Five-year risk of rhegmatogenous retinal detachment following prophylactic laser retinopexy for retinal breaks and lattice degeneration: a real-world analysis.

purposeTo compare the long-term risk of rhegmatogenous retinal detachment (RRD) following prophylactic laser retinopexy across different retinal break phenotypes.

methodsThis retrospective cohort study of de-identified electronic health records used the TriNetX US Collaborative Network to identify adults undergoing prophylactic laser retinopexy for horseshoe tear, unspecified retinal break, round retinal hole, or lattice degeneration between January 2016 and January 2020. Two pairwise propensity score-matched comparisons were performed: horseshoe tear versus unspecified retinal break and round retinal hole versus lattice degeneration. Patients with prior retinal detachment or the same qualifying diagnosis within one year before the index event were excluded. The primary outcome was incident RRD within five years. Secondary outcomes were epiretinal membrane (ERM) formation and vitrectomy with membrane peel.

resultsAfter propensity score matching, 1,259 and 1,588 patients were included per group for the horseshoe tear versus unspecified retinal break and the round hole versus lattice degeneration comparisons, respectively. Five-year RRD rates were similar between horseshoe tears (8.8%) and unspecified retinal breaks (8.9%; hazard ratio [HR], 1.01; 95% CI, 0.75-1.36; P = 0.95). Kaplan-Meier analysis demonstrated that RRD risk was concentrated early, rising most steeply within the first 7 days; the majority of 5-year events accrued within the first 60 days in most phenotypes with cumulative risk largely plateauing by 90 days across all phenotypes. Round retinal holes were associated with lower RRD risk (4.1%) compared with lattice degeneration (7.0%; HR, 0.57; 95% CI, 0.41-0.80; P = 0.001). ERM incidence showed no difference between horseshoe tears (11.2%) and unspecified breaks (10.4%; HR, 1.12; 95% CI, 0.87-1.44; P = 0.36), but was lower following round holes (6.7%) versus lattice degeneration (10.9%; HR, 0.59; 95% CI, 0.46-0.75; P < 0.001). Vitrectomy with membrane peel occurred below reportable thresholds. Contralateral horseshoe tear incidence was 5-6%.

conclusionIn this real-world cohort of patients treated with prophylactic laser retinopexy, five-year RRD rates were less than 10% across all phenotypes, and round holes showed lower observed RRD and ERM risk than lattice degeneration. Most RRD events occurred within the first 60 days, underscoring the clinical importance of early post-treatment surveillance. These results represent outcomes observed after treatment rather than a direct measure of prophylactic efficacy. As such, these observational data may inform individualized treatment decisions, particularly for lesions where evidence supporting prophylaxis remains limited.