Advertisement

Causes of death in intermediate-risk patients: The Randomized Surgical Replacement and Transcatheter Aortic Valve Implantation Trial

Open ArchivePublished:December 13, 2018DOI:https://doi.org/10.1016/j.jtcvs.2018.11.129

      Abstract

      Objectives

      Examine the causes and timing of death in the Surgical Replacement and Transcatheter Aortic Valve Implantation intermediate-risk randomized trial for transcatheter aortic valve replacement (TAVR) and surgical aortic valve replacement (SAVR).

      Methods

      Causes of death were adjudicated by an independent clinical event committee and by post-hoc hierarchical classification. Causes of death were evaluated and characteristics and procedural parameters compared between patients who died and survivors for 3 time periods: early (0-30 days), recovery (31-120 days), and late (121-365 days).

      Results

      All-cause mortality at 1 year was 6.5% after TAVR and 6.7% after SAVR. There were no differences in mortality rates between TAVR and SAVR for any of the 3 time periods. Early mortality was primarily due to technical, procedure-related problems in TAVR and due to complications in SAVR. For TAVR and SAVR, most deaths during recovery were caused by complications. Other causes, including comorbid conditions, accounted for most late deaths.

      Conclusions

      Mortality rates were similar for patients treated with TAVR or SAVR at any time period including at 1 year. Early cause of death was more commonly technical failure after TAVR and due to complications after SAVR. Recovery phase cause of death was dominated by complications from TAVR and SAVR. Late cause of death appeared to be independent of the procedure in both groups.

      Key Words

      Abbreviations and Acronyms:

      AF (atrial fibrillation), AKI (acute kidney injury), AS (aortic stenosis), CEC (Clinical Event Committee), MI (myocardial infarction), PPM (patient-prosthesis mismatch), SAVR (surgical aortic valve replacement), STS PROM (Society of Thoracic Surgeons Predicted Risk of Mortality), SURTAVI (Surgical Replacement and Transcatheter Aortic Valve Implantation), TAVR (transcatheter aortic valve replacement)
      Figure thumbnail fx1
      Hierarchical causes of death after TAVR and SAVR for 3 time periods through 1 year.
      Mortality in intermediate-risk TAVR and SAVR patients is similar and driven by technical, procedural problems in TAVR and complications from SAVR.
      This analysis found similar survival for intermediate-risk patients undergoing TAVR and SAVR during 3 time periods through 1 year. Results suggest that survival may primarily be improved by prevention of technical, procedural problems in TAVR (reducing early death) and prevention of complications after TAVR and SAVR (reducing deaths within 30 days), and during the recovery period after TAVR and SAVR.
      See Commentary on page 729.
      Transcatheter aortic valve replacement (TAVR), first performed in 2002,
      • Cribier A.
      • Eltchaninoff H.
      • Bash A.
      • Borenstein N.
      • Tron C.
      • Bauer F.
      • et al.
      Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description.
      was initially intended as an alternative treatment for severe, symptomatic aortic stenosis (AS) in patients unable to undergo surgery. Randomized controlled trials in patients deemed inoperable, or at extreme or high surgical risk have consistently shown TAVR to have comparable, if not superior, survival compared with medical management or surgery.
      • Adams D.H.
      • Popma J.J.
      • Reardon M.J.
      • Yakubov S.J.
      • Coselli J.S.
      • Deeb G.M.
      • et al.
      Transcatheter aortic-valve replacement with a self-expanding prosthesis.
      • Leon M.B.
      • Smith C.R.
      • Mack M.
      • Miller D.C.
      • Moses J.W.
      • Svensson L.G.
      • et al.
      Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery.
      • Popma J.J.
      • Adams D.H.
      • Reardon M.J.
      • Yakubov S.J.
      • Kleiman N.S.
      • Heimansohn D.
      • et al.
      Transcatheter aortic valve replacement using a self-expanding bioprosthesis in patients with severe aortic stenosis at extreme risk for surgery.
      Recently, 2 randomized trials in patients at intermediate surgical risk have shown TAVR to be noninferior for the end point of mortality or disabling stroke to surgery.
      • Leon M.B.
      • Smith C.R.
      • Mack M.J.
      • Makkar R.R.
      • Svensson L.G.
      • Kodali S.K.
      • et al.
      Transcatheter or surgical aortic-valve replacement in intermediate-risk patients.
      • Reardon M.J.
      • Van Mieghem N.M.
      • Popma J.J.
      • Kleiman M.S.
      • Sondergaard L.
      • Mumtaz M.
      • et al.
      Surgical or transcatheter aortic-valve replacement in intermediate-risk patients.
      As TAVR is applied in patients with lower surgical risk, the differences in outcomes between patients undergoing surgical aortic valve replacement (SAVR) and TAVR may become less pronounced.
      In the CoreValve US Pivotal High Risk Trial, TAVR was associated with superior survival compared with SAVR at 1 year,
      • Adams D.H.
      • Popma J.J.
      • Reardon M.J.
      • Yakubov S.J.
      • Coselli J.S.
      • Deeb G.M.
      • et al.
      Transcatheter aortic-valve replacement with a self-expanding prosthesis.
      and the differences in mortality between the 2 treatment groups were most distinct in patients who died 1 to 4 months postprocedure.
      • Gaudiani V.
      • Deeb G.M.
      • Popma J.J.
      • Adams D.H.
      • Gleason T.G.
      • Conte J.V.
      • et al.
      Causes of death from the randomized CoreValve US pivotal high-risk trial.
      Gaudiani and colleagues
      • Gaudiani V.
      • Deeb G.M.
      • Popma J.J.
      • Adams D.H.
      • Gleason T.G.
      • Conte J.V.
      • et al.
      Causes of death from the randomized CoreValve US pivotal high-risk trial.
      showed that high-risk patients treated with SAVR during this period died primarily due to a failure to adequately repair the AS or the inability to recover from the more invasive procedure, and patients treated with TAVR were influenced by procedure-related complications.
      The Surgical Replacement and Transcatheter Aortic Valve Implantation (SURTAVI) trial randomized patients with severe symptomatic AS at intermediate surgical risk to either TAVR with a self-expanding valve, or SAVR. TAVR was shown to be noninferior to SAVR for the primary end point of all-cause mortality or disabling stroke.
      • Reardon M.J.
      • Van Mieghem N.M.
      • Popma J.J.
      • Kleiman M.S.
      • Sondergaard L.
      • Mumtaz M.
      • et al.
      Surgical or transcatheter aortic-valve replacement in intermediate-risk patients.
      The number of patients in the United States and Germany being treated with TAVR has recently surpassed those treated with SAVR.
      • Eggebrecht H.
      • Mehta R.H.
      Transcatheter aortic valve implantation (TAVI) in Germany 2008-2014: on its way to standard therapy for aortic valve stenosis in the elderly?.
      With TAVR currently being studied in younger patients at low surgical risk (ClinicalTrials.gov identifiers: NCT02675114 and NCT02701283), the proportion of patients undergoing TAVR will likely expand if the trials show positive results. We hypothesized that the causes of death for TAVR and SAVR patients are different at different time periods and understanding this will allow us to focus on mitigating these and improving survival. The objective of this report is to examine the timing and causes of death in the first year postprocedure for patients treated with TAVR or SAVR in the SURTAVI trial.

