Sep 17, 2021 This Week in Cardiology Podcast

John M. Mandrola, MD


September 17, 2021

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.

In This Week’s Podcast

For the week ending September 17, 2021, John Mandrola, MD comments on the following news and features stories.


The big COVID news this week is the debate on boosters. It offers a masterclass of evidence-based medicine. The US Food and Drug Administration (FDA) will discuss the evidence today. The agency struck a cautious tone in released documents last Wednesday.

Cardiac Arrest Care

I don’t know how things are where you are, but in my zip code, the cath lab carries a sort of mythical lofty status of a place where sick patients should go. And one of the main reasons people suffer cardiac arrest is myocardial ischemia, especially acute ischemia. When there is ST segment elevation after an arrest, it’s agreed that cath lab activation for urgent revascularization is warranted.

But what about when there is not ST segment elevation? The 2019 COACT trial studied the immediate angiography vs delayed angiography in patients with cardiac arrest who did not have ST-Segment Elevation Myocardial Infarction (STEMI). Immediate angiography did not significantly reduce the primary outcome of overall survival at 90 days. But the Dutch-led COACT rial included only patients with a shockable rhythm. The TOMAHAWK trial, presented at the European Society of Cardiology (ESC) meeting, published in the New England Journal of Medicine (NEJM), and led by a group from Leipzig, Germany, studied the strategies of immediate vs delayed coronary angiography in 550 patients with successfully-resuscitated, out of hospital cardiac arrest, who did not have STEMI.

  • At 30 days, 54.0% in the immediate-angiography group and 46.0% in the delayed-angiography group had died (hazard ratio (HR), 1.28; 95% confidence interval [CI], 1.00 to 1.63; P=0.06).

  • The composite of death or severe neurologic deficit occurred more frequently in the immediate-angiography group, 64% vs 55% in the delayed group, for a relative risk of 1.16 (95% CI, 1.00 to 1.34)

  • The authors concluded that the strategy of performing immediate angiography provided no benefit over a delayed or selective strategy with respect to the 30-day risk of death from any cause.

Comments. First some specifics:

  • Not only wasn’t the immediate strategy better, death was 28% higher with the near entirety of the CI greater than 1, indicating a high probability of harm. The upper bound of the CI allows for a 63% higher death rate with early catheterization.

  • What’s more, the composite of death or neurologic outcomes was also 16% worse and again, the CIs are essentially all above 1.

As the authors point out, this may be a chance finding, but it also could be that catheterization distracts from or delays early diagnosis and treatment of other conditions. The authors also spend many words explaining why immediate angiography didn’t work. I would say that given the strong concordance of two large RCTs, both looking at the same strategy, and both of which show no benefit and TOMAHAWK trending towards harm, it does not matter why the immediate strategy did not work.

Well-done trials like this teach us so much—not just that immediate angiography doesn’t help this cohort of patients without STEMI, but far more important, the German and Dutch trialists have shown us that accepted strategies of care can and should be tested. And that notion could not be more relevant to our current situation with the pandemic.

Finerenone and CV Outcomes

This summer, the FDA approved the special mineralocorticoid receptor blocker (MRA) finerenone for the prevention of chronic kidney disease (CKD) in patients with type 2 diabetes (T2D). Finerinone is a special MRA because it is nonsteroidal and will not have the antiandrogen effects of spironolactone.

This was based on the FIDELIO-DKD trial, which randomly assigned about 5700 patients with T2D and CKD to finerinone or placebo and found a highly significant, 18% lower risk of CKD progression.

That trial also included cardiovascular (CV) outcomes as key secondary outcomes. Finerinone reduced the risk of the composite CV outcome compared with placebo, with no significant interaction between patients with and without CV death. In this trial, three of the four components of the composite endpoint trended lower with finerinone: CV death, MI, and hospitalization for heart failure (HHF), while stroke was similar.

At ESC, we learned the results of the CV outcomes trial called FIGARO-DKD. About 7200 patients with CKD and T2D were randomly assigned to finerinone vs placebo. The primary outcome was a composite of CV death, MI, stroke, and HHF. Similar to FIDELIO-DKD, patients had to have urinary albumin. In FIGARO, lower levels of albuminuria were allowed.

  • After follow-up of 3.5 years, a primary outcome event occurred in 12.4% in the finerenone group and in 14.2% in the placebo group.

  • The 1.8 absolute risk reduction (ARR) translated to an HR of 0.87, so a 13% reduction, and the 95% CI ranged from 0.76 to 0.98, roughly meaning that the benefit could have been as great as 24% or as little as 2%.

  • The P-value was 0.03. So, if finerinone had no effect, the chance of finding this result or something more extreme was about 1 in 33.

Positive right? Let’s get on the finerinone bandwagon because it reduces cardiac events in patients with T2D.

Not so fast. The first thing to do with these composite outcomes is look at the components. You have some strong ones (CV death, stroke) and some not so strong ones like MI and especially HHF.

