Abstract
Background
Most patients with chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) require maintenance treatment. We aimed to determine whether combined treatment with intravenous immunoglobulin (IVIg) and intravenous methylprednisolone (IVMP) increases remission rates compared with IVIg and placebo.
Methods
We performed a randomised, double-blind, placebo-controlled study. Adults with CIDP who were treatment naive, relapsed after a remission or required continued treatment after a single IVIg dose were included. Participants were treated with IVMP (1000 mg) or placebo (sodium chloride) every 3 weeks for 18 weeks. All participants received IVIg treatment consisting of a loading dose (2 g/kg) and serial maintenance treatments (1 g/kg). The primary outcome was remission, defined as sustained improvement on disability scales at 52 weeks without need for further treatment.
Results
Enrolment was halted after the inclusion of 77 of the 96 planned participants because of safety concerns: four thromboembolic events occurred in the IVIg/IVMP group compared with none in the IVIg/placebo group. 14 of 37 (38%) participants in the IVIg/IVMP group were in remission at 52 weeks compared with 11 out of 40 (28%) in the IVIg/placebo group, with a difference of 10% (95% CI −11 to 31; p=0.47). Several secondary outcomes showed a greater magnitude of improvement in disability and impairment at 18 and 52 weeks in the IVIg/IVMP group.
Conclusions
The study was prematurely stopped due to safety concerns after four thromboembolic events in the IVIg plus IVMP group. Combined treatment did not lead to significantly more frequent remissions than IVIg alone in the doses used in this study.
Trial registration number
ISRCTN15893334.
Keywords: NEUROMUSCULAR, STEROIDS, INTERVENTIONAL, NEUROPATHY, RANDOMISED TRIALS
WHAT IS ALREADY KNOWN ON THIS TOPIC
Both intravenous immunoglobulin (IVIg) and corticosteroids are effective first-line treatments for chronic inflammatory demyelinating polyradiculoneuropathy (CIDP), but most patients require long-term treatment. IVIg generally leads to rapid improvement in responders, and corticosteroids have been shown to induce long-term remissions.
WHAT THIS STUDY ADDS
This is the first randomised controlled study that compares combined induction treatment with IVIg and intravenous methylprednisolone (IVMP) to IVIg treatment alone. This study shows that combined induction treatment with IVIg and IVMP does not lead to more remissions than treatment with IVIg alone and may carry an increased risk of thromboembolic complications. Secondary outcomes suggested that participants treated with IVIg and IVMP improved in disability and impairment to a greater extent than those treated with IVIg alone.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Induction treatment with IVIg and IVMP should not be given with the goal of achieving remissions. Safety concerns associated with combined IVIg and IVMP must be resolved before its potential benefits in improving disability and impairment can be implemented in clinical practice. Some patients on IVIg alone may be undertreated, warranting consideration of novel (combined) treatment regimens.
Introduction
Chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is a heterogeneous autoimmune neuropathy that worsens or relapses over at least 2 months.1,3 Intravenous immunoglobulin (IVIg), as well as corticosteroids in various regimens, is considered an effective induction treatment.4,7 IVIg is an effective induction treatment, as it leads to rapid improvement within 6 weeks in most responders, while improvement with corticosteroids can take up to 4 months.8,12 Alternatively, after induction treatment, most patients treated with IVIg require regular infusions to maintain improvement, while corticosteroids have been reported to induce more long-term remissions.13 Combining IVIg and corticosteroids may lead to both a quick and a lasting response. In a pilot study where treatment-naive participants were treated with combined IVIg and intravenous methylprednisolone (IVMP) every 3 weeks for 18 weeks, 59% of participants who completed treatment were in remission at 52 weeks.14 The aim of this study was to determine whether induction treatment with combined IVIg and IVMP led to more frequent long-term remissions than IVIg treatment alone.
Methods
Study design and participants
This investigator-initiated study took place in neuromuscular and general neurology centres in the Netherlands and the UK. We included adult participants who met the European Federation of Neurological Societies/Peripheral Nerve Society (PNS) 2010 criteria for ‘probable’ or ‘definite’ CIDP.15 Participants were eligible if they were treatment naive, had relapsed after at least 1 year of remission or deteriorated within 3 months after improvement following a single loading dose of IVIg (2 g/kg), reflecting active disease. Full inclusion and exclusion criteria are provided in the online supplemental appendix. The main exclusion criteria were the presence of IgM paraproteinaemia with anti-myelin-associated glycoprotein antibodies or known (para)nodal antibodies, use of any immunosuppressive or immunomodulatory drugs in the previous 6 months and risk factors associated with increased risk of adverse events (AEs) or conditions that could lead to unblinding of treatment (eg, diabetes mellitus).
The published study protocol was amended five times during the study (complete list provided in the online supplemental appendix), including the addition of thromboembolic events (TEE) without anticoagulant therapy as an exclusion criterion.16 All participants provided written informed consent.
Randomisation and masking
Participants were randomly assigned in a 1:1 ratio by blinded investigators using the randomisation software within Castor EDC, stratified by country (Netherlands or UK) as well as CIDP-related treatment status (treatment naive or previously treated), with random permuted blocks of two or four within the strata. Participants were randomised by (blinded) investigators in the coordinating study centres in the Netherlands and the UK. Participants, caregivers and study personnel, including outcome assessors, were blinded to treatment allocations. Primary, secondary and safety outcomes were assessed by the same outcome assessor. Primary and secondary outcomes were assessed prior to safety outcomes to avoid potential unblinding of the assessor. Masking of the IVMP and placebo was performed using standard precautionary measures.16 Blinding was maintained until all participants completed their 52-week visit.
Procedures
During the 18-week treatment phase, participants were treated with either IVIg plus IVMP (1000 mg in 100 mL 0.9% sodium chloride) or IVIg plus placebo (0.9% sodium chloride, 100 mL) every 3 weeks for 18 weeks (seven treatments; figure 1). IVIg consisted of a loading dose of 2 g/kg (capped at 200 g) followed by maintenance infusions of 1 g/kg. All participants were treated with weekly alendronic acid 70 mg and daily calcium/vitamin D 500 mg/800 IU according to national guidelines. Study assessment visits took place every 6 weeks until 24 weeks, at 52 and 104 weeks after baseline. A phone consultation took place after 3 weeks. A preliminary endpoint was reached if participants deteriorated during the 18-week treatment period, did not show improvement at 18 weeks, deteriorated and/or restarted any CIDP treatment between week 18 and week 52 (defined as relapse) or if participants dropped out during the study for whatever reason. In the case of a preliminary endpoint, participants were treated at the discretion of the treating physician and completed study follow-up. The 104-week follow-up will be reported separately.
