Abstract
PURPOSE: To evaluate the role of patient facial masks on the occurrence of post-intravitreal injection (IVI) endophthalmitis in a real-word setting. METHODS: In this retrospective cohort, patients receiving IVIs between 20 February 2019 and 20 February 2021; a 12-month period before the official beginning of COVID-19 epidemic in Iran and a 12-month period thereafter were included. In the pre-COVID era, patients underwent IVI without a facial mask while in the COVID era patients wore an untaped facial mask. Physicians and staff had facial mask in both periods. IVIs were administered in a dedicated operating room without a strict no talk-policy. The main outcome measure was the rate of post-IVI endophthalmitis. RESULTS: A total number of 53,927 injections was performed during the study period: 34,277 in pre-COVID and 19,650 in COVID periods; with a 42.7% decrease in the number of injections. Endophthalmitis occurred in 7 eyes (0.020%) in pre-COVID and 7 eyes (0.036%) in COVID era (p = 0.40). In multivariate analysis, after adjustment for intercorrelations between the eyes and multiple injections in one patient, there was no statistically significant association between wearing facial masks by the patients and risk of endophthalmitis (relative risk = 1.47, 95% confidence interval of 0.97-2.22; p = 0.071). CONCLUSION: Patients' facial masking is not associated with an increased risk of post-injection endophthalmitis.