The ocular pharmacokinetics of topical diclofenac sodium was studied in two experimental models of ocular inflammation and compared to physiological conditions. Keratitis or uveitis were induced by intrastromal injection of clove oil or by intravitreal lipopolysaccharide in rabbits. The control eyes were not inflamed. Simultaneously to the induction of inflammation, 30 microl of 0.1% diclofenac were applied topically in the right eye. Diclofenac levels were measured by HPLC in the cornea, aqueous humor (AH), iris/ciliary body (ICB) and plasma 30 min, 1, 3, 6 and 12 h after application. In physiological conditions, diclofenac reached a peak level in the cornea and ICB at 30 min slowly decreasing afterwards. Low levels of diclofenac were found in AH. In keratitic eyes, two peak levels which were significantly higher than in the controls were found in the cornea 30 min and 3 h after application. Diclofenac concentrations in keratitic AH and ICB were lower than in controls. In uveitic eyes, corneal and ICB levels peaked at 30 min, being significantly higher than in controls, and decreased quickly to very low levels at 1 h after application. In uveitic AH, diclofenac levels were lower than in controls. Plasma levels were very low (less than 0.1 microg/ml) in all experimental groups. It is concluded that the ocular pharmacokinetics of topical diclofenac is affected by inflammatory processes in the eye, reaching higher levels in the target tissues.