Multi-omics data improves one-year mortality prediction in acute heart failure

Abstract

Background

Acute heart failure (AHF) is a complex high-mortality condition, with risks escalating as age increases. Due to the complexity and heterogeneity of AHF, the development of effective prognostic indicators remains challenging. This study aims to assess the prognostic ability of plasma metabolites, gut microbiota, together with clinical indicators, in predicting mortality risk in AHF patients.

Methods

Plasma trimethylamine N-oxide (TMAO) levels were quantified alongside routine clinical parameters. Non-targeted metabolomic profiling was performed, and gut microbiota composition was determined through 16S rRNA gene sequencing. The predictive power of potential biomarkers for 1-year mortality was identified and evaluated using Cox regression analysis and receiver operating characteristic (ROC) curves.

Results

The prognostic value of N-terminal pro-B-type natriuretic peptide (NT-proBNP) alone was shown to have an area under the curve (AUC) of approximately 0.7, while the predictive utility of TMAO was limited. The addition of age and BUN to NT-proBNP enhances prognostic accuracy in AHF. Metabolomic analyses disclosed a subset of 9 metabolites emerged as novel prognostic biomarkers independent of age, BUN, and NT-proBNP. Microbiomic analyses revealed that three differential genera were associated with prognosis in AHF patients. Furthermore, a compact prognostic biomarker panel including NT-proBNP, age, BUN, homoarginine, MHPG sulfate, N-acetylmethionine, methionine methyl ester, leucyl-alanine, proline-hydroxyproline, and Faecalibacterium was developed, achieving an AUC of 0.902 and Brier score of 0.120.

Conclusions

Multi-omics analyses unveiled novel prognostic metabolites and gut microbes associated with 1-year mortality in AHF. Integrating these biomarkers with clinical parameters provided a potential model for improving the prognosis of AHF.