This cohort study examines whether the incidence of retinal artery occlusions and retinal vein occlusions changed after patients were diagnosed with COVID-19 infection.
Key Points
Question
Does the incidence of retinal vascular occlusions change after COVID-19 infection?
Findings
This cohort study of 432 515 patients diagnosed with COVID-19 found that the incidence of retinal vein occlusions, but not retinal artery occlusions, appeared to increase in the 6 months after COVID-19 diagnosis.
Meaning
Patients with COVID-19 infection may have an increased risk of retinal vein occlusion in the 6 months after infection, similar to the increased risk of systemic vascular damage associated with COVID-19, and clinicians need to consider this factor when evaluating these patients.
Abstract
Importance
COVID-19 is associated with systemic vascular damage; however, the risk posed to the retinal vasculature remains incompletely understood.
Objective
To assess if there is a change in the incidence of retinal vascular occlusions after COVID-19 infection.
Design, Setting, and Participants
This cohort study at an integrated health care organization (Kaiser Permanente Southern California) included patients without a history of retinal vascular occlusion who were diagnosed with COVID-19 infection between January 20, 2020, and May 31, 2021. Patients were excluded if they had a history of retinal artery occlusions (RAOs) or retinal vein occlusions (RVOs) more than 6 months before their COVID-19 diagnosis or if they were enrolled in Kaiser Permanente Southern California for less than 6 months before COVID-19 diagnosis.
Exposures
COVID-19 infection.
Main Outcomes and Measures
The change in the average biweekly incidence of new RAOs and RVOs after COVID-19 diagnosis. Adjusted incidence rate ratios (IRRs) were calculated to compare the incidence of retinal vascular occlusions before and after COVID-19 diagnosis after accounting for baseline demographic characteristics, medical history, and hospitalization.
Results
A total of 432 515 patients diagnosed with COVID-19 infection were included in this study. The mean (SD) age was 40.9 (19.2) years, and 231 767 patients (53.6%) were women. Sixteen patients had an RAO (crude incidence rate, 3.00 per 1 000 000 patients), and 65 had an RVO (crude incidence rate, 12.20 per 1 000 000 patients) in the 6 months after COVID-19 diagnosis. The incidence of new RVOs was higher in the 6 months after COVID-19 infection compared with the 6 months before infection after adjusting for age; sex; self-reported race and ethnicity; body mass index; history of diabetes, hypertension, or hyperlipidemia; and hospitalization (adjusted IRR, 1.54; 95% CI, 1.05-2.26; P = .03). There was a smaller increase in the incidence of RAOs after COVID-19 diagnosis (IRR, 1.35; 95% CI, 0.64-2.85; P = .44). The peak incidence of RAOs and RVOs occurred 10 to 12 weeks and 6 to 8 weeks after COVID-19 diagnosis, respectively.
Conclusions and Relevance
The findings of this study suggest that there was an increase in the incidence of RVOs after COVID-19 infection; however, these events remain rare, and in the absence of randomized controls, a cause-and-effect relationship cannot be established. Further large, epidemiologic studies are warranted to better define the association between retinal thromboembolic events and COVID-19 infection.
Introduction
COVID-19 infection can be followed by multiorgan pathology, including vascular damage involving the venous and arterial systems.1,2,3 Optical coherence tomography angiography4,5,6,7 studies have found reductions in retinal vascular density in patients with COVID-19 infection compared with controls. An association between COVID-19 and retinal vascular occlusions seems intuitive given the prothrombotic state induced by this infection and the retinal vascular alterations that have been observed; however, data on a possible association are limited. The aim of this study was to explore whether there was a change in the incidence of retinal vascular occlusions after COVID-19 infection.
