Recovery After Balloon Aortic Valvuloplasty in Patients With Aortic Stenosis and Impaired Left Ventricular Function: Predictors and Prognostic Implications

Download a PDF of this article.

Abstract: Aims. The aim of this study was to evaluate predictors of recovery after balloon aortic valvuloplasty (BAV) among patients with aortic stenosis and depressed left ventricular ejection fraction (LVEF). Predictors for recovery after BAV are not clearly defined. B-type natriuretic peptide (BNP) predicts outcome after surgical and transcatheter aortic valve replacement. Methods and Results. Among 151 consecutive patients treated in our institution by BAV, a total of 59 with poor LVEF underwent an echocardiography at 1 month. In these 59 patients, LVEF significantly improved in 22 patients (group 1) from 27 ± 5% to 45 ± 6% (P<.0001) and remained unchanged in 37 patients (group 2) from 29 ± 8% to 30 ± 11% (P=NS). BNP plasma levels at 24 hours only decreased in group 1 from 2170 ± 967 pg/mL to 1208 ± 662 pg/mL (P=.001). By multivariate analysis, BNP reduction >300 pg/mL was the strongest independent predictor of LVEF improvement at 30 days (hazard ratio, 5.459; 95% confidence interval, 1.580-18.860; P=.007). Kaplan-Meier analysis showed that 1-year survival after BAV was significantly higher in patients of group 1 than in group 2 (95 ± 4% vs 51 ± 8%, respectively; P=.02). Conclusions. BAV in patients with poor left ventricular function resulted in LVEF improvement at 30 days in 37% of cases, which was detected by a reduction of BNP levels already seen at 24 hours. Survival at 1 year was significantly higher in patients with such an improved LVEF after BAV.

J INVASIVE CARDIOL 2013;25(5):235-241

Key words: B-type natriuretic peptide, aortic stenosis, balloon aortic valvuloplasty, left ventricular function  

________________________________________________

Patients with severe aortic stenosis (AS) and depressed left ventricular ejection fraction (LVEF) have a poor prognosis when treated conservatively. Surgical aortic valve replacement (AVR) is recommended by the guidelines¹ despite an increased operative risk of morbidity and mortality.^2-4 Percutaneous aortic balloon valvuloplasty (BAV), first described by Cribier et al⁵ in the mid-1980s, is a less invasive technique, but its use was markedly reduced over time due to the complexity of the procedure and a high mid-term restenosis rate.⁶ Furthermore, improvements in balloon and catheter technology,⁷ use of rapid pacing during balloon inflation,⁸ improved imaging for aortic annulus measurements in guiding balloon size selection, as well as use of vascular closure devices have dramatically diminished the procedural complications rate.⁹ BAV has been shown to improve LVEF^10,11 and has been used either as compassionate therapy or to stabilize patients in critical hemodynamic situations as a bridge therapy to surgical aortic valve replacement (AVR) or transcatheter aortic valve implantation (TAVI).^1,12,13

Predictive factors for clinical recovery and LVEF improvement after BAV are not yet well defined. In a cohort of elderly patients with multiple comorbidities, it could be difficult to evaluate whether BAV would be beneficial. In addition, optimal delay between BAV and TAVI or AVR remains unclear, but it can be estimated that a too-long waiting time would be detrimental to the patient. Therefore, identification of recovery markers at the time of BAV would be useful to predict prognosis and to discriminate the candidates for bridge therapy following the procedure. 

B-type natriuretic peptide (BNP) is a well-known predictor of clinical outcome after AVR and TAVI,^14,15 but has not been investigated in patients undergoing BAV.

The aim of the present study was to identify predictive factors of recovery after BAV among patients with AS and low LVEF. In addition, procedural safety and outcomes were compared with patients who had preserved LVEF.

Methods

Patients. Between January 2006 and April 2011, a total of 151 consecutive patients were treated at our institution with BAV. All patients had severe symptomatic AS (New York Heart Association [NYHA] class II [n = 2], NYHA class III [n = 21], and NYHA class IV [n = 128]) confirmed by transthoracic echocardiography. Patients were referred for BAV as preoperative procedure before urgent non-cardiac surgery (n = 7), for palliation of heart failure symptoms (n = 92), or as a bridge therapy to surgical valve replacement (n = 7) or to TAVI (n = 45). 