  • CVD: 5.3% vs 5.8%; not significant (NS)

  • MI: 2.8% in both; obviously NS

  • Stroke: 2.9% vs 3%; obviously NS

  • HHF: 3.2% vs 4.4%; HR 0.71 (0.56-0.90)

So, you see that HHF drove the 13% benefit seen in the composite endpoints. But, as we went over in the last two weeks in the discussion of EMPEROR-Preserved, if you are going to say HHF is an important endpoint, it had better represent a big chunk of total hospitalizations. Well, if you look at Figure 2 in the NEJM paper, you will see that total hospitalizations or all-cause hospitalizations were 42.7% in the finerinone arm, and 43.8% in the placebo arm. HR a measly 0.97 and totally NS. And that is because in this population of 64-year-old mostly white male patients, HHF represents just 8% of total hospitalizations. Pause there for a moment: how can HHF be an important endpoint when it represents less than 10% of total hospitalizations? The answer is it can’t. This data is totally unconvincing for a CV outcomes drug.

As for the first secondary composite renal outcome of sustained decrease of at least 40% in glomerular filtration rate (GFR), death from renal causes, or kidney failure, there was a 13% reduction with finerinone that just missed statistical significance. As for adverse outcomes, overall, there were no differences, but hyperkalemia was seen in 10.8% vs. 5.3% with placebo.

Comments: Finerenone seems nice because there will likely be better tolerability and maybe less potassium issues. Maybe. And both trials show trends toward improved renal outcomes. Both trials show better CV improvements, though in patients with less severe kidney disease; in FIGARO, the CV benefits were driven by a very weak endpoint of HHF. (Note, I am not saying HHF is always a weak endpoint. It is only weak when it represents a tiny proportion of total hospitalizations).

Another problem for translation of these trials is that only 7% of enrolled patients (diabetic patients with kidney disease) were on SGLT2 inhibitors, so the question remains, how much added benefit will this drug be? While the renal and CV outcomes are evident, they are decidedly modest when applied on an individual level.

Carotid Stenting vs Carotid Endarterectomy – ACST-2 Trial

At ESC, investigators from the United Kingdom presented the ACST-2 RCT of carotid stenting (CAS) vs carotid endarterectomy (CEA) in patients with asymptomatic carotid disease that was felt to require intervention. Lancet published the paper.

There were two primary outcomes: periprocedural risk defined as MI, stroke, or death within 30 days/ long-term rates of disabling or fatal stroke during follow-up of patients. The trial started 13 years ago, in 2008. About 3600 patients were enrolled in 130 centers, with 1800 in each group.

Periprocedural Outcomes:

  • Stroke (total): 3.6% in CAS vs 2.4% in the CEA, P = 0.06

  • MI (total): 5 (0.3%) vs 12 (0.7%), P = 0.15

  • Death: 2 vs 2

  • Death, MI, or any stroke: 3.9% CAS vs 3.2% CEA, P = 0.26

Investigators noted an increase in non-disabling stroke within the first 30 days: 2.7% vs 1.6% (P=0.03). The total rate of any non-procedural stroke: 5.2% CAS vs 4.5% CEA. And the total rate of procedural death or ANY stroke was 8.6% CAS vs 7.0% after CEA. So slightly worse for CAS, but the P-value is 0.09.

The topline results of the trial were that the results are technically similar. The authors say: ”...Main finding was that the effects of the two procedures on disabling or fatal events are approximately equal in terms of procedural hazards (about 1% for each treatment, in line with findings from large, representative registries) and of 5-year disabling stroke rates (which were about 0·5% per year with either procedure).”

I would also add that no CAS patient had a cranial nerve injury whereas 5% in the CEA group had it at one-month.

Three main comments: There remains a great debate on whether a patient with no symptoms who has a carotid lesion should have an intervention. The gamble is that the cumulative in ipsilateral stroke reduction from “fixing that blockage” over and above medical therapy will be greater than the upfront risk. When you look at the as-treated analysis, CAS has > 3% risk of procedural stroke or death, with nearly 3% risk in the CEA arm. To me, that seems a big hurdle. This is quite like the issue with Watchman or any other preventive procedure or surgery—there is a large risk upfront. So, if the gains were great enough, and the patient lives long enough, there may be benefit, but without a medical arm, we just don’t know. The authors cite previous data suggesting that CEA may halve the risk of stroke relative to medical therapy. The problem of course is that these studies were done decades ago; medical therapy has improved, and the rates of cardiovascular events have declined.

Further, as we talked about in an interview I did with ACST-2 author, Richard Bulbulla, the approach to asymptomatic carotid disease varies a lot by country. In some countries it is common for asymptomatic patients to have carotid interventions; in other countries the rate of carotid interventions in asymptomatic patients is nearly zero.