      Methods

      Study Overview

      The present study is a post hoc analysis of the SURTAVI trial. The SURTAVI trial was a randomized, controlled, noninferiority trial performed at 87 sites in Canada, Europe, and the United States, comparing the safety and efficacy of TAVR versus SAVR in patients with symptomatic, severe AS at intermediate surgical risk. Details of the study design and primary outcomes have been previously reported.
      • Reardon M.J.
      • Van Mieghem N.M.
      • Popma J.J.
      • Kleiman M.S.
      • Sondergaard L.
      • Mumtaz M.
      • et al.
      Surgical or transcatheter aortic-valve replacement in intermediate-risk patients.
      In brief, eligible patients were randomized to TAVR or SAVR, with randomization stratified by clinical site and the need for coronary revascularization. An independent Clinical Events Committee (CEC) adjudicated all major clinical events, including cause of death, and study oversight was provided by an independent data safety monitoring board. The SURTAVI study protocol was approved by each local institutional review board or medical ethics committee. All patients provided written informed consent.

      Patients

      Patient selection procedures and details of the study population have been previously described.
      • Reardon M.J.
      • Van Mieghem N.M.
      • Popma J.J.
      • Kleiman M.S.
      • Sondergaard L.
      • Mumtaz M.
      • et al.
      Surgical or transcatheter aortic-valve replacement in intermediate-risk patients.
      In brief, patients with severe symptomatic AS who were deemed to be at intermediate surgical risk (a Society of Thoracic Surgeons Predicted Risk of Mortality [STS PROM] 3%-15%) were eligible to participate in SURTAVI. Other non-STS factors, such as frailty, were also considered. An international screening committee reviewed each patient to determine eligibility for TAVR.

      Procedure Details

      Patients randomized to SAVR underwent surgical valve replacement and coronary revascularization if indicated. The type and size of the bioprosthetic valve were based on operator preference. Patients randomized to the TAVR group received a 23-, 26-, 29-, or 31-mm CoreValve (Medtronic, Minneapolis, Minn) or a 23-, 26-, or 29 mm Evolut R self-expanding bioprosthesis (Medtronic), preferably by transfemoral access. The size of the bioprosthesis was based on preprocedural multidetector computed tomography. If indicated, percutaneous coronary intervention was performed either as a staged procedure before TAVR or as a concomitant intervention during TAVR.

