May 20, 2022 This Week in Cardiology Podcast

John M. Mandrola, MD


May 20, 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 May 20, 2022, John Mandrola, MD comments on the following news and features stories.

Stable CAD – Ischemic Burden vs Anatomic Testing

In Sept 2021, ISCHEMIA trial authors published a sub-analysis of their trial looking at ischemic burden vs anatomic complexity in terms of prediction of events and effect on invasive vs conservative strategies. It is a provocative paper on many levels. This week, Patrice Wendling reported on a letter from heart surgeons who found discrepancies between the number of patients in each anatomic severity group reported in the paper vs those reported in the supplement.

The surgeons found this issue in early April and wrote a letter to Circulation, but the editors decided not to publish it because it came after the 6-week window. (A side note here: I think a 6-week window for letters to the editor is ridiculous; in digital publishing, it ought to be a rolling window.) The authors say it’s a formatting error and to my eye, as of yesterday, the supplement and paper now seem to correspond.

Patrice’s coverage got me looking into this paper and the big idea of ischemic vs anatomic testing. We are doing a ton more computerized tomographic angiograms (CTAs) and I am not sure this is a good idea.

First, recall that the ISCHEMIA trial randomly assigned more than 5000 patients with at least moderate ischemia and documented coronary artery disease (CAD; by coronary CTA) to either an invasive or conservative revascularization strategy. There was no significant difference in the primary endpoint of cardiovascular death (CVD), myocardial infarction (MI), unstable angina, heart failure (HF), or resuscitated cardiac arrest.

The coolest thing about ISCHEMIA was that these similar outcomes in hard endpoints were achieved despite the fact that only 20% in the invasive arm even had revascularization (Supplement Figure s6b). To me, this strongly supports the notion that stable CAD is highly stable and amendable to disease-modifying medical therapy.

Second, the authors of this new subanalysis paper did two major things.

  • First, they looked at how Ischemic burden vs anatomic burden predicted events such as death, CVD, and MI.

  • Second, they evaluated whether ischemic burden vs anatomic burden modified the treatment effect of invasive or delayed therapy

On the prediction front, the authors used mild ischemia as the comparator, and then looked at how mild ischemia compared with moderate or severe ischemia in the prediction of future events. You might think, wait, I thought you had to have moderate to severe ischemia to get into the trial. So how was there “mild” ischemia? Because the site that randomly assigned patients considered it moderate ischemia but the core lab called it mild. This occurred in roughly 15%, so this 15% was the control arm for the prediction.

What they found was that relative to mild ischemia, neither moderate ischemia nor severe ischemia was associated with increased mortality. But it was different when they looked at the anatomic complexity of CAD. Here the authors separated the results of coronary CTA into six categories, called the modified Duke Prognostic Index.

Again, they take the lowest severity as a control, which was one-vessel disease > 50%, and found increasing CAD severity — by anatomy — was associated with a 2.7-fold higher death rate and a 3.7 times higher rate of MI for the most vs least severe CAD subgroup (the confidence intervals (CI) were very wide).

Next, they asked: Did ischemic burden or anatomic severity modify the treatment effect of invasive or conservative. Again, there was no difference in the main trial. Nor was there a signal here: Ischemia severity did not identify a subgroup with treatment benefit on mortality, MI (the trial primary end point), CVD, or MI.

In the most severe anatomic disease group (triple vessel disease > 70% or two-vessel disease with a proximal left anterior descending lesion), the 4-year all-cause mortality was also similar.

Comments. The heart surgeons focused on this topic because they are (rightly) upset about the recent downgrade of coronary artery bypass grafting (CABG) in the severe anatomic category from a Grade I to a IIb recommendation. I struggle to understand why CABG would be downgraded. Surely it is not from the nonsignificant results of ISCHEMIA regarding the invasive approach. I say that because only 20% of the invasive arm had surgical revascularization. The vast majority of revascularization in ISCHEMIA was by percutaneous coronary intervention (PCI) and there is no way to extrapolate findings from that small a subset of patients. Now you might push back and say, come on Mandrola, this substudy you just covered found no benefit of invasive over conservative treatment even in those with the worst CAD. So maybe CABG should be downgraded. Again, I’d say no, because most patients in ISCHEMIA were revascularized with PCI, and the subgroup of severe CAD included 650 patients, which was only about 12%.

Recall that PCI relieves obstructive lesions; however, we don’t know if they are the one that are causing future events. CABG, as the editorialists write, could be considered coronary artery plaque-bypassing surgery.

That said, I would love to see a new trial comparing modern medical therapy vs CABG even in stable patients with left main disease. Medical therapy has improved a lot and it stabilizes all plaques.

As for the discrepancies that sat in a major journal for months, this is problematic. Where was the vetting, the peer review, the editors, etc? To me, this sort of thing is an argument for open data. Why would we not have journal publication contingent on release of the raw data? Especially since ISCHEMIA was publicly funded. But surely open data is even more important for industry-sponsored studies.

Consider what a Neutral Martian would think. She might say, wait, you are publishing science that will affect the care of millions of patients and no one else can see the raw data?

