Please note that the text below is not a full transcript and has not been copyedited. For more insight and commentary on these stories, subscribe to the This Week in Cardiology podcast, download the Medscape app or subscribe on Apple Podcasts, Spotify, or your preferred podcast provider. This podcast is intended for healthcare professionals only.
In This Week’s Podcast
For the week ending March 14, 2025, John Mandrola, MD, comments on the following topics: Listener feedback, resistant hypertension, K-binders for MRA use in heart failure, nutritional epidemiology, and a positive study for vitamin D.
CAUGHT CAD
Last week, I strongly criticized the CAUGHT CAD trial. Mostly on methodologic grounds. The trial tested the use of CAC scoring to reduce plaque volume in intermediate risk patients with a family history of CAD. The trial was positive, but its trial procedures precluded making any reliable clinical judgments.
A Medscape journalist is working on a news piece. He reached out to me for comment regarding comments the senior author made about my appraisal lost week.
“I'm fascinated by the internet phenomenon of opinion propagation. There's no need to propose a solution to a problem, just to denigrate any approaches [other] than the status quo. Nothing will even change his mind, because an outcomes trial with the required 20-year time horizon will never be done.”
I am not sure this comment will make it in the news article, but I include it because it is a comment I have heard from academic researchers. In fact, the former FDA director, Robert Califf, once demeaned non-academic opinion writers on the internet as the peanut gallery, citing the Teddy Roosevelt quote, “The Man in the Arena.” Basically, shut up if you don’t do research.
Well, my response to this was to say that being a researcher or doing research is not a criteria for doing critical appraisal. Critical appraisal is most important for the users of evidence. That sentence is all that I wrote to the journalist, but let me add just a few more comments.
On the matter of the “internet and opinion propagation,” I regret to inform this senior researcher and others that the digital revolution — wherein ideas rise to prominence based on their merit rather than their source (say Harvard, Yale or ACC or AHA) — is more than a decade old. Gone are the days where letters to the editors or lofty discussions from KOLs at medical meetings are how ideas get established.
News flash: We now have an internet. And on this internet are, gulp, blogs and podcasts, and people with ideas. Many of these people are not PhDs or researchers but sometimes that is their strength.
Take late-breaking clinical trials at major meetings. The place to be is not in the audience listening to the congratulatory “trial discussant” but online where robust discussion happens.
My second response, to the comment, “Rather than proposing a solution, and nothing will ever change my mind, or that an outcomes trial will never be done with CAC because of the long time horizon,” I would remind everybody that large outcome studies for screening tests have been done: Axel Diedrichsen studied this in DANCAVAS, screening CT scans have been studied in lung cancer, colonoscopies in colon cancer, and mammography for breast cancer as well as aortic screening have also been submitted to the RCT.
A coronary CT scan, whether for coronary artery calcium or CT angiography, places a patient into the health system. That is no small thing. It should be tested systematically with outcomes. It is not tested because proponents like their place now, where marketing has made the test popular.
Tests that can induce downstream procedures or create fear should be considered potentially harmful as well as beneficial, and given the mountain of evidence against revascularization in stable patients, the CT scan should be randomized. There is a ton of waste coming out of places like NIH, but this is one area where a study could be — and should be — funded.
A New Hypertension (HTN) Drug for Patients With Resistant HTN
Lorundrostat is an aldosterone synthase inhibitor which works by inhibiting CYP11B2, the enzyme responsible for aldosterone production. The company that makes the drug Mineralys Therapeutics announced this week in a press release positive results of the phase 3 RCT called ADVANCE HTN.
Patients were already taking 2-5 blood pressure (BP) meds, one of which had to be a thiazide-like diuretic.
The trial noted a placebo-adjusted reduction of 9 mm Hg after 6 weeks. Patients whose doses were increased to 100 mg per day after 6 weeks had an absolute reduction of 15.7 mm Hg and a placebo-adjusted reduction of 8.4 mm Hg after 12 weeks. One head wind was a relatively high rate of K > 6 of 5.3% and 7.4% in the 50 and 100 mg dose.
This phase 3 study and a phase 2 study called LAUNCH HTN will be presented at the ACC meeting in a few weeks.
As I have said many times on this podcast, I do not love science by press release. We will have to see the details when the trial is published.
But this could be an important development because hyperaldosteronism is present in almost 1 in 5 patients with resistant HT. And a drug that directly reduces aldosterone its production might fill a nice niche. The trial was short in duration at 12 weeks, but if that amount of BP is sustained it is substantially more than renal denervation—and of course patients can avoid a femoral stick.
