Two consecutive groups of high risk patients with left main coronary disease (greater than 50% stenosis) undergoing aortocoronary bypass grafting using different cardioplegia delivery methods were compared retrospectively for perioperative myocardial infarction (MI) and mortality. Group I (July 1979 to June 30, 1982, n = 53) received an initial 1000 ml aortic root cardioplegia infusion from a pressure regulated (85-100 mmHg) delivery system. Regional mid-myocardial wall temperatures determined the distal anastomoses sequence (with the warmest region bypassed first) followed by additional 400 ml cardioplegia infusions. Group II patients (March 1976 to June 30, 1979, n = 47) had their cardioplegia administered by a hand-held syringe without regional temperature mapping. The volume injected varied and was based on cessation of electrical activity or a septal temperature less than 20 degrees C. Background data were similar except that Group I was significantly older than Group II (63.4 +/- 1.4 years vs 57.1 +/- 1.5 years, p less than 0.01). Despite this, there was only 1 (2%) perioperative MI in Group I vs 6 (13%) in Group II (p less than 0.05). There was also a marked reduction in cardiac mortality with 1 (2%) cardiac death in Group I vs 5 (11%) in Group II (p = 0.09). These data suggest that the delivery of adequate volumes of cardioplegia, in conjunction with myocardial temperature directed bypass grafting, can improve myocardial preservation in high risk patients.