Recombinant hirudins are currently under investigation for use in myocardial infarction and unstable angina. In this study the influence of piroxicam on the pharmacodynamics and pharmacokinetics of a recombinant hirudin preparation (CGP 39393) administered intravenously was determined. Twelve healthy, male volunteers received piroxicam 10 mg and matching placebo once daily for 12 days according to a double-blind, randomised cross-over design. On the 12th day, the dose of piroxicam was followed by a 6-hour infusion of hirudin 0.1 mg·kg^–1·h^–1. Plasma concentrations and urinary excretion of hirudin and repeated measurements of the activated partial thromboplastin time (APTT), bleeding time and platelet adhesion index were assessed up to 24 h after the start of the infusion. The maximum APTT was 83 s (placebo) and 84 s (piroxicam), 3 to 4 h after the start of the infusion, and was comparable on both study days. The AUD_0–24 (APTT) came to 913 s·h·kg^–1 under placebo and it was slightly increased to 1,017 s·h·kg^–1 after piroxicam; the 95%-confidence interval according to MOSES ranged from 0.97 to 1.24, and the point estimator was 1.10. Bleeding time was significantly prolonged from 290 s under placebo to 345 s under piroxicam before the start of the infusion of hirudin. No further prolongation was found during or after the infusion. No change was observed in the platelet adhesion index. Responsiveness parameters according to a sigmoidal E_max-model were obtained from the hirudin-plasma concentration/effect (i.e. APTT-prolongation)-curves after placebo and piroxicam. Maximal APTT-prolongation (E_max; i.e. peak APTT minus the baseline value) was 53 s after placebo and 52 s after piroxicam. The EC₅₀ was 34 nmol·l^–1 after placebo and 40 nmol·h·l^–1 after piroxicam. The AUC₀ of hirudin was to 539 nmol·h·l^–1·kg^–1 under placebo and 557 nmol·h·l^–1·kg^–1 after piroxicam coadministration; the 95%- confidence interval according to MOSES ranged from 0.95 to 1.14, and the point estimator was 1.03. No period effect was detected. There were no significant differences between the other pharmacokinetic parameters except V_ss, which was increased slightly from 0.23 l to 0.27 l under piroxicam. The results do not show a clinically relevant pharmacodynamic and/or pharmacokinetic interaction between hirudin and piroxicam.