The Prognostic Value of Entropy from T1 Mapping in Pulmonary Arterial Hypertension

Scritto il 12/10/2026
da Ji Zhao

J Cardiovasc Magn Reson. 2026 Oct 12:102810. doi: 10.1016/j.jocmr.2026.102810. Online ahead of print.

ABSTRACT

BACKGROUND: The prognostic value of left ventricular (LV) entropy from T1 mapping in pulmonary arterial hypertension (PAH) remains unexplored.

STUDY DESIGN AND METHODS: This prospective study enrolled patients with PAH who underwent cardiovascular magnetic resonance (CMR) within 72h of right heart catheterization (RHC). Myocardial native T1, extracellular volume fraction (ECV) and entropy were calculated in mid-ventricular short-axis slice. The primary endpoint was defined as all-cause mortality or lung transplantation. Outcomes were further analyzed across groups stratified by survival status using Cox proportional hazards regression and Kaplan-Meier survival analysis.

RESULTS: A total of 312 patients were analyzed over a median follow-up of 36.10 months, with 59 deaths. Elevated LV entropy was positively correlated with log-transformed NT-proBNP, right ventricular (RV) end-systolic, end-diastolic volume indices, LV native T1 and ECV, and inversely correlated with RV ejection fraction (all P < 0.01). LV entropy did not differ significantly across etiological subgroups, whereas LV entropy without IPs was higher in congenital heart disease-associated PAH. Both LV entropy ≥8.60 and LV entropy without IPs ≥8.30 independently predicted all-cause mortality (HR, 2.207; 95% CI, 1.324-3.678; P = 0.002; and HR, 2.357; 95% CI, 1.405-3.955; P = 0.001, respectively) and were associated with reduced survival (both log-rank P < 0.005). Adding LV entropy improved ESC models (χ²: 55.79 to 60.61, C-statistic: 0.782 to 0.801) and ESC-CMR models (χ²: 35.31 to 41.77, C-statistic: 0.805 to 0.813).

CONCLUSIONS: Elevated LV entropy was related to disease severity, RV dysfunction, and adverse clinical outcomes, and served as an independent predictor of all-cause mortality in patients with PAH.

REGISTRATION: URL: https://www.

CLINICALTRIALS: gov; Unique identifier: ChiCTR1800019314 and ChiCTR1900025518.

PMID:42863762 | DOI:10.1016/j.jocmr.2026.102810