PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S 5 6 (2 0 1 4) 63 5–6 4 5
Available online at www.sciencedirect.com
ScienceDirect www.onlinepcd.com
Challenges and Future Opportunities for Transcatheter Aortic Valve Therapy Martin B. Leona,⁎, Hemal Gadaa , Gregory P. Fontanab a
Center for Interventional Vascular Therapy, Columbia University Medical Center, New York Presbyterian Hospital, New York, NY Lenox Hill Heart & Vascular Institute of New York, Lennox Hill Hospital, North Shore – LIJ Health Care System, Manhasset, NY
b
A R T I C LE I N F O
AB ST R A C T
Keywords:
Background: Transcatheter aortic valve replacement (TAVR) is a novel less-invasive therapy
Transcatheter aortic valve replacement
for high-risk patients with severe aortic stenosis (AS). Despite the impressive clinical
Aortic stenosis
growth of TAVR, there are many challenges as well as future opportunities.
Aortic valve replacement
Results: The heart valve team serves as the central vehicle for determining appropriate case selection. Considerations which impact clinical therapy decisions include frailty assessments and defining clinical “futility”. There are many controversial procedural issues; choice of vascular access site, valve sizing, adjunctive imaging, and post-dilatation strategies. Complications associated with TAVR (strokes, vascular and bleeding events, para-valvular regurgitation, and conduction abnormalities) must be improved and will require procedural and/or technology enhancements. TAVR site training mandates a rigorous commitment to established society and sponsor guidelines. In the future, TAVR clinical indications should extend to bioprosthetic valve failure, intermediate risk patients, and other clinical scenarios, based upon well conducted clinical trials. New TAVR systems have been developed which should further optimize clinical outcomes, by reducing device profile, providing retrievable features, and preventing para-valvular regurgitation. Other accessory devices, such as cerebral protection to prevent strokes, are also being developed and evaluated in clinical studies. Summary: TAVR is a worthwhile addition to the armamentarium of therapies for patients with AS. Current limitations are important to recognize and future opportunities to improve clinical outcomes are being explored. © 2014 Elsevier Inc. All rights reserved.
Background In the past decade, after initial proof-of-concept and subsequent feasibility studies, the application of less-invasive catheter-based approaches to functionally replace diseased aortic valves has been incorporated into the clinical treatment armamentarium in symptomatic high-risk patients with
severe aortic stenosis (AS). Since 2007, in more than 50 countries, over 750 cardiovascular centers have treated almost 100,000 aortic stenosis patients using transcatheter aortic valve replacement (TAVR) technologies. Despite the rapid acceptance and clinical appeal of TAVR, as with any new and novel medical therapy, there are still many challenges to be addressed and future opportunities to be
Statement of Conflict of Interest: see page 643. ⁎ Address reprint requests to Martin B. Leon, MD, Columbia University Medical Center, 161 Ft. Washington Avenue, Herbert Irving Pavilion, 6th Floor, New York, NY 10032. E-mail address: [email protected] (M.B. Leon). 0033-0620/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.pcad.2014.03.004
636
PR O GRE S S I N C ARDI O VAS CU L AR D I S EAS E S 5 6 ( 2 0 14 ) 63 5–6 45
Abbreviations and Acronyms AS = aortic stenosis CT = computerized tomography ICE = intra-cardiac echocardiography LBBB = left bundle branch block LV = left ventricular PARTNER = Placement of Aortic Transcatheter Valves
explored. The purpose of this manuscript is to selectively highlight the crucial challenges of TAVR which are presently under investigation and to direct attention towards expanding clinical applications and new technologies which constitute important future opportunities.
PVR = para-valvular regurgitation RV = right ventricular
Challenges
STS = Society for Thoracic Surgeons
Case selection
TA = transapical TAo = transaortic TAVR = transcatheter aortic valve replacement
Identifying “high-risk” patients
Patient selection under the auspices of a multidisciplinary “Heart Team” TEE = transesophageal echocardiography is crucial to achieve optimal clinical outcomes TF = transfemoral after TAVR. The differenTVT = transcatheter valve tiation between high-risk, therapy “inoperable” (or extreme risk), and prohibitive risk US = United States AS patients has been acVARC = Valve Academic tively debated since reguResearch Consortium latory approval of TAVR and especially during the formulation of the Placement of Aortic Transcatheter Valves (PARTNER) clinical trial.1 Risk assessment has often been guided by standard surgical scoring systems, including the Society of Thoracic Surgery (STS) and EuroSCORE models, which were not fully validated in this high-risk patient population. These on-line risk scores, as designed for everyday use, do not include important co-morbidities such as severe pulmonary hypertension, right ventricular (RV) dysfunction, severe liver disease, home supplemental oxygen, prohibitive anatomy (such as chest deformity or severe aortic calcification), disability, or frailty. Characterization of surgical risk requires direct involvement of experienced surgeons who usually include a number of important co-morbidities when considering the highest risk patients for TAVR: malnutrition and cachexia, physical deconditioning or wheelchair bound, chronic kidney disease on dialysis, history of particular solid tumor malignancies, neurological disorders such as dementia and stroke, and other debilitating conditions that preclude patients from returning to a reasonable functional status. One of the biggest challenges in assessment of patient risk status is developing a validated quantitative algorithm that best defines patient risk from the standpoint of predicting early and late mortality as well as functional recovery in the setting of TAVR. The combined analyses of the PARTNER trials or the new
United States (US) Transcatheter Valve Therapies (TVT) National Registry will hopefully provide sufficient patient data to offer the possibility of a TAVR specific risk algorithm at some point in the future.2
Frailty and futility Not entirely captured in current risk stratification metrics is the attribute of frailty, which has been associated with worse TAVR outcomes. The concept of frailty is crudely defined as an impairment in multiple systems that leads to a decline in resiliency and homeostatic reserve. It is influenced by physical disability and medical co-morbidities, but is not adequately described by just these attributes.3 Green et al have devised a frailty score for TAVR patients, based loosely on criteria established by Fried et al.4 The frailty phenotype, including impairments in gait speed and grip strength, reduced serum albumin, and diminished Katz activities of daily living, was associated with a longer post-TAVR hospital stay, as well as increased 1-year mortality.5 The multicenter FRAILTY-AVR study will compare outcomes of surgical aortic valve replacement (SAVR) and TAVR using several frailty assessment tools in the effort to define which factors are the most predictive of mortality and morbidity in elderly patients. The results of the US CoreValve Pivotal Trial Extreme Risk cohort highlight the need to define and quantify the significance of this interaction, as the only two significant predictors of all-cause mortality or major stroke (the primary endpoint), were STS score of > 15% (p = 0.02) and residence in an assisted living facility (p < 0.01).6 A careful frailty assessment plays a key role in the differentiation of “futility” (“no hope” patients) and high-risk utility patients and should be incorporated into all TAVR risk stratification analyses. The term “Cohort C” describes this subset of futile inoperable patients who have both poor survival (i.e. less than 1 year) and poor quality of life, despite successful TAVR. Simply stated, “Cohort C” or futile patients represent those patients who are dying with aortic stenosis but not from AS. Common clinical characteristics most associated with futile risk patients include extreme comorbidities (e.g. STS score > 15%), extreme frailty usually with a dependent social status, severe pulmonary or liver disease, severe dementia, chronic kidney disease (e.g. dialysis dependent), and hemodynamic instability (especially requiring vasopressors). What remains to be defined is the quantitative interplay of frailty metrics and existing risk stratification models based on age and co-morbid conditions, in accurately determining a “Cohort C” patient.
