[Echocardiographic evaluation of cardiac output in patients with acute myocardial infarction (author's transl)].

Echocardiography is a non invasive technique particularly useful to assess both hemodynamics and left ventricular performances; different methods have been proposed to evaluate echocardiographically the cardiac output. The aim of this work is to suggest a new method (planimetry of the area enclosed by echoes from the mitral valve) stating its advantages and limits and to test three already known methods (cubes method, method of Lalani and Lee, method of Rasmussen and coll.) comparing them to thermodilution. In 24 patients admitted in our Coronary Care Unit for acute myocardial infarction 30 measurements have been performed simultaneously with echocardiography and thermodilution. Statistical analysis (correlation coefficient with the method of the square minimums and Student's 't') of the results has shown that the cubes method is unprecise (P = NS) and unbiased (d = 0.97 +/- 15.46); Lalani-Lee method is unprecise (P = NS) and biased (d = 9.64 +/- 21.36) by a systematic underestimate error concealed by the poor precision; Rasmussen and coll. method is precise (P < 0.01) but biased (d = --3.03 +/- 5.5), the systematic mistake being entirely due to the group of patients with abnormal hemodynamics. Our method is precise (P < 0.01) and unbiased (d = --0.04 +/- 9.62) in patients both with normal and abnormal hemodynamics. Therefore this method offers the following advantages: 1) it is unbiased, even in the presence of abnormal left ventricular performance; 2) it is sufficiently precise, having a correlation coefficient with thermodilution of r = 0.80; 3) it measures directly the amount of flow due to the atrial systole; 4) the measurement is rapid and hardly biased by the reader.