Estimation of cardiac output and peripheral resistance using square-wave-approximated aortic flow signal

Nima Fazeli,Jin-Oh Hahn

doi:10.3389/fphys.2012.00298

Abstract

This paper presents a model-based approach to estimation of cardiac output (CO) and total peripheral resistance (TPR). In the proposed approach, the response of cardiovascular system (CVS), described by the windkessel model, is tuned to the measurements of systolic, diastolic and mean arterial blood pressures (BP) so as to yield optimal individual- and time-specific system time constant that is used to estimate CO and TPR. Unique aspects of the proposed approach are that it approximates the aortic flow as a train of square waves and that it also assumes pressure-dependent arterial compliance, as opposed to the traditional windkessel model in which aortic flow is approximated as a train of impulses and constant arterial compliance is assumed. It was shown that the proposed model encompasses the standard windkessel model as a limiting case, and that it also yields more realistic BP waveform response than the standard windkessel model. The proposed approach has potential to outperform its standard counterpart by treating systolic, diastolic, and mean BP as independent features in estimating CO and TPR, rather than solely resorting to pulse pressure as in the case of the standard windkessel model. Experimental results from in-vivo data collected from a number of animal subjects supports the viability of the proposed approach in that it could achieve approximately 29% and 24% reduction in CO and TPR errors when compared with its standard counterpart.

Highlights

Cardiac output (CO) is one of the most important hemodynamic parameters to be monitored and assessed in ambulatory and critically ill patients (Jansen et al, 1990)
Inspired by its wide acceptance and frequent application in CO and total peripheral resistance (TPR) estimation, this study aims at developing a universal approach that has potential to enhance the efficacy of CO and TPR estimation based on the standard windkessel model (Frank, 1930)
We focus on two main opportunities to enhance the CO and TPR estimation efficacy of the standard windkessel-model-based method: (1) to use a better approximation of aortic flow signal that can result in more realistic blood pressures (BP) waveform(s), and (2) to exploit independent morphological features in the arterial BP waveform more rigorously rather than solely relying on the pulse pressure as in the standard CO estimator implemented with traditional windkessel model

Summary

Introduction

Cardiac output (CO) is one of the most important hemodynamic parameters to be monitored and assessed in ambulatory and critically ill patients (Jansen et al, 1990). It is frequently used for disease diagnostics and monitoring (Heldt, 2006). Direct measurement of CO is extremely difficult. The clinical gold standard accepted for CO measurement is the thermo-dilution (Ganz et al, 1971), but it is known to be a highly invasive procedure that has limited accuracy (Botero et al, 2004) and may incur cardiovascular risk (Manecke et al, 2002). Non-intrusive techniques such as echo-cardiography (Ihlen et al, 1984) and electrical velocimetry (Suttner et al, 2006; Zoremba et al, 2007) are promising alternatives, but often their accuracy is not satisfactory enough yet to be clinically applicable (Siegel and Pearl, 1992)

Objectives

Methods

Results

Discussion

Conclusion