Aug 26, 2022 This Week in Cardiology Podcast

John M. Mandrola, MD


August 26, 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 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 August 26, 2022, John Mandrola, MD comments on the following news and features stories.

Early Rhythm Control

Circulation has published yet another substudy of the EAST-AFNET trial of early rhythm control (ERC) vs standard of care in patients who had atrial fibrillation (AF) for less than a year. This study considered two subgroups — those with high-risk stroke features and those without. To do this, they dichotomized the enrolled patients to a group with a CHADSVASC of 4 or greater vs those with a score of less than 4.

Before I tell you about this paper, let’s quickly review EAST-AFNET. About 2800 patients were enrolled in 135 centers in Europe. The trial randomly assigned patients to ERC or standard care (basically rate control) and measured a primary outcome of cardiovascular death (CVD), stroke, heart failure (HF), or acute coronary syndrome (ACS).

Over the 5-year follow-up, EAST-AFNET found a statistically significant, 21% reduction in this primary endpoint.

Many of my colleagues in electrophysiology (EP) felt relieved that the “negative” results for rhythm control in the 20-year-old AFFIRM trial have been reversed. And since we tend to be proponents of ablation, and ablation is a rhythm-control strategy, EAST-AFNET has provided tail winds to the early AF ablation therapeutic fashion.

While I respect and admire many of the EAST AFNET authors, I have not been enthusiastic about the results for many reasons. Here is the big one:

  • Look at Figure S3 in the supplement.

  • At 2 years, 60% of the standard rate control arm was in sinus rhythm vs 80% of the ERC.

  • Do you really think a 20% difference in sinus rhythm in a cohort of patients of whom 90% are on oral anticoagulants can drive this much benefit in hard outcomes?

If you are like me and you don’t think antiarrhythmic drugs and AF ablation reduce hard outcomes, then you have to explain why this would be. Why was there so much reduction in outcomes with this little difference in rhythm?

I think EAST-AFNET is one of the best examples of performance bias there is.

  • Because the ERC had as its goal maintenance of sinus rhythm, patients were instructed to transmit a patient-operated single-lead electrocardiogram (ECG) twice per week and when they were symptomatic. This surely led to more care of this group, and that is likely what drove the results.

  • There was also a fair amount of missing data, and a three times higher adverse event (AE) rate in the ERC group; that is, if you don’t double count primary efficacy events in the AE table.

  • Finally, EAST-AFNET should not be held up as a reason to do early ablation because less than one in five patients in the ERC arm had ablation.

  • Now to the subgroup analysis from Circulation. The main finding was that ERC reduced the primary endpoint (CVD, stroke, HF, ACS) by a statistically significant 36% in the group with a CHADSVASC of 4 or more, but ERC did not have any notable effect for those with CHADSVASC below 4. Here the hazard ratio (HR) was 0.98.

  • When you go look at the major drivers of the benefit in the high-risk group, you find that stroke was more than halved with ERC. CV death was also substantially lower.

  • To me, the stroke signal is super strange. Remember that more than 90% of the total cohort was on oral anticoagulants. And there was only a 20% difference in sinus rhythm at 2 years.

This could be statistical noise, which is always possible in subgroups because trials are powered for the overall analysis. When you start slicing things up into smaller groups, whether or not it is prespecified, you introduce noise. (Google Richard Peto, ISIS-2 trial, astrological sign.)

But let’s say it isn’t noise; let’s say it is real. I think this strengthens the case for performance bias. Patients with higher disease complexity would be more likely to benefit from more interactions with the health field.


Patients with aortic stenosis (AS) are often older. And older patients with AS often have coronary artery disease (CAD). In the days before transcatheter aortic valve implantation (TAVI), a surgeon who replaced the aortic valve, would bypass these lesions.

TAVI has changed this calculus. Recall that patients with “complex” coronary disease were excluded from the TAVI trials. That is fine, because TAVI trials aimed to compare two ways to treat a valve issue. But in real-life there is a bunch of overlap between AS and CAD.

The journal JACC: -Cardiovascular Interventions has published an observational study from a large Belgian referral center. The study included 600 patients who underwent TAVI between 2008 and 2020.

