wenz iD - Proefschrift Mariam Samim

TRANSCATHETER AORTIC VALVE REPLACEMENT

Optimisation of the technique, assessment of the complications and future perspectives

Mariam Samim

Š Mariam Samim, 2014 Financial support by the University Medical Center Utrecht, Edwards Lifesciences BV, Medtronic Inc. and 3mensio Medical Imaging BV for the publications in this thesis is gratefully acknowledged.

ISBN:

978-94-6295-049-8

Vormgeving: Wendy Schoneveld, www.wenz iD.nl Printed by:

Proefschriftmaken.nl || Uitgeverij BOXPress

Published by: Uitgeverij BOXPress, ’s-Hertogenbosch

TRANSCATHETER AORTIC VALVE REPLACEMENT optimisation of the technique, assessment of the complications and future perspectives

TRANSCATHETER AORTAKLEP IMPLANTATIE optimalisatie van de techniek, bestuderen van de complicaties en toekomst uitdagingen (met een samenvatting in het Nederlands)

PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van rector magnificus, prof. dr. van der Zwaan, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op donderdag 18 december 2014 des middags te 02:30 uur

door Mariam Samim geboren op 22 november 1985 te Herat, Afghanistan

Promotor:

Prof.dr. P.A.F.M. Doevendans

Copromotoren: Dr. P.R. Stella

Dr. P. Agostoni

Financial support for the publication of this manuscript was generously provided by: NWO (Netherlands Organisation for Scientific Research) grant

To my parents and my beloved grandmother

CONTENTS CHAPTER 1

Introduction

9

PART ONE ACCURATE POSITIONING OF TRANSCATHETER AORTIC VALVE CHAPTER 2

Automated 3D analysis of multislice computed tomography to define

15

the line of perpendicularity of the aortic annulus and of the implanted valve: benefit on planning transcatheter aortic valve replacement.

Catheter Cardiovasc Interv. 2014 Jan 1;83(1):E119-27 CHAPTER 3

Automated 3D Analysis of Pre-Procedural MSCT to Predict Annulus Plane

31

Angulation and C-Arm Positioning: Benefit on Procedural Outcome in Patients Referred for TAVR. JACC Cardiovasc Imaging. 2013 Feb;6(2):238-48 CHAPTER 4

Three-dimensional aortic root reconstruction derived from rotational angiography

for

transcatheter

balloon-expandable

aortic

51

valve

implantation guidance.

Int J Cardiol. 2014 Oct 20;176(3):1318-20

PART TWO VALVULAR REGURGITATION AFTER TAVR AND ASSOCIATED FACTORS CHAPTER 5

A prospective "oversizing" strategy of the Edwards SAPIEN bioprosthesis:

69

Results and impact on aortic regurgitation.

J Thorac Cardiovasc Surg. 2013 Feb;145(2):398-405 CHAPTER 6

Impact of aortic valve calcification quantified by computed tomography

87

on procedural and clinical outcomes of transcatheter aortic valve replacement In preparation CHAPTER 7

Rationale and design of the Edwards SAPIEN 3 Periprosthetic Leakage Evaluation versus medtronic Corevalve in Tranfemoral aortic valve implantation (ELECT) trial: a randomized comparison of balloonexpandable versus self-expanding aortic valve prostheses.

In preparation

107

CHAPTER 8

Transcatheter aortic implantation of the Edwards-SAPIEN bioprosthesis:

121

insights on early benefit of TAVR on mitral regurgitation Int J Cardiol. 2011 Oct 6;152(1):124-6

PART THREE ISCHEMIC BRAIN INJURY DURING TAVR CHAPTER 9

Silent ischemic brain lesions after transcatheter aortic valve replacement:

129

lesion distribution and predictors.

Accepted for publication in Clinical Research in Cardiology CHAPTER 10 First-in-man experience with a new embolic deflection device in

145

transcatheter aortic valve interventions.

EuroIntervention. 2012 May 15;8(1):51-6 CHAPTER 11 Use of Embrella embolic protection device in prevention of cerebral

159

ischemic complications during TAVI. J Thorac Cardiovasc Surg. 2014 CHAPTER 12 Use of TriGuard embolic protection device for cerebral protection

177

during transcatheter aortic valve replacement.

In preparation

PART FOUR DISCUSSION CHAPTER 13 General discussion and conclusions

193

APPENIDIX Dutch summary

206

Acknowledgments

215

Curriculum vitae auctoris

219

List of publications

220

CHAPTER

1 INTRODUCTION

CHAPTER 1

INTRODUCTION Aortic valve stenosis is a highly prevalent valvular heart disease in economically developed nations and its prevalence in the community increases with age. Calcific “degeneration� of a bileaflet or trileaflet valve is the most common etiology in western countries. Calcific aortic stenosis is thought to share many features with coronary artery disease, such as lipid accumulation, inflammation, and calcification1. Moreover, an association is found between traditional atherosclerotic risk factors and the development of calcific aortic valve disease2.

For instance, hypercholesterolemia can induce both cellular proliferation and osteoblast phenotype expression that may lead to progression of aortic valve disease3.

Aortic stenosis is an insidious disease with a long latency period, followed by a rapid progression after the appearance of symptoms4-6. When symptoms develop, mortality rises sharply with an average survival of only two to three years among untreated patients. The association of aortic stenosis with clinical features similar to atherosclerosis has led to the hypothesis that aggressive modification of risk factors, as for coronary heart disease, may slow or prevent disease progression in the valve leaflets. So far, no medical treatment has been proved to prevent or delay the disease process. Recent reports using statins have shown mixed results and further evaluation of preventive measures is needed to clarify this issue7.

As for treatment of severe symptomatic aortic valve stenosis, unfortunately no effective drug regimens exists to improve survival and only a few drugs are available to alleviate symptoms.

Patients with evidence of pulmonary congestion can benefit from cautious treatment with digitalis or diuretics, or both1. However, aortic stenosis is obviously a mechanical obstruction and it requires mechanical correction. Therefore the traditional treatment of this condition is surgical aortic valve replacement (SAVR) which seems to improve symptoms, quality of life and survival8, 9. However, a relatively large number of patients (30%-40%) is deemed inoperable or at high surgical risk due to the presence of multiple coexisting conditions9-12.

Percutaneous balloon aortic valvuloplasty was developed in 1985 by Cribier as a less invasive treatment for aortic valve stenosis in patients unsuitable for SAVR. However aortic balloon valvuloplasty does not alter the natural history of aortic stenosis and is considered as a temporary treatment with a high rate of restenosis due to elastic recoil of valve tissue. Hence the development of another treatment option with more permanent results was necessary.

Seventeen years later (in 2002), Cribier demonstrated for the first time the use of transcatheter aortic valve replacement (TAVR), which thereby opened new horizons in the minimally invasive management of valvular heart disease.

The application of TAVR in patients with severe symptomatic aortic stenosis is gaining widespread acceptance with a burgeoning supportive evidence base13-15. Although SAVR remains the standard treatment for aortic stenosis in low or intermediate risk patients, the first large randomised controlled trial (PARTNER 1A) comparing TAVR to SAVR in high risk patients showed similar long-term results with respect to mortality and reduction of symptoms14, 15. In fact, the second and the most recent randomised controlled trial (U.S. CoreValve High Risk Study) in similar patient population showed even a significantly higher rate of survival at 1 year in favour of the TAVR group13. Benefits of TAVR include a shorter procedure time, less post-

10

INTRODUCTION

procedural pain, more rapid mobilization and recovery and a faster reduction of symptoms when compared to SAVR13, 15.

The benefits associated with the application of TAVR, however, are mitigated by the occurrence of more or less disabling adverse events with related increased mortality and early-reduced quality of life. Complications associated with TAVR include bleeding- and vascular complications, kidney dysfunction, thromboembolic complications such as stroke, post-procedural aortic regurgitation and death13, 15. Improvement in TAVR technology, optimal procedural planning using different imaging techniques and insights into the aetiology of complications are important to improve the results of TAVR, especially in the light of shifting the indication for this treatment to younger and lower risk patients. The present thesis aimed to study possible improvements in the patient workup and procedural planning before TAVR and furthermore to provide more insight in potentially devastating complications related to this procedure.

11

CHAPTER 1

REFERENCES 1. Ramaraj R, Sorrell VL. Degenerative aortic stenosis. BMJ. 2008 Mar 8;336(7643):550-5.

2. Glader CA, Birgander LS, Soderberg S, Ildgruben HP, Saikku P, Waldenstrom A, et al. Lipoprotein(a), chlamydia pneumoniae, leptin and tissue plasminogen activator as risk markers for valvular aortic stenosis. Eur Heart J. 2003 Jan;24(2):198-20.

3. Chan KL. Is aortic stenosis a preventable disease? J Am Coll Cardiol. 2003 Aug 20;42(4):593-9.

4. Davies SW, Gershlick AH, Balcon R. Progression of valvar aortic stenosis: A long-term retrospective study. Eur Heart J. 1991 Jan;12(1):10-4.

5. Peter M, Hoffmann A, Parker C, Luscher T, Burckhardt D. Progression of aortic stenosis. role of age and concomitant coronary artery disease. Chest. 1993 Jun;103(6):1715-9.

6. Bonow RO, Carabello BA, Chatterjee K, de Leon AC,Jr, Faxon DP, Freed MD, et al. 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 (writing committee to revise the 1998 guidelines for the management of patients with valvular heart disease). endorsed by the society of cardiovascular anesthesiologists, society for cardiovascular angiography and interventions, and society of thoracic surgeons. J Am Coll Cardiol.

2008 Sep 23;52(13):e1-142.

7. De Vecchis R, Di Biase G, Esposito C, Ciccarelli A, Cioppa C, Giasi A, et al. Statin use for nonrheumatic calcific aortic valve stenosis: A review with meta-analysis. J Cardiovasc Med (Hagerstown). 2013 Aug;14(8):559-67.

8. Lund O. Preoperative risk evaluation and stratification of long-term survival after valve replacement for aortic stenosis. reasons for earlier operative intervention. Circulation. 1990 Jul;82(1):124-39.

9. Kvidal P, Bergstrom R, Horte LG, Stahle E. Observed and relative survival after aortic valve replacement.

J Am Coll Cardiol. 2000 Mar 1;35(3):747-56.

10. Iung B, Cachier A, Baron G, Messika-Zeitoun D, Delahaye F, Tornos P, et al. Decision-making in elderly patients with severe aortic stenosis: Why are so many denied surgery? Eur Heart J. 2005 Dec;26(24):2714-20.

11. O’Brien SM, Shahian DM, Filardo G, Ferraris VA, Haan CK, Rich JB, et al. The society of thoracic surgeons 2008 cardiac surgery risk models: Part 2--isolated valve surgery. Ann Thorac Surg. 2009 Jul;88(1 Suppl):S23-42.

12. Varadarajan P, Kapoor N, Bansal RC, Pai RG. Clinical profile and natural history of 453 nonsurgically managed patients with severe aortic stenosis. Ann Thorac Surg. 2006 Dec;82(6):2111-5.

13. Adams DH, Popma JJ, Reardon MJ, Yakubov SJ, Coselli JS, Deeb GM, et al. Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med. 2014 May 8;370(19):1790-8.

14. Kodali SK, Williams MR, Smith CR, Svensson LG, Webb JG, Makkar RR, et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med. 2012 May 3;366(18):1686-95.

15. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011 Jun 9;364(23):2187-98.

 

12

INTRODUCTION

OUTLINE OF THIS THESIS The outline of this thesis is as follows. In part I, the imaging modalities for accurate positioning and deployment of aortic valve prostheses are presented and discussed. The use of conventional repeated angiography for finding the optimal implantation view is timeconsuming and increases radiation exposure and use of contrast medium during transcatheter aortic valve replacement (TAVR). Chapter 2 and 3 discuss the validity, feasibility and clinical benefit of multislice computed tomography for pre-procedural prediction of the optimal implantation views. Subsequently, Chapter 4 describes the use of rotational angiography for intra-procedural definition of C-arm position, and thereby an optimal implantation view.

Aortic regurgitation (AR) is one of the most frequent complications after TAVR and it is suggested to affect the prognosis of patients undergoing this procedure. Part II discusses the occurrence of AR in our series and also factors that might be associated with it. Chapter 5 aimed to investigate the impact of prosthesis oversizing on the rate and severity of aortic regurgitation after TAVR. Severe calcification of the aortic valve apparatus is a common finding among patients scheduled for TAVR. Chapter 6 studies the association between the severity of aortic valve calcification and the outcome of patients undergoing TAVR. Next, Chapter 7 (study protocol) describes the aims and the design of the randomised controlled trial “Edwards SAPIEN Periprosthetic Leakage Evaluation versus medtronic Corevalve in Tranfemoral aortic valve implantation” (the ELECT trial), which is initiated and started in Utrecht. In short, the latter trial is designed to investigate the difference in the severity of post-TAVR aortic regurgitation between patients undergoing the implantation of the Edwards SAPIEN 3™ bioprosthesis versus patients receiving the Medtronic CoreValve® system.

Chapter 8 concerns the potential benefit of TAVR with the Edwards SAPIEN prosthesis on preexisting mitral regurgitation.

As TAVR procedures are associated with a high intra-procedural embolic load, significant concerns exist with regard to the incidence of thromboembolic cerebral complications after TAVR. Part III of this thesis discusses the frequency of both apparent and asymptomatic cerebral ischemic infarctions associated with TAVR and the benefit of cerebral embolic protection devices for intra-procedural use. Therefore, Chapter 9 describes the rate, distribution, and possible predictors of cerebral ischemic lesions on brain imaging after TAVR. Subsequently Chapter 10, Chapter 11 and Chapter 12 describe the feasibility and benefit of three different embolic deflector devices, the SMT embolic deflection device, the Embrella embolic deflector and the TriGuard™HDH embolic deflection device.

13

PART ONE

ACCURATE POSITIONING OF TRANSCATHET ER AORTIC VALVE

CHAPTER

2 AUTOMATED 3D ANALYSIS OF MULTISLICE COMPUTED TOMOGRAPHY TO DEFINE THE LINE OF PERPENDICULARITY OF THE AORTIC ANNULUS AND OF THE IMPLANTED VALVE: BENEFIT ON PLANNING TRANSCATHETER AORTIC VALVE REPLACEMENT Catheter Cardiovasc Interv. 2014 Jan 1;83(1):E119-27 Samim M1 Juthier F4 Van Belle C4 Agostoni P1 Kluin J2 Stella PR1 Ramjankhan F2 Budde RP3 Sieswerda G1 Algeri E4 Elkalioubie A4 Belkacemi A1 Bertrand ME4 Doevendans PA1 Van Belle E4 Department of Cardiology, 2Department of Cardiothoracic surgery,

1

Department of Radiology, University Medical Center, Utrecht, The Netherlands

3

Department of Cardiology, and EA 2693, University Hospital, Lille, France

4

CHAPTER 2

ABSTRACT Aims We aimed to determine whether preprocedural analysis of multislice computed tomography (MSCT) scan could accurately predict the “line of perpendicularity” (LP) of the aortic annulus and corresponding C-arm angulations required for prosthesis delivery.

Methods and results A three-dimensional (3D) analysis of preprocedural MSCT dedicated to define the LP of the aortic annulus was performed in 60 consecutive patients referred for transcatheter aortic valve replacement (TAVR). In 24 patients, the analysis was performed retrospectively to evaluate reproducibility. In 11 patients of this cohort, additional fluoroscopy and MSCT were performed post-procedurally to compare the LP of the aortic annulus and the LP of the implanted bioprosthesis. In 36 patients, the analysis was performed prospectively and the results were available at the time of the procedure. In those 36 patients, the postprocedure fluoroscopydefined LP of the implanted bioprosthesis was used to validate the LP of the aortic annulus as predicted by MSCT. Intraobserver and interobserver reproducibility of the 3D analysis of MSCT to define the LP of the aortic annulus (κ = 1 and 0.94, respectively) and of the bioprosthesis (κ= 1 and 1, respectively) were excellent. Comparison between the LP of the aortic annulus and the LP of the bioprosthesis showed that the two LPs were virtually identical, demonstrating self-centering of the device during implantation. In the prospective cohort, the ability of MSCT analysis to predict the LP of the aortic annulus was very good (accuracy = 94% and κ = 0.89).

Conclusion Automated 3D analysis of preimplantation MSCT accurately predicts the LP of the aortic annulus and the corresponding C-arm position required for TAVR.

16

MSCT FOR DEFINING THE OPTIMAL IMPLANTATION VIEW DURING TAVR

INTRODUCTION Transcatheter aortic valve replacement (TAVR) is a minimally invasive procedure performed in patients with severe aortic stenosis and who are at high surgical risk.1, 2 Optimal positioning of the prosthetic valve during TAVR is crucial to prevent procedural complications.3 During deployment of the valve prosthesis, an optimal view of the aortic annulus, where the valve will be implanted, has to be determined.1,2 The aim of the optimal view is to position the X-ray tube C-arm perpendicular to the aortic annulus plane to achieve an appropriate delivery of the valve during TAVR. Previous studies have shown that the aortic valve is typically directed in a cranial and anterior fashion with angulation to the right, and that corresponding C-arm positions describe a line in a two-dimensional space. This line is known as the “line of perpendicularity” (LP) of the aortic annulus.4 MSCT provides detailed anatomic assessment of the aortic root and has also a significant potential to assess the aortic root in relation to the body axis.4 Although the use of MSCT analysis has been advocated to predict the LP of the aortic annulus and the angulations of the C-arm for prosthesis positioning during TAVR, its accuracy remains unclear.4,5 This study was designed to evaluate whether automated three-dimensional (3D) analysis of preprocedural MSCT scan could be used to reproducibly and accurately define the LP of the aortic annulus needed for prosthesis positioning and deployment during TAVR. It was further designed to analyze the relationship between the LP of the aortic annulus and the LP of the implanted bioprosthesis.

METHODS Study Population All consecutive patients with severe aortic stenosis considered to be at high or prohibitive surgical risk (as evaluated by a multidisciplinary heart team) and referred to our center for TAVR with a balloon-expendable bioprosthesis between January 2009 and June 2011 were considered for inclusion in this study (Figure 1).

Preprocedural MSCT was performed in all patients within 3 months before the TAVR procedure.

The automated MSCT 3D software analysis (3mensio ValvesTM, 3mensio Medical Imaging BV, The Netherlands, http://www.3mensio.com) was available in our institution starting from December 2009. Analysis of the preprocedural MSCT was performed for all patients.

Twenty-four procedures had already been performed at the time of availability of the MSCT software. In these 24 patients, the analysis with the MSCT 3D software was performed retrospectively (Retrospective Cohort). In addition, for the last 11 of those 24 patients, an additional MSCT scan was performed before hospital discharge and was analyzed with the same software.

In 36 patients treated between December 2009 and June 2011, the MSCT 3D analysis of the LP was performed before the procedure and the results were provided to the operators before the beginning of the procedure. This group constituted the “Prospective Cohort.”

17

CHAPTER 2

Figure 1 | Flowchart of the study. Automated MSCT 3D software (3-mensio) was available since December 2009. All consecutive patients treated between January 2009 and December 2009 constituted the “Retrospective Cohort” (n = 24). In addition, in the last 11 patients of this cohort, postprocedural fluoroscopy and MSCT recording were performed. From January 2010 to June 2011 (n = 36), the MSCT analysis was performed preprocedural and available to the operators. Postprocedural fluoroscopy recordings were performed in the 36 patients.

Multislice Computed Tomography As part of the work-up procedure, the patients underwent MSCT within 3 months before TAVR. In the last 11 patients of the “Retrospective cohort” an additional MSCT scan was performed after implantation of the Edwards-SAPIEN prosthesis before discharge.

The detailed MSCT procedure is described in Supporting Information material.

Automated 3D Image Analysis of MSCT All MSCT images were analyzed by a procedure-independent operator, who was trained to use the dedicated 3D aortic valve analysis 3-mensio software (3mensio ValvesTM, 3mensio Medical Imaging BV, The Netherlands, http://www.3mensio.com). For the 24 patients in the “Retrospective Cohort,” analyses were performed twice on two separate occasions by a dedicated operator. A second operator completed a third series of analysis.

For each patient, the image quality of MSCT was rated by the operators in three groups: excellent, adequate, or inadequate, depending on potential degradations owing to calcification, motion, or contrast issues. The severity of aortic root calcification was classified into four groups: as minimal (<25% of total circumference), mild (25–50%), moderate (50–75%), and severe (75–100%).6 The process to determine the aortic annulus plane and the predicted “LP” of the aortic annulus is described in Supporting Information material and in shown in Figure 2.

18

MSCT FOR DEFINING THE OPTIMAL IMPLANTATION VIEW DURING TAVR

For each patient, the software provides an infinite series of C-arm positions (defined by their combination of Cranio–Caudal and RAO-LAO angulation), where the C-arm is perpendicular to the annulus plane. This series is named the predicted LP of the aortic annulus. For each patient, the predicted “LP” of the annulus can be drawn on a cranio–caudal axis/right–left axis (Figure 3).

In 11 patients, an additional MSCT and a similar 3D analysis were performed after implantation of the prosthesis. In those patients, the plane of the upper edge and the plane of the lower edge of the prosthesis were defined. The LPs of the two additional planes were derived as described previously. For each of the 11 patients, these two LPs were compared to each other, compared to the LP of the bioprosthesis as derived from postprocedural fluoroscopy recordings (see below) and compared to the predicted “LP” of the annulus plane as obtained from the preprocedural MSCT. TAVR Procedure The prosthetic stented Edwards-SAPIEN (until August 2010) or SAPIEN XT (starting from August 2010), valves were mechanically crimped on a balloon catheter immediately before the implantation procedure. The Retroflex or Novaflex delivery systems were used for device implantation via the femoral artery. Until August 2010, the femoral artery was accessed via a surgical cut-down; starting from August 2010, it was performed through a percutaneous puncture. The Ascendra and Ascendra 2 delivery systems were used in case of apical access.

TF and TA-AVR were performed under general anesthesia in the catheterization laboratory and are described in Supporting Information material.7,8 The “Retrospective Cohort” and the “Prospective Cohort”: Specificities In patients in the “Retrospective Cohort” (n = 24), owing to the lack of information on the LP of the aortic annulus by MSCT, the balloon-mounted valve was positioned using conventional angiographic techniques, aiming to locate the unexpanded valve prosthesis mid-way at the level of the virtual line connecting the insertion site of the three cusps to the aortic wall. The search for the LP of the annulus and of the proper position of the C-arm was empirically performed using test angiographies. At the end of the procedure of the last 11 patients of this series, four fluoroscopy recordings of the implanted endoprosthesis were performed at RAO 20°, RAO 10°, LAO 10°, and LAO 20° with adjustment of the C-arm to achieve a perpendicular view of the prosthesis (Figures 1 and 4). These recordings were used, for each patient, to derive the fluoroscopy-defined LP of the prosthesis. For each patient, this fluoroscopy-derived LP of the prosthesis was compared with the LP of the prosthesis derived from the 3D analysis of the MSCT. In patients in the “Prospective Cohort” (n = 36), the position of the C-arm during the prosthesis implantation procedure was chosen among the positions (predicted “LP” of the annulus) proposed by pre-procedural MSCT analysis. At the end of the procedure, four fluoroscopy recordings of the implanted bioprosthesis were performed at RAO 20°, RAO 10°, LAO 10°, and LAO 20° with adjustment of the C-arm to achieve a perpendicular view of the prosthesis (Figures 1 and 4). These recordings were used, for each patient, to derive the fluoroscopy-

19

CHAPTER 2

defined LP of the prosthesis. For each patient, this fluoroscopy-derived LP of the prosthesis was compared with the predicted LP of the annulus derived from the MSCT derived 3D analysis.

Statistical analysis Continuous variables were expressed as means ± standard deviation and categorical variables were reported as frequencies. For comparison between categorical variables, a chi-square test was used and an independent-sample t-test was used to compare continuous values.

Reproducibility of MSCT 3D analysis and agreement between MSCT and X-ray analyses were

Figure 2 | Defining the “LP” of the aortic annulus using automated MSCT 3D software analysis. The valve region is visualized in three double oblique multiplanar reconstruction viewports (A, B) and a volume rendered overview. Two double oblique views show perpendicular cross-sections through the aortic valve (A, B). The other shows the annulus plane. The position of the annulus plane can be changed by dragging the red–yellow icon. The orientation of the plane can be refined by dragging the green and purple handles in the oblique views. Simulated angiogram (C) with corresponding C-Arm angles displayed in the two bottom corners. A virtual C-Arm can be rotated along the aortic centerline by dragging the eye-icon in the left view. The simulated projection is updated interactively. The aortic centerline and the aortic annulus plane are drawn over the simulated angiogram.

20

MSCT FOR DEFINING THE OPTIMAL IMPLANTATION VIEW DURING TAVR

investigated using kappa statistics.9 For these analyses, a 40° range of RAO/LAO angulations, between RAO 20° and LAO 20°, was considered. Cranio–caudal angulation was provided for any 1° increment of RAO/LAO angulation. For any of these 40 positions, if the cranio–caudal angulation was more than 5° outside the prediction, then the whole analysis was considered discordant. Analyses were performed using IBM SPSS Statistics 19.0.

RESULTS Baseline Characteristics of the Study Population A total number of 60 patients were included in this study, 24 patients in the “Retrospective Cohort” and 36 patients in the “Prospective cohort.” The baseline clinical and echocardiography characteristics of the study population as well as the procedural characteristics are summarized in the Table 1. A little more than 50% of the population was female, the mean age was 81 ± 8 years and the mean logistic Euroscore was 18 ± 11. No difference was observed between the “Retrospctive cohort” and the “Prospective Cohort”.

Reproducibility of 3D MSCT Analysis as Performed in the “Retrospective Cohort” The quality of the MSCT images was considered excellent in 85% (51/60) and adequate in 15% (9/60) of cases. In no case, the quality of the MSCT was considered inadequate. Calcification of the aortic root was classified as minimal (5% [n = 3]), mild (18% [n = 11]), moderate (47% [n = 28]), or severe (30% [n = 18]).

Within a margin of 5°, intraobserver (24/24, κ = 1) and interobserver (23/24, κ = 0.94) reproducibility of the 3D, automated MSCT prediction of the “LP” of the annulus were excellent.

In the last 11 of the 24 patients of the Retrospective Cohort, an additional 3D automated MSCT analysis was performed after the procedure to analyze the LP of the bioprosthesis.

As expected, it demonstrated that the LP of the upper edge of the prosthesis was similar to the LP of the lower edge of the prosthesis. Comparison with the fluoroscopy-derived LP of the bioprosthesis demonstrated that the two LPs were within 5° of each other (11/11, κ =1).

Similar comparison with the MSCT-derived LP of the aortic annulus demonstrated that the two LPs were also within 5° of each other (11/11, κ =1). An example of this phenomenon is shown in Figure 3. These findings demonstrate that the planes of the upper edge and of the lower edge of the bioprothesis are parallel to each other and that these two planes are parallel to the plane of the aortic annulus. They also demonstrate that for each patient, the LP of the upper (or lower) edge of the prosthesis can be used as a surrogate of the LP of the annulus.

We will take advantage of this finding to validate the prediction of the 3D analysis of the MSCT in the “Prospective cohort”.

Interindividual Variation of the LP of the Annulus Interindividual variation of the LP of the annulus among the 24 patients in the “Retrospective Cohort” shows the heterogeneity of the orientation of the “LP” of the annulus from patient to patient with a range of 70° in the cranio/caudal axis and 80° in the right/left axis.

21

CHAPTER 2

This demonstrates that for each patient, the proper position of the C-arm can be anywhere in the range of 70° in the cranio/caudal axis for every fixed right/left position, and anywhere in a range of 80° in the right/left axis for every fixed cranio/caudal position.

Table 1 | Clinical, echocardiographic and procedural characteristics

Age (year) Male, n(%)

Retrospective cohort (n = 24)

Prospective cohort (n = 36)

P-value

79 ± 7

82 ± 5

0.08

Total cohort (n = 60) 81 ± 8

12(50)

15(42)

0.60

27(45)

25.6 ± 6.9

26.7 ± 5.5

0.32

26.2 ± 6.0

Hypertension, n(%)

17(71)

27(75)

0.77

44(73)

Dyslipidemia, n(%)

15(62)

16(44)

0.20

31(52)

Diabetes, n(%)

5(21)

12(33)

0.38

17(28)

BMI, mean ± SD

Smoking, n(%)

3(13)

5(14)

0.99

8(13)

CAD, n(%)

16(67)

23(64)

0.82

39(65)

Prior MI, n(%)

8(33)

8(22)

0.38

16(27)

Prior PCI, n(%)

8(33)

11(31)

0.82

19(32)

Prior CABG, n(%)

6(25)

4(11)

0.18

10(17)

CVD, n(%)

9(37)

7(19)

0.14

16(26)

PVD, n(%)

6(25)

7(19)

0.75

13(22)

Renal disease, n(%)

5(21)

14(39)

0.17

19(32)

107 (55)

91(68)

0.35

99 (61)

COPD, n(%)

7(29)

12(33)

0.78

19(32)

Prior malignant disease, n(%)

6(25)

9(25)

0.93

15(25)

Logistic Euroscore

19.7 ± 12.8

18.5 ± 10.8

0.45

19.5 ± 11.8

Peak aortic gradient (mm Hg)

73.0 ± 23.4

66.3 ± 23.6

0.18

70.1 ± 24.5

Aortic valve area (cm )

0.70 ± 0.16

0.71 ± 0.19

0.65

0.71 ± 0.15

LVEF (%)

54.9 ± 12.4

52.4 ± 10.9

0.32

53.9 ± 12.6

23 mm, n(%)

7 (29)

11 (30)

26 mm, n(%)

17 (71)

24 (67)

29 mm, n(%)

0 (0)

1 (3)

Transfemoral, n(%)

11 (46)

26 (72)

Transapical, n(%)

13 (54)

10 (28)

Creatinine (μmol/L), median (IQ)

2

Diameter of implanted valve prosthesis 18 (30) 0.70

41 (68) 1 (2)

TAVR approach 0.07

37 (62) 23 (38)

Data are mean ± SD otherwise stated. BMI, body mass index; MI, myocardial infarction; PCI,= percutaneous coronary intervention; CABG, coronary artery bypass grafting; CVD, cerebrovascular disease; PVD, peripheral vascular disease.

22

MSCT FOR DEFINING THE OPTIMAL IMPLANTATION VIEW DURING TAVR

Validation of the Prediction of the LP in the “prospective Cohort” In this group of patients, the predicted LP defined as the automated 3D analysis of MSCT was used to choose an appropriate angulation of the C-arm to place the bioprosthesis. At the end of the procedure, the actual LP of the upper (and lower) edge of the stent was drawn using fluoroscopy recordings (Figure 4). Validation of the prediction was performed by comparing the predicted LP of the annulus by the automated 3D analysis of MSCT with the actual LP of the upper edge of the stent, as recorded by fluoroscopy (see above). An accurate prediction (within 5°) was observed in 34/36 patients (κ = 0.89) and was not significantly modified by the degree of calcification (P = 0.71). An accurate prediction was observed in 8/8 (κ =1), in 16/17 (κ =0.88), and in 10/11 (κ =0.82) of patients with minimal/mild, moderate, or severe calcifications, respectively. Results in representative patients are shown in Figure 4.

Figure 3 | Comparison between the “LP” of the aortic annulus and the LP of the Edwards-SAPIEN endoprosthesis. Top right: Schemes describing the two ways the endoprosthesis may potentially sit in the aortic annulus after implantation: In the first one (top), the bioprosthesis does not line up to the aortic annulus and the edges of the stent are not parallel to the aortic annulus. In the second one (bottom), the bioprosthesis is lining up to the aortic annulus and the edges of the stent are parallel to the aortic annulus.

Top left: Preprocedural MSCT to define the plane of the aortic annulus (A), postprocedural MSCT to define le the plane of the upper edge (B) and of the lower edge of the endoprosthesis (C). The three derived sets of LPs to each of these three planes as calculated using the 3D software. Bottom: As seen in two different patients, the LPs of the aortic annulus (A), the upper edge (B), and the lower edge of the endoprosthesis (C) are superimposed. This was the case for all patients in whom postprocedural MSCT was performed (n = 11). This demonstrates that the bioprosthesis always line up to the aortic annulus. It further demonstrates that the “LP” of the endoprosthesis is a good surrogate for the “LP” of the aortic annulus.

23

CHAPTER 2

Figure 4 | Validation of MSCT prediction of the LP of the aortic annulus by postimplant fluoroscopy recordings. In a representative patient: Fluoroscopy recordings of the implanted endoprosthesis perpendicular to the stent performed at RAO 20° (A), RAO 10° (B), LAO 10° (C), and LAO 20° (D); Fluoroscopy recording of the implantation of the endoprosthesis (E). Graph illustrating the LP of the C-arm predicted by 3D analysis of the MSCT (black line), the chosen position for implantation (point E) based on the MSCT analysis and the LP of the C-arm derived from the fluoroscopy recording as performed at the end of the procedure (A, B, C, D). As it can be seen, the two lines are nearly superimposed (within 5° of each other).

DISCUSSION Although preprocedural MSCT analysis has been advocated for prediction of the LP of the aortic annulus and the C-arm angulations for prosthesis positioning and deployment during TAVR with balloon-expendable bioprostheses, additional information on the accuracy of this approach are needed to extend its use.4,5 The present study demonstrates that assessment of the LP of aortic annulus on preprocedural MSCT scan and of the LP of the implanted bioprosthesis on the postprocedural MSCT scan is very reproducible. It further demonstrates that the LP of the aortic annulus and the LP of the implanted bioprosthesis are virtually identical. Finally, the prospective part of the study demonstrates that the prediction of the LP of the aortic annulus by the preprocedural MSCT is excellent. Overall, it reinforces the concept that the use of this approach has the potential to improve the planning of TAVR procedures.

24

MSCT FOR DEFINING THE OPTIMAL IMPLANTATION VIEW DURING TAVR

Reproducibility of Automated 3D Analysis to Predict the LP of Aortic Annulus and the LP of the Implanted Bioprosthesis Intraobserver (κ = 1) and interobserver (κ = 0.94) reproducibility of defining the LP of the aortic annulus with the 3Mensio software used in this study was excellent. This information is of importance as, in his preliminary report, Gurvitch et al.4 did not provide any information on the reproducibility of the methods used. It must be noted that the high reproducibility observed in this study was achieved in a population with an excellent or adequate MSCT image quality.

One of the key aspects of this study was also the evaluation of the LP of the implanted bioprosthesis using postprocedural MSCT. As for the analysis of the LP of the aortic annulus, the reproducibility of such analysis was excellent. In addition, the LP of the implanted bioprosthesis as defined by MSCT analysis was similar to the LP found by fluoroscopy.

Furthermore, the comparison between the LP of the aortic annulus and the LP of bioprosthesis as defined by MSCT analysis demonstrated that the two LPs were virtually identical. This information which was never reported earlier is important as it demonstrates that the edges of implanted bioprosthesis lay parallel to the aortic annulus (Figure 3) and suggests that balloon deployment of the bioprosthesis is associated with self-centering of the device which lines up to aortic annulus.

Finally, the demonstration that the LP of the annulus by MSCT, the LP of the edges of the SAPIEN bioprosthesis after implantation by MSCT, and the LP of the edges of the SAPIEN bioprosthesis after implantation by fluoroscopy were actually similar, was the key finding of the prospective part of the study. In particular, it allowed us to use the postprocedural fluoroscopy recording of the LP of the edges of the stent as a surrogate of the LP of the aortic annulus.

Ability of Automated 3D Analysis of Pre-implant MSCT to Accurately Predict the LP of the Aortic Annulus for C-arm Positioning The quality of the prediction by MSCT was very good (94%, 34/36, κ = 0.89), the LP of the aortic annulus predicted by MSCT was within 5° of the LP of the SAPIEN-valve by fluoroscopy.

Importantly, although the annular plane is usually harder to define by angiography in patients with minimal/mild calcifications, the degree of calcification of the aortic root had no impact on the quality of prediction of the MSCT analysis. In particular, in patients with minimal/mild calcifications, the quality of the prediction was excellent (8/8, κ =1).

The quality of the prediction contrasts with the previously reported results by Gurvitch et al.4 in which attempted prediction of the LP by MSCT was considered excellent in only 50% of the population. The possible explanations for our better prediction in this study are: (1) A better imaging quality of the MSCT scans used with 83% of scans considered excellent, whereas this was the case in only 50% of the patients in the study by Gurvitch et al.4; (2) A more extensive data set was provided to the physicians involved in the procedure. Indeed, in the study by Gurvitch et al., only three predefined combinations of angulations were provided to the operator, whereas in this study, the complete LP data set, from RAO 40° to LAO 40° with increments of 5° each, was provided to the operator. Therefore, in this study, the operator had a larger range of possible implantation coordinates to choose from.

25

CHAPTER 2

Magnitude of Interindividual Variation of 3D Aortic Annulus Orientation Although the overall orientation of the “LP” is going from caudal to cranial when the C-arm is moved from right to left, it is an important finding of this study that the magnitude of variation is more than 70° in the cranio–caudal axis and more than 80° in the right–left axis. This interindividual variation probably explains why, when performing empirical positioning of the C-arm during a TAVR procedure, several attempts using contrast injection is usually required to define an appropriate C-arm position; it also explains the reason why pre-procedural definition of the “LP” may be important.

Study Limitations This was a single-center study and patient referral and medical management may have influenced the results. It was also a nonrandomized study; however, the prospective design of the study and the consecutive nature of the population provide accurate information on the ability of pre-procedural MSCT analysis to accurately predict the “LP” of the aortic annulus and the corresponding positions of the C-arm. Finally, as the study was restricted to patients treated with a balloon-expendable bioprosthesis, the benefit of this strategy for the implantation of a self-expendable bioprosthesis remains to be explored.

CONCLUSIONS Based on a prospective and consecutive series, the present report provides important information on the use of preprocedural automated 3D analysis of MSCT to predict the LP of the aortic annulus and the corresponding position of the C-arm for TAVR. Besides the evaluation of vascular access and aortic annulus diameter, these findings reinforce the benefit of performing MSCT to properly plan TAVR procedures.

26

MSCT FOR DEFINING THE OPTIMAL IMPLANTATION VIEW DURING TAVR

REFERENCES 1. Cribier A, Eltchaninoff H, Tron C, Bauer F, Agatiello C, Sebagh L, Bash A, Nusimovici D, Litzler PY, Bessou JP, Leon MB. Early experience with percutaneous transcatheter implantation of heart valve prosthesis for the treatment of end-stage inoperable patients with calcific aortic stenosis. J Am Coll Cardiol 2004;43:698–703.

2. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, Tuzcu EM, Webb JG, Fontana GP, Makkar RR, Williams M, Dewey T, Kapadia S, Babaliaros V, Thourani VH, Corso P, Pichard AD, Bavaria JE, Herrmann HC, Akin JJ, Anderson WN, Wang D, Pocock SJ. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med 2011;364:2187–2198.

3. Tuzcu EM. Transcatheter aortic valve replacement malposition and embolization: Innovation brings solutions also new challenges. Catheter Cardiovasc Interv 2008;72:579–580.

4. Gurvitch R, Wood DA, Leipsic J, Tay E, Johnson M, Ye J, Nietlispach F, Wijesinghe N, Cheung A, Webb JG. Multislice computed tomography for prediction of optimal angiographic deployment projections during transcatheter aortic valve implantation. JACC Cardiovasc Interv 2010;3:1157–1165.

5. Kurra V, Kapadia SR, Tuzcu EM, Halliburton SS, Svensson L, Roselli EE, Schoenhagen P. Pre-procedural imaging of aortic root orientation and dimensions: Comparison between x-ray angiographic planar imaging and 3-dimensional multidetector row computed tomography. JACC Cardiovasc Interv 2010;3:105–113.

6. Rivard AL, Bartel T, Bianco RW, O’Donnell KS, Bonatti J, Dichtl W, Cury RC, Feuchtner GM. Evaluation of aortic root and valve calcifications by multi-detector computed tomography. J Heart Valve Dis 2009;18:662–670.

7. Samim M, Stella PR, Agostoni P, Kluin J, Ramjankhan F, Sieswerda G, Budde R, der Linden M, Samim M, Hillaert M, van Herwerden L, Doevendans PA, van Belle E. Transcatheter aortic implantation of the edwards-sapien bioprosthesis: Insights on early benefit of tavi on mitral regurgitation. Int J Cardiol 2011;152:124–126.

8. Samim M, Stella PR, Agostoni P, Kluin J, Ramjankhan F, Sieswerda G, Budde R, van der Linden M, Juthier F, Banfi C, Hurt C, Samim M, Hillaert M, van Herwerden L, Bertrand ME, Doevendans PA, Van Belle E. A prospective “oversizing” strategy of the edwards sapien bioprosthesis: Results and impact on aortic regurgitation. J Thorac Cardiovasc Surg 2013;145:398--405.

9. Kundel HL, Polansky M. Measurement of observer agreement. Radiology 2003;228:303–308.

27

CHAPTER 2

SUPPLEMENTARY MATERIAL METHODS Multislice Row Computed Tomography Pre-procedurally, all patients underwent a contrast enhanced retrospectively ECG-gated MSCT scan on either a 64- or 256-slice scanner (Brilliance 64 or iCT, respectively, Philips Medical Systems, Best, the Netherlands), according to standard scan protocols that were individually adjusted based on patient body habitus. Tube voltage was 100 or 120 kV, tube current 200400 mAs, collimation 64 or 128 x 0.625 mm and gantry rotation time of 270-420 ms. The scan range was set from the level of the subclavian arteries to the level of the head of the femur. A continuous ECG trace was recorded during image acquisition and images were reconstructed at each 12.5% of the R-R interval, obtaining a total of 8 datasets per scan (including 37.5% for systole and 75% for diastole).

All scans were performed during mid-inspiratory breath-hold, and during injection of iodinated non-ionic contrast agent (Ultravist iopromide - 300 mg/mL, Bayer Schering Pharma AG, Berlin, Germany Healthcare Tarrytown, New York). Beta-blockers were not routinely administered prior to scanning.

Data sets were reconstructed and off-line post-processing of MSCT images was performed on a dedicated workstation.

Automated 3D Image Analysis of MSCT The MSCT data were sent to an external workstation for dedicated analysis. All scans were analyzed using a software package and a dedicated 3D aortic valve analysis workflow (3mensio ValvesTM, 3mensio Medical Imaging BV, The Netherlands, http://www.3mensio.com). Early systolic images of the aortic root reconstructed at 30 to 37.5% of the R-R interval were selected, as recommended.

The first step of the valve analysis workflow is an automatic segmentation of the ascending aorta. Alternatively, placing control points in the aorta, the aortic valve, and in the left ventricle will manually create a centerline. The application now provides an estimated aortic annulus plane location and orientation. In the next step of the workflow, these must be refined. This can be done by positioning and rotating the annulus plane as depicted in Figure 2 in order to define the plane that permits the identification of the 3 aortic sinuses4.

In the final step of the workflow, the latero-lateral (LAO, RAO) and cranio-caudal angles required for a perpendicular orientation of the C-Arm to the aortic annulus plane are determined. This is done by rotating a virtual C-Arm around the aortic centerline at the intersection point with the annulus plane (Figure 2B). A simulated angiogram is interactively updated when the virtual C-Arm is rotated. The respective images are displayed on the screen.

28

MSCT FOR DEFINING THE OPTIMAL IMPLANTATION VIEW DURING TAVR

TAVR procedure TEE was used for all procedures. Patients were premedicated with aspirin and antibiotics.

Heparin was used to maintain an activated clotting time >250 sec. The ACT was reversed with protamine at the end of the procedure.

Both for the TF and TA-AVR approach, the device-delivering sheath was inserted before crossing the valve with any wire. Intra-procedural imaging of the aortic valve was achieved by contrast injection via a pigtail catheter, which was positioned just above the valve itself and introduced via a femoral artery (the contralateral in the case of TF-AVR). Balloon aortic valvuloplasty with a 20- or 23-mm balloon was performed under rapid pacing to pre-dilate the native aortic valve. The prosthesis was subsequently deployed under rapid pacing (180 to 220 beats/min). Exit peripheral angiography was performed to ensure no extravasation of contrast prior to removal of the femoral sheath. Immediately post-procedure, the sheath was removed and surgically closed in the operating room by a surgeon or by using 2 percutaneous closure devices (Perclose, ProGlide, Abbott Vascular).

For the purpose of rapid ventricular pacing during TA-AVR, two unipolar epicardial pacer wires were secured and tested with a high output epicardial pacing system to ensure ventricular capture at rates of 180-220 bpm.

Procedural success was defined as the implantation of a functional prosthetic valve within the aortic annulus at the end of the procedure without in-laboratory mortality. Patients received aspirin (81 mg/day) and clopidogrel (75 mg/day) indefinitely. Warfarin was substituted for clopidogrel in patients with atrial fibrillation.

29

CHAPTER

3 AUTOMATED 3D ANALYSIS OF PRE-PROCEDURAL MSCT TO PREDICT ANNULUS PLANE ANGULATION AND C-ARM POSITIONING: BENEFIT ON PROCEDURAL OUTCOME IN PATIENTS REFERRED FOR TAVR JACC Cardiovasc Imaging. 2013 Feb;6(2):238-48

Samim M1 Stella PR1 Agostoni P1 Kluin J2 Ramjankhan F2 Budde RPJ3 Sieswerda G1 Algeri E4 van Belle C4 Elkalioubie A4 Juthier F4 Belkacemi A1 Bertrand ME4 Doevendans PAFM1 van Belle E4 Department of Cardiology, 2Department of Cardiothoracic surgery,

1

Department of Radiology, University Medical Center, Utrecht, The Netherlands

3

Department of Cardiology, University Hospital, Lille, France

4

CHAPTER 3

ABSTRACT Objectives The aim of this study was to determine whether pre-procedural analysis of multislice computed tomography (MSCT) could accurately predict the â&#x20AC;&#x153;line of perpendicularityâ&#x20AC;? for valve implantation perpendicular to the native aortic valve plane during transcatheter aortic valve replacement (TAVR) and its impact on the outcome of the procedures.

Background Optimal positioning of the transcatheter aortic prosthesis is paramount to procedural success of TAVR procedures.

Methods All patients referred for TAVR at our center underwent a routine pre-procedural MSCT scan.

A MSCT based 3-dimensional (3D) analysis using software dedicated to define the line of perpendicularity and the corresponding C-arm positioning was performed in 71 consecutive patients. In 35 patients, the results of the MSCT analysis were not available at the time of the procedure (angiography cohort). In that cohort the position of the C-arm was determined during the procedure using ad-hoc angiography. In 36 patients, the MSCT analysis was performed pre-procedurally and results were available at the time of the procedure (MSCT cohort). In that cohort the position of the C-arm was derived from the MSCT analysis rather than by ad-hoc angiography.

Results Intraobserver and interobserver reproducibility of MSCT analysis to predict the line of perpendicularity were excellent (kappa = 1 and 0.94, respectively). Patient variations of the perpendicularity line ranged >70°. Compared with the angiography cohort, the MSCT cohort was associated with a significant decrease in implantation time (p = 0.0001), radiation exposure (p = 0.02), amount of contrast (p = 0.001), and risk of acute kidney injury (p = 0.03). Additionally, the combined rate of valve malposition and aortic regurgitation was also reduced (6% vs. 23%, p = 0.03).

Conclusions Automated 3D analysis of pre-implantation MSCT accurately predicts the line of perpendicularity for valve implantation perpendicular to the native valve plane during TAVR. With this approach, the implantation of the balloon-expandable prosthetic valve can be performed safely without a test aortogram in the majority of cases, with a low rate of valve malpositioning and regurgitation.

32

MSCT IN TAVR FOR BETTER IMPLANT ANGLE AND OUTCOMES

INTRODUCTION Transcatheter aortic valve replacement (TAVR) is a minimally invasive procedure performed in patients with severe aortic stenosis and at high surgical risk.1, 2 Optimal positioning and deployment of the prosthetic valve during TAVR is crucial in order to avoid valve embolization, coronary ostial obstruction, perivalvular regurgitation, and conduction disturbance.3 During deployment of the valve prosthesis, an optimal view of the aortic annulus, where the valve will be implanted, has to be determined.1,2 The aim of the optimal view is to position the x-ray tube C-arm perpendicular to the aortic annulus plane in order to achieve an appropriate delivery of the valve during TAVR. Previous studies have shown that the aortic valve is typically directed in a cranial and anterior fashion with angulation to the right, and that corresponding C-arm positions describe a line in a 2-dimensional space. This line is known as the â&#x20AC;&#x153;line of perpendicularityâ&#x20AC;? of the C-arm in relation to the aortic annulus.4 Selecting the best angiographic view usually requires the intraprocedural performance of several angiograms in different angulations of the C-arm, using a considerable amount of contrast with its known possible nephrotoxic effect, and radiation for both patients and operators. An accurate preprocedural assessment of the line of perpendicularity of the annulus would help planning TAVR procedures.

Multislice computed tomography (MSCT) provides detailed anatomic assessment of the aortic root and valve annulus, and assesses the suitability of iliofemoral access. MSCT has also significant potential to assess the aortic root in relation to the body axis.4 Pre-procedural angle prediction of the C-arm with MSCT may decrease the number of angiograms required during the procedure, shortening procedure time, radiation dose, and contrast usage, and may increase the likelihood of an optimal implantation of the valve prosthesis by optimizing the orientation during device placement.

The present study was designed to evaluate whether automated 3-dimensional (3D) analysis of pre-procedural MSCT scans could be used to: 1) reproducibly define the aortic annulus plane and the line of perpendicularity needed for prosthesis positioning and deployment; 2) define the interindividual variation of the line of perpendicularity among patients; and 3) improve safety parameters and clinical outcome of the TAVR procedures.

METHODS Study population All consecutive patients with severe aortic stenosis considered to be at high or prohibitive surgical risk (as evaluated by a multidisciplinary heart team) and referred to our center for TAVR with a balloon-expandable endoprosthesis between January 2009 and June 2011 were included in the study (Figure 1).

Pre-procedural MSCT was performed in all patients within 3 months before TAVR. The automated MSCT 3D software analysis (3Mensio Valves, version 3.0, 3mensio Medical Imaging BV, the Netherlands) has been available in our institution since December 2009. Analysis of the

33

CHAPTER 3

Figure 1 | Flowchart of the Study. All consecutive patients treated between January 2009 and June 2011 were part of the study (n = 71). Automated multislice computed tomography (MSCT) 3-dimensional (3D) software (3-mensio) was available in December 2009. All consecutive patients treated between January 2009 and December 2009 constituted the Angiography Cohort A (n = 24). From January 2010 until June 2011, 1 of every 4 patients was included in the Angiography Cohort B (n = 11), and 3 of every 4 patients were included in the MSCT cohort (n = 36)

pre-procedural MSCT was performed for all patients. Based on the performance and on the availability of the results of this analysis to the operator at the time of the procedure, 2 cohorts were constituted. In the angiography cohort, the analysis of the pre-procedural MSCT was performed after the TAVR procedure had already been performed. In this cohort, the results of MSCT analysis were not available to the operator at the time of the procedure. In the MSCT cohort, the analysis was performed before the procedure, and the results of MSCT analysis were available to the operator at the time of the procedure.

At the time of 3Mensio software availability, 24 procedures had already been performed. These 24 patients were analyzed with the 3Mensio software retrospectively (Angiography Cohort A).

Forty-seven additional patients were treated from December 2009 until June 2011. In 1 of every 4 patients randomly chosen (n = 11), the automated 3D software analysis was not performed before but after the procedure, and the operators were therefore performing the procedure without knowledge of the results of the MSCT 3D analysis. The outcome of this group (Angiography Cohort B, n = 11) was compared with the outcome of the previous group (Angiography Cohort A, n = 24) to evaluate any significant impact on clinical outcome of changes over time in practice and clinical experience of the operators (see the following text).

In 3 of every 4 patients (n = 36), the MSCT 3D analysis was performed before the procedure,

34

MSCT IN TAVR FOR BETTER IMPLANT ANGLE AND OUTCOMES

and the results were provided to the operators before the beginning of the procedure. This group constituted the MSCT cohort. The patients for whom the results of the 3D MSCT analysis (n = 24 + 11 = 35) were not made available to the operators at the time of the procedure constituted Angiography Cohort A+B. Multislice computed tomography As part of the work-up procedure, the patients underwent MSCT within 3 months before TAVR. The detailed MSCT procedure is described in the supplementary materials in the Online Methods. utomated 3D image analysis of MSCT All MSCT images were analyzed by a procedure-independent operator, who was trained to use the dedicated 3D aortic valve analysis 3-mensio software. For the 35 patients in the angiography cohort, analyses were completed on 2 separate occasions by one dedicated operator. A second operator completed a third series of analysis.

For each patient, the image quality of the computed tomography (CT) was rated into 3 groups by the operators: excellent, adequate, or inadequate, depending on potential degradations due to calcification, motion, or contrast issues. The severity of aortic root calcification was classified into 4 groups: minimal (<25% of total circumference), mild (25% to 50%), moderate (50% to 75%), and severe (75% to 100%).5 The process of determination of the aortic annulus plane and the predicted line of perpendicularity of the aortic annulus is described in the supplementary materials in the Online Methods and in Figure 2.

For each patient, the software provides an infinite series of C-arm positions (defined by their combination of cranio-caudal and right anterior oblique/left anterior oblique [RAO-LAO] angulation), where the C-arm is perpendicular to the annulus plane. This series is named the predicted line of perpendicularity of the annulus. For each patient, the predicted â&#x20AC;&#x153;line of perpendicularityâ&#x20AC;? of the annulus can be drawn on a cranio-caudal axis/right-left axis figure (Figure 3).

TAVR procedure The prosthetic stented Edwards SAPIEN (until August 2010) or SAPIEN XT (since August 2010) valves (Edwards Lifesciences, Irvine, California) were mechanically crimped on a balloon catheter immediately before implantation. The Retroflex or Novaflex delivery systems (Edwards Lifesciences) were used for device implantation via the femoral artery. Until August 2010, the femoral artery was accessed via a surgical cut-down; since August 2010, it has been performed through a percutaneous puncture. The Ascendra and Ascendra 2 delivery systems (Edwards Lifesciences) were used in case of apical access.

Transfemoral and transapical aortic valve replacement were performed under general anesthesia in the catheterization laboratory and are described in the supplementary materials (Online Methods).6 ,7 All patients, including those treated through a transapical approach, were indeed placed flat on their backs during the procedure.

