Emperor - Preserved

I am, by nature, a late adopter in life. Whether it’s fads, tech or fashion I tend to let things roll for a bit to see how they pan out before considering my options (although my kids will argue I never actually adopt fashion at any point…). And, I tend to live by the same principles when it comes to prescribing new drug therapy - ‘first do no harm’ and await the evidence base is very much my mantra. However, I have surprised myself over the past couple of years that I’ve been genuinely quite excited about the increasingly wide indications for the SGLT2 inhibitor drugs. Initially, drugs used for type 2 diabetes, they showed impressive improvements in cardiovascular outcomes in people with type 2 diabetes, based on both RCT and observational evidence.

But life in SGLT2i land has moved on swiftly. Improvements in kidney disease and heart failure outcomes are increasingly apparent and we are seeing increasing evidence that these drugs improve outcomes not only in people with type 2 diabetes, but also in people without type 2 diabetes. The DAPA-CKD study and DAPA-HF studies showed improved outcomes in CKD and heart failure with reduced ejection fraction (HFrEF) for dapagliflozin (as well as mortality benefits) in people with and without type 2 diabetes, with the inevitable conclusions that these were ‘landmark’ studies. The EMPEROR-REDUCED trial showed similar results for empagliflozin.

So step forward (drum roll) the recently published EMPEROR-PRESERVED trial. Now, if this trial does what it says on the tin, it truly will be a ‘landmark’ study. Why has there been so much hype about it? All the heart failure trials so far have been focusing on people with heart failure with reduced ejection fraction (HFrEF), but this trial focussed on people with heart failure with PRESERVED ejection fraction (HFpEF). Most of the drug therapy over the years has focussed (with success) to improve outcomes for people with HFrEF (including mortality). But changes to management of acute coronary syndromes and the increase in cardiometabolic disease has led to a substantial change in the make up of people with heart failure. Now somewhere between 40-70% of people with heart failure have normal ejection fractions - this type of heart failure has a different pathophysiology from HFrEF leading to stiff ventricles and impaired relaxation driving the heart failure, rather than poor LV contractility. And currently we have no convincing data that any of the available drugs substantially improve outcomes for HFpEF, as highlighted by a Cochrane review earlier this year. 

But the EMPEROR-PRESERVED trial could change all that. This trial concluded that empagliflozin reduced the combined end point of cardiovascular death or hospitalisation from heart failure by >20% in people with HFpEF (and would be the first trial to do so). But, there is always a but….Remember the Emperors New Clothes? He gets hoodwinked into thinking he has been given a special set of new clothes that are invisible to those more educationally challenged (which he can’t see), when in fact he was sent on his way down the street starkers - all the townsfolk were too scared to say anything to risk looking stupid, until a child calls out the obvious truth. And indeed we can be at risk of getting swept up in the moment with these ‘landmark’ trials when there is clearly something missing.

In this case, the devil is more in the detail. Traditionally, (and indeed as defined by the ESC) HFpEF = ejection fractions ≥50%, HFrEF = ejection fractions of ≤40%, leaving a ‘grey’ area for people with ejection fractions of 41-49% (termed ‘mid range’ or ‘mildly reduced’). In EMPEROR-PRESERVED they included this group with ejection fractions between ~40-50%, when we know those at the lower end with reduced ejection fraction will get benefit, given previous evidence in people with HFrEF, which could skew the results to better outcomes in those with ‘preserved’ ejection fraction. And if you look at the subgroup analyses about a third of the participants had ‘mild’ reduced ejection fraction <50%, and indeed this group did well. But when you look at those with genuinely preserved ejection fractions of >50%, and certainly in those with ejection fractions >60% the benefits disappear. So is this truly a ‘landmark’ study for people with HFpEF? Well, I’m not so sure yet given the inclusion criteria, and I think I’ll go back to my late adopter strategy on this one.