TAVI: Wave of the Future? Or Time for a Deep Breath?
Some practitioners, such as the PARTNER investigators, are advocating that transcatheter aortic valve implantation (TAVI) become the new standard of care for patients with aortic stenosis who are not suitable surgical candidates. However, others question whether the quality of life, practice management, cost effectiveness and ethical considerations of treating such sick patients have been answered definitively before it becomes routine clinical practice in the U.S.

PARTNER unveiled

Approximately 30 percent of U.S. patients with aortic stenosis (AS) who have symptoms are not being treated with surgery—due to extreme age, comorbidities or poor heart function. Currently, transcatheter aortic valves, which could present an alternative for this population, are investigational devices in the U.S., while several have been approved in Europe since 2007.

At TCT10, Martin B. Leon, MD, director of the Center for Interventional Vascular Therapy at New York Presbyterian Hospital/Columbia Medical Center in New York City, presented the much-anticipated results of the PARTNER trial (N Engl J Med 2010;363:1597-1607).  

In the study, investigators randomly assigned 358 patients with AS, whom surgeons determined to be unsuitable for surgery, to standard therapy or TAVI of a balloon-expandable bovine pericardial valve (Sapien, Edwards Lifesciences). The patients underwent randomization at 21 centers (17 in the U.S.) in two arms of 179 patients. The primary endpoint was the rate of death from any cause. The average age of these patients was 83 years, with 20 percent at more than 90 years.

At 30 days, the rate of all-cause mortality was 5 percent in the TAVI arm compared with 2.8 percent in the standard therapy arm. At one year, the rate was 30.7 percent with TAVI compared with 50.7 percent with standard therapy, which confirms the “dire natural history” of patients with untreated AS, says Leon, adding that these results suggest only five patients need to be treated to save a life.

The rate of repeat hospitalization was 5.6 percent at 30 days and 22.3 percent at one year in the TAVI arm, compared with 10.1 percent at 30 days and 44.1 percent at one year in the standard therapy arm.

In the year after TAVI in PARTNER, there was no deterioration in the functioning of the valve, as assessed by evidence of stenosis or regurgitation on an echocardiogram. However, valve durability is an important consideration. “While the median follow-up time for PARTNER was 1.6 years, valve durability needs to be considered out to five, 10 and 15 years,” Leon says, “so the patients require careful follow-up with echocardiograms to ensure that these valves are as durable as surgical valves.”

Despite the promising mortality and repeat hospitalization rates, the TAVI arm experienced higher complication rates. At 30 days, TAVI, compared with standard therapy, was associated with higher incidence of major strokes (5 vs. 1.1 percent) and major vascular complications (16.2 vs. 1.1 percent). At one year, TAVI also was associated with higher incidence of major strokes (7.8 vs. 3.9 percent) and major vascular complications (16.8 vs. 2.2 percent).

While Leon says the stroke outcomes were expected based on the European registries, he suggests that the next generation of devices may help reduce the frequency of procedure-related complications. “In these older patients with vascular disease, it is difficult to insert the larger device used in the PARTNER trial,” he says. “We’re hoping that the next-generation devices, such as the Sapien XT [Edwards] that is 40 percent smaller and more durable or the EU-approved CoreValve [Medtronic], will obviate the vascular complications and reduce the bleeding complications.”

In December 2010, Medtronic commenced its U.S. pivotal trial designed to evaluate the safety and efficacy of the CoreValve system. The study will seek to enroll more than 1,300 patients at 40 clinical sites. The trial includes two studies in different patient populations: one study of patients diagnosed as high risk for aortic valve surgery and a second study of patients diagnosed as extreme risk. Patients deemed at extreme risk will not be randomized to optimal medical management; rather, they will be evaluated against a performance goal derived from contemporary studies. Patients in the high-risk group will be randomized one-to-one to either TAVI with CoreValve or to surgical aortic valve replacement (SAVR). The primary endpoint will be all-cause death or major stroke within 12 months.

Patient Selection

“Patient selection is everything for these procedures,” says Leon. “Within PARTNER, we spent an inordinate amount of time to create a working partnership with our cardiac surgeons to create a heart valve team to establish candidates for TAVI or for surgery.”

