Cohort study of 27 cases of endophthalmitis at a single institution

PURPOSE: To identify potential risk factors associated with post-cataract surgery bacterial endophthalmitis. SETTING: The John A. Moran Eye Center, Salt Lake City, Utah, USA. METHODS: This retrospective cohort study consisted of patients who had surgery for cataract(s) at this eye hospital. A 10% sampling of all patients operated on for cataract surgery from January 1, 1996, to December 31, 2002, were compared with all cases of postcataract surgery bacterial endophthalmitis during this same time period at this institution. The main outcome measure(s) included surgical complication, first postoperative day wound leak, incision placement and location, intraocular lens material, whether a suture was placed, antibiotic used, collagen shield use, and whether the eye was patched. RESULTS: A total of 1525 patients were in the control cohort, and there were 27 cases of endophthalmitis. In a multivariate regression analysis, the factors found to be statistically associated with endophthalmitis were (1) wound leak on the first postoperative day (odds ratio [OR] 44 +/- 42; confidence interval [CI] 6.85 to 287; P<.001); (2) capsular or zonular surgical complication (OR 17.2 +/- 14.2; CI 3.44 to 86.4; P=.001); (3) topical antibiotic started the day after surgery rather than the day of surgery (OR 13.7 +/- 12.9; CI 2.17 to 90.9; P=.005); (4) use of ciprofloxacin rather than ofloxacin topically after surgery (OR 5.3 +/- 3.6; CI 1.41 to 20.0; P=.014); (5) not patching after surgery (OR 7.1 +/- 5.6; CI 1.47 to 36.4; P=.015); and (6) not placing a collagen shield soaked in antibiotic (OR 2.7 +/- 1.3; CI 1.06 to 7.14 P=.037). CONCLUSION: In sutureless cataract surgery, surgical complications and wound leak on the first postoperative day were most strongly associated with endophthalmitis.