Purchase this article with an account.
Michael Ramos, Srinidhi Singuri, Lucy Xu, Julio Castillo, Justis P Ehlers, Peter M Kaiser, Rishi P Singh, Aleksandra Rachitskaya, Sunil K Srivastava, Jonathan E Sears, Andrew Schachat, Amy Babiuch, Sumit Sharma, Daniel F Martin, Careen Y Lowder, Alex Yuan; Presentation and outcomes of patients with post-injection endophthalmitis at a large academic center. Invest. Ophthalmol. Vis. Sci. 2020;61(7):4410.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Intravitreal injections are common treatments for many ocular diseases. While rare, complications such as endophthalmitis can occur. With this study, we report endophthalmitis rate, onset to presentation, clinical setting on presentation (ie: phone, office, or emergency room (ER)), and outcomes. Knowledge of presenting characteristics and overall outcomes can help guide future counseling of patients.
A retrospective, IRB-approved review of the electronic medical record system at a large academic center was performed. Intravitreal injections performed between 2012 to May 2019 were reviewed. Presumed endophthalmitis cases were identified by ICD-10 diagnosis codes H44.001, H44.002, H44.003, or H44.009 within 14 days of injection. Data collected included encounter type, days to presentation, and visual acuity outcomes.
From 2012-May 2019, 84,973 injections were performed, and 33 cases (0.039%) of endophthalmitis relating to an injection were identified. 23 cases were culture-positive. 8 cases were culture-negative. In 2 cases, cultures were unavailable due to a dry tap. Average days to presentation were 4.4 and 4.2 (p=0.88) for culture-positive and –negative/unavailable cases, respectively. At office presentation, vision had decreased 14.2 and 14.1 lines (p=.98). Average line changes at 1, 6, and 12 months after initial presentation were not statistically different, however, vision was decreased from baseline at all timepoints. At 12 months, culture-positive and culture-negative/unavailable cases showed decreases of 7.0 and 6.1 lines, respectively (p=0.85).Initial encounters showed 57.6% of endophthalmitis cases presenting in the office, 27.3% by phone, and 15.2% by ER. All initial phone encounters had a subsequent office visit. A control cohort of all non-infectious complications related to intravitreal injections presented with 65.4% by phone, 32.3% by office, and 2.3% by ER, which was statistically significant compared to the endophthalmitis cohort (X2= 20.18, p=7.0x10-6).
Endophthalmitis following intravitreal injection is uncommon, and the reported incidence (0.039%) is consistent with published reports. No statistical differences were found between culture-positive and –negative/unavailable groups in terms of days to presentation and average line change at 1, 6, and 12 months and both groups failed to return to baseline vision 12 months after encounter.
This is a 2020 ARVO Annual Meeting abstract.
This PDF is available to Subscribers Only