BACKGROUND: The University of Wisconsin solution (UW) has been shown to be superior to the use of traditional types of simple crystalloids in extended heart preservation. This study was designed to determine the necessity of four potentially redundant components in UW. MATERIALS AND METHODS: Forty canine hearts (n = 8 in each group) were preserved for 6 hours under hypothermic (1-2 degrees C), nonperfused conditions in unmodified UW, UW with an extracellular or reversed composition of sodium and potassium (UW-E), UW with chloride instead of lactobionate (UW-L), UW without raffinose (UW-R), and UW without hydroxyethyl starch (UW-HS). The left ventricular functions were assessed in an isolated heart preparation equipped with a computerized servo-volume pump to measure the pressure-volume relation. The systolic function was expressed in elastance derived from the end-systolic pressure-volume relation, and total ventricular performance was expressed in preload recruitable stroke work area derived from the stroke work-end-diastolic volume relation. Diastolic compliance was derived from the end-diastolic pressure-volume pressure relation. The results were compared with a control group with no ischemia. There was no significant deterioration in left ventricular systolic contractile functions as demonstrated by the analysis of elastance and the preload recruitable stroke work area in any of the modifications of UW. However, significant deterioration in the diastolic compliance occurred in the UW-E and UW-HS groups. A similar trend was found in the UW-L group, although it was not statistically significant, whereas the diastolic compliance was preserved in the UW-R group. CONCLUSIONS: The intracellular electrolyte composition and the hydroxyethyl starch are important in the preservation of left ventricular diastolic function, whereas the omission of raffinose was inconsequential. The replacement of lactobionate by chloride had less untoward functional effects compared with studies of other solid organ preservation.