Due to the rigorous patient selection process in PARTNER, David J. Cohen, MD, director of cardiovascular research at St. Luke’s MidAmerica Heart Institute in Kansas City, Mo., hypothesizes that benefits “might be slightly less in real-life clinical practice, but only by a small margin.” He supports this statement by the various subgroup analyses conducted by the investigators to evaluate if certain groups derived greater or less benefit. Results showed that those in these subgroups who underwent the TAVI procedure derived approximately the same benefit.

However, Harold L. Lazar, MD, from the department of cardiothoracic surgery at Boston Medical Center, suggests that if TAVI is approved in the U.S., the patient population considered for this procedure will broaden. “Because facilities will need to maintain a certain patient volume to gain CMS reimbursement as a TAVI center, it could cause providers to include patients who are less sick than the criteria of the PARTNER trial,” he says. “On the other hand, patients who are not surgical candidates, but who have significant co-morbidities which by themselves will limit quality of life, may also be treated with TAVI, resulting in adverse outcomes that will diminish the benefits of this technology.”

To evaluate the procedure in lower risk patients, investigators—many of whom conducted the SYNTAX trial—have created a preliminary design for the European multinational, randomized SURTAVI trial, for which Medtronic is currently seeking approval.

In England, where TAVI is approved, the clinical indications of utilizing TAVI are broader than those used in the PARTNER trial. However, Martyn Thomas, MD, director of cardiothoracic services at Guy’s and St. Thomas’ NHS Foundation Trust in London, says that while the TAVI procedure is “fairly straightforward, the mortality and morbidity rates remain fairly high. After many years, I have learned that it’s the patient’s condition that causes the mortality, not the procedure.”

In the U.K., there has been a great deal of debate on patient selection for TAVI, based on a motivation to ration the number of cases performed, according to Thomas. While some practitioners take into consideration the EuroSCORE or the STS score, at St. Thomas, a multi-disciplinary team (two surgeons, two interventional cardiologists, an imaging specialist and a geriatric specialist) must agree that TAVI is the best option based on all the available data. This decision is made after a 48-hour patient evaluation, which includes a lung function test, carotid Doppler, coronary angiogram, peripheral angiogram, aortic CT, transesophageal and transthoracic echocardiography, as well as a patient consultation.

After this process, 45 percent of patients who are referred to St. Thomas with this condition receive TAVI, 15 percent receive open surgical valve replacement—despite the fact that the referring physician considered them too high risk—and the remainder of the patients receive either medical therapy alone or balloon valvuloplasty.

“In my opinion, this multidisciplinary decision process is currently the most appropriate method, while the EuroSCORE or STS score are just single measures of risk,” Thomas says. “It is particularly important that surgeons are part of this patient selection team because they would most likely opt for surgery when appropriate.

“We are frantically attempting to develop a risk algorithm from the various global registries and databases, but thus far, it has been very difficult to uncover an accurate predictor,” says Thomas, adding that the EuroSCORE is based on experiences with 16,000 patients and the latest STS valve score is based on 80,000 patients. “Based on these numbers, we have a lot more data to collect,” he adds.

Quality of life

Prior to the presentation of the quality-of-life data at AHA10, the impact of TAVI on patients’ functional status and health-related quality of life (QoL)—particularly in comparison with non-surgical therapies—was unknown. Therefore, Cohen et al performed a prospective health-related QoL study for all 358 PARTNER patients at baseline, one, six and 12 months after randomization.

Cohen says their primary measure, the Kansas City Cardiomyopathy Questionnaire (KCCQ), is a “well-validated tool to assess quality of life in patients with heart failure, which is the main syndrome in AS patients. In particular, this instrument has been known to correlate well with survival, hospitalization and cost.” They also used the generic SF-12 Health Status Survey, which provides summary scales for physical and mental health, and the EuroQOL (EQ-5D), which reports utility values rating a patient’s health preference.

For those patients who received TAVI, the findings were “very gratifying,” says Cohen. “These patients demonstrated a substantial benefit in both their heart failure-specific quality of life, as well as their general quality of life.”

The QoL difference between the two arms began to emerge at 30 days, and continued to increase through one year, Cohen notes. Based on the overall KCCQ scores, the TAVI arm had a 21-point improvement in scores by six months, reaching 24.5 points at one year—differences that were both statistically and clinically significant. For example, a 10-point difference in the KCCQ score (a moderately large benefit) was observed in 77 percent of patients in the TAVI group, compared with just 34 percent with standard therapy. A 20-point difference (a large benefit—comparable to a two-level improvement in NYHA functional class) was seen in 62 percent of the TAVI group and just 23 percent of the standard care group.  

