021; Hamdy Salman 2020; Holm 2021; Horby 2021b; Kirenga 2021; Koerper 2021; Korley 2021; Li 2020; Libster 2020; Menichetti 2021; NCT04421404; Ortigoza 2022; O’Donnell 2021; Pouladzadeh 2021; Ray 2022; Sekine 2021; Simonovich 2020; Sullivan 2022; Van den Berg 2022).
Design and sample size
All included studies were RCTs and their sample size ranged from 29 participants in Bajpai 2020 to 11,558 participants in Horby 2021b.
Setting
The included studies differed considerably in their settings.
Six studies were conducted in the USA (Bar 2021; Bennett‐Guerrero 2021; Korley 2021; NCT04421404; Ortigoza 2022; Sullivan 2022). Three were conducted in India (Agarwal 2020; Bajpai 2020; Ray 2022). Two were done in Brazil (De Santis 2022; Sekine 2021), two in Spain (Alemany 2022; Avendano‐Sola 2021), two in Argentina (Libster 2020; Simonovich 2020), and two in the Netherlands (CoV‐Early; Gharbharan 2021). One was carried out in each of China (Li 2020), Bahrain (AlQahtani 2021), Belgium (Devos 2021), Ecuador (Baldeon 2022), Egypt (Hamdy Salman 2020), Germany (Koerper 2021), Iran (Pouladzadeh 2021), Italy (Menichetti 2021), Mexico (Beltran Gonzalez 2021), South Africa (Van den Berg 2022), Sweden (Holm 2021), Uganda (Kirenga 2021), and the UK (Horby 2021b). One study was conducted partly in Brazil, Canada, and the USA (Begin 2021), one was partly in Australia, Belgium, Canada, Croatia, Germany, Hungary, Ireland, Netherlands, New Zealand, Portugal, Romania, Spain, UK and the USA (Estcourt 2021), and one was conducted partly in the USA and partly in Brazil (O’Donnell 2021).
Nine studies are single‐centre studies (Bajpai 2020; Beltran Gonzalez 2021; Bennett‐Guerrero 2021; Hamdy Salman 2020; Kirenga 2021; Pouladzadeh 2021; Ray 2022; Sekine 2021; Sullivan 2022), and 24 are multi‐centre studies (Agarwal 2020; Alemany 2022; AlQahtani 2021; Avendano‐Sola 2021; Baldeon 2022; Bar 2021; Begin 2021; CoV‐Early; De Santis 2022; Devos 2021; Estcourt 2021; Gharbharan 2021; Holm 2021; Horby 2021b; Koerper 2021; Korley 2021; Li 2020; Libster 2020; Menichetti 2021; NCT04421404; Ortigoza 2022; O’Donnell 2021; Simonovich 2020; Van den Berg 2022), with a minimum of two centres for AlQahtani 2021, Bar 2021, Holm 2021 and Ortigoza 2022 and a maximum of 177 centres for Horby 2021b.
Among the RCTs, 29 were performed in an inpatient setting (Agarwal 2020; AlQahtani 2021; Avendano‐Sola 2021; Bajpai 2020; Baldeon 2022; Bar 2021; Begin 2021; Beltran Gonzalez 2021; Bennett‐Guerrero 2021; De Santis 2022; Devos 2021; Estcourt 2021; Gharbharan 2021; Hamdy Salman 2020; Holm 2021; Horby 2021b; Kirenga 2021; Koerper 2021; Korley 2021; Li 2020; Menichetti 2021; NCT04421404; Ortigoza 2022; O’Donnell 2021; Pouladzadeh 2021; Ray 2022; Sekine 2021; Simonovich 2020; Van den Berg 2022). Four studies were performed in an outpatient setting (Alemany 2022; CoV‐Early; Libster 2020; Sullivan 2022).
Participants
The RCTs by Agarwal 2020, AlQahtani 2021, Avendano‐Sola 2021, Baldeon 2022, Holm 2021, Kirenga 2021, Korley 2021, Menichetti 2021 and Simonovich 2020 included participants with moderate disease, and the RCTs by Bar 2021, Beltran Gonzalez 2021, De Santis 2022, Estcourt 2021 and Li 2020 included individuals with severe disease, according to the latest WHO clinical progression score (WHO 2020e). The RCTs by Bajpai 2020, Begin 2021, Bennett‐Guerrero 2021, Devos 2021, Gharbharan 2021, Hamdy Salman 2020, Horby 2021b, Koerper 2021, NCT04421404, O’Donnell 2021, Ortigoza 2022, Pouladzadeh 2021, Ray 2022, Sekine 2021 and Van den Berg 2022 included individuals with both moderate and severe disease, according to the latest WHO clinical progression score (WHO 2020e). The RCTs by Alemany 2022, CoV‐Early, Libster 2020 and Sullivan 2022 included populations with mild disease.
Interventions
Twenty‐five RCTs compared convalescent plasma with standard care, with or without placebo (Agarwal 2020; Alemany 2022; AlQahtani 2021; Avendano‐Sola 2021; Bar 2021; Begin 2021; De Santis 2022; Devos 2021; Estcourt 2021; Gharbharan 2021; Hamdy Salman 2020; Holm 2021; Horby 2021b; Kirenga 2021; Koerper 2021; Korley 2021; Li 2020; Libster 2020; Menichetti 2021; Ortigoza 2022; Pouladzadeh 2021; Ray 2022; Sekine 2021; Simonovich 2020; Van den Berg 2022), and seven RCTs compared convalescent plasma with standard plasma (Bajpai 2020; Baldeon 2022; Bennett‐Guerrero 2021; CoV‐Early; NCT04421404; O’Donnell 2021; Sullivan 2022). One RCT compared convalescent plasma with human immunoglobulin (Beltran Gonzalez 2021).
The dose and volume of plasma given in the studies that we evaluated for efficacy and safety varied. The total volume of convalescent plasma transfused varied between 200 mL and 600 mL of plasma, with participants receiving between one dose of plasma (Alemany 2022; Avendano‐Sola 2021; Baldeon 2022; Bar 2021; Bennett‐Guerrero 2021; CoV‐Early; Gharbharan 2021; Hamdy Salman 2020; Korley 2021; Li 2020; Libster 2020; NCT04421404; O’Donnell 2021; Ortigoza 2022; Simonovich 2020; Sullivan 2022; Van den Berg 2022), one to two doses of plasma (Begin 2021; Pouladzadeh 2021), and two or more doses of plasma (Agarwal 2020; AlQahtani 2021; Bajpai 2020; Beltran Gonzalez 2021; De Santis 2022; Devos 2021; Estcourt 2021; Holm 2021; Horby 2021b; Kirenga 2021; Koerper 2021; Menichetti 2021; Ray 2022; Sekine 2021).
Plasma donors
All included RCTs determined antibody titres in donors, except one (Baldeon 2022). Twelve RCTs reported antibody titres in donors' plasma (AlQahtani 2021; Bajpai 2020; Begin 2021; Beltran Gonzalez 2021; Holm 2021; Kirenga 2021; Libster 2020; Menichetti 2021; O’Donnell 2021; Simonovich 2020; Sullivan 2022; Van den Berg 2022), 12 RCTs reported neutralising antibody titres in donors' plasma (Agarwal 2020; Alemany 2022; Avendano‐Sola 2021; Bar 2021; Bennett‐Guerrero 2021; De Santis 2022; Devos 2021; Gharbharan 2021; Koerper 2021; Korley 2021; Pouladzadeh 2021; Sekine 2021), and eight did not report antibody titre in donors (CoV‐Early; Estcourt 2021; Hamdy Salman 2020; Horby 2021b; Li 2020; NCT04421404; Ortigoza 2022; Ray 2022).
Of the included studies, 30 RCTs reported the donors' eligibility criteria (Agarwal 2020; Alemany 2022; AlQahtani 2021; Avendano‐Sola 2021; Bajpai 2020; Baldeon 2022; Bar 2021; Begin 2021; Beltran Gonzalez 2021; Bennett‐Guerrero 2021; De Santis 2022; Devos 2021; Gharbharan 2021; Hamdy Salman 2020; Holm 2021; Horby 2021b; Kirenga 2021; Koerper 2021; Korley 2021; Li 2020; Libster 2020; Menichetti 2021; Ortigoza 2022; O’Donnell 2021; Pouladzadeh 2021; Ray 2022; Sekine 2021; Simonovich 2020; Sullivan 2022; Van den Berg 2022). They also reported some descriptive information about donors, such as their age, gender and disease severity, and the time from disease recovery or RT‐PCR virus detection, or both. Among those RCTs reporting the sex of donors, in Agarwal 2020, Avendano‐Sola 2021, Gharbharan 2021, Kirenga 2021, Koerper 2021 and Sekine 2021, most of the donors were male (94%, 88%, 91%, 96%, 59% and 65% respectively). In Bajpai 2020, Baldeon 2022 and Holm 2021, all donors were male.
Please refer to the Characteristics of included studies for more detailed information.
Outcomes
A list of outcomes and the applied method for which transformation of data and recalculations were made can be found in Appendix 2.
Efficacy outcomes
We prioritised different efficacy outcomes, based on the setting and the disease severity in participants of the included RCTs (see Types of outcome measures).
Among the RCTs that included individuals with moderate to severe disease, 26 studies reported 28‐day mortality (Agarwal 2020; AlQahtani 2021; Avendano‐Sola 2021; Bajpai 2020; Baldeon 2022; Bar 2021; Begin 2021; Beltran Gonzalez 2021; Bennett‐Guerrero 2021; De Santis 2022; Devos 2021; Estcourt 2021; Gharbharan 2021; Holm 2021; Horby 2021b; Kirenga 2021; Koerper 2021; Korley 2021; Li 2020; Menichetti 2021; O’Donnell 2021; Ortigoza 2022; Ray 2022; Sekine 2021; Simonovich 2020; Van den Berg 2022) and five studies also reported all‐cause mortality during hospital stay (Agarwal 2020; Beltran Gonzalez 2021; Estcourt 2021; Gharbharan 2021; O’Donnell 2021). Nine RCTs reported worsening of clinical status, assessed by the need for invasive mechanical ventilation or death (Agarwal 2020; Alemany 2022; Begin 2021; Estcourt 2021; Horby 2021b; Korley 2021; Menichetti 2021; NCT04421404; Simonovich 2020). Six RCTs reported improvement of clinical status, assessed by the number of participants discharged from hospital (Devos 2021; Gharbharan 2021; Horby 2021b; Li 2020; Sekine 2021; Simonovich 2020). One of the included RCTs reported quality of life (Devos 2021). Nine RCTs reported the safety outcome, adverse event of any grade (Holm 2021; Kirenga 2021; Koerper 2021; NCT04421404; Ortigoza 2022; O’Donnell 2021; Sekine 2021; Simonovich 2020; Van den Berg 2022); six RCTs reported the safety outcome, grade 3 or 4 adverse events (Agarwal 2020; Avendano‐Sola 2021; Begin 2021; Menichetti 2021; Sekine 2021; Simonovich 2020); and 10 RCTs reported serious adverse events (Bar 2021; Begin 2021; Bennett‐Guerrero 2021; Devos 2021; Estcourt 2021; Horby 2021b; Koerper 2021; NCT04421404; O’Donnell 2021; Simonovich 2020).
Among the RCTs that included individuals with asymptomatic or mild disease, four reported 28‐day mortality (Alemany 2022; CoV‐Early; Libster 2020; Sullivan 2022). Three RCTs reported admission to hospital or death within 28 days (Alemany 2022; Sullivan 2022; CoV‐Early). One RCT reported symptom resolution, assessed by 'all initial symptoms resolved' (CoV‐Early), and one RCT reported time to symptom resolution (Alemany 2022). No RCTs for individuals with asymptomatic or mild disease reported quality of life. One RCT reported the safety outcomes, grade 3 or 4 adverse events and serious adverse events (Alemany 2022).
Safety outcomes
Eighteen RCTs (Agarwal 2020; Alemany 2022; AlQahtani 2021; Bar 2021; Begin 2021; Bennett‐Guerrero 2021; Devos 2021; Estcourt 2021; Holm 2021; Kirenga 2021; Koerper 2021; Menichetti 2021; NCT04421404; O’Donnell 2021; Ortigoza 2022; Sekine 2021; Simonovich 2020; Van den Berg 2022), reported adverse events or serious adverse events, or both, for all the participants. From these 18 studies, we extracted safety data from 7953 participants, with safety data for 4913 participants who received convalescent plasma and 3040 participants who did not receive convalescent plasma. All the other included RCTs reported transfusion‐related adverse events for the participants receiving convalescent plasma (Avendano‐Sola 2021; Bajpai 2020; Baldeon 2022; Beltran Gonzalez 2021; CoV‐Early; De Santis 2022; Gharbharan 2021; Hamdy Salman 2020; Horby 2021b; Korley 2021; Li 2020; Libster 2020; Pouladzadeh 2021; Ray 2022; Sullivan 2022), and for the participants receiving standard plasma (Bajpai 2020; Baldeon 2022; CoV‐Early; Sullivan 2022). From these studies, we extracted safety data from 7528 participants who received convalescent plasma only and 903 participants who received the standard plasma.
Please refer to the Characteristics of included studies for more detailed information.
Ongoing studies
Of the 49 ongoing studies, all are RCTs (see Table 8). Of the 49 RCTs, 16 were scheduled to be completed in 2020 and planned to evaluate between 15 and 480 participants, but according to the trial registry, five are not yet recruiting, and nine are still recruiting. Twenty‐three RCTs were expected to be completed in 2021, and planned to evaluate between 15 and 2400 participants. They were scheduled to be completed by the time of writing, but according to the trial registry, five are not yet recruiting, three are still recruiting and 15 are active, but not recruiting. Eight further RCTs are planned to be completed in 2022: CTRI/2020/05/025346, randomising 90 participants; EUCTR2020‐001632‐10 randomising 174 participants; ISRCTN49832318 randomising 210 participants, NCT04333251, randomising 115 participants; NCT04390503, randomising 150 participants; NCT04415086, randomising 120 participants; NCT04558476, randomising 500 participants; NCT05077930, randomising 200 participants.
3. Summary of ongoing convalescent plasma studies: design and planned completion date.
| Study ID | Title | Link | Design | Planned number of participants | Planned completion date | Results available | Other study ID |
| ChiCTR2000030010 | A randomized, double‐blind, parallel‐controlled, trial to evaluate the efficacy and safety of anti‐SARS‐CoV‐2 virus inactivated plasma in the treatment of severe novel coronavirus pneumonia patients (COVID‐19) | www.chictr.org.cn/showproj.aspx?proj=49777 | RCT | 100 | 31 May 2020 | no | |
| ChiCTR2000030179 | Experimental study of novel coronavirus pneumonia rehabilitation plasma therapy severe novel coronavirus pneumonia (COVID‐19) | www.chictr.org.cn/showproj.aspx?proj=50059 | RCT | 100 | 24 April 2020 | no | |
| ChiCTR2000030627 | Study on the application of convalescent plasma therapy in severe COVID‐19 | www.chictr.org.cn/showproj.aspx?proj=50727 | RCT | 30 | 30 May 2020 | no | |
| ChiCTR2000030702 | Convalescent plasma for the treatment of common COVID‐19: a prospective randomized controlled trial | www.chictr.org.cn/showproj.aspx?proj=50537 | RCT | 30 | 15 August 2020 | no | |
| ChiCTR2000030929 | A randomized, double‐blind, parallel‐controlled trial to evaluate the efficacy and safety of anti‐SARS‐CoV‐2 virus inactivated plasma in the treatment of severe novel | www.chictr.org.cn/showproj.aspx?proj=50696 | RCT | 30 | 16 June 2020 | no | |
| CTRI/2020/04/024915 | A phase II, open label, randomized controlled trial to assess the safety and efficacy of convalescent plasma to limit COVID‐19 associated complications | www.ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=43332 | RCT | 100 | 9 May 2021 | no | |
| CTRI/2020/05/025346 | A phase II, open label, randomized controlled trial to assess the safety and efficacy of convalescent plasma in severe COVID‐19 patients | www.ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=43005 | RCT | 90 | 1 June 2022 | no | |
| CTRI/2020/06/026123 | Plasma therapy in corona patients (severe COVID‐19) | www.ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=44667 | RCT | 472 | 25 December 2020 | no | |
| EUCTR2020‐001632‐10 | A randomized open label phase‐II clinical trial with or without infusion of plasma from subjects after convalescence of SARS‐CoV‐2 infection in high‐risk patients with confirmed severe SARS‐CoV‐2 disease | www.clinicaltrialsregister.eu/ctr‐search/trial/2020‐001632‐10/DE | RCT | 174 | NR | no | NCT04433910 |
| EUCTR2020‐001936‐86 | A prospective, randomized, open label phase 2 clinical trial to evaluate superiority of anti‐SARS‐CoV‐2 convalescent plasma versus standard‐of‐care in hospitalized patients with mild COVID‐19 | www.clinicaltrialsregister.eu/ctr‐search/trial/2020‐001936‐86/DE | RCT | 340 | NR | no | |
| EUCTR2020‐002122‐82 | Prospective open‐label randomized controlled phase 2b clinical study in parallel groups for the assessment of efficacy and safety of immune therapy with COVID‐19 convalescent plasma plus standard treatment vs. standard treatment alone of subjects with severe COVID‐19 | www.clinicaltrialsregister.eu/ctr‐search/trial/2020‐002122‐82/DE | RCT | 58 | NR | no | |
| EUCTR2020‐005410‐18 | Multicentre, randomized, double‐blind, placebo‐controlled, non‐commercial clinical trial to evaluate the efficacy and safety of specific anti‐SARS‐CoV‐2 immunoglobulin in the treatment of COVID‐19 | www.clinicaltrialsregister.eu/ctr‐search/trial/2020‐005410‐18/PL | RCT | 480 | NR | no | |
| ISRCTN49832318 | SURCOVID trial: A randomized controlled trial using convalescent plasma early during moderate COVID‐19 disease course in Suriname | https://trialsearch.who.int/Trial2.aspx?TrialID=ISRCTN49832318 | RCT | 210 | 01/04/2022 | no | |
| jRCTs031200374 | An open‐label, randomized, controlled trial to evaluate the efficacy of convalescent plasma therapy for COVID‐19 | https://pesquisa.bvsalud.org/global-literature-on-novel-coronavirus-2019-ncov/resource/en/ictrp-JPRN-jRCTs031200374 | RCT | 200 | NR | ||
| NCT04333251 | Evaluating convalescent plasma to decrease coronavirus associated complications. A phase I study comparing the efficacy and safety of high‐titer anti‐SARS‐CoV‐2 plasma vs best supportive care in hospitalized patients with interstitial pneumonia due to COVID‐19. | clinicaltrials.gov/show/NCT04333251 | RCT | 115 | 31 December 2022 | no | |
| NCT04345289 | Efficacy and safety of novel treatment options for adults with COVID‐19 pneumonia (CCAP) | clinicaltrials.gov/show/NCT04345289 | RCT | 1500 | 15 June 2021 | no | EUCTR2020‐001367‐88 |
| NCT04372979 | Efficacy of convalescent plasma therapy in the early care of COVID‐19 patients | clinicaltrials.gov/show/NCT04372979 | RCT | 80 | May 2021 | no | |
| NCT04374487 | A phase II, open label, randomized controlled trial to assess the safety and efficacy of convalescent plasma to limit COVID‐19 associated complications | clinicaltrials.gov/show/NCT04374487 | RCT | 100 | 9 May 2021 | no | |
| NCT04376788 | Exchange transfusion versus plasma from convalescent patients with methylene blue in patients with COVID‐19 | clinicaltrials.gov/show/NCT04376788 | RCT | 15 | 1 June 2020 | no | |
| NCT04380935 | Effectiveness and safety of convalescent plasma therapy on COVID‐19 patients with acute respiratory distress syndrome | clinicaltrials.gov/show/NCT04380935 | RCT | 60 | 31 August 2020 | no | |
| NCT04385043 | Hyperimmune plasma in patients with COVID‐19 severe infection | clinicaltrials.gov/show/NCT04385043 | RCT | 400 | 15 May 2021 | no | |
| NCT04385186 | Inactivated convalescent plasma as a therapeutic alternative in patients CoViD‐19 | clinicaltrials.gov/show/NCT04385186 | RCT | 60 | 30 November 2020 | no | |
| NCT04388410 | Safety and efficacy of convalescent plasma transfusion for patients with SARS‐CoV‐2 infection | clinicaltrials.gov/show/NCT04388410 | RCT | 410 | 31 December 2020 | no | |
| NCT04390503 | Convalescent plasma for COVID‐19 close contacts | clinicaltrials.gov/ct2/show/NCT04390503 | RCT | 150 | 1 April 2021 | no | |
| NCT04391101 | Convalescent plasma for the treatment of severe SARS‐CoV‐2 (COVID‐19) | clinicaltrials.gov/show/NCT04391101 | RCT | 231 | 31 December 2021 | no | |
| NCT04403477 | Convalescent plasma therapy in severe COVID‐19 infection | clinicaltrials.gov/show/NCT04403477 | RCT | 20 | 30 October 2020 | no | |
| NCT04415086 | Treatment of patients with COVID‐19 with convalescent plasma | clinicaltrials.gov/show/NCT04415086 | RCT | 120 | 22 May 2022 | no | |
| NCT04418518 | A trial of convalescent plasma for hospitalized adults with acute COVID‐19 respiratory illness | clinicaltrials.gov/show/NCT04418518 | RCT | 1200 | 31 December 2021 | no | |
| NCT04425837 | Effectiveness and safety of convalescent plasma in patients with high‐risk COVID‐19 | clinicaltrials.gov/show/NCT04425837 | RCT | 236 | 28 February 2021 | no | |
| NCT04438057 | Evaluating the efficacy of convalescent plasma in symptomatic outpatients infected with COVID‐19 | clinicaltrials.gov/ct2/show/NCT04438057 | RCT | 150 | 6 July 2021 | no | |
| NCT04442191 | Convalescent plasma as a possible treatment for COVID‐19 | clinicaltrials.gov/show/NCT04442191 | RCT | 50 | 31 May 2021 | no | |
| NCT04452812 | Statistical and epidemiological study based on the use of convalescent plasma for the management of patients with COVID‐19 | clinicaltrials.gov/ct2/show/NCT04452812 | RCT | 15 | 1 April 2021 | no | |
| NCT04456413 | Convalescent plasma as treatment for subjects with early COVID‐19 infection | clinicaltrials.gov/show/NCT04456413 | RCT | 306 | 31 July 2021 | no | |
| NCT04483960 | Australasian COVID‐19 trial (ASCOT) | clinicaltrials.gov/ct2/show/NCT04483960 | RCT | 2400 | 12 July 2022 | no | ACTRN12620000445976 |
| NCT04521036 | Convalescent plasma for COVID‐19 patients (CPCP) | clinicaltrials.gov/show/NCT04521036 | RCT | 44 | 30 October 2021 | no | |
| NCT04528368 | Convalescent plasma for treating patients with COVID‐19 pneumonia without iIndication of ventilatory support | clinicaltrials.gov/ct2/show/NCT04528368 | RCT | 60 | 31 December 2020 | no | |
| NCT04558476 | Efficacy of convalescent plasma in patients with COVID‐19 treated with mechanical ventilation | clinicaltrials.gov/ct2/show/NCT04558476 | RCT | 500 | 30 September 2022 | no | |
| NCT04567173 | Convalescent plasma as adjunctive therapy for hospitalized patients with COVID‐19 | clinicaltrials.gov/show/NCT04567173 | RCT | 136 | 30 June 2021 | no | |
| NCT04634422 | Plasma exchange (PLEX) and convalescent plasma (CCP) in COVID‐19 patients with multiorgan failure (COVID‐PLEX) | clinicaltrials.gov/ct2/show/NCT04634422 | RCT | 220 | 30 June 2021 | no | |
| NCT04712344 | Assessment of efficacy and safety of therapy with COVID‐19 convalescent plasma in subjects with severe COVID‐19 (IPCO) (IPCO) | clinicaltrials.gov/ct2/show/NCT04712344 | RCT | 58 | 30 September 21 | no | |
| NCT04730401 | Convalescent plasma in the treatment of COVID‐19 (CP_COVID‐19) | clinicaltrials.gov/ct2/show/NCT04730401 | RCT | 390 | 31 December 2021 | no | |
| NCT04803370 | Efficacy of Reinforcing Standard Therapy in COVID‐19 Patients With Repeated Transfusion of Convalescent Plasma | https://clinicaltrials.gov/ct2/show/NCT04803370 | RCT | 100 | 1 September 2021 | no | |
| NCT05077930 | Convalescent Plasma Therapy for Hospitalized Patients With COVID‐19 | https://clinicaltrials.gov/ct2/show/NCT05077930 | RCT | 200 | January 2022 | no | |
| NL8633 | A randomized, double blinded clinical trial of convalescent plasma compared to standard plasma for treatment of hospitalized non‐ICU patients with COVID‐19 infections | www.trialregister.nl/trial/8633 | RCT | 430 | 1 May 2021 | no | |
| PACTR202006760881890 | Lagos COVID‐19 convalescent plasma trial (LACCPT) | pactr.samrc.ac.za/TrialDisplay.aspx?TrialID=12168 | RCT | 100 | 30 November 2020 | no | |
| PACTR202007653923168 | A clinical trial comparing use of convalescent plasma therapy plus standard treatment to standard treatment alone in patients with severe COVID‐19 infection | pactr.samrc.ac.za/TrialDisplay.aspx?TrialID=11047 | RCT | 206 | 31 December 2021 | no | |
| PER‐013‐20 | Convalescent plasma as treatment for COVID‐19 | www.ins.gob.pe/ensayosclinicos/rpec/recuperarECPBNuevoEN.asp?numec=013‐20 | RCT | 192 | 30 June 2021 | no | |
| PER‐060‐20 | Randomized phase 2 clinical trial to evaluate safety and efficacy of the use of plasma from convalescent plasma with the Coronavirus disease (COVID‐19) for the experimental treatment of patients hospitalized in the Centro Médico Naval "Cirujano Mayor Santiago Távara" | www.ins.gob.pe/ensayosclinicos/rpec/recuperarECPBNuevoEN.asp?numec=060‐20 | RCT | 100 | 7 March 2021 | no | |
| RBR‐7jqpnw | Effect of COVID‐19 convalescent plasma produced by HEMOPE: a randomized study, with a comparative group in several centers | www.ensaiosclinicos.gov.br/rg/RBR‐7jqpnw/ | RCT | 110 | 30 July 2021 | no | |
| RCT: randomised controlled trial | |||||||
Please refer to Characteristics of ongoing studies (Ongoing studies) for more detailed information and Table 8 for further details on the planned completed dates and planned number of participants per study.
Studies awaiting assessment
In the process of finalising the review, two of our tracked ongoing studies were terminated early for futility and the trials stopped recruiting participants (NCT04361253, NCT04539275).
According to the trials registries, 28 RCTs have been completed, or had their recruitment completed, but no results have been published yet (CTRI/2020/05/025299; CTRI/2020/05/025328; CTRI/2020/06/025803; EUCTR2020‐001860‐27‐GB; IRCT20120215009014N353; IRCT20150808023559N21; IRCT20200404046948N1; IRCT20200413047056N1; IRCT20200501047258N1; IRCT20200503047281N1; IRCT20201004048922N1; NCT04332835; NCT04345991; NCT04358783; NCT04362176; NCT04374526; NCT04385199; NCT04405310; NCT04425915; NCT04428021; NCT04442958; NCT04468009; NCT04497324; NCT04521309; NCT04542967; NCT04547127; NCT04649879; NCT04681430). For this reason, we have placed these studies in the category of 'Studies awaiting classification'.
Three studies are platform trials, A platform trial is an adaptive, multistage study design in which numerous interventions can be evaluated through interim analyses. These three trials do not currently include the convalescent plasma intervention, however, in a platform trial new study arms can be added within the study period to examine further interventions (Park 2020). We are tracking these studies, in case they add an arm on convalescent plasma (NCT04501978; NCT04315948; NCT04801940).
Excluded studies
We excluded in total 196 studies that did not match our inclusion criteria as follows.
162 studies are excluded from the review based on unchanged exclusion criteria:
76 studies were single‐arm studies with fewer than 500 participants receiving convalescent plasma (Abdullah 2020; Bradfute 2020; ChiCTR2000029850; ChiCTR2000030039; ChiCTR2000031501; ChiCTR2000033798; CTRI/2020/04/024804; CTRI/2020/08/027285; Donato 2020; Duan 2020; Dulipsingh 2020; Ibrahim 2020; IRCT20151228025732N53; IRCT20200406046968N2; IRCT20200414047072N1; IRCT20200416047099N1; Jin 2020; Liu 2020a; Madariaga 2020; NCT04264858; NCT04292340; NCT04321421; NCT04327349; NCT04332380; NCT04333355; NCT04345679; NCT04346589; NCT04348877; NCT04353206; NCT04354831; NCT04355897; NCT04356482; NCT04365439; NCT04374565; NCT04376034; NCT04377672; NCT04383548; NCT04384497; NCT04388527; NCT04389710; NCT04389944; NCT04390178; NCT04392232; NCT04397523; NCT04407208; NCT04408209; NCT04411602; NCT04412486; NCT04418531; NCT04432103; NCT04438694; NCT04458363; NCT04462848; NCT04471051; NCT04474340; NCT04476888; NCT04502472; NCT04513158; NCT04516954; NCT04535063; NCT04554992; NCT04565197; NCT04569188; NCT04570982; NCT04614012; NCT04616976; NCT04622826; NCT04644198; Olivares‐Gazca 2020; PER‐031‐20; Perotti 2020; RBR‐4vm3yy; RPCEC00000323; Salazar 2020a; Xia 2020; Zeng 2020);
37 studies were single‐arm studies or case series that had not been pre‐registered in a clinical trials registry (Ahn 2020; Anderson 2020; Bao 2020b; Bobek 2020; Cantore 2020; Çınar 2020; Clark 2020; Enzmann 2020; Erkurt 2020; Fan 2020; Figlerowicz 2020; Grisolia 2020; Im 2020; Jamous 2020; Jiang 2020a; Karatas 2020; Kong 2020; Liu 2020b; Martinez‐Resendez 2020; McCuddy 2020; Mira 2020; Niu 2020; Pei 2020; Peng 2020; Salazar 2020b; Shen 2020; Soleimani 2020; Taher 2020; Tan 2020; Wang 2020; Wright 2020; Xu 2020b; Yang 2020; Ye 2020; Zhang 2020a; Zhang 2020b; Zhang 2020c);
17 studies were performed with an intervention other than convalescent plasma (Cao 2020a; Chen 2020b; Chen 2020c; Díez 2020; Hu 2020; ISRCTN86534580; Jiang 2020b; Lin 2020; NCT04261426; NCT04344379; NCT04350580; NCT04368013; Robbiani 2020; Shi 2020; Xie 2020; de Assis 2020; CTRI/2020/10/028547);
five studies pertained to feasibility of collection of convalescent plasma only (Budhai 2020; Hashim 2020; NCT04344015; NCT04344977; NCT04360278);
five studies were cancelled by the investigator before recruiting participants into the study (ChiCTR2000030312; ChiCTR2000030381; ChiCTR2000030442; NCT04325672; NCT04467151);
three studies reported standard operating procedure related to plasma donation (Brasil Ministerio 2020; Franchini 2020; Ministerio de Salud 2020);
Two references were in Chinese (Qiu 2020; Tu 2020); both were translated and assessed by Rujan Shrestha and Ya‐Ying Wang via Cochrane TaskExchange. The papers reported on a generalised collection of information about the COVID‐19 infection of two participants relating to aetiology, pathology, symptoms, clinical presentation and some generalised pharmacological treatment methods;
five studies were controlled, non‐randomised studies with fewer than 500 participants receiving convalescent plasma (Abolghasemi 2020; NCT04347681; NCT04384588; Allahyari 2021; IRCT20200525047562N1);
two studies were controlled studies, but probably not truly randomised (Baklaushev 2020; Rasheed 2020);
two studies were pharmacokinetics studies (NCT04638634; NCT04661839);
two studies were withdrawn or suspended (NCT04377568; RBR‐5r8gv8p);
one study included an irrelevant participant population (participants exposed to COVID‐19; NCT04323800);
one study was a single‐arm study with fewer than 500 participants receiving hyperimmune immunoglobulin (NCT04721236);
one study compared early to deferred convalescent plasma (Balcells 2020);
one study was on plasma donors (NCT04555109);
one study was terminated early and stopped because the sponsor was changed and a new study on convalescent plasma sponsored by the Italian Medicines Agency (AIFA) was started in Italy (NCT04393727);
one study is a completed platform trial that did not include a convalescent plasma arm (NCT04593940).
We excluded 34 more studies based on the updated exclusion criteria for this review version (update 4):
14 studies included hyperimmune immunoblogulin as intervention, which is being investigated in a separate review (CTRI/2020/09/027903; jRCT2031200174; IRCT20200508047346N1; NCT04366245; NCT04395170; NCT04468958; NCT04469179; NCT04514302; NCT04546581; NCT04555148; NCT04573855; Gaborit 2021; NCT04610502; EUCTR2020‐005979‐12‐GR);
10 studies are expanded access studies from the USA (Joyner 2020; NCT04338360; NCT04360486; NCT04363034; NCT04374370; NCT04420988; NCT04445207; NCT04472572; NCT04358211; NCT04372368). These studies are non‐RCTs, and the expanded access programme is an FDA‐initiated, national, multicentre programme providing access to convalescent plasma for patients with serious or life‐threatening COVID‐19 disease, for investigations outside clinical trials when no alternative therapy options are available (US Covid Plasma 2022);
three studies were non‐randomised studies, excluded regardless of their sample size (NCT04492501; NCT04408040; NCT04432272);
two studies were single‐arm studies studies, excluded regardless of their sample size (NCT04352751; NCT04642014);
two studies were observational cohort studies (NCT04497779; NCT04545047) and two were prospective case‐only studies (NCT04463823; NCT04669990), excluded regardless of their sample size;
one study compared high‐titre convalescent plasma to low‐titre convalescent plasma (NCT04524507).
Risk of bias in included studies
We assessed methodological quality and risk of bias for all 33 included RCTs using RoB 2, recommended in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022a). The completed RoB 2 tool with responses to all assessed signalling questions is available online at zenodo.org/record/6685234#.YrMlYEZByHs.
Overall judgements for studies that included individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease
All‐cause mortality
Among those studies reporting a mortality outcome, we rated the overall risk of bias to be of some concern in Agarwal 2020, AlQahtani 2021, Baldeon 2022, Bar 2021, Beltran Gonzalez 2021, De Santis 2022, Gharbharan 2021, Holm 2021, Pouladzadeh 2021 and Ray 2022. We assessed this outcome on a study level at day 28, day 60 and time to event. For Agarwal 2020, there were some inconsistencies in the adherence to the allocated interventions, which could be due to awareness of the intervention in this open‐label trial (see Table 121; Table 124). In AlQahtani 2021, the outcome analysed was not in accordance with the prespecified analysis plan, as the time point of the mortality outcome was not specified in the study protocol (see Table 121). Baldeon 2022 provided no information on the concealment of the allocation sequence (see Table 139; Table 140). In one study (Bar 2021), information on whether the allocation sequence was concealed is missing (see Table 121; Table 123). Also, Beltran Gonzalez 2021 and De Santis 2022 provide no information on the concealment of the allocation sequence (see Table 145; Table 146; Table 147; and see Table 121; Table 122; Table 123). Gharbharan 2021 provided insufficient information on whether co‐interventions were balanced across arms (see Table 121; Table 123; Table 124). For one study (Holm 2021), the time point of measurement is not consistent with the time point indicated in the trial registry (see Table 121). Pouladzadeh 2021 did not publish a protocol and the data that produced this result were not mentioned as predefined outcomes in the trial registration (see Table 122). Ray 2022 did not provide enough information on the randomisation process and allocation concealment, and the trial registry only indicates concealment through "Case Record Numbers" (see Table 121; Table 123).
Risk of bias for analysis 1.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 1.1.1 Individuals with moderate disease | ||||||||||||
| Agarwal 2020 | Low risk of bias | The allocation sequence was random and concealed and there were no differences between intervention groups suggesting a problem with the randomization process. | Some concerns | Five participants in convalescent plasma arm and four in control arm did not receive the allocated intervention and this is probably a deviation from the intended intervention that arose because of the trial context. The analysis was appropriate. | Low risk of bias | Data for this outcome were available for 453 out of 464 participants. Data of six participants in the convalescent plasma arm and of five participants in the control arm were missing. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The data were collected in structured paper case record forms. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | For the outcome "mortality" in this study, there is a low risk of bias from the randomization process, missing outcome data, measurement of the outcome and in selection of the reported result. However, there are some concerns for bias due to deviations from intended interventions. |
| Menichetti 2021 | Low risk of bias | A stratified permuted block randomization procedure with a 1:1 ratio, with blocks of variable sizes and stratification for clinical sites, was generated by using Stata version 16.1 (StataCorp). Eligible patients underwent web‐based treatment allocation through REDcap platform to receive either administration of COVID‐19 CP in addition to standard therapy (ST) or ST alone. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized, in a modified intention‐to‐treat analysis. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | Low risk of bias | ‐ |
| Kirenga 2021 | Low risk of bias | Randomisation was performed using permuted block randomisation with varying block sizes ranging from 4 to 8. Generation of randomisation numbers was done by the study biostatistician, who did not have contact with participants. The biostatistician printed individual randomisation numbers with their corresponding treatment assignment and placed each number in an opaque envelope. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | ‐ |
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized data was available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "mortality", all the domains have a low risk of bias. |
| AlQahtani 2021 | Low risk of bias | Participants were block randomized by computer‐generated random numbering in a 1:1 ratio to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 40 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Some concerns | The data that produced this result was not analysed in accordance with the predefined outcomes stated in the protocol, as the time point of the outcome measurement was not pre‐specified in the study protocol. | Some concerns | For the outcome "mortality", there are some concerns for bias in selection of reported results. However, for all the other domains, there is a low risk of bias. |
| Holm 2021 | Low risk of bias | Patients eligible for inclusion were enrolled and randomized by a study physician (KH, OL, CW, JO, MR), 1:1 using the electronic software REDCap to either receive SOC or SOC with 200–250 mL of CCP administered intravenously during 30 min on three consecutive days. Blocks of ten patients were used for randomization. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Some concerns | No protocol is available online and the time point of all‐cause mortality reported in the article is not consistent with the planed time point indicated in the trial registry (three month in the ClinicalTrial registration). | Some concerns | ‐ |
| Avendano‐Sola 2021 | Low risk of bias | Participants were randomized through a web‐based eCRF system (ORACLE clinical) in a 1:1 ratio to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 359 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | For the outcome "mortality at up to day 28", there is a low risk of bias for all the domains. |
| Korley 2021 | High risk of bias | Baseline differences (characteristics are not balanced) in hospitalisation between groups suggesting a problem with randomization. Participants in the control group were included when they were still hospitalised and not dischared yet, in contrary to the CP group. | Low risk of bias | The participants were blinded to the treatment, but those delivering the intervention were aware of intervention received. There were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received, so it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | High risk of bias | High risk of bias arising from the randomization process. |
| Subgroup 1.1.2 Individuals with severe disease | ||||||||||||
| Li 2020 | Low risk of bias | Participants were block randomized via computer‐generated random numbering in a 1:1 ratio to receive standard treatment coupled with convalescent plasma transfusion or standard treatment alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized, a total of 101 out of 103 participants were included in the analysis. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "mortality at up to day 28", all the domains have a low risk of bias. |
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| De Santis 2022 | Some concerns | The study methods it is stated that this is a "investigator‐initiated multicenter open‐label randomized controlled trial", but no further information on the allocation sequence concealment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | ‐ |
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Subgroup 1.1.3 Individuals with moderate to severe disease | ||||||||||||
| Devos 2021 | Low risk of bias | Patients were randomised through a computerised system (RedCap®, Vanderbilt University, USA, Version 10.6.13) according to a 2:1 allocation scheme stratified by study site using randomly selected block sizes of 6 or 9, to open‐label convalescent plasma combined with standard of care (intervention group) or standard of care alone (control group). | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Ortigoza 2022 | Low risk of bias | A centralized electronic system was used to randomly assign enrolled patients to receive CCP or placebo in a 1:1 ratio stratified by enrollment site and risk status using randomization block sizes of 4 and 6 to maintain balanced group sizes. Allocation was concealed. | Low risk of bias | Patients, treating clinicians, trial personnel, and outcome assessors were blinded to group assignment and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so the interim analysis was no deviation from the protocol and the analysis was appropriate. | Low risk of bias | Data were available for nearly all (921/938) participants. 17 Participants were lost to follow‐up at day 30 (after discharge). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "28‐day mortality", there was a low risk of bias for all the domains. |
| Koerper 2021 | Low risk of bias | Patients were randomized using a Web‐based system with a stratified 1:1 allocation ratio between each stratum (Figure 1). Patients were stratified prior to permutated block randomization by presence or absence of ventilation support, ECMO, or ICU treatment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Sekine 2021 | Low risk of bias | Patients were randomly assigned in a 1:1 ratio to receive two infusions 48 h apart of 300mL of CP plus Standard of Care (SOC) or SOC alone. Randomisation was performed using computer‐generated randomisation with random block sizes of 2 or 4 and stratified according to the unit of hospitalisation on enrolment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Van den Berg 2022 | Low risk of bias | Eligible and consenting participants were randomized (1:1) to receive either a single infusion of 200–250 mL CCP or 200 mL of placebo, together with local standard of care. Random assignment was stratified by study site, age (≥ or < 65 years), and body mass index (BMI) (≥ or < 30 kg/m2). An electronic randomisation application (REDCap) hosted by SANB, was used to generate the treatment allocation. To mask treatment allocation, investigational product (IP) was covered in opaque paper wrapping prior to dispatch from the blood bank. | Low risk of bias | Both participants and those delivering the intervention were probably not aware of intervention received, as treatment allocation was masked (plasma was covered in opaque paper wrapping) and there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized (two participants withdrew consent and two were transferred to alternative facilities prior to transfusion) . | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registry, the protocol was not available. | Low risk of bias | ‐ |
| Ray 2022 | Some concerns | Participants were probably allocated randomly to either the standard of care group alone or standard of care with convalescent plasma group and there were no further information given on the randomization process. There is no information on the allocation concealment, the trial registry only indicates concealment through "Case Record Numbers". There were no baseline imbalances that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 80 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | For the outcome "mortality at up to day 28" in this study, there is a low risk of bias due to deviations from intended interventions, due to missing outcome data, in measurement of the outcome and in selection of the reported result. There are some concerns for bias from the randomization process. |
Risk of bias for analysis 1.4 All‐cause mortality during hospital stay.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 1.4.1 Individuals with moderate disease | ||||||||||||
| Agarwal 2020 | Low risk of bias | The allocation sequence was random and concealed and there were no differences between intervention groups suggesting a problem with the randomization process. | Some concerns | Five participants in convalescent plasma arm and four in control arm did not receive the allocated intervention. This could be due to awareness of the intervention, because this trial was open‐label. | Low risk of bias | Data for this outcome was available for 451 out of 464 participants. Data of eight participants in the convalescent plasma arm and of five participants in the control arm were missing. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The data were collected in structured paper case record forms. | Low risk of bias | There was no information on whether the data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration, as the data was provided on request by the study investigators. | Some concerns | For this outcome "mortality", there is a low risk of bias from the randomization process, missing outcome data, measurement of the outcome and in selection of the reported result. However, there are some concerns for bias due to deviations from intended interventions. |
| AlQahtani 2021 | Low risk of bias | Participants were block randomized by computer‐generated random numbering in a 1:1 ratio to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 40 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | For this outcome, there was no information on whether the data that produced this result was analysed in accordance with the predefined protocol, as the data was provided on request by the study investigators. | Low risk of bias | For the outcome "All‐cause mortality at hospital discharge" all the other domains were at low risk of bias. |
| Subgroup 1.4.2 Individuals with severe disease | ||||||||||||
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | For the outcome "all‐cause mortality at hospital discharge" there is a low risk of bias from the randomization process, due to missing outcome data, in measurement of the outcome and in selection of the reported result. However, there are some concerns for bias due to deviations from intended interventions. |
Risk of bias for analysis 2.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 2.1.1 Individuals with moderate disease | ||||||||||||
| Baldeon 2022 | Some concerns | This was a random double blinded, non‐convalescent plasma (NCP)‐placebo controlled clinical trial. However, there are no information provided on the concealment of the allocation sequence:"A simple randomization scheme was used to allocate participants in the two treatment groups." | Low risk of bias | This study was triple blinded. Hospital personnel at the transfusion medicine services, physicians, and the participants were unaware which plasma was convalescent and which plasma was control. Intention to treat analysis was used. | Low risk of bias | Data was available for all the 158 participants randomised. | Low risk of bias | Clinical outcomes were assessed by investigators who were blinded to the treatment groups. | Low risk of bias | No protocol was available, but the outcome was pre‐specified in the trial registry. | Some concerns | Some concerns due to bias arising from the randomization process. |
| Subgroup 2.1.2 Individuals with moderate to severe disease | ||||||||||||
| Bajpai 2020 | Low risk of bias | Participants were block randomized with concealed allocation using the Sequentially Numbered Opaque Sealed Envelopes (SNOSE) method, to receive either convalescent plasma with standard of care or standard plasma (fresh frozen plasma) with standard of care. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized, in total 29 out of 31 participants (2 participants became PCR negative on the day of plasma transfusion). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | For this outcome "mortality at day 28", there is a low risk of bias for all the domains. |
| Bennett‐Guerrero 2021 | Low risk of bias | Patients were randomized 4:1 to CP or SP using permuted block randomization lists generated using SAS software (version 9.4; SAS Institute, Cary NC) and implemented using an interactive web response randomization tool in Research Electronic Data Capture (REDCap). | Low risk of bias | The individuals who were unblinded were Blood Bank personnel since they needed to label and dispense the masked CP or SP. The bags of CP or SP had an identical label, stating “CP or SP” to preserve blinding. All study personnel who collected data were blinded to study assignment at all times. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Low risk of bias | No protocol available, but the outcome is pre‐defined in Supplemental Detailed Methods. | Low risk of bias | ‐ |
| O’Donnell 2021 | Low risk of bias | Participants were block randomized using a web‐based randomization platform in a 2:1 ratio to receive either convalescent plasma or standard plasma and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 223 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | For the outcome "28‐day mortality", there is a low risk of bias for all the domains. |
Risk of bias for analysis 2.2 All‐cause mortality (time to event).
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Bajpai 2020 | Low risk of bias | Participants were block randomized with concealed allocation using the Sequentially Numbered Opaque Sealed Envelopes (SNOSE) method, to receive either convalescent plasma with standard of care or standard plasma (fresh frozen plasma) with standard of care. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized, in total 29 out of 31 participants (2 participants became PCR negative on the day of plasma transfusion). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | ‐ |
| Baldeon 2022 | Some concerns | This was a random double blinded, non‐convalescent plasma (NCP)‐placebo controlled clinical trial. However, there are no information provided on the concealment of the allocation sequence:"A simple randomization scheme was used to allocate participants in the two treatment groups." | Low risk of bias | This study was triple blinded. Hospital personnel at the transfusion medicine services, physicians, and the participants were unaware which plasma was convalescent and which plasma was control. Intention to treat analysis was used. | Low risk of bias | Data was available for all the 158 participants randomised. | Low risk of bias | Clinical outcomes were assessed by investigators who were blinded to the treatment groups. | Low risk of bias | No protocol was available, but the outcome was pre‐specified in the trial registry. | Some concerns | ‐ |
| Bennett‐Guerrero 2021 | Low risk of bias | Patients were randomized 4:1 to CP or SP using permuted block randomization lists generated using SAS software (version 9.4; SAS Institute, Cary NC) and implemented using an interactive web response randomization tool in Research Electronic Data Capture (REDCap). | Low risk of bias | The individuals who were unblinded were Blood Bank personnel since they needed to label and dispense the masked CP or SP. The bags of CP or SP had an identical label, stating “CP or SP” to preserve blinding. All study personnel who collected data were blinded to study assignment at all times. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Low risk of bias | No protocol available, but in Supplemental Detailed Methods mentioned | Low risk of bias | ‐ |
| O’Donnell 2021 | Low risk of bias | Participants were block randomized using a web‐based randomization platform in a 2:1 ratio to receive either convalescent plasma or standard plasma and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 223 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | ‐ |
Risk of bias for analysis 1.3 All‐cause mortality (time to event).
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 1.3.1 Individuals with moderate disease | ||||||||||||
| Avendano‐Sola 2021 | Low risk of bias | Participants were randomized through a web‐based eCRF system (ORACLE clinical) in a 1:1 ratio to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 359 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | ‐ |
| Kirenga 2021 | Low risk of bias | Randomisation was performed using permuted block randomisation with varying block sizes ranging from 4 to 8. Generation of randomisation numbers was done by the study biostatistician, who did not have contact with participants. The biostatistician printed individual randomisation numbers with their corresponding treatment assignment and placed each number in an opaque envelope. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | ‐ |
| Menichetti 2021 | Low risk of bias | A stratified permuted block randomization procedure with a 1:1 ratio, with blocks of variable sizes and stratification for clinical sites, was generated by using Stata version 16.1 (StataCorp). Eligible patients underwent web‐based treatment allocation through REDcap platform to receive either administration of COVID‐19 CP in addition to standard therapy (ST) or ST alone. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized, in a modified intention‐to‐treat analysis. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | Low risk of bias | ‐ |
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized data was available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "mortality", all the domains have a low risk of bias. |
| Subgroup 1.3.2 Individuals with severe disease | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| De Santis 2022 | Some concerns | The study methods it is stated that this is a "investigator‐initiated multicenter open‐label randomized controlled trial", but no further information on the allocation sequence concealment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | ‐ |
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | Low risk of bias | ‐ |
| Li 2020 | Low risk of bias | Participants were block randomized via computer‐generated random numbering in a 1:1 ratio to receive standard treatment coupled with convalescent plasma transfusion or standard treatment alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 103 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "mortality (time to event)", all the domains have a low risk of bias. |
| Subgroup 1.3.3 Individuals with moderate to severe disease | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for all available | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Devos 2021 | Low risk of bias | Patients were randomised through a computerised system (RedCap®, Vanderbilt University, USA, Version 10.6.13) according to a 2:1 allocation scheme stratified by study site using randomly selected block sizes of 6 or 9, to open‐label convalescent plasma combined with standard of care (intervention group) or standard of care alone (control group). | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | For the outcome "Mortality (time to event)", there is a low risk of bias from the randomization process, due to missing outcome data, in measurement of the outcome and in selection of the reported result. However, there are some concerns for bias due to deviations from intended interventions. |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "mortality (time to event)", there was a low risk of bias for all the domains. |
| Koerper 2021 | Low risk of bias | Patients were randomized using a Web‐based system with a stratified 1:1 allocation ratio between each stratum (Figure 1). Patients were stratified prior to permutated block randomization by presence or absence of ventilation support, ECMO, or ICU treatment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Ray 2022 | Some concerns | Participants were probably allocated randomly to either the standard of care group alone or standard of care with convalescent plasma group and there were no further information given on the randomization process. There is no information on the allocation concealment, the trial registry only indicates concealment through "Case Record Numbers". There were no baseline imbalances that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 80 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Some concerns | The data that produced this result was not analysed in accordance with the predefined outcomes stated in the trial registration. The "time‐to‐event" measurement for mortality was not mentioned in the trial registry. | Some concerns | For the outcome "mortality (time to event)", there is a low risk of bias due to deviations from intended interventions, due to missing outcome data and in measurement of the outcome. There are some concerns for bias in selection of the reported result and from the randomization process. |
| Sekine 2021 | Low risk of bias | Patients were randomly assigned in a 1:1 ratio to receive two infusions 48 h apart of 300mL of CP plus Standard of Care (SOC) or SOC alone. Randomisation was performed using computer‐generated randomisation with random block sizes of 2 or 4 and stratified according to the unit of hospitalisation on enrolment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Van den Berg 2022 | Low risk of bias | Eligible and consenting participants were randomized (1:1) to receive either a single infusion of 200–250 mL CCP or 200 mL of placebo, together with local standard of care. Random assignment was stratified by study site, age (≥ or < 65 years), and body mass index (BMI) (≥ or < 30 kg/m2). An electronic randomisation application (REDCap) hosted by SANB, was used to generate the treatment allocation. To mask treatment allocation, investigational product (IP) was covered in opaque paper wrapping prior to dispatch from the blood bank. | Low risk of bias | Both participants and those delivering the intervention were probably not aware of intervention received, as treatment allocation was masked (plasma was covered in opaque paper wrapping) and there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized (two participants withdrew consent and two were transferred to alternative facilities prior to transfusion) . | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registry, the protocol was not available. | Low risk of bias | ‐ |
Risk of bias for analysis 3.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Beltran Gonzalez 2021 | Some concerns | This is a controlled, randomized, open clinical trial including patients with secondary pneumonia to SARS‐CoV‐2 infection, and fulfilling severe or critical disease criteria, to compare the efficacy and safety of CP administration in comparison with the use of IVIg. No information on the concealment of the allication sequence. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | ‐ |
Risk of bias for analysis 3.2 All‐cause mortality (time to event).
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Beltran Gonzalez 2021 | Some concerns | This is a controlled, randomized, open clinical trial including patients with secondary pneumonia to SARS‐CoV‐2 infection, and fulfilling severe or critical disease criteria, to compare the efficacy and safety of CP administration in comparison with the use of IVIg. No information on the concealment of the allication sequence. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | ‐ |
Risk of bias for analysis 3.3 All‐cause mortality during hospital stay.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Beltran Gonzalez 2021 | Some concerns | This is a controlled, randomized, open clinical trial including patients with secondary pneumonia to SARS‐CoV‐2 infection, and fulfilling severe or critical disease criteria, to compare the efficacy and safety of CP administration in comparison with the use of IVIg. No information on the concealment of the allication sequence. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | ‐ |
Risk of bias for analysis 1.2 All‐cause mortality at up to day 60.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 1.2.1 Individuals with moderate to severe disease | ||||||||||||
| Koerper 2021 | Low risk of bias | Patients were randomized using a Web‐based system with a stratified 1:1 allocation ratio between each stratum (Figure 1). Patients were stratified prior to permutated block randomization by presence or absence of ventilation support, ECMO, or ICU treatment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate, and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Pouladzadeh 2021 | Low risk of bias | This parallel‐group, single‐blind, and randomized controlled trial was designed by randomly assigning patients to the intervention and control groups using a 6‐item randomized block method by computer and an equal allocation ratio (1:1). | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate, and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Some concerns | The data that produced this result was not mentioned as predefined outcomes in the trial registration and there was no protocol available. | Some concerns | ‐ |
| Subgroup 1.2.2 Individuals with severe disease | ||||||||||||
| De Santis 2022 | Some concerns | The study methods it is stated that this is a "investigator‐initiated multicenter open‐label randomized controlled trial", but no further information on the allocation sequence concealment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | ‐ |
We rated the overall risk of bias to be high in Korley 2021, because of baseline differences in hospitalisation between the groups. The participants in the control group were included, while they were still hospitalised, contrary to the intervention group (see Table 121).
We also assessed with a funnel plot the potential risk of publication bias for the outcome all‐cause mortality up to day 28, and there is no indication for publication bias, see Figure 2.
2.
Clinical status
Among those studies reporting at least one of the two outcomes addressing clinical status, we rated the overall risk of bias to be of some concern for Agarwal 2020 and Gharbharan 2021. We assessed clinical status on a study level, including both improvement of clinical status by the number of participants discharged from hospital and worsening of clinical status by need for invasive mechanical ventilation or death. For Gharbharan 2021, we judged the risk of bias for need for invasive mechanical ventilation or death to be of some concern, as the study provided insufficient information on whether co‐interventions were balanced across arms (see Table 126). For Agarwal 2020, we judged the risk of bias for participants discharged from hospital to be of some concern, because of some inconsistencies in adherence to the allocated interventions, which could be due to awareness of the intervention in this open‐label trial, as well as because the outcome analysed was not prespecified in the trials registry and a study protocol was not available (see Table 125).
Risk of bias for analysis 1.6 Clinical improvement: participants discharged from hospital.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 1.6.1 Individuals with moderate disease | ||||||||||||
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized data was available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | The outcome was appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 1.6.2 Individuals with severe disease | ||||||||||||
| Li 2020 | Low risk of bias | Participants were block randomized via computer‐generated random numbering in a 1:1 ratio to receive standard treatment coupled with convalescent plasma transfusion or standard treatment alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized, a total of 101 out of 103 participants were included in the analysis. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 1.6.3 Individuals with moderate to severe disease | ||||||||||||
| Devos 2021 | Low risk of bias | Patients were randomised through a computerised system (RedCap®, Vanderbilt University, USA, Version 10.6.13) according to a 2:1 allocation scheme stratified by study site using randomly selected block sizes of 6 or 9, to open‐label convalescent plasma combined with standard of care (intervention group) or standard of care alone (control group). | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was in accordance with the predefined outcomes stated in the trial protocol: "Duration of hospital and ICU stay: both parameters will be analysed as time‐to‐event parameters with competing risk, whereby the event of interest is discharge from hospital/ICU and the competing risk is hospital/ICU death". | Low risk of bias | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Sekine 2021 | Low risk of bias | Patients were randomly assigned in a 1:1 ratio to receive two infusions 48 h apart of 300mL of CP plus Standard of Care (SOC) or SOC alone. Randomisation was performed using computer‐generated randomisation with random block sizes of 2 or 4 and stratified according to the unit of hospitalisation on enrolment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 1.5 Clinical worsening: need for invasive mechanical ventilation or death at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 1.5.1 Individuals with moderate disease | ||||||||||||
| Agarwal 2020 | Low risk of bias | The allocation sequence was random and concealed and there were no differences between intervention groups suggesting a problem with the randomization process. | Some concerns | Five participants in convalescent plasma arm and four in control arm did not receive the allocated intervention and this is probably a deviation from the intended intervention that arose because of the trial context. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for 453 out of 464 participants. Data of six participants in the convalescent plasma arm and of five participants in the control arm were missing. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The data were collected in structured paper case record forms. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | ‐ |
| Korley 2021 | High risk of bias | Baseline differences (characteristics are not balanced) in hospitalisation between groups suggesting a problem with randomization. Participants in the control group were included when they were still hospitalised and not dischared yet, in contrary to the CP group. | Low risk of bias | The participants were blinded to the treatment, but those delivering the intervention were aware of intervention received. There were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received, so it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | High risk of bias | ‐ |
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized data was available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | The outcome measure was appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 1.5.2 Individuals with severe disease | ||||||||||||
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Subgroup 1.5.3 Individuals with moderate to severe disease | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so the interim analysis was no deviation from the protocol and the analysis was appropriate. | Low risk of bias | Data were available for nearly all (921/938) participants. 17 Participants were lost to follow‐up at day 30 (after discharge). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
In Korley 2021, we rated the overall risk of bias to be high. There are some baseline differences in hospitalisation between the two groups (see Table 125).
