Objectives. We evaluated the prevalence and prognostic significance of transient myocardial ischemia despite beta-adrenergic blockade in patients with coronary artery disease. Background. Persistence of transient ischemia despite therapy may correspond to a subset of high risk patients with coronary disease. The impact of beta-blocker withdrawal in these patients remains unknown. Methods. Patients (n = 313) with documented coronary artery disease and beta-blocker therapy, with (group I, n = 84) or without (group II, n = 229) transient ischemia on ambulatory electrocardiographic monitoring, were followed up during 21 +/- 9 months for cardiac events (death, myocardial infarction, percutaneous transluminal coronary angioplasty, coronary artery bypass surgery and worsening angina). Occurrence of events was compared by log-rank test. Results. The number of coronary stenoses did not differ significantly between groups I and II. Beta-blocker therapy was discontinued more frequently during follow-up in group II (25% vs. 14% in group I, p = 0.04). Cumulative percentage of death or myocardial infarction, or both, tended to be higher in group I at 30 months (17% vs. 5% in group II, p = 0.09). Coronary angioplasty and bypass surgery were significantly more frequent in group I (p = 0.01 and 0.0008, respectively). Transient ischemia was associated with a higher cumulative probability of adverse events (p = 0.004). The number of coronary stenoses, presence of transient ischemia and beta blocker withdrawal sere the only significant prognostic factors of cardiac events in the Cox model. In group I patients, the relative hazard of cardiac events was increased threefold when beta blocker therapy was interrupted. Conclusions. These data suggest that 1) the occurrence of transient ischemia despite beta-blocker therapy identifies a subset of high risk patients with coronary artery disease, and 2) the interruption of beta-blocker therapy increases the risk of adverse cardiac events.