ABSTRACT
To assess whether a simplified cardiac magnetic resonance (CMR)-derived lung water density (LWD) quantification predicted major events in Heart Failure (HF). Single-centre retrospective study of consecutive HF patients with left ventricular ejection fraction (LVEF) < 50% who underwent CMR. All measurements were performed on HASTE sequences in a parasagittal plane at the right midclavicular line. LWD was determined by the lung-to-liver signal ratio multiplied by 0.7. A cohort of 102 controls was used to derive the LWD upper limit of normal (21.2%). The primary endpoint was a composite of time to all-cause death or HF hospitalization. Overall, 290 patients (mean age 64 ± 12 years) were included. LWD measurements took on average 35 ± 4 s, with good inter-observer reproducibility. LWD was increased in 65 (22.4%) patients, who were more symptomatic (NYHA ≥ III 29.2 vs. 1.8%; p = 0.017) and had higher NT-proBNP levels [1973 (IQR: 809-3766) vs. 802 (IQR: 355-2157 pg/mL); p < 0.001]. During a median follow-up of 21 months, 20 patients died and 40 had ≥ 1 HF hospitalization. In multivariate analysis, NYHA (III-IV vs. I-II; HR: 2.40; 95%-CI: 1.30-4.43; p = 0.005), LVEF (HR per 1%: 0.97; 95%-CI: 0.94-0.99; p = 0.031), serum creatinine (HR per 1 mg/dL: 2.51; 95%-CI: 1.36-4.61; p = 0.003) and LWD (HR per 1%: 1.07; 95%-CI: 1.02-1.12; p = 0.007) were independent predictors of the primary endpoint. These findings were mainly driven by an association between LWD and HF hospitalization (p = 0.026). A CMR-derived LWD quantification was independently associated with an increased HF hospitalization risk in HF patients with LVEF < 50%. LWD is a simple, reproducible and straightforward measurement, with prognostic value in HF.