Photo Credit: Rasi Bhadramani

4D flow MRI may be a viable, non-invasive alternative to right heart catheterization for measuring blood flow in patients with pulmonary arterial hypertension.

Authors of a recent study in Diagnostic and Interventional Imaging found a strong correlation between 4D cardiac MRI and the direct Fick method for measuring pulmonary blood flow in patients with pulmonary arterial hypertension associated with congenital heart disease (PAH-CHD), suggesting that 4D flow MRI could be a viable, non-invasive alternative to right heart catheterization.

Clinicians typically perform right heart catheterization using the gold-standard direct Fick method to evaluate cardiac output, disease severity, and shunt correction eligibility in patients with PAH-CHD. However, Estibaliz Valdeolmillos, MD, and colleagues noted that this technique is complex, must be performed by specialists, requires dedicated equipment, and poses risks to patients.

“In day-to-day clinical practice, less accurate methods like thermodilution and the indirect Fick method, which estimates oxygen consumption, are inappropriately applied to patients with cardiac shunts. Furthermore, these invasive methods expose patients to risks such as bleeding, hematomas, arrhythmias, and infection,” the researchers wrote. “Therefore, there has been a significant push towards the development of non-invasive methods over the past few decades.”

The researchers explained that recent advancements have resulted in 4D cardiac flow MRI, which can provide accurate, reproducible data on flow patterns. Previous literature suggests that 4D flow MRI can support diagnosis and prognostication in patients with pulmonary hypertension and CHD. However, data are limited regarding its use in patients with PAH-CHD.

Therefore, the researchers conducted a prospective study comparing 4D flow MRI’s accuracy to that of right heart catheterization.

MRI Protocol & Analysis 

The study involved 64 adult patients with PAH-CHD referred for catheterization at a tertiary center for complex CHDs (Table). Fifty patients (78%) had pre-tricuspid shunts, 8 (12%) had post-tricuspid shunts, and 6 had complex shunts.

The investigators used a 1.5-Tesla MRI system with a 32-channel body coil. A senior radiologist used software supported by deep learning to analyze the images and assess pulmonary flow (Qs) and systemic flow (Qp). The researchers evaluated intra-observer and inter-observer reproducibility by having the same senior radiologist and a second radiologist repeat the measurements on 30 patients two months later.

Meanwhile, patients who were hemodynamically stable underwent right heart catheterizations, with clinicians using the Fick principle to measure Qp and Qs. 

Measurements Show Strong Correlations

There was a significant and excellent correlation between 4D flow MRI and right heart catheterization for measuring Qp (r=0.96; 95% CI, 0.94−0.98; P<0.001). The researchers conducted a Bland Altman analysis and found a minimal bias of 0.15 L/min (range, -0.81 to 1.11), which may indicate that 4D flow MRI slightly underestimated Qp.

Findings indicate a good correlation between the techniques for measuring Qs (r=0.85; 95% CI, 0.76−0.91; P<0.001). With a minor bias of 0.02 L/min (range, -1.37 to 1.41), 4D flow MRI may have marginally underestimated Qs.

In addition, pulmonary-to-systemic flow ratio (Qp/Qs) measurements strongly correlated between MRI and right heart catheterization (r=0.92; 95% CI, 0.88−0.95; P<0.001). The researchers found a bias of 0.08 (range, -0.42 to 0.59), “hinting at a negligible underestimation of Qp/Qs when employing 4D flow MRI.”

Evaluations for inter-observer reproducibility showed intraclass correlation coefficients (ICC) of:

• 97 (95% CI, 0.93−0.99) (mean difference=0.19; max difference=0.60) for Qp;
• 95 (95% CI, 0.92−0.97) (mean difference=0.24; max difference=0.70) for Qs; and
• 96 (95% CI, 0.92−0.98) (mean difference=0.09; max difference=0.31) for the Qp/Qs ratio.

Similarly, intra-observer reproducibility was associated with ICCs of:

• 98 (95% CI, 0.95−0.99) (mean difference=0.11; max difference=0.30) for Qp;
• 98 (95% CI, 0.95−0.99) (mean difference=0.10; max difference=0.20) for Qs; and
• 99 (95% CI, 0.95−0.99) (mean difference=0.06; max difference=0.15) for Qp/Qs ratio.

The study was limited by its single-center design, potential selection bias, small sample size, and the need for specialized training to perform 4D flow MRI. Despite these limitations, the researchers concluded that 4D flow MRI has significant potential as a non-invasive tool for pulmonary flow and shunt fraction estimation in patients with PAH-CHD. They recommended further research to validate its use in broader clinical settings.

“Despite some limitations, our findings underscore the excellent correlation and good agreement with traditional invasive cardiac catheterization methods. This innovative approach could greatly enhance patient comfort and safety by reducing the need for invasive procedures,” Dr. Valdeolmillos and colleagues concluded.