To report the incidence of endophthalmitis after senile cataract surgery and to describe the epidemiology and main risk factors.

All patients were evaluated retrospectively to compare risk factors, epidemiologic factors, and prophylaxis methods related to endophthalmitis. Patient records were used to gather the data. Epidemiologic factors, systemic diseases, other ocular pathologic characteristics, complications during the surgery, technique of cataract surgery, intraocular lens type, method of antibiotic prophylaxis, surgeon experience, vitreous culture, and vision outcome were evaluated in these patients.

One hundred twelve endophthalmitis cases among 480 104 operations reported, equaling an incidence of 0.023%. Patients with diabetes mellitus (14.3%) and of older age (mean age, 81 years), perioperative communication with the vitreous (17.9%), extracapsular cataract surgery procedure (11%), and surgery on the left eye (58.9% vs. 41.1% for right eye; P = 0.03) showed a statistically significant association with endophthalmitis. Short-term treatment with topical or systemic preoperative antibiotics or postoperative subconjunctival injection was associated with a 40% to 50% reduced odds of endophthalmitis compared with no prophylaxis (P = 0.2). No cases of endophthalmitis were observed among the 25 920 patients who received intracameral cefuroxime, suggesting that this approach to antibiotic prophylaxis may be far more effective than traditional topical or subconjunctival approaches.

The incidence of endophthalmitis after cataract surgery in our center was 0.023%, comparable with that of other previously published international studies. Older rural patients with immune suppressive diseases, such as diabetes mellitus, are particularly more prone to endophthalmitis. Vitreous loss at the time of surgery was associated with a significantly increased risk. Whereas antibiotic prophylaxis overall showed a 40% to 50% reduction in risk, intracameral cefuroxime was 100% effective in preventing endophthalmitis in this series.