[Skip to Navigation]
Sign In
Visual Abstract. Transcatheter Aortic Valve Implantation vs Surgical Aortic Valve Replacement and All-Cause Mortality in Patients With Aortic Stenosis
Transcatheter Aortic Valve Implantation vs Surgical Aortic Valve Replacement and All-Cause Mortality in Patients With Aortic Stenosis
Figure 1.  Patient Selection, Allocation, and Flow in the UK Transcatheter Aortic Valve Implantation (UK TAVI) Trial for Aortic Stenosis
Patient Selection, Allocation, and Flow in the UK Transcatheter Aortic Valve Implantation (UK TAVI) Trial for Aortic Stenosis

aPatients were invited to participate after multidisciplinary team review and confirmation of eligibility. Participating sites maintained monthly screening logs of patients recommended for consideration for enrollment; however, some logs were missing, with an estimated overall shortfall of 22% (based on the number of patients randomized but not included in the screening logs). This would imply that the actual number of patients reviewed by the multidisciplinary team and invited to participate was in the region of 1740 rather than the stated figure of 1357, which was derived directly from the screening logs. Data regarding overall national TAVI and surgery activity during the relevant period are available from the relevant national audits.26-28

bRandomization used minimization, including an 80% probabilistic element with stratification for the randomization site, age group (70-79 years vs ≥80 years), and the presence of coronary artery disease considered by the multidisciplinary team to require revascularization if the patient was randomized to receive surgery.

cThree additional participants were treated as randomized but excluded from the per-protocol analysis (2 were treated >1 year after randomization and 1 withdrew from all follow-up).

Figure 2.  Time-to-Event Curves for the Primary Outcome and Major Secondary Outcomes
Time-to-Event Curves for the Primary Outcome and Major Secondary Outcomes

Kaplan-Meier survival analysis at 1 year after randomization. All patients were followed up to the time of an event, withdrawal from the study, or 1 year after randomization. The hazard ratios are specific to the 1-year outcomes. The P values were derived from a Cox proportional hazards model, which was adjusted for randomization minimization factors and used robust SEs to account for clustering of outcomes by randomization site. Cause of death (cardiovascular vs noncardiovascular) and stroke events were adjudicated by the end points and events committee, with reference to outcome definitions based on criteria from the Valve Academic Research Consortium-215 consensus document (eMethods 5 in Supplement 2). TAVI indicates transcatheter aortic valve implantation.

Table 1.  Characteristics of the Participants at Baselinea
Characteristics of the Participants at Baselinea
Table 2.  Noninferiority Analysis of Death From Any Cause at 1 Year (Primary Outcome)
Noninferiority Analysis of Death From Any Cause at 1 Year (Primary Outcome)
Table 3.  Primary Outcome and Event-Related Secondary Outcomes 1 Year After Randomization and Event-Related Secondary Outcomes 30 Days After the Procedure
Primary Outcome and Event-Related Secondary Outcomes 1 Year After Randomization and Event-Related Secondary Outcomes 30 Days After the Procedure

JAMA Deputy Editor Gregory Curfman, MD, gives editorial insight into the UK TAVI Trial results and accompanying editorial published in the May 17, 2022, issue of JAMA.