Figure 1. Treatment protocol. IVIg, intravenous immunoglobulin; IVMP, intravenous methylprednisolone; wks, weeks.
Outcomes
The primary outcome was the proportion of participants in remission at 52 weeks. Remission was defined as improvement sustained without treatment between weeks 18 and 52. Improvement was defined as an improvement in disability of at least the minimal clinical important difference (MCID) on the centile-transformed Inflammatory Rasch-Overall Disability Scale (I-RODS, ranging from 0 to 100 centile logit points with lower scores indicating more disability) and/or on the adjusted Inflammatory Neuropathy Cause and Treatment Disability Scale (aINCAT-DS, ranging from 0 to 10 with higher scores indicating more disability) between baseline and week 18.17 18 An increase of at least 1.96 SE of the mean was considered the MCID on the I-RODS.18 A decrease of 1 point or more was considered MCID improvement on the aINCAT-DS, excluding changes in upper limb function between ‘0’ (normal) and ‘1’ (minor symptoms). ‘Sustained’ was defined as no deterioration (using the same MCID cut-off values) between weeks 18 and 52. Deterioration was defined using the same MCID-based cut-offs on the I-RODS (a decrease of at least 1.96 SE of the mean) and/or aINCAT-DS (an increase of one point or more).
Secondary outcomes included the proportion of participants with improvement in disability at 18 weeks and the time to improvement in disability for these participants (defined as the first week that a MCID on either the I-RODS or aINCAT-DS was reached); the magnitude of change between baseline and week 18 and between baseline and endpoint (week 52, or earlier in case of a preliminary endpoint) measured using I-RODS and aINCAT-DS; grip strength (in kPa using a handheld Martin Vigorimeter, in the most affected hand); Medical Research Council (MRC) sum score (ranging from 0 to 60 with lower scores indicating less muscle strength; in 12 muscle groups); INCAT-sensory sum score (ranging from 0 to 20 with higher values indicating more sensory deficits); Rasch-built Fatigue Severity Scale (R-FSS, ranging from 0 to 21 with higher values indicating more fatigue); Pain Intensity Numeric Rating Scale (PI-NRS, ranging from 0 to 10 for three separate intervals with higher values indicating more pain) and health-related quality of life (using the quality of life visual analogue scale ranging from 0 to 100 with lower values indicating less quality of life). The number of AEs reported by participants or observed by physicians during the first 24 weeks (including a patient-reported questionnaire at week 18) was recorded, as well as AEs observed by the treating physician at 52 weeks (using a tailored report form). AEs were specified and scored as mild, moderate or severe, and any resulting actions were recorded.
Statistical analysis
A sample size of 96 participants was calculated to provide 80% power (with a two-sided alpha of 0.05) to detect a 30% difference in remission rates between the groups, including an assumed 10% attrition rate. We assumed remission rates of 65% in the IVIg/IVMP group and 35% in the IVIg/placebo group, based on previous studies, including the pilot study.7 13 14 An independent data safety monitoring board (DSMB) was convened to perform pre-specified unblinded conduct and safety analyses.16
The primary outcome (between-group difference in proportion of participants in remission at 52 weeks) was analysed using the χ2 test presented with a p value and using the risk difference with 95% CI in the intention-to-treat (ITT) and per-protocol (PP) datasets. A multivariable logistic regression model (including stratification variables in the model) was used to analyse the primary outcome, expressed as an adjusted OR (aOR). The PP datasets included participants who completed the assigned treatment without protocol deviations (ie, excluded those with a change of diagnosis or a preliminary endpoint between baseline and week 18).
The secondary outcome (between-group difference in the proportion of participants with improvement in disability at 18 weeks) was analysed using the risk difference with 95% CI. Other secondary outcomes were presented as median change scores between baseline and 18 weeks (or earlier in case of a preliminary endpoint during treatment) and between baseline and endpoint. Differences between groups were analysed as the absolute difference in means along with 95% CI. Median time to improvement in disability was analysed using Kaplan-Meier and compared using the log-rank test. Frequency of AEs was counted once per patient and compared between the groups. The European Quality of Life five dimensions five-level version will be reported separately. All secondary outcomes were analysed in the ITT dataset. In addition, three secondary outcome measures (proportion of participants with improvement on disability, time to improvement on disability and number of AEs) were analysed in the PP dataset. Tests were two-sided, with significance defined at a p value less than 0.05. The prespecified mixed models to assess the continuous outcomes were not performed due to the preliminary termination of the study.
We added two sensitivity analyses on the primary outcome prior to unblinding. These explored a change of diagnosis and an alternative MCID cut-off value on the I-RODS, based on the updated European Academy of Neurology/PNS guidelines (four or more centile points).4 We added three post hoc analyses prior to unblinding to explore the differences in secondary outcomes found between the groups. One analysed the time to relapse after improvement between the groups using a Kaplan-Meier analysis. The other two analysed differences in secondary outcomes in participants who improved and those in remission.
Results
Between 2 February 2018 and 16 February 2023, we screened participants from 10 neuromuscular and two general neurology centres in the Netherlands and the UK. We assessed 129 consecutive participants for eligibility, of whom 77 were randomly assigned to receive treatment (figure 2). Recruitment was stopped on the recommendation of the DSMB after 77 of 96 intended participants were randomised, following a fourth TEE despite extra mitigation measures (infusion rate cap for IVIg (0.04 mL/kg/min).19 An interim unblinded analysis by the DSMB showed that all three pulmonary embolisms (PEs) and one deep venous thrombosis (DVT) occurred in the IVIg/IVMP group compared with no TEEs in the IVIg/placebo group. Follow-up of participants was completed according to protocol.
Figure 2. Study flowchart. CMT4c, Charcot-Marie-Tooth 4c; IVIg, intravenous immunoglobulin; IVMP, intravenous methylprednisolone; ITT, intention-to-treat; PP, per-protocol.
