Endophthalmitis After Vitrectomy Surgery: A Case–Control Analysis

Abstract

Purpose: To identify risk factors associated with endophthalmitis after pars plana vitrectomy (PPV) surgery. Methods: All eyes that presented with endophthalmitis after PPV from 2005 to 2015 in a community-based vitreoretinal subspecialty practice were evaluated using a retrospective nested case–control study. Exclusion criteria were presentation more than 6 weeks post-PPV, less than 3 months of follow-up, history of globe trauma, age less than 18, and history of trabeculectomy surgery. Controls were randomly selected among a population of patients who underwent PPV and did not develop endophthalmitis. Univariate and multivariate regression analyses were performed to evaluate the following perioperative risk factors: use of sclerotomy suture, use of subconjunctival antibiotics, use of endotamponade, presence of postoperative hypotony, and the use of small- versus large-gauge vitrectomy. Results: Thirty-two patients and 150 controls met inclusion criteria. The average time to presentation with endophthalmitis was 5 days after PPV. Endotamponade usage was the only variable that had significant impact on the risk of developing endophthalmitis in univariate and multivariate analyses. The odds ratio of disease in patients with endotamponade over patients without endotamponade was 0.15 (95% confidence interval: 0.06-40); thus, the odds of disease for patients with endotamponade were 85% lower than those without endotamponade. Median visual acuity significantly improved from 20 of 8000 at the time of diagnosis with endophthalmitis to 20 of 50 at 1-year follow-up ( P < .001). Conclusion: The use of intraocular endotamponade may reduce the rate of endophthalmitis after PPV surgery.