Procedural considerations Access alternatives Factors which may determine preferred TAVR vascular access include peripheral arterial disease (inadequate vessel diameter, severe calcification or extreme tortuosity of the iliofemoral vessels), the presence of extensive calcification of the ascending aorta (i.e. porcelain aorta), hostile chest wall anatomy (either due to ortho-voltage radiation exposure or chest wall deformities), previous coronary bypass graft surgery with mammary conduits adherent to the chest wall,
PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S 5 6 (2 0 1 4) 63 5–6 4 5
and severe lung disease. The four most common techniques for TAVR access are the retrograde transfemoral (TF), antegrade transapical (TA), and the more recently developed direct or trans-aortic (TAo) and subclavian approaches. In the PARTNER trial, using the larger profile SAPIEN transcatheter valve system (outer sheath diameter 9.2 mm for the 26 mm valve) there were frequent major vascular complications associated with TF–TAVR procedures.1 In the PARTNER II trial, the first-generation SAPIEN system was compared to the lower profile SAPIEN XT system (33% lower cross-sectional area) and major vascular complications were reduced from 15.5% to 9.6% (p = 0.04).7 This highlights the importance of lower profile delivery systems in maximally utilizing safe fully percutaneous TF-TAVR as a primary default access strategy. The TA approach avoids peripheral access issues, but also has limitations; increased length of hospitalization, and increased risk of 30-day and 1-year all-cause mortality.8,9 These adverse TA outcomes may have been influenced by differences in underlying baseline co-morbidities between the two populations, given the “TF-first” approach often adopted by clinicians, which relegated only patients with significant peripheral vascular disease to TA-TAVR. Other adverse outcomes associated with the TA approach include a higher likelihood of peri-procedural bleeding, increased risk of hemodynamic instability, and greater patient discomfort, due to pain related to the antero-lateral thoracotomy.10,11 TAo and subclavian access sites for TAVR have been introduced more recently. The largest series of TAo cases12 indicated that compared to a contemporary group of TA patients, there was a lower combined bleeding and vascular event rate (27% vs 46%; p = 0.05), shorter median intensive care unit length of stay (3 vs 6 days; p = 0.01), and a favorable learning curve. Transcarotid access and antegrade transseptal access via the femoral vein have also been described13,14 but there are scant clinical outcome data. The choice of vascular access site for TAVR is an individualized patient-based decision determined by clinical factors, anatomic considerations as well as the experiences and preferences of the Heart Team. Considering the less-invasive nature of fully percutaneous TF access and the increasing availability of lower profile TAVR systems, it is likely that TF–TAVR will be the preferred option for the majority of patients in the foreseeable future.
Valve sizing and positioning Multimodality imaging is essential for patient screening and procedural guidance during TAVR, and has been incorporated into consensus statements, reviews, and guidelines.15,16 Correct valve sizing for either the balloon-expandable or the self-expandable TAVR system requires meticulous attention to three-dimensional imaging, including multi-slice CT and trans-esophageal echocardiography (TEE). Optimal TAVR implantation requires: (1) correct valve sizing based upon established criteria for measuring the annulus dimensions matched to the specific valve type; (2) accurate valve positioning (axial height and alignment) within the annular valve plane. Different TAVR systems mandate specific procedural techniques to determine co-planar implantation views and optimal height and alignment of valve implantation, which can be facilitated by rapid pacing with cine-
637
fluoroscopic and/or TEE guidance. The challenge of valve sizing and positioning cannot be underestimated and requires an intimate understanding of the anatomy of the aortic valvar complex. Ideally, correct sizing and placement of TAVR will result in excellent valve hemodynamics, none or trace para-valvular regurgitation (PVR), low requirements for new pacemakers due to conduction abnormalities, and no evidence of coronary obstruction or annulus injury.
Trans-esophageal echocardiography Intraprocedural TEE can provide a real-time biplane assessment of the annulus before and during deployment, thus fostering more precise valve positioning. TEE can also help to predict and is the gold standard to detect PVR after valve implantation, as well as other complications such as coronary artery obstruction, annulus rupture, pericardial tamponade (e.g. due to chamber perforation), severe mitral regurgitation, aortic dissection (or hematoma), and left ventricular (LV) dysfunction. However, the routine use of TEE during TAVR procedures has become controversial, as TEE usually is associated with general anesthesia and endotracheal intubation, which may introduce additional risks in patients who are hemodynamically unstable or have underlying severe pulmonary disease. Moreover, TEE requires specialized imaging expertise which may not be readily available for all cases and the additional sedation may delay recovery. Therefore, a strong trend in TAVR procedures has been to selectively or systematically favor monitored anesthesia control without intubation, combined with transthoracic echocardiography, as needed. The virtues and drawbacks of routine TEE have been hotly debated and the decision to utilize TEE on a case or sitespecific basis is presently determined by resource availability, perceived clinical need, and personal preferences. The motivation to eliminate the need for general anesthesia has led to an increasing interest in intracardiac echocardiography (ICE) imaging for TAVR procedures. However, single-plane ICE imaging cannot accurately visualize the oval annulus for sizing purposes, measure the coronary artery height, or reliably assess post-implantation PVR.17 New three-dimensional ICE catheters may overcome some of these limitations in the future, thus permitting on-line echo guidance and assessment without the need for general anesthesia.
Pre- and post-dilation Balloon aortic valvuloplasty (BAV) prior to valve deployment has been traditionally performed, especially prior to balloonexpandable transcatheter valve deployment. Pre-dilation BAV allows easier crossing of the valve through the annulus and potentially avoids mechanical complications related to the force and contour of the delivery system. However, BAV carries independent risks of atrioventricular block requiring permanent pacemakers, increased aortic regurgitation, and embolic neurologic events. Recent trends have favored reduced or no pre-dilation with low profile self-expanding valve platforms. Grube et al treated 60 consecutive patients with the Medtronic CoreValve prosthesis without balloon pre-dilation and good hemodynamic performance
638
PR O GRE S S I N C ARDI O VAS CU L AR D I S EAS E S 5 6 ( 2 0 14 ) 63 5–6 45
(post-procedural mean gradients and effective orifice areas).18 In addition, in-hospital major adverse cardiovascular events and need for permanent pacemakers were far less than those reported in the US CoreValve Pivotal Trial Extreme Risk cohort.6 Garcia et al reported the results of no pre-dilation with the SAPIEN XT valve in 10 patients with moderate calcification, homogenous distribution of calcium, and symmetric opening of the valve.19 No important complications were observed and the patients did not require post-dilation for significant PVR which was none or trivial in all patients. Thus, the growing trend of TAVR without pre-dilation appears to be a feasible technique with potential benefits and will be a target for further study in the future. Post-BAV after TAVR has been selectively applied usually in situations of significant PVR, which has been associated with increased late mortality.20 The benefits and risks of post-BAV after TAVR have been hotly debated. TEE data clearly indicate that post-dilation in carefully selected patients importantly reduces PVR. 21 However, post-dilation has been associated with increased complications including embolic neurologic events,22 conduction abnormalities, and the risk of annular rupture. Barbanti et al reported 31 patients receiving balloon-expandable TAVR with annular rupture from a large multicenter TAVR experience 23 and the predictors of annular rupture were subannuluar/LV outflow tract calcification, a higher frequency of ≥ 20% annular area oversizing, and balloon post-dilation. In the future, with improved valve sizing and new technology to prevent PVR, there will be a reduced need for selective post-dilation after TAVR.
