Blue Cross and Blue Shield of Montana (BCBSMT) may consider transcatheter aortic valve replacement, performed via the transfemoral approach, medically necessary for patients with aortic stenosis when all of the following conditions are present:
- Severe aortic stenosis with a calcified aortic annulus; AND
- New York Heart Association (NYHA) heart failure Class II, III or IV symptoms; AND
- Patient is not an operable candidate for open surgery, as judged by at least two cardiovascular specialists (cardiologist and cardiac surgeon).
BCBSMT considers transcatheter aortic valve replacement experimental, investigational and unproven for all other indications, including but not limited to, patients at high risk for open surgery but who are operable candidates.
Transcatheter aortic valve replacement performed via approaches other than transfemoral, including but not limited to the transapical approach for any indication is considered experimental, investigational and unproven.
The evidence on transcatheter aortic valve implantation (TAVI) consists of many uncontrolled case series and one pivotal randomized controlled trial (RCT) - the PARTNER trial. These studies report on two potential populations for TAVI: 1) patients who are not surgical candidates, and 2) patients who are high risk for surgery but still considered to be surgical candidates. The evidence on these two groups of patients will be discussed separately.
Does transcatheter aortic-valve implantation improve outcomes for patients with aortic stenosis who are not suitable candidates for open surgery?
Many case series of TAVI have been published in the last 10 years, the majority of which have included patients who are not candidates for open surgery. However, the selection process for TAVI has largely been subjective, with the expert opinion of the surgeons and/or cardiologists as the main factor determining suitability for open surgery. As a result, there may be some overlap in these series with patients who are surgical candidates, but the distinction cannot be easily made from the reported studies.
An Agency for Healthcare Research and Quality (AHRQ)-sponsored review in 2010 reviewed 84 publications enrolling 2,375 patients. (2) Implantation was successful in 94% of patients overall, with higher success rates reported in more recent publications. The aggregate 30-day survival was 89% across all studies. Adverse event rates were reported in the larger case series, with an estimated 30-day rate of major cardiovascular adverse events and stroke of 8%.
The two largest series included in the AHRQ review reported on 646 patients treated with the Medtronic CoreValve™ (7) and 339 patients treated with the Edwards SAPIEN™ valve. (8) The CoreValve™ study by Piazza et al. study was notable in that it used more objective patient selection criteria than is common in this literature. Their criteria for eligibility included the following: 1) Logistic EuroScore ≥15%; 2) age ≥75 or; 3) age ≥65 with liver cirrhosis, pulmonary insufficiency, pulmonary hypertension, previous cardiac surgery, porcelain aorta, recurrent pulmonary emboli, right ventricular insufficiency, previous chest burns or radiation precluding open surgery, or body mass index (BMI) ≤18kg/m2. Procedural success was 97% and 30-day survival was 92%. The 30-day combined rate of death, myocardial infarction (MI) or stroke was 9.3%. The study by Rodes-Cabou et al. was performed in Canada and used Edwards SAPIEN™ valve. This study had subjective inclusion criteria, relying on the judgment of the participating surgeons to determine eligibility for TAVI. The procedural success rate was 93.3% and the 30-day mortality was 10.4%. The authors also reported a mortality rate of 22.1% at a median follow-up of 8 months.
A second systematic review was published in 2011 by Figulla et al. (9) This review included studies that enrolled symptomatic patients with severe aortic stenosis, had a mean age of 75 years or older, reported on 10 or more patients, and had a follow-up duration of 12 months or more. A total of 12 studies met these criteria and were compared to a group of 11 studies that treated severe aortic stenosis with nonsurgical therapy. The procedural success in these studies ranged from 86-100%, and the 30-day mortality ranged from 5.3-23%. The combined mean survival rate at 1 year was 75.9% (95% confidence interval [CI] 73.3-78.4). This 1-year survival rate compared favorably to medical therapy, which was estimated to be 62.4% (95% CI 59.3-65.5).
