Who Is a TAVR Candidate in 2016?
The assessment of risk and patient selection.
Since the first transcatheter aortic valve replacement (TAVR) procedure was performed by Alain Cribier and colleagues in 2002,1 it is estimated that more than 200,000 people have been treated with transcatheter valve prostheses. The exponential uptake of this exciting treatment modality has been driven by many factors, including excellent trial and clinical results, patient preference, and physician comfort.
Aortic stenosis is predominantly a disease of age and is associated with senile calcific degeneration, particularly in developed nations. It is estimated that more than 4% of people older than 75 years have moderate or severe aortic stenosis, with increasing prevalence with advancing age.2This association with age results in a cohort with a significant burden of comorbidities and higher procedural risk, which prior to the introduction of TAVR, led to approximately 30% of patients not undergoing definitive operative management.3
Due to the increasing age of the population in most Western countries, the number of elderly patients with symptomatic severe aortic stenosis continues to increase. TAVR offers a less invasive means of definitive treatment with a robust evidence base showing it is superior to medical therapy4,5and at least equivalent, if not superior, to surgical valve replacement6,7in appropriately selected patients.
Increasing clinical use of TAVR has, however, been associated with a shift in treated patients toward a lower-risk cohort. It remains important that the technology is used to treat those patients who will receive the greatest clinical benefit, and it is imperative that TAVR be performed in accordance with contemporary trial evidence or within a research framework that will advance the current evidence base.
COHORTS TREATED IN CLINICAL TRIALS
Clinical practice must be built upon a strong evidence base and hence, treated patients should mirror those proven to benefit within clinical trials. At the same time, detailed post hoc analysis of published trials aiming to identify patient subsets that do not benefit from treatment should also be utilized to further refine the clinical population.
PARTNER I Trials
The PARTNER I trials were the first randomized trials to assess the efficacy and safety of TAVR with the Sapien device (Edwards Lifesciences) in high-risk6 and extreme-surgical-risk4 patients with native valve stenosis. Patients were required to have a Society of Thoracic Surgeons Predicted Risk of Mortality (STS PROM) score of ≥ 10% or have comorbidities leading two independent surgeons to estimate a postoperative risk of death at 15% to 30%. The mean STS PROM score in the PARTNER I high-risk cohort undergoing TAVR was 11.8% ± 3.3%, whereas the STS PROM score in the PARTNER I extreme-risk cohort was 11.2% ± 5.8%.
Only patients with aortic stenosis affecting a tricuspid native valve were included, with exclusion criteria including bicuspid valve morphology, significant concomitant valve dysfunction, left ventricular ejection fraction < 20%, recent stroke, and severe renal impairment, resulting in a highly selected group with only 34% of site-screened patients accepted for randomization by the study’s patient review committee.
The trial results were impressive with observed rates of mortality lower than predicted by the STS PROM at 30 days: 3.4% in the high-risk cohort and 5% in the extreme-risk cohort. In this highly selected cohort, the rates of major complications such as stroke, pacing, and vascular injury were also acceptable. At 5 years however, only 32.2%8 of the high-risk TAVR cohort and 28.2%9of the extreme-risk TAVR cohort remained alive. With an estimated procedural cost of > $70,000,10there is room to further improve patient selection.
CoreValve US IDE Trials
The CoreValve US Investigational Device Exemption (IDE) trials are contemporary studies assessing the efficacy of TAVR using the CoreValve prosthesis (Medtronic, Inc.) in high-risk7and extreme-risk5patients. The trial enrolled 394 people into a high-risk TAVR cohort (mean STS PROM, 7.3% ± 3%) and 489 people into the extreme-risk cohort (mean STS PROM, 10.3% ± 5.5%). The lower STS PROM score in the CoreValve US IDE trials may, in part, be due to the inclusion of the frailty assessment in combination with STS PROM in patient selection. Similar to the PARTNER trial, included patients were highly selected with 20% of patients presented to the patient review committee not reaching randomization; only patients with severe native trileaflet aortic stenosis were included. Patients with severe concomitant valve dysfunction, recent stroke, left ventricular ejection fraction < 20%, or severe renal dysfunction were excluded.
