Analysis of the accuracy of continuous thermodilution cardiac output measurement. Comparison with intermittent thermodilution and Fick cardiac output measurement.

Abstract

To evaluate the accuracy of cardiac output measurement obtained by a new continuous thermodilution cardiac output (CCO) pulmonary artery catheter compared to intermittent thermodilution (TCO) and the direct Fick method. Prospective open trial. University hospital, intensive care unit. 23 patients (15 surgical, 8 non-surgical) were monitored with the Intellicath pulmonary catheter. Cardiac output was evaluated by the three methods every 4 to 6h as long as the pulmonary artery catheter was necessary (8-96 h). The correlation coefficient between CCO and TCO was 0.92, no systematic bias was observed, and the relative error increased from 13.9% for a cardiac output of 21/min to 23.7% for an output of 101/min. When comparing CCO and Fick, the correlation coefficient was 0.89, no bias was detected, and the relative error increased from 20.4% for outputs of 21/min to 27.2% for outputs of 101/min. CCO provides clinically acceptable measurements. At high cardiac outputs, the difference with other methods increases and the results must be cautiously interpreted.