Between October 1991 and March 1993, 281 consecutive patients underwent non-emergency isolated coronary artery surgery under the care of one surgeon (A.R.). They were prospectively randomised to receive either intermittent cold (Group I-144 patients) or continuous warm (Group II-137 patients) blood cardioplegia for myocardial protection. There were no significant differences in clinical outcome between the two groups, as judged by operative mortality, rates of peri-operative myocardial infarction, blood loss, need for circulatory support, post-operative neurological deficit, or duration of intensive care or hospital stay. However, sinus rhythm returned spontaneously with greater frequency (91.2% vs 45.8%, P < 0.001) in Group II patients. There was greater transmyocardial oxidative stress in Group I patients, as evidenced by a significant rise in oxidised glutathione in coronary sinus blood on myocardial reperfusion. Also, the serum CKMb isoenzyme level 2 h post-operatively was significantly raised in Group I patients, although this difference had disappeared by the day after surgery. In conclusion this preliminary report suggests that continuous warm blood cardioplegia provides comparable myocardial protection to that achieved with standard hypothermic techniques in patients undergoing coronary artery surgery.