Quality of life
Only one study reported a quality‐of‐life outcome (Devos 2021). In this study, we rated the overall risk of bias to be of some concern, because the assessors were aware of the intervention received and it is possible that the assessment could have been influenced by knowledge of intervention received (see Table 127).
Risk of bias for analysis 1.7 Quality of life, assessed with standardised scales at day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Devos 2021 | Low risk of bias | Patients were randomised through a computerised system (RedCap®, Vanderbilt University, USA, Version 10.6.13) according to a 2:1 allocation scheme stratified by study site using randomly selected block sizes of 6 or 9, to open‐label convalescent plasma combined with standard of care (intervention group) or standard of care alone (control group). | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Some concerns | The measurement of the outcome was appropriate, assessed with the EX‐5D‐5L‐questionnaire and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | Due to bias of measurement outcome |
Safety
Among those studies that reported at least one of the safety outcomes, we rated the overall risk of bias to be of some concern for Agarwal 2020, AlQahtani 2021, Bar 2021, Begin 2021, Bennett‐Guerrero 2021, Devos 2021, Estcourt 2021, Holm 2021, Kirenga 2021, Koerper 2021, Menichetti 2021, Sekine 2021 and Van den Berg 2022. We assessed safety outcomes on a study level and included any adverse events, grade 1 or 2 adverse events, grade 3 to 4 adverse events and serious adverse events.
For Agarwal 2020, we judged the risk of bias for grade 3 to 4 adverse events to be of some concern, because of some inconsistencies in the adherence to the allocated interventions, which could be due to awareness of the intervention in this open‐label trial. For Agarwal 2020 and AlQahtani 2021, the safety data were provided at our request by the study authors, and it is not clear whether data for this outcome were collected from all, or nearly all the participants randomised. There is also no information available on how safety was measured (see Table 130). Bar 2021 gave no information on whether the allocation sequence was concealed and the outcome assessors were aware of the intervention received (see Table 131). In Begin 2021, too, the outcome assessors were aware of the interventions received, but external monitoring was performed at all sites to assess protocol adherence, reporting of adverse events and accuracy of data entry (see Table 130; Table 131). There is not enough information available in Bennett‐Guerrero 2021 to evaluate the measurement of the outcome and the outcome is not predefined in the registry of supplemental material (see Table 144). Devos 2021 and Estcourt 2021did not blind the assessor, so the assessment could have been influenced by knowledge of the intervention received (see Table 131). Holm 2021 did not predefine the outcome in the clinical trials registry and there is no protocol available (see Table 128). Kirenga 2021 did not clearly indicate whether all the participants were included in the analysis and the assessors were aware of the intervention allocation (see Table 128). In Koerper 2021 and in Menichetti 2021, the assessment of the outcome could have been influenced by knowledge of intervention received (see Table 128; Table 130; Table 131). Sekine 2021 did not provide information on whether the outcome measure was appropriate and the outcome assessors were not blinded (see Table 128; Table 129; Table 130). Van den Berg 2022 did not blind the assessors either, the data that produced this result were not analysed in accordance with the predefined outcomes stated in the trials registry and the protocol was not available (see Table 128).
Risk of bias for analysis 1.10 Grades 3 and 4 adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 1.10.1 Individuals with moderate disease | ||||||||||||
| Agarwal 2020 | Low risk of bias | The allocation sequence was random and concealed and there were no differences between intervention groups suggesting a problem with the randomization process. | Some concerns | Five participants in convalescent plasma arm and four in control arm did not receive the allocated intervention and this is probably a deviation from the intended intervention that arose because of the trial context. There is no information on whether the analysis was appropriate. | Low risk of bias | Data for this outcome was available for 451 participants out of 464 participants randomized and the data of eight participants in the convalescent plasma arm and of five participants in the control arm were missing. | Some concerns | There was no information on the method of measuring for this outcome, as the data was provided on request by the study investigator. The outcome assessors were aware of the intervention received and the knowledge of intervention received could have affected outcome measurement, but it is probably not likely. | Low risk of bias | There was no information on whether the data that produced this result was analysed in accordance with a pre‐specified trial registration, as the data was provided on request by the study investigators. | Some concerns | For the outcome "Grade 3 and 4 adverse events", there is a low risk of bias arising from the randomization process and due to missing outcome data and in the in selection of the reported result. But, there are some concerns for bias due to deviations from intended interventions and in measurement of the outcome. |
| AlQahtani 2021 | Low risk of bias | Participants were block randomized by computer‐generated random numbering in a 1:1 ratio to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate.. | Low risk of bias | Data for this outcome was available for all 40 participants randomized. | Some concerns | There is no information available on whether the method of measuring the outcome was appropriate or whether the measurement could have differed between groups, as the data was provided on request by the study investigator and for safety different measurements are possible. The outcome assessors were aware of the intervention received, which is why the knowledge of intervention received could have affected safety outcome assessments, but this is unlikely. | Low risk of bias | For this outcome, there was no information on whether the data that produced this result was analysed in accordance with the predefined protocol, as the data was provided on request by the study investigators. | Some concerns | For the outcome "Grade 3 and 4 adverse events", there is a low risk of bias arising from the randomization process, due to deviations from intended interventions, due to missing outcome data and in the in selection of the reported result. But, there are some concerns for bias in measurement of the outcome. |
| Menichetti 2021 | Low risk of bias | A stratified permuted block randomization procedure with a 1:1 ratio, with blocks of variable sizes and stratification for clinical sites, was generated by using Stata version 16.1 (StataCorp). Eligible patients underwent web‐based treatment allocation through REDcap platform to receive either administration of COVID‐19 CP in addition to standard therapy (ST) or ST alone. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | The data for this outcome was available for all participants randomized. | Some concerns | There is not enough information whether the measurement of the outcome was appropriate. This is an open label trial and the outcome assessors were aware of the intervention allocation, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | ‐ |
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized, data was available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | The outcome measurement is an appropriate measure and it is unlikely that the measurement differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For this outcome "Grade 3 and 4 adverse events", there is a low risk of bias for all the domains. |
| Subgroup 1.10.2 Individuals with moderate to severe disease | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so the interim analysis was no deviation from the protocol and the analysis was appropriate. | Low risk of bias | Data were available for nearly all (921/938) participants. 17 Participants were lost to follow‐up at day 30 (after discharge). | Some concerns | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the interventions received, but external monitoring was performed at all sites to assess protocol adherence, reporting of adverse events and accuracy of data entry, so it is unlikely that the outcome assessment was influenced by the intervention knowledge. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Sekine 2021 | Low risk of bias | Patients were randomly assigned in a 1:1 ratio to receive two infusions 48 h apart of 300mL of CP plus Standard of Care (SOC) or SOC alone. Randomisation was performed using computer‐generated randomisation with random block sizes of 2 or 4 and stratified according to the unit of hospitalisation on enrolment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized | Some concerns | There is no information on whether the outcome measure was appropriate, but it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | ‐ |
Risk of bias for analysis 1.11 Serious adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 1.11.1 Individuals with moderate disease | ||||||||||||
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomised, data were available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | The outcome measurement is an appropriate measure and it is unlikely that the measurement differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For this outcome "Serious adverse events", there is a low risk of bias for all the domains. |
| Subgroup 1.11.2 Individuals with moderate to severe disease | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome were available for (nearly) all participants randomised | Some concerns | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so the interim analysis was no deviation from the protocol and the analysis was appropriate. | Low risk of bias | Data were available for nearly all (921/938) participants. 17 Participants were lost to follow‐up at day 30 (after discharge). | Some concerns | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the interventions received, but external monitoring was performed at all sites to assess protocol adherence, reporting of adverse events and accuracy of data entry, so it is unlikely that the outcome assessment was influenced by the intervention knowledge. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Devos 2021 | Low risk of bias | Patients were randomised through a computerised system (RedCap®, Vanderbilt University, USA, Version 10.6.13) according to a 2:1 allocation scheme stratified by study site using randomly selected block sizes of 6 or 9, to open‐label convalescent plasma combined with standard of care (intervention group) or standard of care alone (control group). | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Some concerns | The measurement of the outcome was appropriate, the serious adverse events are listed in the supplement and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | Due to bias of measurement outcome |
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Some concerns | The measurement of the outcome was appropriate, the serious adverse events are listed in the supplement and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | Due to bias of measurement outcome |
| Koerper 2021 | Low risk of bias | Patients were randomized using a Web‐based system with a stratified 1:1 allocation ratio between each stratum (Figure 1). Patients were stratified prior to permutated block randomization by presence or absence of ventilation support, ECMO, or ICU treatment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Some concerns | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The outcome is predefined in the trial registry. | Some concerns | ‐ |
Risk of bias for analysis 2.6 Serious adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 2.6.1 Individuals with moderate to severe disease | ||||||||||||
| Bennett‐Guerrero 2021 | Low risk of bias | Patients were randomized 4:1 to CP or SP using permuted block randomization lists generated using SAS software (version 9.4; SAS Institute, Cary NC) and implemented using an interactive web response randomization tool in Research Electronic Data Capture (REDCap). | Low risk of bias | The individuals who were unblinded were Blood Bank personnel since they needed to label and dispense the masked CP or SP. The bags of CP or SP had an identical label, stating “CP or SP” to preserve blinding. All study personnel who collected data were blinded to study assignment at all times. | Low risk of bias | Data for this outcome was available for all participants randomized. | Some concerns | There is not enough information available and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Some concerns | The outcome is not pre‐defined in the registry of supplemental material . | Some concerns | ‐ |
| NCT04421404 | Low risk of bias | Subjects enrolled in the study will be randomized using a web based randomization procedure to receive convalescent plasma versus non‐immune plasma at a 1:1 ratio. | Low risk of bias | Both participants and those delivering the intervention were not aware of intervention received, there were no deviations from intended interventions and the analysis was appropriate. The nursing staff may be unblinded to the treatment assignment to ensure proper ABO checking of the plasma unit at bedside per standard transfusion procedures. | Low risk of bias | Data for this outcome was available for all 34 participants randomized. | Low risk of bias | The are no information available (as the study still needs to be published in a journal) on whether the measurement of the outcome was appropriate, but it is unlikely that it differed between intervention groups. The outcome assessors were blinded to the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | Low risk of bias | ‐ |
| O’Donnell 2021 | Low risk of bias | Participants were block randomized using a web‐based randomization platform in a 2:1 ratio to receive either convalescent plasma or standard plasma and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all (219 out of 223) participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | For the outcome "serious adverse events", there is a low risk of bias for all the domains. |
Risk of bias for analysis 1.8 Any grade adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Holm 2021 | Low risk of bias | Patients eligible for inclusion were enrolled and randomized by a study physician (KH, OL, CW, JO, MR), 1:1 using the electronic software REDCap to either receive SOC or SOC with 200–250 mL of CCP administered intravenously during 30 min on three consecutive days. Blocks of ten patients were used for randomization. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized | Some concerns | The safety outcome was not predefined in the trial registry, but the text says: "Patients were monitored once daily for adverse events, blood pressure, oxygen saturation, heart and respiratory rates and temperature. An adverse event was any adverse symptom or clinical sign that was not considered as an obvious symptom of COVID‐19" | Some concerns | The outcome was not pre‐defined in the clinical trial registry and there is no protocol available. | Some concerns | ‐ |
| Kirenga 2021 | Low risk of bias | Randomisation was performed using permuted block randomisation with varying block sizes ranging from 4 to 8. Generation of randomisation numbers was done by the study biostatistician, who did not have contact with participants. The biostatistician printed individual randomisation numbers with their corresponding treatment assignment and placed each number in an opaque envelope. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Some concerns | It is not clearly indicated whethere all the participants were included in the analysis. | Some concerns | There is not enough information whether the measurement of the outcome was appropriate. This is an open label trial and the outcome assessors were aware of the intervention allocation, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registry. | Some concerns | ‐ |
| Koerper 2021 | Low risk of bias | Patients were randomized using a Web‐based system with a stratified 1:1 allocation ratio between each stratum (Figure 1). Patients were stratified prior to permutated block randomization by presence or absence of ventilation support, ECMO, or ICU treatment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Some concerns | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The outcome is predefined in the trial registry. | Some concerns | ‐ |
| Ortigoza 2022 | Low risk of bias | A centralized electronic system was used to randomly assign enrolled patients to receive CCP or placebo in a 1:1 ratio stratified by enrollment site and risk status using randomization block sizes of 4 and 6 to maintain balanced group sizes. Allocation was concealed. | Low risk of bias | Patients, treating clinicians, trial personnel, and outcome assessors were blinded to group assignment and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Sekine 2021 | Low risk of bias | Patients were randomly assigned in a 1:1 ratio to receive two infusions 48 h apart of 300mL of CP plus Standard of Care (SOC) or SOC alone. Randomisation was performed using computer‐generated randomisation with random block sizes of 2 or 4 and stratified according to the unit of hospitalisation on enrolment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized | Some concerns | There is no information on whether the outcome measure was appropriate, but it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | ‐ |
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized, data was available for 228 participants in the convalescent plasma group and 104 participants in the placebo group. The data of two participants in control group was missing, of which one missing participants was due to withdrawal of consent after randomisation and therefore that participant was excluded from the analysis. | Low risk of bias | The outcome measurement is an appropriate measure and it is unlikely that the measurement differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For this outcome "Any grade adverse events", there is a low risk of bias for all the domains. |
| Van den Berg 2022 | Low risk of bias | Eligible and consenting participants were randomized (1:1) to receive either a single infusion of 200–250 mL CCP or 200 mL of placebo, together with local standard of care. Random assignment was stratified by study site, age (≥ or < 65 years), and body mass index (BMI) (≥ or < 30 kg/m2). An electronic randomisation application (REDCap) hosted by SANB, was used to generate the treatment allocation. To mask treatment allocation, investigational product (IP) was covered in opaque paper wrapping prior to dispatch from the blood bank. | Low risk of bias | Both participants and those delivering the intervention were probably not aware of intervention received, as treatment allocation was masked (plasma was covered in opaque paper wrapping) and there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized (two participants withdrew consent and two were transferred to alternative facilities prior to transfusion) . | Some concerns | There is no information whether the measurement of the outcome was appropriate and it could be that it differed between intervention groups. The outcome assessors were aware of the intervention received, and knowledge of intervention received could have affected outcome measurement, but it is unlikely. | Some concerns | The data that produced this result was not analysed in accordance with the predefined outcomes stated in the trial registry and the protocol was not available. | Some concerns | ‐ |
Risk of bias for analysis 1.9 Grades 1‐2 adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Sekine 2021 | Low risk of bias | Patients were randomly assigned in a 1:1 ratio to receive two infusions 48 h apart of 300mL of CP plus Standard of Care (SOC) or SOC alone. Randomisation was performed using computer‐generated randomisation with random block sizes of 2 or 4 and stratified according to the unit of hospitalisation on enrolment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized | Some concerns | There is no information on whether the outcome measure was appropriate, but it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | ‐ |
Overall judgements for studies that included individuals with a confirmed diagnosis of COVID‐19 and mild disease
All‐cause mortality
Among those studies reporting a mortality outcome, we rated the overall risk of bias to be of some concern for CoV‐Early. There is not enough information available on the allocation sequence concealment, as the full‐text article is not published yet. In addition, the trials registry does not provide enough information on missing data (see Table 157).
Risk of bias for analysis 5.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. | Some concerns | There are some concerns, as the data was provided on request and the study results are not published yet. |
| Sullivan 2022 | Low risk of bias | Participants from all sites were randomized by a central web‐based system using blocked assignments to receive ABO compatible control plasma or CCP (both∼250 mL) in a 1:1 ratio. | Low risk of bias | Participants were blinded, but and those delivering the intervention were aware of intervention received. However, there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | A modified intention‐to‐treat was used, so data for this outcome was available for nearly all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of the intervention received could have affected outcome measurement. | Low risk of bias | There was no information on whether the data that produced this result was analysed in accordance with the predefined outcomes, as there is no protocol available and no information on mortality outcome provided in the trial registry. However, for this mortality outcome, the risk of bias remains still low. | Low risk of bias | ‐ |
Admission to hospital or death
Among those studies reporting admission to hospital or death, we rated the overall risk of bias to be of some concern for CoV‐Early. The trials registry did not provide enough information on the allocation sequence concealment and there are not enough information on missing data and the measurement of the outcome (see Table 158).
Risk of bias for analysis 5.2 Admission to hospital or death within 28 days.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on the measurement of the outcome. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. | Some concerns | There are some concerns, as the data was provided on request and the study results are not published yet. |
| Sullivan 2022 | Low risk of bias | Participants from all sites were randomized by a central web‐based system using blocked assignments to receive ABO compatible control plasma or CCP (both∼250 mL) in a 1:1 ratio. | Low risk of bias | Participants were blinded, but and those delivering the intervention were aware of intervention received. However, there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | A modified intention‐to‐treat was used, so data for this outcome was available for nearly all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. COVID‐19 relatedness for hospitalizations was adjudicated by a three physician panel masked to treatment group. | Low risk of bias | No protocol available, but the data that produced this result was analysed in accordance with the predefined outcome in the trial registry. | Low risk of bias | Low risk of bias in all the domains. |
Symptom resolution
We assessed symptom resolution on a study level, including symptom resolution by all initial symptoms resolved (asymptomatic) at day 14, day 28, and at longest follow‐up available, and time to symptom resolution. We rated the overall risk of bias to be of some concern for CoV‐Early for the outcome all initial symptoms resolved (asymptomatic) at day 14 and day 28. There is not enough information available yet on the allocation sequence concealment, as the full‐text article has not yet been published. In addition, the trials registry does not provide enough information on missing data (see Table 160; Table 159)
Risk of bias for analysis 5.4 All initial symptoms resolved (asymptomatic) at up to day 14.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | The data was provided on request by the study investigator. | Some concerns | There are some concerns, as the data was provided on request and the study results are not published yet. |
Risk of bias for analysis 5.3 All initial symptoms resolved (asymptomatic) at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | The data was provided on request by the study investigators. | Some concerns | There are some concerns, as the data was provided on request and the study results are not published yet. |
Quality of life
None of the studies that included individuals with a confirmed diagnosis of COVID‐19 and mild disease examined quality‐of‐life outcomes.
Safety
Among those studies reporting at least one of the safety outcomes, we rated the overall risk of bias to be of some concern for Alemany 2022. This study does not provide enough information to judge whether the outcome was measured appropriately (see Table 155; Table 156).
Risk of bias for analysis 4.8 Grades 3 and 4 adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome available for all participants randomized. | Some concerns | There is not enough information to judge whether the outcome was measured appropriately. | Low risk of bias | Data that produced this result analysed in accordance with a pre‐specified outcome of the protocol. | Some concerns | ‐ |
Risk of bias for analysis 4.9 Serious adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome available for all participants randomized. | Some concerns | This outcome was provided on request by the study investgators. | Low risk of bias | The data that produced this result was provided on request. | Some concerns | ‐ |
Effects of interventions
See: Table 1; Table 2; Table 3; Table 4; Table 5
Individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease
We present the summary of findings and the certainty of the evidence for our prioritised outcomes for individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease in Table 1 (convalescent plasma versus placebo or standard care alone), Table 2 (convalescent plasma versus standard plasma) and Table 3 (convalescent plasma versus human immunoglobulin). We assessed disease severity with the need for respiratory support according to the WHO clinical progression scale (WHO 2020e).
Convalescent plasma versus placebo or standard care alone
Primary outcomes (included in the summary of findings table)
All‐cause mortality
We assessed all‐cause mortality at day 28, day 60, by time until death and during hospital stay.
Twenty‐one studies reported all‐cause mortality for 19,021 participants. Considering the reported event rates across studies, we estimated that 225 of 1000 participants die at up to 28 days when treated with placebo or standard care alone. Treatment with convalescent plasma does not reduce all‐cause mortality at up to day 28 (RR 0.98, 95% CI 0.92 to 1.03; 220 per 1000; 21 studies, 19,021 participants; I² = 1%; high‐certainty evidence; Analysis 1.1); at day 60 (RR 0.74, 95% CI 0.49 to 1.12; 3 studies, 272 participants; I² = 0%; Analysis 1.2); when measured over time (HR 0.98, 95% CI 0.92 to 1.04; 16 studies, 17,070 participants; I² = 0%; Analysis 1.3); or during hospital stay (RR 0.97, 95% CI 0.87 to 1.08; 4 studies, 2556 participants; I² = 0%; Analysis 1.4).
1.1. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28
1.2. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 2: All‐cause mortality at up to day 60
1.3. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 3: All‐cause mortality (time to event)
1.4. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 4: All‐cause mortality during hospital stay
-
Subgroup analyses
Severity of disease: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to all‐cause mortality for people with moderate disease (WHO score 4 and 5) or severe disease (WHO score ≥ 6), according to the WHO Clinical Progression Scale (WHO 2020e; Analysis 1.1; Analysis 1.2; Analysis 1.3; Analysis 1.4).
Antibodies in recipients detected at baseline: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the detection of antibodies in the recipients at baseline (Analysis 6.1).
Length of time since symptom onset: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the length of time since symptom onset(Analysis 7.1).
Other subgroups: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the pre‐existing condition of immunosuppression at baseline (Analysis 10.1); diabetes at baseline (Analysis 12.1); respiratory disease at baseline (Analysis 14.1); or hypertension at baseline (Analysis 15.1); participants' age (Analysis 16.1), sex (Analysis 18.1), or country income groups according to the World Bank definitions (The World Bank 2022; Analysis 20.1).
We could not perform any further planned subgroup analyses for the comparison of convalescent plasma versus standard placebo or standard care alone and the outcome all‐cause mortality.
-
Sensitivity analyses
We summarised the effects of sensitivity analyses in Table 9. Reported effects of our main analysis were robust when we removed studies at high risk of bias or studies that were terminated early. We did not include any preprint articles in the main analysis of this outcome.
6.1. Analysis.
Comparison 6: Subgroup analysis: antibodies in recipients detected at baseline for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28 (random‐effects analysis)
7.1. Analysis.
Comparison 7: Subgroup analysis: length of time since symptom onset for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28 (random‐effects model)
10.1. Analysis.
Comparison 10: Subgroup analysis: pre‐existing condition immunosuppression for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28 (random‐effects model)
12.1. Analysis.
Comparison 12: Subgroup analysis: pre‐existing condition diabetes for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28
14.1. Analysis.
Comparison 14: Subgroup analysis: pre‐existing condition respiratory disease for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28
15.1. Analysis.
Comparison 15: Subgroup analysis: pre‐existing condition hypertension for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28
16.1. Analysis.
Comparison 16: Subgroup analysis: age of participants for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28
18.1. Analysis.
Comparison 18: Subgroup analysis: sex of participants for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28
20.1. Analysis.
Comparison 20: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28
4. Sensitivity analyses for the comparison of convalescent plasma versus placebo or standard care alone for the population of individuals with moderate to severe disease.
| Outcome | Main analysis | Risk of bias (excluding studiesa at high risk of bias) | Publication status (excluding preprintsb) | Study termination (excluding studies terminated earlyc) |
| All‐cause mortality at up to day 28 | RR 0.98 (95% CI 0.92 to 1.03); including 19,021 participants from 21 studies | RR 0.98 (95% CI 0.93 to 1.03); including 18,510 participants from 20 studiesa | RR 0.98 (95% CI 0.92 to 1.03); including 19,021 participants from 21 studies | RR 0.98 (95% CI 0.93 to 1.03); including 17,460 participants from 16 studiesc |
| Clinical worsening: need for invasive mechanical ventilation (for the subgroup of participants not requiring invasive mechanical ventilation at baseline, i.e. WHO ≤ 6) | RR 1.03 (95% CI 0.97 to 1.11); including 14,477 participants from 6 studies | RR 1.02 (95% CI 0.97 to 1.07); including 13,966 participants from 5 studiesa | RR 1.03 (95% CI 0.97 to 1.11); including 14,477 participants from 6 studies | RR 1.02 (95% CI 0.96 to 1.09); including 13,556 participants from 5 studiesc |
| Clinical improvement: participants discharged from hospital | RR 1.00 (95% CI 0.97 to 1.02); including 12,721 participants from 6 studies | RR 1.00 (95% CI 0.97 to 1.02); including 12,721 participants from 6 studies | RR 1.00 (95% CI 0.97 to 1.02); including 12.,21 participants from 6 studies | RR 1.00 (95% CI 0.97 to 1.02); including 12,534 participants from 4 studiesc |
| Quality of life | MD 1.00 (−2.14 to 4.14); including 483 participants from 1 study | MD 1.00 (−2.14 to 4.14); including 483 participants from 1 study | MD 1.00 (−2.14 to 4.14); including 483 participants from 1 study | MD 1.00 (−2.14 to 4.14); including 483 participants from 1 study |
| Grades 3 and 4 adverse events | RR 1.17 (95% CI 0.96 to 1.42); including 2392 participants from 6 studies | RR 1.17 (95% CI 0.96 to 1.42); including 2392 participants from 6 studies | RR 1.17 (95% CI 0.96 to 1.42); including 2392 participants from 6 studies | RR 1.18 (95% CI 0.77 to 1.80); including 1471 participants from 5 studiesc |
| Serious adverse events | RR 1.14 (95% CI 0.91 to 1.44) including 3901 participants from 6 studies | RR 1.14 (95% CI 0.91 to 1.44) including 3901 participants from 6 studies | RR 1.14 (95% CI 0.91 to 1.44) including 3901 participants from 6 studies | RR 1.10 (95% CI 0.81 to 1.50); including 2980 participants from 5 studiesc |
| CI: confidence interval; MD: mean difference; NR: not reported; RR: risk ratio; WHO: World Health Organization | ||||
aExcluded studies with high risk of bias (Korley 2021). bExcluded preprints: no study. cExcluded studies with premature termination (Avendano‐Sola 2021; Begin 2021; Gharbharan 2021; Li 2020; Van den Berg 2022).
Clinical status
We assessed the clinical status of participants by the need for invasive mechanical ventilation or death (worsening of clinical status) and by participants discharged from hospital (improvement of clinical status) up to day 28, day 60, and up to longest follow‐up.
Worsening of clinical status: six studies reported the need for invasive mechanical ventilation or death for 14,477 participants. Considering the reported event rates within the study, we estimated that 287 of 1000 participants not treated with convalescent plasma needed invasive mechanical ventilation or died. Evidence suggests that treatment with convalescent plasma has little to no impact on the need for invasive mechanical ventilation or death when compared to no convalescent plasma (RR 1.03, 95% CI 0.97 to 1.11; 296 per 1000; I² = 13%; high‐certainty evidence; Analysis 1.5).
Improvement of clinical status: six studies reported the number of participants discharged from hospital for 12,721 participants. Considering the reported event rates within the study, we estimated that 665 of 1000 participants not treated with convalescent plasma were discharged from hospital. Evidence suggests that treatment with convalescent plasma has no impact on whether participants are discharged from hospital when compared to no convalescent plasma (RR 1.00, 95% CI 0.97 to 1.02; 665 per 1000; I² = 0%; high‐certainty evidence; Analysis 1.6).
-
Subgroup analyses
Severity of disease: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the need for invasive mechanical ventilation or death for people with moderate disease (WHO score 4 and 5) or severe disease (WHO score ≥ 6), according to WHO Clinical Progression Scale (WHO 2020e; Analysis 1.5). We did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the number of participants discharged from hospital for people with moderate disease (WHO score 4 and 5), or severe disease (WHO score ≥ 6), according to WHO Clinical Progression Scale (WHO 2020e; Analysis 1.6).
Antibodies in recipients detected at baseline: for both clinical status outcomes, the test for subgroup differences suggests significant subgroup differences in the effectiveness of convalescent plasma with regard to antibodies in recipients detected at baseline (P = 0.02; I² = 82.8%; 2 studies; Analysis 6.3); (P = 0.04; I² = 76.9%; 2 studies; Analysis 6.4), however, due to the small number of studies included in both subgroup analyses and the high heterogeneity, we are uncertain whether the analyses produced useful findings.
Length of time since symptom onset: for the outcome 'need for invasive mechanical ventilation or death', we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the length of time since symptom onset onset; Analysis 7.3). For the outcome 'participants discharged from hospital, we identified significant subgroup differences in the effect of convalescent plasma with regard to the length of time since symptom onset (P = 0.09; I² = 65.2%; 2 studies; Analysis 7.4), however, due to the small number of studies included in the subgroup analysis and the high heterogeneity, we are uncertain whether the analysis produced useful findings.
-
Other subgroups
For the outcome 'need for invasive mechanical ventilation or death', we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the pre‐existing condition of diabetes at baseline (Analysis 12.2); or respiratory disease at baseline (Analysis 14.2). We identified significant subgroup differences for the effect of convalescent plasma with regard to participants' age (P = 0.02; I² = 68%; 2 studies; Analysis 16.2) and sex (P = 0.04; 3 studies; I² = 75.6%, Analysis 18.2), however due to the small number of studies included in the subgroup analyses and the high heterogeneity, we are uncertain whether the analyses produced useful findings. We did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to country income groups, according to the World Bank definitions (The World Bank 2022; Analysis 20.2).
For the outcome 'participants discharged from hospital', we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the pre‐existing condition of immunosuppression at baseline (Analysis 10.2); diabetes at baseline (Analysis 12.3); respiratory disease at baseline (Analysis 14.3); or hypertension at baseline (Analysis 15.2); or with regard to participants' age (Analysis 16.3), sex (Analysis 18.3); or country income groups, according to the World Bank definitions (The World Bank 2022; Analysis 20.3).
We could not perform any further planned subgroup analyses for the comparison of convalescent plasma versus placebo or standard care alone for both outcomes summarised under 'clinical status'.
Sensitivity analyses: we summarised the effects of sensitivity analyses in Table 9. Reported effects of our main analyses were robust when we removed studies at high risk of bias, or studies that were terminated early. We did not include any preprint articles in the main analysis of this outcome.
1.5. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 5: Clinical worsening: need for invasive mechanical ventilation or death at up to day 28
1.6. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 6: Clinical improvement: participants discharged from hospital
6.3. Analysis.
Comparison 6: Subgroup analysis: antibodies in recipients detected at baseline for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 3: Clinical worsening: need for invasive mechanical ventilation or death (random‐effects model)
6.4. Analysis.
Comparison 6: Subgroup analysis: antibodies in recipients detected at baseline for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 4: Clinical improvement: participants discharged from hospital
7.3. Analysis.
Comparison 7: Subgroup analysis: length of time since symptom onset for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 3: Clinical worsening: need for invasive mechanical ventilation or death (random‐effects model)
7.4. Analysis.
Comparison 7: Subgroup analysis: length of time since symptom onset for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 4: Clinical improvement: participants discharged from hospital
12.2. Analysis.
Comparison 12: Subgroup analysis: pre‐existing condition diabetes for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 2: Clinical worsening: need for invasive mechanical ventilation or death
14.2. Analysis.
Comparison 14: Subgroup analysis: pre‐existing condition respiratory disease for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 2: Clinical worsening: need for invasive mechanical ventilation or death
16.2. Analysis.
Comparison 16: Subgroup analysis: age of participants for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 2: Clinical worsening: need for invasive mechanical ventilation, or death
18.2. Analysis.
Comparison 18: Subgroup analysis: sex of participants for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 2: Clinical worsening: need for invasive mechanical ventilation or death
20.2. Analysis.
Comparison 20: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 2: Clinical worsening: need for invasive mechanical ventilation, or death at up to day 28
10.2. Analysis.
Comparison 10: Subgroup analysis: pre‐existing condition immunosuppression for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 2: Clinical improvement: participants discharged from hospital
12.3. Analysis.
Comparison 12: Subgroup analysis: pre‐existing condition diabetes for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 3: Clinical improvement: participants discharged from hospital
14.3. Analysis.
Comparison 14: Subgroup analysis: pre‐existing condition respiratory disease for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 3: Clinical improvement: participants discharged from hospital
15.2. Analysis.
Comparison 15: Subgroup analysis: pre‐existing condition hypertension for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 2: Clinical improvement: participants discharged from hospital
16.3. Analysis.
Comparison 16: Subgroup analysis: age of participants for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 3: Clinical improvement: participants discharged from hospital
18.3. Analysis.
Comparison 18: Subgroup analysis: sex of participants for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 3: Clinical improvement: participants discharged from hospital
20.3. Analysis.
Comparison 20: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 3: Clinical improvement: participants discharged from hospital
Quality of life
We assessed quality of life, including fatigue and neurological functioning of participants, if assessed with standardised scales (e.g. WHOQOL‐100) for the following time points: at up to 7 days, up to 28 days, and longest follow‐up available.
One study reported quality of life, assessed with the standardised scale EQ‐5D‐5L questionnaire at day 28, for 483 participants (Devos 2021). Considering the reported event rates within the study, we estimated a mean quality of life of 72 when treated without convalescent plasma. Evidence suggests that treatment with convalescent plasma may have little to no impact on quality of life up to day 28 when compared to no convalescent plasma (MD 1.00, 95% CI −2.14 to 4.14; low‐certainty evidence; Analysis 1.7). Our main reasons for downgrading were very serious imprecision, because of few participants and wide confidence interval.
1.7. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 7: Quality of life, assessed with standardised scales at day 28
Subgroup analyses: we could not perform any of our planned subgroup analyses for the comparison of convalescent plasma versus placebo or standard care alone.
Sensitivity analyses: we summarised the effects of sensitivity analyses in Table 9. We did not include any studies at high risk of bias, preprint articles, or studies terminated early in the main analysis of this outcome.
Adverse events (grade 3 to 4)
We defined the outcome as the number of participants with any grades 3 to 4 adverse event. We summarised data in Table 10, including the potential relationship between the intervention and the adverse event, as reported in the primary studies.
5. Adverse events of any grade.
| Study | Number of participants | Any grade AEs | |
| Convalescent plasma group | Control group | ||
| Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease | |||
| Agarwal 2020 | 227 in CP group and 224 in control group |
|
|
| AlQahtani 2021 | 20 in CP group and 20 in control group |
|
|
| Avendano‐Sola 2021 | 179 in CP group and 172 in control group |
|
|
| Bar 2021 | 40 in CP and 39 in control group |
|
NR |
| Begin 2021 | 614 in CP group, 307 in control group |
|
NR |
| De Santis 2022 | 36 in CP and 71 in control group |
|
NR |
| Devos 2021 | 320 in CP group, 163 in control group |
|
NR |
| Estcourt 2021 | 1078 in CP group, 909 in control group |
|
|
| Gharbharan 2021 | 43 in CP group and 43 in control group |
|
NR |
| Hamdy Salman 2020 | 15 in CP group and 15 in control group |
|
NR |
| Holm 2021 | 17 in CP and 14 in control group |
|
|
| Horby 2021b | 5795 in CP group and 5763 in control group |
|
NR |
| Kirenga 2021 | 69 in CP, 67 in control group |
|
|
| Koerper 2021 | 53 in CP, 52 in control group |
|
NR |
| Korley 2021 | 257 in CP group and 254 in control group |
|
NR |
| Li 2020 | 52 in CP group and 51 in control group |
|
NR |
| Menichetti 2021 | 232 in CP and 241 in control group |
|
NR |
| Ortigoza 2022 | 463 in CP and 463 in control group |
|
|
| Pouladzadeh 2021 | 30 in CP and 30 in control group |
|
NR |
| Ray 2022 | 40 in CP group 40 in control group |
|
NR |
| Sekine 2021 | 80 in CP and 80 in control group |
|
|
| Simonovich 2020 | 228 in CP group and 105 in control group |
|
|
| Van den Berg 2022 | 52 in CP group and 51 in control group |
|
|
| Convalescent plasma versus standard plasma for individuals with moderate to severe disease | |||
| Bajpai 2020 | 14 in CP group and 15 in control group |
|
|
| Baldeon 2022 | 63 in CP group and 95 in control group |
|
NR |
| Bennett‐Guerrero 2021 | 59 in CP, 15 in control group |
|
NR |
| NCT04421404 | 16 in CP group, 18 in control group |
|
|
| O’Donnell 2021 | 147 in CP group and 72 in control group |
|
|
| Convalescent plasma versus human immunoglobulin for individuals with moderate to severe disease | |||
| Beltran Gonzalez 2021 | 130 in CP, 60 in control group |
|
|
| Convalescent plasma versus placebo or standard care alone for individuals with asymptomatic or mild disease | |||
| Libster 2020 | 80 in CP group and 80 in control group |
|
|
| Alemany 2022 | 188 in CP, 188 in control group |
|
|
| Convalescent plasma versus standard plasma for individuals with asymptomatic or mild disease | |||
| Sullivan 2022 | 592 in CP and 589 in control group |
|
|
| AE: adverse event; CP: convalescent plasma; IVIg: intravenous immunoglobulin; NR: not reported; SAE: serious adverse event; SHOT: serious hazards of transfusion; TACO: transfusion‐associated circulatory overload; TAD: transfusion‐associated dyspnoea; TRALI: transfusion‐related acute lung injury | |||
Six studies reported grade 3 to 4 adverse events for both groups in a total of 2392 participants. Considering the reported event rates across studies, we estimated that 181 of 1000 participants not treated with convalescent plasma experience a grade 3 or 4 adverse event. Evidence suggests that treatment with convalescent plasma may have little to no effect on the risk of grades 3 and 4 adverse events (RR 1.17, 95% CI 0.96 to 1.42; 212 per 1000; I² = 31%; low‐certainty evidence; Analysis 1.10). Our main reasons for downgrading were serious indirectness, because some studies did not provide clear definitions of how the adverse events were measured and the number of events are inconsistent, and suspected publication bias, because most studies assessed and reported safety outcomes for the convalescent plasma group only.
1.10. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 10: Grades 3 and 4 adverse events
-
Subgroup analyses
Severity of disease: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to grade 3 or 4 adverse events for people with moderate disease (WHO score 4 and 5), or moderate to severe disease (WHO score ≥ 4), according to WHO Clinical Progression Scale (WHO 2020e; Analysis 1.10).
Other subgroups: we identified significant subgroup differences for the effect of convalescent plasma with regard to country income groups, according to the World Bank definitions (The World Bank 2022; P = 0.07; I² = 68.5%; 5 studies; Analysis 20.4), however due to the small number of studies included in the subgroup analyses and the high heterogeneity, we are uncertain whether the analyses produced useful findings.
We could not perform any further planned subgroup analyses for the comparison of convalescent plasma versus standard plasma for any adverse events.
Sensitivity analyses: we summarised the effects of sensitivity analyses in Table 9. The effects of our main analysis for the outcome grade 3 or 4 adverse events were robust when we removed a study that was terminated early. We did not include any studies at high risk of bias or preprint articles in the main analysis of this outcome.
20.4. Analysis.
Comparison 20: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 4: Grades 3 and 4 adverse events
Serious adverse events
We summarised data on serious adverse events reported only in participants who received convalescent plasma without reporting in the control group, and adverse events with a potential relationship to transfusion in Table 11.
6. Serious adverse events.
| Study | Number of participants | Serious adverse events | |
| CP group | Control group | ||
| Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease | |||
| Agarwal 2020 | 227 in CP group and 224 in control group | NR | NR |
| AlQahtani 2021 | 20 in CP group and 20 in control group | NR | NR |
| Avendano‐Sola 2021 | 38 in CP group and 43 in control group | NR | NR |
| Bar 2021 | 40 in CP group and 39 in control group |
|
|
| Begin 2021 | 614 in CP group, 307 in control group |
|
|
| CoV‐Early | 202 in CP group, 204 in control group | NR | NR |
| De Santis 2022 | 36 in CP group, 71 in control group | NR | NR |
| Devos 2021 | 320 in CP group, 163 in control group |
|
|
| Estcourt 2021 | 1078 in CP group, 909 in control group |
|
|
| Gharbharan 2021 | 43 in group and 43 in control group | NR | NR |
| Hamdy Salman 2020 | 15 in group and 15 in control group | NR | NR |
| Holm 2021 | 17 in CP group and 14 in control group | NR | NR |
| Horby 2021b | 5795 in group and 5763 in control group | NR | NR |
| Kirenga 2021 | 69 in CP group, 67 in control group | NR | NR |
| Koerper 2021 | 53 in CP, 52 in control group |
|
|
| Korley 2021 | 257 in CP group and 254 in control group | NR | NR |
| Menichetti 2021 | 232 in CP group and 241 in control group | NR | NR |
| Ortigoza 2022 | 463 in CP group and 463 in control group | NR | NR |
| Pouladzadeh 2021 | 30 in CP group and 30 in control group | NR | NR |
| Ray 2022 | 40 in CP group and 40 in control group | NR | NR |
| Sekine 2021 | 80 in CP group and 80 in control group | NR | NR |
| Simonovich 2020 | 228 in CP group and 105 in control group |
|
|
| Van den Berg 2022 | 52 in CP group and 51 in control group | NR | NR |
| Convalescent plasma versus standard plasma for individuals with moderate to severe disease | |||
| Bajpai 2020 | 14 in CP group and 15 in control group | NR | NR |
| Baldeon 2022 | 63 in CP group and 95 in control group | NR | NR |
| Bennett‐Guerrero 2021 | 59 in CP group, 15 in control group |
|
|
| NCT04421404 | 16 in CP group, 18 in control group |
|
|
| O’Donnell 2021 | 147 in CP group an 72 in control group |
|
|
| Convalescent plasma versus human immunoglobulin for individuals with moderate to severe disease | |||
| Beltran Gonzalez 2021 | 130 in CP group and 60 in control group | NR | NR |
| Convalescent plasma versus placebo or standard care alone for individuals with mild disease | |||
| Alemany 2022 | 188 in CP group and 188 in control group |
|
|
| Libster 2020 | 80 in CP group and 80 in control group | NR | NR |
| Convalescent plasma versus standard plasma for individuals with mild disease | |||
| Sullivan 2022 | 592 in CP group and 589 in control group | NR | NR |
| CP: convalescent plasma; NR: not reported; TACO: transfusion‐associated circulatory overload; TRALI: transfusion‐related acute lung injury; SAEs: serious adverse events | |||
Six studies reported serious adverse events for both groups and a total of 4901 participants. Considering the event rates reported across studies, we estimated that 118 of 1000 participants not treated with convalescent plasma experienced a serious adverse event. Evidence suggests that treatment with convalescent plasma probably has little to no effect on the risk of serious adverse events (RR 1.14, 95% CI 0.91 to 1.44; 135 per 1000; 6 studies, 3901 participants; I² = 45%%; moderate‐certainty evidence; Analysis 1.11). Our main reason for downgrading was suspected publication bias because most studies assessed and reported transfusion‐related events only; that is, they reported safety data only for the intervention group.
1.11. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 11: Serious adverse events
-
Subgroup analyses
Severity of disease: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to serious adverse events for people with moderate disease (WHO score 4 and 5), or moderate to severe disease (WHO score ≥ 4), according to WHO Clinical Progression Scale (WHO 2020e; Analysis 1.11).
Other subgroups: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to country income groups, according to the World Bank definitions (The World Bank 2022; Analysis 20.5).
We could not perform any further planned subgroup analyses for the comparison of convalescent plasma versus standard plasma for any outcome that summarised adverse events.
-
Sensitivity analyses
We summarised the effects of sensitivity analyses in Table 9. Reported effects of our main analysis for the outcome serious adverse events were robust when we removed a study that was terminated early. We did not include any studies at high risk of bias, or preprint articles in the main analysis of this outcome.
20.5. Analysis.
Comparison 20: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 5: Serious adverse events
Secondary outcomes (not included in the summary of findings table)
Improvement of clinical status
We assessed improvement of clinical status up to day 28, and up to longest follow‐up.
Two studies reported weaning or liberation from invasive mechanical ventilation in surviving patients for 630 participants; the subgroup of participants who were ventilated at baseline (i.e. WHO ≥ 7). Evidence suggests that treatment with convalescent plasma may have little to no impact on being weaned or liberated from invasive mechanical ventilation (RR 1.04, 95% CI 0.57 to 1.93; I² = 75%; Analysis 1.12). Two studies reported ventilator‐free days by day 28 for 1028 participants. Evidence suggests that treatment with convalescent plasma could have little to no impact on ventilator‐free days (MD −0.53, 95% CI −1.90 to 0.84; I² = 0%; Analysis 1.13). Two studies reported liberation from supplemental oxygen in surviving patients for 560 participants; the subgroup of participants requiring any supplemental oxygen or ventilator support at baseline. Evidence suggests little to no difference in the chance of being liberated from supplemental oxygen when treated with convalescent plasma (RR 0.99, 95% CI 0.91 to 1.08; I² = 0%; Analysis 1.14).
1.12. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 12: Clinical improvement: weaning or liberation from invasive mechanical ventilation in surviving participants, for subgroups of participants requiring invasive mechanical ventilation at baseline
1.13. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 13: Clinical improvement: ventilator‐free days by day 28
1.14. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 14: Clinical improvement: liberation from supplemental oxygen in surviving participants, for subgroup of participants requiring any supplemental oxygen or ventilator support at baseline
Need for dialysis
Two studies reported the need for dialysis at up to 28 days for 12,325 participants. Evidence suggests little to no difference between participants receiving convalescent plasma or not (OR 1.03, 95% CI 0.86 to 1.23; I² = 0%; Analysis 1.15).
1.15. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 15: Need for dialysis at up to 28 days
Admission to the ICU
Two studies reported admission to the ICU for 816 participants. Evidence suggests that treatment with convalescent plasma may have little to no impact on being admitted to the ICU (RR 0.93, 95% CI 0.77 to 1.11; I² = 0%; Analysis 1.16).
1.16. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 16: Admission to the intensive care unit (ICU)
Duration of hospitalisation
Two studies reported duration of hospitalisation for 97 participants. Evidence suggests that treatment with convalescent plasma may have little to no impact on the duration of hospitalisation when compared to placebo or standard care alone (MD −1.04, 95% CI −6.87 to 4.79; I² = 96%; Analysis 1.17).
1.17. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 17: Duration of hospitalisation
Viral clearance
We included data of viral clearance if assessed with RT‐PCR test for SARS‐CoV‐2 for the following time points: at baseline, up to 3, 7, and 14 days.
Four studies reported viral clearance for 674 participants. Evidence suggests that more people treated with convalescent plasma may achieve viral clearance at up to day 3 (RR 1.29, 95% CI 0.76 to 2.18; 4 studies, 619 participants; I² = 86%; Analysis 1.18) day 7 (RR 1.24, 95% CI 0.85 to 1.79; 4 studies, 674 participants; I² = 80%; Analysis 1.19), and day 14 (RR 1.46, 95% CI 0.58 to 3.68; 2 studies, 212 participants; I² = 93%; Analysis 1.20).
1.18. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 18: Viral clearance at up to day 3
1.19. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 19: Viral clearance at up to day 7
1.20. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 20: Viral clearance at up to day 14
Adverse events (any grade and grades 1 to 2)
We summarised data in Table 10, including the potential relationship between the intervention and the adverse event, as reported in the primary studies.
Seven studies reported adverse events of any grade for both groups and a total of 1809 participants. Evidence suggests little to no difference in the occurrence of any adverse events when treated with convalescent plasma (RR 1.05, 95% CI 0.95 to 1.17; I² = 0%; Analysis 1.8).
1.8. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 8: Any grade adverse events
One study reported adverse events of grades 1 to 2 for both groups and a total of 160 participants (Sekine 2021). Evidence suggests little to no difference in the occurrence of grades 1 to 2 events when treated with convalescent plasma (RR 1.11, 95% CI 0.87 to 1.41; Analysis 1.9).
1.9. Analysis.
Comparison 1: Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 9: Grades 1‐2 adverse events
Convalescent plasma versus standard plasma
Primary outcomes (included in the summary of findings table)
All‐cause mortality
We assessed all‐cause mortality at day 28, day 60, by time until death and during hospital stay.
Four studies reported all‐cause mortality for 484 participants. Considering the reported event rates within the study, we estimated that 177 of 1000 participants not treated with convalescent plasma die up to 28 days after treatment. We are uncertain whether convalescent plasma reduces or increases all‐cause mortality up to day 28 when compared to standard plasma (RR 0.73, 95% CI 0.45 to 1.19; 129 per 1000; 4 studies, 484 participants; I² = 16%, very low‐certainty evidence; Analysis 2.1); or when measured over time (HR 0.94, 95% CI 0.41 to 2.14; 4 studies, 484 participants; I² = 48%; Analysis 2.2). Our main reasons for downgrading were serious inconsistency because direction of effect was not consistent in both studies, and serious imprecision due to few participants and wide confidence intervals. All‐cause mortality at day 60 and during hospital stay was not reported in the studies.
2.1. Analysis.
Comparison 2: Convalescent plasma versus standard plasma for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28
2.2. Analysis.
Comparison 2: Convalescent plasma versus standard plasma for individuals with moderate to severe disease, Outcome 2: All‐cause mortality (time to event)
-
Subgroup analyses
Length of time since symptom onset: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the length of time since symptom onset (Analysis 8.1).
Other subgroups: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to country income groups, according to the World Bank definitions (The World Bank 2022; Analysis 21.1). We could not perform any further planned subgroup analyses for the comparison of convalescent plasma versus standard plasma and the outcome all‐cause mortality.
Sensitivity analyses: we summarised the effects of sensitivity analyses in Table 12. The effects of our main analysis were robust when we removed preprint articles and studies that were terminated early. We did not include any studies with high risk of bias in the main analysis of this outcome.
8.1. Analysis.
Comparison 8: Subgroup analysis: length of time since symptom onset for the comparison of convalescent plasma versus standard plasma for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28 (random‐effects analysis)
21.1. Analysis.
Comparison 21: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus standard plasma for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28
7. Sensitivity analyses for the comparison of convalescent plasma versus standard plasma for the population of individuals with moderate to severe disease.
| Outcome | Main analysis | Risk of bias (excluding studiesa at high risk of bias) | Publication status (excluding preprintsb) | Study termination (excluding studies terminated earlyc) |
| All‐cause mortality at up to day 28 | RR 0.73 (95% CI 0.45 to 1.19); including 484 participants from 4 studies | RR 0.73 (95% CI 0.45 to 1.19); including 484 participants from 4 studies | RR 0.65 (95% CI 0.43 to 1.01); including 455 participants from 3 studiesb | RR 0.95 (95% CI 0.17 to 5.29); including 252 participants from 2 studiesc |
| Clinical worsening: need for invasive mechanical ventilation (for the subgroup of participants not requiring invasive mechanical ventilation at baseline, i.e. WHO ≤ 6) | RR 5.59 (95% CI 0.29 to 108.38) including 34 participants from 1 study | RR 5.59 (95% CI 0.29 to 108.38) including 34 participants from 1 study | No study includedb | RR 5.59 (95% CI 0.29 to 108.38) including 34 participants from 1 study |
| Clinical improvement: participants discharged from hospital | NR | NR | NR | NR |
| Quality of life | NR | NR | NR | NR |
| Grades 3 or 4 adverse events | NR | NR | NR | NR |
| Serious adverse events | RR 0.80 (95% CI 0.55 to 1.15); including 327 participants from 3 studies | RR 0.80 (95% CI 0.55 to 1.15); including 327 participants from 3 studies | RR 0.77 (95% CI 0.53 to 1.13); including 293 participants from 2 studiesb | RR 0.76 (95% CI 0.51 to 1.14); including 253 participants from 2 studiesc |
| CI: confidence interval;NR: not reported; RR: risk ratio; WHO: World Health Organization | ||||
aExcluded studies with high risk of bias: no study. bExcluded preprints (no peer reviewed results published; Bajpai 2020; NCT04421404) cExcluded studies with premature termination (Baldeon 2022; Bennett‐Guerrero 2021)
Clinical status
We assessed the clinical status of participants by the need for invasive mechanical ventilation or death (worsening of clinical status) and by participants discharged from hospital (improvement of clinical status) up to day 28, day 60, and up to longest follow‐up.
Worsening of clinical status: one study reported the need for invasive mechanical ventilation or death for 34 participants (NCT04421404). Considering the reported event rates within the study, we estimated that 56 of 1000 participants treated with standard plasma needed invasive mechanical ventilation or died. Evidence is uncertain whether treatment with convalescent plasma reduces or increases the need for invasive mechanical ventilation or death when compared to standard plasma (RR 5.59, 95% CI 0.29 to 108.38; 311 per 1000; 1 study, 34 participants; very low‐certainty evidence; Analysis 2.3). Our main reasons for downgrading were extreme imprecision due to very few participants, very few events, and very wide confidence intervals.
Improvement of clinical status: we did not identify any study reporting this outcome.
Subgroup analyses: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to country income groups, according to the World Bank definitions (The World Bank 2022; Analysis 21.2). We could not perform any further planned subgroup analyses for the comparison of convalescent plasma versus standard plasma for the outcome, clinical worsening.
Sensitivity analyses: we summarised the effects of sensitivity analyses in Table 12. We did not see the reported effects of our main analysis when we removed the study with no published peer‐reviewed results. We did not include a study with high risk of bias, or a study that was terminated early in the main analysis of this outcome.
2.3. Analysis.
Comparison 2: Convalescent plasma versus standard plasma for individuals with moderate to severe disease, Outcome 3: Clinical worsening: need for invasive mechanical ventilation or death at up to day 28
21.2. Analysis.
Comparison 21: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus standard plasma for individuals with moderate to severe disease, Outcome 2: Clinical worsening: need for invasive mechanical ventilation or death at up to day 28
Quality of life
We did not identify any study reporting this outcome.
Adverse events (grades 3 to 4)
We defined the outcome as the number of participants with any grades 3 or 4 events. We summarised data in Table 10, including the potential relationship between intervention and adverse event, as reported in the primary studies.
None of the studies reported the number of participants who experienced grades 3 or 4 adverse events. However, O’Donnell 2021 reported separately the number of participants who experienced any grade 3 adverse event (27 events in 147 participants treated with convalescent plasma versus 17 events in 72 participants without convalescent plasma) and any grade 4 adverse event (26 events in 147 participants treated with convalescent plasma versus 15 events in 72 participants without convalescent plasma). We are uncertain whether convalescent plasma reduces or increases the risk of grade 3 or 4 adverse events compared to standard plasma. Our main concerns were serious indirectness, because outcome definitions differed from the definitions used in our review, and very serious imprecision because of few participants and few events.
Serious adverse events
We summarised data on serious adverse events reported only in participants who received convalescent plasma, with no reporting in the control group and including potential relationships between events and transfusion in Table 11.
Three studies reported serious adverse events for both groups and a total of 327 participants. Considering the reported event rates across studies, we estimated that 295 of 1000 participants not treated with convalescent plasma experience a serious adverse event. We are uncertain whether treatment with convalescent plasma reduces or increases the risk of serious adverse events when compared to standard plasma (RR 0.80, 95% CI 0.55 to 1.15; 236 per 1000; 3 studies, 327 participants; I² = 0%; very low‐certainty evidence; Analysis 2.6). Our main reasons for downgrading were serious imprecision because of few participants and few events, and suspected publication bias because most studies assessed and reported transfusion‐related events only; that is, they reported safety data only for the intervention group.
2.6. Analysis.
Comparison 2: Convalescent plasma versus standard plasma for individuals with moderate to severe disease, Outcome 6: Serious adverse events
-
Subgroup analyses
Severity of disease: we could not investigate subgroup differences between participants with moderate and severe disease, because the three studies included in the main analysis for this outcome included participants with moderate disease only (WHO score 4 and 5), according to WHO Clinical Progression Scale (WHO 2020e; Analysis 2.6).
Other subgroups: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to country income groups, according to the World Bank definitions (The World Bank 2022; Analysis 21.3).
We could not perform any further planned subgroup analyses for the comparison of convalescent plasma versus standard plasma for any outcome that summarised adverse events.
Sensitivity analyses: we summarised the effects of sensitivity analyses in Table 12. Reported effects of our main analysis were robust when we removed preprint articles and studies that were terminated early. We did not include any studies with high risk of bias in the main analysis of this outcome.
21.3. Analysis.
Comparison 21: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus standard plasma for individuals with moderate to severe disease, Outcome 3: Serious adverse events
Secondary outcomes (not included in the summary of findings table)
Improvement of clinical status
We did not identify any study reporting this outcome.
Need for dialysis
We did not identify any study reporting this outcome.
Admission to the ICU
We did not identify any study reporting this outcome.
Duration of hospitalisation
Two studies reported duration of hospitalisation for 189 participants. Evidence suggests that treatment with convalescent plasma may have little to no impact on the duration of hospitalisation when compared to standard plasma (MD −2.14, 95% CI −5.24 to 0.95; I² = 52%; Analysis 2.4).
2.4. Analysis.
Comparison 2: Convalescent plasma versus standard plasma for individuals with moderate to severe disease, Outcome 4: Duration of hospitalisation
Viral clearance
We did not identify any study reporting this outcome.
Adverse events (any grade and grade 1 to 2)
We defined the outcome as the number of participants with events of any grade and grades 1 to 2. We summarised data in Table 10, including the potential relationship between the intervention and the adverse event, as reported in the primary studies.
Two studies reported adverse events of any grade for both groups and a total of 253 participants. Evidence suggests an increase of any adverse events when treated with convalescent plasma (RR 1.18, 95% CI 0.93 to 1.50; I² = 0%; Analysis 2.5). None of the studies reported the number of participants who experienced grade 1 or 2 adverse events.
2.5. Analysis.
Comparison 2: Convalescent plasma versus standard plasma for individuals with moderate to severe disease, Outcome 5: Any grade adverse events
Convalescent plasma versus human immunoglobulin
Primary outcomes (included in the summary of findings table)
All‐cause mortality
We assessed all‐cause mortality at day 28, day 60, by time until death and during hospital stay.
One study reported all‐cause mortality for 190 participants (Beltran Gonzalez 2021). Considering the reported event rates across studies, we estimated that 433 of 1000 participants die at up to 28 days when treated with human immunoglobulin. Convalescent plasma may have little to no effect on all‐cause mortality at up to day 28 (RR 1.07, 95% CI 0.76 to 1.50; 464 per 1000; 1 study, 190 participants; low‐certainty evidence; Analysis 3.1); when measured over time (HR 1.14, 95% CI 0.84 to 1.54; 1 study, 190 participants; or low‐certainty evidence; Analysis 3.2); or during hospital stay (RR 1.01, 95% CI 0.76 to 1.34; 1 study, 190 participants; low‐certainty evidence; Analysis 3.3). Our main reason for downgrading was very serious imprecision due to few events and few participants. This study did not report all‐cause mortality at day 60.
3.1. Analysis.
Comparison 3: Convalescent plasma versus human immunoglobulin for individuals with moderate to severe disease, Outcome 1: All‐cause mortality at up to day 28
3.2. Analysis.
Comparison 3: Convalescent plasma versus human immunoglobulin for individuals with moderate to severe disease, Outcome 2: All‐cause mortality (time to event)
3.3. Analysis.
Comparison 3: Convalescent plasma versus human immunoglobulin for individuals with moderate to severe disease, Outcome 3: All‐cause mortality during hospital stay
Subgroup analyses: we could not perform any of our planned subgroup analyses for the comparison of convalescent plasma versus human immunoglobulin.