Since these systematic reviews, there continue to be numerous case series of TAVI published in the literature. The largest of these was a case series from Germany reporting on 697 patients treated with the CoreValve™ system. (10) Procedural success was 98.4% and 30-day mortality was 12.4%. Another large case series from Italy (11) included 663 patients treated with the CoreValve™ device. Procedural success was 98% and mortality at 1 year was 15%. A notable study was published by Gurvitch et al. in 2011 (12) that reported on durability and longer clinical outcomes up to 3 years. Seventy patients who underwent TAVI and survived for greater than 30 days were included. Survival at 1, 2, and 3 years was 81%, 74%, and 61%, respectively. One patient (1.5%) required reoperation during this time period. The valve area decreased from 1.7 ± 0.4 cm² following the procedure to 1.4±0.3 cm² at 3 years. Aortic incompetence was trivial or mild in 84% of patients and did not worsen over time.
The PARTNER trial is a pivotal multicenter RCT of TAVI performed in the U.S., Canada, and Germany, using the SAPIEN™ heart-valve system. Leon et al. 2010 (13) reported results of the cohort of patients from the PARTNER trial with severe aortic stenosis who were not candidates for open surgery. In order to be classified as unsuitable for open surgery, patients had to have a predicted probability of ≥50% for death or a serious irreversible condition at 30 days post-surgery. This probability was determined by two surgeon investigators using clinical judgment and the Society of Thoracic Surgery (STS) risk score. The executive committee of the PARTNER trial reviewed all patient selection decisions and approved the classification of patients as unsuitable for surgery. A total of 3,105 patients were screened for aortic-valve surgery, and 12% of these were eventually included in the cohort of patients deemed unsuitable for surgery.
A total of 358 patients were randomized to TAVI or usual care. TAVI was performed by the transfemoral approach under general anesthesia. Standard therapy was determined by the treating clinicians. In most cases (83.8%), standard treatment included balloon valvuloplasty of the aortic valve. A small number of patients (6.7%) underwent open surgical valve replacement despite the high risk, and another 2.2% of patients underwent TAVI at a center outside the U.S. that was not participating in the trial. The primary outcome was death from any cause over the course of the trial (median follow-up 1.6 years). A “coprimary” endpoint was the composite of time to death from any cause or time to repeat hospitalization related to aortic stenosis or TAVI. Secondary endpoints were cardiovascular mortality, New York Heart Association (NYHA) functional class, the rate of hospitalizations due to aortic stenosis or TAVI, the 6-minute walk test, valve performance as measured by echocardiography, and procedural complications (myocardial infarction [MI], stroke, acute kidney injury, vascular complications, and bleeding).
The mean age of enrolled patients was 83.2 years. There were some baseline imbalances in the patient population indicating that the standard therapy group may have had a higher severity of illness. Standardized scores of surgical risk were higher in the standard therapy group. The Logistic EuroSCORE was significantly higher in the standard therapy group compared to the TAVI group (30.4±19.1 vs. 26.4±17.2, p=0.04) and the Society of Thoracic Surgery (STS) score was numerically higher but did not reach statistical significance (12.1±6.1 vs. 11.2±5.8, p=0.14). Significantly more patients in the standard therapy group had chronic obstructive pulmonary disease (COPD) (52.5% vs. 41.3, p=0.04) and atrial fibrillation (48.8% vs. 32.9%, p=0.04), and there was a nonsignificant trend for more patients in the standard therapy group having a lower ejection fraction (51.1 vs. 53.9%) and frailty, as determined by prespecified criteria (28.0 vs. 18.1%).
Death from any cause at 1 year following enrollment was lower for the TAVI group (30.7% vs. 49.7%, p<0.001). This represents a 19% absolute risk reduction, a 38.2% relative risk reduction, and a number needed to treat of 5.3 to prevent one death over a 1-year follow-up. Most secondary outcomes also favored the TAVI group. Cardiovascular death was lower in the TAVI group (19.6% vs. 44.1%, p<0.001). The composite of all-cause mortality and repeat hospitalizations was reached by 42.5% of the patients in the TAVI group compared with 70.4% in the standard therapy group. Symptoms and functional status were also superior in the TAVI group. The percent of patients in NYHA Class I or II at 1 year was higher for the TAVI group (74.8% vs. 42.0%, p<0.001), and there was a significant improvement in the 6-minute walk test for the TAVI group but not for the standard therapy group (between group comparisons not reported). Subgroup analysis did not report any significant differences in outcomes according to clinical and demographic factors.
