Left ventricular mass (LVM) as assessed by magnetic resonance imaging (MRI, LVM(MRI)) and electrocardiographic (ECG) voltage reflect different pathological features. We hypothesized that ECG voltage is related to the electrical potential of cardiac muscle cells (electrical LVM) and to anatomical LVM as evaluated by MRI, and that the divergence between electrical LVM and anatomical LVM reflects the degree of myocardial damage. Because adipose tissue has high electrical resistance, we previously found a very strong correlation between body-fat-corrected mean ECG voltage (Vfm) and LVM as estimated by echocardiography in patients with essential hypertension. In this study we compared LVM(MRI), Vfm, the ratio of Vfm x 10(2)/LVM(MRI), and the results of 99mTc tetrofosmin scintigraphy in patients with and without myocardial infarction (MI). We studied 33 patients without Ml and 26 patients with Ml. Vfm significantly correlated with LVM(MRI) in patients without MI (r=0.71, p<0.01). The ratio of Vfm x 10(2)/LVM(MRI) apparently reflected the relation between electrical LVM and anatomical LVM. Vfm x 10(2)/LVM(MRI) in patients with MI was smaller than that in patients without MI (0.98+/-0.28 vs. 1.42+/-0.29, p<0.01). Vfm x 10(2)/LVM(MRI) decreased as 99mTc score increased (r=-0.66, p<0.01). Our results indicate that Vfm is a useful index of electrical LVM and that Vfm x 10(2)LVM(MRI) reflects the electrical potential of the viable myocardium in total anatomical LVM.