Besides the use of the C-arm aligned with the plane of the aortic annulus, the operator was free

35

CHAPTER 3

to choose the projection that suited the procedure. An appropriate projection for prosthesis implantation was defined as a projection in which the right coronary cusp was oriented towards the x-ray detector and in which the annulus projection was not superimposed with the spine and not superimposed with the delivery catheter present in the descending aorta.

The angiography cohort and the MSCT cohort: specificities In patients in the MSCT cohort (n = 36), the position of the C-arm was chosen among the positions proposed by the 3Mensio analysis of the MSCT images and included in the predicted line of perpendicularity. No angiography of the aortic root was performed. The balloonmounted bioprosthesis was positioned with the aim to locate the unexpanded bioprosthesis mid-way at the level of the virtual line connecting the insertion sites of the 3 cusps to the aortic

Figure 2 | Defining the Line of Perpendicularity of the Aortic Annulus Using Automated MSCT 3D Software Analysis. After defining the central line of the ascending aorta (yellow line), the valve region is visualized in 3 double oblique multiplanar reconstruction (MPR) viewports (A to C). The 2 double oblique views show perpendicular cross sections through the aortic valve (A and B). The other shows the annulus plane in short axis (C). The level of the annulus plane relative to the central line can be changed by dragging the red icon at the level of the annulus plane. The orientation of the plane can be refined by dragging the green (A) and pink lines (B) in the oblique views and verified by rotating the corresponding green and pink handles on the view of the annulus plane (C). The corresponding 3D view is presented (D). A simulated angiogram with corresponding C-arm angles displayed on the 2 bottom corners is obtained (E). A virtual C-arm can be rotated along the aortic centerline by dragging the eye icon in the left view. The simulated projection is updated interactively. The aortic centerline and the aortic annulus plane are drawn over the simulated angiogram. A full description of the method is provided as supplementary materials (Online Methods).

Abbreviations as in Figure 1.

36

MSCT IN TAVR FOR BETTER IMPLANT ANGLE AND OUTCOMES

wall. The first angiogram was performed as part of the delivery while the balloon-mounted bioprosthesis was sitting within the aortic annulus to validate the height of the bioprosthesis relative to the annulus; rapid pacing was then started, and the bioprosthesis was delivered.

In case the alignment of the annulus appeared to be inaccurate to the operator, the rapid pacing was not started, and delivery was not performed. Additional angiograms were then performed to define a proper C-arm angulation before proceeding to the delivery under rapid pacing.

In patients in the Angiography Cohort A+B (n = 35), the proper position of the C-arm was determined empirically using test angiograms according to standard practice. Once the proper position of the C-arm was defined, the aortic annulus was crossed with the balloonmounted endoprosthesis, and the procedure was carried out in a similar fashion to that in the MSCT cohort using a final angiogram to validate the height of the endoprosthesis just before beginning delivery.

The major difference between the patients included in the Angiography Cohort A and those included in the Angiography Cohort B is the period in which they were treated. The Angiography Cohort B was the most recent and was contemporary with the MSCT cohort. The comparison between the 2 angiography cohorts was used to evaluate the potential impact of a learning curve or procedural changes over time in practice as well as a potential increase in operator expertise.

Although the reduction in contrast and radiation was not mandated by the protocol, it was expected as a consequence of the use of the MSCT-based 3Mensio software for prediction of the optimal implantation view.

Procedural characteristics and follow-up Implantation time was defined as the time between introduction and removal of the delivery sheath. For all procedures, the number of test angiograms, and the amount of contrast (milliliters) and radiation (Greys) delivered to the patient were recorded. The need for postimplantation balloon dilation of the bioprosthesis was also recorded. Valve malposition was defined as an inadequate position of the device requiring the implantation of a second valve or an emergent surgery. Post-procedural aortic regurgitation (AR) was evaluated by transthoracic echocardiography on the day of the procedure and was classified as none/trivial (= 0), mild (= 1), moderate (= 2), moderate-to-severe (= 3), or severe (= 4). An AR grade â&#x2030;Ľ2 was considered significant. Post-procedural acute kidney injury was defined as an increase in serum creatinine by >25% or 44 Îźmol/l (0.5 mg/dl).8 Thirty-day outcomes, including the implantation of a permanent pacemaker or the occurrence of myocardial infarction, stroke, or death were recorded. The primary endpoint was pre-defined as the occurrence of the composite endpoint of valve malposition and/or AR grade â&#x2030;Ľ2 post-procedure. Secondary endpoints were: 1) implantation time; 2) amount of contrast; 3) amount of radiation delivered to the patient during the procedure; 4) occurrence of post-procedural acute kidney injury; and 5) mortality at 30 days.

37

CHAPTER 3

Statistical analysis Continuous variables were expressed as mean ± standard deviation, and categorical variables were reported as frequencies. For comparison between categorical variables, a chi-square test was used, and an independent-sample t test was used to compare continuous values. In order to adjust to potential confounding factors, multivariate linear regression or multivariate logistic regression analyses were performed. Reproducibility of MSCT 3D analysis was investigated using kappa statistics.9 For these analyses, a 40° range of RAO-LAO angulations (between RAO 20° and LAO 20°) was considered. Cranio-caudal angulation was provided for any 1° increment of RAO-LAO angulation. If for any of these 40 positions, the cranio-caudal angulation was >5° outside the prediction, then the whole analysis was considered discordant. Analyses were performed using IBM SPSS Statistics 19.0 (IBM, Armonk, New York).

RESULTS Baseline characteristics of the study population A total number of 71 patients were included in this study, 36 patients in the MSCT cohort and 35 patients in the Angiography Cohort A+B. The baseline clinical and echocardiography characteristics of the population are presented in Table 1. A little more than 50% of the population consisted of women, the mean age was 80 ± 6 years, and the mean logistic EuroSCORE was 18 ± 11.

Within the angiography cohort, no significant difference was observed between the patients treated before (Cohort A, n = 24) or after (Cohort B, n = 11) the availability of 3Mensio software.

Similarly, no difference was observed between the Angiography Cohort A+B and the MSCT cohort.

Reproducibility of 3D MSCT analysis as performed in the Angiography Cohort A+B The quality of the CT scan was considered excellent in 83% (59 of 71) and adequate in 17% (12 of 71) of cases. In no case was the quality of the CT considered inadequate. Calcification of the aortic root was classified as minimal (4% [n = 3]), mild (18% [n = 13]), moderate (48% [n = 34]), or severe (30% [n = 21]).

Within a margin of 5°, intraobserver (35 of 35, kappa = 1) and interobserver (34 of 35, kappa = 0.94) reproducibility of the MSCT based prediction of the line of perpendicularity was excellent.

Interindividual variation of the line of perpendicularity Interindividual variation of the line of perpendicularity of the annulus among patients in the angiography cohort is shown in Figure 3. It shows the heterogeneity of the orientation of the line of perpendicularity of the annulus from patient to patient with a range of 70° in the craniocaudal axis and 80° in the right-left axis. This demonstrates that for each patient, the proper position of the C-arm can be anywhere in a range of 70° in the cranio-caudal axis for every fixed right-left position, and anywhere in a range of 80° in the right-left axis for every fixed cranio-caudal position.

38

MSCT IN TAVR FOR BETTER IMPLANT ANGLE AND OUTCOMES

Men

0.71

p Value

78 ± 9

MSCT Cohort (n = 36)

p Value

79 ± 7

Angiography Cohort A+B (n = 35)

Angiography Cohort B (n = 11)

Age, yrs

Angiography Cohort A (n = 24)

Table 1 | Clinical and Echocardiographic Characteristics at Baseline (N = 71)

79 ± 9

82 ± 5

0.06

12 (50)

5 (45)

0.90

17c(49)

15c(42)

0.56

25.6 ± 6.9

27.3 ± 4.3

0.48

26.2 ± 6.0

26.7 ± 5.5

0.67

Hypertension

17 (71)

9 (82)

0.69

26 (74)

27 (75)

0.96

Dyslipidemia

15 (62)

6 (55)

0.72

21 (60)

16 (44)

0.18

Diabetes

5 (21)

5 (45)

0.23

10 (29)

12 (33)

0.61

Smoking

3 (13)

2 (18)

0.63

5 (14)

5 (14)

0.99

CAD

16 (67)

9 (82)

0.44

25 (71)

23 (64)

0.50

Prior MI

8 (33)

4 (36)

0.91

12 (34)

8 (22)

0.26

Prior PCI

8 (33)

3 (27)

0.90

11 (31)

11 (31)

0.94

Prior CABG

6 (25)

4 (36)

0.69

10 (29)

4 (11)

0.08

CVD

9 (37)

2 (18)

0.44

11 (31)

7 (19)

0.27

PVD

6 (25)

2 (18)

0.93

8 (23)

7 (19)

0.73

Renal disease

5 (21)

3 (27)

0.69

8 (23)

14 (39)

0.22

107 (55)

110 (81)

0.35

109 (46)

91 (68)

0.49

COPD

7 (29)

3 (27)

0.91

10 (29)

12 (33)

0.61

Prior malignant disease

6 (25)

2 (18)

0.93

8 (23)

9 (25)

0.88

Logistic EuroSCORE

19.7 ± 12.8

16.2 ± 11.8

0.30

18.3 ± 11.8

18.5 ± 10.8

0.94

Peak aortic gradient, mm Hg

73.0 ± 23.4

71.4 ± 24.0

0.76

72.5 ± 23.4

66.3 ± 23.6

0.30

0.70 ± 16

0.72 ± 14

0.32

0.71 ± 15

0.71 ± 19

0.91

54.9 ± 12.4

51.3 ± 11.6

0.41

53.7 ± 12.4

52.4 ± 10.9

0.67

BMI, kg/m2

Creatinine, μmol/l, median (IQR)

Aortic valve area, cm

2

LVEF, %

Values are mean ± SD or n (%).

BMI = body mass index; CABG = coronary artery bypass grafting; CAD = coronary artery disease; COPD = chronic obstructive pulmonary disease; CVD = cerebrovascular disease; IQR = interquartile range; LVEF = left ventricular ejection fraction; MSCT = multislice row computed tomography; MI = myocardial infarction; PCI = percutaneous coronary intervention; PVD = peripheral vascular disease.

Procedural characteristics and clinical outcome in the angiography and MSCT cohorts Procedural characteristics are presented in Table 2. Within the angiography cohort, no significant differences were observed in prosthesis diameter, TAVR approach, procedural outcome, and 30-day outcome between the procedures performed before (Cohort A, n = 24) or after (Cohort B, n = 11) the availability of the automated 3D analysis of MSCT. The lack of difference between the 2 populations demonstrated that “time” or “learning curve” were not key predictors of outcome in this series. It also validated the combination of these 2 populations into a single population, the Angiography Cohort A+B.

39

CHAPTER 3

The comparison between the Angiography Cohort A+B and the MSCT cohort demonstrates no difference in choice of the prosthesis diameter. A trend for a reduced use of the transapical approach (p = 0.09) was observed in the MSCT cohort (Table 2).

In the Angiography Cohort A+B, test angiograms were performed in every patient before crossing the valve with the balloon-mounted bioprosthesis. The number of test angiograms ranged from 1 to 5. In 4 of 35 patients, only 1 test angiogram was needed. In 31 of 35 patients, at least 2 test angiograms were performed. In the MSCT cohort, no test angiogram was required in 35 of 36 of the patients. In 1 of 36 patients, the alignment of the annulus appeared to be inaccurate to the operator. In this patient, 1 additional angiogram was performed in order to define a proper C-arm angulation before proceeding to the delivery (Table 2).

The comparison between Angiography Cohort A+B and the MSCT cohort demonstrated that the use of automated 3D analysis of MSCT to predict the line of perpendicularity of the annulus and to choose the appropriate C-arm position to deliver the prosthesis is associated with a reduction of implantation time (46 ± 11 min vs. 63 ± 17 min, p < 0.0001) (Figure 4A),

Figure 3 | Individual Lines of Perpendicularity of the Aortic Annulus Derived From MSCT 3D Analysis in the 35 Patients From the Angiography Cohort.. From patient to patient and for any given RAO-LAO positioning, the appropriate cranio-caudal position can vary within an 80° range. From patient to patient and for any given cranio-caudal positioning, the appropriate RAO-LAO position can vary within a 70° range. LAO = left anterior oblique; RAO = right anterior oblique; other abbreviations as in Figure 1.

40

MSCT IN TAVR FOR BETTER IMPLANT ANGLE AND OUTCOMES

Figure 4 | Clinical Benefit of Using the Pre-Defined Line of Perpendicularity of the Aortic Annulus as Determined by 3D Analysis of MSCT. Pre-procedural 3D analysis of MSCT and pre-procedural prediction of C-arm position (MSCT cohort, green bars) is associated with a decrease in implantation time (A), radiation exposure (B), amount of contrast (C), risk of acute kidney injury (AKI) (D), or the combined endpoint of valve malposition/aortic regurgitation (AR) grade ≥2 (E), and a trend for a decrease in mortality (F) when compared with procedures performed without data from the MSCT analysis available (angiography cohort, pink bars). Abbreviations as in Figure 1.

radiation exposure (1,180 ± 617 Gy vs. 1,671 ± 1,111 Gy, p = 0.02) (Figure 4B), and amount of contrast delivered (110 ± 13 ml vs. 180 ± 119 ml, p = 0.001, Figure 4C). Similarly, the number of patients with post-procedural acute kidney injury was also lower (11% vs. 31%, p = 0.03) (Figure 4D). The use of MSCT-derived predictions was also associated with a reduced need for balloon redilation of the endoprosthesis post-implantation (3% vs. 20%, p = 0.02) and a reduced risk of valve malposition and/or AR grade ≥2 (6% vs. 23%, p = 0.03) (Figure 4E).

41

CHAPTER 3

Finally, a trend for a reduction in 30-day mortality (6% vs. 20%, p = 0.06) (Figure 4F) was also observed. These results did not significantly differ after adjustment for the approach (transapical vs. transfemoral) used for the delivery.

p Value

MSCT Cohort (n = 36)

Angiography Cohort A+B (n = 35)

p Value

Angiography Cohort B (n = 11)

Angiography Cohort A (n = 24)

Table 2 | Procedural Characteristics and 30-Day Outcome

Diameter of implanted valve prosthesis 23 mm

7 (29)

3 (30)

26 mm

17 (71)

7 (29)

29 mm

0 (0)

1 (9)

Transfemoral

11 (46)

7 (64)

Transapical

13 (54)

4 (36)

64 ± 18

61 ± 17

0.76

185 ± 124

170 ± 101

0.37

0.32

10 (28.6)

11 (30.6)

24 (68.6)

24 (66.7)

1 (2.9)

1 (2.8)

18 (51)

26 (72)

17 (49)

10 (28)

63± 17

46 ± 11

<0.0001

180 ± 109

108 ± 34

0.001

0.98

TAVR approach 0.32

0.09

Procedural outcome Implantation time, min Contrast medium, ml Radiation, Gy

1,688 ± 1,185 1,637 ± 999

0.57

1,671 ± 1,111 1,180 ± 617

0.02

Test angiograms before delivery 0

0 (0)

0 (0)

1

3 (12)

1 (9)

≥2

0.77

0 (0)

35 (97)

4 (11)

1 (3)

<0.0001

21 (88)

10 (91)

32 (89)

0 (0)

Valve malposition*

2 (8)

1 (9)

0.94

3 (9)

0 (0)

0.07

AR grade ≥2 by TTE†

4 (17)

2 (18)

0.91

6 (17)

2 (6)

0.11

Valve malposition* and/or AR grade ≥2 by TTE†

6 (25)

3 (27)

0.88

8 (23)

2 (6)

0.03

Post-procedural acute kidney injury 30 day outcome

8 (33)

3 (27)

0.90

11 (31)

4 (11)

0.03

Myocardial infarction

1 (4)

0 (0)

0.99

1 (3)

0 (0)

0.98

Stroke

0 (0)

1 (9)

0.66

1 (3)

1 (3)

0.99

Permanent pacemaker

0 (0)

1 (9)

0.66

1 (3)

1 (3)

0.99

Mortality

5 (21)

2 (18)

0.85

7 (20)

2 (6)

0.06

Values are n (%) or mean ± SD. *Valve malposition requiring implantation of a second valve or emergent surgery; †as evaluated on postprocedural TTE performed on the same day. AR = aortic regurgitation; TAVR = transcatheter aortic valve replacement; TTE = transthoracic echocardiography; other abbreviations as in Table 1

42

MSCT IN TAVR FOR BETTER IMPLANT ANGLE AND OUTCOMES

DISCUSSION Optimal positioning of the transcatheter aortic valve prosthesis is paramount to procedural success, and this requires optimal positioning of the x-ray tube C-arm perpendicular to the aortic annulus, the line of perpendicularity (4). The present study demonstrates that: 1) the magnitude of interindividual variation of the aortic annulus angulation among patients and of the corresponding line of perpendicularity is high; 2) automated 3D analysis of preimplantation MSCT has the ability to reproducibly predict the angulation of the aortic annulus and the corresponding line of perpendicularity for C-arm positioning; and 3) determination of the line of perpendicularity by MSCT before the procedure allows safe implantation of the balloon-expandable prosthetic valve without the need for an aortogram in the majority of cases, with a low rate of valve malpositioning and regurgitation.

Magnitude of interindividual variation of 3D aortic annulus orientation Although difficulties to find the projections resulting in valve implantation perpendicular to the valve plane has been experienced by physicians performing TAVR procedures,4 to the best of our knowledge, the precise interindividual variations have not been reported before.

Although the overall orientation of the line of perpendicularity is going from caudal to cranial when the C-arm is moved from right to left, it is an important finding of the present study that the magnitude of variation is more than 70째 in the cranio-caudal axis and more than 80째 in the right-left axis. These interindividual variations explain why when performing empirical positioning of the C-arm during a TAVR procedure, several attempts using contrast injection may be required to define an appropriate C-arm position; it also explains why pre-procedural definition of the line of perpendicularity (for valve implantation perpendicular to the valve plane) may be important.

Ability of automated 3D analysis of pre-implant MSCT to accurately predict the angulation of the aortic annulus and the corresponding line of perpendicularity for C-arm positioning In the present study, intraobserver (kappa = 1) and interobserver (kappa = 0.94) reproducibility of defining the line of perpendicularity with the 3Mensio software were excellent. This information is of importance, because in their preliminary report, Gurvitch et al.4 did not provide any information on the reproducibility of the methods used. It must be noted that the high reproducibility observed in the present study was achieved in a population with MSCT scans with an excellent or adequate imaging quality.

The quality of the prediction, illustrated by the use of MSCT prediction without correction of the C-arm in 35 of 36 cases, contrasts with previously reported results by Gurvitch et al.4, in which attempted prediction of the line of perpendicularity by MSCT was considered excellent in only 50% of the population. The possible reasons for the better prediction in the present study may be: 1) a better quality of the MSCT images, with 83% of scans considered excellent, whereas this was the case in only 50% of the patients in the study by Gurvitch et al.4; 2 a more

43

CHAPTER 3

extensive dataset was provided to the physicians involved in the procedure. Indeed, although in the study by Gurvitch et al. only 3 predefined combinations of angulations were provided to the operator, in the present study, the complete line of perpendicularity dataset, from RAO 40° to LAO 40° with increments of 5° each, was provided to the operator. Therefore, in the present study, the operator had the opportunity to choose from a wide range of appropriate coordinates for C-arm positioning.

Use of the MSCT based predicted line of perpendicularity is associated with a better procedural and 1-month outcome Safety and peri-procedural morbidity remains one of the major limitations of TAVR10. In the present study, several parameters of morbidity were improved by the use of the predicted line of perpendicularity by pre-procedural MSCT analysis. In particular, the risk of valve malposition and AR grade ≥2 was significantly reduced.

In the present study, precision in valve positioning was improved by using pre-procedurally defined line of perpendicularity. It allowed prevention of valve malposition requiring the need for a second valve or emergent valve surgery or AR. Such benefit is highly important in the light of recent studies showing that post-procedural AR is the major predictor of mortality at 1 year after TAVR.11, 12 This issue will become even more important in the near future if we choose to treat patients with lower surgical risk, in whom the predictability of the result will be a major criterion of the choice between TAVR and conventional surgery.

These results were achieved while using less contrast media during the procedure, which lead to a decrease in the frequency of contrast-induced acute kidney injury. The decrease in contrast use was mainly related to the fact that test angiograms, to search for the line of perpendicularity and for the optimal position of the C-arm, were no longer required. This is particularly important because contrast-induced kidney injury is a major cause of morbidity and mortality after transcatheter interventions,13 and because acute kidney injury has been shown to be associated with an increase in short-term and long-term mortality following TAVR.14, 15 Pre-procedural prediction of optimal implantation coordinates using MSCT, caused a reduction in radiation exposure during TAVR. As for the decrease in contrast media, this benefit was secondary to 1) the decreased need for test angiograms and 2) the reduced time needed to perform the procedure. Because high radiation exposure may increase the risk of skin lesions and cancer, such benefit is important in patients with a potentially longer life expectancy.16, 17 This latter group will increase as in the near future we will use TAVR in younger patients with fewer comorbidities. This benefit is also important for the safety of the operators, especially in high-volume centers or in delivery approaches in which the usual radiation protection glass shield cannot be used (transapical, transaortic).18 Switching from general to local anesthesia is key to reducing the invasiveness of this type of procedure. To achieve this goal, a reduction of the total duration of the procedure is needed.

In that regard, the reduction of the time needed for the procedure by a mean of 17 min is an important step.

44

MSCT IN TAVR FOR BETTER IMPLANT ANGLE AND OUTCOMES

Study limitations This was a single-center study, and patient referral and medical management may have influenced the results. It was also a nonrandomized study; however, the prospective design of the study and the consecutive nature of the population provide accurate information on the ability of pre-procedural MSCT analysis to accurately predict the line of perpendicularity of and the corresponding positions of the C-arm.

Although the safety benefit (reduced contrast and radiation use) was not mandated by the protocol, it was expected to be a consequence of MSCT-based procedural planning.

Nevertheless, because such benefit was not assured, it was important to validate this potential in a clinical situation. Indeed, in case the predicted angulations would have been inaccurate, the operator would have no choice but to perform additional angiograms to correct the position of the C-arm, thus cancelling the expected benefit.

Furthermore, because all patients were placed flat on their backs during the procedure, these results cannot be extrapolated to patients in whom a cushion is used to tilt the left side of the chest for transapical delivery. In that case, some correction of the angulation will have to be applied. Finally, as the study was restricted to patients treated with a balloonexpandable bioprosthesis, the benefit of this strategy for the implantation of a self-expandable bioprosthesis remains to be validated.

Conclusions Clinical implications. Based on a prospective, consecutive, and comparative series, the present report provides important information on the use of pre-procedural automated 3D analysis of MSCT to predict the line of perpendicularity for valve implantation perpendicular to the native valve plane. It further demonstrates that with this approach, the implantation of the balloonexpandable prosthetic valve can be performed without a test aortogram in the majority of cases and still be safe, with a low rate of valve malpositioning and regurgitation. The use of this strategy will also become important when TAVR will be evaluated in younger and lower-risk patients. In the latter patient subset, predictability and safety of the procedure will be critical in order to match the results achieved with conventional surgery.

45

CHAPTER 3

REFERENCES 1

2 3 4

5

6

7

8

9 10

11 12 13 14 15 16

17 18

46

A. Cribier, H. Eltchaninoff, C. Tron et al. Early experience with percutaneous transcatheter implantation of heart valve prosthesis for the treatment of end-stage inoperable patients with calcific aortic stenosis. J Am Coll Cardiol, 43 (2004), pp. 698–703 Smith CR, Leon MB, Mack MJ, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med 364:2187–98.

E.M. Tuzcu. Transcatheter aortic valve replacement malposition and embolization: innovation brings solutions also new challenges. Catheter Cardiovasc Interv, 72 (2008), pp. 579–580 R. Gurvitch, D.A. Wood, J. Leipsic et al. Multislice computed tomography for prediction of optimal angiographic deployment projections during transcatheter aortic valve implantation. J Am Coll Cardiol Intv, 3 (2010), pp. 1157–1165 A.L. Rivard, T. Bartel, R.W. Bianco, K.S. O’Donnell, J. Bonatti, W. Dichtl, R.C. Cury, G.M. Feuchtner.

Evaluation of aortic root and valve calcifications by multi-detector computed tomography. J Heart Valve Dis, 18 (2009), pp. 662–670.

M. Samim, P.R. Stella, P. Agostoni et al. Transcatheter aortic implantation of the Edwards-Sapien bioprosthesis: insights on early benefit of tavi on mitral regurgitation. Int J Cardiol, 152 (2011), pp.

124–126 M. Samim, P. Stella, P. Agostoni et al. A prospective “oversizing” strategy of the Edwards-Sapien bioprosthesis: results and impact on aortic regurgitation. J Thorac Cardiovasc Surg, 145 (2013), pp.

398–405 S.K. Morcos, H.S. Thomsen, J.A. Webb. Contrast-media-induced nephrotoxicity: a consensus report: Contrast Media Safety Committee, European Society of Urogenital Radiology (ESUR). Eur Radiol, 9 (1999), pp. 1602–1613 H.L. Kundel, M. Polansky. Measurement of observer agreement. Radiology, 228 (2003), pp. 303–308 H. Eltchaninoff, A. Prat, M. Gilard, FRANCE Registry Investigators et al. Transcatheter aortic valve implantation: early results of the FRANCE (FRench Aortic National Corevalve and Edwards) registry.

Eur Heart J, 32 (2011), pp. 191–197.

M. Gilard, H. Eltchaninoff, B. Iung, FRANCE 2 Investigators et al. Registry of transcatheter aortic-valve implantation in high-risk patients. N Engl J Med, 366 (2012), pp. 1705–1715 S.K. Kodali, M.R. Williams, C.R. Smith, PARTNER Trial Investigators et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med, 366 (2012), pp. 1686–1695 P.A. McCullough, R. Wolyn, L.L. Rocher, R.N. Levin, W.W. O’Neill. Acute renal failure after coronary intervention: incidence, risk factors, and relationship to mortality. Am J Med, 103 (1997), pp. 368–375 J.M. Sinning, A. Ghanem, H. Steinhauser et al. Renal function as predictor of mortality in patients after percutaneous transcatheter aortic valve implantation. J Am Coll Cardiol Intv, 3 (2010), pp. 1141–1149 Y. Elhmidi, S. Bleiziffer, N. Piazza et al. Incidence and predictors of acute kidney injury in patients undergoing transcatheter aortic valve implantation. Am Heart J, 161 (2011), pp. 735–739 P. Kaul, S. Medvedev, S.F. Hohmann, P.S. Douglas, E.D. Peterson, M.R. Patel. Ionizing radiation exposure to patients admitted with acute myocardial infarction in the united states. Circulation, 122 (2010), pp.

2160–2169 W. Wijns, Y. Popowski. Radiation exposure in patients with myocardial infarction: Another false alarm?

Circulation, 122 (2010), pp. 2113–2115 G.L. Russo, I. Tedesco, M. Russo, A. Cioppa, M.G. Andreassi, E. Picano. Cellular adaptive response to chronic radiation exposure in interventional cardiologists. Eur Heart J, 33 (2012), pp. 408–414

MSCT IN TAVR FOR BETTER IMPLANT ANGLE AND OUTCOMES

SUPPLEMENTARY MATERIAL METHODS Multislice Row Computed Tomography Pre-procedurally, all patients underwent a contrast enhanced retrospectively ECG-gated MSCT scan on either a 64- or 256-slice scanner (Brilliance 64 or iCT, respectively, Philips Medical Systems, Best, the Netherlands), according to standard scan protocols that were individually adjusted based on patient body habitus. Tube voltage was 100 or 120 kV, tube current 200400 mAs, collimation 64 or 128 x 0.625 mm and gantry rotation time of 270-420 ms. The scan range was set from the level of the subclavian arteries to the level of the head of the femur. A continuous ECG trace was recorded during image acquisition and images were reconstructed at each 12.5% of the R-R interval, obtaining a total of 8 datasets per scan (including 37.5% for systole and 75% for diastole).

All scans were performed during mid-inspiratory breath-hold, and during injection of iodinated non-ionic contrast agent (Ultravist iopromide - 300 mg/mL, Bayer Schering Pharma AG, Berlin, Germany, Healthcare Tarrytown, New York). Beta-blockers were not routinely administered prior to scanning.

Data sets were reconstructed and off-line post-processing of MSCT images was performed on a dedicated workstation.

Automated 3D Image Analysis of MSCT The MSCT data were sent to an external workstation for dedicated analysis. All scans were analyzed using a software package and a dedicated 3D aortic valve analysis workflow (3mensio ValvesTM, 3mensio Medical Imaging BV, The Netherlands, http://www.3mensio.com). Early systolic images of the aortic root reconstructed at 30 to 37.5% of the R-R interval were selected, as recommended.

The first step of the valve analysis workflow is an automatic segmentation of the ascending aorta. Alternatively, placing control points in the aorta, the aortic valve, and in the left ventricle will manually create a centerline. The application now provides an estimated aortic annulus plane location and orientation. In the next step of the workflow, these must be refined. This can be done by positioning and rotating the annulus plane as depicted in Figure 2 in order to define the plane that permits the identification of the 3 aortic sinuses4.

In the final step of the workflow, the latero-lateral (LAO, RAO) and cranio-caudal angles required for a perpendicular orientation of the C-Arm to the aortic annulus plane are determined. This is done by rotating a virtual C-Arm around the aortic centerline at the intersection point with the annulus plane (Figure 2B). A simulated angiogram is interactively updated when the virtual C-Arm is rotated. The respective images are displayed on the screen.

47

CHAPTER 3

TAVR procedure TEE was used for all procedures. Patients were premedicated with aspirin and antibiotics.

Heparin was used to maintain an activated clotting time >250 sec. The ACT was reversed with protamine at the end of the procedure.

Both for the TF and TA-AVR approach, the device-delivering sheath was inserted before crossing the valve with any wire. Intra-procedural imaging of the aortic valve was achieved by contrast injection via a pigtail catheter, which was positioned just above the valve itself and introduced via a femoral artery (the contralateral in the case of TF-AVR). Balloon aortic valvuloplasty with a 20- or 23-mm balloon was performed under rapid pacing to pre-dilate the native aortic valve. The prosthesis was subsequently deployed under rapid pacing (180 to 220 beats/min). Exit peripheral angiography was performed to ensure no extravasation of contrast prior to removal of the femoral sheath. Immediately post-procedure, the sheath was removed and surgically closed in the operating room by a surgeon or by using 2 percutaneous closure devices (Perclose, ProGlide, Abbott Vascular).

For the purpose of rapid ventricular pacing during TA-AVR, two unipolar epicardial pacer wires were secured and tested with a high output epicardial pacing system to ensure ventricular capture at rates of 180-220 bpm.

Procedural success was defined as the implantation of a functional prosthetic valve within the aortic annulus at the end of the procedure without in-laboratory mortality. Patients received aspirin (81 mg/day) and clopidogrel (75 mg/day) indefinitely. Warfarin was substituted for clopidogrel in patients with atrial fibrillation.

48

49

CHAPTER

4 THREE-DIMENSIONAL AORTIC ROOT RECONSTRUCTION DERIVED FROM ROTATIONAL ANGIOGRAPHY FOR TRANSCATHETER BALLOON-EXPANDABLE AORTIC VALVE IMPLANTATION GUIDANCE Int J Cardiol. 2014 Oct 20;176(3):1318-20

Mariam Samim1 Pierfrancesco Agostoni1 Freek Nijhoff1 Ricardo P.J. Budde2 Alferso C. Abrahams3 Jolanda Kluin4 Faiz Ramjankhan4 Pieter A. Doevendans1 Pieter R. Stella1 Department of Cardiology, 2Department of Radiology, 3Department of Nephrology,

1

Department of Cardiothoracic surgery, University Medical Center, Utrecht, The Netherlands

4

CHAPTER 4

STRUCTURED ABSTRACT Background Determining the optimal angiographic implantation projection is pivotal for correct positioning of balloon-expandable aortic prosthesis during transcatheter aortic valve replacement (TAVR).

We aimed to assess the value of intra-procedural 3-dimensional (3D) reconstruction of the aortic root derived from rotational angiography images, using dedicated software (DynaCT速), for identification of the ideal projection for prosthesis deployment and to compare it with multislice computed tomography (MSCT).

Methods Fifty-six consecutive patients were included in this study, comprising two cohorts: 41 patients undergoing TAVR with the DynaCT速 technique (DynaCT cohort) and 15 patients in whom MSCT was routinely used preoperatively (historical MSCT cohort). The accuracy of predicted projections was determined post-implantation using angiographic images and graded as excellent, satisfactory or poor.

Results There was a good correlation between the DynaCT速 and MSCT for prediction of optimal implantation views (R=0.84, P<0.001). Excellent final deployment projections, good implantation depth and balanced implanted prosthesis were more frequent in the DynaCTguided compared with MSCT-guided group (93% vs.80%, 95% vs.80%, 93% vs.80%, p=0.33, p=0.11, p=0.32). In 93% of DynaCT cases one aortogram was enough to find the ideal implant angle as opposed to only 20% in the MSCT cohort (p<0.001). DynaCT速 significantly reduced procedural radiation dose (660.5 mGy vs.1497.1 mGy, P<0.001), and showed a trend towards lower moderate/severe periprosthetic aortic regurgitation (PAR) (5% vs.13%, p=0.29).

Conclusions Intra-procedural rotational angiography-derived 3D reconstruction of the aortic root accurately predicts optimal angiographic projections for balloon-expandable TAVR with a significant reduction in radiation dose and number of aortograms during these procedures.

52

DYNACT FOR REAL-TIME DETERMINATION OF THE OPTIMAL IMPLANTATION VIEW DURING TAVR

INTRODUCTION Transcatheter aortic valve replacement (TAVR) is rapidly adopted as a viable and safe treatment strategy for patients with severe symptomatic aortic stenosis considered inoperable or at high surgical risk. For a successful TAVR it is important to accurately position and deploy the valve prosthesis in the 3-dimensional (3D) space of the aortic annulus and root. Adequate positioning of the prosthesis in the aortic annulus is of a greater importance for implantation of balloon expandable transcatheter heart valves (THV), such as the Edwards SAPIEN XT valve (Edwards Lifesciences LLC, Irvine, California, United States), owing to the design of these prostheses. On the other hand, self-expanding THV, such as the Medtronic CoreValve (Medtronic Inc., Minneapolis, Minnesota), provide sufficient safety for adequate positioning, due to the progressive prosthesis delivery and the possibility of readjusting placement during the first steps of deployment1. Incorrect positioning of the valve prosthesis in the aortic annulus may result in aortic regurgitation, heart block, valve embolization, coronary obstruction, or impairment of mitral valve and left ventricular function2-5. In order to decrease the risk of these complications, the valve prosthesis has to be implanted in the optimal implantation view (OIV), which is defined as the fluoroscopic view perpendicular to the native valve, aligning all three sinuses of Valsalva (Figure1A)6. Identifying this OIV conventionally with repeated aortic root angiograms may often cost valuable implantation time, and increased radiation exposure and use of contrast medium2.

An offline, preprocedural 3D reconstruction of the aortic annulus has been shown to be feasible with multislice computed tomography (MSCT) and it has been proven to reasonably predict OIV for TAVR procedures2, 7. The main limitation of MSCT is the lack of real-time determination of the optimal implantation angles during TAVR, whilst in the catheterization laboratory the patient may be positioned differently on the table as compared to patientâ&#x20AC;&#x2122;s position under the MSCT scanner, leading thus to different optimal angles.

More recently rotational angiography has been introduced for determination of the OIV during TAVR. This technique uses C-armâ&#x20AC;&#x2122;s ability to rotate rapidly around the patient to acquire angiographic images at numerous oblique projections around its arch. Three-dimensional angiographic reconstructions of the aortic root derived from rotational C-arm angiography have been evaluated to assess OIV in previous studies6, 8, 9. The aim of this study is to 1) report our experience with the use of rotational angiography-derived 3D reconstructions of the aortic root for improvement in implantation accuracy of balloon-expandable THV 2) evaluate procedural benefits, such as reduction in radiation exposure, number of aortograms needed to find the OIV, and the amount of contrast medium used during TAVR, with this technique as compared to MSCT 3) investigate the agreement between the optimal implantation angles according to conventional angiography, 3D reconstruction images derived from rotational angiography and MSCT.

53

CHAPTER 4

METHODS Consecutive patients with severe symptomatic aortic stenosis undergoing transcatheter implantation of a balloon-expandable Edwards SAPIEN XTâ&#x201E;˘ aortic bioprosthetic valve in our center, were included in this retrospective registry. These patients were subdivided in two cohorts. For patients treated between October 2011 and July 2012, the angles of the OIV were determined pre-procedurally using MSCT scans (historical MSCT cohort). In these patients MSCT was routinely used to find the ideal implant projections, while the learning curve for

Figure 1 | (A) Aortic root angiogram; perpendicular valve projection showing three cusps in line and at equal distance to each other; N indicates noncoronary; R right coronary; and L left coronary cusp. (B) 3D aortic root model derived from rotational angiography; an inappropriate view for valve implantation as the red perpendicularity circle, which is parallel to the plane spanned by the three lowest cusp points, is not degenerated to a straight line. (C) The red straight line below the annular plane indicates that the projection is perpendicular to the aortic valve.

54

DYNACT FOR REAL-TIME DETERMINATION OF THE OPTIMAL IMPLANTATION VIEW DURING TAVR

correct MSCT-based 3D aortic root reconstruction was achieved previously in our institution.

For patients treated between July 2012 and March 2013, besides preprocedural determination of the optimal implantation angles using MSCT, intra-procedural online rotational angiography for aortic root analysis was performed during TAVR and the final valve implantation view was based on the rotational angiography-derived 3D reconstruction (DynaCT cohort). All patients gave written informed consent to undergo TAVR. The presence and severity of periprosthetic aortic regurgitation (PAR) was evaluated in all patients by transthoracic echocardiography on day 4 after TAVR. MSCT All patients underwent a contrast enhanced, electrocardiogram-gated MSCT scan within 3 months before TAVR, as part of the preprocedural work-up. The detailed MSCT procedure and the specifications of MSCT analysis has been described elsewhere (2). Briefly, all MSCT scans were analyzed using a software package and a dedicated 3D aortic valve analysis workflow (3mensio Valves TM, 3mensio Medical Imaging BV, The Netherlands, http://www.3mensio. com). Images of the aortic root reconstructed in systole, at 37.5% of the R-R interval were selected for analysis of the aortic plane. Next, a centerline was automatically placed along the ascending aorta and mark points were placed at the hinge-points of the aortic cusps. The application provides an estimated aortic annulus plane location and orientation, after which the latero-lateral (left anterior oblique [LAO] or right anterior oblique [RAO]) and cranio-caudal angles required for a perpendicular orientation of the C-arm to the aortic annulus plane can be determined. This was done by rotating a virtual C-arm around the aortic centerline at the intersection point with the annulus plane. The MSCT analysis was performed off-line by an experienced observer before TAVR. Rotational angiography Irrespective of the chosen access route for TAVR (transfemoral, transapical or direct aortic) an intraoperative “online” 3D reconstruction of the aortic root and valve was generated from images of a C-arm rotational aortic root angiography using DynaCT® (Siemens AG, Forchheim, Germany). Rotational projection images were acquired during a C-arm rotation through a 200° arch, in 5 seconds time and using rapid pacing of 180 to 200 beats per minute. Angiographic contrast was injected through a 6F pigtail catheter in the ascending aorta above the valvular plane, at the level of the sino-tubular junction (78 mL of diluted contrast, 60% contrast medium with 40% NaCl 0.9%, injected at 15.6 mL/s) using the ACIST™ injector (ACIST Medical Systems Inc., Eden Prairie, Minnesota, USA). Subsequently, the images were automatically transferred to the Aortic Valve Guide (AVG, Siemens AG) software, which is an add-on to the DynaCT® system. This software automatically removes bone and soft tissue and generates an “online” 3D aortic root model in around 10 seconds. In this model the coronary artery ostia, the aortic valve cusps and the most inferior margin of each aortic cusp are automatically identified. Then the software derives a perpendicularity circle parallel to the plane spanned by the three lowest cusp points (see red circle Figure 1B). Visually, this perpendicularity circle degenerates to a straight line only if the three lowest cusp points are aligned (Figure 1C). On an OIV the three

55

CHAPTER 4

lowest cusp points are on a straight line and at equal distance to each other (Figure 1A).

Theoretically three such angles exist as there are three aortic cusps, and AVG automatically chooses the angle closest to the anterior-posterior position as the preferred implantation angle. The integration of the Siemens DynaCT® software and the Siemens C-arm system allows the C-arm to be driven automatically to the displayed angle. An additional standard aortogram has been performed in the indicated position to confirm that the view was satisfactory.

The reconstructed 3D aortic root model derived from rotational angiography images is routinely stored. Afterwards for a comparison with perpendicularity angles obtained using MSCT, multiple optimal implant angles predicted by DynaCT® were documented from LAO 40° through to RAO 40°. The latter was achieved by manual rotation of the reconstructed red perpendicularity circle (Figure 1B) until the 3 lowest cusp point were properly aligned and the circle appeared as a straight line, indicating a view perpendicular to the valve (Figure 1B-C).

Procedural outcome The main aim of this study was to compare procedural outcome after balloon expandable TAVR between the DynaCT cohort and the historical MSCT cohort. Primary endpoint in this study was quality of chosen implantation angle as judged on post-deployment angiographic images, based on previously utilised classification of excellent, satisfactory and poor7. This classification was based on valve perpendicularity and the degree of strut overlap with excellent and satisfactory angiographic projections defined as a gap between superior valve struts within half a cell height and between half to a whole cell height, respectively. Poor angiographic view was described as a gap between superior valve struts greater than the height of a full cell.

Additionally, to evaluate whether DynaCT® facilitates an easier identification of the ideal implant projection, the number of standard aortograms needed to find an OIV was documented in both cohorts. Because optimal placement of the THV along the aorto-ventricular axis is paramount in preventing too high or too low placement of the prosthesis or an oblique placement, post-deployment angiographic views were analysed for implantation depth and obliquity of the prosthesis in the aortic annulus. Implantation depth was defined as the ratio of the average of the two prosthesis edges below the aortic annulus plane to the prosthesis height as seen on the angiographic images after implantation (Figure 2). An ideal implantation depth of the Edwards SAPIEN XT™ was defined as 0.4. An implantation depth between 0.2 and 0.6 was considered as satisfactory to good and any other ratio was called unfavorable10.

The obliquity of the implanted stent in the aortic annulus plane was arbitrarily defined as the ratio of the largest to the smallest protruding edge of the prosthesis in the left ventricle (Figure 2). An obliquity ratio up to 2 was called a balanced position of the implanted prosthesis and a ratio above 2 was a sign of obliquely implanted prosthesis.

Comparison of angles predicted by DynaCT® versus MSCT In order to study the agreement between the angiographic views after valve implantation and the lines of perpendicularity predicted by DynaCT® and MSCT, the cranial/caudal and RAO/ LAO deviations and, using the Pythagorean Theorem, the shortest distance from these lines to

56

DYNACT FOR REAL-TIME DETERMINATION OF THE OPTIMAL IMPLANTATION VIEW DURING TAVR

the angiographic final implantation view and the angiographic post-deployment perpendicular prosthesis view were calculated (Figure 3). Post-deployment perpendicular prosthesis view was defined as a projection with perfectly superimposed anterior and posterior struts, indicating a perpendicular view of the valve prosthesis.

Figure 2 | Post-deployment angiograms showing the position of the prosthesis in the aortic annulus. (A) An example of a balanced position of the prosthesis. (B) Valve prosthesis is implanted obliquely in the aortic annulus, resulting in a high obliquity ratio.

STATISTICAL METHODS All analyses were performed with the use of SPSS software (version 20, SPSS Inc, Chicago, IL).

Continuous variables are expressed as mean ± standard deviation (SD) or median [interquartile range, IQR]. Categorical variables are described by frequencies and percentages. Comparison of continuous variables was performed using t test or Mann-Whitney U test. Comparison of categorical variables was performed using Pearson chi-squared test or Fisher’s exact test.

Linear regression lines by the least-squares method were drawn for every single patient for the perpendicular projection views predicted by DynaCT and MSCT separately. Correlations are expressed by Pearson correlation coefficient. All statistical tests were 2-tailed, and a value of P<0.05 was considered statistically significant.

RESULTS A total of 56 consecutive patients with symptomatic severe aortic stenosis underwent TAVR with a balloon expandable Edwards SAPIEN XT™ aortic prosthesis between October 2011 and

57

CHAPTER 4

Figure 3 | Linear regression line of optimal and perpendicular projection views. (A) Lines of perpendicularity predicted by DynaCT速 and MSCT were drawn for every single patient. Also the chosen implantation C-arm coordinates and post-deployment perpendicular view coordinates were documented. (B) The cranial/ caudal and RAO/LAO deviations and, using the Pythagorean Theorem, the shortest distance from the lines of perpendicularity to the angiographic views were calculated. LAO indicates left anterior oblique; MSCT, multislice computed tomography; and RAO, right anterior oblique.

58

DYNACT FOR REAL-TIME DETERMINATION OF THE OPTIMAL IMPLANTATION VIEW DURING TAVR

March 2013. Figure 4 shows the flow chart of this study. All patients had a pre-procedural MSCT scan, and the most recent 41 cases had an additional intra-procedural 3D angiographic reconstruction of the aortic root using rotational angiography recordings (DynaCT cohort). The first 15 patients with only pre-procedural MSCT were defined as historical MSCT cohort. There was no significant difference in the baseline clinical characteristics among the two cohorts (Table 1).

Figure 4 | Flow chart of patients included.

Table 1 | Baseline characteristics

Age, years

Historical MSCT cohort (n=15)

DynaCT cohort (n=41)

p-value

83.2 ± 6.0

80.6 ± 5.3

0.13

Male

3 (20)

15 (37)

0.34

Coronary artery disease

7 (47)

24 (59)

0.43

Prior CABG

2 (13)

7 (17)

1.00

Hypertension

11 (73)

28 (68)

1.00

Hypercholesterolemia

4 (27)

8 (20)

0.72

Diabetes mellitus

5 (33)

16 (39)

0.70

Cerebrovascular disease

4 (27)

6 (15)

0.43

Peripheral vascular disease

2 (13)

10 (24)

0.48

Creatinine (µmol/L)

164.5 ± 222

122.6 ± 118.8

0.48

History of neoplasm

4 (27)

5 (12)

0.23

17 ± 9.7

14.8 ± 7.3

0.53

Logistic EuroSCORE (%)

Data are shown as mean ± standard deviation or n (%). CABG: Coronary artery bypass grafting.

59

CHAPTER 4

Procedural outcome Device success as defined by VARC-2 definitions was achieved in 12 patients (80%) in the historical MSCT cohort and 39 patients (95%) in the DynaCT cohort (p=0.11). Reasons for not fulfilling the device success criteria were: additional valve implantation in one patient in the historical MSCT cohort due to frozen leaflet phenomenon and moderate PAR in all other patients. No patient developed acute cardiac complications due to rapid ventricular pacing during rotational angiography in the DynaCT cohort. There were no implantation related complications such a coronary ostium obstruction or valve embolisation. On average valve deployment was performed at 11.0±8.9° LAO and 0.6±11.3° caudal. Regarding the accuracy in predicting optimal implantation angles, there was a higher rate of excellent implantation view (Figure 5A) when DynaCT® was used as compared to the MSCT-guided group (93% vs. 80%, respectively, p=0.33). No poor implant angles were documented in either cohort. The number of aortograms needed for finding the optimal implant angles was significantly smaller in the DynaCT cohort as compared to the historical MSCT cohort (1.2 ± 0.6 vs. 3.2 ± 2.2 , respectively, p<0.001). In 12 (80%) cases in the historical cohort more than one aortogram was needed for finding optimal implant angles, whereas that was the case in only 3 (7%) of the DynaCT cohort (p<0.001) (Figure 5B).

Although not significant, less unfavorable implantation depths and more balanced implanted prostheses were found in the DynaCT cohort as compared to the historical cohort (Table 2).

No significant difference was found in the incidence of new pacemaker implantations and PAR after TAVR between the two cohorts (p=0.05 and p=1.00, respectively), although there was a trend towards a lower incidence of moderate/severe PAR in the DynaCT cohort (5% vs.13%, p=0.29). No TAVR-related acute myocardial infarction was reported in the DynaCT cohort. One patient in the historical MSCT cohort developed post-TAVR myocardial infarction, which was unrelated to the position of the implanted aortic valve prosthesis. In the latter case, the patient underwent thoracotomy because of a procedure-related leakage from the left ventricle (this was a transapical case), during which he developed a myocardial infarction due to obstruction of a coronary artery by a suture. Acute kidney injury (AKI), as defined by VARC-2 criteria, occurred in two patients in the DynaCT cohort and one patient in the MSCT cohort (5% vs. 7%, p=1.00).

There was a trend towards higher contrast use despite fewer aortograms in the DynaCT cohort as compared to the historical MSCT cohort (137 mL [137-167] vs. 120 mL [105-165]; p= 0.06). Although fluoroscopy times were comparable for both cohorts (16.0 min [12.6-20.3] in the DynaCT cohort vs. 16.5 min [13.8-20.4] in the historical MSCT cohort, p= 0.36), significantly less radiation dose was used in the DynaCT cohort (660.5 mGy [504.3-860.5] vs. 1497.0 mGy[1154.8-1546.5], p < 0.001).

Optimal implantation view predicted by DynaCT® versus MSCT In DynaCT cohort (in which patients underwent both MSCT and DynaCT) there was a significant correlation between the regression lines obtained using DynaCT® and MSCT (Figure 6A; R: 0.84, P<0.001). The mean difference between the Cranial/Caudal angles predicted by the two different modalities (DynaCT® minus MSCT) was -2.3±0.63°, which means that the regression

60

DYNACT FOR REAL-TIME DETERMINATION OF THE OPTIMAL IMPLANTATION VIEW DURING TAVR

Figure 5 | (A) Implant angle quality in the DynaCT cohort versus the historical cohort. With the use of DynaCT® an excellent final valve implant projection angle was achieved in 93% of cases versus 80% of cases in the historical cohort with the use of MSCT (p = 0.33). (B) Number of aortic root angiograms needed during transcatheter aortic valve replacement to find the ideal implantation projection. In 93% of DynaCT cases one aortogram was enough to find the ideal implant angle as opposed to only 20% in the MSCT cohort.

Figure 6 | The predicted optimal projection views as predicted by DynaCT® and MSCT in the DynaCT cohort with 41 patients.

lines are almost parallel, but the DynaCT line is shifted caudally with respect to the MSCT line.

Regression lines of the predicted angles were drawn for every patient and technique. The slope of regression lines were 0.99 ± 0.01 for DynaCT® in the DynaCT cohort and 0.99 ± 0.01 for MSCT in the historical cohort (p=0.24). Regarding the deviation from the lines of perpendicularity predicted by DynaCT® or MSCT, the coordinates of final implantation views were significantly closer to the DynaCT line as compared to the MSCT line (Table 3).

61

CHAPTER 4

Table 2 | Procedure related characteristics historical MSCT cohort versus DynaCT cohort Historical MSCT cohort (n=15)

DynaCT cohort (n=41)

11 (73)

32 (78)

Transapical

1 (7)

4 (10)

Transaortic

3 (20)

5 (12)

12 (80)

39 (95)

0.11

RAO/LAO

15.4 ± 8.3

9.4 ± 8.2

0.02

Cranial/Caudal

9.6 ± 11.5

-4.3 ± 8.8

<0.001

Good-Satisfactory

12 (80)

39 (95)

0.11

Unfavorable

3 (20)

2 (5)

Balanced

12 (80)

38 (93)

Oblique

3 (20)

3 (7)

None-mild

13 (87)

39 (95)

At least moderate PAR

2 (13)

2 (5)

1 (7)

4 (10)

1.00

168.7 ± 258.1

124.1 ± 104.2

0.80

1 (7)

2 (5)

1.00

Approach

p-value 0.73

Transfemoral

Procedural success Deployment angles

Implantation depth

Obliquity implanted prosthesis 0.32

Periprosthetic regurgitation after TAVR

Pacemaker implantation Creatinine (µmol/L)§ Acute kidney injury (AKIN classification)

0.29

Data are shown as mean ± standard deviation or n (%). § Creatinine assessed 48 hours after TAVR. AKIN: acute kidney injury network ; LAO: left anterior oblique; PAR: periprosthetic aortic regurgitation; RAO: right anterior oblique; TAVR: transcatheter aortic valve replacement.

Table 3| Deviation of angiographic views from the predicted line of perpendicularity MSCT in historical cohort (n=15)

DynaCT® in DynaCT cohort (n=41)

p-value

Deviation from deployment view ∆ RAO (-) / LAO (+)

-6.6±10.7°

1.9±15.5°

0.005

∆ Caudal (-) / Cranial(+)

5.0±7.9°

-2.4±13.1°

0.006

Shortest distance

4.8±5.6°

5.8±8.0°

0.68

Deviation from perpendicular prosthesis view ∆ RAO (-) / LAO (+)

4.1±20.1°

3.48±17.0°

0.75

∆ Caudal (-) / Cranial (+)

-6.0±19.3°

-4.3±15.4°

0.81

Shortest distance

11.4±7.6°

7.8±8.3°

0.06

Data are shown as mean ± standard deviation. All abbreviations can be found in Table 2.

62

DYNACT FOR REAL-TIME DETERMINATION OF THE OPTIMAL IMPLANTATION VIEW DURING TAVR

Learning curve for DynaCT When excluding the first 15 cases in the DynaCT cohort (in order to correct for a potential learning curve for the use of DynaCT®), a further decrease in the deviation from the chosen implantation view to the line of perpendicularity predicted by DynaCT® was found. After this correction the mean distance from the implantation view to the perpendicularity line was only 0.7±7.1° LAO and 0.7±5.8° caudal, which are significantly smaller than the values mentioned above for the MSCT in the MSCT cohort (P=0.003 and P=0.003 respectively).

