Validity of the ECG diagnosis of left ventricular hypertrophy in normotensive and moderately hypertensive men when using the echocardiographic assessment of left ventricular mass index as reference.

The diagnostic validity of ECG criteria for left ventricular hypertrophy (LVH) was assessed in 100 men aged 22-64 (mean 47) years with moderate hypertension (Group 1) and 95 age-matched normotensive men (Group 2) using echocardiographic recordings of LV mass index (MI) as reference. A diagnosis of LVH was made in subjects with LVMI greater than or equal to 125 g/m2. Mean LVMI was 126 +/- 34 g/m2 in Group 1 vs. 100 +/- g/m2 in Group 2 (P less than 0.001), and the prevalence of LVH was 48% and 11% respectively (P less than 0.001). The mean ECG voltage according to Sokolow-Lyon (S-L) was 28 +/- 8 mm in Group 1 and 27 +/- 7 mm in Group 2 (NS); with 19% having LVH in Group 1 and 14% in Group 2 (NS). Using the Cornell criterion Group 1 had on average 15 +/- 6 mm vs. 12 +/- 5 mm in Group 2 (P less than 0.001), but only two Group 1 patients had LVH. In Group 2 a significant negative correlation between age and S-L voltage was found (r = 0.33, P less than 0.001). LVMI was not correlated with any of the two voltage criteria using linear regression analysis whereas multiple regression analysis revealed a weak, but significant correlation between LVMI and S-L voltage in Group 1 (t = 2.06, P = 0.04). No subject had LV strain pattern or LVH according to the Romhilt Estes point score system. In the assessment of possible LVH in normal or moderately hypertensive men less than 65-70 years of age, ECG has limited value.