Fungal colonization and infection in Boston keratoprosthesis

Cornea. 2007 Jan;26(1):9-15. doi: 10.1097/01.ico.0000224650.19837.25.

Abstract

Purpose: To determine the incidence of and risk factors for fungal keratitis and endophthalmitis in patients with a Boston keratoprosthesis (KPro) and to determine whether surveillance cultures were helpful in predicting fungal infection.

Methods: A retrospective chart review was performed of 182 patients (202 eyes) who received a type 1 (through cornea) or type 2 (through cornea and lid) KPro between March 1, 1990, and December 31, 2004, and who were followed for at least 1 month (range, 1 month to 13 years; mean, 2.84 years). There were 148 eyes with type 1 and 54 eyes with type 2. Beginning in late 1999, many eyes were given a prophylactic topical regimen containing vancomycin, and many eyes with type 1 KPro were given therapeutic contact lenses. Cases of fungal keratitis or endophthalmitis were analyzed. To determine the fungal colonization rate, 70 surveillance cultures of the ocular or lid surface around the KPro optic were obtained of 36 uninfected KPro eyes (35 patients) at random time-points over 1 year (August 2002 to July 2003).

Results: There were 4 definite and 1 probable fungal infections in 6893 patient-months of follow-up, or 0.009 fungal infections per patient-year. These included 3 cases of definite or probable Candida endophthalmitis (C. parapsilosis, C. glabrata, and C. albicans) and 2 cases of mold keratitis (Alternaria, Fusarium). The rate was higher in eyes receiving a vancomycin-containing topical prophylactic regimen than those with on a non-vancomycin regimen (5 cases/2774 person-months vs. 0 cases/4119 person-months; P = 0.011). In eyes with type 1 KPro, the rate was higher with therapeutic contact lens wear than without (4/1682 vs. 0/3115 person-months; P = 0.015). Surveillance cultures did not predict fungal infection, and none of the 6 surveillance eyes colonized with fungi (all Candida) developed a fungal infection. The prevalence of fungal colonization in KPro eyes had not changed since our 1996 surveillance study (11% vs. 10%, P > 0.05).

Conclusion: Fungal infections in KPro eyes have appeared since we introduced broad-spectrum antibiotic prophylaxis and therapeutic contact lenses 5 years ago, but the infection rate remains very low in our mostly New England-based patient population. Cleaning or replacing the contact lens on a regular basis and prescribing a short course of topical amphotericin at the first visible signs of fungal colonization may prevent these infections.

Publication types

  • Case Reports

MeSH terms

  • Adult
  • Aged
  • Alternaria / isolation & purification
  • Anti-Bacterial Agents / therapeutic use
  • Candida / isolation & purification
  • Cornea / microbiology
  • Cornea / surgery
  • Drug Therapy, Combination
  • Endophthalmitis / diagnosis
  • Endophthalmitis / drug therapy
  • Endophthalmitis / microbiology*
  • Eye Infections, Fungal / diagnosis
  • Eye Infections, Fungal / drug therapy
  • Eye Infections, Fungal / microbiology*
  • Female
  • Fusarium / isolation & purification
  • Humans
  • Incidence
  • Keratitis / diagnosis
  • Keratitis / drug therapy
  • Keratitis / microbiology*
  • Male
  • Middle Aged
  • Mycoses / diagnosis
  • Mycoses / drug therapy
  • Mycoses / microbiology*
  • Prosthesis Implantation
  • Prosthesis-Related Infections / diagnosis
  • Prosthesis-Related Infections / drug therapy
  • Prosthesis-Related Infections / microbiology*
  • Retrospective Studies
  • Risk Factors
  • Vitreous Body / microbiology

Substances

  • Anti-Bacterial Agents