Sensitivity analyses: we did not summarise the effects of sensitivity analyses in a separate table, as we included only one study in the analysis for this outcome. We did not see the reported effects of our main analysis after we removed the study as a preprint article. We did not judge the study at high risk of bias, nor was it terminated early.
Clinical status
We did not identify any study reporting this outcome.
Quality of life
We did not identify any study reporting this outcome.
Adverse events (grades 3 to 4)
We did not identify any study reporting this outcome.
Serious adverse events
We did not identify any study reporting this outcome.
Secondary outcomes (not included in the summary of findings table)
Improvement of clinical status
We did not identify any study reporting this outcome.
Need for dialysis
We did not identify any study reporting this outcome.
Admission to the ICU
We did not identify any study reporting this outcome.
Duration of hospitalisation
We did not identify any study reporting this outcome.
Viral clearance
We did not identify any study reporting this outcome.
Adverse events (any grade and grades 1 to 2)
We did not identify any study reporting this outcome.
Individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease
We present certainty of the evidence for our prioritised outcomes for individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease in Table 4 and Table 5 (please see 'Summary of findings and assessment of the certainty of the evidence' in Methods). Table 4 includes the comparison of convalescent plasma versus placebo or standard care alone and Table 5 includes the comparison of convalescent plasma versus standard plasma.
Convalescent plasma versus placebo or standard care alone
Primary outcomes (included in the summary of findings table)
All‐cause mortality
We assessed all‐cause mortality at day 28, day 60, by time until death, and at longest follow‐up available.
Two studies reported all‐cause mortality for 536 participants. Considering the reported event rates across studies, we estimated that 22 of 1000 participants die at up to 28 days when treated with placebo or standard care alone. Evidence is uncertain whether or not treatment with convalescent plasma reduces all‐cause mortality at up to day 28 (OR 0.36, 95% CI 0.09 to 1.46; 8 per 1000; 2 studies, 536 participants; I² = 0%; very low‐certainty evidence; Analysis 4.1); or at up to day 60 (OR 0.13, 95% CI 0.01 to 2.16, 1 study, 376 participants; Analysis 4.2). Our main reasons for downgrading were serious indirectness, because in one study (Libster 2020), the authors defined the outcome as deaths associated with COVID‐19, and may not have reported other causes of mortality, and serious imprecision due to few events and wide confidence intervals. None of the studies reported all‐cause mortality measured over time.
4.1. Analysis.
Comparison 4: Convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 1: All‐cause mortality at up to day 28
4.2. Analysis.
Comparison 4: Convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 2: All‐cause mortality at up to day 60
-
Subgroup analyses
Other subgroups: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to country income groups, according to the World Bank definitions (The World Bank 2022; Analysis 22.1).
We could not perform any further planned subgroup analyses for the comparison of convalescent plasma versus placebo or standard care alone for this outcome.
22.1. Analysis.
Comparison 22: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 1: All‐cause mortality at up tp day 28
Admission to hospital or death within 28 days
One study reported admission to hospital or death within 28 days for 376 participants (Alemany 2022). Considering the reported event rates across studies, we estimated that 112 of 1000 participants are admitted to hospital or died within 28 days when treated with placebo or standard care alone. Evidence suggests that the treatment with convalescent plasma may have little to no impact on admission to hospital or death within 28 days (RR 1.05, 95% CI 0.60 to 1.84; 117 per 1000; low‐certainty evidence; Analysis 4.3). Our main reason for downgrading was very serious imprecision, due to very few events, few participants and wide confidence intervals.
4.3. Analysis.
Comparison 4: Convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 3: Admission to hospital or death within 28 days
Subgroup analyses: we could not perform any of our planned subgroup analyses for the comparison of convalescent plasma versus placebo or standard care alone for this outcome.
Symptom resolution
We assessed the outcome symptom resolution by all initial symptoms resolved (asymptomatic) at day 14, day 28, and at longest follow‐up available and by time to symptom resolution.
All initial symptoms resolved: we did not identify any study reporting this outcome.
Time to symptom resolution: one study reported time to symptom resolution for 376 participants (Alemany 2022). Considering the reported event rates across studies, we estimated that 500 of 1000 participants' symptoms resolved after 12 days when treated with placebo or standard care alone. Evidence suggests that treatment with convalescent plasma may have little to no impact on the time to symptom resolution (HR 1.05, 95% CI 0.85 to 1.30; 483 per 1000; low‐certainty evidence; Analysis 4.4). Our main reason for downgrading was very serious imprecision due to very few events, few participants, and wide confidence intervals.
Subgroup analyses: we could not perform any of our planned subgroup analyses for the comparison of convalescent plasma versus placebo or standard care alone for both outcomes of symptom resolution.
4.4. Analysis.
Comparison 4: Convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 4: Time to symptom resolution
Quality of life
We did not identify any study reporting this outcome.
Adverse events (grades 3 to 4)
We defined the outcome as the number of participants with any adverse events of grades 3 to 4. We summarised data in Table 10, including the potential relationship between the intervention and the adverse event, as reported in the primary study.
One study reported grades 3 or 4 adverse events for both groups and a total of 376 participants (Alemany 2022). Considering the reported event rates across studies, we estimated that 112 of 1000 participants not treated with convalescent plasma experience a grade 3 or 4 adverse event. Evidence suggests that treatment with convalescent plasma may have little to no impact on the risk of grade 3 and 4 adverse events (RR 1.29, 95% CI 0.75 to 2.19; 144 per 1000; low‐certainty evidence; Analysis 4.8). Our main reason for downgrading was very serious imprecision, due to few participants, few events and wide confidence intervals. One study reported that convalescent plasma was not associated with any "solicited adverse events" (Libster 2020). Because the definition was unclear and we were unsure whether only drug‐related adverse events were assessed, and we did not receive additional information from the study authors, we did not include this outcome in the analysis. We do not know whether convalescent plasma is associated with a higher risk for grades 3 or 4 adverse events (very low‐certainty evidence).
4.8. Analysis.
Comparison 4: Convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 8: Grades 3 and 4 adverse events
Subgroup analyses: we could not investigate the difference in the effectiveness of convalescent plasma with regard to country income groups, according to the World Bank definitions (The World Bank 2022), as the study included in this analysis was from a high‐income country (Alemany 2022; Analysis 22.2).
We could not perform any of our planned subgroup analyses for the comparison of convalescent plasma versus placebo or standard care alone for the adverse events analyses.
22.2. Analysis.
Comparison 22: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 2: Grades 3 and 4 adverse events
Serious adverse events
One study reported serious adverse events for both groups and a total of 376 participants (Alemany 2022). Considering the reported event rates across studies, we estimated that 117 of 1000 participants not treated with convalescent plasma experience a serious adverse event. Evidence suggests that treatment with convalescent plasma may have little to no impact on the risk of serious adverse events (RR 1.14, 95% CI 0.66 to 1.94; 133 per 1000; low‐certainty evidence; Analysis 4.9). Our main reason for downgrading was very serious imprecision, due to few participants, few events and wide confidence intervals. One study reported that convalescent plasma was not associated with any "solicited serious adverse events" (Libster 2020). Because the definition was unclear and we were unsure whether only drug‐related serious adverse events were assessed, and we did not receive additional information from the study authors, we did not include this outcome in the analysis. We do not know whether convalescent plasma is associated with a higher risk for serious adverse events (very low‐certainty evidence).
4.9. Analysis.
Comparison 4: Convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 9: Serious adverse events
Subgroup analyses: we could not investigate the difference in the effectiveness of convalescent plasma with regard to country income groups, according to the World Bank definitions (The World Bank 2022), as the study included in this analysis was from a high‐income country (Alemany 2022; Analysis 22.3).
We could not perform any of our planned subgroup analyses for the comparison of convalescent plasma versus placebo or standard care alone for the serious adverse events analysis.
22.3. Analysis.
Comparison 22: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 3: Serious adverse events
Secondary outcomes (not included in the summary of findings table)
Worsening of clinical status
We assessed worsening of clinical status at day 28, and up to the longest follow‐up available, by need for hospitalisation with oxygen by mask or nasal prongs, or death, and by need for invasive mechanical ventilation, or death.
Two studies reported the need for hospitalisation with oxygen by mask or nasal prongs, or death, at up to 28 days for 536 participants. Evidence suggests little to no difference between participants receiving convalescent plasma or not (RR 0.76, 95% CI 0.36 to 1.59; I² = 68%; Analysis 4.5). One study reported need for invasive mechanical ventilation or death up to day 28 and up to day 60 for 376 participants (Alemany 2022). Evidence suggests that fewer people treated with convalescent plasma may need invasive mechanical ventilation or die at up to day 28 (OR 0.51, 95% CI 0.10 to 2.55; Analysis 4.6), or at up to day 60 (OR 0.51, 95% CI 0.10 to 2.55; Analysis 4.7)
4.5. Analysis.
Comparison 4: Convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 5: Clinical worsening: need for hospitalisation with at least need of oxygen by mask or nasal prongs, or death
4.6. Analysis.
Comparison 4: Convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 6: Clinical worsening: need for invasive mechanical ventilation or death at up to day 28
4.7. Analysis.
Comparison 4: Convalescent plasma versus placebo or standard care alone for outpatients with mild disease, Outcome 7: Clinical worsening: need for invasive mechanical ventilation or death at up to day 60
Viral clearance (assessed with RT‐PCR)
We did not identify any study reporting this outcome.
Adverse events (any grade and grades 1 to 2)
We identified no study reporting adverse events of any grade or grades 1 to 2.
Convalescent plasma versus standard plasma
Primary outcomes (included in the summary of findings table)
All‐cause mortality
We assessed all‐cause mortality at day 28, day 60, by time until death, and at longest follow‐up available.
Two studies reported all‐cause mortality for 1597 participants. Considering the reported event rates across studies, we estimated that 5 of 1000 participants die at up to 28 days when treated with standard plasma. We are uncertain whether convalescent plasma reduces all‐cause mortality at up to day 28 (OR 0.30, 95% CI 0.05 to 1.75; 2 per 1000; 2 studies, 1597 participants; I² = 19%; very low‐certainty evidence; Analysis 5.1). Our main reason for downgrading was very serious imprecision due to few events and very wide confidence intervals. The studies did not report all‐cause mortality at up to day 60, or measured over time.
5.1. Analysis.
Comparison 5: Convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 1: All‐cause mortality at up to day 28
-
Subgroup analyses
Length of time since symptom onset: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the length of time since symptom onset (Analysis 9.1).
Other subgroups: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the pre‐existing condition of diabetes at baseline (Analysis 13.1), or participant's age (Analysis 17.1). We could not calculate the difference in the effectiveness of convalescent plasma with regard to the sex of participants, as none of the female participants died in either group (out of 43 with convalescent plasma and 52 with standard plasma), and one of the male participants died in each group (out of 163 with convalescent plasma and 158 with standard plasma; Analysis 19.1). We could not investigate the difference in the effectiveness of convalescent plasma with regard to country income groups, according to the World Bank definitions (The World Bank 2022), as both studies included in this analysis were set in high‐income countries (Analysis 23.1).
We could not perform any further planned subgroup analyses for the comparison of convalescent plasma versus standard plasma for this outcome.
9.1. Analysis.
Comparison 9: Subgroup analysis: length of time since symptom onset for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 1: All‐cause mortality at up to day 28
13.1. Analysis.
Comparison 13: Subgroup analysis: pre‐existing condition diabetes for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 1: All‐cause mortality at up to day 28
17.1. Analysis.
Comparison 17: Subgroup analysis: age of participants for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 1: All‐cause mortality at up to day 28
19.1. Analysis.
Comparison 19: Subgroup analysis: sex of participants for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 1: All‐cause mortality at up to day 28
23.1. Analysis.
Comparison 23: Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 1: All‐cause mortality at up to day 28
Admission to hospital or death within 28 days
Two studies reported admission to hospital or death within 28 days for 1595 participants. Considering the reported event rates across studies, we estimated that 73 of 1000 participants treated with standard plasma are admitted to hospital or died within 28 days. Evidence suggests that treatment with convalescent plasma probably reduces admission to hospital or death within 28 days (RR 0.49, 95% CI 0.31 to 0.75; 36 per 1000; I² = 0%; moderate‐certainty evidence; Analysis 5.2). Our main reason for downgrading was serious imprecision because the optimal information size is not met for a power of 0.90.
5.2. Analysis.
Comparison 5: Convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 2: Admission to hospital or death within 28 days
-
Subgroup analyses
Length of time since symptom onset: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the length of time since symptom onset (Analysis 9.2).
Other subgroups: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the pre‐existing condition of diabetes at baseline (Analysis 13.2), or sex of participants (Analysis 19.2). We identified significant subgroup differences for the effect of convalescent plasma with regard to the age of participants (P = 0.02; I² = 80.2%; 2 studies; Analysis 17.2), however due to the small number of studies included in both subgroup analyses and the large heterogeneity, we are uncertain whether the analyses produced useful findings.
We could not perform any further planned subgroup analyses for the comparison of convalescent plasma versus standard plasma for this outcome.
9.2. Analysis.
Comparison 9: Subgroup analysis: length of time since symptom onset for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 2: Admission to hospital or death within 28 days
13.2. Analysis.
Comparison 13: Subgroup analysis: pre‐existing condition diabetes for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 2: Admission to hospital or death within 28 days
19.2. Analysis.
Comparison 19: Subgroup analysis: sex of participants for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 2: Admission to hospital or death within 28 days
17.2. Analysis.
Comparison 17: Subgroup analysis: age of participants for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 2: Admission to hospital or death within 28 days
Symptom resolution
We assessed the outcome symptom resolution by all initial symptoms resolved (asymptomatic) at day 14, day 28, and at longest follow‐up available and by time to symptom resolution.
All initial symptoms resolved: one study reported all initial symptoms resolved for 417 participants (CoV‐Early). Considering the reported event rates across studies, we estimated that all symptoms of 657 of 1000 participants treated with standard plasma would be resolved at up to 28 days. Evidence suggests that treatment with convalescent plasma may have little to no effect on resolution of all initial symptoms up to day 28 (RR 1.12, 95% CI 0.98 to 1.27; 736 per 1000; 1 study, 416 participants; low‐certainty evidence; Analysis 5.3) or at up to day 14 (RR 1.00, 95% CI 0.83 to 1.21; 1 study, 417 participants; Analysis 5.4). Our main reason for downgrading was very serious imprecision due to few events, few participants, and wide confidence intervals.
Time to symptom resolution: we did not identify any study reporting this outcome.
-
Subgroup analyses
Length of time since symptom onset: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to length of time since symptom onset (Analysis 9.3).
Other subgroups: we did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the pre‐existing condition of diabetes at baseline (Analysis 13.3), participants' age (Analysis 17.3), or sex (Analysis 19.3).
We could not perform any further planned subgroup analyses for the comparison of convalescent plasma versus standard plasma for the outcome of all initial symptoms resolved.
5.3. Analysis.
Comparison 5: Convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 3: All initial symptoms resolved (asymptomatic) at up to day 28
5.4. Analysis.
Comparison 5: Convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 4: All initial symptoms resolved (asymptomatic) at up to day 14
9.3. Analysis.
Comparison 9: Subgroup analysis: length of time since symptom onset for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 3: All initial symptoms resolved (asymptomatic) at up to day 28
13.3. Analysis.
Comparison 13: Subgroup analysis: pre‐existing condition diabetes for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 3: All initial symptoms resolved (asymptomatic) at up to day 28
17.3. Analysis.
Comparison 17: Subgroup analysis: age of participants for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 3: All initial symptoms resolved (asymptomatic) at up to day 28
19.3. Analysis.
Comparison 19: Subgroup analysis: sex of participants for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 3: All initial symptoms resolved (asymptomatic) at up to day 28
Quality of life
We did not identify any study reporting this outcome.
Adverse events (grades 3 to 4)
We did not identify any study reporting this outcome.
Serious adverse events
We did not identify any study reporting this outcome.
Secondary outcomes (not included in the summary of findings table)
Worsening of clinical status
We assessed worsening of clinical status at day 28, and up to the longest follow‐up available, by need for hospitalisation with oxygen by mask or nasal prongs, or death, and by need for invasive mechanical ventilation, or death.
One study reported the need for hospitalisation with oxygen by mask or nasal prongs, or death, at up to 28 days for 1181 participants (Sullivan 2022). Evidence suggests that fewer people treated with convalescent plasma may need hospitalisation and receive oxygen by mask or nasal prongs, or die up to day 28 (RR 46, 95% CI 0.23 to 0.90; Analysis 5.5). One study reported need for invasive mechanical ventilation or death up to day 28 for 414 participants (CoV‐Early). Evidence suggests that fewer people treated with convalescent plasma may need invasive mechanical ventilation or die up to day 28 (OR 0.30, 95% CI 0.05 to 1.77; Analysis 5.6).
5.5. Analysis.
Comparison 5: Convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 5: Clinical worsening: need for hospitalisation with need of at least oxygen by mask or nasal prongs, or death
5.6. Analysis.
Comparison 5: Convalescent plasma versus standard plasma for outpatients with mild disease, Outcome 6: Clinical worsening: need for invasive mechanical ventilation or death at up to day 28
Viral clearance (assessed with RT‐PCR)
We did not identify any study reporting this outcome.
Adverse events (any grade and grades 1 to 2)
We did not identify any study reporting adverse events of any grade or grades 1 to 2.
Discussion
Summary of main results
The aim of this review was to assess the effectiveness and safety of convalescent plasma in the treatment of COVID‐19. This is the fifth version of our living systematic review.
We identified 33 RCTs that evaluated 24,861 participants, of whom 11,432 received convalescent plasma. We identified a further 49 ongoing studies evaluating convalescent plasma. We also identified 28 completed but not yet published studies and two studies terminated early for futility, that we categorised as 'Awaiting classification', as well as three platform trials that we have also placed in that category.
Effects of interventions
Individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease
29 RCTs investigated the use of convalescent plasma for 22,728 participants with moderate to severe disease, of which 23 RCTs compared convalescent plasma to placebo treatment or standard care alone, five compared convalescent plasma to standard plasma, and one RCT compared convalescent plasma to human immunoglobulin.
Convalescent plasma versus placebo or standard care alone
Convalescent plasma does not reduce all‐cause mortality at up to day 28 (RR 0.98, 95% CI 0.92 to 1.03; 220 per 1000; 21 RCTs, 19,021 participants; high‐certainty evidence). It has little to no impact on need for invasive mechanical ventilation or death (RR 1.03, 95% CI 0.97 to 1.11; 296 per 1000; 6 RCTs, 14,477 participants; high‐certainty evidence) and has no impact on whether participants are discharged from hospital (RR 1.00, 95% CI 0.97 to 1.02; 665 per 1000; 6 studies, 12,721 participants; high‐certainty evidence). Convalescent plasma may have little to no impact on quality of life, assessed with standardised scale EQ‐5D‐5L questionnaire (MD 1.00, 95% CI −2.14 to 4.14; 1 RCT, 483 participants; low‐certainty evidence). Convalescent plasma may have little to no impact on the risk of grade 3 and 4 adverse events (RR 1.17, 95% CI 0.96 to 1.42; 212 per 1000; 6 RCTs, 2392 participants; low‐certainty evidence). Convalescent plasma probably has little to no effect on the risk of serious adverse events (RR 1.12, 95% CI 0.96 to 1.31; 133 per 1000; 6 RCTs, 4901 participants; moderate‐certainty evidence).
We did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to the detection of antibodies in the recipients at baseline, to length of time since symptom onset, disease severity, country income groups, or certain pre‐existing conditions. For improvement and worsening of clinical status, we identified significant subgroup differences for the effect of convalescent plasma with regard to antibodies in recipients detected at baseline and age of people, but we are uncertain whether the analyses produced useful findings.
Convalescent plasma versus standard plasma
We are uncertain whether convalescent plasma reduces or increases all‐cause mortality at up to day 28 (RR 0.73, 95% CI 0.45 to 1.19; 129 per 1000; 4 studies, 484 participants; very‐low‐certainty evidence). We are uncertain whether convalescent plasma reduces or increases need for invasive mechanical ventilation or death (RR 5.59, 95% CI 0.29 to 108.38; 311 per 1000; 1 study, 34 participants; very low‐certainty evidence). None of the studies reported the number of participants who experienced grade 3 or 4 adverse events, only one study reported the number of participants with any adverse event of grade 3 and grade 4 separately. However, we are uncertain whether convalescent plasma reduces or increases the risk of grade 3 or 4 adverse events and the risk of serious adverse events (RR 0.80, 95% CI 0.55 to 1.15; 236 per 1000; 3 studies, 327 participants; very low‐certainty evidence). We identified no study reporting quality of life.
We did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to length of time since symptom onset, country income groups, or disease severity.
Convalescent plasma versus human immunoglobulin
Convalescent plasma may have little to no effect on all‐cause mortality at up to day 28 (RR 1.07, 95% CI 0.76 to 1.50; 464 per 1000; 1 study, 190 participants; low‐certainty evidence).
We identified no study reporting on clinical status, quality of life or safety.
Individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and mild disease
We identified four RCTs that investigated the use of convalescent plasma for 2133 participants with mild disease, of which two RCTs with 536 participants compared convalescent plasma to placebo or standard care alone, and two RCTs with 1597 participants compared convalescent plasma to standard plasma.
Convalescent plasma versus placebo or standard care alone
We are uncertain whether convalescent plasma reduces all‐cause mortality at up to day 28 (OR 0.36, 95% CI 0.09 to 1.46; 8 per 1000; 2 RCTs, 536 participants; very low‐certainty evidence). Evidence suggests that it may have little to no impact on admission to hospital or death within 28 days (RR 1.05, 95% CI 0.60 to 1.84; 117 per 1000; 1 study, 376 participants; low‐certainty evidence) and on time to COVID‐19 symptom resolution (HR 1.05, 95% CI 0.85 to 1.30; 483 per 1000; 1 study, 376 participants; low‐certainty evidence). Convalescent plasma may have little to no impact on the risk of grade 3 and 4 adverse events (RR 1.29, 95% CI 0.75 to 2.19; 144 per 1000; 1 study, 376 participants; low‐certainty evidence) and the risk of serious adverse events (RR 1.14, 95% CI 0.66 to 1.94; 133 per 1000; 1 study, 376 participants; low‐certainty evidence). We identified no study reporting symptom resolution or quality of life.
We did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to length of time since symptom onset. We were unable to perform any further subgroup analysis.
Convalescent plasma versus standard plasma
We are uncertain whether convalescent plasma reduces all‐cause mortality at up to day 28 (OR 0.30, 95% CI 0.05 to 1.75; 2 per 1000; 2 studies, 1597 participants; very low‐certainty evidence). It probably reduces admission to hospital or death within 28 days (RR 0.49, 95% CI 0.31 to 0.75; 36 per 1000; 2 studies, 1595 participants; moderate‐certainty evidence). Convalescent plasma may have little to no effect on resolution of all initial symptoms at up to day 28 (RR 1.12, 95% CI 0.98 to 1.27; 736 per 1000; 1 study, 416 participants; low‐certainty evidence). We did not identify any study reporting time to symptom resolution, quality of life or safety.
We did not identify any significant subgroup difference in the effectiveness of convalescent plasma with regard to country income groups, pre‐existing condition, or sex. For admission to hospital or death, we identified significant subgroup differences for the effect of convalescent plasma with regard to age, but we are uncertain whether the analyses produced useful findings.
Overall completeness and applicability of evidence
Most of the included participants had received additional treatment options to convalescent plasma or control, including, for instance, antivirals, antimicrobials, corticosteroids, hydroxychloroquine, respiratory support (extracorporeal membrane oxygenation, mechanical ventilation or oxygen), or a combination of those.
Thanks to high‐certainty evidence, we are confident that treatment with convalescent plasma does not reduce all‐cause mortality at up to 28 days, has little to no impact on need for invasive mechanical ventilation or death at day 28, and has no impact on whether participants are discharged from hospital when compared to placebo treatment or standard care alone. For the same comparison, we found low‐certainty evidence. Convalescent plasma may have little to no impact on quality of life and may result in a clinically relevant increased risk of grade 3 or 4 adverse events, but the treatment probably has little to no effect on the risk of serious adverse events (moderate‐certainty evidence). Not all the included RCTs reported adverse events for the control arm. When compared to standard plasma, we are uncertain about the effect of convalescent plasma on all‐cause mortality at up to 28 days, and very uncertain whethe convalescent plasma therapy increases the risk of need for invasive mechanical ventilation or death at day 28. For the same comparison, the evidence for grade 3 and 4 adverse events is uncertain (very‐low evidence), but convalescent plasma therapy may result in little to no impact on serious adverse events (low‐certainty evidence). When comparing the treatment to human immunoglobulin, we have low‐certainty evidence; convalescent plasma may have little to no impact on all‐cause mortality at up to 28 days.
Four RCTs investigated the use of convalescent plasma treatment for individuals with mild disease, of which two RCTs compared the therapy to placebo or standard care alone, and two RCTs compared it to standard plasma. We have low confidence in the evidence; treatment with convalescent plasma may have little to no impact on all‐cause mortality at up to day 28, on admission to hospital or death within 28 days and on time to symptom resolution. For the same comparison, convalescent plasma may result in a clinically relevant increased risk of grade 3 or 4 adverse events and serious adverse events (low‐certainty evidence). When compared to standard plasma, we have low certainty evidence; convalescent plasma may have little to no impact on all‐cause mortality at up to day 28, and on time to symptom resolution. For the same comparison, convalescent plasma probably reduces admission to hospital or death within 28 days (moderate‐certainty evidence).
28 RCTs have been completed, or completed their recruitment, and two RCTs have been terminated early, but no results have been published yet. We are also keeping track of three ongoing platform trials that might potentially add an intervention arm on convalescent plasma. Therefore, we categorised all these as 'Awaiting classification' and we will consider including them in an update of this review, once results are available.
We identified 49 ongoing studies, all RCTs assessing the benefits and safety of convalescent plasma therapy for the treatment of COVID‐19.
Quality of the evidence
Individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease
We included data from 23 RCTs to assess effectiveness and safety of convalescent plasma when compared to treatment with placebo or standard care alone. We were very confident in the identified evidence for effectiveness outcomes, but less confident in the identified evidence for safety outcomes. Our main concerns were that some studies still assessed and reported safety outcomes for the convalescent plasma group only, indicating publication bias and serious imprecision because the optimal information size was not met for a power of 0.90 (the optimal information size indicates the threshold for the number of included participants for an adequately powered trial or meta‐analysis; Guyatt 2011).
We included data from five RCTs to assess effectiveness and safety of convalescent plasma when compared to treatment with standard plasma. Our confidence in the evidence was very low to low for effectiveness and safety outcomes. Our main concerns were serious inconsistency for the outcome mortality at up to day 28, very serious imprecision for mortality and safety outcomes, and extreme serious imprecision for clinical worsening due to wide confidence intervals, few to extremely few participants and events, and because the optimal information size was not met for a power of 0.80. Also, some studies still reported safety outcomes for the convalescent plasma group only, indicating publication bias.
We included data from one RCT to assess effectiveness of convalescent plasma when compared to treatment with human immunoglobulin. We had low confidence in the identified evidence for mortality. Our main concern was very serious imprecision due to few events and few participants.
Individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and mild disease
We included data from two RCTs to assess effectiveness and safety of convalescent plasma when compared to treatment with placebo or standard care alone. We had low confidence in the identified evidence for effectiveness and safety outcomes. Our main concerns were serious to very serious imprecision due to the small information size and serious indirectness for all the reported outcomes, and serious indirectness for mortality, because the outcome definition did not precisely match our outcome definition.
We included data from two RCTs to assess effectiveness of convalescent plasma when compared to treatment with standard plasma. We had moderate to low confidence in the identified evidence for effectiveness. Our main concerns were serious imprecision for admission to hospital or death, because the optimal information size was not met for a power of 0.90. and very serious imprecision for mortality and symptom resolution, due to few events and very wide confidence intervals.
Potential biases in the review process
To avoid potential biases in the review process, we planned to include the best available evidence and adhered to the guidance provided in the Cochrane Handbook for Systematic Reviews of Interventions in all steps of the review (Higgins 2022). Even though COVID‐19 is still a novel disease, we identified more high‐quality evidence and included only RCTs in this version, as we have identified a large number of RCTs since the last update (Piechotta 2021). For this update, we were able to include 33 RCTs. To increase the informative value of our review, we are tracking all registered trials and will continually update this review as more evidence becomes available. There are currently still many new trials being registered in trials registries, as can be seen from the additional 49 RCTs added to the list of ongoing studies in this update of the review.
Two experienced information specialists developed a sensitive search strategy, to identify all ongoing and completed studies. We searched all relevant databases and trials registries, and two review authors conducted all review steps independently and in duplicate.
Another consideration for this rapidly evolving field is the availability of preprint articles that have not yet undergone peer review. We included preprints in this review, however, we are aware of the potentially lower quality of these publications, and investigated the robustness of our results in sensitivity analyses.
The necessary adaptation of review methods to the development of research output, as described in Table 6, is in general a potential source of bias in the review process. Since the available evidence has changed rapidly in a comparably short period of time during the COVID‐19 pandemic, we needed to take this approach to give a comprehensive answer to the review question. Before we start an update, our interdisciplinary team of review authors meets to review the methods and to discuss necessary amendments. We follow the methods we agreed upon before starting the update, and adhere to these decisions throughout each update process.
For this review update, we removed 'hyperimmune immunoglobulin' from the eligibility criteria because we identified several studies that evaluated this intervention treatment and therefore, we decided to assess this as a separate research question in a new review. We do not think that the bias arising from this adaptation was substantial, since the change was driven by objective reasons.
For changes in outcomes and outcome measurement, we specified the outcome 'worsening of clinical status' from standardised scales (WHO 2020e; WHO 2020f), in the third (Chai 2020), and fourth version (Piechotta 2021), to 'worsening of clinical status, assessed by need for invasive mechanical ventilation or death' and added the competing event of death to the outcome. We do not think that this change has led to any bias in the review process. Instead, we think that changing the inclusion criteria for outcome measurement to a standardised scale can facilitate identifying studies with objective and higher‐quality results and, additionally, can contribute to a lower heterogeneity among included studies. We also added a new secondary outcome 'quality of life' from version 2 of this review (Piechotta 2020b) onward, but do not suspect that this had an impact on bias since the outcome was suggested by an external patient representative.
For inclusion criteria regarding different study designs, we tried to anticipate possible changes of the evidence landscape already at protocol stage and therefore excluded study designs of lower‐level evidence, as more RCTs were published. So, for this update, we decided to only include RCTs, as enough RCT data were available to investigate the research question.
In previous versions of this review (Piechotta 2020b), we used RoB 1 for RCTs (Higgins 2011). We started using ROB 2 (Sterne 2019) from the second update (Chai 2020). This led to changes in the risk of bias rating and the GRADE assessment for the mortality, clinical status and safety outcomes of one included study (Li 2020). We think using RoB 2 corrected our judgement from potential personal biases, since it is less sensitive to subjective interpretations and allows for more nuanced assessment of potential bias in open‐label trials.
Agreements and disagreements with other studies or reviews
In the previous version of this systematic review, we compared our findings to two systematic reviews, Janiaud 2021, which included only RCTs, and Klassen 2021, which included RCTs together with other study designs. For the current update, we have been looking for more recent high‐quality, RCT‐only systematic reviews on the effect and safety of convalescent plasma for people with COVID‐19, to compare them to our RCT‐based findings.
Jorda 2022 aggregated data from 16 RCTs with 16,317 participants with moderate to severe disease, except one included trial (Libster 2020), and they found similar results to our review. Their meta‐analysis of 16 RCTs indicated that there is high‐certainty evidence for convalescent plasma not being associated with lower all‐cause morality (RR 0.97, 95% CI 0.90 to 1.04), lower initiation of mechanical ventilation (RR 0.97, 95% CI 0.88 to 1.07), and increased time to hospital discharge (HR 0.95, 95% CI 0.89‐1.02). They did not identify any subgroup differences for disease severity and baseline antibody level in recipients. Their results may be comparable to our findings, as all their included RCTs are also evaluated in our review.
Yang 2022, another systematic review, indicated similar results in their meta‐analysis. They included 14 RCTs with 4543 participants and found that convalescent plasma treatment for patients with severe COVID‐19 infection, who were critically ill, did not reduce mortality risk (RR 0.94, 95% CI 0.85 to 1.03, low‐certainty evidence) nor increase clinical improvement (RR 1.07, 95% CI 0.97 to 1.17, moderate‐certainty evidence). Also, there were no subgroup differences for disease severity.
Fernández‐Lázaro 2022, a systematic review that included six studies (classified as RCTs) in a qualitative synthesis, indicated contradictory results to our review. They found that treatment with convalescent plasma in hospitalised COVID‐19 patients could decrease mortality, viral load and period of infection, without the occurrence of serious adverse events. One potential reason for the difference between their results and ours could be the inclusion of Liu 2020a, Rasheed 2020 and Zeng 2020, which they had classified as RCTs. We did not identify these studies as RCTs, however, and therefore excluded them from our review analyses. Rasheed 2020, for instance, reported that controls were matched to participants according to the disease stage, age, and sex, and assigned participants to convalescent plasma based on ABO compatibility and limited availability of plasma. In our opinion, this does not fit the criteria for a randomised allocation method and we classified it as a controlled non‐randomised study of interventions (NRSI), since we did not receive any further information on their methods.
Also, the FDA and the USA Government considered that there was sufficient evidence of efficacy to widen access to convalescent plasma under the 'Emergency Use Authorization' (EUA) issued on 23 August 2020 (FDA 2020). However, on 11 February 2021, the FDA revised the EUA for convalescent plasma. The authorisation is now limited to the use of high‐titre convalescent plasma for hospitalised individuals at an early stage of disease (FDA 2021).
Furthermore, Bartelt 2022, an RCT available in preprint, evaluated the safety and effectiveness of convalescent plasma with high‐ (> 1:640) compared to lower‐ (≥ 1:160) neutralising antibody titres. The study authors reported that participants treated with high titre convalescent plasma had earlier hospital discharge and lower occurrences of life‐threatening serious adverse events. Even though this study is limited by its small sample size, and further RCTs are needed to confirm their findings, the results might indicate how varying titre levels can have an impact on between‐study differences in our analyses. This could be one potential reason why some of our included RCTs have different safety and effectiveness results.
Authors' conclusions
Implications for practice.
We are very confident in the evidence that convalescent plasma for the treatment of individuals with moderate to severe disease does not reduce mortality and has little to no impact on the need for invasive mechanical ventilation or death (clinical worsening) and participants discharged from hospital (clinical improvement) when compared to placebo or standard care alone. Convalescent plasma probably has little to no effect on the risk of serious adverse events in such patients. Further, we identified very low‐ to low‐certainty evidence for the effects of convalescent plasma when compared to standard plasma in individuals with moderate to severe COVID‐19. Convalescent plasma therapy may result in little to no impact on serious adverse events, in such patients. For the same population, when compared to human immunoglobulin treatment, we identified low‐certainty evidence for mortality.
For individuals with a diagnosis of SARS‐CoV‐2 infection and mild disease, we currently have low confidence in the evidence for the effects of convalescent plasma when compared to placebo or standard care alone. For the same comparison, convalescent plasma may result in a clinically relevant increased risk of grade 3 or 4 adverse events and serious adverse events. When compared to standard plasma, we have low‐ to moderate confidence in the evidence on the effects of convalescent plasma, in such patients.
There is some evidence suggesting subgroup differences with regard to antibodies in recipients at baseline and participants' age, for the treatment of individuals with moderate to severe disease. For individuals with mild disease, there is no evidence of a difference in effect with regard to subgroups on length of time since symptom onset, country income groups, certain pre‐existing conditions, or sex.
Implications for research.
For the fifth version of this living systematic review investigating the use of convalescent plasma for people with COVID‐19, we included data from 33 randomised controlled trials (RCTs) reporting on the effectiveness and safety of convalescent plasma, and in total considered the experience of 24,861 participants. Studies should more consistently report outcomes and, if relevant, consider the competing event of death. Further, studies should consider standard treatment changes during the pandemic to ensure and improve comparability in terms of co‐interventions administered in all study arms. Concerning adverse events, studies should try to blind at least outcome assessors, and report adverse events consistently, and for both the intervention and the control arm.
Given the evidence of subgroup difference with regard to antibodies in recipients at baseline, additional studies evaluating convalescent plasma treatment in asymptomatic participants would be of interest to assess whether the intervention is more effective if given earlier in the course of the disease. Additional data are needed for patients with any disease severity who are immune‐suppressed, as well as subgroup data for plasma from SARS‐CoV‐2 variants, and different ethnicities. Future studies should also report outcomes based on antibody titre in the donor plasma administered.
There are 49 ongoing studies evaluating convalescent plasma and 30 studies reporting in a trials registry as being completed or terminated. We will also keep track of three platform trials. Publication of the results might resolve some of the uncertainties around convalescent plasma therapy for people with asymptomatic or mild disease and for certain subgroups.
What's new
| Date | Event | Description |
|---|---|---|
| 31 January 2023 | New citation required and conclusions have changed | Only RCTs were searched for this update. High‐certainty evidence on further primary outcomes and more safety data available. Subgroup analyses were conducted and further studies on individuals with mild disease were included in the meta‐analysis. |
| 31 January 2023 | New search has been performed | 33 RCTs included |
History
Review first published: Issue 5, 2020
| Date | Event | Description |
|---|---|---|
| 19 March 2021 | New citation required and conclusions have changed | High certainty in the evidence for some of the prioritised outcomes |
| 17 March 2021 | New search has been performed | 12 RCTs and one NRSI included |
| 30 August 2020 | New search has been performed | Two RCTs, eight controlled NRSIs and nine non‐controlled NRSIs included |
| 30 August 2020 | New citation required and conclusions have changed | Additional safety data included (more than 20,000 participants) |
| 3 June 2020 | New citation required and conclusions have changed | We included results from one RCT and three controlled NRSIs and added further safety data from non‐controlled NRSIs. |
| 31 May 2020 | New search has been performed | We included eight new studies. |
Risk of bias
Risk of bias for analysis 1.12 Clinical improvement: weaning or liberation from invasive mechanical ventilation in surviving participants, for subgroups of participants requiring invasive mechanical ventilation at baseline.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants requiring invasive mechanical ventilation at baseline, which was 13 out of 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | There was no information on whether the data that produced this result was analysed in accordance with a pre‐specified analysis plan, as the data was provided on request by the study investigators. | Some concerns | For the outcome "liberation from invasive mechanical ventilation", there is a low risk of bias from the randomization process, due to missing outcome data, in measurement of the outcome and in selection of the reported result. However, there are some concerns for bias due to deviations from intended interventions. |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants requiring invasive mechanical ventilation at baseline, a total of 617 out of 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "Weaning or liberation from invasive mechanical ventilation", there was a low risk of bias for all the domains. |
Risk of bias for analysis 1.13 Clinical improvement: ventilator‐free days by day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so the interim analysis was no deviation from the protocol and the analysis was appropriate. | Low risk of bias | Data were available for nearly all (921/938) participants. 17 Participants were lost to follow‐up at day 30 (after discharge). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| De Santis 2022 | Some concerns | The study methods it is stated that this is a "investigator‐initiated multicenter open‐label randomized controlled trial", but no further information on the allocation sequence concealment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data for this outcome was provided by the study investigators on request. | Some concerns | ‐ |
Risk of bias for analysis 1.14 Clinical improvement: liberation from supplemental oxygen in surviving participants, for subgroup of participants requiring any supplemental oxygen or ventilator support at baseline.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Devos 2021 | Low risk of bias | Patients were randomised through a computerised system (RedCap®, Vanderbilt University, USA, Version 10.6.13) according to a 2:1 allocation scheme stratified by study site using randomly selected block sizes of 6 or 9, to open‐label convalescent plasma combined with standard of care (intervention group) or standard of care alone (control group). | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of the intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined time‐to‐event parameters with competing risk of the outcome "time to clinical improvement" and the planned estimation of the event rates using cumulative incidence functions (CIF), stated in the trial protocol. | Low risk of bias | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants requiring any supplemental oxygen or ventilator support at baseline, being 77 out of 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | For the outcome "liberation from supplemental oxygen", there is a low risk of bias from the randomization process, due to missing outcome data, in measurement of the outcome and in selection of the reported result. However, there are some concerns for bias due to deviations from intended interventions. |
Risk of bias for analysis 1.16 Admission to the intensive care unit (ICU).
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Devos 2021 | Low risk of bias | Patients were randomised through a computerised system (RedCap®, Vanderbilt University, USA, Version 10.6.13) according to a 2:1 allocation scheme stratified by study site using randomly selected block sizes of 6 or 9, to open‐label convalescent plasma combined with standard of care (intervention group) or standard of care alone (control group). | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized data was available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | There is not enough information available, but it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Some concerns | The outcome was not planned or pre‐specified in the protocol. | Some concerns | ‐ |
Risk of bias for analysis 1.18 Viral clearance at up to day 3.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Agarwal 2020 | Low risk of bias | The allocation sequence was random and concealed and there were no differences between intervention groups suggesting a problem with the randomization process. | Some concerns | Five participants in convalescent plasma arm and four in control arm did not receive the allocated intervention, which could be due to awareness of the intervention, as this trial was open‐label. According to the study, a per protocol analysis was performed for this secondary outcome, but there was no potential for a substantial impact (on the result) of the failure to analyse randomized participants. | Some concerns | Data for this outcome was available for 367 out of 451 participants included in the per protocol analysis. Data of 51 participants in the convalescent plasma arm and of 46 participants in the control arm were missing and there is no evidence that the result was not biased by missing outcome data. However, there it is not likely that missingness in the outcome depended on its true value. | High risk of bias | There was no information on whether the method of measuring the outcome was inappropriate, as it was unclear how the outcome was exactly measured and the measurement could have differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | High risk of bias | For this outcome "viral clearance", there is a low risk of bias from the randomization process and in selection of the reported result, there are some concerns for bias due to deviations from intended interventions and due to missing outcome data. However, there is a high risk in measurement of the outcome. |
| Hamdy Salman 2020 | Low risk of bias | Participants were block randomized using website software in a 2:1 ratio to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 30 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Some concerns | The data that produced this result was not analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | For this outcome " viral clearance at day 3", there is a low risk of bias from the randomization process, due to deviations from intended interventions, due to missing outcome data and in measurement of the outcome. However, there are some concerns for bias in selection of the reported result. |
| Kirenga 2021 | Low risk of bias | Randomisation was performed using permuted block randomisation with varying block sizes ranging from 4 to 8. Generation of randomisation numbers was done by the study biostatistician, who did not have contact with participants. The biostatistician printed individual randomisation numbers with their corresponding treatment assignment and placed each number in an opaque envelope. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | ‐ |
| Li 2020 | Low risk of bias | Participants were block randomized via computer‐generated random numbering in a 1:1 ratio to receive standard treatment coupled with convalescent plasma transfusion or standard treatment alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was not available for all, or nearly all, participants randomized (47/52 participants in plasma group and 40/51 participants in control group) and there is no evidence that the result was not biased by missing outcome data. However, it is unlikely that the missingness in the outcome could depend on its true value. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "viral clearance", there is a low risk of bias in all the domains. |
Risk of bias for analysis 1.19 Viral clearance at up to day 7.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Agarwal 2020 | Low risk of bias | The allocation sequence was random and concealed and there were no differences between intervention groups suggesting a problem with the randomization process. | Some concerns | Five participants in convalescent plasma arm and four in control arm did not receive the allocated intervention, which could be due to awareness of the intervention, as this trial was open‐label. According to the study, a per protocol analysis was performed for this secondary outcome, but there was no potential for a substantial impact (on the result) of the failure to analyse randomized participants. | Some concerns | Data for this outcome was available for 342 out of 451 participants included in the per protocol analysis and there is no evidence that the result was not biased by missing outcome data. However, there it is not likely that missingness in the outcome depended on its true value. | High risk of bias | There was no information on whether the method of measuring the outcome was inappropriate, as it was unclear how the outcome was exactly measured and the measurement could have differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | High risk of bias | For this outcome "viral clearance", there is a low risk of bias from the randomization process and in selection of the reported result, there are some concerns for bias due to deviations from intended interventions and due to missing outcome data. However, there is a high risk in measurement of the outcome. |
| Kirenga 2021 | Low risk of bias | Randomisation was performed using permuted block randomisation with varying block sizes ranging from 4 to 8. Generation of randomisation numbers was done by the study biostatistician, who did not have contact with participants. The biostatistician printed individual randomisation numbers with their corresponding treatment assignment and placed each number in an opaque envelope. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | ‐ |
| Li 2020 | Low risk of bias | Participants were block randomized via computer‐generated random numbering in a 1:1 ratio to receive standard treatment coupled with convalescent plasma transfusion or standard treatment alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was not available for all, or nearly all, participants randomized (47/52 participants in plasma group and 40/51 participants in control group) and there is no evidence that the result was not biased by missing outcome data. However, it is unlikely that the missingness in the outcome could depend on its true value. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "viral clearance", there is a low risk of bias in all the domains. |
| Sekine 2021 | Low risk of bias | Patients were randomly assigned in a 1:1 ratio to receive two infusions 48 h apart of 300mL of CP plus Standard of Care (SOC) or SOC alone. Randomisation was performed using computer‐generated randomisation with random block sizes of 2 or 4 and stratified according to the unit of hospitalisation on enrolment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Some concerns | Data for this outcome was only available for 117/160 participants randomized, and the missingness in the outcome could depend on its true value, but it is unlikely. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | ‐ |
Risk of bias for analysis 1.20 Viral clearance at up to day 14.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Kirenga 2021 | Low risk of bias | Randomisation was performed using permuted block randomisation with varying block sizes ranging from 4 to 8. Generation of randomisation numbers was done by the study biostatistician, who did not have contact with participants. The biostatistician printed individual randomisation numbers with their corresponding treatment assignment and placed each number in an opaque envelope. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | ‐ |
| Li 2020 | Low risk of bias | Participants were block randomized via computer‐generated random numbering in a 1:1 ratio to receive standard treatment coupled with convalescent plasma transfusion or standard treatment alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was not available for all, or nearly all, participants randomized (47/52 participants in plasma group and 40/51 participants in control group) and there is no evidence that the result was not biased by missing outcome data. However, it is unlikely that the missingness in the outcome could depend on its true value. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "viral clearance", there is a low risk of bias in all the domains. |
Risk of bias for analysis 2.3 Clinical worsening: need for invasive mechanical ventilation or death at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| NCT04421404 | Low risk of bias | Subjects enrolled in the study will be randomized using a web based randomization procedure to receive convalescent plasma versus non‐immune plasma at a 1:1 ratio. | Low risk of bias | Both participants and those delivering the intervention were not aware of intervention received, there were no deviations from intended interventions and the analysis was appropriate. The nursing staff may be unblinded to the treatment assignment to ensure proper ABO checking of the plasma unit at bedside per standard transfusion procedures. | Low risk of bias | Data for this outcome was available for all 34 participants randomized. | Low risk of bias | The are no information available (as the study still needs to be published in a journal) on whether the measurement of the outcome was appropriate, but it is unlikely that it differed between intervention groups. The outcome assessors were blinded to the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 2.4 Duration of hospitalisation.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Bajpai 2020 | Low risk of bias | Participants were block randomized with concealed allocation using the Sequentially Numbered Opaque Sealed Envelopes (SNOSE) method, to receive either convalescent plasma with standard of care or standard plasma (fresh frozen plasma) with standard of care. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | ‐ | Low risk of bias | ‐ | Low risk of bias | ‐ | Low risk of bias | ‐ | Low risk of bias | ‐ |
| Baldeon 2022 | Some concerns | This was a random double blinded, non‐convalescent plasma (NCP)‐placebo controlled clinical trial. However, there are no information provided on the concealment of the allocation sequence:"A simple randomization scheme was used to allocate participants in the two treatment groups." | Low risk of bias | This study was triple blinded. Hospital personnel at the transfusion medicine services, physicians, and the participants were unaware which plasma was convalescent and which plasma was control. Intention to treat analysis was used. | Low risk of bias | Data was available for all the 158 participants randomised. They report that in the convalescent plasma group 66 participants were analysed, however only 63 were randomised to that group, so probably a tipo. | Low risk of bias | Clinical outcomes were assessed by investigators who were blinded to the treatment groups. | Low risk of bias | No protocol was available, but the outcome was pre‐specified in the trial registry. | Some concerns | Some concerns due to bias arising from the randomization process. |
Risk of bias for analysis 2.5 Any grade adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| NCT04421404 | Low risk of bias | Subjects enrolled in the study will be randomized using a web based randomization procedure to receive convalescent plasma versus non‐immune plasma at a 1:1 ratio. | Low risk of bias | Both participants and those delivering the intervention were not aware of intervention received, there were no deviations from intended interventions and the analysis was appropriate. The nursing staff may be unblinded to the treatment assignment to ensure proper ABO checking of the plasma unit at bedside per standard transfusion procedures. | Low risk of bias | Data for this outcome was available for all 34 participants randomized. | Low risk of bias | The are no information available (as the study still needs to be published in a journal) on whether the measurement of the outcome was appropriate, but it is unlikely that it differed between intervention groups. The outcome assessors were blinded to the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | Low risk of bias | ‐ |
| O’Donnell 2021 | Low risk of bias | Participants were block randomized using a web‐based randomization platform in a 2:1 ratio to receive either convalescent plasma or standard plasma and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all (219 out of 223) participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | For the outcome "any adverse events", there is a low risk of bias for all the domains. |
Risk of bias for analysis 4.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably unaware of the intervention received. | Low risk of bias | Data that produced this result analysed in accordance with a pre‐specified outcome of the protocol. | Low risk of bias | ‐ |
| Libster 2020 | Low risk of bias | Participants were randomized using an electronic system numbering to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | The participants were unaware of their assigned intervention and those delivering the intervention were aware of intervention assigned during the trial, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 160 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably unaware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For this outcome "all‐cause mortality", there is a low risk of bias for all the domains. |
Risk of bias for analysis 4.2 All‐cause mortality at up to day 60.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably unaware of the intervention received. | Low risk of bias | Data that produced this result analysed in accordance with a pre‐specified outcome of the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 4.3 Admission to hospital or death within 28 days.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome available for all participants randomized. | Low risk of bias | This outcome was provided on request by the study investgators. | Low risk of bias | The data that produced this result was provided on request. | Low risk of bias | ‐ |
Risk of bias for analysis 4.4 Time to symptom resolution.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably unaware of the intervention received. | Low risk of bias | Data that produced this result analysed in accordance with a pre‐specified outcome of the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 4.5 Clinical worsening: need for hospitalisation with at least need of oxygen by mask or nasal prongs, or death.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome available for all participants randomized. | Low risk of bias | This outcome was provided on request by the study investgators. | Low risk of bias | The data that produced this result was provided on request. | Low risk of bias | ‐ |
| Libster 2020 | Low risk of bias | Participants were randomized using an electronic system numbering to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | The participants were unaware of their assigned intervention and those delivering the intervention were aware of intervention assigned during the trial, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 160 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably unaware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 4.6 Clinical worsening: need for invasive mechanical ventilation or death at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome available for all participants randomized. | Low risk of bias | This outcome was provided on request by the study investgators. | Low risk of bias | The data that produced this result was provided on request. | Low risk of bias | ‐ |
Risk of bias for analysis 4.7 Clinical worsening: need for invasive mechanical ventilation or death at up to day 60.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome available for all participants randomized. | Low risk of bias | This outcome was provided on request by the study investgators. | Low risk of bias | The data that produced this result was provided on request. | Low risk of bias | ‐ |
Risk of bias for analysis 5.5 Clinical worsening: need for hospitalisation with need of at least oxygen by mask or nasal prongs, or death.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Sullivan 2022 | Low risk of bias | Participants from all sites were randomized by a central web‐based system using blocked assignments to receive ABO compatible control plasma or CCP (both∼250 mL) in a 1:1 ratio. | Low risk of bias | Participants were blinded, but and those delivering the intervention were aware of intervention received. However, there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | A modified intention‐to‐treat was used, so data for this outcome was available for nearly all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. COVID‐19 relatedness for hospitalizations was adjudicated by a three physician panel masked to treatment group. | Low risk of bias | There was no information on whether the data that produced this result was analysed in accordance with the predefined outcomes, as there is no protocol available and no information on this outcome provided in the trial registry. However, for this this outcome, the risk of bias remains still low. | Low risk of bias | Low risk of bias in all the domains. |
Risk of bias for analysis 5.6 Clinical worsening: need for invasive mechanical ventilation or death at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. | Some concerns | There are some concerns, as the data was provided on request and the study results are not published yet. |
Risk of bias for analysis 6.1 All‐cause mortality at up to day 28 (random‐effects analysis).
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 6.1.1 Antibodies detected at baseline | ||||||||||||
| Avendano‐Sola 2021 | Low risk of bias | Participants were randomized through a web‐based eCRF system (ORACLE clinical) in a 1:1 ratio to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all the subgroup participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | For the outcome "mortality at up to day 28", there is a low risk of bias for all the domains. |
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data was awailable for all randomized participants for this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | High because of first domain |
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data was available for the randomized participants wihtin this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome, there was a low risk of bias for all the domains. |
| Ortigoza 2022 | Low risk of bias | A centralized electronic system was used to randomly assign enrolled patients to receive CCP or placebo in a 1:1 ratio stratified by enrollment site and risk status using randomization block sizes of 4 and 6 to maintain balanced group sizes. Allocation was concealed. | Low risk of bias | Patients, treating clinicians, trial personnel, and outcome assessors were blinded to group assignment. ITT was used. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 6.1.2 No antibodies detected at baseline | ||||||||||||
| Avendano‐Sola 2021 | Low risk of bias | Participants were randomized through a web‐based eCRF system (ORACLE clinical) in a 1:1 ratio to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all the subgroup participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | For the outcome "mortality at up to day 28", there is a low risk of bias for all the domains. |
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data was awailable for all randomized participants for this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | High because of first domain |
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data was available for the randomized participants wihtin this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome, there was a low risk of bias for all the domains. |
| Ortigoza 2022 | Low risk of bias | A centralized electronic system was used to randomly assign enrolled patients to receive CCP or placebo in a 1:1 ratio stratified by enrollment site and risk status using randomization block sizes of 4 and 6 to maintain balanced group sizes. Allocation was concealed. | Low risk of bias | Patients, treating clinicians, trial personnel, and outcome assessors were blinded to group assignment. ITT was used. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 6.3 Clinical worsening: need for invasive mechanical ventilation or death (random‐effects model).
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 6.3.1 Antibodies detected at baseline | ||||||||||||
| Agarwal 2020 | Low risk of bias | The allocation sequence was random and concealed and there were no differences between intervention groups suggesting a problem with the randomization process. | Some concerns | Five participants in convalescent plasma arm and four in control arm did not receive the allocated intervention. This could be due to awareness of the intervention, because this trial was open‐label. | Low risk of bias | Data for this outcome was available for participants within this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The data were collected in structured paper case record forms. | Low risk of bias | The outcome was predefined in the trial reistry. | Some concerns | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 6.3.2 No antibodies detected at baseline | ||||||||||||
| Agarwal 2020 | Low risk of bias | The allocation sequence was random and concealed and there were no differences between intervention groups suggesting a problem with the randomization process. | Some concerns | Five participants in convalescent plasma arm and four in control arm did not receive the allocated intervention. This could be due to awareness of the intervention, because this trial was open‐label. | Low risk of bias | Data for this outcome was available for participants within this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The data were collected in structured paper case record forms. | Low risk of bias | The outcome was predefined in the trial reistry. | Some concerns | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 6.4 Clinical improvement: participants discharged from hospital.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 6.4.1 Antibodies detected at baseline | ||||||||||||
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized in this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 6.4.2 No antibodies detected at baseline | ||||||||||||
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized in this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 7.1 All‐cause mortality at up to day 28 (random‐effects model).
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 7.1.1 Duration of symptom onset up to and including 7 days | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for the randomized participants within this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For this outcome, there was a low risk of bias for all the domains. |
| Korley 2021 | High risk of bias | Baseline differences (characteristics are not balanced) in hospitalisation between groups suggesting a problem with randomization. Participants in the control group were included when they were still hospitalised and not dischared yet, in contrary to the CP group. | Low risk of bias | The participants were blinded to the treatment, but those delivering the intervention were aware of intervention received. There were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received, so it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | High risk of bias | High risk of bias arising from the randomization process. |
| Subgroup 7.1.2 Duration of symptom onset more than 7 days | ||||||||||||
| De Santis 2022 | Some concerns | The study methods it is stated that this is a "investigator‐initiated multicenter open‐label randomized controlled trial", but no further information on the allocation sequence concealment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Some concerns | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for the randomized participants within this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For this outcome, there was a low risk of bias for all the domains. |
Risk of bias for analysis 7.2 All‐cause mortality (time to event) (random‐effects model).
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 7.2.1 Duration of symptom onset more than 7 days | ||||||||||||
| De Santis 2022 | Some concerns | The study methods it is stated that this is a "investigator‐initiated multicenter open‐label randomized controlled trial", but no further information on the allocation sequence concealment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | ‐ |
Risk of bias for analysis 7.3 Clinical worsening: need for invasive mechanical ventilation or death (random‐effects model).
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 7.3.1 Duration of symptom onset up to and including 7 days | ||||||||||||
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data was available for randomized participants within this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome, there was a low risk of bias for all the domains. |
| Korley 2021 | High risk of bias | Baseline differences (characteristics are not balanced) in hospitalisation between groups suggesting a problem with randomization. Participants in the control group were included when they were still hospitalised and not dischared yet, in contrary to the CP group. | Low risk of bias | The participants were blinded to the treatment, but those delivering the intervention were aware of intervention received. There were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received, so it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | High risk of bias | ‐ |
| Subgroup 7.3.2 Duration of symptom onset more than 7 days | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | articipants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for all available | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data was available for randomized participants within this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome, there was a low risk of bias for all the domains. |
Risk of bias for analysis 7.4 Clinical improvement: participants discharged from hospital.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 7.4.1 Duration of symptom onset up to 7 days | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 7.4.2 Duration of symptom onset more than 7 days | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 8.1 All‐cause mortality at up to day 28 (random‐effects analysis).