TAVR complications – brief updates, current and future management Strokes Strokes in the setting of TAVR remains a major peri- and postprocedural complication manifesting with a significant deterioration in quality-of-life and increased mortality. Several diffusion weighted MRI studies have shown that the rate of silent cerebral embolism after TAVR approaches 80% of patients.24–26 Van Mieghem et al confirmed these neuro-imaging findings, noting 75% of TAVR procedures producing debris captured in a filterbased cerebral embolic protection device.27 However, a recent meta-analysis of >10,000 patients in 53 studies confirmed that TAVR is associated with in a reasonable peri-procedural stroke rate of 1.5% and a 30-day stroke/transient ischemic attack rate of 3.3%.28 Thus, the discordance between detection of neuroembolic activity with TAVR and subsequent clinical neurologic events requires further resolution. Risk stratification for stroke based on patient characteristics is essential in defining who would likely benefit from cerebral embolic protection devices, as well as tailoring post-procedural management (e.g. surveillance for post-procedural atrial fibrillation and adjunctive pharmacotherapy).
Paravalvular regurgitation The 2-year follow-up of PARTNER Cohort A patients demonstrated that TAVR resulted in significantly worse PVR than SAVR with >50% of TAVR patients had at least mild PVR.20 Moreover,
even mild PVR post-TAVR was associated with 10–15% higher mortality at 2 years than patients with none or trace PVR, as determined by a core echocardiography laboratory. Multiple other studies have shown similar associations between varying degrees of significant PVR and increased late mortality after TAVR.29–31 Accurate diagnosis and clinical impact require a combined assessment of hemodynamics, angiography, and especially echocardiography, which remains the gold standard. The treatment of PVR is based on an understanding of the severity and specific etiology. Valve undersizing or underexpansion, valve mal-alignment (either too high or too low), and severe global and focal aortic valvar complex calcification with mal-apposition are the main causes of PVR and treatment options include strategic post-dilatation, placement of additional TAVR to extend the “seal zone”, or implantation of peri-valve vascular plugs.32 New TAVR systems (see below) have been designed to reduce or eliminate PVR after TAVR in the future by improving sub-annular fixation or with peri-valve spacefilling technology.
Vascular events and bleeding From PARTNER, Genereux et al reported that the 15.3% of inoperable and high-risk A and B patients experiencing Valve Academic Research Consortium (VARC) major vascular complications had significantly higher rates of 30-day and 1-year mortality.33 In this analysis, the only identifiable independent predictor of major vascular complications was female gender (HR 2.31, p = 0.03). Major vascular complications are an independent predictor of major bleeding events, which were found to be the strongest independent predictor of 1-year mortality in PARTNER high-risk patients, although there was a greater prognostic impact in the SAVR arm.34 Hayashida et al, also showed that VARC major vascular complications increased 30-day mortality and were predicted by low procedural experience, femoral calcification, and high sheath-to-femoral artery ratio.35 As mentioned previously, one-year randomized data from PARTNER comparing SAPIEN vs. the lower profile SAPIEN XT7 show a reduction in major vascular complications. Reduction in vascular and bleeding events is dependent on appropriate multi-slice CT screening of vascular anatomy and development of lower profile TAVR systems, in addition to implementation of advanced percutaneous closure techniques.36
Conduction abnormalities There are important differences in the need for permanent pacemakers after TAVR between the balloon-expandable SAPIEN valve and self-expanding CoreValve; 6.5% with SAPIEN vs. 25.8% with CoreValve (p < 0.001) in a large meta-analysis.37 Similarly, the frequency of new-onset left bundle branch block (LBBB) is increased with CoreValve compared with the SAPIEN valve. In a combined analysis of all PARTNER data, Nazif et al showed that persistent, new-onset LBBB occurred in 10.5% of balloonexpandable TAVR patients with normal baseline conduction.38 This finding did not result in increased all-cause mortality, as was indicated in another retrospective study,39 but was associated with a higher rate of subsequent pacemaker implantation and failure of improvement in LV ejection fraction. Device-related factors, such as the depth of the device implant in the LV outflow tract and the continuous radial force exerted by the self-
PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S 5 6 (2 0 1 4) 63 5–6 4 5
expanding CoreValve upon deployment (perhaps causing edema or inflammation of the conduction system in the membranous septum), appear to predispose TAVR patients to conduction abnormalities.
TAVR training and site access issues A four-society expert consensus statement published in 2012 elaborates the recommended criteria for new and existing TAVR programs, pertaining to practitioner (interventionalist or surgeon) and programmatic requirements.40 These guidelines serve as a foundation to maximize the opportunity to provide safe and effective adoption of TAVR into new centers, while maintaining access to care for patients in need of this worthwhile therapy. It has been estimated that approximately 400 of the 1,150 cardiac centers in the US that are currently performing SAVR would meet these initial criteria. 2 Currently, approximately 300 centers are performing TAVR in the US. An expert consensus document, incorporating input from 12 professional societies, detailed all aspects of TAVR and its integration into current clinical practice, highlighting critical published data, including clinical results from the PARTNER trial.16 These consensus documents will become increasingly valuable as more trialbased evidence becomes available and as new device platforms and newer generations of existing device platforms gather clinical data. The expectations regarding training, procedural volume requirements, and anticipated referral patterns appear to be more conservative amongst TAVR trialists than practicing clinical interventionalists. 41 Formal sponsor required and society-based training programs including hands-on exposure to TAVR equipment, simulation training, didactic sessions, imaging workshops, and case presentations are the current foundation for integrating future TAVR practitioners and sites. Thereafter, careful in-person proctoring experiences, maintaining reasonable case volumes to support the maturation of the Heart Valve Team and to reduce “learning curve” concerns, and ongoing advanced training symposia are required to insure optimal clinical outcomes in these high-risk AS patients.
Future opportunities Expanded TAVR clinical indications Surgical bio-prosthetic valve failure (valve-in-valve) Management of patients with acute or chronic structural valve deterioration after surgically implanted bioprosthetic valves is often problematic and can only be successfully treated by repeat SAVR. The availability of a transcatheter less-invasive procedure is an attractive option especially in older patients with co-morbidities or other high-risk characteristics. Early successful cases of TAVR for SAVR failure (called “valve-in-valve”) demonstrated the feasibility of both balloon-expandable and selfexpanding platforms for SAVR failures and the TA balloonexpandable platform for surgical mitral valve failures.42,43 Perhaps most important in TAVR valve-in-valve procedures is a complete knowledge of the subtleties and differences among surgical valve
639
prostheses, as described in recent manuscripts and in a widely used internet application based on the work of Vinayak Bapat.44 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 TAVR placement. Clinical data from a large global TAVR valve-in-valve registry reported by Dvir et al45 indicate that: (1) valve hemodynamics are good, although not as good as TAVR in native valves, especially in small bioprostheses (≤21 mm), wherein the supra-annular CoreValve may have an advantage; (2) PVR is rarely observed; (3) clinical outcomes are generally similar to native valve TAVR accounting for relative differences in patient co-morbidities; (4) both hemodynamics and clinical outcomes are better when the mode of SAVR failure is PVR vs. AS; (5) there is a prohibitively higher frequency of coronary artery obstruction in surgical prostheses with the valve external to the frame. Most thoughtful TAVR specialists (surgeons and cardiologists) agree that TAVR will become the treatment of choice for bioprosthetic SAVR failure in most patients in the future.
