OBJECTIVE--To assess the long-term (four to 10 years) prognostic significance of transient ST segment changes on ambulatory ST segment monitoring after coronary artery bypass grafting (CABG). PATIENTS AND METHODS--76 patients (67 men, nine women) underwent CABG between 1982 and 1984 (n = 31) and between 1987 and 1988 (n = 45) and at a mean age of 57. All underwent 48 hours of ambulatory ST segment monitoring at a mean of 19 weeks after surgery. The results were available for assessment. All general practitioners were contacted and patients' notes reviewed. Patients were contacted by telephone. Details were recorded of intervening events (acute myocardial infarction, unstable angina, need for further revascularisation, and deaths). Event free survival curves were produced for those with and without transient ST segment changes during routine postoperative ambulatory ST segment monitoring. RESULTS--During 3213 hours of monitoring after CABG, 21 (27.6%) of 76 patients had transient ST segment changes, of which 70% were silent. Over a mean 70 month follow up period, patients with such ischaemic changes were no more likely to have either an objective (myocardial infarction or cardiac death) or subjective (unstable angina or another revascularisation) event than those patients without ischaemic changes. This finding was the same in patients operated on between 1987 and 1988 and between 1982 and 1984. CONCLUSIONS--Although ambulatory ST segment monitoring is becoming increasingly popular in some countries as a routine investigation for ischaemia in various coronary subgroups, the findings of such an investigation, when performed after CABG, do not help to identify a subgroup more likely to have an adverse outcome during up to 10 years of follow up. There seems to be no reason to perform this investigation after surgery, and particularly to refer patients for reinvestigation because of the detection of predominantly silent ST segment changes of uncertain relevance.