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 Nov 8, 2024, John Mandrola, MD, comments on the following news and features stories: EARLY TAVR, colchicine, and TRISCEND II.
First, I want to say thank you to my Brazilian colleagues for the kind invitation to the Brazilian Electrophysiology (EP) Society meeting in Curitiba. I gave five talks, including the plenary session on evidence-based medicine. It’s my third time in Brazil and I remain impressed regarding the quality of the EP and interest in critical appraisal in that country. Thank you SOBRAC.
I record to today from the growing city of Rzeszow, Poland. I am speaking at Piotr Futyma’s PQRST EP Congress. This one-day event has become quite a prominent meeting, drawing colleagues from all over Poland. The story of Piotr Futyma is an interesting one. He is a self-made academic and master ablation expert. When not ablating here in Southern Poland, experts in Europe bring him to help with their toughest cases. I had him in my lab recently to help with two patients with left vetricular summit PVCs.
How did Piotr and this congress become so well known? Mostly through social media and his humility regarding his skills. This is my fifth time as faculty at PQRST and it makes me proud to be invited.
The EARLY TAVR Trial
EARLY TAVR is a trial of transcatheter aortic valve replacement (TAVR) vs clinical surveillance in patients with severe but asymptomatic aortic stenosis (AS).
Two comments before I tell you about the trial: one is that few things in medicine worry me more than intervening in people who complain of nothing. Another is that while I love strategy trials, the truth is that they have more noise than typical trials and, therefore, they require strict attention to trial procedures — such things as choosing a proper primary endpoint.
The question being asked in EARLY TAVR is when to operate on patients with severe AS. When we translate the data, we also have to consider which type of aortic surgery — TAVR or surgical AVR (SAVR).
I, like you see patients with severe AS. They may have atrial fibrillation (AF), or a pacemaker, or an implantable cardioverter defibrillator (ICD). AS is on their problem list. I follow their echocardiograms. In days of old, I worried a lot less about their images, because in days of old, we went by the three main symptoms.
Heart failure symptoms, such as dyspnea on exertion, orthopnea, edema, and syncope, especially during exercise, or angina. If patients felt well, then we withheld alarm at the “shadows” on the ultrasound.
But then two small surgical trials, very small trials indeed, RECOVERY and AVATAR, found that early surgical intervention in patients with severe asymptomatic AS proved beneficial. The RECOVERY trial was especially positive as it showed a 91% reduction in the primary endpoint of operative mortality or cardiovascular (CV) death. The problem with these trials were the small numbers of patients and events.
EARLY TAVR was designed to enroll more patients and to test TAVR, not SAVR, in asymptomatic patients. EARLY TAVR was a strategy trial.
Patients had to prove they were asymptomatic. The vast majority had a negative stress test. Patients also had to have suitable anatomy, not only for TAVR, but the balloon-expandable version of TAVR, as the trial was sponsored by Edwards’ Life Sciences, the maker of balloon-expandable valves. In addition, I offer this copy and paste from the NEJM:
The sponsor funded all trial-related activities, participated in site selection, oversaw data collection and monitoring, and performed analyses according to the statistical analysis plan (available with the protocol).
The main problem with this strategy trial came in the choice of the composite primary endpoint of death, stroke, and unplanned hospitalization for CV causes. The latter component of unplanned hospitalization for CV causes is the problem. Of course it is. I will come back to that.
Also note that patients knew that they were in a trial. One group gets a valve intervention because, as they understand it, they have severe valvular heart disease. The other group gets no surgery, despite having severe valvular heart disease.
Unlike the prior surgical trials, which enrolled only 150 patients, EARLY TAVR screened 1600 patients and enrolled 900.
Patients were age 76 years, 30% female, and most notably, they had a very low surgical risk, with an STS (Society of Thoracic Surgeons) score of only 1.8.
