The Take Home: ACC

The American College of Cardiology Scientific Session contained numerous presentations with the potential to change the landscape of interventional cardiology.

Cardiology Today’s Intervention was onsite and spoke to a multitude of experts for their takes on potentially practice-changing data. They included M. Chadi Alraies, MD, from Medstar Washington Hospital Center; Cardiology Today’s Intervention Chief Medical Editor Deepak L. Bhatt, MD, MPH, from Brigham and Women’s Hospital and Harvard Medical School; Chandan Devireddy, MD, from Emory University School of Medicine, Atlanta; Sammy Elmariah, MD, from Massachusetts General Hospital and VA Boston Healthcare System; Cardiology Today’s Intervention Editorial Board Member Howard C. Herrmann, MD, from Hospital of the University of Pennsylvania; Dipti Itchhaporia, MD, from Hoag Memorial Hospital Presbyterian, Newport Beach, California; Farouc A. Jaffer, MD, PhD, from Massachusetts General Hospital and Harvard Medical School; Samir Kapadia, MD, from Cleveland Clinic; Cardiology Today’s Intervention Associate Medical Editor Roxana Mehran, MD, from Icahn School of Medicine at Mount Sinai; Srihari Naidu, MD, from Westchester Medical Center and New York Medical College; Cardiology Today’s Intervention Editorial Board Member William O’Neill, MD, from Henry Ford Hospital, Detroit; Cardiology Today’s Intervention Editorial Board Member Gregg W. Stone, MD, from NewYork-Presbyterian Hospital/Columbia University Medical Center and the Cardiovascular Research Foundation; B. Hadley Wilson, MD, from Sanger Heart & Vascular Institute, Carolinas HealthCare System, Charlotte, North Carolina; and David E. Winchester, MD, MS, from the University of Florida, Gainesville.

SURTAVI

Mehran: This trial is very exciting and offers more excellent data with consistent evidence for transcatheter aortic valve replacement noninferiority compared with surgical AVR for patients with aortic stenosis and intermediate risk, now with a different valve. At 24 months, the estimated incidence of the primary endpoint of all-cause mortality or disabling stroke was 12.6% (95% credible interval, 10.2-15.3) in the TAVR group vs. 14% (95% credible interval, 11.4-17) in the surgery group, translating to a difference of –1.4% (95% credible interval, –5.2 to 2.3) and a posterior probability of noninferiority > 0.999. The proven noninferiority with TAVR in 1,746 patients gives me as a clinician a sense of superiority of TAVR, because there is less hospitalization, better quality of life and less stroke. Those are the superiority aspects — the types of things we may overlook. Having a second valve available in intermediate-risk patients in the United States is going to be incredibly important.

Photo Credit: Cardiology Today’s Intervention

Kapadia: The most important message from SURTAVI is that we confirmed that TAVR and surgical AVR have similar outcomes in intermediate-risk patients. This is an incredibly important landmark for TAVR because it confirms what was found in the PARTNER 2A trial, which now becomes science. It will impact decision-making about who should get TAVR. With two randomized trials showing benefit in this population, the guidelines may be revised to give TAVR the highest indication for use in intermediate-risk patients. It should lead to more confidence in recommending TAVR for intermediate-risk patients, both from the patient perspective and the physician perspective.

The beauty of the results is that the only two complications that were higher in TAVR — paravalvular leak and pacemaker implantation — have been shown to be lower with the newer generation of the device, the CoreValve Evolut R (Medtronic), which is more positionable and has a better seal. A lot of effort has gone in to overcome the limitations of the devices, and the newer versions are already addressing those issues. We also expected to see a lower rate of complications for TAVR in acute kidney injury, AF and bleeding. In its current form, TAVR has improved considerably from the system used in SURTAVI, while surgery, which is an outstanding option, has not changed much.

Patients and physicians should feel more comfortable with TAVR as an alternative to surgery. This is a tremendous outcome.

