One hundred patients scheduled for elective coronary artery bypass grafting (CABG) were randomly allocated to two groups for myocardial preservation:blood cardioplegia (BCP) or crystalloid cardioplegia (CCP). The study protocol comprised recording of the following parameters: mode of resumption of cardiac rhythm, CK-MB analysis, ECG recording, cardiac output measurement, cross-clamping and perfusion times, and clinical outcome. The study period covered the time from commencement of anesthesia to the first postoperative morning. Spontaneous resumption of sinus rhythm was recorded only in the BCP group (22/51 v 0/49, P < 0.001). CK-MB values were similar in both groups, but 1 hour postoperatively the BCP group had lower values (58.8 +/- 26.7 v 74.5 +/- 31.5 U/L, P = 0.0098 by t test). Fifteen patients in the BCP group did not receive any electric countershock; this subgroup had very low CK-MB values. There were four intraoperative myocardial infarctions in the BCP group and two in the CCP group (BCP: 3/51 v CCP: 3/49, P = 0.68). The results suggest better cardioprotection with blood cardioplegia in this subgroup of patients. Spontaneous resumption of normal cardiac rhythm seems to indicate good myocardial preservation, as reflected in markedly lower CK-MB values in this subgroup.