Heart failure with preserved ejection fraction: STEP-HFpEF and the obese phenotype

Half of all heart failure is the kind where the heart squeezes normally but cannot fill properly. In people with obesity, that condition has a specific shape, and STEP-HFpEF was the first trial to treat the obesity rather than the heart.

The obese HFpEF phenotype

Obokata and colleagues compared HFpEF patients with and without obesity using invasive haemodynamics and imaging. The obese group had higher plasma volume, more epicardial fat, greater right heart enlargement, more pericardial restraint on filling and worse exercise capacity, with a haemodynamic profile that differed enough to justify calling it a distinct phenotype (PubMed 28381470). The heart is not failing because of a primary muscle disease; it is failing under a load that obesity imposes from several directions at once.

That framing matters for mechanism. Ventricular muscle does not carry the GLP-1 receptor in validated human mapping (PubMed 24467746). A drug that helps this phenotype is expected to work by removing load, not by acting on the myocardium, and the trial results are consistent with that.

The trial

529 adults with heart failure with preserved ejection fraction (LVEF 45% or higher), BMI 30 or higher and no diabetes were randomised to semaglutide 2.4 mg weekly or placebo for 52 weeks. The dual primary endpoints were change in the Kansas City Cardiomyopathy Questionnaire clinical summary score, a patient-reported measure of symptoms and physical limitation, and change in body weight.

The KCCQ-CSS rose 16.6 points on semaglutide versus 8.7 on placebo, a difference of 7.8 points (95% CI 4.8 to 10.9); a 5-point change is generally considered clinically meaningful. Weight fell 13.3% versus 2.6%. Six-minute walk distance improved by 20.3 m more on semaglutide (95% CI 8.6 to 32.1). C-reactive protein fell 43.5% versus 7.3%, and NT-proBNP, a marker of cardiac wall stress, fell more on semaglutide. Serious adverse events were fewer on semaglutide, 13.3% versus 26.7%, driven by fewer cardiac events (PubMed 37622681). A companion trial in people with diabetes, STEP-HFpEF DM, reported the same pattern.

What it shows and what it does not

It shows that treating obesity improves how people with this phenotype feel and function, with the effect size of a good heart failure drug on the symptom scale. The CRP and NT-proBNP changes are consistent with reduced load and inflammation, though neither proves the pathway.

It does not show reduced mortality or hospitalisation; the trial was too small and too short. The SELECT heart failure composite, hazard ratio 0.82 (95% CI 0.71 to 0.96) in a broader cardiovascular population (PubMed 37952131), is supportive but is not an HFpEF outcome trial. Whether the benefit depends on the amount of weight lost, or on inflammation changes that arrive faster, cannot be separated in these data; this site grades the mechanism D and the symptom effect A.

Compounded semaglutide was not studied, is not FDA approved and is not interchangeable with Wegovy.

Where to go next

• The STEP-HFpEF digest with the full endpoint table: FormBlends Research.
• The cardiovascular outcome trial that frames it: SELECT.
• The receptor argument for indirect cardiac benefit: GLP-1 receptor effects by organ.