Purpose: The emergence of ciprofloxacin-resistant Pseudomonas aeruginosa (CRPA) has created a new therapeutic challenge in ophthalmology. We evaluated ophthalmic antibiotics in vitro and in a rabbit keratitis model to determine effective therapy., Design: Experimental laboratory investigation., Methods: The susceptibilities of 12 CRPA isolates were determined in vitro for amikacin, ceftazidime, tobramycin, polymyxin B, gentamicin, ticarcillin, and the fluoroquinolones (that is, ciprofloxacin, ofloxacin, levofloxacin, gatifloxacin, and moxifloxacin) using E-tests and National Committee of Clinical Laboratory Standards. A rabbit keratitis model was used to determine the reduction in colony counts of CRPA and ciprofloxacin-susceptible P. aeruginosa (CSPA) isolates following topical treatment with polymyxin B/trimethoprim, tobramycin (14 mg/ml), ceftazidime (50 mg/ml), and ciprofloxacin (3 mg/ml)., Results: For 12 CRPA isolates, the susceptibilities and median minimum inhibitory concentrations ([MIC]microg/ml) were as follows: amikacin (92%, 14.0), ceftazidime (75%, 4.0), tobramycin (67%, 1.75), polymyxin B (42%, 7.0), gentamicin (17%, 7.0), ticarcillin (0%, >32.0), and all fluoroquinolones (0%, >32.0). While no antibiotic regimen reduced colony counts in the time frame of the animal model for CRPA, ciprofloxacin alone demonstrated a significant decrease in colony counts for CSPA. Comparing CRPA with CSPA, both tobramycin and ciprofloxacin demonstrated a significant decrease in colony counts for CSPA., Conclusion: Our laboratory studies suggest that current antibiotics may be suboptimal in treating CRPA keratitis. Until new antibiotics are available, combination therapy such as fortified tobramycin and ticarcillin, and others may prove effective in aggressive topical long-term therapy.