Stone: A study of great interest was the SURTAVI trial, in which patients with aortic stenosis at intermediate surgical risk were randomized to either classic surgical aortic valve replacement or to a self-expanding super-angular transcatheter aortic valve. A previous trial, PARTNER II, used a different type of valve and showed that TAVR was at least as good, if not better at relieving mortality and stroke in this controversial patient population because it’s only intermediate risk for surgery.

Essentially, the same findings were seen here. There was a very similar rate of all-cause mortality and stroke at 1 year between the TAVR arm and the surgical aortic valve replacement arm. There was a higher pacemaker need, but on the other hand, there were the usual surgical type of complications that are avoided by the transcatheter approach. Overall, I think this is another option for patients with intermediate surgical aortic stenosis.

ABSORB III

Stone: A highlight from ACC was the 2-year results of the ABSORB III trial, a large-scale, 2,000-patient randomized trial of a bioresorbable vascular scaffold (Absorb GT1, Abbott Vascular) compared with a metallic everolimus-eluting stent. The 2-year rate of target lesion failure was 11% with the BVS vs. 7.9% with the EES (HR = 1.42; 95% CI, 1.04-1.94; P = .03). In patients with reference vessel diameter 2.25 mm by quantitative coronary angiography, 2-year TLF — a composite of cardiac death, target vessel MI and ischemia-driven target lesion revascularization — was 9.4% vs. 7%, respectively (HR = 1.35; 95% CI, 0.93-1.96; P = .11). The rate of definite/probable stent thrombosis was 1.9% with the BVS vs. 0.8% with the EES at 2 years, which was slightly reduced among patients with appropriate-sized vessels (1.3% vs. 0.6%).

The concept of the BVS is that once they completely dissolve at about 3 years, patients will have a better long-term prognosis because they won’t have any stent-related effects by not having the stent in place anymore in the coronary arteries.

The question is, before the dissolution happens, are event rates increased in patients who have received BVS with this first-generation technology? ABSORB was the largest trial done to date and with the data out to 2 years, there was a slight increase in target lesion failure events in patients randomized to Absorb compared with a state-of-the-art metallic DES (Xience, Abbott Vascular).

On the other hand, there was a relatively very small number of patients treated with optimal technique, and we’ve learned that you can’t treat the BVS scaffold the same way you would a metallic DES. Patients who had optimal technique had a much more similar outcome to metallic DES, and we look forward to the ongoing ABSORB IV trial, where we’ve seen some glimpses of the data in which patients are treated more optimally, and it looks like the event rates have been cut down by more than 50%.

Bhatt: ABSORB III generated a lot of discussion. In fact, even before the presentation, the FDA released a warning about the bioresorbable scaffold and it caused a lot of noise here. The FDA raised a concern that there was a significantly higher rate of MACE with the bioresorbable scaffold vs a very good second generation drug eluting stent.

A lot of people discussed the twofold higher rate of stent thrombosis that we saw. Granted, a lot of that risk is more up front in the first year and the P value wasn’t quite statistically significant.

The ABSORB investigators contributed a lot to the learning curve. They learned to not use the bioresorbable scaffold in small vessels, to make sure there’s proper lesion preparation and to make sure there’s ample use of IVUS and/or OCT to verify it’s well-apposed.

With all those maneuvers, it seems like the complication rate can be mitigated, but still, this is a concerning finding. At least to the extent of follow-up to date, there is no advantage to the BVS.

For those doctors that are implanting these stents, since they are FDA-approved for certain indications, be cautious with your implantation technique, with the imaging and with the appropriate use of dual antiplatelet thereafter. Nobody knows what that appropriate duration is, but I would favor longer instead of shorter, given the stent thrombosis finding, which I think is a real finding. I would say be judicious in your decision making in terms of why you want to implant it and what you think the advantage is. They seem to cost more, they’re harder and take longer to put in, and they require IVUS or OCT for optimal placement. Everything’s adding cost to the procedure, so you need to stop and think what the reason is to put it in in the first place.

I would, however, encourage continued research on it. I think we need more trials and longer term follow-up. I think we need better device iteration, and with thinner strut stents, for example, likely the results will be better. However, at the present time, it isn’t clear to me what patient would have a compelling indication for the current-generation bioresorbable scaffold, so, if you’re going to go ahead and implant it, I would just say be cautious, think twice and have a good justification for it.