      Analysis

      The analysis cohort for this report comprised 1660 patients who underwent attempted implant from June 2012 to June 2016 of a transcatheter (n = 864) or surgical (n = 796) bioprosthesis based on randomization. Deaths occurring through July 29, 2017, are reported. In addition to the CEC adjudication, individual patient histories were retrospectively reviewed by 2 cardiovascular surgeons (HA and MJR) to classify the cause of death in a hierarchical fashion, consistent with the method applied by Gaudiani and colleagues.
      • Gaudiani V.
      • Deeb G.M.
      • Popma J.J.
      • Adams D.H.
      • Gleason T.G.
      • Conte J.V.
      • et al.
      Causes of death from the randomized CoreValve US pivotal high-risk trial.
      Recognizing that the final cause of death is often the result of a chain of events that may be initiated before the actual death of a patient, the most probable initiating factor was determined using 5 hierarchical categories (Table 1). Patient deaths were distributed based on 3 time periods: early (0-30 days, with day 0 the day of the procedure), recovery (31-120 days), and late (121-365 days).
      • Gaudiani V.
      • Deeb G.M.
      • Popma J.J.
      • Adams D.H.
      • Gleason T.G.
      • Conte J.V.
      • et al.
      Causes of death from the randomized CoreValve US pivotal high-risk trial.
      The analysis for each time period included patients with known death or survivals at the start and end of the time period.
      Table 1Hierarchical classification of the causes of death
      Reprinted with permission.7
      • Technical problems with the procedure
        • Injury to cardiac structure
        • Lack of myocardial protection
        • Induced ischemia
        • Vascular complication requiring intervention and transfusion of >3 U
      • Failure to repair disease adequately
        • Mean gradient >20 mm Hg
        • Moderate or greater paravalvular leak
        • Failure to correct known significant coronary issues
      • Complications (linked to death)
        • Acute kidney injury
        • Major or life-threatening bleeding with transfusion
        • New atrial fibrillation which does not resolve by discharge
      • Failure to recover
        • Not able to spend ≥14 consecutive days at home (or where patient started from)
      • Other (not directly related to the procedure or disease repair)
        • Malignancy
        • Late falls
        • Late surgeries
        • Accidents
      Reprinted with permission.
      • Gaudiani V.
      • Deeb G.M.
      • Popma J.J.
      • Adams D.H.
      • Gleason T.G.
      • Conte J.V.
      • et al.
      Causes of death from the randomized CoreValve US pivotal high-risk trial.
      Categorical variables are presented as numbers and percentages and compared with Fisher exact test or the χ2 test where appropriate. Continuous variables are presented as mean ± standard deviation and compared with the Student t test. Estimates of all-cause mortality were determined by the Kaplan-Meier method for each time period. The number at risk for each time period included patients with known status at the start of the time period. For each time period, baseline characteristics, procedure outcomes, and clinical outcomes were compared between those who died and survived. The instantaneous hazard function was estimated from the nonparametric Kaplan-Meier survival function using the Epanechnikov kernel smoothing method with a bandwidth of 25 grid points. For any comparison between groups, a P value < .05 was considered to indicate statistical significance. All statistical analyses were performed with the use of SAS software, version 9.4 (SAS Institute Inc, Cary, NC).

      Results

      Overall baseline characteristics have been previously reported and showed no difference between the TAVR and SAVR groups.
      • Reardon M.J.
      • Van Mieghem N.M.
      • Popma J.J.
      • Kleiman M.S.
      • Sondergaard L.
      • Mumtaz M.
      • et al.
      Surgical or transcatheter aortic-valve replacement in intermediate-risk patients.
      The mean age for patients in the SAVR group was 79.7 ± 6.1 years and for the TAVR patients it was 79.9 ± 6.2 years. The mean STS PROM was 4.5% ± 1.6% in the SAVR group and 4.4% ± 1.5% in the TAVR group. There were 438 men (55.0%) in the SAVR group and 498 men (57.6%) in the TAVR group. All patients in the SURTAVI trial now have completed 1-year follow-up. A total of 56 patients in the TAVR group died (6.5%) and 52 patients in the SAVR group died (6.5%) during the first year postprocedure. The instantaneous risk of death, estimated by the hazard function, is shown in Figure 1. The distribution of the hierarchical causes of death after TAVR and SAVR for the 3 time periods are shown in Figure 2 and discussed in detail below.
      Figure thumbnail gr1
      Figure 1Instantaneous hazard curve of all-cause mortality after transcatheter aortic valve replacement (TAVR) and surgical aortic valve replacement (SAVR). The instantaneous hazard function was estimated from the nonparametric Kaplan-Meier survival function using the Epanechnikov kernel smoothing method with a bandwidth of 25 grid points.
      Figure thumbnail gr2
      Figure 2The number of deaths in each follow-up period based on the 5 hierarchical causes of death categories defined by Gaudiani and colleagues
      • Gaudiani V.
      • Deeb G.M.
      • Popma J.J.
      • Adams D.H.
      • Gleason T.G.
      • Conte J.V.
      • et al.
      Causes of death from the randomized CoreValve US pivotal high-risk trial.
      are shown for the transcatheter aortic valve replacement (TAVR) and surgical aortic valve replacement (SAVR) patients. shows the types of deaths include in each of the 5 categories.

