Nov 11, 2022 This Week in Cardiology Podcast

John M. Mandrola, MD


November 11, 2022

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 November 11, 2022, John Mandrola, MD comments on the following news and features stories.

Four Meta Comments on the AHA Meeting in Chicago

The pandemic taught many lessons, but one is surely the inferiority of virtual learning. I’ve given a bunch of in-person talks in the last year, and they are vastly better than virtual. On Zoom, you talk into a screen, you can’t see your audience, you have no idea if you are connecting. I hope all the societies eliminate virtual options. Period. Good riddance.

Another huge positive with the 2022 American Heart association (AHA) meeting was that it looked different than typical cardiology meetings. On the stage and in the halls, there were young people, women, and under-represented groups. This is a very good thing. I’m no expert in diversity but it seems to me that the way to diversity is via diverse mentors.

Two negatives about AHA and most American cardiology meetings. I don’t get the impression that practicing doctors come anymore. I remember the days of old when the halls would be teeming with regular doctors; sessions would be filled up. Now, it seems that the only people who come are the presenters, the faculty, and industry.

No one in my group has come to a meeting purely for education in years. Perhaps it’s the time away, the expense of the hotel room and travel, and the ease of CME from other sources.

If professional societies want to continue to have these meetings, they are going to have to attract regular practicing doctors. Give them something they can’t get from a Power Point CME lecture or a booklet.

Which brings me to the final critique. Cardiology meetings have become way too agreeable. Take the late breakers. It’s like an award ceremony or cheer-leading session. I understand that research is hard and it’s wonderful to see a study through to finish.

But these are practice-moving studies. They deserve serious critical appraisal. The authors rarely focus on their limitations. The discussants spend 1 of their 3 minutes congratulating the authors and then there’s no time for panel discussion. Maybe there’s a deep dive later in the afternoon, but these are just more time for celebration.

You can have deep respect for the evidence generators and still submit these papers to critical appraisal. I see more critical appraisal in a journal editorial, and that’s sad because journal editorials are rarely critical.

Imagine a world where a paper would be presented and then the audience could expect it to be vigorously appraised. Then the authors and panel could engage in debate. This might motivate more practicing doctors to show up.

You may push back and say that if you did that there wouldn’t be enough time to do all the late breakers. To this argument I say you could easily debride one-third of the late-breaking papers and lose nothing.

Science should never be about cheer leading. Debate ought to be center stage--as debate is science.


At AHA, the co-principal investigator of the ISCHEMIA trial, Dr Judith Hochman, presented results of the extended follow-up of the trial. The results were about as clear as any result at AHA. But before I tell you, let’s briefly review ISCHEMA.

Published in the New England Journal of Medicine (NEJM) in 2020, ISCHEMIA randomly assigned more than 5000 patients who had positive stress tests and proven coronary artery disease (CAD) without left main disease to an early invasive strategy or delayed invasive strategy with medical therapy first.

  • At 3 years there was no significant difference in the primary endpoint of cardiovascular (CV) death, myocardial infarction (MI), unstable angina, heart failure (HF), or cardiac arrest.

  • There also were near identical rates of death in both arms.

  • Crucially, the most important finding from ISCHEMIA was tucked away in the supplement. Figures s6a and b show that these similar results were had with 60% to 70% fewer angiographies and percutaneous coronary intervention (PCI).

At AHA, we learned about extended follow-up. This paper included, roughly, 4800 of the original 5200 patients in the trial. Median follow-up was now 5.7 years.

  • After the original trial, 289 patients or 5.6% died. The extended follow-up included 268 additional deaths. More events means more precision.

  • All-cause death was nearly identical with a hazard ratio (HR) of 1.00, and nearly symmetrical Confidence Interval (CI) 0.85 to 1.18.

  • The authors told us that CV deaths were significantly lower with the invasive strategy and non-CV deaths were higher.

  • There were no differences in subgroups. This sounds like jargon, but remember that even with the most severe disease, the early invasive strategy with its higher rates of revascularization did not lead to better survival.

  • Much was made at AHA and in the Circulation paper of the diverging causes of death. In the invasive arm, CV deaths were lower but non-CV deaths were higher.

  • My friends, don’t be fooled. The authors tell us that “cause of death in the extended follow-up is not being centrally adjudicated.” It’s a fool’s errand to try to sort out causes of death, especially in extended follow-up, without central adjudication.

Also, death is a totally appropriate endpoint in such a study. You know why? Because the reason doctors send patients with positive stress tests to the cath lab is because they are afraid a blockage will kill their patients.

ISCHEMIA and its extension study show clearly that even the patients with the most severe CAD can be treated medically, with angiography deferred only for clinical worsening.

Again, the only remaining mystery of stable CAD is why mountains of evidence cannot convince doctors. Because I guarantee you that in most institutions, a patient with a highly positive stress test will soon have a catheter in their wrist.