The authors asked two questions: Did CAD and its complexity associate with worse outcomes and did “fixing” those lesions with a percutaneous coronary intervention (PCI) associate with better outcomes? The results were clear:

  • Yes, as CAD complexity increased, as measured by the Syntax score, 5-year survival worsened, in parallel with increased CV mortality.

  • Neither PCI nor the completeness of revascularization associated with different outcomes.


Well, as I learned in the accompanying editorial, this is not the only piece of evidence suggesting dubious efficacy of percutaneous revascularization before TAVI. A meta-analysis in the Journal of the American Heart Association found no association of benefit, and a potential signal of harm with PCI.

And ACTIVATION, a randomized controlled trial (RCT) comparing PCI vs no-PCI in patients who were to have TAVI failed to reach non-inferiority for PCI. And mortality was numerically higher in the PCI arm, as were rates of bleeding.

How to deal with CAD in the presence of AS is a huge issue. We are doing more TAVIs. Lots more. Yet many of these patients will have CAD. If they were referred to a surgeon, these would be bypassed. If they have TAVI, there will be a temptation to do PCI. But the data for that looks less than reassuring.

I don’t have the answer, but this is super-complex, and you’d have to consider things like life expectancy, what the CAD actually looks like, bleeding risk (PCI will increase bleeding), and whether or not the valve could impair coronary access.

What I worry about is that there is great enthusiasm for TAVI. It’s new. It’s cool. Programs need numbers. Patients want to avoid sternotomy. There are huge therapeutic tailwinds. But perhaps the surgeon’s ability to care for both conditions ought not be forgotten.

ESC 2022

  1. Mandrola Previews His Top 5 Trials From ESC 2022

I am in Barcelona Spain for the European Society of Cardiology (ESC) meeting — the first time it has been held in-person in 2 years. That’s a good thing, because the greatest value of medical meetings is to be with colleagues. This week I will discuss trials presented Friday. Next week I will have more.

TIME. The TIME trial is a pragmatic UK trial that randomly assigned more than 21,000 patients with hypertension. The study compared night-time dosing of antihypertensives with daytime dosing. It has a legit endpoint of myocardial infarction (MI), stroke, or vascular death.

You may be wondering why folks would spend the time and money to study such a thing. I mean, who cares what time a day one takes a long-acting medication? Well, it turns out that numerous trials have reported major reductions in outcomes simply by taking meds at night. The HYGIA trial stood out because it reported a stunning 42% reduction in hard outcomes with night-time vs day-time dosing of meds. But that trial has come under withering criticism.

Also interesting is that just days before TIME was presented, the International Society of Hypertension (ISH) published a systematic review and consensus statement on the matter of nighttime dosing of blood pressure (BP) meds. The European Society of Hypertension and the World Hypertension League have endorsed the document. In their abstract, the TIME trialists note that, “Eight published outcome hypertension studies involved bedtime dosing of antihypertensive drugs, and all had major methodological and/or other flaws and a high risk of bias in testing the impact of bedtime compared to morning treatment.”

Here are their results:

  • The primary endpoint occurred in 362 (3.4%) participants in the evening dosing group and 390 (3.7%) in the morning dosing group giving an unadjusted HR of 0.95 (95% confidence interval [CI] 0.83–1.10; P = 0.53).

  • The results did not vary in pre-specified subgroup analyses.

  • Taking medication in the evening was not harmful.

Wow – we went from a 42% reduction of events in HYGIA to none in TIME. This is remarkable. As per the ISH document, two ongoing RCTs are asking the same question.

I want to point you to what I feel is a very important paper on large effect sizes. Pereira and colleagues did a systematic review of Cochrane reviews looking at statistically significant very large treatment effects of medical interventions when such effects are recorded in the first clinical trial. They found that: Large treatment effects in initial trials (odds ratio > 5) were not rare, occurring in about 10% of all reviews; and when additional evidence is obtained, most of the very large treatment effects become much smaller and many lose their nominal significance.

TIME leaves us with three messages:

  • Patients can take their BP tablets any time of day they want.

  • We should always be skeptical of large effect sizes.

  • Scientific papers that will change practice ought to have to submit source data for independent review.