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 8.1.1 Duration of symptom onset up to and including 7 days | ||||||||||||
| O’Donnell 2021 | Low risk of bias | Participants were block randomized using a web‐based randomization platform in a 2:1 ratio to receive either convalescent plasma or standard plasma and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 217 participants randomized, within this subrgoup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | For the outcome "28‐day mortality", there is a low risk of bias for all the domains. |
| Subgroup 8.1.2 Duration of symptom onset more than 7 days | ||||||||||||
| O’Donnell 2021 | Low risk of bias | Participants were block randomized using a web‐based randomization platform in a 2:1 ratio to receive either convalescent plasma or standard plasma and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 217 participants randomized, within this subrgoup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | For the outcome "28‐day mortality", there is a low risk of bias for all the domains. |
Risk of bias for analysis 9.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 9.1.1 Duration of symptom onset up to and including 7 days | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
| Subgroup 9.1.2 Duration of symptom onset more than 7 days | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
Risk of bias for analysis 9.2 Admission to hospital or death within 28 days.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 9.2.1 Duration of symptom onset up to and including 7 days | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on the measurement of the outcome. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The data was provided on request by the study investigators. | Some concerns | ‐ |
| Subgroup 9.2.2 Duration of symptom onset more than 7 days | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on the measurement of the outcome. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The data was provided on request by the study investigators. | Some concerns | ‐ |
Risk of bias for analysis 9.3 All initial symptoms resolved (asymptomatic) at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 9.3.1 Duration of symptom onset up to and including 7 days | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
| Subgroup 9.3.2 Duration of symptom onset more than 7 days | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
Risk of bias for analysis 10.1 All‐cause mortality at up to day 28 (random‐effects model).
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 10.1.1 Immunosuppression (immune deficiency and cancer) | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Subgroup 10.1.2 No immunosuppression (immune deficiency and cancer) | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
Risk of bias for analysis 10.2 Clinical improvement: participants discharged from hospital.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 10.2.1 Immunosuppression (immune deficiency and cancer) | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Subgroup 10.2.2 No immunosuppression (immune deficiency and cancer) | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
Risk of bias for analysis 11.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 11.1.1 No immunosuppression (immune deficiency and cancer) | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
Risk of bias for analysis 11.2 Admission to hospital or death within 28 days.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 11.2.1 No immunosuppression (immune deficiency and cancer) | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on the measurement of the outcome. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The data was provided on request by the study investigators. | Some concerns | ‐ |
Risk of bias for analysis 11.3 All initial symptoms resolved (asymptomatic) at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 11.3.1 No immunosuppression (immune deficiency and cancer) | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
Risk of bias for analysis 12.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 12.1.1 Diabetes | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Subgroup 12.1.2 No diabetes | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
Risk of bias for analysis 12.2 Clinical worsening: need for invasive mechanical ventilation or death.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 12.2.1 Diabetes | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so the interim analysis was no deviation from the protocol and the analysis was appropriate. | Low risk of bias | Data were available for nearly all (921/938) participants. 17 Participants were lost to follow‐up at day 30 (after discharge). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 12.2.2 No diabetes | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so the interim analysis was no deviation from the protocol and the analysis was appropriate. | Low risk of bias | Data were available for nearly all (921/938) participants. 17 Participants were lost to follow‐up at day 30 (after discharge). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 12.3 Clinical improvement: participants discharged from hospital.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 12.3.1 Diabetes | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Subgroup 12.3.2 No diabetes | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
Risk of bias for analysis 13.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 13.1.1 Diabetes | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
| Subgroup 13.1.2 No diabetes | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
Risk of bias for analysis 13.2 Admission to hospital or death within 28 days.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 13.2.1 Diabetes | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on the measurement of the outcome. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The data was provided on request by the study investigators. | Some concerns | ‐ |
| Sullivan 2022 | Low risk of bias | Participants from all sites were randomized by a central web‐based system using blocked assignments to receive ABO compatible control plasma or CCP (both∼250 mL) in a 1:1 ratio. | Low risk of bias | Participants were blinded, but and those delivering the intervention were aware of intervention received. However, there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | A modified intention‐to‐treat was used, so data for this outcome was available for nearly all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. COVID‐19 relatedness for hospitalizations was adjudicated by a three physician panel masked to treatment group. | Low risk of bias | No protocol available, but the data that produced this result was analysed in accordance with the predefined outcome in the trial registry. | Low risk of bias | ‐ |
| Subgroup 13.2.2 No diabetes | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on the measurement of the outcome. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The data was provided on request by the study investigators. | Some concerns | ‐ |
Risk of bias for analysis 13.3 All initial symptoms resolved (asymptomatic) at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 13.3.1 Diabetes | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
| Subgroup 13.3.2 No diabetes | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
Risk of bias for analysis 14.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 14.1.1 Pulmonary disease | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 14.1.2 No pulmonary disease | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 14.2 Clinical worsening: need for invasive mechanical ventilation or death.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 14.2.1 Respiratory disease | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so the interim analysis was no deviation from the protocol and the analysis was appropriate. | Low risk of bias | Data were available for nearly all (921/938) participants. 17 Participants were lost to follow‐up at day 30 (after discharge). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 14.2.2 No respiratory disease | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so the interim analysis was no deviation from the protocol and the analysis was appropriate. | Low risk of bias | Data were available for nearly all (921/938) participants. 17 Participants were lost to follow‐up at day 30 (after discharge). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 14.3 Clinical improvement: participants discharged from hospital.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 14.3.1 Pulmonary disease | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Subgroup 14.3.2 No pulmonary disease | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
Risk of bias for analysis 15.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 15.1.1 Hypertension | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Subgroup 15.1.2 No hypertension | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
Risk of bias for analysis 15.2 Clinical improvement: participants discharged from hospital.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 15.2.1 Hypertension | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Subgroup 15.2.2 No hypertension | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
Risk of bias for analysis 16.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 16.1.1 < 65 years | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Subgroup 16.1.2 ≥ 65 years | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Subgroup 16.1.3 ≤ 60 years | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Subgroup 16.1.4 > 60 years | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
Risk of bias for analysis 16.2 Clinical worsening: need for invasive mechanical ventilation, or death.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 16.2.1 < 60 years | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so the interim analysis was no deviation from the protocol and the analysis was appropriate. | Low risk of bias | Data were available for nearly all (921/938) participants. 17 Participants were lost to follow‐up at day 30 (after discharge). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 16.2.2 ≥ 60 years | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so the interim analysis was no deviation from the protocol and the analysis was appropriate. | Low risk of bias | Data were available for nearly all (921/938) participants. 17 Participants were lost to follow‐up at day 30 (after discharge). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 16.2.3 < 70 years | ||||||||||||
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 16.2.4 ≥ 70 years | ||||||||||||
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 16.3 Clinical improvement: participants discharged from hospital.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 16.3.1 < 65 years | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Subgroup 16.3.2 ≥ 65 years | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Subgroup 16.3.3 < 70 years | ||||||||||||
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 16.3.4 > 70 years | ||||||||||||
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 17.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 17.1.1 < 65 years | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
| Subgroup 17.1.2 ≥ 65 years | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
Risk of bias for analysis 17.2 Admission to hospital or death within 28 days.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 17.2.1 < 65 years | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on the measurement of the outcome. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The data was provided on request by the study investigators. | Some concerns | ‐ |
| Sullivan 2022 | Low risk of bias | Participants from all sites were randomized by a central web‐based system using blocked assignments to receive ABO compatible control plasma or CCP (both∼250 mL) in a 1:1 ratio. | Low risk of bias | Participants were blinded, but and those delivering the intervention were aware of intervention received. However, there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | A modified intention‐to‐treat was used, so data for this outcome was available for nearly all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. COVID‐19 relatedness for hospitalizations was adjudicated by a three physician panel masked to treatment group. | Low risk of bias | No protocol available, but the data that produced this result was analysed in accordance with the predefined outcome in the trial registry. | Low risk of bias | ‐ |
| Subgroup 17.2.2 ≥ 65 years | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on the measurement of the outcome. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The data was provided on request by the study investigators. | Some concerns | ‐ |
| Sullivan 2022 | Low risk of bias | Participants from all sites were randomized by a central web‐based system using blocked assignments to receive ABO compatible control plasma or CCP (both∼250 mL) in a 1:1 ratio. | Low risk of bias | Participants were blinded, but and those delivering the intervention were aware of intervention received. However, there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | A modified intention‐to‐treat was used, so data for this outcome was available for nearly all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. COVID‐19 relatedness for hospitalizations was adjudicated by a three physician panel masked to treatment group. | Low risk of bias | No protocol available, but the data that produced this result was analysed in accordance with the predefined outcome in the trial registry. | Low risk of bias | ‐ |
Risk of bias for analysis 17.3 All initial symptoms resolved (asymptomatic) at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 17.3.1 < 65 years | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
| Subgroup 17.3.2 ≥ 65 years | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
Risk of bias for analysis 18.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 18.1.1 Female | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Subgroup 18.1.2 Male | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Kirenga 2021 | Low risk of bias | Randomisation was performed using permuted block randomisation with varying block sizes ranging from 4 to 8. Generation of randomisation numbers was done by the study biostatistician, who did not have contact with participants. The biostatistician printed individual randomisation numbers with their corresponding treatment assignment and placed each number in an opaque envelope. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | ‐ |
Risk of bias for analysis 18.2 Clinical worsening: need for invasive mechanical ventilation or death.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 18.2.1 Female | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data available for this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in teh SAP. | Low risk of bias | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Menichetti 2021 | Low risk of bias | A stratified permuted block randomization procedure with a 1:1 ratio, with blocks of variable sizes and stratification for clinical sites, was generated by using Stata version 16.1 (StataCorp). Eligible patients underwent web‐based treatment allocation through REDcap platform to receive either administration of COVID‐19 CP in addition to standard therapy (ST) or ST alone. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for this subgroup, in a modified intention‐to‐treat analysis. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | Low risk of bias | ‐ |
| Subgroup 18.2.2 Male | ||||||||||||
| Begin 2021 | Low risk of bias | Patients were randomised in a 2:1 ratio to receive convalescent plasma or standard of care using a secure, concealed, computer‐generated, web‐accessed randomization sequence (REDCap v11.0.1). Eligible patients were randomly assigned to receive either convalescent plasma or standard of care. | Low risk of bias | Participants and those delivering the intervention were aware of intervention received. The trial was terminated at 78% of planned enrollment after meeting stopping criteria for futility, according to protocol, so there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data available for this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in teh SAP. | Low risk of bias | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for this subgroup. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Menichetti 2021 | Low risk of bias | A stratified permuted block randomization procedure with a 1:1 ratio, with blocks of variable sizes and stratification for clinical sites, was generated by using Stata version 16.1 (StataCorp). Eligible patients underwent web‐based treatment allocation through REDcap platform to receive either administration of COVID‐19 CP in addition to standard therapy (ST) or ST alone. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for this subgroup, in a modified intention‐to‐treat analysis. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 18.3 Clinical improvement: participants discharged from hospital.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 18.3.1 Female | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 18.3.2 Male | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 19.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 19.1.1 Female | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
| Subgroup 19.1.2 Male | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
Risk of bias for analysis 19.2 Admission to hospital or death within 28 days.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 19.2.1 Female | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on the measurement of the outcome. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The data was provided on request by the study investigators. | Some concerns | ‐ |
| Sullivan 2022 | Low risk of bias | Participants from all sites were randomized by a central web‐based system using blocked assignments to receive ABO compatible control plasma or CCP (both∼250 mL) in a 1:1 ratio. | Low risk of bias | Participants were blinded, but and those delivering the intervention were aware of intervention received. However, there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | A modified intention‐to‐treat was used, so data for this outcome was available for nearly all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. COVID‐19 relatedness for hospitalizations was adjudicated by a three physician panel masked to treatment group. | Low risk of bias | No protocol available, but the data that produced this result was analysed in accordance with the predefined outcome in the trial registry. | Low risk of bias | ‐ |
| Subgroup 19.2.2 Male | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Some concerns | The trial registry does not provide enough information on the measurement of the outcome. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The data was provided on request by the study investigators. | Some concerns | ‐ |
| Sullivan 2022 | Low risk of bias | Participants from all sites were randomized by a central web‐based system using blocked assignments to receive ABO compatible control plasma or CCP (both∼250 mL) in a 1:1 ratio. | Low risk of bias | Participants were blinded, but and those delivering the intervention were aware of intervention received. However, there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | A modified intention‐to‐treat was used, so data for this outcome was available for nearly all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. COVID‐19 relatedness for hospitalizations was adjudicated by a three physician panel masked to treatment group. | Low risk of bias | No protocol available, but the data that produced this result was analysed in accordance with the predefined outcome in the trial registry. | Low risk of bias | ‐ |
Risk of bias for analysis 19.3 All initial symptoms resolved (asymptomatic) at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 19.3.1 Female | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
| Subgroup 19.3.2 Male | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. The subgroup data was provided on request by the study investigator. | Some concerns | ‐ |
Risk of bias for analysis 20.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 20.1.1 High‐income countries | ||||||||||||
| Menichetti 2021 | Low risk of bias | A stratified permuted block randomization procedure with a 1:1 ratio, with blocks of variable sizes and stratification for clinical sites, was generated by using Stata version 16.1 (StataCorp). Eligible patients underwent web‐based treatment allocation through REDcap platform to receive either administration of COVID‐19 CP in addition to standard therapy (ST) or ST alone. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized, in a modified intention‐to‐treat analysis. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | Low risk of bias | ‐ |
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| AlQahtani 2021 | Low risk of bias | Participants were block randomized by computer‐generated random numbering in a 1:1 ratio to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 40 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Some concerns | The data that produced this result was not analysed in accordance with the predefined outcomes stated in the protocol, as the time point of the outcome measurement was not pre‐specified in the study protocol. | Some concerns | For the outcome "mortality", there are some concerns for bias in selection of reported results. However, for all the other domains, there is a low risk of bias. |
| Avendano‐Sola 2021 | Low risk of bias | Participants were randomized through a web‐based eCRF system (ORACLE clinical) in a 1:1 ratio to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 359 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | For the outcome "mortality at up to day 28", there is a low risk of bias for all the domains. |
| Holm 2021 | Low risk of bias | Patients eligible for inclusion were enrolled and randomized by a study physician (KH, OL, CW, JO, MR), 1:1 using the electronic software REDCap to either receive SOC or SOC with 200–250 mL of CCP administered intravenously during 30 min on three consecutive days. Blocks of ten patients were used for randomization. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Some concerns | No protocol available online (could be requested), but clinical trial registry is available. All‐cause mortality within 28 days: this was changed from the three month in the ClinicalTrial registration due to a maximum follow‐up time of 28 days | Some concerns | ‐ |
| Koerper 2021 | Low risk of bias | Patients were randomized using a Web‐based system with a stratified 1:1 allocation ratio between each stratum (Figure 1). Patients were stratified prior to permutated block randomization by presence or absence of ventilation support, ECMO, or ICU treatment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "28‐day mortality", there was a low risk of bias for all the domains. |
| Devos 2021 | Low risk of bias | Patients were randomised through a computerised system (RedCap®, Vanderbilt University, USA, Version 10.6.13) according to a 2:1 allocation scheme stratified by study site using randomly selected block sizes of 6 or 9, to open‐label convalescent plasma combined with standard of care (intervention group) or standard of care alone (control group). | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Ortigoza 2022 | Low risk of bias | A centralized electronic system was used to randomly assign enrolled patients to receive CCP or placebo in a 1:1 ratio stratified by enrollment site and risk status using randomization block sizes of 4 and 6 to maintain balanced group sizes. Allocation was concealed. | Low risk of bias | Patients, treating clinicians, trial personnel, and outcome assessors were blinded to group assignment and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that the assessment of the outcome was influenced by knowledge of intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Korley 2021 | High risk of bias | Baseline differences (characteristics are not balanced) in hospitalisation between groups suggesting a problem with randomization. Participants in the control group were included when they were still hospitalised and not dischared yet, in contrary to the CP group. | Low risk of bias | The participants were blinded to the treatment, but those delivering the intervention were aware of intervention received. There were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received, so it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | High risk of bias | High risk of bias arising from the randomization process. |
| Subgroup 20.1.2 Low‐ to middle‐income countries | ||||||||||||
| Van den Berg 2022 | Low risk of bias | Eligible and consenting participants were randomized (1:1) to receive either a single infusion of 200–250 mL CCP or 200 mL of placebo, together with local standard of care. Random assignment was stratified by study site, age (≥ or < 65 years), and body mass index (BMI) (≥ or < 30 kg/m2). An electronic randomisation application (REDCap) hosted by SANB, was used to generate the treatment allocation. To mask treatment allocation, investigational product (IP) was covered in opaque paper wrapping prior to dispatch from the blood bank. | Low risk of bias | Both participants and those delivering the intervention were probably not aware of intervention received, as treatment allocation was masked (plasma was covered in opaque paper wrapping) and there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized (two participants withdrew consent and two were transferred to alternative facilities prior to transfusion) . | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registry, the protocol was not available. | Low risk of bias | ‐ |
| Sekine 2021 | Low risk of bias | Patients were randomly assigned in a 1:1 ratio to receive two infusions 48 h apart of 300mL of CP plus Standard of Care (SOC) or SOC alone. Randomisation was performed using computer‐generated randomisation with random block sizes of 2 or 4 and stratified according to the unit of hospitalisation on enrolment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized data was available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "mortality", all the domains have a low risk of bias. |
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Some concerns | ‐ |
| Li 2020 | Low risk of bias | Participants were block randomized via computer‐generated random numbering in a 1:1 ratio to receive standard treatment coupled with convalescent plasma transfusion or standard treatment alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized, a total of 101 out of 103 participants were included in the analysis. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For the outcome "mortality at up to day 28", all the domains have a low risk of bias. |
| Kirenga 2021 | Low risk of bias | Randomisation was performed using permuted block randomisation with varying block sizes ranging from 4 to 8. Generation of randomisation numbers was done by the study biostatistician, who did not have contact with participants. The biostatistician printed individual randomisation numbers with their corresponding treatment assignment and placed each number in an opaque envelope. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | ‐ |
| Agarwal 2020 | Low risk of bias | The allocation sequence was random and concealed and there were no differences between intervention groups suggesting a problem with the randomization process. | Some concerns | Five participants in convalescent plasma arm and four in control arm did not receive the allocated intervention and this is probably a deviation from the intended intervention that arose because of the trial context. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for 453 out of 464 participants. Data of six participants in the convalescent plasma arm and of five participants in the control arm were missing. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The data were collected in structured paper case record forms. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | For the outcome "mortality" in this study, there is a low risk of bias from the randomization process, missing outcome data, measurement of the outcome and in selection of the reported result. However, there are some concerns for bias due to deviations from intended interventions. |
| Ray 2022 | Some concerns | Participants were probably allocated randomly to either the standard of care group alone or standard of care with convalescent plasma group and there were no further information given on the randomization process. There is no information on the allocation concealment, the trial registry only indicates concealment through "Case Record Numbers". There were no baseline imbalances that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 80 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | For the outcome "mortality at up to day 28" in this study, there is a low risk of bias due to deviations from intended interventions, due to missing outcome data, in measurement of the outcome and in selection of the reported result. There are some concerns for bias from the randomization process. |
| De Santis 2022 | Some concerns | The study methods it is stated that this is a "investigator‐initiated multicenter open‐label randomized controlled trial", but no further information on the allocation sequence concealment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Some concerns | ‐ |
Risk of bias for analysis 20.2 Clinical worsening: need for invasive mechanical ventilation, or death at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 20.2.1 High‐income countries | ||||||||||||
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Korley 2021 | High risk of bias | Baseline differences (characteristics are not balanced) in hospitalisation between groups suggesting a problem with randomization. Participants in the control group were included when they were still hospitalised and not dischared yet, in contrary to the CP group. | Low risk of bias | The participants were blinded to the treatment, but those delivering the intervention were aware of intervention received. There were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received, so it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | High risk of bias | ‐ |
| Subgroup 20.2.2 Low‐ to middle‐income countries | ||||||||||||
| Agarwal 2020 | Low risk of bias | The allocation sequence was random and concealed and there were no differences between intervention groups suggesting a problem with the randomization process. | Some concerns | Five participants in convalescent plasma arm and four in control arm did not receive the allocated intervention. This could be due to awareness of the intervention, because this trial was open‐label. | Low risk of bias | Data for this outcome was available for 451 out of 464 participants. Data of eight participants in the convalescent plasma arm and of five participants in the control arm were missing. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The data were collected in structured paper case record forms. | Low risk of bias | There was no information on whether the data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration, as the data was provided on request by the study investigators. | Some concerns | For this outcome "mortality", there is a low risk of bias from the randomization process, missing outcome data, measurement of the outcome and in selection of the reported result. However, there are some concerns for bias due to deviations from intended interventions. |
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized data was available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | The outcome measure was appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 20.3 Clinical improvement: participants discharged from hospital.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 20.3.1 High‐income countries | ||||||||||||
| Devos 2021 | Low risk of bias | Patients were randomised through a computerised system (RedCap®, Vanderbilt University, USA, Version 10.6.13) according to a 2:1 allocation scheme stratified by study site using randomly selected block sizes of 6 or 9, to open‐label convalescent plasma combined with standard of care (intervention group) or standard of care alone (control group). | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was in accordance with the predefined outcomes stated in the trial protocol: "Duration of hospital and ICU stay: both parameters will be analysed as time‐to‐event parameters with competing risk, whereby the event of interest is discharge from hospital/ICU and the competing risk is hospital/ICU death". | Low risk of bias | ‐ |
| Horby 2021b | Low risk of bias | Participants were randomized through web‐based simple randomization with allocation concealment in a 1:1:1 ratio to a platform trial in a factorial design, receiving either convalescent plasma in addition to the standard therapy or standard of care alone. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 11,558 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Subgroup 20.3.2 Low‐ to middle‐income countries | ||||||||||||
| Gharbharan 2021 | Low risk of bias | Participants were randomized via a web‐based system in a 1:1 ratio to the current standard of care at each hospital with or without the addition of convalescent plasma. There are no baseline differences that would suggest a problem with randomisation. | Some concerns | Both participants and those delivering the intervention were aware of intervention received, but there was no information on deviations from intended intervention, as insufficient information was provided on whether treatments were balanced across arms. The analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 86 participants randomized. | Low risk of bias | The outcome assessors aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement or that the measurement differed between intervention groups. | Some concerns | The data that produced this result was not a pre‐specified outcome from the registry. | Some concerns | ‐ |
| Li 2020 | Low risk of bias | Participants were block randomized via computer‐generated random numbering in a 1:1 ratio to receive standard treatment coupled with convalescent plasma transfusion or standard treatment alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized, a total of 101 out of 103 participants were included in the analysis. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
| Sekine 2021 | Low risk of bias | Patients were randomly assigned in a 1:1 ratio to receive two infusions 48 h apart of 300mL of CP plus Standard of Care (SOC) or SOC alone. Randomisation was performed using computer‐generated randomisation with random block sizes of 2 or 4 and stratified according to the unit of hospitalisation on enrolment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Low risk of bias | ‐ |
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized data was available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | The outcome was appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 20.4 Grades 3 and 4 adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 20.4.1 High‐income countries | ||||||||||||
| AlQahtani 2021 | Low risk of bias | Participants were block randomized by computer‐generated random numbering in a 1:1 ratio to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate.. | Low risk of bias | Data for this outcome was available for all 40 participants randomized. | Some concerns | There is no information available on whether the method of measuring the outcome was appropriate or whether the measurement could have differed between groups, as the data was provided on request by the study investigator and for safety different measurements are possible. The outcome assessors were aware of the intervention received, which is why the knowledge of intervention received could have affected safety outcome assessments, but this is unlikely. | Low risk of bias | For this outcome, there was no information on whether the data that produced this result was analysed in accordance with the predefined protocol, as the data was provided on request by the study investigators. | Some concerns | For the outcome "Grade 3 and 4 adverse events", there is a low risk of bias arising from the randomization process, due to deviations from intended interventions, due to missing outcome data and in the in selection of the reported result. But, there are some concerns for bias in measurement of the outcome. |
| Menichetti 2021 | Low risk of bias | A stratified permuted block randomization procedure with a 1:1 ratio, with blocks of variable sizes and stratification for clinical sites, was generated by using Stata version 16.1 (StataCorp). Eligible patients underwent web‐based treatment allocation through REDcap platform to receive either administration of COVID‐19 CP in addition to standard therapy (ST) or ST alone. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | The data for this outcome was available for all participants randomized. | Some concerns | There is not enough information whether the measurement of the outcome was appropriate. This is an open label trial and the outcome assessors were aware of the intervention allocation, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | ‐ |
| Subgroup 20.4.2 Low‐ to middle‐income countries | ||||||||||||
| Agarwal 2020 | Low risk of bias | The allocation sequence was random and concealed and there were no differences between intervention groups suggesting a problem with the randomization process. | Some concerns | Five participants in convalescent plasma arm and four in control arm did not receive the allocated intervention and this is probably a deviation from the intended intervention that arose because of the trial context. There is no information on whether the analysis was appropriate. | Low risk of bias | Data for this outcome was available for 451 participants out of 464 participants randomized and the data of eight participants in the convalescent plasma arm and of five participants in the control arm were missing. | Some concerns | There was no information on the method of measuring for this outcome, as the data was provided on request by the study investigator. The outcome assessors were aware of the intervention received and the knowledge of intervention received could have affected outcome measurement, but it is probably not likely. | Low risk of bias | There was no information on whether the data that produced this result was analysed in accordance with a pre‐specified trial registration, as the data was provided on request by the study investigators. | Some concerns | For the outcome "Grade 3 and 4 adverse events", there is a low risk of bias arising from the randomization process and due to missing outcome data and in the in selection of the reported result. But, there are some concerns for bias due to deviations from intended interventions and in measurement of the outcome. |
| Sekine 2021 | Low risk of bias | Patients were randomly assigned in a 1:1 ratio to receive two infusions 48 h apart of 300mL of CP plus Standard of Care (SOC) or SOC alone. Randomisation was performed using computer‐generated randomisation with random block sizes of 2 or 4 and stratified according to the unit of hospitalisation on enrolment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized | Some concerns | There is no information on whether the outcome measure was appropriate, but it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | ‐ |
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized, data was available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | The outcome measurement is an appropriate measure and it is unlikely that the measurement differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For this outcome "Grade 3 and 4 adverse events", there is a low risk of bias for all the domains. |
Risk of bias for analysis 20.5 Serious adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 20.5.1 High‐income countries | ||||||||||||
| Bar 2021 | Some concerns | Participants were assigned to treatment or control in a 1:1 ratio using randomization stratified on the use of remdesivir and mechanical ventilation at entry using block randomization with variable block size. There is no information on whether the allocation sequence was concealed. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Some concerns | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the statistical analysis plan (SAP). | Some concerns | ‐ |
| Devos 2021 | Low risk of bias | Patients were randomised through a computerised system (RedCap®, Vanderbilt University, USA, Version 10.6.13) according to a 2:1 allocation scheme stratified by study site using randomly selected block sizes of 6 or 9, to open‐label convalescent plasma combined with standard of care (intervention group) or standard of care alone (control group). | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Some concerns | The measurement of the outcome was appropriate, the serious adverse events are listed in the supplement and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | Due to bias of measurement outcome |
| Estcourt 2021 | Low risk of bias | Participants were randomized via a centralized computer program to each intervention (available locally) starting with balanced assignment. The study will randomly allocate participants to one or more interventions, with each intervention nested within a domain. In this regard, a platform trial is no different to other forms of RCT in that randomization provides the basis for causal inference. However, unlike a conventional RCT, the proportion of participants who are randomized to each available intervention within a domain will not be fixed. Rather, the trial will incorporate RAR. RAR utilizes random allocation with a weighted probability for each intervention, with the weighted probability being proportional to the extent to which similar participants recruited earlier in the trial benefited or not from each particular intervention. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Some concerns | The measurement of the outcome was appropriate, the serious adverse events are listed in the supplement and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial protocol. | Some concerns | Due to bias of measurement outcome |
| Koerper 2021 | Low risk of bias | Patients were randomized using a Web‐based system with a stratified 1:1 allocation ratio between each stratum (Figure 1). Patients were stratified prior to permutated block randomization by presence or absence of ventilation support, ECMO, or ICU treatment. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for (nearly) all participants randomized | Some concerns | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, so the assessment of the outcome could have been influenced by knowledge of intervention received, but it is unlikely. | Low risk of bias | The outcome is predefined in the trial registry. | Some concerns | ‐ |
| Subgroup 20.5.2 Low‐ to middle‐income countries | ||||||||||||
| Simonovich 2020 | Low risk of bias | Participants were randomised through a randomization program (REDCap) in a 2:1 ratio to receive either convalescent plasma or a placebo. The allocation sequence was random and concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received and the analysis was appropriate. | Low risk of bias | From the 334 participants randomized, data was available for 228 participants in the convalescent plasma group and 105 participants in the placebo group. The data of one participant in control group was missing, due to withdrawal of consent after randomisation and therefore the participant was excluded from the analysis. | Low risk of bias | The outcome measurement is an appropriate measure and it is unlikely that the measurement differed between intervention groups. The outcome assessors were not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For this outcome "Serious adverse events", there is a low risk of bias for all the domains. |
Risk of bias for analysis 21.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 21.1.1 High‐income countries | ||||||||||||
| Bennett‐Guerrero 2021 | Low risk of bias | Patients were randomized 4:1 to CP or SP using permuted block randomization lists generated using SAS software (version 9.4; SAS Institute, Cary NC) and implemented using an interactive web response randomization tool in Research Electronic Data Capture (REDCap). | Low risk of bias | The individuals who were unblinded were Blood Bank personnel since they needed to label and dispense the masked CP or SP. The bags of CP or SP had an identical label, stating “CP or SP” to preserve blinding. All study personnel who collected data were blinded to study assignment at all times. | Low risk of bias | Data for this outcome was available for all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Low risk of bias | No protocol available, but in Supplemental Detailed Methods mentioned | Low risk of bias | ‐ |
| Subgroup 21.1.2 Low‐ to middle‐income countries | ||||||||||||
| Bajpai 2020 | Low risk of bias | Participants were block randomized with concealed allocation using the Sequentially Numbered Opaque Sealed Envelopes (SNOSE) method, to receive either convalescent plasma with standard of care or standard plasma (fresh frozen plasma) with standard of care. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | Both participants and those delivering the intervention were aware of intervention received, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for nearly all participants randomized, in total 29 out of 31 participants (2 participants became PCR negative on the day of plasma transfusion). | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of intervention received could have affected outcome measurement. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the trial registration. | Low risk of bias | For this outcome "mortality at day 28", there is a low risk of bias for all the domains. |
| Baldeon 2022 | Some concerns | This was a random double blinded, non‐convalescent plasma (NCP)‐placebo controlled clinical trial. However, there are no information provided on the concealment of the allocation sequence:"A simple randomization scheme was used to allocate participants in the two treatment groups." | Low risk of bias | This study was triple blinded. Hospital personnel at the transfusion medicine services, physicians, and the participants were unaware which plasma was convalescent and which plasma was control. Intention to treat analysis was used. | Low risk of bias | Data was available for all the 158 participants randomised. | Low risk of bias | Clinical outcomes were assessed by investigators who were blinded to the treatment groups. | Low risk of bias | No protocol was available, but the outcome was pre‐specified in the trial registry. | Some concerns | Some concerns due to bias arising from the randomization process. |
Risk of bias for analysis 21.2 Clinical worsening: need for invasive mechanical ventilation or death at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 21.2.1 High‐income countries | ||||||||||||
| NCT04421404 | Low risk of bias | Subjects enrolled in the study will be randomized using a web based randomization procedure to receive convalescent plasma versus non‐immune plasma at a 1:1 ratio. | Low risk of bias | Both participants and those delivering the intervention were not aware of intervention received, there were no deviations from intended interventions and the analysis was appropriate. The nursing staff may be unblinded to the treatment assignment to ensure proper ABO checking of the plasma unit at bedside per standard transfusion procedures. | Low risk of bias | Data for this outcome was available for all 34 participants randomized. | Low risk of bias | The are no information available (as the study still needs to be published in a journal) on whether the measurement of the outcome was appropriate, but it is unlikely that it differed between intervention groups. The outcome assessors were blinded to the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 21.3 Serious adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 21.3.1 High‐income countries | ||||||||||||
| Bennett‐Guerrero 2021 | Low risk of bias | Patients were randomized 4:1 to CP or SP using permuted block randomization lists generated using SAS software (version 9.4; SAS Institute, Cary NC) and implemented using an interactive web response randomization tool in Research Electronic Data Capture (REDCap). | Low risk of bias | The individuals who were unblinded were Blood Bank personnel since they needed to label and dispense the masked CP or SP. The bags of CP or SP had an identical label, stating “CP or SP” to preserve blinding. All study personnel who collected data were blinded to study assignment at all times. | Low risk of bias | Data for this outcome was available for all participants randomized. | Some concerns | There is not enough information available and it is unlikely that it differed between intervention groups. The outcome assessors were unaware of the intervention received. | Some concerns | The outcome is not pre‐defined in the registry of supplemental material. | Some concerns | ‐ |
| NCT04421404 | Low risk of bias | Subjects enrolled in the study will be randomized using a web based randomization procedure to receive convalescent plasma versus non‐immune plasma at a 1:1 ratio. | Low risk of bias | Both participants and those delivering the intervention were not aware of intervention received, there were no deviations from intended interventions and the analysis was appropriate. The nursing staff may be unblinded to the treatment assignment to ensure proper ABO checking of the plasma unit at bedside per standard transfusion procedures. | Low risk of bias | Data for this outcome was available for all 34 participants randomized. | Low risk of bias | The are no information available (as the study still needs to be published in a journal) on whether the measurement of the outcome was appropriate, but it is unlikely that it differed between intervention groups. The outcome assessors were blinded to the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the predefined outcomes stated in the protocol. | Low risk of bias | ‐ |
Risk of bias for analysis 22.1 All‐cause mortality at up tp day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 22.1.1 High‐income countries | ||||||||||||
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome available for all participants randomized. | Low risk of bias | This outcome was provided on request by the study investgators. | Low risk of bias | Data that produced this result analysed in accordance with a pre‐specified outcome of the protocol. | Low risk of bias | ‐ |
| Subgroup 22.1.2 Low‐ to middle‐income countries | ||||||||||||
| Libster 2020 | Low risk of bias | Participants were randomized using an electronic system numbering to receive either convalescent plasma in addition to the standard therapy or standard of care alone and the allocation sequence was concealed. There are no baseline differences that would suggest a problem with randomisation. | Low risk of bias | The participants were unaware of their assigned intervention and those delivering the intervention were aware of intervention assigned during the trial, but there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | Data for this outcome was available for all 160 participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were probably unaware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified analysis plan and the outcome was reported as planned in the protocol. | Low risk of bias | For this outcome "all‐cause mortality", there is a low risk of bias for all the domains. |
Risk of bias for analysis 22.2 Grades 3 and 4 adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 22.2.1 High‐income countries | ||||||||||||
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome available for all participants randomized. | Some concerns | There is not enough information to judge whether the outcome was measured appropriately. | Low risk of bias | Data that produced this result analysed in accordance with a pre‐specified outcome of the protocol. | Some concerns | ‐ |
Risk of bias for analysis 22.3 Serious adverse events.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 22.3.1 High‐income countries | ||||||||||||
| Alemany 2022 | Low risk of bias | This is a prospective, randomized (1:1), double blind study of Convalescent anti‐SARS‐CoV‐2 MBT Plasma plus standard medical treatment (SMT) versus placebo plus SMT in mild or moderate COVID‐19 patients who are non‐hospitalised. However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. | Low risk of bias | Data for this outcome available for all participants randomized. | Some concerns | This outcome was provided on request by the study investgators. | Low risk of bias | The data that produced this result was provided on request. | Some concerns | ‐ |
Risk of bias for analysis 23.1 All‐cause mortality at up to day 28.
| Study | Bias | |||||||||||
| Randomisation process | Deviations from intended interventions | Missing outcome data | Measurement of the outcome | Selection of the reported results | Overall | |||||||
| Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | Authors' judgement | Support for judgement | |
| Subgroup 23.1.1 High‐income countries | ||||||||||||
| CoV‐Early | Some concerns | This trial is a nationwide multicenter, double blind, randomized controlled trial in the Netherlands. Patients will be randomized between the transfusion of 300mL of convP versus regular fresh frozen plasma (FFP). However, there are not enough information avaiable yet on the allocation sequence concealement, as the full text article is not published yet and only the trial registry and requested data are available yet. | Low risk of bias | Both participants and carers and people delivering the intervention were unaware of the assigned intervention received. There is no information available yet ony whether the analysis was appropriate. | Some concerns | The trial registry does not provide enough information on missing data. | Low risk of bias | Mortality is an objective outcome measure, appropriately measured and it is unlikely that the measurement differed between intervention groups. The outcome assessors were probably not aware of the intervention received. | Low risk of bias | The data that produced this result was analysed in accordance with the pre‐specified outcome in the trial registry. | Some concerns | There are some concerns, as the data was provided on request and the study results are not published yet. |
| Sullivan 2022 | Low risk of bias | Participants from all sites were randomized by a central web‐based system using blocked assignments to receive ABO compatible control plasma or CCP (both∼250 mL) in a 1:1 ratio. | Low risk of bias | Participants were blinded, but and those delivering the intervention were aware of intervention received. However, there were no deviations from intended interventions and the analysis was appropriate. | Low risk of bias | A modified intention‐to‐treat was used, so data for this outcome was available for nearly all participants randomized. | Low risk of bias | The measurement of the outcome was appropriate and it is unlikely that it differed between intervention groups. The outcome assessors were aware of the intervention received, but it is unlikely that knowledge of the intervention received could have affected outcome measurement. | Low risk of bias | There was no information on whether the data that produced this result was analysed in accordance with the predefined outcomes, as there is no protocol available and no information on mortality outcome provided in the trial registry. However, for this mortality outcome, the risk of bias remains still low. | Low risk of bias | ‐ |
Acknowledgements
Cochrane Haematology supported the authors in the development of this Review Update. Claire Iannizzi, Nina Kreuzberger, Nicole Skoetz, and Ina Monsef are members of Cochrane Haematology but were not involved in the editorial process or decision making for this review.
The following people conducted the editorial process for this Review Update:
Sign‐off Editor (final editorial decision): Mike Brown, Michigan State University College of Human Medicine, USA
Managing Editor (selected peer reviewers, collated peer‐reviewer comments, provided editorial guidance to authors, edited the article): Anne‐Marie Stephani, Cochrane Central Editorial Service
Editorial Assistant (conducted editorial policy checks and supported editorial team): Lisa Wydrzynski, Cochrane Central Editorial Service
Copy Editor (copy‐editing and production): Denise Mitchell, Evidence, Production & Methods Directorate, Cochrane Central Executive Team
Peer‐reviewers (provided comments and recommended an editorial decision): Michael J. Joyner, MD Department of Anesthesiology & Perioperative Medicine Mayo Clinic USA (clinical/content review), Miquel Lozano, MD, PhD. Department of Hemotherapy and Hemostasis, Hospital Clinic. University of Barcelona. Barcelona. Spain (clinical/content review), Annabel Dawson (consumer review), Rachel Richardson, Associate Editor, Cochrane (methods review), Robin Featherstone, Cochrane Central Editorial Service (search review).
The authors would also like to thank those who contributed to previous versions of this review: Sarah Hodgkinson and Liz Bickerdike (Associate Editors, Cochrane Editorial and Methods Department), Theresa Moore (Methodology Editor, Editorial and Methods Department) for reviewing our risk of bias assessments and the implementation of RoB 2, Gerald Gartlehner and Adrienne Stevens for their advice on rapid review methodology, Carolyn Dorée (Information Specialist, Systematic Review Initiative, NHS Blood and Transplant Oxford) for developing the original search strategy for the first published review version, Analysis of Review Group Output (ARGO) for their comments on the Abstract, and the previous peer reviewer Dr Michael James Ankcorn (Department of Virology, Northern General Hospital, Sheffield Teaching Hospitals NHS Foundation Trust, UK).
We thank the investigators of 11 studies (Agarwal 2020; AlQahtani 2021; Avendano‐Sola 2021; Balcells 2020; Baldeon 2022; Bennett‐Guerrero 2021; De Santis 2022; Gharbharan 2021; Horby 2021b; Li 2020; Rasheed 2020) for providing us with additional information and data.
We thank Rujan Shrestha and Ya‐Ying Wang for translating and assessing Chinese language articles, and Lev E. Korobchenko for translating and assessing articles in Russian for us via Cochrane TaskExchange.
The research was supported by NHS Blood and Transplant and the National Institute for Health Research (NIHR) Oxford Biomedical Research Centre (BRC). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.
This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No. 101015756. The contents of this document reflect only the author's view and the Commission is not responsible for any use that may be made of the information it contains.
Appendices
Appendix 1. Search strategies
MEDLINE
# Searches
1. Coronavirus Infections/ or Coronavirus/ or SARS‐CoV‐2/ or COVID‐19/
2. ("2019 nCoV" or 2019nCoV or coronavir* or coronovir* or COVID or COVID19 or HCoV* or "nCov 2019" or "SARS CoV2" or "SARS CoV 2" or SARSCoV2 or "SARSCoV 2" or anti‐flu* or anti‐influenza* or antiflu* or antinfluenza*).tw,kf.
3. ((corona* or corono*) adj1 (virus* or viral* or virinae*)).tw,kf.
4. "severe acute respiratory syndrome coronavirus 2".tw,kf,nm.
5. or/1‐4
6. Plasma/ or Immunoglobulins/ or Immunoglobulins, Intravenous/ or Immune Sera/
7. ((convalesc* or recovered or cured or rehabilitat* or survivor* or survived or virus‐positive or virus neutrali* or virus inactivated or antibod* or high‐titre* or high‐titer*) adj6 (plasma or blood or serum or sera)).tw,kf.
8. (high‐dos* adj3 (plasma or immunoglobulin* or IVIG* or immune globulin* or globulin* or IgG)).tw,kf.
9. ((plasma adj1 therap*) or gamma‐globulin or gammaglobulin or "y‐Globulin" or hyper‐lg or "C19‐IG").tw,kf.
10. (plasma adj5 (immun* or antibod* or exchange* or donor* or donat* or transfus* or infus*)).tw,kf.
11. ((convalesc* or recovered or cured or rehabilitat* or survivor* or survived or virus‐positive or virus inactivated or antibody‐positive) adj5 (donor* or donat*)).tw,kf.
12. (hyperimmune* or hyper‐immune* or serotherap* or sero‐therap*).tw,kf.
13. exp Immunization, Passive/
14. (passiv* adj3 (antibod* transfer* or immunization* or immunotherap* or immuno‐therap* or vaccin*)).tw,kf.
15. (passiv* adj3 (therap* or treatment* or neutrali?ing or prevent* or protect* or prophylax*)).tw,kf.
16. ((immunoglobulin* or immune globulin*) adj2 (therap* or treatment* or prevent* or protect* or prophylax*)).tw,kf.
17. (passive immunit* or hIVIG or CSL760 or INM005 or XAV‐19 or SAB‐185 or equine or IgY‐110 or IgY110 or GIGA‐2050 or GIGA2050 or GC5131 or 5131A).tw,kf.
18. (((anti‐coronavirus or anticoronavirus) adj1 immunoglobulin*) or ITAC or "Hyperimmune anti‐COVID‐19 IVIG" or C‐IVIG or CIVIG or equine polyclonal antibod* or EpAbs or BSVEQAb or EqAb‐COV‐19 or flebogamma or "F(ab)2").tw,kf.
19. ((bovine adj2 (colostrum* or milk*)) or bioblock*).tw,kf.
20. or/6‐19
21. 5 and 20
22. "Covid‐19 Serotherapy".px.
23. or/21‐22
24. randomized controlled trial.pt.
25. controlled clinical trial.pt.
26. randomi?ed.ab.
27. placebo.ab.
28. drug therapy.fs.
29. randomly.ab.
30. trial.ab.
31. groups.ab.
32. or/24‐31
33. exp animals/ not humans/
34. 32 not 33
35. clinical trial, phase iii/
36. ("Phase 3" or "phase3" or "phase III" or P3 or "PIII").ti,ab,kw.
37. (35 or 36) not 33
38. 34 or 37
39. 23 and 38
40. limit 39 to yr="2020 ‐Current"
41. remove duplicates from 40
Embase
# Searches
1. coronavirinae/ or coronaviridae/ or coronaviridae infection/ or coronavirus disease 2019/ or Coronavirus infection/
2. sars‐related coronavirus/ or "Severe acute respiratory syndrome coronavirus 2"/
3. ((corona* or corono*) adj1 (virus* or viral* or virinae*)).tw,kw.
4. ("2019 nCoV" or 2019nCoV or coronavir* or coronovir* or COVID or COVID19 or HCoV* or "nCov 2019" or "SARS CoV2" or "SARS CoV 2" or SARSCoV2 or "SARSCoV 2").tw,kw.
5. "Severe acute respiratory syndrome coronavirus 2".mp.
6. or/1‐5
7. Plasma Transfusion/ or exp Immunoglobulin/
8. ((convalesc* or recovered or cured or survivor* or survived or rehabilitat* or virus‐positive or virus‐neutrali* or virus inactived or antibody‐rich or high‐tire* or high‐titer*) adj6 (plasma or blood or serum or sera)).mp.
9. ((plasma adj1 therap*) or gamma‐globulin or gammaglobulin or "y‐Globulin" or hyper‐lg or "C19‐IG").tw,kw.
10. (plasma adj5 (immun* or antibod* or exchange* or donor* or donat* or transfus* or infus*)).tw,kw.
11. (high‐dos* adj3 (plasma or immunoglobulin* or IVIG* or immune globulin* or globulin* or IgG)).tw,kw.
12. ((convalesc* or recovered or cured or rehabilitat* or survivor* or survived or virus‐positive or virus inactivated or antibody‐positive) adj5 (donor* or donat*)).tw,kw.
13. (serotherap* or sero‐therap* or hyperimmune* or hyper‐immune*).tw,kw.
14. passive immunization/
15. (passiv* adj3 (therap* or treatment* or neutrali?ing or prevent* or protect* or prophylax*)).tw,kw.
16. (passive immunit* or hIVIG or CSL760 or INM005 or XAV‐19 or SAB‐185 or equine or IgY‐110 or IgY110 or GIGA‐2050 or GIGA2050 or GC5131 or 5131A).tw,kw.
17. (passiv* adj3 (antibod* transfer* or immunization* or immunotherap* or immuno‐therap* or vaccin*)).tw,kw.
18. ((immunoglobulin* or immune globulin*) adj2 (therap* or treatment* or prevent* or protect* or prophylax*)).tw,kw.
19. (((anti‐coronavirus or anticoronavirus) adj1 immunoglobulin*) or ITAC or "Hyperimmune anti‐COVID‐19 IVIG" or C‐IVIG or CIVIG or equine polyclonal antibod* or EpAbs or BSVEQAb or EqAb‐COV‐19 or flebogamma or "F(ab)2").tw,kw.
20. ((bovine adj2 (colostrum* or milk*)) or bioblock*).tw,kw.
21. or/7‐20
22. Randomized controlled trial/
23. Controlled clinical trial/
24. random*.ti,ab.
25. randomization/
26. intermethod comparison/
27. placebo.ti,ab.
28. (compare or compared or comparison).ti.
29. ((evaluated or evaluate or evaluating or assessed or assess) and (compare or compared or comparing or comparison)).ab.
30. (open adj label).ti,ab.
31. ((double or single or doubly or singly) adj (blind or blinded or blindly)).ti,ab.
32. double blind procedure/
33. parallel group*1.ti,ab.
34. (crossover or cross over).ti,ab.
35. ((assign$ or match or matched or allocation) adj5 (alternate or group$1 or intervention$1 or patient$1 or subject$1 or participant$1)).ti,ab.
36. (assigned or allocated).ti,ab.
37. (controlled adj7 (study or design or trial)).ti,ab.
38. (volunteer or volunteers).ti,ab.
39. human experiment/
40. trial.ti.
41. or/22‐40
42. (random$ adj sampl$ adj7 (cross section$ or questionnaire$1 or survey$ or database$1)).ti,ab. not (comparative study/ or controlled study/ or randomi?ed controlled.ti,ab. or randomly assigned.ti,ab.)
43. Cross‐sectional study/ not (randomized controlled trial/ or controlled clinical study/ or controlled study/ or randomi?ed controlled.ti,ab. or control group$1.ti,ab.)
44. (((case adj control$) and random$) not randomi?ed controlled).ti,ab.
45. (Systematic review not (trial or study)).ti.
46. (nonrandom$ not random$).ti,ab.
47. Random field$.ti,ab.
48. (random cluster adj3 sampl$).ti,ab.
49. (review.ab. and review.pt.) not trial.ti.
50. we searched.ab. and (review.ti. or review.pt.)
51. update review.ab.
52. (databases adj4 searched).ab.
53. (rat or rats or mouse or mice or swine or porcine or murine or sheep or lambs or pigs or piglets or rabbit or rabbits or cat or cats or dog or dogs or cattle or bovine or monkey or monkeys or trout or marmoset$1).ti. and animal experiment/
54. Animal experiment/ not (human experiment/ or human/)
55. or/42‐54
56. 41 not 55
57. phase 3 clinical trial/
58. ("Phase 3" or "phase3" or "phase III" or P3 or "PIII").tw,kw.
59. (animal experiment/ or Animal experiment/) not (human experiment/ or human/)
60. (57 or 58) not 59
61. 57 or 58
62. 61 not 59
63. 6 and 21 and (56 or 60)
64. limit 63 to yr="2020 ‐Current"
65. limit 64 to medline
66. 64 not 65
67. remove duplicates from 66
Cochrane COVID‐19 Study Register
plasma or convalesc* or serum or sera or donor* or donation* or serotherapy or "sero‐therapy" or "flu‐IVIG" or "passive immunity" or hyperimmune* or "hyper‐immune" or IVIG or immunoglobulin* or "immune‐globulin" or "immune‐globuline" or globulin* or "gamma‐globulin" or "γ‐Globulin" or "hyper‐Ig" or immunization or immunisation or immunotherap* or "immuno‐therapy" or CSL760 or INM005* or equine* or "XAV‐19" or "SAB‐185" or hIVIG* or INOSARS* or "GIGA‐2050" or GIGA2050 or "IGY‐110" or IGY1109 or "GC5131" or "5131A" or ITAC* or "C‐IVIG" or CIVIG or flebogamma* or EpAbs* or BSVEQAb* or "EqAb‐COV‐19" or "F(ab)2" or "bovine colostrum" or "bovine milk" or "SARS‐CoV‐2‐IG" or "hyper‐Ig" or bioblock*
Study characteristics:
1) Intervention assignment: randomised, unclear
2) Study design: parallel/crossover, unclear"
PubMed
#1 "2019 ncov"[Title/Abstract] OR "2019nCoV"[Title/Abstract] OR "corona virus"[Title/Abstract] OR "corona viruses"[Title/Abstract] OR "Coronavirus"[Title/Abstract] OR "coronaviruses"[Title/Abstract] OR "COVID"[Title/Abstract] OR "COVID19"[Title/Abstract] OR "ncov 2019"[Title/Abstract] OR "SARS‐CoV2"[Title/Abstract] OR "SARS‐CoV‐2"[Title/Abstract] OR "SARSCoV2"[Title/Abstract] OR "SARSCoV‐2"[Title/Abstract] OR "COVID‐19"[MeSH Terms] OR "Coronavirus"[MeSH Terms:noexp] OR "SARS‐CoV‐2"[MeSH Terms] OR "COVID‐19"[Supplementary Concept] OR "severe acute respiratory syndrome coronavirus 2"[Supplementary Concept]
#2 ("convalesc*"[Title/Abstract] OR "recovered"[Title/Abstract] OR "cured"[Title/Abstract] OR "rehabilitat*"[Title/Abstract] OR "survivor*"[Title/Abstract] OR "survived"[Title/Abstract] OR "virus‐positive"[Title/Abstract] OR "virus neutrali*"[Title/Abstract] OR "virus inactivated"[Title/Abstract] OR "antibod*"[Title/Abstract] OR "high titre*"[Title/Abstract] OR "high titer*"[Title/Abstract]) AND ("plasma"[Title/Abstract] OR "blood"[Title/Abstract] OR "donor*"[Title/Abstract] OR "donat*"[Title/Abstract])
#3 ("plasma"[Title] AND ("immun*"[Title/Abstract] OR "transfus*"[Title/Abstract] OR "infus*"[Title/Abstract] OR "exchange*"[Title/Abstract]))
#4 "high dos*"[Title/Abstract] AND ("plasma"[Title/Abstract] OR "immunoglobulin*"[Title/Abstract] OR "ivig*"[Title/Abstract] OR "immune globulin*"[Title/Abstract] OR "globulin*"[Title/Abstract] OR "IgG"[Title/Abstract])
#5 "immunization, passive"[MeSH Terms] OR "passive immunit*"[Title/Abstract] OR "hyperimmune"[Title/Abstract] OR "hyperimmunity"[Title/Abstract] OR "serotherap*"[Title/Abstract] OR "sero therap*"[Title/Abstract] OR "therapeutic plasma"[Title/Abstract] OR "plasma therapy"[Title/Abstract] OR "immune plasma"[Title/Abstract] OR "plasma exchange"[Title/Abstract] OR "serum"[Title] OR "sera"[ Title]
#6 "passiv*"[Title/Abstract] AND (("antibod*"[Title/Abstract] AND "transfer*"[Title/Abstract]) OR "immunisation*"[Title/Abstract] OR "vaccin*"[Title/Abstract] OR "immunization*"[Title/Abstract] OR "immunotherap*"[Title/Abstract] OR "immuno therap*"[Title/Abstract])
#7 ("immunoglobulin*"[Title] OR "immune globulin*"[Title]) AND ("therap*"[Title/Abstract] OR "treat*"[Title/Abstract] OR "prevent*"[Title/Abstract] OR "protect*"[Title/Abstract] OR "prophylax*"[Title/Abstract])
#8 "bovine colostrum"[Title/Abstract] OR "bovine milk"[Title/Abstract] OR "F(ab)2"[Title/Abstract] OR "equine*"[Title/Abstract] OR "Hyperimmune anti‐COVID‐19 IVIG"[Title/Abstract] OR "c ivig*"[Title/Abstract] OR "XAV‐19"[Title/Abstract] OR "5131A"[Title/Abstract] OR "equine polyclonal antibod*"[Title/Abstract] OR "EpAbs"[Title/Abstract] OR "flebogamma*"[Title/Abstract] OR "BSVEQAb"[Title/Abstract] OR "EqAb‐COV‐19"[Title/Abstract] OR "γ‐Globulin"[Title/Abstract] OR "hyper‐Ig"[Title/Abstract] OR Title/Abstract] OR INM005*[Title/Abstract] OR "SAB‐185"[Title/Abstract] OR hIVIG*[Title/Abstract] OR INOSARS*[Title/Abstract] OR "GIGA‐2050"[Title/Abstract] OR GIGA2050[Title/Abstract] OR "IGY‐110"[Title/Abstract] OR "GC5131"[Title/Abstract] OR "5131A"[Title/Abstract] OR ITAC*[Title/Abstract] OR "C‐IVIG"[Title/Abstract] OR CIVIG[Title/Abstract] OR XVR011* [Title/Abstract] OR bioblock*[Title/Abstract] OR "gammaglobulin*"[Title/Abstract] OR "gamma‐globulin"[Title/Abstract] OR "hyper‐Ig"[Title/Abstract]
#9 (("anti‐coronavirus"[Title/Abstract] OR "anticoronavirus"[Title/Abstract]) AND "immunoglobulin*"[Title/Abstract]))
#10 #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9
#11 #1 AND #10
#12 ((randomized controlled trial[pt] OR controlled clinical trial[pt] OR randomized[tiab] OR placebo[tiab] OR drug therapy[sh] OR randomly[tiab] OR trial[tiab] OR groups[tiab] NOT (animals [mh] NOT humans [mh]))
#13 (publisher[sb] OR inprocess[sb] OR pubmednotmedline[sb])
#14 #11 AND #12 AND #13
Filters: from 2020/1/1 ‐ 3000/12/12
World Health Organization COVID‐19 Global literature on coronavirus disease
Advanced search: search fileds: title, abstract subject (ohne Embase, Medline ICTRP, PubMed)
I:
random* or placebo or trial or groups or "phase 3" or "phase3" or p3 or "pIII"
and
convalesc*~6plasma or convalesc*~6blood or convalesc*~6serum or convalesc*~6sera or cured~6plasma or cc‐ured~6blood or cured~6serum or cured~6sera or survivor*~6plasma or survivor*~6blood or survivor*~6serum or survivor*~6sera or survived*~6plasma or survived*~6blood or survived*~6serum or survived*~6sera or rehabilitat*~6plasma or rehabilitat*~6blood or rehabilitat*~6serum or rehabilitat*~6sera or virus‐positive~6plasma or virus‐positive~6blood or virus‐positive~6serum or virus‐positive~6sera or virus‐neutrali*~6plasma or virus‐neutrali*~6blood or virus‐neutrali*~6serum or virus‐neutrali*~6sera or virus inactived~6plasma or virus inactived~6blood or virus inactived~6serum or virus inactived~6sera or antibody‐rich~6plasma or antibody‐rich~6blood or antibody‐rich~6serum or antibody‐rich~6sera or high‐tire*~6plasma or high‐tire*~6blood or high‐tire*~6serum or high‐tire*~6sera or high‐titer*~6plasma or high‐titer*~6blood or high‐titer*~6serum or high‐titer*~6sera or plasma~1therap*
II:
random* or placebo or trial or groups or "phase 3" or "phase3" or p3 or "pIII"
and
gamma‐globulin or "y‐Globulin" or hyper‐lg or plasma~5immun* or plasma~5antibod* or plasma~5exchange* or plasma~5donor* or plasma~5 donat* or plasma~5transfus* or plasma~5infus* or high‐dos*~3plasma or high‐dos*~3immunoglobulin* or high‐dos*~3IVIG* or high‐dos*~3immune globulin* or high‐dos*~3globulin* or high‐dos*~3IgG or convalesc*~5donor* or recovered~5donor* or cured~5donor* or rehabilitat*~5donor* or survivor*~5donor* or survived~5donor* or virus‐positive~5donor* or virus inactivated~5donor* or antibody‐positive~5donor* or convalesc*~5donat or recovered~5donat or cured~5donat or rehabilitat*~5donat or survivor*~5donat or survived~5donat or virus‐positive~5donat or virus inactivated~5donat or antibody‐positive~5donat or serotherap* or sero‐therap* or hyperimmune*
III:
random* or placebo or trial or groups or "phase 3" or "phase3" or p3 or "pIII"
and
hyper‐immune* or passiv*~3therap* or passiv*~3treatment* or passiv*~3neutralising or passiv*~3neutralizing or passiv*~3prevent* or passiv*~3protect* or passiv*~3prophylax* or immunoglobulin*~2therap* or immunoglobulin*~2treat* or immunoglobulin*~2prevent* or immunoglobulin*~2protect* or immunoglobulin*~2prophylax* or immune globulin*~2therap* or immune globulin*~2treat* or immune globulin*~2prevent* or immune globulin*~2protect* or immune globulin*~2 or passive immunit*
IV:
random* or placebo or trial or groups or "phase 3" or "phase3" or p3 or "pIII"
and
hIVIG or CSL760 or INM005 or "XAV‐19" or "SAB‐185" or equine or "IgY‐110" or IgY110 or "GIGA‐2050" or GIGA2050 or GC5131 or 5131A or ITAC or "Hyperimmune anti‐COVID‐19 IVIG" or C‐IVIG or CIVIG or EpAbs or BSVEQAb or "EqAb‐COV‐19" or flebogamma or bovine~2colostrum* or bovine~2milk*
V:
random* or placebo or trial or groups or "phase 3" or "phase3" or p3 or "pIII"
and
passiv*~3antibod* or passiv*~3transfer* or passiv*~3immunization* or passiv*~3immunotherap* or passiv*~3immuno‐therap* or passiv*~3vaccin* or anti‐coronavirus~1immunoglobulin* or anticoronavirus~1immunoglobulin*
Epistemonikos, L*OVE List Coronavirus disease (COVID‐19)
Covid‐19 by PICO
Prevention or treatment: passive immunization: antibody therapies: convalescent plasma
Filtered by primary studies and results by RCT
Prevention or treatment: passive immunization: antibody therapies: immunoglobulin therapy
Filtered by primary studies and results by RCT
Prevention or treatment: passive immunization: Heterologous antibodies
Filtered by primary studies and results by RCT
Appendix 2. Transformations and recalculations of outcomes
Table A8.1: moderate to severe disease
| Convalescent plasma versus placebo, standard care alone, standard plasma, or human immunoglobulin in moderate to severe disease | |||
| Outcome | Study | Source of data | Transformation or recalculation |
| Mortality at longest follow‐up | Beltran Gonzalez 2021 | Beltran Gonzalez 2021 | Recalculation of HR and 95% CI according to methods by Tierney 2007 based on number of events and P value |
| De Santis 2022 | De Santis 2022 | ||
| Kirenga 2021 | Kirenga 2021 | ||
| Koerper 2021 | Koerper 2021 | ||
| Menichetti 2021 | Menichetti 2021 | ||
| O’Donnell 2021 | O’Donnell 2021 | ||
Table A8.2: mild disease
| Convalescent plasma versus placebo, standard care alone, standard plasma, or human immunoglobulin in outpatients with mild disease | |||
| Outcome | Study | Source of data | Transformation or recalculation |
| Mortality up to day 28 | CoV‐Early | Unpublished data provided by the primary study authors | Recalculation based on provided data; any death within 28 days |
Appendix 3. Planned methodology for study designs that are no longer included in this systematic review
Criteria for considering studies for this review
Types of studies
In case of insufficient evidence available from randomised controlled trials (RCTs), we planned to include prospective controlled non‐randomised studies of interventions (NRSIs), including quasi‐RCTs (e.g. assignment to treatment by alternation or by date of birth), controlled before‐and‐after (CBA) studies, and interrupted time series (ITS) studies. We planned to use the methods proposed in the Cochrane Handbook for Systematic Reviews of Interventions for the inclusion of controlled NRSIs in systematic reviews (Reeves 2022).