Over a median follow-up of 3.8 years, the primary endpoint occurred in 26.8% in the TAVR group as compared with 45.3% in the clinical surveillance group (hazard ratio [HR], 0.50; 95% confidence interval [CI], 0.40 to 0.63; P<0.001). Note the massive absolute risk reduction (ARR) of nearly 19%. That’s an NNT of 5 — a hugely positive result!
But we have to look at the individual components.
Death was not significantly different (8.4% vs 9.2%).
Stroke was not significantly different, though it was slightly lower in the TAVR arm, 4.2% vs 6.7%. The 38% lower rate of stroke in the TAVR arm did not reach statistical significance due to the low numbers of events.
The key component of the primary endpoint was the unplanned hospitalizations. 21% TAVR vs 42% in the surveillance arm. That is an HR of 0.43 or a 57% lower rate in the TAVR arm. In the main figure, the Kaplan Meier (KM) curves of unplanned hospitalizations and the primary endpoint were super-imposable.
Secondary outcomes also favored the TAVR arm.
In the surveillance group, conversion to TAVR was fast and robust. More than one in four patients had TAVR at 6 months, nearly half at 1 year, and 72% had TAVR at 2 years.
Figure 2 shows the clinical reasons for conversion to TAVR. It was almost always because of progressive symptoms and signs.
Comments. I’ve said this many times, but I have to say it again: I love trials, we need more trials, but this is not the right way to do trials.
The bias was baked into EARLY TAVR before the first patient was enrolled. It’s all in the trial procedures. These were patients who knew they had severe heart disease (HD). One group got early surgery. The other group, who also knew they had severe valvular HD, did not get treatment.
Christopher Rajkumar and colleagues at Imperial college coined this issue faith healing and subtraction anxiety.
The group that gets the active treatment is faith healed. The group that gets no surgery gets subtraction anxiety. Rajkumar and colleagues used the FAME-2 trial as a classic example. EARLY TAVR is an almost identical issue.
First FAME-2, then EARLY TAVR.
In FAME-2 patients knew they had a serious coronary lesion (as documented by fractional flow reserve [FFR]). One group got percutaneous coronary intervention (PCI; the faith healed), one group got medical therapy and follow-up (the subtracted group).
FAME-2 was positive but the three-component endpoint was driven almost totally by urgent revascularization. Hard endpoints, such as death and myocardial infarction (MI), were similar.
Rajkumar and colleagues were able to describe the reason for urgent revascularization.
In the PCI arm, patients were revascularized in equal proportions because of ECG changes, troponin, and symptoms.
But in the medical arm, the subtracted group, the group who knew they had a severe lesion that was not fixed, urgent revascularization was almost always related to symptoms.
Why? Because they were told they had severe coronary artery disease (CAD) and did not get PCI. They weren’t “fixed”, so the nanosecond they presented with even a twinge of pain, boom, they went into the cath lab.
Plus, the idea of having an unfixed coronary lesion surely created anxiety which may have led to symptoms.
The crazy thing is that FFR was largely established as practice based on FAME-2 and it was a biased trial due to the lack of blinding and choice of endpoint.
The exact same thing happened in EARLY TAVR. Patients were told they had severe AS. The valve was a pinhole. In the trial, one group would get the fix, the other group would not. The early treatment arm believed they would be fixed. They had been healed by faith.
Patients in the clinical surveillance arm knew they had bad disease but were not fixed. They may have felt like they had a time-bomb in their chest.
Two factors nearly prove the existence of subtraction anxiety in the clinical surveillance arm of EARLY TAVR.
First was the near immediate separation of the KM curves for unplanned hospitalizations. That should not happen because these were patients who just passed a formal exercise test.
The second suggestive feature was the observation that most patients converted to the procedure not because of acute problems but progressive signs and symptoms; a near exact replica of FAME-2.
EARLY TAVR found no significant differences in death or stroke. If not for the choice of unplanned hospitalization, clinical surveillance would have performed well, and there would be no enthusiasm for early operations.
