Jan 6, 2023 This Week in Cardiology Podcast

John M. Mandrola, MD


January 06, 2023

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 January 6, 2023, John Mandrola, MD, comments on the following news and features stories.

We are back. Happy New Year everyone.

Damar Hamlin

Five Thoughts on the Damar Hamlin Collapse

I don’t know the numbers, but many millions of people watch “Monday Night Football.” This week, they saw the sudden collapse of Damar Hamlin. It came right after a hard but fairly routine tackle. He stood up momentarily but then passed out stone cold, without protecting himself. He lay immobile.

He was then resuscitated with cardiopulmonary resuscitation and a shock. The public nature of this event, the size of the fanbase of the National Football League, and the fact that this was an important game that was ultimately cancelled, was such big news that it actually overshadowed the chaos in our Congress.

And in US cardiology, everyone was talking about it. My private DM, What’s APP and Signal groups were on fire. In the doctor’s lounge this week, my partner and I could hardly eat lunch because people kept asking us what happened. Thankfully, yesterday, we got news that he could communicate and asked who won the game.

The cause of his cardiac arrest is unknown. It will eventually come out. For now, though, there many important lessons that transcend the cause of his sudden arrhythmia. I wrote about 5 of these on | Medscape Cardiology. It’s a short column. The topics involve:

  • The importance of early and effective intervention;

  • Having trained bystanders;

  • Having AEDs available.

  • I also added an argument on why screening athletes for causes of cardiac arrest remains a bad idea.

Improving survival from cardiac arrest is an area where public health could really shine. Northern European countries have made great progress, and they have shown us a plan. I hope this tragic incident sparks interest in improving the processes of treating fellow humans with out-of-hospital cardiac arrest.

Semaglutide and Obesity in Adolescents

This week the FDA approved the GLP-1 agonist semaglutide for the treatment of obesity in adolescents aged 12 years or older. The once-weekly injection was approved for adults in 2021. The supporting trial is worth a quick look. The New England Journal of Medicine published the STEP-TEENS trial 2 weeks ago.

  • About 200 youngsters were randomly assigned in a blinded fashion to semaglutide or placebo.

  • The mean age was 15 years, and the mean body mass index (BMI) was around 36. Pause there; 36 BMI at age 15. Clearly this is an important problem

  • Semaglutide crushed it. For the primary end point, patients in the semaglutide group had a reduction in BMI of 16 vs 0.6 in the placebo arm.

  • At slightly less than a year and a half, kids in the semaglutide had a 14 times higher odds of having a 5% or more weight loss.

  • The incidence of gastrointestinal (GI) adverse events was greater with semaglutide than with placebo (62% vs. 42%).

  • Five participants (4%) in the semaglutide group and no participants in the placebo group had cholelithiasis.

An expert in pediatric obesity called the result mind-blowing, saying that we are getting close to bariatric surgery results in these adolescent patients with obesity.

Comments. The main weakness of the trial was that it was less than 2 years. These were 15-year-olds; we have implantable cardioverter-defibrillator trials in older people with twice the follow-up.

The data on GLP-1 agonists clearly show they work. But two-thirds have GI adverse events.

  • There are two movies playing in my brain. One is that this drug causes a statistically robust and clinically important loss in weight. And we should consider obesity in kids a health emergency. So that movie tells the story of a short-term success, albeit with the tradeoff of GI complaints.

  • But the other movie playing in my head is, really? This is our solution? As with atrial fibrillation, the problem seems far north of GLP agonism. This movie tells the story of societal norms changing. Things like Big Food companies influencing policy. Lack of walkable neighborhoods, demotion of gym class. You don’t see this problem in other countries.

So, while we celebrate objective gains in weight reduction, can we also keep our minds on upstream causes of obesity. No, the United States is never going to be the Netherlands, but perhaps there are lessons to learn from Northern Europe.

Open Data and PSCK9 inhibitors

We closed the year last month with a beautiful story of a surgical group sharing their data with an independent group. On the December 16 #TWICPodcast, I talked about an academic surgical group that had reported, in a meta-analysis, outsized findings regarding a surgical conduit for bypass. The independent group reanalyzed the exact same data and came to a different conclusion, one more in line with what most surgeons believed. All parties thought the answer to this conundrum was a proper randomized controlled trial. This was made possible because of the generosity and humility of the surgical group that shared their data.

This year begins with another story about open data.

The journal BMJ Open has published a report from a group of authors from Europe and Canada regarding a reanalysis of the FOURIER trial of evolocumab vs placebo. The authors are part of an international effort called RIAT — Restoring Invisible and Abandoned Trials. Their goal is to tackle bias in research reporting.

Evolocumab is a PCSK9 inhibitor and in FOURIER, more than 27,000 patients with established vascular cardiovascular disease, and an LDL cholesterol (LDL-C) > 70 mg/dL on statins, were randomly assigned to evolocumab or placebo.