We further planned to include retrospective controlled NRSIs, in case of insufficient evidence (very low‐certainty evidence or no evidence) available from RCTs and prospective controlled NRSIs and to adapt the methods for the inclusion of controlled NRSIs in systematic reviews as specified by the Cochrane Handbook for Systematic Reviews of Interventions (Reeves 2022).
In case the evidence that we found from RCTs was at high risk of bias and at critical risk of bias for the controlled NRSIs for safety outcomes, we planned to also included safety data from prospectively and retrospectively registered non‐controlled NRSIs, for example, case series, and followed the methodology as specified in the protocol (Piechotta 2020a).
Data collection and analysis
Assessment of risk of bias in included studies
Controlled non‐randomised studies of interventions
As reported above, we planned to include controlled non‐randomised studies of intervention (NRSI) trials if there was insufficient evidence from RCTs.
Two review authors (VP, NS) would have independently assessed eligible studies for methodological quality and risk of bias (using the Risk Of Bias in Non‐randomised Studies ‐ of Interventions (ROBINS‐I) tool; Sterne 2016). The quality assessment strongly depends upon information on the design, conduct and analysis of the trial. The two review authors would have resolved any disagreements regarding quality assessments by discussion, and in case of discrepancies among their judgements, or inability to reach consensus, we had to consult a third review author until consensus could be reached. We asked the Cochrane Editorial and Methods Department (Theresa Moore) to review our judgements for reasonability for previous versions of this review. The categories for risk of bias judgements for controlled NRSIs using ROBINS‐I are 'low risk', 'moderate risk', 'serious risk' and 'critical risk' of bias.
We planned to assess the following domains of bias.
Bias due to confounding
Bias in selection of participants into the study
Bias in classification of interventions
Bias due to deviations from intended interventions
Bias due to missing data
Bias in measurement of outcomes
Bias in selection of the reported result
For every criterion we planned to make a judgement using one of five response options.
Yes
Probably yes
Probably no
No
No information
Measures of treatment effect
Controlled non‐randomised studies of interventions
For dichotomous outcomes, if available, we planned to extract and report the risk ratio (RR) with a 95% confidence interval from statistical analyses adjusting for baseline differences (such as Poisson regressions or logistic regressions) or the ratio of RRs (i.e. the RR post‐intervention/RR pre‐intervention).
For continuous variables, if available, we planned to extract and report the absolute change from a statistical analysis adjusting for baseline differences (such as regression models, mixed models or hierarchical models), or the relative change adjusted for baseline differences in the outcome measures (i.e. the absolute post‐intervention difference between the intervention and control groups, as well as the absolute pre‐intervention difference between the intervention and control groups/the post‐intervention level in the control group; EPOC 2017).
Data synthesis
We planned to not synthesise efficacy data from controlled NRSIs if they were at critical risk of bias. If a meta‐analysis had been feasible for controlled NRSIs we planned to analyse the different types of studies separately. We planned to only analyse outcomes with adjusted effect estimates if these were adjusted for the same factors using the inverse‐variance method as recommended in Chapter 24 of the Cochrane Handbook for Systematic Reviews of Interventions (Reeves 2022).
Summary of findings and assessment of the certainty of the evidence
As we had planned to use the ROBINS‐I tool to assess risk of bias for controlled NRSIs, we planned to follow GRADE guidance 18 to rate the certainty in the evidence for controlled NRSIs; starting from high‐certainty evidence with the opportunity to downgrade by three points for critical risk of bias (Schünemann 2019).
Data and analyses
Comparison 1. Convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 1.1 All‐cause mortality at up to day 28 | 21 | 19021 | Risk Ratio (M‐H, Random, 95% CI) | 0.98 [0.92, 1.03] |
| 1.1.1 Individuals with moderate disease | 8 | 2336 | Risk Ratio (M‐H, Random, 95% CI) | 0.92 [0.70, 1.22] |
| 1.1.2 Individuals with severe disease | 4 | 2265 | Risk Ratio (M‐H, Random, 95% CI) | 0.78 [0.50, 1.22] |
| 1.1.3 Individuals with moderate to severe disease | 9 | 14420 | Risk Ratio (M‐H, Random, 95% CI) | 0.98 [0.93, 1.04] |
| 1.2 All‐cause mortality at up to day 60 | 3 | 272 | Risk Ratio (M‐H, Random, 95% CI) | 0.74 [0.49, 1.12] |
| 1.2.1 Individuals with moderate to severe disease | 2 | 165 | Risk Ratio (M‐H, Random, 95% CI) | 0.63 [0.35, 1.13] |
| 1.2.2 Individuals with severe disease | 1 | 107 | Risk Ratio (M‐H, Random, 95% CI) | 0.87 [0.48, 1.56] |
| 1.3 All‐cause mortality (time to event) | 16 | 17070 | Hazard Ratio (IV, Random, 95% CI) | 0.98 [0.92, 1.04] |
| 1.3.1 Individuals with moderate disease | 4 | 1290 | Hazard Ratio (IV, Random, 95% CI) | 0.81 [0.55, 1.19] |
| 1.3.2 Individuals with severe disease | 4 | 2267 | Hazard Ratio (IV, Random, 95% CI) | 0.77 [0.48, 1.25] |
| 1.3.3 Individuals with moderate to severe disease | 8 | 13513 | Hazard Ratio (IV, Random, 95% CI) | 0.99 [0.93, 1.07] |
| 1.4 All‐cause mortality during hospital stay | 4 | 2556 | Risk Ratio (M‐H, Random, 95% CI) | 0.97 [0.87, 1.08] |
| 1.4.1 Individuals with moderate disease | 2 | 491 | Risk Ratio (M‐H, Random, 95% CI) | 1.12 [0.71, 1.76] |
| 1.4.2 Individuals with severe disease | 2 | 2065 | Risk Ratio (M‐H, Random, 95% CI) | 0.87 [0.56, 1.36] |
| 1.5 Clinical worsening: need for invasive mechanical ventilation or death at up to day 28 | 6 | 14477 | Risk Ratio (M‐H, Random, 95% CI) | 1.03 [0.97, 1.11] |
| 1.5.1 Individuals with moderate disease | 3 | 1308 | Risk Ratio (M‐H, Random, 95% CI) | 1.05 [0.77, 1.43] |
| 1.5.2 Individuals with severe disease | 1 | 1307 | Risk Ratio (M‐H, Random, 95% CI) | 1.09 [0.97, 1.22] |
| 1.5.3 Individuals with moderate to severe disease | 2 | 11862 | Risk Ratio (M‐H, Random, 95% CI) | 1.03 [0.90, 1.18] |
| 1.6 Clinical improvement: participants discharged from hospital | 6 | 12721 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.97, 1.02] |
| 1.6.1 Individuals with moderate disease | 1 | 333 | Risk Ratio (M‐H, Random, 95% CI) | 0.98 [0.86, 1.12] |
| 1.6.2 Individuals with severe disease | 1 | 101 | Risk Ratio (M‐H, Random, 95% CI) | 1.42 [0.90, 2.24] |
| 1.6.3 Individuals with moderate to severe disease | 4 | 12287 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.97, 1.02] |
| 1.7 Quality of life, assessed with standardised scales at day 28 | 1 | 483 | Mean Difference (IV, Random, 95% CI) | 1.00 [‐2.14, 4.14] |
| 1.8 Any grade adverse events | 7 | 1809 | Risk Ratio (M‐H, Random, 95% CI) | 1.05 [0.95, 1.17] |
| 1.9 Grades 1‐2 adverse events | 1 | 160 | Risk Ratio (M‐H, Random, 95% CI) | 1.11 [0.87, 1.41] |
| 1.10 Grades 3 and 4 adverse events | 6 | 2392 | Risk Ratio (M‐H, Random, 95% CI) | 1.17 [0.96, 1.42] |
| 1.10.1 Individuals with moderate disease | 4 | 1311 | Risk Ratio (M‐H, Random, 95% CI) | 1.48 [0.43, 5.02] |
| 1.10.2 Individuals with moderate to severe disease | 2 | 1081 | Risk Ratio (M‐H, Random, 95% CI) | 1.18 [1.02, 1.37] |
| 1.11 Serious adverse events | 6 | 3901 | Risk Ratio (M‐H, Random, 95% CI) | 1.14 [0.91, 1.44] |
| 1.11.1 Individuals with moderate disease | 1 | 333 | Risk Ratio (M‐H, Random, 95% CI) | 1.31 [0.82, 2.09] |
| 1.11.2 Individuals with moderate to severe disease | 5 | 3568 | Risk Ratio (M‐H, Random, 95% CI) | 1.11 [0.85, 1.46] |
| 1.12 Clinical improvement: weaning or liberation from invasive mechanical ventilation in surviving participants, for subgroups of participants requiring invasive mechanical ventilation at baseline | 2 | 630 | Risk Ratio (M‐H, Random, 95% CI) | 1.04 [0.57, 1.93] |
| 1.13 Clinical improvement: ventilator‐free days by day 28 | 2 | 1028 | Mean Difference (IV, Random, 95% CI) | ‐0.53 [‐1.90, 0.84] |
| 1.14 Clinical improvement: liberation from supplemental oxygen in surviving participants, for subgroup of participants requiring any supplemental oxygen or ventilator support at baseline | 2 | 560 | Risk Ratio (M‐H, Random, 95% CI) | 0.99 [0.91, 1.08] |
| 1.15 Need for dialysis at up to 28 days | 2 | 12325 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 1.03 [0.86, 1.23] |
| 1.16 Admission to the intensive care unit (ICU) | 2 | 816 | Risk Ratio (M‐H, Random, 95% CI) | 0.93 [0.77, 1.11] |
| 1.17 Duration of hospitalisation | 2 | 97 | Mean Difference (IV, Random, 95% CI) | ‐1.04 [‐6.87, 4.79] |
| 1.18 Viral clearance at up to day 3 | 4 | 619 | Risk Ratio (M‐H, Random, 95% CI) | 1.29 [0.76, 2.18] |
| 1.19 Viral clearance at up to day 7 | 4 | 674 | Risk Ratio (M‐H, Random, 95% CI) | 1.24 [0.85, 1.79] |
| 1.20 Viral clearance at up to day 14 | 2 | 212 | Risk Ratio (M‐H, Random, 95% CI) | 1.46 [0.58, 3.68] |
Comparison 2. Convalescent plasma versus standard plasma for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 2.1 All‐cause mortality at up to day 28 | 4 | 484 | Risk Ratio (M‐H, Random, 95% CI) | 0.73 [0.45, 1.19] |
| 2.1.1 Individuals with moderate disease | 1 | 158 | Risk Ratio (M‐H, Random, 95% CI) | 0.88 [0.37, 2.11] |
| 2.1.2 Individuals with moderate to severe disease | 3 | 326 | Risk Ratio (M‐H, Random, 95% CI) | 0.75 [0.36, 1.55] |
| 2.2 All‐cause mortality (time to event) | 4 | 484 | Hazard Ratio (IV, Random, 95% CI) | 0.94 [0.41, 2.14] |
| 2.3 Clinical worsening: need for invasive mechanical ventilation or death at up to day 28 | 1 | 34 | Risk Ratio (M‐H, Random, 95% CI) | 5.59 [0.29, 108.38] |
| 2.4 Duration of hospitalisation | 2 | 187 | Mean Difference (IV, Random, 95% CI) | ‐2.14 [‐5.24, 0.95] |
| 2.5 Any grade adverse events | 2 | 253 | Risk Ratio (M‐H, Random, 95% CI) | 1.18 [0.93, 1.50] |
| 2.6 Serious adverse events | 3 | 327 | Risk Ratio (M‐H, Random, 95% CI) | 0.80 [0.55, 1.15] |
| 2.6.1 Individuals with moderate to severe disease | 3 | 327 | Risk Ratio (M‐H, Random, 95% CI) | 0.80 [0.55, 1.15] |
Comparison 3. Convalescent plasma versus human immunoglobulin for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 3.1 All‐cause mortality at up to day 28 | 1 | 190 | Risk Ratio (M‐H, Random, 95% CI) | 1.07 [0.76, 1.50] |
| 3.2 All‐cause mortality (time to event) | 1 | 190 | Hazard Ratio (IV, Random, 95% CI) | 1.14 [0.84, 1.54] |
| 3.3 All‐cause mortality during hospital stay | 1 | 190 | Risk Ratio (M‐H, Random, 95% CI) | 1.01 [0.76, 1.34] |
Comparison 4. Convalescent plasma versus placebo or standard care alone for outpatients with mild disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 4.1 All‐cause mortality at up to day 28 | 2 | 536 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.36 [0.09, 1.46] |
| 4.2 All‐cause mortality at up to day 60 | 1 | 376 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.13 [0.01, 2.16] |
| 4.3 Admission to hospital or death within 28 days | 1 | 376 | Risk Ratio (M‐H, Random, 95% CI) | 1.05 [0.60, 1.84] |
| 4.4 Time to symptom resolution | 1 | 376 | Hazard Ratio (IV, Random, 95% CI) | 1.05 [0.85, 1.30] |
| 4.5 Clinical worsening: need for hospitalisation with at least need of oxygen by mask or nasal prongs, or death | 2 | 536 | Risk Ratio (M‐H, Random, 95% CI) | 0.76 [0.36, 1.59] |
| 4.6 Clinical worsening: need for invasive mechanical ventilation or death at up to day 28 | 1 | 376 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.51 [0.10, 2.55] |
| 4.7 Clinical worsening: need for invasive mechanical ventilation or death at up to day 60 | 1 | 376 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.51 [0.10, 2.55] |
| 4.8 Grades 3 and 4 adverse events | 1 | 376 | Risk Ratio (M‐H, Random, 95% CI) | 1.29 [0.75, 2.19] |
| 4.9 Serious adverse events | 1 | 376 | Risk Ratio (M‐H, Random, 95% CI) | 1.14 [0.66, 1.94] |
Comparison 5. Convalescent plasma versus standard plasma for outpatients with mild disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 5.1 All‐cause mortality at up to day 28 | 2 | 1597 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.30 [0.05, 1.75] |
| 5.2 Admission to hospital or death within 28 days | 2 | 1595 | Risk Ratio (M‐H, Random, 95% CI) | 0.49 [0.31, 0.75] |
| 5.3 All initial symptoms resolved (asymptomatic) at up to day 28 | 1 | 416 | Risk Ratio (M‐H, Random, 95% CI) | 1.12 [0.98, 1.27] |
| 5.4 All initial symptoms resolved (asymptomatic) at up to day 14 | 1 | 417 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.83, 1.21] |
| 5.5 Clinical worsening: need for hospitalisation with need of at least oxygen by mask or nasal prongs, or death | 1 | Risk Ratio (M‐H, Random, 95% CI) | Subtotals only | |
| 5.6 Clinical worsening: need for invasive mechanical ventilation or death at up to day 28 | 1 | 414 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 0.30 [0.05, 1.77] |
Comparison 6. Subgroup analysis: antibodies in recipients detected at baseline for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 6.1 All‐cause mortality at up to day 28 (random‐effects analysis) | 5 | Risk Ratio (M‐H, Random, 95% CI) | Subtotals only | |
| 6.1.1 Antibodies detected at baseline | 5 | 7523 | Risk Ratio (M‐H, Random, 95% CI) | 1.03 [0.94, 1.12] |
| 6.1.2 No antibodies detected at baseline | 5 | 4621 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.79, 1.04] |
| 6.2 All‐cause mortality (time to event) | 2 | Hazard Ratio (IV, Random, 95% CI) | Subtotals only | |
| 6.2.1 Antibodies detected at baseline | 2 | 5997 | Hazard Ratio (IV, Random, 95% CI) | 1.06 [0.94, 1.19] |
| 6.2.2 No antibodies detected at baseline | 2 | 3913 | Hazard Ratio (IV, Random, 95% CI) | 0.83 [0.50, 1.40] |
| 6.3 Clinical worsening: need for invasive mechanical ventilation or death (random‐effects model) | 2 | 9472 | Risk Ratio (M‐H, Random, 95% CI) | 0.98 [0.86, 1.12] |
| 6.3.1 Antibodies detected at baseline | 2 | 5816 | Risk Ratio (M‐H, Random, 95% CI) | 1.06 [0.96, 1.17] |
| 6.3.2 No antibodies detected at baseline | 2 | 3656 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.84, 0.98] |
| 6.4 Clinical improvement: participants discharged from hospital | 1 | 9564 | Risk Ratio (M‐H, Random, 95% CI) | 1.01 [0.94, 1.08] |
| 6.4.1 Antibodies detected at baseline | 1 | 5888 | Risk Ratio (M‐H, Random, 95% CI) | 0.98 [0.95, 1.01] |
| 6.4.2 No antibodies detected at baseline | 1 | 3676 | Risk Ratio (M‐H, Random, 95% CI) | 1.05 [0.99, 1.11] |
6.2. Analysis.
Comparison 6: Subgroup analysis: antibodies in recipients detected at baseline for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 2: All‐cause mortality (time to event)
Comparison 7. Subgroup analysis: length of time since symptom onset for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 7.1 All‐cause mortality at up to day 28 (random‐effects model) | 4 | Risk Ratio (M‐H, Random, 95% CI) | Subtotals only | |
| 7.1.1 Duration of symptom onset up to and including 7 days | 3 | 5007 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.40, 2.05] |
| 7.1.2 Duration of symptom onset more than 7 days | 3 | 7248 | Risk Ratio (M‐H, Random, 95% CI) | 1.04 [0.95, 1.13] |
| 7.2 All‐cause mortality (time to event) (random‐effects model) | 1 | 107 | Hazard Ratio (IV, Random, 95% CI) | 0.81 [0.42, 1.55] |
| 7.2.1 Duration of symptom onset more than 7 days | 1 | 107 | Hazard Ratio (IV, Random, 95% CI) | 0.81 [0.42, 1.55] |
| 7.3 Clinical worsening: need for invasive mechanical ventilation or death (random‐effects model) | 3 | Risk Ratio (M‐H, Random, 95% CI) | Subtotals only | |
| 7.3.1 Duration of symptom onset up to and including 7 days | 2 | 4816 | Risk Ratio (M‐H, Random, 95% CI) | 1.26 [0.50, 3.20] |
| 7.3.2 Duration of symptom onset more than 7 days | 2 | 6686 | Risk Ratio (M‐H, Random, 95% CI) | 1.05 [0.96, 1.14] |
| 7.4 Clinical improvement: participants discharged from hospital | 2 | 11637 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.96, 1.04] |
| 7.4.1 Duration of symptom onset up to 7 days | 2 | 4496 | Risk Ratio (M‐H, Random, 95% CI) | 1.03 [0.98, 1.08] |
| 7.4.2 Duration of symptom onset more than 7 days | 2 | 7141 | Risk Ratio (M‐H, Random, 95% CI) | 0.98 [0.95, 1.01] |
7.2. Analysis.
Comparison 7: Subgroup analysis: length of time since symptom onset for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease, Outcome 2: All‐cause mortality (time to event) (random‐effects model)
Comparison 8. Subgroup analysis: length of time since symptom onset for the comparison of convalescent plasma versus standard plasma for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 8.1 All‐cause mortality at up to day 28 (random‐effects analysis) | 1 | Risk Ratio (M‐H, Random, 95% CI) | Subtotals only | |
| 8.1.1 Duration of symptom onset up to and including 7 days | 1 | 56 | Risk Ratio (M‐H, Random, 95% CI) | 0.72 [0.26, 1.97] |
| 8.1.2 Duration of symptom onset more than 7 days | 1 | 161 | Risk Ratio (M‐H, Random, 95% CI) | 0.51 [0.24, 1.07] |
Comparison 9. Subgroup analysis: length of time since symptom onset for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 9.1 All‐cause mortality at up to day 28 | 1 | 416 | Risk Ratio (M‐H, Random, 95% CI) | 1.15 [0.13, 10.43] |
| 9.1.1 Duration of symptom onset up to and including 7 days | 1 | 401 | Risk Ratio (M‐H, Random, 95% CI) | 3.01 [0.12, 73.57] |
| 9.1.2 Duration of symptom onset more than 7 days | 1 | 15 | Risk Ratio (M‐H, Random, 95% CI) | 0.48 [0.02, 10.07] |
| 9.2 Admission to hospital or death within 28 days | 1 | 414 | Risk Ratio (M‐H, Random, 95% CI) | 0.55 [0.27, 1.12] |
| 9.2.1 Duration of symptom onset up to and including 7 days | 1 | 400 | Risk Ratio (M‐H, Random, 95% CI) | 0.50 [0.24, 1.04] |
| 9.2.2 Duration of symptom onset more than 7 days | 1 | 14 | Risk Ratio (M‐H, Random, 95% CI) | 1.80 [0.14, 22.99] |
| 9.3 All initial symptoms resolved (asymptomatic) at up to day 28 | 1 | 416 | Risk Ratio (M‐H, Random, 95% CI) | 1.01 [0.65, 1.59] |
| 9.3.1 Duration of symptom onset up to and including 7 days | 1 | 401 | Risk Ratio (M‐H, Random, 95% CI) | 1.14 [1.00, 1.29] |
| 9.3.2 Duration of symptom onset more than 7 days | 1 | 15 | Risk Ratio (M‐H, Random, 95% CI) | 0.64 [0.27, 1.54] |
Comparison 10. Subgroup analysis: pre‐existing condition immunosuppression for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 10.1 All‐cause mortality at up to day 28 (random‐effects model) | 3 | 2210 | Risk Ratio (M‐H, Random, 95% CI) | 0.68 [0.45, 1.03] |
| 10.1.1 Immunosuppression (immune deficiency and cancer) | 3 | 172 | Risk Ratio (M‐H, Random, 95% CI) | 0.74 [0.54, 1.02] |
| 10.1.2 No immunosuppression (immune deficiency and cancer) | 3 | 2038 | Risk Ratio (M‐H, Random, 95% CI) | 0.52 [0.20, 1.38] |
| 10.2 Clinical improvement: participants discharged from hospital | 1 | 86 | Risk Ratio (M‐H, Random, 95% CI) | 1.08 [0.83, 1.41] |
| 10.2.1 Immunosuppression (immune deficiency and cancer) | 1 | 14 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.65, 1.53] |
| 10.2.2 No immunosuppression (immune deficiency and cancer) | 1 | 72 | Risk Ratio (M‐H, Random, 95% CI) | 1.14 [0.81, 1.59] |
Comparison 11. Subgroup analysis: pre‐existing condition immunosuppression for the comparison of convalescent plasma vs. standard plasma for outpatients with mild disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 11.1 All‐cause mortality at up to day 28 | 1 | 416 | Risk Ratio (M‐H, Random, 95% CI) | 1.02 [0.06, 16.19] |
| 11.1.1 No immunosuppression (immune deficiency and cancer) | 1 | 416 | Risk Ratio (M‐H, Random, 95% CI) | 1.02 [0.06, 16.19] |
| 11.2 Admission to hospital or death within 28 days | 1 | 414 | Risk Ratio (M‐H, Random, 95% CI) | 0.53 [0.26, 1.08] |
| 11.2.1 No immunosuppression (immune deficiency and cancer) | 1 | 414 | Risk Ratio (M‐H, Random, 95% CI) | 0.53 [0.26, 1.08] |
| 11.3 All initial symptoms resolved (asymptomatic) at up to day 28 | 1 | 416 | Risk Ratio (M‐H, Random, 95% CI) | 1.12 [0.98, 1.27] |
| 11.3.1 No immunosuppression (immune deficiency and cancer) | 1 | 416 | Risk Ratio (M‐H, Random, 95% CI) | 1.12 [0.98, 1.27] |
11.1. Analysis.
Comparison 11: Subgroup analysis: pre‐existing condition immunosuppression for the comparison of convalescent plasma vs. standard plasma for outpatients with mild disease, Outcome 1: All‐cause mortality at up to day 28
11.2. Analysis.
Comparison 11: Subgroup analysis: pre‐existing condition immunosuppression for the comparison of convalescent plasma vs. standard plasma for outpatients with mild disease, Outcome 2: Admission to hospital or death within 28 days
11.3. Analysis.
Comparison 11: Subgroup analysis: pre‐existing condition immunosuppression for the comparison of convalescent plasma vs. standard plasma for outpatients with mild disease, Outcome 3: All initial symptoms resolved (asymptomatic) at up to day 28
Comparison 12. Subgroup analysis: pre‐existing condition diabetes for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 12.1 All‐cause mortality at up to day 28 | 2 | 165 | Risk Ratio (M‐H, Random, 95% CI) | 0.43 [0.17, 1.08] |
| 12.1.1 Diabetes | 2 | 53 | Risk Ratio (M‐H, Random, 95% CI) | 0.47 [0.04, 6.09] |
| 12.1.2 No diabetes | 2 | 112 | Risk Ratio (M‐H, Random, 95% CI) | 0.30 [0.10, 0.89] |
| 12.2 Clinical worsening: need for invasive mechanical ventilation or death | 1 | 921 | Risk Ratio (M‐H, Random, 95% CI) | 1.15 [0.93, 1.43] |
| 12.2.1 Diabetes | 1 | 318 | Risk Ratio (M‐H, Random, 95% CI) | 1.28 [0.89, 1.85] |
| 12.2.2 No diabetes | 1 | 603 | Risk Ratio (M‐H, Random, 95% CI) | 1.09 [0.84, 1.42] |
| 12.3 Clinical improvement: participants discharged from hospital | 1 | 86 | Risk Ratio (M‐H, Random, 95% CI) | 1.13 [0.85, 1.50] |
| 12.3.1 Diabetes | 1 | 21 | Risk Ratio (M‐H, Random, 95% CI) | 0.98 [0.50, 1.96] |
| 12.3.2 No diabetes | 1 | 65 | Risk Ratio (M‐H, Random, 95% CI) | 1.17 [0.86, 1.59] |
Comparison 13. Subgroup analysis: pre‐existing condition diabetes for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 13.1 All‐cause mortality at up to day 28 | 1 | 414 | Risk Ratio (M‐H, Random, 95% CI) | 1.56 [0.07, 34.07] |
| 13.1.1 Diabetes | 1 | 29 | Risk Ratio (M‐H, Random, 95% CI) | 7.33 [0.33, 163.64] |
| 13.1.2 No diabetes | 1 | 385 | Risk Ratio (M‐H, Random, 95% CI) | 0.32 [0.01, 7.76] |
| 13.2 Admission to hospital or death within 28 days | 2 | 511 | Risk Ratio (M‐H, Random, 95% CI) | 0.43 [0.08, 2.44] |
| 13.2.1 Diabetes | 2 | 128 | Risk Ratio (M‐H, Random, 95% CI) | 0.37 [0.01, 24.93] |
| 13.2.2 No diabetes | 1 | 383 | Risk Ratio (M‐H, Random, 95% CI) | 0.48 [0.23, 0.99] |
| 13.3 All initial symptoms resolved (asymptomatic) at up to day 28 | 1 | 414 | Risk Ratio (M‐H, Random, 95% CI) | 1.10 [0.88, 1.37] |
| 13.3.1 Diabetes | 1 | 29 | Risk Ratio (M‐H, Random, 95% CI) | 0.75 [0.35, 1.60] |
| 13.3.2 No diabetes | 1 | 385 | Risk Ratio (M‐H, Random, 95% CI) | 1.13 [0.99, 1.29] |
Comparison 14. Subgroup analysis: pre‐existing condition respiratory disease for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 14.1 All‐cause mortality at up to day 28 | 1 | 95 | Risk Ratio (M‐H, Random, 95% CI) | 0.36 [0.04, 3.24] |
| 14.1.1 Pulmonary disease | 1 | 32 | Risk Ratio (M‐H, Random, 95% CI) | 0.08 [0.00, 1.39] |
| 14.1.2 No pulmonary disease | 1 | 63 | Risk Ratio (M‐H, Random, 95% CI) | 0.77 [0.30, 1.97] |
| 14.2 Clinical worsening: need for invasive mechanical ventilation or death | 1 | 921 | Risk Ratio (M‐H, Random, 95% CI) | 1.16 [0.94, 1.43] |
| 14.2.1 Respiratory disease | 1 | 224 | Risk Ratio (M‐H, Random, 95% CI) | 1.24 [0.81, 1.88] |
| 14.2.2 No respiratory disease | 1 | 697 | Risk Ratio (M‐H, Random, 95% CI) | 1.13 [0.88, 1.45] |
| 14.3 Clinical improvement: participants discharged from hospital | 1 | 86 | Risk Ratio (M‐H, Random, 95% CI) | 1.11 [0.83, 1.49] |
| 14.3.1 Pulmonary disease | 1 | 23 | Risk Ratio (M‐H, Random, 95% CI) | 1.31 [0.78, 2.19] |
| 14.3.2 No pulmonary disease | 1 | 63 | Risk Ratio (M‐H, Random, 95% CI) | 1.03 [0.73, 1.46] |
Comparison 15. Subgroup analysis: pre‐existing condition hypertension for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 15.1 All‐cause mortality at up to day 28 | 2 | 165 | Risk Ratio (M‐H, Random, 95% CI) | 0.47 [0.22, 1.01] |
| 15.1.1 Hypertension | 2 | 75 | Risk Ratio (M‐H, Random, 95% CI) | 0.27 [0.04, 1.87] |
| 15.1.2 No hypertension | 2 | 90 | Risk Ratio (M‐H, Random, 95% CI) | 0.56 [0.22, 1.45] |
| 15.2 Clinical improvement: participants discharged from hospital | 1 | 86 | Risk Ratio (M‐H, Random, 95% CI) | 1.13 [0.78, 1.62] |
| 15.2.1 Hypertension | 1 | 22 | Risk Ratio (M‐H, Random, 95% CI) | 1.50 [0.82, 2.75] |
| 15.2.2 No hypertension | 1 | 64 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.72, 1.39] |
Comparison 16. Subgroup analysis: age of participants for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 16.1 All‐cause mortality at up to day 28 | 2 | 165 | Risk Ratio (M‐H, Random, 95% CI) | 0.45 [0.22, 0.94] |
| 16.1.1 < 65 years | 1 | 45 | Risk Ratio (M‐H, Random, 95% CI) | 0.24 [0.03, 1.98] |
| 16.1.2 ≥ 65 years | 1 | 41 | Risk Ratio (M‐H, Random, 95% CI) | 0.75 [0.28, 1.98] |
| 16.1.3 ≤ 60 years | 1 | 33 | Risk Ratio (M‐H, Random, 95% CI) | 0.15 [0.01, 2.72] |
| 16.1.4 > 60 years | 1 | 46 | Risk Ratio (M‐H, Random, 95% CI) | 0.26 [0.06, 1.13] |
| 16.2 Clinical worsening: need for invasive mechanical ventilation, or death | 2 | 12042 | Risk Ratio (M‐H, Random, 95% CI) | 0.99 [0.87, 1.13] |
| 16.2.1 < 60 years | 1 | 272 | Risk Ratio (M‐H, Random, 95% CI) | 1.23 [0.64, 2.37] |
| 16.2.2 ≥ 60 years | 1 | 649 | Risk Ratio (M‐H, Random, 95% CI) | 1.14 [0.92, 1.42] |
| 16.2.3 < 70 years | 1 | 6972 | Risk Ratio (M‐H, Random, 95% CI) | 1.01 [0.92, 1.11] |
| 16.2.4 ≥ 70 years | 1 | 4149 | Risk Ratio (M‐H, Random, 95% CI) | 0.88 [0.82, 0.95] |
| 16.3 Clinical improvement: participants discharged from hospital | 2 | 11644 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.98, 1.02] |
| 16.3.1 < 65 years | 1 | 45 | Risk Ratio (M‐H, Random, 95% CI) | 1.01 [0.74, 1.38] |
| 16.3.2 ≥ 65 years | 1 | 41 | Risk Ratio (M‐H, Random, 95% CI) | 1.24 [0.74, 2.09] |
| 16.3.3 < 70 years | 1 | 7453 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.97, 1.02] |
| 16.3.4 > 70 years | 1 | 4105 | Risk Ratio (M‐H, Random, 95% CI) | 1.02 [0.96, 1.08] |
Comparison 17. Subgroup analysis: age of participants for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 17.1 All‐cause mortality at up to day 28 | 1 | 416 | Risk Ratio (M‐H, Random, 95% CI) | 1.06 [0.11, 10.46] |
| 17.1.1 < 65 years | 1 | 303 | Risk Ratio (M‐H, Random, 95% CI) | 0.33 [0.01, 7.96] |
| 17.1.2 ≥ 65 years | 1 | 113 | Risk Ratio (M‐H, Random, 95% CI) | 3.39 [0.14, 81.46] |
| 17.2 Admission to hospital or death within 28 days | 2 | 1792 | Risk Ratio (M‐H, Random, 95% CI) | 0.80 [0.29, 2.26] |
| 17.2.1 < 65 years | 2 | 1403 | Risk Ratio (M‐H, Random, 95% CI) | 0.38 [0.21, 0.67] |
| 17.2.2 ≥ 65 years | 2 | 389 | Risk Ratio (M‐H, Random, 95% CI) | 1.89 [0.52, 6.81] |
| 17.3 All initial symptoms resolved (asymptomatic) at up to day 28 | 1 | 416 | Risk Ratio (M‐H, Random, 95% CI) | 1.11 [0.98, 1.26] |
| 17.3.1 < 65 years | 1 | 303 | Risk Ratio (M‐H, Random, 95% CI) | 1.16 [0.99, 1.35] |
| 17.3.2 ≥ 65 years | 1 | 113 | Risk Ratio (M‐H, Random, 95% CI) | 1.03 [0.83, 1.28] |
Comparison 18. Subgroup analysis: sex of participants for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 18.1 All‐cause mortality at up to day 28 | 3 | 257 | Risk Ratio (M‐H, Random, 95% CI) | 0.36 [0.17, 0.77] |
| 18.1.1 Female | 2 | 76 | Risk Ratio (M‐H, Random, 95% CI) | 0.79 [0.10, 6.24] |
| 18.1.2 Male | 3 | 181 | Risk Ratio (M‐H, Random, 95% CI) | 0.27 [0.15, 0.48] |
| 18.2 Clinical worsening: need for invasive mechanical ventilation or death | 3 | 11332 | Risk Ratio (M‐H, Random, 95% CI) | 1.10 [0.88, 1.36] |
| 18.2.1 Female | 3 | 4481 | Risk Ratio (M‐H, Random, 95% CI) | 0.98 [0.88, 1.08] |
| 18.2.2 Male | 3 | 6851 | Risk Ratio (M‐H, Random, 95% CI) | 1.25 [1.01, 1.54] |
| 18.3 Clinical improvement: participants discharged from hospital | 2 | 11644 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.97, 1.02] |
| 18.3.1 Female | 2 | 4152 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.96, 1.05] |
| 18.3.2 Male | 2 | 7492 | Risk Ratio (M‐H, Random, 95% CI) | 0.99 [0.96, 1.02] |
Comparison 19. Subgroup analysis: sex of participants for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 19.1 All‐cause mortality at up to day 28 | 1 | 416 | Risk Ratio (M‐H, Random, 95% CI) | 0.97 [0.06, 15.36] |
| 19.1.1 Female | 1 | 95 | Risk Ratio (M‐H, Random, 95% CI) | Not estimable |
| 19.1.2 Male | 1 | 321 | Risk Ratio (M‐H, Random, 95% CI) | 0.97 [0.06, 15.36] |
| 19.2 Admission to hospital or death within 28 days | 2 | 1595 | Risk Ratio (M‐H, Random, 95% CI) | 0.50 [0.32, 0.77] |
| 19.2.1 Female | 2 | 770 | Risk Ratio (M‐H, Random, 95% CI) | 0.42 [0.21, 0.82] |
| 19.2.2 Male | 2 | 825 | Risk Ratio (M‐H, Random, 95% CI) | 0.56 [0.31, 1.02] |
| 19.3 All initial symptoms resolved (asymptomatic) at up to day 28 | 1 | 416 | Risk Ratio (M‐H, Random, 95% CI) | 1.11 [0.98, 1.26] |
| 19.3.1 Female | 1 | 95 | Risk Ratio (M‐H, Random, 95% CI) | 1.06 [0.78, 1.44] |
| 19.3.2 Male | 1 | 321 | Risk Ratio (M‐H, Random, 95% CI) | 1.12 [0.98, 1.29] |
Comparison 20. Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus placebo or standard care alone for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 20.1 All‐cause mortality at up to day 28 | 20 | 18100 | Risk Ratio (M‐H, Random, 95% CI) | 0.93 [0.82, 1.05] |
| 20.1.1 High‐income countries | 11 | 16530 | Risk Ratio (M‐H, Random, 95% CI) | 0.92 [0.77, 1.10] |
| 20.1.2 Low‐ to middle‐income countries | 9 | 1570 | Risk Ratio (M‐H, Random, 95% CI) | 0.93 [0.74, 1.16] |
| 20.2 Clinical worsening: need for invasive mechanical ventilation, or death at up to day 28 | 5 | 13556 | Risk Ratio (M‐H, Random, 95% CI) | 1.02 [0.96, 1.09] |
| 20.2.1 High‐income countries | 3 | 12759 | Risk Ratio (M‐H, Random, 95% CI) | 1.04 [0.93, 1.16] |
| 20.2.2 Low‐ to middle‐income countries | 2 | 797 | Risk Ratio (M‐H, Random, 95% CI) | 0.99 [0.74, 1.33] |
| 20.3 Clinical improvement: participants discharged from hospital | 6 | 12721 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.97, 1.02] |
| 20.3.1 High‐income countries | 2 | 12041 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.97, 1.02] |
| 20.3.2 Low‐ to middle‐income countries | 4 | 680 | Risk Ratio (M‐H, Random, 95% CI) | 1.02 [0.91, 1.13] |
| 20.4 Grades 3 and 4 adverse events | 5 | 1471 | Risk Ratio (M‐H, Random, 95% CI) | 1.18 [0.77, 1.80] |
| 20.4.1 High‐income countries | 2 | 527 | Risk Ratio (M‐H, Random, 95% CI) | 3.06 [1.00, 9.36] |
| 20.4.2 Low‐ to middle‐income countries | 3 | 944 | Risk Ratio (M‐H, Random, 95% CI) | 1.08 [0.83, 1.40] |
| 20.5 Serious adverse events | 5 | 3980 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.81, 1.24] |
| 20.5.1 High‐income countries | 4 | 3647 | Risk Ratio (M‐H, Random, 95% CI) | 0.94 [0.74, 1.19] |
| 20.5.2 Low‐ to middle‐income countries | 1 | 333 | Risk Ratio (M‐H, Random, 95% CI) | 1.31 [0.82, 2.09] |
Comparison 21. Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus standard plasma for individuals with moderate to severe disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 21.1 All‐cause mortality at up to day 28 | 3 | 261 | Risk Ratio (M‐H, Random, 95% CI) | 0.98 [0.53, 1.83] |
| 21.1.1 High‐income countries | 1 | 74 | Risk Ratio (M‐H, Random, 95% CI) | 0.89 [0.34, 2.31] |
| 21.1.2 Low‐ to middle‐income countries | 2 | 187 | Risk Ratio (M‐H, Random, 95% CI) | 1.15 [0.41, 3.21] |
| 21.2 Clinical worsening: need for invasive mechanical ventilation or death at up to day 28 | 1 | 34 | Risk Ratio (M‐H, Random, 95% CI) | 5.59 [0.29, 108.38] |
| 21.2.1 High‐income countries | 1 | 34 | Risk Ratio (M‐H, Random, 95% CI) | 5.59 [0.29, 108.38] |
| 21.3 Serious adverse events | 2 | 108 | Risk Ratio (M‐H, Random, 95% CI) | 1.14 [0.48, 2.71] |
| 21.3.1 High‐income countries | 2 | 108 | Risk Ratio (M‐H, Random, 95% CI) | 1.14 [0.48, 2.71] |
Comparison 22. Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus placebo or standard care alone for outpatients with mild disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 22.1 All‐cause mortality at up tp day 28 | 2 | 536 | Risk Ratio (M‐H, Random, 95% CI) | 0.40 [0.09, 1.74] |
| 22.1.1 High‐income countries | 1 | 376 | Risk Ratio (M‐H, Random, 95% CI) | 0.20 [0.01, 4.14] |
| 22.1.2 Low‐ to middle‐income countries | 1 | 160 | Risk Ratio (M‐H, Random, 95% CI) | 0.50 [0.09, 2.65] |
| 22.2 Grades 3 and 4 adverse events | 1 | 376 | Risk Ratio (M‐H, Random, 95% CI) | 1.29 [0.75, 2.19] |
| 22.2.1 High‐income countries | 1 | 376 | Risk Ratio (M‐H, Random, 95% CI) | 1.29 [0.75, 2.19] |
| 22.3 Serious adverse events | 1 | 376 | Risk Ratio (M‐H, Random, 95% CI) | 1.14 [0.66, 1.94] |
| 22.3.1 High‐income countries | 1 | 376 | Risk Ratio (M‐H, Random, 95% CI) | 1.14 [0.66, 1.94] |
Comparison 23. Subgroup analysis: income levels of countries for the comparison of convalescent plasma versus standard plasma for outpatients with mild disease.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 23.1 All‐cause mortality at up to day 28 | 2 | 1587 | Risk Ratio (M‐H, Random, 95% CI) | 0.62 [0.03, 12.54] |
| 23.1.1 High‐income countries | 2 | 1587 | Risk Ratio (M‐H, Random, 95% CI) | 0.62 [0.03, 12.54] |
Characteristics of studies
Characteristics of included studies [ordered by study ID]
Agarwal 2020.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Alemany 2022.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
AlQahtani 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Avendano‐Sola 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Bajpai 2020.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Baldeon 2022.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Bar 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Begin 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Beltran Gonzalez 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Bennett‐Guerrero 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
CoV‐Early.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
De Santis 2022.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Devos 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Estcourt 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Gharbharan 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Hamdy Salman 2020.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Holm 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Horby 2021b.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Kirenga 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Koerper 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Korley 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Li 2020.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Libster 2020.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Menichetti 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04421404.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Ortigoza 2022.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
O’Donnell 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Pouladzadeh 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Ray 2022.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Sekine 2021.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Simonovich 2020.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Sullivan 2022.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
Van den Berg 2022.
| Study characteristics | |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
AE: adverse event; ALT: alanine transaminase; ARDS: acute respiratory distress syndrome; AST: aspartate transaminase; ATS: American Thoracic Society; AU: astronomical unit; BAL: bronchoalveolar lavage; BMI: body mass index; CAP: community‐acquired pneumonia; CKD: chronic kidney disease; CoI: conflict of interest; COPD: chronic obstructive pulmonary disease; CP: convalescent plasma; CPK: creatine phosphokinase; CRP: c‐reactive protein; CT: computed tomography; ECMO: extracorporeal membrane oxygenation; ED: emergency department; ELISA: enzyme‐linked immunosorbent assay; EMA: European Medicines Agency; FDA: US Food and Drug Administration; FiO2: fractional inspired oxygen; GFR: glomerular filtration rate; HBV/HCV: hepatitis B/C; HLA: human leukocyte antigen; HNA: human neutrophil antigens; ICU: intensive care unit; IDSA: Infectious Diseases Society of America; IgA (B/G/M): immunoglobulin A (B/G/M);IL‐6: interleukin‐6; IMV: invasive mechanical ventilation;IQR: interquartile range; IV: intravenous; IVIG: intravenous immunoglobulin; LAR: legal authorised representative; LDH: lactate dehydrogenase; LMWH: low molecular‐weight heparin; MAP: mean arterial pressure; NR: not reported; NYHA: New York Heart Association; PaO2: arterial blood oxygen partial pressure; PCR: polymerase chain reaction; PRNT: plaque reduction neutralisation test; QoL: quality of life; RCT: randomised controlled trial; RNA: ribonucleic acid; RT‐PCR: reverse transcription polymerase chain reaction; SAE: serious adverse event; SARS: severe acute respiratory syndrome; SC: subcutaneous; SD: standard deviation; SC: standard care; SOFA: sequential organ failure assessment; SP: standard plasma; SpO2: peripheral capillary oxygen saturation; TACO: transfusion‐associated circulatory overload; TAD: transfusion‐associated dyspnoea; TB: tuberculosis; TRALI: transfusion‐related acute lung injury; UIP: usual interstitial pneumonia; ULN: upper limit of normal; WHO: World Health Organization
Characteristics of excluded studies [ordered by study ID]
| Study | Reason for exclusion |
|---|---|
| Abdullah 2020 | Single arm study; fewer than 500 participants received convalescent plasma |
| Abolghasemi 2020 | Non‐randomised study ; fewer than 500 participants received convalescent plasma |
| Ahn 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Allahyari 2021 | Non‐randomised controlled study |
| Anderson 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Baklaushev 2020 | Controlled study, probably not truly randomised |
| Balcells 2020 | Performed another intervention comparison (early vs deferred plasma) |
| Bao 2020b | Single‐arm study; not pre‐registered in a clinical study registry |
| Bobek 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Bradfute 2020 | Single arm study; fewer than 500 participants received convalescent plasma |
| Brasil Ministerio 2020 | Standard operating procedure |
| Budhai 2020 | Feasibility of plasma collection only |
| Cantore 2020 | Single‐arm study compared to published cases; not pre‐registered in a clinical study registry |
| Cao 2020a | Ineligible intervention |
| Chen 2020b | Ineligible intervention |
| Chen 2020c | Ineligible intervention |
| ChiCTR2000029850 | Single arm study; fewer than 500 participants will receive convalescent plasma |
| ChiCTR2000030039 | Single arm study; fewer than 500 participants will receive convalescent plasma |
| ChiCTR2000030312 | Study cancelled before starting recruitment |
| ChiCTR2000030381 | Study cancelled before starting recruitment |
| ChiCTR2000030442 | Study cancelled before starting recruitment |
| ChiCTR2000031501 | Single arm study; fewer than 500 participants will receive convalescent plasma |
| ChiCTR2000033798 | Single arm study; fewer than 500 participants will receive convalescent plasma |
| Clark 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| CTRI/2020/04/024804 | Single arm study; fewer than 500 participants will receive convalescent plasma |
| CTRI/2020/08/027285 | Single arm study; fewer than 500 participants will receive convalescent plasma |
| CTRI/2020/09/027903 | Ineligible intervention with hyperimmune immunoglobulin |
| CTRI/2020/10/028547 | Ineligible intervention |
| Çınar 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| de Assis 2020 | Ineligible indication |
| Díez 2020 | Ineligible intervention |
| Donato 2020 | Single arm study; fewer than 500 participants received convalescent plasma |
| Duan 2020 | Single arm study; fewer than 500 participants received convalescent plasma |
| Dulipsingh 2020 | Single‐arm study; fewer than 500 participants received convalescent plasma |
| Enzmann 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Erkurt 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| EUCTR2020‐005979‐12‐GR | Ineligible intervention: hyperimmune immunoglobulin |
| Fan 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Figlerowicz 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Franchini 2020 | Standard operating procedure |
| Gaborit 2021 | Ineligible intervention: hyperimmune immunoglobulin |
| Grisolia 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Hashim 2020 | Feasibility of plasma collection only |
| Hu 2020 | Ineligible intervention |
| Ibrahim 2020 | Single‐arm study; fewer than 500 participants received convalescent plasma |
| Im 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| IRCT20151228025732N53 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| IRCT20200406046968N2 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| IRCT20200414047072N1 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| IRCT20200416047099N1 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| IRCT20200508047346N1 | Ineligible intervention: hyperimmune immunoglobulin |
| IRCT20200525047562N1 | Non‐randomised study with fewer than 500 participants receiving convalescent plasma |
| ISRCTN86534580 | Ineligible intervention |
| Jamous 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Jiang 2020a | Single‐arm study; not pre‐registered in a clinical study registry |
| Jiang 2020b | Ineligible intervention |
| Jin 2020 | Single‐arm study; fewer than 500 participants received convalescent plasma |
| Joyner 2020 | Expanded‐access study |
| jRCT2031200174 | Ineligible intervention with hyperimmune immunoglobulin |
| Karatas 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Kong 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Lin 2020 | Ineligible intervention |
| Liu 2020a | Single‐arm study; fewer than 500 participants received convalescent plasma |
| Liu 2020b | Single‐arm study; not pre‐registered in a clinical study registry |
| Madariaga 2020 | Single‐arm study; fewer than 500 participants received convalescent plasma |
| Martinez‐Resendez 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| McCuddy 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Ministerio de Salud 2020 | Standard operating procedure |
| Mira 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| NCT04261426 | Ineligible intervention |
| NCT04264858 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04292340 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04321421 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04323800 | Ineligible participant population (participants exposed to COVID‐19) |
| NCT04325672 | Study cancelled before starting recruitment |
| NCT04327349 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04332380 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04333355 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04338360 | Expanded‐access study |
| NCT04344015 | Feasibility of plasma collection only |
| NCT04344379 | Ineligible intervention |
| NCT04344977 | Feasibility of plasma collection only |
| NCT04345679 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04346589 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04347681 | Controlled, non‐randomised study with fewer than 500 participants receiving convalescent plasma |
| NCT04348877 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04350580 | Ineligible intervention |
| NCT04352751 | Single‐arm study |
| NCT04353206 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04354831 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04355897 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04356482 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04358211 | Expanded‐access study |
| NCT04360278 | Feasibility of plasma collection only |
| NCT04360486 | Expanded‐access studies |
| NCT04363034 | Expanded‐access studies |
| NCT04365439 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04366245 | Ineligible intervention: hyperimmune immunoglobulin |
| NCT04368013 | Ineligible intervention |
| NCT04372368 | Expanded‐access study |
| NCT04374370 | Expanded‐access studies |
| NCT04374565 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04376034 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04377568 | This study was withdrawn |
| NCT04377672 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04383548 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04384497 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04384588 | Controlled, non‐randomised study with fewer than 500 participants receiving convalescent plasma or hyperimmune immunoglobulin |
| NCT04388527 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04389710 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04389944 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04390178 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04392232 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04393727 | Terminated in November 2020 (study was stopped because the sponsor was changed and a new study on convalescent plasma sponsored by the Italian Medicines Agency (AIFA) was started in Italy) |
| NCT04395170 | Ineligible intervention: hyperimmune immunoglobulin |
| NCT04397523 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04407208 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04408040 | Non‐randomised study |
| NCT04408209 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04411602 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04412486 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04418531 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04420988 | Expanded‐access studies |
| NCT04432103 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04432272 | Non‐randomised study with fewer than 500 participants receiving convalescent plasma |
| NCT04438694 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04445207 | Expanded‐access studies |
| NCT04458363 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04462848 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04463823 | Prospective case‐only study |
| NCT04467151 | Study withdrawn |
| NCT04468958 | Ineligible intervention: hyperimmune immunoglobulin |
| NCT04469179 | Ineligible intervention: hyperimmune immunoglobulin |
| NCT04471051 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04472572 | Expanded‐access studies |
| NCT04474340 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04476888 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04492501 | Non‐randomised factorial design |
| NCT04497779 | Prospective cohort study |
| NCT04502472 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04513158 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04514302 | Ineligible intervention: hyperimmune immunoglobulin |
| NCT04516954 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04524507 | Ineligible comparison: high vs low titre |
| NCT04535063 | Single‐arm study; fewer than 500 participants received convalescent plasma |
| NCT04545047 | Observational study |
| NCT04546581 | Ineligible intervention: hyperimmune immunoglobulin |
| NCT04554992 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04555109 | Study of plasma donors |
| NCT04555148 | Ineligible intervention: hyperimmune immunoblogulin |
| NCT04565197 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04569188 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04570982 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04573855 | Ineligible intervention: hyperimmune immunoglobulin |
| NCT04593940 | Completed platform trial without a convalescent plasma arm |
| NCT04610502 | Ineligible intervention: hyperimmune immunoglobulin |
| NCT04614012 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04616976 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04622826 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04638634 | Study on pharmacokinetics |
| NCT04642014 | Single arm study |
| NCT04644198 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| NCT04661839 | Pharmacokinetics study |
| NCT04669990 | Prospective case‐only study |
| NCT04721236 | Single‐arm hyperimmune immunoglobulin study with fewer than 500 participants |
| Niu 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Olivares‐Gazca 2020 | Single‐arm study; fewer than 500 participants received convalescent plasma |
| Pei 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Peng 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| PER‐031‐20 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| Perotti 2020 | Single‐arm study; fewer than 500 participants received convalescent plasma |
| Qiu 2020 | No use of convalescent plasma. Reporting on generalised collection of information about COVID‐19 infection relating to aetiology, pathology, symptoms, clinical presentation and some generalised pharmacological treatment methods. The papers do not contain patient data that we required. Article translated by Rujan Shrestha and Ya‐Ying Wang via Cochrane TaskExchange |
| Rasheed 2020 | Controlled study, probably not truly randomised |
| RBR‐4vm3yy | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| RBR‐5r8gv8p | This study was suspended |
| Robbiani 2020 | Ineligible intervention |
| RPCEC00000323 | Single‐arm study; fewer than 500 participants will receive convalescent plasma |
| Salazar 2020a | Single‐arm study; fewer than 500 participants received convalescent plasma |
| Salazar 2020b | Single‐arm study; not pre‐registered in a clinical study registry |
| Shen 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Shi 2020 | Ineligible intervention |
| Soleimani 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Taher 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Tan 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Tu 2020 | No use of convalescent plasma. Reporting on generalised collection of information about COVID‐19 infection relating to aetiology, pathology, symptoms, clinical presentation and some generalised pharmacological treatment methods. The papers do not contain patient data that we required. Article translated by Rujan Shrestha and Ya‐Ying Wang via Cochrane TaskExchange |
| Wang 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Wright 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Xia 2020 | Single‐arm study; fewer than 500 participants received convalescent plasma |
| Xie 2020 | Ineligible intervention |
| Xu 2020b | Single‐arm study; not pre‐registered in a clinical study registry |
| Yang 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Ye 2020 | Single‐arm study; not pre‐registered in a clinical study registry |
| Zeng 2020 | Single‐arm study; fewer than 500 participants received convalescent plasma |
| Zhang 2020a | Single‐arm study; not pre‐registered in a clinical study registry |
| Zhang 2020b | Single‐arm study; not pre‐registered in a clinical study registry |
| Zhang 2020c | Single‐arm study; not pre‐registered in a clinical study registry |
Characteristics of studies awaiting classification [ordered by study ID]
CTRI/2020/05/025299.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
CTRI/2020/05/025328.
| Methods |
|
| Participants | Inclusion criteria
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
CTRI/2020/06/025803.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
EUCTR2020‐001860‐27‐GB.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
IRCT20120215009014N353.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
IRCT20150808023559N21.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
IRCT20200404046948N1.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
IRCT20200413047056N1.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
IRCT20200501047258N1.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
IRCT20200503047281N1.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
IRCT20201004048922N1.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04315948.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes | Recruitment status: active, not recruiting Prospective completion date: March 2023 Sponsor/funding: Institut National de la Santé et de la Recherche Médicale, France |
NCT04332835.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04345991.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes | during hospital stay
|
| Notes |
|
NCT04358783.
| Methods |
|
| Participants |
Patients with SARS‐CoV‐2 PCR‐confirmed infection with pulmonary infiltrates and hypoxaemia will be screened and invited to participate |
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04361253.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04362176.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04374526.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04385199.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04405310.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes | Recruitment status: completed Prospective completion date: 20 June 2020 Sponsor/funding: Grupo Mexicano para el Estudio de la Medicina Intensiva, Hospital General Naval de Alta Especialidad ‐ Escuela Medico Naval, National Institute of Pediatrics, Mexico, Instituto Nacional de Enfermedades Respiratorias |
NCT04425915.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04428021.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04442958.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04468009.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04497324.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04501978.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04521309.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04539275.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04542967.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04547127.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04649879.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04681430.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
NCT04801940.
