Evaluation of the Corneal Effects of Topical Ophthalmic Fluoroquinolones Using In Vivo Confocal Microscopy

PURPOSE: To compare the effects of several fluoroquinolone antibiotics on the corneal epithelium and stroma using in vivo confocal microscopy. METHODS: Five antibiotic solutions were evaluated: 1) 0.3% ofloxacin (Oflox) solution with 0.005% benzalkonium chloride (BAC); 2) 0.3% gatifloxacin (Gati) solution with 0.005% BAC; 3) 0.3% ciprofloxacin (Cipro) solution with 0.006% BAC; 4) 0.5% levofloxacin (Levo) with 0.005% BAC; and 5) 0.5% moxifloxacin (Moxi) solution with no BAC. Preservative-free artificial tears (Tears) were used as a control. New Zealand white rabbits were used for this study (six per solution group). Ten days prior to exposure to any solution, central corneal epithelial thickness and stromal thickness were measured using in vivo confocal microscopy through focusing. Images of the superficial epithelium were also acquired. Both eyes of each rabbit then received one drop of the assigned solution six times the first day and then four times per day for 6 days. On day 7, in vivo confocal microscopy was repeated. RESULTS: A significant decrease in epithelial thickness was induced by 7 days of exposure to Levo, Gati, Oflox, and Cipro (P < 0.05, two-way repeated-measures ANOVA, Tukey test). Tears and Moxi, which do not contain BAC, did not induce significant changes in epithelial thickness. No significant changes in stromal thickness were detected (P = 0.266), and no keratocyte activation was observed for any of the solutions evaluated. CONCLUSION: We have previously used confocal microscopy to establish a correlation between epithelial thinning (due to superficial cell loss) and slight ocular irritation. The results of this study suggest that Moxi induces less damage to the corneal epithelium than other antibiotic solutions, perhaps because it does not contain BAC.