Background: Standard electrocardiography (ECG) is a specific, but poorly sensitive tool for diagnosis of left ventricular (LV) hypertrophy. In a large population of subjects with hypertension we tested some standard ECG criteria in their sensitivity and specificity for LV hypertrophy. LV mass at echocardiography was the reference standard.
Methods: In the setting of the MAVI (MAssa Ventricolare sinistra nel soggetto Iperteso) study, the ECG and echocardiographic tracings of 947 hypertensive subjects were read blindly in a central office.
Results: Prevalence of LV hypertrophy at ECG was 0.6, 3.0, 4.8, 7.1, 11.1, 11.9 and 18.4%, respectively, using the following criteria: Wilson, typical strain, Romhilt-Estes score > or = 5 points, Gubner-Ungerleider, Sokolow-Lyon, Cornell voltage (S(V3)+R(avL) > 2.8 mV in men or 2.0 mV in women) and Perugia score (positivity of at least one of the following: S(V3)+R(aVL) > 2.4 mV in men or > 2.0 mV in women, a typical strain pattern, or a Romhilt-Estes point score >or = 5). Prevalence of LV hypertrophy at echocardiography ranged from 27.2% (LV mass > 125 g/m2) to 49.9% (LV mass > 51.0 g/m2.7). Using the latter gold standard, sensitivity and specificity of the above ECG criteria were 0.8 and 99.6% (Wilson), 3.8 and 97.9% (strain), 5.9 and 96.4% (Romhilt-Estes), 9.7 and 95.6% (Gubner-Ungerleider), 11.2 and 91.1% (Sokolow-Lyon), 15.2 and 91.4% (Cornell), and 22.2 and 85.4% (Perugia score).
Conclusions: Sensitivity of traditional ECG criteria for LV hypertrophy in subjects with hypertension is poor. However, the combination of three highly specific criteria (Romhilt-Estes, LV strain and Cornell) in a cumulative score produces a rise in sensitivity without excessive deterioration of specificity, with a prevalence of LV hypertrophy at ECG of 18.4%. Traditional interpretation of ECG is valuable and should be reconsidered in the clinical work-up of subjects with hypertension.