2004 OPEN FORUM Abstracts
NON-INVASIVE CARDIAC OUTPUT MONITOR (NICO) vs. BOLUS THERMODILUTION CARDIAC OUTPUT (TDCO), FICK TECHNIQUE, AND CONTINUOUS CARDIAC OUTPUT (CCO)
Desjardins, RRT; Sally Whitten, RRT, Virginia Eddy, MD, Stephen
Prato, MA, Maine Medical Center, Portland, ME
Introduction: Cardiac Output (CO) measurements provide valuable information for the management of patients in the critical care setting. Bolus thermodilution cardiac output (TDCO) measurements have been available for over twenty years. Continuous cardiac output measurements (CCO) via the pulmonary artery catheter have seen more recent clinical application. Currently, technological advances make such analysis possible with both a continuous and non-invasive method. NICO (Novametrix Medical Systems, Wallingford, CT) is a continuous, non-invasive cardiac output monitor based on the partial CO2 rebreathing indirect technique. The ratio of the change in end-tidal CO2 (ETCO2) and CO2 excretion (VCO2), in response to a 50-second period of rebreathing, is used to calculate cardiac output. This study compared the measurement of CO using the NICO rebreathing method to the currently used intermittent bolus (TDCO) and continuous cardiac output (CCO) techniques. Traditional oxygen-Fick technique (FICK), utilizing bedside indirect calorimetry with an arterial blood gas sample, was also compared.
Method: TDCO measurements were obtained from 30 critically ill, mechanically ventilated patients in the Intensive Care Unit (ICU) and Coronary Care Unit (CCU). Abbott Critical Care Systems (Abbott Laboratories, North Chicago, IL) pulmonary artery catheters were inserted according to clinical guidelines and, when required for CCO measurement, were attached to the Abbott CCO computer. NICO Capnostat CO2 rebreathing sensor was placed at the wye of the ventilator circuit, connected to the NICO computer, calibrated to manufacturer’s specifications and left in position for a minimum of 10 minutes. TDCO from the Abbott COM-2 Cardiac Output Computer was obtained via injection of 10 cc of 5% dextrose at room temperature (23-25oC). The recorded value for TDCO was derived from the mean of at least three injections that were within 5% of each other. Following TDCO, if available, a CCO measurement was recorded. NICO results were obtained and recorded. Comparisons between NICO and TDCO were obtained in 30 patients, while 18 patients were compared between NICO and FICK and 11 patients between NICO and CCO. Patients with known valvular disease were excluded from this study.
Results: Cardiac output measurements (for all three comparisons) ranged from 2.8 to 10.l L/min. Mean TDCO measurement was 6.0 ± 1.8 L/min. and mean NICO measurement was 5.5 ± 1.6 L/min., equal to a mean difference (bias) of minus 0.5 L/min., (pRegression=0.0001, R2=.42). Mean FICK measurement was 5.8 ± 1.5 L/min. compared to mean NICO of 5.4 ± 1.5 L/min., for a mean difference of - 0.4 L/min., (pRegression =0.0006, R2=0.53). Mean CCO was 6.2 ± 1.9 L/min., mean NICO was 5.6 ± 1.8 L/min., for a mean difference of - 0.6 L/min., (pRegression =0.02, R2=0.44). In all three cases, NICO generated CO measurements that were lower than the comparative technique by an average of 400 to 600 mL/min.
Conclusion: In the intensive care population, the method of determining cardiac output, whether by TDCO, FICK, CCO or by the new technique of NICO, yields statistically acceptable similarities between results.