Methods
This retrospective cohort study analyzed patients enrolled in an integrated health care organization, Kaiser Permanente Southern California (KPSC), with a confirmed diagnosis of COVID-19 infection between January 20, 2020, and May 31, 2021. Patients were categorized as having COVID-19 infection if they had a positive polymerase chain reaction test result or if they had a diagnosis code in their medical record (International Statistical Classification of Diseases and Related Health Problems, Tenth Revision and internal KPSC codes) for COVID-19. Laboratory tests and diagnoses for COVID-19 performed outside of KPSC were routinely imported into the electronic health record. Patients were excluded if they had a history of retinal artery occlusions (RAOs) or retinal vein occlusions (RVOs) more than 6 months before their COVID-19 diagnosis or if they were enrolled in KPSC for less than 6 months (allowing for a 30-day gap) before COVID-19 diagnosis. Patients were followed up until death, disenrollment from KPSC, diagnosis of RAO or RVO, or the end of the study (July 25, 2021), whichever occurred first. Approval for the study was obtained from the KPSC Institutional Review Board, which waived the need for individual patient consent given the retrospective nature of this study. The study adhered to the tenets of the Declaration of Helsinki and followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Baseline demographic characteristics of patients were collected from the electronic health record, including age (in years, continuous); sex; self-reported race and ethnicity; body mass index (categorical; calculated as weight in kilograms divided by height in meters squared); and history of diabetes, hypertension, and hyperlipidemia before COVID-19 diagnosis. Hospitalizations (including out-of-network claims) 7 days before to 4 weeks after COVID-19 diagnosis were included in this analysis. International Classification of Diseases, Ninth Revision and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes were used to identify patients with RAO, RVO, diabetes, hypertension, and hyperlipidemia (eTable in the Supplement).
The average biweekly incidence of new retinal vascular occlusions was compared between the pre–COVID-19 infection period (26 to 2 weeks before diagnosis) and the postinfection period (2 weeks before to 26 weeks after diagnosis). The 2 weeks before COVID-19 diagnosis was included in the time frame for COVID-19 infection given the typical delay between virus contraction, symptom onset, and testing as well as the possible delays in reporting positive results into the electronic health record.
Descriptive statistics were used to compute the biweekly incidence rates of vascular occlusions. Poisson regression was used to compare the incidence rates across different time intervals. The logged number of eligible patients in each time interval was inserted into the Poisson model as an offset. Adjusted incidence rate ratios (IRRs) were calculated to compare the incidence of retinal vascular occlusions before and after COVID-19 diagnosis after accounting for baseline demographic characteristics, medical history, and hospitalization. In a subsample of patients without any retinal vascular occlusions before the date of COVID-19 diagnosis, a Poisson regression model was used to assess the association between hospitalizations and the incidence of retinal vascular occlusions during the 6 months after infection, with adjustments made for baseline patient characteristics and medical history. The baseline characteristics of patients who experienced a retinal vascular occlusion after COVID-19 were compared with those without RAO or RVO and those with pre–COVID-19 RAO or RVO using the Kruskal-Wallis test for continuous variables, χ2 test for categorical variables, and Fisher exact test for categorical variables with small cell sizes. Data were analyzed using SAS, version 9.4 (SAS Institute Inc). Two-sided P < .05 was considered statistically significant.
Results
A total of 432 515 patients diagnosed with COVID-19 met the inclusion criteria. The mean (SD) age was 40.9 (19.2) years; 231 767 (53.6%) were women and 200 729 (46.4%) were men (19 [0%] had unknown sex); and self-reported race and ethnicity included 29 837 (6.9%) Asian and Pacific Islander, 262 436 (60.7%) Hispanic, 26 150 (6.0%) non-Hispanic Black, and 87 593 (20.3%) non-Hispanic White individuals as well as 26 499 individuals (6.1%) of other (including Native American individuals or those of more than 1 race or ethnicity) or unknown races and ethnicities. There were 12 RAOs (crude incidence rate: 2.31 per 1 000 000 patients) and 43 RVOs (crude incidence rate: 8.30 per 1 000 000 patients) in the 6 months before COVID-19 diagnosis and 16 RAOs (crude incidence rate: 3.00 per 1 000 000 patients) and 65 RVOs (crude incidence rate: 12.20 per 1 000 000 patients) in the 6 months after COVID-19 diagnosis. Table 1 shows the biweekly crude incidence (per 1 000 000 patients) and adjusted IRR of RAOs and RVOs relative to the date of COVID-19 diagnosis. The adjusted IRR accounted for age; sex; self-reported race and ethnicity; body mass index; history of diabetes, hypertension, or hyperlipidemia; and hospitalization. The incidence of RAOs (adjusted IRR, 1.35; 95% CI, 0.64-2.85; P = .44) and RVOs (adjusted IRR, 1.54; 95% CI, 1.05-2.26; P = .03) increased in the 6 months after COVID-19 diagnosis. The immediate post–COVID-19 period (2 weeks before to 12 weeks after diagnosis) had the highest crude incidence rates of RAOs (3.55; 95% CI, 1.20-5.90) and RVOs (12.60; 95% CI, 8.30-17.00) (Table 1). The highest incidence rates of RAOs and RVOs were seen 10 to 12 weeks and 6 to 8 weeks after COVID-19 diagnosis, respectively.