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Notes |
|
AE: adverse event; ALT: alanine aminotransferase; ARDS: acute respiratory distress syndrome; AST: aspartate aminotransferase; BAL: bronchoalveolar lavage; CKD: chronic kidney disease; COPD: chronic obstructive pulmonary disease; CP: convalescent plasma; CPAP: continuous positive airway pressure; CRP: C‐reactive protein; CT: computed tomography; DIC: disseminated intravascular coagulation; ECMO: extracorporeal membrane oxygenation; ED: Emergency Department; ELISA: enzyme‐linked immunosorbent assay; FFP: fresh frozen plasma; FiO2: fractional inspired oxygen; GFR: glomerular filtration rate; ICU: intensive care unit; IgA (B/G/M): immunoglobulin A (B/G/M); IL‐6: interleukin‐6; IMV: invasive mechanical ventilation; IQR: interquartile range; IV: intravenous;IVIG: immunoglobulin; LDH: lactate dehydrogenase; MAP: mean arterial pressure; MRC: Medical Research Council; NEWS: National Early Warning Score; NP: nasopharyngeal; NR: not reported; PaO2: arterial blood oxygen partial pressure; PCR: polymerase chain reaction; PTSD: post‐traumatic stress disorder; QoL: quality of life; RCT: randomised controlled trial; RRT: renal replacement therapy; RT‐PCR: reverse transcription polymerase chain reaction; SAE: serious adverse event; SaO2: oxygen saturation of arterial blook; SC: standard care; SOFA: sequential organ failure assessment; TACO: transfusion‐associated circulatory overload; TAD: transfusion‐associated dyspnoea; TB: tuberculosis; TRALI: transfusion‐related acute lung injury; TTP: thrombotic thrombocytopenia;ULN: upper limit of normal; VA: Veterans Association; WHO: World Health Organization
Characteristics of ongoing studies [ordered by study ID]
ChiCTR2000030010.
| Study name | A randomized, double‐blind, parallel‐controlled, trial to evaluate the efficacy and safety of anti‐SARS‐CoV‐2 virus inactivated plasma in the treatment of severe novel coronavirus pneumonia patients (COVID‐19) |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 19 February 2020 |
| Contact information | Liu Ying, Wuhan Jinyintan Hospital (Wuhan Infectious Diseases Hospital), 1 Yintan Road, Dongxihu District, Wuhan, Hubei, China, 430023, whsjytyy_gcp@163.com Zhang Dingyu, 1 Yintan Road, Dongxihu District, Wuhan, Hubei, China, 430023, 1813886398@qq.com |
| Notes |
|
ChiCTR2000030179.
| Study name | Experimental study of novel coronavirus pneumonia rehabilitation plasma therapy severe novel coronavirus pneumonia (COVID‐19) |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 24 February 2020 |
| Contact information | Liu Wei, The First Affiliated Hospital of Nanchang University, 17 Yongwai Main Street, Nanchang, Jiangxi, China, 330006, cdyfyliuwei@163.com Le Aiping, 17 Yongwai Main Street, Nanchang, Jiangxi, China, 330006, leaiping@126.com |
| Notes |
|
ChiCTR2000030627.
| Study name | Study on the application of convalescent plasma therapy in severe COVID‐19 |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 1 February 2020 |
| Contact information | Guojun Zhang, The First Affiliated Hospital of Zhengzhou University, 1 Jianshe Road East, Zhengzhou, He'nan, China, zlgj‐001@126.com Guojun Zhang, 1 Jianshe Road East, Zhengzhou, He'nan, China, zlgj‐001@126.com |
| Notes |
|
ChiCTR2000030702.
| Study name | Convalescent plasma for the treatment of common COVID‐19: a prospective RCT |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 15 February 2020 |
| Contact information | Liu Zhong, Institute of Blood Transfusion, Chinese Academy of Medical Sciences, 26 Huacai Road, Chenghua District, Chengdu, Sichuan, China, 610000, Liuz@ibt.pumc.edu.cn Cao Bin, 2 Yinghua Street East, Chaoyang District, Beijing, China, 100029, caobin_ben@163.com |
| Notes |
|
ChiCTR2000030929.
| Study name | A randomized, double‐blind, parallel‐controlled trial to evaluate the efficacy and safety of anti‐SARS‐CoV‐2 virus inactivated plasma in the treatment of severe novel coronavirus pneumonia (COVID‐19) |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 17 March 2020 |
| Contact information | Lianghao Zhang, 11443556@qq.com, Sinopharm Wuhan Blood Products Co., Ltd, 1 Golden Industrial Park Road, Zhengdian, Jiangxia District, Wuhan, Hubei, China |
| Notes |
|
CTRI/2020/04/024915.
| Study name | A phase II, open label, randomized controlled trial to assess the safety and efficacy of convalescent plasma to limit COVID‐19 associated complications |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 9 May 2020 |
| Contact information | Corresponding Author: Name: Dr Sangeeta Pathak Affiliation: Blood Bank, Max Super Speciality hospital, Saket (A unit of DevkiDevi Foundation) Full Address: Max Super Speciality Hospital (Devki Devi Foundation), East Block,Blood Bank, 2, Press enclave Road, Saket New DelhiNew DelhiDELHI110017India Email: sangeeta.pathak@maxhealthcare.com |
| Notes |
|
CTRI/2020/05/025346.
| Study name | A phase II, open label, randomized controlled trial to assess the safety and efficacy of convalescent plasma in severe COVID‐19 patients |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 1 June 2020 |
| Contact information | Corresponding author Name: Dr S Anbuselvi Mattuvar Kuzhali Affiliation: Madras Medical College Full Address: Institute of Physiology and Experimental Medicine, Madras MedicalCollege, Chennai 4B, third floor, KGEYES HYACINTH, LB road, Kamaraj Nagar, Thiruvanmiyur, Chennai, Chennai, TAMIL NADU 600003 India Email: dranbuselvimk@gmail.com |
| Notes |
|
CTRI/2020/06/026123.
| Study name | A phase II, open label, randomized controlled trial to assess the safety and efficacy of convalescent plasma in severe COVID‐19 |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 25 June 2020 |
| Contact information | Corresponding Author Name: Dr Sushant Meshram Affiliation: Government Medical College and Hospital, Nagpur Full Address: Dept of Pulmonary Medicine, 4th floor, Super Specialty Hospital, Government Medical College Hospital, Nagpur. Nagpur, MAHARASHTRA 440009 India Email: drsushant.in@gmail.com |
| Notes | Recruitment status: not yet recruiting
|
EUCTR2020‐001632‐10.
| Study name | A randomized open label phase‐II clinical trial with or without infusion of plasma from subjects after convalescence of SARS‐CoV‐2 infection in high‐risk patients with confirmed severe SARS‐CoV‐2 disease |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 04 May 2020 |
| Contact information | Corresponding author Name: Prof. Dr. Carsten Müller‐Tidow Affiliation: University Hospital Heidelberg; Dpt. of Internal Medicine V Hematology, Oncology and Rheumatology Full Address: Im Neuenheimer Feld 410, 69120 Heidelberg, Germany Email: Carsten.Mueller‐Tidow@med.uni‐heidelberg.de |
| Notes |
|
EUCTR2020‐001936‐86.
| Study name | A prospective, randomized, open label phase 2 clinical trial to evaluate superiority of anti‐SARS‐CoV‐2 convalescent plasma versus standard‐of‐care in hospitalized patients with mild COVID‐19 |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 19 October 2020 |
| Contact information | Corresponding author/contact Name: Institute of Transfusion Medicine, Hannover Medical School Affiliation: ‐ Full Address: Carl‐Neuberg‐Str. 1, Hannover, 30625, Germany Email: NR |
| Notes |
|
EUCTR2020‐002122‐82.
| Study name | Prospective open‐label randomized controlled phase 2b clinical study in parallel groups for the assessment of efficacy and safety of immune therapy with COVID‐19 convalescent plasma plus standard treatment vs. standard treatment alone of subjects with severe COVID‐19 |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 12 June 2020 |
| Contact information | Corresponding Author Name: Holger Hackstein Affiliation: Universitätsklinikum Erlangen Full Address: Universitätsklinikum Erlangen, Transfusionsmedizinische Abteilung, Krankenhausstr. 12, Erlangen 91054, Germany Email: holger.hackstein@uk‐erlangen.de |
| Notes |
|
EUCTR2020‐005410‐18.
| Study name | Multicentre, randomized, double‐blind, placebo‐controlled, non‐commercial clinical trial to evaluate the efficacy and safety of specific anti‐SARS‐CoV‐2 immunoglobulin in the treatment of COVID‐19 |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 16 November 2020 |
| Contact information | Samodzielny Publiczny Szpital Kliniczny Nr 1 w Lublinie Samodzielny Publiczny Zakład Opieki Zdrowotnej Krzysztof Tomasiewicz ul. Stanisława Staszica 16 Lublin20‐081 Poland 00488153 49 414564 00488153 49 410 krzysztoftomasiewicz@umlub.pl |
| Notes |
|
ISRCTN49832318.
| Study name | SURCOVID trial: a randomized controlled trial using convalescent plasma early during moderate COVID‐19 disease course in Suriname |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 1 June 2021 |
| Contact information | Name: Rosita Bihariesingh‐Sanchit Address: Kwattaweg 639 ‐ Wanica Suriname Telephone:+596 8753150 Email:rbihariesingh@azp.sr |
| Notes |
|
jRCTs031200374.
| Study name | An open‐label, randomized, controlled trial to evaluate the efficacy of convalescent plasma therapy for COVID‐19 (COVIPLA‐RCT) |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes | Primary study outcome: time‐weighted mean change in the amount of SARS‐CoV‐2 virus in NP swabs from (days 0, 3, 5) Secondary study outcomes
|
| Starting date | 25 February 2021 |
| Contact information | Corresponding author
|
| Notes |
|
NCT04333251.
| Study name | Evaluating convalescent plasma to decrease coronavirus associated complications. A phase I study comparing the efficacy and safety of high‐titer anti‐SARS‐CoV‐2 plasma versus best supportive care in hospitalized patients with interstitial pneumonia due to COVID‐19 |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 1 April 2020 |
| Contact information | NR |
| Notes |
|
NCT04345289.
| Study name | Efficacy and safety of novel treatment options for adults with COVID‐19 pneumonia (CCAP) |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 20 April 2020 |
| Contact information | Thomas Benfield, MD, DMSc: thomas.lars.benfield@regionh.dk |
| Notes |
|
NCT04372979.
| Study name | Evaluation of efficacy of COVID‐19 convalescent plasma versus standard plasma in the early care of COVID‐19 patients hospitalized outside intensive care units |
| Methods |
|
| Participants | Inclusion criteria
Exclusion criteria
|
| Interventions |
|
| Outcomes |
|
| Starting date | May 2020 |
| Contact information |
|
| Notes |
|
NCT04374487.
| Study name | A phase II, open label, randomized controlled trial to assess the safety and efficacy of convalescent plasma to limit COVID‐19 associated complications |
| Methods |
|
| Participants | Inclusion criteria
Or in case of severe or immediately life‐threatening COVID‐19, for example:
Exclusion criteria
|
| Interventions |
|
| Outcomes |
|
| Starting date | 9 May 2020 |
| Contact information |
|
| Notes |
|
NCT04376788.
| Study name | Exchange transfusion versus plasma from convalescent patients with methylene blue in patients with COVID‐19 |
| Methods |
|
| Participants | Inclusion criteria
Exclusion criteria
|
| Interventions |
|
| Outcomes |
|
| Starting date | 6 May 2020 |
| Contact information |
|
| Notes |
|
NCT04380935.
| Study name | Effectiveness and safety of convalescent plasma therapy on COVID‐19 patients with acute respiratory distress syndrome |
| Methods |
|
| Participants | Inclusion criteria
Exclusion criteria
|
| Interventions |
|
| Outcomes |
|
| Starting date | 8 May 2020 |
| Contact information | Robert Sinto, MD: +628158835432, rsinto@yahoo.com |
| Notes |
|
NCT04385043.
| Study name | Phase 2b/3 trial to evaluate the safety and efficacy of plasma transfusion from convalescent patients with SARS‐CoV‐2 infection on severity and mortality of COVID‐19 in hospitalized patients |
| Methods |
|
| Participants | Inclusion criteria
Exclusion criteria:
|
| Interventions |
|
| Outcomes |
|
| Starting date | 1 May 2020 |
| Contact information | Gabriella Talarico, MD0961883111, trasfusionale@aocz.it |
| Notes |
|
NCT04385186.
| Study name | Inactivated convalescent plasma as a therapeutic alternative in hospitalized patients COVID‐19 |
| Methods |
|
| Participants | Inclusion criteria
Exclusion criteria
|
| Interventions |
|
| Outcomes |
|
| Starting date | 20 June 2020 |
| Contact information |
|
| Notes |
|
NCT04388410.
| Study name | Phase 2b/3 trial to evaluate the safety and efficacy of plasma transfusion from convalescent patients with SARS‐CoV‐2 infection on severity and mortality of COVID‐19 in hospitalized patients. |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 1 June 2020 |
| Contact information |
|
| Notes |
|
NCT04390503.
| Study name | A phase 2 randomized, double‐blinded trial to evaluate the efficacy and safety of human anti‐SARS‐CoV‐2 plasma in close contacts of COVID‐19 cases |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | March 2021 (estimated) |
| Contact information |
|
| Notes |
|
NCT04391101.
| Study name | Efficacy of convalescent plasma for the treatment of severe SARS‐CoV‐2 infection: a randomized, open label clinical trial |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | June 2020 |
| Contact information |
|
| Notes |
|
NCT04403477.
| Study name | Convalescent plasma transfusion therapy in severe COVID‐19 patients ‐ a tolerability, efficacy and dose‐response phase II RCT |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 20 May 2020 |
| Contact information |
|
| Notes |
|
NCT04415086.
| Study name | Treatment of patients with COVID‐19 with convalescent plasma transfusion: a multicenter, open‐labeled, randomized and controlled study |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 1 June 2020 |
| Contact information |
|
| Notes |
|
NCT04418518.
| Study name | CONCOR‐1: a randomized open‐label trial of convalescent plasma for hospitalized adults with acute COVID‐19 respiratory illness |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 24 June 2020 |
| Contact information | Celine Arar: 212‐746‐4177; cea4002@med.cornell.edu |
| Notes |
|
NCT04425837.
| Study name | Effectiveness and safety of convalescent plasma in patients with high‐risk COVID‐19: a randomized, controlled study CRI‐CP (Coronavirus Investigation ‐ Convalescent Plasma) |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | July 2020 |
| Contact information |
|
| Notes |
|
NCT04438057.
| Study name | Evaluating the efficacy of convalescent plasma in symptomatic outpatients infected with COVID‐19 |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 6 July 2020 |
| Contact information |
|
| Notes |
|
NCT04442191.
| Study name | Infusion of convalescent plasma for the treatment of patients infected with severe acute respiratory syndrome‐coronavirus‐2 (COVID‐19): a double‐blinded, placebo‐controlled, proof‐of‐concept study |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 5 May 2020 |
| Contact information | Jessica Herrick, Assistant Professor of Clinical Medicine, University of Illinois at Chicago |
| Notes |
|
NCT04452812.
| Study name | Pilot clinical, statistical and epidemiological study on efficacy and safety of convalescent plasma for the management of patients with COVID‐19 |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 6 July 2020 |
| Contact information |
|
| Notes |
|
NCT04456413.
| Study name | Phase II randomized study of convalescent plasma from recovered COVID‐19 donors collected by plasmapheresis as treatment for subjects with early COVID‐19 infection |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 6 November 2020 |
| Contact information |
|
| Notes |
|
NCT04483960.
| Study name | An international multi‐centre randomised clinical trial to assess the clinical, virological and immunological outcomes in patients diagnosed with SARS‐CoV‐2 infection (COVID‐19) |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 21 July 2020 |
| Contact information |
|
| Notes |
|
NCT04521036.
| Study name | Convalescent plasma for COVID‐19 patients (CPCP) |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 1 December 2020 |
| Contact information |
|
| Notes |
|
NCT04528368.
| Study name | Convalescent plasma for treating patients with COVID‐19 pneumonia without indication of ventilatory support |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 18 August 2020 |
| Contact information | Contact: Eduardo M Rego, MD, PhD: edumrego@hotmail.com |
| Notes |
|
NCT04558476.
| Study name | A multicenter randomized trial to assess the efficacy of convalescent plasma therapy in patients with invasive COVID‐19 and acute respiratory failure treated with mechanical ventilation: the CONFIDENT trial |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 1 September 2020 |
| Contact information | Benoit Misset, MD,PhD: benoit.misset@chuliege.be |
| Notes |
|
NCT04567173.
| Study name | A randomized, open‐label, single center clinical trial to assess the efficacy and safety of convalescent plasma to hospitalized adult COVID‐19 patients as adjunctive therapy to reduce the need for ICU admission: Co‐CLARITY trial |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 28 September 2020 |
| Contact information | Deonne Thaddeus V Gauiran, MD: +639088150248.; dvgauiran@up.edu.ph |
| Notes |
|
NCT04634422.
| Study name | Plasma exchange (PLEX) and convalescent plasma (CCP) in COVID‐19 patients with multiorgan failure ‐ the COVID PLEX+CCP Trial |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 18 November 2020 |
| Contact information | Wladimir M Szpirt, MD: 4535451767; mail@covid‐plex.com |
| Notes |
|
NCT04712344.
| Study name | Assessment of efficacy and safety of therapy with COVID‐19 convalescent plasma in subjects with severe COVID‐19 (IPCO) |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 18 January 2021 |
| Contact information | Contact: Mario Schiffer, MD: +49913185 ext 39002; mario.schiffer@uk‐erlangen.de |
| Notes |
|
NCT04730401.
| Study name | Convalescent plasma in the treatment of COVID‐19 (CP_COVID‐19) |
| Methods |
|
| Participants | Inclusion criteria
Exclusion criteria
|
| Interventions |
|
| Outcomes |
|
| Starting date | 27 January 2021 |
| Contact information |
|
| Notes |
|
NCT04803370.
| Study name | Efficacy of reinforcing standard therapy in COVID‐19 patients with repeated transfusion of convalescent plasma |
| Methods |
|
| Participants | Inclusion criteria
Exclusion criteria
|
| Interventions |
|
| Outcomes |
|
| Starting date | 17 March 2021 |
| Contact information | Maria Arrizabalaga Asenjo, 0034871202000, email: marrizab@hsll.es |
| Notes |
|
NCT05077930.
| Study name | Convalescent plasma therapy for hospitalized patients with COVID‐19 |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | October 2021 |
| Contact information | Contact: Tânia P Costa, Master+55 41 3136‐2515tania.p@hospitaldorocio.com.br Contact: Leandro B Agati, PhD+55 11 4040‐8670agati@svriglobal.com |
| Notes |
|
NL8633.
| Study name | A randomized, double blinded clinical trial of convalescent plasma compared to standard plasma for treatment of hospitalized non‐ICU patients with COVID‐19 infections (COV‐PLAS) |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 13 May 2020 |
| Contact information | Name: Jaap Jan Zwaginga Email: j.j.zwaginga@lumc.nl Phone: 0715264006 |
| Notes |
|
PACTR202006760881890.
| Study name | Lagos COVID‐19 convalescent plasma trial (LACCPT) |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 24 September 2020 |
| Contact information |
|
| Notes |
|
PACTR202007653923168.
| Study name | A clinical trial comparing use of convalescent plasma therapy plus standard treatment to standard treatment alone in patients with severe COVID‐19 infection |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 1 August 2020 |
| Contact information |
|
| Notes |
|
PER‐013‐20.
| Study name | Convalescent plasma as treatment for COVID‐19 |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 19 September 2020 |
| Contact information |
|
| Notes |
|
PER‐060‐20.
| Study name | Randomized phase 2 clinical trial to evaluate safety and efficacy of the use of plasma from convalescent patients with the new coronavirus disease (COVID‐19) for the experimental treatment of patients hospitalized in the Centro Médico Naval 'Cirujano Mayor Santiago Távara' |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 19 October 2020 |
| Contact information |
|
| Notes |
|
RBR‐7jqpnw.
| Study name | Therapeutic effectiveness of COVID‐19 convalescent plasma produced by HEMOPE: a multicenter, randomized and controlled clinical trial |
| Methods |
|
| Participants |
|
| Interventions |
|
| Outcomes |
|
| Starting date | 1 July 2020 |
| Contact information |
|
| Notes |
|
ADL: activities of daily living; AE: adverse event; ALT: alanine transaminase; ANC: absolute neutrophil count; ARDS: acute respiratory distress syndrome; AST: aspartate transaminase; BAL: bronchoalveolar lavage; BAT: best available therapy; B(i)PAP: bi‐level positive airway pressure; BMI: body mass index; BP: blood pressure; CDC: Centers for Disease Control and Prevention; COI: conflict of interest; COPD: chronic obstructive pulmonary disease; CAP: community‐acquired pneumonia; CLIA: chemiluminescent immunoassay; CP: convalescent plasma; CPAP: continuous positive airway pressure; CPK: creatine phosphokinase; CRP: C‐reactive protein; CT: computed tomography; DBP: diastolic blood pressure; DIC: disseminated intravascular coagulation; DFPP: double‐filtration plasmapheresis; DIC: disseminated intravascular coagulation; DMARD: disease‐modifying anti‐rheumatic drug; DVT: deep vein thrombosis; ECG: electrocardiogram; ECMO: extracorporeal membrane oxygenation; ED: emergency department; FDA: US Food and Drug Administration; FFP: fresh frozen plasma; FiO2: fractional inspired oxygen; GFR: glomerular filtration rate; HBV/HCV: hepatitis B/C; HCPOA: healthcare power of attorney; HLA: human leukocyte antigen; HNA: human neutrophil antigens; ICU: intensive care unit; IgA (B/G/M): immunoglobulin A (B/G/M); IL‐6: interleukin‐6; IMV: invasive mechanical ventilation; IV: intravenous; IVIG: intravenous immunoglobulin; LAMP: loop‐mediated isothermal amplification; LAR: legal authorised representative; LDH: lactate dehydrogenase; MAP: mean arterial pressure; MCU: medium care unit; NR: not reported; NYHA: New York Heart Association; PaO2: arterial blood oxygen partial pressure; PCR: polymerase chain reaction; PE: pulmonary embolism; QoL: quality of life; RCT: randomised controlled trial; RF: respiratory failure; RNA: ribonucleic acid; RRT: renal replacement therapy; RT‐PCR: reverse transcription polymerase chain reaction; SAE: serious adverse event; SARS: severe acute respiratory syndrome; SBP: systolic blood pressure; SC: standard care; SOFA: Sequential Organ Failure Assessment; SP: standard plasma; SpO2: peripheral capillary oxygen saturation; SRD: severe respiratory disease; TACO: transfusion‐associated circulatory overload; TAD: transfusion‐associated dyspnoea; TB: tuberculosis; TRALI: transfusion‐related acute lung injury; TTP: thrombotic thrombocytopenic purpura; UIP: usual interstitial pneumonia; ULN: upper limit of normal; WBC: white blood cell; WHO: World Health Organization
Differences between protocol and review
In this section, we do not only report the differences between the protocol and the current review version but the changes between each published version of the review. The summary of amendments are also provided in Table 6.
Types of studies
Differences between first and second published review version
(Valk 2020 to Piechotta 2020b)
As the evidence we found from the randomised controlled trial (RCT) was at unclear or high risk of bias and at serious risk of bias for the controlled non‐randomised studies of interventions (NRSIs), and as none of these studies reported safety data for the control arm, we also included safety data from prospective and retrospective non‐comparative study designs (e.g. case series) and followed the methodology as specified in the protocol Piechotta 2020a). Because of the missing comparator, efficacy data of non‐controlled studies cannot be placed in context and therefore do not provide any useful evidence. In contrast to the protocol, we therefore decided to only include safety data of non‐controlled studies.
Differences between second and third published review version
(Piechotta 2020b to Chai 2020)
We decided to include registered non‐controlled NRSIs only to minimise selection bias.
Differences between third and fourth published review version
(Chai 2020 to Piechotta 2021)
Originally we had planned to include different study designs in a top‐down approach: RCTs, prospective and retrospective controlled NRSIs, and prospective and retrospective registered non‐controlled NRSIs. We had planned to include the next lower level in case we had low or very low certainty in the evidence of higher‐quality studies.
However, because large‐scale or expanded access studies could still provide valuable information on the safety of convalescent plasma or hyperimmune immunoglobulins, we decided to include prospectively registered single‐arm studies, even if upcoming RCTs reported safety data for both groups. We decided to consider prospectively registered single‐arm studies only for safety data, and if 500 or more participants were included.
Differences between fourth and current published review version
(Piechotta 2021 to current version)
Since we were aware that more RCTs had become available, we decided to only include data from RCTs in this update to have the best available quality of study design. We identified more safety data, so we decided to exclude prospectively registered single‐arm studies with inclusion of 500 or more participants and expanded access studies.
Types of participants
Differences between third and fourth published version
(Chai 2020 to Piechotta 2021)
After discussion with several attending physicians and clinical experts, we decided to perform separate analyses for populations with asymptomatic infection or mild disease and for hospitalised participants with moderate to severe disease, according to the latest WHO clinical progression score (WHO 2020e). We discussed that patient and study characteristics were not homogeneous enough to be combined and outcomes of interest differ.
Types of intervention
Differences between first and second published review version
(Valk 2020 to Piechotta 2020b)
We added standard immunoglobulin as an eligible control treatment.
Differences between fourth and current published review version
(Piechotta 2021 to current version)
We removed hyperimmune immunoglobulin as an eligible intervention treatment.
Types of outcome measures
Differences between protocol and first published review version
(Piechotta 2020a to Valk 2020)
We revised the secondary outcome 'Improvement of clinical symptoms, assessed through need for respiratory support at up to 7 days; 8 to 15 days; 16 to 30 days and added the fourth bullet point: 'plus high‐flow oxygen', to differentiate from the third bullet point. After revision, it read:
oxygen by mask or nasal prongs
oxygen by NIV (non‐invasive ventilation) or high flow
intubation and mechanical ventilation
extracorporeal membrane oxygenation (ECMO)
Differences between first and second published review version
(Valk 2020 to Piechotta 2020b)
We added the outcome 'quality of life' after discussion with a patient representative.
Differences between second and third published review version
(Piechotta 2020b to Chai 2020)
We renamed the outcome 'time to death' as 'mortality (time to event)'. This did not change the outcome measurement we are interested in.
We revised the secondary outcome 'Improvement of clinical symptoms' according to the revised outcome measure set for COVID‐19 clinical research (COMET 2020). Instead of defining cut‐offs ourselves, we refer to the recommended standardised scales:
Improvement of clinical symptoms, assessed by need for respiratory support with standardised scales (e.g. WHO Clinical Progression Scale (WHO 2020e), WHO Ordinal Scale for Clinical Improvement (WHO 2020f)) at up to 7 days, 8 to 15 days, 16 to 30 days.
We added the outcome 'virological response assessed with reverse transcription‐polymerase chain reaction (RT‐PCR) test for SARS‐CoV‐2 at baseline, up to 3, 7, and 15 days because this was suggested during the peer review of the last version of this review.
Differences between third and fourth published review version
(Chai 2020 to Piechotta 2021)
We divided efficacy outcomes for hospitalised individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease and ambulatory managed individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease, according to WHO clinical progression scale (WHO 2020e).
For individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease, we added the outcomes admission to hospital, development of moderate to severe clinical COVID‐19 symptoms, time to symptom onset, and any grade adverse events; and length of hospital stay for the subgroup of participants being hospitalised in the course of disease.
We revised and redefined outcomes for individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease after discussion with intensive care specialists and the German guideline panel for inpatient therapy of people with COVID‐19. We summarised different outcome measures for all‐cause mortality below one outcome, added sub‐outcomes for clinical improvement, and added clinical worsening to better reflect the course of disease and to detect group differences. We also added the outcome need for dialysis, and extended the definition of quality of life to also include fatigue and functional independence.
We renamed the outcome 'time to discharge from hospital' to 'Duration of hospitalisation, or time to discharge from hospital' to clarify that we are interested in both, continuous and time‐to‐event data. We renamed the outcome 'virological response' to 'viral clearance' to clarify that we are interested in test‐negativity and not in changes of viral load.
Differences between fourth and current published review version
(Piechotta 2021 to current version)
We renamed 'Prioritised outcomes' to 'Primary outcomes' and 'Additional outcomes' to 'Secondary outcomes'.
We revised and redefined outcomes for individuals with a confirmed diagnosis of COVID‐19 and moderate to severe disease. For the primary outcomes, we renamed the outcome 'all‐cause mortality at hospital discharge' to 'all‐cause mortality during hospital stay' and we redefined the outcome of worsening of clinical status to 'new need for invasive mechanical ventilation, or death' (including the competing event of death) and the outcome of improvement of clinical status to 'participants discharged from hospital'. For the secondary outcomes, we added the outcome 'improvement of clinical status' with three sub‐outcomes (weaning or liberation from invasive mechanical ventilation in surviving participants; ventilator‐free days; and liberation from supplemental oxygen in surviving participants). We removed the secondary outcome 'Length of stay on the intensive care unit (ICU) or time to discharge from ICU'.
We revised and redefined outcomes for individuals with a confirmed diagnosis of SARS‐CoV‐2 infection and asymptomatic or mild disease. For the primary outcomes, we added the outcome 'admission to hospital or death' (including the competing event of death), 'symptom resolution' with two sub‐outcomes (all initial symptoms resolved and time to symptom resolution). We removed the outcome 'development of moderate to severe clinical COVID‐19 symptoms' including all the sub‐outcomes. For the secondary outcomes, we removed 'admission to hospital', 'time to symptom onset', 'length of hospital stay', and 'admission to the ICU'. We added the outcome of worsening of clinical status with two sub‐outcomes 'need for hospitalisation with the need for oxygen by mask or nasal prongs, or death' and 'need for invasive mechanical ventilation, or death' (including the competing event of death).
Electronic searches
Differences between protocol and first published review version
(Piechotta 2020a to Valk 2020)
As publication bias might influence all subsequent analyses and conclusions, we searched all potentially relevant trials registries in detail to detect ongoing studies as well as completed but not yet published studies. Nowadays, it is mandatory to provide results at least in the trials registry. In case results were not published elsewhere, we had planned to extract and analyse these data. However, no outcome data had yet been added to the trials registries.
Differences between first and second published review version
(Valk 2020 to Piechotta 2020b)
We decided to exclude individual trials registries from the search strategy because they are already included in the Cochrane COVID‐19 Study Register, which is updated Monday to Friday, and to also exclude the WHO COVID‐19 Global Research Database. The WHO COVID‐19 Global Research Database and LitCov are included in the collection of the Center for Disease Control and Prevention COVID‐19 Research Article Database. The search part for COVID‐19 was updated for the search strategies from IM and CD peer reviewed it.
Differences between third and fourth published review version
(Chai 2020 to Piechotta 2021)
In the list of databases, The Living Overview of Evidence (L*OVE) Covid‐19 provided from Epistemonikos is included due to the variety of sources it contains (since September 2020) and the WHO COVID‐19 global literature on coronavirus disease was added because it integrates the CDC database (since October 2020).
New identified search terms like the MeSH term Immunization, Passive exploded and these search strings (passiv* adj3 (antibod* transfer* or immunization* or immunotherap* or immuno‐therap*)).tw,kf.; ((immunoglobulin* or immune globulin*) adj2 (therap* or treat*)).tw,kf.; (INM005 or CSL760).tw,kf.; (XAV‐19 or SAB‐185 or hIVIG or equine).tw,kf. were searched from November 2020. Due to the availability of more studies, the searches were focused on non‐RCTs and RCTs with adequate study filters (since December 2020). At the beginning of 2021, new MeSH or EMTREE terms were inserted in Medline and Embase, so the whole search strategies were revised and new search terms like IGY‐110 or GIGA‐2050 or GC5131 or 5131A or INOSARS were added. The search string (passive adj2 vaccin*).tw,kf. and new terms for hyperimmune like equine polyclonal antibodies (EpAbs), hyperimmune anti‐COVID‐19 IVIG (C‐IVIG), anti‐coronavirus immunoglobulin (ITAC), flebogamma were included in February 2021.
Differences between fourth and current published review version
(Piechotta 2021 to current version)
From April 2021 onwards the searches were restricted to RCTs with the use of RCT study filters. In May 2021, the new terms F(ab)2, BSVEQAb and EqAb‐COV‐19 were included and the search time was limited from 01 January 2019 to 01 January 2020. Since August 2021, we have run the searches monthly instead of weekly. In September 2021, we added the search terms bovine colostrum and bovine milk.
In March 2022, the search terms bioblock, γ‐Globulin, hyper‐Ig and C19‐IG, gammaglobulin were added. The search lines (Flu‐IVIG or ((anti‐flu* or anti‐influenza* or antiflu* or antinfluenza*) adj5 plasma)).mp. and (anti‐flu* or anti‐influenza* or antiflu* or antifluenza*) from MEDLINE and Embase were deleted. The RCT filter for Embase was replaced by the RCT‐filter Embase Cochrane Highly Sensitive Search Strategy (Lefebvre 2022b) for identifying controlled trials in Embase: (2018 revision); Ovid format. The WHO search strategy was revised and refined by adjacency searching.
Data extraction and management
Differences between protocol and first published review version
(Piechotta 2020a to Valk 2020)
We had planned to extract data using a standardised data extraction form developed in Covidence. However, we could not adapt the standardised form to our needs. Therefore we generated a customised data extraction form in Microsoft Excel (Microsoft Corporation 2018).
Differences between first and second published review version
(Valk 2020 to Piechotta 2020b)
Assessment of risk of bias in included studies
Randomised controlled trials
We had planned to use RoB 2 to judge the risk of bias in the underlying study results (Sterne 2019). However, RoB 2 was not yet available in RevMan Web (RevMan Web 2022), and the Cochrane Editorial and Methods Department recommended that we use the previous risk of bias tool instead (RoB 1; Higgins 2011).
Differences between second and third published review version
(Piechotta 2020b to Chai 2020)
Measures of treatment effect
We had planned to use the Excel tool of the purpose‐built method based on the Parmar and Tierney approach (Parmar 1998; Tierney 2007), to estimate hazard ratios (HRs) with the reported data, if HRs were not available. We were able to read off mortality data from the Kaplan‐Meier curve provided by Gharbharan 2020 [https://revman.cochrane.org/#/660020041013463556/dashboard/htmlCompare/3.6/2.11#STD‐Gharbharan‐2020] per day. Because we did not have the rights to edit the Excel tool to add a greater number of time intervals, we could not use the Excel tool. We therefore used a digitising software (GetData Graph Digitizer [https://revman.cochrane.org/#/660020041013463556/dashboard/htmlCompare/3.6/2.11#REF‐GetData‐Graph‐Digitizer]) to estimate the HR for Gharbharan 2020 [https://revman.cochrane.org/#/660020041013463556/dashboard/htmlCompare/3.6/2.11#STD‐Gharbharan‐2020].
Assessment of risk of bias in included studies
Randomised controlled trials
RoB became available in RevMan Web. We therefore decided to revert to our originally planned methodology for risk of bias assessment in RCTs and used RoB 2.0 for any assessments.
Differences between third and fourth published review version
(Chai 2020 to current Piechotta 2021)
Subgroup analysis and investigation of heterogeneity
We had added subgroup analyses for the following characteristics in this update of the review.
Duration since symptom onset
Level of antibody titre in donors
Level of antibody titre in recipients at baseline
SARS‐CoV‐2 variants
Considering the currently available evidence, we decided to add these subgroups, because their role in the effectiveness of convalescent plasma is currently being discussed and needs to be further investigated.
Differences between fourth and current published review version
(Piechotta 2021 to current version)
Subgroup analysis and investigation of heterogeneity
We had added subgroup analyses for the following characteristics in this update of the review.
Equity impact: sex (divided into female and male)
Equity impact: country income groups, according to the World Bank definitions, divided into high‐ and low‐ or middle‐income countries)
Equity impact: ethnicity
We further defined the following subgroups:
-
Severity of condition for inpatients only (assessed with need for respiratory support according to WHO clinical progression scale (WHO 2020e) are divided into:
moderate when at least 90% of participants are WHO level 4 or above and below WHO level 6
severe disease when at least 90% of participants are WHO level 6, or above, and
moderate to severe when 90% of participants are in both the "moderate" and "severe" category
Level of antibody titre in donors (divided into high and low titres, using the US Food and Drug Administration (FDA) definition for 'low' and 'high' titre, using the definitions in the studies)
In case of missing data, we conducted an available‐case analysis.
Summary of findings and assessment of the certainty of the evidence
Differences between protocol and first published review version
(Piechotta 2020a to Valk 2020)
At protocol stage, we had planned to assess the certainty in the evidence for our primary outcomes (all‐cause mortality at hospital discharge and time to death) only. However, as none of the included studies reported any deaths during their study periods, we decided to assess the certainty in the evidence also for prioritised secondary outcomes (clinical improvement, grades 3 and 4 adverse events, and serious adverse events) to increase the informative value on effectiveness and safety of convalescent plasma therapy.
Differences between first and second published review version
(Valk 2020 to Piechotta 2020b)
For the living systematic review we also prioritised patient quality of life as an important patient outcome and added this outcome to the summary of findings table. We specified in the methods how we graded the certainty of the evidence, especially for non‐randomised controlled trials using ROBINS‐I for risk of bias assessment, for calculation of absolute effects for time‐to‐event outcomes, and for writing informative statements for the findings and certainty of the evidence.
Differences between third and fourth published review version
(Chai 2020 to current Piechotta 2021)
We decided to include two summary of findings tables, one for each population.
We amended the outcomes for inclusion in the summary of findings table, in accordance with redefining the types of outcome measures. We summarised the outcome all‐cause mortality and provided a hierarchy of outcome measures that we would consider for inclusion in the summary of findings table. We added clinical worsening, in addition to clinical improvement, to better reflect the course of disease, and also provide a hierarchy for sub‐outcomes of all‐cause mortality.
Differences between fourth and current published review version
(Piechotta 2021 to current version)
We decided to include a summary of findings table for each comparison and for each population.
We amended the outcomes for inclusion in the summary of findings table, in accordance with redefining the types of outcome measures.
Contributions of authors
CI: methodological expertise, study selection, data extraction and assessment, conception and writing of the manuscript
KLC: clinical expertise, study selection and advice
VP: methodological expertise, study selection and data extraction
SJV: clinical expertise, study selection and advice
CK: clinical expertise, study selection, and advice
EA: study selection, data extraction and assessment
IM: development of the search strategy
EMW: clinical expertise and advice
AL: clinical expertise and advice
DJR: clinical expertise and advice
ZM: clinical expertise and advice
CS‐O: clinical expertise and advice
AJ: clinical expertise and advice
NC: data extraction and assessment
LJE: clinical expertise, and conception and writing of the manuscript
NK: methodological expertise, study selection, data extraction and assessment
NS: methodological expertise, study selection, data extraction and assessment, conception and writing of the manuscript
Sources of support
Internal sources
-
Sanquin Blood Supply, Netherlands
Center for Clinical Transfusion Research
-
University Hospital of Cologne, Germany
Cochrane Cancer, Department I of Internal Medicine
-
Monash University, Australia
Transfusion Research Unit, Department of Epidemiology and Preventive Medicine
-
NHS Blood and Transplant, UK
NHS Blood and Transplant
-
Leukaemia Foundation and HSANZ, Australia
Haematology Society of Australia and New Zealand (HSANZ)
External sources
-
European Union's Horizon 2020 research and innovation programme, Belgium
SUPorting high quality evaluation of COVID‐19 convalescent plasma thrOughouT Europe (SUPPORT‐E)
Declarations of interest
CI: none known, Managing Editor of Cochrane Haematology, but not involved in the editorial process for this review
KLC: HSANZ Leukaemia Foundation PhD scholarship to support studies at Monash University. This is not related to the work in this review.
VP: none known
SJV: is receiving a PhD scholarship from the not‐for‐profit Sanquin blood bank.
CK: none known
EA: none known
IM: none known
EMW: I have received funding support from Australian Medical Research Future Fund for a trial of convalescent plasma. I was not involved in bias assessment, data extraction or interpretation, but served as a content expert.
AL: none known
DJR: investigator on the REMAP‐CAP and RECOVERY trial. I was not involved in bias assessment, data extraction or interpretation, but served as a content expert.
ZM: I have received funding support from Australian Medical Research Future Fund for a trial of convalescent plasma. I was not involved in bias assessment, data extraction or interpretation, but served as a content expert.
CS‐O: is a member of the BEST Collaborative Clinical Study Group and Associate Editor for Transfusion Medicine Journal. I was not involved in bias assessment, data extraction or interpretation, but served as a content expert.
AJ: Investigator of PLACID Trial. I was not involved in bias assessment, data extraction or interpretation, but served as a content expert.
NC: none known
LJE: co‐lead of the COVID‐19 immunoglobulin domain of the REMAP‐CAP trial and investigator on the RECOVERY trial. I was not involved in bias assessment, data extraction or interpretation, but served as a content expert.
NK: none known, staff of Cochrane Haematology
NS: none known; she is Co‐ordinating Editor of Cochrane Haematology, but was not involved in the editorial process for this review.
contributed equally, last author
New search for studies and content updated (conclusions changed)
References
References to studies included in this review
Agarwal 2020 {published data only}
- Agarwal A, Mukherjee A, Kumar G, Chatterjee P, Bhatnagar T, Malhotra P, et al. Convalescent plasma in the management of moderate COVID-19 in adults in India: open label phase II multicentre randomised controlled trial (PLACID trial). BMJ 2020;371:m3939. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Agarwal A, Mukherjee A, Kumar G, Chatterjee P, Bhatnagar T, Malhotra P, et al. Convalescent plasma in the management of moderate COVID-19 in India: an open-label parallel-arm phase II multicentre randomized controlled trial (PLACID Trial). medRxiv [Preprint] 2020. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Agarwal A, Mukherjee A, Kumar G, Chatterjee P, Bhatnagar T, Malhotra P, et al. Erratum: Convalescent plasma in the management of moderate COVID-19 in adults in India: Open label phase II multicentre randomised controlled trial (PLACID Trial) (The BMJ (2020) 371 (m3939) DOI: 10.1136/bmj.m3939). The BMJ 2020;371 (no pagination)(m4232):1-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
Alemany 2022 {published and unpublished data}
- Alemany A, Millat-Martinez P, Corbacho-Monné M, Malchair P, Ouchi D, Ruiz-Comellas A, et al. High-titre methylene blue-treated convalescent plasma as an early treatment for outpatients with COVID-19: a randomised, placebo-controlled trial. Lancet Respiratory Medicine 2022;10(3):278-88. [DOI: 10.1016/S2213-2600(21)00545-2] [DOI] [PMC free article] [PubMed] [Google Scholar]
AlQahtani 2021 {published data only}
- AlQahtani M, Abdulrahman A, Almadani A, Alali SY, Al Zamrooni AM, Hejab A, et al. Randomized controlled trial of convalescent plasma therapy against standard therapy in patients with severe COVID-19 disease. MedRxiv [Preprint] 2020. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
- AlQahtani M, Abdulrahman A, Almadani A, Alali SY, Al Zamrooni AM, Hejab A, et al. Randomized controlled trial of convalescent plasma therapy against standard therapy in patients with severe COVID-19 disease. Scientific Reports 2021;11(1):9927. [DOI: 10.1038/s41598-021-89444-5] [DOI] [PMC free article] [PubMed] [Google Scholar]
Avendano‐Sola 2021 {published data only}
- Avendaño-Solá C, Ramos-Martínez A, Muñez-Rubio E, Ruiz-Antorán B, Malo de Molina R, Torres F, et al. A multicenter randomized open-label clinical trial for convalescent plasma in patients hospitalized with COVID-19 pneumonia. Journal of Clinical Investigation 2021;131(20):e152740. [DOI: 10.1172/JCI152740] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Avendano-Sola C, Ramos-Martinez A, Munez-Rubio E, Ruiz-Antoran B, Malo de Molina R, Torres F, et al. Convalescent plasma for COVID-19: a multicenter, randomized clinical trial. medRxiv [Preprint] 2020. [DOI: ] [Google Scholar]
- Diago-Sempere E, Bueno JL, Sancho-Lopez A, Munez-Rubio E, Torres F, Malo de Molina R, et al. Evaluation of convalescent plasma versus standard of care for the treatment of COVID-19 in hospitalized patients: study protocol for a phase 2 randomized, open-label, controlled, multicenter trial. Trials 2021;22(1):70. [DOI] [PMC free article] [PubMed]
Bajpai 2020 {published data only}
- Bajpai M, Kumar S, Maheshwari A, Chabra K, Kale P, Gupta A, et al. Efficacy of convalescent plasma therapy compared to fresh frozen plasma in severely ill COVID-19 patients: a pilot randomized controlled trial. Medrxiv [Preprint] 2020. [DOI: ] [Google Scholar]
Baldeon 2022 {published data only}
- Baldeón ME, Maldonado A, Ochoa-Anrade M, Largo C, Pesantez M, Herdoiza M. Effect of convalescent plasma as complementary treatment inpatients with moderate COVID-19 infection. Transfusion Medicine 2022;32(2):153-61. [DOI: 10.1111/tme.12851] [DOI] [PubMed] [Google Scholar]
Bar 2021 {published data only}
- Bar KJ, Shaw PA, Choi GH, Aqui N, Fesnak A, Yang JB, et al. A randomized controlled study of convalescent plasma for individuals hospitalized with COVID-19 pneumonia. Journal of Clinical Investigation 2021;131(24):e155114. [DOI: 10.1172/JCI155114] [DOI] [PMC free article] [PubMed] [Google Scholar]
Begin 2021 {published data only}
- Bégin P, Callum J, Jamula E, Cook R, Heddle NM, Tinmouth A, et al. Author correction: Convalescent plasma for hospitalized patients with COVID-19: an open-label, randomized controlled trial. Nature Medicine 2022;28:212. [DOI: 10.1038/s41591-021-01667-1] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bégin P, Callum J, Jamula E, Cook R, Heddle NM, Tinmouth A, et al. Convalescent plasma for hospitalized patients with COVID-19: an open-label, randomized controlled trial. Nature Medicine 2021;27:2012-24. [DOI: 10.1038/s41591-021-01488-2] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Begin P, Callum J, Heddle N, Cook R, Zeller MP, Timmouth A, et al. Convalescent plasma for adults with acute COVID-19 respiratory illness (CONCOR-1): study protocol for an international, multicentre, randomized, open-label trial. Trials 2021;22(1):323. [DOI: 10.1186/s13063-021-05235-3] [DOI] [PMC free article] [PubMed] [Google Scholar]
- The CONCOR-1 Study Group, Bégin P, Callum J, Jamula E, Cook R, Heddle N M, Tinmouth A, et al. Convalescent plasma for hospitalized patients with COVID-19 and the effect of plasma antibodies: a randomized controlled, open-label trial. MedRxiv [Preprint] 2021. [DOI: ]
Beltran Gonzalez 2021 {published data only}
- Beltran Gonzalez JL, Gonzalez Gamez M, Mendoza Enciso EA, Esparza Maldonado RJ, Hernandez Palacios D, Duenas Campos S, et al. Efficacy and safety of convalescent plasma and intravenous immunoglobulin in critically ill COVID-19 patients. A controlled clinical trial. MedRxiv [Preprint] 2021. [DOI: ] [Google Scholar]
Bennett‐Guerrero 2021 {published data only}
- Bennett-Guerrero E, Romeiser JL, Talbot LR, Ahmed T, Mamone LJ, Singh SM, et al. Severe acute respiratory syndrome coronavirus 2 convalescent plasma versus standard plasma in coronavirus disease 2019 infected hospitalized patients in New York. Critical Care Medicine 2021;49(7):1015-25. [DOI: 10.1097/CCM.0000000000005066] [DOI] [PMC free article] [PubMed] [Google Scholar]
CoV‐Early {published data only}
- NCT04589949. Early convalescent plasma therapy for high-risk patients with COVID-19 in primary care (the CoV-Early Study). Available from clinicaltrials.gov/show/NCT04589949 (first received 19 October 2020).
De Santis 2022 {published data only}
- De Santis GC, Correa Oliveira L, Garibaldi PM, Almado CE, Croda J, Arcanjo GG, et al. High-dose convalescent plasma for treatment of severe COVID-19. Emerging Infectious Diseases 2022;28(3):548-55. [DOI: 10.3201/eid2803.212299] [DOI] [PMC free article] [PubMed] [Google Scholar]
Devos 2021 {published data only}
- Devos T, Geukens T, Schauvlieghe A, Ariën K, Barbezange C, Cleeren M, et al. A randomized, multicentre, open-label phase II proof-of-concept trial investigating the clinical efficacy and safety of the addition of convalescent plasma to the standard of care in patients hospitalized with COVID-19: the Donated Antibodies Working against nCoV (DAWn-Plasma) trial. Trials 27 November 2020;21:981. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Devos T, Geukens T, Schauvlieghe A, Ariën K, Barbezange C, Cleeren M, et al. Correction to: a randomized, multicentre, open-label phase II proof-of-concept trial investigating the clinical efficacy and safety of the addition of convalescent plasma to the standard of care in patients hospitalized with COVID-19: the Donated Antibodies Working against nCoV (DAWn-Plasma) trial. Trials 14 December 2020;21:981. [DOI: 10.1186/s13063-020-04876-0] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Devos T, Geukens T, Schauwvlieghe A, Ariën K, Barbezange C, Cleeren M, et al. A randomized, multicentre, open-label phase II proof-of-concept trial investigating the clinical efficacy and safety of the addition of convalescent plasma to the standard of care in patients hospitalized with COVID-19: the Donated Antibodies Working against nCoV (DAWn-Plasma) trial. ResearchSquare [Preprint] 2020. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Devos T, Van Thillo Q, Compernolle V, Najdovski T, Romano M, Dauby N, et al. Early high antibody-titre convalescent plasma for hospitalised COVID-19 patients: DAWn-plasma. European Respiratory Journal 2021;59(2):2101724. [DOI: 10.1183/13993003.01724-2021] [DOI] [PMC free article] [PubMed] [Google Scholar]
Estcourt 2021 {published data only}
- The REMAP-CAP Investigators, Estcourt LJ. Convalescent plasma in critically ill patients with COVID-19. MedRxiv [Preprint] 2021. [DOI: ] [Google Scholar]
- Writing committee for the REMAP-CAP Investigators, Estcourt LJ, Turgeon AF, McQuilten ZK, McVerry BJ, Al-Beidh F, Annane D, et al. Effect of convalescent plasma on organ support-free days in critically ill patients with COVID-19: a randomized clinical trial. JAMA 2021;326:1690-702. [DOI: 10.1001/jama.2021.18178.] [DOI] [PMC free article] [PubMed] [Google Scholar]
Gharbharan 2021 {published and unpublished data}
- Gharbharan A, Jordans C, GeurtsvanKessel C, Hollander J, Karim F, Mollema F, et al. Effects of potent neutralizing antibodies from convalescent plasma in patients hospitalized for severe SARS-CoV-2 infection. ResearchSquare [Preprint] 2020. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gharbharan A, Jordans CC, GeurtsvanKessel C, den Hollander JG, Karim F, Mollema FP, et al. Convalescent plasma for COVID-19. A randomized clinical trial. medRxiv [Preprint] 2020. [DOI: ] [Google Scholar]
- Gharbharan A, Jordans CC, GeurtsvanKessel C, den Hollander JG, Karim F, Mollema FP, et al. Effects of potent neutralizing antibodies from convalescent plasma in patients hospitalized for severe SARS-CoV-2 infection. Nature Communications 2021;21:3189. [DOI: 10.1038/s41467-021-23469-2] [DOI] [PMC free article] [PubMed] [Google Scholar]
Hamdy Salman 2020 {published data only}
- Hamdy Salman O, Ail Mohamed HS. Efficacy and safety of transfusing plasma from COVID-19 survivors to COVID-19 victims with severe illness. A double-blinded controlled preliminary study. Egyptian Journal of Anaesthesia 2020;36(1):264-72. [DOI: 10.1080/11101849.2020.1842087] [DOI] [Google Scholar]
Holm 2021 {published data only}
- Holm K, Lundgren MN, Kjeldsen-Kragh J, Ljungquist O, Böttiger B, Wikén C, et al. Convalescence plasma treatment of COVID-19: results from a prematurely terminated randomized controlled open-label study in southern Sweden. BMC Research Notes 2021;14(1):440. [DOI: 10.1186/s13104-021-05847-7] [DOI] [PMC free article] [PubMed] [Google Scholar]
Horby 2021b {published data only}
- Horby PW, Estcourt L, Peto L, Emberson JR, Staplin N, Spata E, et al. Convalescent plasma in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial. medRxiv [Preprint] 2021. [DOI: ] [Google Scholar]
- RECOVERY Collaborative Group. Convalescent plasma in patients admitted to hospital with COVID-19 (RECOVERY): a randomised controlled, open-label, platform trial. Lancet 2021;397(10289):2049-59. [DOI: 10.1016/S0140-6736(21)00897-7] [DOI] [PMC free article] [PubMed] [Google Scholar]
Kirenga 2021 {published data only}
- Kirenga B, Byakika-Kibwika P, Muttamba W, Kayongo A, Loryndah NO, Mugenyi L, et al. Efficacy of convalescent plasma for treatment of COVID-19 in Uganda. BMJ 2021;8:e001017. [DOI: 10.1136/bmjresp-2021-001017] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Muttamba W, Lusiba J, Namakula LO, Byakika-Kibwika P, Ssali F, Ddungu H, et al. Feasibility of collecting and processing of COVID-19 convalescent plasma for treatment of COVID-19 in Uganda. PLoS One 2021;16(6):e0252306. [DOI: 10.1371/journal.pone.0252306] [DOI] [PMC free article] [PubMed] [Google Scholar]
Koerper 2021 {published data only}
- Koerper S, Jahrsdoerfer B, Appl T. Convalescent plasma for treatment of severe COVID-19: rationale and designing of a randomized, open-label clinical trial of convalescent plasma compared to best supportive care (CAPSID Trial). Transfusionsmedizin 2020;10(3):143-9. [Google Scholar]
- Koerper S, Weiss M, Zickler D, Wiesmann T, Zacharowski K, Corman VM, et al. Results of the CAPSID randomized trial for high-dose convalescent plasma in severe COVID-19 patients. Journal of Clinical Investigation 2021;131(20):e152264. [DOI: 10.1172/JCI152264] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Koerper S, Weiss M, Zickler D, Wiesmann T, Zacharowski K, Corman VM et al. High dose convalescent plasma in COVID-19: results from the randomized trial CAPSID. medRxiv [Preprint] 2021. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
Korley 2021 {published data only}
- Korley FK, Durkalski-Mauldin V, Yeatts SD, Schulman K, Davenport RD, Dumont LJ, et al. Early convalescent plasma for high-risk outpatients with COVID-19. New England Journal of Medicine 2021;385(21):1951-60. [DOI: 10.1056/NEJMoa2103784] [DOI] [PMC free article] [PubMed] [Google Scholar]
Li 2020 {published data only}
- Li L, Zhang W, Hu Y, Tong X, Zheng S, Yang J, et al. Effect of convalescent plasma therapy on time to clinical improvement in patients with severe and life-threatening COVID-19: a randomized clinical trial. JAMA 3 June 2020;324(5):460-70. [DOI: 10.1001/jama.2020.10044] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Li L, Zhang W, Hu Y. Correction to data in trial of convalescent plasma treatment for COVID-19. JAMA 2020;324(5):519. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Liu Z. Errors in trial of effect of convalescent plasma therapy on time to clinical improvement in patients with severe and life-threatening COVID-19. JAMA 2020;324(5):518-9. [DOI] [PubMed] [Google Scholar]
Libster 2020 {published data only}
- Libster R, Marc GP, Wappner D, Coviello S, Bianchi A, Braem V, et al. Early high-titer plasma therapy to prevent severe COVID-19 in older adults. New England Journal of Medicine 2021;384(7):610-18. [DOI: 10.1056/NEJMoa2033700] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Libster R , Marc GP , Wappner D , Coviello S , Bianchi A , Braem V, et al. Prevention of severe COVID-19 in the elderly by early high-titer plasma. medRxiv [Preprint] 2020. [DOI: ] [Google Scholar]
Menichetti 2021 {published data only}
- Menichetti F, Popoli P, Puopolo M, Alegiani Spila S, Tiseon G, Bartolini A, et al. Effect of high-titer convalescent plasma on progression to severe respiratory failure or death in hospitalized patients with COVID-19 pneumonia: a randomized clinical trial. JAMA Network Open 2021;4(11):e2136246. [DOI: 10.1001/jamanetworkopen.2021.36246] [DOI] [PMC free article] [PubMed] [Google Scholar]
NCT04421404 {published data only}
- NCT04421404. Effects of COVID-19 convalescent plasma (CCP) on coronavirus-associated complications in hospitalized patients. Available from clinicaltrials.gov/show/NCT04421404 (first received 9 June 2020).