1.
Cribier  A, Eltchaninoff  H, Bash  A,  et al.  Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description.   Circulation. 2002;106(24):3006-3008. doi:10.1161/01.CIR.0000047200.36165.B8PubMedGoogle ScholarCrossref
2.
Leon  MB, Smith  CR, Mack  M,  et al; PARTNER Trial Investigators.  Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery.   N Engl J Med. 2010;363(17):1597-1607. doi:10.1056/NEJMoa1008232PubMedGoogle ScholarCrossref
3.
Smith  CR, Leon  MB, Mack  MJ,  et al; PARTNER Trial Investigators.  Transcatheter versus surgical aortic-valve replacement in high-risk patients.   N Engl J Med. 2011;364(23):2187-2198. doi:10.1056/NEJMoa1103510PubMedGoogle ScholarCrossref
4.
Mack  MJ, Leon  MB, Smith  CR,  et al; PARTNER 1 Trial Investigators.  5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial.   Lancet. 2015;385(9986):2477-2484. doi:10.1016/S0140-6736(15)60308-7PubMedGoogle ScholarCrossref
5.
Adams  DH, Popma  JJ, Reardon  MJ,  et al; US CoreValve Clinical Investigators.  Transcatheter aortic-valve replacement with a self-expanding prosthesis.   N Engl J Med. 2014;370(19):1790-1798. doi:10.1056/NEJMoa1400590PubMedGoogle ScholarCrossref
6.
Gleason  TG, Reardon  MJ, Popma  JJ,  et al; CoreValve US Pivotal High Risk Trial Clinical Investigators.  5-year outcomes of self-expanding transcatheter versus surgical aortic valve replacement in high-risk patients.   J Am Coll Cardiol. 2018;72(22):2687-2696. doi:10.1016/j.jacc.2018.08.2146PubMedGoogle ScholarCrossref
7.
Généreux  P, Webb  JG, Svensson  LG,  et al; PARTNER Trial Investigators.  Vascular complications after transcatheter aortic valve replacement: insights from the PARTNER (Placement of AoRTic TraNscathetER Valve) trial.   J Am Coll Cardiol. 2012;60(12):1043-1052. doi:10.1016/j.jacc.2012.07.003PubMedGoogle ScholarCrossref
8.
Conte  JV, Hermiller  J  Jr, Resar  JR,  et al.  Complications after self-expanding transcatheter or surgical aortic valve replacement.   Semin Thorac Cardiovasc Surg. 2017;29(3):321-330. doi:10.1053/j.semtcvs.2017.06.001PubMedGoogle ScholarCrossref
9.
O’Brien  SM, Shahian  DM, Filardo  G,  et al; Society of Thoracic Surgeons Quality Measurement Task Force.  The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 2—isolated valve surgery.   Ann Thorac Surg. 2009;88(1)(suppl):S23-S42. doi:10.1016/j.athoracsur.2009.05.056PubMedGoogle ScholarCrossref
10.
Society of Thoracic Surgeons. STS short-term risk calculator. Accessed February 5, 2022. https://www.sts.org/resources/risk-calculator
11.
Nashef  SA, Roques  F, Sharples  LD,  et al.  EuroSCORE II.   Eur J Cardiothorac Surg. 2012;41(4):734-744. doi:10.1093/ejcts/ezs043PubMedGoogle ScholarCrossref
12.
EuroSCORE Study Group. EuroSCORE II calculator. Accessed February 5, 2022. http://www.euroscore.org/calc.htm
13.
Fried  LP, Tangen  CM, Walston  J,  et al; Cardiovascular Health Study Collaborative Research Group.  Frailty in older adults: evidence for a phenotype.   J Gerontol A Biol Sci Med Sci. 2001;56(3):M146-M156. doi:10.1093/gerona/56.3.M146PubMedGoogle ScholarCrossref
14.
Rockwood  K, Song  X, MacKnight  C,  et al.  A global clinical measure of fitness and frailty in elderly people.   CMAJ. 2005;173(5):489-495. doi:10.1503/cmaj.050051PubMedGoogle ScholarCrossref
15.
Kappetein  AP, Head  SJ, Généreux  P,  et al.  Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document.   J Am Coll Cardiol. 2012;60(15):1438-1454. doi:10.1016/j.jacc.2012.09.001PubMedGoogle ScholarCrossref
16.
Bonita  R, Beaglehole  R.  Recovery of motor function after stroke.   Stroke. 1988;19(12):1497-1500. doi:10.1161/01.STR.19.12.1497PubMedGoogle ScholarCrossref
17.
Campeau  L.  Letter: grading of angina pectoris.   Circulation. 1976;54(3):522-523. doi:10.1161/circ.54.3.947585PubMedGoogle ScholarCrossref
18.
Criteria Committee of the American Heart Association Affiliate NYC.  Nomenclature and Criteria for Diagnosis of Diseases of the Heart and Great Vessels. 9th ed. Little Brown & Co; 1994.
19.
Lincoln  NB, Gladman  JR.  The extended activities of daily living scale: a further validation.   Disabil Rehabil. 1992;14(1):41-43. doi:10.3109/09638289209166426PubMedGoogle ScholarCrossref
20.
Lipkin  DP, Scriven  AJ, Crake  T, Poole-Wilson  PA.  Six minute walking test for assessing exercise capacity in chronic heart failure.   Br Med J (Clin Res Ed). 1986;292(6521):653-655. doi:10.1136/bmj.292.6521.653PubMedGoogle ScholarCrossref
21.
Folstein  MF, Folstein  SE, McHugh  PR.  “Mini-Mental State”: a practical method for grading the cognitive state of patients for the clinician.   J Psychiatr Res. 1975;12(3):189-198. doi:10.1016/0022-3956(75)90026-6PubMedGoogle ScholarCrossref
22.
EuroQol Group. EuroQol EQ-5D-5L. Accessed February 5, 2022. https://euroqol.org/eq-5d-instruments/
23.
Rector  TS, Kubo  SH, Cohn  JN.  Patients’ self-assessment of their congestive heart failure, part 2: content, reliability and validity of a new measure, the Minnesota Living with Heart Failure Questionnaire.   Heart Fail. 1987;3:198-209.Google Scholar
24.
Bridgewater  B, Kinsman  R, Walton  P,  et al.  Demonstrating Quality: The Sixth National Adult Cardiac Surgery Database Report. Dendrite Clinical Systems Ltd; 2009.
25.
Kleinman  LC, Norton  EC.  What’s the risk? a simple approach for estimating adjusted risk measures from nonlinear models including logistic regression.   Health Serv Res. 2009;44(1):288-302. doi:10.1111/j.1475-6773.2008.00900.xPubMedGoogle ScholarCrossref
26.
National Institute for Cardiovascular Outcomes Research. National Adult Cardiac Surgery Audit: 2020 summary report (2016/17-2018/19 data). Accessed February 5, 2022. https://www.nicor.org.uk/wp-content/uploads/2020/12/National-Adult-Cardiac-Surgery-Audit-NACSA-FINAL.pdf
27.
Society for Cardiothoracic Surgery in Great Britain and Ireland. National cardiac surgery activity and outcome report 2002-2016. Accessed February 5, 2022. https://scts.org/_userfiles/pages/files/sctscardiacbluebook2020_11_20tnv2.pdf
28.
British Cardiovascular Intervention Society. United Kingdom national audit of transcatheter aortic valve implantation. Accessed February 5, 2022. https://www.bcis.org.uk/wp-content/uploads/2021/04/BCIS-Audit-2019-20-data-TAVI-subset-as-26-04-2020-for-web.pdf
29.
Leon  MB, Smith  CR, Mack  MJ,  et al; PARTNER 2 Investigators.  Transcatheter or surgical aortic-valve replacement in intermediate-risk patients.   N Engl J Med. 2016;374(17):1609-1620. doi:10.1056/NEJMoa1514616PubMedGoogle ScholarCrossref
30.
Reardon  MJ, Van Mieghem  NM, Popma  JJ,  et al; SURTAVI Investigators.  Surgical or transcatheter aortic-valve replacement in intermediate-risk patients.   N Engl J Med. 2017;376(14):1321-1331. doi:10.1056/NEJMoa1700456PubMedGoogle ScholarCrossref
31.
Thyregod  HG, Steinbrüchel  DA, Ihlemann  N,  et al.  Transcatheter versus surgical aortic valve replacement in patients with severe aortic valve stenosis: 1-year results from the all-comers NOTION randomized clinical trial.   J Am Coll Cardiol. 2015;65(20):2184-2194. doi:10.1016/j.jacc.2015.03.014PubMedGoogle ScholarCrossref
32.
Mack  MJ, Leon  MB, Thourani  VH,  et al; PARTNER 3 Investigators.  Transcatheter aortic-valve replacement with a balloon-expandable valve in low-risk patients.   N Engl J Med. 2019;380(18):1695-1705. doi:10.1056/NEJMoa1814052PubMedGoogle ScholarCrossref
33.
Popma  JJ, Deeb  GM, Yakubov  SJ,  et al; Evolut Low Risk Trial Investigators.  Transcatheter aortic-valve replacement with a self-expanding valve in low-risk patients.   N Engl J Med. 2019;380(18):1706-1715. doi:10.1056/NEJMoa1816885PubMedGoogle ScholarCrossref
34.
Siontis  GCM, Overtchouk  P, Cahill  TJ,  et al.  Transcatheter aortic valve implantation vs surgical aortic valve replacement for treatment of symptomatic severe aortic stenosis: an updated meta-analysis.   Eur Heart J. 2019;40(38):3143-3153. doi:10.1093/eurheartj/ehz275PubMedGoogle ScholarCrossref
35.
Kolkailah  AA, Doukky  R, Pelletier  MP, Volgman  AS, Kaneko  T, Nabhan  AF.  Transcatheter aortic valve implantation versus surgical aortic valve replacement for severe aortic stenosis in people with low surgical risk.   Cochrane Database Syst Rev. 2019;12(12):CD013319. doi:10.1002/14651858.CD013319.pub2PubMedGoogle ScholarCrossref
36.
Makkar  RR, Fontana  G, Jilaihawi  H,  et al.  Possible subclinical leaflet thrombosis in bioprosthetic aortic valves.   N Engl J Med. 2015;373(21):2015-2024. doi:10.1056/NEJMoa1509233PubMedGoogle ScholarCrossref
37.
Chakravarty  T, Søndergaard  L, Friedman  J,  et al; RESOLVE; SAVORY Investigators.  Subclinical leaflet thrombosis in surgical and transcatheter bioprosthetic aortic valves: an observational study.   Lancet. 2017;389(10087):2383-2392. doi:10.1016/S0140-6736(17)30757-2PubMedGoogle ScholarCrossref
38.
Blackman  DJ, Saraf  S, MacCarthy  PA,  et al.  Long-term durability of transcatheter aortic valve prostheses.   J Am Coll Cardiol. 2019;73(5):537-545. doi:10.1016/j.jacc.2018.10.078PubMedGoogle ScholarCrossref
39.
Makkar  RR, Thourani  VH, Mack  MJ,  et al; PARTNER 2 Investigators.  Five-year outcomes of transcatheter or surgical aortic-valve replacement.   N Engl J Med. 2020;382(9):799-809. doi:10.1056/NEJMoa1910555PubMedGoogle ScholarCrossref
Original Investigation
May 17, 2022

Effect of Transcatheter Aortic Valve Implantation vs Surgical Aortic Valve Replacement on All-Cause Mortality in Patients With Aortic Stenosis: A Randomized Clinical Trial

The UK TAVI Trial Investigators
JAMA. 2022;327(19):1875-1887. doi:10.1001/jama.2022.5776
Key Points

Question  Is transcatheter aortic valve implantation (TAVI) noninferior to surgical aortic valve replacement (surgery) in patients aged 70 years or older with severe, symptomatic aortic stenosis and moderately increased operative risk?

Findings  In this randomized clinical trial that included 913 patients at moderately increased operative risk due to age or comorbidity, all-cause mortality at 1 year was 4.6% with TAVI vs 6.6% with surgery, a difference that met the prespecified noninferiority margin of 5%.

Meaning  Among patients aged 70 years or older with severe, symptomatic aortic stenosis and moderately increased operative risk, treatment with TAVI was noninferior to surgery with respect to all-cause mortality at 1 year.

Abstract

Importance  Transcatheter aortic valve implantation (TAVI) is a less invasive alternative to surgical aortic valve replacement and is the treatment of choice for patients at high operative risk. The role of TAVI in patients at lower risk is unclear.

Objective  To determine whether TAVI is noninferior to surgery in patients at moderately increased operative risk.

Design, Setting, and Participants  In this randomized clinical trial conducted at 34 UK centers, 913 patients aged 70 years or older with severe, symptomatic aortic stenosis and moderately increased operative risk due to age or comorbidity were enrolled between April 2014 and April 2018 and followed up through April 2019.

Interventions  TAVI using any valve with a CE mark (indicating conformity of the valve with all legal and safety requirements for sale throughout the European Economic Area) and any access route (n = 458) or surgical aortic valve replacement (surgery; n = 455).

