Which management option for the treatment of hyperkalemia in patients with HFrEF has the least limitations?

• A. Low potassium diet
• B. Discontinuation or dose reduction of RAASi-based therapy
• C. Traditional potassium binders (sodium polystyrene sulphonate)
• D. Novel selective potassium binders (sodium zirconium cyclosilicate (SZC) or patiromer)

The correct answer is D.

[music]

Hello everybody, we're going to start by reviewing a case of a gentleman with heart failure and reduced ejection fraction, who has a scenario that's relatively common when he develops elevated potassium levels in a setting of optimizing guideline-directed medical therapy.

This is a 50-year-old gentleman who does not have prior history of cardiovascular disease but presents to the emergency department with progressive shortness of breath for the past several months. In particular of the past few weeks, the shortness of breath has been getting progressively worse and worse. In the past, he was asymptomatic. His BMI is 31, blood pressure is elevated, and on exam, he has rales bilaterally and also degree of low extremity edema. You can see his laboratory findings here, especially elevated NT-proBNP as well as impaired kidney function with the eGFR of 43. The serum potassium is mildly elevated at 5.3 millimoles per liter. He is on no current medications. The ECG is clearly abnormal, showing sinus rhythm and presence of left bundle branch block with wide QRS complex, and to the chest X-ray not surprisingly, we see evidence of congestion and volume overload consistent with heart failure. Here's echocardiogram, and we can clearly appreciate that it's markedly abnormal, left ventricle is dilated, and hyperkinetic ejection fraction is severely reduced to 20%. There is also biatrial dilatation and presence of moderate mitral and tricuspid regurgitation. Then subsequently with cardiac MRI, we can see no evidence of late enhancement, which excludes possibility of myocarditis. Again, just like an echocardiogram, left ventricle is dilated, diffusely hyperkinetic with severely reduced ejection fraction.

So, what do we do here? Of course, what needs to be done is starting the patient on guideline-directed medical therapy. Here is this patient three months later, the patient is now on sacubitril/valsartan at a reasonable dose of 49/51 milligrams twice daily, a good dose of carvedilol, patient is on loop diuretic, as well as SGLT2 inhibitor, and low dose spironolactone. Now, why low dose spironolactone? Because, as you recall, the potassium was already mildly elevated to the 5.3 millimoles per liter, and now after 12½ milligrams of spironolactone, potassium is now moderately elevated at 5.8 millimoles per liter, making first up-titration of MRAs quite challenging.

What do we do in a situation like this? Again, not uncommon. This is something that we probably should have expected, for the elevation on potassium level with initiation of spironolactone, and the patient already has mild hyperkalemia and significantly impaired kidney function with GFR in the CKD stage 3B category.

What do we do next? There are several therapeutic options here. The options that unfortunately is always that common is down-titration and discontinuation of renin angiotensin aldosterone system inhibitors, so is a discontinuation of spironolactone, lack of up-titration of spironolactone, or reduction in dose of sacubitril/valsartan as a reaction to this elevated potassium level. The alternatives are, of course, to try to get hyperkalemia treated. With that, we can use a low potassium diet, we can potentially think about increasing loop diuretics, we can think about adding a traditional potassium binder such as SPS, or sodium polystyrene sulfonate, or perhaps a novel potassium binder. There are other options as well, do nothing, which is probably not a great option, as well as maybe conferring with nephrologists to see what their recommendations would be. Which way do we go?

Let's just go through those management options. First of all, a low-potassium diet certainly could help here, but there is an issue with that, it's difficult to adhere to, patients don't like it, it takes a lot of foods that we normally recommend that our patients actually eat if they have evidence of heart disease out of consideration. Probably, the most important thing is that long-term efficacy is really suboptimal because of low adherence rates.

Diuretics, while the patient is already on a loop diuretic, should we just ramp up loop diuretics to try to bring the potassium levels down? That's not really a great option either because if the patient is decongested, effectively, which this patient now is after three months of GDMT up-titration, just giving diuretics to lower potassium, first of all, it's not very efficacious for potassium lowering itself. Volume deflation could potentially rev up neurohormones, and actually be counterproductive from a heart failure standpoint in the long term. Also, importantly, can actually lead to worsening kidney function.

We already have discussed that discontinuation or lowering the dose of RAASi is not a great option either because, of course, the reason we have patient on RAASi, to begin with, is because it saves lives and prevents hospitalizations in this patient population.

Then, traditional binders have a number of important limitations as well. In the short-term, sometimes, it's not an unreasonable option, long-term, efficacy has never really been evaluated. We have problems with long-term tolerability due to gastrointestinal side effects, and they are just generally not palatable, easy to take medications. Just to demonstrate that, if you look at a median duration of sodium polystyrene sulfonate treatment tends to be about seven days, again, for the reasons we just talked about.

Do we have other options? Of course, we do now with novel potassium binders. We have two potassium binders: patiromer, which is an organic polymer and sodium zirconium cyclosilicate, which is an inorganic crystal. We have efficacy and safety data for both of these potassium binder options.