Speaking of Aldosterone
We need to talk about a trial of a K binding agent called the REALIZE K trial.
JACC published an RCT of sodium zirconium cyclosilicate, which I will subsequently abbreviate SZC, for the management of hyperkalemia during spironolactone optimization in patients with heart failure (HF).
You know the background: MRA drugs (mineralocorticoid receptor antagonists) are hugely helpful in HFrEF and probably HFpEF too if you do a careful analysis of the TOPCAT trial. For super young learners, it is worth pulling up the RALES trial of spironolactone vs placebo. RALES shows one of the largest effect sizes in all of cardiology — for a generic $4 per month drug.
The problem, of course, is that MRA drugs are often not given out of fear of hyperkalemia. K binding agents hold the promise of allowing more MRA use because binding the K may allow MRA use without getting high K.
The DIAMOND trial of patiromer, published in EHJ in 2022, found very modest reductions in K levels and 31 placebo patients stopped MRA vs 20 patients in the patiromer arm. A difference of only 11 patients. So, the $1000 per month drug surely has little effect on HF therapy.
The question in the REALIZE K trial is would SZC fare any better? SZC is an orally administered, inorganic crystal that has high affinity for the K+ ion and exchanges K+ for hydrogen and sodium. Prior trials have demonstrated that SZC is effective at rapidly lowering K+ and maintaining normal K+ levels long term in patients with hyperkalemia. Most of these studies did not enroll patients with left ventricular dysfunction (LVD), however.
Patients in REALIZE had HFrEF on stable meds and had to have high K or be at risk for MRA-induced hyperkalemia. Then patients meeting all eligibility criteria entered an open-label run-in period (4-6 weeks) for optimization of spironolactone dosing and normalization of serum K+ with use of SZC orally as needed.
After normalization of serum K+, spironolactone was initiated or uptitrated to a target dose of 50 mg/day with further adjustment of SZC dosing (5 g every other day up to 15 g once daily orally) as needed to maintain K+ in the normal range.
After the run-in there was a double-blind, placebo-controlled, randomized-withdrawal phase in which they were randomly assigned 1:1 to continued treatment with the same doses of spironolactone and SZC orally or to continued treatment with spironolactone and matching placebo orally for 6 months.
The primary endpoint was called optimal treatment response (normokalemia on spironolactone ≥ 25 mg/day without rescue therapy for hyperkalemia [months 1-6]). There were 5 secondary endpoints tested hierarchically each measuring K outcomes.
There was also an exploratory outcome of time to first CVD or HF event.
Results
There were only 200 patients randomized so this was not a HF outcomes trial.
The results were positive. The odds were more than 4x that SZC treated patients would have an optimal response of normal K on S 25 mg. SZC vs placebo also had positive results on the other secondary endpoints regarding K and S dosing.
Kansas City Cardiomyopathy Questionnaire (KCCQ) scores were similar. Adverse events were similar.
However, the composite of cardiovascular (CV) death or worsening HF occurred in 11% participants in the SZC group vs 3% participants in the placebo group, and that log-rank nominal P value was 0.034.
The authors concluded that SZC led to large improvements in the percentage of patients with normal K levels on optimum spironolactone dose as well as a reduced incidence of high K or down-titration/discontinuation of S.
They also write that, “Although underpowered for clinical outcomes, more participants had HF events with SZC than placebo, which should be factored into the clinical decision making.”
Comments
This is very interesting. They included the clinical outcomes Kaplan-Meier (KM) curve in the central illustration and featured it prominently in the conclusions even though it was a difference of 11 vs 3 events.
This makes me think. Was the worsening of event rates just noise or was it real. A nominal P-value was positive, but again, it’s an exploratory endpoint with low numbers.
Before I say anything I want to highlight the accompanying editorial from my friend Milton Packer, who, is an extremely clear writer and thinker. His editorial is titled: “Do Potassium Binders Block Both the Benefits and the Risks of Mineralocorticoid Receptor Antagonists in Heart Failure?: The Four Anti-Aldosterone Myths”
Myth 1: Many patients with HF cannot receive MRAs because of K. Packer cites the RALES and EMPHASIS trials and notes that clinically meaningful hyperkalemia when MRAs are dosed appropriately which is at 25mg is rare. The mean dose of spironolactone in RALES, which I’ve already said had a huge effect size in mortality, was only 26 mg. Uptitration to 50 mg was uncommon and rarely done in the trial. “More than 80% of patients with heart failure should tolerate highly effective (lower) doses of MRAs, but in practice, only 15% to 30% are receiving MRAs.”