“Similarly, there were benefits for the TAVI arm on each of the other components of the KCCQ, as well as the general health measures,” Cohen says. The difference on the SF-12 scale was approximately five points, which is equivalent to the differences that are seen with 10 years of aging, he explains.  

However, Lazar points out that the PARTNER trial did not record or define which patients were living independently or living in a rehabilitation facility or which specific activities these patients could undertake after TAVI. “To assess whether this procedure is appropriate for our patients, physicians need to understand how many patients can function on their own post-procedurally, he says. “Practically, it’s important to know if this procedure permits these very sick patients with symptoms of aortic stenosis to live at home or spend time with their grandchild.”

Thomas notes that since there has been such tremendous focus on mortality, the European registries and patient data documentation for QoL has been “poor.” However, he anecdotally has witnessed “impressive improvements” in QoL for his patients, and hypothesizes that these data will undergo better documentation in the coming years.

Cost effective?

Because the valves are not currently priced in the U.S., a true cost-effectiveness analysis for the American market is not possible. In Europe, the valves currently cost approximately $27,000 to $30,000 USD.

“These valves are much more expensive than the typical surgical valves in the U.S., which are closer to $5,000 to $6,000, but those costs include far more bells and whistles that allow for this very complicated procedure,” Cohen says. He and his PARTNER investigators are in the beginning stages of evaluating the cost data of TAVI versus standard therapy. “While nothing conclusive can be stated now, the initial data look promising for TAVI to be reasonably cost effective, just based on mortality benefits alone, even though the procedure is quite expensive,” he says.

Also, Thomas says that cost effectiveness will be integrally tied with patient selection because “neither surgery nor TAVI will have a great advantage over medical therapy for patients who are too high-risk. Similarly, if TAVI was utilized in a lower risk population, in whom surgical valve replacement is very effective, it most likely won’t have any better outcomes for mortality or quality of life. Until the cost of the device drops, it won’t be cost-effective in low-risk populations.”  

An additional cost consideration for the hospital is whether or not to keep an operating room (OR) or hybrid OR available for these procedures. Building a hybrid OR can cost approximately $5 million, which might only be used for backup.

“Now that the complication rate with angioplasty is down to approximately 0.4 percent, it wouldn’t be economically feasible to keep the staff and OR on standby,” Lazar says. “However, with facilities beginning to open structural heart programs, the institutional costs are not just based on consultation fees from surgeons, but also for the necessity of the institution to keep a back-up OR available.

“Also, the larger ethical-economic question is whether 90-year-old women with a series of comorbidities—from which they are likely to die—should undergo these complex, expensive procedures,” he says. Lazar suggests that CMS and third-party payors will assess this when considering whether to reimburse approximately $50,000 to $60,000, including hospitalization costs, for TAVI procedures.

Thomas sees the bigger concern as how money is distributed within the U.S. healthcare system, particularly with regards to the potentially unethical appearance or nature of an individual American physician who could reap financial benefit from recommending patients undergo TAVI procedures. In contrast, within the NHS, physician salaries are fixed, regardless of what procedures are recommended.

Real-life clinical practice

Traditionally, operator experience has been critical to successfully performing such difficult procedures, yet Leon notes that PARTNER incorporated many U.S. centers that were on the earliest arc of the learning curve. Of the 17 U.S. centers, only three had any experience with TAVI prior to PARTNER. However, it is difficult to know whether their success will translate into real-life clinical practice.

Many physicians question who will eventually perform these procedures. “While surgeons are being trained in the procedure, the technical skills of interventional cardiologists with catheter-related procedures make them the likely candidates to perform the procedures,” Lazar says. “Also, the guidelines need to determine which patients will go to the operating room emergently, if the TAVI procedure fails.”

While there has been some discussion of a bridge profession, requiring surgeons to acquire interventional skills, Lazar sees the two specialties as having distinct skill sets.

Thomas concurs that the bridge profession might not be necessary, as the team approach has proven effective for decision making, as well as having both surgeons and interventionalists present during transapical and transfemoral procedures.

“While some may question the necessity of having interventionalists and surgeons present during TAVI procedures as well as its costliness, we’re still in the beginning stages of understanding this condition and undertaking the TAVI technique, and we continually learn how helpful it is to rely on each other’s skill sets,” he concludes.