OBJECTIVE--To examine the cardioprotective efficacy of allopurinol in patients undergoing elective coronary artery surgery. DESIGN--Prospective randomised trial. SETTING--London teaching hospital. PATIENTS--Twenty patients with at least moderately good left ventricular function undergoing elective coronary artery surgery and requiring at least two bypass grafts. INTERVENTIONS--Patients were randomised to receive allopurinol (1200 mg in two divided doses) or to act as controls. MAIN OUTCOME MEASURE--The primary determinant of the efficacy of myocardial protection was serial measurement (preoperatively and subsequently at one, six, 24, and 72 hours after the end of cardiopulmonary bypass) of cardiac troponin T (cTnT) a highly sensitive and specific marker of myocardial damage. Additional evidence was provided by serial measurement of the MB-isoenzyme of creatine kinase (CK-MB) and myoglobin, ECG changes, and clinical outcome. RESULTS--There was no significant difference in age, ejection fraction, number of grafts, bypass times, or cross clamp times between the two groups. In both groups there was a highly significant (p < 0.01) rise in cTnT, CK-MB, and myoglobin. Peak concentrations were reached between one (CK-MB and myoglobin) and six hours (cTnT) after the end of cardiopulmonary bypass. At 72 hours cTnT concentrations were six times higher than baseline concentrations whereas CK-MB and myoglobin were approximately double baseline concentrations. There was no significant difference in cTnT, CK-MB, or myoglobin between the allopurinol and control groups at any time. There was no diagnostic ECG evidence of perioperative infarction in any patient. CONCLUSION--Unlike previous reports this study did not show that allopurinol had a cardioprotective effect in patients with good left ventricular function undergoing elective coronary artery surgery.