Comparison of left ventricular size by computed tomography with magnetic resonance imaging measures of left ventricle mass and volumes: The multi-ethnic study of atherosclerosis

Background: A non-contrast-enhanced cardiac computed tomography (NCE-CCT) scan for assessing coronary artery calcification (CAC) is being increasingly used for assessing underlying burden of atherosclerosis. Although many studies document the potential value of measuring CAC, little is known about the other measures such as left ventricular (LV) geometry that can be obtained from the same scan data. Objectives: We sought to evaluate the accuracy of noncontrast CT-derived LV size (LVS; sum of LV volume and mass) compared with magnetic resonance imaging (MRI)-derived measures as the clinical reference standard. Methods: Participants (n = 5004) in the Multi-Ethnic Study of Atherosclerosis (MESA) who underwent cardiac MRI studies from August 2000 to September 2002 were included. CT-derived LVS was defined as the sum of LV mass and LV intracavitary volume. The calculated LVS was taken from a measurement of a single slice from noncontrast images. Multivariate analysis adjusting for demographics was used to identify predictors of the relation between CT LVS and MRI LVS, and Bland-Altman analysis was performed comparing MRI-derived measures with CT-derived measure of LVS. Results: The mean CT LVS was 187.8 ± 56.8 mL (range, 33.6-486.4 mL). The correlation was 0.73 (P = 0.01) for MRI-derived LV volume and 0.74 (P = 0.01) for MRI-derived LV mass. The correlation between CT LVS and MRI-derived LV end-diastolic total volume (mass + volume) was 0.79. Conclusion: A single NCE-CT scan used to detect and quantify coronary calcification can also estimate LVS with reasonable accuracy compared with MRI. This provides a new method to study ventricular size in epidemiologic studies and potentially provide additional information for clinical screening of cardiac risk.