Changes in body-surface electrocardiograms from geometric remodeling with obesity

Both diabetes and obesity cause cardiac dysfunction. To separate consequences of geometric changes due to obesity from electrophysiological ones, we investigated how changes in cardiac and torso geometry affected body-surface ECGs. For this study, we modified the realistic heart and torso models of the simulation package ECGSIM. ECGs were calculated from action potentials on the heart surface using our bidomain forward-problem solution. These ECGs were studied using spectral- and principal-component analyses and isopotential and energy maps. We found relative errors over the body-surface during the QT interval of 12%, 14%, and 68% for hypertrophy of the heart, extension of the abdomen, and heart displacement with obesity, respectively. The major change to the standard 12-lead set also occurred with heart displacement. The mean relative error over the QT interval in the precordial leads was 78% with heart displacement. These results demonstrate the limitations of using standard lead sets to characterize electrocardiographic changes in obese subjects and point to the need for more inclusive measures, such as body-surface mapping and inverse electrocardiography, to describe electrical remodeling in the presence of habitus changes due to obesity.