427 randomized patients undergoing coronary artery bypass graft (CABG) were divided into two groups. In Group A (269 patients), myocardial protection was achieved using cardioplegic solution (Bretschneider). Group B (158 patients) was operated with intermittent aortic cross-clamping. Between both groups there was no significant difference in early postoperative mortality rates, incidence of postoperative low output syndrome, perioperative myocardial infarctions, length of ICU-stay, length of intubation, incidence of arrhythmia, and inotropic requirements. The ratio of reperfusion time divided by aortic cross-clamping time (R/C-ratio) in Group B (1.06 + 1.17) was larger than in Group A (0.49 + 0.17), which could explain this result. In Group A significantly larger incidence of postoperative arrhythmia was recognized, if reperfusion time was shorter than 20 minutes. Also in Group A, longer postoperative intubation was required with R/C-ratio under 0.60. In Group B longer intubation was needed postoperatively with aortic cross-clamping time longer than 30 minutes, reperfusion time shorter than 30 minutes, or R/C-ratio under 0.80. In conclusion, reperfusion should be continued at least for 20 minutes and if possible, for the length equivalent to R/C-ratio over 0.60 in CABG with presently widely used cardioplegic arrest.