O’Neill: The ABSORB III findings are incredibly good news for patients that need to have stents implanted because it shows how incredibly safe the traditional DES are, and also it shows the safety of the BVS. There were 1-year target vessel failure rates in the single digits, which are almost unheard of. Probably to the disappointment of Abbott, the event rates for the Xience were so low that no specific advantage for the BVS could be shown.

A technology’s success depends on finding “the killer app,” the place where a device becomes so essential that if I didn’t have that device in the cath lab, then I wouldn’t do the case. Here, I don’t really see where there’s an essential need for the BVS. There has always been some hope that the patients would prefer it, but the outcomes speak for themselves in the long-term event rates: zero acute thrombosis in the follow-up for Xience. It’s hard to say if we could ever show that a device would be safer and more effective than the Xience stent right now. The one place where there might be some value for the BVS would be in patients who have left anterior descending (LAD) artery disease, with the understanding that if they have recurrent restenosis, or progress in their disease, they might be a candidate for a left internal mammary artery graft in the future. If you stent throughout the mid LAD, then it is much less possible to do a bypass operation later on. That’s the one theoretical case I can think of. Usually the places we have problems with patients are in small vessels and diffuse disease, and unfortunately this iteration of the BVS doesn’t work at all in small vessels. In the places where there is the greatest need, it seems to be the least beneficial.

An avenue for future research might be in the acute MI setting, marrying some sort of antithrombotic to the BVS. But right now, most of us are scratching our heads trying to figure out where this product is going to go.

Itchhaporia: I think that we’re going through yet another revolution in the intervention world: the introduction of bioresorbable scaffolds and stents. In any sort of revolution, there are lessons to be learned. And I think we learned many lessons in this ABSORB database. Maybe we need to do a better job of prepping the vessels and imaging them and post-dilating them ... maybe ... PSP may reduce some of the adverse outcomes. We’ve learned that maybe we need to avoid the very small vessels. You can’t argue against the data, which is that when the gold standard has zero thrombosis, it’s hard to accept any amount of scaffold thrombosis. ... Given these data, am I going to run out and implant more Absorb stents? Probably not. But I also think this technology is in its infancy and these are the growing pains we have with any new technology where we learn lessons and then we get better. Maybe the long-term outcomes will be better when this scaffold is completely absorbed. And that’s what we’re going to be looking for. But it’s a long time to wait, and most of us are not quite that patient. But I think that’s what we’re going to have to do: wait and see.

COMPARE-ACUTE

Wilson: Although it will probably not change guidelines at this point, COMPARE-ACUTE offers more data and information suggesting that it’s reasonable for clinicians to consider pressure-wire fractional flow reserve of the noninfarct arteries in patients who present with STEMI or acute MI and let that guide them as to whether or not those arteries that were not involved in the MI but are significant or seriously blocked should also undergo PCI or stenting at that same time.

The study of 885 patients basically shows that it’s safe to perform this additional procedure. The primary outcome, defined as a composite of all-cause mortality, nonfatal MI, revascularization and cerebrovascular events at 12 months, occurred in 7.8% of patients who underwent FFR-guided complete revascularization vs. 20.5% of patients who underwent infarct-related artery-only revascularization (HR = 0.35; 95% CI, 0.22-0.55). Also, it shows that if the procedure is performed and it guides the clinician toward putting additional stents in other arteries at the time of the original MI, those patients actually do better over time and have fewer combined events like recurrent MI, need for other interventions or revascularization, fewer strokes and fewer overall combined endpoints.

Jaffer: This was a very important trial because it demonstrates that additional FFR-guided nonculprit PCI, immediately following culprit PCI for STEMI, improves clinical outcomes. Whether or not to treat and the timing of treatment of nonculprit lesions during STEMI remains a debate. Prior studies (eg, PRAMI, CvLPRIT and DANAMI-3-PRIMULTI) have shown a range of outcomes. This trial strengthens the evidence base that it’s going to be useful to treat significant nonculprit lesions that are present in the setting of STEMI.