      Early Mortality (0-30 Days)

      Within the first 30 days postprocedure 31 patients died: 18 patients in the TAVR group (2.1%) and 13 patients in the SAVR group (1.6%) (P = .50) (Table 2). There were 846 survivors in the TAVR group and 778 survivors in the SAVR group (Table 3). Within the immediate postprocedure period (≤72 hours) 9 patients died: 5 in the TAVR group and 4 in the SAVR group. More than 1 valve was implanted in 58 patients (6.7% of the TAVR group); 57 in the patients treated with the CoreValve bioprosthesis and 1 treated with the Evolut R valve. A detailed comparison of baseline characteristics and procedural outcomes between patients who died and those who survived during each of the follow-up periods is presented in the Table E1. There were no differences in baseline characteristics for the patients in the TAVR group who died in the first 30 days postprocedure compared with those who survived. For the patients in the SAVR group who died versus those who survived within the first 30 days, more patients who died had a history of cerebrovascular disease (46.2% vs 15.8%; P = .003) and home oxygen use (15.4% vs 2.4%; P = .04).
      Table 2Mortality rates for patients in the transcatheter aortic valve replacement (TAVR) and surgical aortic valve replacement (SAVR) groups for each of the 3 time periods
      0 to 30 d31 to 120 d121 to 365 d
      TAVRSAVRP value
      Based on log-rank test.
      TAVRSAVRP value
      Based on log-rank test.
      TAVRSAVRP value
      Based on log-rank test.
      All-cause mortality18 (2.1)13 (1.6).5010 (1.2)16 (2.1).1628 (3.4)23 (3.1).76
      Cardiovascular17 (2.0)13 (1.6).617 (0.8)12 (1.6).1816 (2.0)16 (2.2).76
      Noncardiovascular1 (0.1)0 (0.0).343 (0.4)4 (0.5).6212 (1.5)7 (1.0).36
      Values are presented as n (Kaplan-Meier rates as percentage). TAVR, Transcatheter aortic valve replacement; SAVR, surgical aortic valve replacement.
      Based on log-rank test.
      Table 3Clinical outcomes for patients in the transcatheter aortic valve replacement (TAVR) and surgical aortic valve replacement (SAVR) groups for patients who died and survived for the 3 time periods
      Outcome0 to 30 d31 to 120 d121 to 365 d
      TAVRSAVRTAVRSAVRTAVRSAVR
      Dead (n = 18)Alive (n = 846)Dead (n = 13)Alive (n = 778)Dead (n = 10)Alive (n = 833)Dead (n = 16)Alive (n = 753)Dead (n = 28)Alive (n = 777)Dead (n = 23)Alive (n = 691)
      Stroke5 (67.2)
      P < .05 for dead versus alive.
      23 (2.7)
      P < .05 for dead versus alive.
      2 (19.2)41 (5.3)1 (14.3)33 (3.1)5 (66.1)
      P < .05 for dead versus alive.
      45 (5.6)
      P < .05 for dead versus alive.
      2 (9.2)36 (4.6)5 (23.9)
      P < .05 for dead versus alive.
      38 (5.5)
      P < .05 for dead versus alive.
       Disabling4 (45.3)
      P < .05 for dead versus alive.
      6 (0.7)
      P < .05 for dead versus alive.
      2 (19.2)
      P < .05 for dead versus alive.
      17 (2.2)
      P < .05 for dead versus alive.
      1 (14.3)
      P < .05 for dead versus alive.
      9 (0.8)
      P < .05 for dead versus alive.
      4 (62.5)
      P < .05 for dead versus alive.
      19 (2.1)
      P < .05 for dead versus alive.
      2 (9.2)
      P < .05 for dead versus alive.
      11 (1.4)
      P < .05 for dead versus alive.
