We examined the prevalence and prognostic importance of silent myocardial ischemia detected by continuous electrocardiographic monitoring in 70 patients with unstable angina. All the patients received intensive medical treatment with nitrates, beta-blockers, and calcium-channel blockers. Continuous electrocardiographic recordings were made during the first two days in the coronary care unit to quantify the frequency and duration of asymptomatic ischemic episodes, defined as a transient ST-segment shift of 1 mm or more. Thirty-seven patients (Group 1) had at least one episode of silent ischemia, and the other 33 patients had no silent ischemia (Group 2). Over the subsequent month, myocardial infarction occurred in 6 patients in Group 1 and in only 1 in Group 2 (P less than 0.01); bypass surgery or angioplasty was required for recurrent symptomatic angina in 10 patients in Group 1 and only 3 in Group 2 (P = 0.02). Survival-curve analysis demonstrated that silent ischemia was associated with these outcomes (P less than 0.002), and multivariate analysis showed that silent ischemia was the best predictor of these outcomes among the 15 variables tested (P less than 0.002). Patients in Group 1 with 60 minutes or more of silent ischemia per 24 hours had a worse prognosis than those with under 60 minutes per 24 hours (P = 0.04). Silent ischemia occurred in more than 50 percent of our patients with unstable angina, despite intensive medical therapy, and it identified a subset who were at high risk for early unfavorable outcomes.