To verify the hypothesis, formulated on the basis of data emerging from animal experiments, that branched-chain amino acids (BCAA) exert a protective effect on the heart during ischaemia, eight patients immediately before aortocoronary reconstruction were provided 400 ml of a 3% BCAA solution. A control group comprised another eight patients. Per- and postoperative myocardial status and myocardial enzyme levels were assessed. Myocardial biopsy was performed during surgery to determine glycogen levels. Creatine kinase (CK) levels were invariably higher in the BCAA group, with statistically significant differences in samples obtained immediately after surgery (10.6 +/- 3.35 mu kat/l vs. 4.07 +/- 0.59, p < 0.0004), in the evening after surgery (14.2 +/- 5.92 vs. 5.91 +/- 2.21, p < 0.06) and in the morning of the first postoperative day (18.0 +/- 10.1 vs. 7.5 +/- 4.76, p < 0.025) when aspartate aminotransferase (AST) levels were likewise higher (1.35 +/- 0.28 vs. 1.00 +/- 0.26, p < 0.035). There were no differences between the groups in the number of defibrillations after ischaemia, myocardial glycogen content, peroperative ischaemia, incidence of arrhythmia and catecholamine support. We conclude that BCAA at the above indicated doses did not raise myocardial glycogen content or improve myocardial status after cardiac surgery. Their administration resulted in a rise in CK and AST in the postoperative period.