Ortigoza 2022 {published data only}
- Ortigoza MB, Yoon H, Goldfeld KS, Troxel AB, Daily JP, Wu Y, et al. Efficacy and safety of COVID-19 convalescent plasma in hospitalized patients: a randomized clinical trial. JAMA Internal Medicine 2022;182(2):115-26. [DOI: 10.1001/jamainternmed.2021.6850] [DOI] [PMC free article] [PubMed] [Google Scholar]
O’Donnell 2021 {published data only}
- Eckhardt CM, Cummings MJ, Rajagopalan KN, Borden S, Bitan ZC, Wolf A, et al. Correction to: evaluating the efficacy and safety of human anti-SARS-CoV-2 convalescent plasma in severely ill adults with COVID-19: a structured summary of a study protocol for a randomized controlled trial. Trials 2020;21(1):927. [DOI: 10.1186/s13063-020-04877-z] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eckhardt CM, Cummings MJ, Rajagopalan KN, Borden S, Bitan ZC, Wolf A, et al. Evaluating the efficacy and safety of human anti-SARS-CoV-2 convalescent plasma in severely ill adults with COVID-19: a structured summary of a study protocol for a randomized controlled trial. Trials 2020;21(1):499. [DOI] [PMC free article] [PubMed] [Google Scholar]
- O’Donnell M, Grinsztejn B, Cummings MJ, Justman J, Lamb MR, Eckhardt CM, et al. A randomized, double-blind, controlled trial of convalescent plasma in adults with severe COVID-19. MedRxiv [Preprint} 2021. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
- O’Donnell M, Grinsztejn B, Cummings MJ, Justman J, Lamb MR, Eckhardt CM, et al. A randomized double-blind controlled trial of convalescent plasma in adults with severe COVID-19. Journal of Clinical Investigation 2021 [Epub ahead of print];11:150646. [DOI: 10.1172/JCI150646] [DOI] [PMC free article] [PubMed] [Google Scholar]
Pouladzadeh 2021 {published data only}
- Pouladzadeh M, Safdarian M, Eshghi P, Abolghasemi H, Bavani AG, Sheibani B, et al. A randomized clinical trial evaluating the immunomodulatory effect of convalescent plasma on COVID-19-related cytokine storm. Internal and Emergency Medicine 2021;16:2181-91. [DOI: 10.1007/s11739-021-02734-8] [DOI] [PMC free article] [PubMed] [Google Scholar]
Ray 2022 {published data only}
- Bandopadhyay P, Rozario R, Lahiri A, Sarif J, Ray Y, Paul SR, et al. Nature and dimensions of the cytokine storm and its attenuation by convalescent plasma in severe COVID-19. medRxiv [Preprint] 2020. [DOI: ] [Google Scholar]
- Ray Y, Paul SR, Bandopadhyay P, D'Rozario R, Sarif J, Lahiri A, et al. Clinical and immunological benefits of convalescent plasma therapy in severe COVID-19: insights from a single center open label randomised control trial. medRxiv [Preprint] 2020. [DOI: ] [Google Scholar]
- Ray Y, Paul SR, Bandopadhyay P, D'Rozario R, Sarif J, Raychaudhuri D, et al. A phase 2 single center open label randomised control trial for convalescent plasma therapy in patients with severe COVID-19. Nature Communications 2022;13:383. [DOI: 10.1038/s41467-022-28064-7] [DOI] [PMC free article] [PubMed] [Google Scholar]
Sekine 2021 {published data only}
- Sekine L, Arns B, Fabro BR, Cipolatt MM, Machado RR, Durigon EL, et al. Convalescent plasma for COVID-19 in hospitalised patients: an open-label, randomised clinical trial. European Respiratory Journal 2021;59(2):2101471. [EMBASE: 10.1183/13993003.01471-2021] [DOI] [PMC free article] [PubMed] [Google Scholar]
Simonovich 2020 {published data only}
- Simonovich VA, Burgos PL, Scibona P, Beruto MV, Vallone MG, Vázquez C, et al. A randomized trial of convalescent plasma in COVID-19 severe pneumonia. New England Journal of Medicine 2020;384(7):619-29. [DOI: 10.1056/NEJMoa2031304] [DOI] [PMC free article] [PubMed] [Google Scholar]
Sullivan 2022 {published data only}
- Sullivan DJ, Gebo KA, Shoham S, Bloch EM, Lau B, Shenoy AG, et al. Early outpatient treatment for COVID-19 with convalescent plasma. New England Journal of Medicine March 2022 [Epub ahead of print];Moa2119657:NP. [DOI: 10.1056/NEJMoa2119657] [DOI] [PMC free article] [PubMed] [Google Scholar]
Van den Berg 2022 {published data only}
- Van den Berg K, Glatt TN, Vermeulen M, Little F, Swanevelder R, Barrett C, et al. Convalescent plasma in the treatment of moderate to severe COVID-19 pneumonia: a randomized controlled trial (PROTECT-Patient Trial). Scientific Reports 2022;12:2552. [DOI: 10.1038/s41598-022-06221-8] [DOI] [PMC free article] [PubMed] [Google Scholar]
References to studies excluded from this review
Abdullah 2020 {published data only}
- Abdullah HM, Hama-Ali HH, Ahmed SN, Ali KM, Karadakhy KA, Mahmood SO, et al. A severe refractory COVID-19 patient responding to convalescent plasma; a case series. Annals of Medicine and Surgery 2020;56:125-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
Abolghasemi 2020 {published data only}
- Abolghasemi H, Eshghi P, Cheraghali AM, Imani Fooladi AA, Bolouki Moghaddam F, Imanizadeh S, et al. Clinical efficacy of convalescent plasma for treatment of COVID-19 infections: results of a multicenter clinical study. Transfusion and Apheresis Science 2020 July 15 [Epub ahead of print];59(5):102875. [DOI: 10.1016/j.transci.2020.102875] [DOI] [PMC free article] [PubMed] [Google Scholar]
Ahn 2020 {published data only}
- Ahn JY, Sohn Y, Lee SH, Cho Y, Hyun JH, Baek YJ, et al. Use of convalescent plasma therapy in two COVID-19 patients with acute respiratory distress syndrome in Korea. Journal of Korean Medical Science 2020;35(14):e149. [DOI] [PMC free article] [PubMed] [Google Scholar]
Allahyari 2021 {published data only}
- Allahyari A, Seddigh-Shamsi M, Mahmoudi M, Amel Jamehdar S, Amini M, Mozdourian M, et al. Efficacy and safety of convalescent plasma therapy in severe COVID-19 patients with acute respiratory distress syndrome. International Immunopharmacology 2021;93:107239. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
- IRCT20200409047007N1. Effect of COVID 19 survivors plasma in COVID 19 patients with ARDS. Available from en.irct.ir/trial/47058 (first received 12 April 2020).
Anderson 2020 {published data only}
- Anderson J, Schauer J, Bryant S, Graves CR. The use of convalescent plasma therapy and remdesivir in the successful management of a critically ill obstetric patient with novel coronavirus 2019 infection: a case report. Case Reports in Women's Health 2020;27:e00221. [DOI: 10.1016/j.crwh.2020.e00221] [DOI] [PMC free article] [PubMed] [Google Scholar]
Baklaushev 2020 {published data only}
- Baklaushev VP, Averyanov AV, Sotnikova AG, Perkina AS, Ivanov AV, Yusubalieva GM, et al. Safety and efficacy of convalescent plasma for COVID-19: the preliminary results of a clinical trial. Journal of Clinical Practice 2020;11(2):38-50. [Google Scholar]
Balcells 2020 {published data only}
- Balcells ME, Rojas L, Le Corre N, Martínez-Valdebenito C, Ceballos ME, Ferrés M, et al. Early anti-SARS-CoV-2 convalescent plasma in patients admitted for COVID-19: a randomized phase II clinical trial. medRxiv [Preprint] 2020. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
Bao 2020b {published data only}
- Bao Y, Lin SY, Cheng ZH, Xia J, Sun YP, Zhao Q, et al. Clinical features of COVID-19 in a young man with massive cerebral hemorrhage—case report. SN Comprehensive Clinical Medicine 2020 May 23 [Epub ahead of print];2:703–9. [DOI: 10.1007/s42399-020-00315-y] [DOI] [PMC free article] [PubMed] [Google Scholar]
Bobek 2020 {published data only}
- Bobek I, Gopcsa L, Reti M, Beko G, Hancz L, Lakatos B, et al. Successful administration of convalescent plasma in critically ill COVID-19 patients in Hungary: the first two cases. Orvosi Hetilap 2020;161(27):1111-21. [DOI] [PubMed] [Google Scholar]
Bradfute 2020 {published data only}
- Bradfute SB, Hurwitz I, Yingling AV, Ye C, Cheng Q, Noonan TP, et al. SARS-CoV-2 neutralizing antibody titers in convalescent plasma and recipients in New Mexico: an open treatment study in COVID-19 patients. Journal of Infectious Diseases 2020;11:11. [DOI] [PMC free article] [PubMed] [Google Scholar]
Brasil Ministerio 2020 {published data only}
- Brasil Ministério da Saúde, Secretaria de Ciência. NA [Tratamento farmacológico para casos internados com SARS-COV-2, do Hospital das Clínicas da Faculdade de Medicina de Ribeirão Preto da Universidade de São Paulo]. Available from fi-admin.bvsalud.org/document/view/27t7v (accessed 25 June 2020).
Budhai 2020 {published data only}
- Budhai A, Wu AA, Hall L, Strauss D, Paradiso S, Alberigo J, et al. How did we rapidly implement a convalescent plasma program? Transfusion 2020 May 25 [Epub ahead of print];60:1348-55. [DOI: 10.1111/trf.15910] [DOI] [PMC free article] [PubMed] [Google Scholar]
Cantore 2020 {published data only}
- Cantore I, Valente P. Convalescent plasma from COVID 19 patients enhances intensive care unit survival rate. A preliminary report. Transfusion and Apheresis Science 2020 Jun 10 [Epub ahead of print];59(5):102848. [DOI: 10.1016/j.transci.2020.102848] [DOI] [PMC free article] [PubMed] [Google Scholar]
Cao 2020a {published data only}
- Cao W, Liu X, Bai T, Fan H, Hong K, Song H, et al. High-dose intravenous immunoglobulin as a therapeutic option for deteriorating patients with coronavirus disease 2019. Open Forum Infectious Diseases 2020;7(3):ofaa102. [DOI] [PMC free article] [PubMed] [Google Scholar]
Chen 2020b {published data only}
- Chen X, Li Y, Wang J, Cai H, Cao H, Sheng J. Pregnant women complicated with COVID-19: a clinical analysis of 3 cases. Zhejiang da Xue Xue Bao. Yi Xue Ban = Journal of Zhejiang University. Medical Sciences 2020;49(2):240-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
Chen 2020c {published data only}
- Chen Q, Quan B, Li X, Gao G, Zheng W, Zhang J, et al. A report of clinical diagnosis and treatment of nine cases of coronavirus disease 2019. Journal of Medical Virology 2020;92(6):683-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
ChiCTR2000029850 {published data only}
- ChiCTR2000029850. Efficacy and safety of convalescent plasma treatment for severe patients with novel coronavirus pneumonia (COVID-19): a prospective cohort study. Available from www.chictr.org.cn/showproj.aspx?proj=49533 (first received 15 February 2020).
ChiCTR2000030039 {published data only}
- ChiCTR2000030039. Clinical study for infusing convalescent plasma to treat patients with new coronavirus pneumonia (COVID-19). Available from www.chictr.org.cn/showproj.aspx?proj=49544 (first received 21 February 2020).
ChiCTR2000030312 {published data only}
- ChiCTR2000030312. Cancelled, due to modify the protocol A single-center, open-label and single arm trial to evaluate the efficacy and safety of anti-SARS-CoV-2 inactivated convalescent plasma in the treatment& [A single-center, open-label and single arm trial to evaluate the efficacy and safety of anti-SARS-CoV-2 inactivated convalescent plasma in the treatment of novel coronavirus pneumonia (COVID-19) patient]. Available from www.chictr.org.cn/showproj.aspx?proj=50258 (first received 23 April 2020).
ChiCTR2000030381 {published data only}
- ChiCTR2000030381. Cancelled by investigator. A randomized, open-label, controlled and single-center trial to evaluate the efficacy and safety of anti-SARS-CoV-2 inactivated convalescent plasma in the treatment of novel coronavirus pneumonia (COVID-19) patient [A randomized, open-label, controlled and single-center trial to evaluate the efficacy and safety of anti-SARS-CoV-2 inactivated convalescent plasma in the treatment of novel coronavirus pneumonia (COVID-19) patient]. Available from www.chictr.org.cn/showproj.aspx?proj=50290 (first received 23 April 2020).
ChiCTR2000030442 {published data only}
- ChiCTR2000030442. Combination of tocilizumab, IVIG and CRRT in severe patients with novel coronavirus pneumonia (COVID-19). Available from www.chictr.org.cn/showproj.aspx?proj=50380 (first received 23 April 2020).
ChiCTR2000031501 {published data only}
- ChiCTR2000031501. The efficacy of convalescent plasma in patients with critical novel coronavirus pneumonia (COVID-19): a pragmatic, prospective cohort study. Available from www.chictr.org.cn/showproj.aspx?proj=50254 (first received 2 April 2020).
ChiCTR2000033798 {published data only}
- ChiCTR2000033798. The efficacy and safety of convalescent plasma therapy in novel coronavirus pneumonia (COVID-19): a medical records based retrospective cohort study. Available from www.chictr.org.cn/showproj.aspx?proj=55194 (first received 12 June 2020).
Clark 2020 {published data only}
- Clark E, Guilpain P, Filip L, Pansu N, Le Bihan C, Cartron G, et al. Convalescent plasma for persisting COVID-19 following therapeutic lymphocyte depletion: a report of rapid recovery. British Journal of Haematology 2020;27:27. [DOI] [PMC free article] [PubMed] [Google Scholar]
CTRI/2020/04/024804 {published data only}
- CTRI/2020/04/024804. Evaluation of safety and efficacy of convalescent plasma in COVID-19 patients. Available from ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=42849 (first received 22 April 2020).
CTRI/2020/08/027285 {published data only}
- CTRI/2020/08/027285. Safety of convalescent plasma (CVP) drawn from mild symptomatic COVID-19 patients. Available from www.ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=45050 (first received 21 August 2020).
CTRI/2020/09/027903 {published data only}
- CTRI/2020/09/027903. Testing the efficacy and safety of a blood product COVID-19 hyper-immuneglobulin (human) solution in participants with active COVID-19. Available from ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=47147 (first received 18 September 2020).
CTRI/2020/10/028547 {published data only}
- CTRI/2020/10/028547. IND02 for prevention against SARS-CoV-2 infection: a randomized controlled study in moderate to high risk population. Available from ctri.nic.in/Clinicaltrials/showallp.php?mid1=48708&EncHid=&userName=2020/10/028547.
Çınar 2020 {published data only}
- Çınar OE, Sayınalp B, Karakulak EA, Karataş AA, Velet M, İnkaya AÇ, et al. Convalescent (immune) plasma treatment in a myelodysplastic COVID-19 patient with disseminated tuberculosis. Transfusion and Apheresis Science 2020;59(5):102821. [DOI: 10.1016/j.transci.2020.102821] [DOI] [PMC free article] [PubMed] [Google Scholar]
de Assis 2020 {published data only}
- Assis RR, Jain A, Nakajima R, Jasinskas A, Felgner J, Obiero JM, et al. Analysis of SARS-CoV-2 antibodies in COVID-19 convalescent plasma using a coronavirus antigen microarray. bioRxiv [Preprint] 2020. [DOI: 10.1101/2020.04.15.043364] [DOI] [PMC free article] [PubMed] [Google Scholar]
Díez 2020 {published data only}
- Díez JM, Romero C, Gajardo R. Currently available intravenous immunoglobulin contains antibodies reacting against severe acute respiratory syndrome coronavirus 2 antigens. Immunotherapy 2020 May 13 [Epub];12:571-6. [DOI: 10.2217/imt-2020-0095] [DOI] [PMC free article] [PubMed] [Google Scholar]
Donato 2020 {published data only}
- Donato M, Park S, Baker M, Korngold R, Morawski A, Geng X, et al. Clinical and laboratory evaluation of patients with SARS-CoV-2 pneumonia treated with high-titer convalescent plasma: a prospective study. medRxiv [Preprint] 2020. [DOI: 10.1101/2020.07.20.20156398] [DOI] [PMC free article] [PubMed] [Google Scholar]
Duan 2020 {published data only}
- Duan K, Liu B, Li C, Zhang H, Yu T, Qu J, et al. Effectiveness of convalescent plasma therapy in severe COVID-19 patients. Proceedings of the National Academy of Sciences of the United States of America 2020;117:9490-6. [DOI: 10.1073/pnas.2004168117] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Duan K, Liu B, Li C, Zhang H, Yu T, Qu J, et al. The feasibility of convalescent plasma therapy in severe COVID-19 patients: a pilot study. medRxiv [Preprint] 2020. [DOI: 10.1101/2020.03.16.20036145] [DOI] [Google Scholar]
Dulipsingh 2020 {published data only}
- Dulipsingh L, Danyal I, Crowell R, Diffenderfer M, Williams K, Lima C, et al. SARS-CoV-2 serology and virology trends in donors and recipients of convalescent plasma. Transfus Apher Sci 2020;59:102922. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
Enzmann 2020 {published data only}
- Enzmann MO, Erickson MP, Grindeland CJ, Lopez SM, Hoover SE, Leedahl DD. Treatment and preliminary outcomes of 150 acute care patients with COVID-19 in a rural health system in the Dakotas. Epidemiology and Infection 2020;148:e124. [DOI] [PMC free article] [PubMed] [Google Scholar]
Erkurt 2020 {published data only}
- Erkurt MA, Sarici A, Berber I, Kuku I, Kaya E, Ozgul M. Life-saving effect of convalescent plasma treatment in COVID-19 disease: clinical trial from eastern Anatolia. Transfusion and Apheresis Science 2020 Jun 27 [Epub ahead of print];59(5):102867. [DOI: 10.1016/j.transci.2020.102867] [DOI] [PMC free article] [PubMed] [Google Scholar]
EUCTR2020‐005979‐12‐GR {published data only}
- EUCTR2020-005979-12-GR. Efficacy and safety of XAV-19 for the treatment of moderate to severe COVID-19. Available from https://www.clinicaltrialsregister.eu/ctr-search/trial/2020-005979-12/GR (first received 08 March 2021).
Fan 2020 {published data only}
- Fan O, Qiang F, Shuhong G, Haibing Y, Xiangyang L, Min T, et al. Recovery from critical COVID-19 despite delays in diagnosis and respiratory treatment: a cautionary tale. Signa Vitae 2020;16(1):193-8. [Google Scholar]
Figlerowicz 2020 {published data only}
- Figlerowicz M, Mania A, Lubarski K, Lewandowska Z, Sluzewski W, Derwich K, et al. First case of convalescent plasma transfusion in a child with COVID-19-associated severe aplastic anemia. Transfusion and Apheresis Science 2020 July 1 [Epub ahead of print];59(5):102866. [DOI: 10.1016/j.transci.2020.102866] [DOI] [PMC free article] [PubMed] [Google Scholar]
Franchini 2020 {published data only}
- Franchini M, Marano G, Velati C, Pati I, Pupella S, Liumbruno GM. Operational protocol for donation of anti-COVID-19 convalescent plasma in Italy. Vox Sanguinis 2021;116(1):136-7. [DOI: 10.1111/vox.12940] [DOI] [PMC free article] [PubMed] [Google Scholar]
Gaborit 2021 {published data only}
- EUCTR202000257427. A randomized, double-blind, placebo-controlled phase 2a and 2b study to evaluate the safety and efficacy of XAV-19 in patients with COVID-19 induced moderate pneumonia. Available from www.clinicaltrialsregister.eu/ctr-search/trial/2020-002574-27/FR (first received 01 July 2020).
- Gaborit B, Vanhove B, Vibet M A, Le Thuaut A, Lacombe K, Dubee V, et al. Evaluation of the safety and efficacy of XAV-19 in patients with COVID-19-induced moderate pneumonia: study protocol for a randomized, double-blinded, placebo-controlled phase 2 (2a and 2b) trial. Trials 2021;22(199). [DOI: 10.1186/s13063-021-05132-9] [DOI] [PMC free article] [PubMed] [Google Scholar]
- NCT04453384. Study to evaluate the safety and efficacy of XAV-19 in patients with COVID-19 induced moderate pneumonia. Available from clinicaltrials.gov/show/NCT04453384 (first received 01 July 2020).
Grisolia 2020 {published data only}
- Grisolia G, Franchini M, Glingani C, Inglese F, Garuti M, Beccaria M, et al. Convalescent plasma for coronavirus disease 2019 in pregnancy: a case report and review. American Journal of Obstetrics and Gynecology 2020;2(3):100174. [DOI] [PMC free article] [PubMed] [Google Scholar]
Hashim 2020 {published data only}
- Hashim HT. Convalescent plasma to treat COVID-19: its challenges in Iraq situation. Ethics, Medicine and Public Health 2020 Jul 16 [Epub ahead of print];15:100564. [DOI: 10.1016/j.jemep.2020.100564] [DOI] [PMC free article] [PubMed] [Google Scholar]
Hu 2020 {published data only}
- Hu H, Ma F, Wei X, Fang Y. Coronavirus fulminant myocarditis saved with glucocorticoid and human immunoglobulin. European Heart Journal 2020;16:16. [DOI] [PMC free article] [PubMed] [Google Scholar]
Ibrahim 2020 {published data only}
- Ibrahim D, Dulipsingh L, Zapatka L, Eadie R, Crowell R, Williams K, et al. Factors associated with good patient outcomes following convalescent plasma in COVID-19: a prospective phase II clinical trial. Infectious Diseases and Therapy 2020 [Epub ahead of print];9:913–926. [DOI: 10.1007/s40121-020-00341-2] [DOI] [PMC free article] [PubMed] [Google Scholar]
Im 2020 {published data only}
- Im JH, Nahm CH, Baek JH, Kwon HY, Lee JS. Convalescent plasma therapy in coronavirus disease 2019: a case report and suggestions to overcome obstacles. Journal of Korean Medical Science 2020;35(26):e239. [DOI] [PMC free article] [PubMed] [Google Scholar]
IRCT20151228025732N53 {published data only}
- IRCT20151228025732N53. Therapeutic effects of plasma of recovered people from COVID-19 on hospitalized patients with this disease. Available from en.irct.ir/trial/46931 (first received 10 April 2020).
IRCT20200406046968N2 {published data only}
- IRCT20200406046968N2. Efficacy of convalescent plasma transfusion of COVID-19 survivors on the treatment of respiratory failure of these patients. Available from www.irct.ir/trial/47115 (first received 22 April 2020).
IRCT20200414047072N1 {published data only}
- IRCT20200414047072N1. Use of convalescent plasma in the treatment of COVID-19. Available from en.irct.ir/trial/47163 (first received 28 April 2020).
IRCT20200416047099N1 {published data only}
- IRCT20200416047099N1. Plasma therapy in patient with COVID-19. Available from en.irct.ir/trial/47266 (first received 21 April 2020).
IRCT20200508047346N1 {published data only}
- IRCT20200508047346N1. Evaluation of the effectiveness of rabbit antibody against coronavirus in patients. Available from www.irct.ir/trial/47953 (first received 13 May 2020).
IRCT20200525047562N1 {published data only}
- IRCT20200525047562N1. Treatment of COVID-19 patients with convalescent plasma. Available from en.irct.ir/trial/48493 (first received 14 June 2020).
ISRCTN86534580 {published data only}
- ISRCTN86534580. A trial evaluating treatments for suspected coronavirus infection in people aged 50 years and above with pre-existing conditions and those aged 65 years and above. Available from www.isrctn.com/ISRCTN86534580 (first received 20 March 2020).
Jamous 2020 {published data only}
- Jamous F, Meyer N, Buus D, Ateeli H, Taggart K, Hanson T, et al. Critical illness due to COVID-19: a description of the surge in a single center in Sioux Falls. South Dakota Medicine 2020;73(7):312-7. [PubMed] [Google Scholar]
Jiang 2020a {published data only}
- Jiang J, Miao Y, Zhao Y, Lu X, Zhou P, Zhou X, et al. Convalescent plasma therapy: helpful treatment of COVID-19 in a kidney transplant recipient presenting with severe clinical manifestation and complex complications. Clinical Transplantation 2020 Jun 30 [Epub ahead of print];34:e14025. [DOI: 10.1111/ctr.14025] [PMID: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
Jiang 2020b {published data only}
- Jiang Y, He S, Zhang C, Wang X, Chen X, Jin Y, et al. Clinical characteristics of 60 discharged cases of 2019 novel coronavirus-infected pneumonia in Taizhou, China. Annals of Translational Medicine 2020;8(8):547. [DOI] [PMC free article] [PubMed] [Google Scholar]
Jin 2020 {published data only}
- Jin C, Gu J, Yuan Y, Long Q, Zhang Q, Zhou H, et al. Treatment of 6 COVID-19 patients with convalescent plasma. medRxiv [Preprint] 2020. [DOI: 10.1101/2020.05.21.20109512] [DOI]
Joyner 2020 {published data only}
- Joyner MJ, Bruno KA, Klassen SA, Kunze KL, Johnson PW, Lesser ER, et al. Safety update: COVID-19 convalescent plasma in 20,000 hospitalized patients. Mayo Clinical Proceedings 2020;95(9):1888-97. [DOI] [PMC free article] [PubMed] [Google Scholar]
jRCT2031200174 {published data only}
- jRCT2031200174. An international multicenter, adaptive, randomized double-blind, placebo-controlled trial of the safety, tolerability and efficacy of anti-coronavirus hyperimmune intravenous immunoglobulin for the treatment of adult hospitalized patients at onset of clinical progression of COVID-19. Available from who.int/trialsearch/Trial2.aspx?TrialID=JPRN-jRCT2031200174 (first received 26 October 2020).
Karatas 2020 {published data only}
- Karatas A, Inkaya AC, Demiroglu H, Aksu S, Haziyev T, Cinar OE, et al. Prolonged viral shedding in a lymphoma patient with COVID-19 infection receiving convalescent plasma. Transfusion and Apheresis Science 2020 October [Epub ahead of print ];59(5):102871. [DOI: 10.1016/j.transci.2020.102871] [DOI] [PMC free article] [PubMed] [Google Scholar]
Kong 2020 {published data only}
- Kong Y, Cai C, Ling L, Zeng L, Wu M, Wu Y, et al. Successful treatment of a centenarian with coronavirus disease 2019 (COVID-19) using convalescent plasma. Transfusion and Apheresis Science 2020 May 21 [Epub ahead of print];59:102820. [DOI: 10.1016/j.transci.2020.102820] [DOI] [PMC free article] [PubMed] [Google Scholar]
Lin 2020 {published data only}
- Lin JH, Chen YC, Lu CL, Hsu YN, Wang WJ. Application of plasma exchange in association with higher dose CVVH in cytokine storm complicating COVID-19. Journal of the Formosan Medical Association 2020;119(6):1116-8. [DOI: 10.1016/j.jfma.2020.04.023] [DOI] [PMC free article] [PubMed] [Google Scholar]
Liu 2020a {published data only}
- Liu ST, Lin HM, Baine I, Wajnberg A, Gumprecht JP, Rahman F, et al. Convalescent plasma treatment of severe COVID-19: a propensity score–matched control study. Nature Medicine 2020;26:1708–13. [DOI: ] [DOI] [PubMed] [Google Scholar]
Liu 2020b {published data only}
- Liu M, Chen Z, Dai MY, Yang JH, Chen XB, Chen D, et al. Lessons learned from early compassionate use of convalescent plasma on critically ill patients with COVID-19. Transfusion 2020 Aug 8 [Epub ahead of print];60:2210-16. [DOI: 10.1111/trf.15975] [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
Madariaga 2020 {published data only}
- Madariaga ML, Guthmiller J, Schrantz S, Jansen M, Christenson C, Kumar M, et al. Clinical predictors of donor antibody titer and correlation with recipient antibody response in a COVID-19 convalescent plasma clinical trial. medRxiv [Preprint] 2020. [DOI: 10.1101/2020.06.21.20132944] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Madariaga ML, Schrantz S, Jansen MO, Christensen C, Kumar M, Prochaska M, et al. Integrated COVID-19 convalescent plasma treatment and antibody research program at a single academic medical center. SSRN [Preprint] 2020. [DOI: 10.2139/ssrn.3605131] [DOI] [Google Scholar]
Martinez‐Resendez 2020 {published data only}
- Martinez-Resendez MF, Castilleja-Leal F, Torres-Quintanilla A, Rojas-Martinez A, Garcia-Rivas G, Ortiz-Lopez R, et al. Initial experience in Mexico with convalescent plasma in COVID-19 patients with severe respiratory failure, a retrospective case series. medRxiv [Preprint] 2020. [DOI: 10.1101/2020.07.14.20144469] [DOI] [Google Scholar]
McCuddy 2020 {published data only}
Ministerio de Salud 2020 {published data only}
- Ministerio de Salud - Instituto Nacional de Salud. NA [Lineamientos técnicos para uso de plasma convaleciente en pacientes con COVID-19]. Available from fi-admin.bvsalud.org/document/view/nruba 2020;1:20.
Mira 2020 {published data only}
- Mira E, Yarce OA, Ortega C, Fernández S, Pascual N, Gómez C, et al. Rapid recovery of a SARS-CoV-2 infected X-linked agammaglobulinemia patient after infusion of COVID-19 convalescent plasma. Journal of Allergy and Clinical Immunology: In Practice 2020;8(8):2793-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
NCT04261426 {published data only}
- NCT04261426. The efficacy of intravenous immunoglobulin therapy for severe 2019-nCoV infected pneumonia. Available from clinicaltrials.gov/ct2/show/NCT04261426 (first received 23 April 2020).
NCT04264858 {published data only}
- ChiCTR2000030841. Treatment of acute severe COVID-19 with immunoglobulin from cured COVID-19 patients. Available from www.chictr.org.cn/showproj.aspx?proj=51072 (first received 15 March 2020).
- NCT04264858. An exploratory clinical study on the treatment of acute severe 2019-nCoV pneumonia with immunoglobulin from cured 2019-nCoV pneumonia patients. Available from clinicaltrials.gov/show/NCT04264858 (first received 11 February 2020).
NCT04292340 {published data only}
- NCT04292340. The efficacy and safety of anti-SARS-CoV-2 inactivated convalescent plasma in the treatment of novel coronavirus pneumonia patient (COVID-19): an observational study. Available from clinicaltrials.gov/show/NCT04292340 (first received 3 March 2020).
NCT04321421 {published data only}
- NCT04321421. Hyperimmune plasma for critical patients with COVID-19. Available from clinicaltrials.gov/ct2/show/NCT04321421 (first received 28 May 2020).
NCT04323800 {published data only}
- NCT04323800. Convalescent plasma to stem coronavirus: a randomized, blinded phase 2 study comparing the efficacy and safety human coronavirus immune plasma (HCIP) vs. control (SARS-CoV-2 non-immune plasma) among adults exposed to COVID-19. Available from clinicaltrials.gov/show/NCT04323800 (first received 23 April 2020).
NCT04325672 {published data only}
- NCT04325672. Convalescent plasma to limit coronavirus associated complications: an open label, phase 2A study of high-titer anti-SARS-CoV-2 plasma in hospitalized patients with COVID-19. Available from clinicaltrials.gov/show/NCT04325672 (first received 23 April 2020).
NCT04327349 {published data only}
- NCT04327349. Investigating effect of convalescent plasma on COVID-19 patients outcome: a clinical trial. Available from clinicaltrials.gov/show/NCT04327349 (first received 31 March 2020).
NCT04332380 {published data only}
- NCT04332380. Convalescent plasma for patients with COVID-19: a pilot study. Available from clinicaltrials.gov/show/NCT04332380 (first received 2 April 2020).
NCT04333355 {published data only}
- NCT04333355. Phase 1 study to evaluate the safety of convalescent plasma as an adjuvant therapy in patients with SARS-CoV-2 infection. Available from clinicaltrials.gov/show/NCT04333355 (first received 3 April 2020).
NCT04338360 {published data only}
- NCT04338360. Expanded access to convalescent plasma for the treatment of patients with COVID-19. Available from clinicaltrials.gov/show/NCT04338360 (first received 8 April 2020).
NCT04344015 {published data only}
- NCT04344015. COVID-19 plasma collection. Available from clinicaltrials.gov/show/NCT04344015 (first received 23 April 2020).
NCT04344379 {published data only}
- NCT04344379. Prevention of SARS-CoV-2 in hospital workers exposed to the virus. Available from clinicaltrials.gov/show/NCT04344379 (first received 14 April 2020).
NCT04344977 {published data only}
- NCT04344977. COVID-19 plasma collection. Available from clinicaltrials.gov/ct2/show/NCT04344977 (first received 14 April 2020).
NCT04345679 {published data only}
- NCT04345679. Anti COVID-19 convalescent plasma therapy. Available from clinicaltrials.gov/show/NCT04345679 (first received 14 April 2020).
NCT04346589 {published data only}
- NCT04346589. Convalescent antibodies infusion in critically ill COVID 19 patients. Available from clinicaltrials.gov/ct2/show/NCT04346589 (first received 15 April 2020).
NCT04347681 {published data only}
- Albalawi M, Zaidi SZ, AlShehry N, AlAskar A, Zaidi AR, Abdallah RN, et al. Safety and efficacy of convalescent plasma to treat severe COVID-19: protocol for the Saudi collaborative multi-center phase II study. JMIR Research Protocol 2020;9(10):e23543. [DOI] [PMC free article] [PubMed] [Google Scholar]
- NCT04347681. Potential efficacy of convalescent plasma to treat severe COVID-19 and patients at high risk of developing severe COVID-19. Available from clinicaltrials.gov/show/NCT04347681 (first received 15 April 2020).
NCT04348877 {published data only}
- NCT04348877. Plasma rich antibodies from recovered patients from COVID19. Available from clinicaltrials.gov/show/NCT04348877 (first received 16 April 2020).
NCT04350580 {published data only}
- NCT04350580. Polyvalent immunoglobulin in COVID-19 related ARDS. Available from clinicaltrials.gov/show/NCT04350580 (first received 17 April 2020).
NCT04352751 {published data only}
- NCT04352751. Experimental use of convalescent plasma for passive immunization in current COVID-19 pandemic in Pakistan in 2020. Available from clinicaltrials.gov/show/NCT04352751 (first received 20 April 2020).
NCT04353206 {published data only}
- NCT04353206. Convalescent plasma in ICU patients with COVID-19-induced respiratory failure. Available from clinicaltrials.gov/show/NCT04353206 (first received 20 April 2020).
NCT04354831 {published data only}
- NCT04354831. A study evaluating the efficacy and safety of high-titer anti-SARS-CoV-2 plasma in hospitalized patients with COVID-19 infection. Available from clinicaltrials.gov/ct2/show/NCT04354831 (first received 21 April 2020).
NCT04355897 {published data only}
- NCT04355897. COVID-19 plasma in treatment of COVID-19 patients. Available from clinicaltrials.gov/ct2/show/NCT04355897 (first received 21 April 2020).
NCT04356482 {published data only}
- NCT04356482. Convalescent plasma for ill patients by COVID-19. Available from clinicaltrials.gov/show/NCT04356482 (first received 22 April 2020).
NCT04358211 {published data only}
- NCT04358211. Expanded access to convalescent plasma to treat and prevent pulmonary complications associated with COVID-19. Available from clinicaltrials.gov/show/NCT04358211 (first received 24 April 2020).
NCT04360278 {published data only}
- NCT04360278. Plasma collection from convalescent and/or immunized donors for the treatment of COVID-19. Available from clinicaltrials.gov/show/NCT04360278 (first received 24 April 2020).
NCT04360486 {published data only}
- NCT04360486. Treatment of COVID-19 with anti-SARS-CoV-2 convalescent plasma (ASCoV2CP). Available from clinicaltrials.gov/show/NCT04360486 (first received 24 April 2020).
NCT04363034 {published data only}
- NCT04363034. Arkansas expanded access COVID-19 convalescent plasma treatment program. Available from clinicaltrials.gov/ct2/show/NCT04363034 (first received 27 April 2020).
NCT04365439 {published data only}
- NCT04365439. Convalescent plasma for COVID-19. Available from clinicaltrials.gov/show/NCT04365439 (first received 28 April 2020).
NCT04366245 {published data only}
- NCT04366245. Clinical trial to evaluate the efficacy of treatment with hyperimmune plasma obtained from convalescent antibodies of COVID-19 infection. Available from clinicaltrials.gov/show/NCT04366245 (first received 28 April 2020).
NCT04368013 {published data only}
- NCT04368013. Host-pathogen interactions, immune response, and clinical prognosis at COVID-19 - the CoVUm trial. Available from clinicaltrials.gov/show/NCT04368013 (first received 20 April 2020).
NCT04372368 {published data only}
- NCT04372368. Convalescent plasma for the treatment of patients with COVID-19. Available from clinicaltrials.gov/show/NCT04372368 (first received 04 May 2020).
NCT04374370 {published data only}
- NCT04374370. SARSCoV2 (COVID-19) convalescent plasma (CP) expanded access protocol (EAP). Available from clinicaltrials.gov/show/NCT04374370 (first received 5 May 2020).
NCT04374565 {published data only}
- NCT04374565. Convalescent plasma for treatment of COVID-19 patients with pneumonia. Available from clinicaltrials.gov/show/NCT04374565 (first received 5 May 2020).
NCT04376034 {published data only}
- NCT04376034. Convalescent plasma collection and treatment in pediatrics and adults. Available from clinicaltrials.gov/show/NCT04376034 (first received 6 May 2020).
NCT04377568 {published data only}
- NCT04377568. Efficacy of human coronavirus-immune convalescent plasma for the treatment of COVID-19 disease in hospitalized children. Available from clinicaltrials.gov/show/NCT04377568 (first received 6 May 2020).
NCT04377672 {published data only}
- NCT04377672. Human convalescent plasma for high-risk children exposed or infected with SARS-CoV-2. Available from clinicaltrials.gov/show/NCT04377672 (first received 6 May 2020).
NCT04383548 {published data only}
- NCT04383548. Clinical study for efficacy of anti-corona VS2 immunoglobulins prepared from COVID19 convalescent plasma prepared by VIPS mini-pool IVIG medical devices in prevention of SARS-CoV-2 infection in high risk groups as well as treatment of early cases of COVID. Available from clinicaltrials.gov/show/NCT04383548 (first received 12 May 2020).
NCT04384497 {published data only}
- NCT04384497. Convalescent plasma for treatment of COVID-19: an exploratory dose identifying study. Available from clinicaltrials.gov/ct2/show/NCT04384497 (first received 12 May 2020).
NCT04384588 {published data only}
- NCT04384588. COVID19-convalescent plasma for treating patients with active symptomatic COVID 19 infection (FALP-COVID). Available from clinicaltrials.gov/show/NCT04384588 (first received 12 May 2020).
NCT04388527 {published data only}
- NCT04388527. COVID-19 convalescent plasma for mechanically ventilated population. Available from clinicaltrials.gov/show/NCT04388527 (first received 14 May 2020).
NCT04389710 {published data only}
- NCT04389710. Convalescent plasma for the treatment of COVID-19. Available from clinicaltrials.gov/show/NCT04389710 (first received 15 May 2020).
NCT04389944 {published data only}
- NCT04389944. Amotosalen-ultraviolet a pathogen-inactivated convalescent plasma in addition to best supportive care and antiviral therapy on clinical deterioration in adults presenting with moderate to severe COVID-19. Available from clinicaltrials.gov/show/NCT04389944 (first received 15 May 2020).
NCT04390178 {published data only}
- NCT04390178. Convalescent plasma as treatment for acute coronavirus disease (COVID-19). Available from clinicaltrials.gov/show/NCT04390178 (first received 15 May 2020).
NCT04392232 {published data only}
- NCT04392232. A phase 2 study of COVID 19 convalescent plasma in high risk patients with COVID 19 infection. Available from clinicaltrials.gov/show/NCT04392232 (first received 18 May 2020).
NCT04393727 {published data only}
- NCT04393727. Transfusion of convalescent plasma for the early treatment of pneumonIa due to SARSCoV2. Available from clinicaltrials.gov/show/NCT04393727 (first received 19 May 2020).
NCT04395170 {published data only}
- NCT04395170. Convalescent plasma compared to anti-COVID-19 human immunoglobulin and standard treatment (TE) in hospitalized patients. Available from clinicaltrials.gov/show/NCT04395170 (first received 20 May 2020).
NCT04397523 {published data only}
- NCT04397523. Efficacy and safety of COVID-19 convalescent plasma. Available from clinicaltrials.gov/show/NCT04397523 (first received 21 May 2020).
NCT04407208 {published data only}
- NCT04407208. Convalescent plasma therapy in patients with COVID-19. Available from clinicaltrials.gov/show/NCT04407208 (first received 29 May 2020).
NCT04408040 {published data only}
- Use of convalescent plasma for COVID-19. Available from clinicaltrials.gov/ct2/show/NCT04408040 (first received 29 May 2020).
NCT04408209 {published data only}
- NCT04408209. Convalescent plasma for the treatment of patients with severe COVID-19 infection. Available from clinicaltrials.gov/show/NCT04408209 (first received 29 May 2020).
NCT04411602 {published data only}
- NCT04411602. Feasibility study of anti-SARS-CoV-2 plasma transfusions in COVID-19 patients with SRD. Available from clinicaltrials.gov/show/NCT04411602 (first received 2 June 2020).
NCT04412486 {published data only}
- NCT04412486. COVID-19 convalescent plasma (CCP) transfusion. Available from clinicaltrials.gov/show/NCT04412486 (first received 2 June 2020).
NCT04418531 {published data only}
- NCT04418531. Convalescent antibodies infusion in COVID 19 patients. Available from clinicaltrials.gov/show/NCT04418531 (first received 5 June 2020).
NCT04420988 {published data only}
- Investigational COVID-19 convalescent plasma infusion for severely or life-threateningly ill COVID-19 patients. Available from clinicaltrials.gov/ct2/show/NCT04420988 (first received 09 June 2020).
NCT04432103 {published data only}
- NCT04432103. Treatment of severe and critical COVID-19 pneumonia with convalescent plasma. Available from clinicaltrials.gov/show/NCT04432103 (first received 16 June 2020).
NCT04432272 {published data only}
- NCT04432272. Antibody-level based analysis of COVID convalescent serum (ABACCuS). Available from clinicaltrials.gov/ct2/show/NCT04432272 (first received 16 June 2020).
NCT04438694 {published data only}
- NCT04438694. Use of convalescent plasma for treatment of patients with COVID-19 infection. Available from clinicaltrials.gov/show/NCT04438694 (first received 19 June 2020).
NCT04445207 {published data only}
- Experimental expanded access treatment with convalescent plasma for the treatment of patients with COVID-19. Available from clinicaltrials.gov/ct2/show/NCT04445207 (first received 24 June 2020).
NCT04458363 {published data only}
- NCT04458363. Convalescent plasma in pediatric COVID-19. Available from clinicaltrials.gov/ct2/show/NCT04458363 (first received 07 July 2020).
NCT04462848 {published data only}
- NCT04462848. COVID-19 convalescent plasma as prevention and treatment for children with underlying medical conditions. Available from clinicaltrials.gov/show/NCT04462848 (first received 8 July 2020).
NCT04463823 {published data only}
- "NORPLASMA" COVID-19 convalescent plasma treatment monitoring study. Available from clinicaltrials.gov/ct2/show/NCT04463823 (first received 09 July 2020).
NCT04467151 {published data only}
- NCT04467151. Administration of anti-SARS-CoV-2 convalescent plasma in hospitalized, non-ICU patients with COVID-19. Available from clinicaltrials.gov/show/NCT04467151 (first received 10 July 2020).
NCT04468958 {published data only}
- NCT04468958. Safety, tolerability, and pharmacokinetics of SAB-185 in healthy participants. Available from clinicaltrials.gov/ct2/show/NCT04468958 (first received 13 July 2020).
NCT04469179 {published data only}
- NCT04469179. Safety, tolerability, and pharmacokinetics of SAB-185 in ambulatory participants with COVID-19. Available from clinicaltrials.gov/show/NCT04469179 (first received 13 July 2020).
NCT04471051 {published data only}
- NCT04471051. An observational cohort trial of outcomes and antibody responses following treatment with COVID19 convalescent plasma in hospitalized COVID-19 patients. Available from clinicaltrials.gov/show/NCT04471051 (first received 14 July 2020).
NCT04472572 {published data only}
- NCT04472572. Expanded access to convalescent plasma for treatment of COVID-19. Available from clinicaltrials.gov/show/NCT04472572 (first received 15 July 2020).
NCT04474340 {published data only}
- NCT04474340. COVID-19 convalescent plasma treatment in SARS-CoV-2 infected patients: multicenter interventional study. Available from clinicaltrials.gov/show/NCT04474340 (first received 16 July 2020).
NCT04476888 {published data only}
- NCT04476888. Convalescent plasma treatment in COVID-19. Available from clinicaltrials.gov/show/NCT04476888 (first received 20 July 2020).
NCT04492501 {published data only}
- Investigational treatments for COVID-19 in tertiary care hospital of Pakistan. Available from clinicaltrials.gov/ct2/show/NCT04492501 (first received 30 July 2020).
NCT04497779 {published data only}
- Evaluation of coronavirus disease 19 (COVID-19) convalescent plasma. Available from clinicaltrials.gov/ct2/show/NCT04497779 (first received 04 August 2020).
NCT04502472 {published data only}
- NCT04502472. Open-label treatment of severe coronavirus disease 2019 (COVID-19) with convalescent plasma. Available from clinicaltrials.gov/show/NCT04502472 (first received 06 August 2020).
NCT04513158 {published data only}
- NCT04513158. Convalescent plasma in the early treatment of high-risk patients with SARS-CoV-2 (COVID-19) infection. Available from clinicaltrials.gov/show/NCT04513158 (first received 14 August 2020).
NCT04514302 {published data only}
- NCT04514302. Safety and efficacy of anti-SARS-CoV-2 equine antibody fragments (INOSARS) for hospitalized patients with COVID-19. Available from clinicaltrials.gov/show/NCT04514302 (first received 14 August 2020).
NCT04516954 {published data only}
- NCT04516954. Convalescent plasma for COVID-19 patients. Available from clinicaltrials.gov/show/NCT04516954 (first received 18 August 2020).
NCT04524507 {published data only}
- NCT04524507. COVID-19 antibody plasma research study in hospitalized patients (UNC CCP RCT). Available from clinicaltrials.gov/show/NCT04524507 (first received 24 August 2020).
NCT04535063 {published data only}
- NCT04535063. Convalescent plasma as potential therapy for severe COVID-19 pneumonia. Available from clinicaltrials.gov/show/NCT04535063 (first received 01 September 2020).
NCT04545047 {published data only}
- NCT04545047. Observational study of convalescent plasma for treatment of veterans with COVID-19. Available from clinicaltrials.gov/show/NCT04545047 (first received 10 September 2020).
NCT04546581 {published data only}
- EUCTR2020-002542-16-GR. Treatment of patients with coronavirus infection with immunoglobulin. Available from who.int/trialsearch/Trial2.aspx?TrialID=EUCTR2020-002542-16-GR (first received 14 September 2020).
- NCT04546581. Inpatient treatment with anti-coronavirus immunoglobulin (ITAC). Available from clinicaltrials.gov/show/NCT04546581 (first received 14 September 2020).
NCT04554992 {published data only}
- NCT04554992. Convalescent plasma for the treatment of COVID-19. Available from clinicaltrials.gov/show/NCT04554992 (first received 18 September 2020).
NCT04555109 {published data only}
- NCT04555109. Convalescent plasma for COVID-19 research donor study. Available from clinicaltrials.gov/show/NCT04555109 (first received 18 September 2020).
NCT04555148 {published data only}
- NCT04555148. COVIDIG (COVID-19 Hyper-ImmunoGlobulin). Available from clinicaltrials.gov/show/NCT04555148 (first received 18 September 2020).
NCT04565197 {published data only}
- NCT04565197. Convalescent plasma therapy for COVID-19 patients. Available from clinicaltrials.gov/show/NCT04565197 (first received 25 September 20209.
NCT04569188 {published data only}
- NCT04569188. Convalescent plasma in COVID-19 elderly patients. Available from clinicaltrials.gov/show/NCT04569188 (first received 29 September 2020).
NCT04570982 {published data only}
- NCT04570982. Clinical protocol for convalescent plasma and remdesivir therapy in Nepal. Available from clinicaltrials.gov/show/NCT04570982 (first received 30 September 2020).
NCT04573855 {published data only}
- NCT04573855 - unclear. Treatment with anti-SARS-CoV-2 immunoglobulin in patients with COVID-19. Available from clinicaltrials.gov/show/NCT04573855 (first received 05 October 2020).
NCT04593940 {published data only}
- NCT04593940. Immune modulators for treating COVID-19 (ACTIV-1 IM). Available from clinicaltrials.gov/ct2/show/NCT04593940 (first retrieved 14 June 2021).
NCT04610502 {published data only}
- NCT04610502. Efficacy and safety of two hyperimmune equine anti Sars-CoV-2 in COVID-19 patients. Available from clinicaltrials.gov/show/NCT04610502 (first received 30 October 2020).
NCT04614012 {published data only}
- NCT04614012. Hyperimmune plasma for patients with COVID-19. Available from clinicaltrials.gov/show/NCT04614012 (first received 03 November 2020).
NCT04616976 {published data only}
- NCT04616976. COVID-19 with convalescent plasma. Available from clinicaltrials.gov/show/NCT04616976 (first received 05 November 2020).
NCT04622826 {published data only}
- NCT04622826. plasmApuane CoV-2: efficacy and safety of immune COVID-19 plasma in COVID-19 pneumonia in non ITU patients. Available from clinicaltrials.gov/show/NCT04622826 (first received 10 November 2020).
NCT04638634 {published data only}
- NCT04638634. Pharmacokinetics, safety, and tolerability of CSL760, an anti- COVID-19 hyperimmune intravenous immunoglobulin, in healthy adult subjects. Available from clinicaltrials.gov/ct2/show/NCT04638634 (first received 20 November 2020).
NCT04642014 {published data only}
- NCT04642014. Application of convalescent plasma in the treatment of SARS CoV-2 disease (COVID-19) with evaluation of therapy effectiveness. Available from clinicaltrials.gov/show/NCT04642014 (first received 24 November 2020).
NCT04644198 {published data only}
- NCT04644198. Convalescent plasma transfusion in severe COVID-19 patients in Jamaica. Available from clinicaltrials.gov/ct2/show/NCT04644198 (first received 25 November 2020).
NCT04661839 {published data only}
- NCT04661839. Phase 1 study to evaluate safety and pharmacokinetics of COVID-HIGIV administered as a single dose or a repeat dose in healthy adults. Available from clinicaltrials.gov/show/NCT04661839 (first received 10 December 2020).
NCT04669990 {published data only}
- NCT04669990. Remdesivir and convalescent plasma therapy for treatment of COVID-19 infection in Nepal: a registry study. Available from clinicaltrials.gov/ct2/show/NCT04669990 (first received 17 December 2020).
NCT04721236 {published data only}
- NCT04721236. Early use of hyperimmune plasma in COVID-19 (COV-II-PLA). Available from clinicaltrials.gov/ct2/show/NCT04721236 (first received 22 January 2021).
Niu 2020 {published data only}
- Niu A, McDougal A, Ning B, Safa F, Luk A, Mushatt DM, et al. COVID-19 in allogeneic stem cell transplant: high false-negative probability and role of CRISPR and convalescent plasma. Bone Marrow Transplantation 2020;15:15. [DOI] [PMC free article] [PubMed] [Google Scholar]
Olivares‐Gazca 2020 {published data only}
- Olivares-Gazca JC, Priesca-Marin JM, Ojeda-Laguna M, Garces-Eisele J, Soto-Olvera S, Palacios-Alonso A, et al. Infusion of convalescent plasma is associated with clinical improvement in critically ill patients with COVID-19: a pilot study. Revista de Investigation Clinica 2020;72(3):159-64. [DOI] [PubMed] [Google Scholar]
Pei 2020 {published data only}
- Pei S, Yuan X, Zhimin ZZ, Run YR, Xie Y, Minxue SM, et al. Convalescent plasma to treat COVID-19: Chinese strategy and experiences. medRxiv [Preprint] 2020. [DOI: 10.1101/2020.04.07.20056440] [DOI]
Peng 2020 {published data only}
- Peng H, Gong T, Huang X, Sun X, Luo H, Wang W, et al. A synergistic role of convalescent plasma and mesenchymal stem cells in the treatment of severely ill COVID-19 patients: a clinical case report. Stem Cell Research and Therapy 2020;11(1):291. [DOI] [PMC free article] [PubMed] [Google Scholar]
PER‐031‐20 {published data only}
- PER-031-20. Phase 2 Study of efficacy and safety of plasma from convalescent patients with COVID-19 in patients with moderate disease (AUNA 20-01). https://www.ins.gob.pe/ensayosclinicos/rpec/recuperarECPBNuevoEN.asp?numec=031-20 (first received 06 July 2020).
Perotti 2020 {published data only}
- Perotti C, Baldanti F, Bruno R, Delfante C, Seminari E, Casari S, et al. Mortality reduction in 46 severe COVID-19 patients treated with hyperimmune plasma. A proof of concept single arm multicenter interventional trial. medRxiv [Preprint] 2020. [DOI: 10.1101/2020.05.26.20113373] [DOI] [PMC free article] [PubMed]
- Perotti C, Baldanti F, Bruno R, Del Fante C, Seminari E, Casari S, et al. Mortality reduction in 46 severe COVID-19 patients treated with hyperimmune plasma. A proof of concept single arm multicenter trial. Haematologica 2020 July 23 [Epub ahead of print];105:2834-40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Perotti C, Del Fante C, Baldanti F, Franchini M, Percivalle E, Vecchio Nepita E, et al. Plasma from donors recovered from the new coronavirus 2019 as therapy for critical patients with COVID-19 (COVID-19 plasma study): a multicentre study protocol. Internal and Emergency Medicine 2020;15(5):819-24. [DOI: 10.1007/s11739-020-02384-2] [DOI] [PMC free article] [PubMed] [Google Scholar]
Qiu 2020 {published data only}
- Qiu T, Wang J, Zhou J, Zou J, Chen Z, Ma X, et al. The report of two cases infection with novel coronavirus (2019-NCcoV) after kidney transplantation and the association literature analyzation. Chinese Journal of Organ Transplantation 2020;41(0):E004. [Google Scholar]
Rasheed 2020 {published data only}
- Rasheed AM, Fatak DF, Hashim HA, Maulood MF, Kabah KK, Almusawi YA, et al. The therapeutic effectiveness of convalescent plasma therapy on treating COVID-19 patients residing in respiratory care units in hospitals in Baghdad, Iraq. medRxiv [Preprint] 2020. [DOI: 10.1101/2020.06.24.20121905] [DOI] [PubMed]
RBR‐4vm3yy {published data only}
- RBR-4vm3yy. Effect of convalescent plasma in patients with severe COVID-19. www.ensaiosclinicos.gov.br/rg/RBR-4vm3yy/ (first received 11 May 2020).
RBR‐5r8gv8p {published data only}
- RBR-5r8gv8p. Clinical trial with convalescent plasma for COVID-19 therapy. REBEC (ensaiosclinicos.gov.br) (first received 07 July 2020).
Robbiani 2020 {published data only}
- Robbiani DF, Gaebler C, Muecksch F, Cetrulo LJ, Wang Z, Cho A, et al. Convergent antibody responses to SARS-CoV-2 infection in convalescent individuals. bioRxiv [Preprint] 2020. [DOI: 10.1101/2020.05.13.092619] [DOI] [PMC free article] [PubMed]
RPCEC00000323 {published data only}
- RPCEC00000323. Plasma treatment to asymptomatic patient with COVID-19 infection. rpcec.sld.cu/en/trials/RPCEC00000323-En (first received 03 July 2020).
Salazar 2020a {published data only}
- Salazar E, Christensen PA, Graviss EA, Nguyen DT, Castillo B, Chen J, et al. Treatment of COVID-19 patients with convalescent plasma reveals a signal of significantly decreased mortality. American Journal of Pathology 2020;11:11. [DOI] [PMC free article] [PubMed] [Google Scholar]
Salazar 2020b {published data only}
- Salazar E, Perez KK, Ashraf M, Chen J, Castillo B, Christensen PA, et al. Treatment of COVID-19 patients with convalescent plasma. American Journal of Pathology 2020 May 27 [Epub ahead of print];190:1680-90. [DOI: 10.1016/j.ajpath.2020.05.014] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Salazar E, Perez KK, Ashraf M, Chen J, Castillo B, Christensen PA, et al. Treatment of COVID-19 patients with convalescent plasma in Houston, Texas. medRxiv [Preprint] 2020. [DOI: 10.1101/2020.05.08.20095471] [DOI] [PMC free article] [PubMed]
Shen 2020 {published data only}
- Shen C, Wang Z, Zhao F, Yang Y, Li J, Yuan J, et al. Treatment of 5 critically ill patients with COVID-19 with convalescent plasma. JAMA 2020;323(16):1582-9. [DOI: 10.1001/jama.2020.4783] [DOI] [PMC free article] [PubMed] [Google Scholar]
Shi 2020 {published data only}
- Shi H, Zhou C, He P, Huang S, Duan Y, Wang X, et al. Successful treatment of plasma exchange followed by intravenous immunoglobulin in a critically ill patient with 2019 novel coronavirus infection. International Journal of Antimicrobial Agents 2020;56(2):105974. [DOI: 10.1016/j.ijantimicag.2020.105974] [DOI] [PMC free article] [PubMed] [Google Scholar]
Soleimani 2020 {published data only}
- Soleimani Z, Soleimani A. ADRS due to COVID-19 in midterm pregnancy: successful management with plasma transfusion and corticosteroids. Journal of Maternal-Fetal & Neonatal Medicine 2020 Jul 26 [Epub ahead of print];26:1-4. [DOI: 10.1080/14767058.2020.1797669] [DOI] [PubMed] [Google Scholar]
Taher 2020 {published data only}
- Taher A, Alalwan AA, Naser N, Alsegai O, Alaradi A. Acute kidney injury in COVID-19 pneumonia: a single-center experience in Bahrain. Cureus 2020;12(8):e9693. [DOI] [PMC free article] [PubMed] [Google Scholar]
Tan 2020 {published data only}
- Tan L, Kang X, Zhang B, Zheng S, Liu B, Yu T, et al. A special case of COVID-19 with long duration of viral shedding for 49 days. medRxiv [Preprint] 2020. [DOI: 10.1101/2020.03.22.20040071] [DOI] [PMC free article] [PubMed]
Tu 2020 {published data only}
- Tu Y, Wu X, Liu F, Wang J, Luo Y, Cai Z, et al. Two clinical cases of novel coronavirus pneumonia (NCP) in renal transplant recipients. Chinese Journal of Organ Transplantation 2020;41(0):E005. [Google Scholar]
Wang 2020 {published data only}
- Wang Y, Zhang Y, Yu Q, Zhu K. Convalescent plasma coupled with medications for the treatment of a severe COVID-19 patient: drugs analysis and pharmaceutical care based on the newly established guidelines for COVID-19 remedy. Frontiers in Pharmacology 2020;11:966. [DOI: 10.3389/fphar.2020.00966] [DOI] [PMC free article] [PubMed] [Google Scholar]
Wright 2020 {published data only}
- Wright Z, Bersabe A, Eden R, Cap A. Successful use of COVID-19 convalescent plasma in a patient recently treated for follicular lymphoma. Clinical Lymphoma, Myeloma and Leukemia 2020 Jun 25 [Epub ahead of print];21(1):66-8. [DOI: 10.1016/j.clml.2020.06.012] [DOI] [PMC free article] [PubMed] [Google Scholar]
Xia 2020 {published data only}
- Xia X, Li K, Wu L, Wang Z, Zhu M, Huang B, et al. Improved clinical symptoms and mortality on severe/critical COVID-19 patients utilizing convalescent plasma transfusion. Blood 2020;23:23. [DOI] [PMC free article] [PubMed] [Google Scholar]
Xie 2020 {published data only}
- Xie Y, Cao S, Li Q, Chen E, Dong H, Zhang W, et al. Effect of regular intravenous immunoglobulin therapy on prognosis of severe pneumonia in patients with COVID-19. Journal of Infection 2020;81(2):318-56. [DOI: 10.1016/j.jinf.2020.03.044] [DOI] [PMC free article] [PubMed] [Google Scholar]
Xu 2020b {published data only}
- Xu TM, Lin B, Chen C, Liu LG, Xue Y. Non-optimal effectiveness of convalescent plasma transfusion and hydroxychloroquine in treating COVID-19: a case report. Virology Journal 2020;17(1):80. [DOI] [PMC free article] [PubMed] [Google Scholar]
Yang 2020 {published data only}
- Yang X, Sui Y, Liu F, Kang Z, Wu S, Zhao J, et al. Clinical characteristics and convalescent plasma therapy in severe and critically ill COVID-19 patients. Social Science Research Network 2020 May 5 [Preprint]. [DOI: 10.2139/ssrn.3576894] [DOI] [Google Scholar]
Ye 2020 {published data only}
- Ye M, Fu D, Ren Y, Wang F, Wang D, Zhang F, et al. Treatment with convalescent plasma for COVID-19 patients in Wuhan, China. Journal of Medical Virology 2020 April 15 [Epub ahead of print];92:1890-901. [DOI: 10.1002/jmv.25882] [DOI] [PMC free article] [PubMed] [Google Scholar]
Zeng 2020 {published data only}
- Zeng QL, Yu ZJ, Gou JJ, Li GM, Ma SH, Zhang GF, et al. Effect of convalescent plasma therapy on viral shedding and survival in COVID-19 patients. Journal of Infectious Diseases 2020;222(1):38-43. [DOI: 10.1093/infdis/jiaa228] [DOI] [PMC free article] [PubMed] [Google Scholar]
Zhang 2020a {published data only}
- Zhang B, Liu S, Tan T, Huang W, Dong Y, Chen L, et al. Treatment with convalescent plasma for critically ill patients with severe acute respiratory syndrome coronavirus 2 infection. Chest 2020;158(1):e9-e13. [DOI] [PMC free article] [PubMed] [Google Scholar]
Zhang 2020b {published data only}
- Zhang B, Liu S, Tan T, Huang W, Dong Y, Chen L, et al. Treatment with convalescent plasma for critically ill patients with SARS-CoV-2 infection. Chest 2020 Mar 31 [Epub ahead of print];158:e9-e13. [DOI: 10.1016/j.chest.2020.03.039] [DOI] [PMC free article] [PubMed] [Google Scholar]
Zhang 2020c {published data only}
- Zhang L, Pang R, Xue X, Bao J, Ye S, Dai Y, et al. Anti-SARS-CoV-2 virus antibody levels in convalescent plasma of six donors who have recovered from COVID-19. Aging 2020 Apr 22 [Epub ahead of print];12:6536–42. [DOI: 10.18632/aging.103102] [DOI] [PMC free article] [PubMed] [Google Scholar]
References to studies awaiting assessment
CTRI/2020/05/025299 {published data only}
- CTRI/2020/05/025299. Convalescent plasma to limit coronavirus associated complications: an open label clinical study of anti-SARS-CoV-2 plasma in hospitalized patients with COVID-19. Available from www.ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=43752 (first received 21 May 2020).
