Bacterial Keratitis after Penetrating Keratoplasty:
Incidence, Microbiological Profile, Graft Survival, and Visual Outcome
Purpose
To determine the incidence, microbiological profile, graft survival, and factors influencing graft survival after the development of bacterial keratitis after penetrating keratoplasty (PK).
Design
Retrospective case series.
Participants
One hundred two patients (102 eyes) treated at a single center during a 5-year period.
Methods
Retrospective review of the medical records of every patient treated for culture-positive keratitis between January 1, 1998 and December 31, 2002 who previously had undergone penetrating keratoplasty at the King Khaled Eye Specialist Hospital.
Main Outcome Measures
Graft survival and visual outcome.
Results
There were 2103 PKs performed and 102 (4.9%) cases of culture-positive keratitis during the study period. There were 168 bacterial isolates, of which 140 (83.3%) were gram positive, 28 (16.7%) were gram negative, and 1 (0.6%) was acid fast. Only 38 (37.3%) grafts remained clear after a mean follow-up of 985 days (range, 82–2284). The best graft survival was in eyes with PK for keratoconus (83.7%), whereas the poorest grat survival was for previously failed grafts (5.6%). By Kaplan–Meier analysis, there was an immediate steep decline in graft survival to 54.9%, followed by a slow decline to 47.2% by 1 year and 35.8% after 4 years. Factors associated with an increased risk of graft failure were the surgical indication for PK (P<0.001), increasing patient age (P = 0.004), smaller donor (P = 0.001) and recipient (P = 0.0003) graft size, history of previous microbial keratitis (P = 0.02) or endothelial rejection episodes (P = 0.02), and coexisting glaucoma (P = 0.001). The visual outcome was ≥20/40 in only 8 (8.2%) eyes and better than 20/200 in only 21 (21.6%) eyes.
Conclusion
The development of bacterial keratitis after PK is a serious complication that is associated with a high incidence of graft failure and poor visual outcome.
To access this article, please choose from the options below
Manuscript no. 2006-711.
The authors have no conflicts of interest with or proprietary interest in any of the topics presented in the article.
PII: S0161-6420(06)01372-8
doi:10.1016/j.ophtha.2006.10.015
© 2007 American Academy of Ophthalmology. Published by Elsevier Inc. All rights reserved.