37 (48%) participants received IVIg/IVMP and 40 (52%) received IVIg/placebo (figure 2). Groups were balanced with the exception of the duration of disease, which was longer in the IVIg/placebo group (8.0 (5.0–30.0) months in the IVIg/IVMP group and 28 (8.8–69.0) months in the IVIg/placebo group (table 1). Four (11%) participants in the IVIg/IVMP group and eight (20%) in the IVIg/placebo group did not complete treatment (figure 2). Participants who deteriorated during treatment were rescued with additional IVIg or plasma exchange, the latter usually combined with pulsed corticosteroids (dexamethasone 40 mg once daily for 4 days, monthly for 3–6 months).7 In four participants, the diagnosis was changed (to autoimmune nodopathy in three and Charcot-Marie-Tooth 4c (CMT4c) in one). All of the participants were included in the ITT dataset.
Table 1. Baseline characteristics.
| Characteristic | IVIg+IVMP (n=37) | IVIg+placebo (n=40) |
|---|---|---|
| Men | 24 (65%) | 28 (70%) |
| Age (years) | 63.0 (59.0–70.0) | 65.0 (53.5–71.3) |
| Duration of disease (months) | 8.0 (5.0–30.0) | 28 (8.8–69.0) |
| European Federation of Neurological Societies/Peripheral Nerve Society 2010 criteria | ||
| Definite | 35 (95%) | 39 (98%) |
| Probable | 2 (5%) | 1 (2%) |
| Chronic inflammatory demyelinating polyradiculoneuropathy subtype | ||
| Typical (sensorimotor) | 32 (86%) | 31 (77%) |
| Pure motor | 1 (3%) | 1 (3%) |
| Pure sensory | 0 (0%) | 1 (3%) |
| Multifocal acquired demyelinating sensory and motor neuropathy | 3 (8%) | 6 (15%) |
| Distal acquired demyelinating sensory neuropathy | 1 (3%) | 1 (3%) |
| Treatment status | ||
| Treatment naive | 31 (84%) | 33 (82%) |
| Previously treated | 2 (5%) | 5 (13%) |
| Single IVIg dose prior to inclusion) | 4 (11%) | 2 (5%) |
| Adjusted INCAT-Disability Scale (range: 0–10)* | 4.0 (3.0–6.0) | 4.0 (3.0–5.0) |
| Inflammatory Rasch-Overall Disability Scale (range: 0–100)† | 51.0 (39.0–58.0) | 59.0 (46.8–67.0) |
| Grip strength (kPa)† | 40.0 (20.0–56.0) | 50.0 (28.8–69.3) |
| Medical Research Council sum score (range: 0–60)† | 50.0 (46.0–54.0) | 53.5 (49.8–56.3) |
| INCAT sensory sum score (range: 0–20)* | 8.0 (5.0–10.0) | 7.0 (4.0–10.0) |
| Rasch-built Fatigue Severity Scale (range: 0–21)* | 18.0 (15.8–20.0) | 17.0 (12.8–20.0) |
| Pain Intensity Numeric Rating Scale (range: 0–30)* | 5.0 (0.0–15.0) | 10.0 (0.0–18.3) |
| European Quality of Life five dimensions five-level version–European Quality of Life Visual Analogue Scale† | 50.0 (40.0–70.0) | 50.0 (40.0–65.0) |
| HbA1c (%)‡ | 5.4 (5.3–5.7) | 5.4 (5.4–5.8) |
Data are presented as n (%) or median (IQR). R-FSS data missing for one patient in the IVIg/IVMP group. HbA1c data missing for one patient in the IVIg/placebo group.
Higher values indicate greater disability, impairment or pain.
Higher values indicate lesser disability, impairment or quality of life.
HbA1c values in mmol/mol were converted to % using the National Glycohaemoglobin Standardisation Programme formula: HbA1c (%)=(0.0915×mmol/mol)+2.15.
HbA1c, haemoglobin A1c; INCAT, Inflammatory Neuropathy Cause and Treatment; IVIg, intravenous immunoglobulin; IVMP, intravenous methylprednisolone.
Primary outcome
14 (38%) participants in the IVIg/IVMP group and 11 (28%) in the IVIg/placebo group were in remission at 52 weeks, with a 10% difference not reaching statistical significance (95% CI −11% to 31%, p=0.47). The aOR of reaching remission corrected for inclusion country and CIDP treatment status at 52 weeks was 1.7 (95% CI 0.62 to 4.5; online supplemental appendix includes results of the full model). The PP and sensitivity analyses were comparable with the ITT analysis (online supplemental table 1).
Secondary outcomes
27 (73%) participants in the IVIg/IVMP group and 27 (68%) in the IVIg/placebo group improved in disability on either the I-RODS or aINCAT-DS after completing the 18-week treatment (risk difference 5.0%, 95% CI −15 to 26) (table 2). The median time to improvement was 6 weeks (95% CI 3 to 12 weeks) in the IVIg/IVMP group and 12 weeks (95% CI 6 to 18) in the IVIg/placebo group (log-rank test: p=0.40; table 2 and figure 3). 13 of 27 (48%) participants in the IVIg/IVMP group and 16 of 27 (59%) in the IVIg/placebo group relapsed. Among the participants who improved at 18 weeks, the magnitude of change from baseline in various domains was greater in the IVIg/IVMP relative to the IVIg/placebo group (table 2, figure 4 and online supplemental table 2).
