Two-dimensional (2D) echocardiographic estimation of infarcted mass is limited by having only a few selected nonparallel views for data analysis. Volume-rendered three-dimensional (3D) echocardiography may be able to overcome the above limitations, because it uses multiple, parallel 2D images to derive quantitative data. Previous experimental studies demonstrated that 3D echocardiography is an accurate and reproducible method to assess dysfunctional mass. To estimate the accuracy of 3D echocardiography in humans, we evaluated 10 patients who had a single myocardial infarction. All patients underwent 2D and 3D echocardiography using the transesophageal approach, and contrast (gadolinium) magnetic resonance imaging (MRI), considered a reference standard for infarcted tissue detection. The mean extent of dysfunctional mass by MRI was 28 +/- 13 g and by 3D echocardiography was 30 +/- 12 g; the mean difference was 1.9 +/- 2.3 g (p = not significant). Linear regression analysis between the 2 measurements was y = 0.97x - 1.12, r = 0.98. Dysfunctional mass derived from 3D echocardiography reflects the real site and extension of damaged myocardium.