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

A comprehensive national registry of transcatheter aortic valve replacement (TAVR) encompassing both major transcatheter bioprostheses and multiple access routes has yielded good periprocedural and encouraging midterm survival. However, according to results published online October 19, 2011, ahead of print in the Journal of the American College of Cardiology, there was considerable attrition between 30 days and 1 year

Investigators led by Neil E. Moat, MBBS, MS, of the Royal Brompton and Harefield National Health Service Foundation Trust (London, United Kingdom), evaluated outcomes in 870 patients undergoing 877 TAVR procedures at 25 centers in the United Kingdom between January 2007 and December 2009. The cohort, which represents all procedures performed in the country  and enrolled in the United Kingdom Transcatheter Aortic Valve Implantation registry, received similar numbers of Edwards Sapien (n = 410; Edwards Lifesciences, Irvine, CA) and CoreValve (n = 459; Minneapolis, MN) devices; in 8 cases the type of bioprosthesis was unknown.

TAVR was successful in 97.2% of patients. The transcatheter procedure was converted to surgical replacement in 6 patients, all of whom were receiving an Edwards valve delivered via the transapical approach. 

Strong Early Survival, Then Marked Risk Out to 1 Year 

Overall survival at 30 days was 92.9%, with a lower mortality rate among patients treated with a transfemoral compared with other approaches (5.5% vs. 10.7%; P = 0.006).  Postprocedural complications included stroke, MI, moderate/severe aortic regurgitation, major vascular complications, and new permanent pacemaker implantation. Patients receiving CoreValve devices were more likely to need pacemaker implantation or to experience moderate or severe aortic regurgitation (table 1). Major vascular complications were more common in the transfemoral group.

Table 1. Major Complications at 30 Days

 
There was a marked decline in survival between 30 days and 6 months (9.6% mortality) and between 6 months and 1 year (4.7% mortality). In Kaplan-Meier estimates, 1-year survival was 78.6% and 2-year survival was 73.7%. No differences in 1-year survival were seen between the Sapien and CoreValve cohorts. However, patients who underwent nontransfemoral implantation experienced higher 1-year mortality than those who received transfemoral procedures (22.7% vs. 18.5%), a difference that persisted at 2 years (36.7% vs. 22.5%; P = 0.0013).

Predictors of Midterm Mortality

In multivariate analysis, independent predictors of 1-year mortality were:

• LVEF less than 30%
• COPD
• Moderate or severe aortic regurgitation

In both univariate and multivariate analyses, age, New York Heart Association (NYHA) class, and peripheral vascular disease were not predictive.

Moreover, there was no difference in 1-year survival between patients with a logistic EuroScore of 0 to 20 and 21 to 40 but there was a marked reduction in midterm survival for those with a logistic EuroScore greater than 40 (P = 0.0001). According to the authors, this finding “reaffirms the relative lack of utility in this scoring system in risk/outcome prediction for this group of patients and confirms the need for more sophisticated and procedure specific (rather than generic) scoring systems.”

In an email communication with TCTMD, Dr. Moat said it is important to identify which patients are at especially high risk of dying between 30 days and 1 year and to try to reduce this attrition through better medical management. If that cannot be achieved, perhaps these high-risk patients should not receive TAVR, he added.

The investigators note that these data encompass the learning curve with this technique across the United Kingdom, and the fact that 30-day and midterm mortality rates were equivalent between proctored and nonproctored cases and for the first 20 cases compared with subsequent cases at each center “reflects well” on both device companies’ mentoring programs. “Further growth and dissemination of [TAVR] should retain this robust training structure in addition to a strict multidisciplinary approach,” they add.  

A Glimpse of the Future

In an accompanying editorial, Alec Vahanian, MD, and colleagues of Hôpital Bichat (Paris, France), observe that because this registry covers a diverse experience, it “could give a sense of what will happen in the future when most centers will use varied devices and approaches … to offer the most appropriate treatment to the individual patient.”

Addressing the predictors of midterm mortality, the editorial notes that with regard to low LVEF, TAVR may be more effective than surgery at improving LV function in aortic stenosis. However, this possibility requires further investigation since currently low LVEF is a relative contraindication to TAVR, the editorial adds.

On the other hand, identification of COPD as an independent predictor should prompt consideration of whether a patient’s incapacity is due primarily to aortic stenosis or to COPD, Dr. Vahanian and colleagues write. Moreover, participation of noncardiologists in patient selection can help ensure that TAVR is not performed in those with a limited life expectancy or ability to recover, they suggest.

The adverse implications of moderate or severe aortic regurgitation also highlights the importance of patient selection, in particular the use of imaging to match the size of the annulus to the bioprosthesis and to evaluate the magnitude and location of calcification.

“All our efforts to pursue the development of [TAVR] should aim at improving patient selection both by a dedicated medicosurgical team and by improving procedural performance through careful training and improvement in technology,” Dr. Vahanian and colleagues conclude.

Study Details 

The majority (69%) of valves were implanted by the transfemoral route. But while almost 90% of CoreValve procedures were transfemoral, more than half of Edwards implantations were transapical. The overall mean logistic EuroScore was 18.5%; it was higher in the nontransfemoral than the transfemoral group (21.4% vs. 17.1%; P < 0.0001).

Patients treated with Edwards Sapien valves tended to be older and were more likely to be in NYHA functional class III/IV and to have coronary and peripheral vascular disease. 

Peripheral vascular disease, CAD, prior cardiac surgery, renal dysfunction, and NYHA class III/IV were more common in patients undergoing TAVR via a nontransfemoral than a transfemoral approach.

2. Vahanian, A, Himbert, D, Iung B. Transcatheter aortic valve implantation. A snapshot from the United Kingdom. J Am Coll Cardiol. 2011;Epub ahead of print.

Related Stories:

• Comprehensive French TAVI Registry Shows Promising Results
• SOURCE Registry Data Published: TAVI Stays Strong at 1 Year
• TAVI with CoreValve Effective, Safe at 2-Years, Though Mortality a Concern