Echocardiography. All patients underwent a transthoracic echocardiography before the BAV procedure (within 24 hours for the majority of them; delay up to 3 days in less than 1% of the cohort). Data acquisition was standardized and included LVEF calculated with the Simpson method, transvalvular pressure gradient determined by the Bernoulli formula, and aortic valve area calculated by the continuity equation. Aortic valve stenosis was considered as severe whether the aortic valve area was <1 cm² (or <0.6 cm²/m² body surface area) and/or peak gradient >40 mm Hg and/or peak velocity >4 m/sec. A reduced left ventricular function was defined as LVEF <40%.

A significant improvement in LVEF was defined as an increase of at least 10% of LVEF at 30 days, measured by echocardiography using the Simpson method. LVEF improvement was evaluated independently by two physicians experienced in echocardiography; after analysis of interobserver agreement, discordant findings (n = 6) between the two readers were resolved by consensus reading.

Balloon aortic valvuloplasty procedure. BAV was performed under local anesthesia according to standard technique via the retrograde femoral approach. A Swan-Ganz catheter (Edwards LifeSciences) was placed in the pulmonary artery from the femoral vein for hemodynamic assessment and cardiac output measurement using the thermodilution technique; a 6 Fr Soloist pacemaker lead (Medtronic) was placed into the right ventricle from the femoral vein for rapid pacing during the balloon inflation; a Crystal balloon (Balt) was introduced in a 10 Fr sheath in the femoral artery. Heparin (30-60 U/kg) was given after femoral sheath insertion. The balloon size was selected after aortic annulus measurement by transthoracic echocardiography, in order to obtain a balloon/annulus ratio close to 1. Mean and peak transvalvular aortic gradients were measured at baseline and after dilatation by a simultaneous pressure measurement in the left ventricle and in the aorta. Aortic valve area was measured at baseline and at the end of the procedure using the Gorlin formula. A contrast angiogram of the ascending aorta was performed before crossing the valve and after balloon dilatation in order to assess the grade of aortic regurgitation and to exclude aortic dissection. Arterial puncture site was closed using an 8 Fr Angio-Seal (St Jude Medical) and venous puncture sites were closed by manual compression.

The only contraindication for BAV was a baseline aortic regurgitation ≥ grade 3 determined by contrast aortogram. Major procedural complications were defined as death, stroke, tamponade, severe arrhythmias requiring electrical shock, complete atrioventricular block requiring pacing, profound hypotension requiring resuscitation, aortic regurgitation > grade 2, aortic dissection, myocardial infarction, renal failure, and serious vascular complication requiring intervention and/or transfusion.

Measurement of neurohormone levels. Blood was drawn from an antecubital vein into chilled EDTA tubes and the serum BNP level was determined using the fluorescent immunoassay (Triage BNP Test; Biosite Diagnostics, Inc). The assay analytical sensitivity was 5 pg/mL. The measuring range was 5-5000 pg/mL, with a variability of 8.8%. Blood samples were obtained at baseline as well as 24 hours after the BAV procedure.

Study protocol. The data of all the BAV procedures and in-hospital clinical events were prospectively entered into a dedicated database. Clinical follow-up was obtained by review of medical records or telephone interview of the patient or the referring cardiologist. Analysis of LVEF improvement was only performed in patients undergoing echocardiography in our institution both at baseline and at 30-day follow-up. The investigation conformed to the principles outlined in the Declaration of Helsinki. All subjects gave written informed consent to participate in the study, which was approved by our local ethical committee.

Statistical analysis. Data are presented as mean ± 1 standard deviation unless otherwise stated. Differences between patient groups were analyzed with a Student t-test for continuous variables or χ² test for categorical variables. The interobserver agreement for LVEF improvement was evaluated using the κ statistic, where κ <0.4 indicated poor agreement, 0.4< κ <0.6 indicated fair agreement and κ >0.6 indicated good agreement. The receiver-operating characteristic (ROC) curve was determined to evaluate the performance of BNP to predict LVEF improvement. Optimal cut-off points for serum levels of BNP were chosen when the sensitivity and specificity were maximized. Survival curves according to LVEF improvement and bridge therapy were established by the Kaplan-Meier estimation method. Multivariate analysis of the predictive factors of LVEF improvement and death were performed using the Cox proportional hazards method. A P-value <.05 was considered statistically significant. Analyses were performed using the XLSTAT software (version Pro 2009 for Windows; Addinsoft).

