Effect of transcatheter aortic valve implantation vs surgical aortic valve replacement on all-cause mortality in patients with aortic stenosis: a randomized clinical trial
Selected in JAMA : Journal of the American Medical Association by D. Giacoppo
The UK TAVI trial sought to compare TAVI with SAVR in patients with severe, symptomatic aortic stenosis and moderately increased surgical risk.
References
Authors
UK TAVI Trial Investigators,; William D Toff, David Hildick-Smith, Jan Kovac, Michael J Mullen, Olaf Wendler, Anita Mansouri, Ines Rombach, Keith R Abrams, Simon P Conroy, Marcus D Flather, Alastair M Gray, Philip MacCarthy, Mark J Monaghan, Bernard Prendergast, Simon Ray, Christopher P Young, David C Crossman, John G F Cleland, Mark A de Belder, Peter F Ludman, Stephen Jones, Cameron G Densem, Steven Tsui, Manoj Kuduvalli, Joseph D Mills, Adrian P Banning, Rana Sayeed, Ragheb Hasan, Douglas G W Fraser, Uday Trivedi, Simon W Davies, Alison Duncan, Nick Curzen, Sunil K Ohri, Christopher J Malkin, Pankaj Kaul, Douglas F Muir, W Andrew Owens, Neal G Uren, Renzo Pessotto, Simon Kennon, Wael I Awad, Saib S Khogali, Maciej Matuszewski, Richard J Edwards, Bandigowdanapalya C Ramesh, Miles Dalby, Shahzad G Raja, Giovanni Mariscalco, Clinton Lloyd, Ian D Cox, Simon R Redwood, Mark G Gunning, Paul D Ridley
Reference
JAMA. 2022;327(19):1875-1887. doi:10.1001/jama.2022.5776
Published
17 may 2022
Link
Read the abstractReviewer
My Comment
Why this study – the rationale/objective?:
In recent years, high-quality evidence on the comparison between transcatheter aortic valve implantation (TAVI) and surgical aortic valve replacement (SAVR) has substantially increased.
Randomized clinical trials of patients with symptomatic severe aortic stenosis and intermediate surgical risk have generally demonstrated that TAVI with balloon- or self-expandable valves - are associated with noninferior outcomes compared with SAVR.
Subsequent trials of patients with low surgical risk have confirmed the noninferiority of TAVI to SAVR and sometimes indicated a possible early clinical improvement associated with the transcatheter approach.
The UK TAVI trial sought to compare TAVI with SAVR in patients with severe, symptomatic aortic stenosis and moderately increased surgical risk.
How was it executed? - the methodology:
The UK TAVI trial was an investigator-initiated, randomized clinical trial conducted at 34 centres in the United Kingdom, designed to test the noninferiority of TAVI to SAVR in terms of 12-month all-cause death.
Between April 2014 and April 2018, a total of 913 patients older than 70 years with severe, symptomatic aortic stenosis were randomly assigned to TAVI (n = 458) or SAVR (n = 455).
Patients randomized to TAVI could receive any transcatheter aortic valve with a CE mark, implanted by any access route.
Events were adjudicated by a clinical events committee which was unaware of the assigned treatment.
What is the main result?
The median age was 81 years, 46.4 % of the participants were women, and the median Society of Thoracic Surgeons (STS) mortality risk score was 2.6 % [interquartile range 2.0 % - 3.4 %].
In the TAVI group, general anaesthesia was used in 30.4 % of patients. In the SAVR group, midline sternotomy was performed in 89.5 % of patients.
In the TAVI group, 7.3% of patients underwent coronary revascularization during the same hospitalization or as a staged procedure, while in the SAVR group 21.5 % of patients underwent concurrent coronary artery bypass grafting.
In the TAVI group, 5 patients crossed over to surgery and 3 did not receive treatment. In the SAVR group, 17 patients crossed over to TAVI and 19 did not receive treatment.
