Improved accuracy and precision of thermodilution cardiac output measurement using a dual thermistor catheter system.

OBJECTIVES To assess whether thermodilution cardiac output determination based on measurement of injectate temperature in vivo leads to more accurate and precise estimates and to study the influence of chilled injectate on test performance. BACKGROUND Cardiac output measurement via right heart catheterization is used extensively for hemodynamic evaluation in a variety of diagnostic, perioperative and critical care settings. Maximizing accuracy is essential for optimal patient care. METHODS This prospective study of 960 thermodilution cardiac output measurements was conducted using conventional and dual thermistor techniques. Specialized dual thermistor right heart catheters were constructed using a second thermistor positioned to measure injectate temperature in vivo just prior to entry into the right atrium. To eliminate interinjection variability, a custom set-up was developed that permitted output measurement using both techniques simultaneously. Both ambient temperature injections and cooled injections were investigated. RESULTS The dual thermistor technique demonstrated significantly less measurement variability than the conventional technique for both ambient temperature (precision = 0.41 vs. 0.55 L/min, p < 0.001) and cooled (precision = 0.35 vs. 0.43 L/min, p = 0.01) injections. Similarly, the average range of cardiac output values obtained during five sequential injections in each patient was less using the dual thermistor approach (1.05 vs. 1.55 L/min, p < 0.001). The use of cooled injectate reduced the mean error of the dual thermistor technique but actually increased the mean error of the conventional technique. Even with ambient temperature injections, injectate warming during catheter transit varied considerably and unpredictably from injection to injection (2 SD range = -0.22 to 5.74 degrees C). Conventional ambient temperature and cooled measurements significantly overestimated Fick cardiac output measurements by 0.32 and 0.50 L/min, respectively (p < 0.001). In contrast, dual thermistor measurements were statistically similar (-0.08 and -0.08 L/min, p = 0.34) to Fick measurements. CONCLUSIONS This new dual thermistor approach results in a significant improvement in both precision and accuracy of thermodilution cardiac output measurement.