Relative toxicity of four ophthalmic antibiotics studied

Fort Lauderdale, FL-In a comparison study that has implications for wound healing in surface ablation procedures, investigators found that four commonly used topical ophthalmic antibiotics-moxifloxacin (Vigamox, Alcon), levofloxacin (Quixin, Santen), gatifloxacin (Zymar, Allergan), and ofloxacin (Ocuflox, Allergan)-all showed evidence of toxicity on human corneal epithelial cells, with moxifloxacin causing the least amount of toxicity.

The comparison of relative toxicity was conducted in a tissue culture model, and differences in corneal epithelial healing with the four antibiotics were also studied in chicken eyes following PRK. Enrollment has begun for a comparison of the relative toxicity of the four compounds in patients, said Richard W. Yee, MD, director of corneal and external diseases and refractive surgery, Hermann Eye Center, Houston, who was senior author of the study.

"The study is completed but the saga goes on," said Dr. Yee, who is also clinical professor, Department of Ophthalmology and Visual Science, University of Texas Health Science Center, Houston. He presented the results of the tissue culture study at the annual meeting of the Association for Research in Vision and Ophthalmology.

"When we choose antibiotics in refractive surgery and specifically in surface ablation procedures such as LASEK, epi-LASIK, or PRK, it is critical that everything we use on the ocular surface is ideal to promote wound healing and will not delay the healing, and that it is not toxic to the cornea," Dr. Yee said. "Toxicity is one component that could induce keratocytes in the stroma to become activated and trigger scar formation."

The findings could apply to the treatment of abrasions, corneal ulcers, keratitis, and conjunctivitis as well as to healing following surface ablation procedures, Dr. Yee added.

He and his colleagues tested levofloxacin, gatifloxacin, moxifloxacin, and ofloxacin with a relative toxicity model, which Dr. Yee emphasized is to be used within itself.

"It is not to be used with human data or data from animal models. It is an internal system that shows how these antibodies compare in terms of toxicity. You can't compare one system with another," he explained.

To perform the tissue culture screening, investigators used 160 tissue wells containing 10.014 pRSV-T human corneal epithelial cells (2,000 to 3,000 cells per well). The wells were divided into 10 groups, and each was exposed to 100 µl of undiluted antibiotics or polyethylene glycol and propylene glycol (Systane eye drops, Alcon). They were evaluated after incubation for either 5 or 15 minutes. Live/dead reagents were used to quantify the effects of each drug on culture cells compared with cells grown in normal media. Relative levels of live/dead cells were then evaluated using a Biotek FL600 Microplace Fluorescence Reader.

Results of the testing showed that the least amount of epithelial cell death occurred in the untreated (control) tissue cultures and in tissue cultures treated with Systane eye drops. Among the antibiotics, moxifloxacin showed the least amount of toxicity. Gatifloxacin, levofloxacin, and ofloxacin were similar in epithelial cell toxicity and showed a greater percentage of cell death than moxifloxacin.

At 5 minutes, the mean percentage of cell death was: tissue culture media, 70.9 ± 6.3; Systane, 70.2 ± 4.0; moxifloxacin, 83.6 ± 6.4; levofloxacin, 89.5 ± 7.6; gatifloxacin, 88.2 ± 7.8; and ofloxacin, 90.9 ± 6.7.

At 15 minutes, the percentages were: tissue culture media, 70.9 ± 6.3; Systane, 70.7 ± 6.9; moxifloxacin, 82.3 ± 6.1; levofloxacin, 88.9 ± 5.5; gatifloxacin, 91.2 ± 6.6; and ofloxacin, 89.6 ± 6.3. According to Dr. Yee, the results were statistically significant.