Transcatheter aortic valve replacement - Evidence and Indications update

Interesting links: Webinar on Valvular Heart disease, and Guidelines on Valvular Heart Disease (Management of)

Background

Symptomatic aortic stenosis amongst the elderly is associated with a dramatic increase in morbidity and mortality.(1,2) One-year mortality rates may reach up to 50% in untreated patients.(2,3,4) Nonetheless, more than 30% of patients with symptomatic severe stenosis do not undergo surgical aortic valve replacement (SAVR), mainly due to advanced age and the accumulation of concomitant comorbidities. (5) Since 2002, (6,7) transcatheter aortic valve implantation (TAVI) has been established as an emerging therapeutic approach for patients with unacceptable surgical risk and a less invasive alternate treatment option for patients at high risk for open-heart surgery. (8,9)
Meanwhile, more than 60,000 patients have undergone TAVI procedures worldwide. So far, data from two randomised controlled trials and several large multi-centre TAVI registries have built the cornerstones for our clinical decision-making process. (10,11,12,13). New technology advances promise to simplify TAVI and to improve outcome by reducing the rate of TAVI-specific complications such as peri-prosthetic aortic regurgitation, acute kidney injury, vascular complications, and conduction disturbances. (14,15,16,17,18) 

I - Update on Clinical Evidence

Registries

• UK TAVI 

The main purpose of the UK TAVI program was to define the characteristics of a real-world TAVI patient population, regardless of technology or access route, and to capture and report clinical outcomes on TAVI procedures performed within the United Kingdom. Data were collected prospectively on 870 patients until December 31, 2009. The procedure was performed with the use of the Medtronic CoreValve® (Medtronic, Minneapolis, Minnesota, USA) (52%) or the Edwards-SAPIEN THV (Edwards Lifesciences, Irvine, California, USA) (48%). The majority of the TAVI implants (69%) were performed via the transfemoral approach. 
Outcomes of TAVI patients in the UK TAVI registry at 30 days, 1 year, and 2 years were encouraging with mortality rates of 7.1%, 21.4%, and 26.3%, respectively. 

• FRANCE 2 

In the recently published FRANCE 2 registry, 3,195 patients were enrolled prospectively between January 2010 and October 2011 in 34 centers in France. Edwards SAPIEN and Med-tronic CoreValve devices were implanted in 66.9% and 33.1% of patients, respectively. The primary end point of this registry was death from any cause. Approaches were either transarterial (transfemoral, 74.6%; subclavian, 5.8%; and other, 1.8%) or transapical (17.8%). Rates of death at 30 days and 1 year were 9.7% and 24.0%, respectively. At 1 year, the incidence of stroke was 4.1%, and the incidence of peri-prosthetic aortic regurgitation was 64.5%. In a multivariate analysis, this registry showed that patients with a more-than-mild peri-prosthetic aortic regurgitation following TAVI had a 2.5-times increased 1-year mortality risk. 

• German TAVI 

The prospective, multi-centre German TAVI registry was initiated in 2009 to evaluate indication, intervention, and clinical outcome and assess quality of life after TAVI in routine clinical practice. (19) The majority of patients underwent TAVI with use of the Medtronic CoreValve (82%); the remainder were treated with the Edwards-SAPIEN THV (18%). Thirty-day and 1-year mortality rate were 8.2% and 20.2%, respectively (Table 1). (20)
 

• German Aortic Valve (GARY)

Only recently was the first data from the GARY Registry published. The GARY registry is a nationwide complete survey of patients with aortic valve stenosis undergoing invasive procedures including surgical (AVR), catheter-based (TAVI) transfemoral, catheter-based (TAVI) transapical procedures, and valvuloplasty. The aim of this unique registry initiated by cardiologists and heart surgeons together is to evaluate catheter-based procedures in comparison to surgical aortic valve replacement. In detail, the registry will allow for the development of criteria for adequate patient selection for the best treatment modality. From 01/01/2011 to 31/12/2011, 13,860 patients were included of whom 6,523 received SAVR without CABG, 3,462 SAVR with CABG, 2,694 transvascular TAVI, and 1,181 transapical TAVI. Outcome parameters are available for 1 year and show a continuous increase in mortality after hospital discharge, predominately in high-risk groups. In low and intermediate risk groups, surgical AVR without CABG seems superior to TAVI while in high-risk groups, surgical AVR and TAVI show equal outcome. Interestingly, this large cohort allowed the development of a new aortic valve score (AV score) which better predicts early outcome compared to the log Euro-Score or the STS score. Further analyses will help to improve risk stratification and to find the optimal treatment option for patients with symptomatic aortic stenosis.