Results

Patients. Baseline characteristics of patients are listed in Table 1. LVEF was <40% at baseline among 69 patients, of which 59 underwent an echocardiography at our institution at 1-month follow-up. As compared with the cohort of preserved left ventricular function (group A; n = 82), patients with reduced LVEF (group B; n = 69) were less frequently female, had significantly higher logistic EuroSCORE, and a lower transvalvular gradient. Among patients of group B, all baseline characteristics were similar between patients with LVEF improvement (group 1; n = 22) and those without LVEF improvement (group 2; n = 37) at 30 days. Nine patients underwent redo BAV at a median time interval of 398 days.

Procedural outcome. Immediately after BAV, aortic valve area (using the Gorlin formula) increased from 0.6 ± 0.2 cm² to 0.9 ± 0.3 cm² and invasive mean gradient reduced from 35 ± 18 mm Hg to 17 ± 9 mm Hg (P<.005 for both). Significant reduction of transvalvular gradient and significant improvement of aortic valve area were observed in all subgroups of patients immediately after the balloon dilatation (Table 2). The baseline invasive mean gradients were significantly higher among patients of group A, and were more importantly reduced after BAV than in patients of group B. There were no differences between groups 1 and 2 regarding valve area and gradient values before or after BAV. Coronary angioplasty was associated among 9% of patients, in the same proportion of patients from both groups.

The total rate of major procedural complications was 13%; the main adverse events during BAV were arrhythmias (7%) including ventricular fibrillation/tachycardia (4%) and complete atrioventricular block (3%), followed by tamponade (2%), aortic regurgitation > grade 2 (observed in 2 cases; 1%) and femoral pseudoaneurysm/hematoma (1%). All complete atrioventricular blocks observed at the end of BAV recovered within 24 hours with no need for permanent pacemaker implantation. The rate of complications was similar in all subgroups except for ventricular arrhythmias, which were more frequently observed in group B than in group A. 

LVEF recovery. Among 59 patients with baseline reduced LVEF <40% undergoing an echocardiography at 30 days, LVEF significantly improved in 22 patients (group 1) from 27 ± 5% to 45 ± 6% (P<.0001) and remained unchanged in 37 patients (group 2) from 29 ± 8% to 30 ± 11% (P=NS) (Figure 1). The interobserver agreement between the two readers for LVEF improvement was good (κ= 0.79); the intraobserver variability was 5.5%. BNP levels at baseline were similar between groups 1 and 2 but significantly reduced 24 hours after BAV only in group 1 (Table 3).

By multivariate analysis, the change of BNP levels at 24 hours after BAV was an independent predictor of LVEF improvement at 30 days after the procedure (Table 4). ROC analysis showed that the optimal cut-off value for change of BNP level 24 hours after BAV to predict LVEF improvement at 30 days was 300 pg/mL, with a sensitivity of 84% and a specificity of 85%. Area under the curve was 0.85 (Figure 2). The extent of coronary artery disease — presence of coronary artery disease (73% vs 65%; P=.57); previous myocardial infarction (41% vs 38%; P=.1); previous bypass graft (14% vs 24%; P=.50) — was not significantly different between groups 1 and 2.

Clinical outcome. Symptom relief was immediately observed in a vast majority of patients (83%); 30-day and 1-year survival rates of patients were significantly higher among patients with clinical improvement than in those without (97% and 60% vs 65% and 19%, respectively). 

In-hospital adverse events were: death in 4 patients (2 refractory cardiogenic shock, 1 pulmonary sepsis, and 1 severe aortic regurgitation complicating BAV); 1 major ischemic stroke; and renal failure in 11 patients (7%) with no need for new dialysis. 

Overall 30-day and 1-year survival rates following BAV were 91% and 54%, respectively. Causes of death were cardiovascular in 78% of cases (54 recurrent heart failure and/or sudden death, 1 major stroke) and non-cardiovascular in 22% of cases (due to the comorbidities of the patients). By univariate analysis, survivors at 1 year had a lower prevalence of renal failure (28% vs 50%; P<.01), a lower logistic EuroSCORE (37% vs 44%; P<.05), and a lower baseline BNP level (1092 pg/mL vs 1809 pg/mL, P<.01); they also more often underwent a bridge therapy after BAV (56% vs 12%; P<.001). Multivariate analysis showed that renal failure, bridge therapy, and baseline BNP levels were independent predictors of 1-year mortality (Table 5).

