GARY: Transapical TAVR Appears Inherently Riskier Than Transvascular

BARCELONA, Spain—One-year mortality is higher for patients who undergo transcatheter aortic valve replacement (TAVR) via the transapical rather than the transvascular approach. Device type, on the other hand, appears to have no bearing on 1-year mortality after multivariable adjustment for possible confounders, according to data from the GARY registry presented September 2, 2014, at the European Society of Cardiology Congress.

“This is the first large-scale registry comparing valve types over 1-year follow-up,” said researcher Christian Hamm, MD, of Kerckhoff Clinic (Bad Nauheim, Germany), noting that TAVR outcomes in the high-risk population are comparable with those seen in randomized trials.

Baseline characteristics differed substantially between the transvascular and transapical TAVR groups. For example, those undergoing transvascular procedures were older, more likely to be women, and less likely to have CAD or prior cardiac surgery, and they had higher mean ejection fraction. Transapical TAVR was more commonly elective, used less contrast dye, involved more bleeding, and resulted in longer hospital stays. Although periprocedural stroke was similar between groups, transvascular TAVR was associated with higher rates of periprocedural TIA and vascular complications.

At 1 year, transapical TAVR carried higher risk of mortality compared with transvascular TAVR, even after adjusting for confounders (OR 1.398; 95% CI 1.217-1.606). Stroke and TIA remained equivalent for the 2 approaches (table 1).

Table 1. One-Year Outcomes by TAVR Approach

More than half of transvascular procedures (55.9%) used the CoreValve device (Medtronic; Minneapolis, MN) rather than the Sapien (41.3%; Edwards Lifesciences; Irvine, CA) or other devices (2.6%). The majority of transapical procedures (79.1%) used Sapien, with the rest involving:

• ACURATE valve (Symetis; Lausanne, Switzerland): 9%
• JenaValve (JenaValve Technology; Munich, Germany): 6.1%
• Engager (Medtronic): 3.5%
• Other valve: 2.3%

In the transvascular group, many baseline characteristics were different between patients treated with CoreValve vs Sapien. CoreValve patients were younger, were more likely to be men, had lower ejection fraction, and had higher aortic valve area. They were more likely to be undergoing urgent/emergent treatment but less likely to have limiting secondary disease.

Accordingly, many outcomes also differed between the CoreValve and Sapien groups in patients who underwent transvascular TAVR. Crude 1-year mortality for CoreValve was higher than for Sapien, though the difference lost significance after multivariable adjustment (OR 1.075; 95% CI 0.919-1.258). Need for new pacemaker/ICD and aortic regurgitation both were increased with CoreValve, while stroke was more common with Sapien (table 2).

Table 2. Transvascular TAVR: Outcomes by Device Type

In transapical TAVR, mortality was similar among the 4 identified devices (log-rank P = .093).

Dr. Hamm concluded that TAVR “is safe and effective in this high-risk population. The transfemoral approach should be preferred over the transapical approach, if this is feasible. Valve types at this stage are not different, but I think this needs further observation over time, with more patients.”

No Obvious Explanations for Risk Variation

Panelist Sarah Catherine Clarke, MD, of Papworth Hospital (Cambridge, England), inquired about variations in the degree of coronary disease among groups—and how such disease was managed pre-TAVR—asking whether it might influence outcome.

“There were some differences in terms of coronary disease between the transapical and transvascular approach, interestingly,” Dr. Hamm replied, noting that the specifics of staged procedures can be difficult to pin down. “But we didn’t see any difference with the CoreValve and Sapien [groups].” 

Dr. Hamm also noted that there is no obvious explanation why 1-year mortality would be higher for transapical TAVR after multivariable adjustment. “Maybe they’re more invasive procedures,” he suggested. “It’s just a fact. We tried to correct for all confounders, all the risk variables we know, and still it remains a higher-risk procedure.”

Another question, raised by Geneviève Anne Derumeaux, MD, PhD, of Université Paris-Est Créteil (Créteil, France), in the discussion was whether the TAVR approaches and device types were evenly distributed across participating centers. 

Dr. Hamm pointed out that while many centers do both transapical and transvascular TAVR, some only offer transvascular. Case volume also varied, and was as low as 20-50 procedures at some centers, he added. Lower-volume centers were likely to only use 1 valve, yet volume did not appear to affect the relationship between device type and mortality.

An audience member asked whether the transapical patients died from complications specifically related to the approach used, such as damage to the mitral valve. “We tried to look at it and so far, we have not found any indicators of why they died,” Dr. Hamm replied, mentioning the possibilities of more bleeding and longer hospital stay, likely due to complications. “It must be related somehow to that.”

Source: 
Hamm C. German Aortic Valve Registry: differential 1-year outcomes in TAVI procedures. Presented at: European Society of Cardiology Congress; September 2, 2014; Barcelona, Spain.

Disclosures:

• The GARY registry is an investigator initiated study with unrestricted grants from Edwards, JenaValve, Medtronic, Sorin, St. Jude, and Symetis. It receives support from the German Cardiac Society, German Society for Thoracic and Cardiovascular Surgery, and the German Heart Foundation.
• Dr. Hamm reports receiving lecture fees from Edwards Lifesciences and Medtronic and honoraria for advisory board activities from Medtronic.

Related Stories:

• GARY: Majority of Patients Do Well With TAVR or Surgery at 1 Year
• TAVR Improves Quality of Life, Reduces Hospitalization in German Registry
• German Valve Registry Shows Advantage of Surgery in Lower Risk Patients