At 30 days, the observed mortality in the high-risk cohort was 3.3% and 8.4% in the extreme-risk cohort. At 12 months, the observed rates were 14.2% and 24.3%, respectively, with superiority over surgical valve replacement demonstrated in the highly selected, high-risk cohort.
The CoreValve US IDE and PARTNER trial 1-year results differed in a number of important areas, such as stroke rates and all-cause mortality. Although the CoreValve US IDE trial achieved superiority of the TAVR arm compared to surgical AVR, while the PARTNER IA trial achieved noninferiority, direct comparison between these trials is difficult and potentially misleading. As previously mentioned, the mean STS PROM scores were significantly lower in the CoreValve US IDE trials, particularly in the high-risk cohort. The CoreValve trial commenced enrolling patients nearly 4 years after the PARTNER trials started and a year after the PARTNER trial results were released. It is possible that lessons learned from the PARTNER trials led to improvements in patient selection, awareness of the importance of minimizing paravalvular leak, and attention to access site management, all contributing to the outstanding results seen in the CoreValve US IDE trial.
Nonrandomized Next-Generation Device Clinical Trials
A number of newer-generation TAVR devices are entering trial and clinical use. With design features purported to reduce procedural complications and improve outcomes, most have been studied in single-arm trials with comparative US IDE trials currently enrolling. These trials have included highly selected cohorts with inclusion and exclusion criteria similar to the PARTNER and CoreValve US IDE trials, although with a trend to lower reported STS PROM scores (Figure 1).
The Lotus Valve system (Boston Scientific Corporation) was studied in the single-arm REPRISE suite of trials. The REPRISE II trial enrolled 120 patients with a mean STS PROM score of 7.1%.11 The Direct Flow prosthesis (Direct Flow Medical) was studied in the DISCOVER trial12that enrolled 100 patients with a mean STS PROM score of 9.7%. The CoreValve Evolut R device (Medtronic, Inc.), an iteration of the CoreValve was studied in a first-in-human CE Mark trial of 100 patients, with a mean STS PROM score of 7%.13
These trials have seen a gradual trend to include patients who are deemed high-risk, despite a lower average STS PROM score. This is due to assessment of a number of additional variables, such as STS Plus risk factors, frailty indices (ie, hand grip, gait speed, Katz Index), and the Charlson Comorbidity Index when selecting appropriate patients. This trend reflects an acknowledgement that these additional factors have been shown in analyses of previous study and registry cohorts to offer additional predictive benefit over surgical risk scores alone.
COHORTS TREATED IN CLINICAL REGISTRIES
While the majority of clinical trials published to date have focused on TAVR for treatment of native trileaflet aortic stenosis in elderly high- and extreme-surgical-risk patients, the large number of clinical registries have provided evidence for treatment of a number of additional cohorts, often without the strict inclusion and exclusion criteria of clinical trials. Registries offer a view of real-world practice.
The creep of STS PROM scores toward a lower-risk cohort, seen in contemporary clinical trials, has been even more pronounced in published registries (Figure 1). The lower STS PROM scores seen in these registries may be partially due to the inclusion of other metrics in risk assessment, which are not included in surgical risk scores. Retrospective analyses of large clinical trials and registries have shown that a number of other unmeasured factors are as important, or potentially of greater predictive value, as surgical risk scores alone. These measures include frailty indices, a broader range of comorbid conditions, such as liver, renal, and cognitive dysfunction, together with a detailed anatomical assessment and awareness of how anatomic variables may increase procedural complexity. Incorporating all of these variables may identify patients who, despite low or intermediate STS PROM scores, are better served by a less invasive treatment modality or identify patients with modest or high scores, but significant other variables that make even TAVR a potentially futile procedure. Together, treatment of appropriate patients with low surgical risk scores and exclusion of futile patients with higher surgical risk scores will lead to a justifiable lowering of the overall mean surgical risk score.
