Serial studies of adaptation to aortic regurgitation (AR) were undertaken to determine whether sonomicrometry and echocardiography could be combined to measure changes in left ventricular (LV) cavitary volume (V-1V) and wall mass using the geometric formula [V-1V = K pi b(2)a - wall volume], where K is a constant depending on the geometric model and a and b are epicardial major- and minor-axis diameters, respectively. Postmortem studies were performed in six normal dogs and in nine with AR; ultrasonic ventricular dimensions were measured as V-1v varied with an intracavitary balloon. Three models were tested: 1) ellipsoid (model I; K=1/6), 2) cylinder-ellipsoid (model II; K= 5/24), and 3) cylinder (model III; K = 1/4). The slope of the relationship between calculated V-1v and balloon volume varied between models (I, 0.71 +/- 0.11; II, 0.89 +/- 0.14; III, 1.07 +/- 0.17), and empiric determination of K to produce a slope of 1.0 resulted in a value of 0.26 +/- 0.04, not significantly different from the cylindrical model. Serial measurements of LV dimensions in 10 chronically instrumented conscious dogs revealed no significant change in end-diastolic or end-ejection LV shape after up to 16 wk of AR. Sonomicrometry and echocardiography can be integrated using a cylindrical geometric model to accurately estimate changes in end-diastolic or end-ejection V-1v during chronic volume overload.