Two-dimensional echocardiographic evaluation of left ventricular ejection fraction by the ellipsoid single-plane algorithm: a reliable method for assessing low or very low ejection fraction values?

The reliability of two-dimensional (2D) echocardiographic estimation of left ventricular ejection fraction (EF) is commonly recognized, but no satisfactory data are available about the accuracy of low or very low EF values determined by 2D echocardiography (ECHO-EF). The purpose of our study was to assess the reliability of low ECHO-EF values obtained using a simple time-economical algorithm such as the ellipsoid single-plane area-length method. Radionuclide angiography (RAD-EF) was taken as the standard of comparison. We studied 59 consecutive patients (31 women and 28 men) referred to our echocardiographic laboratory. Both 2D echocardiography and radionuclide angiography were blindly performed within 48 h of one another. EF was calculated by the two methods and then compared. Data were globally analyzed. Furthermore, data were divided and analyzed according to the ECHO-EF cut-off point of 50%. An ECHO-EF value of 50% was chosen to conventionally distinguish between low ECHO-EF values and normal-high ones. Data were plotted, and the line of equality and the regression lines were drawn. Regression line slopes, correlation coefficients, means and standard deviations were calculated. The agreement was analyzed by calculating the mean difference (RAD-EF-ECHO-EF) and the standard deviation of the differences. ECHO-EF was linearly related to RAD-EF even when data were split. In particular, as regards ECHO-EF < or = 50%, the regression line practically overlapped the line of equality, and the two methods showed both a strong correlation and a good degree of agreement.