Correlation between Global Longitudinal Strain and QRS Voltage on Electrocardiogram in Patients with Left Ventricular Hypertrophy

PURPOSE: Left ventricular hypertrophy (LVH) is as an independent risk factor. Discrepancies were reported between LV mass (LVM) estimated by echocardiography and electrocardiography (ECG) findings. We hypothesized that QRS voltage criteria may reflect not only anatomical changes (LVM) but also changes in LV function and we tested the relationship between QRS voltage and echocardiographic parameters of LV function in patients (pts) with different types of LVH. METHODS: We prospectively enrolled pts with LVH and preserved ejection fraction (LVEF >50%): 20 pts with isolated arterial hypertension, HTN, 20 pts with severe aortic stenosis, AS (indexed aortic valve area <0.6 cm(2)/m(2)), and 20 pts with symmetric hypertrophic cardiomyopathy, HCM. Standard 12-lead ECG (including Sokolow and Cornell voltage indices) and a comprehensive two-dimensional (2D) echocardiography were performed in all. Left ventricular mass was calculated according to Devereux formula. Global longitudinal strain (GLS) was assessed by speckle tracking echocardiography. RESULTS: A significant correlation was found between both ECG indices and LVM assessed by echocardiography. Moreover, significant correlations were found between Sokolow-Lyon voltage and LVEF (r = 0.26; P = 0.03), GLS (r = 0.59; P < 0.001) and E/e' average (r = 0.43; P < 0.001). Cornell voltage index correlated significantly only with GLS. In multivariable analysis GLS emerged as the only independent correlate of both Sokolow-Lyon (ß = 0.6, P < 0.001) and Cornell voltage indices (ß = 0.45, P < 0.001). CONCLUSION: These findings suggest that in pts with LVH, ECG should no longer be used only as a surrogate method for LVM estimation (structural changes only), but rather as an investigation complementary to imaging, incorporating information on overall LV remodeling (changes in structure and function).