Although ophthalmic timolol is generally well tolerated, a significant fraction of topically administered timolol can be systemically absorbed. We investigated the effect of the strong CYP2D6 inhibitor paroxetine on the pharmacokinetics of timolol after ophthalmic administration. In a four-phase crossover study, 12 healthy volunteers ingested either paroxetine (20 mg) or placebo daily for 3 days. In phases 1-2, timolol 0.1% gel, and in phases 3-4, timolol 0.5% drops were administered to both eyes. Paroxetine increased the plasma concentrations of timolol with both timolol formulations to a similar degree. The geometric mean ratio (95% confidence interval) of timolol peak concentration was 1.53-fold (1.23-1.91) with 0.1% timolol and 1.49-fold (0.94-2.36) with 0.5% timolol, and that of timolol area under the plasma concentration-time curve (AUC) from time 0 to 12 hours was 1.61-fold (1.26-to 2.06-fold) and 1.78-fold (1.21-2.62), respectively. During paroxetine administration, six subjects on 0.5% timolol drops, but none on 0.1% timolol gel, had plasma timolol concentrations exceeding 0.7 ng/ml, which can cause systemic adverse effects in patients at risk. There was a positive correlation between the AUC from time 0 to 13 hours of paroxetine and the placebo phase AUC from time 0 to infinity of timolol after timolol 0.5% drops (P < 0.05), and a nonsignificant trend after timolol 0.1% gel, consistent with the role of CYP2D6 in the metabolism of both agents. In the orthostatic test, heart rate immediately after upright standing was significantly lower (P < 0.05) during the paroxetine phase than during the placebo phase at 1 and 3 hours after 0.5% timolol dosing. In conclusion, paroxetine and other CYP2D6 inhibitors can have a clinically important interaction with ophthalmic timolol, particularly when patients are using 0.5% timolol formulations.