Is ischemia dead as a predictor of events? Should we just go to anatomic imaging with coronary CTA, as my friends from Edinburgh suggest in the accompanying editorial? I would say no, and respectfully disagree with my friends in Scotland. Here is why: in the editorial, Drs Newby, Williams, and Dweck argue that this ISCHEMIA substudy counters a COURAGE substudy that found that PCI may be more beneficial in patients with high ischemic burden.

Their point: unlike in COURAGE, this substudy of ISCHEMIA did not find that ischemia predicted benefit of early intervention, nor did it find that ischemic burden predicted events. They said anatomy looked better. Here’s the first problem with their argument:

  • That substudy of COURAGE that “tentatively suggested” revascularization may have some benefit in those with high ischemic burden actually found no significant effect of PCI in high ischemic burden after adjustment for baseline differences.

  • The other problem with the tenet that COURAGE may have shown that PCI may offer benefit in patients with high ischemia burden is another substudy, by the same group, which found opposite results:

    “At baseline, moderate to severe ischemia occurred in more than one-quarter of patients. Death or MI was similar in those with moderate to severe ischemia, and there was no gradient increase in events for the overall cohort with the extent of ischemia.”

The second problem with the argument by Newby and colleagues is that they then cite two other trials that they say back up the value of anatomy: PROMISE and SCOT-HEART. I am scratching my head because PROMISE was nonsignificant.

  • PROMISE was an RCT of functional (stress testing) vs coronary CTA (anatomy) in patients with symptoms of CAD, but PROMISE was stone cold negative: Major adverse coronary events (MACE) occurred in 3.3% of the CTA group vs 3% in functional stress test arm. How does this back up anatomy as better than ischemic testing?

  • And as for SCOT-HEART, you have to believe that an imaging test reduced major clinical outcomes by 40% when only a small fraction more patients in the CTA arm received statins and ASA. I wrote a column explaining my concerns with SCOT-HEART.

Thus, this ISCHEMIA substudy does not warrant favoring CTA over ischemic testing. I see CTA as having a role in ruling out left main disease. I wish you could simply block at all the rest of the scan. I say that because in the United States, the presence of significant disease often leads to the cascade of catheterization and revascularization.

I know this may not be true at places like Columbia University, but I’d estimate in 90% of America, cardiologists don’t say, yep, you’ve got multivessel CAD, here are some tablets. Positive CTAs go to the lab. And from there it is a short trip to a metal cage.

In my zip code, it’s not uncommon for my interventional cardiology colleagues to text me “Hey John, I did another cath for positive coronary artery calcium (CAC) scan.” If CAC is driving angiography, then you know CTA will.

I know things may be different in Europe but my worry with CTA as a first test is incidental disease and overtreatment of CAD. Plus, as an electrophysiologist, I see tremendous value in the other things one gets from an exercise stress: functional capacity and rhythm issues for example.


JAMA has published the large pragmatic UK TAVI vs SAVR for patients with severe aortic stenosis (AS). This was a special and neat trial:

  • 34 UK centers; 913 patients; and transaortic valve implantation (TAVI) could be with any approved valve.

  • The primary outcome was all-cause death at 1 year.

  • It was a noninferiority (NI) trial that used a risk difference of 5% for the NI margin. There were oodles of secondary endpoints.

Every center that did TAVI in the United Kingdom were included as centers. This was a publicly funded trial with very few exclusion criteria. The trial was conceived in 2009, but things changed, and patients in this trial were actually similar to those in the low-risk PARTNER and EVOLUT low-risk trials. The median STS (Society of Thoracic Surgeons) risk score in the UK trial was only 2.6%, so the trial population would conventionally be classified as low risk, even though the mean age was 81 years.

The results:

  • At 1 year, there were 21 deaths (4.6%) in the TAVI group and 30 deaths (6.6%) in the surgery group, with an adjusted absolute risk difference of −2.0%; the upper bound or worst-case scenario was 1.2% which was way less than the 5% NI margin, so TAVI easily met NI.

  • The relative risk reduction: The adjusted hazard ratio (HR) for death from any cause was 0.69 (95% CI, 0.38 to 1.26)

  • Of 30 prespecified secondary outcomes reported, 24 showed no significant difference at 1 year.

  • Stroke was nearly 2 times higher and it nearly met significance; HR 1.98 with CI 0.95 – 4.11 and P of 0.07.

  • TAVI had fewer major bleeds, shorter hospital stay but higher pacer rates, more vascular complications, and more aortic insufficinecy. Exactly as you’d expect.

Comments. You’ve got to love the National Health Service for trials like these. Pragmatic trials with few exclusion criteria that include all implanting centers and any valve are valuable because of their generalizability or high external validity. So, kudos to the authors here.

Sanjay Kaul sent a slide to our What’sApp group that considered the pluses and minuses if 1000 patients like those in UK TAVI were treated with TAVI vs surgery.

On the plus side, TAVI over surgery:

  • 20 deaths prevented (2% x 1000) (note this is not significant for superiority);

  • 130 fewer major bleeds (13% ARR x 1000);

  • Shorter hospital stays;

  • No sternotomy.

