Reporting and Extent of Glucocorticoid Use in Systemic Sclerosis Randomized Trials: A Meta-Research Study

To summarize the extent of glucocorticoid (GC) use in systemic sclerosis (SSc) randomized controlled trials (RCTs) and assess the adequacy of reporting.

We searched PubMed for RCTs on SSc published between 2000 and 2024 investigating pharmacologic treatment. We recorded key trial characteristics (country, phase, intervention, funding source, and sample size) and information on planned and observed GC use. Data were summarized as number (percentage) for categorical variables and median (range, interquartile range) for continuous variables.

We included 76 studies, mostly single-country (72.4%), industry-funded (71.4%) trials investigating treatment for skin (30.3%), Raynaud's phenomenon/digital ulcers (28.9%), and lung complications (14.4%). Glucocorticoid use was mentioned in the eligibility criteria or other sections of the publication or protocol in 48 (63.1%) studies. Forty-six (65.7%) studies did not provide information about whether GCs were allowed, and 50 (65.8%) did not specify at what dosage. Information on whether the intake of GCs could have been modified during the study was missing in 43 of 48 RCTs. Among the 48 trials that mentioned GC use, baseline data on the number of patients receiving GCs per arm and mean baseline dosage were missing in 23 (47.9%) and 20 (41.6%) studies, respectively. In trials that reported the number of patients receiving GCs at randomization, the median proportion of users was 44.1% (range, 5.9%-100%) in the overall sample, 50.8% (range, 13.6%-100%) in skin-related studies, 50% (range, 23.4%-100%) in lung fibrosis studies, and 28.6% (range, 5.9%-37.5%) in Raynaud's phenomenon/digital ulcer studies.

Glucocorticoid use is widespread in SSc-RCTs but is often inadequately documented. Variable and incomplete reporting of GC utilization may introduce cointervention bias (bias from unequal or uncontrolled concurrent treatments), complicating interpretation of treatment effects.