  • The primary endpoint was a composite of the first event of cardiovascular (CV) death, myocardial infarction (MI), stroke, hospitalization for unstable angina, or coronary revascularization.

  • The trial was planned for about 5 years but was terminated early for benefit at only 2.2 years.

  • In 2017, FOURIER reported that evolocumab reduced the incidence of the primary composite endpoint; hazard ration (HR) 0.85 (95% confidence interval [CI], 0.79 to 0.92).

  • Evolocumab also reduced LDL-C by 59% whereas LDL-C did not change in the placebo group.

  • There were, however, some curiosities of the components of the composite endpoint.

    • CV death and overall death were numerically higher in the evolocumab group. Not statistically significant, but weird, that the reductions in MI, stroke, coronary revascularization were not enough to reduce CV death or overall death.

    • Another curious thing, not nefarious, just curious, was that the vast majority of the CV deaths were listed as “other” CV death, not as CV death due to MI or stroke, but other.

Well, the authors of the BMJ Open paper got hold of the Clinical Study Reports(CSRs), which provide more details than published in the NEJM. CSRs are technical documents prepared by the manufacturer and submitted to regulatory agencies as part of the approval.

They contain information about the trial’s protocol, amendments, inclusion and exclusion criteria, outcome definitions and measurement, efficacy and safety results and statistical analysis plan. The main FOURIER CSR is over 25,000 pages. The first story is how these authors got hold of such documents. They asked the US Food and Drug Administration (FDA) and FDA said it would take years. They then asked Health Canada and boom, they got access.

Here are the main findings:

  • FOURIER had 870 deaths; 41% of the time the cause of death determined by the clinical events committee differed from that declared by the local investigator. Pause there. That seems like a lot of discrepancy.

  • The CSR info, when compared with the 2017 publication in NEJM, yielded 11 more deaths from MI in the evolocumab group vs three fewer deaths from MI in the placebo group.

  • In the CSR, the number of deaths due to cardiac failure in the evolocumab group was almost double those in the placebo group (31 vs 16).

  • After the authors readjudication, they noted that cardiac deaths were numerically, but nonsignificantly, higher in the evolocumab group (113) than in the placebo group (88; relative risk [RR] 1.28, 95% CI 0.97 to 1.69, P=0.078), whereas non-cardiac vascular deaths were similar between groups

  • The reported HR for cardiovascular mortality in the original trial analysis was 1.05 (95% CI 0.88 to 1.25); after readjudication, we found a greater (although still non-significant) relative increase in cardiovascular mortality in the evolocumab treatment group (RR 1.20, 95% CI 0.95 to 1.51, p=0.13).

The authors overstepped in their conclusions, I believe, saying that there was possible cardiac harm and that clinicians should be skeptical about prescribing evolocumab for patients with established atherosclerotic cardiovascular disease.

But they were right to claim that, at the time the trial was terminated early, a non-significantly higher risk of cardiovascular mortality was observed with evolocumab, which was numerically greater in the readjudication

Comments. One view of this is that there is nothing to see. A group of authors who had less information than the trialists looked at the CSR’s and came up with differences in causes of death. And statistically speaking, the results of the trial were upheld — that evolocumab reduced nonfatal outcomes but had no significant effect on overall or CV mortality. And I am sympathetic to those who called the authors for oversized conclusions.

The other view, which I favor, is this:

  • Nearly half of the time, the clinical events committee disagreed with the local investigator on cause of death. And CV death was an endpoint. That’s a lot.

  • Also, what if the independent review is correct? They found a substantially higher CV death rate in the evolocumab group.

  • The RR went from 5% higher in the main trial publication, to 20% higher in the reanalysis.

  • And in the reanalysis the 95% CI went from 0.95 to 1.51, which holds that it could be 51% higher. I may be naïve but that seems like a lot of flex in the results of a major trial.

Finally, and this may be super important, my colleague in Egypt, Professor Ahmed Bendary Tweeted that this might be the biggest story of 2023, writing, “trust me.” Especially if you took this data together with some other data pointing to possible development of left ventricular hypertrophy and abnormal cardiac lipid metabolism in PCSK9-knockout mice.

The European Heart Journal paper he cites concludes: “PCSK9 deficiency impacts cardiac lipid metabolism in an LDL-Receptor independent manner and contributes to the development of HFpEF.”

  • On the curiosity scale, I find this whole story a 10 out of 10. Come on. Look at the LDL reduction in FOURIER. Evolocumab smashes LDL compared with placebo. And we all know the importance of LDL. Plus, these were high risk patients with established heart disease,

  • Why then, does evolocumab not reduce vascular events enough in these high-risk patients to reduce CV death or overall death? Why are CV deaths higher? Is it noise or signal?

  • Perhaps it is because drug effects are complex and LDL reduction may be only one effect.

  • 27,000 patients with heart disease and no CV death or overall death reduction.

  • And in a re-analysis a signal of higher CV death.