CTRI/2020/05/025328 {published data only}
- CTRI/2020/05/025328. Study to assess the safety and efficacy of convalescent plasma on outcome of COVID-19 associated complications. Available from ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=43703 (first received 23 May 2020).
CTRI/2020/06/025803 {published data only}
- CTRI/2020/06/025803. Effect of convalescent plasma in COVID-19 patients. Available from ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=44478 (first received 11 June 2020).
EUCTR2020‐001860‐27‐GB {published data only}
- EUCTR2020-001860-27-GB. AGILE: seamless phase I/IIa platform for the rapid evaluation of candidates for COVID-19 treatment. Available from https://www.clinicaltrialsregister.eu/ctr-search/trial/2020-001860-27/GB (first retrieved 14 June 2021).
IRCT20120215009014N353 {published data only}
- IRCT20120215009014N353. Effect of plasma of patients recovered from COVID-19 versus control group on treatment of COVID-19: a randomized clinical trial. Available from www.irct.ir/trial/47501 (first received 27 April 2020).
IRCT20150808023559N21 {published data only}
- IRCT20150808023559N21. The effect of convalescent plasma therapy on patients with 19-COVID. Available from en.irct.ir/trial/47594 (first received 9 May 2020).
IRCT20200404046948N1 {published data only}
- IRCT20200404046948N1. Efficacy and safety of convalescent plasma in the treatment of COVID-19. Available from en.irct.ir/trial/46973 (first received 15 April 2020).
IRCT20200413047056N1 {published data only}
- IRCT20200413047056N1. Comparison between the efficacy of intravenous immunoglobulin and convalescent plasma in COVID-19. Available from en.irct.ir/trial/47212 (first received 17 April 2020).
IRCT20200501047258N1 {published data only}
- IRCT20200501047258N1. Effects of convalescent plasma in COVID-19. Available from en.irct.ir/trial/47629 (first received 04 May 2020).
IRCT20200503047281N1 {published data only}
- IRCT20200503047281N1. Evaluation of convalescent plasma therapy for COVID-19 patients. Available from en.irct.ir/trial/47632 (first received 25 July 2020).
IRCT20201004048922N1 {published data only}
- IRCT20201004048922N1. Evaluation of the effectiveness of intravenous infusion of human COVID-19 hyperimmune plasma with specific antibody titer in hospitalized patients with COVID-19: a randomized clinical trial. Available from www.irct.ir/trial/51443 (first received 19 July 2020).
NCT04315948 {published data only}
- NCT04315948. Trial of treatments for COVID-19 in hospitalized adults (DisCoVeRy). Available from clinicaltrials.gov/ct2/show/NCT04315948 (first received 20 March 2020).
NCT04332835 {published data only}
- NCT04332835. Convalescent plasma for patients with COVID-19: a randomized, open label, parallel, controlled clinical study. Available from clinicaltrials.gov/show/NCT04332835 (first received 3 April 2020).
NCT04345991 {published data only}
- NCT04345991. Efficacy of convalescent plasma to treat COVID-19 patients, a nested trial in the CORIMUNO-19 cohort. Available from clinicaltrials.gov/show/NCT04345991 (first received 15 April 2020).
NCT04358783 {published data only}
- NCT04358783. Convalescent plasma compared to the best available therapy for the treatment of SARS-CoV-2 pneumonia. Available from clinicaltrials.gov/show/NCT04358783 (first received 24 April 2020).
NCT04361253 {published data only}
- NCT04361253. Evaluation of SARS-CoV-2 (COVID-19) antibody-containing plasma therapy. Available from clinicaltrials.gov/show/NCT04361253 (first received 24 April 2020).
NCT04362176 {published data only}
- NCT04362176. Passive immunity trial of Nashville II. Available from clinicaltrials.gov/show/NCT04362176 (first received 24 April 2020).
- Self WH, Stewart TG, Wheeler AP, Atrouni WE, Bistran-Hall AJ, Casey JD et al. Passive immunity trial for our nation (PassITON): study protocol for a randomized placebo-control clinical trial evaluating COVID-19 convalescent plasma in hospitalized adults (preprint). Available from www.researchsquare.com/article/rs-227796/v1 [Preprint]. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
NCT04374526 {published data only}
- NCT04374526. Early transfusIon of convalescent plasma in elderly COVID-19 patients to prevent disease progression. Available from clinicaltrials.gov/show/NCT04374526 (first received 5 May 2020).
- Teofili L, Landolfi R, Cingolani A, Antinori A, Vecchiet J, Sanguinetti M, et al. Early transfusion of convalescent plasma in older patients with COVID-19 to prevent disease progression: a structured summary of a study protocol for a randomised controlled trial. Trials 2020;21(1):875. [DOI: 10.1186/s13063-020-04821-1] [DOI] [PMC free article] [PubMed] [Google Scholar]
NCT04385199 {published data only}
- NCT04385199. Convalescent plasma for patients with COVID-19. Available from clinicaltrials.gov/show/NCT04385199 (first received 12 May 2020).
NCT04405310 {published data only}
- NCT04405310. Convalescent plasma of COVID-19 to treat SARS-COV-2 a randomized double blind 2 center trial (CPC-SARS). Available from clinicaltrials.gov/show/NCT04405310 (first received 28 May 2020).
NCT04425915 {published data only}
- NCT04425915. Efficacy of convalescent plasma therapy in patients with COVID-19. Available from clinicaltrials.gov/show/NCT04425915 (first received 11 June 2020).
NCT04428021 {published data only}
- NCT04428021. Standard or convalescent plasma in patients with recent onset of COVID-19 respiratory failure. Available from clinicaltrials.gov/show/NCT04428021 (first received 11 June 2020).
NCT04442958 {published data only}
- NCT04442958. Effectiveness of convalescent immune plasma therapy. Available from clinicaltrials.gov/show/NCT04442958 (first received 23 June 2020).
NCT04468009 {published data only}
- NCT04468009. Treatment of critically ill patients with COVID-19 with convalescent plasma. Available from clinicaltrials.gov/show/NCT04468009 (first received 13 July 2020).
NCT04497324 {published data only}
- NCT04497324. Peruconplasma: evaluating the use of convalescent plasma as management of COVID-19. Available from clinicaltrials.gov/ct2/show/NCT04497324 (first received 04 August 2020).
- PER-016-20. PERUCONPLASMA: Evaluating the use of convalescent plasma as managment of COVID-19. Available from www.ins.gob.pe/ensayosclinicos/rpec/recuperarECPBNuevoEN.asp?numec=016-20 (first received 01 June 2020).
- Soto A, Krapp F, Vargas A, Cabrejos L, Argumanis E, Garcia P L, et al. Randomized clinical trial to evaluate safety and efficacy of convalescent plasma use among hospitalized patients with COVID-19 (PERUCONPLASMA): a structured summary of a study protocol for a randomized controlled trial. Trials 2021;22:342. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
NCT04501978 {published data only}
- NCT04501978. ACTIV-3: therapeutics for inpatients with COVID-19 (TICO). Available from clinicaltrials.gov/ct2/show/NCT04501978 (first received 6 August 2020).
NCT04521309 {published data only}
- Ali S, Luxmi S, Anjum F, Muhaymin SM, Uddin SM, Ali A, et al. Hyperimmune anti-COVID-19 IVIG (C-IVIG) therapy for passive immunization of severe and critically ill COVID-19 patients: a structured summary of a study protocol for a randomised controlled trial. Trials 2020;21(905). [DOI] [PMC free article] [PubMed] [Google Scholar]
- SARS-CoV-2 antibodies based IVIG therapy for COVID-19 patients. Available from clinicaltrials.gov/ct2/show/NCT04521309 (first received 20 August 2020).
NCT04539275 {published data only}
- NCT04539275. COVID-19 (VA CURES-1). Available from clinicaltrials.gov/show/NCT04539275 (first received 04 September 2020).
NCT04542967 {published data only}
- NCT04542967. Study on the safety and efficacy of convalescent plasma in patients with severe COVID-19 disease. Available from clinicaltrials.gov/show/NCT04542967 (first received 09 September 2020).
NCT04547127 {published data only}
- Ferrer Roca R, Llamas P, Manez R, Galban C, Quintana M, Sanchez-Garcia M, et al. Design of a study to evaluate the safety and efficacy of convalescent plasma to treat COVID-19 in critically ill patients. Intensive Care Medicine Experimental. Conference: 33rd European Society of Intensive Care Medicine Annual Congress, ESICM 2020;8 Suppl 2. [Google Scholar]
- NCT04547127. A study to evaluate safety and efficacy of convalescent methylene blue treated (MBT) plasma from donors recovered from coronavirus disease 2019 (COVID-19). Available from clinicaltrials.gov/show/NCT04547127 (first received 14 September 2020).
NCT04649879 {published data only}
- NCT04649879. Convalescent plasma for treatment of COVID-19: an open randomised controlled trial. Available from clinicaltrials.gov/show/NCT04649879 (first received 02 December 2020).
NCT04681430 {published data only}
- Keitel V, Jensen B, Feldt T, Fischer JC, Bode JG, Matuschek C, et al. Reconvalescent plasma/camostat mesylate in early SARS-CoV-2 Q-PCR positive high-risk individuals (RES-Q-HR): a structured summary of a study protocol for a randomized controlled trial. Trials 2021;22:343. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
- NCT04681430. Reconvalescent plasma/camostat mesylate early in SARS-CoV-2 Q-PCR (COVID-19) positive high-risk individuals (RES-Q-HR). Available from clinicaltrials.gov/ct2/show/NCT04681430 (first received 23 December 2020).
NCT04801940 {published data only}
- NCT04801940. Helping alleviate the longer-term consequences of COVID-19 (HEAL-COVID). Available from clinicaltrials.gov/ct2/show/NCT04801940 (first retrieved 14 June 2021).
References to ongoing studies
ChiCTR2000030010 {published data only}
- ChiCTR2000030010. A randomized, double-blind, parallel-controlled, trial to evaluate the efficacy and safety of anti-SARS-CoV-2 virus inactivated plasma in the treatment of severe novel coronavirus pneumonia patients (COVID-19). Available from www.chictr.org.cn/showproj.aspx?proj=49777 (first received 19 February 2020).
ChiCTR2000030179 {published data only}
- ChiCTR2000030179. Experimental study of novel coronavirus pneumonia rehabilitation plasma therapy severe novel coronavirus pneumonia (COVID-19). Available from www.chictr.org.cn/showproj.aspx?proj=50059 (first received 24 February 2020).
ChiCTR2000030627 {published data only}
- ChiCTR2000030627. Study on the application of convalescent plasma therapy in severe COVID-19. Available from www.chictr.org.cn/showproj.aspx?proj=50727 (first received 8 March 2020).
ChiCTR2000030702 {published data only}
- ChiCTR2000030702. Convalescent plasma for the treatment of common COVID-19: a prospective randomized controlled trial. Available from www.chictr.org.cn/showproj.aspx?proj=50537 (first received 10 March 2020).
ChiCTR2000030929 {published data only}
- ChiCTR2000030929. A randomized, double-blind, parallel-controlled trial to evaluate the efficacy and safety of anti-SARS-CoV-2 virus inactivated plasma in the treatment of severe novel coronavirus pneumonia (COVID-19). Available from www.chictr.org.cn/showproj.aspx?proj=50696 (first received 17 March 2020).
CTRI/2020/04/024915 {published data only}
- CTRI/2020/04/024915. A phase II, open label, randomized controlled trial to assess the safety and efficacy of convalescent plasma to limit COVID-19 associated complications. Available from ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=43332 (first received 29 April 2020).
CTRI/2020/05/025346 {published data only}
- CTRI/2020/05/025346. A phase II, open label, randomized controlled trial to assess the safety and efficacy of convalescent plasma in severe COVID-19 patients. Available from ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=43005 (first received 5 May 2020).
CTRI/2020/06/026123 {published data only}
- CTRI/2020/06/026123. Plasma therapy in corona patients(Severe COVID-19). Available from ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=44667 (first received 24 June 2020).
EUCTR2020‐001632‐10 {published data only}
- EUCTR2020-001632-10. A randomized open label phase-II clinical trial with or without infusion of plasma from subjects after convalescence of SARS-CoV-2 infection in high-risk patients with confirmed severe SARS-CoV-2 disease. Available from www.clinicaltrialsregister.eu/ctr-search/trial/2020-001632-10/DE (first received 20 April 2020). [DOI] [PMC free article] [PubMed]
- Janssen M, Schakel U, Djuka Fokou U, Krisam J, Stermann J, Kriegsmann K, et al. A randomized open label phase-II clinical trial with or without infusion of plasma from subjects after convalescence of SARS-CoV-2 infection in high-risk patients with confirmed severe SARS-CoV-2 disease (RECOVER): a structured summary of a study protocol for a randomised controlled trial. Trials 2020;21(1):828. [DOI] [PMC free article] [PubMed] [Google Scholar]
EUCTR2020‐001936‐86 {published data only}
- EUCTR2020-001936-86. A prospective, randomized, open label Phase 2 clinical trial to evaluate superiority of anti-SARS-CoV-2 convalescent plasma versus standard-of-care in hospitalized patients with mild COVID-19. Available from www.clinicaltrialsregister.eu/ctr-search/trial/2020-001936-86/DE (first received 20 August 2020).
EUCTR2020‐002122‐82 {published data only}
- EUCTR2020-002122-82. Prospective open-label randomized controlled phase 2b clinical study in parallel groups for the assessment of efficacy and safety of immune therapy with COVID-19 convalescent plasma plus standard treatment alone of subjects with severe COVID-19. Available from www.clinicaltrialsregister.eu/ctr-search/trial/2020-002122-82/DE (first received 07 May 2020).
EUCTR2020‐005410‐18 {published data only}
- EUCTR2020-005410-18. Multicentre, randomized, double-blind, placebo-controlled, non-commercial clinical trial to evaluate the efficacy and safety of specific anti-SARS-CoV-2 immunoglobulin in the treatment of COVID-19. Available from www.clinicaltrialsregister.eu/ctr-search/trial/2020-005410-18/PL.
ISRCTN49832318 {published data only}
- SURCOVID trial: a randomized controlled trial using convalescent plasma early during moderate COVID-19 disease course in Suriname. Ongoing study. 1 June 2021. Contact author for more information.
jRCTs031200374 {published data only}
- jRCTs031200374. An open-label, randomized, controlled trial to evaluate the efficacy of convalescent plasma therapy for COVID-19. https://pesquisa.bvsalud.org/global-literature-on-novel-coronavirus-2019-ncov/resource/en/ictrp-JPRN-jRCTs031200374 (first received 24 February 2021).
NCT04333251 {published data only}
- NCT04333251. Evaluating convalescent plasma to decrease coronavirus associated complications. A phase I study comparing the efficacy and safety of high-titer anti-Sars-CoV-2 plasma vs best supportive care in hospitalized patients with interstitial pneumonia due to COVID-19. Available from clinicaltrials.gov/show/NCT04333251 (first received 3 April 2020).
NCT04345289 {published data only}
- EUCTR2020-001367-88-DK. Efficacy and safety of novel treatment options for adults with COVID-19 pneumonia. Available from apps.who.int/trialsearch/Trial2.aspx?TrialID=EUCTR2020-001367-88-DK (first received 14 April 2020).
- NCT04345289. Efficacy and safety of novel treatment options for adults with COVID-19 pneumonia (CCAP). Available from clinicaltrials.gov/show/NCT04345289 (first received 14 April 2020).
NCT04372979 {published data only}
- NCT04372979. Efficacy of convalescent plasma therapy in the early care of COVID-19 patients. Available from clinicaltrials.gov/show/NCT04372979 (first received 04 May 2020).
NCT04374487 {published data only}
- NCT04374487. A phase II, open label, randomized controlled trial to assess the safety and efficacy of convalescent plasma to limit COVID-19 associated complications. Available from clinicaltrials.gov/show/NCT04374487 (first received 5 May 2020).
NCT04376788 {published data only}
- NCT04376788. Exchange transfusion versus plasma from convalescent patients with methylene blue in patients with COVID-19. Available from clinicaltrials.gov/show/NCT04376788 (first received 6 May 2020).
NCT04380935 {published data only}
- NCT04380935. Effectiveness and safety of convalescent plasma therapy on COVID-19 patients with acute respiratory distress syndrome. Available from clinicaltrials.gov/show/NCT04380935 (first received 6 May 2020).
NCT04385043 {published data only}
- NCT04385043. Hyperimmune plasma in patients with COVID-19 severe infection. Available from clinicaltrials.gov/show/NCT04385043 (first received 12 May 2020).
NCT04385186 {published data only}
- NCT04385186. Inactivated convalescent plasma as a therapeutic alternative in patients COVID-19. Available from clinicaltrials.gov/show/NCT04385186 (first received 12 May 2020).
NCT04388410 {published data only}
- NCT04388410. Safety and efficacy of convalescent plasma transfusion for patients with SARS-CoV-2 infection. Available from clinicaltrials.gov/show/NCT04388410 (first received 14 May 2020).
NCT04390503 {published data only}
- NCT04390503. Convalescent plasma for COVID-19 close contacts. Available from clinicaltrials.gov/ct2/show/NCT04390503 (first received 15 May 2020).
NCT04391101 {published data only}
- NCT04391101. Convalescent plasma for the treatment of severe SARS-CoV-2 (COVID-19). Available from clinicaltrials.gov/show/NCT04391101 (first received 18 May 2020).
NCT04403477 {published data only}
- Chowdhury FR, Hoque A, Chowdhury FU, Amin MR, Rahim A, Rahman MM, et al. Convalescent plasma transfusion therapy in severe COVID-19 patients- a safety, efficacy and dose response study: a structured summary of a study protocol of a phase II randomized controlled trial. Trials 2020;21(1):883. [DOI: 10.1186/s13063-020-04734-z] [DOI] [PMC free article] [PubMed] [Google Scholar]
- NCT04403477. Convalescent plasma therapy in severe COVID-19 infection. Available from clinicaltrials.gov/show/NCT04403477 (first received 27 May 2020).
NCT04415086 {published data only}
- NCT04415086. Treatment of patients with COVID-19 with convalescent plasma. Available from clinicaltrials.gov/show/NCT04415086 (first received 4 June 2020).
NCT04418518 {published data only}
- NCT04418518. A trial of convalescent plasma for hospitalized adults with acute COVID-19 respiratory illness. Available from clinicaltrials.gov/show/NCT04418518 (first received 5 June 2020).
NCT04425837 {published data only}
- NCT04425837. Effectiveness and safety of convalescent plasma in patients with high-risk COVID-19. Available from clinicaltrials.gov/show/NCT04425837 (first received 11 June 2020).
NCT04438057 {published data only}
- NCT04438057. Evaluating the efficacy of convalescent plasma in symptomatic outpatients infected with COVID-19. Available from clinicaltrials.gov/show/NCT04438057 (first received 18 June 2020).
NCT04442191 {published data only}
- NCT04442191. Convalescent plasma as a possible treatment for COVID-19. Available from clinicaltrials.gov/show/NCT04442191 (first received 22 June 2020).
NCT04452812 {published data only}
- NCT04452812. Statistical and epidemiological study based on the use of convalescent plasma for the management of patients with COVID-19. Available from clinicaltrials.gov/show/NCT04452812 (first received 30 June 2020).
NCT04456413 {published data only}
- NCT04456413. Convalescent plasma as treatment for subjects with early COVID-19 infection. Available from clinicaltrials.gov/show/NCT04456413 (first received 2 July 2020).
NCT04483960 {published data only}
- NCT04483960. Australasian COVID-19 trial (ASCOT). Available from clinicaltrials.gov/show/NCT04483960 (first received 23 July 2020).
NCT04521036 {published data only}
- NCT04521036. Convalescent plasma for COVID-19 patients (CPCP). Available from clinicaltrials.gov/show/NCT04521036 (first received 20 August 2020).
NCT04528368 {published data only}
- NCT04528368. Convalescent plasma for treating patients with COVID-19 pneumonia without indication of ventilatory support. Available from clinicaltrials.gov/show/NCT04528368 (first received 27 August 2020).
NCT04558476 {published data only}
- Misset B, Hoste E, Donneau AF, Grimaldi D, Meyfroidt G, Moutschen M, et al. A multicenter randomized trial to assess the efficacy of CONvalescent plasma therapy in patients with Invasive COVID-19 and acute respiratory failure treated with mechanical ventilation: the CONFIDENT trial protocol. BMC Pulmonary Medicine 2020;20(317). [DOI] [PMC free article] [PubMed] [Google Scholar]
- Misset B, Hoste E, Donneau AF, Grimaldi D, Meyfroidt G, Moutschen M, et al. Correction to: A multicenter randomized trial to assess the efficacy of CONvalescent plasma therapy in patients with Invasive COVID-19 and acute respiratory failure treated with mechanical ventilation: the CONFIDENT trial protocol. BMC Pulmonary Medicine July 2021;248. [DOI: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
- NCT04558476. Efficacy of convalescent plasma in patients with COVID-19 treated with mechanical ventilation. Available from clinicaltrials.gov/show/NCT04558476 (first received 22 September 2020).
NCT04567173 {published data only}
- NCT04567173. Convalescent plasma as adjunctive therapy for hospitalized patients with COVID-19. Available from clinicaltrials.gov/show/NCT04567173 (first received 28 September 2020).
NCT04634422 {published data only}
- NCT04634422. Plasma exchange (PLEX) and convalescent plasma (CCP) in COVID-19 patients with multiorgan failure. Available from clinicaltrials.gov/show/NCT04634422 (first received 18 November 2020).
NCT04712344 {published data only}
- NCT04712344. Assessment of efficacy and safety of therapy with COVID-19 convalescent plasma in subjects with severe COVID-19 (IPCO). Available from clinicaltrials.gov/ct2/show/NCT04712344 (first received 15 January 2021).
NCT04730401 {published data only}
- NCT04730401. Convalescent plasma in the treatment of COVID-19 (CP_COVID-19). Available from clinicaltrials.gov/ct2/show/NCT04730401 (first received 27 January 2021).
NCT04803370 {published data only}
- NCT04803370. Efficacy of reinforcing standard therapy in COVID-19 patients with repeated transfusion of convalescent plasma. Available from clinicaltrials.gov/ct2/show/NCT04803370 (first received 14 June 2021).
NCT05077930 {published data only}
- NCT05077930. Convalescent plasma therapy for hospitalized patients with COVID-19. https://clinicaltrials.gov/ct2/show/NCT05077930.
NL8633 {published data only}
- NL8633. A randomized, double-blinded clinical trial of convalescent plasma compared to standard plasma for treatment of hospitalized non-ICU patients with COVID-19 infections. Available from trialregister.nl/trial/8633 (first received 13 May 2020).
PACTR202006760881890 {published data only}
- PACTR202006760881890. Lagos COVID-19 convalescent plasma trial (LACCPT). Available from pactr.samrc.ac.za/TrialDisplay.aspx?TrialID=12168.
PACTR202007653923168 {published data only}
- PACTR202007653923168. A clinical trial comparing use of convalescent plasma therapy plus standard treatment to standard treatment alone in patients with severe COVID-19 infection. Available from pactr.samrc.ac.za/TrialDisplay.aspx?TrialID=11047 (first received 16 July 2020).
PER‐013‐20 {published data only}
- PER-013-20. Convalescent plasma as treatment for COVID-19. Available from ins.gob.pe/ensayosclinicos/rpec/recuperarECPBNuevoEN.asp?numec=013-20 (first received 25 June 2020).
PER‐060‐20 {published data only}
- PER-060-20. Randomized phase 2 clinical trial to evaluate the safety and efficacy of the use of plasma from convalescent patients with the new coronavirus disease (COVID-19). Available from who.int/trialsearch/Trial2.aspx?TrialID=PER-060-20 (first received 21 September 2020).
RBR‐7jqpnw {published data only}
- RBR-7jqpnw. Effect of COVID-19 convalescent plasma produced by HEMOPE: a randomized study, with a comparative group in several centers. Available from ensaiosclinicos.gov.br/rg/RBR-7jqpnw/ (first received 11 May 2020).
Additional references
Adjei 2022
- Adjei S, Hong K, Molinari NM, Bull-Otterson L, Ajani UA, Gundlapalli AV, et al. Mortality risk among patients hospitalized primarily for COVID-19 during the omicron and delta variant pandemic periods — United States, April 2020–June 2022. Morbidity and Mortality Weekly Report 2022;71(37):1182–9. [DOI: 10.15585/mmwr.mm7137a4] [DOI] [PMC free article] [PubMed] [Google Scholar]
Bao 2020a
- Bao L, Deng W, Gao H, Xiao C, Liu J, Xue J, et al. Reinfection could not occur in SARS-CoV-2 infected rhesus macaques. bioRxiv [Preprint] 2020. [DOI: 10.1101/2020.03.13.990226] [DOI]
Bartelt 2022
- Bartelt LA, Markmann A, Nelson B, Keys J, Root H, Henderson HI, et al. Outcomes of convalescent plasma with defined high- versus lower-neutralizing antibody titers against SARS-CoV-2 among hospitalized patients: CoronaVirus Inactivating Plasma (CoVIP), double-blind phase 2 study. medRxiv [Preprint] 2022. [DOI: 10.1101/2022.04.29.22274387] [DOI] [PMC free article] [PubMed] [Google Scholar]
Beigel 2017
- Beigel JH, Tebas P, Elie-Turenne MC, Bajwa E, Bell TE, Cairns B, et al. Immune plasma for the treatment of severe influenza: an open-label, multicentre, phase 2 randomised study. Lancet Respiratory Medicine 2017;5(6):500-11. [DOI: 10.1016/S2213-2600(17)30174-1] [DOI] [PMC free article] [PubMed] [Google Scholar]
Beigel 2019
- Beigel JH, Aga E, Elie-Turenne M-C, Cho J, Tebas P, Clark CL, et al. Anti-influenza immune plasma for the treatment of patients with severe influenza A: a randomised, double-blind, phase 3 trial. Lancet Respiratory Medicine 2019;7(11):941-50. [DOI: 10.1016/S2213-2600(19)30199-7] [DOI] [PMC free article] [PubMed] [Google Scholar]
Beigel 2020
- Beigel JH, Tomashek KM, Dodd LE, Mehta AK, Zingman BS, Kalil AC, et al. Remdesivir for the treatment of COVID-19 — preliminary report. New England Journal of Medicine 2020;383(10):992-4. [DOI: 10.1056/NEJMoa2007764] [DOI] [PubMed] [Google Scholar]
Bloch 2020
- Bloch EM, Shoham S, Casadevall A, Sachais BS, Shaz B, Winters JL, et al. Deployment of convalescent plasma for the prevention and treatment of COVID-19. Journal of Clinical Investigation 2020;130(6):2757-65. [DOI] [PMC free article] [PubMed] [Google Scholar]
Brandal 2021
- Brandal LT, MacDonald E, Veneti L, Ravlo T, Lange H, Naseer U, et al. Outbreak caused by the SARS-CoV-2 omicron variant in Norway, November to December 2021. European Surveillance 2021;26(50):2101147. [DOI: 10.2807/1560-7917.ES.2021.26.50.2101147] [DOI] [PMC free article] [PubMed] [Google Scholar]
Casadevall 2020
- Casadevall A, Pirofski L. The convalescent sera option for containing COVID-19. Journal of Clinical Investigation 2020;130(4):1545-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
CDC 2020
- Centers for Disease Control and Prevention (CDC). Interim clinical guidance for management of patients with confirmed coronavirus disease (COVID-19). Available from www.cdc.gov/coronavirus/2019-ncov/hcp/clinical-guidance-management-patients.html (accessed 22 March 2021).
CDC 2022
- Centers for Disease Control and Prevention. Vaccines for COVID-19. Available from https://www.cdc.gov/coronavirus/2019-ncov/vaccines (accessed 25 May 2022).
Chavez‐MacGregor 2022
- Chavez-MacGregor M, Lei X, Zhao H, Scheet P, Giordano SH. Evaluation of COVID-19 mortality and adverse outcomes in US patients with or without cancer. JAMA Oncology 2022;8(1):69-78. [DOI: 10.1001/jamaoncol.2021.5148] [DOI] [PMC free article] [PubMed] [Google Scholar]
Chen 2020a
- Chen L, Xiong J, Bao L, Shi Y. Convalescent plasma as a potential therapy for COVID-19. Lancet Infectious Diseases 2020;20(4):398-400. [DOI] [PMC free article] [PubMed] [Google Scholar]
Chun 2016
- Chun S, Chung CR, Ha YE, Han TH, Ki CS, Kang ES, et al. Possible transfusion-related acute lung injury following convalescent plasma transfusion in a patient with Middle East respiratory syndrome. Annals of Laboratory Medicine 2016;36(4):393-5. [DOI: 10.3343/alm.2016.36.4.393] [DOI] [PMC free article] [PubMed] [Google Scholar]
COMET 2020
- Core outcome set developers’ response to COVID-19 (2nd April 2020). Available from www.comet-initiative.org/Studies/Details/1538 (accessed 9 April 2020).
Covidence [Computer program]
- Covidence. Version accessed 20 April 2020. Melbourne, Australia: Veritas Health Innovation, 2020. Available at covidence.org.
Deeks 2022
- Deeks JJ, Higgins JP, Altman DG, editor(s). Chapter 10: Analysing data and undertaking meta-analyses. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al, editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.3 (updated February 2022). Cochrane, 2022. Available from www.training.cochrane.org/handbook.
Driggin 2020
- Driggin E, Madhavan MV, Bikdeli B, Chuich T, Laracy J, Biondi-Zoccai G, et al. Cardiovascular considerations for patients, health care workers, and health systems during the COVID-19 pandemic. Journal of the American College of Cardiology 2020;75(18):2352-71. [DOI] [PMC free article] [PubMed] [Google Scholar]
Eibl 2008
- Eibl MM. History of immunoglobulin replacement. Immunology and Allergy Clinics of North America 2008;28(4):737-64. [DOI: 10.1016/j.iac.2008.06.004] [DOI] [PubMed] [Google Scholar]
Eldridge 2016
- Eldridge S, Campbell M, Campbell M, Dahota A, Giraudeau B, Higgins JT, et al. Revised Cochrane risk of bias tool for randomized trials (RoB 2.0) additional considerations for cluster-randomized trials. Available from www.riskofbias.info/welcome/rob-2-0-tool/archive-rob-2-0-cluster-randomized-trials-2016 (accessed 22 March 2021).
EPOC 2017
- Cochrane Effective Practice and Organisation of Care (EPOC). What study designs can be considered for inclusion in an EPOC review and what should they be called? EPOC Resources for review authors. Available from epoc.cochrane.org/resources/epoc-resources-review-authors (accessed 22 March 2021).
FDA 2020
- US Food and Drug Administration. Recommendations for investigational COVID-19 convalescent plasma. Available from www.fda.gov/vaccines-blood-biologics/investigational-new-drug-ind-or-device-exemption-ide-process-cber/recommendations-investigational-covid-19-convalescent-plasma (accessed 22 March 2021).
FDA 2021
- US Food and Drug Administration. Recommendations for investigational COVID-19 convalescent plasma. Available from www.fda.gov/vaccines-blood-biologics/investigational-new-drug-ind-or-device-exemption-ide-process-cber/recommendations-investigational-covid-19-convalescent-plasma (accessed 22 March 2021).
Fernández‐Lázaro 2022
- Fernández-Lázaro D, Ortega CD, Sánchez-Serrano N, Beddar Chaib F, Jerves Donoso D, Jiménez-Callejo E, et al. Convalescent plasma therapy, therapeutic formulations of repurposed drugs in 20th century epidemics against COVID-19: a systematic review. Pharmaceutics 2022;14(5):1020. [DOI: 10.3390/pharmaceutics14051020] [DOI] [PMC free article] [PubMed] [Google Scholar]
Focosi 2021
- Focosi D, Franchini M. Impact of pathogen-reduction technologies on COVID-19 convalescent plasma potency. Transfusion Clinique et Biologique 2021;28(2):132-4. [DOI: 10.1016/j.tracli.2021.02.005] [DOI] [PMC free article] [PubMed] [Google Scholar]
Fung 2020
- Fung M, Babik JM. COVID-19 in immunocompromised hosts: what we know so far. Clinical Infectious Diseases 2020;72:340-50. [DOI: 10.1093/cid/ciaa863] [DOI] [PMC free article] [PubMed] [Google Scholar]
GRADEpro GDT [Computer program]
- GRADEpro GDT. Version accessed 20 April 2020. Hamilton (ON): McMaster University (developed by Evidence Prime), 2020. Available at gradepro.org.
Guyatt 2011
- Guyatt GH, Oxman AD, Kunz R, Brozek J, Alonso-Coello P, Rind D, et al. GRADE guidelines 6. Rating the quality of evidence—imprecision. Journal of Clinical Epidemiology 2011;64(12):1283-93. [DOI: 10.1016/j.jclinepi.2011.01.012] [DOI] [PubMed] [Google Scholar]
Higgins 2003
- Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557-60. [DOI] [PMC free article] [PubMed] [Google Scholar]
Higgins 2011
- Higgins JP, Altman DG, Sterne JA, editor(s). Chapter 8: Assessing risk of bias in included studies. In: Higgins JP, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available from www.handbook.cochrane.org.
Higgins 2022
- Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.3 (updated February 2022). Cochrane, 2022. Available from www.training.cochrane.org/handbook 2022.
Higgins 2022a
- Higgins JP, Savović J, Page MJ, Elbers RG, Sterne JA. Chapter 8: Assessing risk of bias in a randomized trial. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al, editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.3 (updated February 2022). Cochrane, 2022. Available from training.cochrane.org/handbook 2022.
Higgins 2022b
- Higgins J, Eldridge S, Li T. Chapter 23: Including variants on randomised trials. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al, editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.3 (updated February 202s). Cochrane, 2022. Available from training.cochrane.org/handbook 2022.
Higgins 2022c
- Higgins JP, Li T, Deeks JJ, editor(s). Chapter 6: Choosing effect measures and computing estimates of effect. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al, editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.3 (updated February 2022). Cochrane, 2022. Available from training.cochrane.org/handbook 2022.
Ho 2005
- Ho MS, Chen WJ, Chen HY, Lin SF, Wang MC, Di J, et al. Neutralizing antibody response and SARS severity. Emerging Infectious Diseases 2005;11(11):1730-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
Horby 2020
- Horby P, Lim WS, Emberson J, Mafham M, Bell J, Linsell L, et al. Effect of dexamethasone in hospitalized patients with COVID-19: preliminary report. medRxiv [Preprint] 2020. [DOI: ] [Google Scholar]
Horby 2021a
- Horby PW, Pessoa-Amorim G, Peto L, Brightling CE, Sarkar R, Thomas K, et al. Tocilizumab in patients admitted to hospital with COVID-19 (RECOVERY): preliminary results of a randomised, controlled, open-label, platform trial. medRxiv [Preprint] 2021. [DOI: ] [Google Scholar]
Huang 2020
- Huang C, Wang Y, Li X, Ren L, Zhao J, Hu Y, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet 2020;395(10223):497-506. [DOI: 10.1016/S0140-6736(20)30183-5] [DOI] [PMC free article] [PubMed] [Google Scholar]
Iwasaki 2021
- Iwasaki Akiko. What reinfections mean for COVID-19. Lancet Infectious Diseases 2021;21(1):3-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
Izquierdo‐Dominguez 2020
- Izquierdo-Dominguez A, Rojas-Lechuga MJ, Mullol J, Alobid I. Olfactory dysfunction in the COVID-19 outbreak. Journal of Investigational Allergology & Clinical Immunology 2020;30(5):317-26. [DOI: 10.18176/jiaci.0567] [DOI] [PubMed] [Google Scholar]
Janiaud 2021
- Janiaud P, Axfors C, Schmitt AM, Gloy V, Ebrahimi F, Hepprich M, et al. Association of convalescent plasma treatment with clinical outcomes in patients with COVID-19: a systematic review and meta-analysis. JAMA 2021:e212747. [DOI: 10.1001/jama.2021.2747] [DOI] [PMC free article] [PubMed] [Google Scholar]
Johns Hopkins 2022
- Johns Hopkins University and Medicine. Mortality analyses. Available from https://coronavirus.jhu.edu/data/mortality (last accessed 21 September 2022).
Jorda 2022
- Jorda A, Kussmann M, Kolenchery N, Siller-Matula J M, Zeitlinger M, Jilma B, et al. Convalescent plasma treatment in patients with COVID-19: a systematic review and meta-analysis. Frontiers in Immunology 2022;13:817829. [DOI: 10.3389/fimmu.2022.817829] [DOI] [PMC free article] [PubMed] [Google Scholar]
Joyner 2021
- Joyner MJ, Carter RE, Senefeld JW, Klassen SA, Mills JR, Johnson PW, et al. Convalescent plasma antibody levels and the risk of death from COVID-19. New England Journal of Medicine 2021;384:1015-27. [DOI: 10.1056/NEJMoa2031893] [DOI] [PMC free article] [PubMed] [Google Scholar]
Kim 2020
- Kim D-H, Choe YJ, Jeong J-Y. Understanding and interpretation of case fatality rate of coronavirus disease 2019. Journal of Korean Medical Science 2020;35(12):e137. [DOI: 10.3346/jkms.2020.35.e137] [DOI] [PMC free article] [PubMed] [Google Scholar]
Klassen 2021
- Klassen SA, Senefeld JW, Johnson PW, Carter RE, Wiggins CC, Shoham S, et al. Evidence favoring the efficacy of convalescent plasma for COVID-19 therapy. medRxiv [Preprint] 2021. [DOI: 10.1101/2020.07.29.20162917] [DOI] [Google Scholar]
Kostin 2021
- Kostin AI, Lundgren MN, Bulanov AY, Ladygina EA, Chirkova KS, Gintsburg AL, et al. Impact of pathogen reduction methods on immunological properties of the COVID-19 convalescent plasma. Vox Sang 2021;116(6):665-72. [DOI: 10.1111/vox.13056] [DOI] [PMC free article] [PubMed] [Google Scholar]
Larrea 2022
- Larrea L, Castro E, Navarro L, Vera B, Francés-Gómez C, Sánchez-Sendra B, et al. Preservation of anti-SARS-CoV-2 neutralising antibodies in convalescent plasma after pathogen reduction with methylene blue and visible light. Blood Transfusion 2022;20(3):206-12. [DOI: 10.2450/2021.0136-21] [DOI] [PMC free article] [PubMed] [Google Scholar]
Lauer 2020
- Lauer SA, Grantz KH, Bi Q, Jones FK, Zheng Q, Meredith HR, et al. The incubation period of coronavirus disease 2019 (COVID-19) from publicly reported confirmed cases: estimation and application. Annals of Internal Medicine 2020;172(9):577-82. [DOI: 10.7326/M20-0504] [DOI] [PMC free article] [PubMed] [Google Scholar]
Lee 2022
- Lee AR, Wong SY, Chai LY, Lee SC, Lee MX, Muthiah MD, et al. Efficacy of COVID-19 vaccines in immunocompromised patients: systematic review and meta-analysis. BMJ 2022;376:e068632. [DOI: 2022;376:e068632] [DOI] [PMC free article] [PubMed] [Google Scholar]
Lefebvre 2022a
- Lefebvre C, Glanville J, Briscoe S, Featherstone R, Littlewood A, Marshall C, et al. Chapter 4: Searching for and selecting studies. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al, editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.3 (updated February 2022). Cochrane, 2022. Available from www.training.cochrane.org/handbook.
Lefebvre 2022b
- Lefebvre C, Glanville J, Briscoe S, Featherstone R, Littlewood A, Marshall C, et al. Technical Supplement to Chapter 4: Searching for and selecting studies. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.3 (updated February 2022). Cochrane, 2022. Available from www.training.cochrane.org/handbook. Available from www.training.cochrane.org/handbook 2022.
Li 2022
- Li T, Higgins JP, Deeks JJ, editor(s). Chapter 5: Collecting data. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al, editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.3 (updated February 2022). Cochrane, 2022. Available from www.training.cochrane.org/handbook.
Liang 2020
- Liang W, Guan W, Chen R, Wang W, Li J, Xu K, et al. Cancer patients in SARS-CoV-2 infection: a nationwide analysis in China. Lancet Oncology 2020;21(3):335-7. [DOI: 10.1016/S1470-2045(20)30096-6] [DOI] [PMC free article] [PubMed] [Google Scholar]
Luke 2006
- Luke TC, Kilbane EM, Jackson JL, Hoffman SL. Meta-analysis: convalescent blood products for Spanish influenza pneumonia: a future H5N1 treatment? Annals of Internal Medicine 2006;145(8):599-609. [DOI] [PubMed] [Google Scholar]
Mair‐Jenkins 2015
- Mair-Jenkins J, Saavedra-Campos M, Baillie JK, Cleary P, Khaw FM, Lim WS, et al. The effectiveness of convalescent plasma and hyperimmune immunoglobulin for the treatment of severe acute respiratory infections of viral etiology: a systematic review and exploratory meta-analysis. Journal of Infectious Diseases 2015;211(1):80-90. [DOI: 10.1093/infdis/jiu396] [DOI] [PMC free article] [PubMed] [Google Scholar]
McKenzie 2022
- McKenzie JE, Brennan SE, Ryan RE, Thomson HJ, Johnston RV, Thomas J. Chapter 3: Defining the criteria for including studies and how they will be grouped for the synthesis. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al editor(s). Cochrane Handbook for Systematic Reviews of Interventions version 6.3 (updated February 2022). Cochrane, 2022. Available from www.training.cochrane.org/handbook. Available from www.training.cochrane.org/handbook.
Microsoft Corporation 2018 [Computer program]
- Microsoft Excel. Microsoft Corporation, 2018. Available from office.microsoft.com/excel.
Mo 2006
- Mo H, Zeng G, Ren X, Li H, Ke C, Tan Y, et al. Longitudinal profile of antibodies against SARS-coronavirus in SARS patients and their clinical significance. Respirology 2006;11(1):49-53. [DOI: 10.1111/j.1440-1843.2006.00783.x] [DOI] [PMC free article] [PubMed] [Google Scholar]
Moher 2009
- Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Journal of Clinical Epidemiology 2009;62(10):1006-12. [DOI] [PubMed] [Google Scholar]
Morens 1994
- Morens DM. Antibody-dependent enhancement of infection and the pathogenesis of viral disease. Clinical Infectious Diseases 1994;19(3):500-12. [DOI] [PubMed] [Google Scholar]
Nyberg 2022
- Nyberg T, Ferguson NM, Nash SG, Webster HH, Flaxman S, Andrews N, et al. Comparative analysis of the risks of hospitalisation and death associated with SARS-CoV-2 omicron (B.1.1.529) and delta (B.1.617.2) variants in England: a cohort study. Lancet 2022;399:1303-12. [DOI: 10.1016/S0140-6736(22)00462-7] [DOI] [PMC free article] [PubMed] [Google Scholar]
ONS 2022
- Office for National Statistics (ONS). Coronavirus (COVID-19) Infection Survey, characteristics of people testing positive for COVID-19. www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/coronaviruscovid19infectionsurveycharacteristicsofpeopletestingpositiveforcovid19uk/21september2022#cite-this-statistical-bulletin 21 September 2022.
Otrock 2017
- Otrock ZK, Liu C, Grossman BJ. Transfusion‐related acute lung injury risk mitigation: an update. Vox Sanguinis 2017;112(8):694-703. [DOI] [PubMed] [Google Scholar]
Pandey 2012
- Pandey S, Vyas GN. Adverse effects of plasma transfusion. Transfusion 2012;52 Suppl 1:65S-79S. [DOI] [PMC free article] [PubMed] [Google Scholar]
Park 2020
- Park JJ, Harari O, Dron L, Lester RT, Thorlund K, Mills EJ. An overview of platform trials with a checklist for clinical readers. Journal of Clinical Epidemiology 2020;125:1-8. [DOI] [PubMed] [Google Scholar]
Parmar 1998
- Parmar MK, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Statistics in Medicine 1998;17(24):2815-34. [DOI] [PubMed] [Google Scholar]
Payne 2016
- Payne DC, Iblan I, Rha B, Algasrawi S, Hin A, Al Nsour M, et al. Persistence of antibodies against Middle East respiratory syndrome coronavirus. Emerging Infectious Disease Journal 2016;22(10):1824-6. [DOI: 10.3201/eid2210.160706] [DOI] [PMC free article] [PubMed] [Google Scholar]
Reeves 2022
- Reeves BC, Deeks JJ, Higgins JP, Shea B, Tugwell P, Wells GA. Chapter 24: Including non-randomized studies on intervention effects. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al, editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.3 (updated February 2022). Cochrane, 2022 . Available from www.training.cochrane.org/handbook.
RevMan Web 2022 [Computer program]
- Review Manager Web (RevMan Web). Version 4.21.1. The Cochrane Collaboration, 2022. Available from revman.cochrane.org.
Ricke 2020
- Ricke D, Malone R. Medical countermeasures analysis of 2019-nCoV and vaccine risks for antibody-dependent enhancement (ADE). Preprints [Preprint] 2020. [DOI: 10.20944/preprints202003.0138.v1] [DOI] [Google Scholar]
Robbins 1995
- Robbins JB, Schneerson R, Szu SC. Perspective: hypothesis: serum IgG antibody is sufficient to confer protection against infectious diseases by inactivating the inoculum. Journal of Infectious Diseases 1995;171(6):1387-98. [DOI] [PubMed] [Google Scholar]
Rock 2011
- Rock G. A comparison of methods of pathogen inactivation of FFP. Vox Sanguinis 2011;100(2):169-78. [DOI] [PubMed] [Google Scholar]
Santesso 2020
- Santesso N, Glenton C, Dahm P, Garner P, Akl A, Alper B, et al. GRADE guidelines 26: informative statements to communicate the findings of systematic reviews of interventions. Journal of Clinical Epidemiology 2020;119:126-35. [DOI] [PubMed] [Google Scholar]
Schünemann 2019
- Schünemann HJ, Cuello C, Akl EA, Mustafa RA, Meerpohl JJ, Thayer K, et al. GRADE guidelines: 18. How ROBINS-I and other tools to assess risk of bias in nonrandomized studies should be used to rate the certainty of a body of evidence. Journal of Clinical Epidemiology 2019;111:105-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
Schünemann 2022
- Schünemann HJ, Higgins JP, Vist GE, Glasziou P, Akl EA, Skoetz N, et al. Chapter 14: Completing ‘Summary of findings’ tables and grading the certainty of the evidence. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.3 (updated February 2022). Cochrane, 2022 . Available from www.training.cochrane.org/handbook.
Skoetz 2020
- Skoetz N, Goldkuhle M, Van Dalen EC, Akl EA, Trivella M, Mustafa RA, et al. GRADE guidelines 27: how to calculate absolute effects for time-to-event outcomes in summary of findings tables and evidence profiles. Journal of Clinical Epidemiology 2020;118:124-31. [DOI] [PubMed] [Google Scholar]
Sterne 2016
- Sterne JA, Hernán MA, Reeves BC, Savović J, Berkman ND, Viswanathan M, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016;355:i4919. [DOI] [PMC free article] [PubMed] [Google Scholar]
Sterne 2019
- Sterne JA, Savovic J, Page MJ, Elbers RG, Blencowe NS, Boutron I, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019;366:l4898. [DOI] [PubMed] [Google Scholar]
Team 2020
- Team NCPERE. Vital surveillances: the epidemiological characteristics of an outbreak of 2019 novel coronavirus diseases (COVID-19) – China. China CDC Weekly 2020;2(8):113-22. [DOI: 10.3760/cma.j.issn.0254-6450.2020.02.003] [DOI] [PMC free article] [PubMed] [Google Scholar]
Teh 2022
- Teh JS, Coussement J, Neoh ZC, Spelman T, Lazarakis S, Slavin MA, et al. Immunogenicity of COVID-19 vaccines in patients with hematologic malignancies: a systematic review and meta-analysis. Blood Advances 2022;6(7):2014-34. [DOI: 10.1182/bloodadvances.2021006333] [DOI] [PMC free article] [PubMed] [Google Scholar]
The World Bank 2022
- The World Bank. Countries and Economies. Available from https://data.worldbank.org/country (accessed 28 March, 2022).
Tierney 2007
- Tierney JF, Stewart LA, Ghersi D, Burdett S, Sydes MR. Practical methods for incorporating summary time-to-event data into meta-analysis. Trials 2007;8:16. [DOI] [PMC free article] [PubMed] [Google Scholar]
Tolouian 2020
- Tolouian R, Vahed SZ, Ghiyasvand S, Tolouian A, Ardalan M. COVID-19 interactions with angiotensin-converting enzyme 2 (ACE2) and the kinin system; looking at a potential treatment. Journal of Renal Injury Prevention 2020;9(2):e19. [DOI: 10.34172/jrip.2020.19] [DOI] [Google Scholar]
Ulloa 2022
- Ulloa AC, Buchan SA, Daneman N, Brown KA. Estimates of SARS-CoV-2 omicron variant severity in Ontario, Canada. JAMA 2022;327(13):1286–8. [DOI: 10.1001/jama.2022.2274] [DOI] [PMC free article] [PubMed] [Google Scholar]
US Covid Plasma 2022
- Convalescent plasma COVID-19 (coronavirus) treatment. Available from www.uscovidplasma.org (accessed 22 September 2022).
Van de Veerdonk 2020
- Van de Veerdonk F, Netea MG, Van Deuren M, Van der Meer JW, De Mast Q, Bruggemann RJ, et al. Kinins and cytokines in COVID-19: a comprehensive pathophysiological approach. Preprints [Preprint] 2020. [DOI: 10.20944/preprints202004.0023.v1] [DOI] [Google Scholar]
Wan 2020
- Wan Y, Shang J, Sun S, Tai W, Chen J, Geng Q, et al. Molecular mechanism for antibody-dependent enhancement of coronavirus entry. Journal of Virology 2020;94(5):e02015-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
Wang 2014
- Wang S-F, Tseng S-P, Yen C-H, Yang J-Y, Tsao C-H, Shen C-W, et al. Antibody-dependent SARS coronavirus infection is mediated by antibodies against spike proteins. Biochemical and Biophysical Research Communications 2014;451(2):208-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
WHO 2007
- World Health Organization (WHO). Cumulative number of reported probable cases of SARS. Available from www.who.int/csr/sars/country/2003_07_11/en/ (accessed 22 March 2021).
WHO 2019
- World Health Organization (WHO). Middle East respiratory syndrome coronavirus (MERS-CoV). Available from www.who.int/emergencies/mers-cov/en/ (accessed 22 March 2021).
WHO 2020a
- World Health Organization (WHO). Report of the WHO‐China Joint Mission on coronavirus disease 2019 (COVID‐19); February 2020. Available from www.who.int/docs/default-source/coronaviruse/who-china-joint-mission-on-covid-19-final-report (accessed 22 March 2021).
WHO 2020b
- World Health Organization (WHO). Rolling updates on coronavirus diseases (COVID‐19). Available from www.who.int/emergencies/diseases/novel-coronavirus-2019/events-as-they-happen (accessed 22 March 2021).
WHO 2020c
- World Health Organization (WHO). Coronavirus disease (COVID-19) situation report-209; 16 August 2020. Available from https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200816-covid-19-sitrep-209.pdf?sfvrsn=5dde1ca2_2 (accessed 22 March 2021).
WHO 2020d
- World Health Organization (WHO). Clinical management of severe acute respiratory infection (SARI) when COVID19 disease is suspected: interim guidance. World Health Organization 2020;WHO/2019-nCoV/clinical/2020.4.
WHO 2020e
- WHO Working Group on the Clinical Characterisation and Management of COVID-19 infection. A minimal common outcome measure set for COVID-19 clinical research. Lancet Infectious Diseases 2020;20(8):e192-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
WHO 2020f
- World Health Organization. COVID-19 therapeutic trial synopsis. Available from www.who.int/blueprint/priority-diseases/key-action/COVID-19_Treatment_Trial_Design_Master_Protocol_synopsis_Final_18022020.pdf (accessed 22 March 2021).
WHO 2022a
- Therapeutics and COVID-19: living guideline. Available from www.who.int/publications/i/item/WHO-2019-nCoV-therapeutics-2022.4 (accessed 16 October 2022).
WHO 2022b
- World Health Organization. Tracking SARS-CoV-2 variants. Awailable from www.who.int/en/activities/tracking-SARS-CoV-2-variants/ (accessed 17 October 2022).
Williamson 2020
- Williamson E, Walker AJ, Bhaskaran KJ, Bacon S, Bates C, Morton CE, et al. OpenSAFELY: factors associated with COVID-19-related hospital death in the linked electronic health records of 17 million adult NHS patients. medRxiv [Preprint] 2020. [DOI: ] [Google Scholar]
Wu 2020a
- Wu C, Chen X, Cai Y, Xia J, Zhou X, Xu S, et al. Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in Wuhan, China. JAMA Internal Medicine 2020;180:934-43. [DOI: 10.1001/jamainternmed.2020.0994] [DOI] [PMC free article] [PubMed] [Google Scholar]
Wu 2020b
- Wu F, Wang A, Liu M, Wang Q, Chen J, Xia S, et al. Neutralizing antibody responses to SARS-CoV-2 in a COVID-19 recovered patient cohort and their implications. medRxiv [Preprint] 2020. [DOI: ] [Google Scholar]
Xu 2020a
- Xu Z, Shi L, Wang Y, Zhang J, Huang L, Zhang C, et al. Pathological findings of COVID-19 associated with acute respiratory distress syndrome. Lancet Respiratory Medicine 2020;8(4):420-2. [DOI: 10.1016/S2213-2600(20)30076-X] [DOI] [PMC free article] [PubMed] [Google Scholar]
Yang 2022
- Yang P, Wang J, Zheng R, Tan R, Li X, Liu X, et al. Convalescent plasma may not be an effective treatment for severe and critically ill COVID-19 patients: a systematic review & meta-analysis of randomized controlled trials. Heart & lung : the journal of critical care 2022;53:51-60. [DOI: 10.1016/j.hrtlng.2022.01.019] [DOI] [PMC free article] [PubMed] [Google Scholar]
Zhao 2020
- Zhao J, Yuan Q, Wang H, Liu W, Liao X, Su Y, et al. Antibody responses to SARS-CoV-2 in patients of novel coronavirus disease 2019. medRxiv [Preprint] 2020. [DOI: 10.1101/2020.03.02.20030189] [DOI] [PMC free article] [PubMed] [Google Scholar]
References to other published versions of this review
Chai 2020
- Chai KL, Valk SJ, Piechotta V, Kimber C, Monsef I, Doree C, et al. Convalescent plasma or hyperimmune immunoglobulin for people with COVID‐19: a living systematic review. Cochrane Database of Systematic Reviews 2020, Issue 10. Art. No: CD013600. [DOI: 10.1002/14651858.CD013600.pub3] [DOI] [PMC free article] [PubMed] [Google Scholar]
Piechotta 2020a
- Piechotta V, Valk SJ, Chai KL, Wood EM, Lamikanra A, Kimber C, et al. Safety and effectiveness of convalescent plasma or hyperimmune globulin for people with COVID-19: a rapid review. available at: doi.org/10.17605/OSF.IO/DWF53 2020. [DOI: 10.17605/OSF.IO/DWF53] [DOI] [PMC free article] [PubMed]
Piechotta 2020b
- Piechotta V, Chai KL, Valk SJ, Doree C, Monsef I, Wood EM, et al. Convalescent plasma or hyperimmune immunoglobulin for people with COVID-19: a living systematic review. Cochrane Database of Systematic Reviews 2020, Issue 7. Art. No: CD013600. [DOI: 10.1002/14651858.CD013600.pub2] [DOI] [PMC free article] [PubMed] [Google Scholar]
Piechotta 2021
- Piechotta V, Iannizzi C, Chai KL, Valk SJ, Kimber C, Dorando E. Convalescent plasma or hyperimmune immunoglobulin for people with COVID-19: a living systematic review. Cochrane Database of Systematic Reviews 2021, Issue 5. Art. No: CD013600. [DOI: 10.1002/14651858.CD013600.pub4] [DOI] [PMC free article] [PubMed] [Google Scholar]
Valk 2020
- Valk SJ, Piechotta V, Chai KL, Doree C, Monsef I, Wood EM, et al. Convalescent plasma or hyperimmune immunoglobulin for people with COVID-19: a rapid review. Cochrane Database of Systematic Reviews 2020, Issue 5. Art. No: CD013600. [DOI: 10.1002/14651858.CD013600] [DOI] [PMC free article] [PubMed] [Google Scholar]