Intermediate risk AS patients Although categorical risk profiling appears contrived in situations where surgical risk is a continuous occurrence, for regulatory and other purposes, the SAVR population has been partitioned into low, intermediate, and high risk subgroups. Based upon current data in high-risk AS patients and evidence suggesting improved clinical outcomes with recent modifications in procedural factors and technology, intermediate risk AS patients would be the next logical use extension for TAVR. The intermediate risk surgical cohort represents between one-quarter and one-third of surgically eligible patients, and using the STS quantitative risk scoring system as a guidepost, an STS score between 3 or 4% and 8% approximates intermediate risk for most AS patients. A multicenter propensity risk adjustment study in intermediate risk patients comparing TAVR and SAVR has indicated similar early and late mortality.46 Similarly, among the initial 7,710 TAVR patients enrolled in the US Transcatheter Valve Therapy registry,2 the median baseline STS score was only 7%, suggesting that a significant subgroup was likely intermediate risk. The overall observed in-hospital mortality and stroke rates were 5.5% and 2.0% respectively, both very acceptable outcomes for recently trained centers. Two important large randomized clinical trials, PARTNER 2A (SAPIEN XT valve) and SURTAVI (CoreValve), in intermediate risk AS patients (STS score ~4–8%) comparing TAVR vs. SAVR are ongoing and future analyses of the more than 4,000 randomized patients from these studies should help to clarify questions regarding advisability of TAVR in this risk strata. Nevertheless, clinical practice around the world has already evolved, as elderly AS patients (>80 years old) with none or one co-morbidity are often treated with TAVR strategies.
Other possible clinical indications Although, several other patient subgroups and clinical indications would seem reasonable candidates for TAVR therapy, ultimate decisions await careful assessments of clinical need and evaluations of results from rigorous clinical trials. For instance, subset analyses from PARTNER47 indicate
640
PR O GRE S S I N C ARDI O VAS CU L AR D I S EAS E S 5 6 ( 2 0 14 ) 63 5–6 45
that patients with low flow–low gradient AS may benefit from TAVR and could be an alternative to SAVR, especially in high-risk patients. Patients with AS and concomitant coronary disease, mandating combined AVR and coronary revascularization, may do as well or better with the combination of TAVR and percutaneous coronary angioplasty. Asymptomatic “very severe” AS (peak velocity > 5 liters/sec) patients may be well suited for preemptive TAVR rather than watchful waiting strategies and there are already some early data in selected patients with predominant aortic regurgitation who were successfully treated using self-expanding TAVR systems.48,49 Clearly, the temptation to generalize TAVR to all current SAVR situations must be resisted until confirmation of transcatheter valve durability and compelling clinical evidence dictates a change in clinical practice.
New TAVR systems Important features of new TAVR systems Despite the success of “first generation” TAVR systems, several device design limitations have been identified which have contributed to suboptimal clinical outcomes (Table 1). The major limitation of early TAVR technologies was the requirement of excessively large diameter TAVR delivery sheaths. The overall outer diameter profile of the SAPIEN balloon-expandable FDA-approved TAVR system is > 8 mm for the 23 mm valve size and > 9 mm for the 26 mm valve. This creates a significant femoral artery– sheath size mismatch in many patients resulting in both frequent vascular complications and the frequent use of nonTF access sites. In the future, to support successful TF access in the vast majority of TAVR-eligible patients (especially women), an outer sheath diameter of less than 18 French for all valve sizes is advisable. Smaller TAVR system profiles are also important to negotiate tortuous vascular anatomy, facilitate native valve crossing, minimize trauma to the aorta and the native valve, allow the option of no pre-dilation before deployment, and improve alignment and positioning accuracy during implantation. Another important limitation of early and current TAVR systems is the lack of consistent and precise positioning during deployment which may result in valve embolization (or “pull through”), obstruction of the
Table 1 – Design limitations of first generation TAVR systems. 1. Large diameter delivery sheaths and catheters resulting in frequent vascular complications and non-transfemoral access alternatives 2. Imprecise valve positioning during deployment which may cause para-valvular regurgitation or interfere with aortic valvar complex structures (aortic root, coronary arteries, conduction system, and mitral valve) 3. Absence of valve retrieval and repositioning features 4. Definitive approach to reduce or eliminate para-valvular regurgitation (either improved sub-annular fixation or external space-filling materials. or both) 5. Unknown durability of the frame and valves (material composition and thickness, valve geometry, and effects of crimping)
coronary arteries (too high placement), interference with the conduction system or the mitral valve (too low placement), and increased PVR (either too high or too low placement). Ideally, a slow and controlled valve deployment, allowing for positioning adjustments before final implantation is preferred. This may be limited by the need for transient rapid RV pacing during deployment of balloon-expandable valves. The availability of partial or complete valve retrieval is being incorporated into many of the newer TAVR systems, which provides the operator a “second chance” if the initial attempts at precise positioning were suboptimal. The major difference between SAVR and TAVR has been the greater frequency and severity of PVR in most currently available TAVR systems. To address this issue, newer devices have explored improvements in sub-annular fixation and coaxial alignment, as well as the addition of external space-filling materials to reduce or eliminate incomplete circumferential apposition of the valve frame against the aortic annulus. Finally, durability of the frame and valve itself remains a concern, especially if long-term implants in younger patients are being contemplated.
New TAVR systems with significant clinical data (and CE-approval) Most of the new TAVR systems in early stages of clinical practice have creatively attempted to incorporate design changes which reduce many of the aforementioned limitations. Importantly, both Edwards SAPIEN and Medtronic CoreValve TAVR technologies have similarly evolved and current iterations of these landmark devices should be compared with other new TAVR systems. The version of the balloon-expandable TAVR system most commonly used around the world is the SAPIEN XT (Fig 1A). This device was completely redesigned with important changes in the frame (reduced metal, different geometry, and cobalt alloy material), the valve itself (geometry allowing partially closed configuration and 29 mm size) and enhanced tissue processing to improve durability. The delivery system is much lower in profile (18 and 20 French), which represents a 33% cross-sectional area reduction, in part due to in situ docking of the stent valve on the balloon. A large multicenter European registry (SOURCE XT) and a randomized multicenter US trial comparing SAPIEN with SAPIEN XT (PARTNER 2B) have confirmed improved ease-of-use and reduced complications with SAPIEN XT in high-risk AS patients.7,50 The newest version of the balloon expandable platform, SAPIEN 3 (Fig 1B), has just received CE-approval and has been studied in an early US registry. This TAVR system incorporates a further refinement in frame strut pattern, additional changes in valve geometry, even lower profile delivery systems (all valves delivered via 14 and 16 French expandable sheaths), more precise positioning features prior to deployment, and an external fabric (polyethylene terephthalate) skirt which prevents PVR.51 The new Edwards self-expanding TAVR system, Centera (Fig 1C), has just initiated clinical trials.52 The contoured short frame height, treated bovine pericardial valve, and 14 French motorized delivery catheter allowing the valve to be fully retrieved and redeployed before final implantation are distinguishing characteristics. The CoreValve Evolut R (Fig 2) is a next generation selfexpanding TAVR system with several enhancements, including
PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S 5 6 (2 0 1 4) 63 5–6 4 5
A
B
C
641
D
Fig 1 – Edwards balloon-expandable and self-expanding TAVR systems: A – balloon expandable SAPIEN. B – balloon expandable SAPIEN XT. C – balloon expandable SAPIEN 3. D – self-expanding Centera.
a redesigned and shortened outflow section, more consistent radial force, an extended inflow skirt to elongate the landing zone (should reduce PVR), a lower profile in-line sheath, and full retrievability during deployment.53 Presently, clinical data using the Evolut R TAVR system are being obtained in Europe. The Medtronic Engager TAVR system (Fig 3) is a TA device (29 French) with a self-expanding short nitinol frame and polyester skirt, control arms which are placed outside the native leaflets, a supra-annular bovine pericardial tissue valve, and commissural alignment features. A 125 patient registry in high-risk AS patients indicated good clinical outcomes with very low (
Available online at www.sciencedirect.com
ScienceDirect www.onlinepcd.com
Challenges and Future Opportunities for Transcatheter Aortic Valve Therapy Martin B. Leona,⁎, Hemal Gadaa , Gregory P. Fontanab a
Center for Interventional Vascular Therapy, Columbia University Medical Center, New York Presbyterian Hospital, New York, NY Lenox Hill Heart & Vascular Institute of New York, Lennox Hill Hospital, North Shore – LIJ Health Care System, Manhasset, NY
b
A R T I C LE I N F O
AB ST R A C T
Keywords:
Background: Transcatheter aortic valve replacement (TAVR) is a novel less-invasive therapy
Transcatheter aortic valve replacement
for high-risk patients with severe aortic stenosis (AS). Despite the impressive clinical
Aortic stenosis
growth of TAVR, there are many challenges as well as future opportunities.