Complication rates were higher for the TAVI group. Stroke or transient ischemic attack (TIA) at 1 year was more than twice as frequent for the TAVI group (10.6% vs. 4.5%, p=0.04). Major bleeding and vascular complications occurred in a substantial percent of patients undergoing TAVI, and were significantly higher than in the standard therapy group (22.3% vs. 11.2%, p=0.007; and 32.4% vs. 7.3%, p<0.001 respectively).
Discussion Numerous case series have demonstrated feasibility and short-term efficacy for TAVI in patients who are not surgical candidates. In the PARTNER B trial, there was a large decrease in all-cause mortality and cardiovascular mortality at 1 year for TAVI compared to standard therapy. Baseline group differences were present, indicating that the TAVI group may have been healthier. While these differences are unlikely to account for the degree of mortality benefit reported, they may have resulted in an overestimation of the mortality benefit. Longer term follow-up would be needed to determine whether this mortality benefit is maintained over time, given that the patient population was very elderly with a high burden of comorbidities. Also, longer follow-up is needed to determine the durability of the valve implanted by TAVI. The benefit in mortality was accompanied by an increased stroke risk, as well as substantial increases in vascular complications and major bleeding. There is also uncertainty concerning the generalizability of these results, since patient selection was primarily determined by the judgment of the cardiovascular surgeons and/or cardiologists. It is not known whether this type of decision making by surgeons and cardiologists is reliable across the range of practicing clinicians.
Does transcatheter aortic-valve implantation improve outcomes for high-risk patients with aortic stenosis as an alternative to open surgery?
Results from the cohort of patients in the PARTNER trial who were high risk for open surgery, but still suitable candidates, were published in June 2011. (14) The inclusion and exclusion criteria were generally the same as for the prior cohort, except that these patients were classified as high risk for surgery rather than unsuitable for surgery. For high risk, patients had to have a predicted perioperative mortality of ≥15%, as determined by a cardiac surgeon and cardiologist using clinical judgment. An STS score of ≥10 was included as a guide for high-risk, but an STS score threshold was not a required criterion for enrollment. The executive committee of the PARTNER trial reviewed all patient selection decisions and approved the classification of patients as high risk for surgery. A total of 3,105 patients were screened for aortic valve surgery, and 22.5% of these were eventually included in the cohort of patients deemed high-risk for surgery.
A total of 699 patients were randomized to TAVI or usual care. The primary hypothesis was that TAVI was non-inferior to open aortic valve replacement (AVR), using a one-sided non-inferiority boundary of 7.5% absolute difference in mortality at 1 year. TAVI was performed under general anesthesia using the transfemoral approach if possible (n=492). If the transfemoral approach was not possible, transapical approach was used (n=207). The comparison group underwent open AVR. Details of the open procedure were not provided in presentation slides.
The primary outcome was death from any cause at 1-year follow-up. A second powered endpoint was non-inferiority at 1 year for the patients undergoing TAVI by the transfemoral approach. Secondary endpoints were cardiovascular mortality, NYHA functional class, rehospitalizations, the 6-minute walk test, valve performance as measured by echocardiography, and procedural complications (MI, stroke, acute kidney injury, vascular complications, and bleeding). The mean age of enrolled patients was 83.6 years in the TAVI group and 84.5 years in the open AVR group. Other baseline demographics and clinical characteristics were generally well-balanced, except for a trend toward an increased percent of patients in the TAVI group with a creatinine level >2.0 (11.1% vs. 7.0%, p=0.06).
Death from any cause at 1 year following enrollment was 24.2% for the TAVI group compared to 26.8% for the open AVR group (p=0.44 for difference between groups). The upper limit of the 95% confidence interval (CI) for the difference between groups was a 3.0% excess mortality in the TAVI group, which was well within the non-inferiority boundary of 7.5%. Thus the criterion of non-inferiority was met, with a p value of 0.001. For the subgroup of patients who underwent TAVI by the transfemoral approach, results were similar with 22.2% mortality in the TAVI group compared with 26.4% mortality in the open AVR group (p=0.002 for non-inferiority). The secondary outcomes of cardiovascular mortality (14.3% vs. 13.0%, p=0.63) and rehospitalizations (18.2% vs. 15.5%, p=0.38) were not significantly different for the TAVI versus open AVR groups. The percent of patients in NYHA Class I or II at 1 year was similar between groups at 1 year, as was the improvement in the 6-minute walk test. On subgroup analysis, there was a significant effect for gender, with women deriving greater benefit than men (p=0.045), and a significant effect for prior coronary artery bypass graft (CABG), with patients who had not had prior CABG deriving greater benefit in the TAVI group.