One aspect that wasn’t evaluated in this study but remains an open question is whether the deferring nonculprit FFR-guided PCI for a few days might be a safer option for STEMI patients. There is still a concern that when nonculprit lesions are treated in the setting of STEMI, there may be extra contrast use or additional ischemia, which can be harmful to certain patients with STEMI. It is important to note that patients excluded from this trial included those with advanced HF, left main CAD, chronic total occlusion and suboptimal culprit PCI outcomes.

The primary benefit of this study appears to be that of fewer repeat revascularizations, rather than hard events such as death or MI. The study may have been underpowered for these latter endpoints. For patients meeting the inclusion criteria, this study is attractive because treating the culprit and nonculprit lesions at the same time is efficient and avoids a repeat second invasive procedure. However, another consideration is that interventionalists who adopt FFR-guided nonculprit PCI during STEMI cases may now experience longer case times, which could affect workflow and staffing for such emergency cases.

Stone: The COMPARE-ACUTE study was the fifth randomized trial that has asked the question of what should we do with patients who have STEMI, who have primary angioplasty, but have multivessel disease with other moderately severe lesions that may or may not need revascularization. This trial randomly assigned those patients to either FFR guidance — in other words, do FFR of those intermediate lesions, and right then and there in the same procedure if the FFR 0.8, do PCI, and if it’s not, defer vs. a more conservative approach, where patients would only get those lesions treated if they developed spontaneous ischemia or ischemia on a treadmill test. In this trial, there was actually an improved prognosis by using FFR guidance and doing acute multi-vessel intervention, which was required in about half of the patients where half could be safely deferred.

Specifically, there was a lower need for unplanned revascularization, there was less symptom recurrence, and less need for unplanned hospitalization, but the difference in death or MI was not statistically significant between the two strategies. We await the results of two other large, ongoing randomized trials.

DEFINE-FLAIR and iFR-SWEDEHEART

Bhatt: DEFINE-FLAIR and iFR-SWEDEHEART are particularly relevant to interventional practice. These were two very similar trials with a combined 4,529 patients examining guidance by instantaneous wave-free ratio (iFR) vs. guidance by FFR. Both were set up as noninferiority trials and both easily met noninferiority. The primary endpoint in both studies was a composite of all-cause mortality, nonfatal MI or unplanned revascularization. In DEFINE-FLAIR, the primary endpoint occurred in 6.79% of patients who underwent iFR-guided revascularization and 7.02% of those who underwent FFR-guided revascularization, with a difference of 0.2 percentage points (95% CI, 2.3 to 1.8; P < .001 for noninferiority; HR = 0.95; 95% CI, 0.68-1.33). In iFR-SWEDEHEART, rates of the primary endpoint were similar between the iFR group and the FFR group (6.7% vs. 6.1%), with a difference in event rates of 0.7 percentage points (95% CI, 1.5 to 2.8; P = .007 for noninferiority; HR = 1.12; 95% CI, 0.79-1.58). In fact, when the data are pooled together in a meta-analysis, it’s a really solid noninferiority with a relatively tight 95% CI.

Why should we be excited about noninferiority? Well, the reason I’m excited is because iFR is a lot easier to do than FFR. It’s quicker, it’s easier and it confers less patient discomfort. The iFR calculates the measurement on flow and diazoline using a particular algorithm, so there is no need to give adenosine.

For example, if you wanted to do a multivessel assessment with FFR, you’d have to give adenosine for each vessel. With iFR, you just wire it and you get your measurements over a few cycles of diazoline and you just repeat it for another artery. If you want, repeat it for another lesion. Serial lesions are a bit of a challenge with FFR. They can be assessed, but I think a lot of operators don’t totally understand the interplay between the two lesions and how the flow through one affects flow and potentially, FFR measurements to the other. The IFR technology facilitates both the multi-lesion and multivessel assessment.

Also in those studies, there was less stent usage. Maybe not all interventionalists are excited about that, but it is good for the patient and health care costs if the need for stenting is reduced without hurting patient outcomes.