      5 (23.9)
      P < .05 for dead versus alive.
      14 (2.0)
      P < .05 for dead versus alive.
      Atrial fibrillation
      New or worsening atrial fibrillation not adjudicated by the Clinical Events Committee.
      4 (41.3)
      P < .05 for dead versus alive.
      105 (12.4)
      P < .05 for dead versus alive.
      6 (58.3)334 (42.9)0 (0.0)133 (14.9)12 (76.6)
      P < .05 for dead versus alive.
      330 (43.4)
      P < .05 for dead versus alive.
      8 (32.0)135 (17.4)11 (71.7)310 (44.9)
      Reintervention1 (7.1)
      P < .05 for dead versus alive.
      6 (0.7)
      P < .05 for dead versus alive.
      0 (0.0)1 (0.1)0 (0.0)14 (1.3)2 (16.7)
      P < .05 for dead versus alive.
      0 (0.0)
      P < .05 for dead versus alive.
      1 (6.7)15 (1.9)0 (0.0)1 (0.1)
      Major bleed0 (0.0)58 (6.9)0 (0.0)27 (3.5)0 (0.0)72 (8.2)3 (18.8)
      P < .05 for dead versus alive.
      36 (4.1)
      P < .05 for dead versus alive.
      6 (22.2)
      P < .05 for dead versus alive.
      73 (9.4)
      P < .05 for dead versus alive.
      2 (10.0)38 (5.5)
      LTDB12 (100)
      P < .05 for dead versus alive.
      37 (4.4)
      P < .05 for dead versus alive.
      3 (23.1)
      P < .05 for dead versus alive.
      44 (5.7)
      P < .05 for dead versus alive.
      1 (10.0)44 (4.9)5 (100)
      P < .05 for dead versus alive.
      48 (5.8)
      P < .05 for dead versus alive.
      3 (12.9)46 (5.9)3 (27.0)45 (6.5)
      Transfusion >1 U
      Any homologous blood product.
      12 (66.7)
      P < .05 for dead versus alive.
      109 (12.9)
      P < .05 for dead versus alive.
      10 (76.9)371 (47.7)2 (20.0)106 (12.7)11 (68.8)355 (47.1)9 (32.1)
      P < .05 for dead versus alive.
      93 (12.0)
      P < .05 for dead versus alive.
      16 (69.6)
      P < .05 for dead versus alive.
      322 (46.6)
      P < .05 for dead versus alive.
      Major vascular complication7 (39.4)
      P < .05 for dead versus alive.
      44 (5.2)
      P < .05 for dead versus alive.
      1 (7.7)
      P < .05 for dead versus alive.
      8 (1.0)
      P < .05 for dead versus alive.
      2 (32.5)
      P < .05 for dead versus alive.
      45 (5.3)
      P < .05 for dead versus alive.
      0 (0.0)8 (1.1)3 (10.7)40 (5.1)0 (0.0)7 (1.0)
      AKI8 (61.0)
      P < .05 for dead versus alive.
      40 (4.7)
      P < .05 for dead versus alive.
      6 (48.1)
      P < .05 for dead versus alive.
      121 (15.6)
      P < .05 for dead versus alive.
      1 (10.0)40 (4.7)6 (37.5)
      P < .05 for dead versus alive.
      114 (15.1)
      P < .05 for dead versus alive.
      2 (7.1)36 (4.6)6 (26.1)101 (14.6)
      MI3 (25.3)
      P < .05 for dead versus alive.
      4 (0.5)
      P < .05 for dead versus alive.
      2 (15.4)
      P < .05 for dead versus alive.
      5 (0.6)
      P < .05 for dead versus alive.
      1 (10.0)
      P < .05 for dead versus alive.
      6 (0.7)
      P < .05 for dead versus alive.
      2 (12.5)
      P < .05 for dead versus alive.
      7 (0.8)
      P < .05 for dead versus alive.
      2 (8.4)
      P < .05 for dead versus alive.
      10 (1.3)
      P < .05 for dead versus alive.
      0 (0.0)6 (0.9)
      Cardiac perforation6 (33.3)
      P < .05 for dead versus alive.
      8 (0.9)
      P < .05 for dead versus alive.