SECURE Trial: The Polypill. The SECURE trial is a European cardiac outcomes trial of a polypill of aspirin (100 mg), ramipril (2.5,5, or 10 mg), and atorvastatin (20 to 40 mg) vs standard care in about 2500 older patients.

SECURE was secondary prevention. Patients had a history of Type 1 MI within the previous 6 months. This is a super-important aspect. Patients with plaque rupture are high risk, and medical therapy likely has the greatest absolute risk reduction in these patients. That is, if they take the pills. Polypills improve the adherence problem by decreasing pill burden treatment complexity. Previous trials show that adherence associates with better outcomes—obviously. The average age of these patients was 75 years.

The results:

  • After 3 years of follow-up, primary outcome event of CVD, MI, stroke, or urgent revascularization occurred in 9.5% of patients in the polypill group vs 12.7% in standard care.

  • This 3.2% absolute reduction resulted in a HR of 0.76 or 24% reduction. CIs ranged from 39% reduction to a 4% reduction and were statistically significant at a P-value of 0.02. The curves separate early and get wider over time.

  • The good news for the PP was that all the hard outcomes — CVD, MI, stroke — favored the PP group. Only urgent revascularizations were similar. You like it when the hardest endpoint drives the benefit (unlike the sacubitril/valsartan and SGLT2 inhibitor story in HF with preserved ejection fraction, where the softer HF hospitalizations (HHF) rather than CVD drove the positive results.

  • There were no variations in outcomes among countries.

Caveats. These are older patients who are post plaque rupture and there were no differences in overall death. That speaks to the competing cause of mortality in patients in their mid-70s. You can reduce CVD but it did not translate into gains in lifespan.

This was an open-label trial. This could have led patients in the polypill arm to behave in a more heart-healthy way. Strengthening the argument for this is that there was only a small, less than 10% difference in compliance, and no significant differences in BP and cholesterol. Polypills work via their simplicity. Polypills improve compliance. Here the very modest improvement in did not drive BP or lipids. Then how exactly did the polypills reduce outcomes by a 24%? The authors point to the pleiotropic effects of statins and possibly higher rate of aspirin use in the polypill group. Maybe.

All in all, here is what I think: There were no safety issues, taking lots of pills increases the work of being a patient, and SECURE found a large effect size that was statistically robust. I’ve been skeptical of combining this many drugs in one pill, but my initial impression of SECURE is positive. Dr Fuster made an important comment in the press conference. He said there was great tension in medicine between simplicity and personalized medicine. Personalized medicine has its virtues, but in the case of post-MI patients, the simplicity of a polypill may be the better philosophy.

CAPLA Trial (embargoed until 10:30 am EDT Friday). CAPLA may seem like highly EP-centric trial. CAPLA enrolls patients with persistent AF — a group known to have lower AF ablation success rates when compared with those with paroxysmal AF. It then compares the standard approach to AF ablation, pulmonary vein isolation (PVI) to PVI plus isolation of the posterior wall of the left atrium.

But I am not highlighting CAPLA because it might inform a specific niche area of EP. That is important because this is a multicenter randomized trial, it’s the best way to get to the bottom of treatment efficacy and safety. I highlight CAPLA because there has been a strong contingent of key opinion leaders who have advocated for posterior wall (PW) isolation. In other words, PW ablation has largely been eminence-based. The data to support this practice is weak. Plus, PW isolation entails ablating even more in the back wall of the atrium. That’s the area of the esophagus. And thermal injury to the esophagus can be deadly. These are the main reasons I’ve not adopted PW as a routine strategy.

  • CAPLA randomly assigned about 340 patients with persistent AF.

  • They were drug-refractory; this was a first-time ablation; and AF had to have been for more than 7 days and less than 3 years.

  • The primary endpoint of freedom from AF/atrial tachycardia off antiarrhythmic drugs at 12 months was the same in both groups (HR 1.01). The Kaplan Meier curves super-imposed. Not a shred of difference.

  • Procedure times were longer in the PW arm.

The Australian authors, led by the famous Melbourne group write that “these findings do not support the empiric inclusion of posterior wall isolation for persistent AF ablation.”

I want to go further. I would say these findings support the power of the RCT. It supports evidence-based practice over eminence-based practice. It should infuse us with caution about single-center or observational studies.

