Irvine, CA - More physicians in Europe and the US are getting the chance to try out the Sapien transcatheter aortic valve developed by the industry heavyweight, Edwards Lifesciences, and most are still excited about the technology, despite recent reports highlighting the mortality associated with different device-delivery methods, the tremendous skill required to implant it, and nuances of appropriate patient selection. In Europe, the Sapien has been available since late 2007, while in the US, interventionalists and surgeons are getting their hands on the device as the pivotal US PARTNER trial ramps up enrollment.

The Sapien transcatheter aortic valve [Source: Edwards Lifesciences]

Their enthusiasm speaks to the unmet medical need for a percutaneous solution for aortic-valve disease. Surgical valve replacement is the gold standard, but it is an open-heart procedure, meaning that many patients are deemed too old or too sick to undergo valve replacement. Studies suggest that just one-third of patients with severe, symptomatic aortic-valve stenosis end up getting a surgical valve, and those who don't face mortality rates as high as 60% at one year. Of all the percutaneous valve replacement technologies under development, aortic-valve technologies are far more advanced than devices for other forms of valve disease, and more than a dozen companies are actively developing transcatheter aortic-valve devices, with Edwards being the closest to having its transcatheter valve approved in the US. But as results trickle in from early patient series, physicians and market-watchers are increasingly divided as to which delivery method for the Edwards valve is superior and how this may affect mortality outcomes both in and beyond the PARTNER trial.

Edwards has developed two delivery techniques; both are being tested in PARTNER. In the transfemoral approach, the valve is delivered fully percutaneously, while the transapical approach involves a minithoracotomy and an incision in the apex of the heart. In Europe, the Sapien valve was first approved for sale in September 2007 for the transfemoral procedure and in December for the newer transapical procedure, which requires a different delivery system.

Click here for an animation [Source: Edwards Lifesciences] Transfemoral approach.

Transapical approach. Click here for an animation [Source: Edwards Lifesciences]

Several recently published, consecutive patient series examining outcomes with the transapical approach are stoking debate over whether this approach is less safe than the fully percutaneous method. Two published patient series from Germany, both with cardiac surgeon Dr Thomas Walther (Herzzentrum Leipzig, Germany) as first author, have reported excellent success rates with transapical delivery [1,2]. But by contrast, a US feasibility study, conducted as a preliminary step for PARTNER at three sites in the US, reported much lower one-month survival—the critical period for considering device/procedure success [3]. Results for the US series, published in the Annals of Thoracic Surgery in July 2008 by Cleveland Clinic cardiac surgeon Dr Lars G Svensson and colleagues, are in part responsible for rumors that the transapical approach may not be as good as the transfemoral approach. By way of comparison, 30-day survival rates in 50 patients treated using the transfemoral approach by interventionalist Dr John Webb (St Paul's Hospital, Vancouver, BC) and colleagues were 84% at the beginning of the series but rose to 92% in the second half of the study [4].

Commenting on the results of his group's study to heartwire, Svensson explained that even to take part in the US feasibility study, patients had to have a predicted risk from conventional surgery of 15% or higher. "We showed that, as a first step, it is feasible to do this procedure, but the risk of death was about 17% at 30 days. But, on the other hand, we did not have a stroke in anybody. So from our point of view, this is a necessary first step with a new procedure."

He also points out that the patients enrolled in the US study were extremely high risk; half were deemed ineligible for the transfemoral procedure, usually due to severe iliofemoral arterial disease. Mean EuroSCORE, a measure of predicted operative mortality, among patients in the US trial was also considerably higher than in the European studies or in Webb et al's transfemoral series: 37% vs 27% and 28%, respectively.

Walther, for his part, was reluctant to compare his results with those of other institutions, but he acknowledged that the inclusion criterion for his transapical study was an "all-comers" approach—the EuroSCORE would naturally be lower.

But despite their different results, both Svensson and Walther are convinced that the transapical procedure is an excellent approach and that mortality rates will come down as operators become more adept at the technique and better at patient selection.

"I think the transapical approach, at least in my experience, is at least as good as the transfemoral approach and is associated with a much lower risk of stroke," Walther said. "We've had zero strokes in more than 150 patients—this is a very, very strong argument in favor of transapical."

Valve aficionados are now looking ahead to Sapien data that will be unveiled at the upcoming European Association for Cardiothoracic Surgery (EACTS) and TCT meetings, especially six-month results from the 130-patient PARTNER EU trial, including both transapical and transfemoral patients. Other registry data are also due to be released. But rumors of higher-than-anticipated mortality rates in these as-yet-unseen data sets are also rippling through the cardiology and CV surgery communities.