Aortic valve replacement
Results: The heart valve team serves as the central vehicle for determining appropriate case selection. Considerations which impact clinical therapy decisions include frailty assessments and defining clinical “futility”. There are many controversial procedural issues; choice of vascular access site, valve sizing, adjunctive imaging, and post-dilatation strategies. Complications associated with TAVR (strokes, vascular and bleeding events, para-valvular regurgitation, and conduction abnormalities) must be improved and will require procedural and/or technology enhancements. TAVR site training mandates a rigorous commitment to established society and sponsor guidelines. In the future, TAVR clinical indications should extend to bioprosthetic valve failure, intermediate risk patients, and other clinical scenarios, based upon well conducted clinical trials. New TAVR systems have been developed which should further optimize clinical outcomes, by reducing device profile, providing retrievable features, and preventing para-valvular regurgitation. Other accessory devices, such as cerebral protection to prevent strokes, are also being developed and evaluated in clinical studies. Summary: TAVR is a worthwhile addition to the armamentarium of therapies for patients with AS. Current limitations are important to recognize and future opportunities to improve clinical outcomes are being explored. © 2014 Elsevier Inc. All rights reserved.
Background In the past decade, after initial proof-of-concept and subsequent feasibility studies, the application of less-invasive catheter-based approaches to functionally replace diseased aortic valves has been incorporated into the clinical treatment armamentarium in symptomatic high-risk patients with
severe aortic stenosis (AS). Since 2007, in more than 50 countries, over 750 cardiovascular centers have treated almost 100,000 aortic stenosis patients using transcatheter aortic valve replacement (TAVR) technologies. Despite the rapid acceptance and clinical appeal of TAVR, as with any new and novel medical therapy, there are still many challenges to be addressed and future opportunities to be
Statement of Conflict of Interest: see page 643. ⁎ Address reprint requests to Martin B. Leon, MD, Columbia University Medical Center, 161 Ft. Washington Avenue, Herbert Irving Pavilion, 6th Floor, New York, NY 10032. E-mail address: [email protected] (M.B. Leon). 0033-0620/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.pcad.2014.03.004
636
PR O GRE S S I N C ARDI O VAS CU L AR D I S EAS E S 5 6 ( 2 0 14 ) 63 5–6 45
Abbreviations and Acronyms AS = aortic stenosis CT = computerized tomography ICE = intra-cardiac echocardiography LBBB = left bundle branch block LV = left ventricular PARTNER = Placement of Aortic Transcatheter Valves
explored. The purpose of this manuscript is to selectively highlight the crucial challenges of TAVR which are presently under investigation and to direct attention towards expanding clinical applications and new technologies which constitute important future opportunities.
PVR = para-valvular regurgitation RV = right ventricular
Challenges
STS = Society for Thoracic Surgeons
Case selection
TA = transapical TAo = transaortic TAVR = transcatheter aortic valve replacement
Identifying “high-risk” patients
Patient selection under the auspices of a multidisciplinary “Heart Team” TEE = transesophageal echocardiography is crucial to achieve optimal clinical outcomes TF = transfemoral after TAVR. The differenTVT = transcatheter valve tiation between high-risk, therapy “inoperable” (or extreme risk), and prohibitive risk US = United States AS patients has been acVARC = Valve Academic tively debated since reguResearch Consortium latory approval of TAVR and especially during the formulation of the Placement of Aortic Transcatheter Valves (PARTNER) clinical trial.1 Risk assessment has often been guided by standard surgical scoring systems, including the Society of Thoracic Surgery (STS) and EuroSCORE models, which were not fully validated in this high-risk patient population. These on-line risk scores, as designed for everyday use, do not include important co-morbidities such as severe pulmonary hypertension, right ventricular (RV) dysfunction, severe liver disease, home supplemental oxygen, prohibitive anatomy (such as chest deformity or severe aortic calcification), disability, or frailty. Characterization of surgical risk requires direct involvement of experienced surgeons who usually include a number of important co-morbidities when considering the highest risk patients for TAVR: malnutrition and cachexia, physical deconditioning or wheelchair bound, chronic kidney disease on dialysis, history of particular solid tumor malignancies, neurological disorders such as dementia and stroke, and other debilitating conditions that preclude patients from returning to a reasonable functional status. One of the biggest challenges in assessment of patient risk status is developing a validated quantitative algorithm that best defines patient risk from the standpoint of predicting early and late mortality as well as functional recovery in the setting of TAVR. The combined analyses of the PARTNER trials or the new
United States (US) Transcatheter Valve Therapies (TVT) National Registry will hopefully provide sufficient patient data to offer the possibility of a TAVR specific risk algorithm at some point in the future.2
Frailty and futility Not entirely captured in current risk stratification metrics is the attribute of frailty, which has been associated with worse TAVR outcomes. The concept of frailty is crudely defined as an impairment in multiple systems that leads to a decline in resiliency and homeostatic reserve. It is influenced by physical disability and medical co-morbidities, but is not adequately described by just these attributes.3 Green et al have devised a frailty score for TAVR patients, based loosely on criteria established by Fried et al.4 The frailty phenotype, including impairments in gait speed and grip strength, reduced serum albumin, and diminished Katz activities of daily living, was associated with a longer post-TAVR hospital stay, as well as increased 1-year mortality.5 The multicenter FRAILTY-AVR study will compare outcomes of surgical aortic valve replacement (SAVR) and TAVR using several frailty assessment tools in the effort to define which factors are the most predictive of mortality and morbidity in elderly patients. The results of the US CoreValve Pivotal Trial Extreme Risk cohort highlight the need to define and quantify the significance of this interaction, as the only two significant predictors of all-cause mortality or major stroke (the primary endpoint), were STS score of > 15% (p = 0.02) and residence in an assisted living facility (p < 0.01).6 A careful frailty assessment plays a key role in the differentiation of “futility” (“no hope” patients) and high-risk utility patients and should be incorporated into all TAVR risk stratification analyses. The term “Cohort C” describes this subset of futile inoperable patients who have both poor survival (i.e. less than 1 year) and poor quality of life, despite successful TAVR. Simply stated, “Cohort C” or futile patients represent those patients who are dying with aortic stenosis but not from AS. Common clinical characteristics most associated with futile risk patients include extreme comorbidities (e.g. STS score > 15%), extreme frailty usually with a dependent social status, severe pulmonary or liver disease, severe dementia, chronic kidney disease (e.g. dialysis dependent), and hemodynamic instability (especially requiring vasopressors). What remains to be defined is the quantitative interplay of frailty metrics and existing risk stratification models based on age and co-morbid conditions, in accurately determining a “Cohort C” patient.