Table 2. Secondary outcomes in the intention-to-treat dataset.
| Outcomes | IVIg+IVMP* (n=37) |
IVIg+placebo* (n=40) |
Difference† mean (95% CI) |
|---|---|---|---|
| Improvement in disability, n (%)‡ | 27 (73%) | 27 (68%) | 5% (−15 to 26) |
| Time to improvement in disability in (weeks)ठ| 6.0 (3.0 to 12.0) | 12.0 (6.0 to 18.0) | NA |
| aINCAT-DS | |||
| Week 18 | −1.0 (−3.0 to 0.0) | −1.0 (−2.0 to 0.0) | −0.5 (−1.4 to 0.3) |
| Endpoint | −2.0 (−3.0 to −0.3) | −1.0 (−1.8 to −0.3) | −0.8 (−1.7 to 0.0) |
| I-RODS | |||
| Week 18 | 13.0 (0.0 to 33.0) | 4.0 (−1.0 to 11.0) | 12.5 (3.8 to 21.1) |
| Endpoint | 16.0 (2.0 to 37.0) | 2.0 (−3.0 to 10.0) | 16.4 (5.5 to 27.2) |
| Grip strength | |||
| Week 18 | 22.0 (8.0 to 36.0) | 13.0 (4.0 to 20.0) | 12.7 (0.9 to 24.5) |
| Endpoint | 18.0 (5.0 to 30.0) | 10.0 (0.0 to 18.0) | 12.5 (0.0 to 25.0) |
| MRC sum score | |||
| Week 18 | 7.0 (2.0 to 11.0) | 2.0 (0.0 to 5.5) | 4.7 (1.3 to 8.0) |
| Endpoint | 6.5 (2.0 to 11.8) | 2.0 (1.0 to 3.8) | 4.3 (1.8 to 6.9) |
| INCAT sensory sum score | |||
| Week 18 | −3.0 (−6.3 to 0.0) | −1.0 (−5.0 to 0.0) | −0.9 (−2.8 to 1.0) |
| Endpoint | −2.0 (−5.8 to −0.3) | 0.0 (−4.0 to 3.0) | −2.3 (−4.3 to −0.3) |
| R-FSS | |||
| Week 18 | −5.0 (−10.0 to 0.0) | 0.0 (−2.5 to 0.5) | −4.0 (−6.9 to −1.2) |
| Endpoint | −4.0 (−9.3 to 2.0) | 0.0 (−2.8 to 1.8) | −2.6 (−6.3 to 1.2) |
| PI-NRS | |||
| Week 18 | 0.0 (−4.0 to 0.0) | 0.0 (−3.8 to 0.8) | −1.1 (−4.3 to 2.1) |
| Endpoint | 0.0 (−1.0 to 1.5) | 0.0 (−2.5 to 1.0) | 0.5 (−4.2 to 5.2) |
| European Quality of Life five dimensions five-level version | |||
| Week 18 | 17.0 (0.0 to 35.0) | 10.0 (−1.5 to 22.5) | 4.3 (−7.1 to 15.6) |
| Endpoint | 12.0 (0.0 to 23.0) | 5.0 (−3.0 to 13.0) | 2.8 (−9.1 to 14.6) |
Data are changes from baseline and presented as median (IQR) unless indicated otherwise. An overview of missing data is provided in the online supplemental appendix.
Higher aINCAT-DS and sensory sum scores indicate more disability or sensory impairment. Lower I-RODS scores indicate more disability. Higher MRC sum scores indicate more strength. Higher PI-NRS scores indicate more pain. Higher R-FSS scores indicate more severe fatigue.
52 patients reached a preliminary endpoint, 23 between baseline and week 18 (10 in IVIg/IVMP group and 13 in IVIg/placebo group), 29 between week 18 and week 52 (13 in IVIg/IVMP group and 16 in placebo group). A total of 25 patients reached 52 weeks without a preliminary endpoint (14 in the IVIg/IVMP group and 11 in the IVIg/placebo group).
Difference between the IVIg/IVMP and IVIg/placebo group, presented as the absolute difference in proportion for the number of patients that improved in disability and the absolute difference in means (95% CI) for other outcomes. The 95% CIs were determined based on the Wald distribution for number of patients that improved in disability and based on the t-distribution for other outcomes.
Minimal clinically important difference based improvement in either the I-RODS or aINCAT, presented as n (%).
Presented as the number of weeks (with 95% CIs) when cumulative improvement was 50% based on Kaplan-Meier analysis.
aINCAT-DS, adjusted Inflammatory Neuropathy Cause and Treatment-Disability Scale; I-RODS, Inflammatory Rasch Disability Scale; IVIg, intravenous immunoglobulin; IVMP, intravenous methylprednisolone; MRC, Medical Research Council; PI-NRS, Pain Intensity Numeric Rating Scale; R-FSS, Rasch-built Fatigue Severity Scale.
Figure 3. Time to improvement in disability. Improvement in disability is defined as improvement on the adjusted Inflammatory Neuropathy Cause and Treatment Disability Scale and/or the Inflammatory Rasch-Overall Disability Scale by a minimal clinical important difference. IVIg, intravenous immunoglobulin; IVMP, intravenous methylprednisolone.
Figure 4. Changes from baseline on secondary outcome measures, difference between IVIg/IVMP and IVIg/placebo groups (intention-to-treat dataset). Data are the differences between the IVIg/IVMP group and the IVIg/placebo group in the changes from baseline. Black dots represent means along with 95% CIs. The horizontal axis represents the numerical values of each scale. aINCAT-DS, adjusted Inflammatory Neuropathy Cause and Treatment Disability Scale; EQ-5D-5L, European Quality of Life five dimensions five-level version; GS, grip strength (units in kPa); INCAT-SS, Inflammatory Neuropathy Cause and Treatment sensory sum score; I-RODS, Inflammatory Rasch Disability Scale; IVMP, intravenous methylprednisolone; MRC, Medical Research Council; PI-NRS, Pain Intensity Numeric Rating Scale; R-FSS, Rasch-built Fatigue Severity Scale.
Post hoc analyses
There was no difference in time to relapse after improvement between the groups (online supplemental figure 1). The differences in magnitude of improvement in favour of the IVIg/IVMP group were even greater in post hoc subgroup analysis of participants who had improved or who were in remission, whereas these differences were not seen in patients who had not responded to treatment (online supplemental tables 3–5 and online supplemental figures 2 and 3). The differences in favour of the IVIg/IVMP group were seen in the mean, median, 25th and 75th centiles, suggesting a greater magnitude of improvement in almost all those who responded to treatment and not just in a subset with very large improvement.
Safety analysis
34 (92%) participants in the IVIg/IVMP group and 31 (78%) in the IVIg/placebo group reported AEs (p=0.65; table 3 and online supplemental tables 2 and 6). In total, 296 AEs were reported, with 152 in the IVIg/IVMP group and 144 in the IVIg/placebo group. Six AEs were classified as severe by the treating physician (two in the IVIg/IVMP group).