Table 1. Average Incidence of RAOs and RVOs Relative to the Date of COVID-19 Diagnosis.
| Outcome | Time period | Crude rate, per 1 million patients (95% CI) | Unadjusted IRR (95% CI)a | P value | Adjusted IRR (95% CI)a | P value |
|---|---|---|---|---|---|---|
| RAO | Pre–COVID-19 infectionb | 2.31 (1.00 to 3.60) | 1 [Reference] | 1 [Reference] | ||
| Post–COVID-19 infectionc | 3.00 (1.50 to 4.50) | 1.30 (0.61 to 2.74) | .50 | 1.35 (0.64 to 2.85) | .44 | |
| Immediate post–COVID-19 infectiond | 3.55 (1.20 to 5.90) | 1.53 (0.65 to 3.64) | .33 | 1.58 (0.67 to 3.76) | .30 | |
| Late post–COVID-19 infectione | 2.50 (0.70 to 4.40) | 1.08 (0.43 to 2.75) | .87 | 1.13 (0.44 to 2.87) | .80 | |
| RVO | Pre–COVID-19 infectionb | 8.30 (5.80 to 10.80) | 1 [Reference] | 1 [Reference] | ||
| Post–COVID-19 infectionc | 12.20 (9.20 to 15.10) | 1.47 (1.00 to 2.16) | .05 | 1.54 (1.05 to 2.26) | .03 | |
| Immediate post–COVID-19 infectiond | 12.60 (8.30 to 17.00) | 1.52 (0.96 to 2.41) | .07 | 1.58 (1.00 to 2.50) | .049 | |
| Late post–COVID-19 infectione | 11.80 (7.80 to 15.80) | 1.42 (0.90 to 2.24) | .13 | 1.50 (0.95 to 2.36) | .08 |
Abbreviations: IRR, incidence rate ratio; RAO, retinal artery occlusion; RVO, retinal vein occlusion.
The IRRs were obtained from unadjusted and adjusted Poisson regression. Adjusted analyses accounted for age at index date (in years); sex (male or female patient or unknown); self-reported race and ethnicity (Asian and Pacific Islander, Hispanic, non-Hispanic Black, and non-Hispanic White individuals as well as individuals of other [including Native American individuals or those of more than 1 race or ethnicity] or unknown race and ethnicity); history of diabetes, hypertension, or hyperlipidemia; most recent body mass index within 1 year before the index date (calculated as weight in kilograms divided by height in meters squared and categorized as underweight, healthy weight, overweight, obese, and missing or unknown body mass index); and whether a patient was hospitalized within 7 days before to 1 month after the date of COVID-19 diagnosis.
Period 26 to 2 weeks before diagnosis.
Period 2 weeks before to 26 weeks after diagnosis.
Period 2 weeks before to 12 weeks after diagnosis.
Period 12 to 26 weeks after diagnosis.
Typical systemic vascular risk factors (eg, diabetes, hypertension, and hyperlipidemia) were more common in patients who had a retinal vascular occlusion (Table 2). The adjusted IRR for hospitalization was not found to be associated with any retinal vascular occlusion category (Table 3).