Procedural considerations Access alternatives Factors which may determine preferred TAVR vascular access include peripheral arterial disease (inadequate vessel diameter, severe calcification or extreme tortuosity of the iliofemoral vessels), the presence of extensive calcification of the ascending aorta (i.e. porcelain aorta), hostile chest wall anatomy (either due to ortho-voltage radiation exposure or chest wall deformities), previous coronary bypass graft surgery with mammary conduits adherent to the chest wall,
PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S 5 6 (2 0 1 4) 63 5–6 4 5
and severe lung disease. The four most common techniques for TAVR access are the retrograde transfemoral (TF), antegrade transapical (TA), and the more recently developed direct or trans-aortic (TAo) and subclavian approaches. In the PARTNER trial, using the larger profile SAPIEN transcatheter valve system (outer sheath diameter 9.2 mm for the 26 mm valve) there were frequent major vascular complications associated with TF–TAVR procedures.1 In the PARTNER II trial, the first-generation SAPIEN system was compared to the lower profile SAPIEN XT system (33% lower cross-sectional area) and major vascular complications were reduced from 15.5% to 9.6% (p = 0.04).7 This highlights the importance of lower profile delivery systems in maximally utilizing safe fully percutaneous TF-TAVR as a primary default access strategy. The TA approach avoids peripheral access issues, but also has limitations; increased length of hospitalization, and increased risk of 30-day and 1-year all-cause mortality.8,9 These adverse TA outcomes may have been influenced by differences in underlying baseline co-morbidities between the two populations, given the “TF-first” approach often adopted by clinicians, which relegated only patients with significant peripheral vascular disease to TA-TAVR. Other adverse outcomes associated with the TA approach include a higher likelihood of peri-procedural bleeding, increased risk of hemodynamic instability, and greater patient discomfort, due to pain related to the antero-lateral thoracotomy.10,11 TAo and subclavian access sites for TAVR have been introduced more recently. The largest series of TAo cases12 indicated that compared to a contemporary group of TA patients, there was a lower combined bleeding and vascular event rate (27% vs 46%; p = 0.05), shorter median intensive care unit length of stay (3 vs 6 days; p = 0.01), and a favorable learning curve. Transcarotid access and antegrade transseptal access via the femoral vein have also been described13,14 but there are scant clinical outcome data. The choice of vascular access site for TAVR is an individualized patient-based decision determined by clinical factors, anatomic considerations as well as the experiences and preferences of the Heart Team. Considering the less-invasive nature of fully percutaneous TF access and the increasing availability of lower profile TAVR systems, it is likely that TF–TAVR will be the preferred option for the majority of patients in the foreseeable future.
Valve sizing and positioning Multimodality imaging is essential for patient screening and procedural guidance during TAVR, and has been incorporated into consensus statements, reviews, and guidelines.15,16 Correct valve sizing for either the balloon-expandable or the self-expandable TAVR system requires meticulous attention to three-dimensional imaging, including multi-slice CT and trans-esophageal echocardiography (TEE). Optimal TAVR implantation requires: (1) correct valve sizing based upon established criteria for measuring the annulus dimensions matched to the specific valve type; (2) accurate valve positioning (axial height and alignment) within the annular valve plane. Different TAVR systems mandate specific procedural techniques to determine co-planar implantation views and optimal height and alignment of valve implantation, which can be facilitated by rapid pacing with cine-
637
fluoroscopic and/or TEE guidance. The challenge of valve sizing and positioning cannot be underestimated and requires an intimate understanding of the anatomy of the aortic valvar complex. Ideally, correct sizing and placement of TAVR will result in excellent valve hemodynamics, none or trace para-valvular regurgitation (PVR), low requirements for new pacemakers due to conduction abnormalities, and no evidence of coronary obstruction or annulus injury.
Trans-esophageal echocardiography Intraprocedural TEE can provide a real-time biplane assessment of the annulus before and during deployment, thus fostering more precise valve positioning. TEE can also help to predict and is the gold standard to detect PVR after valve implantation, as well as other complications such as coronary artery obstruction, annulus rupture, pericardial tamponade (e.g. due to chamber perforation), severe mitral regurgitation, aortic dissection (or hematoma), and left ventricular (LV) dysfunction. However, the routine use of TEE during TAVR procedures has become controversial, as TEE usually is associated with general anesthesia and endotracheal intubation, which may introduce additional risks in patients who are hemodynamically unstable or have underlying severe pulmonary disease. Moreover, TEE requires specialized imaging expertise which may not be readily available for all cases and the additional sedation may delay recovery. Therefore, a strong trend in TAVR procedures has been to selectively or systematically favor monitored anesthesia control without intubation, combined with transthoracic echocardiography, as needed. The virtues and drawbacks of routine TEE have been hotly debated and the decision to utilize TEE on a case or sitespecific basis is presently determined by resource availability, perceived clinical need, and personal preferences. The motivation to eliminate the need for general anesthesia has led to an increasing interest in intracardiac echocardiography (ICE) imaging for TAVR procedures. However, single-plane ICE imaging cannot accurately visualize the oval annulus for sizing purposes, measure the coronary artery height, or reliably assess post-implantation PVR.17 New three-dimensional ICE catheters may overcome some of these limitations in the future, thus permitting on-line echo guidance and assessment without the need for general anesthesia.
Pre- and post-dilation Balloon aortic valvuloplasty (BAV) prior to valve deployment has been traditionally performed, especially prior to balloonexpandable transcatheter valve deployment. Pre-dilation BAV allows easier crossing of the valve through the annulus and potentially avoids mechanical complications related to the force and contour of the delivery system. However, BAV carries independent risks of atrioventricular block requiring permanent pacemakers, increased aortic regurgitation, and embolic neurologic events. Recent trends have favored reduced or no pre-dilation with low profile self-expanding valve platforms. Grube et al treated 60 consecutive patients with the Medtronic CoreValve prosthesis without balloon pre-dilation and good hemodynamic performance
638
PR O GRE S S I N C ARDI O VAS CU L AR D I S EAS E S 5 6 ( 2 0 14 ) 63 5–6 45
(post-procedural mean gradients and effective orifice areas).18 In addition, in-hospital major adverse cardiovascular events and need for permanent pacemakers were far less than those reported in the US CoreValve Pivotal Trial Extreme Risk cohort.6 Garcia et al reported the results of no pre-dilation with the SAPIEN XT valve in 10 patients with moderate calcification, homogenous distribution of calcium, and symmetric opening of the valve.19 No important complications were observed and the patients did not require post-dilation for significant PVR which was none or trivial in all patients. Thus, the growing trend of TAVR without pre-dilation appears to be a feasible technique with potential benefits and will be a target for further study in the future. Post-BAV after TAVR has been selectively applied usually in situations of significant PVR, which has been associated with increased late mortality.20 The benefits and risks of post-BAV after TAVR have been hotly debated. TEE data clearly indicate that post-dilation in carefully selected patients importantly reduces PVR. 21 However, post-dilation has been associated with increased complications including embolic neurologic events,22 conduction abnormalities, and the risk of annular rupture. Barbanti et al reported 31 patients receiving balloon-expandable TAVR with annular rupture from a large multicenter TAVR experience 23 and the predictors of annular rupture were subannuluar/LV outflow tract calcification, a higher frequency of ≥ 20% annular area oversizing, and balloon post-dilation. In the future, with improved valve sizing and new technology to prevent PVR, there will be a reduced need for selective post-dilation after TAVR.