In the TAVI group, the common access route was transfemoral (92.0 %) and the most commonly implanted transcatheter aortic valve was the balloon-expandable Sapien 3 valve (45.1 %), followed by the self-expandable Evolut R valve (13.6 %) and the mechanically-expandable Louts valve (9.6 %).
At 12 months, 21 patients (4.6 %) in the TAVI group had died compared with 30 patients (6.6 %) in the SAVR group, resulting in an upper limit of the 1-sided 97.5 % confidence interval of the absolute risk difference (1.2 %) lower than the prespecified noninferiority margin (pnoninferiority < 0.001).
The hazard ratio (HR) was 0.69 [95 % confidence interval (CI) 0.39-1.20].
Sensitivity (per-protocol, missing data, adjustment for age, coronary artery disease requiring revascularization, and centre) and subgroup analyses showed consistent results.
At 12 months, the secondary composite endpoints of all-cause death or nonfatal stroke (8.5 % vs. 8.8 %, p = 0.88; HR 0.96, 95 % CI 0.62-1.50) and all-cause death, nonfatal stroke, or reintervention (14.2 % vs. 16.3 %, p = 0.41; HR 0.86, 95 % CI 0.62-1.20) did not significantly differ between TAVI and SAVR.
Among the individual outcomes of the composite endpoints, any-type nonfatal stroke tended to be more frequent after TAVI than SAVR (5.2 % vs. 2.6 %, p = 0.07), while no significant difference in reintervention was observed between TAVI and SAVR (6.6 % vs. 8.1 %, p = 0.31).
Among the other individual secondary outcomes, cardiovascular death was not significantly different between TAVI and SAVR (2.8 % vs. 3.3 %, p = 0.69).
The need for permanent pacing was significantly higher after TAVI compared with SAVR (14.2 % vs. 7.3 %, p < 0.001). Patients assigned to TAVI showed lower incidences of major bleeding (7.2 % vs. 20.2 %, p < 0.001), but more frequent occurrence of vascular complications (10.3 % vs. 2.4 %, p < 0.001).
Critical reading and the relevance for clinical practice:
The UK TAVI trial convincingly demonstrates that TAVI is noninferior to SAVR with respect to all-cause death at 12 months and enhances accumulated evidence on the comparison between TAVI vs. SAVR in patients with intermediate-to-low surgical risk.
At 12 months, the incidence of all-cause death was low in both TAVI and SAVR groups and, except for a numerical increase in any-type stroke associated with TAVI, no significant differences or possible signals of harm were noticed between treatment groups in terms of composite endpoints.
Consistently with previous studies, TAVI was associated with a higher need for permanent pacing, a higher occurrence of major vascular complications, and lower rates of major bleeding compared with SAVR.
Although the main results of the UK TAVI trial generally support TAVI safety and efficacy, some considerations are required.
Statistical power was likely insufficient to test the noninferiority of TAVI compared to SAVR for the individual endpoint of all-cause death and an absolute noninferiority margin of 5 % is quite large by considering the number of events observed at 12 months.
More in detail, the initial sample size was based on an assumed 12-month mortality after surgery of 15 % and a noninferiority margin of 7.5 %.
At the interim analysis, a lower than expected incidence of all-cause death led to a revision of the sample size with an assumed 12-month mortality after surgery of 7.5 % and a noninferiority margin of 5 %.
At 12 months, however, all-cause death occurred only in 6.6 % of patients assigned to surgery and the selected noninferiority margin implies that the primary hypothesis could be met when the upper limit of the 1-sided 97.5 % confidence interval was not above 12.5 %.
The observation of outcomes incidences lower than expected may be linked to the inclusion of patients with lower surgical risk. Indeed, although the trial was designed to include patients with moderate surgical risk, median STS mortality (2.0 %) and EuroSCORE II (2.0 %) risk scores were consistent with a low surgical risk population on the basis of the definition generally adopted in previous studies (i.e., < 4.0 %). Only 17 % (n = 155) of patients had an STS score < 4.0 % and 13 % (n = 122) had an EuroSCORE II <4.0 %.