Table 3. Adverse events reported by participants during the first 24 weeks after start of treatment.
| Adverse event | IVIg+IVMP (n=37) |
IVIg+placebo (n=40) |
P value |
|---|---|---|---|
| Any adverse event | 34 (92%) | 31 (78%) | 0.65* |
| Adverse events specified | n=34 | n=31 | |
| Change of taste | 18 (49%) | 7 (21%) | 0.02 |
| Unrest, agitation or irritability | 12 (32%) | 6 (18%) | 0.18 |
| Feeling of euphoria | 5 (14%) | 2 (6%) | 0.43 |
| Red or swollen face | 15 (41%) | 4 (12%) | 0.01 |
| Palpitations | 6 (16%) | 5 (15%) | 1.0 |
| Gastro-intestinal symptoms | 5 (14%) | 6 (18%) | 0.75 |
| Headache | 22 (60%) | 23 (68%) | 0.62 |
| Weight gain | 13 (35%) | 9 (27%) | 0.45 |
| Mood disorders | 5 (14%) | 11 (32%) | 0.09 |
| Fever or chills | 3 (8%) | 4 (12%) | 0.7 |
| Rash | 11 (30%) | 16 (47%) | 0.15 |
| Muscle pain | 8 (22%) | 14 (41%) | 0.12 |
| Polydipsia | 10 (27%) | 7 (21%) | 0.59 |
| Polyuria | 14 (38%) | 13 (38%) | 1.0 |
| Other | 5 (14%) | 5 (15%) | 1.0 |
| Total adverse events | 152 | 144 | 0.63◦ |
| Adverse event severity | |||
| Mild | 136 (90%) | 115 (80%) | 0.08 |
| Moderate | 14 (9%) | 25 (17%) | |
| Severe | 2 (1%) | 4 (3%) | |
| Consequence for study drug administration | |||
| Temporary discontinuation | 2 (1%) | 1 (1%) | 0.85 |
Data are presented as n (%). Differences between groups were analysed using χ2 or Fisher test, in case of low counts in a subgroup. Data on adverse events were missing in six patients in the IVIg/placebo group. Percentages were based on total patients with available data (n=37 in the IVIg/IVMP group and n=34 in the IVIg/placebo group). Four adverse events met the trial criteria for a SAE (three in the IVIg/IVMP group).
Analysis performed on the median number of adverse events per patient (IVIg/IVMP 4 (IQR 0–8) compared with IVIg/placebo 4 (IQR 0–9) using Wilcoxon rank-sum test).
IVIg, intravenous immunoglobulin; IVMP, intravenous methylprednisolone; SAE, serious adverse event.
In the IVIg/IVMP group, severe AEs occurred in two participants with symptomatic PEs. One other case of PE and a DVT were classified as moderate severity in the IVIg/IVMP group. In the three participants with PE, symptoms consisted of dyspnoea and chest pain and were reported 5–10 days after the loading dose, during the second and fourth treatments. One patient was admitted for oxygen therapy and discharged after 4 days. The participant with DVT experienced swelling and pain in the lower leg 20 days after the loading dose. The participants with TEEs were all male, aged 50–74, with a body mass index ranging from 23.6 to 28.4, and no one smoked. Mobility was mildly affected in all four participants at baseline and in the weeks leading up to the event (aINCAT-DS leg score of 1). Two participants had (well-regulated) hypertension. No underlying (acquired) coagulopathies or malignancies were found. All participants were started on direct oral anticoagulants and recovered without residual symptoms. In the IVIg/placebo group, severe AEs consisted of migraine (one participant in two different visits), a lower respiratory infection and a skin rash.
More participants in the IVIg/IVMP group reported a change of taste and a red or swollen face. We did not find an increased rate of known long-term steroid-related AEs (online supplemental table 7).
Blinding
70% of participants in the IVIg/IVMP group (21/30) and 71% in the IVIg/placebo group (24/34) guessed their treatment allocation correctly at endpoint. Nine (43%) of these participants in the IVIg/IVMP group and eight (33%) in the IVIg/placebo group were in remission at 52 weeks. Data of participants reporting a change of taste and a red or swollen face are provided in the online supplemental appendix.
Discussion
This study is the first randomised controlled trial (RCT) investigating whether IVMP in addition to IVIg leads to more remissions from CIDP than treatment with IVIg alone. The trial was prematurely stopped for safety concerns after four TEEs in the IVIg/IVMP group and none in the IVIg/placebo group. In this underpowered study, we did not find a statistically significant difference between the groups in the primary endpoint of remission rates. Several secondary outcomes suggested a greater magnitude of improvement in disability and impairment at 52 weeks in the IVIg/IVMP group.
We found that 38% of participants treated with IVIg/IVMP were in remission at 52 weeks, which was lower than the anticipated remission rate of 65% we found in our pilot study. The remission rate is lower than remission rates reported in the literature, ranging from 40% to 59% in patients with CIDP treated with corticosteroids.13 14 20 The PREDICT trial compared daily oral prednisolone with pulsed dexamethasone and showed that 56% of participants were in remission at 52 weeks when excluding those with a changed diagnosis.7 13 This is higher compared with the 39% in our study when excluding the one participant with a changed diagnosis.13 These differences may in part be explained by different patient characteristics (the IMC trial included known responders to IVIg or corticosteroids and did not include patients with atypical CIDP), different outcome measures and definitions of remission (assessment of improvement at different timepoints during follow-up) and, finally, duration of treatment-free sustained improvement.7 20 Our definition of remission can be considered conservative but relevant, as both sustained improvement and not needing long-term treatment are important aims of treatment. In this trial, we used the same dose as in our pilot study, which differed from previous trials. The cumulative dose of steroids in this trial was a cumulative 7 g of IVMP over 18 weeks, which fell within the range of doses used in the two previous RCTs (dexamethasone dose equivalent to 5.1 g of IVMP used in the PREDICT trial and 12 g of IVMP used in the IMC trial, both for 6 months).7 20
The proportion of participants who improved was similar to the rates reported in other studies. Surprisingly, participants treated with IVIg/IVMP showed a greater magnitude of improvement in disability and impairment and showed less residual disability and impairment at 52 weeks. A post hoc analysis showed that these differences were driven mainly by the greater magnitude of improvement in those participants who improved and in those who were in remission at 52 weeks. These exploratory data suggest that IVIg alone, in the doses used in this trial, did not generate the full potential improvement in disability and impairment in all participants and thus suggest that some responders to IVIg are ‘undertreated’. This is underscored by another study that reported a higher proportion of responders in participants treated with 2 g/kg IVIg compared with 1 g/kg, although differences were not statistically significant.21 Most studies in CIDP that measure improvement in disability compare proportions of patients reaching specific thresholds that are designed to capture the minimal degree of change on outcome scales to define improvement in disability. Arguably, the degree of improvement in disability is more important to patients than reaching a minimal threshold of change. We consider the difference in the magnitude of improvement on our secondary outcomes clinically relevant, and for future trials, we would advocate more frequent use of the magnitude of change (between groups) to define the optimal treatment effect rather than focusing on proportions that focus on much lower thresholds of improvement than those observed in patients improving in this study.
