A clinical comparative study of the TruCCOMS and the Vigilance continuous cardiac output monitors in cardiac surgery

Abstract

The Vigilance cardiac output monitor averages cardiac output over several minutes which may critically delay appropriate treatment. The TruCCOMS has been shown experimentally to provide near instantaneous detection of changes in pulmonary blood flow. In the TruCCOMS, blood flow is derived from the power required to maintain a 2° gradient between proximally and distally placed thermistors. The aim of this study was to compare the performances of these monitors with the thermodilution method. After Local Research Ethics Committee approval and informed consent, 30 patients undergoing cardiac surgery were randomised to TruCCOMS (First Generation 2.3, Aortech, Scotland) or Vigilance (VGS2, Baxter, USA) groups, requiring placement of a TruCCOMS or Baxter pulmonary artery (PA) catheter respectively. The time taken for continuous cardiac output to be displayed on the monitor after activating this mode was recorded at two time points: (i) after insertion of the PA catheter and (ii) immediately after separation from cardiopulmonary bypass. Six sets of continuous and thermodilution cardiac output were acquired at defined times: two before the start of surgery, two during surgery and two postoperatively. Haemodynamics were also recorded. Continuous outputs were downloaded every 10 s on to a laptop. Statistical analysis was with paired t-test, Pearson correlation coefficient and Bland-Altman analysis for assessing agreement between two methods of clinical measurement. No problems were associated with the use of either catheter. One TruCCOMS catheter failed and was replaced during the study. Mean (SD) time to display continuous cardiac output before surgery was 86.7 (11.5) s for the TruCCOMS and 224.7 (73.7) s for the Vigilance (p < 0.05); after separation from bypass these times were 86.9 (13.7) s and 525.6 (176.9) s respectively. Six sets of measurements were lost through protocol violations leaving 174 sets for analysis. Correlation coefficients were 0.312 (p = 0.003) for the TruCCOMS and 0.781 (p < 0.01) for the Vigilance. Bland Altman analysis showed the precision (2 × SD of differences) was 2.2 l.min−1 for both monitors; the bias (mean difference) for the TruCCOMS was 0.16 l.min−1 with 95% limits of agreement −2.06 to 2.37 l.min−1 and for the Vigilance it was 0.29 (−1.9 to 2.5) l.min−1. The TruCCOMS displayed the cardiac output faster than the Vigilance and was unaffected by recent separation from bypass. In contrast, the Vigilance was significantly delayed. This may be due to afterdrop in blood temperature. The Vigilance showed a better correlation with thermodilution than the TruCCOMS, although the bias and precision of the two monitors were comparable. In this study, the TruCCOMS overestimated low cardiac outputs and underestimated high outputs. Further studies are required to confirm these findings as previous investigators have found better correlations [1]. This study was supported by Aortech, Scotland.