But now, and solely because of the choice of a biased endpoint in an industry-sponsored trial, guidelines will change, key opinion leaders will lecture, review articles will be published, and cardiologists the world over will begin sending patients for early surgeries.
Sadly, I have more. The infusion of bias because of the endpoint choice is bad. But that is not all the bad news about EARLY TAVR.
The other glaring issue is why there were only two groups in this trial. Why would the authors exclude SAVR? The trial enrolled relatively young patients with low-comorbidity (they walked on a treadmill and had low STS-PROM scores).
I do not understand, nor did I read in the manuscript, why a surgical valve replacement was not considered. In fact, the two previous trials that established the equipoise underpinning EARLY TAVR were surgical trials.
Longer-term results (5 years) of the PARTNER-3 trial comparing balloon-expandable TAVR vs SAVR clearly show a catch-up benefit for SAVR. Death, stroke, and hospitalization rates at longer-term follow-up were similar. SAVR had the advantage of lower rates of aortic insufficiency and valve thrombosis.
The point is not to relitigate TAVR vs SAVR, but to question the trialists’ decision not to include surgical valve replacement in these patients.
I wish it weren’t so. But the design and choices made in EARLY TAVR tempt me to be cynical about the scientific enterprise. Choosing unplanned hospitalization in this trial guaranteed a “positive” result.
That choice and the omission of a surgical aortic valve option causes a failure to inform the decision regarding the timing of aortic valve intervention in patients with asymptomatic but severe AS.
Colchicine
It is almost as if doctors sensed something about colchicine. Despite two positive trials (COLCOT in post-MI patients and LoDoCo in chronic CAD), prescription of colchicine for secondary prevention was low. I don’t think I have ever seen a patient on low-dose colchicine for prevention of heart disease.
The reduction in the composite primary endpoint of COLCOT and LoDoCo were 23% and 31%, respectively. Both were statistically significant but with small absolute risk reductions, and neither trial showed a difference in CV death. Only non-fatal endpoints.
A meta-analysis, including these trials, and four others, also showed a similar risk reduction with colchicine for secondary prevention.
Things took a different turn at the recent Transcatheter Cardiovascular Therapeutics (TCT) meeting in Washington.
Professor Jolly, from McMaster University in Canada presented the non-significant results of the CLEAR SYNERGY trial.
The randomized controlled trial, from 104 sites on five continents randomly assigned more than 7000 patients shortly after an MI and drug-eluting stent to receive colchicine vs placebo. It was actually a 2x2 factorial design with a spironolactone arm, which we did not hear about.
CLEAR SYNERGY enrolled 60-year-old patients, about 80% men (similar characteristics to COLCOT). The primary endpoint was a composite of CV death, MI, stroke, or ischemia-driven revascularization. Follow-up was about 3.5 years.
The primary endpoints were similar to the previous trials but follow-up in CLEAR-Synergy was longer and there were twice as many events as both previous trials. What’s more, there was evidence of decreased inflammation –as shown by greater reduction of CRP values in the colchicine arm.
CLEAR SYNERGY was powered to find a 25% reduction in the primary endpoint, assuming a 9% control event. And they guessed correctly. There was a 9.3% rate of primary outcome event in the control arm, and nearly the exact same rate of events in the colchicine arm, at 9.1%.
CLEAR SYNERGY found absolutely no difference in outcomes, despite double the number of events. CV death, MI, stroke, and ischemia-driven revascularization were nearly identical. The on-treatment analysis was also no difference.
Adverse events overall were similar but there was more diarrhea in the colchicine arm, 10.6% vs 6.6%.
Subgroup analysis showed no signal of heterogenous treatment effect based on selected subgroups.
Comments. This is hard, isn’t it? CLEAR SYNERGY is clearly negative, but absolutely not in synergy with the previous two secondary prevention trials.