AE numbers were similar between the groups, and most were considered mild. However, the occurrence of four TEEs in the IVIg/IVMP group (three PEs and one DVT) needs to be emphasised. Both IVIg and systemic corticosteroids are known risk factors for TEEs. 22,25The increased incidence of TEEs in patients with inflammatory neuropathies treated with IVIg was demonstrated previously.20 Vascular risk factors were more frequent in those with TEEs, suggesting these patients are most at risk. The development of TEEs following IVIg is thought to arise from increased serum viscosity, but other mechanisms, such as platelet activation, may be involved as well.21 The risk of TEEs for corticosteroids, attributed to increased clotting factor activity and fibrinogen levels, seems more pronounced within 90 days of treatment initiation in new users and in those prescribed higher doses, with risks decreasing between 90 and 365 days after cessation.25 It is unclear whether combined IVIg and pulsed corticosteroids increase the cumulative risk of TEEs. In previous RCTs in participants with Guillain-Barré syndrome, myasthenia gravis and Kawasaki disease treated with IVIg and corticosteroids over shorter periods and with differing dose regimens, no TEEs were reported despite including participants less mobile than ours.26,28 TEEs occurred in 11% of the IVIg/IVMP group in the current study, despite inclusion criteria and implemented mitigating strategies designed to limit the risk of these occurring. The proportion of participants with TEEs in the IVIg/IVMP group in this study is notable and suggests that the combined treatment contributed to this risk, especially taking the two additional TEEs found in our pilot study into account.14 Even though our inclusion criteria were designed to limit the number of participants with cardiovascular risk factors, such as previous stroke and diabetes, using the QRISK3 model to estimate 10-year stroke risk, we found that only one of the participants with a TEE had a low risk.29 Two others had an estimated >10% risk, and one had a >20% risk of stroke in 10 years. Using the QThrombosis risk calculator, a model designed to predict risk of venous thrombosis, 5-year risk scores ranged from 0.6% to 2.8%.30 If the combined treatment increased the risk of TEEs, the question arises whether a lower dose of corticosteroids or an alternative way of administration might have decreased the risk of TEEs or whether other preventive strategies, such as the temporary use of oral anticoagulants, should be advised when using a combination of IVIg and high-dose corticosteroids.
The main limitation of our study was its termination before reaching the statistically powered sample size, resulting in an underpowered analysis of the primary outcome. We included only 80% of the planned number of participants. Even in the most optimistic scenario, where all additional IVIg/IVMP participants achieved remission and none in the IVIg/placebo group did, the full sample would still not have yielded the planned 30% difference. Importantly, these results do not exclude a smaller difference in remission rates, as the study was powered to detect a relatively large difference in remission rates. Of note, the IVIg/placebo group had a longer symptom duration of pre-existing disease that may be associated with more axonal damage and less pronounced treatment response. Considering the similar proportion of participants who improved in both groups, this imbalance seems unlikely to have influenced the results, but it may have impacted the magnitude of change on the secondary outcome measures. In four participants (three in the IVIg/placebo group), the diagnosis was changed to autoimmune nodopathy and CMT4c during the study. These diseases may mimic CIDP and have been shown to respond poorly (in the case of autoimmune nodopathy) to IVIg or require different management altogether (in the case of CMT4c), reflecting the challenges associated with performing CIDP trials.4 31 We expected a certain proportion of participants with a change in diagnosis, in particular autoimmune nodopathy, for which reason we included a per protocol analysis in which participants with a change of diagnosis were excluded. Altogether, we consider the number of participants with a change of diagnosis in this study to be low compared with other studies.7 13
Another factor to consider is the substantial proportion of participants who correctly guessed their treatment allocation, as well as the imbalance in participants in the IVIg/IVMP group who reported a red or swollen face or change in taste in the first 24 weeks. Measuring the success of blinding in a trial cannot be done with validity, because it is not possible to distinguish blindness from hunches about efficacy.32 33 More important than whether participants and/or physicians remained blinded throughout the study is the extent to which potential bias introduced by unblinding may have influenced the results. There was no difference in the proportion of participants who reached remission after 52 weeks between participants who correctly guessed treatment allocation (the difference between the groups remained 10%). Taking this into consideration, along with the fact that we assessed remission rates 8 months after cessation of treatment, we consider it unlikely that unblinding significantly influenced the primary outcome. However, we cannot rule out some degree of confounding influencing both primary and secondary outcomes.
In conclusion, our prematurely terminated study indicates that the addition of IVMP to IVIg as induction treatment in CIDP, in the doses used in this study, does not lead to significantly higher remission rates than treatment with IVIg and placebo. Our exploratory secondary outcomes suggest a place for additional treatment beyond IVIg induction alone. However, we cannot recommend the addition of corticosteroids until the suggested increased risk of TEEs with combined treatment is better understood.
Supplementary material
Acknowledgements
MPL is supported by the University College London Hospitals NHS Foundation Trust Biomedical Research Centre. We would like to acknowledge the effort and logistical work provided by Hobert Smink and Janneke Zwiers (former employees of Sanquin Plasma Products). We also acknowledge the contribution of the independent DSMB, which included Thomas Harbo (chair), Gabor Linthorst and Joke Dijk. Sanquin Plasma Products (Dutch Blood Supply Foundation) provided logistical support.
