Predictors of Retinal Detachment and Visual Outcome in Acute Retinal Necrosis: A 14-Year Retrospective Cohort Publisher



Tabatabaei S A ; Soleimani M ; Ghochani G ; Daneshvar K ; Mahdizad Z ; Ghasemi S ; Mehrabi Bahar M ; Mohammadi N ; Esmaeili K ; Riazi Esfahani H ; Aminizade M ; Karimi M ; Baharvand A ; Valizadeh A Show All Authors
Source: Graefe's Archive for Clinical and Experimental Ophthalmology Published:2026

Abstract

Purpose: To identify predictors of retinal detachment (RD) and final visual outcome in acute retinal necrosis (ARN). Method: This retrospective cohort study included ARN patients who were treated at a tertiary center between 2009 and 2022. The incidence and risk factors of RD and predictors of final vision loss (< 20/200) were analyzed. Result: 156 ARN patients (58.3% male, mean age: 47.74 ± 15.49 years) were included in our study. RD occurred in 51 eyes (32.7%), at a median of 8.0 weeks from ARN onset. Extensive retinitis (four vs. one quadrant involvement) independently predicted RD (adjusted OR 3.45, p = 0.042). Poor final visual outcome (< 20/200) occurred in 79% of eyes. Worse baseline visual acuity was the strongest predictor of poor final vision (adjusted OR 5.58, p = 0.003). In contrast, presence of retinal hemorrhages at presentation (adjusted OR 0.185, p = 0.011) were associated with better final visual outcomes. Conclusion: In ARN, severe initial disease, characterized by panretinal necrosis and poor presenting acuity, predisposes to RD and profound vision loss. Most detachments occur within two months of onset, highlighting the need for vigilant early monitoring. Baseline visual acuity is a powerful prognostic indicator, whereas adjunctive treatments (prophylactic laser, intravitreal antivirals) did not independently improve outcomes. To our knowledge, this is the largest cohort of ARN patients reported from the Middle East and one of the few studies globally to incorporate time-to-event analysis and ROC-based visual prognostication. These findings provide region-specific data that refine global risk models for ARN. These findings can guide risk stratification and management decisions in ARN. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2026.