However, the comparison of 12-month mortality rates between the UK TAVI trial and previous studies did not reflect a low surgical risk population. Indeed, mortality in the UK TAKI trial (4.6 % vs. 6.6 %) was higher than in the PARTNER 3 (1.0% vs. 2.5 %) and Evolut Low Risk trials (2.4 % vs. 4.6 %) – low surgical risk trials. Yet, in the UK TAVI trial, 12-month mortality was lower than in the PARTNER 2 trial (12.3 % vs. 12.9 %) and numerically similar to the SURTAVI trial (6.7 % vs. 6.8 %) - intermediate surgical risk trials.
The accrual rate in the TAVI UK trial - involving every centre performing TAVI in the UK - was lower than in studies on the same topic and the absence of information related to the number and characteristics of patients not considered for randomization does not allow assessing the rigorousness of the selection.
In addition, substantial changes in guidelines and transcatheter aortic valves during the enrolment may have played a role.
The results of the PARTNER 2 and SURTAVI trials as well as the long-term analyses of the PARTNER and CoreValve US High Risk trials have broadened the indications for TAVI during UK TAVI trial conduction and the use of different valve technologies and iterations may have influenced the results observed in the TAVI group. Indeed, early transcatheter aortic valves have generally shown a lower performance compared with contemporary iterations, some transcatheter aortic valves used in the trial have not resulted to be noninferior to the more widely used controls in other studies, and a transcatheter aortic valve was withdrawn from the market due to issues in the delivery technology and unfavourable outcomes. Possible technological advantages of contemporary high-performance valves might improve some outcomes observed in the TAVI group.
Finally, studies on TAVI vs. SAVR require prolonged observation to draw conclusions accounting for time-dependent effects related to the type of intervention (e.g., bioprosthesis degeneration). The promising results of the UK TAVI trial warrant confirmation at very long-term follow-up.
References
- Leon et al. Transcatheter or Surgical Aortic-Valve Replacement in Intermediate-Risk Patients. N Engl J Med 2016; 374:1609-1620 doi: 10.1056/NEJMoa1514616
- Reardon et al. Surgical or Transcatheter Aortic-Valve Replacement in Intermediate-Risk Patients. N Engl J Med 2017; 376:1321-1331 doi: 10.1056/NEJMoa1700456
- Mack et al. Transcatheter Aortic-Valve Replacement with a Balloon-Expandable Valve in Low-Risk Patients. N Engl J Med 2019; 380:1695-1705 doi: 10.1056/NEJMoa1814052
- Popma et al. Transcatheter Aortic-Valve Replacement with a Self-Expanding Valve in Low-Risk Patients. N Engl J Med 2019; 380:1706-1715 doi: 10.1056/NEJMoa1816885
- Mack et al. 5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1). Lancet 2015;385:2477-84
doi: 10.1016/S0140-6736(15)60308-7 - Gleason et al. 5-Year Outcomes of Self-Expanding Transcatheter Versus Surgical Aortic Valve Replacement in High-Risk Patients. J Am Coll Cardiol 2018;72:2687-2696 doi: 10.1016/j.jacc.2018.08.2146
- Makkar et al. Self-expanding intra-annular versus commercially available transcatheter heart valves in high and extreme risk patients with severe aortic stenosis (PORTICO IDE). Lancet 2020;396:669-683 doi: 10.1016/S0140-6736(20)31358-1
- Lanz et al. Safety and efficacy of a self-expanding versus a balloon-expandable bioprosthesis for transcatheter aortic valve replacement in patients with symptomatic severe aortic stenosis. Lancet 2019;394:1619-1628 doi: 10.1016/S0140-6736(19)32220-2
- Tamburino et al. Comparison of Self-Expanding Bioprostheses for Transcatheter Aortic Valve Replacement in Patients With Symptomatic Severe Aortic Stenosis. Circulation 2020;142:2431-2442
doi: 10.1161/CIRCULATIONAHA.120.051547
No comments yet!