On the negative side:

  • 26 more strokes;

  • 78 more vascular complications;

  • 69 excess pacers;

  • 283 more cases of mild to moderate AI.

Finally, the main limitation here is that this is 1-year data. We know from the 2-year PARTNER 3 data that early death advantages of TAVI lessen over time. My take on TAVI vs surgical aortic valve replacement (SAVR) remains the same: For high surgical risk patients with limited life expectancy, TAVI is a godsend; for younger lower-risk patients, I just don’t think we have enough long-term data. I totally understand that most patients want to avoid sternotomy, but I don’t think we can discount the higher rates of AI, pacers, and vascular complications with TAVI.

TAVI for Aortic Insufficiency?

It’s fun to watch iterations. For years, we have been talking TAVI only for AS. But at the EuroPCR meeting in Paris, we heard preliminary results of a single-arm feasibility study for TAVI for AI, in 45 patients, mean age, 77 years.

  • Patients had moderate to severe AI and a very high surgical risk. Half had an ejection fraction (EF) of 50% or less and three-fourths had Class 3 or 4 HF.

  • At discharge, the average aortic valve gradient was low and the aortic valve area large at 2.62 cm2.

  • No patient had moderate or severe paravalvular regurgitation and 91% had no aortic regurgitation (AR) or just trace AR at discharge (55.5% none, 35.5% trace, and 9% mild).

  • But 23% got pacers.

Patrice Wendling has great news coverage. We learn that this particular valve system has CE marking (approval) in Europe – which is shocking from so little data. She tells us that there is an ongoing trial in the United States. It’s called the JenaValve ALIGN-AR Pivotal Trial (ALIGN-AR). Of course, it’s not really a trial because there is no comparator and no blinding.

I won’t say much more. We are in the first 10k of a marathon when it comes to treating AI without surgery. Obviously, AI is a tougher problem for transcatheter valves. But bioengineering is pretty amazing.


Not every day but many days, even in an electrophysiology clinic, the question of how best to keep stents from clotting off comes up. My friends, stents for ST-elevation MI (STEMI) is one of cardiology’s greatest inventions, but there remains the challenges of keeping clots off these metal cages. EuroPCR had an interesting paper on shortening the course of dual antiplatelet therapy (DAPT) after stenting. It’s a bit like Icarus. You’d don’t want to fly too close to the sun with anti-thrombotics, as bleeding will also cause bad outcomes.

MASTER-DAPT was a big trial that studied older patients who had high bleeding risk.

Patients: 40% with stable angina; 25%, non-STEMI; 12 % STEMI; 11% UA — a mixed PCI presentation. They used a biodegradable-polymer sirolimus-eluting stent –which is not used in the United States. Therapy: Abbreviated vs standard therapy DAPT after PCI. Patients had to go 1 month after a PCI with no events. Then they were randomized to abbreviated — going from DAPT to single APT (SAPT) vs standard therapy which meant continued DAPT for at least 5 additional months (6 months after the index procedure) or, for those receiving oral anticoagulants (OAC), for at least 2 additional months (3 months after the index procedure) with the continuation of SAPT thereafter.

The three ranked primary outcomes were:

  • Net adverse clinical events (NACE; a composite of death, MI, stroke, or major bleeding);

  • MACCE (major adverse cardiac or cerebral events,a composite of death, MI, or stroke);

  • Major or clinically relevant nonmajor bleeding; cumulative incidences were assessed at 1 year.

The first two outcomes were tested for NI and the third, bleeding, or safety, was tested for superiority.

The main trial found that 1 month of DAPT was noninferior to the continuation of therapy for at least 2 additional months with regard to the occurrence of NACE and MACCE. Abbreviated therapy also resulted in a lower incidence of major or clinically relevant nonmajor bleeding.

The most recent trial considered the question of abbreviated DAPT in patients who had complex PCI. Surely, goes the conventional thinking, those who have complex PCI, numerous vessels, numerous stents, etc, would benefit from longer DAPT. Well, about a third of the 4400 patients in the main trial had complex PCI.

Now you see where they are going. They can do a comparison of outcomes for the two groups — complex and noncomplex PCI. It was a prespecified analysis, meaning they planned to do this before the trial. And boom, the complexity of the PCI in these high bleeding risk patients did not matter.

DAPT discontinuation was associated with similar NACE and MACCE and lower bleeding rates compared with standard DAPT, regardless of PCI or patient complexity.

Comments. The simple conclusion is that you are tempted to just make it simple: shorter DAPT for any patient with high bleeding risk. But there are caveats, and many of them are captured beautifully by journalist Sue Hughes.

  • This was a special biodegradable DES stent not used in the United States. Would the results be similar with a different stent?

  • There is no universally accepted definition of complex PCI.

  • The NI margins are wide, meaning there is uncertainty.

  • MASTER-DAPT had small numbers of STEMI. Can we extrapolate this to STEMI patients?

  • MASTER DAPT used a lot of clopidpgrel monotherapy rather than aspirin.

I’m still calling my interventional cardiology buddies a lot. But it sure looks like less-is-more is proving its mettle once again.


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.