Sensitivity distributions of impedance cardiography using band and spot electrodes analyzed by a three-dimensional computer model

Impedance cardiography (ICG) offers a safe, noninvasive, and inexpensive method to track stroke volume estimates over long periods of time. Several modified ICG measurement configurations have been suggested where for convenience or improved performance the standard band electrodes are replaced with electrocardiogram electrodes. This report assesses the sensitivity of the conventional and three modified ICG methods in detecting regional conductivity changes in the simulated human thorax. The theoretical analyses of the measurement sensitivity employ the reciprocity theorem and the lead field theory with a highly detailed, anatomically accurate, three-dimensional computer thorax model. This model is based on the finite-difference element method and the U.S. National Library of Medicine's Visible Human Man anatomy data. The results obtained indicate that the conventional four-band ICG is not specifically sensitive to detect conductivity changes in the region of the heart, aortas, and lungs. Analyzed modified electrode configurations do not reproduce exactly the measurement sensitivity distribution of the conventional four-band ICG. Thus, although the signals measured with modified spot arrangements may appear similar to the four-band configuration, the distribution of the signal origin may not be the same. Changing from band to spot electrodes does not overcome the methodological problems associated with ICG. (C) 1998 Biomedical Engineering Society.