RESUMO
BACKGROUND: The analysis of the arterial pulse contour obtained by means of a non-invasive device (Finapres) seems to be an ideal method to measure cardiac output (CO). An individual calibration factor (Z) dimensionally equal to aortic impedence is the necessary pre-requisite to calculate CO by pulse contour analysis. To verify the reliability of non-invasive pulse contour method, we compared the COs measured from Finapres tracings with those measured from thermodilution method in Intensive Care patients. METHODS: In 9 patients undergoing cardiac and general surgery, CO was measured from thermodilution (COTD) within 24 hours of postoperative period (total of 67 measurements). During COTD measurements, Finapres tracings were recorded and then analysed to calculate CO by two different procedures. In the former (COA), Z was calculated from an algorithm which takes into account heart rate, mean arterial pressure and age of the patient. In the latter procedure (COB), Z was experimentally determined from the initial COTD measure and then updated for the hemodynamic conditions of the patient. RESULTS: COTD ranged between 3.5 and 9.5 L.min-1 (mean value 5.53 +/- 1.29 L.min-1). The mean difference between COTD e COA was 0.485 +/- 1.537 L.min-1 and the mean percentage error was 25.1 +/- 14.5%. The experimental determination of Z reduced the mean difference and the mean percentage error between thermodilution and Finapres method to--0.002 +/- 1.056 L.min-1 and 15.5 +/- 11.0%, respectively. The regression line between COTD and COB turned out to be: COB = 0.68 + 0.88.COTD (r = 0.73). CONCLUSIONS: The pulse contour analysis applied to Finapres tracing allows to calculate CO with reasonable accuracy in the intensive care patients. An initial experimental determination of Z is recommended to improve the accuracy of Finapres method.
