Non-invasive estimation of cardiac output in critical care patients

Objective. This study was carried out to compare cardiac output measurement s determined by thermodilution and by Portapres, a non-invasive system. Des ign, patients and setting. Eighty-seven non-invasive blood pressure measure ments were performed in 46 patients in our critical care unit utilising the new, non-invasive Portapres system. Cardiac output values were obtained fr om these blood pressure values using an aortic impendance model and compare d to cardiac output values estimated by the thermodilution technique. Measu rements and main results. Statistically significant (p < 0.01) differences (2.3 l/min; limits of agreement +/- 5 l/min) were noted between invasive an d non-invasive cardiac output measurements. Differences in measured cardiac outputs increased for patients receiving catecholamine therapy, in patient s with hemodynamic instability (e.g., sepsis and cardiac insufficiency), in patients with artificial ventilation, in patients with long duration of in tensive care, in younger (< 60 yr) patients and in women. We found no influ ence of the body mass index (BMI) on the accuracy of Portapres results. In only one single subgroup, 10 patients with pulmonary diseases, Portapres me asurements were not statistically significant different from reference resu lts. Conclusions. To date, Portapres measurements cannot replace thermodilu tion cardiac output estimations. Fluctuations of finger arterial perfusion due to hemodynamic instability, hypothermia and catecholamines may be respo nsible for problems of Portapres use in critically ill patients.