Improved reproducibility of left atrial and left ventricular measurements by guided three-dimensional echocardiography

Abstract

Objectives: The objective of this study was to determine whether guided three-dimensional echocardiography could improve the reproducibility of left atrial and left ventricular anteroposterior measurements over that of standard unguided two-dimensional echocardiography.

Background: Although these measurements are standard indexes for evaluating chamber size, their use is limited by significant interobserver variability largely due to variable image plane positioning. To improve measurement accuracy and reproducibility, we have developed a three-dimensional echocardiograph that displays the line of intersection of the real-time image with a previously saved orthogonal reference image. This display shows the relation of the real-time image to anatomic landmarks in its third, nonvisualized dimension and may be used to guide image positioning.

Methods: Three pairs of operators independently performed unguided two-dimensional and guided three-dimensional examinations on three groups of 10 patients each. The left atrium was measured in a plane through the inferior surface of the aortic cusps and the left ventricle in a plane perpendicular to its long axis 1 cm below the mitral leaflet tips. Interobserver variability of these measurements on unguided parasternal long-axis images and on guided short-axis images was assessed.

Results: The standard unguided two-dimensional examination was associated with an interobserver variability of 14.6% and 9.1% for atrial and ventricular measurements, respectively. Guided three-dimensional echocardiography significantly reduced interobserver variability to 5.0% and 3.1%, respectively, for the same measurements (p < 0.005 by McNemar's test).

Conclusions: Significant interobserver variability occurs with standard unguided two-dimensional echocardiographic measurement of left atrial and left ventricular dimensions. Guided three-dimensional echocardiography achieves a nearly threefold improvement of reproducibility of these measurements and provides the basis for improved serial evaluation and comparison of atrial and ventricular size by different operators.