Transcatheter aortic valve implantation (TAVI) versus surgical aortic valve replacement (SAVR) for patients with severe aortic stenosis and low surgical risk and across surgical risk groups: a health technology assessment
Health technology assessment
|Published
In August 2019 the Ordering Forum Regional Health Authority commissioned the Norwegian Institute of Public Health (NIPH) to perform a health technology assessment of transcatheter aortic valve implantation/ replacement (TAVI/TAVR) compared with surgical aortic valve replacement (SAVR) in patients with severe aortic stenosis across surgical risk groups.
Key message
In August 2019 the Ordering Forum Regional Health Authority commissioned the Norwegian Institute of Public Health (NIPH) to perform a health technology assessment of transcatheter aortic valve implantation/ replacement (TAVI/TAVR) compared with surgical aortic valve replacement (SAVR) in patients with severe aortic stenosis across surgical risk groups.
We conducted an overview of systematic reviews that included the two newest randomised trials on TAVI in low risk group published in May 2019. We included 15 systematic reviews (2 covering all risk groups, 11 the low risk group, and 2 the intermediate and low risk groups). Based on evidence from eight randomised trials, we conclude that TAVI compared with SAVR in patients with severe aortic stenosis across all surgical risk groups:
- probably improves all-cause mortality or disabling stroke up to two years
- may slightly reduce major bleeding, new-onset fibrillation and acute kidney injury
- probably increases transient ischemic attacks, major vascular complications, permanent pacemaker implantation, re-intervention and paravalvular leak
- may make little or no difference for all-cause and cardiovascular mortality, myocardial infarction and stroke at long-term follow-up.
Health economic analysis was limited to the low surgical risk group, as the intermediate risk group was evaluated in a 2019 NIPH report. The cost-utility analysis in a lifetime perspective indicated that TAVI was more effective (gain of 0.05 QALYs) and less costly (saving of NOK 35 000) than SAVR for patients with severe aortic stenosis at low surgical risk. The analysis is based on 1-year follow-up data from the PARTNER 3 study and long-term mortality and adverse events for TAVI and SAVR beyond this period remain unclear. The results are sensitive to variations in procedure costs.
The budget impact analysis indicates that the introduction of TAVI for low risk patients is likely to be cost-neutral in the short run. We have not accounted for the costs of the capacity expanding.
Summary
Background
Heart failure due to aortic stenosis is an increasing health problem with increasing age, and hence in an aging society. In general, medical therapy does not treat severe aortic stenosis, but may be used to optimise blood flow and to alleviate symptoms in patients with symptomatic severe aortic stenosis. Therefore, until a few years ago, surgical treatment was the treatment of choice for patients with severe aortic stenosis. This changed with the introduction of transcatheter aortic valve implantation (TAVI), deploying a bioprosthesis in the aortic valve using a catheter. In contrast to traditional open-heart surgery or surgical aortic valve replacement, the procedure is less invasive and can be performed with light sedation and without cardiopulmonary bypass. With increasing clinical use and established effect and safety for TAVI in patients with severe aortic stenosis at high/intermediate surgical risk, the focus of TAVI producers shifted to patients at low surgical risk. In August 2019, in light of two newly completed RCTs including patients with severe aortic stenosis at low surgical risk, the Ordering Forum RHA commissioned the Norwegian Institute of Public Health to perform an assessment across all risk groups.
Objective
The objective of this health technology assessment is to update and summarise current knowledge on effectiveness and safety with transcatheter aortic valve implantation/ replacement (TAVI/TAVR) compared with surgical aortic valve replacement (SAVR) in the treatment of patients with severe aortic stenosis across surgical risk groups, including patients with severe aortic stenosis and high surgical risk.
The aim of the health economic evaluation is to assess the cost-effectiveness and budget impact of TAVI for patients with severe aortic stenosis and low surgical risk compared with open surgery and to evaluate the intervention against the priority setting criteria applicable in Norway. This information will supplement the 2019 report on the intermediate risk group.
Method
We conducted an overview of systematic reviews guided by the methodology handbook used at the Division for Health Services at the Norwegian Institute of Public Health. We excluded reviews published before April 2019, before the publication of the newest studies on patients with low surgical risk. We assessed the quality of identified systematic reviews with a 9-point checklist from our methodology handbook. We reported on the most updated reviews of acceptable quality, and communicated their findings, including GRADE assessment of the confidence in the effect estimates; both across all risk groups and for the low risk group specifically.