So, for a variety of reasons, this is a big win and could be practice-changing and increase the overall use of invasive physiologic assessment in the cath lab.

Stone: These trials, which were very similar, randomly assigned a total of 4,500 patients with borderline lesions in the cath lab to decision making with either iFR or FFR. Both trials essentially found the exact same thing: That IFR was noninferior to FFR, leading to very similar outcomes in terms of death, MI and the need for repeat revascularization. The IFR procedure was easier — it cut off anywhere from 2 to 5 minutes from the procedure — but more importantly, it was associated with about a tenfold reduction in symptoms that the patients experienced and was linked to less arrhythmias. Later data will reveal the cost-effectiveness of this approach; saving the need for using adenosine will likely save several hundred dollars.

Itchhaporia: It’s intriguing to have a technology that we now know is noninferior to FFR, which is the established therapy. There is going to be a higher adoption rate of iFR, especially since it promises improved patient comfort and shorter procedural time. As a result, we may see more people adopting physiologic testing of intermediate patients with this type of technology.

The issue is not whether we will convert to iFR. The issue is more basic: Can we make the technology easy enough that we’re going to be able to get more people to use it and perform more physiologic testing? With this easier technology, it may be easier to get more people to do it.

Jaffer: These are two remarkable studies conducted with two multicenter groups that are both well respected in the field of coronary physiology. They effectively show us that using iFR-guided management vs. FFR-guided management performed similarly, and both studies met their endpoints for noninferiority.

Clinically, the ability to eliminate adenosine during functional assessment intuitively makes a lot of sense. We were encouraged by an adenosine-free iFR concept when it was introduced, but we needed more clinical outcomes data. These two trials now provide outcome evidence showing that if we opt to use iFR, we are going to be doing as good a job as FFR. Adenosine-related side effects can be significant for some patients, so it is nice that iFR avoids them as well as the cost of adenosine. An iFR option might now increase the use of coronary physiological assessment in stable CAD, which is underused in decision making for intermediately stenotic lesions.

With these two new studies, we will probably see a growth in iFR use, and that is a big deal because FFR has been established for more than 10 years.

Alraies: DEFINE-FLAIR and IFR-SWEDEHEART added more evidence to the current literature that FFR is the tool of choice to evaluate myocardial ischemia. These two studies also showed that iFR is non-inferior to FFR. Moreover, iFR is associated with less side effects since no adenosine, a hyperemic agent, is used which improved patient experience and minimized side effects (ie, flushing and arrhythmia or AV blocks).

There are several notable aspects of these studies. No. 1, DEFINE-FLAIR and iFR SWEDEHEART are the first studies to offer outcome data on iFR compared with FFR. In both studies, the researchers evaluated hard outcomes including all-cause mortality, MACE, unplanned revascularization and nonfatal MI at 1 year.

No. 2, both studies, which are robust and include a large number of patients, add to the existing knowledge about FFR while also proving that iFR performs as well as FFR with less side effects. For instance, in DEFINE-FLAIR, the researchers found no difference in the primary endpoint (all-cause death, MI, or unplanned revascularization at 12 months) between the two interventions — 6.79% for patients who underwent iFR-guided revascularization and 7.02% for those who underwent FFR-guided revascularization.

No. 3, I find it intriguing that the number of patients who were deferred from coronary intervention was higher in the iFR group than in the FFR group; yet, these patients in whom treatment was deferred had the same outcome. Meaning if a lesion is considered normal using iFR, then the clinician does not need to worry about it. However, if iFR is performed and the results are equivocal or questionable, then FFR is justified (hybrid approach).

Despite these findings, DEFINE-FLAIR and IFR-SWEDEHEART didn’t incorporate cost in to the design of the study. A cost-effectiveness study comparing FFR with iFR is important since iFR eliminated the use of adenosine and deferred more patients from coronary interventions. This is a critical question that needs to be answered as a next step.

