A meta-analysis of intravitreal ranibizumab versus laser photocoagulation for the treatment of retinopathy of prematurity.

Abstract

Purpose: Laser photocoagulation (LPC) has been a traditional treatment for retinopathy of prematurity (ROP). However, intravitreal anti-VEGF agents such as bevacizumab and ranibizumab (IVR) have also been increasingly used. This meta-analysis aims to rigorously compare IVR to LPC in the treatment of ROP.

Methods: Medline, Embase and Cochrane CENTRAL were used to identify studies comparing IVR monotherapy to LPC (PROSPERO ID: CRD42023390855). The primary outcome was ROP regression. Secondary outcomes included likelihood of additional treatment, time from treatment to reactivation or re-treatment, refractive outcomes and adverse events such as retinal detachment, cataract, macular dragging/ectopia, vitreous or retinal hemorrhage, glaucoma, and endophthalmitis. A random effects meta-analysis was designed.

Results: 2361 articles were identified. 1947 eyes from 7 cohort studies, 1 case-control study and 2 RCTs were included with a median follow-up of 21 months (range: 11-75 months). There was no significant difference in disease regression between IVR and LPC (risk ratio [RR]=0.96, 95% confidence interval [CI] [0.83, 1.10], p=0.52), however, eyes that underwent IVR were associated with a higher likelihood of requiring additional treatment (RR= 2.70, CI= [1.55, 4.68], p<0.001). Although less frequent, retreatment occurred earlier with LPC compared to IVR (weighted mean difference (WMD)= -4.29 weeks, CI= [-6.48, -2.10], p<0.001). Furthermore, eyes that received IVR had a lower refractive error, with a WMD of -0.93 diopters (CI= [-1.54, -0.32], p=0.003) at a median age of assessment of 5.0 years (range 1.5-6.3 years). There was no difference in the rate of adverse events between LPC and IVR (p>0.05 for RD, MDR, VH and cataract). Quality of evidence was rated moderate for likelihood and time of additional treatment, as well as refractive error, but was considered low for disease regression and adverse events.

Conclusion: Compared to LPC, IVR was associated with a higher likelihood of requiring additional treatment but a lower risk of myopia. More studies are needed to evaluate dose-response relationships and temporal trends in ROP regression following these treatments.