Main Outcomes and Measures  The primary outcome was all-cause mortality at 1 year. The primary hypothesis was that TAVI was noninferior to surgery, with a noninferiority margin of 5% for the upper limit of the 1-sided 97.5% CI for the absolute between-group difference in mortality. There were 36 secondary outcomes (30 reported herein), including duration of hospital stay, major bleeding events, vascular complications, conduction disturbance requiring pacemaker implantation, and aortic regurgitation.

Results  Among 913 patients randomized (median age, 81 years [IQR, 78 to 84 years]; 424 [46%] were female; median Society of Thoracic Surgeons mortality risk score, 2.6% [IQR, 2.0% to 3.4%]), 912 (99.9%) completed follow-up and were included in the noninferiority analysis. At 1 year, there were 21 deaths (4.6%) in the TAVI group and 30 deaths (6.6%) in the surgery group, with an adjusted absolute risk difference of −2.0% (1-sided 97.5% CI, −∞ to 1.2%; P < .001 for noninferiority). Of 30 prespecified secondary outcomes reported herein, 24 showed no significant difference at 1 year. TAVI was associated with significantly shorter postprocedural hospitalization (median of 3 days [IQR, 2 to 5 days] vs 8 days [IQR, 6 to 13 days] in the surgery group). At 1 year, there were significantly fewer major bleeding events after TAVI compared with surgery (7.2% vs 20.2%, respectively; adjusted hazard ratio [HR], 0.33 [95% CI, 0.24 to 0.45]) but significantly more vascular complications (10.3% vs 2.4%; adjusted HR, 4.42 [95% CI, 2.54 to 7.71]), conduction disturbances requiring pacemaker implantation (14.2% vs 7.3%; adjusted HR, 2.05 [95% CI, 1.43 to 2.94]), and mild (38.3% vs 11.7%) or moderate (2.3% vs 0.6%) aortic regurgitation (adjusted odds ratio for mild, moderate, or severe [no instance of severe reported] aortic regurgitation combined vs none, 4.89 [95% CI, 3.08 to 7.75]).

Conclusions and Relevance  Among patients aged 70 years or older with severe, symptomatic aortic stenosis and moderately increased operative risk, TAVI was noninferior to surgery with respect to all-cause mortality at 1 year.

Trial Registration  isrctn.com Identifier: ISRCTN57819173

Introduction

Transcatheter aortic valve implantation (TAVI) is a less invasive alternative to surgical aortic valve replacement for patients with severe, symptomatic aortic stenosis requiring intervention. The first clinical use of TAVI was in 20021 and evidence from randomized clinical trials has supported its adoption as the treatment of choice for patients who are unfit for conventional surgery2 or who are at high operative risk.3-6 Early trials used first-generation TAVI devices, which were associated with a high rate of procedural complications.7,8 Technological developments, procedural refinements, and increased operator experience have subsequently resulted in improved outcomes, and there is increasing interest in the use of TAVI in patients at lower operative risk. The UK Transcatheter Aortic Valve Implantation (UK TAVI) trial was conducted to compare TAVI with surgical aortic valve replacement in patients with severe, symptomatic aortic stenosis and moderately increased operative risk due to age or comorbidity.

Methods
Trial Design and Oversight

This was an investigator-initiated, pragmatic, multicenter, randomized clinical trial involving all National Health Service hospitals performing TAVI in the UK. Details of the participating sites and investigators appear in the eAppendix in Supplement 1. The trial was designed by the investigators and overseen by an independent trial steering committee and an independent data monitoring committee (eMethods 1 in Supplement 2). The TAVI valves and the surgical valves were procured through standard National Health Service commissioning. The trial protocol was approved by the London Stanmore research ethics committee. All participants gave written informed consent. The trial protocol appears in Supplement 3 and the statistical analysis plan appears in Supplement 4. One-year outcomes are presented herein. Follow-up to a minimum of 5 years is ongoing.

Participants

Eligible patients were aged 70 years or older with severe, symptomatic aortic stenosis and increased operative risk due to comorbidity or age. Age alone was a sufficient criterion for inclusion of patients aged 80 years or older. Eligibility was determined by a multidisciplinary team at each site based on clinical equipoise regarding the choice of intervention. Society of Thoracic Surgeons predicted mortality9,10 and European System for Cardiac Operative Risk Evaluation II11,12 risk scores were calculated but used in a discretionary manner, with no prespecified thresholds for inclusion. Patients requiring coronary revascularization were included unless only surgical revascularization was considered appropriate. Inclusion and exclusion criteria are listed in eMethods 2 in Supplement 2, with further details of the patient identification process in eMethods 3 in Supplement 2.

Ethnicity data were recorded to assess the diversity of the study population, which may have implications for the external validity of the trial. Ethnicity was determined by the staff at each site based on discussion with the participant or review of hospital records using a list of prespecified options that corresponded to those used by the UK Office for National Statistics.

Randomization

Participants were randomly assigned in a 1:1 ratio to receive TAVI or surgical aortic valve replacement (surgery). Randomization was performed using an electronic web-based system developed and hosted by the Centre for Healthcare Randomised Trials at the University of Aberdeen. The randomization used minimization, including an 80% probabilistic element with stratification for the randomization site, age group (70-79 years vs ≥80 years), and the presence of coronary artery disease considered by the multidisciplinary team to require revascularization if the patient was randomized to receive surgery. Participants and site staff were unblinded to the treatment assigned.

Interventions

Participants randomized to TAVI were treated using any valve with a CE mark (indicating conformity of the valve with all legal and safety requirements for sale throughout the European Economic Area). All aspects of the TAVI procedure, including the choice of local or general anesthesia, access route, and prior or concurrent revascularization were determined by the local clinical team. For participants randomized to surgical aortic valve replacement (surgery), the use of any commercially available valve was permitted apart from sutureless valves. All aspects of the surgical procedure and perioperative care were determined by the local clinical team. The use and choice of anticoagulant and antithrombotic therapy were at the discretion of the responsible physician.

Participants underwent clinical assessment at baseline, 6 weeks after undergoing the intervention, and 1 year after randomization. Interim telephone follow-up was performed 3 months after undergoing the intervention and 6 months after randomization. Frailty at baseline was assessed using the Fried criteria13 and the Canadian Study of Health and Aging Clinical Frailty Scale.14

Outcomes

The primary outcome was all-cause mortality at 1 year (death from any cause within 1 year from randomization). Secondary outcomes included cardiovascular death; stroke; reintervention; a composite of death or stroke; a composite of death or disabling stroke; a composite of death, disabling stroke, or reintervention; vascular complications; major bleeding events; conduction disturbances requiring permanent pacing; myocardial infarction; kidney replacement therapy; and infective endocarditis.

A list of all prespecified outcomes (eMethods 4) and the outcome definitions (eMethods 5) based on criteria from the Valve Academic Research Consortium-215 consensus document appear in Supplement 2. Only 30-day and 1-year outcomes are reported herein; longer-term follow-up is ongoing. Six of the 36 secondary outcomes are not included in this report (identified in eMethods 4 in Supplement 2). Disability after stroke was assessed using the modified Rankin Scale at 90 days.16 Outcome events were adjudicated by an end points and events committee, which was aware of the assigned treatment. The data presented are based on adjudicated outcomes.

Symptoms and functional capacity were assessed using the Canadian Cardiovascular Society angina grading system,17 the New York Heart Association classification system,18 the Nottingham Extended Activities of Daily Living Scale,19 and the 6-minute walk test.20 Cognitive function was assessed using the Mini-Mental State Examination.21 Quality of life was assessed using the 5-level EuroQol (EQ-5D-5L)22 instrument and the Minnesota Living with Heart Failure Questionnaire.23 Details of these instruments, including their ranges, directionality, and minimal clinically important differences appear in eMethods 5 in Supplement 2. Participants underwent echocardiography at baseline, 6 weeks, and 1 year. Images were analyzed by an independent core laboratory. Assessors were not informed of the scan time point or the treatment allocation, but the presence of a prosthetic valve usually will have been evident and the type of valve often will have been identifiable, so blinding was incomplete.

Sample Size

The initial sample size was 808 patients based on an assumed 1-year mortality of 15% after surgery (based on age-specific data for 2004-2008 from the UK National Adult Cardiac Surgery database24), with an absolute difference noninferiority margin of 7.5% for the evaluation of whether TAVI is noninferior to surgery. However, a prespecified interim analysis of pooled data showed lower 1-year mortality than expected and the sample size was increased based on the recommendation of the trial steering committee.