DECISION-CTO

Wilson: DECISION-CTO was a randomized trial of 805 patients with a coronary chronic total occlusion comparing optimal medical therapy alone to optimal medical therapy plus CTO PCI. Over a period of more than 5 years, there was really no difference in the patients who received complex PCI and stenting of these CTOs vs. those who were treated medically. The rate of the primary endpoint of death, MI, stroke or any revascularization at 3 years in the intention-to-treat population of patients with a CTO was 19.6% in the optimal medical therapy group vs. 20.6% in the PCI group (P for noninferiority = .008). However, in the discussion after the presentation, it was noted that this was an older trial starting way back in 2010, and there certainly have been newer devices and techniques available in the last 5 years that may have made a difference. Also, there perhaps weren’t enough patients that were followed long enough to see whether there can be differences. It still remains that PCI should be done for patients with CTO to try and help relieve symptoms, but at this time we can’t really show that there’s a mortality difference or benefit.

Stone: The DECISION CTO trial has sparked a lot of controversy. This was a trial that was designed to determine whether revascularizing CTO, which can now be done with a high success rate at selected centers, improves hard outcomes in patients with CTOs compared with optimal medical therapy. This was a multicenter trial in Korea, which was initially planned to enroll 1,200 patients and planned as a superiority trial, but because of difficulties of recruitment, it was changed to about 830 patients and converted into a noninferiority trial, so the statistics were a little bit confusing.

The bottom line is, with 834 patients, they did not see any difference in death, MI or the need for repeat revascularization between the two approaches. There was a very small signal for lower mortality with PCI, but it wasn’t statistically significant. Whether this was due to the design, inclusion criteria, the crossover rates — which were over 20% for medical therapy to PCI — or other methodologic issues remains unknown and unresolved, but where it’s left us is that we don’t know for sure whether patients with CTO will benefit from PCI and if so, which patients are most likely to benefit.

Bhatt: DECISION-CTO, a Korean trial which looked at PCI in patients with CTOs, didn’t find any benefit in terms of hard outcomes. It was billed here as kind of a negative trial.

While the study has limitations, for all those CTO enthusiasts out there, the results point to a need for some data to show that opening up CTOs really does affect hard outcomes. For sure, there are patients with bad angina where you want to open up a CTO just for that purpose, but in terms of the ability to influence hard outcomes, we do need more data, especially given that the EXPLORE-CTO trial also didn’t find benefit in terms of LV function of opening up CTOs in patients with STEMI.

RESOLVE/SAVORY

Herrmann: This is essentially the largest study to date of CT scans in people after TAVR and surgical valve replacement. However, although it’s the largest, it is not a perfect study.

Prevalence of subclinical leaflet thrombosis was considerably higher among the 752 patients who underwent TAVR compared with the 138 who underwent surgical AVR (13.4% vs. 3.6%; P = .001). The researchers also found that leaflet thickness (5.01 mm vs. 1.85 mm; P = .0004) and the percentage of leaflet motion restriction were greater in the TAVR group vs. the surgery group (71% vs. 56.9%; P = .0004).

There are two caveats: These were not consecutive patients and not all CT scans were done at exactly the same time point. With those caveats, it’s hard to be sure of what CIs to put around those incidences of leaflet thrombosis and leaflet thickening.

That said, there are a couple of takeaway points. First, it appears that the incidence of subclinical leaflet thrombosis was higher with TAVR than surgery. Second, treatment with anticoagulants is good and warfarin seemed to be equivalent to novel oral anticoagulants. Third, most of the events were asymptomatic and did not translate into clinical events. That suggests that some of these subclinical leaflet thromboses may resolve without therapy since we are not seeing problems of this magnitude years out in the follow-up studies.

So, the point is that it may still be too early to treat all patients with warfarin or novel oral anticoagulants because the event rate was somewhere between 4% and 13%, of which only a minority of patients have clinical events. That would mean we would be treating more than 90% of patients with a blood thinner that they do not need or from which they would not benefit. The goal is therefore finding out where the risk-benefit is, how long to treat patients and which patients should be treated, which is the aim of trials like GALILEO and ATLANTIS.

