Comparative incidence of endophthalmitis after intravitreal dexamethasone implant versus anti-VEGF injections: a retrospective study

Endophthalmitis is a rare but vision-threatening complication of intravitreal injections. Intravitreal dexamethasone implants and anti–vascular endothelial growth factor (anti-VEGF) agents are widely used, yet comparative real-world endophthalmitis rates remain incompletely defined. This study aimed to compare the incidence of endophthalmitis following intravitreal dexamethasone implant injections with that following anti-VEGF injections administered in outpatient retina practices. This retrospective cohort study included all intravitreal injections performed between January 2019 and January 2024 at participating practices. Endophthalmitis events consistent with post-injection exogenous endophthalmitis occurring within two weeks of intravitreal injection were identified through clinical records. Chi-square testing was used to compare incidence rates between dexamethasone implant and anti-VEGF injections. Patient characteristics, visual acuity outcomes, and treatment approaches were assessed for confirmed endophthalmitis cases. A total of 330,572 intravitreal injections were analyzed, including 318,618 anti-VEGF injections and 11,954 dexamethasone implant injections. Endophthalmitis occurred more frequently after dexamethasone implant injections (0.125%, 15/11,954; 1.25 per 1,000 injections) than after anti-VEGF injections (0.033%, 106/318,618; 0.33 per 1,000 injections). This difference was statistically significant (χ² = 23.45, p = 2.3 × 10⁻⁷). Among dexamethasone-associated cases, 46.7% recovered to baseline or better visual acuity by six months, while the remaining patients experienced persistent visual deficits. Management commonly included intravitreal antibiotics with or without adjunct corticosteroids. Intravitreal dexamethasone implant injections were associated with a significantly higher incidence of endophthalmitis compared with anti-VEGF injections in this large outpatient cohort. Although nearly half of affected patients regained baseline vision, many had lasting impairment. These findings highlight the need for careful risk–benefit assessment when selecting intravitreal dexamethasone implants and underscore the importance of further study into modifiable risk factors and prevention strategies. A key limitation to consider is that clinically diagnosed endophthalmitis may include sterile intraocular inflammatory reactions following dexamethasone implant injection.