The revised sample size was based on an assumed 1-year mortality of 7.5% after surgery and a noninferiority margin of 5%. It was estimated that at least 890 participants would provide 80% power to show that the upper limit of the 1-sided 97.5% CI for the treatment difference would not be above the noninferiority margin, allowing for a 2% dropout rate.

The chosen noninferiority margin was based on the principle of balancing clinical preference for the lowest possible margin with the feasibility of recruitment in an acceptable time frame. Five percent was considered to be an acceptable margin in the collective opinion of the trial steering committee, which included clinical and lay members, noting that the margin relates to the upper limit of the 1-sided 97.5% CI for the difference in mortality that would be accepted for TAVI to be considered noninferior to surgery.

Statistical Analysis

For the primary statistical analysis, participants were included in the groups as randomly assigned. The analysis data set included all randomized participants; however, those with unknown vital status at 1 year due to withdrawal from the trial were excluded. For the primary outcome, the absolute risk difference was derived from a logistic regression model using delta method–estimated SEs.25 The logistic regression model was adjusted for randomization minimization factors and used robust SEs to account for the clustering of outcomes by randomization site. Noninferiority was met if the upper limit of the 1-sided 97.5% CI was less than 5% for the adjusted absolute difference in mortality between TAVI and surgery.

The robustness of the conclusions was assessed by a per-protocol analysis, which included the subset of participants who were treated as randomly assigned (ie, went to the catheter laboratory or operating theater for their randomly assigned intervention within 1 year of randomization even if the procedure was subsequently abandoned or converted to an alternative intervention). A sensitivity analysis also was performed in which participants with unknown vital status at 1 year were assumed to have died if randomized to TAVI and to have survived if randomized to surgery.

Event-related outcomes at 30 days after the procedure and at 1 year after randomization were analyzed using Cox proportional hazards regression models adjusted for randomization minimization factors and using robust SEs to account for the clustering of outcomes by randomization site. If applicable, participants were censored at their date of withdrawal or death. Log-log survival plots and Schoenfeld residuals were used to check the assumption of proportionality. Kaplan-Meier plots were constructed for time-to-event analyses.

Risk differences with 95% CIs also were calculated. The descriptive statistics are based on all available data. Exploratory subgroup analyses for the primary outcome were performed using logistic regression models and included covariates for the treatment, the relevant subgroup, and an interaction term for both. Unless otherwise specified, statistical analyses were performed using Stata version 15.1 (StataCorp).

For the statistical analysis of the secondary outcomes that were not event-related outcomes, continuous outcome variables were summarized as mean (SD) or median (IQR). Treatment effects were estimated using a multilevel mixed-effects model, including repeated measures of the secondary outcome variables at the relevant time points after randomization, nested within participants. The model was adjusted for the randomization factors in line with the primary analysis model, as well as for baseline values of the outcomes if appropriate. Time was added to the model as a categorical variable and treatment × time interactions were included. Robust SEs were used to account for clustering of the outcomes by randomization site. Categorical outcomes were analyzed similarly using multilevel, mixed-effects logistic regression models. The analyses used the available cases subset and the participants were included in the groups as randomly assigned, regardless of the treatment they received.

A post hoc sensitivity analysis also was performed using a fully adjusted Bayesian hierarchical joint model, which simultaneously adjusted for intermittent missing data and dropout due to death as well as randomization minimization factors. Additional details appear in eMethods 6 in Supplement 2.

The secondary outcomes were tested at a 2-sided significance level of .05. Because of the potential for type I error due to multiple comparisons, the findings for the analyses of the secondary outcomes should be interpreted as exploratory.

Results
Participants

Between April 9, 2014, and April 30, 2018, 913 participants were randomly assigned to undergo either TAVI (n = 458) or surgery (n = 455) (Figure 1). In the TAVI group, 450 participants underwent TAVI, 5 crossed over to surgery, and 3 did not receive treatment. In the surgery group, 419 participants underwent surgery, 17 participants crossed over to TAVI, and 19 participants did not receive treatment.

Baseline characteristics were well balanced between the groups (Table 1 and eTable 1 in Supplement 2). The median age was 81 years (IQR, 78-84 years) and 46.4% of the participants were women. The median Society of Thoracic Surgeons mortality risk score was 2.6% (IQR, 2.0%-3.4%). Coronary artery disease that was considered by the multidisciplinary team to require revascularization if the patient was assigned to receive surgery was present in 19.8% of the participants.

Interventions

The median time from randomization to treatment was 40 days (IQR, 22-69 days) in the TAVI group and 37 days (IQR, 21-63 days) in the surgery group. In the TAVI group, a valve was deployed in 443 of the 450 participants (98.4%) who were treated as randomized. More than 1 valve was used in 5 participants. Details of the types of valves and access routes appear in eTable 2 in Supplement 2. Of these 450 participants, conscious sedation or local (or regional) anesthesia was used in 313 (69.6%) and general anesthesia in 137 (30.4%). The median procedure duration was 82 minutes (IQR, 63-113 minutes). Coronary revascularization was performed as a staged procedure or during the same hospital admission in 33 of these 450 participants (7.3%).

In the surgery group, a valve was implanted in 416 of the 419 participants (99.3%) who were treated as randomized. Of these 419 participants, midline sternotomy was performed in 375 (89.5%) and minimally invasive surgery in 44 (10.5%). Details of the surgical valves appear in eTable 3 in Supplement 2. The median procedure duration was 182 minutes (IQR, 150-230 minutes), the median cardiopulmonary bypass time was 85 minutes (IQR, 66-106 minutes), and the median cross-clamp time was 63 minutes (IQR, 50-80 minutes). Concurrent coronary revascularization was performed in 90 of these 419 participants (21.5%).

After TAVI, the median stay in the intensive care unit was 0 days (IQR, 0-0 days) and was 0 days (IQR, 0-1 days) in the high-dependency unit. After surgery, the median stay in the intensive care unit was 1 day (IQR, 1-3 days) and was 1 day (IQR, 0-3 days) in the high-dependency unit. Further details appear in eTable 4 in Supplement 2. The percentage of participants discharged to home was 94.2% in the TAVI group and 82.6% in the surgery group. Details of anticoagulant and antithrombotic medication use at discharge and during follow-up appear in eTables 5-6 in Supplement 2.

Primary Outcome

For the primary outcome, data were unavailable for 1 patient who withdrew from all follow-up. At 1 year, 21 participants (4.6%) in the TAVI group had died compared with 30 participants (6.6%) in the surgery group (adjusted absolute risk difference, −2.0% [1-sided 97.5% CI, −∞ to 1.2%]) (Table 2). The upper limit of the 1-sided 97.5% CI (1.2%) was less than the prespecified noninferiority margin (5%), consistent with noninferiority of TAVI with respect to death from any cause at 1 year (P<.001 for noninferiority). The findings also were consistent with noninferiority in the sensitivity analysis allowing for missing data and in the per-protocol population (Table 2). The adjusted hazard ratio (HR) for death from any cause was 0.69 (95% CI, 0.38 to 1.26; Table 3). The Kaplan-Meier plot for the primary outcome appears in Figure 2A. The treatment effect was consistent across prespecified subgroups (eFigure in Supplement 2). Details of deaths are provided in eTables 7-9 in Supplement 2.

Secondary Outcomes

The median duration of the hospital stay was 3 days (IQR, 2-5 days) after TAVI and was 8 days (IQR, 6-13 days) after surgery. The event-related secondary clinical outcomes appear in Table 3. After the procedure, there were significantly fewer major bleeding events at 30 days in the TAVI group (5.5%) compared with in the surgery group (19.5%) (adjusted HR, 0.27 [95% CI, 0.19-0.37]; P < .001) and the incidence of major bleeding events at 1 year after randomization was 7.2% vs 20.2%, respectively (adjusted HR, 0.33 [95% CI, 0.24-0.45]; P < .001).

