Comparison between cardiac output measured by thermodilution technique and calculated by O2 and modified CO2 Fick methods using a new metabolic monitor.

Objective: To calculate cardiac ouptut from dual oximetry with carbon dioxide production (VCO2) and oxygen consumption (VO2) measured by a new metabolic monitor, and to compare these values with measurements made simultaneously using the thermodilution method during the steady state condition.
Design: Prospective, comparative clinical study.
Setting: The adult postsurgical intensive care unit (ICU) of a University Hospital.
Patients: Twenty mechanically ventilated postsurgical patients (70.7 +/- 7.8 years of age; range 50-84).
Measurements and Results: A new metabolic monitor (Puritan-Bennett 7250, Carlsbard, USA) connected to a ventilator (Puritan-Bennett 7200) was used to measure VCO2 and VO2. Measurements of arterial (SaO2) and mixed venous (SvO2) oxygen saturations were made using pulse and venous fiberoptic oximeters. Cardiac output starting from VCO2 (COVCO2) was obtained according to Mahutte's formula: COVCO2 = VCO2/[k (SaO2-SvO2)], where k represents a constant. The value for each patient was determined from the initial measurements of thermodilution cardiac output (COtd), VCO2, SaO2 and SvO2. COVCO2 calculated from the previous equation was compared to the COtd. Cardiac output calculated from the traditional O2 Fick equation (COVO2) was compared to the COtd. All patients were studied over a period of 120 min at 15-min intervals in reasonably stable conditions. COVCO2 was closely related to COtd (r = 0.94; SEE = 0.79; p = 0.0001; n = 180) with a bias of -0.10 and a precision of 0.45 l/min. The mean percent difference between the two methods was -2.2 +/- 8.3%. COVO2 was related to COtd (r = 0.77; SEE = 0.79; p = 0.0001; n = 180) with a bias of -0.57 and precision of 0.86 l/min. The mean percent difference between the two methods was -10.8 +/- 16.0%.
Conclusions: In stable patients, cardiac output measurements obtained from dual oximetry with VO2 and VCO2 measured by this new metabolic monitor, show good correlation with measurements made using the thermodilution method. The values of cardiac output calculated from VCO2 are more accurate and precise than values from VO2. The validity of these measurements in hemodynamically unstable patients and during various modes of mechanical ventilation seems warranted.