A large atrial contribution to left ventricular (LV) filling (%oA) in patients with LV hypertrophy has been assumed by some to indicate abnormal LV compliance. We tested this assumption by examining the influence of short-and long-term changes in load on compliance and filling dynamics using nitroprusside to decrease load in 11 patients with severe aortic stenosis (AS) and ergonovine to increase load in nine normal subjects. LV angiographic volume was analyzed frame-by-frame simultaneous with micromanometer pressure recordings. Operative LV chamber compliance (dV/VdP) and a time constant for isovolumic relaxation rate were computed using three-constant exponential equations fit to the data. Compared with normal subjects, resting left ventricular end-diastolic pressure was increased and dV/VdP was reduced in AS, but %oA was not different. %oA was inversely related to left ventricular end-diastolic pressure (r= -0.48, p =0.02) and positively correlated with dVNVdP (r=0.90, p<0.001) within the AS group. Nitroprusside infusion reduced LV peak systolic pressure by 11%, end-diastolic pressure by 38%, and end-diastolic volume by 12% (p<0.004 for each) and tended to increase dV/VdP by 26% (p=0.23). These alterations in load resulted in a 21% decrease (-16 ml) in the early filling volume (p <0.05) and variable increases (mean, +7 ml; p=NS) in the late atrial filling volume and in the percent atrial contribution to ventricular filling (26±19% to 35+25% for the AS group, p=NS) that were related to changes in compliance. Changes in filling dynamics with load augmentation by ergonovine in normal subjects were characterized by a 25% increase in early filling (p=0.03) and a 37% decrease in late atrial filling (p=0.01), with a 49% decrease (p=0.04) in operative compliance. In conclusion, load-induced decreases in compliance in normal subjects results in increased early filling and reduced late atrial filling; in aortic stenosis, %oA was inversely related to left ventricular end-diastolic pressure and positively correlated with dVNVdP; and the positive correlation between operative compliance and %oA is opposite to the stated assumption. (Circulation 1990;81:101-106) Although diastolic dysfunction may have important clinical consequences, it is difficult to measure with precision, even under the most rigorous conditions.12 Recently, many investigators have purported to detect diastolic dysfunction by studies of left ventricular filling dynamics using contrast angiography, radionuclide angiography, or Doppler transmitral flow recordings. Having observed the ratio of late (atrial)-to-early filling velocity to be higher than normal in the elderly,3 diabetics,4 hypertensives,5 and patients