doi: 10.1111/j.1472-8206.2004.00285.x ORIGINAL ARTICLE Interactions between aspirin and COX-2 inhibitors or NSAIDs in a rat thrombosis model Anwar Umar a,b , Michel Boisseau c , Abdiryim Yusup d , Halmurat Upur b,d , Bernard Be ´gaud a , Nicholas Moore a * a Department of Pharmacology, Universite ´ Victor Segalen, 33076 Bordeaux, France b Department of Pharmacology, Xinjiang medical University, Urumqi, Xinjiang 830054, China c Department of Hematology, Universite ´ Victor Segalen, 33076 Bordeaux, France d Institute of traditional Uighur Medicine, Urumqi, Xinjiang 830001, China INTRODUCTION The possible prothrombotic effect of selective inhibitors of cyclo-oxygenase 2 (COX-2) and the possible interaction between aspirin used at cardioprotective doses and COX-2 selective or non-selective NSAIDs have been the object of much recent debate: during the VIGOR study comparing rofecoxib and naproxen [1] in patients with rheumatoid arthritis, there was an apparent doubling of the risk of myocardial infarction in the rofecoxib-treated patients, compared with the patients in the naproxen arm. This was not found in the CLASS trial comparing celecoxib and ibuprofen or diclofenac [2] in patients with osteoarthritis or rheumatoid arthritis. Concerns were raised about the cardiac safety of COX-2 inhibitors [3]. The reassessment of the cardiovascular safety of these drugs showed that there did not seem to be an excess risk of myocardial infarction with the available selective COX-2 inhibitors [4,5], and that the effect found in the VIGOR study was probably related to a protective effect of naproxen. In the course of this investigation, Catella- Lawson et al. [6] showed an inhibition by a single dose of ibuprofen, but not by rofecoxib, acetaminophen or diclofenac of the effects of aspirin on serum thromboxane B2 and ex vivo platelet aggregation induced by arachi- donic acid in man [6]. The inhibition by ibuprofen of aspirin’s irreversible effect on platelets has been known for over 20 years, but apparently forgotten [7]. Some population-based studies have found an increased overall and cardiovascular mortality in concomitant users of Keywords aspirin, cyclo-oxygenase inhibitors, experimental thrombosis, inhibition, interaction Received 14 January 2004; revised 9 June 2004; accepted 10 June 2004 2 *Correspondence and reprints: nicholas.moore@pharmaco. u-bordeaux2.fr ABSTRACT Recent in vitro studies, clinical trials and epidemiological studies have suggested possible interactions between aspirin and other cyclo-oxygenase (COX) inhibitors, such as ibuprofen of the COX-2 inhibitors celecoxib and rofecoxib. The objective of this study was to test the effects of aspirin (1, 2.5 and 5 mg/kg), and ibuprofen (4 and 15 mg/kg), diclofenac (2.5 mg/kg), flurbiprofen (2 mg/kg), celecoxib (7.5 mg/kg), and rofecoxib (1 mg/kg), alone or combined on a rat model of arterial thrombosis. Drugs were given orally daily for 7 days, before insertion of an arterio-venous shunt thrombosis system, left in place for 15 min. Main parameter was thrombus weight. Five to 12 rats were used per experiment, and 35 controls overall. Aspirin inhibited thrombus formation in a dose-dependent manner. All NSAIDS given alone also inhibited thrombus formation to approximately the same level as aspirin 1 mg/kg/day. Ibuprofen, celecoxib and rofecoxib inhibited the effects of aspirin, but not diclofenac or flurbiprofen. The interactions with aspirin do not seem to affect all NSAIDs to equal levels. The clinical impact of this needs to be confirmed in adequately powered clinical trials or pharmaco-epidemiological studies. Ó 2004 Blackwell Publishing Fundamental & Clinical Pharmacology 18 (2004) 559–563 559