Isolated stenosis of the aortic valve leads to left ventricular hypertrophy which makes myocardial protection difficult during cardiac, surgery and the choice of optimal cardioplegia remains controversial. The authors compared three protocols of cardioplegia in patients operated for isolated aortic stenosis with left ventricular hypertrophy. Sixty consecutive patients with these criteria were randomly attributed to one of the three following groups (20 in each group): cardioplegia with continuous warm blood; cardioplegia with intermittent cold blood with warm reperfusion; cardioplegia with intermittent cristalloid using SLF11 solution. The preoperative data was comparable in three groups. There were no deaths. Patients undergoing cardioplegia with warm blood came off cardio-pulmonary bypass more quickly (15 mn vs 21 mn for the other groups, p = 0.03). Cristalloid cardioplegia was associated with major acidosis in coronary sinus blood when the aorta was declamped (7.11 vs 7.38 for cardioplegia with cold blood and 7.39 for cardioplegia with warm blood, p < 0.0001) but with a low postoperative CPK-MB rise. Cardioplegia with cold blood induced higher CPK-MB liberation than the other forms of cardioplegia (at H-, 63 mcg/L vs 33 for warm blood and 45 for cristalloid cardioplegia, p = 0.0019). None of the protocols tested prevented myocardial lactate production at aortic declamping. Cardioplegia with warm blood offers therefore the best protection for hypertrophied myocardium during simple aortic valve replacement but it does not maintain strictly aerobic metabolism.