In jurisdictions in which TAVR has entered routine clinical practice, patient and physician preference for less invasive treatment may also contribute to inclusion of a lower-risk cohort. This is particularly pronounced in countries where data reporting and heart team involvement is not mandated or linked to funding.
In the United States, where funding is linked to reporting data to the TVT Registry, the mean STS PROM score has reduced slightly from 7.1% in 2012 to 6.7% in 2014,14whereas in France, a temporal reduction in mean surgical risk score from 14.4% in the FRANCE II Registry to 11.5% in the FRANCE TAVI Registry has also been seen. Even within the FRANCE TAVI Registry, the mean STS PROM score has decreased from 12.5% in 2013 to 10.7% in 2014.15A trend to the reduction of surgical risk scores is, however, not universally seen. In the UK TAVI registry analysis of 3,980 patients treated between 2007 and 2012, no variation in the surgical risk score was shown over time, as assessed by the logistical EuroSCORE.16
The evidence for efficacy of TAVI in the treatment of native aortic valve regurgitation is largely limited to registry data. Coexisting aortoventricular interface dilatation and lack of aortic valve calcification may lead to difficulty in achieving prosthesis stability and is likely to have contributed to the relatively high frequency of ectopic valve deployment and the need for valve-in-valve therapy, as seen in the few published cohorts.17
The JenaValve (JenaValve Technology), which uses a clipping mechanism to grasp the valve leaflets and overcome the difficulty of anchoring a transcatheter prosthesis in a noncalcified annulus, has demonstrated some efficacy in native valve regurgitation and has obtained CE Mark approval for the treatment of aortic stenosis and regurgitation. While the clipping mechanism has distinct advantages in noncalcified valves, the ability to clip heavily calcified leaflets, particularly with protuberant nodules, may be more difficult. The clip, similar to a paperclip, can function even if the clip components are held apart by a significant amount of tissue. While this could conceivably increase the risk of paravalvular regurgitation, the ongoing JUPITER postmarket registry will provide a larger sample size to further assess safety and efficacy in patients with aortic regurgitation and aortic stenosis.
Bioprosthetic Valve Dysfunction
Although native valve regurgitation is associated with intrinsic difficulties for transcatheter treatment, the presence of rigid sewing rings and frames on bioprostheses offers a platform for transcatheter prosthesis anchoring. The Valve-in-Valve International Data registry was established to assess the safety and efficacy of TAVR prostheses for the treatment of bioprosthetic valve dysfunction.
Reported surgical bioprosthesis failure rates vary from 10% to 30% at 10 years and 30% to 60% at 15 years, with durability affected by variables such as age at implantation, type of prosthesis, and comorbidities. Valve failure can be due to stenosis from calcific degeneration, pannus formation, or subclinical leaflet thrombosis, or due to regurgitation from leaflet wear or infection, or a combination of stenosis and regurgitation. Traditionally, valve failure was treated by repeat surgical valve replacement though increasing age at the time of valve failure; additional comorbidities and the complexities of repeat sternotomy may result in prohibitively high morbidity and mortality. In this population, TAVR has provided an alternate treatment modality for patients with either stenosis or regurgitation.
In 459 patients treated at 55 centers, the 1-year survival after valve-in-valve procedures was 83.2%, with baseline predictors of mortality including stenosis as the mode of bioprosthesis failure and small initial bioprosthesis size.18These data add weight to the implantation of surgical bioprostheses rather than mechanical prostheses in a slightly younger cohort than that traditionally selected, with the understanding that should the bioprosthesis fail, TAVR is a safe and effective second procedure.
TRIAL EVIDENCE FOR TREATMENT OF LOWER-RISK PATIENTS
The PARTNER II Sapien 3 intermediate-risk cohort enrolled 1,076 patients with a mean STS PROM score of 5.3%. This was an elderly cohort with a mean age of 81.9 years. At 30 days, there was a low 1.1% rate of mortality and a 2.6% incidence of stroke.19Longer-term follow-up has not yet been reported, and comparative safety and efficacy outcomes with surgical AVR (SAVR) remain to be studied. The PARTNER IIA trial will randomize a similar intermediate-risk cohort (STS PROM ≥ 4%) to receive treatment by TAVR with the Sapien XT prosthesis (Edwards Lifesciences) or SAVR. The results of this trial will provide randomized data regarding the safety and efficacy of TAVR in a lower-risk cohort.