TAVR complications – brief updates, current and future management Strokes Strokes in the setting of TAVR remains a major peri- and postprocedural complication manifesting with a significant deterioration in quality-of-life and increased mortality. Several diffusion weighted MRI studies have shown that the rate of silent cerebral embolism after TAVR approaches 80% of patients.24–26 Van Mieghem et al confirmed these neuro-imaging findings, noting 75% of TAVR procedures producing debris captured in a filterbased cerebral embolic protection device.27 However, a recent meta-analysis of >10,000 patients in 53 studies confirmed that TAVR is associated with in a reasonable peri-procedural stroke rate of 1.5% and a 30-day stroke/transient ischemic attack rate of 3.3%.28 Thus, the discordance between detection of neuroembolic activity with TAVR and subsequent clinical neurologic events requires further resolution. Risk stratification for stroke based on patient characteristics is essential in defining who would likely benefit from cerebral embolic protection devices, as well as tailoring post-procedural management (e.g. surveillance for post-procedural atrial fibrillation and adjunctive pharmacotherapy).
Paravalvular regurgitation The 2-year follow-up of PARTNER Cohort A patients demonstrated that TAVR resulted in significantly worse PVR than SAVR with >50% of TAVR patients had at least mild PVR.20 Moreover,
even mild PVR post-TAVR was associated with 10–15% higher mortality at 2 years than patients with none or trace PVR, as determined by a core echocardiography laboratory. Multiple other studies have shown similar associations between varying degrees of significant PVR and increased late mortality after TAVR.29–31 Accurate diagnosis and clinical impact require a combined assessment of hemodynamics, angiography, and especially echocardiography, which remains the gold standard. The treatment of PVR is based on an understanding of the severity and specific etiology. Valve undersizing or underexpansion, valve mal-alignment (either too high or too low), and severe global and focal aortic valvar complex calcification with mal-apposition are the main causes of PVR and treatment options include strategic post-dilatation, placement of additional TAVR to extend the “seal zone”, or implantation of peri-valve vascular plugs.32 New TAVR systems (see below) have been designed to reduce or eliminate PVR after TAVR in the future by improving sub-annular fixation or with peri-valve spacefilling technology.
Vascular events and bleeding From PARTNER, Genereux et al reported that the 15.3% of inoperable and high-risk A and B patients experiencing Valve Academic Research Consortium (VARC) major vascular complications had significantly higher rates of 30-day and 1-year mortality.33 In this analysis, the only identifiable independent predictor of major vascular complications was female gender (HR 2.31, p = 0.03). Major vascular complications are an independent predictor of major bleeding events, which were found to be the strongest independent predictor of 1-year mortality in PARTNER high-risk patients, although there was a greater prognostic impact in the SAVR arm.34 Hayashida et al, also showed that VARC major vascular complications increased 30-day mortality and were predicted by low procedural experience, femoral calcification, and high sheath-to-femoral artery ratio.35 As mentioned previously, one-year randomized data from PARTNER comparing SAPIEN vs. the lower profile SAPIEN XT7 show a reduction in major vascular complications. Reduction in vascular and bleeding events is dependent on appropriate multi-slice CT screening of vascular anatomy and development of lower profile TAVR systems, in addition to implementation of advanced percutaneous closure techniques.36
Conduction abnormalities There are important differences in the need for permanent pacemakers after TAVR between the balloon-expandable SAPIEN valve and self-expanding CoreValve; 6.5% with SAPIEN vs. 25.8% with CoreValve (p < 0.001) in a large meta-analysis.37 Similarly, the frequency of new-onset left bundle branch block (LBBB) is increased with CoreValve compared with the SAPIEN valve. In a combined analysis of all PARTNER data, Nazif et al showed that persistent, new-onset LBBB occurred in 10.5% of balloonexpandable TAVR patients with normal baseline conduction.38 This finding did not result in increased all-cause mortality, as was indicated in another retrospective study,39 but was associated with a higher rate of subsequent pacemaker implantation and failure of improvement in LV ejection fraction. Device-related factors, such as the depth of the device implant in the LV outflow tract and the continuous radial force exerted by the self-
PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S 5 6 (2 0 1 4) 63 5–6 4 5
expanding CoreValve upon deployment (perhaps causing edema or inflammation of the conduction system in the membranous septum), appear to predispose TAVR patients to conduction abnormalities.
TAVR training and site access issues A four-society expert consensus statement published in 2012 elaborates the recommended criteria for new and existing TAVR programs, pertaining to practitioner (interventionalist or surgeon) and programmatic requirements.40 These guidelines serve as a foundation to maximize the opportunity to provide safe and effective adoption of TAVR into new centers, while maintaining access to care for patients in need of this worthwhile therapy. It has been estimated that approximately 400 of the 1,150 cardiac centers in the US that are currently performing SAVR would meet these initial criteria. 2 Currently, approximately 300 centers are performing TAVR in the US. An expert consensus document, incorporating input from 12 professional societies, detailed all aspects of TAVR and its integration into current clinical practice, highlighting critical published data, including clinical results from the PARTNER trial.16 These consensus documents will become increasingly valuable as more trialbased evidence becomes available and as new device platforms and newer generations of existing device platforms gather clinical data. The expectations regarding training, procedural volume requirements, and anticipated referral patterns appear to be more conservative amongst TAVR trialists than practicing clinical interventionalists. 41 Formal sponsor required and society-based training programs including hands-on exposure to TAVR equipment, simulation training, didactic sessions, imaging workshops, and case presentations are the current foundation for integrating future TAVR practitioners and sites. Thereafter, careful in-person proctoring experiences, maintaining reasonable case volumes to support the maturation of the Heart Valve Team and to reduce “learning curve” concerns, and ongoing advanced training symposia are required to insure optimal clinical outcomes in these high-risk AS patients.
Future opportunities Expanded TAVR clinical indications Surgical bio-prosthetic valve failure (valve-in-valve) Management of patients with acute or chronic structural valve deterioration after surgically implanted bioprosthetic valves is often problematic and can only be successfully treated by repeat SAVR. The availability of a transcatheter less-invasive procedure is an attractive option especially in older patients with co-morbidities or other high-risk characteristics. Early successful cases of TAVR for SAVR failure (called “valve-in-valve”) demonstrated the feasibility of both balloon-expandable and selfexpanding platforms for SAVR failures and the TA balloonexpandable platform for surgical mitral valve failures.42,43 Perhaps most important in TAVR valve-in-valve procedures is a complete knowledge of the subtleties and differences among surgical valve
639
prostheses, as described in recent manuscripts and in a widely used internet application based on the work of Vinayak Bapat.44 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 TAVR placement. Clinical data from a large global TAVR valve-in-valve registry reported by Dvir et al45 indicate that: (1) valve hemodynamics are good, although not as good as TAVR in native valves, especially in small bioprostheses (≤21 mm), wherein the supra-annular CoreValve may have an advantage; (2) PVR is rarely observed; (3) clinical outcomes are generally similar to native valve TAVR accounting for relative differences in patient co-morbidities; (4) both hemodynamics and clinical outcomes are better when the mode of SAVR failure is PVR vs. AS; (5) there is a prohibitively higher frequency of coronary artery obstruction in surgical prostheses with the valve external to the frame. Most thoughtful TAVR specialists (surgeons and cardiologists) agree that TAVR will become the treatment of choice for bioprosthetic SAVR failure in most patients in the future.