However, there was a significantly higher incidence of vascular complications at 30 days after the procedure in the TAVI group (10.1%) compared with the surgery group (2.3%) (adjusted HR, 4.43 [95% CI, 2.53-7.78]; P < .001) and the incidence of vascular complications at 1 year after randomization was 10.3% vs 2.4%, respectively (adjusted HR, 4.42 [95% CI, 2.54-7.71]; P < .001). The incidence of conduction disturbances requiring permanent pacing at 30 days after the procedure was 11.0% in the TAVI group vs 6.7% in the surgery group (adjusted HR, 1.72 [95% CI, 1.13-2.61]; P = .01) and the incidence at 1 year after randomization was 14.2% vs 7.3%, respectively (adjusted HR, 2.05 [95% CI, 1.43-2.94]; P < .001).

There was no significant difference in the rate of stroke at 30 days (2.4% in the TAVI group vs 2.3% in the surgery group; adjusted HR, 1.05 [95% CI, 0.35-3.17]; P = .94) or at 1 year (5.2% vs 2.6%, respectively; adjusted HR, 1.98 [95% CI, 0.95-4.11]; P = .07). There was no significant difference in the rate of cardiovascular death, the composite of death from any cause or nonfatal stroke, or for other clinical outcomes at 30 days after the procedure or at 1 year after randomization (Figure 2B, C, and D and Table 3).

Echocardiographic findings are reported in eTables 10-11 in Supplement 2. At 6 weeks, the mean aortic valve mean gradient was 10.36 mm Hg in the TAVI group vs 10.01 mm Hg in the surgery group (adjusted difference, 0.31 mm Hg [95% CI, −0.53 to 1.15 mm Hg]; P = .47) and the mean aortic valve effective orifice area was 1.53 cm2 vs 1.51 cm2, respectively (adjusted difference, 0.04 cm2 [95% CI, −0.02 to 0.09 cm2]; P = .22). These hemodynamic improvements were sustained and not significantly different between the groups at 1 year (eTable 10 in Supplement 2).

At 6 weeks, mild aortic regurgitation was significantly more prevalent in the TAVI group (43.7%) than in the surgery group (12.3%) as well as moderate aortic regurgitation (2.4% vs 0.9%, respectively) (adjusted odds ratio for mild, moderate, or severe aortic regurgitation combined vs none, 5.37 [95% CI, 3.86 to 7.46]; P < .001). No instances of severe aortic regurgitation were reported. At 1 year, mild aortic regurgitation was significantly more prevalent in the TAVI group (38.3%) than in the surgery group (11.7%) as well as moderate aortic regurgitation (2.3% vs 0.6%, respectively) (adjusted odds ratio for mild, moderate, or severe aortic regurgitation combined vs none, 4.89 [95% CI, 3.08 to 7.75]; P < .001).

There was a reduction in the prevalence of angina in both groups at 6 weeks and at 1 year with no statistically significant between-group difference (eTable 12 in Supplement 2). There was a significantly greater improvement in New York Heart Association class, Minnesota Living with Heart Failure Questionnaire scores, and 6-minute walk distance in the TAVI group at 6 weeks; however, there was no significant between-group difference at 1 year (eTables 13-15 in Supplement 2). There was significantly greater independence in activities of daily living after TAVI at 6 weeks but no significant difference at 1 year (eTable 16 in Supplement 2). There were no significant differences in cognitive function (eTable 17 in Supplement 2). The EuroQol EQ-5D-5L utility and EQ visual analog scale scores improved within 2 weeks after TAVI and the benefits were sustained at 1 year. In the surgery group, quality of life was diminished at 2 weeks. Quality of life improved after 6 weeks but utility and visual analog scale scores were significantly higher in the TAVI group and the utility score remained so at 1 year (eTable 18 in Supplement 2).

Adverse Events

There were a total of 483 serious adverse events in the TAVI group and 545 in the surgery group. The number of participants with at least 1 serious adverse event was 252 (55%) in the TAVI group and 255 (56%) in the surgery group (eTable 19 in Supplement 2). Details of relatedness to the interventions and the type of events (classified using the Medical Dictionary for Regulatory Activities) appear in eTables 20-21 in Supplement 2.

Discussion

In this trial that enrolled patients aged 70 years or older with severe, symptomatic aortic stenosis and moderately increased operative risk, TAVI was noninferior to surgery with respect to all-cause mortality at 1 year. These findings are concordant with those from other trials in patients with intermediate risk29,30 or low risk31-33 and in recent meta-analyses.34,35

In contrast to previous trials, this trial was pragmatic, publicly funded, and designed to compare a TAVI strategy using any valve type and access route vs a conventional surgical strategy in a broad range of patients. Inclusion was based on clinical equipoise regarding the treatment options and not bound by prespecified risk scores. Entry to the trial thus reflected a site-specific assessment of risk, encompassing factors not reflected in the risk scores such as frailty. This approach also allowed for the temporal evolution of clinicians’ individual perspectives on the risk threshold for considering TAVI as an alternative to surgery, with increasing local and global experience of the procedure as recruitment progressed. The inclusion of every center performing TAVI in the UK and having few exclusion criteria further increased the likelihood that the trial outcomes reflected effectiveness in routine clinical practice in the UK rather than efficacy under optimal conditions.

When the trial was conceived in 2009, it was envisaged that it would recruit patients at intermediate or high operative risk. However, clinical practice had evolved by the time enrollment commenced in 2014. With a median Society of Thoracic Surgeons mortality risk score of 2.6%, the trial population would conventionally be classified as low risk. This was reflected in the procedural outcomes, with 30-day mortality of 0.9% in the surgery group, which is similar to the 1.1% in the surgery group in the PARTNER 3 (Placement of Aortic Transcatheter Valves 3) trial32 and the 1.3% in the surgery group in the Evolut (Evolut Surgical Replacement and Transcatheter Aortic Valve Implantation in Low Risk Patients) trial.33 However, the 1-year surgical mortality in the current trial (6.6%) was higher than in the PARTNER 3 trial (2.5%) and in the Evolut trial (3.0%) most likely reflecting the older age, increased comorbidity, and increased prevalence of frailty in the patients in the current trial. The treatment effect was consistent in a subgroup analysis assessing the interaction of these factors and in patients with vs those without the need for coronary revascularization.

Improvements in aortic valve area and gradient were similar between the groups but mild and moderate aortic regurgitation were more prevalent after TAVI. Aortic regurgitation was predominantly mild and may partly reflect the use of earlier-generation TAVI valves in the initial recruitment phase and the inclusion of patients with bicuspid valves. The prognostic significance of mild aortic regurgitation is uncertain and long-term follow-up is required to determine its clinical effect. Concerns have been raised about the increased frequency of subclinical valve leaflet thrombosis after TAVI compared with surgery.36,37 This was not examined and no specific antithrombotic or anticoagulant regimen was mandated to prevent subclinical valve leaflet thrombosis. However, the early improvements in valve areas and gradients were sustained in both groups at 1 year. There was some late divergence of the event curves for stroke with a higher frequency in the TAVI group at 1 year, but the number of events was small and the difference was not statistically significant.

Limitations

This study has several limitations. First, the lack of site-specific screening data makes it difficult to determine what proportion of the total referrals for aortic valve replacement the trial population represents and how selected was the group that was reviewed by the multidisciplinary team at each site. Background data on all patients treated with TAVI or surgery are available from the relevant national registries26-28 and a comprehensive analysis is planned.

Second, the data presented in this report only address 1-year outcomes. Further follow-up is required to monitor clinical outcomes and the need for reintervention in the long-term. There is some uncertainty about the long-term durability of TAVI valves. Data from the UK TAVI registry found the incidence of moderate structural valve deterioration to be 8.7% and severe structural valve deterioration to be 0.4% after a median follow-up of 5.8 years; however, the data predominantly relate to early-generation devices and their reliability is limited by high mortality and possible survival bias.38 Five-year follow-up from the intermediate-risk PARTNER 2 trial showed more frequent aortic valve reintervention after TAVI (3.2%) compared with surgery (0.8%).39 Pending long-term follow-up, treatment selection should be individualized and take account of these uncertainties, particularly in younger patients with longer life expectancies.

Conclusions

Among patients aged 70 years or older with severe, symptomatic aortic stenosis and moderately increased operative risk, TAVI was noninferior to surgery with respect to all-cause mortality at 1 year.

Back to top
Article Information

Corresponding Author: William D. Toff, MD, Department of Cardiovascular Sciences, University of Leicester, Groby Road, Glenfield Hospital, Clinical Sciences Wing, Leicester LE3 9QP, England (w.toff@le.ac.uk).

