The evaluation of left ventricular function in man. A comparison of methods.

Abstract

Comparisons of the sensitivities of parameters for assessing left ventricular performance in man were made in 38 patients. The parameters compared were the ejection fraction, ventriculographic contraction patterns, the left ventricular end-diastolic pressure, and the contractile indices including the contractile element velocity at 10 mm Hg (Vce 10) and maximal contractile element velocity (Vmax). The contractile indices were obtained by catheter tip manometry, utilizing developed pressure (DP) to calculate the velocity of contractile element shortening (Vce) from the formula: dp/dt divided by 32 DP. Vce 10 was measured directly and Vmax was derived by linear manual extrapolation of the pressure-velocity plot to 0 mm Hg. Vmax values derived from linear manual extrapolation were compared with values obtained by computer least squares fitting of the Vce and developed pressure data points to single and double exponential equations. The Vce and developed pressure data points fit the single exponential equation better than the double exponential equation but the use of either equation resulted in slightly higher values for Vmax than obtained with linear manual extrapolation. The effect of heart rate on myocardial contractility was eliminated by making comparisons at both a basal and atrial paced rate of 100. Utilizing all methods, 24 patients were identified to have ventricular dysfunction. The contractile indices were significantly less sensitive than any other parameter (P smaller than 0.05) and identified seven patients while the left ventricular end-diastolic pressure, ejection fraction, and presence of asynergy identified 15, 15, and 12 patients, respectively. The use of a common atrial paced rate of 100 did not increase the sensitivity of the contractile indices. Since there was only partial overlapping between parameters in the identification of left ventricular dysfunction, the combination of different parameters was more sensitive than any single parameter alone. It is concluded that several methods are required to identify all patients with left ventricular dysfunction and that the contractile indices are the least sensitive indicator of left ventricular dysfunction.