Background: Left atrial (LA) volume is a predictor of cardiovascular events. Information on LA volume is available on contrast-enhanced electrocardiogram (EGC)-gated multidetector computed tomography (MDCT) scans.
Objective: To assess interobserver and intraobserver reproducibility of 3-dimensional threshold-based volume (3DTV) and 2-dimensional (2D) measurements for the assessment of LA volumes with contrast-enhanced cardiac 64-slice MDCT.
Methods: Contrast-enhanced 64-slice MDCT (0.6-mm slice thickness, 120 kVp, 850 mAseff) was performed in 96 consecutive subjects (mean age 52 years; 48% women) as a subset of the Rule Out Myocardial Infarction using Computer Assisted Tomography trial. Two observers independently measured maximal (LAV(max)) and minimal (LAV(min)) LA volumes with (1) a modified Simpson's method (3DTV) based on delineation of LA areas in axial slices and (2) estimated LA volumes typically used in 2D echocardiography (area length and prolate ellipse). Interobserver and intraobserver reproducibility for each method as well as correlations between the methods were calculated.
Results: Interobserver (n = 96) and intraobserver (n = 20) variability was significantly lower for 3DTV (8%) than for area length (13%; P < 0.001) or prolate ellipse (16%; P < 0.001). 2D-based measurements rendered significantly lower LA volumes than did 3DTV (area length: -17% and -22%; prolate ellipse: -43% and -46% for LAV(max) and LAV(min), respectively; P < 0.001 for all). By 3DTV, mean LA volume was 90.4 +/- 24.5 mL for LAV(max) and 52.5 +/- 17.6 mL for LAV(min).
Conclusion: ECG-gated contrast-enhanced cardiac MDCT offers volumetric assessment of LA volume with excellent reproducibility without additional contrast administration or radiation exposure. 3D measures of LA volume are more reproducible and render larger volumes than 2D-derived estimates, typically used in echocardiography.