Aortic Regurgitation Confirmed as Powerful Mortality Predictor after TAVR

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

Moderate to severe post-procedural aortic regurgitation (AR) after transcatheter aortic valve replacement (TAVR) is the strongest predictor of 1-year mortality in such patients, according to results from the FRANCE 2 registry published online February 24, 2014, ahead of print in Circulation.

The study was first presented in an abstract session at the European Society of Cardiology Congress in Amsterdam, the Netherlands in September 2013.

Eric Van Belle, MD, PhD, of University Hospital (Lille, France), performed postprocedural transthoracic echocardiography (TTE) in 2,769 patients who underwent successful TAVR at 34 French centers and were enrolled in the FRANCE 2 registry. More than two-thirds (67.6%) of patients received the balloon-expandable Sapien or Sapien XT devices (Edwards Lifesciences, Irvine, CA; n = 1,872), while the remainder received the self-expandable CoreValve (Medtronic, Minneapolis, MN; n = 897).

Strong Independent Predictor

At a median follow-up duration of 306 days, moderate/severe postprocedural AR  (≥ grade 2) was seen in 15.8% of patients and was more frequent in those receiving a self-expandable than a balloon-expandable device (21.5% vs 13%; P = 0.0001).

On multivariable analysis, use of a self-expandable device (OR 2.03; P < 0.0001) and femoral approach (OR 1.72; P = 0.0003) were independent predictors of moderate/severe postprocedural AR. When the devices were analyzed separately, 8 independent predictors of moderate/severe postprocedural AR were identified for balloon-expandable devices. However, the presence of AR > grade 2 at baseline (OR 1.45; P = 0.04) and use of a femoral approach (OR 2.25; P = 0.003) were the only independent correlates for the self-expandable device.

Additionally, moderate/severe postprocedural AR was a strong independent predictor of 1-year mortality for both balloon-expandable (HR 2.50; P = 0.0001) and self-expandable (HR 2.11; P = 0.0001) device use.

While moderate/severe postprocedural AR was well tolerated in those who had the condition at baseline, the complication was associated with increased mortality in several subgroups characterized by:

  • Renal failure (43%)
  • AR > grade 2 at baseline (31%)
  • Low transaortic gradient (35%)
  • Non-femoral access (45%)

While no interaction was observed between the type of device and most of the predictors of moderate/severe postprocedural AR, there was a relationship between type of device and both aortic annulus (P = 0.04) and prosthesis (P = 0.003) diameter.

During follow-up, there were 312 deaths (11.3%), including 175 (6.3%) from cardiovascular causes. Compared with no AR, moderate/severe postprocedural AR was associated with higher all-cause and cardiovascular morality (table 1).

Table 1. Mortality: Moderate/Severe vs No Postprocedural AR


HR (95% CI)

P Value

All-cause Mortality

2.33 (1.82-2.99)

< 0.0001

CV Mortality

2.36 (1.70-3.29)

< 0.0001

Highlights Role of the LV

The study authors say the fact that minimal impact on 1-year survival was seen in patients with significant baseline aortic regurgitation “may suggest that the LV of patients with preexisting AR is already adapted to volume overload and can thus better tolerate postprocedural AR.” Conversely, in patients without preexisting AR, “the pronounced concentrically remodeled and small LV cavity, with reduced compliance and impaired filling, will poorly tolerate postprocedural AR,” they write.

Regarding the difference in AR rate when a femoral vs non-femoral approach is used, they say this “may suggest that catheter control during delivery, which is typically more difficult during a femoral than a non-femoral procedure, is important for minimizing the risk of AR with the [self-expandable] device. This observation, combined with the trapezoidal shape of the lower end of the [self-expandable] device, may also suggest that femoral delivery, because it makes the depth of implantation more difficult to control, creates additional uncertainty regarding the ‘true’ diameter of the prosthesis at the level of the aortic annulus, and thus increases the risk of AR. This further suggests that, in order to reduce the risk of AR with [self-expandable] devices, future directions might include a modified shape and/or a repositionable ability.”

Confirmatory Data

In an editorial accompanying the study, E. Murat Tuzcu, MD, of the Cleveland Clinic (Cleveland, OH), and colleagues say the new report confirms previously published data from a meta-analysis and several European registries.

As for patients treated non-femorally, they note that it is “possible that these patients whose comorbidity profile is worse are challenged by the stresses of the procedure even though the surgical trauma is limited. Whether there is any consequence of apical suturing and possible alteration of diastolic dysfunction as a mechanism for not being able to tolerate AR remains to be studied.”

However, Dr. Tuzcu and colleagues also observe that generalizability of the findings is limited by TTE assessment, no core lab oversight, lack of post discharge follow-up, and limited information on the 3D annular size.  

Mortality Findings at Odds

In a telephone interview with TCTMD, Ted Feldman, MD, of Evanston Hospital (Evanston, IL), said “the challenge is reconciling these findings with the recent CoreValve extreme-risk presentation that didn’t show an impact of moderate AR on mortality at a year.” He explained that while those CoreValve data were abbreviated, 1-year mortality in the extreme-risk CoreValve group was almost identical to that of the moderate AR group in the current analysis.

“So while the CoreValve [extreme-risk] report did not find a statistical difference between trivial, mild, or moderate AR and death at 1 year, there’s a numerical difference,” he added. Dr. Feldman said the new data showing how different subgroups are affected by post-TAVR AR may provide an explanation for the lack of predictive effect in the extreme risk cohort, but only further mining of the data will be able to confirm this theory.

However, he continued, given that next-generation TAVR devices are “so effective at minimizing paravalvular leak,” the discussion is becoming less and less relevant.

Study Details 

Half of the population was female (48.9%), the mean age was 83 ± 7 years, 75% were NYHA 3 or 4, and the mean logistic EuroScore was 21.5 ± 13.8. The procedure was performed transfemorally in 75.3% of cases.


1. Van Belle E, Juthier F, Susen S, et al. Post-procedural aortic regurgitation in balloon-expandable and self-expandable TAVR procedures: analysis of predictors and impact on long-term mortality: insights from the FRANCE 2 registry. Circulation. 2014;Epub ahead of print.

2. Tuzcu EM, Kapadia SR, Svensson LG. Valve design and paravalvular aortic regurgitation: new insights from the French registry [editorial]. Circulation. 2014;Epub ahead of print.



Related Stories:

  • The FRANCE 2 Registry was supported by the French Ministry of Health and grants from Edwards Life Science and Medtronic.
  • Drs. Van Belle and Tuzcu report no relevant conflicts of interest.
  • Dr. Feldman reports receiving research grants from and serving as a consultant to Abbott and Boston Scientific.