Table 2. Baseline Characteristics of Patients With COVID-19 Infection Stratified by Retinal Vascular Occlusion Status.
| Characteristic | No. (%) | Group 2 – Group 1 | Group 3 – Group 1 | Group 3 – Group 2 | |||||
|---|---|---|---|---|---|---|---|---|---|
| Group 1: no RAO or RVO (n = 432 379) | Group 2: pre–COVID-19 RAO or RVO (n = 55) | Group 3: post–COVID RAO or RVO (n = 81) | Mean or % difference | P valuea | Mean or % difference | P valuea | Mean or % difference | P valuea | |
| Age at index date, y | |||||||||
| Mean (SD) | 40.8 (19.18) | 63.4 (13.06) | 62.4 (13.29) | 22.5 (17.5 to 27.6) | <.001 | 21.5 (17.4 to 25.7) | <.001 | –0.99 (–5.6 to 3.6) | .64 |
| Median (Q1-Q3) | 40 (26.0 to 55.0) | 64 (55.0 to 72.0) | 63 (53.0 to 72.0) | NA | NA | NA | NA | NA | NA |
| Range | 0 to 110.0 | 27.0 to 89.0 | 31.0 to 91.0 | NA | NA | NA | NA | NA | NA |
| Sex | |||||||||
| Female | 231 692 (53.6) | 29 (52.7) | 46 (56.8) | –0.9 (–14.0 to 12.3) | .90 | 3.2 (–7.7 to 14.1) | .56 | 4.1 (–13.2 to 21.4) | .64 |
| Male | 200 668 (46.4) | 26 (47.3) | 35 (43.2) | 0.9 (–12.3 to 14.0) | .90 | –3.2 (–14.1 to 7.7) | .57 | –4.1 (–21.4 to 13.2) | .64 |
| Unknown | 19 | 0 | 0 | 0 (–0.2 to 0.2) | .99 | 0 (–0.1 to 0.1) | .99 | 0 | NA |
| Self-reported race and ethnicity | |||||||||
| Asian and Pacific Islander | 29 828 (6.9) | 3 (5.5) | 6 (7.4) | –1.4 (–8.1 to 5.3) | .99 | 0.5 (–5.0 to 6.0) | .83 | 1.9 (–6.7 to 10.6) | .74 |
| Hispanic | 262 361 (60.7) | 31 (56.4) | 44 (54.3) | –4.3 (–17.2 to 8.6) | .51 | –6.4 (–17.0 to 4.3) | .24 | –2.0 (–19.4 to 15.3) | .81 |
| Non-Hispanic Black | 26 128 (6.0) | 8 (14.5) | 14 (17.3) | 8.5 (2.2 to 14.8) | .008 | 11.2 (6.1 to 16.4) | <.001 | 2.7 (–10.1 to 15.6) | .67 |
| Non-Hispanic White | 87 566 (20.3) | 13 (23.6) | 14 (17.3) | 3.4 (–7.2 to 14.0) | .53 | –3.0 (–11.7 to 5.8) | .51 | –6.4 (–20.2 to 7.5) | .36 |
| Other or unknownb | 26 496 (6.1) | 0 | 3 (3.7) | –6.1 (–12.5 to 0.2) | .049 | –2.4 (–7.7 to 2.8) | .49 | 3.7 (–1.4 to 8.8) | .27 |
| History of diabetes | |||||||||
| No | 366 367 (84.7) | 25 (45.5) | 45 (55.6) | NA | NA | NA | NA | NA | NA |
| Yes | 66 012 (15.3) | 30 (54.5) | 36 (44.4) | 39.3 (29.8 to 48.8) | <.001 | 29.2 (21.3 to 37.0) | <.001 | –10.1 (–27.4 to 7.2) | .25 |
| History of hypertensionc | |||||||||
| No | 327 404 (75.7) | 10 (18.2) | 27 (33.3 | NA | NA | NA | NA | NA | NA |
| Yes | 104 975 (24.3) | 45 (81.8) | 54 (66.7) | 57.5 (46.2 to 68.9) | <.001 | 42.4 (33.1 to 51.7) | <.001 | –15.1 (–30.4 to 0.1) | .05 |
| History of hyperlipidemiac | |||||||||
| No | 293 036 (67.8) | 16 (29.1) | 21 (25.9) | NA | NA | NA | NA | NA | NA |
| Yes | 139 343 (32.2) | 39 (70.9) | 60 (74.1) | 38.7 (26.3 to 51.0) | <.001 | 41.9 (31.7 to 52.0) | <.001 | 3.2 (–12.3 to 18.7) | .68 |
| Hospitalized | |||||||||
| No | 405 119 (93.7) | 46 (83.6) | 69 (85.2) | NA | NA | NA | NA | NA | NA |