Myth 2: Enabling the Use of MRAs With Potassium Binders Can Be Expected to Favorably Impact Morbidity and Mortality in Heart Failure. Packer notes that trials of K binding drugs like DIAMOND tend to focus on patients asking “excessive” doses of MRA, because hyperkalemia is so uncommon at 25 mg per day is low. In DIAMOND, Packer notes, the difference in MRA discontinuation of the placebo vs patiromer group was only 11 patients which would have a minimal effect on HF events over 3 years.
Myth 3: Potassium Binders Do Not Inhibit the Effect of MRAs to Reduce Morbidity and Mortality in Heart Failure. Here, Packer cites the results I just highlighted in REALIZE K — that is, that the SZC group tripling of HF events vs placebo. Why did this occur despite enhancing MRA use? Because, Packer writes, the SCZ adds 800 mg per day of sodium — thus imposing a huge dietary salt load.
Myth 4: Innovations in Anti-aldosterone Pharmacology Will Improve the Benefit-to-Risk Profile of MRAs in Heart Failure. Relevant to my first story on a new molecule for HTN, lorundrostat, the aldosterone synthase inhibitor, Packer notes that these drugs are being developed based on the belief that aldosterone can exert adverse effects through mechanism other than the mineralocorticoid receptor. The problem with this, Packer writes, is that MRAs antagonize numerous genomic and nongenomic pathways that may contribute to their benefit in HF.
The problem is that relevance of these many different issues is that we would need trials that compare aldosterone synthase inhibitors to MRA in patients with HF. Packer laments the fact that these trials are “being strategically avoided.”
Sadly, this is exactly what the makers of finerenone did. Instead of studying finerenone vs placebo the HF trials should have studied finerenone vs spironolactone.
I want to be clear about my thoughts here. The problem is not industry. Industry does what industry does — they develop products to maximize profits and that is not nefarious. The problem is that so much of our science is funded by industry, which will never fund trials that could minimize units sold. Industry-funded science can advance our field but only when the interests of our field are confluent with profits for the company.
Industry-funded science is why we had no trials of conduction system pacing (CSP), because CSP could reduce CRT units sold. Fortunately, a number of tireless scientists have now convinced government to fund such a trial.
I will cover an awful study later in this podcast that was partially funded by NIH. Obviously, NIH funding is in the news. It’s important because we need NIH to fund important studies in our field that industry will not fund. Such a comparison of aldosterone inhibitor drugs is a great example. But if NIH is burning money and time on low-quality studies there is less bandwidth for important outcomes trials.
Packer concludes by writing: “Importantly, the need for alternative anti-aldosterone agents has been substantially dampened by the fact that both neprilysin and sodium-glucose cotransporter 2 inhibitors mitigate the risk of hyperkalemia in patients with heart failure receiving MRAs.
I agree with all these comments and feel it is sad that MRAs are under-used and under thought of. RALES is an impressive trial. And it appears that the K-binders are neither necessary nor beneficial when it comes to outcomes.
Butter vs Plant-Based Oils and Mortality
JAMA-Internal Medicine has published another one of those nutritional observational studies comparing butter and plant-based oil with mortality as an endpoint. It’s led by a group at Harvard Medical School.
Obviously, seed oils, and their danger, are also a hot topic. The Altmetric score of this study is already over 2000. The pattern of this study is common. Three large cohorts were used. The primary exposures were self-reports of butter or plant-based oils, which could include safflower, soybean, corn, canola, and mysteriously, olive oil.
Main outcome measure is total mortality. Secondary is mortality due to cancer and CVD.
Follow-up was 33 years. There were 221,000 adults included. Participants were categorized into quartiles based on their butter or plant-based oil intake.
And the results: After adjusting for potential confounders, the highest butter intake was associated with a 15% higher risk of total mortality compared to the lowest intake (hazard ratio [HR], 1.15; 95% CI, 1.08-1.22; P for trend < .001).
In contrast, the highest intake of total plant-based oils compared to the lowest intake was associated with a 16% lower total mortality (HR, 0.84; 95% CI, 0.79-0.90; P for trend < .001).
Every 10-g/d increment in plant-based oils intake was associated with an 11% lower risk of cancer mortality (HR, 0.89; 95% CI, 0.85-0.94; P for trend < .001) and a 6% lower risk of CVD mortality (HR, 0.94; 95% CI, 0.89-0.99; P for trend = .03), whereas a higher intake of butter was associated with higher cancer mortality (HR, 1.12; 95% CI, 1.04-1.20; P for trend < .001).
