Controlled metabolic studies were used to gauge the relative efficacy of three cardioplegic techniques in 41 patients undergoing multiple coronary artery bypass grafts. Normal-volume (1,946 +/- 155 ml) crystalloid cardioplegia (NVCC) (14 patients) was compared to high-volume (4,961 +/- 282 ml) crystalloid cardioplegia (HVCC) (14 patients) and to blood cardioplegia (BC) (1,672 +/- 127 ml) (13 patients). Measurements of coronary blood flow, coronary vascular resistance, coronary arteriovenous oxygen difference, myocardial oxygen consumption and extraction, and myocardial lactate and potassium extraction and release were all measured in the isolated, vented, paced, beating heart, before and for 20 minutes after a 1 hour arrest interval during which revascularization was completed. Additionally, during administration of the cardioplegic solution, infusion flow rate, myocardial oxygen consumption and extraction, and lactate and potassium release and uptake were noted. The results indicate that during cardioplegic administration, myocardial oxygen consumption is 1 ml O2/min with crystalloid infusion and 2.6 ml O2/min during BC infusion. The volume of crystalloid solution administered contributed to increased oxygen utilization during HVCC compared to NVCC, whereas BC promoted the highest oxygen utilization of the three groups. Potassium absorption was nearly three times greater during BC than during crystalloid administration. During myocardial reperfusion, oxygen extraction was maintained at prearrest levels only in the HVCC group. Following both NVCC and BC, oxygen extraction was depressed during the first 5 minutes of reperfusion, and the difference between the latter two groups and HVCC was significant (p less than 0.01). The rapid recovery in normal metabolic function seen with HVCC allows early discontinuation of cardiopulmonary bypass without myocardial metabolic depression.