The strongest trial at AHA had the most apt name. But before I tell you anything about this outcomes trial in the heart failure space, which was published in The Lancet, I want to tell you the results.

  • The active arm of this trial led to 34% relative and 8.1% absolute risk reduction (ARR) in the composite of death or HF readmission.

    • CV death, 26% lower

    • All-cause death, 16% lower

    • HF readmission, 44% lower

If this were a trial of a drug or a device the booth would take up a fifth of the expo space and it would be covered in the New York Times and the Wall Street Journal. There would be banners going up the stairs at the congress hall. But it wasn’t a drug. It was a strategy of care after discharge from the hospital for an acute HF episode.

STRONG-HF enrolled about 1600 patients with acute heart failure who were not treated with optimal medications from 87 hospitals in 14 countries. Nearly 90% were recruited from Africa and Russia. Patients were randomly assigned to a high-intensity treatment arm or usual care.

The high-intensity arm was intense.

  • Patients first were given half the optimal dose of heart failure meds while still hospitalized. These included renin-angiotensin blockers, beta-blockers, and mineralocorticoid receptor antagonists.

  • At week 1 after discharge, patients were checked for tolerance of these meds. At week 2, meds were up titrated to full dose. The week 3 visit was used to check tolerance of full-dose meds, and patients were checked again after 6 weeks.

  • Notably, cardiologists performed these "safety" visits, using only a history, exam, and measurement of basic labs.

  • Quality of life measures were better in the active arm, as were signs of congestion.

  • Boy did medical management differ: Far more patients in the high-intensity arm were on optimal doses of the three classes of meds (SGLT2 inhibitors were not yet used during the trial).

Of course, at AHA, more medicines were given all the credit, and no doubt that helped. But four visits in 6 weeks surely entailed lots of education and caring.

One of the criticisms I heard at AHA was that because most patients were enrolled from Russia and Africa, the control arm had sub-optimal care, so good care should always beat sub-optimal care.

I want to push back on that because I could cite oodles of papers showing how poor guideline directed medical therapy uptake is in the United States. We don’t do much better, outside of clinical trials. Another criticism was that these were young ambulatory patients with few co-morbidities and most patients admitted for HF are older and have more co-morbidities. This is true, but who is to say that a STONG-HF like protocol might not work for them as well, not so much to maximize tablets but ensure treatment of all co-morbid conditions, including the provision of palliative care.

STRONG HF is provocative. It shows that if you get people four weekly visits after an HF admission in which patients are maximized on meds, they do better. This can’t be done in our system right now. There is nowhere near enough people power. I suspect we are representative. This is what I wrote in my column:

If we use evidence to guide practice, health systems have their work cut out for them if they want to implement a STRONG-HF protocol.

It will require a massive investment in people power and will not come at a low cost. Health leaders love to talk about better outcomes. Now it's time to bolster talk with action.

I look forward to watching how STRONG-HF influences care systems.

Experts vs Practicing Docs and HTN

The Diuretic Comparison Project (DCP) compared cardiovascular outcomes with HCTZ vs chlorthalidone (CTD) in patients in the Veterans Admnistration (VA) system. There are two hooks to this trial that make it fun.

First is that American College of Cardiology (ACC)/AHA guideline writers favor CTD over HCTZ. Yet every practicing doctor I know uses HCTZ. Observational studies comparing the two drugs will usually have 9 times more patients taking HCTZ than CTD. What do practicing doctors know that experts don’t? Why the discrepancy and recommendations vs practice?

The second hook was the way the authors accomplished the comparisons. Namely, they embedded the trial right into the electronic health record (EHR) and the normal practice patterns of the doctors. There were no special research visits.

After the authors got permission from the primary care clinicians, veterans who were taking HCTZ were called and asked to participate in the head-to-head comparison.

  • Half the patients were randomly assigned to continue HCTZ and the other half to switch to CTD. About 6000 patients were in each arm.

  • After 5 years there were no differences in the primary outcome of stroke, heart attack, non-cancer death, HF, or need for coronary procedures. (HR, 1.04; 95% CI, 0.94 - 1.16).

  • More adverse events occurred in the chlorthalidone arm, primarily low potassium levels, though this was not enough to affect the primary outcome.

Gosh this is a cool study. Think about it — if the experts were right about their first-line choice, and there was even a tiny signal of benefit, this would be important to know because so many people are taking HCTZ.

The other thing that I love is that the experts were wrong and working-stiff doctors were right. That’s provocative, but I wouldn’t take it too far. I wouldn’t posit that it was cumulative wisdom of the those who see patients every day.

