Comparative effectiveness and safety of adjunctive methotrexate in rhegmatogenous retinal detachment complicated by or at high risk for proliferative vitreoretinopathy: a systematic review and meta-analysis

Abstract Purpose To evaluate the efficacy and safety of adjunctive methotrexate (MTX) in vitreoretinal surgery for rhegmatogenous retinal detachment (RRD) complicated by or at high risk for proliferative vitreoretinopathy (PVR). Methods This PRISMA-compliant systematic review and meta-analysis (PROSPERO CRD420261281220) searched PubMed, Embase, CENTRAL, and Web of Science through October 16, 2025. Randomized trials, comparative observational studies, and case series were included. Comparative and single-arm data were analyzed separately using random-effects models. Primary outcomes were final retinal reattachment and best-corrected visual acuity (BCVA); postoperative PVR and adverse events were secondary outcomes. Results Sixteen reports met the eligibility criteria, with 14 unique studies contributing to quantitative synthesis. MTX was associated with a lower risk of postoperative PVR (6 studies; 131 MTX vs. 138 controls; RR 0.56, 95% CI 0.40–0.77; p = 0.004; I²=0%) and better BCVA in the primary analysis (MD − 0.20 logMAR, 95% CI − 0.34 to − 0.06; I²=12.3%), but not with improved final retinal reattachment (8 studies; 215 MTX vs. 207 controls; RR 1.07, 95% CI 0.90–1.28; p = 0.412; I²=56.7%). However, the association with better BCVA was attenuated and no longer statistically significant after studies at critical risk of bias were excluded. Single-arm analyses yielded a pooled reattachment rate of 80% (95% CI 75–85%) and postoperative PVR rate of 22% (95% CI 15–31%). Adverse events were predominantly mild, reversible corneal complications, reported mainly with repeated MTX injections. Certainty of evidence was very low across major efficacy outcomes. Conclusions Adjunctive MTX was associated with reduced postoperative PVR but not with improved final retinal reattachment. Although the primary analysis suggested modestly better postoperative BCVA, this finding was not robust to the exclusion of studies at critical risk of bias. Given the very low certainty of evidence, these findings are hypothesis-generating rather than practice-changing. Larger randomized trials with standardized MTX regimens and outcome definitions are needed.