Which Measurement Technique for Cardiac Output?

Abstract & Commentary

Synopsis: Simultaneous measurements of QT by thermodilution, Fick method, and TEE in critically ill patients varied widely, showing that these methods cannot be used interchangeably.

Source: Axler C, et al. J Crit Care 1996;11:109-116.

Axler et al at st. paul’s hospital, vancouver, BC, simultaneously measured cardiac output (QT) by three currently available techniques in 13 critically ill, mechanically ventilated patients, before and after the rapid infusion of 500 mL of saline. The techniques used were thermodilution, using a pulmonary artery catheter; the Fick method, in which oxygen consumption was determined using a metabolic monitor; and transesophageal echocardiography (TEE), in which QT was calculated by multiplying heart rate by the difference between left-ventricular end-diastolic and end-systolic volumes.

Several steps were taken to assure maximal patient stability during the period of measurement, and the means of three measurements at five-min intervals, before and after fluid administration, were calculated for each method.

Mean QT by the three methods ranged from 8.0 ± 3.4 to 8.4 ± 3.3 L/min, and were not significantly different from each other. However, although the results obtained with the three techniques correlated significantly (r2 = 0.40-0.42, P < 0.00001), these correlations were only moderately good, and the differences between them were substantial when assessed by the method of Bland and Altman (Bland JM, Altman DG. Lancet 1986; 1:307-310). The 95% confidence intervals for agreement between methods exceeded 4 L/min for some measurements in all three comparisons (e.g., Thermodilution vs Fick, Thermodilution vs TEE, TEE vs Fick).


This study shows that agreement among simultaneous measurements of QT by thermodilution, Fick, and TEE in critically ill patients receiving ventilatory support is poor, with variations in the values obtained lying well into the range of clinical significance. Values obtained by TEE and Fick methods had no better agreement than those by thermodilution as compared with either of the former, prompting the authors to conclude that all three techniques give variable results. Thus, at least for ICU patients such as those studied, there does not appear to be a single "gold standard" for QT.

Axler et al offer the following advantages and disadvantages for each of the available techniques for measuring QT. Thermodilution, the method most often used in critical care, is easy, fast, and can be repeated frequently, although it requires a pulmonary artery catheter and its results are operator-dependent. Thermodilution overestimates the true QT in the presence of low QT, and underestimates it with tricuspid regurgitation, right-to-left shunt, and major dysrhythmias.

The Fick method is also reasonably easy to use and has less operator dependency. It also requires a pulmonary artery catheter, plus an expensive metabolic cart or other oxygen monitor, which takes about 30 minutes to set up for use and is less and less accurate with increasing FIO2 above 0.6. Measuring QT meaningfully by the Fick method requires that the patient be stable metabolically and may be affected by pulmonary oxygen consumption.

Although less invasive than a pulmonary artery catheter, TEE requires an expensive echo machine and is highly operator-dependent. It yields additional diagnostical information, so that the physiological cause of altered QT may be discernible with the same procedure, but repeated QT measurements are less easily done than with either of the other methods.

Whichever method the clinician uses to measure QT in a critically ill patient, that method should also be used in assessing acute responses to changes in ventilator, fluid, or pharmacological therapy. None of the three methods can substitute for the integrated knowledge, experience, and judgment of a good clinician.