Key points from the evidence

1.Six meta-analyses were selected for review based on their methodological quality, length of follow up, outcomes and primary study designs. They included an overlapping set of five RCTs and eight observational studies reporting all-cause mortality after TAVI compared with SAVR in people with symptomatic severe aortic stenosis who are at low surgical risk.

• A Cochrane review of four RCTs (n=2,818) reported no statistically significant difference in all-cause mortality at 30 days follow up (relative risk (RR) 0.60, 95% confidence interval (CI) 0.33 to 1.44) or 1 year follow up (RR 0.70, 95% CI 0.44 to 1.11).
• A meta-analysis of three RCTs (n=2,748) found no statistically significant difference in all-cause mortality at 4–5 years follow up (odds ratio (OR) 0.94, 95% CI 0.65 to 1.37). One of the RCTs, the NOTION trial reported consistent findings at 8 and 10 years follow up.
• A time to event meta-analysis of three RCTs and five observational studies (n=5,444) found no statistically significant difference in all-cause mortality over the first 2 years of follow up (hazard ratio (HR) 1.08, 95% CI 0.89 to 1.31). Meta-analysis of the observational studies found a statistically significantly higher risk of all-cause mortality for the TAVI group compared with the SAVR group between 2 and 8 years follow up (HR 1.51, 95% CI 1.14 to 2.00, n=2,696). These longer term results should be interpreted with caution because observational studies have inherent biases compared with RCTs.

2.Three recently published RCTs (UK TAVI, DEDICATE, NOTION-2) also concluded there were no statistically significant differences in all-cause mortality in comparisons of TAVI and SAVR at 1 year follow up.

3. Four out of five RCTs that reported length of hospital stay found that patients who had a TAVI procedure needed to stay in hospital for significantly fewer days than patients who had SAVR. The only RCT that reported a longer hospital stay for TAVI patients (STACCATO) used a higher risk vascular access route for performing TAVI.

4. A meta-analysis of studies in people at low or intermediate surgical risk (n=3,681 low risk patients) found a statistically significant improvement in quality of life for people who had a TAVI procedure compared with people who had SAVR at 30 days follow up. At 1 year follow up there was no statistically significant between group difference in quality of life scores.

5. The six meta-analyses reported that TAVI was associated with statistically significantly higher risks of new pacemaker implantation and paravalvular leaks. In the same meta-analyses, TAVI was associated with significantly lower risks of new onset atrial fibrillation, acute kidney injury and bleeding.

• A Cochrane review (n=2,748) demonstrated a statistically significantly higher risk of new permanent pacemaker implantation in the TAVI group compared with the SAVR group (HR 3.65, 95% CI 1.50 to 8.87).
• Two meta-analyses (n=2,611 and n=2,219) reported a seven- to nine-fold higher risk of paravalvular leaks in patients who had TAVI compared with patients who had SAVR. It was unclear whether the severity of these paravalvular leaks affected patient outcomes.
• The Cochrane review also found statistically significantly lower risks of new onset atrial fibrillation (HR 0.21, 95% CI 0.15 to 0.30), acute kidney injury (HR 0.30, 95% CI 0.16 to 0.58) and any bleeding event (HR 0.31, 95% CI 0.16 to 0.72) in the TAVI group compared with the SAVR group.
• Safety results were consistent over the long term (more than 1 year follow up), across more recent RCTs and in meta-analyses of observational studies.

6. A systematic review of six economic analyses comparing TAVI with SAVR in people with
symptomatic severe aortic stenosis who are at low surgical risk, found that TAVI was
dominant (more effective and less costly) in one study, cost effective in three studies and not cost effective in one study. The study that reported TAVI was not cost effective was based on older (2010 vs 2018) TAVI devices that may have a different effectiveness and cost profile.

7. Nine primary economic evaluations compared TAVI (using the SAPIEN 3 valve) with SAVR.

• Seven analyses conducted in European countries using local health system cost data found the SAPIEN 3 TAVI valve to be dominant or cost effective (that is, below willingness to pay thresholds) compared with SAVR. The analyses used outcome data from the PARTNER 3 trial, combined with local costs. Results were stable across time horizons of up to 30 years. TAVI was not cost effective when the analyses were restricted to a time horizon of 5 years.
• Economic evaluations from the United States (US) and Japanese payer perspectives found that TAVI was dominant or cost effective compared with SAVR.
• Key drivers of cost effectiveness attributed to TAVI are greater survival benefit, fewer complications and lower long term management costs. Higher device and 1 year follow up costs for TAVI are offset by lower healthcare costs over a lifetime time horizon.

8. In 2021, the National Institute for Health and Care Excellence (NICE) published an economic evaluation using data from RCTs and the UK TAVI registry. The NICE model found that TAVI was not cost effective compared with SAVR in people at low surgical risk over a 15-year time horizon.

• The conflicting results of the NICE evaluation compared with other published economic evaluations can mainly be attributed to higher device procurement costs in the UK (£17,500) compared to other countries.
• The device price in the UK would need to be lower than £14,800 for TAVI to be cost effective at a willingness to pay threshold of £30,000 per quality adjusted life-year (QALY).

9. An economic analysis from the perspective of the UK NHS (2024) used 4-year follow up data from the Evolut Low Risk trial to compare TAVI using the Evolut self-expanding valve with SAVR. TAVI was cost effective at an assumed UK TAVI valve list price of £17,500 over a lifetime horizon.

• The improved cost effectiveness of TAVI in this analysis compared with the NICE evaluation can be attributed to the availability of longer follow up data from the Evolut Low Risk trial, which suggests improved survival and outcomes after TAVI that leads to a ten-fold increase in QALYs over a 15-year time horizon.

10. This section describes patient and social aspects relating to TAVI among people who have symptomatic severe aortic stenosis regardless of their surgical risk because the literature rarely reports the level of surgical risk applicable to study participants.

11. A retrospective cohort study (n=10,069) in NHS England found inequalities in access to aortic valve replacement procedures (TAVI or SAVR) for women, people living in more deprived areas and people of African or south Asian descent.

12. A systematic review (n=353) explored patient experiences of having a TAVI procedure.

• Having an individualised care plan was important to patients because it helped mitigate any disappointment or frustration relating to outcomes after the procedure. When people felt their treatment goals had been met, they reported improved quality of life even when objective measures suggested minimal change in functional outcomes.
• Some patients experience persistent psychosocial and physical symptoms after their procedure, often because of pre-existing comorbidities. For many people this led to feelings of disappointment, isolation and vulnerability.

13. A systematic review (n=1,096), a qualitative study (n=18) and a cross-sectional study (n=98) explored people’s reasons for undergoing treatment for aortic stenosis and factors influencing their decision. Patients were often motivated by a strong desire to regain some semblance of a normal life and to continue living independently. The published literature suggests this could be achieved more quickly after TAVI compared with SAVR.