• SOURCE 

The 2,307-patient SOURCE (Edwards-SAPIEN Aortic Bioprosthesis European Outcome) Regis-try was created to obtain clinical data as post-marketing registry during the early phase of commercialisation in Europe. The majority of the patients (60.1%) underwent TAVI with the first generation Edwards-SAPIEN THV via the transapical approach; all other patients were treated transfemorally. (21)
In the SOURCE registry, predominantly inoperable and high-risk patients were included, reflected by a mean logistic EuroSCORE of 23.9 ± 14.2% among transfemoral TAVI patients and of 27.6 ± 16.1% among transapical patients. One and 2-year survival in the transfemoral TAVI patients were slightly better than in the transapical TAVI patients (80.1% vs. 72.2% and 72.2% vs. 64.7%, respectively) which might be explained, at least in part, by the higher logistic EuroSCORE of the transapical patients. In multivariable analysis, the prevalence of renal insufficiency before TAVI and a high logistic EuroSCORE were the strongest independent predictors of mortality. 

• ADVANCE 

One thousand fifteen patients undergoing TAVI with use of the Medtronic CoreValve prosthesis were enrolled. (22) Clinical endpoints were recorded according to VARC criteria. Primary endpoint was a major adverse cardiac or cerebrovascular event (MACCE) at 30-days post procedure defined as a composite of all-cause mortality, myocardial infarction, emergent cardiac surgery or percutaneous re-intervention, and stroke.
Patients showed a mean logistic EuroSCORE of 19.2. At 30 days, major adverse cardiac and cerebrovascular events (MACCE) had occurred in 8.3% of patients. Rates of total mortality and cardiac mortality were 4.5% and 3.4%, respectively. Strokes occurred in just 2.9% of patients. At six months, all-cause mortality was 12.8% and cardiovascular mortality was 8.4%. 

Randomised Trial Data

• PARTNER 

To date, the Placement of Aortic Transcatheter Valves (PARTNER) Trial is the only randomised controlled trial in TAVI patients. The trial consisted in two different patient cohorts: patients who were considered to be at high surgical risk (Society of Thoracic Surgeons [STS] score ≥10) were included in the PARTNER cohort A; patients who were not considered to be suitable candidates for SAVR were enrolled into PARTNER cohort B. 
The primary endpoint was the rate of mortality from any cause over the duration of 1 year. All TAVI patients were treated with the first generation of the balloon-expandable Edwards-SAPIEN THV. TAVI procedures were performed under general anaesthesia and transfemoral access was achieved by surgical cutdown in all patients.
The results of PARTNER cohort A confirmed the non-inferiority of TAVI compared to SAVR. Mortality at 30 days was lower than expected in both arms of the trial. In the intention-to-treat-analysis, a non-significant trend towards a lower 30-day mortality in the transcatheter subgroup could be observed compared with the surgical group (3.4% vs. 6.5%; P=0.07). The 30-day mortality (3.4%) in the TAVI group was the lowest reported in any series, despite an early generation device and limited operator experience. In a non-pre-specified subgroup analysis, TAVI was especially beneficial among women and patients without prior CABG. Furthermore, the clinical benefits of TAVI over SAVR included significantly shorter stays in the intensive care unit and in hospital overall, and a more rapid improvement in New York Heart Association (NYHA) functional class at 30 days and 6 months. 
The 2-year follow-up data of patients in the PARTNER trial cohort A supported the use of TAVI as an alternative to surgery in high-risk patients with a comparable 2-year mortality rate (33.9% vs. 35.0%; P=0.78). The early increase in the risk of stroke with TAVI was attenuated over time (7.7% vs. 4.9%; P=0.17). In this analysis, mild to severe peri-prosthetic aortic regurgitation following TAVI was more frequent after TAVI and associated with a more than two times increased 2-year mortality risk. Surprisingly, also mild peri-prosthetic aortic regurgitation had a negative impact on survival after TAVI.
In PARTNER cohort B, inoperable patients were randomly assigned to either transfemoral TAVI or standard medical therapy including balloon aortic valvuloplasty (BAV). The PARTNER cohort B has shown that TAVI is superior to standard medical therapy including BAV. In the fist year, only five patients needed to undergo TAVI to prevent one death. At 2 years, TAVI remained superior to standard therapy and markedly reduced the rates of all-cause mortality (43.3% vs. 68.0%; P<0.001; Number-needed-to-treat [NNT]=4.0 patients), cardiovascular mortality (31.0% vs. 62.4%; P<0.001; NNT=3.2 patients), and repeat hospitalisation (35.0% vs. 72.5%; P<0.001; NNT=2.7 patients).(23) In a landmark analysis, the mortality benefit of TAVI was determined separately for the second year post procedure and was again significantly superior compared to conservative management including BAV (18% vs. 35%; P=0.019). Furthermore, TAVI patients still had an improved functional status at 2 years and decreased heart failure symptoms (NYHA class III/IV) compared to standard therapy (16.8% vs. 57.5%; P<0.001). A subgroup analysis according to the surgical STS risk score suggested that the most pronounced benefit of TAVI was found in patients without extreme clinical comorbidities.