Accepted for Publication: March 28, 2022.

The UK TAVI Trial Investigators Authorship Group: William D. Toff, MD; David Hildick-Smith, MD; Jan Kovac, MD; Michael J. Mullen, MD; Olaf Wendler, MD, PhD; Anita Mansouri, MSc; Ines Rombach, DPhil; Keith R. Abrams, PhD; Simon P. Conroy, MB, ChB, PhD; Marcus D. Flather, MB, BS; Alastair M. Gray, PhD; Philip MacCarthy, MB, ChB, PhD; Mark J. Monaghan, PhD; Bernard Prendergast, DM; Simon Ray, MD; Christopher P. Young, MD; David C. Crossman, MD; John G. F. Cleland, MD; Mark A. de Belder, MD; Peter F. Ludman, MD; Stephen Jones, BA; Cameron G. Densem, MD; Steven Tsui, MD; Manoj Kuduvalli, MCh; Joseph D. Mills, MD; Adrian P. Banning, MD; Rana Sayeed, BM, BCh, PhD; Ragheb Hasan, ChM; Douglas G. W. Fraser, DM; Uday Trivedi, MB, BS; Simon W. Davies, MB, BS; Alison Duncan, MB, BS, PhD; Nick Curzen, BM (Hons), PhD; Sunil K. Ohri, MD; Christopher J. Malkin, MD; Pankaj Kaul, MS, MCh; Douglas F. Muir, MB, ChB; W. Andrew Owens, MD; Neal G. Uren, MD; Renzo Pessotto, MD; Simon Kennon, MD; Wael I. Awad, MD; Saib S. Khogali, MD; Maciej Matuszewski, MD; Richard J. Edwards, MB, BS, PhD; Bandigowdanapalya C. Ramesh, MS, MCh; Miles Dalby, MD; Shahzad G. Raja, MB, BS; Giovanni Mariscalco, MD, PhD; Clinton Lloyd, MB, ChB; Ian D. Cox, MD; Simon R. Redwood, MD; Mark G. Gunning, MD; Paul D. Ridley, MD.

Affiliations of The UK TAVI Trial Investigators Authorship Group: Department of Cardiovascular Sciences, University of Leicester, Leicester, England (Toff, Kovac, Mariscalco); National Institute for Health Research Leicester Biomedical Research Centre, Glenfield Hospital, Leicester, England (Toff, Kovac, Mariscalco); Sussex Cardiac Centre, Brighton and Sussex University Hospitals NHS Trust, Brighton, England (Hildick-Smith, Trivedi); Institute of Cardiovascular Science, University College London, London, England (Mullen); Department of Cardiothoracic Surgery, King’s College Hospital NHS Foundation Trust, London, England (Wendler); Oxford Clinical Trials Research Unit, Nuffield Department of Orthopedics, Rheumatology, and Musculoskeletal Sciences, University of Oxford, Oxford, England (Mansouri, Rombach); Centre for Health Economics, University of York, York, England (Abrams); Department of Statistics, University of Warwick, Coventry, England (Abrams); Department of Health Sciences, University of Leicester, Leicester, England (Abrams, Conroy); Norwich Medical School, University of East Anglia, Norwich, England (Flather); Health Economics Research Centre, Nuffield Department of Population Health, University of Oxford, Oxford, England (Gray); Department of Cardiology, King’s College Hospital NHS Foundation Trust, London, England (MacCarthy, Monaghan); Department of Cardiology, St Thomas’ Hospital, London, England (Prendergast); Department of Cardiology, Manchester University NHS Foundation Trust, Manchester, England (Ray); Department of Cardiothoracic Surgery, St Thomas’ Hospital, London, England (Young); School of Medicine, University of St Andrews, Fife, Scotland (Crossman); Robertson Centre for Biostatistics and Glasgow Clinical Trials Unit, Institute of Health and Wellbeing, University of Glasgow, Glasgow, Scotland (Cleland); National Institute for Cardiovascular Outcomes Research, Barts Health NHS Trust, London, England (de Belder); Institute of Cardiovascular Sciences, Birmingham University, Birmingham, England (Ludman); Surgical Intervention Trials Unit, Nuffield Department of Surgical Sciences, University of Oxford, Oxford, England (Jones); Department of Cardiology, Royal Papworth Hospital, Cambridge, England (Densem); Department of Cardiothoracic Surgery, Royal Papworth Hospital, Cambridge, England (Tsui); Department of Cardiothoracic Surgery, Liverpool Heart and Chest Hospital NHS Foundation Trust, Liverpool, England (Kuduvalli); Department of Cardiology, Liverpool Heart and Chest Hospital NHS Foundation Trust, Liverpool, England (Mills); Department of Cardiology, John Radcliffe Hospital, Oxford University Hospitals NHS Foundation Trust, Oxford, England (Banning); Department of Cardiothoracic Surgery, John Radcliffe Hospital, Oxford University Hospitals NHS Foundation Trust, Oxford, England (Sayeed); Department of Cardiothoracic Surgery, Manchester University NHS Foundation Trust, Manchester, England (Hasan); Department of Cardiovascular Medicine, University of Manchester, Manchester, England (Fraser); Cardiac Department, Royal Brompton Hospital, Royal Brompton and Harefield NHS Foundation Trust, London, England (Davies, Duncan); Wessex Cardiothoracic Centre, University Hospital Southampton, Southampton, England (Curzen, Ohri); Department of Cardiology, Leeds Teaching Hospitals NHS Trust, Leeds, England (Malkin); Department of Cardiac Surgery, Leeds Teaching Hospitals NHS Trust, Leeds, England (Kaul); Department of Cardiology, James Cook University Hospital, South Tees Hospitals NHS Foundation Trust, Middlesbrough, England (Muir); Department of Cardiothoracic Surgery, James Cook University Hospital, South Tees Hospitals NHS Foundation Trust, Middlesbrough, England (Owens); Edinburgh Heart Centre, Royal Infirmary of Edinburgh, Edinburgh, Scotland (Uren, Pessotto); Barts Heart Centre, Barts Health NHS Trust, London, England (Kennon, Awad); Heart and Lung Centre, New Cross Hospital, Wolverhampton, England (Khogali, Matuszewski); Cardiothoracic Department, Newcastle upon Tyne Hospitals, Newcastle upon Tyne, England (Edwards, Ramesh); Department of Cardiology, Harefield Hospital, Royal Brompton and Harefield NHS Foundation Trust, London, England (Dalby); Department of Cardiac Surgery, Harefield Hospital, Royal Brompton and Harefield NHS Foundation Trust, London, England (Raja); Department of Cardiothoracic Surgery, Derriford Hospital, Plymouth, England (Lloyd); Department of Cardiology, Derriford Hospital, Plymouth, England (Cox); Cardiovascular Division, King’s College London, British Heart Foundation Centre of Research Excellence, Rayne Institute, St Thomas’ Hospital, London, England (Redwood); Cardiology Department, Royal Stoke University Hospital, Stoke-on-Trent, England (Gunning); Department of Cardiothoracic Surgery, Royal Stoke University Hospital, Stoke-on-Trent, England (Ridley).

Author Contributions: Ms Mansouri and Dr Rombach had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Toff, Hildick-Smith, Kovac, Mullen, Wendler, Abrams, Conroy, Flather, Gray, MacCarthy, Monaghan, Prendergast, Ray, Young.

Acquisition, analysis, or interpretation of data: Toff, Hildick-Smith, Kovac, Mullen, Wendler, Mansouri, Rombach, Abrams, Conroy, Flather, Gray, MacCarthy, Monaghan, Prendergast, Ray, Young, Crossman, Cleland, de Belder, Ludman, Jones, Densem, Tsui, Kuduvalli, Mills, Banning, Sayeed, Hasan, Fraser, Trivedi, Davies, Duncan, Curzen, Ohri, Malkin, Kaul, Muir, Owens, Uren, Pessotto, Kennon, Awad, Khogali, Matuszewski, Edwards, Ramesh, Dalby, Raja, Mariscalco, Lloyd, Cox, Redwood, Gunning, Ridley.

Drafting of the manuscript: Toff, Rombach, Abrams.