Footnotes
Funding: This study was funded by grants from Princess Beatrix Spierfonds (the Dutch neuromuscular foundation, grant number W.OR17-03); ZonMW (Netherlands Organisation for Health Research and Development for health-related funding, grant number 848016011); GBS/CIDP Foundation International; and GAIN UK. None of the funders was involved in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review or approval of the manuscript; and decision to submit the manuscript for publication.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study involves human participants. The study was approved by the ethics committees of all participating centres (study sponsor MREC reference number: 2017_316#B201821) and registered in the ISRCTN registry (ISRCTN15893334) and EU clinical trial register (2017-002511-34). Participants gave informed consent to participate in the study before taking part.
Data availability free text: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. FE (corresponding author) had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Collaborators: OPTIC study group: Janneke GJ Hoeijmakers, Maastricht University Medical Centre+ (Maastricht UMC+), Department of Neurology, Maastricht, The Netherlands; Nicol Voermans, Department of Neurology, Radboud University Medical Centre (RUMC), Nijmegen, The Netherlands; Paul W Wirtz, Haga Hospital, Department of Neurology, The Hague, The Netherlands; Mariëlle Padberg, Martini Hospital Groningen, Department of Neurology, Groningen, The Netherlands; Maarten de Rijk, Catharina Hospital Eindhoven, Department of Neurology, Eindhoven, The Netherlands; Adája Baars, Erasmus MC, University Medical Center, Department of Neurology, Rotterdam, The Netherlands; Govind Chavada, Queen Elizabeth University Hospital, Department of Neurology, Glasgow, UK; James Miller, Royal Victoria Infirmary, Department of Neurology, Newcastle upon Tyne, UK; James KL Holt, The Walton Centre, Department of Neurology, Liverpool, UK; Stephen W Reddel, Department of Neurology, Concord Repatriation General Hospital, Sydney, Australia.
Contributor Information
On behalf of the OPTIC study group:
Janneke GJ Hoeijmakers, Nicol Voermans, Paul W Wirtz, Mariëlle Padberg, Maarten de Rijk, Adája Baars, Govind Chavada, James Miller, James KL Holt, and Stephen W Reddel
Data availability statement
Data are available upon reasonable request.
References
- 1.Doneddu PE, Cocito D, Manganelli F, et al. Atypical CIDP: diagnostic criteria, progression and treatment response. Data from the Italian CIDP Database. J Neurol Neurosurg Psychiatry . 2019;90:125–32. doi: 10.1136/jnnp-2018-318714. [DOI] [PubMed] [Google Scholar]
- 2.Lunn MP, Manji H, Choudhary PP, et al. Chronic inflammatory demyelinating polyradiculoneuropathy: a prevalence study in south east England. J Neurol Neurosurg Psychiatry . 1999;66:677–80. doi: 10.1136/jnnp.66.5.677. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Mathey EK, Park SB, Hughes RAC, et al. Chronic inflammatory demyelinating polyradiculoneuropathy: from pathology to phenotype. J Neurol Neurosurg Psychiatry . 2015;86:973–85. doi: 10.1136/jnnp-2014-309697. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Van den Bergh PYK, van Doorn PA, Hadden RDM, et al. European Academy of Neurology/Peripheral Nerve Society guideline on diagnosis and treatment of chronic inflammatory demyelinating polyradiculoneuropathy: Report of a joint Task Force-Second revision. Eur J Neurol. 2021;28:3556–83. doi: 10.1111/ene.14959. [DOI] [PubMed] [Google Scholar]
- 5.Bus SR, de Haan RJ, Vermeulen M, et al. Intravenous immunoglobulin for chronic inflammatory demyelinating polyradiculoneuropathy. Cochrane Database Syst Rev. 2024;2024:CD001797. doi: 10.1002/14651858.CD001797.pub4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.van Lieverloo GGA, Peric S, Doneddu PE, et al. Corticosteroids in chronic inflammatory demyelinating polyneuropathy : A retrospective, multicentre study, comparing efficacy and safety of daily prednisolone, pulsed dexamethasone, and pulsed intravenous methylprednisolone. J Neurol. 2018;265:2052–9. doi: 10.1007/s00415-018-8948-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.van Schaik IN, Eftimov F, van Doorn PA, et al. Pulsed high-dose dexamethasone versus standard prednisolone treatment for chronic inflammatory demyelinating polyradiculoneuropathy (PREDICT study): a double-blind, randomised, controlled trial. Lancet Neurol. 2010;9:245–53. doi: 10.1016/S1474-4422(10)70021-1. [DOI] [PubMed] [Google Scholar]
- 8.Eftimov F, Winer JB, Vermeulen M, et al. Intravenous immunoglobulin for chronic inflammatory demyelinating polyradiculoneuropathy. Cochrane Database Syst Rev. 2013;12:CD001797. doi: 10.1002/14651858.CD001797.pub3. [DOI] [PubMed] [Google Scholar]
- 9.Adrichem ME, Lucke IM, Vrancken AFJE, et al. Withdrawal of intravenous immunoglobulin in chronic inflammatory demyelinating polyradiculoneuropathy. Brain (Bacau) 2022;145:1641–52. doi: 10.1093/brain/awac054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Vallat JM, Sommer C, Magy L. Chronic inflammatory demyelinating polyradiculoneuropathy: diagnostic and therapeutic challenges for a treatable condition. Lancet Neurol. 2010;9:402–12. doi: 10.1016/S1474-4422(10)70041-7. [DOI] [PubMed] [Google Scholar]
- 11.van Doorn PA. Treatment of Guillain-Barré syndrome and CIDP. J Peripher Nerv Syst. 2005;10:113–27. doi: 10.1111/j.1085-9489.2005.0010203.x. [DOI] [PubMed] [Google Scholar]
- 12.van Doom PA. Intravenous Immunoglobulin Treatment in Patients With Chronic Inflammatory Demyelinating Polyneuropathy. Arch Neurol . 1991;48:217. doi: 10.1001/archneur.1991.00530140113024. [DOI] [PubMed] [Google Scholar]
- 13.Eftimov F, Vermeulen M, van Doorn PA, et al. Long-term remission of CIDP after pulsed dexamethasone or short-term prednisolone treatment. Neurology (ECronicon) 2012;78:1079–84. doi: 10.1212/WNL.0b013e31824e8f84. [DOI] [PubMed] [Google Scholar]