More to Come From ESC

Here is a brief teaser for next week’s #TWICpodcast.

The big study I am looking at is DANCAVAS trial. Danish researchers have done an RCT of 45,000 men on the Island of Funen. One-third are offered screening and two-thirds get standard Danish care. And this isn’t Mickey Mouse screening. It’s serious: CT scans to look for coronary artery calcium and aneurysms, ankle-brachial index to look for peripheral artery disease, rhythm assessment looking for AF, and blood tests for lipids and glucose. The primary endpoint is death or cardiac events. The authors will also do cost-efficacy analyses.

I eagerly await this trial. Gosh it is important, because there is a heck of a lot of screening going on in the United States.

The REVIVED-BCIS trial, another UK trial, randomly assigned about 700 patients who had ischemic cardiomyopathy (ejection fraction [EF] ≤ 35%) due to multivessel CAD to either PCI vs medical therapy. I say it’s a best-case scenario because you had patients with amenable anatomy, high levels of viability, and left ventricular (LV) dysfunction.

But the thing is that modern-day, disease-modifying medical therapy is quite good, for both LV dysfunction and CAD.

The POST-PCI trial will study the strategy of doing routine surveillance testing with non-invasive stress testing in high-risk post-PCI patients.

For my entire career, the strategy of post-revascularization stress testing has lined the pockets of cardiologists. There’s no question that dualities of interest drove this low-value therapeutic fashion. Mostly it has stopped. But more and more, PCI is being done on high-risk patients. Stents at proximal bifurcation points or in the left main. Perhaps these patients will benefit from looking more intensely.

I’m glad this idea is being tested in an RCT. And if POST-PCI fails to find benefit, this should completely shut down the practice of post-revascularization stress testing. Period.

The DELIVER trial of dapagliflozin in patients with HF and an EF > 40%. We know, from a press release, it met its primary endpoint of CV death or HHF. The key finding will be what drove the endpoint? Look at the breakdown of endpoints.

Will DELIVER be like the EMPEROR-Preserved trial of empagliflozin, a positive trial driven only by HHF, which were only a small fraction of total hospitalizations.

I love SGLT2 inhibitors but you really want a costly drug to move CV death or you want the reduction in HHF to be significant enough to reduce total hospitalizations.

The ADVOR trial, a Belgian trial, will test the old drug acetazolamide in IV form in addition to loop diuretics for the treatment of patients with decompensated heart failure. The primary endpoint is “successful decongestion,” defined as the absence of signs of volume overload in the first 3 days and without an indication for escalation of decongestive therapy.

An inexpensive drug that aids acute HF therapy would be a big advance, because acute HF is common. One limitation to look for: the trial was designed before SGLT2 inhibitors were widely available. And, if positive, the question will be how do the results translate to patients taking or indicated for SGLT2 inhibitors?

The BOX trial from Copenhagen will test both BP and oxygenation targets in comatose survivors of cardiac arrest. In the oxygen target study, it is a test of PAO2 of 69 to 75 or 98 to 105. In the BP arm, it was a mean arterial pressure of 63 vs 77.

I love trials that test strategies, especially those that test less aggressive measures. If the lower oxygen and BP targets lead to similar results, it’s a big gain because, of course, it takes less supplemental oxygen and less pressors to achieve the same goal.

There is more, but I will close with INVICTUS-VKA, which is a an RCT of 4500 patients with rheumatic heart disease (RHD) and AF, comparing rivaroxaban with vitamin K antagonists (VKA) — mostly warfarin. While the current recommendation for patients with RHD is VKA, the problem is that the quality of anticoagulation in developing countries is often poor. Rivaroxaban makes it easy with once daily dosing.

The trial enrolled patients in 23 countries in Africa, Asia, and South America. It was a noninferiority (NI) trial, and a shining example of when NI should be used. You don’t need rivaroxaban to be superior to VKA. If it is just as good, that is perfect, because it offers a much easier way to anticoagulated.

The primary outcome is also perfect. Unlike some recent NI trials, INVICTUS authors studied thrombotic events—stroke and systemic embolism. Safety was a separate endpoint. While RHD is not common in the North America or Western Europe, worldwide it is a major issue and I commend the authors and look forward to the results.


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.