Systematic review of cardiac output measurements by echocardiography vs. thermodilution: the techniques are not interchangeable

Abstract

Echocardiography is frequently used in the hemodynamic evaluation of critically ill patients, but inaccurate measurements may lead to wrong clinical decisions. The aim of our systematic review was to investigate the interchangeability of echocardiography with thermodilution technique in measuring cardiac output and its changes. In August 2015 we systematically searched electronic databases and included studies investigating the echocardiographic measurement of cardiac output compared with thermodilution technique using the Bland–Altman method. Two authors independently reviewed the studies and extracted data on type of measurements, clinical setting and characteristics, and those of the Bland–Altman and trending ability analyses. We identified 13,834 citations and included 24 studies in the final analysis. The median number of participants was 32 (range 8–65). Most of the studies assessed left-sided heart structures and the majority had small bias, wide limits of agreement, and high percentage error between echocardiography and thermodilution. In only two of the 24 studies the precision of each technique (echocardiography and thermodilution) was assessed before comparing them. In the single study evaluating trending ability using valid methodology, agreement was observed between echocardiography and thermodilution in detecting the directional changes in cardiac output, but the magnitude of changes varied considerably. The majority of studies comparing echocardiography with thermodilution were difficult to interpret, but current evidence does not support interchangeability between these techniques in measuring cardiac output. The techniques may be interchangeable in tracking directional changes in cardiac output, but this has to be confirmed in large high-quality studies.