Warm cardioplegia produced by essentially continuous infusion has been used as an alternative to traditional cold intermittent infusion techniques during cardiac surgery, but its effects on postoperative left ventricular function have not been defined. We performed a randomized clinical trial to assess the effects of warm and cold blood cardioplegia on load-independent indices of ventricular function. Fifty-three patients were randomized to warm (n = 27) or cold (n = 26) cardioplegia. Myocardial oxygen consumption, lactate production, adenine nucleotides, and adenine nucleotide degradation products were measured during cardioplegia and reperfusion. In 13 patients per group, pressure-volume loops were constructed and ventricular function was assessed 3 hours after the operation. Warm cardioplegia resulted in greater myocardial lactate production but improved recovery of oxygen consumption during reperfusion. Depletion of adenosine triphosphate was similar between groups, but total adenine nucleotides (adenosine triphosphate + adenosine diphosphate + adenosine monophosphate) fell further during warm cardioplegia. Cold cardioplegia was associated with an accumulation of adenosine diphosphate and adenosine monophosphate. Creatine kinase MB isoenzyme release was reduced in the warm group. Three hours after the operation, end-systolic elastance and preload-recruitable stroke work index were increased after warm cardioplegia, and early diastolic relaxation was also increased. Increased systolic function after warm cardioplegia may have been related to improved myocardial protection, elevated arterial lactate concentrations, or increased circulating catecholamine levels. Altered diastolic compliance in the warm group may reflect greater active relaxation during early diastolic filling.