The effect of intravenous nisoldipine (0.12 μg/kg/min) on diastolic left ventricular (LV) dysfunction was studied from simultaneous hemodynamic and radionuclide measurements in 12 patients with New York Heart Association class II to IV cardiac failure. The initial LV filling fraction was low, the peak LV filling rate normalized for end-diastolic volume was decreased, and the pulmonary capillary wedge pressure (PCWP) was high and associated with clinical shortness of breath. Nisoldipine produced an increase in LV filling fraction from 37 ± 17% to 43 ± 20% (p = 0.003). The increase in filling took place in both early and late diastole: peak early filling rate (PFR) increased in 11 of the 12 patients (p = 0.02) and late diastolic filling rate (atrial [A] wave in eight of them (NS). When the determinants of these changes, were examined further, it was found that in the control state PFR was inversely related to LV end-systolic volume (r = 0.77), whereas the A wave was related in exponential fashion to PCWP (preload) (r = 0.79). Nisoldipine did not change the slope of these relationships, and it did not alter the end-diastolic pressure-volume relationship, implying that inherent myocardial relaxation and distensibility were unaltered by the drug. In summary, nisoldipine improved measurements of diastolic LV dysfunction in patients with cardiac failure. This study illustrates the importance of considering ventricular loading conditions when analyzing and interpreting measurements of diastolic ventricular dysfunction. The measured changes in diastolic LV function during infusion of nisoldipine appear to be due to alterations in ventricular loading conditions rather than to a direct myocardial effect of the drug.