It is hypothesized that because of its potential to increase coronary flow and simultaneously decrease myocardial performance and O2 consumption, midazolam would minimize regional metabolic impairment during myocardial ischemia. Therefore, the hemodynamic and regional metabolic effects of systemic midazolam administration were compared during moderate and severe constrictions of the left anterior descending artery (LADa) to nontreated but ischemic animals in a canine model of acute coronary occlusion. In 16 anesthetized, ventilated, surgically prepared, and catheterized dogs, resting flow in the LADa was decreased by 50% and 75% for 15 minutes with 1 hour of normal flow in between. By arbitrary assignment, 7 dogs received midazolam (0.3 mg/kg and then 0.05 mg/kg/min) before thoracotomy. In all dogs, heart rate, electrocardiogram, LADa flow, left ventricular (LV) first time-derivative, and aortic, pulmonary artery, LADa, and LV pressures were measured continuously. Before and during constrictions, cardiac output by thermodilution and regional myocardial blood flow by microspheres were measured and blood was sampled for analysis. Data (mean +/- SEM) were compared within and between groups using ANOVA. Before placement of the LADa ligature, midazolam decreased heart rate and mean aortic pressure. Before ischemia, heart rate and LADa pressure were lower with midazolam than without it, but baseline metabolic variables were similar between the two groups (except for O2 consumption in the ischemic zone, which was lower with midazolam than without it). During 75% constriction with midazolam, LV end-diastolic pressure, coronary resistance, and ischemic zone O2 consumption were lower than without midazolam. Ischemic zone O2 delivery/consumption ratio was higher.(ABSTRACT TRUNCATED AT 250 WORDS)