Abstract
Background and objectiveRelapse after treatment discontinuation or reduction frequently occurs in pulmonary sarcoidosis (PS), but reported rates vary widely. We conducted a systematic review and meta-analysis to estimate relapse prevalence and evaluate risk factors.MethodsWe searched PubMed, Embase, Scopus, and Google Scholar from inception through January 2026 for studies reporting relapse prevalence in PS. Two reviewers independently extracted data on relapse prevalence, timing, definitions, and predictors, and assessed study quality using the Hoy et al. tool. We calculated pooled prevalence using random effects models, assessed heterogeneity using I2, and performed subgroup analyses and meta-regression.ResultsWe included 51 studies comprising 6076 patients; 3682 were followed up, and 1442 relapsed. Pooled relapse prevalence was 0.39 (95% CI, 0.33-0.45) with substantial heterogeneity (I2=94%). Relapse prevalence was lower in prospective (0.31, 95% CI: 0.25-0.37) than retrospective (0.45, 95% CI: 0.36-0.54) studies. Only 23 (45%) studies provided explicit relapse definitions, which varied across clinical, radiological, functional, and biomarker criteria. Advanced radiographic stage was associated with higher relapse odds than stage I (log OR 0.55-1.04, p<0.05), and Black race with higher odds than White race (log OR 0.91, p<0.01). Relapse prevalence did not differ between clinical-only versus composite definitions.ConclusionsApproximately two-fifths of PS patients relapse after treatment reduction or discontinuation. Marked heterogeneity in relapse definitions and observational study designs underscores the need for standardized criteria. Clinical assessment appears comparable to complex definitions for assessing relapse.