In the total population, Kaplan-Meier analysis (Figure 3) showed that overall 1-year survival was significantly higher in patients

undergoing bridge therapy (84 ± 5%) than in those undergoing BAV as a palliative option (37 ± 5%; P<.0001).

Among patients with impaired left ventricular function, improvement of LVEF 30 days after BAV was an independent predictor of 1-year survival (hazard ratio, 0.270; 95% confidence interval, 0.078-0.933; P=.039). Kaplan-Meier analysis (Figure 4) showed that 1-year survival was significantly higher in patients with LVEF improvement following BAV versus those without (95 ± 4% in group 1 vs 51 ± 8% in group 2, respectively; P=.02). 

Discussion

The salient findings of this study including 151 patients who underwent BAV are: 

• Safety of BAV is similarly high in patients with impaired versus preserved left ventricular function, except for ventricular arrhythmias.
• BAV is leading to an improvement in LVEF among 37% of patients with baseline LVEF <40%.
• Reduction of BNP levels 24 hours after BAV predicts LVEF improvement at 30 days.
• One-year survival is significantly higher among patients with such an LVEF improvement following BAV.

Procedural safety. High complication and mortality rates after BAV were reported early in the experience,^5,6 documenting complications in 25% of patients, procedural death in up to 3%, and in-hospital death ranging from 4.6%-10%. Multiple modifications have been made to this technique over the years. Rapid ventricular pacing enables precise and stable balloon positioning,⁸ subsequently reducing the risk of calcium embolization and aortic root damage. Better large-sized balloons that allow smaller arterial sheaths, the use of closure devices, and assessment of peripheral vascular tree by computed tomography have contributed to the reduction of vascular complications. In our series, we experienced a 13% major complication rate, including only 1% of vascular complications, 1% of significant aortic regurgitation, and no procedural deaths, which is consistent with data more recently reported.^9,10,17 Furthermore, it has to be emphasized that low LVEF increases the risk of cardiovascular interventions, including those for AS treatment.^2-4 In our study, the safety profile of BAV was very good among patients with low LVEF and those with preserved LVEF, despite a higher rate of ventricular arrhythmias in the group with impaired left ventricular function. 

LVEF improvement. Among patients with depressed left ventricular function, surgical AVR^2,4 and TAVI^18,19 have been shown to be associated with left ventricular function recovery by relief of the valvular obstruction and thereby the pressure overload imposed on the left ventricle. Improvement in aortic valve hemodynamics is less well achieved by BAV. The residual gradient and only moderate increase in aortic valve area that are usually observed after BAV could be limiting factors for LVEF recovery. Nevertheless, flexibility of the native aortic leaflets can be improved by balloon dilatation; intraleaflet fractures within calcified nodular deposits, cleavage planes along collagenized stroma, or separation of fused leaflets have been proposed as mechanisms of BAV on the aortic valve.¹⁶ Because the left ventricle with depressed systolic function is highly sensitive to an increase in afterload, greater aortic valve leaflet flexibility achieved by balloon dilatation could have an impact on LVEF recovery after BAV. 

Dworakowski et al¹⁰ published a significant ejection fraction improvement from 27% to 30% in 43 patients undergoing BAV alone. Doguet et al¹² observed an increase in LVEF from 40% to 44% among 25 patients undergoing BAV as a bridge to AVR. Safian et al¹¹ measured an increase in LVEF from 37% to 44% in 13/28 patients, whereas 15 other patients failed to show improvement. In our series, among 59 patients with impaired left ventricular function at baseline, LVEF significantly improved from 28% to 36% at 30 days. There was no difference between patients with versus those without a significant LVEF improvement with respect to baseline characteristics or procedural results; in particular, change of gradient and valve areas obtained by BAV were similar, like in the series by Safian et al, suggesting that these parameters are not the main mechanism of ventricular function recovery after BAV.

