Association Between Retinal Vein Occlusion and Epiretinal Membrane Development

Abdulkader Almosa; Ahmed M Alshaikhsalama; Ahmed Abdi; Amer F Alsoudi; Melissa Yuan; Asad Loya; Ehsan Rahimy; Tamer H Mahmoud

doi:10.1016/j.ajo.2026.02.030

Association Between Retinal Vein Occlusion and Epiretinal Membrane Development

Am J Ophthalmol. 2026 Feb 24:286:108-122. doi: 10.1016/j.ajo.2026.02.030. Online ahead of print.

Authors

Abdulkader Almosa¹, Ahmed M Alshaikhsalama², Ahmed Abdi², Amer F Alsoudi³, Melissa Yuan⁴, Asad Loya³, Ehsan Rahimy⁴, Tamer H Mahmoud⁵

Affiliations

¹ From the School of Medicine (A.A.), Texas Tech University Health Sciences Center, Lubbock, Texas, USA.
² UT Southwestern Medical Center (A.M.A., A. Abdi), Dallas, Texas, USA.
³ Oakland University William Beaumont School of Medicine (A.F.A., T.H.M.), Royal Oak, Michigan, USA.
⁴ Byers Eye Institute (M.Y.), Horngren Family Vitreoretinal Center, Department of Ophthalmology, Stanford University School of Medicine, Palo Alto, California, USA; Department of Ophthalmology (M.Y.), Palo Alto Medical Foundation, Palo Alto, California, USA.
⁵ Oakland University William Beaumont School of Medicine (A.F.A., T.H.M.), Royal Oak, Michigan, USA. Electronic address: tmahmoud@arcpc.net.

PMID: 41748056
DOI: 10.1016/j.ajo.2026.02.030

Abstract

Purpose: To evaluate the temporal association between retinal vein occlusion (RVO) and the development of epiretinal membrane (ERM) including subsequent ERM peel.

Design: Retrospective cohort study.

Participants: After propensity score matching (PSM) and applying inclusion/exclusion criteria, 19,172 patients with branch retinal vein occlusion (BRVO) and 14,974 with central retinal vein occlusion (CRVO) were compared with matched controls.

Methods: Data was extracted using a national clinical database. Patients with BRVO or CRVO were compared with individuals without RVO (control) for the development of main outcomes measures. Secondary analyses examined RVO cohorts with anti-VEGF injections: BRVO-T (anti-VEGF-treated) and CRVO-T versus their respective untreated counterparts.

Main outcome measures: Relative risk (RR) of incident ERM formation and ERM peel at multiple time points from 3 months to 5 years.

Results: During the 5-year study period, BRVO and CRVO cohorts had the highest risk for ERM formation at 3 months compared with their respective controls (BRVO: RR = 4.54; CRVO: RR = 4.90; P < .0001). The risk of ERM peel was greatest at one year for BRVO (RR, 3.83; P < .0001) and 3 years for CRVO (RR, 3.91; P < .0001). In the secondary analysis, anti-VEGF treatment was associated with higher ERM rates at 3 months in BRVO-T (RR, 5.60; P < .0001) and CRVO-T (RR, 6.80; P < .0001) cohorts. The incidence of ERM peel peaked later in treated eyes at 5 years (BRVO-T: RR, 3.30; P = .0004; CRVO-T: RR, 2.60; P = .0075) compared with untreated eyes.

Conclusions: ERM formation typically occurs during the first 3 months following RVO, while surgical intervention peaks later, as early as one year in untreated eyes, and 5 years in eyes treated with anti-VEGF agents. Treated patients also exhibited an elevated risk for ERM development, which could be influenced by differences in baseline disease severity rather than a clear treatment effect.

Published by Elsevier Inc.