An acute heart failure risk calculator integrating comorbidities and gut-derived biomarkers to stratify risk of rehospitalisation and death has been developed and validated by a multinational research team.
Led by investigators from the University of Leicester and collaborating European centres, the team aimed to address limitations of existing heart failure risk scores, which largely neglect the contribution of non-cardiac comorbidities and the emerging role of the gut–heart axis.
According to the British Heart Foundation, which part funded the research, there is growing evidence that diet-influenced gut microbes can impact heart health by causing inflammation in the body.
Published in the European Journal of Preventative Cardiology, the analysis combined two independent cohorts: a single-centre UK cohort of 806 patients admitted with acute heart failure between 2004 and 2011, and 1,265 inpatients from the multicentre BIOSTAT-CHF cohort recruited across 12 European countries.
The pooled population (n = 2,071) was randomly split into training (70%; n=1,444) and validation (30%; n=627) datasets. Patients had a median age of 74 years, 68% were male and 80% were in New York Heart Association (NYHA) class III/IV.
The primary endpoint was a composite of all-cause mortality or heart failure rehospitalisation at one year, occurring in 36% of patients.
Understanding the gut–heart axis
Using backward logistic regression, the investigators derived an 11-variable model incorporating age, prior heart failure hospitalisation, NYHA class, NT-proBNP, diastolic blood pressure, creatinine, diabetes, chronic obstructive pulmonary disease (COPD), loop diuretic use, beta-blocker non-use, and combined gut microbiome metabolites.
Several gut-derived biomarkers were measured, including gamma‑butyrobetaine and acetyl-L-carnitine, which tend to rise in people who eat a Western‑style diet high in red meat and eggs.
Importantly, these biomarkers were independently associated with the primary outcomes, with each incremental rise in gut metabolite levels linked to an approximate 8% increase in the risk of death or heart failure rehospitalisation (odds ratio 1.08; 95% CI 1.03–1.13; P < 0.001), highlighting the prognostic contribution of the gut–heart axis.
The model demonstrated moderate discriminatory performance, with a C-statistic of 0.71 in the training cohort and 0.70 in validation (P < 0.001). This level of accuracy is similar to current risk calculators.
A simplified scoring system stratified patients into low-, mid- and high-risk groups, with corresponding event rates of 17%, 40% and 72% in the training cohort, and clear separation confirmed in validation analyses.
Compared with established risk models, the new score showed improved net reclassification, particularly versus older tools such as ADHERE and GWTG-HF.
The inclusion of gut-derived metabolites, reflecting the gut–heart axis, provided additional prognostic value beyond conventional clinical and biochemical markers.
‘Powerful influence’ of the gut microbiome in heart failure
The researchers noted that the benefits of eating a balanced and varied diet go beyond maintaining a healthy weight and cholesterol, adding that eating foods that feed ‘good’ gut microbes could prove to be an effective way maintaining heart health for cardiovascular patients.
Professor Toru Suzuki, professor of cardiovascular medicine at the University of Leicester, who led the study, added: ‘This study provides clear further evidence of the powerful influence the gut microbiome has over cardiovascular health. Our research has now found a way to harness that power, improving the tools doctors use to predict risk when treating seriously ill patients.
‘We hope the new calculator can be adopted in management guidelines for heart failure within the next few years, giving doctors the ability to better support their patients and save lives.’
Further validation required
The researchers also acknowledged several limitations, including the fact that the cohorts were historical and heterogeneous, with differences in enrolment periods and healthcare settings.
Dichotomisation of variables may have reduced statistical power and limited generalisability of cut-offs, and as the model was primarily derived from inpatient populations, it was potentially biased towards more severe heart failure.
Furthermore, variability in predictive performance across subgroups such as COPD and diabetes suggested that disease heterogeneity may influence accuracy.
Nevertheless, the researchers concluded that the model could support clinical decision-making by enabling more precise risk stratification in acute and worsening heart failure.
They highlighted the need for further validation in broader and outpatient populations, as well as refinement across heart failure subtypes, to enhance generalisability and clinical implementation.
Professor Bryan Williams, chief scientific and medical officer at the British Heart Foundation, said: ‘Research continues to reveal a potential link between gut health and heart health. This study evaluated whether the assessment of gut health might help predict patient outcomes if they have heart failure. It also raises the intriguing question as to whether taking action to improve gut health might help improve the outcome of patients with heart failure. This would require further research.’
Reference
Israr MZ et al. Risk calculator of multimorbid risk of rehospitalisation and death from heart failure: including the contribution of the gut microbiome. Eur J Prev Cardiol 2026 Feb 3;33(2):202-211.