Critical changes in left and right ventricular function immediately after myocardial revascularization may affect the success of the procedure, morbidity, and mortality. To delineate these changes and identify vulnerable patient populations and times of highest risk, ventricular function was studied for 24 h in 22 patients undergoing myocardial revascularization. Preoperative ejection fractions ranged from 0.26 to 0.81. For each patient, eight left and eight right ventricular function curves (LVFC and RVFC) were generated by altering preload during the 24-h perioperative period. Central venous pressure ranged from 0 to 19 mmHg and pulmonary capillary wedge pressure from 0 to 31 mmHg. In all patients, significant (P less than 0.05) left and right ventricular dysfunction occurred at 15 min following bypass, LVFCs and RVFCs being depressed 35-75% of control. The degree of depression and the pattern of recovery could be predicted best (stepwise logistic regression) by two preoperative indices: the ejection fraction and degree of dyssynergy. Patients with ejection fractions greater than 0.55 and no significant dyssynergy (n = 11) had postbypass LVFCs and RVFCs that were 75% and 60% of control, respectively. However, these depressions were transient, and recovery to 90% of control occurred within 4 h of revascularization. In contrast, patients having preoperative ejection fractions less than 0.45 or dyssynergy (n = 11) had more severely depressed ventricular function (LVFC = 40% and RVFC = 30% of control) that persisted for 24 h after revascularization, resulting in only 60% recovery of ventricular function. In conclusion, the preoperative indices--ejection fraction and degree of dyssynergy--best identified patients most likely to have significant and prolonged biventricular dysfunction after revascularization.