The efficacy of retrograde coronary sinus cardioplegia (RCSC) administered through the right atrium compared with aortic root cardioplegia (ARC) has not been examined critically in patients undergoing coronary artery bypass grafting (CABG). Twenty patients having elective CABG were randomized prospectively to receive cold blood ARC (Group I, 10 patients) or cold blood RCSC (Group II, 10 patients). Patient demographics were similar in both groups. Ventricular function was assessed preoperatively by radionuclide ventriculography and postoperatively by simultaneous hemodynamic and radionuclide ventriculographic studies with volume loading. There was no change in ejection fraction (EF) (preoperative versus postoperative value) in Group I (50 +/- 6% versus 53 +/- 6%) but in group II, at similar peak systolic pressure and similar left ventricular end-diastolic volume index (LVEDVI), LVEF improved significantly (49 +/- 6% versus 60 +/- 12%, p less than 0.05). Postoperative ventricular function (stroke work index versus EDVI) for the left ventricle and right ventricle were similar in both groups. Evaluation of postoperative LV systolic function (end-systolic blood pressure versus end-systolic volume index) and diastolic function (pulmonary capillary wedge pressure versus EDVI) were also similar in both groups. Retrograde coronary sinus cardioplegia is as effective as ARC for intraoperative myocardial protection, and provides excellent postoperative function in patients undergoing elective CABG.