I wouldn’t say that because history has proven us wrong so often: Premature ventricular contractions after MI, HRT in post-menopausal women, Swan-Ganz catheters and cardiac cooling in survivors of cardiac arrest. I could go on.

The way to know what works is not to rely on cumulative wisdom. Instead, do exactly what Areef Ishani, MD, and colleagues did: Make the choice to use two therapies randomly. You do a randomized controlled trial.

#TWICPodcast wouldn’t be true to itself if we didn’t mention some of the limitations. Academic cardiologist Jim Stein, from the University of Wisconsin sent me some of his concerns:

  • DCP was open label, so people likely were treated differently – at a minimum some people knew they were switched vs not switched. People change behaviors and are treated differently when they know “things;” what things we don’t know, but we know those things lead to differential care.

  • BPs were the same in both arms throughout the study – why wouldn’t they be lower on CTD? Is this null bias?

  • Outcomes were from the her, which is not sensitive for endpoints.

  • This study pretty much was optimized to show low potassium on CTD (EHR is good for that).

These are all good criticisms. I would add that we haven’t seen the full paper yet. Here’s the thing though: DCP was a pragmatic trial. It’s embedded in the system. It’s much lower cost and burden than a standard trial. Pragmatic trials are always going to be a bit noisier with, as Dr. Stein says, less precision.

No trial is perfect. But DCP had an HR of 1.04 in a study of more than 12,000 patients. Seeing the paper would help sort out the total events, but gosh, on the surface, it seems unlikely that we are missing a truly positive result from CTD — a false negative or type 2 error.

The biggest contribution of DCP is not only that HCTZ looks like the better thiazide type diuretic, but that they have shown us a framework for gaining knowledge in the practice of medicine. Imagine a future where instead of guessing which drug or procedure is best, doctors have the option to randomize patients into trials that are embedded into the normal practice patterns.

Transform HF

AHA featured another pragmatic trial of torsemide vs furosemide in patients discharged after a HF admission. Dr Robert Mentz presented results in the first late breaker. Regardless of ejection fraction (EF), patients at 60 US sites were randomly assigned to either of the two loop diuretics. Clinicians took care of dosing and the primary outcome was all-cause death as assessed by the National Death Index. This was another pragmatic trials were there were no in-person visits.

This was an event-driven trial powered to find a 20% mortality reduction with torsemide, which has theoretic benefits, such as better absorption and possible beneficial pleiotropic effects. Think about that, I will come back to it.

  • There were about 1400 patients per group; about one-fifth had an EF greater than 50%.

  • There was no difference in mortality with an HR of 1.02and CI 0.891.18. Since the lower bound was 0.89, that’s an 11% reduction, the results pretty much exclude a 20% reduction in death.

  • There was also no significant difference in total hospitalizations (good on the authors for looking at total hospitalizations and not heart failure hospitalizations).

  • The problem with the trial, again, is the noise inherent in a pragmatic trial: Unknown loop diuretic status was reported in about 700 patients. There were crossovers in about 6% to 8% of patients.

  • I also wonder why the authors thought torsemide would reduce mortality by 20% over furosemide. We haven’t seen a reduction in HF mortality since RALES in 1999. Overestimating benefit allows for enrolling fewer patients but if you enroll too few patients you end up not answering a question.

These criticisms should not diminish the spirit of doing pragmatic trials. We need more of these trials, noisy as they are.

Triglyceride Lowering – the PROMINENT Trial

We have long known that high triglyceride (TG) levels associate with increased risk of CV events. The problem is that most trials have failed to show that lowering TG levels reduces events. At AHA we heard results of the PROMINENT trial, an RCT of a novel TG-lowering drug called pemafibrate, (which rhymes with fenofibrate). NEJM published the CV outcomes trial of more than 10,000 patients. 10,000 patients means the trialists expect a small benefit because you don’t need 10,000 patients for a therapy that has big effects. These patients had to have type 2 diabetes, mild to moderately high TG levels and HDL cholesterol levels below 40 mg/dL. Patients were already on lipid-lowering therapy, mostly statins. The mean TG level was about 273 mg/dL.

A note on pemafibrate — it’s a fairly novel drug that has been shown to reduce TG levels.

  • Pemafibrate shredded TG levels by more than 25%;

  • It had absolutely no effect on the primary outcome of MI, stroke, revascularization, or CV death.

  • Given the previous trials that failed to find any signal that lowering TG improves outcomes, and now this huge trial of a drug that reduced TG but does not improve outcomes, perhaps it’s time to conclude that TG levels are a marker of risk but not a causal factor in CV events.

That would mean that we can reduce polypharmacy by stopping drugs that change only a lab value but provide no CV benefit to patients; we could save research dollars and time by studying other theories, perhaps ways to get people to eat better and exercise more; and use TGs as a classic teaching case that moving surrogate markers doesn’t necessarily translate to benefits.


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.