Infectious endophthalmitis in adult eyes receiving Boston type I keratoprosthesis

To report the clinical characteristics of infectious endophthalmitis after Boston type I keratoprosthesis (K-Pro) implantation.

Retrospective study.

One hundred forty-one adult eyes receiving a K-Pro at a single institution from May 2004 through July 2008.

A retrospective chart review was performed of all adult eyes receiving a K-Pro at the University of Rochester from May 2004 through July 2008. Those patients identified as having been treated for exogenous bacterial endophthalmitis were reviewed for demographic data, indication for K-Pro, bandage contact lens use, prophylactic antibiotic use, timing and clinical presentation of endophthalmitis, gram stain and culture results of intraocular fluid, timing and presentation of any subsequent episodes of endophthalmitis (recurrent endophthalmitis), and preoperative and postoperative visual acuity through August 2010.

Incidence of endophthalmitis, time to occurrence, recurrence rates, visual outcomes, and risk factors associated with K-Pro endophthalmitis.

Ten (7.1%) of 141 eyes of 130 adult patients were diagnosed and treated for bacterial endophthalmitis. Average time to endophthalmitis developing after K-Pro was 9.8 months (standard deviation [SD], 6.2 months; range, 2-25 months). Coagulase-negative staphylococci were identified in 7 eyes. In 7 of the 10 eyes, recurrent endophthalmitis developed that occurred at a mean of 4 months (SD, 3.9 months; range, 1-13 months) after resolution of the initial episode. At each episode of endophthalmitis, no eye was receiving vancomycin ophthalmic drops and most eyes were receiving only fluoroquinolone ophthalmic drops for prophylaxis.

Infectious endophthalmitis after K-Pro implantation has a higher incidence, delayed onset, and high risk for recurrence compared with postoperative endophthalmitis associated with more common intraocular procedures such as cataract surgery. The concurrent use of topical vancomycin is recommended because it seems to be important in reducing the incidence and recurrence of endophthalmitis and because fluoroquinolone ophthalmic drops do not seem to be sufficient prophylaxis in these eyes.