Dr. Yancy on Recent Heart Failure Studies

Fishing for Signals of Benefit

Dr. Walton-Shirley: Exactly why do you think it worked? Is it just a matter of preload reduction? Could you do the same thing with an extra 40 mg of furosemide daily to prevent hospitalization?

Dr. Yancy: Let's get to that next threshold: Did it work, or did it not work? There is a signal -- maybe even a message -- that giving a low dose of spironolactone (an aldosterone antagonist) on top of the other indicated therapies, with already controlled blood pressure, (a very important proviso) of a possible benefit on hospitalizations.

There are a lot of ways to interpret this study. It's not a simple yes or no. If one looks at this and understands it in the context of the safety that was exhibited in giving these lower doses of spironolactone, then you can begin to make the argument that it is something that certainly won't make it to guidelines (that I can vouch for), but something that a practitioner might say, "Here is something that I can use in a patient population for whom I have nothing else." This trial informs my decision-making and says that a low dose of spironolactone -- independent of the diuretic effect -- may be beneficial, in terms of fibrosis, stiffness, and diastolic performance.

Dr. Walton-Shirley: Speaking of that, what do you think we will glean from the subset analysis, because as you pointed out, this is a heterogeneous group of patients. What do you think it is going to show us?

Dr. Yancy: I love the way that Drs. Pfeffer and Pitt portrayed this. The solid data are the clinical trial results that didn't meet the primary endpoint. You get on thin ice when you start looking at subgroups, and then you get on even thinner ice when you start looking at post hoc non-prespecified analyses.

But if you allow yourself to undergo what we might call a "thought experiment" and ask, "Is there something here that percolates to the top and gives us reason to have some optimism?" I think there is, because there was a group of patients who were outside of the Americas (North and South America) in Russia and Georgia, who seemingly had a very different disease state. Their placebo event rate was down in the 9%-10% range. The expected placebo event rate was 2 or 3 times that, so they had seemingly a very different kind of disease. Not only were the outcomes not seen at all in that group, which was about 45% of the patients, but it may have been a very different kind of disease. To be fair, you can't put that as a top-line result, but you can say it is tantalizing. And then if you look in the group that had the biomarker-positive assay for heart failure, we again see something that looks like a signal.

Here we go: We have a trend in the primary results, although not statistical significance. We have these geographic differences that look specifically like a benefit in the Americas, not just in reduced hospitalization but maybe even in mortality. That is very soft, and then we have this other component of a biomarker-positive phenotype, and maybe that too reflects something positive. You can say we are fishing, but you know what? For HFpEF, we need to do some serious fishing, because we have a clear unmet need.

Dr. Walton-Shirley: I thought it was interesting. I looked at the Russian sodium intake to see whether there was a difference, and they actually have a long way to go on sodium regulation, so it wasn't dietary intake that made that regional variation.

Hyperkalemia: Elephant or Chipmunk in the Room?

Dr. Walton-Shirley: Let's talk a little bit about the elephant in the room: hyperkalemia. There was a doubling of hyperkalemia incidence in this study in patients who were on treatment. How do you handle the patient who starts with a normal glomerular filtration rate (GFR) and a normal potassium level, when you throw in spironolactone on top of an angiotensin receptor blocker? When do you make your next follow-up laboratory assessment?

Dr. Yancy: That elephant in the room is more like a chipmunk, because even though we saw more hyperkalemia in TOPCAT from a safety analysis, if you look at all-cause hospitalization (not just heart failure hospitalizations), all-cause hospitalization was not different. If that hyperkalemia had been overwhelmingly important clinically, we should have seen a difference there. Secondarily, when you look at all-cause death in the safety analysis, it did not show a difference.

Why are we concerned about hyperkalemia? Because it drives people into the hospital, and perhaps even unfortunately causes death. In a very large trial -- 3000 patients -- there wasn't a signal.

It's not the elephant anymore, but it does prompt the necessity for surveillance and monitoring, and more important, for rational decision-making. If this is a patient whom you can't follow appropriately and longitudinally, then this is not the therapy for them. When the risk and benefits are this narrow -- that is, the benefits are presumptive and the risks are real -- you want to be pretty clear about the patient in whom you start this therapy. I want to be pretty confident that I can follow them longitudinally.

Comments

Commenting is limited to medical professionals. To comment please Log-in.

Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.