My second comment is that while the rates were close, there does seem to be numerically higher stroke rates in the CAS group, both peri-procedurally, and later. In the supplement, the figure that caught my eye was the rate of procedural death, procedural stroke, or non-procedural ipsilateral stroke. Here the rate was 5.3% vs 3.6% in favor of CEA. Almost 2% higher in stent arm. But CAS had fewer MIs, no cranial nerve injuries, and shorter stays.

It seems that this information could be incorporated into a decision support tool, and with the guidance of a wise clinician skilled in both types of procedure, decisions could be individualized based on the patient’s unique features and goals. But to translate this evidence to the bedside, you have to have some reassurance that performance of either procedure in the real world, would be similar to the carefully selected centers in ACST-2. Given the improvements in medical therapy, which leads to lower baseline rates of stroke, any slight increase in upfront harm negates the probability benefit.

Final issue: I wonder whether technology might help adjudicate which of these bad lesions may benefit more from revascularization—things such as transcranial doppler, intraplaque hemorrhage seen on MRI, reduced cerebral blood flow reserve.

Protection after TAVI when AF is present

Four facts:

  • Atrial fibrillation (AF) and aortic stenosis are common conditions that afflict older patients.

  • Patients who have AF benefit from anticoagulation (AC).

  • Direct oral ACs (DOACs) beat warfarin in trials of patients with AF.

  • Patients who have transcatheter aortic valve replacement (TAVR) require some degree of anti-thrombotic meds.

Since DOACs appear useful against warfarin, how will they do after TAVR? DOACs haven’t looked good post-TAVR in patients without AF. For instance, GALILEO compared rivaroxaban 10 mg vs aspirin (ASA) post- transcatheter aortic valve implantation (TAVI) and rivaroxaban had a higher risk of death or first thromboembolic event.

ATLANTIS compared apixaban vs warfarin (or ASA) post-TAVI. In the control arm, if there was an indication for AC, patients got warfarin; if not, they got ASA. There was no significant difference in primary outcome of death, stroke, MI, valve thrombosis, pulmonary embolus, venous thromboembolism, or major bleeding: 18.4% vs 20.1%. There were numerically higher numbers of secondary endpoints including death, stroke, heart attack, or systemic embolism in the apixaban group compared with the standard care group. Worse, in the subgroup of patients without a DOAC indication, apixaban had an 86% higher rate of mortality.

The ENVISAGE-TAVI-AF trial compared edoxaban vs warfarin in post-TAVI patients who had AF. About 1400 patients were randomly assigned to edoxaban vs warfarin. Mean age was 82 years. (I love trials that enroll older patients). Half were woman, too. Edoxaban could be dose adjusted and warfarin was targeted to an INR of 2-3 but could be adjusted to 1.6-2.6 for patients older than 70 years. Antiplatelet drugs were allowed at the doctor’s choice. Nearly all had AF.

The primary efficacy outcome was important in its breath: incidence of net adverse clinical events, defined as the composite of death, MI, ischemic stroke, systemic embolism, valve thrombosis, or major bleeding. This endpoint was tested as a non-inferiority of edoxaban if the upper boundary of 95% CI did not exceed 1.38. In other words, edoxaban could be 38% worse and still be noninferior.

The primary safety outcome was major bleeding. The primary efficacy outcome was tested first.

  • The rate of the composite primary efficacy outcome was 17.3 per 100 person-years in the edoxaban group vs 16.5 per 100 person-years in the warfarin group.

  • The point estimate of the hazard ratio was 1.05; 5% worse. And the 95% CI went from 0.85 to 1.31. Since the upper bound of 1.31 was less than 1.38, edoxaban met noninferiority against warfarin.

  • However, rates of major bleeding were 9.7 per 100 person-years in the edoxaban group vs 7.0 per 100 person-years with warfarin, respectively (HR 1.40; 95% CI, 1.03 to 1.91)

What to say about this? First, it is abundantly clear that edoxaban is inferior to warfarin in the bleeding endpoint. But what about efficacy? Well, the choices regarding noninferiority in the first question are debatable; 1.38 as a margin is pretty high. It’s based on vitamin K antagonist studies that used only stroke or systemic embolism as an efficacy endpoint. But in this trial, the primary endpoint was like every bad thing that can happen. In ATLANTIS, the noninferiority margin was 1.2, and if this were the case, ENVISAGE-TAVI would not have met noninferiority.

Noninferiority trials are supposed to be used to test whether a new treatment is just as good, if that new treatment has some other benefit—say improved safety. This DOAC was tested for noninferiority against warfarin for any efficacy—stroke and systemic embolism. That is it. Then they did a superiority analysis for bleeding.

In the case of ENVISAGE-TAVI, you have so many components of the endpoints that they could cancel each other out and thus make noninferiority easier to reach.

I think the message of ENVISAGE-TAVI is clearly that warfarin is the preferred anticoagulant in patients with AF. And putting the message of all three trials together, it sure looks like DOACs are not the preferred agents. What is new is not always better.

I did not get to MASTERDAPT; I underestimated how much study it would take to get a handle on post-PCI antiplatelet regimens. Stay tuned


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.
Post as: