At the American Heart Association (AHA) virtual meeting, an analysis of the EMPEROR-Preserved trial showed that the SGLT2 inhibitor empagliflozin (Jardiance) reduced cardiovascular death or heart failure hospitalization by 17% among those with a left ventricular ejection fraction (LVEF) of at least 50%.
In this exclusive Ƶ video, study author Stefan Anker, MD, of the Charité Campus Virchow-Klinikum in Berlin, discusses the results of the trial and his takeaway message.
Following is a transcript of his remarks:
At the European Society of Cardiology meeting, we presented, for the first time, the main results and published them in the .
But many people have said, "well, you included the patients with ejection fraction above 40%, what about the patients that -- by the guidelines, by many other standards -- have 'true' heart failure with preserved ejection fraction, true HFpEF." And that is the patient with 50% and higher ejection fraction.
This analysis was presented at the American Heart Association late-breaking science session, and it's based on the fact that we actually randomized the patients into EMPEROR-Preserved by stratifying the patients according to exactly that cut point of 50% and higher as one group, and below 50% [as the other group], which in our trial was 41 to 49% ejection fraction. And so the stratified randomization allows for what I think is a valid ... group analysis.
In the presentation, we focused really on the group of 50% and higher, showed the baseline characteristics of the patients, and highlighted the fact that these are somewhat older than the patients with lower ejection fraction, on average 73 years, and have exactly 50% women in this cohort.
Otherwise, very similar as in the main trial. Half the patients were diabetic, half the patients with atrial fibrillation. And also interesting if you think about the group having ejection fraction of 50% and higher, these patients had treatment, for the most part, like HFrEF patients, ACE inhibitors and ARBs [angiotensin receptor blockers], to some smaller degree ARNIs [angiotensin receptor-neprilysin inhibitors], more than 80% had beta-blockers, and even MRAs [mineralocorticoid receptor antagonists], 33% in this cohort of patients. Many also have asked how many patients were on diuretics, and this is about 80%.
Now, when you look for the overall results of the study, I can report that the primary endpoint of the trial is also significant for the patients with 50% and higher ejection fraction. There's a 17% benefit, the P value is 0.024. So, there is a significance for the overall results for the subgroup of patients with true HFpEF. There is no significant interaction for this outcome for the patients with low ejection fraction, but the magnitude of benefit was somewhat bigger in those with lower ejection fractions.
There's also a significant benefit for the first hospitalization of heart failure, which is the key driver for this primary endpoint result being significant. For first hospitalization with heart failure, there was a 22% reduction. Cardiovascular death was 11% reduced, not significant with the hazard ratios crossing 1. But, overall, we can say that the results in this subgroup are similar to the results for the overall trial population.
If you look at, very quickly, for the patients 41 to 49%, they have a 29% reduction in the primary endpoint, they have a 42% reduction for the first hospitalization for heart failure, and a 43% reduction for recurring heart failure hospitalization.
Now, as you know, we have a discussion ongoing whether ejection fraction interacts with the results, and we have done for this particular presentation also a cut point analysis by 5% increments. And it's quite interesting to see that the group of patients with 65 to 69% have indeed a hazard ratio that is well above 1 for the primary outcome, it's 1.3. However, then the patients with more than 70% ejection fraction have a hazard ratio that is well below 1, it's 0.6. So we now think seeing these results in that kind of granularity that the ejection fraction issue maybe is one of random variation.
We don't know why the group that is 65 to 69% has this less good response. Otherwise, true HFpEF patients also benefit for quality of life, benefit also for the kidney function outcome, and for symptom results. And, if you wonder, the body weight loss for these patients is 1 kg, and there's only 1-mm mercury reduction in the blood pressure.
I could say now as a conclusion that the results are positive for this subgroup. We compared them also, particularly this subgroup, with other previous trials. So here we have for first events of cardiovascular deaths and heart failure hospitalization about a 17% reduction. In other previous trials like CHARM ... etc., including also PARAGON, it was 4 to 8%, so it's really at least twice as big and significant in itself in this group. It leads me to say empagliflozin works in HFpEF patients, regardless of how you define them. And I would suggest that maybe this should become a new paradigm for treatment in these patients.