This study was designed to examine the effect of benzalkonium chloride/ethylenediaminetetraacetic acid (BAK/EDTA) on the ocular bioavailability (F-ocular) of ketorolac tromethamine after ocular instillation to normal and de-epithelialized corneas of rabbits both in vitro and in vivo. The in vitro F-ocular of the formulations was measured in flow-through perfusion chambers. For in vivo studies, a 35 mu L dose of 0.5% ketorolac tromethamine with or without BAK/EDTA was instilled into rabbit eyes with intact or de-epithelialized corneas. At 0.5, 1, 2, 4, 6, and 8 h postdose, rabbits were euthanized, and the corneas and aqueous humor were collected from both eyes. The ketorolac concentrations from both in vivo and in vitro samples were quantified by reversed-phase high-performance liquid chromatography. The in vitro study results indicated that BAK/EDTA statistically significantly increased the F-ocular of ketorolac through de-epithelialized corneas but not through intact corneas. The in vivo study results showed that BAK/EDTA had no effect on the F-ocular of ketorolac in rabbits with intact corneas, based on the values of the area under the aqueous humor concentration versus time curves (AUC(0-6h)) of ketorolac. AS expected, de-epithelialization of the corneas produced a faster and greater ocular absorption of ketorolac as evidenced by the smaller T-max and larger AUC values compared to those for the intact corneas in vivo. However, BAK/EDTA decreased the ocular absorption of ketorolac in rabbits with de-epithelialized corneas. The half-lives (t(1/2)) of ketorolac in corneal tissue and aqueous humor were longer in rabbits with intact corneas than those in rabbits with de-epithelialized corneas. In conclusion, the in vivo F-ocular of ketorolac was not altered by BAK/EDTA in rabbits with intact corneas, but it was decreased by BAK/EDTA in rabbits with de-epithelialized corneas. Therefore, the formulation with ketorolac alone may be better as a postoperative ocular analgesic.