- 14.Adrichem ME, Bus SR, Wieske L, et al. Combined intravenous immunoglobulin and methylprednisolone as induction treatment in chronic inflammatory demyelinating polyneuropathy (OPTIC protocol): a prospective pilot study. Euro J of Neurology. 2020;27:506–13. doi: 10.1111/ene.14096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Van den Bergh PYK, Hadden RDM, Bouche P, et al. European Federation of Neurological Societies/Peripheral Nerve Society guideline on management of chronic inflammatory demyelinating polyradiculoneuropathy: report of a joint task force of the European Federation of Neurological Societies and the Peripheral Nerve Society - first revision. Eur J Neurol. 2010;17:356–63. doi: 10.1111/j.1468-1331.2009.02930.x. [DOI] [PubMed] [Google Scholar]
- 16.Bus SRM, Zambreanu L, Abbas A, et al. Intravenous immunoglobulin and intravenous methylprednisolone as optimal induction treatment in chronic inflammatory demyelinating polyradiculoneuropathy: protocol of an international, randomised, double-blind, placebo-controlled trial (OPTIC) Trials. 2021;22:155. doi: 10.1186/s13063-021-05083-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hughes RAC, Donofrio P, Bril V, et al. Intravenous immune globulin (10% caprylate-chromatography purified) for the treatment of chronic inflammatory demyelinating polyradiculoneuropathy (ICE study): a randomised placebo-controlled trial. Lancet Neurol. 2008;7:136–44. doi: 10.1016/S1474-4422(07)70329-0. [DOI] [PubMed] [Google Scholar]
- 18.van Nes SI, Vanhoutte EK, van Doorn PA, et al. Rasch-built Overall Disability Scale (R-ODS) for immune-mediated peripheral neuropathies. Neurology (ECronicon) 2011;76:337–45. doi: 10.1212/WNL.0b013e318208824b. [DOI] [PubMed] [Google Scholar]
- 19.Aggarwal R, Charles-Schoeman C, Schessl J, et al. Trial of Intravenous Immune Globulin in Dermatomyositis. N Engl J Med. 2022;387:1264–78. doi: 10.1056/NEJMoa2117912. [DOI] [PubMed] [Google Scholar]
- 20.Nobile-Orazio E, Cocito D, Jann S, et al. Intravenous immunoglobulin versus intravenous methylprednisolone for chronic inflammatory demyelinating polyradiculoneuropathy: a randomised controlled trial. Lancet Neurol. 2012;11:493–502. doi: 10.1016/S1474-4422(12)70093-5. [DOI] [PubMed] [Google Scholar]
- 21.Cornblath DR, van Doorn PA, Hartung H-P, et al. Randomized trial of three IVIg doses for treating chronic inflammatory demyelinating polyneuropathy. Brain (Bacau) 2022;145:887–96. doi: 10.1093/brain/awab422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kapoor M, Spillane J, Englezou C, et al. Thromboembolic risk with IVIg: Incidence and risk factors in patients with inflammatory neuropathy. Neurology (ECronicon) 2020;94:e635–8. doi: 10.1212/WNL.0000000000008742. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kapoor M, Hunt I, Spillane J, et al. IVIg-exposure and thromboembolic event risk: findings from the UK Biobank. J Neurol Neurosurg Psychiatry . 2022;93:876–85. doi: 10.1136/jnnp-2022-328881. [DOI] [PubMed] [Google Scholar]
- 24.Huerta C, Johansson S, Wallander M-A, et al. Risk factors and short-term mortality of venous thromboembolism diagnosed in the primary care setting in the United Kingdom. Arch Intern Med. 2007;167:935–43. doi: 10.1001/archinte.167.9.935. [DOI] [PubMed] [Google Scholar]
- 25.Johannesdottir SA, Horváth-Puhó E, Dekkers OM, et al. Use of glucocorticoids and risk of venous thromboembolism: a nationwide population-based case-control study. JAMA Intern Med. 2013;173:743–52. doi: 10.1001/jamainternmed.2013.122. [DOI] [PubMed] [Google Scholar]
- 26.Bril V, Szczudlik A, Vaitkus A, et al. Randomized Double-Blind Placebo-Controlled Trial of the Corticosteroid-Sparing Effects of Immunoglobulin in Myasthenia Gravis. Neurology (ECronicon) 2023;100:e671–82. doi: 10.1212/WNL.0000000000201501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.van Koningsveld R, Schmitz PIM, Meché FGA, et al. Effect of methylprednisolone when added to standard treatment with intravenous immunoglobulin for Guillain-Barré syndrome: randomised trial. Lancet. 2004;363:192–6. doi: 10.1016/s0140-6736(03)15324-x. [DOI] [PubMed] [Google Scholar]
- 28.Kobayashi T, Saji T, Otani T, et al. Efficacy of immunoglobulin plus prednisolone for prevention of coronary artery abnormalities in severe Kawasaki disease (RAISE study): a randomised, open-label, blinded-endpoints trial. Lancet. 2012;379:1613–20. doi: 10.1016/S0140-6736(11)61930-2. [DOI] [PubMed] [Google Scholar]
- 29.Hippisley-Cox J, Coupland C, Brindle P. Development and validation of QRISK3 risk prediction algorithms to estimate future risk of cardiovascular disease: prospective cohort study. BMJ. 2017;357:j2099. doi: 10.1136/bmj.j2099. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hippisley-Cox J, Coupland C. Development and validation of risk prediction algorithm (QThrombosis) to estimate future risk of venous thromboembolism: prospective cohort study. BMJ. 2011;343:d4656. doi: 10.1136/bmj.d4656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Querol L, Devaux J, Rojas-Garcia R, et al. Autoantibodies in chronic inflammatory neuropathies: diagnostic and therapeutic implications. Nat Rev Neurol. 2017;13:533–47. doi: 10.1038/nrneurol.2017.84. [DOI] [PubMed] [Google Scholar]
- 32.Hróbjartsson A, Forfang E, Haahr MT, et al. Blinded trials taken to the test: an analysis of randomized clinical trials that report tests for the success of blinding. Int J Epidemiol. 2007;36:654–63. doi: 10.1093/ije/dym020. [DOI] [PubMed] [Google Scholar]
- 33.Sackett DL. Commentary: Measuring the success of blinding in RCTs: don’t, must, can’t or needn’t? Int J Epidemiol. 2007;36:664–5. doi: 10.1093/ije/dym088. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available upon reasonable request.