Comparison of Repetitive Cardiac Output Measurements at Rest and End-Exercise by Direct Fick Using Pulse Oximetry vs. Blood Gases in Patients With Pulmonary Hypertension

Abstract

Background: Exact and simultaneous measurements of mean pulmonary artery pressure (mPAP) and cardiac output (CO) are crucial to calculate pulmonary vascular resistance (PVR), which is essential to define pulmonary hypertension (PH). Simultaneous measurements of mPAP and CO are not feasible using the direct Fick (DF) method, due to the necessity to sample blood from the catheter-tip. We evaluated a modified DF method, which allows simultaneous measurement of mPAP and CO without needing repetitive blood samples. Methods: Twenty-four patients with pulmonary arterial or chronic thromboembolic PH had repetitive measurements of CO at rest and end-exercise during three phases of a crossover trial. CO was assessed by the original DF method using oxygen uptake, measured by a metabolic unit, and arterial and mixed venous oxygen saturations from co-oximetry of respective blood gases served as reference. These CO measurements were then compared with a modified DF method using pulse oximetry at the catheter- and fingertip. Results: The bias among CO measurements by the two DF methods at rest was -0.26 L/min with limits of agreement of ±1.66 L/min. The percentage error was 28.6%. At the end-exercise, the bias between methods was 0.29 L/min with limits of agreement of ±1.54 L/min and percentage error of 16.1%. Conclusion: Direct Fick using a catheter- and fingertip pulse oximetry (DFp) is a practicable and reliable method for assessing CO in patients with PH. This method has the advantage of allowing simultaneous measurement of PAP and CO, and frequent repetitive measurements are needed during exercise. Clinical Trial Registration: https://clinicaltrials.gov/ct2/show/NCT02755259, identifier: NCT02755259.

Keywords: cardiac output; direct Fick; exercise; pulmonary hypertension; pulse oximetry.

Figure 1

Identity plot of cardiac output (CO) measurement methods. Relationship by regression, such as data points from rest and end-exercise; the dashed line represents the line of equality. The black solid line represents the regression line, while the dots stand for the corresponding measurements. (A) Direct Fick using pulse oximetry (DFp) vs. direct Fick using blood gases (DF): The regression line shows an R² of 0.92, a slope of 1.07 with an intercept of −0.50 L/min and no displacement over the line of identity over a broad range of CO values. (B) Intermittent thermodilution (TD) vs. direct Fick using blood gases (DF): The regression line shows an R² of 0.72, a slope of 1.07 with an intercept of −0.20 L/min and no displacement over the line of identity over a broad range of CO values.

Figure 2

Distribution of cardiac output (CO) measurements by different measurement methods at rest and end-exercise are shown as boxplots. The horizontal line represents the median, the box the 25th to 75th percentile, the whiskers 5th to 95th percentile, and the dots represent the individual values, including outliers. (A) CO measurement at rest: direct Fick using pulse oximetry (DFp) vs. direct Fick using blood gases (DF)*. (B) CO measurements at end-exercise: direct Fick using pulse oximetry (DFp) vs. direct Fick using blood gases (DF)*. (C) CO measurements at rest: intermittent thermodilution (TD) vs. direct Fick using blood gases (DF)*. (D) CO-measurements at end-exercise: intermittent thermodilution (TD) vs. direct Fick using blood gases (DF)*. *if the corresponding CO measurement was not available or missing, both measurements were excluded.

Figure 3

Bland-Altman plot of the comparison of cardiac output (CO) measurement methods by direct Fick using pulse oximetry (DFp) vs. direct Fick using blood gases (DF). Y-axis was calculated as the mean of the two methods being compared. The solid line represents no difference in the means by the two methods. The dot-dashed line around zero represents the measured difference of the means, while the dotted lines stand for the 95% CI. The broken lines above and under it, represent the limits of agreements. (A) Measurements at rest: the difference in the means (bias) was −0.26 L/min (95% CI, −0.62 to 0.09 L/min), with the lower and upper limits of the agreement being −1.92 and 1.39 L/min, respectively. (B) Measurements at end-exercise: bias was 0.29 L/min (95% CI, −0.08 to 0.67 L/min), with the lower and upper limits of the agreement being −1.25 and 1.83 L/min, respectively.

Figure 4

Bland-Altman plot of the comparison of cardiac output (CO) measurement methods by intermittent thermodilution (TD) vs. direct Fick using blood gases (DF). Y-axis was calculated as the mean of the two methods being compared. The solid line represents no difference in the means by the two methods. The dot-dashed line around zero represents the measured difference of the means, while the dotted lines stand for the 95% CI. The broken lines above and under it, represent the limits of agreements. (A) Measurements at rest: the difference in the means (bias) was −0.45 L/min (95% CI, −0.98 to 0.08 L/min), with the lower and upper limits of the agreement being −2.93 and 2.03 L/min, respectively. (B) Measurements at end-exercise: bias was 1.39 L/min (95% CI, 0.37–2.42 L/min), with the lower and upper limits of the agreement being −2.84 and 5.62 L/min, respectively.