The national cooperative randomized trial for the medical or surgical treatment of unstable angina pectoris is reviewed and its findings examined as a basis for recommendations for management. There was no difference in mortality between medical and surgical groups at a mean follow-up period of 2 1/2 years. Approximately one third of patients initially treated medically received surgery later, and this prevented definitive conclusions about the relative effects of medical or surgical therapy on long-term survival. However, the study does indicate that the clinical condition of most patients with the syndrome of unstable angina pectoris can be stabilized and acutely managed by a medical program consisting of long-acting nitrates and propranolol to control symptoms or to achieve therapeutic control of blood pressure and pulse rate. Management may then proceed to performing coronary arteriography. Early surgery can be recommended when there are symptoms of continuing severe angina or when there is left main coronary artery disease, which is present in approximately 10% of patients with unstable angina. Otherwise surgery may be performed electively and with no greater risk than in those who have urgent surgery. It may be possible to control symptoms medically in a subgroup of patients, but those with three- and also two-vessel disease and those with continuing angina pectoris will probably require surgery to control symptoms.