DISCUSSION Finding the OIV before prosthesis deployment is paramount for procedural success during balloon expandable TAVR. Inaccurate positioning of the valve prosthesis might lead to devastating complications such as prosthesis migration, conduction system disturbances and coronary obstruction. Repeated aortography is the most common approach in determining the optimal angiographic implantation view. However, this method may often increase implantation time, radiation exposure and use of contrast medium. Our study demonstrates a good correlation between MSCT and DynaCT® for prediction of optimal implantation angles for aortic prosthesis positioning. However, both the chosen implantation views and the post-deployment perpendicular prosthesis views were closer to the line of perpendicularity predicted by DynaCT® than by MSCT. This suggests a greater agreement between final angiographic views and lines of perpendicularity predicted by DynaCT®. Furthermore, in contrast with MSCT, DynaCT® allows for real-time determination of the OIV during TAVR. As rotational angiography and DynaCT® assessments are performed in the same session as TAVR implantation, the predicted optimal C-arm angles reflect more precisely the current state of the aortic root as compared to those predicted by MSCT before TAVR. Patient positioning during MSCT and the TAVR procedure may not be identical, especially in transapical and direct aortic procedures, as the patient is usually tilted towards the right side with a small pillow under the left hemithorax to facilitate the work of the surgeon. Of paramount importance is the fact that DynaCT® can be used irrespective of the chosen access site for TAVR. Moreover, given that the combination of DynaCT® and AVG generates the OIV automatically, observer related errors do not play a role.

Findings in this study suggest a trend towards more accurate valve implantation when DynaCT® is used for prediction of the optimal valve deployment views for balloon expandable THV implantation. Two other studies comparing DynaCT® with MSCT for prediction of the optimal implantation angles, reported similar results with an increased accuracy of prosthesis deployment with the use of DynaCT®6, 9. Our study demonstrates that DynaCT® is able to reliably predict the optimal implantation projection and thereby it significantly reduces the need for repeated aortic root angiography. In the historical MSCT cohort in 80% of cases more than one aortogram was needed for finding the OIV, whereas in only 7% of the DynaCT cohort additional angiograms were needed.

63

CHAPTER 4

No significant difference was found in the number of new pacemaker implantations, AKI and incidence of moderate/severe PAR between the DynaCT-guided and MSCT-guided cohorts.

However a trend towards less moderate to severe PAR was found in the DynaCT group. One previous study showed that excellent implant angles occur more often in DynaCT-guided TAVR procedures than in TAVR without DynaCT® guidance and that excellent implant angles are significantly more likely to be associated with no PAR than non-excellent angles9. The latter suggests that optimization of implant angles in order to ensure a perpendicular aortoventricular axis may play an important role in preventing PAR after TAVR. Given the impact of even mild PAR on mortality after TAVR, its reduction may have important implications for the outcome of patients undergoing these procedures and larger studies may clarify the significance of implantation position for the incidence and severity of PAR11.

Fluoroscopy is the main imaging modality used to capture real time images on a C-arm system to provide guidance to the operator. Use of DynaCT® significantly decreased radiation exposure during TAVR procedures in our center. This reduction was due to fewer aortograms needed to find the ideal implantation projection in the DynaCT cohort. Because high radiation exposure increases the risk of skin lesions and cancer12, 13, such benefit is important in patients with a potentially longer life expectancy and is off course also beneficial for the operators. In the present study the amount of contrast medium needed for DynaCT-guided procedures was slightly larger (137 mL vs. 120 mL; p= 0.06) as compared to the MSCT-guided procedures, which is probably caused by the additional 47 mL contrast medium needed for rotational angiography. This difference in the volume of contrast medium was not associated with a higher incidence of AKI in the DynaCT cohort as compared to the historical MSCT cohort (5% vs. 7%, p=1.00).

Placement of the pigtail catheter in the ascending aorta, above the valvular plane, is of significant importance during rotational angiography, since placing it in the sinuses of Valsava might cause an inhomogeneous distribution of the contrast fluid. As a result the three aortic cusps are visualized unequally and the reconstruction of the valve with DynaCT® may be suboptimal. Also the timing of contrast injection needed for the rotational angiography is important for optimal angiographic images and hence 3D aortic root reconstruction. A delay of 2 seconds between the start of contrast injection and the rotational angiography run improves reconstruction because a sufficient amount of contrast medium already fills the three aortic cusps and a better visualization is thereby achieved. The aforementioned described phenomena were noticed during the first few TAVR procedures with DynaCT®. After excluding the first 15 DynaCT cases, an improved agreement was found between the chosen implantation view and the line of perpendicularity predicted by DynaCT®.

Although previous reports and the present study have demonstrated the significant value of real-time determination of the optimal implantation projection using DynaCT® instead of “off line” pre-procedural prediction using MSCT, nevertheless MSCT remains an important part of the workup procedure before TAVR and provides valuable anatomic assessment of the aortic root, annulus plane, distance to the coronary ostia, calcifications and iliofemoral access. Thus, MSCT remains a key imaging modality for planning TAVR procedures. However intra-procedural rotational angiography based DynaCT®, on the top of preprocedural MSCT,

64

DYNACT FOR REAL-TIME DETERMINATION OF THE OPTIMAL IMPLANTATION VIEW DURING TAVR

is of additional value for a successful implantation of the balloon expandable aortic valve prosthesis. Although the present study investigated the benefit of DynaCT® for implantation of this type of THV, Poon et al. have previously demonstrated the benefit of this system also in the implantation of self-expandable Medtronic CoreValve system9. Besides its benefit for an increased accuracy of prosthesis implantation, DynaCT® 3D reconstruction of 2-dimensional fluoroscopic images can also be used for post-procedural evaluation of prosthesis position in difficult cases14. Its easy integration with angiographic imaging during TAVR and its ability to provide valuable 3D reconstructions of anatomical structures make DynaCT® an important adjunct to the imaging techniques needed for a successful TAVR. Study limitations The historical cohort in this study consisted of only 15 patients for comparative reasons. They were the last patients at our institution in whom MSCT was used routinely for defining the ideal implantation projections, and in whom the learning curve for those assessments was achieved previously. Considering the technical message of this paper, the small size of this historical cohort might not distort the reported results.

To investigate the accuracy of rotational angiography for the prediction of OIV, we compared DynaCT® assessments with post-deployment perpendicular prosthesis views, whereas these views may not represent the original valve plane. However both DynaCT® and MSCT were compared with the same principle, which might justify the use of these perpendicular prosthesis views as reference.

CONCLUSIONS Intra-procedural rotational angiography-derived 3D reconstruction of the aortic root using DynaCT® accurately predicted optimal angiographic projections for balloon-expandable THV implantation, irrespective of the chosen access site. In addition, intra-procedural use of DynaCT® resulted in significant reduction in the number of aortograms needed to find the OIV and in the radiation dose used during TAVR. Moreover, a trend towards a lower incidence of moderate to severe post-procedural PAR was found when DynaCT® was used during TAVR.

65

CHAPTER 4

REFERENCES 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

66

Descoutures F, Himbert D, Radu C, Iung B, Cueff C, Messika-Zeitoun D, et al. Transarterial medtronic CoreValve system implantation for degenerated surgically implanted aortic prostheses. Circ Cardiovasc Interv. 2011 Oct 1;4(5):488-94.

Samim M, Stella PR, Agostoni P, Kluin J, Ramjankhan F, Budde RP, et al. Automated 3D analysis of preprocedural MSCT to predict annulus plane angulation and C-arm positioning: Benefit on procedural outcome in patients referred for TAVR. JACC Cardiovasc Imaging. 2013 Feb;6(2):238-4.

Wong DR, Ye J, Cheung A, Webb JG, Carere RG, Lichtenstein SV. Technical considerations to avoid pitfalls during transapical aortic valve implantation. J Thorac Cardiovasc Surg. 2010 Jul;140(1):196-202.

Al Ali AM, Altwegg L, Horlick EM, Feindel C, Thompson CR, Cheung A, et al. Prevention and management of transcatheter balloon-expandable aortic valve malposition. Catheter Cardiovasc Interv. 2008 Oct 1;72(4):573-8.

Masson JB, Kovac J, Schuler G, Ye J, Cheung A, Kapadia S, et al. Transcatheter aortic valve implantation: Review of the nature, management, and avoidance of procedural complications. JACC Cardiovasc Interv. 2009 Sep;2(9):811-20.

Binder RK, Leipsic J, Wood D, Moore T, Toggweiler S, Willson A, et al. Prediction of optimal deployment projection for transcatheter aortic valve replacement: Angiographic 3-dimensional reconstruction of the aortic root versus multidetector computed tomography. Circ Cardiovasc Interv. 2012 Apr;5(2):24752.

Gurvitch R, Wood DA, Leipsic J, Tay E, Johnson M, Ye J, et al. Multislice computed tomography for prediction of optimal angiographic deployment projections during transcatheter aortic valve implantation. JACC Cardiovasc Interv. 2010 Nov;3(11):1157-65.

Maeda K, Kuratani T, Torikai K, Shimamura K, Sawa Y. Transcatheter aortic valve replacement using DynaCT. J Card Surg. 2012 Sep;27(5):551-3.

Poon KK, Crowhurst J, James C, Campbell D, Roper D, Chan J, et al. Impact of optimising fluoroscopic implant angles on paravalvular regurgitation in transcatheter aortic valve replacements - utility of three-dimensional rotational angiography. EuroIntervention. 2012 Sep;8(5):538-45.

Nijhoff F, Agostoni P, Samim M, Ramjankhan FZ, Kluin J, Doevendans PA, et al. Optimisation of transcatheter aortic balloon-expandable valve deployment: The two-step inflation technique.

EuroIntervention. 2013 Sep 22;9(5):555-63.

Kodali SK, Williams MR, Smith CR, Svensson LG, Webb JG, Makkar RR, et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med. 2012 May 3;366(18):1686-95.

Small GR, Chow BJ, Ruddy TD. Low-dose cardiac imaging: Reducing exposure but not accuracy. Expert Rev Cardiovasc Ther. 2012 Jan;10(1):89-104.

Hung MC, Hwang JJ. Cancer risk from medical radiation procedures for coronary artery disease: A nationwide population-based cohort study. Asian Pac J Cancer Prev. 2013;14(5):2783-7.

Incani A, Lee JC, Poon KK, Crowhurst JA, Raffel OC, Walters DL. Normal functioning of a constrained CoreValve with DynaCT imaging demonstrating incomplete stent frame expansion. Int J Cardiol. 2013 Feb 10;163(1):e9-10.

67

PART TWO

VALVULAR REGURGITATION AFTER TAVR AND FACTORS ASSOCIATED WITH IT

CHAPTER

5 A PROSPECTIVE OVERSIZING STRATEGY OF THE EDWARDS SAPIEN BIOPROSTHESIS. RESULTS AND IMPACT ON AORTIC REGURGITATION The Journal of Thoracic and Cardiovascular Surgery Volume 145, Issue 2, February 2013, Pages 398-405

Mariam Samim1 Pieter R. Stella1 Pierfrancesco Agostoni1 Jolanda Kluin2 Faiz Ramjankhan2 Gertjan Sieswerda1 Ricardo Budde3 Marijke van der Linden1 RN, Francis Juthier4 Carlo Banfi4 Christopher Hurt4 Morsal Samim1 Marieke Hillaert1 Lex van Herwerden2 Michel E. Bertrand4 Pieter A. M. Doevendans1 Eric Van Belle4 Department of 1Cardiology, 2Department of Cardiothoracic surgery, 3Department of Radiology, University Medical Center, Utrecht, The Netherlands Department of Cardiology, University Hospital, Lille, France

4

CHAPTER 5

ABSTRACT Objective Moderate to severe aortic regurgitation occurs in 20% to 30% of cases after transcatheter aortic valve replacement (TAVR).

Methods The purpose of the study was to investigate the impact of a prospective policy of “oversizing” the Edwards SAPIEN bioprosthesis (Edwards Lifesciences LLC, Irvine, Calif) relative to the diameter of the aortic annulus on the rate and severity of aortic regurgitation in 28 consecutive patients considered eligible for TAVR on the basis of angiography, multislice computed tomography, and transthoracic echocardiography. This policy included the systematic use of transesophageal echocardiography (TEE) to determine the aortic annulus diameter more accurately and modification of pre-procedurally determined prosthesis size according to intraprocedural balloon dilation, choosing the largest prosthesis size possible.

Results Twenty-eight patients were initially considered eligible for TAVR. Based on TEE, 6 of 28 patients (21%) were excluded due to an annulus diameter >24 mm, making them ineligible for the implantation of a 26 mm prosthesis. In 6 of 22 included patients, the procedure was adapted according to our “oversizing” policy. As a result, the “prosthesis/annulus cover index” was 12.4% ± 4.3%. The procedure was successful in 21 of 22 patients (95%), and 18 patients were available for echocardiography at 1 month. Although a moderate to severe aortic regurgitation was observed pretreatment in 4 of 18 patients (22%), it was no longer the case at 1 month (0/18, 0%; P = .03). The improvement was secondary to a disappearance of the aortic regurgitation in all 7 patients with a significant aortic regurgitation at pretreatment, whereas the new aortic regurgitations appearing in 5 of the 11 patients, with no aortic regurgitation at pretreatment, were only mild.

Conclusions In patients undergoing implantation of an Edwards SAPIEN valve, a simple “oversizing” policy with a systematic use of transesophageal echocardiography and modification of prosthesis size according to balloon pre-dilation may prevent the occurrence of moderate and severe aortic regurgitations.

70

PROSPECTIVE OVERSIZING STRATEGY IN BALLOON EXPANDABLE TAVR

INTRODUCTION Transcatheter aortic valve replacement (TAVR) is a developing technique to treat patients with severe symptomatic aortic valve stenosis. 1,2 Because questions remain concerning safety and durability, its use currently is limited to patients contraindicated or at very high risk for conventional aortic valve replacement. After TAVI, a moderate to severe paravalvular aortic regurgitation (AR) can occur in approximately 20% to 30% of cases after implantation.3,4 Although this is usually considered as acceptable in the elderly population, it is a limitation to a broader use of this technique. Implantation of an undersized prosthesis has been proposed as a major cause of this issue.3,5-7 Although it is usually recommended to ‘‘oversize’’ the implantation of the Edwards SAPIEN bioprosthesis (Edwards Lifesciences LLC, Irvine, Calif) relative to the annulus diameter as measured by transesophageal echocardiography (TEE) by 2 to 4 mm, this rule is not strictly applied, particularly in patients with a large annulus or borderline vascular accesses.3 To date, the benefit of a strict ‘‘oversizing’’ policy on the severity of AR has never been investigated. When starting our TAVR program with the Edwards SAPIEN bioprosthesis, we decided to implement a prospective policy of systematic ‘‘oversizing’’ the bioprosthesis relative to the diameter of the aortic annulus and to test its impact on the rate and severity of AR. This issue was studied by means of serial 2-dimensional transthoracic echocardiography (TTE) performed before the procedure, before discharge, and at 1 month.

MATERIALS AND METHODS Patient Population and Selection Patients with severe aortic stenosis considered to be at high or prohibitive surgical risk were considered to undergo TAVR with the Edwards SAPIEN valve prosthesis at University Medical Center, Utrecht, The Netherlands. Consensus was achieved between cardiologists and cardiac surgeons regarding the surgical risk, and all patients provided informed written consent. To determine the initial eligibility for TAVR and to choose between the transfemoral (TF) aortic valve replacement (AVR) or the transapical (TA)-AVR approach, all patients underwent coronary angiography, aortobifemoral angiography, multislice computed tomography, and TTE. Potential candidates for TAVR demonstrated severe symptomatic degenerative aortic valve stenosis with at least New York Heart Association class 2 symptoms and an aortic valve area of smaller than 1.0 cm2 or mean gradient greater than 40 mm Hg or peak gradient greater than 80 mm Hg. All candidates were contraindicated for conventional surgery or considered a high surgical risk with an operative mortality risk of greater than 20% as assessed by at least 2 cardiovascular surgeons and 2 cardiologists.8 Patients were excluded from TAVI if they had at least 1 of the following: (1) aortic valve annulus diameter smaller than 16 mm or greater than 24 mm by TTE; (2) congenital unicuspid or bicuspid valve; (3) noncalcified valve; (4) untreated clinically significant proximal coronary artery disease; (5) severe left ventricular (LV) dysfunction with an ejection fraction of less than 20%; or (6) hemodynamic instability requiring inotropic

71

CHAPTER 5

support. All patients were considered as potential candidates for TF-AVR unless they had at least 1 of the following: (1) mobile aortic arch atheroma greater than 5 mm, or (2) iliofemoral dimensions, morphology, or calcifications that would preclude insertion of a 22 F sheath or 24 F introducer sheath based on a detailed assessment of aortobifemoral angiography and multislice computed tomography, in which case they were considered for TA-AVR. On the basis of this initial selection process, 28 patients were considered potential candidates. All patients underwent TEE before the final planning of the procedure. As part of our ‘‘oversizing’’ policy (see below), those with an annulus diameter greater than 24 mm by TEE were further excluded and did not undergo the procedure.

Choice of Bioprosthesis Size and ‘‘Oversizing’’ Policy The Edwards SAPIEN transcatheter valve is a catheter-delivered heart valve prosthesis that is composed of a stainless-steel balloon-expandable stent, with an integrated trileaflet tissue valve and a polyethylene terephthalate fabric cuff. The valve tissue is made from 3 equal sections of bovine pericardium treated with a proprietary tissue treatment. During the time of the study, the bioprosthesis was available in 2 sizes (23 and 26 mm) and could be delivered through the TF or TA approach. Our ‘‘oversizing’’ policy for the implantation of the Edwards SAPIEN bioprosthesis was defined as follows: (1) All patients with an aortic annulus greater than 24 mm by TEE during the final step of the screening were strictly excluded (see above); (2) a 26-mm prosthesis was implanted in all patients with an annulus greater than 21 mm, and the TA approach was used every time it was needed; and (3) in patients with an annulus 21 mm or less, a 26-mm prosthesis was also implanted every time the balloon predilatation with a 23-mm balloon associated with contrast angiography at maximum balloon inflation demonstrated a significant aortic regurgitation. To appreciate the degree of ‘‘oversizing,’’ the ‘‘prosthesis/annulus cover index’’ expressed as 100 x ([prosthesis diameter – TTE annulus diameter]/ prosthesis diameter) was used, as suggested by Detaint and colleagues.3 Transcatheter Aortic Valve Replacement The prosthetic stented valve was mechanically crimped on a balloon catheter immediately before implantation. The Retroflex delivery system (Edwards Lifesciences) was used for device implantation.6,9,10 TF and TA-AVR were performed under general anesthesia in the catheterization laboratory as previously described.6,9,10 TEE was used for all procedures. Patients were premedicated with aspirin and antibiotics. Heparin was used to maintain an activated clotting time greater than 300 seconds. The activated clotting time was reversed with protamine at the end of the procedure.

For TF-AVR, our technique was similar to that in previously published reports.6,9 A surgical cutdown of the femoral artery was performed, and a 14 F sheath was initially placed. A temporary transvenous electrode was introduced into the right ventricle. Balloon aortic valvuloplasty with a 20- or 23-mm balloon was performed. The balloon-mounted valve was positioned using angiographic techniques and echocardiographic guidance, and subsequently deployed under rapid pacing (180–220 beats/min). Exit peripheral angiography was performed to ensure no

72

PROSPECTIVE OVERSIZING STRATEGY IN BALLOON EXPANDABLE TAVR

Figure 1 | Patient flow chart. TEE, Transesophageal echocardiography; TTE, transthoracic echocardiography.

extravasation of contrast before removal of femoral sheath. Subsequently, the sheath was removed and the access site was closed surgically in the operating room by a surgeon. The TA-AVR procedure was performed as previously described.10 For the purpose of rapid ventricular pacing, 2 unipolar epicardial pacer wires were secured and tested with a high-output epicardial pacing system to ensure ventricular capture at rates of 180 to 220 beats/min. Procedural success was defined as the implantation of a functional prosthetic valve within the aortic annulus at the end of the procedure without inlaboratory mortality. Patients received aspirin (81 mg/d) and clopidogrel (75 mg/d) indefinitely. Warfarin was substituted for clopidogrel in patients with atrial fibrillation.

73

74 17(61) 27±6 21(75) 12(43) 5(18) 16(57) 8(29) 11(39) 7(25) 8(29) 3(11) 6(21) 8(29)

Male, n(%)

BMI (kg/m2)

Hypertension, n(%)

Atrial fibrillation, n(%)

Diabetes, n(%)

Coronary artery disease, n(%)

Prior myocardial infarction, n(%)

Percutaneous coronary intervention, n(%)

Coronary artery bypass, n(%)

Cerebrovascular events, n(%)

Peripheral vascular disease, n(%)

Renal disease, n(%)

COPD, n(%) 0(0) 3(11) 22(78) 3(11)

I

II

III

IV

NYHA Class, n(%)

80±8

Age (year)

Total

1(17)

4(66)

1(17)

0(0)

2(33)

1(17)

1(17)

1(17)

1(17)

2(33)

1(17)

3(50)

1(17)

2(33)

5(83)

28±4

6(100)

80±9

Excluded N=6

Table 1 | Clinical and echocardiography characteristics of the 28 patients

2(9)

18(82)

2(9)

0(0)

6(27)

5(23)

2(9)

7(32)

6(27)

9(41)

7(32)

13(59)

4(18)

10(45)

16(73)

26±5

11(50)

79±7

Included N = 22

0.74

0.77

0.74

0.61

0.45

0.58

0.73

0.45

0.69

0.93

0.59

0.58

0.04

0.02

0.71

P

0(0)

9(90)

1(10)

0(0)

3(30)

2(20)

0(0)

3(30)

2(20)

2(20)

2(20)

3(30)

2(20)

5(50)

7(70)

26±7

4(40)

78±7

Trans femoral

2(17)

9(75)

1(8)

0(0)

3(25)

3(25)

2(17)

4(33)

4(33)

7(58)

5(42)

10(83)

2(17)

5(42)

9(75)

26±4

7(58)

80±8

Trans apical

0.40

0.79

0.78

0.48

0.87

0.64

0.06

0.38

0.02

0.84

0.70

0.80

0.85

0.66

0.61

P

CHAPTER 5

1.33±0.82

3(50)

3(50)

24.9 ± 0.4

22.5 ± 1.1

0.69±0.14

73 ± 21

44 ± 15

47.5±5.3

1(17)

23.2±16.3

1.55±1.01

9(41)

13(59)

22.0 ± 1.3

21.6 ± 1.3

0.73±0.19

70 ± 26

41 ± 16

45.7±5.3

5(23)

21.3±14.1

0.35

0.69

0.01

0.10

0.25

0.65

0.55

0.43

0.74

0.81

1.80±1.40

5(50)

5(50)

22.1 ± 1.3

21.5 ± 1.4

0.73±0.17

70 ± 27

40 ± 16

45.3±6.1

2(20)

18.1±15.5

1.33±0.49

4(33)

8(67)

22.0 ± 1.3

21.6 ± 1.5

0.73±0.21

71 ± 26

41 ± 17

46.2±4.8

3(25)

24.1±12.8

0.29

0.72

0.86

0.91

0.96

0.95

0.88

0.71

0.78

0.33

BMI, Body mass index; COPD, chronic obstructive pulmonary disorder; NYHA, New York Heart Association; euroSCORE, European System for Cardiac Operative Risk Evaluation; LVEF, left ventricular ejection fraction; LV, left ventricular; TTE, transthoracic echocardiography; AR, aortic regurgitation. Data are mean-standard deviation otherwise stated.

1.55±0.95

12(43)

Aortic regurgitation grade(0-4)

16(57)

Significant AR, n(%)

22.6 ± 1.4

Annulus diameter by TEE (mm)

No significant AR, n(%)

21.8 ± 1.2

Annulus diameter by TTE (mm)

71 ± 24 0.72±0.19

Peak aortic gradient (mmHg)

Aortic valve area (cm2)

42 ± 15

Mean aortic gradient (mmHg)

6(21) 46.2±4.5

LVEF <35%, n(%)

LV End-diastolic diameter (mm)

21.8±13.3

Logistic Euroscore

Table 1 Continued

PROSPECTIVE OVERSIZING STRATEGY IN BALLOON EXPANDABLE TAVR

75

CHAPTER 5

Echocardiography Follow-up TEE was the final step of the screening process and was performed during the procedure in all patients. TTE was performed during the screening process, pretreatment (last echocardiography study before TAVR), posttreatment (at discharge), and postdischarge (first outpatient clinic visit) using a Philips 5500 or ie33 system (Philips, Eindhoven, The Netherlands).

Complete echocardiographic studies were performed for each patient in a standard fashion.

Standard parameters were recorded.11 By using the parasternal long-axis view and M-mode or 2-dimensional echocardiography, LV diameters, function, and outflow track diameter were obtained. To assess the severity of aortic stenosis peak aortic velocity, peak instantaneous gradient, mean transaortic gradient, and velocity-time integral were measured. Aortic valve area was estimated using the continuity equation approach.12 The color-flow Doppler signal was used to assess aortic regurgitation. According to the European Society of Cardiology guidelines and American Society of Echocardiography recommendations, AR was graded in 5 groups as none (=0), trivial (=1), mild (=2), moderate (=3), or severe (=4).13,14 AR grade 2 or higher was considered significant. More specifically, regurgitation was assessed by visual inspection and by using color-flow mapping of the regurgitation jet as described by Zoghbi and colleagues14 and Helmcke and colleagues.15 One experienced and independent cardiologist, blinded to the patient clinical status, graded AR. Statistical Analysis Continuous variables were presented as mean Âą standard deviation. Discrete variables were presented as absolute numbers and percentages.

For comparison between categorical variables, a chi-square test was used. A 1-way analysis of variance test was used for comparison between continuous variables. A 1-way repeatedmeasures Friedman test with post hoc analysis was used to evaluate changes over time.

Analyses were performed using SPSS 18.0 (SPSS Inc, Chicago, lll).

RESULTS Characteristics of the Study Population and Procedure The baseline clinical and echocardiography characteristics of the study population are shown in Table 1. Based on the measurement of the annulus diameter by TTE (<24 mm), 28 patients were initially considered eligible for TAVR. On the basis of the subsequent measurement of the annulus diameter by TEE (>24 mm) during the final step of the screening process, 6 of 28 patients (21%) were excluded. According to TEE, the excluded 6 patients had a larger aortic annulus diameter as compared to the included patients (P = 0.01, Table 1). This difference was not detectable by TTE (P = 0.10, Table 1). The 6 excluded patients were male (100%) and had a higher body mass index (P = 0.04) than the remaining 22 patients (Table 1). The medical decision in the excluded 6 patients who would not undergo TAVR with the Edwards SAPIEN valve prosthesis, was as follows: One patient underwent conventional aortic valve replacement and was alive at 12 months follow-up; 1 patient was eligible for the implantation of a 29 mm

76

PROSPECTIVE OVERSIZING STRATEGY IN BALLOON EXPANDABLE TAVR

CoreValve (Medtronic Inc, Minneapolis, Minn) (the procedure was successful, and the patient died at 7 months follow-up); the last 4 patients underwent conservative medical management (2 patients died within 12 months and 2 patients remained alive at 1 year under conservative treatment; 1 of the last 2 patients recently underwent the implantation of a 29 mm Edwards SAPIEN bioprosthesis). Among the 22 remaining patients, 12 underwent TA-AVR and 10 underwent TF-AVR. Patients who underwent TA-AVR rather than TF-AVR were more likely to have a history of coronary artery disease (P = 0.02, Table 1). The other characteristics were not different between the groups. The relatively low mean annulus diameter (22.0 ± 1.3 mm, Tables 1 and 2) of the population in whom the procedure was performed reflects the exclusion of all patients with an annulus greater than 24 mm by TEE as part of our ‘‘oversizing’’ policy.

Furthermore, in 6 of 22 patients the procedure was modified: In 3 patients with an annulus diameter greater than 21 mm and suitable for a 22 F but not for a 24 F iliofemoral sheath, a TA approach was used. In 3 other patients with an annulus diameter of 21 mm or smaller by TTE, a 26 mm rather than a 23 mm prosthesis was implanted based on the results of balloon predilatation. Consequently, the ‘‘prosthesis/annulus cover index’’ was 12.4% ± 4.3%. Using a 23 mm prosthesis in the 6 excluded cases mentioned above would have led to a significantly lower ‘‘cover index’’ (9.9% ± 3.6%, P = 0.02). The procedure was successful in 21 of 22 patients (95%, Table 2). The patient with unsuccessful prosthesis implant was an 85-year-old man with a history of myocardial infarction, bypass surgery, and peripheral vascular disease. The technical failure was related to the inability to perform appropriate hemostasis at the level of the purse

Table 2 | Procedural Characteristics and 30 day outcome Total N=22

Transfemoral N=10

Transapical N=12

P

Annulus diameter (mm), mean ± SD

22.0 ± 1.3

22.1 ± 1.3

22.0 ± 1.3

0.86

Aortic valve area(cm2) , mean ± SD

0.73 ± 0.19

0.73 ± 0.17

0.73 ± 0.21

0.95

23 mm

6(27)

3(30)

3(25)

0.79

26 mm

16(73)

7(70)

9(75)

Prosthesis diameter, n(%)

Cover Index, %, mean ± SD

12.4± 4.3

11.8± 5.5

12.8± 3.2

0.58

Successful valvuloplasty, n(%)

22(100)

10(100)

12(100)

1

Successful valve deployment, n(%)

21(95)

10(100)

11(92)

0.35

0(0)

0(0)

0(0)

1

Myocardial infarction

1(4)

0(0)

1(8)

0.35

Stroke, n (%)

0(0)

0(0)

0(0)

1

Intraprocedural death, n(%) 30 day outcome

Permanent pace maker

1(4)

0(0)

1(8)

0.35

Mortality, n(%)

4(18)

1(10)

3(25)

0.35

SD, Standard deviation. Cover index as 100 X (prosthesis diameter – annulus diameter by TEE/prosthesis diameter).

77

CHAPTER 5

Figure 2 | A, Changes in aortic regurgitation in the 18 patients with echocardiography follow-up (P = .03 by 1-way repeated-measures Friedman test). B, Individual changes in aortic regurgitation from pretreatment to postdischarge. AR, Aortic regurgitation

at the apex of the left ventricle related to pericardial adherences. The procedure was aborted and converted to open surgery for conventional valve replacement. Although the surgical intervention was successful, the patient died of multiorgan failure on day 11. Another patient died of right ventricular failure secondary to an inferior myocardial infarction in hospital on day 3 after TAVR. These 2 in-hospital deaths were observed among the 12 patients who underwent a TA-AVR procedure. No procedural failure and no in-hospital deaths were observed in the 10 patients undergoing a TF-AVR procedure. One patient who underwent TA-AVR required a permanent pacemaker implantation before discharge. No rupture of the aortic annulus was observed. No case of severe AR was observed immediately after balloon valvuloplasty by

78

PROSPECTIVE OVERSIZING STRATEGY IN BALLOON EXPANDABLE TAVR

Table 3 | Echocardiography findings during follow-up (n=18) Pretreatment

Post-treatment

Postdischarge

P

LVEF < 35%, n (%)

3 (23)

2 (15)

3 (17)

.57

TF

1 (11)

1 (11)

1 (11)

TA

2 (22)

1 (11)

2 (25)

LV end-diastolic diameter (mm)

45.7 ± 5.9

45.2 ± 5.5

44.7 ± 5.3

TF

45.0 ± 6.3

44.7 ± 5.5

44.3 ± 4.9

TA

46.4 ± 5.5

45.8 ± 5.6

45.1 ± 5.7

Mean aortic gradient (mm Hg)

42 ± 17

9±4

8±3

TF

41 ± 17

12 ± 4

9±3

TA

44 ± 18

7±2

7±2

Peak aortic gradient (mm Hg)

73 ± 26

17 ± 6

15 ± 6

TF

73 ± 28

21 ± 6

17 ± 8

75 ± 27

13 ± 4

13 ± 4

Aortic valve area (cm )

0.72 ± 0.21

1.87 ± 0.45

1.81 ± 0.31

TF

0.73 ± 0.18

1.87 ± 0.50

1.86 ± 0.34

TA

0.71 ± 0.24

1.88 ± 0.43

1.75 ± 0.30

Aortic regurgitation grade (0–4)

1.56 ± 1.10

1.17 ± 0.71

0.89 ± 0.83

TF

1.78 ± 1.48

1.22 ± 0.67

1.11 ± 0.78

TA

1.33 ± 0.50

1.11 ± 0.78

0.67 ± 0.87

TA 2

.03

.0001

.0001

.0001

.03

All parameters are mean ± standard deviation, otherwise stated. P reflects the difference among the 3 time points by 1-way repeated measures Friedman test. LVEF, Left ventricular ejection fraction; TF, transfemoral; TA, transapical; LV, left ventricular. *P < .05 compared with “pretreatment” by post hoc analysis of the Friedman test

TEE. After implantation of the bioprosthesis, fluoroscopy did not show any distortion of the edges of the stent supporting the valve. Likewise, TEE performed in short axis showed that the circularity of the stent was preserved. Overall, 20 patients (91%) underwent a successful procedure and were discharged alive from the hospital. Within the first 30 days, 2 additional patients died of pulmonary infection in the TA-AVR group (n=1) and sudden death (n=1) in the TF-AVR group. No stroke was observed. Eighteen patients were available for echocardiography after discharge (Figure 1).

Echocardiography Analysis The echocardiography findings of the 18 patients with an echocardiography at 1 month are summarized in Table 3. There was a statistically significant decrease in the peak (pretreatment 73 ± 26 mm Hg vs post-treatment 17 ± 6 mm Hg vs 1 month 15 ± 6 mm Hg, P<.0001) and mean gradient (pretreatment 42 ± 16 mm Hg vs post-treatment 9 ± 4 mm Hg vs 1 month 8 ± 3 mm Hg, P<.0001) after Edwards SAPIEN prosthesis implantation, which was associated with an increase in the aortic valve area (pretreatment 0.72 ± 0.21 cm2 vs post-treatment

79

CHAPTER 5

AR during balloon inflation with a 23–mm balloon

Edwards SAPIEN* diameter (mm)

AR before treatment

AR at 1 mo

TF

No

23

Mild

None/trivial

TA

No

26

None/trivial

None/trivial

TA

Yes

26

None/trivial

None/trivial

Yes

TF

Yes

26

None/trivial

Mild

20.3

No

TA

No

23

None/trivial

NA

23.5

Yes

TF

Yes

26

None/trivial

Mild

23.8

No

TA

Yes

26

None/trivial

Mild

20.8

Yes

TF

Yes

26

Mild

None/trivial

20.5

20.8

No

TA

No

23

None/trivial

Mild

22.5

22.9

No

TA

No

26

None/trivial

None/trivial

Male†

20.1

20.7

Yes

TF

Yes

26

Moderate

None/trivial

Male

22.8

23.6

Yes

TF

Yes

26

Moderate

None/trivial

Female

20.4

21.0

22F only

TF

No

23

Mild

NA

Female†

20.9

21.4

22F only

TA

No

26

Severe

None/trivial

Female†

20.2

20.6

No

TA

Yes

26

Moderate

None/trivial

Male

23.2

23.7

No

TA

Yes

26

None/trivial

None/trivial

Female

19.0

19.9

No

TA

No

23

None/trivial

None/trivial

Male

21.9

22.5

No

TA

No

26

None/trivial

NA

Male

22.7

23.2

No

TA

No

26

Mild

NA

Female

20.3

20.9

22F only

TF

No

23

Mild

None/trivial

Male

23.0

23.4

Yes

TF

Yes

26

None/trivial

Mild

Female

21.7

22.9

Yes

TF

No

26

None/trivial

None/trivial

TTE

TEE

Female

19.9

20.6

Yes

Male†

21.5

21.9

22F only

Female†

22.9

23.5

22F only

Female

22.8

23.5

Female

19.8

Male

22.9

Male

23.5

Female†

20.2

Male Male

Gender

Femoral approach possible

Annulus diameter (mm)

Approach used for delivery

Table 4 | Procedural approach for each patient and impact on aortic regurgitation

TTE, Transthoracic echocardiography; TEE, transesophageal echocardiography; AR, aortic regurgitation; TF, ransfemoral; TA, transapical; NA, not available. * Edwards Lifesciences LLC, Irvine Calif. † Patients in whom the procedure was modified

80

PROSPECTIVE OVERSIZING STRATEGY IN BALLOON EXPANDABLE TAVR

1.87 ± 0.45 cm2 vs 1 month 1.81 ± .32 cm2, P<.0001). Ventricular function did not change significantly during follow-up. No difference was observed between patients treated through the TF or TA approach. Before treatment, a significant AR was observed in 9 of 22 patients (41%, Table 1). AR was mild, moderate, and severe in 5 patients (23%), 3 patients (13%), and 1 patient (5%), respectively. The degree of AR before and 1 month after Edwards SAPIEN valve implantation in the 18 patients with echocardiography follow-up is shown in Table 3 and Figure 2, A. Although a moderate to severe AR was observed in 4 of 18 patients (22%) before treatment, it was no longer the case at 1 month (P = 0.03, Figure 2, A). The improvement was secondary to a disappearance of the AR in all 7 patients with a significant AR before treatment, whereas the new ARs appearing in 5 of the 11 patients with no AR at pretreatment were only mild (Table 4; Figure 2, B).

DISCUSSION The present study is the first to show that in patients referred for TAVI for aortic stenosis, a simple prospective strategy of ‘‘oversizing’’ the implantation of the Edwards SAPIEN bioprosthesis may prevent the occurrence of moderate and severe ARs. Impact of ‘‘Oversizing’’ the Edwards SAPIEN Bioprosthesis on Aortic Regurgitation The key finding of this study is that the use of a simple ‘‘oversizing’’ policy was associated with no occurrence of moderate or severe ARs at 1 month with no case of rupture of the aortic annulus and no prosthesis–patient mismatch. This benefit was the result of 3 simple measures, including (1) exclusion of all patients with an aortic annulus greater than 24 mm by TEE during the screening process, (2) use of a 26-mm prosthesis in all patients with aortic annulus greater than 21, and (3) use of a 26 mm prosthesis in patients with an aortic annulus 21 or less whenever sizing based on balloon valvuloplasty with a 23 mm balloon and contrast angiography suggested that it was feasible. Our results are consistent with a recent retrospective analysis of determinants of AR after Edwards SAPIEN bioprosthesis implantation by Detaint and colleagues.3 They observed that oversizing the prosthesis, as estimated by the ‘‘prosthesis/annulus cover index,’’ was the major factor associated with the absence of severe AR after implantation.

Of note, in our population the cover index was twice as high as in the population studied

by Detaint and colleagues (12.4 ± 4.3 vs 6.7 ± 4). Detaint and colleagues also found that a moderate to severe AR was never observed in patients with a ‘‘cover index’’ greater than 8 or an aortic annulus less than 22 mm. Of note, this combination was present in 20 of 22 patients (91%) in our population but in only half of the population studied by Detaint and colleagues.3 Analysis of the respective role of the criteria used in the selection process demonstrates that the exclusion of patients with an annulus greater than 24 mm would have already resulted in a high cover index of 9.9% ± 3.6%. The use of the 2 additional criteria further increased the cover index of our population to 12.4% ± 4.3%. The exclusion of patients with an aortic annulus greater than 24 mm is already recommended as a common practice. However, this rule is not strictly applied, and in a recently published series, Detaint and colleagues3 reported that an

81

CHAPTER 5

Edwards SAPIEN was implanted in patients with an annulus greater than 24 mm in more than one third of the cases.

In some cases, this is an unanticipated finding of the TEE performed at the time of the intervention in a patient in whom no TEE was performed during the screening. In this context, our observation that 6 of 28 patients (21%) were finally excluded on the basis of the TEE, whereas they were initially considered good candidates for an Edwards SAPIEN prosthesis implantation based on TTE, advocates the systematic use of TEE as part of the screening process. This also strengthens the need for larger aortic valve prostheses. The recent availability of the 29 mm Edwards SAPIEN valve prosthesis will allow this technology to be offered to patients with an aortic annulus greater than 24 mm without decreasing the ‘‘cover index’’ and taking the risk to induce a moderate or severe AR. In addition, this approach was not associated with any stent deformation or prosthesis–patient mismatch as shown by the low transvalvular gradient (8 ± 3 mmHg). This observation combined with the absence of aortic rupture in our series suggests that despite the presence of calcification, the aortic annulus keeps some of its ability to be stretched. Our study also suggests that it might be important to switch from the TF to the TA approach in case an appropriately sized valve prosthesis cannot be delivered. Because the new device generation has been downsized, switching to the TA approach to accommodate the 26 mm bioprosthesis will be less frequent. This decision will remain critical in patients with an annulus diameter of approximately 24 mm in whom the choice to implant a 29 mm bioprosthesis will imply the switch from the TF to the TA approach.

In our study, the increment of the intended size of the bioprosthesis after predilation combined with aortic angiography in 3 patients strengthens the importance of this step as the final critical step to size the annulus and to choose the appropriate diameter of the bioprosthesis.

Although the strategy applied in our population was associated with no occurrence of moderate to severe AR, it did not prevent the occurrence of mild AR in 5 of 11 patients without AR at pretreatment. Further improvements in bioprosthesis design or implantation strategy are still needed to avoid this complication. In this series, no event associated with the oversizing strategy was observed. In particular, no case of annulus rupture or stroke was observed. The rates of myocardial infarction (4%) and heart block requiring pacemaker (4%) are on par with those described in previous series and in the recently published PARTNER trial.16 Fluoroscopy and TEE performed at the end of the procedure did not show distortion of the struts of the stent. Likewise, TTE performed before discharge or at 1-month follow-up did not show high transvalvular gradient suggestive of prosthesis–patient mismatch. Of note, in this series, a benefit on LV dimensions was observed as early as 1 month after the procedure.

Although it takes several months to achieve near complete LV remodeling after conventional aortic surgery for aortic stenosis,17 series including early echocardiography follow-up after aortic replacement for aortic stenosis or aortic regurgitation demonstrated changes in LV dimensions within days of surgery.18,19 The results of our series in which no case of moderate to severe aortic regurgitation was observed after TAVR are consistent with those previous observations. In addition, the disappearance of moderate to severe aortic regurgitation that was present at pretreatment in approximately 25% of our patient population may have played a role.

82

PROSPECTIVE OVERSIZING STRATEGY IN BALLOON EXPANDABLE TAVR

Study Limitations This was a single-center study, and patient referral and medical management may have influenced the results. However, the prospectively designed nature of our policy of ‘‘oversizing’’ in the implantation of the Edwards SAPIEN bioprosthesis, the consecutive nature of the study population, and the individual patient analysis of the serial echocardiography followup provide useful insights into the benefit of TAVR with the Edwards SAPIEN valve on aortic regurgitation. The lack of information on wall thickness and myocardial mass and the lack of a longer echocardiography follow-up may limit the interpretation of the benefit of the strategy on LV remodeling. Finally, although no event related to the oversizing strategy was observed, we must acknowledge that the number of patients included in the present study is too small to draw a definite conclusion on the safety of this approach.

CONCLUSIONS On the basis of an individual and serial patient analysis, the present report provides important information on the outcome of AR after implantation of an Edwards SAPIEN valve in the aortic position. In particular, the results suggest that a simple ‘‘oversizing’’ policy may prevent the occurrence of moderate and severe ARs. Although these findings need confirmation in studies including more patients and with a longer follow-up, they support the use of larger valve prostheses (29 mm) to adequately treat patients with an annulus greater than 24 mm.

83

CHAPTER 5

REFERENCES 1. Sun JC, Ghanta RK, Davidson MJ. Highlights from the transcatheter cardiovascular therapeutics conference 2010: Washington, DC, September 21-25, 2010. J Thorac Cardiovasc Surg. 2011;142:468-71.

2. Higgins J, Ye J, Humphries KH, Cheung A,Wood DA,Webb JG, et al. Early clinical outcomes after transapical aortic valve implantation: a propensity-matched comparison with conventional aortic valve replacement. J Thorac Cardiovasc Surg. 2011;142:e47-52.

3. Detaint D, Lepage L, Himbert D, Brochet E, Messika-Zeitoun D, Iung B, et al.Determinants of significant paravalvular regurgitation after transcatheter aortic valve: Implantation impact of device and annulus discongruence. JACC Cardiovasc Interv. 2009;2:821-7.

4. Moss RR, Ivens E, Pasupati S, Humphries K, Thompson CR, Munt B, et al. Role of echocardiography in percutaneous aortic valve implantation. JACC Cardiovasc Imaging. 2008;1:15-24.

5. Takagi K, Latib A, Al-Lamee R, Mussardo M, Montorfano M, Maisano F, et al.Predictors of moderate-tosevere paravalvular aortic regurgitation immediately after CoreValve implantation and the impact of postdilatation. Catheter Cardiovasc Interv. 2011;78:432-43.

6. Webb JG, Pasupati S, Humphries K, Thompson C, Altwegg L, Moss R, et al. Percutaneous transarterial aortic valve replacement in selected high-risk patients with aortic stenosis. Circulation. 2007;116:755-63.

7. Descoutures F, Himbert D, Lepage L, Iung B, Detaint D, Tchetche D, et al. Contemporary surgical or percutaneous management of severe aortic stenosis in the elderly. Eur Heart J. 2008;29:1410-7.

8. Vahanian A, Alfieri O, Al-Attar N, Antunes M, Bax J, Cormier B, et al. Transcatheter valve implantation for patients with aortic stenosis: a position statement from the European Association of Cardio-Thoracic Surgery (EACTS) and the European Society of Cardiology (ESC), in collaboration with the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur Heart J. 2008;29:1463-70.

9. Cribier A, Eltchaninoff H, Tron C, Bauer F, Agatiello C, Nercolini D, 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;47:1214-23.

10. Walther T, Simon P, Dewey T,Wimmer-Greinecker G, Falk V, Kasimir MT, et al. Transapical minimally invasive aortic valve implantation: multicenter experience. Circulation. 2007;116:I240-5.

11. Tzikas A, Piazza N, van Dalen BM, Schultz C, GeleijnseML, van Geuns RJ, et al. Changes in mitral regurgitation after transcatheter aortic valve implantation. Catheter Cardiovasc Interv. 2009;75:43-9.

12. Oh JK, Taliercio CP, Holmes DR Jr, Reeder GS, Bailey KR, Seward JB, et al. Prediction of the severity of aortic stenosis by Doppler aortic valve area determination: prospective Doppler-catheterization correlation in 100 patients. J Am Coll Cardiol. 1988;11:1227-34.

13. Vahanian A, Baumgartner H, Bax J, Butchart E, Dion R, Filippatos G, et al. Guidelines on the management of valvular heart disease: the task force on the management of valvular heart disease of the European Society of Cardiology. Eur Heart J. 2007;28:230-68.

14. Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA, Kraft CD, Levine RA, et al. Recommendations for evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr. 2003;16:777-802.

15. Helmcke F, Nanda NC, Hsiung MC, Soto B, Adey CK, Goyal RG, et al. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation. 1987;75:175-83.

16. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010;363:1597-607.

17. Yarbrough WM, Mukherjee R, Ikonomidis JS, Zile MR, Spinale FG. Myocardial remodeling with aortic stenosis and after aortic valve replacement: mechanisms and future prognostic implications. J Thorac Cardiovasc Surg. 2011 Jul 13 [Epub ahead of print].

18. Petrov G, Regitz-Zagrosek V, Lehmkuhl E, Krabatsch T, Dunkel A, Dandel M, et al. Regression of myocardial hypertrophy after aortic valve replacement: faster in women? Circulation. 2010;122:S23-8.

19. Carroll JD, Gaasch WH, Zile MR, Levine HJ. Serial changes in left ventricular function after correction of chronic aortic regurgitation. Dependence on early changes in preload and subsequent regression of hypertrophy. Am J Cardiol. 1983;51:476-82.

84

85

CHAPTER

6 IMPACT OF AORTIC VALVE CALCIFICATION QUANTIFIED BY COMPUTED TOMOGRAPHY ON PROCEDURAL AND CLINICAL OUTCOMES OF TRANSCATHETER AORTIC VALVE REPLACEMENT In preparation

Mariam Samim1 Pieter Stella1 Ricardo Budde2 Catherina H.Z Li1 Freek Nijhoff1 Pieter A Doevendans1 Pierfrancesco Agostoni1 Department of Cardiology, 2Department of Radiology,

1

University Medical Center Utrecht, the Netherlands

CHAPTER 6

ABSTRACT Background Aortic valve calcification (AVC) volume and distribution may have an impact on transcatheter aortic valve replacement (TAVR) procedure and its outcome. This study sought to investigate the association between the severity of AVC and the outcome of patients undergoing TAVR. Methods Consecutive patients undergoing TAVR with the Edwards SAPIEN or Medtronic CoreValve system, between 2008 and 2013, were included in this study. Pre-procedurally, all patients underwent contrast enhanced multislice computed tomography. Aortic annulus dimensions were assessed and AVC and its distribution were quantified. To standardise the amount of AVC, a valve calcification index (VCI) was calculated: [AVC] / annulus perimeter. Procedural complications (including residual aortic regurgitation (AR)), "extra manoeuvres" to optimize results (such as post-dilatation and second valve implantation), 30-days clinical outcomes (composite of mortality, cerebrovascular events, myocardial infarction, pacemaker implantation, and acute kidney injury), and mid-term mortality were documented for all patients.

Results Overall, 263 patients were included (age 80.3 ± 7.3 years). Baseline VCI, but not asymmetrical distribution of AVC, was a significant independent predictor of post-procedural AR severity (odds ratio per 10 ul/mm VCI 10.27 [95% confidence interval: 10.05-10.49], p=0.015). Patients with the need for an “extra manoeuvre” to optimize the TAVR result showed significantly higher VCI (n=49, VCI=27.21±13.48 μl/mm) than patients without the need for additional measures (n=214, VCI=21.91±11.19 μl/mm, p= 0.007). Despite its effect on procedural results, baseline VCI had no significant impact on the composite 30-days clinical outcomes (p=0.594) or midterm mortality.

Conclusion The extent of AVC predicts the occurrence of post-procedural aortic regurgitation and consequently the need for additional manoeuvres to optimize TAVR. On the other hand, AVC is not associated with short-term clinical events and mid-tem mortality.

88

IMPACT OF AORTIC VALVE CALCIFICATION ON TAVR OUTCOME

INTRODUCTION Transcatheter aortic valve replacement (TAVR) is an accepted treatment alternative to surgery for inoperable and high risk patients with severe symptomatic aortic valve stenosis. As the indication for TAVR is broadening toward lower risk patients, attention has been shifted from feasibility of this technique to reduction of significant complications related to this treatment.

One clinically relevant complication is post-procedural aortic regurgitation (AR). It seems that even mild AR may be related to increased late mortality after TAVR1.

Suboptimal placement of the prosthesis with incomplete sealing of the annulus by the skirt, an incomplete apposition of the stent frame in the device landing zone, and undersizing of the valve prosthesis relative to the dimensions of the aortic annulus, are all potential mechanisms of post-TAVR AR. In contrast to surgical aortic valve replacement, during TAVR the diseased native aortic valve remains in situ, which might lead to insufficient prosthesis expansion or to significant AR due to extensive amount of calcium.

Severe aortic valve calcification is a common finding among patients undergoing TAVR. The presence of a significant amount of aortic valve calcification (AVC) may prevent perfect apposition between the prosthesis and the aortic annulus, which might result in paravalvular AR. Furthermore, in patients with heavily calcified valves, non-circular deployment of the prosthesis has been reported, a phenomenon that also increases the risk of post-TAVR AR2.

In addition, a higher degree of AVC may also increase the risk of dislodgement of calcific material during prosthesis implantation, which might lead to post-procedural cerebrovascular incidents, coronary artery occlusion, acute kidney injury (AKI), and eventually death. Previous studies reported contradictory results in regard to the impact of AVC on TAVR outcome, one showing a significant association between the amount of AVC at baseline and post-TAVR mortality (at 30 days, 2 years and 3 years) and the second reporting no such association3,4.

Contrast-enhanced multislice computed tomography (MSCT), which may be part of the preprocedural screening of patients undergoing TAVR, provides excellent visualization of AVC and its precise location on aortic valve cusps. Using contrast-enhanced MSCT for detailed quantitative calcium measurement, we aimed to evaluate in a large cohort of TAVR patients the association between AVC and: I) the degree of post-TAVR AR, II) procedural outcomes (need for “extra manoeuvres” to optimize TAVR result) and III) clinical outcomes.

METHODS Study population and data collection Patients with severe symptomatic aortic stenosis scheduled for TAVR were included in our institutional prospective registry. The TAVR procedures were performed with two currently commercially available bioprostheses, the Edwards SAPIEN (SAPIEN or SAPIEN XT™) valve prosthesis (Edwards Lifesciences, Irvine, CA, USA) and the Medtronic CoreValve system (CoreValve Revalving Technology, Medtronic, Minneapolis, MN, USA). Patient data were collected prospectively using a dedicated database including baseline clinical, laboratory,

89

CHAPTER 6

echocardiographic and MSCT parameters, as well as procedural and clinical follow-up data.

For this study, data from patients treated between August 2008 and November 2013 were analysed retrospectively. The study was conformed to the guiding principles of the Decleration of Helsinki and patients consented to clinical evaluation.

Multislice computed tomography Pre-procedurally, all patients underwent a contrast enhanced retrospectively ECG-gated MSCT scan on either a 64- or 256-slice scanner (Brilliance 64 or iCT, respectively, Philips Medical Systems, Best, the Netherlands), according to standard scan protocols that were individually adjusted based on patient body habitus. Tube voltage was 100 or 120 kV, tube current 200400 mAs, collimation 64 or 128 x 0.625 mm and gantry rotation time of 270-420 ms. The scan range was set from the level of the subclavian arteries to the level of the head of the femur. A continuous ECG trace was recorded during image acquisition and images were reconstructed at each 12.5% of the R-R interval, obtaining a total of 8 datasets per scan (including 37.5% for systole and 75% for diastole). All scans were performed during mid-inspiratory breath-hold, and during injection of iodinated non-ionic contrast agent (Ultravist iopromide - 300 mg/mL, Bayer Schering Pharma AG, Berlin, Germany, Healthcare Tarrytown, New York). Beta-blockers were not routinely administered prior to scanning. Data sets were reconstructed and off-line post-processing of MSCT images was performed on a dedicated workstation. The images were post-processed and analysed by the software application 3Mensio Valves™ (3mensio Valves™, 3mensio Medical Imaging BV, The Netherlands, http://www.3mensio.com). After segmentation of the ascending aorta (Figure 1A), the aortic annulus was rendered by determination of the anchor points of the three aortic leaflets (marking the aortic annulus) using multiplanar reformation planes followed by a computer-guided reconstruction of the resulting annular transverse plane. Subsequently, the area, the perimeter and the maximum and minimum diameters of the aortic valve annulus were measured on this annular transverse plane (Figure 1B). For every patient an oversizing index and an eccentricity index were calculated as described in previous reports5,6.

