|| Checking for direct PDF access through Ovid
Background: The ASE/EAE guidelines concluded that echocardiographic LV mass is more accurate and reproducible than septal thickness for the estimation of LVH. However, the relative prognostic utility of these two methods has never been compared. Methods: 2545 adults referred for any indication had LVH assessed measuring LV mass by the ASE/EAE formula using LV linear dimensions indexed to BSA. Multivariable Cox models were used to assess the association between LVH degree and the risk of death. ROC curves were generated to assess the prognostic performance of septal thickness and indexed LV mass. Results: 2545 subjects (mean age 61.9±15.8, 53% women, mean septal thickness 10.3±2.2 mm and mean indexed LV mass 107±37g/m2). 1335 (52.5%) subjects had LVH using the septal thickness and 1198 (47%) by indexed LV mass. After 2.5±1.2 years 121 (4.7%) deaths occurred. Survival was worse with greater LVH. This relation persisted after adjusting for age, sex, LVEF, WMSI, significant valvular disease and atrial fibrillation. The adj HR for all cause death was 1.7 for mild (95%CI 1.14-2.54, p=0.009), 2.33 for moderate (95%CI 1.34-4.06, p= 0.003) and 3.00 for severe (95%CI 0.92-9.71, p=0.06) abnormal septal thickness. The adj HR for all cause death was 2.17 for mild (95%CI 1.23-3.81, p=0.007), 3.04 for moderate (95%CI 1.76-5.24, p<0.001), and 3.81 for severe (95%CI 2.43-5.97, p<0.001) abnormal indexed LV mass. The area under the ROC curve for the model including the 4 degrees of LVH by indexed LV mass was superior (AUC=0.66) to that based on the septal thickness (AUC=0.58), p=0.0004, Fig. Conclusion: In a cohort study of unselected outpatients the categorization of LV mass by the ASE/EAE recommendations offered prognostic information beyond that provided by the linear measurements such as septal thickness.