Search

Recurrence of Pseudomonas keratitis during treatment with corticosteroids has been reported previously in humans. Rabbits with keratitis due to Pseudomonas aeruginosa or Streptococcus pneumoniae were treated with antibiotics and either vehicle, methylprednisolone acetate, flurbiprofen, or nordihydroguaiaretic acid (NDGA). Cultures performed after 7 days were negative, and antibiotics were discontinued. Two weeks later, Pseudomonas keratitis recurred in 6 of 7 (85.7%) steroid-treated rabbits, 1 of 8 (12.5%) flurbiprofen-treated rabbits, 1 of 8 (12.5%) NDGA-treated rabbits, and none of 8 vehicle-treated rabbits. None of the 31 rabbits infected with Streptococcus pneumoniae experienced recurrence. These data confirm the clinical observation that Pseudomonas keratitis may recur if antibiotic therapy is discontinued and corticosteroids are administered; the risk of recurrence appears to be much less with nonsteroidal antiinflammatory agents.

Corneal scarring as a consequence of bacterial keratitis is an important cause of visual loss and a major indication for penetrating keratoplasty. Anti-inflammatory agents might be useful in this condition for limiting corneal damage, but benefit from adjunctive anti-inflammatory therapy has never been demonstrated. In this limited pilot study, we compared the effect on clinical outcome of treating Pseudomonas keratitis in guinea pigs with prednisolone (a corticosteroid), flurbiprofen (a cyclo-oxygenase inhibitor), nordihydroguaiaretic acid (a lipoxygenase inhibitor), and a leukotriene antagonist, SKF104353 [R-(R*, S*)]-beta-[(2-carboxyethyl) thio-alpha-hydroxy-2-(8-phenyloctyl) benzenepropanoic acid, zinc salt]. None of the anti-inflammatory agents prevented sterilization of ulcers with antibiotic (ofloxacin) therapy. Therapy with the leukotriene antagonist appeared to reduce infiltrate size more quickly and produce a more rapid reduction in lesion size, but the differences were not statistically significant. Sample size calculations suggest that very large numbers of animals would be required to prove efficacy. The role of anti-inflammatory agents in reducing the stromal destruction caused by bacterial keratitis remains controversial.

Ofloxacin, a new quinolone antibiotic with a broad spectrum of activity, is very effective against Pseudomonas aeruginosa in vitro. Its effectiveness was studied in a rabbit model of tobramycin-sensitive P. aeruginosa. Treatment groups received either vehicle, tobramycin 0.3%, or ofloxacin 0.3%. Twelve hours of treatment decreased the bacterial counts from a mean of 2.2 +/- 0.7 x 10(6) colony forming units (cfu) per cornea in the vehicle group to means of 513 +/- 670 and 435 +/- 524 cfu in the tobramycin and ofloxacin groups, respectively. This decrease in bacterial counts was statistically significant (p = 0.001 for tobramycin and ofloxacin each compared with control, p = 0.86 for tobramycin compared with ofloxacin). After seven days, all antibiotic-treated corneas were sterile and the epithelial defects healed at comparable rates. Aqueous humor drug levels were higher in infected eyes without an intact epithelium (p = 0.02); in eyes with intact epithelium, concentrations of ofloxacin were higher than were those of tobramycin (p = 0.002). In this animal model, ofloxacin proved to be an effective antibiotic with no evidence of toxicity.

Topical steroids potentiate Pseudomonas keratitis in the absence of concomitant antibiotic therapy. The effects of other anti-inflammatory agents on microbial keratitis are unknown, but if these agents do not intensify the infection, they may be useful in the initial management of microbial keratitis by limiting ulceration and thus the size of the resultant corneal scar. In rabbit models of untreated Pseudomonas keratitis and pneumococcal keratitis, topical 1% prednisolone phosphate, 0.03% flurbiprofen sodium, and vehicle were applied hourly. In cases of Pseudomonas keratitis, prednisolone worsened the clinical disease, and flurbiprofen further worsened the disease. When combined with effective antibiotic therapy, neither anti-inflammatory agent worsened the disease. In pneumococcal keratitis, neither anti-inflammatory agent worsened the disease. Topical anti-inflammatory agents should be used with caution in cases of microbial keratitis at least until effective antibiotic therapy has been instituted.