      1 (7.7)
      P < .05 for dead versus alive.
      5 (0.6)
      P < .05 for dead versus alive.
      0 (0.0)8 (1.0)0 (0.0)5 (0.7)0 (0.0)8 (1.0)0 (0.0)5 (0.7)
      Cardiac tamponade5 (27.8)
      P < .05 for dead versus alive.
      7 (0.8)
      P < .05 for dead versus alive.
      1 (7.7)
      P < .05 for dead versus alive.
      9 (1.2)
      P < .05 for dead versus alive.
      0 (0.0)8 (1.0)1 (6.3)9 (1.2)0 (0.0)8 (1.0)2 (8.7)
      P < .05 for dead versus alive.
      7 (1.0)
      P < .05 for dead versus alive.
      Endocarditis0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)1 (0.1)1 (7.7)
      P < .05 for dead versus alive.
      2 (0.1)
      P < .05 for dead versus alive.
      0 (0.0)2 (0.3)2 (28.8)
      P < .05 for dead versus alive.
      1 (0.1)
      P < .05 for dead versus alive.
      Valve thrombosis0 (0.0)1 (0.1)0 (0.0)0 (0.0)0 (0.0)1 (0.1)0 (0.0)0 (0.0)0 (0.0)2 (0.3)0 (0.0)0 (0.0)
      AV hospitalization0 (0.0)24 (2.8)0 (0.0)32 (4.1)2 (40.0)
      P < .05 for dead versus alive.
      54 (5.8)
      P < .05 for dead versus alive.
      2 (12.5)43 (5.6)11 (50.5)
      P < .05 for dead versus alive.
      57 (7.3)
      P < .05 for dead versus alive.
      3 (25.5)48 (6.9)
      PPI
      Including patients with a pacemaker at baseline.
      2 (14.9)218 (25.8)1 (14.3)50 (6.4)1 (10.0)230 (27.1)2 (12.5)55 (6.9)8 (28.6)224 (28.8)2 (8.7)58 (8.4)
      Coronary artery obstruction1 (5.6)
      P < .05 for dead versus alive.
      1 (0.1)
      P < .05 for dead versus alive.
      0 (0.0)0 (0.0)0 (0.0)1 (0.1)0 (0.0)0 (0.0)0 (0.0)1 (0.1)0 (0.0)0 (0.0)
      Procedural conversion4 (23.6)
      P < .05 for dead versus alive.
      7 (0.8)
      P < .05 for dead versus alive.
      1 (7.7)
      P < .05 for dead versus alive.
      1 (0.1)
      P < .05 for dead versus alive.
      1 (10.0)
      P < .05 for dead versus alive.
      6 (0.7)
      P < .05 for dead versus alive.
      0 (0.0)1 (0.1)1 (3.6)5 (0.6)0 (0.0)1 (0.1)
      To open surgery3 (17.9)
      P < .05 for dead versus alive.
      4 (0.5)
      P < .05 for dead versus alive.
      0 (0.0)0 (0.0)0 (0.0)4 (0.5)0 (0.0)0 (0.0)1 (3.6)3 (0.4)0 (0.0)0 (0.0)
      TAVR to other percutaneous0 (0.0)1 (0.1)0 (0.0)0 (0.0)0 (0.0)1 (0.1)0 (0.0)0 (0.0)0 (0.0)1 (0.1)0 (0.0)0 (0.0)
      SAVR to alternative procedure0 (0.0)0 (0.0)1 (7.7)
      P < .05 for dead versus alive.
      0 (0.0)
      P < .05 for dead versus alive.
      0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)0 (0.0)
      Aborted procedure1 (5.6)
      P < .05 for dead versus alive.
      2 (0.2)
      P < .05 for dead versus alive.
      0 (0.0)1 (0.1)0 (0.0)0 (0.0)0 (0.0)1 (0.1)0 (0.0)1 (0.1)0 (0.0)1 (0.1)
      Values are presented as mean ± standard deviation or number of patients with an event (Kaplan-Meier rates as percentage). TAVR, Transcatheter aortic valve replacement; SAVR, surgical aortic valve replacement; LTDB, life-threatening or disabling bleeding; AKI, acute kidney injury; MI, myocardial infarction; AV, aortic valve; PPI, permanent pacemaker implantation.
      P < .05 for dead versus alive.
      New or worsening atrial fibrillation not adjudicated by the Clinical Events Committee.
      Any homologous blood product.
      § Including patients with a pacemaker at baseline.

