Zdravniški Vestnik (Dec 2001)
EVALUATION OF CONTINUOUS THERMODILUTION METHOD FOR CARDIAC OUTPUT MEASUREMENT
Abstract
Background. Continuous monitoring of haemodynamic variables is often necessary for detection of rapid changes in critically ill patients. In our patients recently introduced continuous thermodilution technique (CTD) for cardiac output measurement was compared to bolus thermodilution technique (BTD) which is a »golden standard« method for cardiac output (CO) measurement in intensive care medicine.Methods. Ten critically ill patients were included in a retrospective observational study. Using CTD method cardiac output was measured continuously. BTD measurements using the same equipment were performed intermittently. The data obtained by BTD were compared to those obtained by CTD just before the BTD (CTD-before) and 2–3 minutes after the BTD (CTD-after). The CO values were divided into three groups: all CO values, CO > 4.5 L/min, CO < 4.5 L/min. The bias (mean difference between values obtained by two methods), standard deviation, 95% confidence limits and relative error were calculated and the linear regression analysis was performed. t-test for pared data was used to compare the biases for CTD-before and CTD-after for an individual group. The p value of less than 0.05 was considered statistically significant.Results. A total of 60 data triplets were obtained. CTD-before ranged from 1.9 L/min to 12.6 L/min, CTD-after from 2.0 to 13.2 L/min and BTD from 1.9 to 12.0 L/min. For all CO values the bias for CTD-before was 0.13 ± 0.52 L/min (95% confidence limits 1.17–0.91 L/min), relative error was 3.52 ± 15.20%, linear regression equation was CTD-before = 0.96 × BTD + 0.01 and Pearson’s correlation coefficient was 0.95. The values for CTD-after were 0.08 ± 0.46 L/min (1.0–0.84 L/min), 2.22 ± 9.05%, CTD-after = 0.98 × BTD + 0.01 and 0.98 respectively. For all CO values there was no statistically significant difference between biases for CTD-before and CTD-after (p = 0,51). There was no statistically significant difference between biases for CTD-before and CTD-after for CO > 4.5 L/min and CO < 4.5 L/min (p = 0.90 and p = 0.24 respectively).Conclusions. The continuous thermodilution cardiac output measurement shows clinically acceptable agreement with the bolus thermodilution technique in a wide range of CO values. There was no statistically significant difference between biases for CTD values before and after internal recalibration. CTD offers a reliable method of cardiac output monitoring in critically ill patients