| Yes | 27 260 (6.3) | 9 (16.4) | 12 (14.8) | 10.1 (3.6 to 16.5) | .002 | 8.5 (3.2 to 13.8) | .002 | –1.6 (–14.1 to 11.0) | .81 |
| BMId | |||||||||
| Underweight | 13 500 (3.1) | 0 | 0 | –3.1 (–7.7 to 1.5) | .42 | –3.1 (–6.9 to 0.7) | .19 | 0 | NA |
| Healthy weight | 59 896 (13.9) | 7 (12.7) | 12 (14.8) | –1.1 (–10.3 to 8.0) | .81 | 1.0 (–6.6 to 8.5) | .80 | 2.1 (–10.0 to 14.2) | .73 |
| Overweight | 88 889 (20.6) | 14 (25.5) | 28 (34.6) | 4.9 (–5.8 to 15.6) | .37 | 14.0 (5.2 to 22.8) | .002 | 9.1 (–6.9 to 25.1) | .26 |
| Obese | 137 771 (31.9) | 30 (54.5) | 33 (40.7) | 22.7 (10.4 to 35.0) | <.001 | 8.9 (–1.3 to 19.0) | .09 | –13.8 (–31.0 to 3.4) | .11 |
| Unknown | 132 323 (30.6) | 4 (7.3) | 8 (9.9) | –23.3 (–35.5 to –11.2) | <.001 | –20.7 (–30.8 to –10.7) | <.001 | 2.6 (–7.3 to 12.5) | .75 |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); NA, not applicable; Q, quartile; RAO, retinal artery occlusion; RVO, retinal vein occlusion.
Obtained from Kruskal-Wallis test for continuous variables, χ2 test for categorical variables, and Fisher exact test for categorical variables with small cell sizes.
Other includes the following races and ethnicities: Native American individuals or those of more than 1 race or ethnicity.
Whether a patient had the diagnosis any time before COVID-19 diagnosis.
Obtained from the most recent record within 1 year before the index date. Underweight was categorized as BMI<18.5; healthy weight, BMI 18.5 to 24.9; overweight, BMI 25 to 25.9; and obese, BMI ≥30.
Table 3. Unadjusted and Adjusted IRRs of Retinal Vascular Occlusion by Hospitalization Post–COVID-19 Diagnosis.
| Outcome | Hospitalization anytime from 7 d to 1 mo after COVID-19 diagnosis | Unadjusted IRR (95% CI)a | P value | Adjusted IRR (95% CI)a | P value |
|---|---|---|---|---|---|
| RAO | No | 1 [Reference] | 1 [Reference] | ||
| Yes | 1.23 (0.16-9.3) | .84 | 0.34 (0.04-2.70) | .31 | |
| RVO | No | 1 [Reference] | 1 [Reference] | ||
| Yes | 3.75 (1.96-7.17) | <.001 | 1.27 (0.64-2.50) | .50 |
Abbreviations: IRR, incidence rate ratio; RAO, retinal artery occlusion; RVO, retinal vein occlusion.
The IRRs were obtained from unadjusted and adjusted Poisson regression. Adjusted analyses accounted for age at index date (in years); sex (male or female patient or unknown); self-reported race and ethnicity (Asian and Pacific Islander, Hispanic, non-Hispanic Black, and non-Hispanic White individuals as well as individuals of other [including Native American individuals or those of more than 1 race or ethnicity] or unknown race and ethnicity); history of diabetes, hypertension, or hyperlipidemia; and most recent body mass index within a year before the index date (calculated as weight in kilograms divided by height in meters squared and categorized as underweight, healthy weight, overweight, obese, and missing or unknown weight).