      Procedure differences

      More patients in the TAVR group who died within the first 30 days required revascularization (41.2% vs 19.1%; P = .02), had more than 1 valve implanted (23.5% vs 6.4%; P = .02) and had a significantly longer procedure time (93.8 ± 62.3 minutes vs 51.6 ± 31.5 minutes; P = .02). There were no significant procedure-related differences in the SAVR group, yet 1 patient died in the early period due to annular disruption.

      Clinical outcomes at 30 days

      Major clinical outcomes for patients who died and survived are shown in Table 3. Compared with survivors, patients who died in the TAVR group had a higher incidence of new stroke, atrial fibrillation (AF), myocardial infarction (MI), acute kidney injury (AKI), major vascular complications, coronary obstruction, and procedural conversion. Patients in the TAVR group who died during this early period also more often received a transfusion 66.7% vs 12.9%; P < .001). Patients who died in the SAVR group more often had a new disabling stroke, major vascular complication, AKI, MI, and procedural conversion (Table 3).

      Causes of death

      The causes of death based on the CEC adjudication are shown in Table 4. All early deaths were due to cardiovascular causes, except for 1 patient in the TAVR group who died due to infection/sepsis. In both groups, procedural complications were the most frequent cause of death, but this cause was more common in the TAVR group than in the SAVR group (61.1% vs 38.5%). By hierarchical classification (Table 4), the most frequent causes of death among TAVR patients were technical procedure-related issues in 11 patients (61.1%) followed by fatal complications in 6 patients (33.3%), while 11 of 13 (84.6%) among SAVR patients were due to complications (Figure 2).
      Table 4Clinical Events Committee (CEC) adjudication and hierarchical classification of deaths per follow-up period
      0 to 30 d31 to 120 d121 to 365 d
      TAVR (n = 864)SAVR (n = 796)TAVR (n = 846)SAVR (n = 778)TAVR (n = 833)SAVR (n = 753)
      No. of deaths181310162823
      CEC adjudication
       Cardiovascular17 (94.4)13 (100)7 (70.0)12 (75.0)16 (57.1)16 (69.6)
      Valve-related0001 (6.3)00
      Complications of the procedure11 (61.1)5 (38.5)2 (20.0)3 (18.8)00
      Sudden/unwitnessed1 (5.6)01 (10.0)3 (18.8)2 (7.1)1 (4.3)
      Worsening heart failure001 (10.0)2 (12.5)3 (10.7)1 (4.3)
      Myocardial infarction1 (5.6)1 (7.7)001 (3.6)0
      Arrhythmia01 (7.7)0001 (4.3)
      Cardiac tamponade1 (5.6)00000
      Neurological event2 (11.1)1 (7.7)1 (10.0)2 (12.5)2 (7.1)2 (8.7)
      Other vascular disease1 (5.6)00001 (4.3)
      Other02 (15.4)1 (10.0)002 (8.7)
      Unknown03 (23.1)1 (10.0)1 (6.3)8 (28.6)8 (34.8)
       Noncardiovascular1 (5.6)03 (30.0)4 (25.0)12 (42.9)7 (30.4)
      Malignancy0002 (12.5)01 (4.3)
      Accidental00002 (7.1)0
      Infection/sepsis1 (5.6)03 (30.0)05 (17.9)5 (21.7)
      Renal disease00002 (7.1)0
      COPD000001 (4.3)
      Other0002 (12.5)3 (10.7)0
      Hierarchical classification
       Technical11 (61.1)2 (15.4)1 (10.0)2 (12.5)00
       Failure to repair1 (5.6)0001 (2.7)0
       Complications6 (33.3)11 (84.6)9 (90.0)7 (43.8)01 (4.3)
       Failure to recover0001 (6.3)01 (4.3)
       Other0006 (37.5)27 (97.3)21 (91.3)
      Values are presented as n (%). TAVR, Transcatheter aortic valve replacement; SAVR, surgical aortic valve replacement; CEC, Clinical Events Committee; COPD, chronic obstructive pulmonary disease.

      Recovery Period Mortality (31-120 Days)

      During the recovery period, 10 patients in the TAVR group (1.2%) and 16 patients in the SAVR group (2.1%) died (P = .16) (Table 2). There were 833 survivors in the TAVR group and 753 survivors in the SAVR group (Table 3). Patients who died in the TAVR group during the recovery period had a significantly higher STS PROM score compared with those who survived (5.5% ± 2.0% vs 4.4% ± 1.5%; P = .02) (Table E1).

      Clinical outcomes from 31 to 120 days

      Compared with survivors, patients who died in the TAVR group had a higher incidence of disabling stroke, vascular complication, MI, and aortic valve-related hospitalization (Table 3). Patients in the SAVR group who died had a higher incidence of reintervention, stroke, major bleeding, endocarditis, new or worsening AF, AKI, and MI.

      Causes of death from 31 to 120 days

      Most deaths were due to cardiovascular causes (70% and 75% in the TAVR and SAVR group, respectively) with none of the subcategories being highly predominant (Table 4). All 3 noncardiovascular deaths in the TAVR group were due to infection/sepsis. By hierarchical classification (Table 4), complications were the most frequent cause of death in both groups (90.0% and 43.8% in the TAVR and SAVR group, respectively), although 37.5% of deaths in SAVR patients were classified as Other (ie, not directly related to the procedure or disease repair).

      Late Mortality (121-365 Days)

      During the late period, 28 patients (3.4%) in the TAVR group died and 23 patients (3.1%) in the SAVR group died (P = .76) (Table 2). There were 777 survivors in the TAVR group and 691 survivors in the SAVR group (Table 3). The only differences in baseline characteristics for those who died compared to survivors was a significantly higher STS PROM score in the TAVR patients who died (5.1% ± 1.6% vs 4.4% ± 1.5%; P = .01) and more patients who died in the SAVR group had a history of AF (56.5% vs 25.6%; P = .001).

      Clinical outcomes from 121 to 365 days

      In the TAVR group, the incidence of disabling stroke, MI, major bleeding, and aortic valve-related hospitalizations were significantly higher in the patients who died 121 to 365 days postprocedure (Table 3). For the SAVR group, the incidence of stroke and endocarditis were significantly higher in the patients who died.

