Cardiac MRI-derived mean right atrial pressure and its prognostic importance

Background

Right atrial pressure (RAP) is a key variable that cardiac MRI (CMR) cannot currently measure. We aimed to develop a model to estimate mean RAP (mRAP) using CMR and assess the prognostic value of CMR-derived mRAP in an independent patient cohort.

Methods

The derivation cohort consisted of patients investigated for heart failure symptoms with right heart catheterisation and CMR. Right atrial and ventricular CMR measurements were correlated with invasive mRAP to inform multivariable linear regression models incorporating patient characteristics. CMR-derived mRAP was tested as a predictor for clinical outcomes (lower-limb oedema, heart failure hospitalisation and all-cause mortality) on an independent cohort of patients receiving CMR. Both cohorts were derived from hospital registries.

Results

In the derivation cohort (n=672), invasive mRAP was >8 mm Hg in 56% of patients. Right atrial end-systolic volume (RAESV) had the strongest correlation with invasive mRAP (Pearson’s coefficient 0.58, p<0.01). RAESV was as accurate as more complex models for mRAP prediction (p>0.05). CMR-derived mRAP ≥10 mm Hg was better associated with outcomes than mRAP ≥8 mm Hg in the clinical cohort (n=101) with diagnostic power for peripheral oedema (area under the curve (AUC) 0.75, p=0.02) and heart failure hospitalisation (AUC 0.93, p<0.01). Kaplan-Meier analysis demonstrated elevated CMR-derived mRAP (≥10 mm Hg) was associated with reduced survival compared with mRAP <10 mm Hg (χ2=5, p=0.02) over a mean follow-up of 6.8 years.

Conclusion

mRAP can be estimated by CMR. Raised CMR-derived mRAP is predictive of lower-limb oedema, heart failure hospitalisation and all-cause mortality.