  • At minimum, the authors are right in their future directions paragraph: Independent scrutiny of medical literature is crucial, especially when it is practice changing;

  • Raw data should be made available;

  • Great results will pass muster.

It shocks me to this day that there is so little independent scrutiny of major medical evidence.

TAVR Trials and Bias

Here is another story about re-analysis of data. This week a group of authors, and full disclosure, I was one, published a paper in JAMA Network Open looking into potential biases in the transcatheter aortic valve replacement (TAVR) vs surgical aortic valve replacement (SAVR) trials.

The lead author was Fabio Barilli, and most of us are part of a group called INTEGRITTY -- International Evidence Grading Research Initiative Targeting Transparency and Quality (INTEGRITTY) Collaborators. This is multidisciplinary group of academic professionals brought together by a common objective: critical appraisal of evidence to support better and optimized patient management.

As you know, the TAV implantation (TAVI) community has been exemplary in performing trials, first in high risk, then intermediate risk, and now low-risk patients with aortic stenosisn (AS).

I hold these trials up in lectures as an example of studies done not for marketing but to answer a crucially important question regarding the best approach to AS — surgical or trans-aortic? But that said, no trial is perfect, and translation of data requires consideration of potential biases in trials.

Led by the tireless Fabio Barilli, our paper was a meta-analysis and systematic review of the eight TAVI vs SAVR trials, looking specifically at three forms of bias.

  • One is deviation from assigned treatment, meaning patients who were assigned to TAVI and did not get it or vice versa. Again, no trial is perfect, so you’d ideally want all patients assigned to a treatment to get that treatment. Then, of course, the intent to treat (ITT) analysis would be the same as the per-protocol and as-treated arm.

  • Second is loss-to-follow-up. Ideally, you would like this to be low and, crucially, balanced in the two arms. If one arm, has more or less lost to follow-up, then you worry about a systemic bias, called attrition bias.

  • Third was additional treatments. In an ideal trial, you’d want both groups to differ only in the one treatment. If one group gets more or less additional treatments, you worry about performance bias.

What we found and reported was:

  • There was substantial deviation from treatment favoring TAVI. The RR was highly significant at 0.16.

  • There was also a moderate amount of loss-to-follow-up at 4.8%. and there was an imbalance of loss-to-follow-up favoring TAVI, the RR was significant at 0.39.

  • Finally, there was also an imbalance of additional treatments wherein many more patients in the SAVR arm received additional valve and revascularization procedures; 5% vs 16%.

In sum, despite randomization, which is supposed to balance things, there were significantly lower proportions of deviations from therapy, loss-to-follow-up, and receipt of additional procedures in the TAVI groups.

The most important of these may be the loss-to-follow-up issue. Almost all the pivotal RCTs have a high imbalance of loss to follow-up, with a constant pattern favoring TAVI, reaching a loss of 10% in the TAVI groups and more than 20% in the SAVR groups at 5 years.

The deviation from assigned group issue gets complicated. The difference in ITT (where the analysis is done based on what group a patient is in regardless of whether he or she has the assigned treatment) and as-treated (where the analysis is done only on patients who have the assigned treatment) ought to be minimal.

This came up in the CABANA trial of AF ablation vs drugs. In the ITT analysis there was no difference in outcomes. But there were substantial numbers of subjects who deviated from their treatment assignment. Some didn’t get ablation, and there were a lot crossovers from drugs to ablation. So the authors analyzed it according to the as-treated and ablation looked better. But, of course, those patients may be different in other ways, so as-treated analyses are biased,as they are no longer random.

When ITT and as-treated numbers are different, you have to look at whether it’s a superiority or non-inferiority design. In a superiority trial, using the ITT tends to favor the null hypothesis, because some patients did not receive the treatment but are still counted.

But in a non-inferiority design, the alternate hypothesis is that there is no difference, so the ITT is no longer conservative, because if you have patients not receiving their assigned treatment you bias to no difference. Seven of the eight trials had a noninferiority design.

As for the additional procedures, many more patients had extra surgical procedures. This is normal surgical practice. It would have been hard to limit these things, because then we’d have an external validity, or generalizability issue. But the point is that doing more additional procedures in SAVR arm could increase the risk of bad outcomes in the short run but pay dividends in the long run. However, most of the TAVI trials had short term follow-up.

Here is the take-home:

  • All we know from this analysis is that there were challenges to the internal validity of these trials.

  • We did not have individual patient level data, so you can’t say what direction these biases leaned.

  • The answer to sorting through these biases would be to open the data, and allow independent and transparent review.

  • You can hold two views of TAVI. I do.

    • One is that it is an amazing procedure and triumph of innovation, one that allows high surgical risk patients with AS to live and prosper for longer than they would have otherwise. And the trialists have done a fantastic job confirming this view.

    • The other view is that before we replace the established norm of surgical valve replacement in lower-risk patients we ought to have high level, long-term, and independently verified evidence of equivalency of the newer approach.


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.