Relative oversizing by area [%] = [(nominal prosthesis area/cross-sectional area) – 1] x 100. It has been suggested that there should be a minimum of 10% prosthesis oversizing of the aortic annulus dimensions, to help reduce significant AR after TAVR. Eccentricity index= 1 – (sagittal diameter/coronal diameter). Annulus was considered ellipsoid when the eccentricity index was ≥ 0.1.

Semi-automatic analysis of MSCT images for assessment of AVC volume Images of the aortic root reconstructed in systole, at 37.5% of the R-R interval were selected for analysis of the aortic plane and measurements of aortic valve calcification using 3Mensio software. Next, aortic leaflet calcification was measured in a region starting from the aortic annulus to 20 mm above that, marking the most cranial aspect of the aortic leaflets.

Quantitative measurement of AVC on contrast-enhanced MSCT is challenging due to different contrast levels used among individual patients. In this study we propose a method for quantitation of aortic valve calcification on contrast enhanced MSCT, with consideration of different contrast levels across the patient population. Because of the inter-patient difference

90

IMPACT OF AORTIC VALVE CALCIFICATION ON TAVR OUTCOME

Figure 1 | A, Automated segmentation of aortic root with centre line using the dedicated 3Mensio Valve software. B, The aortic annulus level is defined manually by determination of the anchor points of the three aortic leaflets. C, The mean aortic attenuation value (HU Aorta) and standard deviation is measured in the ascending aorta 25 mm above the level of the aortic annulus to calculate the threshold for calcium detection (HU aorta + 2 SD). D, The software detects automatically area of calcium according to the predetermined threshold set by the investigator. E, The aortic valve is automatically divided into 3 cusps and manual adjustment of the divisions between the cusps is possible.

in the amount of contrast used for contrast-enhanced MSCT, the threshold for detecting calcification in the aortic valve apparatus was chosen on an individual basis, using the method described by Mylonas and colleagues for quantification of coronary artery calcification: for each patient, using axial images, a region of interest was placed in the ascending aorta 25 mm above the level of the aortic annulus7. The mean aortic attenuation value (Hounsfield units (HU) aorta)

91

CHAPTER 6

and standard deviation (SD) was measured at this level (Figure 1C). Using these measures, the threshold for calcium detection was calculated as 2 SD above the mean attenuation in the aorta (HU aorta + 2 SD). The calculated threshold among our study population was 484.47 ± 130.00 HU. Areas of aortic valve calcium were highlighted and surrounding areas of calcium were excluded to generate a composite score for the aortic valve (Figure 1 D). As previously reported, quantification of AVC using the Agatston score requires a non-linear weighting factor in its derivation and thus has been shown to exhibit greater variability than the volumetric quantification of AVC8,9. Therefore, we did not measure AVC severity using the Agatston score, but quantified it in cubic millimetres.

To standardise the volume of AVC per millimetre of valvular annulus perimeter, we calculated a valve calcification index (VCI) as follow: [AVC] / annulus perimeter. For instance, an AVC volume of 1200 μl and a perimeter of 80 mm results in a VCI of 15 μl/mm. Furthermore, in order to evaluate the impact of an unequal distribution of calcium across the three aortic valve cusps on the occurrence of post-procedural AR, we calculated an arbitrarily defined symmetry index for every patient. Because we expected that the difference in cubic millimetres between the cusp with the maximum amount of calcium (cMax) and the cusp with the minimum amount of calcium (cMin) will have the greatest impact on the risk of post-procedural AR, the calcium symmetry index (CSI) was calculated as follow: [cMax - cMin] / [cMax + cMin]. The CSI will be between 0 and 1. The smaller the CSI is, the more symmetrical is the AVC distribution. For example, when the most severely calcified cusp has 900 μl calcium and the least calcified cusp 200 μl, that will result in: [900 – 200] / [900 + 200] = 0.64 All MSCT images were analysed by a procedure-independent operator, who was trained to use the dedicated 3-mensio software and who was blinded for the outcome and all other patient data.

Aortic valve regurgitation after prosthesis implantation Aortic regurgitation was assessed during the procedure by supra-aortic angiograms immediately after device deployment. Interventional options in case of unsatisfactory result included postdilatation and/or implantation of a second prosthesis. Application of these “secondary manoeuvres” was at operator’s discretion.

Post-procedural evaluation of the prosthesis function and the presence and severity of AR was performed using transthoracic echocardiography (TTE) examination at day 5 (+1) after TAVR. The AR was graded according to VARC-2 definitions as 0 (no or minimal regurgitation), 1 (mild), 2 (moderate), and 3 (severe).

Endpoints and definitions In the present study, clinical endpoints including AKI, transient ischemic attack (TIA), and stroke are defined according to the VARC-2 document10. In order to evaluate the possible impact of AVC on short-term outcome, a 30-day safety endpoint was defined as a composite of allcause mortality, cerebrovascular events (TIA or stroke according to VARC-2), acute myocardial infarction, new pacemaker implantation and post-procedural AKI grade II or III.

92

IMPACT OF AORTIC VALVE CALCIFICATION ON TAVR OUTCOME

Statistical analysis Due to the poor quality of MSCT images, in 28 (10.6 %) patients the measurements of aortic annulus dimensions and AVC were not available. In order to reduce the risk of bias due to missing data, multiple imputation was implemented using the Statistical Package for Social Sciences version 20.0 (SPSS, Chicago, Illinois).

Continuous variables are expressed as mean ± standard deviation and categorical variables as proportions. Continuous variables were analysed with Student t test or Mann-Whitney U test and the one-way ANOVA with Tukey’s post hoc test. Categorical variables were compared with the χ2 test. Survival at one year follow-up was assessed using Kaplan-Meier analysis. Adjusted survival analysis using semi parametric Cox proportional hazard modelling was utilized to evaluate the impact of AVC on survival, taking also into account the influence of other possible predictors reported to be associated with post-TAVR long-term mortality by previous studies: Logistic EuroScore and discharge AR, stroke or TIA and AKI. To test whether VCI is an independent predictor of post-TAVR AR, generalized linear models (GLM) for an ordinal outcome was used. Receiver operating characteristic curves were subsequently generated to evaluate the predictive value of VCI for different degrees of postprocedural AR. All statistical tests were 2-tailed, with a type I error rate of 0.05. All statistical analysis was carried out using SPSS version 20.0.

RESULTS Two hundred and sixty-three patients were included in this study. Population and procedural characteristics are shown in Table 1. Device success according to VARC-2 was achieved in 227 patients (86.3%) of cases. Prosthesis malpositioning according to VARC-2 criteria occurred in 8 (3.0%) patients. Prosthesis embolization to the left ventricle, requiring urgent surgery, occurred in two patients receiving a 26 mm Edwards SAPIEN prosthesis (one transfemoral and one transapical case). Of these patients, one underwent a successful surgical aortic valve replacement after removal of the transcatheter aortic valve prosthesis. In the second patient, the transcatheter valve prosthesis was removed during surgery, however due to his poor condition and because the aortic valve area was sufficiently increased by balloon dilatation during TAVR, no surgical replacement of the aortic valve was performed. This patient died on day 11 after TAVR due to multiple organ failure. In 4 other patients undergoing transfemoral Medtronic CoreValve implantation, the prosthesis moved towards ascending aorta during prosthesis deployment, necessitating valve-in-valve secondary prosthesis implantation which resulted in satisfactory prosthesis function in all case. Valve migration was observed in two patients after correct (transfemoral) positioning of a Medtronic CoreValve, moving upwards in one case requiring a secondary prosthesis implantation and moving downwards in the second case, which was solved by snaring and pulling the prosthesis upwards to it correct position.

Intra-procedural death occurred in one patient (0.4%) due to vascular complication during transapical implantation of a 26 mm Edwards SAPIEN prosthesis. In 49 (18.6%) patients, additional manoeuvres after prosthesis deployment were necessary due to either prosthesis

93

CHAPTER 6

malposition or more-than-mild aortic regurgitation. Forty-three (16.3%) of these patients underwent post-dilatation of the valve prosthesis. Additional 11 (4.2%) patients underwent an implantation of a second (9 patients) and a third (2 patients) valve prosthesis. Thirty-day survival rate was 92.4% (243 of 263 patients).

Table 1 | Characteristics of patients and TAVR procedures

Aortic valve annulus dimensions

Patient characteristics and medical history

and valvular cusp calcification

N=263

Age (years)

80.3 ± 7.3

Aortic annulus had an ellipsoid

Female sex

143 (54.4)

shape in 240 patients (91.3%, ec-

Diabetes mellitus

87 (33.1)

centricity index: 0.21±0.06), with

Hypertension

154 (58.6)

Hyperlipidaemia

79 (30.0)

Previous acute myocardial infarction

54 (20.5)

Previous PCI

107 (40.7)

Previous CABG

54 (20.5)

Atrial fibrillation

88 (33.5)

Peripheral vascular Disease

58 (22.1)

Stroke or TIA

49 (18.6)

Left ventricular function

a sagittal diameter of 22.11±2.43 mm and a coronal diameter of 28.12±2.83 mm. The remaining 23 patients presented a round shape (eccentricity index: 0.03±0.10, sagittal diameter 25.77±2.77, coronal diameter 26.77±2.56). According to the aortic annulus area measurements using MSCT, prosthesis sizing was at least 10% oversized

Good

129 (49.0)

Impaired

77 (29.3)

oversized in the remaining. No

Moderate

40 (15.2)

difference was observed in the

in 149 (56.7%) patients and less

Poor

17 (6.5)

occurrence of ≥ grade 2 AR after

Cancer

62 (23.6)

TAVR among patients with rela-

eGFR < 60 ml/min/1.73m2

149 (56.7)

tive prosthesis oversizing < 10% as

Logistic EuroSCORE (%)

16.1 ± 9.3

compared to patients with ≥ 10%

Procedural characteristics

oversizing (9.0% vs 8.0%, p=0.657,

TAVR approach

Figure 2).

Transfemoral

206 (78.3)

Transapical

44 (16.7)

Transaortic

12 (4.6)

Trans-subcliavian

1 (0.4)

TAVR prosthesis type Edward SAPIEN

186 (70.7)

Medtronic CoreValve

77 (29.3)

Predilatation

256 (97.4)

Postdilatation

43 (16.3)

Data are shown as mean ± SD or n (%). CABG: coronary artery bypass surgery; eGFR: estimated glomerular filtration rate (ml/min/m2); PCI: Percutaneous Coronary Intervention; TAVR: transcatheter aortic valve replacement

94

Concerning

AVC,

total

calcium

volume was significantly higher in the non-coronary cusp (700.5 ul) as compared to the right coronary (598.2 ul) and left coronary (570.1 ul) cusps (p=0.029 and p=0.004, respectively).

Impact of aortic valve calcification on post-TAVR aortic valve regurgitation Overall, 121 (46.0%) patients had grade 0, 119 (45.2%) had grade

IMPACT OF AORTIC VALVE CALCIFICATION ON TAVR OUTCOME

Figure 2 | Occurrence of post-TAVR AR according to the extent of prosthesis oversizing. No difference was observed in the occurrence of significant AR after TAVR among patients with relative prosthesis oversizing < 10% as compared to patients with ≥ 10% oversizing

1 and 23 (8.7%) had ≥ grade 2 post-TAVR AR at discharge transthoracic echocardiography.

Eleven patients (4.2%) were found to have isolated central aortic regurgitation post-TAVR. No difference was observed in the occurrence of AR between patients undergoing Edwards SAPIEN prosthesis or a Medtronic CoreValve system (Figure 3). The mean volume of VCI in patients without AR was 20.8±12.3 ul/mm, 24.9±11.8 ul/mm with grade 1 AR and 24.7±16.9 ul/mm with AR grade ≥ 2 (p=0.028) (Table 2A). At univariate analysis, VCI appeared to be a significant predictor of post-procedural AR severity (odds ratio (OR; per 10 ul/mm VCI) 10.27 [95% confidence interval (CI): 10.05-10.49, p=0.015). Calcification symmetry index on the other hand had no predictive value for post-TAVR AR severity (OR 0.92 [95% CI: 0.63-1.34,], p=0.673).

Subsequently in the multivariate analysis using GLM, including annulus eccentricity index and relative prosthesis oversizing as other possible predictors of post-TAVR AR, VCI remained the only independent predictor of AR severity (Table 2B). Furthermore, increasing VCI was a stronger predictor for AR grade 1 (AUC 0.608, p=0.004) than AR grade 2 (r=0.505, p=0.950).

95

CHAPTER 6

Figure 3 | Severity of AR among patients undergoing an Edwards SAPIEN prosthesis versus A Medtronic CoreValve system.

Table 2A | Univariate analysis of baseline aortic valve characteristics with respect the severity of postprocedural aortic valve regurgitation Variable

None (n=121)

Grade 1 (n=119)

Grade 2 (n=23)

p

Total calcification volume (ul)

1676.0±1003.1

2052.5±998.5

2038.5±1474.3

0.012

20.8±12.3

24.9±11.8

24.7±16.9

0.028

Calcification left coronary cusp

542.0±574.6

596.3±445.4

617.8±609.0

0.530

Calcification right coronary cusp

523.0±354.2

667.5±392.3

673.5±509.2

0.008

Calcification noncoronary cusp

611.0±424.9

788.8±428.2

747.2±701.4

0.007

0.3864±0.20878

0.3521±0.206041

0.4170±0.253983

0.249

VCI (ul/mm)

Asymmetry index

Table 2B | Multivariate analysis on predictors of AR severity after TAVR OR [95.0% CI]

P

Native aortic annulus eccentricity index

1.797 [0.596-5.426]

0.143

Relative prosthesis oversizing

0.990 [0.997-1.004]

0.196

VCI (ul/mm)

1.027 [1.006-1.048]

0.018

Data are shown as mean ± SD. TAVR: transcatheter aortic valve replacement; VCI: valve calcification index.

96

IMPACT OF AORTIC VALVE CALCIFICATION ON TAVR OUTCOME

Prosthesis malposition and post-dilatation No significant difference was found in VCI among patients with prosthesis malposition (n=8) as compared to patients without this periprocedural complication (22.85 ± 15.99, median 14.57 [IQR 11.08-41.49] vs 22.94 ± 11.01, median 21.79 [IQR 14.99-29.43], p=0.459). Also CSI appeared to be similar in these two groups of patients (0.41 ± 0.20, median 0.35 [IQR 0.300.65] vs 0.37±0.21, median 0.30 [IQR 0.20-0.50], p=0.600). Interestingly, when the VCI values were divided into 4 percentiles, prosthesis malposition among patients who received Edwards SAPIEN valve prosthesis was only observed in the first percentile group (Figure 4). Also among patients undergoing implantation of a Medtronic CoreValve system, the occurrence of prosthesis malposition decreased with the increasing VCI value (Figure 4).

Patients with the need for post-dilatation (n=43) showed significantly higher VCI and total calcium volume (27.06±12.91 ul/mm and 2264.89±1093.61ul, respectively) than patients without need for post-dilatation (22.11±10.72 ul/mm and 1790.78±895.75 ul, p=0.008 and p=0.002, respectively), however no difference was observed in the CSI (0.35±0.23 and 0.38±0.22, p=0.54). Furthermore, patients (n=49) with the need for an “extra manoeuvre” (postdilatation and/or additional valve-in-valve implantation) showed significantly higher VCI and total calcium volume (26.83±12.96 ul/mm and 2252.44±1096.92 ul, respectively) than all other patients (22.03±10.58 ul/mm and 1781.40 ± 885.50 ul, p= 0.007 and p=0.001, respectively), however no difference in the CSI (0.37±0.20 and 0.37±0.21, p=0.841).

Clinical outcome The composite 30-days safety endpoint occurred in 60 (22.8%) cases, including 11 (4.2%) patients with AKI > grade I, 11 (4.2%) with cerebrovascular events (10 patients with stroke, 1 patient with TIA), 2 (0.8%) with acute myocardial infarction, 24 (9.1%) needing a new pacemaker implantation, and 20 (7.6%) patients who did not survive beyond 30 days after TAVR. Baseline VCI was similar between patients who met and patients who did not meet the composite 30days safety endpoint (23.60 ± 11.23 vs 22.72 ± 11.24, p=0.594).

Impact of aortic valve calcification on mid-term survival Overall 12-months mortality was 17.2%. In the univariate analysis none of the aortic valve calcification parameters (AVC, VCI or CSI) were of predictive value for one-year mortality (Table 3A). Moreover, multivariate analysis including logistic EuroScore, baseline VCI and post-TAVR AKI class II or III, stroke or TIA and aortic regurgitation more than mild (all according to VARC-2) revealed that only AKI class II or III is significantly associated with one-year mortality (HR 3.725 [1.421-9.766, 95.0% CI], Table 3a-b).

97

CHAPTER 6

Table 3A | Univariate analysis of baseline aortic valve characteristics parameters with respect to oneyear mortality HR [95.0% CI]

P

AVC volume (ul)

1.000 [1.000-1.000]

0.949

VCI (ul/mm)

0.999 [0.972-1.027]

0.953

CSI

0.678 [0.145-3.161]

0.621

Table 3 B | Multivariate analysis of possible predictors of post-TAVR one-year mortality HR [95.0% CI]

P

Logistic EuroScore

1.026 [1.002-1.041]

0.061

VCI (ul/mm)

0.996 [0.970-1.024]

0.795

AKI class II or III

3.725 [1.421-9.766]

0.008

Stroke or TIA

2.235 [1.217-4.106]

0.186

Aortic regurgitation more than mild

1.581 [0.978-2.558]

0.341

Post-TAVR

AKI: Acute Kidney Injury; AVC: aortic valve calcification; CSI: calcification symmetry index; TAVR: transcatheter aortic valve replacement; VCI: valve calcification index.

DISCUSSION In the present study, we investigated the impact of the volume and distribution of aortic valve calcium, measured quantitatively using contrast-enhanced MSCT, on the outcome of TAVR procedures. VCI appeared to be a significant predictor of post-procedural AR and of the need for “secondary manoeuvres” during TAVR. However, no association was found between aortic valve calcification at baseline and short-term clinical events and mid-tem mortality after TAVR. Measurement of aortic valve calcification Contrast enhanced MSCT provides excellent visualization of AVC and its precise location on aortic valve cusps. For quantification of AVC on contrast-enhanced MSCT, no adequate calcification scoring system has been validated. As previously reported8,9, using the Agatston score for quantification of AVC on contrast-enhanced MSCT requires a non-linear weighting factor in its derivation and thus has been shown to exhibit great interscan variability. In this study we proposed a scoring method for quantitation of aortic valve calcification on contrast enhanced MSCT, with consideration of different contrast levels across the patient population.

Using 3Mensio software, the appropriate threshold for detection of calcium on contrastenhanced MSCT was determined on individual patient-basis. We believe that our method improves interscan comparability in quantification of calcium on contrast-enhanced MSCT. One previous paper by Ewe and colleagues, performing quantitative measurement of AVC on MSCT using the 3Mensio software, used a threshold of ≥ 800 HU to detect calcium in the region of interest, while in their study the luminal contrast enhancement ranged from 250

98

IMPACT OF AORTIC VALVE CALCIFICATION ON TAVR OUTCOME

Figure 4 | The occurrence of prosthesis malposition across VCI percentiles for the Medtroic CoreValve and Edwards SAPIEN prostheses.

to 760 HU11. Using similar threshold in our series led to missing area of calcification in the aortic valve apparatus. Hence, due to their strategy, and the fact that the treshold for calcium detection in our study was much lower (484.47 HU), Ewe et al might have missed areas of calcium in case of low luminal contrast enhancement.

Post-procedural aortic valve regurgitation Much of the current research for improvements in TAVR technique focuses on the predictors of AR and methods to prevent this frequently occurring phenomenon which is shown to affect the outcome of TAVR12-14. In theory, incomplete expansion of the prosthesis frame due to the presence of a significant amount of calcification, prosthesis undersizing, or suboptimal device positioning - too low or too high compared with the annulus level- can result in aortic

99

CHAPTER 6

regurgitation after device deployment. The presence of bulky calcification at the level of the commissures and on the cusps probably prevents adequate alignment of the stented prostheses against the aortic wall, with a resultant defective seal between these structures.

Hence, the implanted prosthesis may not be tightly in contact with the severely calcified native aortic wall throughout its circumference. This phenomenon induces gaps that in turn cause several diastolic paravalvular regurgitation jets that might add up to a relevant regurgitation.

Incomplete or non-circular prosthesis expansion due to the presence of unequal and asymmetrical distribution of calcification throughout the aortic valve, posing resistance to prosthesis deployment, is believed to be one of the other contributing factors15. Delgado and colleagues studied the geometry of deployed prosthesis at one month post-TAVR and reported a significantly higher degree of aortic valve calcification in patients with non-circular deployment of the valve prosthesis15.

In our series, 54.0% of patients had some degree of aortic valve regurgitation. In a large patient group, our analysis could confirm previous observations that aortic valve calcium correlates significantly and positively with the degree of post-procedural AR11,16-19. Moreover, our data demonstrated VCI as a stronger predictor for AR grade 1 than for AR grade 2 or higher. The latter suggests that higher amounts of calcium increases the risk of significant AR to a lesser extent than the risk of AR grade 1. A high amount of calcium probably has a ‘sealing’ effect between the aortic annulus and the metallic stent struts, preventing severe paravalvular AR. In regard to calcium distribution across the aortic valve apparatus, CSI was not predictive for significant post-procedural AR. This finding confirms the earlier reports showing no correlation between the distribution of calcium and AR severity after TAVR17,19-21.

Furthermore, the extent of oversizing and the eccentricity index of the native aortic valve annulus did not prove to be independent predictors of AR according to our data. This finding is in line with a few previous studies reporting no significant association between the shape of native aortic annulus (circular or more ellipsoid) or oversizing of the valve prosthesis and post-TAVR aortic valve regurgitation12,19,22,23. On the other hand, as suggested by Delgado and colleagues, an ellipsoid shape of the implanted prosthesis is associated with significant postprocedural AR15.

Interestingly, our data showed that according to the aortic annulus area measurements on pre-procedural MSCT, prostheses were ≥ 10% oversized in only 56.7% of patients. However, the ≥ 10% prosthesis oversizing rule has been suggested based on the 2-dimensional echocardiographic measurements and may not apply to annular measurements derived from MSCT24. The observation that occurrence of significant AR was not different between patients with or without relative prosthesis oversizing ≥ 10% advocates for that.

Second manouvers Limited amount of aortic valve calcification is likely to increase the risk of prosthesis malposition, either migration or embolization, while too large amount of aortic valve calcium might induce eccentric prosthesis deployment and post-procedural aortic regurgitation. In case of prosthesis malposition or significant post-procedural AR, a “secondary manoeuvre”, usually post-dilatation or implantation of a second prosthesis will be necessary. Indeed our

100

IMPACT OF AORTIC VALVE CALCIFICATION ON TAVR OUTCOME

data showed a higher VCI and total calcification volume in patients needing a â&#x20AC;&#x153;secondary manoeuvreâ&#x20AC;? compared to patients without need for post-dilatation or additional prosthesis implantation during TAVR. Impact on clinical outcome In regard to post-TAVR survival, only one previous study by Koos and colleagues, with a relatively small study population size (n=76) reported a significant independent positive association between Agatston AVC score and mortality at 30-days, two year and three years post-TAVR (3). However, our study showed no association between one-year mortality and VCI volume in a large cohort of patients.

For a successful TAVR, a certain amount of calcification in the area of aortic valve annulus is necessary for a firm fixation of the implantable valve prosthesis. During the implantation process, the aortic valve calcium is sandwiched between the frame of the valve prosthesis and the aortic wall and will probably not have much impact anymore on the long-term outcome of TAVR after the procedure. The highest impact of AVC on the outcome of patients undergoing TAVR is to be expected during the index procedure, as dislodgement of the calcification material during balloon dilatation or during prosthesis deployment may cause distal embolization and embolic complications for the brain or other essential organs such as the kidney. The latter hypothesis is however not supported by our results, as we did not find a significant difference between the VCI among patients who met and patients who did not meet the 30-days safety endpoint (composite of post-procedural AKI grade II or III, cerebrovascular events, acute myocardial infarction, new pacemaker implantation and all-cause mortality).

Interestingly, in contrary to a few recent publications, the occurrence of post-procedural AR was not a significant predictor of one-year mortality in our study. Also Koos et al did not find a significant independent relationship between mortality and post-TAVR paravalvular regurgitation3.

Aortic cusp calcification In the present study, the noncoronary of the aortic valve exhibited the highest calcium volume and also appeared to increase the most across increasing degrees of AR as compared to the right coronary and the left coronary cusps. A few previous reports described similar findings with the highest amount of calcification in the noncoronary cusp17,19,21. Aortic valve lesions are likely initiated by endothelial disruption due to increased mechanical or decreased shear stress, similar to that seen in early atherosclerotic lesions. Shear stress across the endothelium of the noncoronary cusp is lower than the left and right coronary cusps because of the absence of diastolic coronary flow, which likely explains why the noncoronary cusp carries the highest amount of calcification25. Furthermore, given the left posterior angulation of the ascending aorta, the posteriorly located noncoronary cusp is probably exposed the most to retrograde turbulances from aortic valve jet than more anteriorly located left and right coronary cusps26.

In addition, the noncoronary cusp is usually slightly larger than the left and the right cusps.

Another factor might be the continuity of the noncoronary cusp with interatrial septum which provides structural support and restricting this cusp in its movement a bit more than the right

101

CHAPTER 6

and the left cusps. However these hypothesis need further evaluation as also contradictory evidence has been reported suggesting the left coronary cusp to show the highest amount of calcium16,20.

CONCLUSION A high amount of calcification in the native aortic valve may prevent perfect apposition between the prosthesis and aortic wall, hence resulting in post-procedural AR. In a relatively large patient population, our data confirmed the relation between AVC and the occurrence of post-procedural AR. Calcification symmetry index on the other hand had no predictive value for post-procedural AR severity.

Furthermore, a higher VCI and total calcification volume was observed in patients needing post-dilatation or additional prosthesis implantation compared to patients without need for these “secondary manoeuvres” during TAVR. Therefore, evaluation of the severity of AVC before TAVR might allow the physician to predict the risk of significant AR after prosthesis deployment and the need for “secondary manoeuvres”.

Lastly, our data suggest no association between VCI at baseline and short-term clinical events and one-year mortality after TAVR. LIMITATIONS Although the evaluation of post-procedural AR after TAVR using echocardiography was performed according to current VARC-2 recommendations, an observer-related bias still remains. However, an accurate assessment of the severity of the post-procedural aortic regurgitation remains difficult due to the absence of validated methods to quantify paravalvular leakage.

102

IMPACT OF AORTIC VALVE CALCIFICATION ON TAVR OUTCOME

REFERENCES 1. Kodali SK, Williams MR, Smith CR, Svensson LG, Webb JG, Makkar RR, Fontana GP, Dewey TM, Thourani VH, Pichard AD, Fischbein M, Szeto WY, Lim S, Greason KL, Teirstein PS, Malaisrie SC, Douglas PS, Hahn RT, Whisenant B, Zajarias A, Wang D, Akin JJ, Anderson WN, Leon MB, PARTNER Trial Investigators. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med 2012;366:1686-1695.

2. Zegdi R, Ciobotaru V, Noghin M, Sleilaty G, Lafont A, Latremouille C, Deloche A, Fabiani JN. Is it reasonable to treat all calcified stenotic aortic valves with a valved stent? Results from a human anatomic study in adults. J Am Coll Cardiol 2008;51:579-584.

3. Koos R, Reinartz S, Mahnken AH, Herpertz R, Lotfi S, Autschbach R, Marx N, Hoffmann R. Impact of aortic valve calcification severity and impaired left ventricular function on 3-year results of patients undergoing transcatheter aortic valve replacement. Eur Radiol 2013;23:3253-3261.

4. Staubach S, Franke J, Gerckens U, Schuler G, Zahn R, Eggebrecht H, Hambrecht R, Sack S, Richardt G, Horack M, Senges J, Steinberg DH, Ledwoch J, Fichtlscherer S, Doss M, Wunderlich N, Sievert H, German Transcatheter Aortic Valve Implantation-Registry Investigators. Impact of aortic valve calcification on the outcome of transcatheter aortic valve implantation: results from the prospective multicenter German TAVI registry. Catheter Cardiovasc Interv 2013;81:348-355.

5. Doddamani S, Grushko MJ, Makaryus AN, Jain VR, Bello R, Friedman MA, Ostfeld RJ, Malhotra D, Boxt LM, Haramati L, Spevack DM. Demonstration of left ventricular outflow tract eccentricity by 64-slice multi-detector CT. Int J Cardiovasc Imaging 2009;25:175-181.

6. Blanke P, Willson AB, Webb JG, Achenbach S, Piazza N, Min JK, Pache G, Leipsic J. Oversizing in transcatheter aortic valve replacement, a commonly used term but a poorly understood one: dependency on definition and geometrical measurements. J Cardiovasc Comput Tomogr 2014;8:67-76.

7. Mylonas I, Alam M, Amily N, Small G, Chen L, Yam Y, Hibbert B, Chow BJ. Quantifying coronary artery calcification from a contrast-enhanced cardiac computed tomography angiography study. Eur Heart J Cardiovasc Imaging 2014;15:210-215.

8. Callister TQ, Cooil B, Raya SP, Lippolis NJ, Russo DJ, Raggi P. Coronary artery disease: improved reproducibility of calcium scoring with an electron-beam CT volumetric method. Radiology 1998;208:807-814.

9. Ohnesorge B, Flohr T, Fischbach R, Kopp AF, Knez A, Schroder S, Schopf UJ, Crispin A, Klotz E, Reiser MF, Becker CR. Reproducibility of coronary calcium quantification in repeat examinations with retrospectively ECG-gated multisection spiral CT. Eur Radiol 2002;12:1532-1540.

10. Kappetein AP, Head SJ, Genereux P, Piazza N, van Mieghem NM, Blackstone EH, Brott TG, Cohen DJ, Cutlip DE, van Es GA, Hahn RT, Kirtane AJ, Krucoff MW, Kodali S, Mack MJ, Mehran R, Rodes-Cabau J, Vranckx P, Webb JG, Windecker S, Serruys PW, Leon MB, Valve Academic Research Consortium-2.

Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. J Thorac Cardiovasc Surg 2013;145:6-23.

11. Ewe SH, Ng AC, Schuijf JD, van der Kley F, Colli A, Palmen M, de Weger A, Marsan NA, Holman ER, de Roos A, Schalij MJ, Bax JJ, Delgado V. Location and severity of aortic valve calcium and implications for aortic regurgitation after transcatheter aortic valve implantation. Am J Cardiol 2011;108:1470-1477.

12. Marwan M, Achenbach S, Ensminger SM, Pflederer T, Ropers D, Ludwig J, Weyand M, Daniel WG, Arnold M. CT predictors of post-procedural aortic regurgitation in patients referred for transcatheter aortic valve implantation: an analysis of 105 patients. Int J Cardiovasc Imaging 2013;29:1191-1198.

13. Tamburino C, Capodanno D, Ramondo A, Petronio AS, Ettori F, Santoro G, Klugmann S, Bedogni F, Maisano F, Marzocchi A, Poli A, Antoniucci D, Napodano M, De Carlo M, Fiorina C, Ussia GP. Incidence and predictors of early and late mortality after transcatheter aortic valve implantation in 663 patients with severe aortic stenosis. Circulation 2011;123:299-308.

103

CHAPTER 6

14. Kodali SK, Williams MR, Smith CR, Svensson LG, Webb JG, Makkar RR, Fontana GP, Dewey TM, Thourani VH, Pichard AD, Fischbein M, Szeto WY, Lim S, Greason KL, Teirstein PS, Malaisrie SC, Douglas PS, Hahn RT, Whisenant B, Zajarias A, Wang D, Akin JJ, Anderson WN, Leon MB, PARTNER Trial Investigators. Twoyear outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med 2012;366:16861695.

15. Delgado V, Ng AC, van de Veire NR, van der Kley F, Schuijf JD, Tops LF, de Weger A, Tavilla G, de Roos A, Kroft LJ, Schalij MJ, Bax JJ. Transcatheter aortic valve implantation: role of multi-detector row computed tomography to evaluate prosthesis positioning and deployment in relation to valve function. Eur Heart J 2010;31:1114-1123.

16. Haensig M, Lehmkuhl L, Rastan AJ, Kempfert J, Mukherjee C, Gutberlet M, Holzhey DM, Mohr FW. Aortic valve calcium scoring is a predictor of significant paravalvular aortic insufficiency in transapical-aortic valve implantation. Eur J Cardiothorac Surg 2012;41:1234-40; discussion 1240-1.

17. Koos R, Mahnken AH, Dohmen G, Brehmer K, Gunther RW, Autschbach R, Marx N, Hoffmann R. Association of aortic valve calcification severity with the degree of aortic regurgitation after transcatheter aortic valve implantation. Int J Cardiol 2011;150:142-145.

18. John D, Buellesfeld L, Yuecel S, Mueller R, Latsios G, Beucher H, Gerckens U, Grube E. Correlation of Device landing zone calcification and acute procedural success in patients undergoing transcatheter aortic valve implantations with the self-expanding CoreValve prosthesis. JACC Cardiovasc Interv 2010;3:233-243.

19. Wendt D, Plicht B, Kahlert P, Hartmann K, Al-Rashid F, Price V, Konorza T, Erbel R, Jakob H, Thielmann M. A novel calcium scoring system accurately predicts likelihood and location of post-TAVI paravalvular leak. J Cardiovasc Surg (Torino) 2014;55:423-433.

20. Feuchtner G, Plank F, Bartel T, Mueller S, Leipsic J, Schachner T, Muller L, Friedrich G, Klauser A, Grimm M, Bonaros N. Prediction of paravalvular regurgitation after transcatheter aortic valve implantation by computed tomography: value of aortic valve and annular calcification. Ann Thorac Surg 2013;96:1574-1580.

21. John D, Buellesfeld L, Yuecel S, Mueller R, Latsios G, Beucher H, Gerckens U, Grube E. Correlation of Device landing zone calcification and acute procedural success in patients undergoing transcatheter aortic valve implantations with the self-expanding CoreValve prosthesis. JACC Cardiovasc Interv 2010;3:233-243.

22. Wood DA, Tops LF, Mayo JR, Pasupati S, Schalij MJ, Humphries K, Lee M, Al Ali A, Munt B, Moss R, Thompson CR, Bax JJ, Webb JG. Role of multislice computed tomography in transcatheter aortic valve replacement. Am J Cardiol 2009;103:1295-1301.

23. Colli A, D'Amico R, Kempfert J, Borger MA, Mohr FW, Walther T. Transesophageal echocardiographic scoring for transcatheter aortic valve implantation: impact of aortic cusp calcification on postoperative aortic regurgitation. J Thorac Cardiovasc Surg 2011;142:1229-1235.

24. Walther T, Dewey T, Borger MA, Kempfert J, Linke A, Becht R, Falk V, Schuler G, Mohr FW, Mack M. Transapical aortic valve implantation: step by step. Ann Thorac Surg 2009;87:276-283.

25. Freeman RV, Otto CM. Spectrum of calcific aortic valve disease: pathogenesis, disease progression, and treatment strategies. Circulation 2005;111:3316-3326.

26. Viscardi F, Vergara C, Antiga L, Merelli S, Veneziani A, Puppini G, Faggian G, Mazzucco A, Luciani GB. Comparative finite element model analysis of ascending aortic flow in bicuspid and tricuspid aortic valve. Artif Organs 2010;34:1114-1120.

104

105

CHAPTER

7 RATIONALE AND DESIGN OF THE EDWARDS SAPIEN 3 PERIPROSTHETIC LEAKAGE EVALUATION VERSUS MEDTRONIC COREVALVE IN TRANFEMORAL AORTIC VALVE IMPLANTATION (ELECT) TRIAL: A RANDOMIZED COMPARISON OF BALLOON-EXPANDABLE VERSUS SELF-EXPANDING AORTIC VALVE PROSTHESES In preparation

Mariam Samim1 Pieter R. Stella1 Steven A. Chamuleau1 Freek Nijhoff1 Kim Urgel1 Jeroen Hendrikse2 Tim Leiner2 Alferso C. Abrahams3 Bart van der Worp4 Pieter A. Doevendans1 Pierfrancesco Agostoni1 Departments of Cardiology, 2Department of Radiology, 3Department of Nephrology,

1

Department of Neurology, University Medical Centre Utrecht, the Netherlands

4

CHAPTER 7

ABSTRACT Background Transcatheter aortic valve replacement (TAVR) is a valid treatment for patients with severe symptomatic aortic valve stenosis at high risk for surgery. Periprosthetic aortic regurgitation (AR) after TAVR remains an important limitation and possible differences among different TAVR devices currently available have been suggested.

Objective To evaluate potential differences in regard to prosthesis function and clinical outcome between the balloon-expandable Edwards SAPIEN 3 (ES) bioprosthesis and the self-expanding Medtronic CoreValve速 system (MCS) with main focus on post-TAVR AR. Methods and design The ELECT study is an investigator-initiated, single-centre, randomised controlled trial in highrisk or inoperable patients with severe symptomatic aortic valve stenosis and an anatomy suitable for transfemoral TAVR. One hundred and eight patients will be randomly allocated in a one-to-one ratio to receive the Edwards SAPIEN 3 (n=54) or the Medtronic CoreValve速 system (n=54). The primary endpoint is post-TAVR AR measured with several different imaging modalities (angiography, transthoracic and (3-dimensional) transesophageal echocardiography, cardiac magnetic resonance imaging (MRI)). Secondary objectives of this study include clinical endpoints (according to VARC-2 definitions), quality of life at mid-term follow up. Peri-procedural cerebral MRI and 24-hour urine collection were performed for a careful evaluation of post-TAVR new cerebral infarction or kidney function impairment, respectively.

Discussion The ELECT trial is the first randomised controlled trial to quantitatively compare the extent of post-procedural AR between MCV system and the ES prosthesis. Furthermore, it will evaluate potential differences between the two prostheses in regard to mid-term clinical outcome and quality of life.

Trial registration ClinicalTrials.gov: NCT01982032 Registration date: 2 August 2013.

108

BALLOON-EXPANDABLE VERSUS SELF-EXPANDING AORTIC VALVE PROSTHESES: THE ELECT TRIAL

BACKGROUND Transcatheter aortic valve replacement (TAVR) is a valid treatment strategy for patients with severe symptomatic aortic stenosis regarded at high risk or inoperable by open-heart surgery. Two transcatheter aortic valve prostheses based on different technical concepts, balloon-expandable or self-expanding , have been developed and widespread used: the Edwards SAPIEN (ES, Edwards Lifesciences) and the Medtronic CoreValve (MCV, Medtronic Inc) prosthesis. Both valves have shown excellent clinical results, but each has specific features, advantages and disadvantages 1, 2.

Currently available evidence suggests that TAVR is feasible and provides long-term hemodynamic and clinical improvements, but questions remain concerning safety and durability of this technique. A few important complications of TAVR have to be resolved in order to warrant the wider use of this procedure. Significant concerns are raised around the high incidence of post-procedural periprosthetic aortic regurgitation (AR), which is associated with an increased mortality 1, 3-5. It is currently estimated that up to 41-100% of patients have some degree of AR following TAVR 4, 6-12. Even mild post-procedural AR is associated with 1015% higher mortality at 2 years than patients with none or trace AR, as shown by results of PARTNER cohort A 1.

Suboptimal placement of the prosthesis, with incomplete sealing of the annulus by the skirt, an incomplete apposition of the stent frame owing to calcification in the device landing zone, and undersizing of the valve prosthesis relative to the dimensions of the aortic annulus, are mechanisms of post-TAVR AR. However, the impact of prosthesis type on the risk of post-TAVR AR is less clear. A few recent reports, among which the only randomized trial comparing ES XT with MCV (CHOICE study), have suggested differences in the hemodynamic performance of both prostheses with the MCV associated with a higher rate of residual periprosthetic AR 13, 14.

However, the results from observational registries might be hampered by bias and confounding factors, such as operatorsâ&#x20AC;&#x2122; familiarity with the device. Furthermore, periprosthetic AR assessment using 2-dimensional (2D) echocardiography and according to invalidated Valve Academic Research Consortium (VARC), or subjective AR assessment using unidimensional angiography images (such as in the CHOICE study) may have biased the results of the previous reports.

Hence, an adequately powered randomized study with quantitative AR assessment may clarify the difference in AR severity between ES and MCV more reliably. 3-dimensional transesophageal echocardiography (3DTEE), Q-flow (phase-contrast pulse sequences) magnetic resonance imaging (MRI), and contrast aortography are proposed for quantification of post-TAVR AR 15-17.

Recently, the Edwards SAPIEN 3 (ES3) was introduced, which carries an additional outer skirt in order to minimize post-TAVR AR. However, whether this new feature is effective in preventing post-TAVR AR and whether it makes ES3 superior to MCV in that regard, remains yet to be clarified. Therefore, we designed a clinical trial for randomized comparison of MCV and ES3 in high-risk or inoperable patients with severe symptomatic aortic valve stenosis and eligible for transfemoral TAVR: Edwards SAPIEN Periprosthetic Leakage Evaluation versus medtronic Corevalve in Tranfemoral aortic valve implantation (ELECT) trial focusing on post-TAVR AR measured with innovative imaging modalities that allow a more accurate quantification of the

109

CHAPTER 7

AR. Secondary aims of this study are comparing clinical endpoints as defined by VARC-2 (18) and the quality of life at one year post-TAVR, between MCV and ES3.

METHODS AND DESIGN The ELECT study is a single centre prospective two-arm randomised controlled trial. No external financial support exists. Consecutive adult males or females, judged eligible for transfemoral TAVR (TF-TAVR) by the heart- team, and who meet the inclusion criteria and none of the exclusion criteria, will be approached for inclusion in this study. The ELECT trial conforms to the Ethical Principles of the Declaration of Helsinki, the Ethics Committee of the University Medical Centre Utrecht gave full approval to the study and patients who agree to participate will be asked for their written informed consent.

Inclusion criteria In order to be eligible to participate in this study, subjects must meet all of the following criteria: • Patient is 18 years of age or older and diagnosed with severe symptomatic aortic stenosis, judged inoperable or at high surgical risk (EuroSCORE > 15%) and deemed eligible for TAVR by a consensus among cardiologists and cardiac surgeons (heart-team) or a patient who is considered to be operable by the heart-team, but who chooses to undergo TAVR instead of conventional surgery.

• Aortic annulus diameter ≥18 and ≤28 mm as assessed with multislice computed tomography (MSCT). Patients with an annulus diameter outside this range cannot participate in randomisation for the Edwards SAPIEN 3 and Medtronic CoreValve® bioprostheses. Until now no Edwards SAPIEN 3 prosthesis is available for patients with an annulus diameter >28.

• Patients eligible for TF-TAVR. Exclusion criteria • Patients unable or unwilling to give informed consent Patients who are excluded from this study because of the size of their aortic annulus diameter No TAVR prostheses is available for patients with an annulus diameter < 18 mm, and therefore they cannot undergo a TAVR procedure. Patients with an annulus diameter >28 mm may be candidates for the implantation of a MCV 31 mm prosthesis (which is only allowed for annulus diameters up to 29 mm). The latter will be decided by the heart-team.

Sample size calculation A sample size calculation is performed for the primary endpoint post-TAVR AR. A literature search is performed for difference in the extent of AR after TAVR between MCV and the ES prosthesis. No appropriate studies were found with quantitative AR comparison between MCV and ES. We used data from 25 patients treated with TAVR at our centre for a sample size calculation. AR quantification in these patients was performed using cardiac MRI. Because the distribution of these volume data was skewed to the right, we performed a log transformation

110

BALLOON-EXPANDABLE VERSUS SELF-EXPANDING AORTIC VALVE PROSTHESES: THE ELECT TRIAL

on the original data. The log transformed values were as follow: an “overall” mean regurgitant volume of 0.48, an “overall” median regurgitant volume of 0.48, an “overall” standard deviation (SD) of 0.53, a mean regurgitant volume of 0.32 (±0.35 SD) for ES, and a mean regurgitant volume of 0.63 (±0.69) for MCV. As even mild AR is shown to be associated with long-term mortality, any significant difference in post-TAVR AR volume between MCV and ES was considered clinically important 19. For the sample size calculation, a value of 0.80 is used for the power, and α (type I error rate) is set at 0.05. The sample size calculation, including the above mentioned log transformed values for mean regurgitant volumes and standard deviations, yielded a sample size of 49 patients in every arm to show superiority of the ES3 prothesis over de MCV prothesis. As we took into account an expected dropout rate of 10%, this yielded a total sample size of 54 patients in every arm.

Randomization and interventions Patients who meet all the inclusion criteria and none of the exclusion criteria will be randomized in consecutive order of qualification. In the heart catheterization lab and using sealed envelopes, after successful puncture of the femoral artery chosen as entry site for TAVR, subjects will be randomised in a 1:1 fashion to receive a MCV of an ES3.

Both prostheses (ES and MCV) are approved in the European Union and received CE-Mark approval for the treatment of severe aortic stenosis. Prosthesis size determination was based on the aortic annulus diameter. The dimensions of the aortic annulus are measured using preprocedural contrast enhanced MSCT. The self-expanding MCV is available in four sizes: 23 mm, 26 mm, 29 mm and 31mm (for native aortic annulus diameters 18-20 mm, 20-23 mm, 23-27 mm, and 27-29 mm, respectively). The ES3 is provided in three sizes: 23 mm, 26 mm and 29mm (for native aortic annulus diameters 18-22 mm, 21-25 mm, and 24-28 mm, respectively). In order to be able to randomize patients for receiving MCV of ES3, only patients with an annulus diameter range between 18 mm and 28 mm will be included in this study, thus excluding patients with an annulus diameter of 29 mm (who will be treated with MCV 31 mm).

TAVR procedure All procedures are performed by a highly experienced team and trial operators have an overall experience of performing > 250 TAVR using both devices. Details of implantation technique for both ES and MCV have been previously reported 20-22. In regard to the implantation of ES3 prostheses, the implantation technique is similar to that of the ES XT, except for the additional radio-opaque marker located in the central part of the balloon, assisting in valve positioning22.

The procedure is performed under conscious sedation and with local anaesthesia (without endotracheal intubation). Fluoroscopy and intracardiac echocardiography are used for procedural guidance. Balloon pre-dilation is performed in all cases as per centre routine. The valve prostheses are deployed as per routine under rapid pacing (Edwards – 180 beats/min) or “stabilizing” pacing (CoreValve – 120 beats/min). Post-TAVR patient are monitored for at least 72 hours and discharged on a regimen of life-long low-dose aspirin (80-100 mg) or oral anticoagulant (in case of clinical indication for it) and 3 months clopidogrel (75 mg).

111

CHAPTER 7

End point assessment Follow-up assessments for the measurements of the primary and secondary endpoints will be performed at specified time points (Figure 1).

Primary endpoint Post-TAVR aortic regurgitation is assessed using several techniques, among which innovative imaging modalities that allow a more accurate quantification of the AR: • At the end of each TAVR procedure, a contrast angiography is performed in order to measure the degree of post-implantation AR by contrast densitometry (CAAS A-valve quantitative regurgitation analysis; Pie Medical Imaging, Maastricht, The Netherlands) (Figure 2). Technical details of this approach have been reported previously 17.

• At day 4 (+1) a transthoracic echocardiography (TTE) and a 3DTEE are performed as per routine. Using TTE, colour Doppler evaluation is performed just below the prosthesis for periprosthetic jets, and at the coaptation point of the leaflets for central regurgitation.

Imaging windows include the parasternal short-axis view for assessment of the number and severity of paravalvular jets. In addition, 3DTEE is used to analyze the complete morphology of the AR color flow stream in the region of its origin. Consequently, the view from any level can be obtained, and the direction and extension of periprosthetic AR jets can be assessed 23. New dedicated analysis software (Personal Space Technologies B.V., the Netherlads) designed for the visualization and analysis of 3D volumetric data, will be used for quantitative grading of AR severity in this study.

• At day 4 (+1) after TAVR, patients without contraindications for MRI, will undergo cardiac Q-flow MRI imaging for quantitative grading of AR. Secondary endpoints Patients without contraindications for MRI will undergo cerebral diffusion weighted MRI at day 4(+1) after TAVR, for detection of new cerebral ischemic injury. In order to detect kidney injury related to TAVR, kidney function is assesses using 24-hour urine collection at three different moments: I) within 48 hours before TAVR, II) within 5 days after TAVR, and III) at 6 months follow-up.

In order to investigate possible mid-term changes in post-TAVR AR, valvular function is evaluated at 6 months using TTE and 3DTEE as describe above. Finally, the EuroQol EQ-5D and SF-36 quality of life questionnaires will be used to measure health-related quality of life at one year follow-up as compared to baseline.

112

BALLOON-EXPANDABLE VERSUS SELF-EXPANDING AORTIC VALVE PROSTHESES: THE ELECT TRIAL

Figure 1 | Flowchart of the study.

113

CHAPTER 7

Figure 2 | Contrast densitometry after transcatheter prosthesis implantation. As shown by the top panel, the region of interest is drawn manually which includes aortic root and the left ventricle, marked with dotted yellow lines. The bottom panel shows the 5 curves generated by the software (Pie Medical Imaging) for the aortic root reference area (red), the left ventricle base (purple), left ventricle apex (green), midsection (light blue) and overall (yellow). Left ventricle contrast densitometry on the aortograms provides a quantitative measure for post-TAVR AR (qRA), ranging from 0 (absent) to 3 (severe). In this example qRA is 1.2, which corresponds to mild AR.

STATISTICAL ANALYSIS As the ELECT trial is designed to address whether one prosthesis type is better than the other, it is a superiority trial. The recommended method in superiority trials to avoid any bias is an intention to treat (ITT) analysis, which will be the primary analysis in this study. Descriptive statistics will be used to ascertain any marked imbalance between the arms at baseline. The extent of missing data will be reported and baseline factors will be compared for completers and non-completers to assess the extent of any bias that may result. Continuous variables will be presented as mean (ÂąSD). The difference between the means of the two arms will be calculated using an unpaired T-test. For the analysis of the difference in AR between the two prostheses, a log transformation will be performed on the regurgitation volume data before performing a T-test. The follow-up analysis for changes in the endpoints in the time within the arms will be performed using the paired T-test and the repeated measurements ANOVA. Dichotomous variables will be expressed as numbers (percentages) and compared using Fischer exact chisquare test. A two-tailed p value of less than 0,05 will be regarded as statistically significant.

114

BALLOON-EXPANDABLE VERSUS SELF-EXPANDING AORTIC VALVE PROSTHESES: THE ELECT TRIAL

Figure 3 | Planned flow diagram during ELECT trial.

DISCUSSION The ELECT trial is the first randomised controlled trial to quantitatively compare the magnitude of post-procedural AR between the MCV system and the ES 3 prosthesis which is designed to minimize periprosthetic regurgitation. Other important objectives of this trial include comparison of clinical endpoints between the two prostheses and assessment of the value of angiography, transthoracic and (3D) transesophageal echocardiography and cardiac MRI for the measurement of post-TAVR AR severity.

Regardless of the prosthesis type, post-procedural AR is a common complication after TAVR associated with increased mortality 24. A meta-analysis by Athappan and colleagues, including 45 studies, reported an overall incidence of moderate or severe AR of 11.7% within 30 days

115

CHAPTER 7

after TAVR (24). The self-expanding MCV and the balloon-expandable ES are two transcatheter heart valves that have been in widespread use worldwide. Few previous studies, comparing the hemodynamic performance of these two prosthesis types, have suggested a higher incidence of moderate or severe post-TAVR AR with the implantation of MCV 13, 24.

An important aspect of MCV that might increase the risk of post-TAVR periprosthetic AR (and thus explain the aforementioned data) is the intrinsic radial strength of its nitinol frame which may not be sufficient for complete apposition of the prosthesis to the native annulus.

This incomplete apposition might create periprosthetic gaps, especially in the presence of calcification along the aortic wall. Furthermore, an extreme angulation between the left ventricular outflow tract and the ascending aorta (so called horizontal aorta) may reduce the ability of the MCV prosthesis to seal the paravalvular space 24. Therefore oversizing and balloon post-dilatation are more important for MCV implantation compared to ES implantation. In addition, owing to the noncylindrical shape of the MCV system, its effective area inside the aortic annulus depends on the depth of the prosthesis in the left ventricular outflow tract.

Therefore sealing of the paravalvular space by the prosthesis in case of the MCV depends also on its implantation depth. However, besides the factors mentioned above which make MCV susceptible to post-TAVR AR, we also need to keep in mind that the self-expanding nitinol frame has the potential to expand overtime. Therefore, the apposition of the prosthesis frame to the aortic annulus has the potential to improve gradually, decreasing the extent of postTAVR AR with time. Interestingly, the recently published results of the CoreValve US Pivotal Trial showed that the incidence of any AR after the implantation of MCV decreases overtime, with 42.1% at discharge and 31.9% at one year. The latter was also accompanied by a reduction in the incidence of more-than-mild AR during follow-up, with 13.8% at discharge, 10.1% at 6 months, and 6.4% at one year. This suggests that measurement of post-TAVR AR in case of MCV in the few first days after TAVR will probably overestimate its long-term severity. In the ELECT study, the severity of AR is also measured at 6 months follow-up using transthoracic and 3D transesophageal echocardiography, which allows for a more reliable comparison between the two prostheses.

One previously published meta-analysis showed a higher risk of moderate or severe postprocedural AR after MCV implantation (16%) as compared to ES implantation (9.1%, p=0.005) (24). The CHOICE study, which is the only randomized head-to-head trial to date comparing the balloon expandable ES XT to the MCV system, reported a significant difference in the frequency of any degree of AR (38% in ES XT group vs 65% in the MCV group, p < 0.001) and more-than mild AR (4.1% in ES XT group vs 18.3 % in the MCV group, p < 0.001 by angiography) favouring the balloon-expandable valve. On the other hand, the PRAGMATIC study reported a very low and comparable incidence of more-than-mild aortic regurgitation with both MCV and ES (2.0% and 1.8%, respectively) in a large multicentre propensity score matched study 25.

