Left ventricular systolic and diastolic function were studied before and after surgical revascularization in a group of 24 patients with stable angina who all had an excellent clinical response to surgery. With use of micromanometer left ventricular pressure measurements and ventricular volumes, calculated from biplane cineangiograms, left ventricular function at rest and during exercise before and after surgery was compared. Before surgery all patients had exercise-induced ischemia with new asynergy, a fall in ejection fraction from 57% to 49% (p less than .001), and a rise in left ventricular end-diastolic pressure from 23 to 37 mm Hg (p less than .001). Postoperative exercise resulted in no new asynergy and ejection fraction rose from 59% to 61% (p less than .05). Left ventricular end-diastolic pressure still rose from 17 to 25 mm Hg (p less than .01). Left ventricular pressure decay during exercise was greatly improved after revascularization and allowed maintenance of reduced early diastolic pressures. The early diastolic pressure nadir before surgery rose from 9 to 21 mm Hg (p less than .001); the postoperative nadir was 5 mm Hg at rest and 6 mm Hg during exercise. All patients had an upward shift in the diastolic pressure-volume relationship during preoperative exercise. After revascularization there was no upward shift in some patients and a much smaller shift in others. The postoperative increase in left ventricular end-diastolic pressure was due to increased end-diastolic volume, not altered compliance. There was an increase in mean right atrial pressure during exercise either before (6 to 11 mm Hg) or after surgery (4 to 10 mm Hg). These increases were quite variable, suggesting no consistent role of pericardial restraint during exercise. Early diastolic peak filling rate during exercise was greater after surgery (1260 vs 950 ml/sec, p less than .001). In fact, during postoperative exercise early diastolic filling rates were greater than normal, reflecting the persistence of abnormally high atrial pressures for filling. As at preoperative study, late diastolic filling during exercise was restricted after revascularization when compared with that in a control group. Postoperatively patients undergoing bypass procedures with a good clinical result showed significantly improved left ventricular diastolic and systolic function. Persistent elevation of end-diastolic and atrial pressures and other abnormalities of diastolic function may reflect chronic structural changes and need to be taken into account when evaluating patients after bypass surgery.