Prognostic significance of MRI-derived lung water density in cardiac amyloidosis

Background

Lung Water Density (LWD), quantified by thoracic magnetic resonance imaging (MRI), is a validated non-invasive marker of pulmonary congestion. A cut-off of >21.2% was previously established and externally validated in heart failure (HF). However, its prognostic value in cardiac amyloidosis, characterized by restrictive physiology, remains unclear.

Aims

To evaluate the prognostic value of LWD in cardiac amyloidosis.

Methods

We performed a retrospective study of patients with cardiac amyloidosis (TTR – transthyretin amyloidosis; and AL – light-chain amyloidosis) who underwent standardized thoracic MRI between 2018 and 2025. Quantitative LWD was calculated using a validated equations as the lung-to-liver ratio multiplied by 0.714. The intensity signal from the region of interest of the lung and liver were obtained from HASTE sequences. Patients were stratified according to the established LWD >21.2% HF threshold. The primary endpoint was a composite of all-cause death or HF hospitalization. The main secondary endpoints included 30-month survival and composite event-free survival. Kaplan–Meier curves were truncated at 30 months, and Cox proportional hazards regression assessed the prognostic impact of LWD. LWD was also analyzed as a continuous variable, expressed as the percentage (per 1% increment) above the upper limit of normal (>21.2%) and imputed as zero within the normal range (0–21.2%).

Results

A total of 101 patients were included (mean age 81 ± 8 years; 87% male; 93% ATTR). 20 patients had elevated LWD (median 26%, interquartile range [IQR] 23–28) and 81 had LWD <21.2% (median 15%, IQR 12–18). Baseline characteristics were similar between groups, except for left ventricular ejection fraction (LVEF), which was higher in patients with low LWD (48 ± 11 vs. 40 ± 17, p=0.020).

During 30 months follow-up, 29 patients died and 17 were hospitalized for HF. Patients with elevated LWD had significantly higher rates of the 30-month composite endpoint (68.8% vs. 16.0%, p<0.001). In the multivariable model, each 1% increase in LWD above 21.2% was independently associated with a significantly higher risk of the composite primary endpoint (HR 1.38 per 1% increase; 95% CI 1.10–1.75; p = 0.006), adjusted for NYHA class, clinical congestion, and LVEF. Multivariable models with LWD as a continuous variable, imputing null values to LWD within normal range, and models combining LWD with other variables of interest – including amyloid subtype (ATTR-CM vs. AL-CM) - yielded consistent results.

Conclusion

Elevated MRI-derived lung water emerged as a powerful independent predictor of adverse outcomes in patients with cardiac amyloidosis, strongly associated with reduced survival and a substantially higher incidence of cardiovascular events. These findings support LWD as a clinically meaningful, objective biomarker of congestion and risk stratification in cardiac amyloidosis.Central figureFor image description, please refer to the figure legend and surrounding text.