This study investigated the effect of intravenous cocaine (0.5 to 2 mg/kg body weight) on the coronary circulation and systemic hemodynamics in closed chest sedated dogs. The role of alpha- and beta-adrenoceptor stimulation in mediating these effects was also investigated. Cocaine produced dose-dependent increases in mean arterial pressure and rate-pressure product. Although the lower doses of cocaine had no significant effect on the coronary circulation, the 2 mg/kg dose produced a 55 +/- 14% increase in coronary vascular resistance (p less than 0.05 versus baseline) and a 19 +/- 3% reduction in diameter of the left anterior descending coronary artery (p less than 0.05 versus baseline). Despite these potentially deleterious effects on the coronary circulation (occurring at a time of markedly increased myocardial oxygen demand), the electrocardiogram did not demonstrate ischemic changes and there was no myocardial lactate production. Cocaine-induced coronary vasoconstriction was abolished by pretreatment with the alpha-adrenoceptor antagonist phentolamine, but not by pretreatment with the beta-adrenoceptor antagonist propranolol. The findings that cocaine did not change systemic vascular resistance in dogs without adrenergic blockade, reduced systemic vascular resistance in dogs after alpha-blockade (p less than 0.05) and increased systemic vascular resistance in dogs after beta-blockade (p = 0.06) suggest that epinephrine (rather than norepinephrine) is primarily responsible for the peripheral vascular actions of cocaine. Thus, in this canine preparation with normal coronary arteries, cocaine produced vasoconstriction of both epicardial and coronary resistance vessels that was not associated with evidence of myocardial ischemia. The pharmacologic mechanism for the effect of cocaine on the coronary circulation is alpha-adrenoceptor stimulation, whereas systemic hemodynamic effects are mediated by combined alpha- and beta-adrenoceptor stimulation.