The authors concluded:
Higher intake of butter was associated with increased mortality, while higher plant-based oils intake was associated with lower mortality. Substituting butter with plant-based oils may confer substantial benefits for preventing premature deaths.
Funding support came from numerous grants from NIH.
Comments:
The editorialists write that this study has several strengths—the large sample size, the long follow-up and the repeated assessment of diet and adjustment for confounders.
I disagree with all of these. As for sample size, it is hugely important, perhaps the most important lesson you take from today’s podcast, is that sample size in these observational studies may have no bearing on mitigating bias. When there is a systemic bias — such as butter eaters are different from olive oil eaters — no number of patients can fix this problem.
Larger numbers of patients fix the problem I just mentioned in the REALIZE K study, where 11 vs 3 patients had HF events, but a million patients does not fix a confounding bias.
I would also disagree that adjustment has any chance of sorting out baseline characteristic differences. Because you can only adjust for what is on a spreadsheet and there are many other relevant factors.
Another problem: The authors make choices in their analytic method. As I have reported before, the work of Dena Zeraatkar, PhD, from McMaster University, who showed with a meat consumption and mortality study that there literally a quadrillion different ways to analyze a data sheet, and some results are positive, and some are negative, and most show no effect.
Baseline characteristic issues and analytic flexibility are not the only issues with this study. You also have recall bias of food consumption, and the fact that olive oil is mixed in with other seed oils. Plus, there is the matter of implausibility. It belies any sense of biology to posit that one food component could have this large an effect on mortality!
Finally, I want to share a somewhat funny but enlightening nutritional study published in 2013 by Jon Schoenfeld and John Ioannidis. The title: “Is Everything we eat associated with cancer? A systematic cookbook review.” I will link to it. It’s in the American Journal of Clinical Nutrition.
The researchers selected 50 random ingredients from a cookbook and searched the scientific literature for studies about these ingredients' relationships with cancer. They found that 80% (40) of the ingredients had been studied for cancer risk, with 36 of these ingredients having at least one study claiming either increased or decreased cancer risk.
Key findings include:
Of the hundreds of single studies, 72% concluded that foods were associated with cancer risk (39% increased risk, 33% decreased risk).
Most of these claims (75%) were based on weak statistical evidence.
Single studies often reported implausibly large effect sizes, with median relative risks of 2.20 for increased risk and 0.52 for decreased risk.
Meta-analyses (36 of them) presented more conservative results, with only 26% reporting increased or decreased risk, and effects generally closer to null (median RR: 0.96).
The exposure contrasts used across studies were highly variable and inconsistent, making comparisons difficult.
The two authors argue that the nutritional epidemiology field may suffer from several biases:
Publication bias favoring "positive" results
Selective reporting of analyses that yield significant findings
Flexibility in how analyses are conducted and reported
Overinterpretation of weak statistical evidence
My point here is that such studies are totally useless. You could even argue that such studies are worse than useless; they might even be net harmful. Harmful how, you might wonder. Well, perhaps seed oils are worse for you, and this paper might cause increased consumption.
A more likely way this harms society is that papers like this shred trust in medical science. People look at this and see the flaws, and ask why would Harvard-based researchers do such a study? Why would the JAMA-IM publish such flawed work? Where were the peer reviewers? And if we cannot rely on journals like JAMA-IM, who can we rely on?
It’s bad bad bad.
Vitamin D
I’ve lost track how many times I have mentioned vitamin D on this cardiology podcast. Every time it’s because of a negative trial.
Well, it’s off topic, but JAMA has published an actual positive trial for vitamin D supplementation. I won’t say much because the trial tested high-dose vitamin D for the treatment of patients with newly diagnosed multiple sclerosis (MS).
F. Perry Wilson has a nice recap on his Medscape feature called the Impact Factor. The placebo-controlled trial seemed to show a clinically important and statistically robust effect size. Subgroups suggest patients with lower vitamin D levels at baseline benefit more. Benefit also seemed to be strongest in patients with milder disease.
I note the D-Lay trial for two reasons: one is that I think it’s a first for vitamin D, and secondly, as Perry notes, perhaps some of the replacement trials have been negative because of patient selection or not enough dose. This trial used very high doses very early in the disease course.
© 2025 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: Mar 14, 2025 This Week in Cardiology Podcast - Medscape - Mar 14, 2025.
Comments