BNP. In the present study, early reduction of BNP observed 24 hours after BAV was the strongest predictor of LVEF improvement at 30 days. BNP is mainly produced by ventricular cardiomyocytes in response to increased wall stress, afterload, and myocardial hypertrophy. Berger-Klein et al¹⁴ reported that BNP was related to LVEF and Neverdal et al²⁰ found a correlation between left ventricular mass index and BNP, suggesting the role of BNP to detect both systolic and diastolic left ventricular dysfunction in patients with AS. Reverse remodeling has been observed after aortic valve replacement and is associated with a reduction of serum BNP levels after surgery.²¹ In our series of patients at prohibitive risk and low EF, baseline BNP levels were at a very high level around 2000 pg/mL, but significantly reduced only among patients experiencing an improvement of left ventricular function after BAV. The optimal delay for considering a bridge therapy following BAV is not yet well defined and ranges in the literature between a few days and several months.^6,13 Early discrimination of patients who may benefit from BAV with a BNP blood dosage at 24 hours post procedure would be useful in the future decision-making process. 

Survival. Our study confirms a poor prognosis of patients with severe and symptomatic AS undergoing BAV only, with a mortality rate of 63% at 1 year; this observation is in accordance with rates previously reported.^9,22 The use of BAV as a bridge to surgery in hemodynamically unstable patients is the strongest guideline recommendation for this procedure.¹ Bridge therapy after BAV has been reported to be a reasonable approach in patients considered at too high a risk for immediate TAVI.¹³ Doguet et al¹² showed a decreased operative risk after BAV allowing a good outcome of patients with low EF treated by AVR 8-14 weeks later. In the present study, the outcome of patients who had BAV as a bridge to TAVI or AVR was much better when compared with patients who had BAV only, in accordance with data published by Ben-Dor et al⁹ and Ussia et al.¹³ Predictors of overall mortality in our study are renal failure and baseline BNP levels. Renal insufficiency was reported to be associated with worse outcomes after AVR, TAVI, and BAV.^9,23,24 Preoperative values of BNP have been shown to predict outcomes after AVR and TAVI;^14,15 the present study shows that a high level of BNP before BAV is also associated with poor prognosis.

In addition, our study shows that in cases of baseline LVEF <40%, the outcomes of patients experiencing an improvement in LVEF after BAV are much better in comparison with patients with no change in left ventricular function.

Conclusion

In a high-risk population of patients with AS and impaired left ventricular function, BAV improves LVEF among 37% of cases with a very good safety profile, not different from patients with preserved left ventricular function. LVEF improvement at 30 days after BAV, which can be detected by a reduction of BNP levels already 24 hours after the procedure, is a predictor of survival at 1 year. Long-term survival is limited after BAV, when not followed by AVR or TAVI. Our findings suggest that BNP could be useful for early discrimination of patients who have benefited from the procedure, in view of bridge therapy following BAV.  