We narratively summarised the findings of our earlier reports and supplemented our former findings with the newly identified literature where possible and relevant.
In the economic evaluation, we performed a cost-utility analysis (CUA) comparing TAVI with open surgery for patients at low surgical risk, where we accounted for all relevant cost and health outcomes related to both procedures. The costs were expressed in 2020 Norwegian kroner (NOK), and effects in quality-adjusted life-years (QALYs). The results were expressed as mean incremental cost-effectiveness ratio (ICER). The Markov model was developed and analysed in TreeAge Pro ® 2020. The uncertainty in model parameters were handled by performing probabilistic sensitivity analyses (PSA). The analyses were performed from the healthcare perspective. Both costs and effects were discounted using an annual discount rate of 4%.
In accordance with the Government White Paper about priority setting, (Meld. St. 34 2015–2016), and its recommendations related to quantification of the severity criterion, we estimated absolute shortfall for patients with severe aortic stenosis and low surgical risk.
Premised on assumptions based on registry data about adoption rates for TAVI as well as cost data derived from the Markov model, we calculated likely budgetary consequences of extending TAVI as a routine treatment onto patients with severe aortic stenosis and lower risk groups.
Results
Of the 78 identified references, we assessed all titles and abstracts against the inclusion criteria and considered 15 as possibly relevant. We assessed the quality of all 15 reviews. The reviews cover a total of eight randomised trials, including the two most recent trials on patients with low surgical risk published in 2019.
Based on evidence from eight randomised trials captured in several systematic reviews, we conclude that TAVI compared with SAVR in patients with severe aortic stenosis across all surgical risk groups
- probably improves all-cause mortality or disabling stroke up to two years
- may slightly reduce major bleeding, new-onset fibrillation and acute kidney injury
- probably increases transient ischemic attacks, major vascular complications, permanent pacemaker implantation, re-intervention and paravalvular leak
- may make little or no difference for all-cause and cardiovascular mortality, myocardial infarction and stroke at long-term follow-up.
The results of the cost-utility analysis in the base-case scenario show that TAVI for patients at low risk is associated with a higher QALY-gain (incremental QALY 0.05) and lower cost (incremental costs – NOK 35 000) when compared to surgical aortic valve replacement (SAVR). These results are most sensitive to changes in estimates of the procedure costs.
The expansion of use of TAVI onto patients with lower surgical risk is likely to be cost-neutral in the short run. This expansion would imply a doubling in the numbers of TAVI procedures performed within the next five years. The costs of the capacity expanding were not included in the analyses.
The calculated absolute shortfall for patients with severe aortic stenosis and low surgical risk is equal to 2 QALYs.
Conclusion
Based on available evidence from eight RCTs, captured in several systematic reviews, we conclude that for patients with severe aortic stenosis across all surgical risk groups TAVI compared with SAVR probably improves all-cause mortality or disabling stroke up until 2 years. TAVI may slightly reduce incidences of major bleeding, new-onset fibrillation, and acute kidney injury. On the other hand, TAVI probably increases the incidence of transient ischemic attacks, major vascular complications, permanent pacemaker implantation, reintervention, and paravalvular leak. Moderate-quality evidence suggests that TAVI may make little or no difference for the incidences of all-cause and cardiovascular mortality, myocardial infarction, and stroke after two years; based on the limited long-term data. The clinical decision for either option may benefit from a broader evaluation of the patient’s medical state and their life expectancy due to uncertainty regarding long term effects.
The results of our cost-utility analysis based on 1-year follow-up data from the PARTNER 3 study indicate, that TAVI for patients at low surgical risk is slightly more effective (0.05 QALYs gained) and less costly (saving of NOK 35 000) than SAVR. The results are sensitive to variations in procedure costs. The budget impact analysis indicated that the extension of use of TAVI to patients at low surgical risk is likely to be cost-neutral in the short run.
The calculated absolute shortfall for patients with severe aortic stenosis and low surgical risk relative to their age cohort in the general population is equal to 2 QALYs, categorising these patients into severity class 1, which is the least severe of the six classes suggested by the Magnussen group. These findings can help decision makers appraise the intervention against the official priority setting criteria in health care sector applicable in Norway.