You can approach this problem two ways: one is to say the larger trial with twice the number of events takes precedence. Statistically, CLEAR SYNERGY is stronger than either of the two previous trials (COLCOTT and LoDoCo). Adding to this argument would be the incredibly strong post-MI medical care in the CLEAR SYNERGY trial — 97% on aspirin, nearly all on a P2Y12 inhibitor, 97% on a statin, and nearly 80% on an ACE or ARB. Background medical therapy plus a drug eluting stent during MI leaves little room to remove residual risk.
I am surprised this hasn’t been done yet, but adding CLEAR SYNERGY to a meta-analysis would likely drive the relative risk reduction (RRR) to the null, since the 7000 patients in the trial were nearly the same as the total in the meta-analysis of the previous trials.
I suspect that the general opinion for colchicine as a clinically meaningful risk reducer before CLEAR SYNERGY was pretty pessimistic, so the trial will reinforce this consensus.
But, of course, there is another way to translate the results: you could say that CLEAR SYNERGY was the clear outlier. COLCOT and LoDoCo outputted very similar RRR). In the meta-analysis of six trials, the Forest plot shows little heterogeneity with RRR all lining up in favor of colchicine. In such a Forest plot, CLEAR SYNERGY would be the outlier.
You could go with the six previous trials.
I favor the former argument — namely that colchicine may have had a statistically significant RRR, but the absolute risk reductions were not terribly clinically meaningful, at least not enough to justify an extra pill, which could cause GI upset.
Perhaps if you had a pt with severe CAD and high levels of CRP, colchicine could be considered, but in the average post-MI patient, I think CLEAR SYNERGY supports the general cardiology community’s pessimism regarding this drug as a residual-risk reducer.
TRISCEND II Trial
Treating patients with tricuspid regurgitation (TR) can be a challenge. In the vast majority, the TR is secondary to left sided processes. We say the TR is a secondary.
In some cases, and it is hard to know exactly, the TR is more of a primary process. When this happens, the patient can suffer with venous congestion. Back up occurs in the abdomen, leading to ascites or hepatic congestion, and/or in the periphery, causing profound edema.
Before there were structural interventions for TR, surgeons would try operating on the valve. But primary tricuspid valve (TCV) surgery came with a high morbidity and mortality.
In recent years, innovators have introduced two interventions; one is tricuspid transcatheter edge-to-edge repair, using clips to reduce TR. It has the advantage of being less invasive than the second approach, which is a full transcatheter valve replacement (TVR).
Edwards Life Sciences makes the EVOQUE transcatheter tricuspid valve replacement (TVR). The full results of the TRISCEND II trial were presented at the recent TCT meeting, and were published in NEJM, along with two other papers in JACC and JAMA-Cardiology.
TRISCEND II randomly assigned 400 patients with symptomatic severe TR in a 2:1 fashion to TVR or medical therapy. These were 79-year-old patients, three-quarters female, and more than 90% with AF. About 70% had a class 3 or 4 New York Heart Association (NYHA) functional class. Half had chronic kidney disease, and nearly 40% already had a pacing device.
The cause of TR was determined to be secondary in three-quarters of patients and about one-eighth had primary TR with the rest being mixed.
Before I tell you the primary endpoint of this experiment, pause to think how you would compare the two treatment strategies of valve replacement vs tablets, mostly diuretics. Would you measure unbiased endpoints like death or CV death? Or would you introduce more subjective endpoints, like heart failure hospitalizations (HHF), quality of life (QOL), and walk distances, etc?
Well, the way the trialists did it was to measure many endpoints and combine them in a big composite, to be analyzed in a stepwise hierarchical fashion.
First was death from any cause;
Second, implantation of a right ventricular assist device or heart transplantation;
Third, post index tricuspid-valve intervention;
Fourth, HHF;
Fifth, an improvement of at least 10 points in the score on the Kansas City Cardiomyopathy Questionnaire overall summary (KCCQ-OS);
Sixth, an improvement of at least one NYHA functional class;
Seventh, an improvement of at least 30 meters on the 6-minute walk distance.
