To test thehypothesis that impaired cardiac performance in some patients with pressure-overload hypertrophy is due to inappropriately high wall stress, rather thandepressed contractility, the impor-tance ofhemodynamic and geometric factors was assessed in 14 patients with isolated aortic stenosis and vari-ous degrees of left ventricular failure (ejection fraction range 0.19-0.85). There was poor correlation between either aortic valve area, peak left ventricular systolic pressure, or left ventricular mass, and measures of ven-tricular function. In contrast, there were close correlations between circumferential wall stress and both ejec-tion fraction (r= 0.96) and velocity of fiber shortening (r= 0.91) in patients with aortic stenosis. Force-velocity-shortening relationships in six normal control subjects fell on the same regression line as that defined by the patients with aortic stenosis, while force-velocity-shortening relationships of patients with primary myocardial failure clearly differed. A major determinant of wall stress was the ratio of left ventricular wall thickness to cavity radius (h/R). Patients with h/R ratios> 0.36 had higher values for ejection fraction (0.61±0.06 vs 0.36 i 0.07, p< 0.05), Vcf (0.79±0.10 vs 0.39±0.04 sec', p< 0.05) and stroke work index (71±10 vs 45 9 gm/m2, p< 0.005) than those with lower ratios. The results indicate that left ventricular wall thickness and geometry are closely correlated withventricular performance in patients with pressure-overload hypertrophy due to aortic stenosis. Poor cardiac performance in some such patients may be due to inadequate hypertrophy (or inappropriate geometry) rather than to depression of myocardialcontractility.