Chronic aortic regurgitation leads to progressive left ventricular dilatation and hypertrophy. It is important to be able to measure these variables for clinical decision making on the timing of aortic valve replacement. M-mode and two-dimensional echocardiography are widely utilized in clinical practice for a qualitative assessment of left ventricular size and function. Recent recommendations have been proposed for the use of quantitative assessment by left ventricular size and left ventricular mass by two-dimensional echocardiography; there has been no study examining the validity of these measurements in patients with aortic regurgitation. The purpose of this study was to prospectively examine the various geometric models of echocardiographically determined left ventricular mass and volumes in patients with chronic aortic regurgitation compared with cine-computed tomographic scanning. Twenty-two patients with chronic aortic regurgitation were prospectively identified and underwent two-dimensional echocardiographic and cine-computed tomographic scanning. M-mode and two-dimensional echocardiographic images were analyzed on an off-line measurement digital system. Seven previously described geometric models were used in this study for quantitative analysis. Both left ventricular mass and left ventricular volume were calculated from real-time ultrafast tomographic scans. Left ventricular mass determinations by two-dimensional echocardiography correlated more than M-mode determinations (r = 0.84 vs. r = 0.75, standard error of the estimate (SEE) = 38 g vs. 139 g). Biplane two-dimensional echocardiography formulas using Simpson's rule construct yielded more accurate values for left ventricular end-diastolic volumes (r = 0.92, SEE = 24 ml) than either the single-plane Simpson's rule (r = 0.58, SEE = 73 mL) or methods using predetermined models of left ventricular shape (r = 0.80, SEE = 38 ml).(ABSTRACT TRUNCATED AT 250 WORDS)