Asymptomatic ("silent") ischaemia has been shown to be of prognostic significance in patients with stable and unstable angina and more recently in patients recovering after myocardial infarction. No therapeutic regimen has yet been shown to improve the prognosis of patients with silent ischaemia after infarction, which can be found in as many as a third of these patients. Attempts to achieve therapeutic revascularisation in all these patients may be undesirable, but early revascularisation could be especially beneficial in some selected high risk patients. Two hundred and fifty consecutive clinically stable survivors of myocardial infarction who had predischarge submaximal exercise tests were followed up for a year. Silent ischaemia was found in 27% of these patients; 15% had symptomatic ischaemia. Patients with a positive exercise test were prescribed a beta blocker before discharge. Mortality in patients with silent (9.4%) and symptomatic (5.4%) ischaemia in the first year after infarction was not significantly different. Patients with symptomatic ischaemia were more likely to have undergone coronary artery bypass grafting in the first year. Patients with silent ischaemia were, however, significantly more likely to die than patients with a negative exercise test (relative odds 12:1). Patients with silent ischaemia and an abnormal blood pressure response or who could not complete a submaximal exercise protocol were at particularly high risk, being 32 times more likely to die than those with a negative test (95% confidence interval from 3.3 to 307 times more likely). First year mortality in this group was 22%. The benefits of therapeutic revascularisation in this high risk group need to be studied.