The SURTAVI clinical trial similarly aims to determine the safety and efficacy of the CoreValve device compared to SAVR in patients with an estimated 30-day periprocedural mortality of 3% to 15%. Recruitment was initially slow, which was at least partially attributed to patient and physician preference for treatment by TAVR in the nontrial clinical setting, rather than potential randomization to surgery if they were to enter the study.
The NOTION trial20 randomized an all-comer population of 280 patients aged > 70 years to treatment by TAVR using the CoreValve device versus SAVR. The mean STS PROM score was 2.9% in the TAVR arm and 3.1% in the SAVR arm. The primary outcome measure of all-cause death, stroke, or myocardial infarction at 12 months was not significantly lower in the TAVR arm (13.1% vs 16.3%; P = .43). There were, however, significantly lower rates of major bleeding, stage II/III kidney injury, and postprocedural atrial fibrillation, but higher rates of new pacemaker implantation. Echocardiographic data demonstrated a statistically, although potentially not clinically, higher effective orifice area and lower mean transprosthetic gradient after TAVR.
What remains elusive from trial evidence is long-term follow-up and demonstration of equivalent treatment durability. While the PARTNER I trial has reported 5-year results, the inclusion of elderly high-risk patients has resulted in nearly 70% mortality8and hence an insufficient cohort to prove treatment longevity. Evidence is still a number of years off and will come from ongoing follow-up of large registries, which included treatment of younger and lower-risk patients and even more distant results from lower-risk trials such as NOTION, SURTAVI, and PARTNER II.
LIMITATIONS OF SURGICAL RISK SCORES
With the frequent categorization of patients into low-, intermediate-, high-, and extreme-procedural-risk groups, it would seem an easy feat to identify those best served by each treatment modality; however, surgical risk scores have a number of limitations. The STS and EuroSCORE were designed to predict perioperative mortality and were formulated from, and validated in, surgical cohorts. There is no weight given within these scoring systems to frailty or many comorbid conditions, which are of increasing frequency in an aging cohort.
The use of surgical risk scores to select patients also tends to rely heavily on the potential for periprocedural mortality to select patients. However, for many elderly patients, mortality may not be as important a complication as morbidity and the potential impact on quality of life.
Acknowledging the limitations of surgical risk scores, frailty has emerged as a strong predictor of procedural risk and is increasingly incorporated into patient selection. Frailty is the risk of significant decline a person is exposed to due to declining health and physical reserve, often seen in and associated with aging. Because frailty is a syndrome, diagnostic criteria are varied and not universally agreed upon. A number of metrics have, however, been attempted to quantitate frailty; these metrics include 5-m gait speed, hand-grip strength, physical activity questionnaires, exhaustion questionnaires, self-reported weight loss, serum albumin levels, and activities of daily living dependency.
Although frailty may be an indication for percutaneous, rather than surgical, valve replacement, it is also a predictor of morbidity and mortality and hence, a discriminator for those who are at too high risk even for TAVR. This appears to hold true regardless of the means of frailty assessment. The ABC study demonstrated that slow 5-m gait speed led to a two- to threefold increase in the rate of mortality or major morbidity after surgery regardless of baseline STS PROM score.21Similarly, a substudy from the PARTNER cohort demonstrated that elevated frailty assessed by albumin, gait speed, grip strength, and activities of daily living dependency was independently predictive of mortality.22Even when frailty is not formally assessed by using any of these suggested metrics, a subjective ‘end-of-bed’ assessment of frailty by the treating clinician has proven to be an independent predictor of late mortality after TAVR.23
The frailty assessment, together with the burden of comorbidities, must be used to identify patients who are at too high risk for surgical valve replacement, but also to identify patients who are also at too high risk for TAVR. Essentially, patients who are dying from aortic stenosis and who would benefit from treatment must be differentiated from those who are dying from other comorbidities with aortic stenosis.