In a recent equity research report, JP Morgan market analyst Michael Weinstein predicted, on the basis of talks with unnamed cardiologists and cardiac surgeons, that the mortality data for transapical patients in the study would be 40% at six months—as compared with 20% in the transfemoral group. Partly due to these mortality-rate predictions, which he called "higher than expected," he has downgraded his rating on Edwards stock. But another analyst, Larry Biegelsen of Wachovia, who issued a much more upbeat report, points out that a 30% to 45% mortality rate for the transapical approach is in keeping with previously reported data, and slightly higher rates would naturally reflect the fact that the device was being used for the first time by inexperienced operators.

"There are obviously reports in some newspapers and even from the stock market that are pretty negative, but I see this completely differently," Walther told heartwire. "First, I'm a positive person, but I think our results are positive as well. The negative part comes, in my opinion, from the fact that healthier patients, who have good femoral arteries, go in the transfemoral arm of these studies. The sicker ones, those who have more disease, go in the transapical arm, so the transapical cohort accumulates a higher-risk profile, and they will get worse results, of course. At our hospital, we have similar outcomes, and results are at least as good with transapical."

Others, however, more readily admit that the transapical results are trailing behind those of the transfemoral cases, although how much that has to do with patient selection, operator skill, or other factors remains unclear.

Webb, whose Vancouver group has been involved in the evolution of both delivery methods, insists he had seen no new data to support the rumors but acknowledged: "The results with the transapical procedure aren't as good as the femoral so far, except at some centers. But I think there's a learning curve with these procedures, so that people who don't have much experience have some problems with them initially."

Svensson says his group has already identified problems that drove up mortality rates in their published series and believes the patients now enrolling in PARTNER trial will reap the benefits of this learning process. "I think we were learning a lot about where to position the valve—that was clearly one issue we had and resulted in some of the problems. And second, in retrospect, we probably selected some patients for the procedure whom we shouldn't have, whom we then subsequently had problems with."

Dr Christopher Thompson (St Paul's Hospital), an echocardiographer on Webb's team, says this is part of the learning process.

"There's no question in my mind that this technology works, and the delivery systems and our understanding of the procedure are improving," he told heartwire. "There will be a role for this device, but these are the early days of a novel technology that is still finding its place. . . . That's what the PARTNER trial is for, establishing the risk/benefit [ratio] vs the best available alternative."

If the mortality is as rumored, then I think it's going to be very difficult for transapical to survive.

Almost all of the surgeons who spoke with heartwire defended the transapical approach and were adamant that it will have a role in the future treatment of aortic-valve disease. No surprise here: the surgeon is both the team captain and star player during a transapical procedure but more or less stays on the sidelines during the transfemoral procedure, subbing in when emergencies arise. But interventionalists have long believed that the future for this device will be a fully transfemoral approach and that the only barrier to this for now is the size of the delivery catheters—22 F or 24 F for the Sapien's RetroFlex II delivery catheter.

"I really do believe the future will be fully transfemoral," Dr Eberhard Grube (Helios Heart Center, Siegburg, Germany), an interventional cardiologist, told heartwire. "If the mortality is as rumored, then I think it's going to be very difficult for transapical to survive."

Grube's expertise is with Edwards's competitors, the CoreValve device, also on the market in Europe, and new devices in development by Sadre and Direct Flow. He says he is able to use a transfemoral approach with the CoreValve, which has a smaller delivery catheter than the Sapien, in 98% of cases. For the few remaining patients in whom a transfemoral approach is impossible, "you have to ask yourself whether this patient should be getting a procedure at all," he said.

CoreValve data flow in The CoreValve Revalving Aortic Bioprosthesis [Source: CoreValve] CoreValve's Revalving percutaneous aortic-valve replacement system also holds CE Mark approval in Europe and is now being used in 70 European centers; upwards of 1500 valves have already been implanted, including patients treated in preapproval studies, according to company president and CEO Dan Lemaitre. The company also intends to present new data at the European Society of Cardiology, TCT, and EACTS meetings. The company is still in talks with the FDA about what sort of trial will be required to gain US approval, but the European arm of CoreValve's 18F International Revalving trial (testing the company's new smaller delivery system) has already completed enrollment and is finishing up recruitment in Canada, where the trial had been on hold. CoreValve's product is a transfemoral system, and Lemaitre said diplomatically that the company is "not fully convinced" about the transapical procedure, although it anticipates developing other ways that the procedure could also "embrace" surgeons. -SW

Grube and others believe that interspecialty politics may in some cases be driving procedure choice and, ultimately, patient outcomes. "At centers where surgeons are very strong, they will be the gatekeepers for this procedure, and the transapical will be the dominant procedure, so there will be a negative bias toward using the transfemoral approach," Grube told heartwire. At centers where interventionalists are stronger, procedure numbers will be higher for the fully percutaneous procedure, he said.

