PARTNER Shows Continued Benefits of TAVR, but STACCATO Introduces Questions About Transapical Use

MIAMI BEACH, FLA.—New data from the nonrandomized PARTNER continued access registry for transfemoral and transapical approaches demonstrate that the favorable results achieved with TAVR in high-risk patients in the randomized, controlled trial are reproducible in a real-world setting, according to presentations at TCT 2012. Data from the prematurely terminated STACCATO trial, however, raised concerns about the effects of TAVR specifically via transapical access.

Beyond PARTNER

Cohort A of the PARTNER trial compared TAVR using the Edwards Sapien valve with conventional surgery in high-risk patients. Of the 348 patients in cohort A, 104 underwent the transapical approach at 14 centers, Todd M. Dewey, MD, of Southwest Cardiothoracic Surgeons, Dallas, TX, said at TCT 2012.

Researchers used Kaplan-Meier estimates to create an intention-to-treat analysis comparing PARTNER cohort A to the nonrandomized PARTNER Continued Access cohort (n=988 at 22 sites). Thirty-day in-hospital outcomes showed significantly lower rates of stroke, stroke/TIA, major vascular injury and valve embolization in the nonrandomized registry vs. the randomized trial (see Table 1), with similar mortality, cardiac mortality and rehospitalization rates. At one year, stroke and the composite of death/major stroke were lower in the Continued Access registry, with mortality showing a trend favoring the registry, and cardiac mortality showing similar rates in both groups (see Table 2).

Two-year data also showed a higher rate of stroke (12.5% vs. 6.1%) in the randomized trial, but no difference in rehospitalization after undergoing the transapical approach. Overall mortality showed a trend favoring the registry at 2 years (40.7% vs. 33.6%; P=.16).

Researchers also observed a significant improvement in return to NYHA functional class after 30 days in registry patients.

Transfemoral TAVR

Susheel K. Kodali, MD, of Columbia University, compared data from transfemoral patients in the nonrandomized PARTNER Continued Access registry (n=423) with results from transfemoral patients in the randomized PARTNER trial (n=1,080). Kodali stressed it is important to note that the FDA initially limited enrollment to 39 cases per month, allowing for one to two cases per site per month.

Data on 30-day in-hospital outcomes showed a lower rate of stroke or TIA in the nonrandomized registry (6.1% vs. 3.8%; P=.05). One-year outcomes also favored the nonrandomized cohort in terms of stroke and death or major stroke (see Table 3).

Procedural mortality was similar between groups, but neurological complications were less frequent in the nonrandomized trial (5.4% vs. 3.3%; P=.06).

Reflecting on STACCATO

Leif Thuesen, DMSc, of the Aarhus University Hospital in Denmark, discussed the prospective, randomized STACCATO trial comparing transapical TAVR with surgical aortic valve replacement in elderly patients. The study was prematurely terminated after reviewing the results of 70 patients due to advice from the Data Safety Monitoring Board.

Thuesen reported that at 1 month, there was one death among those on the waiting list in the TAVR group; one left main occlusion in a patient who was operated on and died; two major strokes; and one case of renal failure in the surgical group. There was only one major stroke between 1 and 3 months; the patient experienced a TIA that later developed into major stroke and partially recovered. There was one case of severe paravalvular leakage, and the patient died postoperatively. Between months 3 and 12, there was one major stroke, another paravalvular leakage.

Thuesen said the major issue was with paravalvular leakage — 43% of patients had minor leakage and 13% had moderate or severe leakage in the TAVR group vs. virtually no problems in the surgical group. He noted, however, that this may change with improved catheter-based devices and better pre-procedural assessment.