Certain complication rates showed significant differences between groups. Stroke or TIA at 1 year was higher for the TAVI group (8.3% vs. 4.3%, p=0.04). Vascular complications occurred in 18.0% percent of patients undergoing TAVI, compared with 4.8% in the open AVR group (p=0.01), and major vascular complications were also higher in the TAVI group (11.3% vs. 3.5%, p=0.01). On the other hand, major bleeding was more common in the open group compared to TAVI (25.7% vs. 14.7%, p=0.01).
Discussion. This trial reports equivalence between TAVI and open AVR in terms of mortality at 1 year and most major secondary outcomes. The non-inferiority boundaries for this trial included an upper limit of 7.5% absolute increase in mortality, but in actuality, the reported mortality for the TAVI group was lower than for the open group, although not significantly different. Stroke or TIA was significantly more common for the TAVI group, occurring at a rate of almost 2 times that reported for open surgery. Other secondary outcomes was similar between groups, except for higher rates of vascular complications in the TAVI group and higher rates of major bleeding in the open surgery group. As in the first PARTNER cohort, there is concern for generalizability of results given that the patient selection process relied largely on the judgment of surgeons and cardiologists participating in the trial.
Does transcatheter aortic-valve implantation by alternative approaches (e.g. transapical or transaxillary) achieve similar outcomes to those reported from the transfemoral approach?
The majority of all patients treated with TAVI, and all the patients enrolled in the PARTNER B trial, have been by the transfemoral approach. Other approaches, such as the transapical approach, have been used in patients with inadequate femoral access. There is a limited amount of evidence comparing outcomes from different approaches. In the PARTNER A trial, slightly less than one-third of procedures were performed by the transapical approach, (13) and there were no substantial differences in outcomes between the 2 approaches. Ewe et al. (15) compared outcomes from transfemoral (n=47) versus transapical TAVI (n=60) in 107 consecutive patients undergoing TAVI. There were marginally less vascular complications in the transapical group compared to the transfemoral group (185 vs. 5%, p=0.05). Otherwise, there were no differences in outcomes between the 2 approaches.
Other studies have reported higher complication rates with the transapical approach. Thomas et al. (16) used data from a European registry to compare patients undergoing TAVI by the transfemoral (n=463) with patients undergoing TAVI by the transapical approach (n=575). Complications were more frequent in the transapical group, but the transapical group also may have been more severely ill as judged by a higher Eurogual risk score. A publication from the UK TAVI registry (17) evaluated risk factors for adverse outcomes in 877 TAVI procedures. On univariate analysis, TAVI by the transapical approach was associated with lower survival, although this relationship did not persist after controlling for demographic and clinical factors.
TAVI addresses an unmet need for patients with severe aortic stenosis who require intervention, but who are a high or prohibitive risk for open surgery. The Sapien Total Heart Valve System received FDA-approval in November 2011 for patients who are not operable candidates.
For patients who are not surgical candidates due to excessive surgical risk, the PARTNER B trial reported results for patients treated with TAVI by the transfemoral approach compared to continued medical care with or without balloon valvuloplasty. There was a large decrease in mortality for the TAVI patients at 1 year compared to medical care. This trial also reported improvements on other relevant clinical outcomes for the TAVI group. There was an increased risk of stroke and vascular complications in the TAVI group. Despite these concerns, the overall balance of benefits and risks from this trial indicate that health outcomes are improved. Therefore TAVI may be considered medically necessary for patients with severe aortic stenosis who are not surgical candidates.
For patients who are high risk for open surgery, but are operable candidates, the PARTNER A trial reported non-inferiority for survival at 1 year compared to open surgery. In this trial, TAVI patients also had higher risks for stroke and vascular complications. The results of this trial do not establish that TAVI is superior to open surgery, and the complication rate may be higher with TAVI. Therefore, TAVI for patients at high risk for surgery but who are surgical candidates, is considered experimental, investigational and unproven.
