French Registry Provides Window into Real-World TAVR Use

Download this article's Factoid in PDF (& PPT for Gold Subscribers)



A large French registry capturing the country’s entire experience with transcatheter aortic valve replacement (TAVR) shows that the procedure excels in actual practice, with 1-year mortality matching what has been seen in a clinical trial setting. The findings appear in the May 3, 2012, issue of the New England Journal of Medicine.

Researchers led by Martine Gilard, MD, PhD, of CHU de la Cavale Blanch (Brest, France), analyzed data from the prospective FRANCE 2 (French Aortic National CoreValve and Edwards) registry on 3,195 patients with severe aortic stenosis treated at 33 centers in France and 1 in Monaco from January 2010 to October 2011. Most were highly symptomatic, with 77.8% having New York Heart Association (NYHA) class III or IV. Mean age was 82.7 years, with women comprising 49% of the cohort.

Mortality Predicted by Patient, Procedural Factors

In all, 66.9% of patients received either the Sapien or Sapien XT device (Edwards Lifesciences, Irvine, CA) and 33.1% the CoreValve device (Medtronic, Minneapolis, MN). The majority of cases were performed transfemorally (74.6%), with the remainder using the transapical (17.8%), subclavian (5.8%), or transaortic or transcarotid (1.8%) approach. Procedural success was achieved in 96.9% of cases, and the average hospital stay lasted 11.1 ± 8.0 days.

Overall, 15.6% required a new pacemaker and 4.5% experienced major bleeding. The researchers documented major outcomes at 30 days, 6 months, and 1 year, with death from any cause serving as the primary endpoint (table 1). By 1 year, 89.5% of surviving patients were asymptomatic or mildly symptomatic with NYHA class I or II.

Table 1. FRANCE 2 Registry: Outcomes After TAVR

 

30 Days

6 Months

1 Year

All-Cause Death

9.7%

18.6%

24.0%

Stroke

3.4%

3.8%

4.1%

Periprosthetic Regurgitation

64.5%

62.9%

66.9%


Multivariable analyses found several independent predictors of 1-year mortality:

  • Logistic EuroScore (HR 1.37 per 1% increase; 95% CI 1.19-1.58)
  • NYHA class III or IV (HR 1.49 vs. class I or II; 95% CI 1.09-2.03)
  • Use of the transapical approach (HR 1.45 vs. transfemoral; 95% CI 1.09-1.92)
  • Periprosthetic regurgitation grade 2 or higher (HR 2.49 vs. lower grade; 95% CI 1.91-3.25)

Several differences were noted between the years 2010 (n = 1,551) and 2011 (n = 1,640), including decreases in baseline logistic EuroScore (from 22.8 ± 14.8 to 20.9 ± 13.8), use of the transapical approach (from 19.5% to 16.2%), use of general anesthesia (from 73.4% to 65.0%; P < 0.001 for all comparisons). Thirty-day mortality remained the same over the course of the study. The segment of patients preferring TAVR to surgery, meanwhile, increased from 14.4% in 2010 to 17.1% in 2011 (P = 0.04).

The Inevitable Comparison

While the 30-day death rate of 9.7% falls in the range of what had been reported by other registries, the paper notes, it exceeds what was seen in the PARTNER trial (5.2% and 5.0% in high-risk patients who were and were not eligible for surgery, respectively). But by 1 year, the death rate in FRANCE 2 was 24.0%, similar to the 24.2% rate achieved at 1 year in surgical candidates in PARTNER’s cohort A.

 

The risk of stroke also was similar to what was seen in PARTNER, Dr. Gilard and coauthor Hélène Eltchaninoff, MD, of the University of Rouen (Rouen, France), confirmed to TCTMD in an e-mail communication, but the need for permanent pacemaker was higher in FRANCE 2: 11.5% with Sapien and 24.2% with CoreValve. PARTNER evaluated only the Sapien valve, with less than 4% of patients needing pacemakers. Major bleeding and complication rates, though, were higher in PARTNER, they said, perhaps due to varying endpoint definitions, some trial sites being less experienced, and the use of larger size introducers.

Overall, Drs. Gilard and Eltchaninoff commented, “[c]omplication rates seem to be acceptable considering the frequent comorbidities of this population. TAVI appears to be a reasonable option for a selected population of inoperable or high-risk patients with severe symptomatic aortic stenosis.”

Patient selection is crucial to improving procedural safety and subsequent outcomes, they said. “Procedural modifications [such as the smaller introducer size now available for use with Sapien XT] will facilitate the use of [the transfemoral] approach by decreasing the size of introducers to 16-18 Fr and should further decrease the risk of complications.”

Peter C. Block, MD, of Emory University Hospital (Atlanta, GA), told TCTMD in a telephone interview that this cohort is a “mixed bag, so to make the comparison with PARTNER is not fair, but what it does tell us is when you release this [into] the real world [the] outcomes certainly do not become inferior to a very controlled clinical trial in high-risk patients.” As would be expected, the FRANCE 2 patients fared slightly better than PARTNER’s cohort B, which enrolled patients truly ineligible for surgery and had 1-year mortality of 30.7%, he added.

The crucial question is how to identify patients likely to die within 1 year, Dr. Block noted. “It would be good if we just simply did not [perform TAVR] in those patients,” he observed, adding that stratifying according to Society of Thoracic Surgeons (STS) score might be helpful, as “the sicker you are when you go in, the less chance you have of coming out.” Dr. Block reported that medical therapy performs just as well as TAVR in the highest-risk patients.

US Experience Largely Unknown

Asked how FRANCE 2 compares with the US experience, Dr. Block said that little is known about commercial use of TAVR in this country since the Edwards Sapien device’s approval by the Food and Drug Administration in November 2011. “The problem with the commercial valves here,” he noted, “is that people have limited their use primarily to . . . truly nonoperative patients,” making it difficult to gauge these outcomes against more varied populations like FRANCE 2.

The study authors related that European operators benefit not only from a longer TAVR experience but also from technological improvements.

“We in the United States have this obsession with randomized trials,” Dr. Block concluded, “which is a very good thing in many ways but handicaps us severely in others. Europeans have taken the art of registries to lovely levels. [The FRANCE 2 study] once again demonstrates that registry data, though perhaps not as perfect as randomized trial data, seem to repetitively show that things come out about the same. [It] probably means that there’s truth there.”

Study Details

Most patients were at high surgical risk (logistic EuroScore 21.9 ± 14.3 and STS score 14.4 ± 11.9). A quarter of patients (26.1%) were at lower risk but had various coexisting conditions that contraindicated surgery, such as severe calcification of the aorta, chest-wall deformity, or irradiation damage. A small proportion (15.8%) opted to undergo TAVR rather than conventional surgery.

 


Source:
Gilard M, Eltchaninoff H, Iung B, et al. Registry of transcatheter aortic-valve implantation in high-risk patients. N Engl J Med. 2012;366:1705-1715.

 

 

Related Stories:


Click here for a listing of companies that provide support to the Cardiovascular Research Foundation, owner and operator of TCTMD.

Disclosures
  • Disclosure information for Dr. Gilard and colleagues is available online at www.nejm.org.
  • The study was supported by Edwards Lifesciences and Medtronic.
  • Dr. Block reports being an investigator for the PARTNER trials sponsored by Edwards Lifesciences and serving as a proctor for Edwards.

Comments