A Study of the Human Heart as a Multiple Dipole Electrical Source

Using the method previously described, the time-varying strengths of 12 dipoles (representing the depolarization of the ventricles) were calculated from body surface potential measurements. Dipole activities (DA) were obtained for each dipole by time integration of its strength. Seventy-two patients with angiographically determined left ventricular muscle weight (LVMW) were studied. Fifty-two of 53 patients with increased LVMW had increased DA in the left ventricle and septum (LVSDA). Thus, diagnostic sensitivity was 98%. Sixteen of 19 patients with normal LVMW had normal LVSDA (specificity, 84%). The correlation between LVSDA and LVMW was r 2 =86%, and the standard error of the LVMW estimate ( see ) was±49 g. A subgroup with pure left ventricular hypertrophy (LVH) consisting of 27 patients over 25 years of age with isolated aortic valve disease was studied separately. The correlation between LVSDA and LVMW was r 2 =92% and see was ±31 g. After correcting for the uncertainty in the angiographic LVMW measurements, the see for the complete group was reduced to ±37 g. The diagnosis and quantitation of LVH with this method was considerably better than with conventional electrocardiography or vectorcardiography.