Abstract
Bloodless method for measuring human cardiac stroke volume (SV) was proposed by Kubicek et al. in 1966 (Aerosp. Med.). If it is assumed that the variation of the recorded thoracic impedance (Zt) is attributable to that of the blood volume, SV is given by the equation, SV=A|dZt/dt|maxVET (A; constant, VET; ventricular ejection time). The Zt, however, usually involves a slow respiration-synchronized component. This slow component might be due to the volume-change in the air or in the blood or both, which obscures quantitative assessment of the beat-to-beat SV change. It should be noted that the respiration itself affects SV. Without an autonomic regulation of the heart rate, both SV and |dZt/dt|max may fluctuate with respiratory variation of the venous return. Thus it seems essential to hold breath during recording and then calculate an average of several |dZt/dt|maxs to exclude the erroneous component. In the present study we tried to specify the origin of the slow component of Zt under various recording conditions with different placements of electrodes. The beat-to-beat VET was measured by comparing dZt/dt with heart sound, which was not significantly changed by breathing. Thus it was suggested that the errors in measuring SV might be due mainly to the respiratory variation of |dZt/dt|max. It will be shown that the erroneous component is minimized in some recording conditions. Limitations of the non-invasive SV measurement will be discussed. [Jpn J Physiol 55 Suppl:S95 (2005)]
