Low rate of endophthalmitis in a large series of open globe injuries

To determine the percentage of patients in whom endophthalmitis developed after open globe injury.

Retrospective, noncomparative, consecutive case series.

Charts of all patients (675 in total) treated surgically for open globe injury at the Massachusetts Eye and Ear Infirmary (MEEI) between January 1, 2000 and July 31, 2007 were reviewed. Cases with at least 30 days of follow-up were included in statistical analyses (558 in total). A standardized treatment protocol was used in all cases. Intravenous vancomycin and ceftazidime were started on admission and were stopped after 48 hours. Patients were discharged on topical antibiotics, corticosteroids, and cycloplegia. Surgical repairs were performed by the chief of trauma, a full-time position rotating yearly, who is on call for all open globe trauma. Data collection variables included timing of injury and repair, mechanism of injury, details of surgical repair, and details of follow-up such as duration, presence of complications, and vision. A primary outcome measure of endophthalmitis and secondary outcome measure of risk factors for endophthalmitis were studied.

During 7.5 years, 675 open globe injuries were treated at MEEI. Of these, 558 had at least 30 days of follow-up (mean, 11 months) and were used in statistical analyses. The overall percentage of endophthalmitis was 0.9% (3 culture-positive cases and 2 culture-negative cases). Four of the 5 cases achieved final acuity of 20/80 or better. Risk factors for endophthalmitis included intraocular foreign body (P = .03; odds ratio, 7.52) and primary intraocular lens placement (P = .05).

A standardized protocol including surgical repair by a dedicated eye trauma service and 48 hours of intravenous antibiotics was associated with a posttraumatic endophthalmitis percentage of less than 1%.