Ocular Microbiology and Immunology Group
Back to OMIG Main Page

2012 Agenda and Abstracts | < Previous | Next >

2012 OMIG Abstract 3

Boston Keratoprosthesis-associated Endophthalmitis:
Global Incidence and Recommendations for Prevention
I. Behlau, K.V. Martin, J.N. Martin, R. Walcott-Harris, J. Chodosh, and C.H. Dohlman
Department of Ophthalmology, Massachusetts Eye and Ear Infirmary
Harvard Medical School, Boston, MA

Purpose: To determine the incidence and etiologies of infectious endophthalmitis post-Boston Keratoprosthesis Type I/II implantation, correlate with antimicrobial prophylactic regimens, and provide guidance to safeguard its use.

Methods: The incidence and risk factors for endophthalmitis associated with the Boston Keratoprosthesis were determined by two retrospective methods: 1) electronic medical database systematic literature review and meta-analysis from 1990 through Feb 17, 2012, and 2) endophthalmitis surveillance survey sent to all KPro surgeons who implanted Boston KPro from 1990 through December 2010 with one year follow-up. The incidence of endophthalmitis, visual outcomes post-endophthalmitis, antimicrobial prophylactic regimens, and microbiologic results were assessed.

Results: The number of Boston KPros implanted worldwide through Dec. 2010 was 4728 as determined by invoices and/or confirmed by a reporting surgeon. The U.S. surgeon response rate was 95.2% (199/209) with 99.5% (3483/3500) implanted KPros accounted for. The overall U.S. cumulative incidence of endophthalmitis is 3.17% (111/3500). The international (non-US) surgeon response rate was 66.7% (106/159) with 86.2% (1059/1228) implanted KPros accounted for. The international incidence of endophthalmitis is no less than 4.97% (61/1228). In Boston, the incidence of bacterial endophthalmitis has declined from 11% to 1% over subsequent decades with an emergence of 1.7% fungal keratitis/endophthalmitis, often treatable, during the latter decade.

Conclusions: In select large KPro centers, the risk of endophthalmitis has declined to a range of 1%. Given the majority of endophthalmitis is due to Gram-positive organisms, antimicrobial coverage must address this group of organisms, but be inclusive of Gram-negative coverage. Selection of prophylactic regimens should be tailored to local antibiotic susceptibility patterns, be cost-effective, and should not promote the emergence of antimicrobial resistance nor encourage fungal colonization.  Examples of effective regimens will be suggested.

Disclosure: S    Fight-for-Sight Grant-in-Aid (IB) and MEEI KPro Fund

2012 Agenda and Abstracts | < Previous | Next >