Abstract
Purpose :
Endophthalmitis (ENDO) is a serious complication of intravitreal injections (IVI), often resulting in complete vision loss. The standard preventive measure is the application of 5% povidone-iodine (PI) to the conjunctival surface before injection, which significantly reduces the risk of infection. However, some patients refuse PI use due to the associated pain, preferring no antiseptic at all. PI has also been shown to cause corneal irritation and ocular surface abnormalities. This study aims to compare the risk of ENDO for IVIs with a reduced concentration of 2.5% PI to no PI use. For patients who decline standard PI treatment due to discomfort, it is crucial to understand how the use of lower concentrations of PI could balance the risk of infection with patient tolerance.
Methods :
We analyzed ten years of longitudinal data from the Barnet Dulaney Perkins Eye Center (2013-2023). Patients who have received IVIs from two providers, one who uses 2.5% PI and the other who does not use any PI, were included in the study. The primary outcome is the development of either culture positive or culture negative ENDO. Preoperative variables of potential risk factors of IVI infection were also collected. Chi-square and relative risk ratios were used to evaluate associations between pre-operative and/or intra-operative variables with post-operative outcomes.
Results :
During the study period, a total of 84,498 IVIs were administered to 3,572 patients. The mean age of patients was 76.2 years ± 5 (mean ± SD) with 52% of patients being male. 51 IVIs led to ENDO (0.0604%). The rate of ENDO in patients given 2.5% PI and no PI was 0.01178% and 0.1340% respectively. The rate of ENDO in no PI use was higher than that in a 2.5% PI protocol (Relative Risk: 11.36, p ≤ 0.0001).
Conclusions :
Our results show a significant increase in ENDO rate when no PI is used compared to 2.5% PI. Providers who want to forgo the use of PI for patient comfort may inadvertently increase their patient’s risk of ENDO. We show that the use of 2.5% PI instead could reduce ocular irritation and minimize the rate of ENDO. Future directions include expanding the study to other surgical centers and providers to reduce the effect of confounding variables, such as surgeon technique.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.