Intermediate risk AS patients Although categorical risk profiling appears contrived in situations where surgical risk is a continuous occurrence, for regulatory and other purposes, the SAVR population has been partitioned into low, intermediate, and high risk subgroups. Based upon current data in high-risk AS patients and evidence suggesting improved clinical outcomes with recent modifications in procedural factors and technology, intermediate risk AS patients would be the next logical use extension for TAVR. The intermediate risk surgical cohort represents between one-quarter and one-third of surgically eligible patients, and using the STS quantitative risk scoring system as a guidepost, an STS score between 3 or 4% and 8% approximates intermediate risk for most AS patients. A multicenter propensity risk adjustment study in intermediate risk patients comparing TAVR and SAVR has indicated similar early and late mortality.46 Similarly, among the initial 7,710 TAVR patients enrolled in the US Transcatheter Valve Therapy registry,2 the median baseline STS score was only 7%, suggesting that a significant subgroup was likely intermediate risk. The overall observed in-hospital mortality and stroke rates were 5.5% and 2.0% respectively, both very acceptable outcomes for recently trained centers. Two important large randomized clinical trials, PARTNER 2A (SAPIEN XT valve) and SURTAVI (CoreValve), in intermediate risk AS patients (STS score ~4–8%) comparing TAVR vs. SAVR are ongoing and future analyses of the more than 4,000 randomized patients from these studies should help to clarify questions regarding advisability of TAVR in this risk strata. Nevertheless, clinical practice around the world has already evolved, as elderly AS patients (>80 years old) with none or one co-morbidity are often treated with TAVR strategies.
Other possible clinical indications Although, several other patient subgroups and clinical indications would seem reasonable candidates for TAVR therapy, ultimate decisions await careful assessments of clinical need and evaluations of results from rigorous clinical trials. For instance, subset analyses from PARTNER47 indicate
640
PR O GRE S S I N C ARDI O VAS CU L AR D I S EAS E S 5 6 ( 2 0 14 ) 63 5–6 45
that patients with low flow–low gradient AS may benefit from TAVR and could be an alternative to SAVR, especially in high-risk patients. Patients with AS and concomitant coronary disease, mandating combined AVR and coronary revascularization, may do as well or better with the combination of TAVR and percutaneous coronary angioplasty. Asymptomatic “very severe” AS (peak velocity > 5 liters/sec) patients may be well suited for preemptive TAVR rather than watchful waiting strategies and there are already some early data in selected patients with predominant aortic regurgitation who were successfully treated using self-expanding TAVR systems.48,49 Clearly, the temptation to generalize TAVR to all current SAVR situations must be resisted until confirmation of transcatheter valve durability and compelling clinical evidence dictates a change in clinical practice.
New TAVR systems Important features of new TAVR systems Despite the success of “first generation” TAVR systems, several device design limitations have been identified which have contributed to suboptimal clinical outcomes (Table 1). The major limitation of early TAVR technologies was the requirement of excessively large diameter TAVR delivery sheaths. The overall outer diameter profile of the SAPIEN balloon-expandable FDA-approved TAVR system is > 8 mm for the 23 mm valve size and > 9 mm for the 26 mm valve. This creates a significant femoral artery– sheath size mismatch in many patients resulting in both frequent vascular complications and the frequent use of nonTF access sites. In the future, to support successful TF access in the vast majority of TAVR-eligible patients (especially women), an outer sheath diameter of less than 18 French for all valve sizes is advisable. Smaller TAVR system profiles are also important to negotiate tortuous vascular anatomy, facilitate native valve crossing, minimize trauma to the aorta and the native valve, allow the option of no pre-dilation before deployment, and improve alignment and positioning accuracy during implantation. Another important limitation of early and current TAVR systems is the lack of consistent and precise positioning during deployment which may result in valve embolization (or “pull through”), obstruction of the
Table 1 – Design limitations of first generation TAVR systems. 1. Large diameter delivery sheaths and catheters resulting in frequent vascular complications and non-transfemoral access alternatives 2. Imprecise valve positioning during deployment which may cause para-valvular regurgitation or interfere with aortic valvar complex structures (aortic root, coronary arteries, conduction system, and mitral valve) 3. Absence of valve retrieval and repositioning features 4. Definitive approach to reduce or eliminate para-valvular regurgitation (either improved sub-annular fixation or external space-filling materials. or both) 5. Unknown durability of the frame and valves (material composition and thickness, valve geometry, and effects of crimping)
coronary arteries (too high placement), interference with the conduction system or the mitral valve (too low placement), and increased PVR (either too high or too low placement). Ideally, a slow and controlled valve deployment, allowing for positioning adjustments before final implantation is preferred. This may be limited by the need for transient rapid RV pacing during deployment of balloon-expandable valves. The availability of partial or complete valve retrieval is being incorporated into many of the newer TAVR systems, which provides the operator a “second chance” if the initial attempts at precise positioning were suboptimal. The major difference between SAVR and TAVR has been the greater frequency and severity of PVR in most currently available TAVR systems. To address this issue, newer devices have explored improvements in sub-annular fixation and coaxial alignment, as well as the addition of external space-filling materials to reduce or eliminate incomplete circumferential apposition of the valve frame against the aortic annulus. Finally, durability of the frame and valve itself remains a concern, especially if long-term implants in younger patients are being contemplated.
New TAVR systems with significant clinical data (and CE-approval) Most of the new TAVR systems in early stages of clinical practice have creatively attempted to incorporate design changes which reduce many of the aforementioned limitations. Importantly, both Edwards SAPIEN and Medtronic CoreValve TAVR technologies have similarly evolved and current iterations of these landmark devices should be compared with other new TAVR systems. The version of the balloon-expandable TAVR system most commonly used around the world is the SAPIEN XT (Fig 1A). This device was completely redesigned with important changes in the frame (reduced metal, different geometry, and cobalt alloy material), the valve itself (geometry allowing partially closed configuration and 29 mm size) and enhanced tissue processing to improve durability. The delivery system is much lower in profile (18 and 20 French), which represents a 33% cross-sectional area reduction, in part due to in situ docking of the stent valve on the balloon. A large multicenter European registry (SOURCE XT) and a randomized multicenter US trial comparing SAPIEN with SAPIEN XT (PARTNER 2B) have confirmed improved ease-of-use and reduced complications with SAPIEN XT in high-risk AS patients.7,50 The newest version of the balloon expandable platform, SAPIEN 3 (Fig 1B), has just received CE-approval and has been studied in an early US registry. This TAVR system incorporates a further refinement in frame strut pattern, additional changes in valve geometry, even lower profile delivery systems (all valves delivered via 14 and 16 French expandable sheaths), more precise positioning features prior to deployment, and an external fabric (polyethylene terephthalate) skirt which prevents PVR.51 The new Edwards self-expanding TAVR system, Centera (Fig 1C), has just initiated clinical trials.52 The contoured short frame height, treated bovine pericardial valve, and 14 French motorized delivery catheter allowing the valve to be fully retrieved and redeployed before final implantation are distinguishing characteristics. The CoreValve Evolut R (Fig 2) is a next generation selfexpanding TAVR system with several enhancements, including
PR O G RE S S I N C ARDI O V A S CU L A R D I S EA S E S 5 6 (2 0 1 4) 63 5–6 4 5
A
B
C
641
D
Fig 1 – Edwards balloon-expandable and self-expanding TAVR systems: A – balloon expandable SAPIEN. B – balloon expandable SAPIEN XT. C – balloon expandable SAPIEN 3. D – self-expanding Centera.
a redesigned and shortened outflow section, more consistent radial force, an extended inflow skirt to elongate the landing zone (should reduce PVR), a lower profile in-line sheath, and full retrievability during deployment.53 Presently, clinical data using the Evolut R TAVR system are being obtained in Europe. The Medtronic Engager TAVR system (Fig 3) is a TA device (29 French) with a self-expanding short nitinol frame and polyester skirt, control arms which are placed outside the native leaflets, a supra-annular bovine pericardial tissue valve, and commissural alignment features. A 125 patient registry in high-risk AS patients indicated good clinical outcomes with very low (
