Endophthalmitis rates after macular hole and epiretinal membrane surgery: a multicentre TriNetX study

Abstract Purpose Previous work has postulated that a post-vitrectomy gas bubble disrupts capillary action between the external eye and vitreous cavity, thereby reducing the risk of endophthalmitis. To test this theory, we will investigate the rates of endophthalmitis between macular hole surgery versus membrane peeling for epiretinal membrane. Methods Data was extracted from TriNetX, a de-identified global electronic health record database. Patients were queried for specific Current Procedural Terminology (CPT) codes 67041 and 67042, and International Classification of Disease (ICD) 10 codes H35.34x (Macular hole) and H35.37x (macular puckering/epiretinal membrane). Outcome was defined as any diagnosis of endophthalmitis (H44.0x) within 14 days of initial surgery. Propensity score matching (PSM) was performed to balance baseline demographic and clinical characteristics. Patients were excluded if they had any prior history of endophthalmitis. Results 23 490 eligible epiretinal membrane (ERM) surgeries and 15 346 eligible macular hole (MH) surgeries were included. After PSM, 14 545 matched surgeries in each cohort were analysed. In the ERM group, there were a total of 52 patients who developed endophthalmitis. In the MH group, there were 20. This yielded a risk ratio of 2.60 (95% CI: 1.55, 4.35; p-value of 0.000). Conclusion Pars plana vitrectomy for ERM appears to have a higher risk for subsequent endophthalmitis as compared to vitrectomy for MH. This study is limited by an inability to control for vitrectomy gauge and confirm use of gas or air tamponade.