Given the complexity involved in assessing all domains that contribute to a patient’s suitability for TAVR, the involvement of a heart team is essential. The use of a heart team is recommended in both European24and American25guidelines and, in many countries, it is linked to the funding of the procedure. However, what constitutes a heart team is varied, with many heart teams made up of only the treating physician and surgeon. We would argue that the more diverse the team, the more successful it will be in selecting appropriate patients and, potentially more importantly, excluding patients who are best served by medical therapy, surgery, treatment of other comorbidities, or even palliative management.
WHO IS A TAVR CANDIDATE IN 2016?
In 2016, TAVR has proven efficacy in elderly high- and extreme-risk patients with symptomatic severe native trileaflet valve aortic stenosis. The treatment of patients who have lower STS PROM scores, but who are deemed to be at high risk by a heart team due to comorbidities and/or frailty, is also appropriate and supported by registry and increasing trial evidence. The treatment of patients with bicuspid aortic stenosis, predominant aortic regurgitation, or bioprosthesis dysfunction when they have appropriate aortoventricular interface anatomy, and are deemed to be at high risk for surgical intervention by a heart team is also appropriate, although supported by a lower level of registry evidence. The treatment of patients with these conditions should be performed by experienced operators and centers where specific preprocedural assessments, devices, and procedural techniques are required. For example, device sizing in bicuspid anatomy requires assessment of metrics, such as intercommissural distance, while specific TAVR devices with annular sealing mechanisms may be more appropriate in pure aortic regurgitation.
Treatment of “intermediate-risk” elderly patients is likely safe and effective, although the long-term durability of results cannot be ensured at present. This is of greater concern in a population with a longer life expectancy. In 2016, we would recommend that these patients, if treated by TAVR, are treated in a clinical trial or properly reported registry framework in order to build the required evidence base to prove safety, efficacy, and longevity in these cohorts.
Perhaps more importantly, in 2016, we must do more to identify who is not a candidate for TAVR. Within the PARTNER I trial, at 5 years, more than 70% of the cohort was deceased. While TAVR may have allowed a number of these patients to enjoy improved quality of life, a significant number of people in contemporary trials do not report quantitative or qualitative improvement in function or quality of life. Until a well-validated, TAVR-specific mortality and morbidity score is created, it remains essential that an experienced heart team is available to make a comprehensive patient assessment of benefit versus futility.
1. Cribier A, Eltchaninoff H, Bash A, et al. Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description. Circulation. 2002;106:3006-3008.
2. Nkomo VT, Gardin JM, Skelton TN, et al. Burden of valvular heart diseases: a population-based study. Lancet. 2006;368:1005-1011.
3. Iung B, Cachier A, Baron G, et al. Decision-making in elderly patients with severe aortic stenosis: why are so many denied surgery? Eur Heart J. 2005;26:2714-2720.
4. Leon MB, Smith CR, Mack M, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010;363:1597-1607.
5. Popma JJ, Adams DH, Reardon MJ, et al. Transcatheter aortic valve replacement using a self-expanding bioprosthesis in patients with severe aortic stenosis at extreme risk for surgery. J Am Coll Cardiol. 2014;63:1972-1981.
6. Smith CR, Leon MB, Mack MJ, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011;364:2187-2198.
7. Adams DH, Popma JJ, Reardon MJ. Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med. 2014;371:967-968.
8. Mack MJ, Leon MB, Smith CR, et al. 5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER I): a randomized controlled trial. Lancet. 2015;385:2477-2484.
9. Kapadia SR, Leon MB, Makkar RR, et al. 5-year outcomes of transcatheter aortic valve replacement compared with standard treatment for patients with inoperable aortic stenosis (PARTNER I): a randomized controlled trial. Lancet. 2015;385:2485-2491.
10. Reynolds MR, Magnuson EA, Lei Y, et al. Cost-effectiveness of transcatheter aortic valve replacement compared with surgical aortic valve replacement in high-risk patients with severe aortic stenosis: results of the PARTNER (Placement of Aortic Transcatheter Valves) trial (Cohort A). J Am Coll Cardiol. 2012;60:2683-2692.
