Met-HeFT found no significant cardiovascular benefit from metformin in heart failure

Metformin did not significantly reduce deaths or major clinical events in a randomized study of people with chronic heart failure and diabetes or elevated diabetes risk, according to results presented at the European Society of Cardiology congress on August 30, 2026. The finding matters for healthspan research because it directly tests whether a familiar metabolic treatment translates into fewer serious illnesses and deaths.

This retrospective analysis examines evidence available by September 6. The result concerns a defined heart-failure population; it cannot settle whether metformin affects aging in otherwise healthy adults.

What the trial tested

Met-HeFT was an investigator-initiated, double-blind, placebo-controlled trial conducted at 23 Danish centers within the DANHEART research program. Investigators randomized 940 participants with symptomatic chronic heart failure, a left ventricular ejection fraction of 40% or less, and diabetes, prediabetes or increased diabetes risk. Their mean age was 70, and 16% were women.

The statistical analysis plan, dated April 1, 2026, specified a primary comparison based on participants’ original randomized assignments. It measured time to the first qualifying event. Those events included death, hospitalization or urgent treatment for worsening heart failure, heart transplantation, implantation of a ventricular assist device, heart attack or stroke.

That endpoint gives the study particular relevance to healthy longevity: it measures serious clinical consequences. Changes in glucose metabolism alone would not establish a benefit on this outcome.

A neutral result with remaining uncertainty

Over a mean follow-up of 3.7 years, the primary endpoint occurred in 25.1% of participants assigned metformin and 23.4% assigned placebo. The ESC reported a hazard ratio of 1.10 and a p value of 0.48, indicating no statistically significant difference.

The numerically higher event rate with metformin does not establish that the drug caused harm. Equally, the trial did not demonstrate equivalence between treatments. A nonsignificant comparison leaves uncertainty about the true size and direction of an effect.

The American College of Cardiology’s conference report also described no significant effects on secondary outcomes, including all-cause death, unplanned heart-failure events, new diabetes diagnoses or the cardiac stress marker NT-proBNP. An accompanying analysis of seven randomized placebo-controlled trials likewise reported no significant cardiovascular benefit in established heart failure or ischemic heart disease.

That pooled analysis included Met-HeFT itself. It therefore provides broader context but should not be mistaken for an entirely independent replication of the new trial.

Who the evidence represents

Only about 10% of participants had established type 2 diabetes. Investigators attributed the limited enrollment partly to existing metformin use and reluctance to discontinue it for trial participation. This selection limits how confidently the result can be generalized to the wider population living with both diabetes and heart failure.

A separate conference assessment on Herzmedizin.de reported that recruitment and event accumulation were lower than planned. It also described participants as generally receiving extensive background heart-failure treatment. Those features matter: the study evaluated additional benefit in that treatment setting, and its ability to detect modest effects was limited.

Women were underrepresented, and the findings concern reduced-ejection-fraction heart failure. Extending the conclusion to other heart-failure types or healthy older adults would exceed the evidence.

The implication for healthspan claims

The defensible conclusion is that Met-HeFT did not establish additional cardiovascular protection in the population studied. It does not erase metformin’s glucose-lowering role or provide grounds for a general claim that the medicine is ineffective.

For longevity research, the lesson is methodological: plausible metabolic effects need confirmation against clinical outcomes in the intended population. These conference findings strengthen that distinction, while a full results publication would enable closer assessment of adherence, adverse events and subgroup estimates.