Risk Factors for Retinal Pigment Epithelium Tears in Patients with Neovascular Age-Related Macular Degeneration: A Systematic Review and Meta-Analysis.

Abstract

Topic: To identify and synthesize clinical, imaging, and treatment-related risk factors associated with retinal pigment epithelium (RPE) tears in neovascular age-related macular degeneration (nAMD).

Clinical relevance: RPE tears are a vision-threatening complication of nAMD, occurring most often in eyes with pigment epithelial detachments (PEDs) and frequently temporally associated with initiation of anti-vascular endothelial growth factor (anti-VEGF) therapy. Improved recognition of high-risk PED features may enhance counseling and risk stratification.

Methods: This systematic review and meta-analysis followed PRISMA guidelines and was prospectively registered in PROSPERO (CRD420251075278). MEDLINE, Embase, and CENTRAL were searched from inception to December 15, 2025. Randomized and non-randomized studies of adults (≥50 years) with nAMD evaluating candidate predictors of RPE tear were included. Two reviewers independently screened, extracted data, and assessed risk of bias (RoB 2, ROBINS-I). Random-effects meta-analyses were performed for baseline PED height and area, with per-study weighting, standardized mean difference (SMD), and leave-one-out sensitivity analyses. Certainty was assessed using GRADE.

Results: Sixteen studies comprising 4,043 eyes were included. Five studies (55 tear eyes; 370 controls; 425 eyes) showed greater baseline PED height in eyes developing tears (mean difference [MD], 212.39 µm; 95% confidence interval [CI], 133.87-290.91; I² = 24.6%; SMD, 1.09). Four studies (48 tear eyes; 1,312 controls; 1,360 eyes) showed larger baseline PED area (MD, 5.80 mm²; 95% CI, 3.48-8.12; I² = 2.7%; SMD, 0.88). Both persisted across all leave-one-out iterations; omitting the study contributing 43.5% of the area weight increased the effect (MD, 7.11 mm²). Fibrovascular PED, greater basal diameter and volume, subretinal fluid, sub-RPE clefts, radial hyperreflective lines, and occult or hemorrhagic macular neovascularization were also associated with tears. Anti-VEGF agent selection was not consistently associated. Most studies were at moderate to serious risk of bias; certainty was very low.

Conclusion: Baseline PED morphology, particularly increased height, is the most consistently reported association with RPE tear in nAMD. Because the evidence is predominantly retrospective, largely unadjusted, and of very low certainty, these are associations rather than validated predictive markers. Prospective studies using standardized imaging protocols are needed.