Financial and other constraints, such as operative risk, may prevent older patients being considered for coronary arterial bypass grafting. Grafting was performed in 315 elderly patients (244 males, 71 females, age 65-79, mean 69 years) between 1981-1986. All patients had limiting angina, 38% had rest pain, 90% were housebound and 80% had triple-vessel disease. Impairment of left ventricular function was absent in 46%, mild in 20%, moderate in 23% and severe in 10% of patients. Grafts (saphenous vein or internal mammary artery) were inserted into 3 vessels (52%), 4 vessels (42%), 5 vessels (6%), 6 vessels (0.5%). Death during surgery occurred in 1.6% and a further 3.5% of patients died later during the same admission (70% of deaths were among the 33% with preoperative moderate or severe left ventricular impairment). Surgical complications included myocardial infarction (8%), cerebrovascular accident (1%), transient cerebral vascular ischaemia (5%), chest infection (10%) and wound infection (4%). Median stay on the intensive care unit was 1 day and median total hospital stay 12 days. 299 patients therefore survived to leave hospital and follow-up data are available for 217 (72%) of these. 96% were subjectively improved by surgery, 88% being free of angina on no antianginal drugs a median of 72 weeks (range 8-307) and a further 8% not limited by angina on medical therapy a median of 85 weeks (range 9-302) after surgery. We conclude that coronary arterial surgery is an effective treatment for angina in the elderly. This will have consequences for future resource allocation if the elderly are not to be denied effective therapy because of financial rather than clinical restraints.