II - Cost effectiveness

Given the advanced age and multiple comorbid conditions that characterise patients with high surgical risk for SAVR, the question of whether TAVI can provide meaningful cost-effective health benefits is of particular importance. A pre-planned health economic study of the PARTNER trial showed that TAVI increases life expectancy in elderly patients who are not considered candidates for SAVR at an incremental cost per life year gained well within accepted values for commonly used cardiovascular technologies. With approximately $50,000 per year of life gained, TAVI has a favourable cost-effectiveness compared to the dialysis benchmark of approximately $70,000 per year of life gained. (24) However, further studies will be needed to compare the cost-effectiveness of TAVI with SAVR, especially in lower risk patient populations.
A word of caution is needed in the context of the PARTNER trial: as usually in this kind of trial, the patient cohort is a highly selected and important patient subgroups, e.g. those requiring coronary revascularisation, suffering from severe renal insufficiency, or severe peripheral arterial disease were excluded per protocol. Furthermore, the rate of vascular complications might not be transferable to real-life scenarios, as with the second generation of the Edwards-SAPIEN THV the transfemoral sheath size has been reduced with most TAVI procedures in Europe now performed percutaneously. 
CoreValve US Pivotal Trial
The results of the CoreValve US Pivotal trial, which will enroll more than 1,600 patients at 44 sites in the US, are eagerly awaited and will have to be compared with the encouraging results of the PARTNER Trial. In a randomised fashion, TAVI with use of the third generation Medtronic CoreValve bioprosthesis will be compared with SAVR in high-risk surgical and extreme risk patients in analogy to the PARTNER trial. The composite primary endpoint of this study includes all-cause mortality and the occurrence of major stroke at 1 year.

Clinical Evidence - Summary

Compared to conventional SAVR, short- and medium-term results of TAVI are non-inferior with a 30-day survival rate of 92-97% and 1-year survival rate of 69-79% in high-risk patients. 

Update on TAVI Indications

The evidence for TAVI in high-risk and inoperable patients is striking. The logical next step is to evaluate intermediate risk patients. A number of studies have addressed this issue:

• BERMUDA Triangle study

In the retrospective three-centre BERMUDA (BERn MUnich rotterDAm) Triangle study, 510 matched intermediate risk patients with STS scores of 3-8% (TAVI: n=255 vs. SAVR: n=255) had comparable 1-year mortality with TAVI (18.8%) compared to a surgical cohort (17.0%). (25)

• STACCATO Trial

The STACCATO Trial (A Prospective, Randomised Trial of Transapical Transcatheter Aortic Valve Implantation versus Surgical Aortic Valve Replacement in Operable Elderly Patients with Aortic Stenosis) from Denmark randomised intermediate-risk patients to transapical TAVI or SAVR (logistic EuroSCORE: 9.4±3.9% vs. 10.3±5.8%, P=0.25; mean STS score 3.1±1.5 vs. 3.4±1.2, P=0.43).(26) The primary endpoint was a composite of all-cause mortality, major stroke, and/or renal failure at 30 days. The trial was terminated prematurely after the inclusion of 70 patients due to an excess of adverse events in the transapical TAVI group compared to patients undergoing SAVR (35.3% vs. 8.3%). This unacceptable high event rate in transapical cases was completely unexpected, since the previous experience of both centers was almost 0% in high-risk TAVI patients. 

• SURTAVI Trial and PARTNER II Trial

The Surgical Replacement and Transcatheter Aortic Valve Implantation (SURTAVI) trial will randomize patients ≥70 years with aortic stenosis and an STS score of 3–8% to TAVI or SAVR. This trial will be the counterpart of the PARTNER II trial for the CoreValve bioprosthesis and is currently recruiting. The primary endpoint is a composite of death and major stroke at 2 years. (27)
The PARTNER II Trial will address an intermediate risk cohort (STS score 4-8%) comparing TAVI versus SAVR in a randomised 1:1 fashion. After assessment of transfemoral access, patients will either undergo transfemoral TAVI with the Edwards-SAPIEN XT™ or transapical TAVI with the Ascendra 2™ system. With the composite primary endpoint of all-cause mortality and major stroke at two years, 1,500 to 2,000 patients will be needed to prove non-inferiority. Importantly, patients suffering from symptomatic aortic stenosis and concomitant CAD will be sub-stratified to compare different strategies (TAVI + PCI vs. SAVR + CABG).

Update on Indications - Summary

To summarize, the evidence for TAVI in high-risk and inoperable patients is striking. Data on intermediate risk patients from non-randomised trial are encouraging. However, until the results of the randomised PARTNER II trial and the SURTAVI trial are available, strict adherence to the current ESC guidelines is highly recommended. The non-inferiority of TAVI in lower risk cohorts compared to conventional surgical aortic valve replacement need to be proven first in randomized trials.

Table - Procedural outcomes of current randomised controlled TAVI trials and multicenter TAVI registries

Conclusion

TAVI is the new standard-of-care in inoperable patients. In high-risk patients, TAVI has shown non-inferiority compared with SAVR. Data from all national multi-centre registries are encouraging and fuel speculation about the expansion of TAVI to intermediate-risk patients. Randomised studies are eagerly awaited comparing TAVI vs. SAVR in intermediate risk patients. Important questions remain: i) which access route provides the best clinical results - transvascular or transapical, ii) which access strategy should be preferred when a trans-femoral approach is denied, and iii) which of the upcoming valve types offers superior results with regard to ease of implantation, vascular access, paravalvular leakage, pacemaker requirement, and durability.