Ongoing Clinical Trials
A search of online site ClinicalTrials.gov returned 17 ongoing trials of TAVI in various stages of evolution. The majority of these are single-arm trials evaluating the safety and efficacy of TAVI, using various types of valves, delivery systems, ancillary treatments, and outcomes. Five RCTs were identified with the following characteristics:
- NCT01057173. Transcatheter compared to surgical valve implantation in patients with severe aortic stenosis. This is an RCT underway in Europe that is comparing TAVI with open surgical valve replacement using the Medtronic CoreValve. Estimated completion date is December 2018.
- NCT01314313. The PARTNER II Trial: Placement of Aortic Transcatheter Valves. This is an RCT underway in the U.S. that is comparing 2 types of the Edwards SAPIEN Valve system, the SAPIEN valve with RetroFlex3 and the SAPIEN XT with NovaFlex. Estimated completion date is December 2017.
- NCT00986193. Transapical transcatheter treatment versus conventional surgery in patients with native aortic valve stenosis. This is an RCT underway in Europe that is comparing TAVI using the transapical approach and the Edwards SAPIEN valve with open surgical valve replacement. Estimated completion date December 2015.
- NCT01251328. Near Infrared Spectroscopy in Aortic Valve Replacement. This is an RCT underway in Europe that is comparing the influence of these 2 anesthesiologic managements, general anesthesia and conscious sedation on the cerebral oxygen saturation during the "rapid pacing"-period during valvuloplasty. Estimated completion date not given.
- NCT01240902. Safety and Efficacy Study of the Medtronic CoreValve® System in the Treatment of Symptomatic Severe Aortic Stenosis in High Risk and Very High Risk Subjects Who Need Aortic Valve Replacement. This is an RCT underway in the U.S. that is comparing TAVI with open surgical valve repair using the Medtronic CoreValve in patients who are at high risk for open surgery. Estimated completion date November 2017.
Practice Guidelines and Position Statements
A “Professional Society Overview” on transcatheter valve therapy was published July 2011 by the American College of Cardiology Foundation and the Society of Thoracic Surgeons. (18) The purpose of this document was to enumerate the core issues that will be anticipated in integrating TAVI into general clinical care. As part of this document, a list of necessary components for the successful introduction of Transcatheter Heart Valve Therapies was developed:
- Specialized heart centers with experienced multidisciplinary physicians and paramedical personnel
- Professional multidisciplinary heart team:
- Primary cardiologists
- Cardiac surgeons
- Interventional cardiologists
- Echocardiographers and imaging specialists
- Heart failure specialists
- Proper procedure and facilities
- Modified conventional cardiac laboratory
- Hybrid operating room
- Development of and participation in clinical database and registries
- Knowledge of, and evaluation of, evidence-based medical literature concerning patient selection, procedural performance, and complication management
- Specific standardized protocols for management strategies, procedural performance, problem-solving, and complication management
The National Institute for Health and Clinical Excellence (NICE). Interventional procedure guidance on transcatheter aortic valve implantation for aortic stenosis noted that the evidence on transcatheter aortic valve implantation for aortic stenosis is limited to small numbers of patients who were considered to be at high risk for conventional cardiac surgery. It shows good short-term efficacy, but there is little evidence on long-term outcomes. There is a potential for serious complications; however, the patients on whom this procedure has been used have a poor prognosis without treatment and are at high risk if treated by open heart surgery. Clinicians wishing to use this procedure should do so only with special arrangements for clinical governance, consent and for audit or research. (19)
Disclaimer for coding information on Medical Policies
Procedure and diagnosis codes on Medical Policy documents are included only as a general reference tool for each policy. They may not be all-inclusive.
The presence or absence of procedure, service, supply, device or diagnosis codes in a Medical Policy document has no relevance for determination of benefit coverage for members or reimbursement for providers. Only the written coverage position in a medical policy should be used for such determinations.
Benefit coverage determinations based on written Medical Policy coverage positions must include review of the member’s benefit contract or Summary Plan Description (SPD) for defined coverage vs. non-coverage, benefit exclusions, and benefit limitations such as dollar or duration caps.