Intravitreal methotrexate as a prophylactic strategy for proliferative vitreoretinopathy in high-risk retinal detachment

Abstract

Abstract Background Proliferative vitreoretinopathy (PVR) remains the major cause of surgical failure in rhegmatogenous retinal detachment (RRD). This study evaluates the anatomical and functional outcomes, as well as the safety, of a standardized intraoperative intravitreal methotrexate infusion protocol during pars plana vitrectomy (PPV) in eyes with early stage PVR (grade B) or with advanced PVR (grade C). Methods A retrospective case series was conducted on 60 consecutive eyes of 59 patients with RRD and PVR grade B ( n = 29) or grade C ( n = 31) undergoing PPV with adjunct intraoperative methotrexate between November 2024 and February 2026 at a German tertiary center. Methotrexate 40 mg was administered via a 500 mL balanced salt solution infusion. Functional and structural testing included best-corrected visual acuity (BCVA), intraocular pressure, multifocal electroretinogram, perimetry, and spectral-domain optical coherence tomography. The primary outcome measure was single-surgery anatomical success (SSAS) at 12 weeks after primary surgery or after silicone oil removal. Secondary outcomes included BCVA changes and complication rates. Results SSAS was achieved in 81.7% of all eyes, with a significantly higher success rate in eyes with PVR grade B compared to PVR grade C (93.1% versus 71.0%; p = 0.043). At 12 weeks, BCVA improved significantly in macula-off cases ( p < 0.001) and remained stable in macula-on cases. In an exploratory subset of 17 patients postoperative multifocal electroretinogram and perimetry demonstrated preserved central retinal function. No ocular or systemic adverse events related to methotrexate occurred. Conclusions The standardized intraoperative intravitreal methotrexate infusion protocol was safe, feasible, and associated with high single-surgery reattachment rates, particularly in PVR grade B eyes. Anatomical outcomes in PVR grade C eyes were comparable to other complex PVR cohorts, without any drug-related local or systemic toxicity. Trial registration Not applicable (retrospective observational study).