Validation of cardiac index measured by four-dimensional cardiac magnetic resonance flow against the invasively measured cardiac index in patients with pulmonary hypertension

The purpose of this study was to validate cardiac index (CI) measured noninvasively by four-dimensional (4D) and two-dimensional (2D) cardiovascular magnetic resonance (CMR) flows against the invasively measured CI by right heart catheterization (RHC) in patients with pulmonary hypertension (PH). Thirty patients with PH (mean age: 32 +/- 10 years) were included. 4D and 2D flow measured CI within 24 h from RHC measured CI invasively. Qualitative analysis of 4D pulmonary flow (vortex presence and eccentricity of flow) was performed.All patients had helical right-sided flow with vortex formation; the mean vortex diameter was 29 +/- 7 mm, occupying 69% of the main pulmonary artery (MPA) lumen. MPA was dilated (42 +/- 9 mm). Mean CI measured by 4D flow CMR was closer to mean CI measured invasively (indirect Fick method CI = 2.1 +/- 0.8 L/min/m(2) vs. PA 4D flow = 2.3 +/- 0.7 L/min/m(2) "bias 0.22 +/- 0.25 L/min, p = 0.001," and Ao 4D flow = 2.3 +/- 0.7 L/min/m(2) "bias 0.2 +/- 0.28 L/min, p = 0.001"), while 2D flow had a higher mean CI (PA 2D flow = 2.5 +/- 0.7 L/min/m(2) "bias 0.45 +/- 0.7 L/min, p = 0.001" and Ao 2D flow = 2.5 +/- 0.8 L/min/m(2) "bias 0.45 +/- 0.67 L/min, p = 0.001"). The correlation coefficients among the different comparisons of CI showed: a low correlation between the 2D flow-indirect Fick method (Ao r(2) = 0.37, PA r(2) = 0.32) and a high correlation between the 4D flow-indirect Fick method (Ao r(2) = 0.86, PA r(2) = 0.89).There is an excellent agreement between CI measured by 4D flow and CI measured invasively. 4D flow, a noninvasive imaging technique, could accurately measure CI better than the conventional 2D flow in patients with PH.