Devireddy: The combined imaging and outcomes data of the RESOLVE and SAVORY registries provide further weight to the concern that TAVR may pose a risk of subclinical leaflet thrombosis in certain patients. Is it, however, time to sound the sirens of alarm? It is difficult to claim the sky is falling on TAVR from this study alone. Without prospective consecutively collected data, there is a great risk of bias in comparing TAVR vs. surgical AVR in this manner.

Marked differences are noted in the time from valve implantation to CT imaging between TAVR and surgical AVR which may reflect some inherent confounding in the types of patients enrolled in the registry.

Independent of whether TAVR patients experience more leaflet thrombosis than surgical AVR patients, the RESOLVE and SAVORY investigators have provided firm proof that subclinical leaflet thrombosis is a real complication and that 4-D volume rendered imaging with CT is a valuable tool in diagnosing it. Based on these registries, it would seem wise to have a low threshold to acquire 4-D imaging to assess leaflet thickness and mobility in any TAVR patient with an unexplained sudden rise in gradient.

In order to learn if TAVR patients (or potentially all AVR patients) would benefit from more aggressive anticoagulation, we can only hope that the currently enrolling ATLANTIS and GALILEO studies will shed further light. If benefit is found with anticoagulation, difficult decisions remain. Which patients stand to benefit the most? Is one transcatheter valve platform more likely to clot than another? Can anticoagulation ever be discontinued in those who need it? TAVR has revolutionized our ability to treat valvular heart disease. Hopefully, such studies will help us optimize the longevity and performance of this technology which has already improved so many lives.

BICUSPID VS. TRICUSPID STENOSIS in TAVR

Elmariah: This was a thorough and well-executed observational study investigating the safety and efficacy of TAVR in patients with symptomatic severe bicuspid aortic stenosis as compared to those with tricuspid aortic stenosis. The study represents the largest analysis to date evaluating TAVR in bicuspid aortic stenosis, draws from a large number of experienced international sites, and employs propensity-score matching in an attempt to adjust for measured confounders. In addition, within subgroup analyses, the study evaluates and compares the performance of older vs. new-generation transcatheter valve prostheses for bicuspid aortic stenosis.

All-cause mortality was 17.2% in those with bicuspid aortic stenosis and 19.4% in those with tricuspid aortic stenosis at 2 years (P = .28).

This is the first study to evaluate and systematically compare TAVR performance in bicuspid aortic stenosis to that in tricuspid aortic stenosis. While the authors confirm prior concerns that TAVR in bicuspid aortic stenosis is associated with lower rates of device success, higher rates of conversion to surgery, and higher rates of moderate to severe paravalvular aortic regurgitation and aortic root injury, subgroup analyses suggest that these differential results do not impact newer-generation transcatheter valves. Moreover, cumulative 2-year mortality did not differ between bicuspid and tricuspid aortic stenosis groups, further supporting the safety of TAVR in patients with bicuspid aortic stenosis.

The results of this study support the safety and efficacy of TAVR in patients with bicuspid aortic stenosis; however, it remains concerning that TAVR for bicuspid aortic stenosis is associated with more frequent procedural complications, reduced device success and greater paravalvular aortic regurgitation. While such results may be acceptable for high- or extreme-risk patients with limited alternative therapeutic options, greater safety is needed before the widespread application of TAVR for bicuspid aortic stenosis can be endorsed in healthier patients. The more favorable outcomes observed with newer-generation TAVR valves are certainly encouraging and warrant further investigation and confirmation.

Two primary factors must be considered when evaluating treatment options for patients with bicuspid aortic stenosis. No. 1 is patient risk. Until greater device success and safety can be consistently attained with TAVR, surgical AVR should remain the treatment of choice for low- and intermediate-risk patients with bicuspid aortic stenosis. No. 2, the presence of an associated aortopathy should strongly favor surgery unless the predicted risk of mortality is felt to be too great.