Obviously, there is a discrepancy in the incidence of AR across different studies that is most probably related to the challenges in identification and quantification of post-TAVR AR. Also the VARC-2 document did not propose new diagnostic criteria for adequate assessment of post-TAVR AR. The lack of standardised and validated method for evaluation of post-TAVR AR is a major limitation in comparing echocardiographic AR analysis performed in different studies

116

BALLOON-EXPANDABLE VERSUS SELF-EXPANDING AORTIC VALVE PROSTHESES: THE ELECT TRIAL

and meta-analyses. Aortic regurgitation after TAVR usually consists of multiple eccentric jets that are non-parallel and irregular in shape 16, 26. The eccentric jets in turn are frequently entrained along the left ventricular wall with fanning of the jets as they regurgitate 24. Therefore, the eccentric aspect of post-TAVR AR makes the assessment of its severity challenging. Also acoustic shadowing from the calcifications and Doppler attenuation form the prosthesis can obscure regurgitant jets and thus result in underestimation of its severity 26. On the other hand, assessment of AR severity using aortic root angiography relies on subjective assessment of unidimensional images, and can be affected by interobserver and intraobserver variability.

The extent of the periprosthetic AR in the CHOICE trial was measured using aortic root angiography immediately after prosthesis implantation, which is an important limitation of this clinical trial. In the ELECT trial, the severity of periprosthetic AR is measured quantitatively using several different modalities, from dedicated softwares for angiography directly postTAVR implantation to echocardiography either transthoracic and transesophageal (with 3D reconstruction) to cardiac MRI including phase-contrast pulse sequences. A secondary aim of the ELECT trial will be the comparison of the different imaging modalities in assessing AR. This comparison is an important aspect of the present trial as until now no validated tool exists for reliable measurement of post-TAVR AR.

117

CHAPTER 7

REFERENCES 1. Kodali SK, Williams MR, Smith CR, Svensson LG, Webb JG, Makkar RR, et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med. 2012 May 3;366(18):1686-95.

2. Adams DH, Popma JJ, Reardon MJ, Yakubov SJ, Coselli JS, Deeb GM, et al. Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med. 2014 May 8;370(19):1790-8.

3. Gotzmann M, Korten M, Bojara W, Lindstaedt M, Rahlmann P, Mugge A, et al. Long-term outcome of patients with moderate and severe prosthetic aortic valve regurgitation after transcatheter aortic valve implantation. Am J Cardiol. 2012 Aug 2.

4. Sinning JM, Hammerstingl C, Vasa-Nicotera M, Adenauer V, Lema Cachiguango SJ, Scheer AC, et al.

Aortic regurgitation index defines severity of peri-prosthetic regurgitation and predicts outcome in patients after transcatheter aortic valve implantation. J Am Coll Cardiol. 2012 Mar 27;59(13):1134-41.

5. Moat NE, Ludman P, de Belder MA, Bridgewater B, Cunningham AD, Young CP, et al. Long-term outcomes after transcatheter aortic valve implantation in high-risk patients with severe aortic stenosis: The U.K. TAVI (united kingdom transcatheter aortic valve implantation) registry. J Am Coll Cardiol. 2011 Nov 8;58(20):2130-8.

6. Abdel-Wahab M, Zahn R, Horack M, Gerckens U, Schuler G, Sievert H, et al. Aortic regurgitation after transcatheter aortic valve implantation: Incidence and early outcome. results from the german transcatheter aortic valve interventions registry. Heart. 2011 Jun;97(11):899-906.

7. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010 Oct 21;363(17):1597-60.

8. Schultz CJ, Tzikas A, Moelker A, Rossi A, Nuis RJ, Geleijnse MM, et al. Correlates on MSCT of paravalvular aortic regurgitation after transcatheter aortic valve implantation using the medtronic CoreValve prosthesis. Catheter Cardiovasc Interv. 2011 Sep 1;78(3):446-55.

9. Sherif MA, Abdel-Wahab M, Beurich HW, Stocker B, Zachow D, Geist V, et al. Haemodynamic evaluation of aortic regurgitation after transcatheter aortic valve implantation using cardiovascular magnetic resonance. EuroIntervention. 2011 May;7(1):57-63.

10. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011 Jun 9;364(23):2187-98.

11. Lerakis S, Hayek SS, Douglas PS. Paravalvular aortic leak after transcatheter aortic valve replacement: Current knowledge. Circulation. 2013 Jan 22;127(3):397-40.

12. Tamburino C, Capodanno D, Ramondo A, Petronio AS, Ettori F, Santoro G, et al. Incidence and predictors of early and late mortality after transcatheter aortic valve implantation in 663 patients with severe aortic stenosis. Circulation. 2011 Jan 25;123(3):299-308.

13. Abdel-Wahab M, Mehilli J, Frerker C, Neumann FJ, Kurz T, Tolg R, et al. Comparison of balloonexpandable vs self-expandable valves in patients undergoing transcatheter aortic valve replacement: The CHOICE randomized clinical trial. JAMA. 2014 Apr 16;311(15):1503-14.

14. Athappan G, Patvardhan E, Tuzcu EM, Svensson LG, Lemos PA, Fraccaro C, et al. Incidence, predictors, and outcomes of aortic regurgitation after transcatheter aortic valve replacement: Meta-analysis and systematic review of literature. J Am Coll Cardiol. 2013 Apr 16;61(15):1585-9.

15. Sherif MA, Abdel-Wahab M, Beurich HW, Stocker B, Zachow D, Geist V, et al. Haemodynamic evaluation of aortic regurgitation after transcatheter aortic valve implantation using cardiovascular magnetic resonance. EuroIntervention. 2011 May;7(1):57-63.

16. Zamorano JL, Badano LP, Bruce C, Chan KL, Goncalves A, Hahn RT, et al. EAE/ASE recommendations for the use of echocardiography in new transcatheter interventions for valvular heart disease. Eur Heart J. 2011 Sep;32(17):2189-214.

118

BALLOON-EXPANDABLE VERSUS SELF-EXPANDING AORTIC VALVE PROSTHESES: THE ELECT TRIAL

17. Schultz CJ, Slots TL, Yong G, Aben JP, Van Mieghem N, Swaans M, et al. An objective and reproducible method for quantification of aortic regurgitation after TAVI. EuroIntervention. 2014 Jul 20;10(3):35563.

18. Kappetein AP, Head SJ, Genereux P, Piazza N, van Mieghem NM, Blackstone EH, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: The valve academic research consortium-2 consensus document. J Thorac Cardiovasc Surg. 2013 Jan;145(1):6-23.

19. Kodali SK, Williams MR, Smith CR, Svensson LG, Webb JG, Makkar RR, et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med. 2012 May 3;366(18):1686-95.

20. Grube E, Schuler G, Buellesfeld L, Gerckens U, Linke A, Wenaweser P, et al. Percutaneous aortic valve replacement for severe aortic stenosis in high-risk patients using the second- and current thirdgeneration self-expanding CoreValve prosthesis: Device success and 30-day clinical outcome. J Am Coll Cardiol. 2007 Jul 3;50(1):69-76.

21. Webb JG, Chandavimol M, Thompson CR, Ricci DR, Carere RG, Munt BI, et al. Percutaneous aortic valve implantation retrograde from the femoral artery. Circulation. 2006 Feb 14;113(6):842-50.

22. Binder RK, Rodes-Cabau J, Wood DA, Mok M, Leipsic J, De Larochelliere R, et al. Transcatheter aortic valve replacement with the SAPIEN 3: A new balloon-expandable transcatheter heart valve. JACC Cardiovasc Interv. 2013 Mar;6(3):293-300.

23. Goncalves A, Almeria C, Marcos-Alberca P, Feltes G, Hernandez-Antolin R, Rodriguez E, et al. Threedimensional echocardiography in paravalvular aortic regurgitation assessment after transcatheter aortic valve implantation. J Am Soc Echocardiogr. 2012 Jan;25(1):47-55.

24. Athappan G, Patvardhan E, Tuzcu EM, Svensson LG, Lemos PA, Fraccaro C, et al. Incidence, predictors, and outcomes of aortic regurgitation after transcatheter aortic valve replacement: Meta-analysis and systematic review of literature. J Am Coll Cardiol. 2013 Apr 16;61(15):1585-9.

25. Chieffo A, Buchanan GL, Van Mieghem NM, Tchetche D, Dumonteil N, Latib A, et al. Transcatheter aortic valve implantation with the edwards SAPIEN versus the medtronic CoreValve revalving system devices: A multicenter collaborative study: The PRAGMATIC plus initiative (pooled-RotterdAm-milanotoulouse in collaboration). J Am Coll Cardiol. 2013 Feb 26;61(8):830-6.

26. Raffa GM, Malvindi PG, Settepani F, Ornaghi D, Basciu A, Cappai A, et al. Aortic valve replacement for paraprosthetic leak after transcatheter implantation. J Card Surg. 2012 Jan;27(1):47-51.

119

CHAPTER

8 TRANSCATHETER AORTIC IMPLANTATION OF THE EDWARDS-SAPIEN BIOPROSTHESIS: INSIGHTS ON EARLY BENEFIT OF TAVR ON MITRAL REGURGITATION Int J Cardiol. 2011 Oct 6;152(1):124-6

Samim M1 Stella PR1 Agostoni P1 Kluin J2 Ramjankhan F2 Sieswerda G1 Budde R3 der Linden M1 Samim M1 Hillaert M1 van Herwerden L2 Doevendans PA1 van Belle E1 Department of Cardiology, 2Department of Cardiothoracic surgery, 3Department of

1

Radiology, University Medical Center, Utrecht, The Netherlands

CHAPTER 8

Mitral regurgitation is observed in about 2/3 of patients with aortic stenosis. Following conventional aortic valve replacement improvement of mitral regurgitation (MR) has been observed in about 50% of cases.1 The impact of transcatheter aortic valve replacement (TAVR) on MR is controversial. Two recent publications have reported improvement in MR grades following TAVR with the Edwards-SAPIEN valve system.2,3 These findings were not replicated with the Medtronic CoreValve system.4 Furthermore, the time course of improvement in MR grades after Edwards SAPIEN valve implantation has however not been described on an individual patient basis and the potential mechanisms of the benefit are unclear.

In the present study the potential benefit of TAVR with the Edwards SAPIEN valve on mitral regurgitation and insights on the mechanisms were studied by means of serial transthoracic echocardiography.

From September 2008 to September 2009, 22 consecutive patients with severe aortic stenosis, underwent TAVR with the Edwards SAPIEN prosthesis in our institution. All candidates were contraindicated for conventional surgery or considered a high surgical risk with an operative mortality risk of > 20% as assessed by at least 2 cardiovascular surgeons and 2 cardiologists.

The procedures were performed as previously described.2 Transthoracic echocardiography was performed pretreatment, post-treatment, and postdischarge, using a Philips 5500 (Philips, The Netherlands). MR was graded in 5 groups as none (= 0), trivial (= 1), mild (= 2), moderate (= 3), or severe (= 4).5 A valvular insufficiency ≥ 2 was considered significant. MR was defined as organic in the presence of calcifications or myxomatous degeneration of the mitral annulus and/or leaflets and functional in case of LV dysfunction and absence of morphological abnormalities of the mitral apparatus.4 For comparison between categorical variables a chi square test was used. A one-way ANOVA test was used for comparison between continuous variables. A one-way repeated measures Friedman test with post-hoc analysis was used to evaluate changes over time. A P-value of < 0.05 was considered statistically significant.

The baseline clinical and echocardiographic characteristics of the study population are presented in Table 1. Patients who underwent a transapical aortic valve replacement (TA-AVR) rather than a transfemoral (TF-AVR) procedure were more likely to have a history of coronary artery disease (p = 0.02). The procedure was successful in 21/22 patients (95%). The patient with an unsuccessful valve implant underwent emergent conventional surgery and died inhospital at day 11. Another patient died in hospital at day 3 from right ventricular failure secondary to an inferior myocardial infarction. Within the first 30 days, 2 additional patients died, one from pulmonary infection in the TA-AVI group and one from sudden death in the TF-AVI group. No stroke was observed. Eighteen patients were available for echocardiography after discharge.

The echocardiography findings of the 18 patients with an echocardiography at 1 month are summarized in Table 1. At baseline a significant MR was observed in 16/22 patients (73%, Table 1). It was mild, moderate or severe respectively in 8 patients (36%), 7 patients (32%) and 1 patient (4%). It was organic in 6 patients (27%) and functional in 10 patients (46%). The left ventricular (LV) end-diastolic diameter was larger in patients with a significant MR compared to those without (48.1 ± 4.4 mm vs 39.8 ± 1.7 mm; p = 0.001). The degree of MR before and

122

EARLY BENEFIT OF TAVR ON MITRAL REGURGITATION

Table 1 | Patients characteristics at baseline and 1 month follow-up Clinical and echocardiography characteristics at baseline Total N = 22

Transfemoral N = 10

Transapical N = 12

P

Age (year), mean ± SD

79 ± 7

78 ± 7

80 ± 8

0.61

Male, n (%)

11 (50)

4 (40)

7 (58)

0.66

BMI, mean ± SD

26 ± 5

26 ± 7

26 ± 4

0.85

Hypertension, n (%)

16 (73)

7 (70)

9 (75)

0.80

Atrial fibrillation, n (%)

10 (45)

5 (50)

5 (42)

0.70

Diabetes, n (%)

4 (18)

2 (20)

2 (17)

0.84

Coronary artery disease, n (%)

13 (59)

3 (30)

10 (83)

0.02

Prior myocardial infarction, n (%)

7 (32)

2 (20)

5 (42)

0.38

Percutaneous coronary intervention, n (%)

9 (41)

2 (20)

7 (58)

0.06

Coronary artery bypass, n (%)

6 (27)

2 (20)

4 (33)

0.64

Cerebrovascular events, n (%)

7 (32)

3 (30)

4 (33)

0.87

Peripheral vascular disease, n (%)

2 (9)

0 (0)

2 (17)

0.48

Renal disease, n(%)

5 (23)

2 (20)

3 (25)

0.78

COPD, n (%)

6 (27)

3 (30)

3 (25)

0.79

21.3 ± 14.1

18.1 ± 15.5

24.1 ± 12.8

0.33

5 (23)

2 (20)

3 (25)

0.78

LV End-diastolic diameter (mm)

45.7 ± 5.3

45.3 ± 6.1

46.2 ± 4.8

0.71

Aortic valve area (cm ), mean ± SD

0.73 ± 0.19

0.73 ± 0.17

0.73 ± 0.21

0.96

16 (73)

6 (60)

10 (83)

0.34

Organic, n (%)

6 (27)

2 (20)

4 (33)

0.78

Functional, n (%)

10 (46)

4 (40)

6 (50)

Logistic Euroscore, mean ± SD LVEF < 35%, n (%) 2

Significant MR, n (%) Type of MR

Echocardiography findings during follow-up (N = 18) Pre-treatment Post-treatment Post-discharge LV EF < 35%, n (%)

3 (23)

2 (15)

3 (17)

Transfemoral

1 (11)

1 (11)

1 (11)

Transapical

2 (22)

1 (11)

2 (25)

Transfemoral

45.0 ± 6.3

44.7 ± 5.5

44.3 ± 4.9

Transapical

46.4 ± 5.5

45.8 ± 5.6

45.1 ± 5.7

Mean Aortic gradient (mmHg)

42 ± 17

9±4

8±3

Transfemoral

41 ± 17

12 ± 4

9±3

Transapical

44 ± 18

7±2

7±2

Mitral regurgitation grade (0–4)

2.06 ± 0.94

1.51 ± 1.09

1.41 ± 0.92§

Transfemoral

1.89 ± 0.93

1.59 ± 0.93

1.23 ± 0.71

Transapical

2.22 ± 0.97

1.43 ± 1.22

1.59 ± 1.05

P* 0.57

LV end-diastolic diameter (mm) 0.03 0.0001

0.02

All parameters are mean ± SD, otherwise stated. P* for reflects the difference between the 3 time points by 1-way repeated measures Friedman test. § P < 0.05 compared to “Pre-treatment” by posthoc analysis of the Friedman test

123

CHAPTER 8

after Edwards SAPIEN valve implantation in the 18 patients with echocardiography follow-up is shown in Table 1 and in Figure 1A. An improvement in MR grades was observed in these patients (p = 0.02, Figure 1A). This benefit was secondary to a reduction in MR grades in 7 of the 12 patients (58%) with an MR at baseline without worsening in the other 5 patients with MR at baseline. No occurrence of MR was observed in the 6 patients without MR at baseline (Figure 1B). A trend for a greater improvement in MR grade was observed in patients with functional MR (n = 7, — 1.00 ± 1.0) compared to those with an organic MR (n = 5, — 0.29 ± 0.24; p = 0.10). Similarly a greater reduction in LV end-diastolic diameter was associated with a greater improvement in MR as illustrated by the higher reduction in LV end-diastolic diameter in the 6 patients with a reduction in MR compared to the 6 without reduction of MR (2.2 ± 1.1 mm vs 0.3 ± 0.9 mm; p = 0.009).

The present study is the first to demonstrate that the previously reported benefit in MR grades after implantation of the Edwards prosthesis is secondary to a reduction in MR grades in 50% of patients with MR at baseline without deleterious effects in the other 50% or in those without MR.2, 3 These results are in contrast with those recently reported with the CoreValve system.4 Two recent publications have reported improvement in MR grades following TAVR with the Edwards-SAPIEN valve system.2,

3

However in these 2 studies the improvement was not

investigated on an individual patient basis and the mechanism(s) of the benefit was unclear.

The present study confirms these previous observations and demonstrates that the benefit is secondary to a reduction in MR grades in 50% of patients with MR without deleterious effects on the other 50% or in those without MR. It further demonstrates that the benefit can be observed as early as 1 month post implantation and is associated with favourable LV diameter remodelling secondary to the disappearance of the aortic stenosis.

These results need also be interpreted in light of a recently published report with serial echocardiography follow-up of 46 patients who underwent implantation of the Medtronic CoreValve system.4 Although in their series an improvement in MR grades was observed in 25% of patients, it was counterbalanced by a worsening of MR in 22%. More importantly a new MR was observed in 42% of patients. This deleterious effect could be secondary to the design of the Medtronic CoreValve system in which the average depth of implantation is about 8 mm below the base of the aortic root and could alter the normal motion of anterior mitral leaflet.4 The Edwards SAPIEN valve, because of its minimal protrusion in the aortic root, might have no or minimal impact on mitral function, thus preventing induction or worsening of preprocedural MR.2 The lack of worsening of MR or occurrence of new MR in our series supports this hypothesis.

The present findings provide a potential explanation for the greater benefit of TAVR with the Edwards-SAPIEN valve in patients with MR as observed in the PARTNER trial.6 Finally, they provide additional support to the hypothesis that the Edwards SAPIEN valve and the Medtronic CoreValve system might have a different impact on the functionality of the mitral valve, due to the differences in their design. This issue will need to be tested in a randomized controlled trial.

124

EARLY BENEFIT OF TAVR ON MITRAL REGURGITATION

Figure 1 | A: Changes in mitral regurgitation in the 18 patients with echocardiography follow-up (p=0.02 by 1-way repeated measures Friedman test). B: Individual changes in mitral regurgitation from pre-treatment to post discharge.

125

CHAPTER 8

REFERENCES 1

2

3

4 5

6

126

Barreiro CJ, Patel ND, Fitton TP, Williams JA, Bonde PN, Chan V, et al. Aortic valve replacement and concomitant mitral valve regurgitation in the elderly: impact on survival and functional outcome.

Circulation 2005;112:I443–7.

Webb JG, Pasupati S, Humphries K, Thompson C, Altwegg L, Moss R, et al. Percutaneous transarterial aortic valve replacement in selected high-risk patients with aortic stenosis. Circulation 2007;116:755– 63.

Osten MD, Feindel C, Greutmann M, Chamberlain K, Meineri M, Rubin B, et al. Transcatheter aortic valve implantation for high risk patients with severe aortic stenosis using the Edwards Sapien balloonexpandable bioprosthesis: a single centre study with immediate and medium-term outcomes.

Catheter Cardiovasc Interv 2009.

Tzikas A, Piazza N, van Dalen BM, Schultz C, Geleijnse ML, van Geuns RJ, et al. Changes in mitral regurgitation after transcatheter aortic valve implantation. Catheter Cardiovasc Interv 2009;75:43–9.

Vahanian A, Baumgartner H, Bax J, Butchart E, Dion R, Filippatos G, et al. Guidelines on the management of valvular heart disease: The task force on the management of valvular heart disease of the European Society of Cardiology. Eur Heart J 2007;28:230–68.

Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, Tuzcu EM,Webb JG, Fontana GP, Makkar RR, Brown DL, Block PC, Guyton RA, Pichard AD, Bavaria JE, Herrmann HC, Douglas PS, Petersen JL, Akin JJ, Anderson WN, Wang D, Pocock S. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med;363:1597–607

127

PART THREE

ISCHEMIC BRAIN INJURY DURING TAVR

CHAPTER

9 SILENT ISCHEMIC BRAIN LESIONS AFTER TRANSCATHETER AORTIC VALVE REPLACEMENT: LESION DISTRIBUTION AND PREDICTORS Accepted for publication in Clinical Research in Cardiology

Mariam Samim1 Jeroen Hendrikse2 H. Bart van der Worp3 Pierfrancesco Agostoni1 Freek Nijhoff1 Pieter A. Doevendans1 Pieter R. Stella1 Department of Cardiology, 2Department of Radiology, 3Department of Neurology and

1

Neurosurgery, Brain Center Rudolf Magnus, University Medical Centre Utrecht, the Netherlands

CHAPTER 9

ABSTRACT Aims Silent ischemic brain lesions and ischemic stroke are known complications of transcatheter aortic valve replacement (TAVR). We aimed to investigate the occurrence and distribution of TAVR-related silent ischemic brain lesions using diffusion-weighted magnetic resonance imaging (DWI).

Methods Consecutive patients with severe aortic valve stenosis treated with TAVR underwent cerebral DWI within 5 days after the index procedure. DWI scans were analysed for the occurrence and distribution of new ischemic lesions post-TAVR. Results Forty-two patients were enrolled in this study. After TAVR, a total of 276 new cerebral ischemic lesions were detected in 38 (90%) patients, with a median of 4.5 [interquartile range, 2.0-7.0] lesions per patient. A total of 129 (47%) lesions were detected in the cortical regions, 97 (35%) in the subcortical regions, and 50 (18%) in the cerebellum or brainstem. The median lesion volume was 20.2 Âľl [10.0, 42.7] and the total ischemic lesion volume 132.3 Âľl [42.8, 336.9].

The new ischemic brain lesions were clinically silent in 37 (97%) patients; the other patient had a transient ischemic attack. Age (B=0.528, p=0.015), hyperlipidaemia (B=5.809, p=0.028) and post-dilatation of the implanted prosthesis (B=7.196, p=0.029) were independently associated with the number of post-TAVR cerebral DWI lesions. In addition, peak transaortic gradient was independently associated with post-procedural total infarct volume.

Conclusion Clinically silent cerebral infarcts occurred in 90% of patients following TAVR, most of which were small (<20 ul) and located in the cortical regions of the cerebral hemispheres. An independent association was found between age, hyperlipidaemia and balloon post-dilatation and the number of post-TAVR ischemic brain lesions. Only peak transaortic gradient was independently associated with post-procedural total infarct volume.

130

SILENT ISCHEMIC BRAIN LESIONS AFTER TRANSCATHETER AORTIC VALVE REPLACEMENT

INTRODUCTION More than ten years after its first introduction in 2002 by Cribier, transcatheter aortic valve replacement (TAVR) has evolved to the new standard-of-care treatment for patients with a severe symptomatic aortic stenosis considered inoperable or at high surgical risk. In spite of many improvements in the technique and the valve prostheses, questions remain related to the safety and durability of this technique. Significant concerns exist with regard to the incidence of cerebral complications after TAVR. Recent large multicenter series and national registries reported a rate of stroke or TIA of 3.3% in the first 30 days after TAVR, with the majority being major strokes (2.9%) 1. These rates are among the highest in the field of interventional cardiology. Furthermore, the average 30-day mortality in TAVR patients with post-procedural stroke was more than 3.5-fold higher than in non-stroke patients (25.5 vs. 6.9) 1. In addition to the clinically apparent ischemic brain lesions, several cerebral magnetic resonance imaging studies have shown a very high (58%-91%) incidence of new ischemic lesions after TAVR, regardless of the transcatheter valve type and approach (Table 1) 2-6. These small, usually asymptomatic, DWI lesions are most probably caused by micro-embolic particles arising during the valve implantation procedure. Hypoperfusion is thought to contribute to the occurrence of brain infarcts in TAVR patients only under the circumstances of cardiac arrest or severe systemic underperfusion. Several steps during the TAVR procedure may increase the embolic load towards the cerebral circulation. Previous studies using transcranial Doppler (TCD) have shown that cerebral microemboli are most common at the time of interaction of the prosthesis with the native valve, specifically during positioning and expansion of the prosthesis 7.

Table 1 | Previous studies investigating cerebral ischemic injury after TAVR using DWI Study

Number of patients

Arnold2

Valve Type

Approach

25

Sapien

TA

Ghanem

22

CoreValve

TF

Kahlert6

32

Sapien: [N=22) CoreValve : (N=10)

TF

Rodes-Cabau4

60

Sapien

Fairbairn5

31

Knipp20

12

Astarci15

35

3

Ischemic lesions

Number of lesions

68%

N/A

73%

2.5 [1.0-5.5]

Sapien: 86% CoreValve: 0%

Sapien: 4 [2.1-6.0] CoreValve: 2.6 [0.3-4.9]

TF (N=29), TA (N=31)

71%

TF: 3 [1-7] TA: 4 [2-9]

CoreValve

TF

77%

2 [1-5]

Sapien

TA

58%

1.8 Âą 1.9

Sapien

TF (N=21), TA (N=14)

TF: 90%, TA: 93%

TF: 4, TA: 4.5

Data are shown as median [interquartile range] or means Âą standard deviations TAVR: transcatheter aortic valve replacement; DWI: diffusion weighted magnetic resonance imaging; TA: transapical; TF: transfemoral; N/A: not available.

131

CHAPTER 9

Although the consequences of silent ischemic brain lesions are still uncertain, in some studies they have been associated with cognitive decline and an increased risk of dementia and depression 8-10. Furthermore, they have been shown to indicate an increased risk of subsequent stroke in patients with previous minor stroke and atrial fibrillation 10. In order to understand the causal mechanism of silent cerebral ischemic lesions and their consequence for the cerebral function, exploring the characteristics of concurrent lesions and their location in human brain is pivotal. The typically silent aspect of the majority of TAVR-related DWI lesions might be explained by their location in the brain and their size. Furthermore, data on risk factors associated with new silent ischemic brain lesions after TAVR are scarce. In this paper, we aimed to report our findings with regard to cerebral ischemic injury contributable to TAVR in patients undergoing this procedure at our center, as post-procedural cerebral DWI is part of standard patient care at our center. We intended 1) to investigate the topographic pattern of cerebral ischemic lesions on DWI in patients undergoing TAVR, 2) to evaluate the risk factors that may be associated with the number of new lesions after TAVR and 3) to evaluate the risk factors that may be associated with the total infarct volume.â&#x20AC;&#x192;

METHODS Patient population and TAVR procedure From September 2012 to September 2013, 65 consecutive patients with severe symptomatic aortic stenosis who underwent TAVR at our center were screened for inclusion in this prospective registry. The TAVR procedures were performed with the two currently commercially available bioprostheses, the Edwards SAPIEN XTâ&#x201E;˘ valve prosthesis (Edwards Lifesciences, Irvine, CA, USA) and the Medtronic CoreValve system (CoreValve Revalving Technology, Medtronic, Minneapolis, MN, USA). Patients were excluded when contraindications for MRI were present (such as a pacemaker). Between May and July 2013 a cerebral embolic protection device was used during TAVR in 15 consecutive patients and these patients were also excluded from this registry. Before the TAVR procedure, all patients were evaluated carefully for possible cardiac sources of embolism (atrial fibrillation, left ventricular thrombus, etc.) with the use of an electrocardiogram and echocardiography, which were mandatory in all patients as part of the pre-procedural TAVR evaluation protocol. After the TAVR procedures, all patients were evaluated, inter alia, for clinical signs of a new cerebrovascular event. The study was conformed to the guiding principles of the Declaration of Helsinki and patients consented to clinical evaluation. Because during the study period cerebral DWI was part of standard postprocedural care at our center, ethics approval was waived.

Diffusion weighted magnetic resonance imaging Magnetic resonance imaging was performed within four days after TAVR, using a 3 Tesla system (Philips Medical Systems, the Netherlands). The imaging protocol included a diffusion-weighted single-shot spin echo echoplanar sequence (diffusion gradient b values of 0 and 1000 s/mm2, repetition time (TR): 3307 ms, echo time (TE): 68 ms, 26 slices with a slice thickness of 4 mm, field

132

SILENT ISCHEMIC BRAIN LESIONS AFTER TRANSCATHETER AORTIC VALVE REPLACEMENT

of view: 230 mm, matrix: 256 x 205) and a turbo fluid attenuated inversion recovery (FLAIR; TR/ TE 11000/125 ms). The acquisition time for the diffusion weighted sequences was 69 seconds.

All DWI images were assessed by two skilled observers blinded to neurological status and procedure. Number, volume, location and vascular territories of all focal diffusion abnormalities (bright lesions on DWI), signifying acute cerebral ischemia, were documented. All supratentorial infarcts were divided into subcortical infarcts and cortical infarcts. Subcortical infarcts were further divided into deep grey matter infarcts, internal border zone infarcts and true subcortical infarcts. Deep grey matter infarcts were defined as lesions within the region of basal ganglia or thalamus. Internal border zone infarcts were defined as lesions located in the white matter of the centrum semiovale or corona radiate, at the border zone between lenticulostriate perforators and the deep penetrating cortical branches of the middle cerebral artery (MCA) or at the border zone of deep white matter branches of the MCA and the anterior cerebral artery. Cortical infarcts were subclassified into cortical border zone infarcts and cortical territorial infarcts. The distinction between cortical border zone and territorial locations was based on templates of arterial flow territories 11 (Figure 1). A cortical border zone territory was defined as the area between major cerebral arteries. Finally, infratentorial infarcts were divided into brainstem infarcts and cerebellar infarcts.

Preoperative assessment of calcification on contrast enhanced MSCT All patients underwent a contrast-enhanced, electrocardiogram-gated MSCT scan within three months before TAVR, as part of the pre-procedural work-up. All MSCT scans were analyzed using a software package and a dedicated 3D aortic valve analysis workflow (3mensio Valves TM, 3mensio Medical Imaging BV, The Netherlands, http://www.3mensio.com). Images of the aortic root reconstructed in systole, at 37.5% of the R-R interval were selected for analysis of the aortic plane and measurements of aortic valve calcification. Next, a centerline was automatically placed along the ascending aorta and mark points were placed at the anchor points of the three aortic leaflets, marking the aortic annulus. Subsequently, aortic leaflet calcification was measured in a region starting from the aortic annulus to 20 mm above that, marking the most cranial aspect of the aortic leaflets. Because of the inter-patient difference in the amount of contrast used for contrast-enhanced MSCT, the threshold for detecting calcifica¬tion in the aortic valve apparatus was chosen on an individual basis, using the method described by Mylonas 12 and colleagues for quantification of coronary artery calcification: for each patient, using axial images, a region of interest was placed in the ascending aorta 25 mm above the level of the aortic annulus. The mean aortic attenuation value (HU aorta) and standard deviation (SD) was measured at this level. Using these measures, the threshold for calcium detection was calculated as 2 SD above the mean attenuation in the aorta (HU Aorta + 2 SD).

Statistical Analysis Categorical data are presented as frequencies and percentages and compared between groups with Pearson chi-squared test or Fisher’s exact test. Continuous variables are presented as mean ± SD or medians and interquartile range [IQR] and compared between groups with the t-test or Mann-Whitney U test.

133

CHAPTER 9

Figure 1 | Cerebral arterial territory. ACA, anterior cerebral artery; ACHA, anterior choroideal artery; AICA, anterior inferior cerebellar artery; bBA, branches from basilar artery; bVA, branches from vertebral arteries; LSA, lenticulostriate arteries; MCA, middle cerebral artery; PCA, posterior cerebral artery; PICA: posterior inferior cerebellar artery; SCA, superior cerebellar artery.

134

SILENT ISCHEMIC BRAIN LESIONS AFTER TRANSCATHETER AORTIC VALVE REPLACEMENT

Linear regression analysis was used to identify the relationship between patient and procedural factors and the number and the total volume of new ischemic brain lesions following TAVR. Univariate analysis was used to identify individual predictors. Variables were reviewed to exclude collinearity in the multivariate model before testing. Variables with a univariate significance of p<0.2 were entered into a multivariate regression analysis to determine the independence of these predictors. A p-value of <0.05 was considered statistically significant.

RESULTS Patient and technical characteristics The flowchart of this study is shown in Figure 2. Of all screened patients, twenty-three were excluded (one due to claustrophobia and hence no MRI, 6 due to pre-TAVR permanent pacemaker therapy, 15 due to the use of a cerebral protection device during TAVR and one due to post-operative hemodynamic instability). Forty-two patients (65%) were included in this study, of whom 26 received the balloon-expandable Edwards SAPIEN XTâ&#x201E;˘ valve prosthesis and 16 the self-expandable Medtronic CoreValve system. Clinical and procedural characteristics are presented in table 2. Pre-procedural atrial fibrillation was observed in 15 patients (36%).

Transthoracic echocardiography showed no signs of ventricular thrombus in all cases.

Device success as defined by VARC-2 definitions 13 was achieved in 36 patients (86%). Reasons for not fulfilling the device success criteria were: additional valve implantation in one patient due to too low placement of a Medtronic CoreValve prosthesis and hence severe paravalvular regurgitation, and residual moderate paravalvular regurgitation after valve implantation in 5 other patients (3 Edwards SAPIEN XTâ&#x201E;˘ and 2 Medtronic CoreValve prostheses).

Figure 2 | Study flow diagram.

135

CHAPTER 9

Cerebral ischemic brain lesions on DWI

Table 2 | Baseline and procedural characteristics

Post-procedural DWI findings are shown

Patient characteristics

N = 42

in Table 3 and Table 4. Representative DWI

Age (years)

82 [78, 84]

examinations are depicted in Figure 3. After

Female sex

17 (40)

TAVR, a total of 276 new ischemic lesions

Diabetes mellitus

11 (26)

Hypertension

24 (57)

Hyperlipidaemia

12 (29)

Coronary artery disease

30 (71)

Previous acute myocardial infarction

7 (17)

were detected in 38 patients (90%), with a median of 4.5 [IQR, 2.0-7.0] lesions per patient. These foci were typically multiple and dispersed in both hemispheres, suggesting a cardio-embolic or aortic origin of these lesions. Of all lesions, 226 (82%) were supratentorial and 50 (18%) infratentorial (Table 3B). Moreover, 129 (57%) of the supratentorial lesions were located in the left cerebral hemisphere and 97 (43%) in the right hemisphere (p=0.03). The majority (129; 57%) of these supratentorial lesions was located in the cortical regions: 24 (19%) in the territory of the anterior cerebral artery, 49 (38%) in that of the middle cerebral artery, 24 (19%) in that of the posterior cerebral artery, and 32 (25%) in the cortical border zones.

Of all infratentorial lesions, 46 (92%) was

Previous PCI

24 (57)

Previous CABG

11 (26)

Atrial fibrillation

15 (36)

Carotid disease

3 (7)

Peripheral vascular disease

5 (12)

Prior stroke or TIA

7 (17)

Logestic EuroSCORE (%)

13.1 [8.3, 18.1]

Procedural characteristics Access site Femoral artery

37 (88)

Apex

4 (10)

Subclavian artery

1 (2)

Prosthesis type

located in the cerebellum and 4 (8%) in the

Edwards SAPIEN

26 (62)

brainstem (pons/midbrain).

Medtronic CoreValve System

16 (38)

The median lesion volume was 20.2 µl

Prosthesis size, mm, n (%)

[10.0, 42.7] and the median total ischemic

23

6 (14)

volume 132.3 µl [42.8, 336.9]. Half of the

26

19 (45)

lesions (138) had a volume of 20 µl or

29

10 (24)

31

7 (17)

smaller (Figure 4). New DWI lesions were found more frequently in patients treated with the Medtronic CoreValve than in patients treated with the Edwards SAPIEN XT™ ( 6.0 [2.0, 13.5] vs. 4.0 [1.0, 6.3], P=0.05; Table 3).

136

Pre-dilatation

42 (100)

Post-dilatation

7 (17)

Data are shown as median [interquartile range] or n (%).

SILENT ISCHEMIC BRAIN LESIONS AFTER TRANSCATHETER AORTIC VALVE REPLACEMENT

Demographic

Table 3 | Findings on post-TAVR DWI N= 42 patients

and

procedural

risk

factor assessment

Patient with new lesions

38 (90)

Risk factors for a higher number and

Patient with single lesion

4 (10)

greater volume of new ischemic brain le-

Patients with multiple lesions

34 (81)

sions are reported in Table 5 and Table

Lesions per patient

4.5 [2.0, 7.0]

Right side of the brain

2.0 [0.0, 4.0 ]

Left side of the brain

2.0 [1.0, 5.0]

Edwards SAPIEN

4.0 [1.0, 6.3]

Medtronic CoreValve

6.0 [2.0, 13.5]

Lesion volume (Âľl) Mean per patient

25.5 [10.3, 65.8]

6, respectively. In the univariate analysis, there were no associations between the presence of diabetes mellitus, hypertension, hyperlipidaemia, coronary artery disease, atrial fibrillation, carotid disease, peripheral vascular disease and prior stroke or TIA, and the number or the total volume of new ischemic brain lesions after TAVR.

Total infarct volume

132.3 [42.8, 336.9]

Edwards SAPIEN

116.9 [18.4, 398.4]

Medtronic CoreValve

163.6 [52.5, 350.2]

TAVR (peak gradient; [R=0.412, p=0.008]

Right side of the brain

60.3 [0.0, 136.2]

volume of aortic valve calcification was not

Left side of the brain

67.6 [8.8, 185.2]

related to the number or the total volume

Data are shown as median [interquartile range] or n (%).

Pre-procedural transaortic gradients were associated with both the number and the volume of cerebral ischemic lesions after and [R=0.415, p=0.008], respectively). The

of new ischemic brain lesions (p=0.142 and p=0.443, respectively). Also none of the procedural characteristics were associated with either the number or the total volume of cerebral ischemic lesions.

Figure 3 | DWI images of three different patients four days after TAVR showing multiple small acute ischemic lesions in different territories. A, two micro-infarcts located in the territory of the right temporal lobe (arrow head) and the right cerebellar hemisphere (arrow). B, small DWI lesion located in the putamen (arrow). C, multiple cortical (arrow) and subcortical (arrowhead) micro-infarcts.

137

CHAPTER 9

Figure 4 | Incidence of different sizes of cerebral DWI lesions after TAVR. DWI lesions were categorized according to volume.

Half of the lesions was 20 µl or smaller. Only 14 (5%) of the 276 lesions were larger than 140 µl.

The univariate variables (age, hyperlipideamia, coronary artery disease, peak transaortic gradient, aortic valve calcification and post-dilatation) were then entered into a multivariate regression model, where age, hyperlipideamia and post-dilatation of the implanted valve prosthesis remained independent predictors of the number of new ischemic brain lesions (p=0.015, p=0.028, p=0.029, respectively) (Table 5 ). Peak transaortic gradient was the only variable independently associated with the total volume of new ischemic brain lesions (p=0.039) (Table 6).

Neurological performance During in-hospital follow-up, no patient developed a new focal neurological deficit suggestive of stroke, except for one patient with temporary dysphasia, which appeared 30 minutes after the procedure, and was diagnosed as TIA. Cerebral MRI performed after TAVR confirmed a new ischemic lesion (40µl) in the site corresponding to the symptoms in this patient (Wernicke’s area).

DISCUSSION The present study adds valuable information to the existing knowledge about silent cerebral ischemic lesions after TAVR procedures. We found new ischemic lesions on post-procedural DWI in 90% of all cases. In this series, half of these cerebral ischemic lesions was 20 µl or smaller and about half were located in the cortex of the cerebral hemispheres. As these lesions were small, they were less likely to lead to new focal neurological deficits14. That is probably also part of the explanation for the absence of convincing evidence for a relationship between the size or the number of new ischemic cerebral lesions on post-TAVR DWI and clinical neurocognitive deficits.

138

SILENT ISCHEMIC BRAIN LESIONS AFTER TRANSCATHETER AORTIC VALVE REPLACEMENT

Table 4 | Lesion distribution and frequency of microembolism DWI lesions (N=276) Supratentorial infarcts Cortical infarcts

226 (82) 129 (57)

Anterior cerebral artery

24 (19)

Middle cerebral artery

49 (38)

Posterior cerebral artery

24 (19)

Border zone

32 (25)

Subcortical infarcts

The

majority

of

the

supratentorial

lesions were located in the left cerebral hemisphere. A comparable observation was reported by a two other reports

3, 5,

, with the largest portion of DWI lesions

15

located on the left side of the brain. Two different hypotheses might explain this finding: 1) three major arteries, originating from the aortic arch, are at risk of receiving embolic

particles

from

the

calcified

97 (43)

aortic valve and ascending aorta: the

68 (70)

brachiocephalic, the left common carotid

Anterior cerebral artery

8 (12)

and the left subclavian arteries. The last

Middle cerebral artery

14 (21)

two major arteries, mainly supplying the

Posterior cerebral artery

9 (13)

left side of the brain, might together receive

Border zone

37 (54)

twice as many emboli as compared to the

Internal border zone

16 (16)

brachiocephalic artery, supplying the right

Deep grey matter

13 (13)

True subcortical infarcts

Nucleus caudatus

9 (69)

Putamen

1 (8)

Thalamus

3 (23)

Infratentorial infarcts

50 (18)

Cerebellum Brainstem (pons/midbrain)

46 (92) 4 (8)

Data are shown as n (%).

side of the brain and the right arm; 2) in addition, the left common carotid and the left subclavian arteries also receive emboli dislodged from the aortic arch due to the scratching of the aortic wall by the catheters and the devices. None of these hypotheses are supported by evidence to date and future studies focusing on the embolization process, for instance with the use of TCD during TAVR, are needed. The

higher incidence of cerebral DWI lesions at the left side of the brain raises the question of the usefulness of cerebral protection devices, designed specifically for usage during TAVR, that do not protect efficiently the left-sided branch arteries of the aortic arch. For instance, the Embrella Embolic Deflector System (Edwards Lifesciences Ltd., Irvine, CA, USA) and the Claret Medical protection devices lack complete coverage of the left vertebral arterial system, which may reduce the usefulness of these types of cerebral protection devices. Future refinements to cerebral protection devices may offer full coverage of all three branch arteries of the aortic arch and hence better cerebral protection.

With regard to potential risk factors for the development of new cerebral infarcts during TAVR, Fairbairn5 and colleagues reported old age and atherosclerotic burden of the aorta as independent predictors of the number of new cerebral ischemic lesions after these procedures. Astarci et al. also found an association between age and the number of postTAVR cerebral DWI lesions. In line with these previous reports our study found an independent association between age and the number of post-TAVR cerebral DWI lesions. In addition, our study showed an independent association between hyperlipidaemia at baseline and post-

139

CHAPTER 9

dilatation after prosthesis implantation and the number of new ischemic brain lesions. The association between post-dilatation of the aortic valve prosthesis might be due to an increased interaction between the stent frame of the valve prosthesis and the native aortic valve, which might indeed favor the dislodgment of calcific particles from the native valve. One previous study by Nombela-Franco and colleagues 16 reported that further stretching of the calcified native valve during balloon post-dilatation is independently associated with a 2-fold risk of cerebrovascular events immediately or within the first few hours after the procedure. With regard to procedural risk factors, two previous studies 17 showed that high-intensity transient signals (HITS) observed with TCD occurred during all procedural intervals in TAVR, however the embolic events appeared to peak during prosthesis deployment. Transient expansion and

Table 5 | Linear regression analysis for the prediction of the number of new cerebral infarcts (on post-TAVR DWI) Univariate

Multivariate

R

P Value

B

t

P Value

Age (years)

0.230

0.143

0.528

2.571

0.015

Logistic EuroSCORE (%)

-0.131

0.408

Diabetes mellitus

0.127

0.422

Hypertension

0.172

0.275

Hyperlipidaemia

0.237

0.131

5.809

2.306

0.028

Coronary artery disease

-0.251

0.108

1.841

0.708

0.484

Atrial fibrillation

-0.125

0.429

Carotid disease

-0.123

0.436

Peripheral vascular disease

-0.079

0.620

Prior stroke or TIA

-0.167

0.292

0.412

0.008

0.095

1.639

0.111

Mean transaortic gradient (mm Hg)

0.372

0.021

AVA (cm2)

<0.001

0.999

-0.236

0.142

-0.001

-1.757

0.089

Success device

0.022

0.892

Fluoroscopy time (min)

-0.106

0.514

Transfemoral

0.120

0.451

Transapical

-0.081

0.611

Transsubclavian

-0.098

0.536

Post-dilatation

0.258

0.099

7.196

2.295

0.029

Comorbidities

Pre-TAVR echocardiography Peak transaortic gradient (mm Hg)

Pre-TAVR MSCT Aortic valve calcification (ul) Procedural characteristics

140

SILENT ISCHEMIC BRAIN LESIONS AFTER TRANSCATHETER AORTIC VALVE REPLACEMENT

recoil of a metallic stent frame within a partially disrupted native valve (due to pre-dilatation) is probably a particularly efficient way to generate embolic particles, although the prosthesis itself may provide some embolic protection. In regard to risk factors associated with the total infarct volume after TAVR, only peak transaortic gradient at baseline was found to be independently associated. One previous report by Kahlert et al, studying cerebral embolization during TAVR by use of TCD, identified mean transaortic gradient at baseline as an independent predictor for the number of HITS during TAVR. These findings may be expected as a higher transaortic gradient often reflects a more severe aortic valve stenosis, which is often accompanied by a higher amount of aortic valve calcification. The latter increases the risk of dislodgment of calcific microdebris from the degenerative leaflets.

Table 6 | Linear regression analysis for the prediction of the total volume of new cerebral infarcts (on post-TAVR DWI) Univariate

Multivariate

R

P Value

B

t

P Value

Age (years)

0.102

0.521

Logistic EuroSCORE (%)

-0.170

0.281

Diabetes mellitus

-0.057

0.721

Hypertension

0.176

0.264

Hyperlipidaemia

0.279

0.074

Coronary artery disease

-0.236

0.132

2.573

1.053

0.299

-1.716

-0.668

0.509

Atrial fibrillation

-0.170

0.281

Carotid disease

-0.085

0.593

Peripheral vascular disease

-0.054

0.736

Prior stroke or TIA

-0.135

0.393

Peak transaortic gradient (mm Hg)

0.415

0.008

0.128

Mean transaortic gradient (mm Hg)

0.398

0.013

2.141

0.039

AVA (cm2)

-0.004

0.982

-0.125

0.443

Success device

0.007

0.964

Fluoroscopy time (min)

-0.190

0.240

Transfemoral

0.047

0.769

Transapical

-0.031

0.847

Transsubclavian

-0.040

0.800

Post-dilatation

-0.007

0.966

Comorbidities

Pre-TAVR echocardiography

Pre-TAVR MSCT Aortic valve calcification (ul) Procedural characteristics

141

CHAPTER 9

Catheter manipulation of the calcified stenotic valve increases this risk, and it has been shown previously that cerebral embolism as detected by DWI occurs even after valve passage with relatively soft diagnostic catheters 18. In our study however, aortic valve calcification was not significantly associated with either the number or volume of the new cerebral ischemic lesions.

Until now, no convincing evidence has related the size or the number of new ischemic lesions on post-TAVR brain imaging to clinical neurocognitive deficits. Although the clinical relevance of these DWI lesions remains uncertain, they may cause decline in cognitive function as suggested by previous reports on silent cerebral lesions involving patients other than the typical TAVR patients 8, 9, 19. For instance, the population-based Rotterdam Scan Study showed that elderly people with silent brain infarcts have an increased risk of dementia and a steeper decline in cognitive function than those without such lesions 8, 9, 19. Therefore based on these previous reports, the high frequency of ischemic lesions on post-TAVR cerebral imaging calls for procedural and technical developments to reduce the risk of peri-procedural embolization.

Less traumatic devices, avoidance of extensive manipulation of the calcified aortic valve, omission of post-dilatation after prosthesis implantation, and use of effective cerebral protection devices are currently under consideration. Furthermore, meticulous attention must be paid to valve preparation and thorough sheath and catheter flushing to reduce thrombus formation and gaseous embolization.

CONCLUSION Multiple, clinically silent, small ischemic brain lesions was detected on post-TAVR DWI in 90% of patients, with half of the lesions being very small (<20 ul) and half located in the cortical regions of the cerebral hemispheres. An independent association was found between age, hyperlipidaemia at baseline and balloon post-dilatation and the number of post-TAVR ischemic brain lesions. Only peak transaortic gradient was independently associated with post-procedural total infarct volume.

LIMITATIONS First, the number of patients included in this study is limited, affecting the sensitivity to find independent predictors of the occurrence of ischemic brain lesions. Second, some (clinical) factors that might be associated with the extent of cerebral ischemic injury during TAVR, such as blood pressure change during the procedure and the presence of aortic atheroma could not be assessed sufficiently due to the retrospective design of this study. In addition, our study lacks data on cognitive performance after TAVR.â&#x20AC;&#x192;

142

SILENT ISCHEMIC BRAIN LESIONS AFTER TRANSCATHETER AORTIC VALVE REPLACEMENT

REFERENCES 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

Eggebrecht H, Schmermund A, Voigtlander T et al. Risk of stroke after transcatheter aortic valve implantation (TAVI): a meta-analysis of 10,037 published patients. EuroIntervention 2012;8(1):129-38.

Arnold M, Schulz-Heise S, Achenbach S et al. Embolic cerebral insults after transapical aortic valve implantation detected by magnetic resonance imaging. JACC Cardiovasc Interv 2010;3(11):1126-32.

Ghanem A, Muller A, Nahle CP et al. Risk and fate of cerebral embolism after transfemoral aortic valve implantation: a prospective pilot study with diffusion-weighted magnetic resonance imaging. J Am Coll Cardiol 2010;55(14):1427-32.

Rodes-Cabau J, Dumont E, Boone RH et al. Cerebral embolism following transcatheter aortic valve implantation: comparison of transfemoral and transapical approaches. J Am Coll Cardiol 2011;57(1):18-2.

Fairbairn TA, Mather AN, Bijsterveld P et al. Diffusion-weighted MRI determined cerebral embolic infarction following transcatheter aortic valve implantation: assessment of predictive risk factors and the relationship to subsequent health status. Heart 2012;98(1):18-23.

Kahlert P, Knipp SC, Schlamann M et al. Silent and apparent cerebral ischemia after percutaneous transfemoral aortic valve implantation: a diffusion-weighted magnetic resonance imaging study.

Circulation 2010;121(7):870-8.

Reinsfelt B, Westerlind A, Ioanes D et al. Transcranial Doppler microembolic signals and serum marker evidence of brain injury during transcatheter aortic valve implantation. Acta Anaesthesiol Scand 2012;56(2):240-7.

Knecht S, Oelschlager C, Duning T et al. Atrial fibrillation in stroke-free patients is associated with memory impairment and hippocampal atrophy. Eur Heart J 2008;29(17):2125-32.

Vermeer SE, Prins ND, den Heijer T et al. Silent brain infarcts and the risk of dementia and cognitive decline. N Engl J Med 2003;348(13):1215-22.

Silent brain infarction in nonrheumatic atrial fibrillation. EAFT Study Group. European Atrial Fibrillation Trial. Neurology 1996;46(1):159-65.

Tatu L, Moulin T, Vuillier F et al. Arterial territories of the human brain. Front Neurol Neurosci 2012;30:99-110.

Mylonas I, Alam M, Amily N et al. Quantifying coronary artery calcification from a contrast-enhanced cardiac computed tomography angiography study. Eur Heart J Cardiovasc Imaging 2013.

Kappetein AP, Head SJ, Genereux P et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. J Thorac Cardiovasc Surg 2013;145(1):6-23.

Hassell ME, Nijveldt R, Roos YB et al. Silent cerebral infarcts associated with cardiac disease and procedures. Nat Rev Cardiol 2013;10(12):696-70.

Astarci P, Glineur D, Kefer J et al. Magnetic resonance imaging evaluation of cerebral embolization during percutaneous aortic valve implantation: comparison of transfemoral and trans-apical approaches using Edwards Sapiens valve. Eur J Cardiothorac Surg 2011;40(2):475-9.

Nombela-Franco L, Webb JG, de Jaegere PP et al. Timing, predictive factors, and prognostic value of cerebrovascular events in a large cohort of patients undergoing transcatheter aortic valve implantation. Circulation 2012;126(25):3041-53.

Erdoes G, Basciani R, Huber C et al. Transcranial Doppler-detected cerebral embolic load during transcatheter aortic valve implantation. Eur J Cardiothorac Surg 2012;41(4):778,83; discussion 783-4.

Omran H, Schmidt H, Hackenbroch M et al. Silent and apparent cerebral embolism after retrograde catheterisation of the aortic valve in valvular stenosis: a prospective, randomised study. Lancet 2003;361(9365):1241-6.

Daneault B, Kirtane AJ, Kodali SK et al. Stroke associated with surgical and transcatheter treatment of aortic stenosis: a comprehensive review. J Am Coll Cardiol 2011;58(21):2143-50.

Knipp SC, Kahlert P, Jokisch D et al. Cognitive function after transapical aortic valve implantation: a single-centre study with 3-month follow-up. Interact Cardiovasc Thorac Surg 2013;16(2):116-22.

143

CHAPTER

10 FIRST-IN-MAN EXPERIENCE WITH A NEW EMBOLIC DEFLECTION DEVICE IN TRANSCATHETER AORTIC VALVE INTERVENTIONS EuroIntervention 2012 May 15;8(1):51-6.

Kevin Onsea1 Pierfrancesco Agostoni1 Mariam Samim1 Michiel Voskuil1 Jolanda Kluin2 Ricardo Budde3 Jeroen Hendrikse3 Faiz Ramjankhan2 Jan van Klarenbosch4 Pieter Doesburg4 Gertjan Sieswerda1 Pieter Stella1 K. Onsea and M. Samim have contributed equally to this paper.

Department of Cardiology, 2Department of Cardiothoracic Surgery, 3Department of Radiology,

1

Department of Anaesthesiology, University Medical Center Utrecht, Utrecht, The Netherlands

4

CHAPTER 10

ABSTRACT Aims To report our first-in-man experience with a new cerebral embolic deflection device (SMT Embolic Deflection Device) during transcatheter aortic valve implantation (TAVI). A significant number of strokes and brain infarcts are caused by embolisation of atherosclerotic material, clots and other debris during various phases of invasive cardiac procedures, especially TAVI. The application of a temporary filter in the aortic arch averting dislodged emboli from entering the cerebral circulation might prevent this.

