Bacterial isolates and antimicrobial susceptibilities in equine bacterial ulcerative keratitis (1993–2004).

Abstract: Bacterial ulcerative keratitis is a common and often vision-threatening problem in horses. Emerging bacterial resistance to commonly used topical antibiotics has been demonstrated. Previous antibiotic use may alter the antimicrobial susceptibility of bacterial isolates. Objective: To document aerobic bacterial isolates and associated bacterial susceptibilities from horses with ulcerative keratitis treated at the University of Tennessee between January 1993 and May 2004 and determine whether prior antibiotic therapy affected antimicrobial susceptibility of the isolates. Methods: Medical records from horses with ulcerative keratitis and positive aerobic bacterial cultures and antimicrobial susceptibility were evaluated. Clinical history regarding antibiotic therapy prior to culture was documented. Results: Fifty-one aerobic bacterial isolates from 43 horses were identified. Streptococcus equi subspecies zooepidemicus was the most commonly isolated organism, accounting for 33.3% of all isolates, followed by Pseudomonas aeruginosa (11.8%), Staphylococcus spp. (11.8%) and Gram-negative nonfermenting rods (7.8 %). No resistance was noted amongst S. equi ssp. zooepidemicus to cephalothin, chloramphenicol or ciprofloxacin. Only 64 % of S. equi ssp. zooepidemicus isolates were sensitive to bacitracin. No resistance was noted among P. aeruginosa to gentamicin, tobramycin or ciprofloxacin. Antibiotic therapy with neomycin-polymixin B-bacitracin prior to presentation and culture was documented in 11/17 horses in which S. equi ssp. zooepidemicus was isolated and in 4/6 horses in which P. aeruginosa was isolated. Three horses received topical corticosteroids prior to culture, of which 2 had polymicrobial infections. Conclusions: S. equi ssp. zooepidemicus and P. aeruginosa were the most frequently isolated bacterial organisms in equine ulcerative keratitis. No significant trends in aminoglycoside or fluoroquinolone resistance were noted among these organisms. However, the resistance of S. equi ssp. zooepidemicus to bacitracin with common use of this antibiotic suggests that previous antibiotic therapy probably affects antimicrobial resistance. Conclusions: Therapy prior to culture may play an important role in antimicrobial susceptibility of corneal bacterial isolates. Corticosteroid use may increase the risk of polymicrobial infections of corneal ulcers, leading to a worse prognosis. Although significant fluoroquinolone resistance has not been documented in the veterinary literature, these antimicrobials should be reserved for known infected corneal ulcers and not used for prophylaxis. Empirical antibiotic therapy should not only be guided by clinical signs, but also take into consideration previous antimicrobial and corticosteroid therapy.