The prevalence and characteristics of silent myocardial ischemia as detected by 24-hour ambulatory electrocardiography ST-segment depression were prospectively assessed in 94 patients examined early (1 to 3 months) and 184 patients examined late (12 months) after coronary artery bypass grafting (CABG), and followed for a mean of 48 +/- 11 (range 4 to 62) months. The relation of ambulatory electrocardiographic silent ischemia to evidence of completeness of revascularization as defined by cardiac angiography performed 1 and 12 months after CABG, and to prognosis by follow-up of adverse clinical events was analyzed. Silent ischemia was detected early in 20% (19 of 94) and late in 27% (50 of 184) of patients, and showed a mean frequency of episodes ranging from 6 to 10 episodes/24 hours with a mean duration ranging from 15 to 23 minutes. The circadian distribution of episodes disclosed a significant peak of ischemic activity during the period of 6 A.M. to noon and a secondary peak between 6 P.M. and midnight (p less than 0.01 and p less than 0.001, respectively). Silent ischemia was not found by univariate analysis to be associated with graft or anastomotic site occlusions, low graft flow rates, grafted arteries with significant distal residual stenoses or ungrafted stenotic native coronary arteries. Kaplan-Meier analysis of time to cardiac event showed that silent ischemia was not predictive of an adverse clinical event in the early years after CABG. Cox regression analysis of 30 covariates only disclosed age (relative risk 1.06 [95% confidence interval, 1.01 to 2.94]) as having an effect on time to adverse clinical event.(ABSTRACT TRUNCATED AT 250 WORDS)