Association of heart failure and liver T1 mapping in cardiac magnetic resonance imaging

Heart failure culminates in liver congestion. As the liver can be assessed by cardiac magnetic resonance T1 mapping sequences, it represents a promising imaging biomarker for heart failure.

To evaluate the association between heart failure and magnetic resonance liver T1 mapping, as a non-invasive imaging surrogate of liver congestion.

Patients with cardiac magnetic resonance including T1 mapping sequences on a single 1.5T scanner were consecutively included between 01/2017 and 12/2019. Liver T1 was measured in the left lobe, right lobe and caudate lobe on short axis view and in the liver dome on four chamber view. Echocardiographic and laboratory parameters were analyzed within 90 days of the cardiac magnetic resonance exam. Left ventricular filling pressures were assessed according to the European Association of Cardiovascular Imaging guidelines [1]. Liver T1 and echocardiographic parameters were compared between patients with and without elevated NT-proBNP levels (>125 pg/ml) using a Mann-Whitney U test. A multivariate logistic regression model was used to correct for age, sex, body mass index, left ventricular ejection fraction and the presence of atrial fibrillation.

A total of 397 patients were included (median age 56, 127 females), of whom 138 patients (35%) presented with elevated NT-proBNP levels. Patients with elevated NT-proBNP levels had a larger indexed end-diastolic left ventricular volume (92 vs. 81 ml/m², p<0.001), lower left ventricular ejection fraction (60 vs. 50%, p<0.001) and a higher E/E' ratio (11 vs. 8, p<0.001). Liver T1 was higher in patients with elevated NT-proBNP in the right liver lobe (670 vs. 596 ms, p<0.001) and the caudate lobe (664 vs. 598 ms, p<0.001), but not in the left liver lobe (571 vs. 568 ms, p=0.068) and in the liver dome (590 vs. 560 ms, p=0.1). Echocardiographic evaluation showed similar results with significant difference in T1 times between patients with elevated and non-elevated left ventricular filling pressures in caudate and right liver lobe. Liver T1 retained its predictive value when corrected for age, left ventricular ejection fraction and the presence of atrial fibrillation, when measured in the caudate lobe (adjusted odds ratio 1.013, 95% confidence interval 1.004–1.023, p=0.005) and in the right lobe (adjusted odds ratio 1.012, 95% confidence interval 1.003–1.021, p=0.009).

Elevated liver T1 in cardiac magnetic resonance imaging is associated with heart failure and represents an independent non-invasive imaging surrogate for liver congestion.

Type of funding sources: None.