To evaluate the clinical significance of mathematical coupling on the correlation between cardiac output and right ventricular end-diastolic volume (RVEDV) through measurement of cardiac output by two independent techniques.Design
Prospective, observational study.Setting
Surgical intensive care unit in a level 1 trauma center.Patients
Twenty-eight critically ill surgical patients who received mechanical ventilation and hemodynamic monitoring with a pulmonary artery catheter.Interventions
A pulmonary artery catheter designed to measure right ventricular ejection fraction (RVEF) and cardiac output by the intermittent bolus thermodilution (TDCO) method and continuous cardiac output by the pulsed thermal energy technique was placed. A computerized data logger was used to collect data simultaneously from the RVEF/TDCO system and the continuous cardiac output system.Measurements and Main Results
Two hundred forty-nine data sets from 28 patients were compared. There is statistical correlation between TDCO and continuous cardiac output measurements (r = 0.95, p < 0.0001) with an acceptable bias (−0.11 L/min) and precision (±0.74 L/min). The correlation was maintained over a wide range of cardiac outputs (2.3–17.8 L/min). There is a high degree of correlation between RVEDV and both TDCO (r = 0.72, p < 0.0001) and independently measured continuous cardiac output (r = 0.68, p < 0.0001). These correlation coefficients are not statistically different (p = 0.15).Conclusions
The continuous cardiac output technique accurately approximates cardiac output measured by the TDCO method. RVEDV calculated from TDCO correlates well with both TDCO and independently measured continuous cardiac output. Because random measurement errors of the two techniques differ, mathematical coupling alone does not explain the correlation between RVEDV estimates of preload and cardiac output.