Methods and results In 15 patients (mean age 79 years) with severe aortic stenosis undergoing percutaneous transfemoral or transapical aortic valve implantation, the SMT Embolic Deflection Device was advanced utilising the contralateral femoral artery access using a 9 Fr delivery sheath. Once deployed in the aortic arch, a porous membrane shields the supraaortic-cerebral trunks by deflecting emboli away from the cerebral circulation. Embolic material is not contained or removed by the device. A 6 Fr pigtail catheter can be used through the same sheath throughout the whole procedure. Brain diffusion weighted (DW)-MRI was obtained in 10 patients before and at 4 days after (Âą2 days) the procedure and retrospectively compared to 20 patients previously undergoing TAVI without a protection device. Successful placement of the embolic protection device was achieved in all patients. Additional procedural time due to the use of the device was 7 min (Âą2 min). There were no procedural complications. No patient developed new neurological symptoms or clinical findings of stroke except one patient who suffered from a transient ischaemic attack (TIA) two days after the procedure. DW-MRI showed 3.2 new cerebral lesions per patient, compared to 7.2 new lesions per patient in the group without SMT filter.

Conclusions In this first-in-man experience, the feasibility of a new embolic deflection device is demonstrated.

Larger randomised, prospective studies are required to confirm these findings and prove safety and efficacy by reducing the incidence of cerebral embolism and stroke after TAVI. â&#x20AC;&#x192;

146

SMT EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

INTRODUCTION Transcatheter aortic valve implantation (TAVI) has emerged as a treatment option for inoperable or high-risk surgical patients with severe aortic stenosis. While TAVI reduces 1-year mortality as compared to conservative treatment and reduces perioperative morbidity as compared to surgery with similar mortality outcomes, there is an increased risk of neurological complications. In high-risk surgical candidates in the PARTNER trial, TAVI was associated with an approximately two-fold increased incidence of stroke or transient ischaemic attack (TIA) (5.5% vs. 2.4%, p=0.04) compared with surgical aortic valve replacement (SAVR) at 30 days1.

In contrast to percutaneous treatment of coronary artery disease which has managed to achieve a very low incidence (0.22%) of periprocedural stroke2, the percutaneous treatment of a diseased aortic valve clearly still comes at a higher price. The passage of bulky devices through a diseased aorta3 and the aggressive manipulation of a stenotic, calcified aortic valve may cause embolisation of atherosclerotic material, clots and other calcified debris. Indeed, as already previously demonstrated during diagnostic angiogram as work up for SAVR, left catheterisation with retrograde crossing of the aortic valve has been associated with a higher incidence of new focal diffusion-imaging abnormalities on magnetic resonance imaging (MRI) as compared to catheterisation without passage of the valve, with new lesions in 22% of patients (suggesting new cerebral embolic events), and a 3% rate of clinically relevant neurological deficits4. A study utilising transcranial Doppler ultrasound during TAVI demonstrated the occurrence of cerebral microemboli in each of 52 patients, mainly during direct manipulation of the diseased valve and crushing of the leaflets during implantation5. Only a minority of these patients suffered from clinical stroke or TIA, but new foci of restricted cerebral perfusion on diffusion weighted (DW)-MRI are reported in 58% to 91% of TAVI patients6,7,8,9,10. The clinical impact of these new, asymptomatic DW-MRI lesions is uncertain but, in epidemiological studies, their presence has been associated with declines in cognitive and physical function, frailty, development of dementia and an increased risk of subsequent stroke11. Embolic protection devices have already demonstrated their efficacy in the interventional treatment of saphenous vein graft disease12 and carotid lesions13. The application of a similar concept during TAVI could minimise both clinical and sub-clinical cerebral emboli. We report our first-in-man experience with the SMT Embolic Deflection Device (SMT Research and Development Ltd., Herzliya, Israel), a retrievable low-profile filter, designed for percutaneous delivery to the aortic arch, for the prevention of stroke caused by emboli originating in the heart and aorta.

METHODS Patients The SMT filter was used during aortic valve intervention in 15 patients with symptomatic severe aortic stenosis. These patients were formally discussed in the heart team and considered inoperable or at high risk for SAVR due to age, a high logistic EuroSCORE, porcelain aorta, malignancy, frailty or severe comorbidities. Informed consent for use of the SMT device was obtained.

147

CHAPTER 10

SMT Embolic deflection device (figure 1 and figure 2) The SMT Device is a sterile, biocompatible filter, implanted in a transcatheter procedure, through a needle puncture in either common femoral artery, and is located under fluoroscopy in the aortic arch. While the filter does not block or decrease normal blood flow to the brain via the aortic branches and vertebral artery, it diverts emboli/particulate matter downstream, where they can be treated effectively or probably cause less harm, although clinical impact on kidney and other end-organ function has to be further established. The SMT device is made of fine nitinol #1 (nickel titanium alloy) wires, which exhibit superelasticity so that the filter can be crimped into an 8 or 9 Fr sheath, the latter providing the possibility to use simultaneously a 6 Fr pigtail through the same sheath, hence avoiding additional groin punctures. Upon deployment the filter unfolds and regains its original shape The SMT device consists of 5 functional parts (figure 1 and figure 2) 1. A dual wire nitinol frame that anchors the device in the desired location in the aortic arch.

2. A thin nitinol mesh designed to allow blood flow through, while diverting clinically significant emboli towards the descending aorta.

3 and 4. Two stabilisers that facilitate the positioning of the filter. They lock into position by retrograde traction and gradual release from the delivery shaft. The filter is anchored in the aortic arch by the upper stabiliser in the innominate artery ostium which prevents filter retrograde migration and by the lower stabiliser that push the filter in apposition with the upper wall of the aortic arch.

5. The tail end (distal from the heart) of the SMT filter is a connection, by which the SMT filter remains securely attached to the plunger (â&#x20AC;&#x153;pusherâ&#x20AC;?) during the procedure.

The filter is coated to shield the blood from the underlying medical device material. The chemical and physical properties of the coating reduce the likelihood for blood components to adhere and activate, thus reducing the formation of thrombus or emboli. Interestingly, the SMT filter was originally developed as a permanent surgical implant for patients with high risk of stroke by cardiac emboli, however subsequently technically modified for temporary transcatheter implants.

Delivery system The delivery system includes the following components: The 9 Fr delivery sheath (Cook Medical, Bloomington, IN, USA) is a 75 cm long catheter, curved at the distal end. The direction of the curve is opposite to the sidearm fitting, which allows flushing the sheath even when a catheter or the plunger is inside.

The dilator is a matching tube (also part of the Cook Medical set) with a 1 mm lumen (for a guidewire). It fits snugly into the 9 Fr delivery sheath and has a female luer connector at the proximal end, for rinsing.

The loading tube with a female luer hub at the proximal end is used for crimping and loading the SMT filter. It connects to the delivery sheath by being inserted (the smooth distal end) into the septum hub. It is attached to a 3-way Y-connector homeostasis valve at its proximal end to prevent blood dripping and allow access to the delivery sheath.

148

SMT EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

The crimper is a funnel that enables easy loading of the SMT device into the cartridge. It mounts on the distal end of the cartridge prior to loading, and it is removed after loading.

The plunger (delivery shaft) is a flexible wire used for advancing the filter from the cartridge into the delivery sheath and through it to the site of deployment. It is also used for retrieving the SMT filter back into the 9 Fr delivery sheath.

Figure 1 | SMT embolic deflection device. The SMT device consists of 5 functional parts: a dual wire nitinol frame (white arrow), a thin Nitinol mesh (asterisk), an upper and lower stabiliser (black arrows) and the tail end (dotted arrow) (see text for further detail).

Figure 2 | A) SMT device attached to plunger. The SMT filter upon deployment, attached to the plunger of the delivery system (arrow) inside a 9 Fr sheath (arrowhead) and with simultaneous advancement of a 6 Fr pigtail catheter (dotted arrow). B) Animated illustration of a deployed SMT device in the aortic arch.

149

CHAPTER 10

Figure 3 | Fluoroscopic images of an SMT device in the aortic arch during transfemoral TAVI. The SMT filter (arrow) is deployed in the aortic arch at the beginning of the procedure through a 9 Fr delivery sheath (asterisk). Unhampered passage of the Edwards SAPIEN valve (arrowhead) through the aortic arch via the Retroflex delivery system (Edwards Lifesciences, Inc., Irvine, CA, USA) (dotted arrow). After successful aortic valve implantation (arrowhead), the SMT device (arrow) is retrieved in the 9 Fr delivery sheath (asterisk).

Aortic valve implantation All TAVI procedures were routinely performed by a team, consisting of 2 interventional cardiologists, one cardiologist specialised in echocardiography, one cardiac surgeon, one anaesthetist and 2 specialised nurses. After deployment of the SMT filter (Figure 3), an Edwards SAPIEN (Edwards Lifesciences, Inc., Irvine, CA, USA) transcatheter aortic valve was implanted in 15 patients, who were all pretreated with aspirin and clopidogrel and received heparin during the procedure in order to maintain an activated clotting time above 250 ms. Nine patients were approached via the transfemoral way, the remaining 6 underwent a transapical procedure according to standard techniques14,15. The size of the valve was chosen according to echocardiographic estimates of aortic annulus diameter and varied between 23 and 26 mm. All procedures were done under general anaesthesia and with transoesophageal echocardiography guidance. The stenotic valve was predilated with an undersized balloon to facilitate valve implantation and subsequent valve deployment was accomplished during rapid pacing in the right ventricle. If necessary, additional post-dilation was performed in case of relevant paravalvular regurgitation. At the end of the procedure, the SMT filter was retrieved into the delivery sheath and removed.

MRI MRI scanning of the brain before TAVI and <1 week after was performed in 10 patients undergoing TAVI with cerebral protection device and retrospectively compared to 20 patients who had undergone TAVI without cerebral protection previously in our institution. Scanning was performed according to current standard practice. The MRI scan included a diffusion weighted imaging (DWI) scan sequence. Scans were scored by two blinded observers in consensus. New ischaemic lesions were defined as new areas of high signal intensity on the post-procedural DWI images. On each scan the number of lesions with high signal intensity on DWI images (representing ischaemic lesions) was counted.

150

SMT EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

Statistical analysis Due to the observational single arm nature of this registry, only descriptive statistics has been performed. Data are presented as mean ±standard deviation if continuous or as number (percentage) if dichotomous.

RESULTS Baseline patient characteristics (table 1) The SMT device was used in 15 patients, 11 women and four men, with a mean age of 79.3 years (±9.9). Risk factors for stroke included previous stroke/TIA (5 patients), severe peripheral vascular disease (6 patients) with presence of a porcelain aorta in 3 patients, diabetes mellitus (4 patients), hypertension (10 patients) and dyslipidaemia (10 patients). Three patients underwent percutaneous coronary intervention at least 1 month before TAVI because of unstable angina or myocardial infarction and 3 patients had a history of CABG. One patient had a very poor left ventricular systolic function (ejection fraction <30%).

Procedural and clinical outcome (table 1) In all cases, the SMT filter was deployed without complications across the aortic arch via a 9 Fr femoral arterial sheath at the beginning of the procedure and withdrawn at the end of the procedure (demonstrated in Figure 3). No vascular or bleeding complications occurred at the femoral access site, which was closed successfully in all patients with an 8 Fr Angio-Seal closure device (St. Jude Medical, St Paul, MN, USA). Additional procedural time due to the use of the device was 7 min (±2 min). On the basis of fluoroscopic images, the device immediately after deployment seemed to cover the ostia of the three supraaortic trunks (brachiocephalic, left carotid and left subclavian) in all cases. In case of transfemoral TAVI, subsequent passage of pigtail, stiff wires, predilatation balloon and valve prosthesis was not hampered by the device. During in-hospital follow-up, no patient developed new neurological symptoms or clinical findings of stroke except for one patient who suffered from a TIA 2 days after the procedure. No patient suffered from clinically relevant peripheral embolisation towards nonbrain regions.

MRI results MRI scanning of the brain in 10 patients with SMT filter showed on average 3.2 new DW lesions per patient, as compared to 7.2 new lesions per patient in the historical comparison group without SMT filter. Lesions occurred almost equally in left and right cerebral and cerebellar hemispheres in both treatment groups.

151

CHAPTER 10

90

32.97

TA

23

aHTN, HC, PVD, CAD, TIA

seccessful

no

2

F

89

12.8

TA

23

Mild cognitive dysfunction

seccessful

TIA

3

M

67

13.52

TA

26

aHTN, HC, PVD, porcelain aorta, CABG, COPD

seccessful

no

4

F

87

12.08

TF

26

aHTN, HC, CAD, COPD,

seccessful

no

5

M

52

10.88

TF

26

aHTN, HC, PVD, porcelain aorta, CABG

seccessful

no

6

F

84

15.46

TA

23

COPD

seccessful

no

7

M

72

9.01

TF

26

aHTN, HC, CABG, stroke

seccessful

no

8

M

80

15.6

TA

26

aHTN, HC, eGFR < 30, CAD, LVEF < 35%, COPD

seccessful

no

9

F

79

11.66

TF

26

TIA, malignancy, COPD

seccessful

no

10

F

79

11.66

TA

23

aHTN, HC, PVD, porcelain aorta, eGFR < 30, stroke, DM

seccessful

no

11

F

83

24

TF

23

aHTN, HC, PVD, DM, COPD

seccessful

no

12

F

75

9

TF

23

aHTN, HC, DM, COPD

seccessful

no

13

F

88

12.8

TF

23

aHTN, DM

seccessful

no

14

F

85

20.58

TF

26

PVD, malignancy

seccessful

no

15

F

80

7.46

TF

26

HC, stroke, sarcoid

seccessful

no

TIA/ stroke

Valve size

SMT deployment

TA/TF

F

ES (log)

m/f

1

Age

n

Comorbidities

Table 1 |

m: male; f: female; ES(log): logistic EuroSCORE; TA: transapical; TF: transfemoral; TIA: transient ischaemic attack â&#x20AC;&#x201C; aHTN: arterial hypertension; HC: hypercholesterolemia; PVD: peripheral vascular disease; CAD: coronary artery disease; CABG: coronary artery bypass surgery; COPD: chronic obstructive pulmonary disease; DM: diabetes mellitus; eGFR: estimated glomerular filtration rate (ml/min/m2); LVEF: left ventricular ejection fraction

152

SMT EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

DISCUSSION Clinical and subclinical cerebral embolism during TAVI is a matter of concern and a drawback for the ubiquitous introduction of this revolutionary treatment. Alongside vascular access complications, it obviously represents one of the “weakest links” of TAVI for which the interventional community has to seek a solution. Indeed, while periprocedural clinical stroke and TIA probably only represent the tip, underneath is an iceberg of subclinical new cerebral lesions caused by TAVI which may be associated in the long term with accelerated cognitive and functional decline. Especially if TAVI has the potential to become a treatment for younger patients with aortic stenosis in the foreseeable future, this issue has to be tackled.

While downsizing of valve delivery equipment16, increased operator experience and better patient selection may all contribute to a reduction in neurological complications, mechanical manipulation of a diseased and calcified aortic valve will still cause a shower of debris towards the brain which necessitates some kind of sentinel at the entrance of the cerebral circulation, allowing blood to pass but capturing or deflecting all other “unwanted visitors”. Several embolic protection/deflection devices (e.g., Embrella [Edwards Lifesciences, Inc., Irvine, CA, USA], Claret [Claret Medical, Inc. Santa Rosa, CA, USA]) for use during TAVI are currently being tested17.

The SMT device is a retrievable low-profile filter designed for transfemoral percutaneous delivery at the level of the aortic arch to reduce the occurrence of stroke by deflecting emboli that occur during instrumentation of the heart, aortic root and branching vessels. It is made from nitinol (nickel titanium alloy) wires, which exhibit super elasticity so that the filter can be crimped into a 8 Fr calibre catheter and regains its original shape upon deployment. A potentially major advantage of the SMT filter over the Embrella and Claret devices is the fact that the last two cover only the innominate and right carotid arteries, while the SMT filter covers all the supraaortic trunks. Possible drawbacks are that it is, like the Embrella, an embolic deflection instead of capture device, and it is larger, necessitating femoral access, while both Claret and Embrella devices can be delivered through a 6 Fr sheath.

In 15 patients, of whom 6 were known to have peripheral vascular disease including 3 with a porcelain aorta, we managed to position the device without complications in the aortic arch, where it shielded the ostia of all major arteries arising from the aortic arch according to the fluoroscopic images. The average total procedural time was only lengthened by 7 (±2) minutes, indicating the ease and user-friendliness with which the device can be deployed. Apart from one patient who suffered a TIA 2 days after TAVI, all patients remained free from clinical neurological events or symptomatic peripheral embolism. Although not the aim of this small feasibility study, we found that the number of new brain lesions on DW-MRI, which may prognosticate future cognitive dysfunction, was numerically reduced by more than half as compared to an historical group undergoing TAVI without the SMT deflection device at our centre. Whether this reduction is associated with better long-term neurological performance remains to be proven in larger, prospective randomised trials comparing protection versus no protection. Interestingly, new lesions were found substantially equally distributed in the whole central nervous system, in both left and right cerebral and cerebellar hemispheres. This finding supports the idea that protection should be provided to all cerebral vessels, whereas partial protection may be insufficient.

153

CHAPTER 10

STUDY LIMITATIONS Due to the low number of patients, this study can only be considered as â&#x20AC;&#x153;feasibility-testâ&#x20AC;?, paving the way first for a safety CE approval study (ongoing) and subsequently for a large, prospective randomised trial with enough power to demonstrate clinical efficacy and safety. While delivery and deployment of the deflection device was successful in all our patients, certain aortic arch anatomies may be inaccessible for the device or render efficacious deployment impossible, with persistent access for debris towards the cerebral circulation. It can be expected that, with the use of aortic protection devices, preprocedural computed tomography imaging will become even more important for the planning of TAVI procedure, not only to assess the aortic valve (calcifications, measurement of the annulus) and the ileofemoral system (calcifications, tortuosity), but also to evaluate the aortic arch and its potential to harbour a specific protective filter and meanwhile facilitate easy valve-device crossing.

The patients in our study did not undergo formal neurological assessment pre- and postprocedure, implying that subtle neurological deficits after TAVI may be missed. Future studies investigating the use of TAVI protection devices will require pre- and post-procedural patient evaluation by a neurologist and long-term patient follow-up to understand the potential clinical impact of a reduction in DW-MRI cerebral lesions.

Finally, the SMT device is a deflection device that redirects particulate matter away from the brain towards the descending aorta and peripheral organs. Although in our study this was not associated with any obvious clinical events, larger studies are needed to understand its potential negative clinical impact on kidney, liver or gastrointestinal function.

CONCLUSION The use of the transfemoral SMT deflection device during transfemoral or transapical TAVI seems feasible and is associated with less cerebral embolism on DW-MRI in this small proofof-concept study. Larger, randomised trials are required to confirm these findings and to understand the clinical risk reduction of (silent) cerebral embolism and stroke.

CONFLICT OF INTEREST STATEMENT P.R. Stella is member of the scientific advisory board of SMT Medical. All other authors have no conflict of interest.

154

SMT EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

REFERENCES 1. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, Tuzcu EM, Webb JG, Fontana GP, Makkar RR, Williams M, Dewey T, Kapadia S, Babaliaros V, Thourani VH, Corso P, Pichard AD, Bavaria JE, Herrmann HC, Akin JJ, Anderson WN, Wang D, Pocock SJ; PARTNER Trial Investigators. Transcatheter versus Surgical Aortic-Valve Replacement in High-Risk Patients. N Engl J Med. 2011;364:2187-98.

2. Aggarwal A, Dai D, Rumsfeld JS, Klein LW, Roe MT; American College of Cardiology National Cardiovascular Data Registry. Incidence and predictors of stroke associated with percutaneous coronary intervention. Am J Cardiol. 2009;104:349-53.

3. Keeley EC, Grines CL. Scraping of aortic debris by coronary guiding catheters: a prospective evaluation of 1,000 cases. J Am Coll Cardiol. 1998;32:1861-5.

4. Omran H, Schmidt H, Hackenbroch M, Illien S, Bernhardt P, von der Recke G, Fimmers R, Flacke S, Layer G, Pohl C, Lüderitz B, Schild H, Sommer T. Silent and apparent cerebral embolism after retrograde catheterisation of the aortic valve in valvular stenosis: a prospective, randomised study.

Lancet. 2003;361:1241-6.

5. Kahlert P, Doettger P, Mori K, Al-Rashid F, Thielmann M, Wendt D, Jakob H, Erbel R, Eggebrecht H. TVT-95. Cerebral Embolization during Transcatheter Aortic Valve Implantation: A Transcranial Doppler Study. Paper presented at: Transcatheter Cardiovascular Therapeutics, 2010; Washington, DC. 6. Rodés-Cabau J, Dumont E, Boone RH, Larose E, Bagur R, Gurvitch R, Bédard F, Doyle D, De Larochellière R, Jayasuria C, Villeneuve J, Marrero A, Côté M, Pibarot P, Webb JG. Cerebral embolism following transcatheter aortic valve implantation: comparison of transfemoral and transapical approaches. J Am Coll Cardiol. 2011;57:18-28.

7. Astarci P, Glineur D, Kefer J, D’Hoore W, Renkin J, Vanoverschelde JL, El Khoury G, Grandin C. Magnetic resonance imaging evaluation of cerebral embolization during percutaneous aortic valve implantation: comparison of transfemoral and trans-apical approaches using Edwards Sapiens valve.

Eur J Cardiothorac Surg. 2011;40:475-9.

8. Kahlert P, Knipp SC, Schlamann M, Thielmann M, Al-Rashid F, Weber M, Johansson U, Wendt D, Jakob HG, Forsting M, Sack S, Erbel R, Eggebrecht H. Silent and apparent cerebral ischemia after percutaneous transfemoral aortic valve implantation: a diffusion-weighted magnetic resonance imaging study. Circulation. 2010;121:870-8.

9. Ghanem A, Muller A, Nahle CP, Kocurek J, Werner N, Hammerstingl C, Schild HH, Schwab JO, Mellert F, Fimmers R, Nickenig G, Thomas D. Risk and fate of cerebral embolism after transfemoral aortic valve implantation: a prospective pilot study with diffusion-weighted magnetic resonance imaging. J Am Coll Cardiol. 2010;55:1427-32.

10. Arnold M, Schulz-Heise S, Achenbach S, Ott S, Dörfler A, Ropers D, Feyrer R, Einhaus F, Loders S, Mahmoud F, Roerick O, Daniel WG, Weyand M, Ensminger SM, Ludwig J. Embolic cerebral insults after transapical aortic valve implantation detected by magnetic resonance imaging. JACC Cardiovasc Interv. 2010;3: 1126-32.

11. Vermeer SE, Longstreth WT Jr, Koudstaal PJ. Silent brain infarcts: a systematic review. Lancet Neurol.

2007;6:611-9.

12. Agostoni P, Voskuil M, Onsea K, Vermeersch P, Stella PR. Tools & Techniques: Percutaneous intervention of saphenous vein graft lesions. EuroIntervention. 2011;7:878-9.

13. Henry M, Polydorou A, Henry I, Polydorou AD, Hugel M. Carotid angioplasty and stenting under protection: advantages and drawbacks. Expert Rev Med Devices. 2008;5:591-603.

14. Webb JG, Chandavimol M, Thompson CR, Ricci DR, Carere RG, Munt BI, Buller CE, Pasupati S, Lichtenstein S. Percutaneous aortic valve implantation retrograde from the femoral artery. Circulation.

2006;113:842-50.

15. Walther T, Dewey T, Borger MA, Kempfert J, Linke A, Becht R, Falk V, Schuler G, Mohr FW, Mack M. Transapical aortic valve implantation: step by step. Ann Thorac Surg. 2009;87:276-83.

155

CHAPTER 10

16. Grube E, Schuler G, Buellesfeld L, Gerckens U, Linke A, Wenaweser P, Sauren B, Mohr FW, Walther T, Zickmann B, Iversen S, Felderhoff T, Cartier R, Bonan R. Percutaneous aortic valve replacement for severe aortic stenosis in high-risk patients using the second- and current third-generation selfexpanding CoreValve prosthesis: device success and 30-day clinical outcome. J Am Coll Cardiol.

2007;50:69-76.

17. Nietlispach F, Wijesinghe N, Gurvitch R, Tay E, Carpenter JP, Burns C, Wood DA, Webb JG. An embolic deflection device for aortic valve interventions. JACC Cardiovasc Interv. 2010;3: 1133-8

156

157

CHAPTER

11 EMBRELLA EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TRANSCATHETER AORTIC VALVE REPLACEMENT J Thorac Cardiovasc Surg. 2014

Mariam Samim1 Pierfrancesco Agostoni1 Jeroen Hendrikse2 Ricardo P.J. Budde2 Freek Nijhoff1 Jolanda Kluin3 Faiz Ramjankhan3 Pieter A. Doevendans1 Pieter R. Stella1 Department of Cardiology, 2Department of Radiology, 3Department of Cardiothoracic

1

surgery, University Medical Centre Utrecht, the Netherlands

CHAPTER 11

ABSTRACT Aims To compare the extent of cerebral ischemic injury after transcatheter aortic valve replacement (TAVR) with the use of an Embrella Embolic Deflector System versus unprotected TAVR. Methods Fifteen patients with severe symptomatic aortic stenosis underwent TAVR with usage of Embrella Embolic Deflector System for cerebral protection. Cerebral diffusion weighted magnetic resonance imaging (DWI) was obtained in all patients at day 4 after the procedure and retrospectively compared to 37 patients previously undergoing TAVR without a protection device (TAVR-only group).

Results Successful placement of the Embrella device was achieved in all patients. DWI revealed an increase in the number of ischemic lesions in the Embrella group as compared to the TAVRonly group (9.0 vs. 5.0, p=0.044). The use of the Embrella device was however associated with a significant reduction in single lesion volume, 9.7 µl [5.8, 18.4] versus 17.8 µl [9.5, 38.7] (p <0.001). Moreover, total infarct volumes of more than 1000 µl were only seen in the TAVR-only group. More lesions occurred in the right side of the brain in the Embrella group, whereas in the TAVR-only group lesions were distributed equally between left and right. One patient in the TAVR-only group suffered from a transient ischemic attack. Postoperative evaluation was clinically uneventful in the Embrella group.

Conclusion The use of the Embrella device during TAVR increased the number of cerebral ischemic lesions on post-procedural brain imaging. This increase in number was however accompanied by a significant reduction in single lesion volume and the absence of large total infarct volumes.

ULTRAMINI ABSTRACT We evaluated the benefit of the Embrella Embolic Deflector System for cerebral protection during transcatheter aortic valve replacement (TAVR). The use of this device during TAVR increased the number of cerebral ischemic lesions on post-procedural brain imaging, although it significantly reduced lesion volume as compared to procedures performed without cerebral protection.

160

EMBRELLA EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

INTRODUCTION Transcatheter aortic valve replacement (TAVR) is an accepted treatment option for inoperable or high-risk patients with severe aortic stenosis. Although good procedural success and favourable clinical outcomes have been reported 1, 2, issues remain regarding the relatively high complication rate. One of the most important drawbacks of TAVR is the risk of intraprocedural cerebral embolization causing brain injury. In high-risk surgical candidates, TAVR is associated with an approximately two-fold increased incidence of stroke or transient ischemic attack (TIA) (5.5% vs. 2.4%, p=0.04) at 30 days as compared with surgical aortic valve replacement 3.

Moreover, new foci of restricted cerebral perfusion on diffusion weighted magnetic resonance imaging (DWI) are reported in 58% to 91% of patients undergoing TAVR 4-8.The clinical impact of these new, usually asymptomatic DWI lesions is still uncertain. However epidemiological studies have reported an association between asymptomatic cerebral infarctions and frailty, decline in cognitive and physical functions, early development of dementia and an increased risk of subsequent stroke 9, 10. In order to reduce the risk of these potentially devastating adverse events and in the light of progressive movement of TAVR towards younger lower-risk patients, significant research has focused on identifying risk factors of TAVR-related cerebral embolization and methods for reducing the risk of cerebral ischemic injury. Embolic protection devices (EPD) have already demonstrated their benefit in carotid artery interventions 11. This would suggest that the application of a similar concept for brain protection could minimize the embolic burden during TAVR as well.

We report our experience with the Embrella Embolic Deflector System (Edwards Lifesciences Ltd., Irvine, CA, USA), an umbrella-like device, designed for percutaneous delivery to the aortic arch, for prevention of cerebral ischemic injury caused by emboli originating in the heart and aorta. Proof-of-concept was shown in a first-in-human study with 4 patients 12. We aimed to evaluate the effectiveness of the Embrella device in diminishing the number of post-TAVR cerebral infarcts on DWI, in patients undergoing TAVR with the use of this protection device as compared to TAVR procedures without the use of an EPD.

METHODS Patients Between September 2012 and July 2013, 58 patients with symptomatic severe aortic stenosis underwent transfemoral TAVR in our institution. During this period, brain magnetic resonance imaging (MRI) has been performed routinely as part of post-TAVR standard care at our center in all patients without contraindications for MRI. From May 2013 onwards, the Embrella Embolic Deflector System was used during TAVR in 15 consecutive patients, without implementing eligibility requirements. These patients were retrospectively compared to the all 37 patients who had undergone TAVR in the previously mentioned period of time, without the use of an EPD and who underwent a post-procedural brain MRI (TAVR-only group). All patients were formally discussed in the heart team and considered inoperable or at high risk

161

CHAPTER 11

for surgical aortic valve replacement due to age, high logistic EuroSCORE, porcelain aorta, malignancy, frailty or severe comorbidities. As the Embrella device is CE-marked in May 2010, commercially available and used in clinical practice in our series and as cerebral DWI is part of post-procedural standard care at our center, institutional approval of waiver of informed consent was obtained.

Device and Procedure The Embrella Embolic Deflector System is an umbrella-like device that consists of 2 polyurethane membranes mounted on a nitinol frame (Figure 1A). This device is designed to deflect rather than to capture embolic particles. The polyurethane membrane has 100-Îźm pores to ensure proper blood circulation downstream of the device. The device can be folded, sheathed and loaded into a 6F long delivery sheath. It is attached to a 0.035-inch nitinol delivery cable and can be introduced in the horizontal segment of the aortic arch through either right radial, ulnar or brachial arteries. Subsequently, the device consisting of two petals is released from the sheath, is pulled back and is positioned at the outer curvature of the aortic arch such that the petals cover the left carotid and the innominate arteries (Figure 1B). In some patients it will further cover (sometimes only partially) the left subclavian artery.

Before beginning the TAVR procedure, apposition of the filter frame in the aortic arch was confirmed with angiography to ensure protection of the cerebral vascular circulation.

Sitting at the outer curvature of the aortic arch, the device does not interfere with the TAVR procedure, and in particular there is no interference with the large valve delivery system (Figure 1C). Once the procedure is terminated, the device is re-sheathed using the 6F delivery sheath.

TAVR was performed in all cases via the transfemoral approach with the Medtronic CoreValve system (CoreValve Revalving Technology, Medtronic, Minneapolis, MN, USA) or Edwards SAPIEN XTTM valve prosthesis (Edwards Lifesciences, Irvine, CA, USA). Regardless of the prosthesis type balloon aortic valvuloplasty was performed under rapid pacing to pre-dilate the native aortic valve. The valve prostheses were subsequently deployed under rapid pacing (Edwards â&#x20AC;&#x201C; 180 beats/min) or slow pacing (CoreValve â&#x20AC;&#x201C; 120-140 beats/min).

Magnetic resonance imaging Magnetic resonance imaging was performed within four days after TAVR, using a 3 Tesla system (Philips Medical Systems, the Netherlands).

The imaging protocol included a diffusion-weighted single-shot spin echo echoplanar sequence (diffusion gradient b values of 0 and 1000 s/mm2, repetition time (TR): 3307 ms, echo time (TE): 68 ms, 26 slices with a slice thickness of 4 mm, field of view: 230 mm, matrix: 256 x 205) and a turbo fluid attenuated inversion recovery (FLAIR; TR/TE 11000/125 ms). The acquisition time for the diffusion-weighted sequences was 69 seconds.

All MRI images were assessed by two skilled observers blinded to neurological status and procedure. Diffuse alterations in the diffusion weighted image were not regarded as embolic types of lesions. Number of ischemic lesions, overall single lesion volume, mean lesion volume per patient, total ischemic volume per patient and location and vascular territories of all focal diffusion abnormalities (bright lesions on DWI) were documented.

162

EMBRELLA EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

Figure 1 | A. The Embrella Embolic Deflector System, consisting of two petals and a delivery cable. This device is placed at the outer curvature of the aortic arch and the two petals cover the brachiocephalic trunk and the left carotid artery. B. Once positioned in the aortic arch, a contrast injection through the delivery sheet confirms proper placement of the device. Note that in this patient also the left subclavian artery is covered by the Embrella petal. C. Unhampered passage of the valve delivery system through the aortic arch in the presence of Embrella device.

Endpoints and definitions The primary endpoint was the number of new ischemic lesions on cerebral DWI. Our secondary endpoints included technical success, defined as successful delivery and retrieval of the Embrella Embolic Defector System, volume and distribution of post-procedural DWI lesions and clinical outcomes including post-TAVR rates of TIA and stroke and periprocedural cerebrovascular events. According to the VARC-2 criteria, stroke is defined as an acute episode of focal or global neurological dysfunction caused by the brain, spinal cord, or retinal vascular injury as a result of haemorrhage or infarction 13. According to the same criteria, TIA is defined as a transient episode (< 24 h) of focal neurological dysfunction caused by the brain, spinal cord, or retinal ischemia, without acute infarction. Silent brain infarction was defined as new bright lesion on post-TAVR DWI in absence of clinical symptoms.

Statistical methods All analyses were performed with the use of SPSS software (version 20, SPSS Inc, Chicago, IL).

Continuous variables are expressed as mean Âą standard deviation (SD) or median [interquartile range, IQR]. Categorical variables are described by frequencies and percentages. Comparison of continuous variables was performed using t test or Mann-Whitney U test. Comparison of categorical variables was performed using Pearson chi-squared test or Fisherâ&#x20AC;&#x2122;s exact test and a value of P<0.05 was considered statistically significant.

163

CHAPTER 11

RESULTS Patient Characteristics and technical outcome Clinical characteristics are presented in table 1. Technical success was 100%. Correct placement of the Embrella device with the deflecting petals covering the brachiocephalic and left carotid artery ostia was achieved in all but one patient. One patient showed anomaly of the aortic arch, with the right carotid and an aberrant right subclavian artery (arteria lusoria) arising directly from the aorta (Online Figure 1). In this patient left and right carotid and left subclavian arteries were not covered by the Embrella device. All other patients had type II aortic arch anatomy with the right subclavian and right carotid artery originating from the brachiocephalic artery. The access site was radial in 13 (87%) and ulnar in 2 (13%) patients. In 6 cases (40%) also the ostium of the left subclavian artery was covered.

With regard to the implantation procedure of the aortic valve prosthesis, advancement of the transfemoral valve delivery system alongside the Embrella device was associated with minimal interaction without significant interference or dislodgement of the device. Device success as defined by VARC-2 definitions was achieved in all patients in the Embrella group and 35 patients (95%) in the TAVR-only group (p=1.00). Reasons for not fulfilling the device success criteria were: additional valve implantation in one patient in the TAVR-only group due too low placement of a Medtronic CoreValve prosthesis and hence severe paravalvular regurgitation, and residual moderate paravalvular regurgitation after (Edwards SAPIEN XT™) valve implantation in a second patient in the same group.

After valve implantation and eventual additional attempts to optimize the result, the Embrella device was withdrawn by “collapsing” the 2 nitinol petals into the delivery sheath. Haemostasis of the radial or ulnar access site was accomplished with a standard compression band. No vascular and bleeding complications occurred. The total fluoroscopy time was significantly longer in the Embrella group as compared to the TAVR-only group (23.2 ± 7.0 vs. 18.0 ± 6.9, Median: 22.0 [18.8,26.0] vs. 17.8 [12.9,20.0], p=0.005) (Table 1).

Primary, DWI outcome As summarized in Table 2, post-TAVR DWI revealed new ischemic lesions in all patients in the Embrella group and in 35 (95%) cases in the TAVR-only group. Lesions were typically multiple in both groups with a significantly higher number of lesions in the Embrella group: a median of 9.0 lesions [4.0 to 12.0] in the Embrella group and 5.0 lesions [2.0 to 7.0] in the TAVR-only group (p=0.044). The use of the Embrella device was however associated with a significant reduction in single lesion volume: 9.7 µl [5.8, 18.4] versus 17.8 µl [9.5, 38.7] (p <0.001). Ischemic lesions in the Embrella group were more frequently of the smallest size (<20 µl) compared to the TAVRonly group (80% vs. 54%, p <0.001) (Figure 2A). Concerning the total volume of new ischemic lesions (total infarct volume) no significant difference was observed between the two groups: 111.6 µl [65.0 to 184.1] in the Embrella group and 129.4 µl [53.0 to 357.0] in the TAVR-only group (p=0.98). However, total infarction volume of more than 1000 µl occurred only in the TAVR-only group (in 8.1% of cases) (Figure 2B).

164

EMBRELLA EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

New ischemic lesions were located in the cerebellum and cortex, deep white matter, and deep grey matter in the affected cerebral hemisphere. A higher frequency of cortical lesions in the perfusion territory of anterior cerebral artery was detected in the Embrella group as compared to the TAVR-only group (28 (18%) vs. 24 (9%), p=0.02)(Table 3). No other differences were observed in the distribution pattern of DWI lesions between the two groups. Whereas lesions were detected almost equally in left and right cerebral and cerebellar hemispheres in the TAVR-only group, more lesions were detected in the right side of the brain in the Embrella group (Figure 2C).

Table 1 | Characteristics of patients and TAVR procedures TAVR alone (n= 37 )

TAVR + Embrella (n= 15)

p

Age (years)

81 [78, 84]

84 [73, 87]

0.29

Female sex

16 (43)

7 (47)

0.82

Diabetes mellitus

9 (24)

5 (33)

0.51

Hypertension

21 (57)

8 (53)

0.82

Hyperlipidaemia

10 (27)

4 (27)

1.00

Coronary artery Disease

25 (68)

8 (53)

0.33

Previous acute myocardial infarction

7 (19)

1 (7)

0.41

Previous PCI

20 (54)

5 (33)

0.18

Previous CABG

9 (24)

4 (27)

1.00

2(5)

1(7)

1.00

Atrial fibrillation

15 (41)

6 (40)

0.97

Carotid disease

1 (3)

2 (13)

0.20

Peripheral vascular Disease

3 (8)

1 (7)

1.00

Patient characteristics

Porcelain aorta according to VARC-2

Prior stroke or TIA

5 (14)

3 (20)

0.68

13.1 [8.0, 18.1]

16.6 [10.7, 22.6]

0.12

Fluoroscopy time (min)

17.8 [12.9, 20.0]

22.0 [18.8,26.0]

0.005

Contrast medium (ml)

145.0 [125.3, 180.0]

160.0 [150.0,190.0]

0.12

Edward SAPIEN XTâ&#x201E;˘

22 (59)

11 (73)

0.35

Medtronic CoreValve

15 (41)

4 (27)

0.35

Logistic EuroSCORE (%) Procedural characteristics

TAVR prosthesis type

Data are shown as median [interquartile range] or n (%). CABG: coronary artery bypass grafting; DWI: diffusion weighted magnetic resonance imaging; PCI: percutaneous coronary intervention; TAVR: transcatheter aortic valve replacement; TIA: transient ischemic attack.

165

CHAPTER 11

Table 2 | Findings on DWI after TAVR TAVR alone (n= 37 )

TAVR + Embrella (n= 15)

p

Patient with new lesions

35 (95)

15 (100)

1.00

Patient with single lesion

4 (11)

0 (0)

0.31

Patients with multiple lesions

31 (84)

15 (100)

0.17

5.0 [2.0,7.0]

9.0 [4.0, 12.0]

0.044

Right hemisphere and cerebellum

2.0 [0.5, 4.0]

5.0 [2.0, 10.0]

0.008

Left hemisphere and cerebellum

2.0 [1.0, 5.0]

3.0 [2.0, 6.0]

0.34

Lesions per patient

Lesion volume (µl) Mean per patient

25.1 [11.0, 61.0]

15.2 [10.9, 21.7]

0.092

129.4 [53.0, 357.0]

111.6 [65.0, 184.1]

0.98

Total ischemic volume right side

60.3 [1.2, 132.4]

68.4 [20.6, 126.8]

0.63

Total ischemic volume left side

65.3 [10.4, 191.3]

51.3 [32.1, 83.8]

0.80

Total ischemic volume per patient

Data are shown as median [interquartile range] or n (%). DWI: diffusion weighted magnetic resonance imaging; TAVR: transcatheter aortic valve replacement.

Table 3 | DWI lesion analysis

Cortical infarcts

TAVR alone (n = 254)

TAVR + Embrella (n = 159)

p

123 (48)

86 (54)

0.26

Anterior cerebral artery

24 (9)

28 (18)

0.02

Middle cerebral artery

47 (19)

27 (17)

0.70

Posterior cerebral artery

22 (9)

9 (6)

0.26

Border zone

30 (12)

22 (14)

0.55

75 (30)

37 (23)

0.16

Subcortical infarcts

12 (5)

3 (2)

0.13

Deep grey matter

Internal border zone

11 (4)

13 (8)

0.10

Infratentorial

45 (18)

23 (14)

0.39

Cerebellum

41 (16)

21 (13)

0.42

Brainstem

4 (2)

2 (1)

1.00

Single lesion volume (µl)

17.8 [9.5, 38.7]

9.7 [5.8, 18.4)

<0.001

Right side

17.6 [9.2, 35.0]

9.5 [5.1, 19.3]

<0.001

Left side

18.6 [9.8, 39.6]

10.4 [7.3,15.8]

<0.001

Data are shown as median [interquartile range] or n (%).

DWI: diffusion weighted magnetic resonance imaging; TAVR: transcatheter aortic valve replacement.

166

EMBRELLA EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

Secondary, clinical outcome No procedural complications occurred, in particular there were no complications related to insertion of the Embrella device. During in-hospital follow-up, no patient developed new neurological symptoms or clinical findings of stroke except for one patient in the TAVR-only group who suffered from dysphasia, diagnosed as TIA, 30 minutes after the procedure. MRI performed after TAVR confirmed a new ischemic lesion (40µl) in the site corresponding to the symptoms in this patient (Online Figure 2). No patient suffered from clinically relevant peripheral embolization towards non-brain regions.

A

B

C

Figure 2 | A. Incidence of different sizes of cerebral DWI lesions in the Embrella group compared to the TAVR-only group. DWI lesions were categorized according to volume. The majority of lesions in the Embrella group have a volume of less than 20 µl (80%), which is the case in only 54% of TAVR-only group.

B. Lesion volumes of more than 1000 µl occurred only in the TAVR-only group. C. Left/right distribution of DWI lesions. A shift towards a higher number of lesions in the right side of the brain was observed when an Embrella Embolic Deflector System was used during TAVR.

167

CHAPTER 11

DISCUSSION Minimizing the risk of procedure-related cerebral injury is of paramount importance to the overall success of TAVR, especially if this procedure has the potential to become a treatment option also for younger patients with severe aortic stenosis. The passage of bulky devices through a diseased aorta and the aggressive manipulation of a stenotic, calcified aortic valve could cause embolization of atherosclerotic material, clots and other calcified debris. The need to reduce TAVR-related cerebral complications has spurned significant research and the development of embolic protection devices. The use of an EPD during TAVR is not a novel idea since the benefit of these types of devices for brain protection has been demonstrated in carotid intervention 11. A systematic review including 896 carotid stenting cases with EPD and 2537 cases without EPD found that EPD usage was associated with significant lower stroke rates (1.8% vs. 5.5%, P < 0.001)(14).

In the present study we investigated the benefit of the use of Embrella embolic deflector system for brain protection during TAVR procedures. Initial animal work have demonstrated that Embrella was effective in deflecting emboli ranging in size from 150 Îźm to 600 Îźm away from the carotid circulation15. Retrieved particulate matter was reduced in the carotid filtration circuit from 19% (during unprotected aortic injection of prepared human atheroma) to 1.3% with Embrella usage (P < 0.001). Nietlispach et al. 12 described previously their initial in-human experience with the Embrella device in three transfemoral TAVR patients (with the Edwards Sapien valve) and one balloon aortic valvuloplasty case. No clinically overt stroke was found, but post-procedure DWI revealed a new ischemic lesion in the balloon valvuloplasty patient.

DWI data in our study revealed a significantly higher number of lesions in the Embrella group as compared to the TAVR-only group. An increase in the number of ischemic lesions may be caused by the introduction and recapture of the Embrella device, manipulations that may inadvertently lead to cerebral showering caused by dislodgement of atherocalcific material from the brachiocephalic artery and aortic arch. Dislodgment of atherocalcific material in the brachiocephalic artery may cause direct cerebral embolization. The latter would theoretically cause an increased number of embolic DWI lesions in the right side of the brain. Our data support the aforementioned hypothesis as we showed a significant increase in the number of DWI lesions in the right half of the brain in the Embrella group as compared to the TAVR-only group. Also a trend towards higher total infarct volumes was found in the right side of the brain in the Embrella group. This phenomenon was not observed in the TAVR-only group (Table 2).

These observations suggest a possible increase in the ischemic burden for the right side of the brain due to right radial/brachial access. Considering that the elderly TAVR patient candidates are at high risk of having thrombogenic plaques in brachiocephalic artery and aortic arch, transfemoral introduction of the EPD might be a safer approach for the brain.

Interestingly, the present study showed a significant reduction in lesion volume on DWI and a trend towards lower total infarct volume due to the use of an Embrella device and especially the absence of large volume lesions. The latter observation may be explained by deflection of large embolic particles, originating from the calcified aortic valve, by the Embrella device during TAVR. Moreover, small emboli may arise distal to the Embrella device, during

168

EMBRELLA EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

its introduction and recapture. This could explain the larger number of smaller lesions in the Embrella group. Reduction in total infarct volumes in the Embrella group may decrease the risk of future neurocognitive impairment, though not supported by the available evidence and needs further investigation.

Almost all studies to date investigating the benefit of the embolic protection devices for brain protection during TAVR have reported post-procedural occurrence of new DWI lesions, even with the use of an EPD (Online Table 1)12, 16-18. This (partial) lack in full effectiveness could be explained by different factors. First of all, lack of coverage of the left vertebral arterial system by the Embrella device and the Claret Medical diminishes the usefulness of these types of EPD. Secondly, thromboembolic events that cause cerebral injury are not merely limited to the actual TAVR procedure. Interestingly, as stated previously by other reports, the risk of stroke, though highest in the intraprocedural period, persists following successful TAVR 3, 19.

Therefore, in addition to the use of an effective EPD, there may be also an important role for optimal medication regimen after TAVR in order to decrease the risk of post-procedural cerebrovascular events. The optimal post-TAVR drug therapy to achieve this aim remains however unclear. Most physicians routinely continue dual antiplatelet therapy (aspirin and clopidogrel) for a number of months. Although not indicated due to lack of evidence, there may be a role for systemic anticoagulation in a role similar to surgical aortic valve replacement with bioprosthethic valves. However, a large number of TAVR patients, given their comorbidities, are likely to be at an increased risk of haemorrhagic complications. Future studies are necessary to address these issues.

In addition to the search for an efficient cerebral protection during and after TAVR, more attention needs to be paid to preprocedural patient evaluation and assessment. A thorough evaluation of risk factors for cerebrovascular events, with multimodality imaging may aid risk stratification and guide the approach for valve delivery. For instance, significant atheroma in the aortic arch may favour a transapical approach, although this has not been proved by DWI or clinical reports to date. Nonetheless, further refinements to EPD technique may offer a modality for significant reduction of brain injury during TAVR. Future studies need to elucidate the benefit of improved EPD technology for usage during TAVR on the one hand and the impact of usually silent DWI lesions on neurocognitive function of patients undergoing these procedures on the other hand.

LIMITATIONS This study has a number of limitations. First, due to the nonrandomized trial design, a selection bias cannot be excluded. Second, the number of patients treated with the Embrella device is small and therefore a definitive conclusion regarding the effectiveness of this device cannot be made. Finally, our study lacks formal neurocognitive assessment before and after the procedures in order to capture subtle changes in neurological and cognitive status of patients. A few small studies have recently focused on the benefit of EPD usage during TAVR procedures, while the real clinical impact of asymptomatic DWI lesions is still debatable. Future studies are needed to address this issue more thoroughly.

169

CHAPTER 11

CONCLUSIONS The current study shows that the use of the Embrella Embolic Deflector System during TAVR procedures is associated with a higher number of post-procedural cerebral DWI lesions as compared to procedures without EPD usage. This increase in number is however accompanied by a significant reduction in single lesion volume and the absence of large total infarct volumes (>1000 Âľl), reflecting the potential of the Embrella device to prevent large volume ischemic lesions. The reduction in single lesion volume may reduce cerebral ischemic burden during TAVR and needs further investigation in future studies with a larger number of patients.

170

EMBRELLA EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

REFERENCES 1. Rodes-Cabau J, Webb JG, Cheung A, Ye J, Dumont E, Feindel CM, et al. Transcatheter aortic valve implantation for the treatment of severe symptomatic aortic stenosis in patients at very high or prohibitive surgical risk: Acute and late outcomes of the multicenter canadian experience. J Am Coll Cardiol. 2010 Mar 16;55(11):1080-9.

2. Gilard M, Eltchaninoff H, Iung B, Donzeau-Gouge P, Chevreul K, Fajadet J, et al. Registry of transcatheter aortic-valve implantation in high-risk patients. N Engl J Med. 2012 May 3;366(18):1705-1.

3. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010 Oct 21;363(17):1597-60.

4. Arnold M, Schulz-Heise S, Achenbach S, Ott S, Dorfler A, Ropers D, et al. Embolic cerebral insults after transapical aortic valve implantation detected by magnetic resonance imaging. JACC Cardiovasc Interv. 2010 Nov;3(11):1126-32.

5. Ghanem A, Muller A, Nahle CP, Kocurek J, Werner N, Hammerstingl C, et al. Risk and fate of cerebral embolism after transfemoral aortic valve implantation: A prospective pilot study with diffusionweighted magnetic resonance imaging. J Am Coll Cardiol. 2010 Apr 6;55(14):1427-32.

6. Rodes-Cabau J, Dumont E, Boone RH, Larose E, Bagur R, Gurvitch R, et al. Cerebral embolism following transcatheter aortic valve implantation: Comparison of transfemoral and transapical approaches. J Am Coll Cardiol. 2011 Jan 4;57(1):18-2.

7. Fairbairn TA, Mather AN, Bijsterveld P, Worthy G, Currie S, Goddard AJ, et al. Diffusion-weighted MRI determined cerebral embolic infarction following transcatheter aortic valve implantation: Assessment of predictive risk factors and the relationship to subsequent health status. Heart. 2012 Jan;98(1):18-23.

8. Kahlert P, Knipp SC, Schlamann M, Thielmann M, Al-Rashid F, Weber M, et al. Silent and apparent cerebral ischemia after percutaneous transfemoral aortic valve implantation: A diffusion-weighted magnetic resonance imaging study. Circulation. 2010 Feb 23;121(7):870-8.

9. Knecht S, Oelschlager C, Duning T, Lohmann H, Albers J, Stehling C, et al. Atrial fibrillation in strokefree patients is associated with memory impairment and hippocampal atrophy. Eur Heart J. 2008 Sep;29(17):2125-32.

10. Vermeer SE, Prins ND, den Heijer T, Hofman A, Koudstaal PJ, Breteler MM. Silent brain infarcts and the risk of dementia and cognitive decline. N Engl J Med. 2003 Mar 27;348(13):1215-22.

11. Atkins MD, Bush RL. Embolic protection devices for carotid artery stenting: Have they made a significant difference in outcomes? Semin Vasc Surg. 2007 Dec;20(4):244-51.

12. Nietlispach F, Wijesinghe N, Gurvitch R, Tay E, Carpenter JP, Burns C, et al. An embolic deflection device for aortic valve interventions. JACC Cardiovasc Interv. 2010 Nov;3(11):1133-8.

13. Kappetein AP, Head SJ, Genereux P, Piazza N, van Mieghem NM, Blackstone EH, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: The valve academic research consortium-2 consensus document. J Thorac Cardiovasc Surg. 2013 Jan;145(1):6-23.

14. Kastrup A, Groschel K, Krapf H, Brehm BR, Dichgans J, Schulz JB. Early outcome of carotid angioplasty and stenting with and without cerebral protection devices: A systematic review of the literature.

Stroke. 2003 Mar;34(3):813-9.

15. Carpenter JP, Carpenter JT, Tellez A, Webb JG, Yi GH, Granada JF. A percutaneous aortic device for cerebral embolic protection during cardiovascular intervention. J Vasc Surg. 2011 Jul;54(1):174,181.e1.

16. Etienne PY, Papadatos S, Pieters D, El Khoury E, Alexis F, Price J, et al. Embol-X intraaortic filter and transaortic approach for improved cerebral protection in transcatheter aortic valve implantation. Ann Thorac Surg. 2011 Nov;92(5):e95-6.

17. Onsea K, Agostoni P, Samim M, Voskuil M, Kluin J, Budde R, et al. First-in-man experience with a new embolic deflection device in transcatheter aortic valve interventions. EuroIntervention. 2012 May 15;8(1):51-6.

171

CHAPTER 11

18. Naber CK, Ghanem A, Abizaid AA, Wolf A, Sinning JM, Werner N, et al. First-in-man use of a novel embolic protection device for patients undergoing transcatheter aortic valve implantation. EuroIntervention.

2012 May 15;8(1):43-50.

19. Rodes-Cabau J, Webb JG, Cheung A, Ye J, Dumont E, Feindel CM, et al. Transcatheter aortic valve implantation for the treatment of severe symptomatic aortic stenosis in patients at very high or prohibitive surgical risk: Acute and late outcomes of the multicenter canadian experience. J Am Coll Cardiol. 2010 Mar 16;55(11):1080-9.

172

EMBRELLA EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

% patients with DWI lesions

Average DWI lesions per patient

25%

0.25

Samim et al.

15

0%

100%

10.8

Onsea et al.17

15

-

100%

3.2

140 µm

Naber et al.

