How many patients would be misclassified using M-mode and two-dimensional estimates of left atrial size instead of left atrial volume? A three-dimensional echocardiographic study

Objectives Left atrial size has shown prognostic importance in a variety of cardiac conditions. Diameters, area, and volume derived from M-mode and two-dimensional (2D) echocardiography are commonly used to estimate left atrial size. However, M-mode and 2D measures of left atrial size rely on various geometrical assumptions and their accuracy remains to be determined. To address this issue, we compared M-mode and 2D parameters routinely used to estimate left atrial size with three-dimensional (3D) echo measured left atrial volume (LAV) as a reference standard. Methods We studied 104 patients (55% males, 62 +/- 15 years, range 10-87 years), presenting for a routine echocardiographic evaluation. Results The mean 3D LAV for the study population was 90 +/- 68 ml (range 24-458 ml). We found highly significant (P<0.0001) correlations between 3D LAV and left atrial anterior-posterior (r=0.78, 95% CI=0.69-0.85), superior-inferior (r=0.74, 95% CI=0.63-0.81) and medial-lateral (r=0.91, 95% CI=0.86-0.93) diameters. A highly significant correlation was also found between 3D LAV and left atrial area (r=0.94, 95% CI=0.91-0.96). However, using M-mode anterior-posterior diameter or left atrial area would have misclassified 57% and 70% of our study patients, respectively, regarding the degree of left atrial dilatation. Closer correlations and narrower confidence intervals were found between 3D LAV and single-plane (r=0.98; 95% CI=0.94-0.97) and biplane (r=0.97; 95% CI=0.96-0.98) 2D LAVs. Conclusion Left atrial diameters and area measurements were poor predictors of 3D LAV, especially in the enlarged left atria. Therefore, these parameters can be misleading in assessing the severity of left atrial dilatation. Two-dimensional LAVs are accurate in estimating 3D LAV. The small additional accuracy obtained by using the biplane instead of the single-plane area-length method, and the fact that the biplane method is more technically demanding and time consuming, may allow the use of the area-length for routine clinical use.