Comparison of time domain algorithms for estimating aortic characteristic impedance in humans

Using data obtained intraoperatively from 134 patients (262 data sets), ten algorithms for estimating aortic characteristic impedance from the relative slopes of the aortic pressure and flow (electromagnetic flow probe) waveforms during early systole were compared to the estimate obtained by averaging input impedance modulus values for frequencies between 2 and 16 Hz (Zc). Results clearly confirmed the relationship between these slopes and Zc:r>or=0.80, y intercepts approximately=0.0 and slopes approximately=1.0 for all algorithms tested. However, four algorithms yielded estimates with r >or=0.95. The common trait of these four algorithms was their dependence on portions of the pressure and flow waveforms independently centered around peak derivatives. Results imply that Zc can be estimated successfully in real time, which would be advantageous in critical postoperative periods. Furthermore, cumbersome computational procedures can be eliminated whenever Zc is the only spectral-related parameter to be calculated. >