 References

1. Bonow R, Carabello B, Chatterjee K, et al; on behalf of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. 2008 focused update incorporated into the ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2008;52(13):e1-e142.
2. Connolly H, Oh J, Schaff H, et al. Severe aortic stenosis with low transvalvular gradient and severe left ventricular dysfunction: result of aortic valve replacement in 52 patients. Circulation. 2000;101(16):1940-1946.
3. Monin JL, Quéré JP, Monchi M, et al. Low-gradient aortic stenosis: operative risk stratification and predictors for long-term outcome: a multicenter study using dobutamine stress hemodynamics. Circulation. 2003;108(3):319-324.
4. Pereira JJ, Lauer MS, Bashir M, et al. Survival after aortic valve replacement for severe aortic stenosis with low transvalvular gradients and severe left ventricular dysfunction. J Am Coll Cardiol. 2002;39(8):1356-1363.
5. Cribier A, Savin T, Saoudi N, Rocha P, Berland J, Letac B. Percutaneous transluminal valvuloplasty of acquired aortic stenosis in elderly patients: an alternative to valve replacement? Lancet. 1986;1(8472):63-67.
6. Percutaneous balloon aortic valvuloplasty. Acute and 30-day follow-up results in 674 patients from the NHBLI balloon valvuloplasty registry. Circulation. 1991;84(6):2383-2397.
7. Yamen E, Daniels D, Van H, Yeung A, Fearon W. Use of a low-profile, compliant balloon for percutaneous aortic valvuloplasty. Catheter Cardiovasc Interv. 2010;75(5):794-798.
8. Witzke C, Don C, Cubeddu R, et al. Impact of rapid ventricular pacing during percutaneous balloon aortic valvuloplasty in patients with critical aortic stenosis: should we be using it?  Catheter Cardiovasc Interv. 2010;75(3):444-452.
9. Ben-Dor I, Pichard A, Satler L, et al. Complications and outcome of balloon aortic valvuloplasty in high-risk or inoperable patients. J Am Coll Cardiol. 2010;3(11):1150-1156.
10. Dworakowski R, Bhan A, Brickham B, Monaghan M, MacCarthy P. Effectiveness of balloon aortic valvuloplasty is greater in patients with impaired left ventricular function.  Int J Cardiol. 2011;150(1):103-105.
11. Safian R, Warren S, Berman A, et al. Improvement in symptoms and left ventricular performance after balloon aortic valvuloplasty in patients with aortic stenosis and depressed left ventricular ejection fraction. Circulation. 1988;78(5 Pt 1):1181-1191.
12. Doguet F, Godin M, Lebreton G, et al. Aortic valve replacement after percutaneous valvuloplasty — an approach in otherwise inoperable patients. Eur J Cardiothorac Surg. 2010;38(4):394-399.
13. Ussia GP, Capodanno D, Barbanti M, et al. Balloon aortic valvuloplasty for severe aortic stenosis as a bridge to high-risk transcatheter aortic valve implantation. J Invasive Cardiol. 2010;22(4):161-166.
14. Bergler-Klein J, Klaar U, Heger M, et al. Natriuretic peptides predict symptom-free survival and postoperative outcome in severe aortic stenosis. Circulation. 2004;109(19):2302-2308.
15. Kefer J, Beauloye C, Astarci P, et al. Usefulness of B-type natriuretic peptide to predict outcome of patients treated by transcatheter aortic valve implantation. Am J Cardiol. 2010;106(12):1782-1786.
16. Hara H, Pedersen WR, Ladich E, et al. Percutaneous balloon aortic valvuloplasty revisited: time for a renaissance? Circulation. 2007;115(12):334-338.
17. Eltchaninoff H, Cribier A, Tron C, et al. Balloon aortic valvuloplasty in elderly patients at high risk for surgery, or inoperable. Eur Heart J. 1995;16(8):1079-1084.
18. Cribier A, Eltchaninoff H, Tron C, et al. Treatment of calcific aortic stenosis with the percutaneous heart valve: mid-term follow-up from the initial feasibility studies: the French experience. J Am Coll Cardiol. 2006;47(6):1214-1223.
19. Clavel MA, Webb JG, Rodés-Cabau J, et al. Comparison between transcatheter and surgical prosthetic valve implantation in patients with severe aortic stenosis and reduced left ventricular ejection fraction. Circulation. 2010;122(19):1928-1936.
20. Neverdal N, Knudsen C, Husebye T, et al. The effect of aortic valve replacement on B-type natriuretic peptide in patients with severe aortic stenosis — one year follow-up. Eur J Heart Fail. 2006;8(3):257-262.
21. Poulsen S, Sogaard P, Nielsen-Kudsk J, Egeblad H. Recovery of left ventricular systolic longitudinal strain after valve replacement in aortic stenosis and relation to natriuretic peptides. J Am Soc Echocardiogr. 2007;20(7):877-884.
22. Leon MB, Smith CR, Mack M, et al; PARTNER Trial Investigators. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010;363(17):1597-1607.
23. Grossi EA, Schwartz CF, Yu PJ, et al. High-risk aortic valve replacement: are the outcomes as bad as predicted? Ann Thorac Surg. 2008;85(1):102-106.
24. Bagur R, Webb JG, Nietlispach F, et al. Acute kidney injury following transcatheter aortic valve implantation: predictive factors, prognostic value, and comparison with surgical aortic valve replacement. Eur Heart J. 2010;31(7):865-874.

___________________________________________

From the Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Division of Cardiology, Brussels, Belgium.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.

Manuscript submitted October 17, 2012, provisional acceptance given November 13, 2012, final version accepted January 11, 2013.

Address for correspondence: Joelle Kefer, MD, PhD, Cliniques Universitaires Saint-Luc, Université Catholique de Louvain, Division of Cardiology, Avenue Hippocrate, 10, 1200 Brussels, Belgium. Email: joelle.kefer@uclouvain.be