Incidence, Risk Factors, and Outcomes of Rhegmatogenous Retinal Detachment after Intravitreal Injections of Anti-VEGF for Retinal Diseases

Correspondence: phgabrielle@gmail.comInternational audiencePURPOSE: To report the estimated incidence, probability, risk factors and one-year outcomes of rhegmatogenous retinal detachment (RRD) in eyes receiving intravitreal injections (IVT) of vascular endothelial growth factor (VEGF) inhibitors for various retinal conditions in routine clinical practice. DESIGN: Retrospective analysis of data from a prospectively designed observational outcomes registry: the Fight Retinal Blindness! PARTICIPANTS: Eyes starting IVT with VEGF inhibitors (ranibizumab, aflibercept or bevacizumab) for neovascular age-related macular degeneration (nAMD), diabetic macular edema (DME) or retinal vein occlusion (RVO) from 1 January 2006 to 31 December 2020. All eyes that developed RRD within 90 days of an intravitreal injection were defined as RRD cases and were matched with control eyes. METHODS: Estimated incidence, probability, and hazard ratios (HR) of RRD were measured using Poisson regression, Kaplan-Meier survival curve and Cox-proportional hazards models. Locally weighted scatterplot smoothing curves were used to compare visual acuity (VA) between cases and matched controls. MAIN OUTCOME MEASURES: Estimated incidence of RRD. RESULTS: We identified 16915 eyes of 13792 patients who collectively received 265781 IVT over 14 years. Thirty-six eyes were reported to develop RRD over the study period. The estimated incidence (95% confidence interval [95%CI]) per year per 1000 patients and per 10000 injections was 0.77 (0.54, 1.07) and 1.36 (0.95, 1.89), respectively. The probability of RRD did not increase significantly at each successive injection (P = 0.95) with the time of follow-up. Older patients (hazard ratio [HR] [95%CI] = 1.81 [1.21, 3.62] for every decade increase in age, P < 0.01) were at higher risk of RRD, while patients with good presenting VA (HR = 0.85 [0.70, 0.98] for every 10-letter increase in VA, P = 0.02) were at a lower risk. Neither the type of retinal disease (P = 0.52) nor the VEGF inhibitor (P = 0.09) were significantly associated with RRD risk. RRD cases lost three lines of vision on average compared to the prior-RRD VA and had significantly fewer injections than matched controls over the year following the RRD. CONCLUSIONS: RRD is a rare complication of VEGF inhibitor IVT in routine clinical practice with poor visual outcomes at one year