Comparison of corneal surface effects of gatifloxacin and moxifloxacin using intensive and prolonged dosing protocols.

PURPOSE: To compare the effects of the commercial formulations of moxifloxacin and gatifloxacin on rabbit corneal epithelium using two dosing protocols: high-frequency dosing for bacterial keratitis and cataract surgery prophylaxis. METHODS: Forty eyes of 20 New Zealand white rabbits were randomized to receive topical gatifloxacin, topical moxifloxacin, or no drops as controls. Eighteen eyes received 1 drop of antibiotic every 5 minutes for 15 minutes followed by 1 drop every 15 minutes for 4 hours. Twelve eyes received topical gatifloxacin or topical moxifloxacin 4 times a day for 10 days. Rabbits were euthanized, eyes enucleated, and the corneas separated, washed, and fixed; scanning electron microscopy was performed. Photomicrographs of three separate areas from the corneal apex were taken at x1200 and x3000 for each cornea. Two masked, experienced examiners then graded the corneal epithelial damage. RESULTS: In the high-frequency dosing group, mean corneal damage score for eyes treated with gatifloxacin was 1.593, that for moxifloxacin was 1.407, and control was 1.000. No statistically significant difference was found between gatifloxacin and moxifloxacin (P = 0.41), gatifloxacin and control (P = 0.14), or moxifloxacin and control (P = 0.23). In the cataract surgery prophylaxis group, mean corneal damage scores for eyes treated with gatifloxacin was 1.167, that for moxifloxacin was 1.368, and control was 1.000. No statistically significant difference was found between gatifloxacin and moxifloxacin (P = 0.23), gatifloxacin and control (P = 0.08), or moxifloxacin and control (P = 0.23). CONCLUSION: Despite differences in formulations, with short-term dosing neither gatifloxacin nor moxifloxacin appears to be toxic to the corneal epithelium in this rabbit model.

Comparative penetration of moxifloxacin and gatifloxacin in rabbit aqueous humor after topical dosing.

PURPOSE: To evaluate the aqueous penetration of the fourth-generation fluoroquinolones moxifloxacin and gatifloxacin. SETTING: University of Arizona, Tucson, Arizona, USA. METHODS: Forty eyes of 20 New Zealand white rabbits were divided into 2 experimental groups. In Experiment I rabbits (20 eyes), a commercial preparation of topical gatifloxacin 0.3% was administered to 9 eyes and moxifloxacin 0.5% to 9 eyes; 2 eyes served as a control. Eyes were dosed according to a keratitis protocol; ie, every 15 minutes for 4 hours. The aqueous humor was sampled 10 minutes after the last dose. Experiment II rabbits (20 eyes) were dosed according to a cataract prophylaxis protocol; ie, 4 times a day for 10 days. The aqueous humor was sampled 1 hour after the last dose of antibiotic in 12 eyes and 24 hours after the last dose in 8 eyes. High-performance liquid chromatography was used to determine the fluoroquinolone concentration. RESULTS: In the keratitis dosing protocol, the mean concentration of moxifloxacin in the aqueous (n=9) was 11.057 microg/mL (range 7.66 to 18.87 microg/mL), which was significantly higher than the mean concentration of gatifloxacin (n=8) (7.570 microg/mL [range 4.75 to 10.86 microg/mL]) (P=.030). In the cataract prophylaxis dosing protocol, the mean aqueous concentration of moxifloxacin (n=6) was 1.745 microg/mL (range 0.92 to 3.87 mg/mL). The mean concentration of gatifloxacin (n=6) was 1.207 microg/mL (range 0.44 to 2.44 microg/mL). The difference was not statistically significant (P=.359). CONCLUSIONS: Higher mean levels (x1.46) of aqueous penetration were achieved with moxifloxacin than with gatifloxacin in the keratitis-dosing model. There was no statistically significant difference between the 2 drugs in the cataract prophylaxis dosing model. Both antibiotics had aqueous levels in excess of the minimum inhibitory concentration for most pathogenic organisms in both models.

Corneal and conjunctival changes caused by commonly used glaucoma medications.

PURPOSE: To evaluate the extent of epithelial corneal and conjunctival changes associated with prolonged use of topical glaucoma medications. METHODS: Thirty eyes of 15 New Zealand white rabbits were randomized to 1 of 6 treatment groups: artificial tears (Refresh Tears, carboxymethyl cellulose 0.5%) BID, brimonidine Purite 0.15% BID, bimatoprost 0.03% QD, dorzolamide 2% BID, timolol maleate 0.5% BID, or latanoprost 0.005% QD for 30 days. Corneal damage was evaluated by scanning electron microscopy and graded on a standard scale by a masked observer. Conjunctival inflammation was evaluated with light microscopy, and inflammatory cells were counted in the epithelium and superficial and deep stroma by a masked observer according to a standard protocol. RESULTS: In the cornea, artificial tears produced significantly less damage than dorzolamide or latanoprost (P = 0.001), and brimonidine Purite produced significantly less damage than dorzolamide, timolol, or latanoprost (P = 0.001). The mean damage scores with bimatoprost were significantly lower than with dorzolamide, timolol, or latanoprost (P = 0.002). In the conjunctiva, the number of inflammatory cells in the epithelium was significantly lower in eyes treated with artificial tears or brimonidine Purite than in eyes treated with timolol or latanoprost (P = 0.042). CONCLUSIONS: Although the adverse effects of glaucoma medications on the ocular surface are likely multifactorial, 1-month treatment with glaucoma medications containing higher levels of benzalkonium chloride (BAK) resulted in greater corneal damage and conjunctival cell infiltration than medications preserved with Purite or with lower levels of BAK. Using glaucoma medications with alternative preservatives or low levels of BAK may help preserve ocular health.

Diode laser contact transscleral cyclophotocoagulation: getting the most from the G-probe.

BACKGROUND AND OBJECTIVE: To determine the multi-use behavior of the G-probe that is traditionally marketed as a single-use cyclophotocoagulation instrument. MATERIALS AND METHODS: A diode laser equipped with a G-probe was used to perform cyclophotocoagulation of the ciliary body in 4 human cadaver eyes and 15 porcine eyes (1,750 mW x 2 seconds). A determination of G-probe effectiveness was made following measurements of G-probe energy output, scanning electron microscopy examination of probe tips, and histologic examination of treated tissue. RESULTS: The mean energy output of 12 of the 15 G-probes remained constant at 3.10 J (range, 3.04 to 3.14 J) for 20 treatment cycles of 20 applications each. Scanning electron microscopy demonstrated minimal surface change. Three G-probes suffered a significant drop-off in energy output during the experiment, and scanning electron microscopy showed significant surface change. Histologic examination of human eyes treated with laser revealed disruption of the ciliary body stroma and separation of pigmented and nonpigmented layers of the ciliary epithelium. More pronounced tissue necrosis and disruption was observed in eyes treated with newer G-probes. Milder coagulative damage with cellular vacuolization was observed in G-probes used for higher numbers of treatment cycles. CONCLUSION: Every G-probe tested delivered more than 100 consecutive applications of laser energy before any loss of efficiency was noted. This suggests that in the absence of obvious structural damage, a G-probe may be safely used for treatment of at least 5 eyes.