Critical revision of the manuscript for important intellectual content: Toff, Hildick-Smith, Kovac, Mullen, Wendler, Mansouri, Rombach, Abrams, Conroy, Flather, Gray, MacCarthy, Monaghan, Prendergast, Ray, Young, Crossman, Cleland, de Belder, Ludman, Jones, Densem, Tsui, Kuduvalli, Mills, Banning, Sayeed, Hasan, Fraser, Trivedi, Davies, Duncan, Curzen, Ohri, Malkin, Kaul, Muir, Owens, Uren, Pessotto, Kennon, Awad, Khogali, Matuszewski, Edwards, Ramesh, Dalby, Raja, Mariscalco, Lloyd, Cox, Redwood, Gunning, Ridley.

Statistical analysis: Mansouri, Rombach, Abrams.

Obtained funding: Toff, Hildick-Smith, Kovac, Mullen, Wendler, Abrams, Conroy, Flather, Gray, MacCarthy, Monaghan, Prendergast, Ray, Young.

Administrative, technical, or material support: Monaghan, Ray, Crossman, Cleland, Jones.

Supervision: Toff, Hildick-Smith, Kovac, Mullen, Wendler, Abrams, Conroy, Flather, Gray, MacCarthy, Monaghan, Prendergast, Ray, Young, Crossman.

Conflict of Interest Disclosures: All authors reported receiving grant funding from the National Institute for Health Research that was awarded to their institution during the conduct of the study. Dr Hildick-Smith reported receiving personal fees from Edwards Lifesciences, Boston Scientific, Medtronic, and Abbott. Dr Kovac reported proctoring and receiving personal fees from Boston Scientific, Medtronic, and Edwards Lifesciences. Dr Mullen reported receiving grants and personal fees from Edwards Lifesciences and Abbott Vascular. Dr Abrams reported being a partner/director and receiving personal fees from Visible Analytics Ltd; receiving grants from Duchenne UK and Swiss Precision Diagnostics; being a member of the National Institute for Health and Care Excellence diagnostics advisory committee; and being a National Institute for Health and Care Research senior investigator emeritus. Dr MacCarthy reported proctoring and receiving educational grants and personal fees from Edwards Lifesciences. Dr Prendergast reported receiving unrestricted educational and research grants to his institution from Edwards Lifesciences; receiving speaker fees from Edwards Lifesciences, Medtronic, and Abbott; and receiving consultancy fees from Anteris and Microport. Dr Cleland reported receiving personal fees from Abbott and Medtronic for serving on advisory boards and data and safety monitoring committees; receiving support from Abbott for a health economic analysis of the MitraClip device; nonfinancial support from Boston Scientific (access to data from a clinical trial); and receiving grants to his institution from Medtronic for a trial of a subcutaneous monitoring device. Dr Banning reported receiving an educational grant to his institution from Boston Scientific. Dr Sayeed reported serving as a company director and receiving dividends from Oxford Heart Surgery Ltd. Dr Fraser reported proctoring and receiving personal fees and speaker fees from Medtronic and receiving speaker fees from Edwards Lifesciences. Dr Duncan reported receiving personal fees from Edwards Lifesciences, Medtronic, and Abbott Vascular. Dr Curzen reported receiving grants from Boston Scientific, Haemonetics, Heartflow, and Beckmann Coulter and receiving nonfinancial support from Edwards Lifesciences, Biosensors, and Boston Scientific. Dr Malkin reported receiving personal fees from Medtronic, Abbott, and Boston Scientific. Dr Muir reported receiving personal fees from Edwards Lifesciences and Abbott Vascular. Dr Uren became an employee of Edwards Lifesciences in May 2021 and had no further input in the trial thereafter, apart from confirming approval of the final version of the submitted work. Dr Pessotto reported proctoring and receiving personal fees from Edwards Lifesciences. Dr Khogali reported serving as a consultant and proctor and receiving personal fees from Boston Scientific and Medtronic. Dr Dalby reported proctoring and receiving personal fees from Medtronic and receiving personal fees from Edwards Lifesciences and Boston Scientific. Dr Redwood reported proctoring and receiving personal fees from Edwards Lifesciences and serving as an advisory board member for Medtronic. No other disclosures were reported.

Funding/Support: The study was funded by the National Institute for Health Research Health Technology Assessment Programme (project reference 09/55/63). Additional support was provided by the National Institute for Health Research Clinical Research Network. Excess treatment costs were paid by the NHS England in England, by the National Institute for Social Care and Health Research in Wales, and by the Chief Scientist Office and NHS Research Scotland in Scotland. The transcatheter aortic valves and surgical valves were procured through standard National Health Service commissioning processes. The research governance sponsor for the study was the University of Leicester.

Role of the Funder/Sponsor: The funders/sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Group Information: A complete list of the UK TAVI Trial Investigators appears in Supplement 1 and a list of the study center staff and other individuals (nonauthor collaborators) who participated in the conduct of the trial appears in Supplement 5.

Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect those of the National Institute for Health Research or the Department of Health and Social Care.

Meeting Presentation: Presented in part at the American College of Cardiology Scientific Session/World Congress of Cardiology virtual meeting; March 29, 2020. The data have been amended since the presentation.

Data Sharing Statement: See Supplement 6.

