To define and compare predictors of postoperative normalization of diastolic left ventricular dimension after aortic valve replacement, echocardiographic indexes of left ventricular size, function, degree of hypertrophy and systolic wall stress were examined in 43 patients with chronic and 14 with acute aortic insufficiency. In all of the latter 14 patients, left ventricular diastolic dimension returned to normal (mean 5.2 +/- 0.4 cm) in the postoperative follow-up period (mean 8.0 months). In contrast, of those with chronic insufficiency, 28 (group A) had postoperative normalization of diastolic dimension whereas the remaining 15 (group B) had persistent enlarged diastolic dimension. Preoperative end-systolic dimension, diastolic radius/thickness ratio, mean radius/thickness ratio, mean wall stress and end-systolic stress were 84 to 93 percent accurate in predicting normalization versus persistence of left ventricular dilatation postoperatively, and were superior to preoperative end-diastolic dimension and shortening fraction. Postoperatively, group A had complete normalization of end-systolic dimension and of mean and end-systolic wall stresses with persistence of a normal shortening fraction. Group B continued to have increases in end-systolic dimension, mean wall stress and end-systolic stress with a reduction in shortening fraction. Postoperatively there was a 43 and 29 percent incidence rate of heart failure and death by heart failure, respectively, in group B versus none in group A (p less than 0.01). These findings support the concept that inappropriate hypertrophy in chronic aortic insufficiency is associated with progressive increases in wall stress and end-systolic dimension and a reduction in shortening fraction that eventually result in irreversible cardiac dilatation and failure. Accurate and clinically relevant determination of reversible and irreversible alterations in left ventricular size and function may be obtained with these echocardiographic indexes.