Recent reports (1-4) of adverse cardiac events following the administration of sumatriptan have raised concern among doctors and patients. The following case report underlines the difficulties in establishing a causal association between drugs and acute medical events. A 44-year-old Dutch woman suddenly developed chest pain and collapsed while on holiday in Belgium. She was found in ventricular fibrillation when the ambulance staff arrived and received several DC shocks. She had an inferior postero-lateral myocardial infarction, was treated with tissue plasminogen activator and admitted to the University Hospital; she never regained consciousness and died 10 h after the initial event. On autopsy, an acute infarction of most of the postero-lateral cardiac wall and a 70% single atheromatous stenosis of the initial second segment of the right coronary artery were found. Since it appeared from a rudimentary history taken on admission that this patient was a known migraineur on sumatriptan, her record was tentatively classified as acute myocardial infarction secondary to sumatriptan-induced vasospasm of a stenotic coronary artery.In accordance with the available data this diagnosis appeared plausible. Sumatriptan has possibly promoted coronary vasospasm in at least two women in the absence of any other underlying cardiovascular disease and of any concomitant drug use (3-first patient, 4). It has been shown during diagnostic coronary arteriography that 6 mg subcutaneous sumatriptan can reduce the mean coronary artery diameter by an average of 17% (5). However, when this patient's husband and general practitioner were questioned in more detail about her medical history and recent drug utilization, it appeared most unlikely that sumatriptan was responsible for the cardiac event. Although she had been taking sumatriptanregularly for 18 months, with total satisfaction and without any cumbersome adverse event, she had not used this drug, nor any other antimigraine medication since her arrival in Belgium one week prior to her death. With the exception of contraceptive pills (gestoden-ethinylestradiol), no drug was found in her personal belongings and sumatriptan was not on the Belgian market at the time. Since plasma half-life of sumatriptan (2 h) and its pharmacodynamic effects on both cerebral (6) and coronary (5) arteries are short-lived (less than 1 h), we felt at last that sumatriptan was not guilty and that this patient's acute myocardial infarction had been brought on by the association of heavy smoking and hormonal contraception.