Discussion
Findings of this study suggest that there may be an increased risk for RVO after COVID-19 infection. This increased risk could be associated with secondary events, such as worsening control of systemic blood pressure or lack of physical activity. A cause-and-effect association could not be established in this retrospective, nonrandomized study because selection bias and unaccounted-for confounders may make the change in RVO incidence coincidental to COVID-19.8
The time between COVID-19 diagnosis and the increase in retinal vascular occlusion diagnoses may be associated with delays in patients seeking care after visual changes or may indicate an extended prothrombotic state after infection. Furthermore, it is possible that the previously described vascular changes after COVID-19 infection may reduce the vascular reserve of patients at risk for retinal vascular occlusions and predispose them to later acute retinal vascular occlusions. Patients with RAOs or RVOs before or after COVID-19 infection were more likely to have typical systemic risk factors (eg, diabetes, hypertension, and hyperlipidemia) than those who did not have retinal vascular occlusions. Of note, the systemic features, including demographic characteristics and medical history (eg, diabetes, hypertension, and hyperlipidemia), of patients who experienced vascular occlusions before and after COVID-19 were similar. This study did not find that hospitalization was associated with an increased risk of retinal vascular occlusions; however, these events may be underestimated in severely ill patients because they may be unable to provide a history of vision changes that would elicit an ophthalmic evaluation (ie, intubated or altered mental status). There was an association between COVID-19 infection and RVOs but not RAOs in this study, which is noteworthy because this outcome may further support the notion that venous disease may predominate over arterial disease in patients with COVID-19 infection.3,9
Strengths of this study include its large size, diverse population, and longitudinal follow-up, which may increase its generalizability. Limitations include that patients with COVID-19 who did not undergo testing were not captured in this analysis; therefore, patients with mild or asymptomatic disease may have been underrepresented.
Clinicians need to consider the potential association between COVID-19 infection and retinal vascular occlusions when evaluating patients after infection. The findings provide further evidence of the prothrombotic state induced by COVID-19 and indicate that the postinfection impacts may last several weeks. Large epidemiologic studies are warranted to better define the association between retinal thromboembolic events and COVID-19 infection.
eTable. International Classification of Diseases, Ninth Revision and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision Codes Used.
References
- 1.Go AS, Reynolds K, Tabada GH, et al. COVID-19 and risk of VTE in ethnically diverse populations. Chest. 2021;160(4):1459-1470. doi: 10.1016/j.chest.2021.07.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Piazza G, Campia U, Hurwitz S, et al. Registry of arterial and venous thromboembolic complications in patients with COVID-19. J Am Coll Cardiol. 2020;76(18):2060-2072. doi: 10.1016/j.jacc.2020.08.070 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lodigiani C, Iapichino G, Carenzo L, et al. ; Humanitas COVID-19 Task Force . Venous and arterial thromboembolic complications in COVID-19 patients admitted to an academic hospital in Milan, Italy. Thromb Res. 2020;191:9-14. doi: 10.1016/j.thromres.2020.04.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Guemes-Villahoz N, Burgos-Blasco B, Vidal-Villegas B, et al. Reduced retinal vessel density in COVID-19 patients and elevated D-dimer levels during the acute phase of the infection. Med Clin (Barc). 2021;156(11):541-546. doi: 10.1016/j.medcli.2020.12.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Turker IC, Dogan CU, Guven D, Kutucu OK, Gul C. Optical coherence tomography angiography findings in patients with COVID-19. Can J Ophthalmol. 2021;56(2):83-87. doi: 10.1016/j.jcjo.2020.12.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Abrishami M, Emamverdian Z, Shoeibi N, et al. Optical coherence tomography angiography analysis of the retina in patients recovered from COVID-19: a case-control study. Can J Ophthalmol. 2021;56(1):24-30. doi: 10.1016/j.jcjo.2020.11.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zapata MÁ, Banderas García S, Sánchez-Moltalvá A, et al. Retinal microvascular abnormalities in patients after COVID-19 depending on disease severity. Br J Ophthalmol. 2020;bjophthalmol-2020-317953. doi: 10.1136/bjophthalmol-2020-317953 [DOI] [PubMed] [Google Scholar]
- 8.Jampol LM, Tauscher R, Schwarz HP. COVID-19, COVID-19 vaccinations, and subsequent abnormalities in the retina: causation or coincidence? JAMA Ophthalmol. 2021;139(10):1135-1136. doi: 10.1001/jamaophthalmol.2021.3483 [DOI] [PubMed] [Google Scholar]
- 9.Tan BK, Mainbourg S, Friggeri A, et al. Arterial and venous thromboembolism in COVID-19: a study-level meta-analysis. Thorax. 2021;76(10):970-979. doi: 10.1136/thoraxjnl-2020-215383 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable. International Classification of Diseases, Ninth Revision and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision Codes Used.