35

0%

-

-

120 µm

Etienne et al.16

1

0%

0%

0

100 µm

SMT

Femoral

Deflecting

~200 µm

Claret Medical Radial/brachial

Capturing

Embol-X

Capturing

Transaortic

Study

Deflecting

Access site Radial/ulnar/ brachial

Device Embrella

# of cases in the study

0%

Pore Size

Nietlispach et al.12 4

Mechanism of work

TIA/Stroke

Online Table 1 | Embolic protection devices studied for usage during TAVR

18

DWI: diffusion weighted magnetic resonance imaging; TAVR: transcatheter aortic valve replacement; TIA: transient ischemic attack.

Figure E1 | Rotational angiography image showing Embrella device (arrow) placed in the aortic arch of a patient with an arteria lusoria: anomaly of the aortic arch, with the right carotid and an aberrant right subclavian artery arising directly from the aorta. The arteria lusoria (arrow head) arises beyond the left subclavian artery. In this patient left and right carotid and left subclavian arteries were not covered by the Embrella device.

LCCA indicates left common carotid artery; LSA, left subclavian artery; RCCA, right common carotid artery; RSA, right subclavian artery.

173

CHAPTER 11

Figure E2 | DWI showing a TIArelated lesion of 40 µl in the Wernick’s area (arrow head). The patient suffered from transient signs of dysphasia.

174

175

CHAPTER

12 TRIGUARD™HDH EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TRANSCATHETER AORTIC VALVE REPLACEMENT: THE RESULTS OF “DEFLECT II TRIAL”

In preparation

Mariam Samim1 Bart van der Worp2 Pierfrancesco Agostoni1 Jeroen Hendrikse3 Ricardo P.J. Budde3 Freek Nijhoff1 Faiz Ramjankhan4 Pieter A. Doevendans1 Pieter R. Stella1 Department of Cardiology, 2Department of Neurology, 3Department of Radiology,

1

Department of Cardiothoracic surgery, University Medical Centre Utrecht, the Netherlands

4

CHAPTER 12

ABSTRACT Aims To evaluate the safety and performance of the embolic deflection device TriGuard™HDH in patients undergoing transcatheter aortic valve replacement (TAVR).

Methods This prospective, single arm feasibility pilot study included 15 patients with severe symptomatic aortic stenosis scheduled for TAVR. Cerebral diffusion weighted magnetic resonance imaging (DWI) was planned in all patients one day before and at day 4 (±2) after the procedure. Major adverse cerebral and cardiac events (MACCEs) and neurological status, including NIH Stroke Scale (NIHSS) and the modified Rankin Scale (mRS) scores, were recorded for all patients.

Primary endpoints of this study were I) device performance success defined as complete coverage of the three aortic arch takeoffs throughout the entire TAVR procedure and II) MACCE occurrence. Secondary endpoints included the number and the volume of new cerebral ischemic lesions on DWI. Results Fourteen patients underwent transfemoral TAVR and one patient a transapical procedure.

Edwards SAPIEN valve prosthesis was implanted in 9 (60%) patients and Medtronic CoreValve prosthesis in the remaining 6 (40%). Predefined performance success of the TriGuard™HDH device was achieved in 10 (67 %) patients. The composite endpoint MACCE occurred in none of the patients. NIH Stroke Scale scores were 0 in all patients on admission and remained unchanged during hospital stay. Modified Rankin Scale scores ranged from 0 to 3 (average 2.1) on admission and remained unchanged during hospitalisation. Post-procedural DWI was performed in 12 patients. Comparing the DWI of these patients to a historical control group showed no reduction in number [median 5.5 vs 5.5, p=0.96], however a trend towards a decrease in mean lesion volume per patient [median 12.4 vs 25.1, p=0.11] and total ischemic volume [median 98.9 vs 129.4, p=0.16].

Conclusion The use of the TriGuard™HDH for cerebral protection during TAVR is safe. Device performance success was achieved in 67% of all cases. Our data indicate that the use of this protection device might decrease cerebral ischemic burden during TAVR.

178

TRIGUARDâ&#x201E;˘HDH EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

INTRODUCTION Transcatheter aortic valve replacement (TAVR) is increasingly used to treat patients with severe symptomatic aortic valve stenosis who are considered inoperable or too high a risk for surgical aortic valve replacement (SAVR). As shown by the first randomized controlled trials in high risk patients, TAVR is non-inferior to surgery in terms of improvements in cardiac symptoms and long-term mortality.1, 2 Despite its clinical benefit, TAVR is associated with the risk of clinically manifest transient or irreversible neurological impairment. Moreover, diffusion weighted magnetic resonance imaging (DWI) studies have revealed a high incidence of clinically silent cerebral ischemic lesions after TAVR in 58% to 91% of patients undergoing TAVR.3-7 These high rates of subclinical cerebral infarcts after TAVR is up to double of that seen in isolated SAVR and quadruple that of diagnostic retrograde aortic valve catheterization.21, 22 Although the clinical impact of these cerebral lesions is still uncertain, several studies have reported an association between asymptomatic cerebral infarctions and frailty, premature decline in cognitive and physical functions, early development of dementia and an increased risk of subsequent stroke.8, 9 TAVR entails several procedural steps that can cause cerebral embolization of atherosclerotic/ calcific particles, notably advancement of bulky catheters through the aortic arch, balloon valvuloplasty, and most importantly deployment of the valve prosthesis in the calcified aortic annulus. The devastating consequences of periprocedural stroke, as well as concern regarding potential long-term neurocognitive sequelae of subclinical cerebral embolic events, have prompted investigation into methods to minimize the risk of periprocedural cerebral injury among patients undergoing TAVR. Efforts have predominantly focused on the use of dedicated embolic protection devices (EPD) and early clinical data have shown considerable promise for these devices in terms of technical success and safety.13-17 The DEFLECT I trial investigating the feasibility of the first-generation TriGuardâ&#x201E;˘ cerebral protection device (Keystone Heart, Herzliya, Israel) demonstrated good procedural success and safety, as well as a reduction in average lesion volume and total ischemic volume when compared with unprotected TAVR.18. After the initial experiences with the previous design, the device has been further optimized with smaller mesh (filter) pore size, improved (x-ray) visibility, and better manoeuvrability. The DEFLECT-II trial was designed to confirm feasibility and safety of this new design.

METHODS Patients All patients with severe symptomatic aortic valve stenosis scheduled for TAVR between January and June 2014 were screened for inclusion in the DEFLECT II pilot study. Key exclusion criteria were TAVR approach other than transfemoral and transapical, acute myocardial infarction within 72 hours before the procedure, impaired kidney function with glomerular filtration rate < 30 ml/min, history of stroke or transient ischemic attack within 6 months prior to TAVR,

179

CHAPTER 12

severe peripheral arterial disease that precludes the introduction of 9 French sheath through the femoral artery, documented friable or mobile atherosclerotic plaque in the aortic arch, and torturous or unsuitable anatomy of the aortic arch that may interfere with deployment of the TriGuard™HDH device. Study protocol included cerebral DWI and neurological evaluation, using the NIH stroke scale (NIHSS) and modified Rankin Scale (mRS), one day before and 4 (+2) days after TAVR. In case of transient ischemic attack (TIA) or stroke, both mRS and NIHSS were to be performed at day 90 post-TAVR to determine the severity of these events. The study was approved by the hospital’s ethics committee and performed with the consent of all study participants.

The TriGuard™HDH embolic deflection device The Keystone Heart TriGuard™HDH is a cerebral embolic protection device designed to deflect embolic particles from entering the three aortic arch takeoffs, thereby it theoretically protects all arterial territories of the brain (Figure 1A). Embolic particles are deflected toward the descending aorta and downstream, where they are either harmless or can be treated effectively without significant sequelae (Figure 1B).

The TriGuard™HDH deflector is comprised of several functional components (Figure 1C): a Nitinol frame (1) that is stabilized in the aortic arch by the lower stabilizer leg, (2) which pushes the frame towards the upper wall of the aortic arch, and by an upper stabilizer (3) that protrudes into the innominate artery ostium. The filter (4) is coated with a biocompatible heparin-hydrophilic coating, which serves to reduce the risk of thrombus formation. The frame includes 4 radiopaque markers for a better fluoroscopic visualization, on the proximal tip (5) of the device, on the top of the upper stabilizer (6), on the left splitting point of the upper stabilizer (7), and of the left splitting point of lower stabilizer (8). The 30 mm flexible tail (9) enables the deflector to conform to a variety of aortic arch anatomies. A connector (10) with 180⁰ rotational freedom connects the delivery tether with the deflector, preventing undesired torque of the deflector in response to unintended rotation of the tether.

The TriGuard™HDH device used in the DEFLECT II study includes a few changes as compared to the original device (TriGuard™) in order to improve its performance: I) reduced mesh pore size (130 X 250μm vs. 250μm X 250μm, respectively) to improve protection, II) improved visibility by the addition of 4 radiopaque markers as mentioned above, and III) improved manoeuvrability of the filter by changing the filter-tether connection (part 9, Figure 1C).

Use of the TriGuard™HDH device during TAVR The TriGuard™HDH device is a retrievable filter which is introduced towards the aortic arch through a 9 French sheath via the contralateral (non TAVR) femoral artery. Upon deployment from the sheath, the deflector unfolds and regains its original shape due to its shape memory property. Once the device is in position, procedures such as movement of guidewires, angiocatheters and valve delivery systems can be advanced over the aortic arch and below the deflector. Correct positioning of the TriGuard™HDH device during different steps of TAVR was confirmed with angiography, to ensure adequate protection of the aortic arch takeoffs during the entire procedure. Special care was taken in order to manipulate the pigtail and stiff wire

180

TRIGUARD™HDH EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

A

B

C

Figure 1 | A. Threedimensional reconstruction of the aortic arch using DynaCT technique, showing the ostia of the three supraaortic trunks (brachiocephalic, left carotid and left subclavian) (arrowheads) and a deployed TriGuard™HDH device in the aortic arch (arrow). B. This device diverts emboli/particulate matter downstream. C. The TriGuard™HDH deflector is comprised of a Nitinol frame (1) that is stabilized in the aortic arch by the lower stabilizer leg (2), which pushes the frame towards the upper wall of the aortic arch, and by an upper stabilizer (3) that protrudes into the innominate artery ostium. The frame includes 4 radiopaque markers for a better fluoroscopic visualization, on the proximal tip (5) of the device, on the top of the upper stabilizer (6), on the left splitting point of the upper stabilizer (7), and of the left splitting point of lower stabilizer (8) (see text for further detail).

under or below the device in both LAO and RAO views (Figures 2). Once the procedure is terminated, the device can be retrieved within the 9F delivery sheath.

TAVR was performed in all cases via the transfemoral approach with the Medtronic CoreValve system (CoreValve Revalving Technology, Medtronic, Minneapolis, MN, USA) or Edwards SAPIEN (SAPIEN XT™ or SAPIEN 3) valve prosthesis (Edwards Lifesciences, Irvine, CA, USA).

Regardless of the prosthesis type balloon predilatation of the aortic valve was performed in all cases as standard routine at our centre. The valve prostheses were subsequently deployed as per routine under rapid pacing (Edwards – 180 beats/min) or slow pacing (CoreValve – 120140 beats/min).

181

CHAPTER 12

Magnetic resonance imaging Magnetic resonance imaging was performed one day before and 4 (+1) days after TAVR. The imaging protocol included a diffusion-weighted single-shot spin echo echoplanar sequence (diffusion gradient b values of 0 and 1000 s/mm2, repetition time (TR): 3307 ms, echo time (TE): 68 ms, 26 slices with a slice thickness of 4 mm, field of view: 230 mm, matrix: 256 x 205) and a turbo fluid attenuated inversion recovery (FLAIR; TR/TE 11000/125 ms). The acquisition time for the diffusion-weighted sequences was 69 seconds.

All MRI images were assessed by a core laboratory blinded to neurological status and procedure. Diffuse alterations in the diffusion weighted image were not regarded as embolic types of lesions. Number of ischemic lesions, overall single lesion volume, mean lesion volume per patient, total ischemic volume per patient and location of all focal diffusion abnormalities (bright lesions on DWI) were documented.

Subsequently the DWI data of DEFLECT II study participants were retrospectively compared to 37 patients who had undergone TAVR between September 2012 and May 2013, without the use of an EPD (historical group). Patients in the historical group underwent post-procedural cerebral DWI as part of standard post-TAVR clinical care.

Endpoints and definitions The primary endpoints of this trial included device performance success and a composite safety endpoint during and immediately after TAVR. Successful device performance was defined as meeting all four of the following criteria: I) accessing the aortic arch with the delivery catheter; II) deployment of the TriGuard™HDH filter unit from the delivery catheter into the aortic arch; III) positioning the TriGuard™HDH to cover the cerebral vessels (as verified by angiography) without obstruction of the blood flow and without interference with the TAVR procedure (defined as ability to successfully advance, deploy, and retrieve the TAVR device); and IV) retrieving and removing the TriGuard™HDH system intact. The primary safety endpoint in this study was major adverse cardiac and cerebrovascular event (MACCE) during hospitalization defined as the composite of the following TAVR-related complications: cardiovascular mortality, disabling stroke (according to VARC-2), life-threatening bleeding (according to VARC2), distal embolization (noncerebral) from a vascular source requiring surgery or resulting in irreversible end-organ damage, major vascular or access site injury, and lastly need for acute cardiovascular surgery. The secondary endpoints of the present study were the number and the volume of new ischemic lesions on post-procedural cerebral DWI. Statistical methods All analyses were performed with the use of SPSS software (version 20, SPSS Inc, Chicago, IL).

Continuous variables are expressed as mean ± standard deviation (SD) or median [interquartile range, IQR]. Categorical variables are described by frequencies and percentages. Comparison of continuous variables was performed using the Mann-Whitney U test. Comparison of categorical variables was performed using the Fisher’s exact test and a value of P < 0.05 was considered statistically significant.

182

TRIGUARD™HDH EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

RESULTS Patient characteristics Twenty-four patients were screened for inclusion in this study. Nine patients were excluded because of consent denial (n=4), contraindication for MRI (n=1), mobile atherosclerotic plaque in the aortic arch (n=1) and unsuitable aortic arch anatomy making stabilization of the TriGuard™HDH device challenging (n=3). Fifteen patients were included in this pilot study, 10 (67%) women and 5 (33%) men, with a mean age of 82 ±4 years. Baseline patient characteristics are presented in Table 1.

Table 1 | Characteristics of patients included in DEFLECT II

Procedural and clinical outcome

N = 15

Device performance success was

Age (years)

82 ± 4

achieved in 10 (67%) patients (Table

Female sex

10 (67)

2). Although in the remaining 5

Diabetes mellitus

4 (27)

(33%) patients the TriGuard™HDH

Hypertension

9 (60)

Hyperlipidaemia

5 (33)

Coronary artery Disease

8 (53)

Previous acute myocardial infarction

3 (20)

Previous PCI

6 (40)

Previous CABG

2 (13)

Porcelain aorta according to VARC-2

0 (0)

Atrial fibrillation

4 (27)

Carotid disease

1 (7)

Peripheral vascular Disease

2 (13)

Prior stroke or TIA

2 (13)

was 28.5 ± 7.7 minutes. The primary

Logistic EuroSCORE (%)

17 ± 9

safety endpoint MACCE did not

Data are shown as mean ± SD or n (%). CABG: coronary artery bypass grafting; PCI: percutaneous coronary intervention; TAVR: transcatheter aortic valve replacement; TIA: transient ischemic attack.

was initially positioned in the aortic arch, it was removed before valve deployment due to an irreversible dislocation of the protection device in 4 patients and owing to a small dissection in the wall of ascending aorta (caused by the 9 French delivery sheath) in another patient.

In this last case, the dissection could be treated conservatively.

The average total fluoroscopy time

occur in the study population. NIH Stroke Scale scores were 0 in all patients on admission and remained unchanged during hospital stay.

Furthermore, modified Rankin Scale scores ranged from 0 to 3 (average 2.1) on admission and remained unchanged during hospitalisation.

183

CHAPTER 12

Device performance success

Peri-procedural TIA or Stroke

Post-procedural cerebral DWI

Nr of lesions

Volume per lesion (ul)

Total infarct volume (ul)

No

Yes

11

22.19

244.12

29

No

No

Yes

9

12.45

112.01

23

Yes

No

Yes

1

11.78

11.78

ES

23

Yes

No

Yes

0

0

0

ES

29

No

No

Yes

1

5.04

5.04

TA

ES

23

Yes

No

No

-

-

-

TF

CV

26

Yes

No

No

-

-

-

8

TF

CV

29

No

No

Yes

12

17.18

206.16

Valve Type

Yes

TF/TA

31

n

Valve size

Table 2 | Procedural characteristics and cerebral DWI data per patient

1

TF

CV

2

TF

CV

3

TF

ES

4

TF

5

TF

6 7 9

TF

ES

26

No

No

No

-

-

-

10

TF

CV

29

Yes

No

Yes

8

16.65

133.16

11

TF

CV

29

No

No

Yes

7

13.49

94.43

12

TF

ES

26

Yes

No

Yes

4

38.18

152.72

13

TF

ES

23

Yes

No

Yes

1

12.40

12.40

14

TF

ES

23

Yes

No

Yes

4

6.45

25.82

15

TF

ES

26

Yes

No

Yes

9

11.49

103.38

Cerebral ischemic lesions after TAVR Pre-procedural DWI was performed in all 15 patients and none showed focal diffusion abnormalities. Post-procedural DWI was performed in 12 patients. Median number of DWI lesions per patient was 5.5 [1.0-9.0], with 12.4 [7.7-17.0] μl mean lesion volume and 98.9 [11.9147.8] μl total ischemic volume (Table 3).

Among DEFLECT II patients with device performance success, both the median number (4.0 vs 8.0) and median total ischemic volume (64.6 μl vs 103.2 μl) seemed to be smaller comparing to patients with unsuccessful device performance (Table 3).

Moreover, a trend towards smaller mean lesion volume per patient (12.4 [7.7-17.0] vs 25.1 [11.0, 61.0], p=0.110) and smaller total ischemic volume (98.9 [11.9-147.8] vs 129.4 [53.0, 357.0], p=0.156) was seen among DEFLECT II patients when compared to the historical unprotected comparison group (Table 3). No difference was observed in the median number of ischemic DWI lesions between DEFELCT II patients and the historical comparison group (5.5 vs 5.0, p=0.963).

184

8 (67)

Patients with multiple lesions

13.0 [6.9-16.3]

5.0 [1.5-8.5]

1.5 [0-5.3]

8.0 [2.5-11.3]

3 (75)

1 (25)

4 (100)

Unsuccessful device performance (n=4)

12.1 [7.7-20.8]

1.0 [0.3-2.8]

2.0 [0.3-6.3]

4.0 [1.0-8.8]

5 (63)

2 (25)

7 (88)

Device performance success (n=8)

25.1 [11.0-61.0]

2.0 [1.0-5.0]

2.0 [0.5-4.0]

5.0 [2.0-7.0]

31 (84)

4 (11)

35 (95)

Historical group (n= 37 )

23.7 [6.9-103.8]

Total ischemic volume left side

64.7 [12.5-153.1]

16.8 [0-66.2] 18.0 [3.1-87.3]

28.8 [1.8-72.8]

65.3 [10.4-191.3]

60.3 [1.2-132.4]

0.264

0.105

0.156

0.110

0.540

0.869

0.963

0.233

0.340

1.000

Overall vs HG

0.181

0.269

0.172

0.095

0.166

0.940

0.571

0.326

0.286

0.452

DPS vs HG

DEFLECT II vs HG

Data are shown as median [interquartile range] or n (%). PS: device performance success; DWI: diffusion weighted magnetic resonance imaging; HG: historical group; TAVR: transcatheter aortic valve replacement.

22.7 [0-72.8]

98.9 [11.9-147.8] 103.2 [27.4-182.6] 64.6 [11.9-147.8] 129.4 [53.0-357.0]

Total ischemic volume right side

Total ischemic volume per patient

Mean per patient

12.4 [7.7-17.0]

1.5 [1.0-6.0]

Left hemisphere and cerebellum

Lesion volume (µl)

2.0 [0-5.5]

Right hemisphere and cerebellum

5.5 [1.0-9.0]

3 (25)

Patient with single lesion

Lesions per patient

11 (92)

Patient with new lesions

Overall (n=12)

DEFLECT II

Table 3 | Findings on DWI after TAVR in DEFLECT II study as compared to a historical group with patients who underwent unprotected TAVR.

TRIGUARD™HDH EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

185

CHAPTER 12

DISCUSSION The present pilot study showed that the use of TriGuard™HDH embolic deflection device for cerebral protection during TAVR is safe with a device performance success of 67%.

Asymptomatic new ischemic lesions were detected in 92% of DEFLECT II study patients who underwent post-TAVR cerebral DWI. Furthermore we found a trend towards reduction in mean lesion volume and total ischemic volume in DEFLECT II study population comparing to historical control group.

The initial positioning of the TriGuard™HDH in the aortic arch was successful in all study patients. However, the protection device dislocated irreversibly before valve deployment step of TAVR in 5 patients. The latter was caused by the interaction between the TriGuardTMHDH device and the valve delivery system. The current TriGuard™HDH device does not have additional features for a better stabilization when compared to the previous TriGuard™ device.

The DEFLECT I study, which investigated the performance of the TriGuard™ device, showed a device performance success of 80% among 37 patients. As the intended 100% device success was not reached in both DEFLECT I and DEFLECT II studies, the TriGuard™HDH device clearly needs technical improvements in order to ameliorate its stabilization in the aortic arch. The latter is necessary in order to facilitate an easy application of the device and to increase the efficiency of its use during TAVR. Obviously, intraprocedural use of an EPD is proposed in order to minimize the risk of cerebral ischemic injury during TAVR. However, the TriGuard™HDH evaluated in the present study did not decrease the frequency of new ischemic lesions on post-TAVR brain imaging. Comparable observation was reported by the previous DEFLECT I study. Thus, according to our data improvements in the previous version of the Keystone heart device, including a smaller pore size (from 250 um to 140 um), has not led to reduction in number of lesions compared to unprotected TAVR. Interestingly, even the Embrella embolic deflection device (Edwards LifeSciences, Irvine, CA), which has the smallest pore size (100 µm) among all tested cerebral protection devices for usage during TAVR, has shown no reduction in the number of DWI lesions after TAVR.17. Further reduction of the pore size might be undesirable, as that might jeopardize blood flow to the cerebral circulation.

In regard to the size of the ischemic injury, the DEFLECT I study reported 61% reduction in single lesion volume and 57% in total lesion volume (compared to a historical control group) among patients with device performance success. Similar to the DEFLECT I study, our study showed a trend toward reduction in both single lesion (52% reduction) and total ischemic volume (50% reduction) in the protected (patients with device performance success) TAVR group as compared to the historical control group. However, the clinical significance of such reduction in the volume of ischemic infarctions is unknown. Indeed in the light of broadening the indication for TAVR to younger and lower risk patients, reduction in cerebral ischemic burden to any extent is probably favorable.

Several other types of cerebral embolic protection devices, either deflecting or capturing embolic particles, have been tested in patients undergoing TAVR (Table 4). The Embrella deflector and the Montage 2 (Claret Medical, Santa Rosa, CA) are both delivered via radial

186

TRIGUARD™HDH EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

# of cases in the study

TIA/Stroke

% patients with DWI lesions

Average DWI lesions per patient

Nietlispach et al.13

4

0%

25%

0.25

Samim et al.

15

0%

100%

10.8

15

-

100%

3.2

≥ 5.4% 79%

5.1

Study

Pore Size

Mechanism of work

Access site

Device

Table 4| Embolic protection devices studied for usage during TAVR

Embrella

Radial/ulnar/ brachial

Deflecting 100 µm

SMT

Femoral

Deflecting ~200 µm Onsea et al.14

TriGuard™

Femoral

Deflecting 250 µm

Lansky et al.

28

TriGuard™HDH

Femoral

Deflecting 140 µm

Samim et al.

15

0%

Claret Medical

Radial/brachial Capturing 140 µm

Naber et al.

35

0%

-

-

Embol-X

Transaortic

Etienne et al.16

1

0%

0%

0

Capturing 120 µm

17

18

15

88%

5.5

DWI: diffusion weighted magnetic resonance imaging; TAVR: transcatheter aortic valve replacement; TIA: transient ischemic attack.

artery access, and offer protection to the brachiocephalic and left common carotid arteries.

Conversely, the keystone Heart embolic deflection devices, the Shimon Embolic protection filter or SHEF, and both TriGuard™ and TriGuart™HDH are delivered via femoral access and are designed to protect all three aortic arch takeoffs. The benefit of femoral access is avoiding passage through the innominate artery, where dislodgment of atherosclerotic material might increase the embolic load in the right sided brain circulation.

Although limited, early clinical data have shown considerable promise for the embolic deflection devices, the Embrella and the Keystone heart deflection devices, at least in terms of reduction of post-procedural cerebral ischemic volume. Hence, further refinements to EPD technique may offer a modality for significant reduction of brain injury during TAVR. In addition, adequately powered and randomized controlled trials, implementing competent neurocognitive tests, are needed to investigate the benefit of EPD usage during TAVR. The ongoing randomised controlled DEFLECT III trial, investigating the benefit of the TriGuard™HDH device in a large group of patients may clarify the clinical benefit of the use of this cerebral protection device during TAVR. In addition to the increased intraprocedural embolic load, new onset atrial fibrillation occurring in 14% to 32% of patients undergoing TAVR, is shown to be a predictor of subacute cerebrovascular events, with in the largest study on the topic showing an odds ratio of 2.76 (95% confidence interval, 1.11–6.83).24-27 Therefore, adequate monitoring for detection of even short durations of NOAF and the institution of anticoagulant therapy in its occurrence are as important as intra-procedural use of EPD. Whether the institution of anticoagulant therapy in case of NOAF should occur as triple therapy (aspirin, clopidogrel, and heparin), dual therapy (aspirin and heparin), or heparin alone is yet to be evaluated.

187

CHAPTER 12

LIMITATIONS The main limitation of this study is the small number of patients included, although it was designed as a feasibility pilot study. A second limitation is the lack of a prospective control group.

However, comparing the DWI results of the DEFLECT II study population to a historical control group, consisting of patients treated in the same institution and under similar procedural circumstances provides important information about the usefulness of the TriGuard™HDH protection device during TAVR. Further clarification will come with the results of the ongoing randomised controlled DEFLECT III trial, investigating the benefit of the TriGuard™HDH device in a large group of patients.

CONCLUSIONS The current study shows that the use of the TriGuard™HDH for cerebral protection during TAVR is safe. Device performance success was achieved in 67% of all cases. The susceptibility for dislocation due to interaction with the TAVR system is an important limitation of this device. No reduction was observed in the number of post-procedural cerebral ischemic lesions when TriGuard™HDH was used. However, its use led to a reduction in infarction volumes, which may reduce cerebral ischemic burden during TAVR. The results of the ongoing DEFLECT III trial will offer more insights in the benefit of intra-procedural use of TriGuard™HDH for cerebral protection.

188

TRIGUARDâ&#x201E;˘HDH EMBOLIC DEFLECTION DEVICE FOR CEREBRAL PROTECTION DURING TAVR

REFERENCES 1. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011 Jun 9;364(23):2187-98.

2. Adams DH, Popma JJ, Reardon MJ, Yakubov SJ, Coselli JS, Deeb GM, et al. Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med. 2014 May 8;370(19):1790-8.

3. Arnold M, Schulz-Heise S, Achenbach S, Ott S, Dorfler A, Ropers D, et al. Embolic cerebral insults after transapical aortic valve implantation detected by magnetic resonance imaging. JACC Cardiovasc Interv. 2010 Nov;3(11):1126-32.

4. Ghanem A, Muller A, Nahle CP, Kocurek J, Werner N, Hammerstingl C, et al. Risk and fate of cerebral embolism after transfemoral aortic valve implantation: A prospective pilot study with diffusionweighted magnetic resonance imaging. J Am Coll Cardiol. 2010 Apr 6;55(14):1427-32.

5. Rodes-Cabau J, Dumont E, Boone RH, Larose E, Bagur R, Gurvitch R, et al. Cerebral embolism following transcatheter aortic valve implantation: Comparison of transfemoral and transapical approaches. J Am Coll Cardiol. 2011 Jan 4;57(1):18-2.

6. Fairbairn TA, Mather AN, Bijsterveld P, Worthy G, Currie S, Goddard AJ, et al. Diffusion-weighted MRI determined cerebral embolic infarction following transcatheter aortic valve implantation: Assessment of predictive risk factors and the relationship to subsequent health status. Heart. 2012 Jan;98(1):18-23.

7. Kahlert P, Knipp SC, Schlamann M, Thielmann M, Al-Rashid F, Weber M, et al. Silent and apparent cerebral ischemia after percutaneous transfemoral aortic valve implantation: A diffusion-weighted magnetic resonance imaging study. Circulation. 2010 Feb 23;121(7):870-8.

8. Knecht S, Oelschlager C, Duning T, Lohmann H, Albers J, Stehling C, et al. Atrial fibrillation in strokefree patients is associated with memory impairment and hippocampal atrophy. Eur Heart J. 2008 Sep;29(17):2125-32.

9. Vermeer SE, Prins ND, den Heijer T, Hofman A, Koudstaal PJ, Breteler MM. Silent brain infarcts and the risk of dementia and cognitive decline. N Engl J Med. 2003 Mar 27;348(13):1215-22.

10. Fanning JP, Walters DL, Platts DG, Eeles E, Bellapart J, Fraser JF. Characterization of neurological injury in transcatheter aortic valve implantation: How clear is the picture? Circulation. 2014 Jan 28;129(4):504-15.

11. Rodes-Cabau J, Webb JG, Cheung A, Ye J, Dumont E, Feindel CM, et al. Transcatheter aortic valve implantation for the treatment of severe symptomatic aortic stenosis in patients at very high or prohibitive surgical risk: Acute and late outcomes of the multicenter canadian experience. J Am Coll Cardiol. 2010 Mar 16;55(11):1080-9.

12. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010 Oct 21;363(17):1597-60.

13. Nietlispach F, Wijesinghe N, Gurvitch R, Tay E, Carpenter JP, Burns C, et al. An embolic deflection device for aortic valve interventions. JACC Cardiovasc Interv. 2010 Nov;3(11):1133-8.

14. Onsea K, Agostoni P, Samim M, Voskuil M, Kluin J, Budde R, et al. First-in-man experience with a new embolic deflection device in transcatheter aortic valve interventions. EuroIntervention. 2012 May 15;8(1):51-6.

15. Naber CK, Ghanem A, Abizaid AA, Wolf A, Sinning JM, Werner N, et al. First-in-man use of a novel embolic protection device for patients undergoing transcatheter aortic valve implantation. EuroIntervention.

2012 May 15;8(1):43-50.

16. Etienne PY, Papadatos S, Pieters D, El Khoury E, Alexis F, Price J, et al. Embol-X intraaortic filter and transaortic approach for improved cerebral protection in transcatheter aortic valve implantation. Ann Thorac Surg. 2011 Nov;92(5):e95-6.

17. Samim M, Agostoni P, Hendrikse J, Budde RP, Nijhoff F, Kluin J, et al. Embrella embolic deflection device for cerebral protection during transcatheter aortic valve replacement. The Journal of thoracic and cardiovascular surgery. 2014.

189

CHAPTER 12

18. Lansky A, Baumbach A, Meller S, Brickman A, Gambone L, Margolis P, et al. <br />Deflect I trial. A prospective single arm feasibility study to evaluate the safety and performance of the keystone heart TriGuard embolic defletion device in patients undergoing transcatheter aortic valve replacement (TAVR). EuroPCR. In press 2013.

19. Van Mieghem NM, Schipper ME, Ladich E, Faqiri E, van der Boon R, Randjgari A, et al. Histopathology of embolic debris captured during transcatheter aortic valve replacement. Circulation. 2013 Jun 4;127(22):2194-201.

20. Eggebrecht H, Schmermund A, Voigtlander T, Kahlert P, Erbel R, Mehta RH. Risk of stroke after transcatheter aortic valve implantation (TAVI): A meta-analysis of 10,037 published patients.

EuroIntervention. 2012 May 15;8(1):129-38.

21. Omran H, Schmidt H, Hackenbroch M, Illien S, Bernhardt P, von der Recke G, et al. Silent and apparent cerebral embolism after retrograde catheterisation of the aortic valve in valvular stenosis: A prospective, randomised study. Lancet. 2003 Apr 12;361(9365):1241-6.

22. Astarci P, Glineur D, Kefer J, D'Hoore W, Renkin J, Vanoverschelde JL, et al. Magnetic resonance imaging evaluation of cerebral embolization during percutaneous aortic valve implantation: Comparison of transfemoral and trans-apical approaches using edwards sapiens valve. Eur J Cardiothorac Surg. 2011 Aug;40(2):475-9.

23. Ghanem A, Kocurek J, Sinning JM, Wagner M, Becker BV, Vogel M, et al. Cognitive trajectory after transcatheter aortic valve implantation. Circ Cardiovasc Interv. 2013 Dec 1;6(6):615-24.

24. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011 Jun 9;364(23):2187-98.

25. Nuis RJ, Van Mieghem NM, Schultz CJ, Moelker A, van der Boon RM, van Geuns RJ, et al. Frequency and causes of stroke during or after transcatheter aortic valve implantation. Am J Cardiol. 2012 Jun 1;109(11):1637-43.

26. Nombela-Franco L, Webb JG, de Jaegere PP, Toggweiler S, Nuis RJ, Dager AE, et al. Timing, predictive factors, and prognostic value of cerebrovascular events in a large cohort of patients undergoing transcatheter aortic valve implantation. Circulation. 2012 Dec 18;126(25):3041-53.

27. Amat-Santos IJ, Rodes-Cabau J, Urena M, DeLarochelliere R, Doyle D, Bagur R, et al. Incidence, predictive factors, and prognostic value of new-onset atrial fibrillation following transcatheter aortic valve implantation. J Am Coll Cardiol. 2012 Jan 10;59(2):178-8.

190

191

PART FOUR

GENERAL DISCUSSION AND CONCLUSIONS

CHAPTER

13 GENERAL ISSUES DISCUSSED IN THIS THESIS

CHAPTER 13

Before the development of transcatheter aortic valve replacement (TAVR), inoperable patients with severe symptomatic aortic valve stenosis were left with no therapeutic option to improve survival. At its first introduction in 2002, TAVR was embraced by some and received with suspicion by others. However the results of large registries and randomised controlled trials have demonstrated superiority to medical therapy (including balloon valvuloplasty) and non-inferiority to surgical aortic valve replacement (SAVR) in terms of mid- and long-term mortality.1-4 Technological advances, innovative techniques and the miniaturization of TAVR enabling devices have simplified the procedure, reduced the risk of complications and set the stage for broadening its indication. Today, 12 years after the first introduction of TAVR, more than 100.000 procedures have been performed worldwide. The benefits associated with the application of this technique, however, are still mitigated by the occurrence of substantial complications associated with increased mortality and reduced quality of life. The growing practice of TAVR worldwide has increased the need to understand and to overcome the limitations of this treatment. Improvements in valve implantation technology, optimisation of procedural planning using different imaging techniques and insights into the complicating factors are important to improve the results of TAVR, especially in the light of shifting the indication for this treatment to younger and lower risk patients. The present thesis studies possible improvements in the planning of TAVR and it furthermore sought to provide key insights on a few devastating complications related to this procedure.

Accurate positioning of transcatheter aortic valve For a successful TAVR it is important to accurately position and deploy the valve prosthesis in the 3-dimensional (3D) space of the aortic annulus and root. Incorrect positioning and deployment of the valve prosthesis may result in aortic regurgitation, heart block, valve embolization, coronary obstruction, or impairment of mitral valve and left ventricular function.5-7 In order to decrease the risk of these complications, the valve prosthesis has to be implanted in the optimal implantation view (OIV), which is defined as the fluoroscopic view perpendicular to the native valve, aligning all three sinuses of Valsalva.8 Fluoroscopy (using X-ray) is the main imaging modality used to capture real time images with a C-arm system to provide guidance to the operator during TAVR. Identifying the OIV and C-arm positioning is conventionally accomplished with repeated aortic root angiograms, which often costs valuable implantation time, increases radiation exposure and leads to higher contrast loads (Chapter 3). In chapter 2 and 3 we describe the feasibility and clinical benefit of OIV prediction using multislice computed tomography (MSCT) for prosthesis positioning and deployment during TAVR. The implementation of MSCT in the procedural planning decreased procedural time, and the amount of radiation and contrast used during TAVR in our center. It also led to an improved procedural safety expressed in a lower rate of valve embolization and significant aortic regurgitation (AR) probably reflecting a more appropriate valve positioning and deployment. Although MSCT is shown to be a valuable method for pre-procedural determination of OIV, it may be contra-indicated in some patients due to their kidney function and its use is only reliable in good MSCT image quality. Furthermore, another limitation of MSCT is the lack of real-time determination of the optimal implantation angles during TAVR,

194

GENERAL DISCUSSION

whilst in the catheterization laboratory the patient may be positioned differently on the table as compared to patient’s position under the MSCT scanner, leading thus to different angles.

More recently, rotational angiography has been introduced for determination of the OIV during TAVR. This technique uses the C-arm’s ability to rotate rapidly around the patient to acquire angiographic images at numerous oblique projections around its arch. As demonstrated in Chapter 5, the fully automatic software, called the DynaCT®, first generates a 3D reconstruction of the aortic valve and root and it subsequently determines a wide range of appropriate OIVs that the operator can choose from. As demonstrated in Chapter 5, DynaCT® is able to accurately predict optimal angiographic projections for balloon-expandable TAVR. The same chapter showed a good correlation between the coordinates of OIVs found by MSCT and DynaCT®, however when compared to MSCT, intra-procedural use of DynaCT® resulted in a significant reduction in the number of aortograms needed to find the OIV and in the radiation dose used during TAVR. As many centers in the world do not have the DynaCT® technology available in their catheterization lab, MSCT is still a valuable method for C-arm positioning during TAVR, although missing the important real-time aspect of the DynaCT®.

Even when DynaCT® is available for intra-procedural use, MSCT remains an important part of the workup procedure before TAVR and provides valuable anatomic assessment of the aortic root, annulus plane, distance to the coronary ostia, calcifications and iliofemoral access.

However intra-procedural rotational angiography based DynaCT®, on the top of pre-procedural MSCT, is of additional value for a successful implantation of the balloon expandable aortic valve prosthesis. Although Chapter 5 shows the benefit of the DynaCT® technique for only Edwards SAPIEN implantation, previous studies have also shown the benefit of this system for the implantation of self-expandable Medtronic CoreValve system.9 Besides its benefit for an increased accuracy of prosthesis implantation, DynaCT® 3D reconstruction of 2-dimensional fluoroscopic images can also be used for post-procedural evaluation of prosthesis position in difficult cases.10 Its easy integration with angiographic imaging during TAVR and its ability to provide valuable 3D reconstructions of anatomical structures make DynaCT® an important adjunct to the imaging techniques needed for a successful TAVR. Valvular regurgitation after TAVR and associated factors One of the most frequent complications after TAVR is post-procedural AR with > 50% of patients having at least mild AR.3 However, even mild post-procedural AR is associated with 1015% higher mortality at 2 years than patients with none or trace AR, as suggested by results of PARTNER cohort A. 3 Moreover, several other studies have shown similar associations between varying degrees of significant AR and increased late mortality after TAVR.11-13 Therefore much of the current research for improvements in TAVR technique focuses on the predictors of AR and methods to prevent this frequently occurring phenomenon. Contributing factors include valve undersizing or underexpansion, too high or too low placement and valve mal-alignment.14 One of the frequent causes of AR is inaccurate prosthesis sizing. As opposed to SAVR, during which exposure of the aortic valve enables direct measurement of the annulus dimensions by the surgeon, prosthesis sizing in TAVR is highly dependent on imaging modalities. In the initial years after the introduction of TAVR (at the moment of writing chapter 5), transthoracic

195

CHAPTER 13

and transesophageal echocardiography were the conventional tools for assessment of aortic valve annulus diameters needed for prosthesis sizing. However, both are two-dimensional imaging modalities and given the oval shape of the aortic annulus, they usually measure the smallest (sagittal) aortic annulus diameter. Therefore, basing prosthesis sizing on these imaging techniques frequently caused adverse results such as prosthesis migration and significant AR after TAVR. Operators were looking for strategies to overcome the sizing problem. As proposed in Chapter 5, excluding patients with an annulus diameter at the higher border of the range recommended by the manufacturer (borderzone annulus diameter) for a particular prosthesis size, and oversizing of the valve prosthesis according to intra-procedural balloon valvuloplasty, were the first steps to overcome the undersizing problem. In the more recent years, MSCT is shown to be a more reliable tool for measurements of aortic annulus dimensions and is generally recommended to be used for the conformation of the choice of prosthesis size. Hence, in the past years the incidence of prosthesis undersizing attributable to inaccurate annulus measurements decreased and as a consequence also the rate of related complications such as prosthesis malposition and aortic valve regurgitation. At the moment of writing, 10% prosthesis oversizing is still recommended in order to prevent complications; however this knowledge is based on earlier studies using 2D echocardiography for prosthesis sizing. As computed tomography has become the gold standard for annulus measurements and prosthesis sizing based on the area and perimeter of the aortic annulus, the current 10% oversizing rule needs validation in the future studies. In addition to extended knowledge regarding more accurate imaging techniques, the introduction of a wider range of valve prostheses facilitates more accurate prosthesis sizing and may make oversizing less relevant.

Also aortic valve calcification, as shown in Chapter 6, is associated with the severity of postprocedural AR. The presence of bulky calcification at the level of the commissures and on the cusps probably prevents adequate alignment of the stented prostheses against the aortic wall, with a resultant defective seal between these structures. In regard to calcification distribution, our data showed no association between an asymmetrical calcification distribution in the aortic valve and post-TAVR AR severity. In Chapter 6 we have also shown that lack of a sufficient amount of calcium in the aortic valve apparatus, might increase the risk of prosthesis malposition as calcium is needed to stabilize/fixate the valve prosthesis in the aortic annulus.

In case of prosthesis malposition or significant post-procedural AR, a “secondary maneuver” (usually post-dilatation or implantation of a second prosthesis) will be necessary. Indeed our data showed a higher amount of calcification volume in patients needing a “secondary maneuver” compared to patients without the need for post-dilatation or additional prosthesis implantation during TAVR. Therefore, in addition to annulus dimensions, careful evaluation of aortic valve apparatus calcification might provide important information about the risk of aortic valve regurgitation or prosthesis malposition. The amount of calcification was however not associated with the 30-days safety endpoint, which was a composite of AKI, cerebrovascular events, pacemaker implantations and death.

New TAVR systems have been designed to reduce AR after TAVR and the introduction of SAPIEN 3 prosthesis by Edwards Lifesiences is a good example of these recent developments. In this latter valve prosthesis, an outer skirt is designed to minimize paravalvular regurgitation after

196

GENERAL DISCUSSION

TAVR. In chapter 7, we describe the aims and the design of the randomised controlled trial “Edwards SAPIEN Periprosthetic Leakage Evaluation versus medtronic Corevalve in Tranfemoral aortic valve implantation” (the ELECT trial), which is initiated and started in Utrecht. In the ELECT trial we aim to investigate the difference in the severity of post-TAVR aortic regurgitation between patients undergoing the implantation of the Edwards SAPIEN 3™ bioprosthesis versus patients receiving the Medtronic CoreValve® system. The results of this trial are expected in 2016 and will provide us with important insights on differences in prosthesis function and clinical (according to Valve Academic Research Consortium-2 definitions) outcome between these two different types of percutaneous valves which dominate the TAVR market at this moment.

Ischemic brain injury during TAVR Although the incidence of stroke is relatively low after TAVR, with 1.5% at 24 hours, 3.3% at 1 month, 4.3% at 6 months and 5.2% at 1 year follow-up, it still is a devastating complication that calls for preventive measures.15 In addition, clinically apparent cerebrovascular events represent only the tip-of-the-iceberg. As reported in Chapter 9, subclinical and silent cerebral ischemic lesions are reported in more than 90% of patients undergoing TAVR. The small size (50% <20 microliter) of these cerebral lesions (Chapter 9), makes them less likely to cause new focal neurological deficits. The clinical significance of these silent ischemic lesions remains controversial, with ambiguous results across studies. On the short term, no clear correlation has been established between silent brain infarction and measurable impairment in neurocognitive function or apparent neurological events within the first 3 months after TAVR 16. Furthermore, additional evidence suggests also no impact on midterm and long-term prognosis, in terms of cognitive performance, mortality, self-sufficiency, or health-related quality of life outcomes.17, 18 The significance of the subclinical cerebral ischemic lesions may however become apparent when TAVR indication is broadened to younger lower risk patients with less pre-existing physical and neurocognitive impairments and higher demands for cognitive function. Furthermore, previous research involving patients other than the typical TAVR population has reported a significant association between silent brain infarctions and frailty, decline in cognitive and physical functions, early development of dementia and an increased risk of subsequent stroke. For instance, the population-based Rotterdam Scan Study showed that elderly people with silent brain infarcts have an increased risk of dementia and a steeper decline in cognitive function than those without such lesions.19 Expansion of the understanding of neurological injury associated with TAVR has fuelled the interest in strategies for their prevention. In Chapter 9 we found an independent association between age, hyperlipidaemia at baseline and balloon post-dilatation and the number of postTAVR ischemic brain lesions. Only peak transaortic gradient was independently associated with post-procedural total infarct volume. According to these data, omitting balloon post-dilatation will already decrease intraprocedural cerebral ischemic burden. Recently, much effort was put into the use of dedicated embolic protection devices (EPD) for cerebral protection during TAVR. Several types of EPD, either deflecting of capturing embolic particles, have been subject of research in the past recent years. Two different types of deflecting EPD have

197

CHAPTER 13

been studied for cerebral protection in the present thesis: the Embrella Embolic deflector system (Edwards LifeSciences, Irvine, CA), and the Keystone Heart embolic deflector devices (SMT embolic deflection device and TriGuardâ&#x201E;˘HDH). Three important features distinguish different types of protection devices from each other, namely the pore size, the access site of device introduction, and capturing versus deflecting. In general, the devices evaluated in this thesis and all other cerebral protection devices that have been tested up to now, have shown limited value for reduction of the number of embolic lesions. Although, as shown in Chapter 11 and 12, a reduction was observed in single lesion volume with the use of the Embrella device and an encouraging trend towards a reduction in both single lesion and total cerebral ischemic burden in case of the TriGuardâ&#x201E;˘HDH device. Obviously, the studies to date on the field of EPD for TAVR are feasibility studies and lack adequate power and design to proof effectiveness in terms of clinical endpoints. The on-going randomised controlled studies DEFLECT III and CLARET, investigating the TriGuard deflection device and the Claret capturing device, respectively, for cerebral protection during TAVR, will provide more solid information about the effectiveness of these devices. Although, before paying much more attention to EPD, the clinical significance of the small subacute cerebral lesions need to be clarified, as the incidence of stroke is low and is even decreasing with improvements in the TAVR technique with less traumatic devices.

Future challenges and opportunities Despite the rapid acceptance and clinical appeal of TAVR, as with any novel medical therapy, there are still many challenges to be addressed and opportunities to be explored. Current research is primarily aimed at reducing post-TAVR complications and extending TAVR to a broader group of patients.

Together with the growing experience and technical improvements, a trend towards broadening the indication for TAVR is observed. Although TAVR originally was indicated in high risk or inoperable patients, clinical practice around the world has already evolved as elderly patients (>80 years) with none or one co-morbidity are often treated with TAVR. The next logical use extension would be younger intermediate risk patients (STS 4-8%) with severe symptomatic aortic stenosis, representing one-quarter to one-third of surgical patients.

Supporting the latter, two propensity risk adjustment studies in intermediate risk patients comparing TAVR and SAVR has indicated similar early and late mortality.20, 21 The results of two ongoing important large randomized clinical trials, PARTNER 2A (SAPIEN XT valve) and SURTAVI (CoreValve), in intermediate risk patients comparing TAVR versus SAVR will help to clarify the questions regarding the benefits of TAVR in this risk strata. However, in order to implement the current TAVR technique to lower risk and younger patients, a reduced risk of stroke, vascular injury, and post-procedural regurgitation is needed together with an improved durability of the frame and valve itself.

Several other patient subgroups and clinical indications would seem reasonable candidates for TAVR therapy. Recently, transcatheter heart valve implantation within failed surgically implanted bioprosthesis has been proven feasible and clinical outcomes are generally similar to native-valve TAVR accounting for relative differences in patient co-morbidities.22-26 Probably

198

GENERAL DISCUSSION

one of the most important issues in valve-in-valve-procedures is a complete knowledge of the differences among surgical valve prostheses, as the true inner diameter of the surgical valve, the location of the sewing ring, fluoroscopic landmarks and the placement of the valve relative to the frame (inside or outside) are some of the critical features which impact the results of valve-in-valve implantation.

The TAVR technology is also evolving. Intra-procedural complication rate will decrease with down-sizing of the device systems and the introduction of new types of valve prostheses with retrievable features and advanced design for preventing post-procedural regurgitation.

Moreover, adequately powered randomized clinical trials with a long follow-up are necessary to determine if routine or selective use of EPD for cerebral protection will improve clinical outcome after TAVR. In addition, more attention should be paid to the management of new onset atrial fibrillation (NOAF) occurring in 14% to 32% of patients undergoing TAVR. NOAF is shown to be a predictor of subacute cerebrovascular events, with in the largest study on the topic showing an odds ratio of 2.76 (95% confidence interval, 1.11â&#x20AC;&#x201C;6.83).27-30 Therefore, adequate monitoring for detection of even short durations of NOAF and the immediate institution of anticoagulant therapy in its occurrence are as important as intra-procedural use of EPD. Whether the institution of anticoagulant therapy in case of NOAF should occur as triple therapy (aspirin, clopidogrel, and heparin), dual therapy (aspirin and heparin), or heparin alone is yet to be evaluated.

Other possible mechanisms of brain protection during TAVR need future evaluation.

Hemodynamic instability during TAVR may cause systemic hypotension which can impair cerebral perfusion pressure beyond cerebral autoregulatory capacity, resulting in hypoperfusion. Although in itself a cause for ischemia, low cerebral flow magnifies the effects of microemboli by impairing their clearance and permitting small emboli to lodge.31 Based on the foregoing, one may expect that increasing the cerebral blood flow may allow an improvement in the washout of emboli. Furthermore, that might enhance dissolution of emboli due to an increase in flow velocity and level of fibrinolytic factors at the site of the occlusion. The foregoing hypothesis needs to be explored in future studies.

Aortic valve regurgitation is one of the most frequent complications after TAVR, while previous research has suggested an association between this complication and late mortality.

However, evaluation of TAVR is limited by a lack of validated techniques for quantification of the regurgitant volume. Paravalvular AR jets are often eccentric, small and irregular, and thus qualitative assessment methods may underestimate the severity of regurgitation. Quantitative assessment of total AR using advanced imaging modalities may be a feasible way to overcome this problem. The use of 3D transesophageal echocardiography or Q-flow cardiac magnetic resonance imaging are currently investigated for this aim in the previously mentioned ELECT trial and will probably clarify this issue in the future.

As moderate or severe mitral regurgitation is a common finding in patients with severe symptomatic aortic valve stenosis undergoing TAVR (Chapter 8), combined or staged transcatheter mitral interventions would improve the clinical outcome in a large group of patients. Although a few cases have been performed worldwide in the past few years, more experience with the technique and further developments in the mitral interventions will

199

CHAPTER 13

probably increase the rate of combined aortic and mitral valve interventions in the near future.

In short, with the increasing experience, technical improvements, the results of on-going clinical trials and broadening its indication, TAVR seems to follow the footstep of percutaneous coronary interventions.

200

GENERAL DISCUSSION

REFERENCES 1. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010 Oct 21;363(17):1597-60.

2. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011 Jun 9;364(23):2187-98.

3. Kodali SK, Williams MR, Smith CR, Svensson LG, Webb JG, Makkar RR, et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med. 2012 May 3;366(18):1686-95.

4. Adams DH, Popma JJ, Reardon MJ, Yakubov SJ, Coselli JS, Deeb GM, et al. Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med. 2014 May 8;370(19):1790-8.

5. Wong DR, Ye J, Cheung A, Webb JG, Carere RG, Lichtenstein SV. Technical considerations to avoid pitfalls during transapical aortic valve implantation. J Thorac Cardiovasc Surg. 2010 Jul;140(1):196-202.

6. Al Ali AM, Altwegg L, Horlick EM, Feindel C, Thompson CR, Cheung A, et al. Prevention and management of transcatheter balloon-expandable aortic valve malposition. Catheter Cardiovasc Interv. 2008 Oct 1;72(4):573-8.

7. Masson JB, Kovac J, Schuler G, Ye J, Cheung A, Kapadia S, et al. Transcatheter aortic valve implantation: Review of the nature, management, and avoidance of procedural complications. JACC Cardiovasc Interv. 2009 Sep;2(9):811-20.

8. Binder RK, Leipsic J, Wood D, Moore T, Toggweiler S, Willson A, et al. Prediction of optimal deployment projection for transcatheter aortic valve replacement: Angiographic 3-dimensional reconstruction of the aortic root versus multidetector computed tomography. Circ Cardiovasc Interv. 2012 Apr;5(2):24752.

9. Poon KK, Crowhurst J, James C, Campbell D, Roper D, Chan J, et al. Impact of optimising fluoroscopic implant angles on paravalvular regurgitation in transcatheter aortic valve replacements - utility of three-dimensional rotational angiography. EuroIntervention. 2012 Sep;8(5):538-45.

10. Incani A, Lee JC, Poon KK, Crowhurst JA, Raffel OC, Walters DL. Normal functioning of a constrained CoreValve with DynaCT imaging demonstrating incomplete stent frame expansion. Int J Cardiol. 2013 Feb 10;163(1):e9-10.

11. Moat NE, Ludman P, de Belder MA, Bridgewater B, Cunningham AD, Young CP, et al. Long-term outcomes after transcatheter aortic valve implantation in high-risk patients with severe aortic stenosis: The U.K. TAVI (united kingdom transcatheter aortic valve implantation) registry. J Am Coll Cardiol. 2011 Nov 8;58(20):2130-8.

12. Sinning JM, Hammerstingl C, Vasa-Nicotera M, Adenauer V, Lema Cachiguango SJ, Scheer AC, et al.

Aortic regurgitation index defines severity of peri-prosthetic regurgitation and predicts outcome in patients after transcatheter aortic valve implantation. J Am Coll Cardiol. 2012 Mar 27;59(13):1134-41.

