Nongeometric Doppler stroke volume determination

Abstract

Accurate noninvasive measurement of cardiac output is a long-sought goal with obvious advantages. Numerous methods have yielded varying success. Recent Doppler echocardiographic studies have produced promising results. 1,2 These studies depend on geometric measurements, that is, the cross-sectional flow area of a valve or vessel. Stroke volume is then calculated as the product of the mean blood velocity by Doppler (flow velocity integral or “stroke distance”) and orifice area. Although correlations with hemodynamically measured stroke volume and cardiac output have been satisfactory, many limitations are inherent in the requirement for accurate cross-sectional flow area, including (1) changes in the flow area itself with changing cardiac output; (2) inaccurate true orifice measurement because of inaccurate edge detection by M-mode or 2-dimensional echocardiography due to poor resolution, assumption of a circular crosssection or because the functional area may be less than the measured area (seen as a halo around a color flow jet in cross-section); and (3) nonlaminar flow without a flat flow profile at the site of measurement. A method to eliminate the need for geometric measurements when measuring blood velocity at a point where flow is laminar and velocity profile flat would be advantageous. We developed a method with these characteristics and preliminary results are encouraging. We report the initial investigation of this approach. The basis of the nongeometric method is the relation: stroke volume = ejection time × ejection rate (SV = ET × ER). Thus, if the ejection time and ejection rate are known, stroke volume can be calculated. The left ventricular ejection time can be measured with precision from Doppler flow traces. 4 The ejection rate can be determined indirectly—we recently demonstrated the close relation between mean velocity calculated from the Doppler spectral envelope as blood is ejected and the volume ejection rate determined by thermodilution. 5 The latter was calculated from the relation: ejection rate = stroke volume/ejection time (LVER = SV/LVET). Of note, in the validation investigations the ejection time was determined using both the carotid pulse and the Doppler tracing independently. 5 To ensure laminar flow with a flat velocity profile, the Doppler sample volume was placed in the center of the aortic valve orifice. 6 The relation between Doppler-derived mean ejection flow velocity and thermodilution-derived volume ejection rate yielded the following regression equation 5: y = 494x − 66 ml/s, where y is mean left ventricular volume ejection rate in ml/s and x is Doppler mean flow velocity in m/s. Thus, by measuring mean Doppler velocity at the aortic valve orifice, the regression equation yielded the left ventricular ejection rate. 5