References
1.
Cribier  A, Eltchaninoff  H, Bash  A,  et al.  Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description.   Circulation. 2002;106(24):3006-3008. doi:10.1161/01.CIR.0000047200.36165.B8PubMedGoogle ScholarCrossref
2.
Leon  MB, Smith  CR, Mack  M,  et al; PARTNER Trial Investigators.  Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery.   N Engl J Med. 2010;363(17):1597-1607. doi:10.1056/NEJMoa1008232PubMedGoogle ScholarCrossref
3.
Smith  CR, Leon  MB, Mack  MJ,  et al; PARTNER Trial Investigators.  Transcatheter versus surgical aortic-valve replacement in high-risk patients.   N Engl J Med. 2011;364(23):2187-2198. doi:10.1056/NEJMoa1103510PubMedGoogle ScholarCrossref
4.
Mack  MJ, Leon  MB, Smith  CR,  et al; PARTNER 1 Trial Investigators.  5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial.   Lancet. 2015;385(9986):2477-2484. doi:10.1016/S0140-6736(15)60308-7PubMedGoogle ScholarCrossref
5.
Adams  DH, Popma  JJ, Reardon  MJ,  et al; US CoreValve Clinical Investigators.  Transcatheter aortic-valve replacement with a self-expanding prosthesis.   N Engl J Med. 2014;370(19):1790-1798. doi:10.1056/NEJMoa1400590PubMedGoogle ScholarCrossref
6.
Gleason  TG, Reardon  MJ, Popma  JJ,  et al; CoreValve US Pivotal High Risk Trial Clinical Investigators.  5-year outcomes of self-expanding transcatheter versus surgical aortic valve replacement in high-risk patients.   J Am Coll Cardiol. 2018;72(22):2687-2696. doi:10.1016/j.jacc.2018.08.2146PubMedGoogle ScholarCrossref
7.
Généreux  P, Webb  JG, Svensson  LG,  et al; PARTNER Trial Investigators.  Vascular complications after transcatheter aortic valve replacement: insights from the PARTNER (Placement of AoRTic TraNscathetER Valve) trial.   J Am Coll Cardiol. 2012;60(12):1043-1052. doi:10.1016/j.jacc.2012.07.003PubMedGoogle ScholarCrossref
8.
Conte  JV, Hermiller  J  Jr, Resar  JR,  et al.  Complications after self-expanding transcatheter or surgical aortic valve replacement.   Semin Thorac Cardiovasc Surg. 2017;29(3):321-330. doi:10.1053/j.semtcvs.2017.06.001PubMedGoogle ScholarCrossref
9.
O’Brien  SM, Shahian  DM, Filardo  G,  et al; Society of Thoracic Surgeons Quality Measurement Task Force.  The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 2—isolated valve surgery.   Ann Thorac Surg. 2009;88(1)(suppl):S23-S42. doi:10.1016/j.athoracsur.2009.05.056PubMedGoogle ScholarCrossref
10.
Society of Thoracic Surgeons. STS short-term risk calculator. Accessed February 5, 2022. https://www.sts.org/resources/risk-calculator
11.
Nashef  SA, Roques  F, Sharples  LD,  et al.  EuroSCORE II.   Eur J Cardiothorac Surg. 2012;41(4):734-744. doi:10.1093/ejcts/ezs043PubMedGoogle ScholarCrossref
12.
EuroSCORE Study Group. EuroSCORE II calculator. Accessed February 5, 2022. http://www.euroscore.org/calc.htm
13.
Fried  LP, Tangen  CM, Walston  J,  et al; Cardiovascular Health Study Collaborative Research Group.  Frailty in older adults: evidence for a phenotype.   J Gerontol A Biol Sci Med Sci. 2001;56(3):M146-M156. doi:10.1093/gerona/56.3.M146PubMedGoogle ScholarCrossref
14.
Rockwood  K, Song  X, MacKnight  C,  et al.  A global clinical measure of fitness and frailty in elderly people.   CMAJ. 2005;173(5):489-495. doi:10.1503/cmaj.050051PubMedGoogle ScholarCrossref
15.
Kappetein  AP, Head  SJ, Généreux  P,  et al.  Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document.   J Am Coll Cardiol. 2012;60(15):1438-1454. doi:10.1016/j.jacc.2012.09.001PubMedGoogle ScholarCrossref
16.
Bonita  R, Beaglehole  R.  Recovery of motor function after stroke.   Stroke. 1988;19(12):1497-1500. doi:10.1161/01.STR.19.12.1497PubMedGoogle ScholarCrossref
17.
Campeau  L.  Letter: grading of angina pectoris.   Circulation. 1976;54(3):522-523. doi:10.1161/circ.54.3.947585PubMedGoogle ScholarCrossref
18.
Criteria Committee of the American Heart Association Affiliate NYC.  Nomenclature and Criteria for Diagnosis of Diseases of the Heart and Great Vessels. 9th ed. Little Brown & Co; 1994.
19.
Lincoln  NB, Gladman  JR.  The extended activities of daily living scale: a further validation.   Disabil Rehabil. 1992;14(1):41-43. doi:10.3109/09638289209166426PubMedGoogle ScholarCrossref
20.
Lipkin  DP, Scriven  AJ, Crake  T, Poole-Wilson  PA.  Six minute walking test for assessing exercise capacity in chronic heart failure.   Br Med J (Clin Res Ed). 1986;292(6521):653-655. doi:10.1136/bmj.292.6521.653PubMedGoogle ScholarCrossref
21.
Folstein  MF, Folstein  SE, McHugh  PR.  “Mini-Mental State”: a practical method for grading the cognitive state of patients for the clinician.   J Psychiatr Res. 1975;12(3):189-198. doi:10.1016/0022-3956(75)90026-6PubMedGoogle ScholarCrossref
22.
EuroQol Group. EuroQol EQ-5D-5L. Accessed February 5, 2022. https://euroqol.org/eq-5d-instruments/
23.
Rector  TS, Kubo  SH, Cohn  JN.  Patients’ self-assessment of their congestive heart failure, part 2: content, reliability and validity of a new measure, the Minnesota Living with Heart Failure Questionnaire.   Heart Fail. 1987;3:198-209.Google Scholar
24.
Bridgewater  B, Kinsman  R, Walton  P,  et al.  Demonstrating Quality: The Sixth National Adult Cardiac Surgery Database Report. Dendrite Clinical Systems Ltd; 2009.
25.
Kleinman  LC, Norton  EC.  What’s the risk? a simple approach for estimating adjusted risk measures from nonlinear models including logistic regression.   Health Serv Res. 2009;44(1):288-302. doi:10.1111/j.1475-6773.2008.00900.xPubMedGoogle ScholarCrossref
26.
National Institute for Cardiovascular Outcomes Research. National Adult Cardiac Surgery Audit: 2020 summary report (2016/17-2018/19 data). Accessed February 5, 2022. https://www.nicor.org.uk/wp-content/uploads/2020/12/National-Adult-Cardiac-Surgery-Audit-NACSA-FINAL.pdf
27.
Society for Cardiothoracic Surgery in Great Britain and Ireland. National cardiac surgery activity and outcome report 2002-2016. Accessed February 5, 2022. https://scts.org/_userfiles/pages/files/sctscardiacbluebook2020_11_20tnv2.pdf
28.
British Cardiovascular Intervention Society. United Kingdom national audit of transcatheter aortic valve implantation. Accessed February 5, 2022. https://www.bcis.org.uk/wp-content/uploads/2021/04/BCIS-Audit-2019-20-data-TAVI-subset-as-26-04-2020-for-web.pdf
29.
Leon  MB, Smith  CR, Mack  MJ,  et al; PARTNER 2 Investigators.  Transcatheter or surgical aortic-valve replacement in intermediate-risk patients.   N Engl J Med. 2016;374(17):1609-1620. doi:10.1056/NEJMoa1514616PubMedGoogle ScholarCrossref
30.
Reardon  MJ, Van Mieghem  NM, Popma  JJ,  et al; SURTAVI Investigators.  Surgical or transcatheter aortic-valve replacement in intermediate-risk patients.   N Engl J Med. 2017;376(14):1321-1331. doi:10.1056/NEJMoa1700456PubMedGoogle ScholarCrossref
31.
Thyregod  HG, Steinbrüchel  DA, Ihlemann  N,  et al.  Transcatheter versus surgical aortic valve replacement in patients with severe aortic valve stenosis: 1-year results from the all-comers NOTION randomized clinical trial.   J Am Coll Cardiol. 2015;65(20):2184-2194. doi:10.1016/j.jacc.2015.03.014PubMedGoogle ScholarCrossref
32.
Mack  MJ, Leon  MB, Thourani  VH,  et al; PARTNER 3 Investigators.  Transcatheter aortic-valve replacement with a balloon-expandable valve in low-risk patients.   N Engl J Med. 2019;380(18):1695-1705. doi:10.1056/NEJMoa1814052PubMedGoogle ScholarCrossref
33.
Popma  JJ, Deeb  GM, Yakubov  SJ,  et al; Evolut Low Risk Trial Investigators.  Transcatheter aortic-valve replacement with a self-expanding valve in low-risk patients.   N Engl J Med. 2019;380(18):1706-1715. doi:10.1056/NEJMoa1816885PubMedGoogle ScholarCrossref
34.
Siontis  GCM, Overtchouk  P, Cahill  TJ,  et al.  Transcatheter aortic valve implantation vs surgical aortic valve replacement for treatment of symptomatic severe aortic stenosis: an updated meta-analysis.   Eur Heart J. 2019;40(38):3143-3153. doi:10.1093/eurheartj/ehz275PubMedGoogle ScholarCrossref
35.
Kolkailah  AA, Doukky  R, Pelletier  MP, Volgman  AS, Kaneko  T, Nabhan  AF.  Transcatheter aortic valve implantation versus surgical aortic valve replacement for severe aortic stenosis in people with low surgical risk.   Cochrane Database Syst Rev. 2019;12(12):CD013319. doi:10.1002/14651858.CD013319.pub2PubMedGoogle ScholarCrossref
36.
Makkar  RR, Fontana  G, Jilaihawi  H,  et al.  Possible subclinical leaflet thrombosis in bioprosthetic aortic valves.   N Engl J Med. 2015;373(21):2015-2024. doi:10.1056/NEJMoa1509233PubMedGoogle ScholarCrossref
37.
Chakravarty  T, Søndergaard  L, Friedman  J,  et al; RESOLVE; SAVORY Investigators.  Subclinical leaflet thrombosis in surgical and transcatheter bioprosthetic aortic valves: an observational study.   Lancet. 2017;389(10087):2383-2392. doi:10.1016/S0140-6736(17)30757-2PubMedGoogle ScholarCrossref
38.
Blackman  DJ, Saraf  S, MacCarthy  PA,  et al.  Long-term durability of transcatheter aortic valve prostheses.   J Am Coll Cardiol. 2019;73(5):537-545. doi:10.1016/j.jacc.2018.10.078PubMedGoogle ScholarCrossref
39.
Makkar  RR, Thourani  VH, Mack  MJ,  et al; PARTNER 2 Investigators.  Five-year outcomes of transcatheter or surgical aortic-valve replacement.   N Engl J Med. 2020;382(9):799-809. doi:10.1056/NEJMoa1910555PubMedGoogle ScholarCrossref
×