Incidence, microbiology, and outcomes of endophthalmitis after 111,876 pars plana vitrectomies at a single, tertiary eye care hospital

PURPOSE: To describe the incidence, risk factors, clinical presentation, causative organisms, and outcomes in patients with endophthalmitis following pars plana vitrectomy (20G and minimally invasive vitrectomy surgery (MIVS). METHODS: Of 111,876 vitrectomies (70,585 20-G 41,291 MIVS) performed, 45 cases developed acute-onset, postoperative endophthalmitis. RESULTS: The rate of culture positive and culture negative endophthalmitis was 0.021% (2.1/10,000 surgeries) and 0.019% (1.9/10,000 surgeries) overall, 0.031% (3.1/10,000 surgeries) and 0.025% (2.5/10,000 surgeries) in 20G, and 0.005% (0.5/10,000 surgeries) and 0.007% (0.7/10,000 surgeries) in the MIVS group respectively. Potential predisposing factors were as follows: diabetes, 46.7%; vitrectomy for vascular retinopathies, 44.4%; and vitrectomy combined with anterior segment surgeries, 35.5%. The culture proven rates were 53.3% overall, 55.0% for 20G and 40.0% for MIVS. The most common organism was Pseudomonas aeruginosa for 20G. Klebsiella and Staphylococcus aureus were isolated in the two culture positive cases in MIVS group. The follow-up period for the patients with endophthalmitis was 586.14 ± 825.15 days. Seven were lost to follow up beyond one week. Of the remaining 38, 13 (34.2%) cases had a favorable visual outcome (i.e., best-corrected visual acuity [BCVA] > 5/200) and 24 (63.2%) had unfavorable visual outcome (BCVA < 5/200). Group with culture test results negative had significantly better outcomes (P < 0.05) as compared to those with positive. CONCLUSIONS: MIVS does not increase the risk of endophthalmitis. Outcomes are poor despite appropriate treatment, particularly in cases with culture results positive.