Related Experiment Video
Updated: Jan 7, 2026
Intravitreal Injection and Quantitation of Infection Parameters in a Mouse Model of Bacterial Endophthalmitis
Published on: February 6, 2021
Incidence and Microbiology of Endophthalmitis in Eyes With a Type 1 Boston Keratoprosthesis: A Systematic Review and
David M Morcos1, Thomas H Dohlman2, Marlene L Durand2,3
1Department of Ophthalmology and Visual Sciences, Washington University, St. Louis, MO.
Purpose:
To evaluate recent literature to determine the incidence and microbiology of endophthalmitis in eyes with the type 1 Boston Keratoprosthesis (BKPro).
Methods:
A literature review was performed using PubMed, Embase, and Cochrane databases. Studies were included if published ≥2010, included ≥20 BKPro eyes, and reported the mean follow-up time. Studies with duplicate data sets or limited to 1 type of underlying condition or age group were excluded. A meta-analysis was performed to calculate endophthalmitis incidence (cases per 100 eye-years).
Results:
Of 501 studies reviewed, 34 met inclusion criteria; these included 2732 BKPro eyes. The mean follow-up (range 1.4-8.9 years) was <5 years in 85% of studies. A total of 161 eyes developed endophthalmitis. Most studies reported the proportion of their cohort that developed endophthalmitis (0%-22%) but not the incidence. The pooled mean incidence of endophthalmitis was 2.09 cases per 100 eye-years (95% CI, 1.63-2.67). In 44 endophthalmitis cases with complete microbiology, 54.5% were bacterial (two thirds because of gram-positive bacteria), 29.5% fungal (primarily Candida ), and 16% no growth. Some cases occurred despite topical antibiotics effective against the endophthalmitis pathogen. Risk factors included antecedent keratitis, noncompliance with prophylactic antibiotics, corneal melt around the BKPro, and an exposed glaucoma drainage device.
Conclusions:
The incidence of endophthalmitis in BKPro eyes is 2 cases per 100-eye years, based on 34 recent studies with relatively short follow-up times. Future studies should report endophthalmitis incidence rather than proportion. More studies with long follow-up times will be helpful in determining the 5- and 10-year risk of endophthalmitis.