First, let's talk about patiromer. We know that based on onset of duration study with patiromer is that it starts working after about four to seven hours. We know that when used in patients with mild to moderate hyperkalemia, it can normalize potassium levels within a few days, and then keep potassium levels well-controlled for up to four weeks. Then, in AMETHYST trial, using it long-term in patients mild to moderate hyperkalemia, again, it brings the potassium levels down in those both with mild or moderate degrees of hyperkalemia, and it can keep it down for a duration of as long as one year. It tends to be reasonably well-tolerated as well with a relatively small proportion of patients having gastrointestinal side effects.

Do we actually have data to suggest that use of patiromer can lead to better RAASi optimization in patients with heart failure and reduce dejection fraction? We actually do from the DIAMOND trial. DIAMOND was initially designed as a cardiovascular outcome trial but was heavily impacted by COVID pandemic, and due to low enrolment as well as low event rates, was ultimately changed, the protocol was amended, with the primary endpoint being potassium levels. The way the study was designed is that patients with either hyperkalemia at baseline or at high risk for it enter it in the open-label running phase where they were optimized on patiromer and MRAs. Then, there was a randomized withdrawal period where the patients either continue patiromer or were randomized to withdrawal to placebo.

The primary endpoint in terms of potassium levels was in favor of patiromer, but probably more important endpoints are looking at the time to hyperkalemic event. You can see that patiromer was clearly superior to placebo with a hazard ratio of 0.63 in terms of time to hyperkalemic event. Also, if you look at the time having to reduce the dose of an MRA, there was also a benefit of patiromer hazard ratio of 0.62. Interestingly enough, because lots of patients in the DIAMOND trial were not hyperkalemic at baseline, actually, majority of patients, even those patients that were in placebo were able to be successfully optimized on the MRA even without the need for potassium binder. Again, indicating that many patients out there that are not getting optimal MRAs could probably tolerate it even without the need for potassium binder, but certainly, potassium binder makes it even more likely to get those patients an optimal MRA dose.

What about sodium zirconium cyclosilicate, the inorganic crystal, with high affinity for potassium? We certainly have efficacy and tolerability data, where SZC now as well from a number of trials in the HARMONIZE trial, where patients initially enter it in the open-label correction phase. SZC actually significantly reduced potassium level within one hour. By the end of the running phase, vast majority of patients actually were able to get into the normokalemic range. In fact, it didn't matter if these patients had CKD, heart failure and diabetes, if they were or not on RAASi, and even those patients with severe hyperkalemia were able to successfully achieve reduction in potassium levels and they were able to get into the normokalemic range within a short period of time. In a randomized maintenance phase, SZC successfully maintained potassium levels in the normal range, there was actually an extension phase of the study where they went for up to 11 months of maintaining normal potassium levels. Once the drug was discontinued, the potassium levels went back up, not surprisingly. In the subsequent long-term tolerability study 005, which was a single arm, long-term, study looking at SZC, we see that, again, in patients with, mild and moderate hyperkalemia, average potassium levels of 5.6, SZC successfully reduced potassium levels into the normal range and maintained the normal range for up to one year. Again, once the medication was discontinued, potassium levels go back up.

Do we actually have data that using SZC can optimize RAASi? We're working on the trial right now, the trial is called REALIZE-K. In REALIZE-K, patients that have hyperkalemia at baseline or considered to be at higher risk enter into open-label phase. During that open-label phase, if they're hyperkalemic already, they get started on SZC and they're being optimized on the MRA. These patients are not on optimal MRA at baseline by study design. If those patients are at high risk, they actually get started on MRA up-titration with spironolactone and then once they develop hyperkalemia, they get SZC. That's the open-label phase. Once they're optimized on MRA and are normokalemic because of a potassium binder, they get randomized through the continuation of SZC or withdrawal to placebo. The primary endpoint is proportion of patients that are optimized on spironolactone and they are normokalemicwithout requiring rescue treatment. The results of safety, this about seven days afterwards, so we will know when the trial is completed whether SZC can be used successful in patients with heart failure and reduced ejection fraction in terms of optimizing GDMT, which we know improves patient outcomes, and of course we're going to be looking very carefully, at the tolerability and safety of SZC in this trial as well.

In summary, we know that hyperkalemia is common in patient populations that we reviewed in the beginning, like the patient case I presented. We know it limits optimization of RAASi and GDMT, which ultimately saves lives and prevents hospitalizations. The typical decision making, unfortunately, is to down-titrate or discontinue RAASi, but we know that's not the optimal solution for the patient, and leads to marked under use of GDMT in this patient group. Novel potassium binders really give us another option to optimize GDMT and try to make our patients, hopefully, live longer and feel better. As a result of that, the guidelines have acknowledged that novel potassium binders is a another treatment option that we should consider. We will have additional evidence regarding potassium binder use as a potential way of enabling optimal GDMT and RAASi in patients with HFrEF, and hopefully that will advance their adoption in clinical practice even further. Thank you very much.

Share this page with your colleagues and friends: