Abstract
PURPOSE: This study characterizes the epidemiology, microbiological profile, management and outcomes of endophthalmitis over 18 years at a tertiary referral center in regional New South Wales (NSW), Australia. METHODS: A retrospective observational study of all endophthalmitis cases at John Hunter Hospital (November 2006 to December 2024) identified via ICD-10 codes and confirmed via record review. Data included demographics, clinical presentation, etiology, microbiology, management and surgical interventions. RESULTS: In total, 232 cases (227 patients) were identified (median age 75.6 years). Annual incidence declined significantly (slope = -0.17 cases/year, 95% CI -0.29 to -0.052); p = 0.005). Exogenous endophthalmitis accounted for 87.1% of cases, most commonly following cataract surgery or intravitreal injection (38.1% each). Endogenous cases (12.9%) were mainly associated with bacteremia (56.7%) and immunosuppression (20%). Culture positivity was 56%, with 97.7% of isolates being bacterial. Gram positive cocci were identified in 84% of culture-proven cases, predominantly coagulase-negative staphylococci (42%), while gram-negative organisms comprised 9.4%. Intravitreal antibiotics were administered in 97.3% of cases and pars plana vitrectomy (PPV) performed in 41.5%. Overall, 52.5% of eyes improved in visual acuity. The Win Ratio for improvement with vitrectomy versus antibiotics alone was 1.31 in favor of vitrectomy (95% CI 0.77-2.22; p = 0.31). In eyes presenting with light perception vision, the Win Ratio was 2.28 in favor of vitrectomy (95% CI 0.79-6.55; p = 0.13). CONCLUSION: Exogenous endophthalmitis remains the leading cause of endophthalmitis at a major regional tertiary referral center. Findings align with the Endophthalmitis Vitrectomy Study, suggesting vitrectomy benefits mainly eyes presenting with light perception only vision.