Hemodynamic Risk Assessment by Thermodilution and Direct Fick Measurement of Cardiac Output in Pulmonary Hypertension

Abstract

Background

Accurate measurement of cardiac output (CO) is critical in the evaluation and monitoring of pulmonary hypertension (PH). We assessed the accuracy of thermodilution (TD) CO vs direct Fick (DF) CO among patients with PH and evaluated whether the method of CO measurement affected diagnosis or risk assessment.

Research Question

Does using Thermodilution CO as compared to Direct Fick CO alter hemodynamic risk status in PH?

Study Design and Methods

We included patients who had undergone a right heart catheterization with both TD CO and DF CO measurements at University of California, Los Angeles between January 2021 and January 2023. Based on the cardiac index, patients were classified into low-, intermediate-, or high-risk hemodynamic status according to the European Society of Cardiology/European Respiratory Society guidelines.

Results

The analysis included 116 patients with PH. Of the patients, 55% were on PH therapy at the time of catheterization. The median age was 59 years (25th-75th percentile, 50-69), and 63% were female. The median TD CO and DF CO were 4.6 L/min (25th-75th percentile, 3.6-6.0) and 5.3 L/min (25th-75th percentile, 4.2-7.0) (P = .007), respectively. Bland-Altman analysis revealed a mean bias of −0.64 L/min. Median DF pulmonary vascular resistance and TD pulmonary vascular resistance were 4.7 Wood units (25th-75th percentile, 2.7-6.6) and 5.6 Wood units (25th-75th percentile, 3.0-8.0), respectively. Among patients with a low TD cardiac index, almost 40% had a preserved DF cardiac index. There was 78% agreement between DF and TD hemodynamic risk status. Using TD over DF reclassified 8% of patients with precapillary PH (n = 101) from low-risk into intermediate- or high-risk hemodynamic status. TD had a sensitivity of 97% for appropriately risk stratifying patients into intermediate-/high-risk status but a specificity of 73%. Overall, there was a strong correlation between DF CO and TD CO (concordance correlation coefficient, 0.81; 25th-75th percentile, 0.74-0.86).

Interpretation

Hemodynamic risk status was concordant between TD and DF measurements in almost 80% of patients. Oxygen consumption measurement should be considered if available on index right heart catheterization in patients with PH to aid in hemodynamic risk stratification or in whom strict pulmonary vascular resistance calculations are required.