11. Meredith I. Transcatheter aortic valve replacement for severe symptomatic aortic stenosis using a repositionable valve system. J Am Coll Cardiol. 2014;64:1339-1348.
12. Lefevre T, Colombo A, Tchetche D, et al. Prospective multicenter evaluation of the direct flow medical transcatheter aortic valve system: 12-month outcomes of the evaluation of the direct flow medical percutaneous aortic valve 18F system for the treatment of patients with severe aortic stenosis (DISCOVER) study. JACC Cardiovasc Interv. 2016;9:68-75.
13. Manoharan G, Walton AS, Brecker SJ, et al. Treatment of symptomatic severe aortic stenosis with a novel resheathable supra-annular self-expanding transcatheter aortic valve system. JACC Cardiovasc Interv. 2015;8:1359-1367.
14. Holmes DR Jr, Nishimura RA, Grover FL, et al. Annual outcomes with transcatheter valve therapy: from the STS/ACC TVT registry. J Am Coll Cardiol. 2015;66:2813-2823.
15. Auffret V, Bedossa M, Boulmier D, et al. From FRANCE 2 to FRANCE TAVI: are indications, technique and results of transcatheter aortic valve replacement the same? Presse Med. 2015;44:752-760.
16. Ludman PF, Moat N, de Belder MA, et al. Transcatheter aortic valve implantation in the United Kingdom: temporal trends, predictors of outcome, and 6-year follow-up: a report from the UK Transcatheter Aortic Valve Implantation (TAVI) registry, 2007 to 2012. Circulation. 2015;131:1181-1190.
17. Roy DA, Schaefer U, Guetta V, et al. Transcatheter aortic valve implantation for pure severe native aortic valve regurgitation. J Am Coll Cardiol. 2013;61:1577-1584.
18. Dvir D, Webb JG, Bleiziffer S, et al. Transcatheter aortic valve implantation in failed bioprosthetic surgical valves. JAMA. 2014;312:162-170.
19. Kodali S. Clinical and echocardiographic outcomes at 30 days with the SAPIEN 3 TAVR system in inoperable, high-risk and intermediate-risk AS patients. Presented at: ACC 2015; March 15, 2015; San Diego.
20. Thyregod HG, Steinbruchel DA, Ihlemann N, et al. Transcatheter versus surgical aortic valve replacement in patients with severe aortic valve stenosis: 1-year results from the all-comers NOTION randomized clinical trial. J Am Coll Cardiol. 2015;65:2184-2194.
21. Afilalo J, Eisenberg MJ, Morin JF, et al. Gait speed as an incremental predictor of mortality and major morbidity in elderly patients undergoing cardiac surgery. J Am Coll Cardiol. 2010;56:1668-1676.
22. Green P, Arnold SV, Cohen DJ, et al. Relation of frailty to outcomes after transcatheter aortic valve replacement (from the PARTNER trial). Am J Cardiol. 2015;116:264-269.
23. Rodes-Cabau J, Webb JG, Cheung A, et al. Long-term outcomes after transcatheter aortic valve implantation: insights on prognostic factors and valve durability from the Canadian multicenter experience. J Am Coll Cardiol. 2012;60:1864-1875.
24. Joint Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology (ESC), European Association for Cardio-Thoracic Surgery (EACTS), Vahanian A, et al. Guidelines on the management of valvular heart disease (version 2012). Eur Heart J. 2012;33:2451-2496.
25. Nishimura RA, Otto CM, Bonow RO, et al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2014;63:2438-2488.
Robert P. Gooley, MBBS (Hons)
MonashHeart, Monash Health
Southern Clinical School, Monash University
Disclosures: Consultant for Boston Scientific Corporation.
Ian T. Meredith, MBBS (Hons), PhD
MonashHeart, Monash Health
Southern Clinical School, Monash University
Disclosures: Consultant for Boston Scientific Corporation; serves on the Strategic Advisory Board for Boston Scientific Corporation and Medtronic, Inc.