Seven endpoints. Analyzed by the win-ratio, which compares patient pairs.
Not surprising to you was that TRISCEND II was a positive trial. The overall win-ratio, determined by dividing the total wins by total losses, was 2.02 and the CIs were 1.56 to 2.62.
Death at one year, 12.6% vs 15.2%. This was a win-ratio so we don’t get individual statistics. I don’t know whether the 3% absolute RR is significant using typical CI or P-value.
HHF was also slightly lower for the TVR group, 21% vs 26%.
When you look at the KM plots on 0 to 100 scale as shown, it’s hard to see much difference.
Most of the benefit was seen in subjective measures — the KCCQ, the NYHA class, and the 6-minute walk. I could go through the actual numbers but suffice to say they all were better in the group that got the bigger treatment, as you would expect in any unblinded trial.
Echo parameters also favored the valve group: 72.6% of the patients in the valve-replacement group had no residual TR, 22% had mild regurgitation, 4% had moderate regurgitation, and 1% had severe regurgitation.
Notably, the reduction in TR with the valve replacement is better than when the clips are used.
Now we need to talk safety, because there were notable differences. Safety was assessed at 30 days, not 1 year.
There were nine deaths in the valve replacement arm vs none in the control group. That is not a misspeak. Nine vs 0zero; eight of the nine were CV deaths.
Severe bleeding: 27 patients (10%) in the valve replacement arm vs two patients (1.5). Wow.
Need for pacing: 41 patients (16%) vs zero patients. Double wow. The risk of needing a pacer was 4 -fold.
The overall major adverse event rate is not listed in the main manuscript, but in the supplement it is 27%. You don’t have to worry about the observation that nearly one in three patients have a safety issue because the authors along with the US Food and Drug Administration set a performance goal of 70%.
Comments. I am not sure about this. I realize there is an unmet need for patients with severe TR that is resistant to diuretics, and for treatment of left-sided pathology. These patients struggle. But the number of patients with isolated severe TR for which tricuspid intervention would help is quite low.
Still, we want to have treatment options for these patients.
The problem is testing the devices properly.
I realize the news coverage and spin regarding results are positive. The conclusions read: “Transcatheter tricuspid-valve replacement was superior to medical therapy alone for the primary composite outcome, driven primarily by improvements in symptoms and quality of life.” There is no mention of the near 30% complication rate.
I don’t see much to like here.
First, the bias resistant endpoints were not reduced. The claimed efficacy comes from patient-reported QOL measures, which are severely biased in an open label trial. Again, if we, as a scientific field, were rigorous, we would test these devices with placebo (or sham) controls. It would be entirely possible to do.
Second, though the echo parameters are better, echo parameters are a surrogate outcome. Echo images are only scientifically useful if they indicate or correlate with improvement in harder outcomes (like lower death rates or HHF). Here they did not. Proponents could say the degree of TR reduction correlates with QOL improvement, but that has to be rejected without a sham arm.
Third, I don’t care what the made-up performance goal was, this was a highly morbid procedure.
Nine vs no deaths in the first 30 days. Five-fold higher rates of bleeding. 41 times higher rates of pacing, which is not a small thing, because remember, if you have a TVR, you have to get creative with pacing, such as conduction system pacing, leadless pacing, or even epicardial pacing.
What’s more, and the authors barely mention this, 40% of patients had an existing pacer or ICD. That means that these patients now have entrapped leads. This may result in pacing failure, or perhaps worse; if an infection ever occurred it would not be possible to extract the lead.
This is horrible. Please, to all my colleagues in structural cardiology, call your EP before entrapping a lead.
Overall, I am not persuaded by this data. I am not sure how it got approved. Indeed, it is a first-generation device, but as a neutral observer, there is little signal for efficacy in unbiased endpoints, and huge safety concerns.
© 2024 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: Nov 08, 2024 This Week in Cardiology Podcast - Medscape - Nov 08, 2024.
Comments