13. Tamburino C, Capodanno D, Ramondo A, Petronio AS, Ettori F, Santoro G, et al. Incidence and predictors of early and late mortality after transcatheter aortic valve implantation in 663 patients with severe aortic stenosis. Circulation. 2011 Jan 25;123(3):299-308.

14. Genereux P, Head SJ, Hahn R, Daneault B, Kodali S, Williams MR, et al. Paravalvular leak after transcatheter aortic valve replacement: The new achilles' heel? A comprehensive review of the literature. J Am Coll Cardiol. 2013 Mar 19;61(11):1125-36.

15. Eggebrecht H, Schmermund A, Voigtlander T, Kahlert P, Erbel R, Mehta RH. Risk of stroke after transcatheter aortic valve implantation (TAVI): A meta-analysis of 10,037 published patients.

EuroIntervention. 2012 May 15;8(1):129-38.

16. Kahlert P, Knipp SC, Schlamann M, Thielmann M, Al-Rashid F, Weber M, et al. Silent and apparent cerebral ischemia after percutaneous transfemoral aortic valve implantation: A diffusion-weighted magnetic resonance imaging study. Circulation. 2010 Feb 23;121(7):870-8.

201

CHAPTER 13

17. Ghanem A, Muller A, Nahle CP, Kocurek J, Werner N, Hammerstingl C, et al. Risk and fate of cerebral embolism after transfemoral aortic valve implantation: A prospective pilot study with diffusionweighted magnetic resonance imaging. J Am Coll Cardiol. 2010 Apr 6;55(14):1427-32.

18. Ghanem A, Kocurek J, Sinning JM, Wagner M, Becker BV, Vogel M, et al. Cognitive trajectory after transcatheter aortic valve implantation. Circ Cardiovasc Interv. 2013 Dec 1;6(6):615-24.

19. Vermeer SE, Prins ND, den Heijer T, Hofman A, Koudstaal PJ, Breteler MM. Silent brain infarcts and the risk of dementia and cognitive decline. N Engl J Med. 2003 Mar 27;348(13):1215-22.

20. Latib A, Maisano F, Bertoldi L, Giacomini A, Shannon J, Cioni M, et al. Transcatheter vs surgical aortic valve replacement in intermediate-surgical-risk patients with aortic stenosis: A propensity scorematched case-control study. Am Heart J. 2012 Dec;164(6):910-7.

21. Piazza N, Kalesan B, van Mieghem N, Head S, Wenaweser P, Carrel TP, et al. A 3-center comparison of 1-year mortality outcomes between transcatheter aortic valve implantation and surgical aortic valve replacement on the basis of propensity score matching among intermediate-risk surgical patients.

JACC Cardiovasc Interv. 2013 May;6(5):443-51.

22. Webb JG, Wood DA, Ye J, Gurvitch R, Masson JB, Rodes-Cabau J, et al. Transcatheter valve-in-valve implantation for failed bioprosthetic heart valves. Circulation. 2010 Apr 27;121(16):1848-57.

23. Gurvitch R, Cheung A, Ye J, Wood DA, Willson AB, Toggweiler S, et al. Transcatheter valve-in-valve implantation for failed surgical bioprosthetic valves. J Am Coll Cardiol. 2011 Nov 15;58(21):2196-209.

24. Piazza N, Bleiziffer S, Brockmann G, Hendrick R, Deutsch MA, Opitz A, et al. Transcatheter aortic valve implantation for failing surgical aortic bioprosthetic valve: From concept to clinical application and evaluation (part 1). JACC Cardiovasc Interv. 2011 Jul;4(7):721-32.

25. Piazza N, Bleiziffer S, Brockmann G, Hendrick R, Deutsch MA, Opitz A, et al. Transcatheter aortic valve implantation for failing surgical aortic bioprosthetic valve: From concept to clinical application and evaluation (part 2). JACC Cardiovasc Interv. 2011 Jul;4(7):733-42.

26. Dvir D, Webb J, Brecker S, Bleiziffer S, Hildick-Smith D, Colombo A, et al. Transcatheter aortic valve replacement for degenerative bioprosthetic surgical valves: Results from the global valve-in-valve registry. Circulation. 2012 Nov 6;126(19):2335-44.

27. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011 Jun 9;364(23):2187-98.

28. Nuis RJ, Van Mieghem NM, Schultz CJ, Moelker A, van der Boon RM, van Geuns RJ, et al. Frequency and causes of stroke during or after transcatheter aortic valve implantation. Am J Cardiol. 2012 Jun 1;109(11):1637-43.

29. Nombela-Franco L, Webb JG, de Jaegere PP, Toggweiler S, Nuis RJ, Dager AE, et al. Timing, predictive factors, and prognostic value of cerebrovascular events in a large cohort of patients undergoing transcatheter aortic valve implantation. Circulation. 2012 Dec 18;126(25):3041-53.

30. Amat-Santos IJ, Rodes-Cabau J, Urena M, DeLarochelliere R, Doyle D, Bagur R, et al. Incidence, predictive factors, and prognostic value of new-onset atrial fibrillation following transcatheter aortic valve implantation. J Am Coll Cardiol. 2012 Jan 10;59(2):178-8.

31. Caplan LR, Hennerici M. Impaired clearance of emboli (washout) is an important link between hypoperfusion, embolism, and ischemic stroke. Arch Neurol. 1998 Nov;55(11):1475-82.

202

203

APPENDIX DUTCH SUMMARY ACKNOWLEDGMENTS CURRICULUM VITAE AUCTORIS LIST OF PUBLICATIONS

APPENDIX

DUTCH SUMMARY Inleiding Voor de ontwikkeling van transcatheter aortaklep implantatie (internationale term is transcatheter aortic valve replacement: TAVR), bestond er geen goede therapeutische mogelijkheid voor inoperabele patiënten met ernstige aortaklepstenose. Toen TAVR in 2002 voor het eerst werd geïntroduceerd, waren de reacties erg verschillend. Sommigen zagen deze behandeling als een potentiele alternatieve behandeling voor chirurgische aortaklep vervanging (SAVR) in oudere patiënten. Anderen waren daarentegen erg sceptisch.

Grote internationale studies hebben in de afgelopen jaren laten zien dat TAVR superieur is aan conservatieve therapie (inclusief ballonvalvuloplastiek) en zeker niet inferieur is aan chirurgische aortaklepvervanging wat betreft middellange en lange termijn mortaliteit. Door verbeteringen in de techniek, nieuwe innovaties en miniaturisatie van katheters en sheets die tijdens de procedures worden gebruik, is TAVR toegankelijker geworden en de kans op intraprocedurele complicaties is in de loop van jaren kleiner geworden. Daarmee is er de basis gelegd voor de uitbreiding van de indicatie van TAVR naar een bredere patiëntengroep.

Nu 12 jaar na de eerste klinische toepassing van TAVR, zijn er wereldwijd meer dan 100000 procedures uitgevoerd. TAVR heeft een aantal belangrijke voordelen, waaronder het minder invasieve karakter en het snellere klinische herstel, waardoor tegenwoorden ook operabele patiënten zelf voor deze procedure kiezen. Ondanks de vele voordelen van TAVR, zijn er een aantal belangrijke procedure-gerelateerde complicaties die ondanks een lage incidentie toch aandacht behoeven omdat ze de post-procedurele kwaliteit van leven en mortaliteit in de negatieve zin beïnvloeden. Gezien er wereldwijd steeds meer TAVR procedures worden uitgevoerd, is het belangrijk om aandacht te besteden aan de beperkingen van deze techniek en om het risico op complicaties te verkleinen. Verbeteringen in de klepimplantatietechnieken, optimalisatie van procedureplanning met behulp van verschillende beeldvormende technieken en kennis van complicerende factoren zijn belangrijk voor het verbeteren van de resultaten van TAVR. Alleen dan kan de indicatie voor TAVR worden uitgebreid naar jongere patiënten met een lager risicoprofiel. Dit proefschrift bestudeert de mogelijke verbeteringen in de planning van TAVR en het draagt daarnaast bij aan belangrijke inzichten in een aantal significante complicaties van dit type procedure.

Nauwkeurige plaatsing van percutane aortakleppen Voor een succesvolle TAVR is het zeer belangrijk om de klepprothese nauwkeurig in de aortaklepannulus te positioneren en te ontplooien. Onnauwkeurige klepplaatsing kan leiden tot aortaklepregurgitatie (AR), geleidingsstoornissen, embolisatie van de aortaklep, obstructie van de coronair arteriën, beschadiging van de mitralisklep of/en aantasting van linker ventrikel functie5-7. Om het risico op deze complicaties te verkleinen, is het van cruciaal belang om de klepprothese in de optimale fluoroscopische projectie (OFP) te implanteren. OFP is gedefinieerd is als de fluoroscopische projectie loodrecht op de natieve aortaklep, waarbij de drie sinussen van Valsalva op een lijn liggen8. Fluoroscopie (waarbij röntgenstraling wordt gebruikt voor beeldvorming) is de belangrijkste beeldvormende techniek tijdens TAVR en wordt

206

APPENDIX

gebruikt om real-time beelden te maken. Tot voor kort was voor het vinden van de OFP tijdens TAVR het maken van multipele angiogrammen noodzakelijk. Het zoeken naar OFP met slechts angiogrammen kost vaak te veel tijd en patiënten worden daarbij aan relatief grote hoeveelheid contrast en straling blootgesteld (Hoofdstuk 3). In hoofdstukken 2 en 3 beschrijven we de procedurele en klinische voordelen van het gebruik van multislice computertomografie (MSCT) voor pre-procedurele bepaling van de OFP voor kleppositionering tijdens TAVR. Zoals beschreven in hoofdstuk 3, het gebruik van MSCT voor dit doel heeft de proceduretijd en de hoeveelheid straling en contrast tijdens TAVR verkleind in ons centrum. Daarnaast heeft het gebruik van MSCT geleid tot veiligere procedures met een lagere incidentie van embolisatie van klepprotheses en significante AR. Deze bevindingen zijn waarschijnlijk gerelateerd aan een nauwkeurigere positionering van de aortaklepprothese. Hoewel MSCT een goede methode is voor pre-procedurele voorspelling van OFPs, het gebruik ervan is alleen betrouwbaar bij goede beeldkwaliteit van de MSCT. Daarnaast is MSCT gecontra-indiceerd bij patiënten met een ernstige nierfunctiestoornis. Een andere limitatie van de bepaling van OFP middels MSCT is het feit dat de metingen niet real-time gedaan kunnen worden. De ligging van de patiënt onder de pre-procedurele MSCT scanner en de ligging op tafel tijdens TAVR zouden van elkaar kunnen verschillen en dat zou kunnen leiden tot het berekenen van onnauwkeurige coördinaten van de OFPs. Recent is rotationele angiografie geïntroduceerd voor de real-time bepaling van de OFP gedurende TAVR. Deze techniek werkt als volgt: de C-arm draait 159 graden om de patiënt heen en stelt een driedimensionale (3D) dataset samen die vergelijkbaar is met MSCT beeldvorming. Zoals besproken in hoofdstuk 5, een volledig automatische software die DynaCT wordt genoemd produceert in de eerste instantie een 3D reconstructie van de aortaklep en aortawortel om vervolgens een reeks van OFPs te genereren waaruit de operateur kan kiezen.

Hoofdstuk 5 laat zien dat DynaCT in staat is om nauwkeurige OFPs te voorspellen voor de implantatie van de ballon-expandeerbare aortaklep. Datzelfde hoofdstuk laat verder zien dat er een goede correlatie is tussen de coördinaten van de OFP voorspeld door MSCT en gegenereerd door DynaCT. Echter vergeleken met MSCT, het intraprocedurele gebruik van DynaCT resulteerde in een significante vermindering van het aantal controle aortogrammen en van de hoeveelheid straling gebruikt tijdens TAVR. Omdat veel centra wereldwijd geen beschikking hebben over de DynaCT techniek, zal MSCT een belangrijk modaliteit blijven voor pre-procedurele bepaling van de OFP voor TAVR. Zelfs in aanwezigheid van DynaCT blijft MSCT een onmisbare beeldvormende techniek voor TAVR omdat daarmee de dimensies van de aortawortel en aortaklepannulus, de afstand tot de coronair arterie ostia, hoeveelheid calcificatie en de toegankelijkheid via het iliofemoraal traject gemeten kunnen worden. Hoewel hoofdstuk 5 alleen het gebruik van DynaCT voor de ballon expandeerbare aortaklep laat zien, echter eerdere studies hebben de toepassing ervan laten zien voor de implantatie van de Medtronic CoreValve9. Daarnaast kan DynaCT gebruikt worden voor post-procedurele evaluatie van hoe de klepprothese geplaatst is10. De makkelijke integratie van DynaCT met angiografietechniek in de hartkatheterisatiekamer en de toepassing ervan voor het maken van real-time 3D reconstructie van anatomische structuren, maken deze techniek een waardevolle toevoeging aan de beeldvormende technieken die nodig zijn voor een succesvolle TAVR.

207

APPENDIX

Aortaklepregurgitatie na TAVR en factoren die daarmee geassocieerd zijn Een van de meest frequente complicaties na TAVR is post-procedurele AR. Meer dan 50% van patiënten hebben post-TAVR op zijn minst milde AR3. Zoals gebleken uit de resultaten van PARTNER cohort A, zelfs milde post-TAVR AR is geassocieerd is met 10-15% hogere mortaliteit na 2 jaar vergeleken met patiënten die geen of alleen een spoortje AR hebben(3. Het is daarom ook niet vreemd dat veel onderzoekers geïnteresseerd zijn in de voorspellers van post-TAVR AR en daarnaast methodes om deze complicatie te voorkomen. Factoren die de kans op postTAVR AR vergroten zijn een te kleine klepprothese ten opzichte van de grootte van de natieve annulus (ondersizing), onvolledige ontplooiing van de klepprothese, een te hoge of te lage plaatsing van de klepprothese en een scheve plaatsing van de klepprothese in de 3D structuur van de aortaklep14.

Ondersizing met een discrepantie tussen de gekozen klepmaat en de grootte van de natieve annulus is veelal het gevolg van een inaccurate meting van de dimensies van de aortaklepannulus.

In tegenstelling tot de SAVR waarbij de dimensies van de natieve aortaklepannulus real-time gemeten kunnen worden onder directe visualisatie van de aortaklep, het bepalen van de klepmaat in het geval van TAVR is erg afhankelijk van adequate beeldvormende technieken.

In de eerste jaren na de introductie van TAVR (op het moment van het schrijven van hoofdstuk 5), transthoracale en transoesofageale echocardiografie waren de conventionele beeldvormingsmodaliteiten voor de meting van annulusdimensies en dus voor het kiezen van de maat van de klepprothese. Echter beide technieken zijn 2D modaliteiten en meten de kleinste (sagittale) diameter van de ovaalvormige aortaklepannulus. Het baseren van de klepmaat op deze beeldvormingsmodaliteiten veroorzaakte dan ook regelmatig complicaties zoals klepmigratie en belangrijke aortaklepregurgitatie. Zoals beschreven in hoofdstuk 5, het excluderen van patiënten met een annulusdiameter aan de bovengrens van de range aangegeven door de fabrikant (borderzone annulusdiameter) voor een bepaalde klepmaat, en “oversizing” van de klepprothese op basis van intraprocedurele ballonvalvuloplastiek, waren de eerste stappen om het probleem van ondersizing te verhelpen.

Recent is MSCT een betere methode gebleken voor het meten van de dimensies van de aortaklepannulus en heeft tegenwoordig de voorkeur voor de bepaling van de klepmaat.

Sindsdien is de incidentie van ondersizing van de klepmaat tijdens TAVR gedaald en daarmee de incidentie van complicaties zoals klepmalpositie en post-procedurele AR. Op dit moment wordt aangeraden om de klepmaat 10% groter te kiezen dan de gemeten aortaklepannulus diameter. Dit laatste is echter gebaseerd op relatief oude studies waarbij 2D echocardiografie is gebruikt voor de bepaling van de klepmaat. Omdat tegenwoordig de oppervlakte en de omtrek van de aortaklepannulus op MSCT worden gebruikt voor het bepalen van de klepmaat, moet deze 10% oversizing regel met toekomstige studies worden gevalideerd. Naast het gebruik van MSCT bij de bepaling van de klepmaat, heeft de introductie van meer klepmaten een preciezere klepsizing mogelijk gemaakt, waarbij oversizing minder belangrijk wordt.

Zoals besproken in hoofdstuk 6, is de mate van aortaklepcalcificatie geassocieerd met de ernst van postprocedure AR. Calcificatie ter hoogte van de commissuren van het klepapparaat en op de klepbladen voorkomen waarschijnlijk dat de aortaklepprothese volledig ontplooit en goed aansluit tegen de aortawand waardoor er ruimtes ontstaan voor AR. Onze data

208

APPENDIX

laat verder geen relatie zien tussen de asymmetrie van kalkdistributie en de ernst van postprocedurele AR. Wanneer er na klepimplantatie sprake is van significante AR of malpositie van de klepprothese, dan is vaak een extra behandeling geïndiceerd. Deze extra behandeling

is meestal

balloondilatatie van de klepprothese of implantatie van een additionele klepprothese.

Hoofdstuk 6 liet zien dat wanneer er een extra behandeling nodig is, dat er sprake is van meer kalk in het aortaklepapparaat. Derhalve pre-procedurele evaluatie van aortaklepcalcificatie geeft ons informatie over risico op post-TAVR AR of malpositie van de aortaklepprothese.

De hoeveelheid aortaklepcalcificatie was in onze studie verder niet geassocieerd met 30-dagen veiligheidseindpunt bestaande uit acute nierschade, cerebrovasculaire accidenten, pacemakerimplantatie en dood.

Nieuwe ontwikkelingen op het gebied van percutane aortakleppen, zoals de introductie van Edwards SAPIEN 3, moeten post-procedurele AR doen verminderen. Zo heeft Edwards SAPIEN 3 een extra buitenste rok die bedoeld is om paravalvulaire AR tegen te gaan.

In hoofdstuk 7 bespreken we de opzet van de gerandomiseerde trial “ Edwards SAPIEN periprosthetic Leakage Evaluation versus medtronic Corevalve in Transfemoral aortic valve implantation” (the ELECT trial) in Utrecht. De ELECT studie heeft als doel om de Edwards SAPIEN klepprothese te vergelijken met de Medtronic CoreValve voor verschil in de ernst van post-TAVR AR. De resultaten van deze trial worden verwacht in 2016 en zullen ons meer inzicht geven in verschillen in klepfunctie en klinische uitkomsten tussen de twee percutane aortaklepprotheses die op dit moment de TAVR markt domineren.

Ischemische hersenschade tijdens TAVR Hoewel de incidentie van herseninfarct relatief laag is na TAVR, met 1.5% op 24 uur, 3.3% op 1 maand, 4.3 % op 6 maanden en 5.2% op 1 jaar follow-up, toch maakt deze invallerende complicatie preventieve maatregelen noodzakelijk (15). Daarnaast zijn klinische cerebrovasculaire accidenten alleen een klein onderdeel van het probleem. Zoals besproken in Hoofdstuk 9, subklinische en stille cerebrale infarcten komen voor in meer dan 90% van patiënten die een TAVR ondergaan. Deze subklinische ischemische infarcten zijn echter over het algemeen klein (50% <20 microliter) (Hoofdstuk 9), en de kans dat ze neurologisch uitval veroorzaken, is gering. Het klinische belang van deze stille infarcten blijft daarom ook controversieel, met ambigue resultaten in verschillende studies. De studies tot nu toe hebben geen duidelijke associatie gevonden tussen post-TAVR stille herseninfarcten en waarneembare afwijkingen in neurocognitieve functies op korte termijn of klinische neurologische stoornissen binnen 3 maanden na TAVR

. Verder zijn er ook geen aanwijzingen gevonden voor een

(16

negatieve impact van deze stille herseninfarcten op lange termijn prognose wat betreft de cognitieve functie, mortaliteit, zelfredzaamheid en kwaliteit van leven17, 18. Het belang van de stille herseninfarcten zou echter zichtbaar kunnen worden wanneer TAVR indicatie wordt uitgebreid naar jongere patiënten met een lager risicoprofiel, met minder pre-existente fysieke beperkingen en neurocognitieve stoornissen en bij wie het behoud van neurocognitieve functie nog belangrijker is. Daarnaast hebben oudere studies, met patiëntenpopulatie anders dan de typische TAVR patiënten, een significantie associatie laten zien tussen stille herseninfarcten en

209

APPENDIX

“frailty”, vermindering van cognitieve en fysieke functies op lange termijn, vroege ontwikkeling van dementie en een verhoogd risico op een recidief herseninfarct. De Rotterdam Scan Study heeft bijvoorbeeld laten zien dat oudere patiënten met stille herseninfarcten een verhoogde kans hebben op dementie en een verdere achteruitgang van cognitieve functie dan patiënten zonder deze hersenlaesies19.

Gezien de hoge incidentie van post-TAVR stille herseninfarcten is er steeds meer aandacht voor strategieën om deze te voorkomen. In hoofdstuk 9 beschrijven we dat er een onafhankelijke relatie is tussen leeftijd, hyperlipidemie, balloon post-dilatatie van de klepprothese en het aantal post-TAVR ischemische cerebrale laesies. Volgens deze data zou het achterwege laten van post-dilatatie van de aortaklep al het risico op intra-procedurele cerebrale emobolisatie verkleinen. Wat betreft de totale volume aan herseninfarct bleek alleen de maximale gradiënt over de aortaklep een onafhankelijke relatie daarmee te hebben.

Recent is er veel aandacht voor het gebruik van een protectie device (PD) voor bescherming van de hersenen tijdens TAVR. Verschillende soorten PD zijn onderzocht in de afgelopen jaren.

Drie belangrijke kenmerken onderscheiden de verschillende soorten PD van elkaar, namelijk de poriegrootte, de introductieplaats van het device, en het werkingsmechanisme: opvangen of afkaatsen van embolieën. Twee verschillende soorten afkaatsende PDs voor gebruik tijdens TAVR zijn bestudeerd in het huidige proefschrift: de “Embrella embolic deflector system” (Edwards LifeSciences, Irvine, CA) (Hoofdstuk 11), en de “Keystone heart embolic deflector” (SMT embolic deflection device en TriGuard™HDH) (Hoofdstuk 10 en 12).

In het algemeen, hebben de PDs geëvalueerd in deze thesis en alle andere cerebrale PDs die tot nu toe getest zijn voor gebruik tijdens TAVR slechts een beperkte waarde laten zien voor het verminderen van het aantal cerebrale infarcten. Zoals besproken in hoofdstuk 11 en 12, er was een vermindering in het volume van laesies bij gebruik van de Embrella device tijdens TAVR en een tendens tot vermindering van zowel laesie volume als totale volume aan cerebrale ischemie bij gebruik van het TriGuard™HDH device. Het gebruik van EPD tijdens TAVR is tot nu toe slechts in pilotstudies bestudeerd en grotere studies met voldoende power en een adequaat design zijn nodig om het belang van EPD tijdens TAVR te onderzoeken. De lopende gerandomiseerde studies DEFLECT III (TriGuard device) en CLARET (Claret device) zullen in de nabije toekomst meer informatie opleveren over de effectiviteit van deze EPDs. Daarnaast, voordat we veel meer aandacht besteden aan EPD, moet de klinische betekenis van de kleine “stille” cerebrale infarcten worden onderzocht. Dit laatste is vooral belangrijk omdat de incidentie van klinische herseninfarct laag is en zelfs afneemt met verbeteringen in de TAVR-techniek zoals minder traumatische katheters.

Future challenges and opportunities Sinds de introductie van TAVR, is er een snelle groei geweest van het aantal procedures wereldwijd. In de afgelopen jaren is er veel kennis en inzicht opgedaan met betrekking tot optimale beeldvormende technieken, screening en selectie van patiënten voor TAVR, de klinische uitkomsten op korte en lange termijn en het optreden en de preventie van TAVR-

210

APPENDIX

gerelateerde complicaties. Huidige studies zijn er vooral op gericht om de post-TAVR complicaties te verminderen, zodat er een kleinere drempel is om TAVR uit te breiden naar een grotere groep patiënten.

Met de toenemende ervaring en de technische verbeteringen wordt langzamerhand de gelegenheid gecreëerd voor het uitbreiden van TAVR indicatie. Hoewel TAVR oorspronkelijk is ontwikkeld voor hoog risico of inoperabele patiënten, ziet men tegenwoordig dat oudere patiënten (> 80 jaar) met geen of een co-morbiditeit steeds vaker behandeld worden met TAVR. De volgende logische stap zou zijn het uitbreiden van TAVR indicatie naar iets jongere patiënten met een matig risicoprofiel (STS 4-8%), een groep die een kwart tot een derde van chirurgische patiënten beslaat.

Twee studies met een propensity score gematchte vergelijking tussen TAVR en SAVR in patiënten met een matig operatie risico laten vergelijkbare vroege en late mortaliteit zien in beide behandelgroepen20, 21. Twee lopende grote gerandomiseerde klinische studies, PARTNER 2A (SAPIEN XT valve) en SURTAVI (CoreValve), waarbij TAVR wordt vergeleken met SAVR in patiënten met een matig risicoprofiel, zullen waardevolle informatie verschaffen over de toepasbaarheid van TAVR onder patiënten met een matig operatierisico. Met het oog op een bredere toepassing van TAVR in jongere lager risico patiënten, zijn het verlagen van het risico op beroerte, vaatschade en post procedurele aortaklepregurgitatie en daarnaast verbeterde duurzaamheid van de bioprothese essentiële stappen.

Naast een uitbreiding van TAVR indicatie naar patiënten met een lager risicoprofiel, zijn ook nieuwe toepassingen voor dit type procedure denkbaar in de nabije toekomst. Implantatie van een percutane klepprothese in een degeneratieve chirurgische bioprothese is een nieuwe indicatiegebied voor TAVR en de eerste klinische resultaten zijn uitstekend gebleken na correctie voor de comorbiditeiten van de chirurgische groep22-26.

Met de verbeteringen in de TAVR technologie, minimalisering van de implantatiesystemen en de komst van nieuwe en verbeterde klepprotheses wordt het aantal complicaties minder.

Het belang van EPD voor cerebrale protectie tijdens TAVR wordt momenteel in gerandomiseerde klinische trials met een groot aantal patiënten onderzocht. Daarnaast, moet meer aandacht worden besteed aan boezemfibrilleren de novo (NOAF), een fenomeen dat voorkomt in 14% tot 32% van patiënten die een TAVR ondergaan. NOAF lijkt een belangrijke voorspeller te zijn van subacute cerebrale infarcten na TAVR, en de grootste studie op dat gebied laat een odds ratio zien van 2,76 (95% betrouwbaarheidsinterval, 1.11-6.83)27-30. Opsporing van zelfs korte perioden van NOAF en de instelling van een adequate therapie met orale anticoagulantia zijn dus even belangrijk als intra-procedurele gebruik van EPD. Of men in geval van NOAF moet kiezen voor triple therapie (aspirine, clopidogrel en orale antistolling), of voor duale therapie (aspirine en orale antistolling), of orale antistolling alleen moet nog worden geëvalueerd.

Andere mogelijke mechanismen van hersenbescherming tijdens TAVR moeten in de toekomst worden onderzocht. Hemodynamische instabiliteit tijdens TAVR kan systemische hypotensie veroorzaken en daardoor ook cerebrale hypoperfusie. Hoewel op zichzelf een oorzaak voor ischemie, cerebrale hypoperfusie vergroot de kans dat micro-embolieën cerebrale schade veroorzaken door een slechtere uitwas van kleine embolieën in de cerebrale circulatie31.

Op basis van het voorgaande, kan men verwachten dat verhoging van de cerebrale doorbloeding

211

APPENDIX

een betere uitwas van embolieën kan bewerkstelligen. Bovendien zou daardoor ontbinding van embolieën beter mogelijk zijn door een toename van aanvoer van fibrinolytische factoren op de plaats van bloedvatafsluiting door een micro-embolie. Deze hypothese moet worden onderzocht in toekomstige studies.

Aortaklepregurgitatie is één van de meest voorkomende complicaties na TAVR en voorgaande studies hebben laten zien dat er een associatie is tussen deze complicatie en late sterfte.

Evaluatie van post-TAVR AR wordt echter beperkt door een gebrek aan gevalideerde technieken voor de kwantificatie van het regurgitatievolume. Post-TAVR AR bestaat vaak uit multiple jets die paravalvulair zijn. Daardoor kunnen kwalitatieve tweedimensionale evaluatiemethoden de ernst van paravalvulaire AR onderschatten. Kwantitatieve beoordeling van totale AR met behulp van geavanceerde beeldvormende technieken zou dit probleem kunnen oplossen. Het gebruik van 3D transoesofageale echocardiografie of Q-flow cardiale MRI worden momenteel onderzocht voor dit doel.

Een nieuwe uitdaging op het gebied van TAVR zou zijn het combineren van aortaklep- en mitralisklepinterventie. Hoewel er al een paar combinatieprocedures wereldwijd zijn uitgevoerd in de afgelopen jaren, meer ervaring met de techniek en verdere ontwikkelingen op het gebied van mitralisklepinterventies zullen de combinatieprocedures wellicht beter mogelijk maken.

Concluderend, met de toename van ervaring met de procedure, de technische verbeteringen, daarnaast meer kennis en inzicht in mogelijkheden voor preventie van complicaties en met het uitzicht op verbreding van indicatie, lijkt TAVR in de voetstappen te treden van percutane coronaire interventies.

212

APPENDIX

REFERENTIES 1. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010 Oct 21;363(17):1597-60.

2. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011 Jun 9;364(23):2187-98.

3. Kodali SK, Williams MR, Smith CR, Svensson LG, Webb JG, Makkar RR, et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med. 2012 May 3;366(18):1686-95.

4. Adams DH, Popma JJ, Reardon MJ, Yakubov SJ, Coselli JS, Deeb GM, et al. Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med. 2014 May 8;370(19):1790-8.

5. Wong DR, Ye J, Cheung A, Webb JG, Carere RG, Lichtenstein SV. Technical considerations to avoid pitfalls during transapical aortic valve implantation. J Thorac Cardiovasc Surg. 2010 Jul;140(1):196-202.

6. Al Ali AM, Altwegg L, Horlick EM, Feindel C, Thompson CR, Cheung A, et al. Prevention and management of transcatheter balloon-expandable aortic valve malposition. Catheter Cardiovasc Interv. 2008 Oct 1;72(4):573-8.

7. Masson JB, Kovac J, Schuler G, Ye J, Cheung A, Kapadia S, et al. Transcatheter aortic valve implantation: Review of the nature, management, and avoidance of procedural complications. JACC Cardiovasc Interv. 2009 Sep;2(9):811-20.

8. Binder RK, Leipsic J, Wood D, Moore T, Toggweiler S, Willson A, et al. Prediction of optimal deployment projection for transcatheter aortic valve replacement: Angiographic 3-dimensional reconstruction of the aortic root versus multidetector computed tomography. Circ Cardiovasc Interv. 2012 Apr;5(2):24752.

9. Poon KK, Crowhurst J, James C, Campbell D, Roper D, Chan J, et al. Impact of optimising fluoroscopic implant angles on paravalvular regurgitation in transcatheter aortic valve replacements - utility of three-dimensional rotational angiography. EuroIntervention. 2012 Sep;8(5):538-45.

10. Incani A, Lee JC, Poon KK, Crowhurst JA, Raffel OC, Walters DL. Normal functioning of a constrained CoreValve with DynaCT imaging demonstrating incomplete stent frame expansion. Int J Cardiol. 2013 Feb 10;163(1):e9-10.

11. Moat NE, Ludman P, de Belder MA, Bridgewater B, Cunningham AD, Young CP, et al. Long-term outcomes after transcatheter aortic valve implantation in high-risk patients with severe aortic stenosis: The U.K. TAVI (united kingdom transcatheter aortic valve implantation) registry. J Am Coll Cardiol. 2011 Nov 8;58(20):2130-8.

12. Sinning JM, Hammerstingl C, Vasa-Nicotera M, Adenauer V, Lema Cachiguango SJ, Scheer AC, et al.

Aortic regurgitation index defines severity of peri-prosthetic regurgitation and predicts outcome in patients after transcatheter aortic valve implantation. J Am Coll Cardiol. 2012 Mar 27;59(13):1134-41.

13. Tamburino C, Capodanno D, Ramondo A, Petronio AS, Ettori F, Santoro G, et al. Incidence and predictors of early and late mortality after transcatheter aortic valve implantation in 663 patients with severe aortic stenosis. Circulation. 2011 Jan 25;123(3):299-308.

14. Genereux P, Head SJ, Hahn R, Daneault B, Kodali S, Williams MR, et al. Paravalvular leak after transcatheter aortic valve replacement: The new achilles' heel? A comprehensive review of the literature. J Am Coll Cardiol. 2013 Mar 19;61(11):1125-36.

15. Eggebrecht H, Schmermund A, Voigtlander T, Kahlert P, Erbel R, Mehta RH. Risk of stroke after transcatheter aortic valve implantation (TAVI): A meta-analysis of 10,037 published patients.

EuroIntervention. 2012 May 15;8(1):129-38.

16. Kahlert P, Knipp SC, Schlamann M, Thielmann M, Al-Rashid F, Weber M, et al. Silent and apparent cerebral ischemia after percutaneous transfemoral aortic valve implantation: A diffusion-weighted magnetic resonance imaging study. Circulation. 2010 Feb 23;121(7):870-8.

213

APPENDIX

17. Ghanem A, Muller A, Nahle CP, Kocurek J, Werner N, Hammerstingl C, et al. Risk and fate of cerebral embolism after transfemoral aortic valve implantation: A prospective pilot study with diffusionweighted magnetic resonance imaging. J Am Coll Cardiol. 2010 Apr 6;55(14):1427-32.

18. Ghanem A, Kocurek J, Sinning JM, Wagner M, Becker BV, Vogel M, et al. Cognitive trajectory after transcatheter aortic valve implantation. Circ Cardiovasc Interv. 2013 Dec 1;6(6):615-24.

19. Vermeer SE, Prins ND, den Heijer T, Hofman A, Koudstaal PJ, Breteler MM. Silent brain infarcts and the risk of dementia and cognitive decline. N Engl J Med. 2003 Mar 27;348(13):1215-22.

20. Latib A, Maisano F, Bertoldi L, Giacomini A, Shannon J, Cioni M, et al. Transcatheter vs surgical aortic valve replacement in intermediate-surgical-risk patients with aortic stenosis: A propensity scorematched case-control study. Am Heart J. 2012 Dec;164(6):910-7.

21. Piazza N, Kalesan B, van Mieghem N, Head S, Wenaweser P, Carrel TP, et al. A 3-center comparison of 1-year mortality outcomes between transcatheter aortic valve implantation and surgical aortic valve replacement on the basis of propensity score matching among intermediate-risk surgical patients.

JACC Cardiovasc Interv. 2013 May;6(5):443-51.

22. Webb JG, Wood DA, Ye J, Gurvitch R, Masson JB, Rodes-Cabau J, et al. Transcatheter valve-in-valve implantation for failed bioprosthetic heart valves. Circulation. 2010 Apr 27;121(16):1848-57.

23. Gurvitch R, Cheung A, Ye J, Wood DA, Willson AB, Toggweiler S, et al. Transcatheter valve-in-valve implantation for failed surgical bioprosthetic valves. J Am Coll Cardiol. 2011 Nov 15;58(21):2196-209.

24. Piazza N, Bleiziffer S, Brockmann G, Hendrick R, Deutsch MA, Opitz A, et al. Transcatheter aortic valve implantation for failing surgical aortic bioprosthetic valve: From concept to clinical application and evaluation (part 1). JACC Cardiovasc Interv. 2011 Jul;4(7):721-32.

25. Piazza N, Bleiziffer S, Brockmann G, Hendrick R, Deutsch MA, Opitz A, et al. Transcatheter aortic valve implantation for failing surgical aortic bioprosthetic valve: From concept to clinical application and evaluation (part 2). JACC Cardiovasc Interv. 2011 Jul;4(7):733-42.

26. Dvir D, Webb J, Brecker S, Bleiziffer S, Hildick-Smith D, Colombo A, et al. Transcatheter aortic valve replacement for degenerative bioprosthetic surgical valves: Results from the global valve-in-valve registry. Circulation. 2012 Nov 6;126(19):2335-44.

27. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011 Jun 9;364(23):2187-98.

28. Nuis RJ, Van Mieghem NM, Schultz CJ, Moelker A, van der Boon RM, van Geuns RJ, et al. Frequency and causes of stroke during or after transcatheter aortic valve implantation. Am J Cardiol. 2012 Jun 1;109(11):1637-43.

29. Nombela-Franco L, Webb JG, de Jaegere PP, Toggweiler S, Nuis RJ, Dager AE, et al. Timing, predictive factors, and prognostic value of cerebrovascular events in a large cohort of patients undergoing transcatheter aortic valve implantation. Circulation. 2012 Dec 18;126(25):3041-53.

30. Amat-Santos IJ, Rodes-Cabau J, Urena M, DeLarochelliere R, Doyle D, Bagur R, et al. Incidence, predictive factors, and prognostic value of new-onset atrial fibrillation following transcatheter aortic valve implantation. J Am Coll Cardiol. 2012 Jan 10;59(2):178-8.

31. Caplan LR, Hennerici M. Impaired clearance of emboli (washout) is an important link between hypoperfusion, embolism, and ischemic stroke. Arch Neurol. 1998 Nov;55(11):1475-82.

214

APPENDIX

ACKNOWLEDGMENTS Met de laatste woorden in dit proefschrift wil ik mijn dank betuigen aan iedereen die, direct of indirect, heeft bijgedragen aan de totstandkoming van dit proefschrift.

De promotiecommissie Prof. P.A.M.F. Doevendans, geachte promotor, beste professor, dit proefschrift is voort gekomen uit uw enthousiaste reactie op mijn vraag of ik in het vierde jaar van mijn studie geneeskunde in uw divisie onderzoek mocht doen. U liet mij toen een artikel zien over cerebro-embolische complicaties na percutane coronaire interventies en u verwees me daarmee door naar Prof. dr. Van Belle. In de jaren daarna heb ik onder uw motiverende supervisie met veel plezier klinisch onderzoek gedaan op de afdeling interventiecardiologie. Met uw enthousiasme heeft u mijn interesse in de cardiologie nog meer aangewakkerd. In de laatste fase van mijn studie geneeskunde heeft u mij de kans gegeven om de meer fundamentele kant van cardiologie te zien. Zo heb ik een tijdje laboratoriumonderzoek gedaan in uw groep in het UMC Utrecht en later in San Diego, door u beschreven als een door God zelf geschapen “Paradijs op aarde”.

Daarover ben ik het zeker met u eens. Het is een eer om deel uit te maken van de grote groep promovendi die onder uw hoede hun promotietraject heeft volbracht.

Geachte dr. P.R. Stella, beste Pieter, dank voor de goede begeleiding tijdens mijn promotietraject.

Je hebt me vanaf het begin van mijn promotietraject bijgestaan met je kennis op het gebied van percutane aortaklepimplantatie en je hebt me geleerd efficiënt en met focus te werken.

Ik heb het vertrouwen dat je in mij hebt gesteld altijd zeer gewaardeerd; het heeft me de ruimte gegeven om mijn eigen ideeën uit te werken. Een extra dankzegging bovendien voor je ondersteuning tijdens de eindsprint. Op naar het volgende proefschrift!

Geachte dr. P. Agostoni, beste Ago, als geneeskundestudent had ik veel bewondering voor je enthousiasme voor het vak interventiecardiologie en het wetenschappelijk onderzoek. Toen ik net was begonnen, was je nog niet getrouwd en ik weet nog goed dat je na je klinische werkzaamheden tot laat in de avond in het UMC aan wetenschappelijke artikelen werkte.

Overleg buiten kantoortijden over projecten was dan ook geen probleem. Dat is tegenwoordig (gelukkig voor Catherina) anders, om 19:00 uur ben je vaak niet meer bereiken op je sein.

Je was regelmatig een strenge begeleider. Eten en slapen kon ik volgens jou beter afleren, hetgeen mij tegen het eind van mijn promotietraject bijna gelukt is. Ik heb met veel plezier van je geleerd en met jou samengewerkt en gediscussieerd over het onderzoek. Daarnaast wil ik je hartelijk danken voor je vriendschap en de goede tijd die ik heb gehad als promovendus onder jouw begeleiding.

De leden van de beoordelingscommissie, Prof. dr. T. Leiner, Prof. dr. H. Suryapranata, Prof. dr.

L.A. van Herwerden, Prof. dr. R.J. de Winter en Prof. dr. N.H.J. Pijls, dank voor uw interesse in dit proefschrift.

215

APPENDIX

De onderzoeksgroep en externe begeleiders Prof. dr. R. Goldschmeding, beste Roel, bedankt voor de begeleiding toen ik als groentje een start maakte aan mijn wetenschappelijke carrière tijdens mijn studie geneeskunde. Ook na het afstuderen en tijdens mijn promotietraject kon ik altijd bij je terecht voor je wijze adviezen en daar ben ik je enorm dankbaar voor.

Prof. dr. T. Leiner, beste Tim, hartelijk dank voor je enthousiasme wanneer ik weer eens met een vraag of onderzoekidee naar jou kwam, je goede adviezen en daarnaast je bereidheid het manuscript te beoordelen en in de commissie zitting te nemen.

Dear Prof. E. van Belle, thank you for your guidance at the beginning of my PhD period. It was great working with you and I wish you all the best in France.

Beste Freek, we hebben ongeveer twee jaar gewerkt aan soortgelijke projecten in een gezamenlijke luxe grote kamer in het Q-gebouw. Naast interessante discussies over de wetenschappelijke projecten, hebben we ook samen congressen bezocht, met een aantal memorabele hoogtepunten. Zo hebben we dit jaar even vastgezeten in de metro in Parijs, om vervolgens in congresoutfit een onsuccesvolle sprint te trekken over de Champs Élysée.

Uiteindelijk waren we dankzij jou toch net op tijd bij het congresgebouw: ik was nog nooit zo blij om ergens te staan presenteren. Verder, je kritische blik en je behulpzaamheid hebben mij veel geholpen tijdens mijn promotietraject. Ik ben je dankbaar als collega, als vriend, en daarnaast als paranimf.

Lieve Anneliene, erg bedankt voor je luisterend oor, je steun en je gezelligheid in de afgelopen jaren. Je bent op dit moment supergoed bezig met je promotietraject en je gaat het helemaal maken. We moeten weer eens een Disney-uitje afspreken, Orlando dit keer?

Lieve Judith, in de afgelopen maanden hebben we samen ons promotieonderzoek afgerond en daarna zijn we de uitdaging in de kliniek aangegaan. Je mooie aanstekelijke lach gaf steeds veel geruststelling en vertrouwen dat het allemaal goed zou komen. Dank voor alle steun.

Lieve Rosemarijn, mijn partner in crime, ik heb geluk met jou als vriendin en als collega, op jou kan ik altijd rekenen, behalve als het gaat om kaartlezen. Bedankt voor je wijze adviezen, je gezelligheid en je vriendschap. Je bent hard op weg om “3D echo-koningin” te worden en dat gaat je zeker lukken!

Lieve Jetske, dankjewel voor je vriendschap, je steun, de spinning- en de salsa-avondjes, en alle gezelligheid. Naast een fijne collega, ben je een lieve vriendin en je hebt altijd klaargestaan voor mij. We hebben samen veel gelachen, maar ook de frustraties van onderzoek doen met elkaar gedeeld. Heel veel succes met het afronden van je promotie in de komende maanden, daarna gaan we samen weer feestvieren.

216

APPENDIX

Natuurlijk wil ik ook mijn Q-genoten, mede-promovendi van de afdeling cardiologie, Marloes, Thomas, Gijs, Ing Han, René, Peter-Paul, Remco, Manon, Sofieke, Cheyenne, Sanne, Frebus, Cas, Martine, Hamza, Mieke, Willemien, Wouter en Zeleyha bedanken voor de goede sfeer en alle gezelligheid en ontspanning naast werk en promotie. Ik wens jullie uiteraard allemaal veel succes in de komende tijd met het vervolg van jullie promoties.

Ook de dames van “research and development” van de cardiologie, Manon Kuikhoven, Yvonne Breuer-Otten, Jeannette Visser-Bouman, Ellie van Schaik-Brood en Astrid Links, wil ik graag bedanken voor al hun hulp bij de inclusie van patiënten en het verzamelen van studiedata, zonder jullie hulp had ik het nooit op tijd gered.

Alle medewerkers van de hartcatheterisatiekamer wil ik bedanken voor hun medewerking en steun bij mijn projecten. In het bijzonder dank aan Herman Bouwman, Marion van den Bergen en Mathilde Hissink.

Alle arts-assistenten en cardiologen uit het UMC Utrecht wil ik bedanken voor hun collegialiteit, samenwerking en gezelligheid tijdens mijn ANIOS-stage in het UMC Utrecht.

Tamara Rietveld, Jantine Nieuwkoop en Sylvia van der Straten, ik wil jullie graag bedanken voor jullie organisatorische ondersteuning. Met jullie hulp heb ik al het papierwerk aan het eind van mijn promotietraject op tijd kunnen afronden.

Ook hartelijke dank aan alle co-auteurs voor hun bijdrage aan de verschillende stukken in dit proefschrift.

Ik ben de commissie van het Alexander Suerman stipendium bijzonder dankbaar voor de steun en inspirerende masterclasses. Ook wil ik graag de NWO bedanken voor de financiering van een belangrijk deel van mijn wetenschappelijke carrière.

Vrienden en familie Lieve Nuray, samen hebben we als nieuwkomers in NL vanaf de internationale schakelklas ons een weg naar de universiteit gebaand. In al die jaren was jij mijn maatje in gelukkige en moeilijkere tijden. Bedankt voor je liefde, je luisterend oor en je wijze woorden in al die jaren.

In oktober ben ik tante geworden van jouw mooie kind, love you both.

Veel andere vrienden hebben me tijdens mijn promotietraject bijgestaan met hun gezelligheid en dat waardeer ik enorm. In het bijzonder veel dank daarvoor aan Mona Charaghvandi, Allert de Vries, Yi Kie Lam, Nynke Koning, Hogaei Oriakheil, Danny Young-Afat, Dino Colo, Pauline van Kempen en Maryam Nadem.

Liefste Mors, als zusje, beste vriendin, huisgenootje en mijn persoonlijke adviseur op veel gebieden ben ik je dankbaar voor heel veel dingen. Je bent er altijd voor mij en ik kan altijd op

217

APPENDIX

je rekenen, daar heb ik me altijd gelukkig voor geprezen. Bedankt voor alle liefde, je luisterend oor, de stoom-afblaasmomenten en heel belangrijk: je wijze adviezen. Tenslotte, dank dat je me ook nu als paranimf wilt bijstaan.

Liefste jongste zusje Mehri, na veel twijfel heb je uiteindelijk toch voor geneeskunde gekozen als vak, en ook jij in Utrecht. De drie Samim zussen (M.Samim) zorgen voor veel verwarring in het UMC Utrecht. Dank dat je er altijd voor me bent, dat ik mijn blije tijden en ook mijn frustraties met jou kan bespreken, je wijze adviezen ondanks het grote leeftijdsverschil tussen ons twee, alle gezellige zussenavonden en je hulp bij de literatuurstudie die nodig was voor hoofdstuk 10 van dit proefschrift.

Lieve Fais, dank voor al je steun, je zorgzaamheid en onze lange gesprekken. Ik heb een superfamilie, want ook jou kon ik dag en nacht vragen om van alles, jij stond altijd voor me klaar, niks ging te ver, zelfs toen je op duizenden kilometers afstand woonde. Dank voor alles broertje!

Liefste pap en mam, dank voor al jullie steun, de liefde en toewijding waarmee jullie mij in de loop van mijn carrière tot nu toe hebben begeleid, de mogelijkheden die jullie mij hebben geboden en het feit dat jullie in al mijn keuzes altijd 200% achter mij hebben gestaan. Mijn doorzettingsvermogen heb ik van jullie, jullie zijn een grote inspiratiebron voor al je kinderen en zonder jullie hulp had ik dit allemaal niet bereikt. Bedankt voor alles en ik draag mijn proefschrift op aan jullie.

218

APPENDIX

CURRICULUM VITAE AUCTORIS Mariam Samim was born on the 22th of November 1985 in Herat, Afghanistan. She moved together with her parents and siblings to the Netherlands in 1999. In 2005 she finished secondary school at the “Etty Hillesum Lyceum” in Deventer. She studied medicine from 2006 until 2012 at the University of Utrecht. During her studies she spent some time on clinical research at the department of interventional cardiology of University Medical Center Utrecht.

She mainly worked on projects concerning transcatheter aortic valve replacement (TAVR).

In the last year of medicine, she spent 6 months in San Diego, California working on a research project on the field of developmental cardiology. At her return in the Netherlands, she applied for PhD grants for reaserch on the field of TAVR and in 2012 her research proposal was awarded with a NWO grant (Mozaïek) and the Alexandre Suerman MD/PhD Stipedium.

Following that, Mariam started working as a PhD student at the department of interventional cardiology under supervision of Prof. dr. P.A. Doevendans, dr. P.R. Stella en dr. P. Agostoni.

The results of her work are included in this thesis. Mariam is currently working as a resident at the Cardiology department, University Medical Center in Utrecht.

219

APPENDIX

LIST OF PUBLICATIONS Embrella embolic deflection device for cerebral protection during transcatheter aortic valve replacement. Samim M, Agostoni P, Hendrikse P, Budde RP, Nijhoff F, Kluin J, Ramjankhan F, Doevendans PA, Stella PR. J Thorac Cardiovasc Surg. 2014. In press.

Silent ischemic brain lesions after transcatheter aortic valve replacement: lesion distribution and predictors. Samim M, Hendrikse J, Van der Worp HB, Agostoni P, Nijhoff F, Doevendans PA, Stella PR. Cinical research in cardiology. 2014. In press.

Three-dimensional aortic root reconstruction derived from rotational angiography for transcatheter balloon-expandable aortic valve implantation guidance. Samim M, Agostoni P, Nijhoff F, Budde RP, Abrahams AC, Kluin J, Ramjankhan F, Doevendans PA, Stella PR. Int J Cardiol. 2014 Oct 20;176(3):1318-20.

Transcatheter aortic valve implantation in patients with severe aortic valve stenosis and large aortic annulus, using the self-expanding 31-mm Medtronic CoreValve prosthesis: first clinical experience. Nijhoff F, Agostoni P, Amrane H, Latib A, Testa L, Oreglia JA, De Marco F, Samim M, Bedogni F, Maisano F, Bruschi G, Colombo A, Van Boven AJ, Stella PR. J Thorac Cardiovasc Surg.

2014 Aug;148(2):492-9.e1.

Same wrist intervention via the cubital (ulnar) artery in case of radial puncture failure for percutaneous cardiac catheterization or intervention: the multicenter SWITCH registry.

Agostoni P, Zuffi A, Faurie B, Tosi P, Samim M, Belkacemi A, Voskuil M, Stella PR, Romagnoli E, Biondi-Zoccai G. Int J Cardiol. 2013 Oct 25;169(1):52-6.

Optimisation of transcatheter aortic balloon-expandable valve deployment: the two-step inflation technique. Nijhoff F, Agostoni P, Samim M, Ramjankhan FZ, Kluin J, Doevendans PA, Stella PR. EuroIntervention. 2013 Sep;9(5):555-63.

Reply: MSCT in TAVR for better implant angle and outcomes. Van Belle E, Samim M, El Kalioubie A, van Belle C, Stella PR, Agostoni P, Kluin J, Ramjankhan F, Budde RP, Sieswerda G, Algeri E, Juthier F, Belkacemi A, Bertrand ME, Doevendans PA. JACC Cardiovasc Imaging. 2013 Aug;6(8):923.

Automated 3D analysis of pre-procedural MSCT to predict annulus plane angulation and C-arm positioning: benefit on procedural outcome in patients referred for TAVR. Samim M, Stella PR, Agostoni P, Kluin J, Ramjankhan F, Budde RP, Sieswerda G, Algeri E, van Belle C, Elkalioubie A, Juthier F, Belkacemi A, Bertrand ME, Doevendans PA, Van Belle E. JACC Cardiovasc Imaging. 2013 Feb;6(2):238-48.

220

APPENDIX

Infective complications after transcatheter aortic valve implantation: results from a single centre. Onsea K, Agostoni P, Voskuil M, Samim M, Stella PR. Neth Heart J. 2012 Sep;20(9):360-4.

First-in-man experience with a new embolic deflection device in transcatheter aortic valve interventions. Onsea K, Agostoni P, Samim M, Voskuil M, Kluin J, Budde R, Hendrikse J, Ramjankhan F, van Klarenbosch J, Doesburg P, Sieswerda G, Stella P. EuroIntervention. 2012 May 15;8(1):51-6.

Optical coherence tomography assessment of early stent strut coverage in patients treated with a thin-strut bare cobalt-chromium stent coated with silicon carbide. Samim M, Agostoni P, Voskuil M, Belkacemi A, Doevendans PA, Stella PR. Int J Cardiol. 2012 May 31;157(2):291-2.

Automated 3D analysis of multislice computed tomography to define the line of perpendicularity of the aortic annulus and of the implanted valve: benefit on planning transcatheter aortic valve replacement. Samim M, Juthier F, Van Belle C, Agostoni P, Kluin J, Stella PR, Ramjankhan F, Budde RP, Sieswerda G, Algeri E, Elkalioubie A, Belkacemi A, Bertrand ME, Doevendans PA, Van Belle E. Catheter Cardiovasc Interv. 2014 Jan 1;83(1):E119-27.

A prospective "oversizing" strategy of the Edwards SAPIEN bioprosthesis: results and impact on aortic regurgitation. Samim M, Stella PR, Agostoni P, Kluin J, Ramjankhan F, Sieswerda G, Budde R, van der Linden M, Juthier F, Banfi C, Hurt C, Samim M, Hillaert M, van Herwerden L, Bertrand ME, Doevendans PA, Van Belle E. J Thorac Cardiovasc Surg. 2013 Feb;145(2):398-405.

Transcatheter aortic implantation of the Edwards-SAPIEN bioprosthesis: insights on early benefit of TAVI on mitral regurgitation. Samim M, Stella PR, Agostoni P, Kluin J, Ramjankhan F, Sieswerda G, Budde R, der Linden Mv, Samim M, Hillaert M, van Herwerden L, Doevendans PA, van Belle E. Int J Cardiol. 2011 Oct 6;152(1):124-6.

221

222