Origin of cardiac-related thoracic electrical impedance variations in lambs

Cardiac-related deflections in thoracic electrical impedance have been thought to correlate sufficiently well with cardiac stroke volume to be used as the basis for a noninvasive estimation of cardiac output. To determine more precisely the physiological origin of the impedance deflection (DZ), we regarded right ventricular stroke volume (SVa) as the sum of two components: 1) that part of SVa responsible for the transient increment in pulmonary blood volume within a cardiac cycle, SVa-v and 2) the remaining part of SVa, (SVa-SVa-v). SVa-v was measured in lambs by integration of the difference between pulmonary arterial and pulmonary venous flow. SVa and its components were varied experimentally by opening and closing an aorticocaval shunt or by inflating and deflating a cuff implanted around the pulmonary artery. DZ was measured using a tetrapolar disk electrode system. Multivariate linear regression analysis revealed that SVa-v had a significant positive effect on DZ, and, at the same time, (SVa-SVa-v) had a significant negative effect on DZ. In the pulmonary artery occluder model, the positive effect of SVa-v dominated the opposing negative effect of (SVa - SVa-v) so that the net effect of SVa on DZ was positive and significant. In the aorticocaval shunt model, these effects opposed each other to the extent that there was no significant correlation between SVa and DZ. These results shed new light on the physiological origin of DZ. They also demonstrate that use of DZ to measure acute changes in cardiac output may yield misleading results. Changes or the lack of changes in thoracic electrical impedance do not necessarily reflect cardiac output status.