Incidence and Visual Outcomes of Culture-Proven Endophthalmitis Following Dexamethasone Intravitreal Implant

Importance The rate of endophthalmitis following dexamethasone intravitreal implant (DEX) has varied in large clinical trials. Furthermore, to our knowledge, the optimal management of eyes with endophthalmitis associated with DEX has not been established. Objective To report the incidence of culture-proven endophthalmitis in a single vitreoretinal practice over the course of 3 years and describe the clinical outcomes associated with each case of endophthalmitis. Design, Setting, and Participants All patients who received DEX between January 14, 2013, and August 31, 2016, were included in this retrospective single-center case series at a private vitreoretinal practice. The patients were identified during a search of the billing records over the period of interest. Cases of endophthalmitis associated with DEX were also identified. Exposures Treatment with DEX. Main Outcomes and Measures Development of endophthalmitis following DEX and the clinical management and outcomes of each case of endophthalmitis. Results Of the 1051 participants who collectively received 3593 injections of DEX, 4 patients developed endophthalmitis; all 4 patients were white, female, and 60 years or older (mean [SD] age, 75.6 [13] years). Two patients had culture-proven bacterial endophthalmitis after DEX monoinjections (0.06% of injections and 0.2% of patients). Three other cases of endophthalmitis developed after coinjection with bevacizumab (aggregate rate: 0.14% of injections and 0.38% of patients), of which 2 were culture positive. One patient developed endophthalmitis on 2 separate occasions. Vitrectomy was performed in 2 patients, and in 1 of these patients, the implant was removed. All 4 patients were treated with injection of intravitreous vancomycin and ceftazidime. Conclusions and Relevance These data suggest that endophthalmitis is a rare event following injection of DEX. However, given the rarity of endophthalmitis following DEX and the heterogeneity among our reported cases, it remains unclear whether the DEX endophthalmitis rate approximates that of intravitreous anti–vascular endothelial growth factor therapy. These data also suggest that vitrectomy with removal of DEX may not be necessary in all cases of DEX-associated endophthalmitis.