HEMODYNAMIC SUPPORT in PCI for MI with CARDIOGENIC SHOCK

Naidu: I would like to applaud the researchers for putting together a consortium to tackle the issue of cardiogenic shock in a system-wide approach, which is unusual. They put the following approach together to see if it would move the needle on cardiogenic shock: rapid transfer with Impella support, done pre-PCI, followed by PCI and invasive hemodynamic monitoring to determine escalation and weaning of therapies, with reduced use of inotropes when possible. This observational study shows that that approach is feasible. A group of hospitals can get together, at least with a small number of patients, and do all of these things with a high survival rate. What the results don’t tell us is what the survival rate would have been if any of the steps didn’t happen. It’s not a proof of concept that these steps improve survival, but that they are safe and feasible to do; they did not seem to cause harm. The overall survival of more than 80% is markedly higher than we would expect in an acute MI/cardiogenic shock population needing assist devices, but it cannot be ultimately proven in this trial.

It is reasonable to try to standardize this approach in a larger registry series compared with other types of patients who come in with similar risk factors to see if these patients are actually improving in survival. Factors such as how fast the patient came in, whether support was used before or after PCI and the number of inotropes used can be controlled for. Perhaps a case-control study between the different approaches in patients at similar risk would be warranted. It’s very hard to do that with 37 patients, so a larger number of patients would be needed to determine an actual improvement in survival.

Ideally, we’d like to see a randomized trial, but it has always been very difficult to do randomized controlled trials for shock. Some have been done, but with relatively small numbers of patients. It’s hard to compare this approach to “standard care” when there would probably be significant portions of patients stabilized with devices ahead of time, with some but not all of the features of the Detroit initiative’s approach.

These findings are impressive, especially because they were produced by collaboration among competing hospitals. It was encouraging to see survival was high with this approach. The findings are further confounded by the fact that these are highly skilled operators who are quite good at taking care of patients with cardiogenic shock. Whether this can be generalized throughout the country therefore also becomes an issue.

MR-INFORM

Winchester: This study provides interesting data on the possibility for using MR for guiding a revascularization strategy. It’s always nice to have an additional noninvasive technology that helps us make those sorts of decisions. The authors are to be commended for a job well done.

Revascularization occurred in 44.2% of the FFR-guided group vs. 36% of the MR-guided group (P = .0053). The event rate of the primary outcome of composite of all-cause mortality, nonfatal MI and re-revascularization of a vessel targeted at the index revascularization procedure at 1 year was 3.9% in the FFR-guided group and 3.33% in the MR-guided group (absolute risk difference, –0.56; 95% CI, –2.98 to 1.86; HR = –0.852; 95% CI, –0.43 to 1.69).

The data suggest that perhaps revascularization is less with the MR-guided strategy vs. the FFR-guided strategy, with no difference in cardiac events. This is in line with what we’ve seen in other studies of stable ischemic heart disease such as BARI-2D and COURAGE.

For those with access to MR and the expertise to perform it, this strategy may be a viable option. That said, MR is not available in a large number of facilities and more than 16,000 patients were screened but only 918 were randomized. This suggests it might not be a technique that is best for all-comers with stable angina.

Disclosures: Alraies, Elmariah and Naidu report no relevant financial disclosures. Bhatt and Stone report financial ties with various pharmaceutical and device companies. Devireddy reports serving on an advisory board for Medtronic. Herrmann reports receiving research funding from most of the transcatheter valve companies for participation in their trials and acting as a consultant to Edwards. He is also an investigator for the GALILEO trial. Itchhaporia reports receiving research grants from St. Jude Medical. Jaffer reports consulting for Abbott Vascular and Boston Scientific and receiving research funding from Canon and Siemens. Kapadia reports serving on steering committees for trials funded by Abbott Vascular, Boston Scientific and Edwards Lifesciences. Mehran reports receiving consultant fees or honoraria from Janssen Pharmaceuticals, Medscape, Osprey, The Medicines Company, Watermark Research Partners, WebMD, Wiley Blackwell Publishing Company; and research grants from AstraZeneca, Bayer Healthcare Pharmaceuticals, Eli Lilly/Daiichi Sankyo and OrbusNeich. O’Neill reports consulting for Abbott and Medtronic. Wilson reports receiving consultant fees or honoraria from Boston Scientific. Winchester reports receiving a research grant from Roche Diagnostics.