Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 17, 2026Molecular MedicineOpen Access

A 10-metabolite model predicts ~23-fold greater 4-year mortality in HFpEF for the highest tertile.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Accurate prediction of mortality risk in patients with HFpEF remains an unmet clinical need.

Does a 10-metabolite prognostic model improve the prediction of 4-year all-cause mortality compared to a clinical risk score in hospitalized Chinese Han patients with HFpEF?

Population

500 hospitalized Chinese Han patients with HFpEF from a national multicenter prospective cohort

Comparison

10-metabolite prognostic model vs clinical risk score

Design

Prospective cohort study with metabolomic and transcriptomic analyses

Follow-up

4 years

Key result

A 10-metabolite model predicted 4-year all-cause mortality in HFpEF, with the highest tertile having a 23-fold increased risk compared to the lowest tertile (HR 22.95; 95% CI 11.07-57.54).

Authors

YLYinchu LiYHYun HongJLJiapeng Lu

Discussion

Loading...

Member takes

Overview

May refine mortality prediction beyond clinical scores in HFpEF; hypothesis-generating and requires prospective validation before practice change.

Key Points

  • Develop and validate a plasma metabolite-based risk stratification model to predict 4-year all-cause mortality and identify associated biological pathways in patients with heart failure with preserved ejection fraction.
  • Analyzed plasma metabolomic profiles in 500 hospitalized Chinese Han patients with HFpEF from a national multicenter prospective cohort, randomly split 1:1 into training and validation sets.
  • Selected candidate metabolites using logistic regression, LASSO regression, and the Boruta algorithm, followed by validation and bootstrap analysis.
  • Integrated mRNA microarray data from 291 cohort patients using weighted gene co-expression network analysis (WGCNA) and multi-omics enrichment analysis to determine prognostic pathways.
  • In the validation set, the 10-metabolite model discriminated 4-year all-cause mortality with an AUC of 0.878 (95% CI 0.835–0.921), surpassing the clinical risk score (AUC 0.714, 95% CI 0.649–0.778).
  • The model significantly enhanced risk classification (NRI = 0.585, 95% CI 0.469–0.701, P < 0.001; IDI = 0.413, 95% CI 0.344–0.482, P < 0.001), with the highest risk tertile showing marked mortality risk compared to the lowest (HR = 22.95, 95% CI 11.07–57.54).
  • Multi-omics analysis revealed metabolic pathway alterations contributing to prognosis, notably glycerophospholipid metabolism, tyrosine metabolism, the TCA cycle, and AMPK signaling.

Study Design

Type

Cohort (n=500)

Multicenter

Yes

Structured PICO

Does a 10-metabolite prognostic model improve the prediction of 4-year all-cause mortality compared to a clinical risk score in hospitalized Chinese Han patients with HFpEF?

P
Population
500 hospitalized Chinese Han patients with heart failure with preserved ejection fraction from a national multicenter prospective cohort, followed for 4 years.
E
Exposure
10-metabolite prognostic model for risk stratification
C
Comparator
Clinical risk score
O
Outcome
4-year all-cause mortalityhard clinical

Main Result

Hazard Ratio: 22.95 (95% CI 11.07–57.54)

A novel 10-metabolite prognostic model significantly improves the prediction of 4-year all-cause mortality in hospitalized Chinese Han patients with HFpEF compared to standard clinical risk scores.

Limitations

  • Requires further validation in other populations before being used in routine clinical practice
  • requires further validation in other populations before being used in routine clinical practice

Cite This Study

Li et al. (2026) conducted a cohort in Heart failure with preserved ejection fraction (HFpEF) (n=500). Highest tertile of 10-metabolite score vs. Lowest tertile of 10-metabolite score was evaluated on 4-year all-cause mortality (HR 22.95, 95% CI 11.07-57.54). A 10-metabolite model predicted 4-year all-cause mortality in HFpEF, with the highest tertile having a 23-fold increased risk compared to the lowest tertile (HR 22.95; 95% CI 11.07-57.54).

synapsesocial.com/papers/6aabb75d5f706d05830e664dhttps://doi.org/10.1186/s10020-026-01644-9
View Full Paper
Ask AI
Bookmark
Share