Likewise, Dr Grayson Wheatley (Arizona Heart Institute, Phoenix), a surgeon, believes that subtle turf wars may end up affecting the outcomes from the PARTNER trial.

"There is a little bit of investigator pushing going on, and I think the surgeons, to be honest, are pushing subtly toward a transapical approach because they are more involved and they maybe feel a little more comfortable," Wheatley said. "Whereas cardiologists feel pretty good about the femoral arteries, and they say, why should we subject the patients to the surgical risk when we can do this transfemorally, something we do on a regular basis? I think it's a disruptive force within the trial and within the use of this technology."

I think the surgeons, to be honest, are pushing subtly toward a transapical approach because they are more involved and they maybe feel a little more comfortable.

One issue may be that while interventionalists do not appear to be clamoring for scalpels and sutures, many surgeons are learning catheter techniques for the first time. "I do know that some of the surgeons who have been doing the transapical procedure have had a little bit of trouble because they weren't really skilled with the catheters and wires," Wheatley said. "Maybe there's a bigger learning curve with the transapical approach because it involves different sheaths, different wires, and sometimes people with lesser catheter or wire skills are doing those procedures."

Wheatley worries that unless the learning curve for physicians performing the transapical procedure in the PARTNER trial improves and the mortality rates come down, the "cohort A" data results may be difficult for the FDA to interpret.

"If we could somehow disregard all of the transapical data and just focus on the transfemoral, I think we would be much better off," he predicts. "I think introducing this whole transapical approach way downstream from the initial launch of the technology has been very confusing, disrupting, and challenging for physicians to make sense of things. We had a nice thing going, we had a nice clean data set, the device was in the hands of good clinicians, and we had very consistent errors. But there are so many variables going into transapical: the learning curve, different sheaths, different wires, different ways of sizing the valve, different visualization. I think we've introduced a whole other level of variables to try to sort out how to make sense of the technology, rather than focusing on the original, well-defined data set that was progressing nicely."

Almost everyone agrees that the higher mortality rates in the transapical studies reflect the fact that training a percutaneous valve "team" made up of multiple medical specializations takes time, and maintaining proficiency is essential. Walther pointed to the fact that in Svensson's series there was a four-month gap in the middle of the study, when the FDA was approving the enrollment of the next 20 patients. "I think they kind of lost their experience during this gap and they had to restart," Walther suggested. "You need to do this regularly. At our center, we now do valve cases one day per week; this allows us to retain experience as to how to deal with problems and how to handle very sick patients postoperatively, and that's how you obtain better results."

Overall, says Svensson, optimism remains "high" for percutaneous valves. He predicts that after some initial growing pains, the transapical results will be comparable to the transfemoral. "Everybody is going to try to second guess what is going to happen, but my expectation is that these procedures will continue to have something like a 4% to 7% 30-day mortality rate, and that's partly going to be dependent on the procedure and the problems we have with doing it, and also the patient group that we're including."

As for what the ideal patient group might be, that's the million-dollar question. Webb has pointed out in the past that this new procedure has, appropriately, battled to prove itself in so-called "no-hope" patients, and the high mortality rates in this group have inevitably raised questions about whether this technology was worth pursuing.

"Until now, we've been using this procedure in patients who are extremely high risk for surgery," Webb reminded heartwire. "And the early data seem to show that this procedure might work very well in low-risk patients and even be a desirable procedure in patients who are low risk for surgery."

But that's a question to be answered down the road, presumably. What the FDA is hoping to learn from PARTNER is whether those "15% or higher" surgical risk patients can benefit from one or both of these transcatheter approaches. Learning the true potential of any novel technology means putting it to the test in very ill patients, but physicians involved in the field are starting to admit that enthusiasm for the new procedure may mean that it is being tested in patients who are proving too sick or too frail to survive. US physicians honing their skills in the PARTNER trial and European physicians receiving ever-increasing referrals for percutaneous valves may have to make peace with the fact that despite the promise of this new technology, it's not for everyone.

"Doctors naturally want to help," Thompson told heartwire. "When they see a patient who has no other options, they keep extending and extending the limits in higher- and higher-risk patients. Ultimately you have to stop and say, we've gone too far and pull back to the 'right' population."