      Causes of death from 121 to 365 days

      Like the recovery period, deaths were predominantly cardiovascular, with no specific subcategory being a leading cause of death (Table 4). In nearly 50% of patients in each group, the specific underlying cause of death could not be determined. The inciting cause of death for patients in the TAVR group who died from complications was frequently due to respiratory complications; pneumonia and aspiration were the most common causes.
      In most of the deaths, hierarchical classification (Table 4) identified causes other than those directly related to the procedure or disease repair (96.4% and 91.3% for TAVR and SAVR patients, respectively).

      Discharge Disposition

      Overall, patients undergoing TAVR were more frequently discharged home than SAVR patients (85.1% vs 54.3%) (Table 5). More SAVR patients were discharged to a rehabilitation facility (22.2% vs 6.7% of TAVR patients).
      Table 5Discharge location
      TAVR (n = 830)SAVR (n = 688)P value
      Discharge location<.001
      Died in hospital13 (1.5)11 (1.4)
      Home735 (85.1)432 (54.3)
      Rehabilitation clinic58 (6.7)177 (22.2)
      Skilled nursing facility34 (3.9)106 (13.3)
      Another hospital11 (1.3)29 (3.6)
      Other13 (1.5)41 (5.2)
      Values are presented as n (%).

      Discussion

      In this intermediate-risk patient population, TAVR with the CoreValve and Evolut R self-expanding valves or SAVR have similar all-cause mortality during all time periods examined, including at 1 year,
      • Van Mieghem N.
      • Reardon M.J.
      • Popma J.J.
      • Windecker S.
      • Gleason T.
      • LEe J.
      • et al.
      Transcatheter aortic valve replacement with a self-expanding prosthesis or surgical aortic valve replacement in intermediate-risk patients: complete 1-year outcome from the SURTAVI trial [abstract TCT-107].
      but with different inciting causes. Insights into the causes of death following cardiac surgery or intervention may help improve outcomes. Recognizing that death after cardiovascular procedures may follow a cascade of events, we applied the methodology of categorizing the inciting cause of death in a hierarchical manner as previously reported in patients from the CoreValve US Pivotal High-Risk Trial.
      • Gaudiani V.
      • Deeb G.M.
      • Popma J.J.
      • Adams D.H.
      • Gleason T.G.
      • Conte J.V.
      • et al.
      Causes of death from the randomized CoreValve US pivotal high-risk trial.
      Unlike the striking temporal differences in mortality risk reported in the CoreValve US Pivotal High-Risk Trial,
      • Gaudiani V.
      • Deeb G.M.
      • Popma J.J.
      • Adams D.H.
      • Gleason T.G.
      • Conte J.V.
      • et al.
      Causes of death from the randomized CoreValve US pivotal high-risk trial.
      the instantaneous mortality hazards in the SURTAVI trial (Figure 1) were similar for patients undergoing TAVR and SAVR, highlighting the narrowing of differences between patients undergoing these 2 different methods for replacing an aortic valve. Using our hierarchical method of categorizing the causes of mortality, the primary cause of death in the early postprocedure period for the TAVR patients was due to technical reasons and due to complications in the midterm recovery period. During the immediate postprocedure period (<72 hours)
      • Kappetein A.P.
      • Head S.J.
      • Genereux P.
      • Piazza N.
      • van Mieghem N.M.
      • Blackstone E.H.
      • et al.
      Updated standardized endpoint definitions for transcatheter aortic valve implantation: the valve academic research consortium-2 consensus document (VARC-2).
      the number of patients who died in each group was similar (4 SAVR group deaths and 5 TAVR group deaths). The primary cause of death in SAVR patients was due to complications in the early and recovery periods, consistent with the adjudication of the CEC. The most notable finding is that the proportion of SAVR patients (30 out of 67 [44.8%]) who died due to failure to alleviate the aortic stenosis or failure to recover seen with the CoreValve US Pivotal High-Risk Trial,
      • Gaudiani V.
      • Deeb G.M.
      • Popma J.J.
      • Adams D.H.
      • Gleason T.G.
      • Conte J.V.
      • et al.
      Causes of death from the randomized CoreValve US pivotal high-risk trial.
      was greatly reduced in SURTAVI (2 out of 52 [3.8%]), highlighting the ability of this lower-risk patient group to recover from the physiological insult of surgery.
      The major cause of death for SAVR patients was due to complications in the early and midterm recovery periods. Complications most frequently included AKI, AF, bleeding, and cardiogenic shock. These complications were often associated with the need for inotropic support and the final event was often multiorgan dysfunction and sepsis.
      AKI was significantly higher in the TAVR patients who died early in our study and has been previously reported to be associated with higher mortality following TAV.
      • Arsalan M.
      • Squiers J.J.
      • Farkas R.
      • Worley C.
      • Herbert M.
      • Stewart W.
      • et al.
      Prognostic usefulness of acute kidney injury after transcatheter aortic valve replacement.
      Preconditioning and limiting or dilution of contrast use may offset acute procedure-related impact on kidney function. AKI has also been previously identified as an independent predictor of mortality following cardiac surgery.
      • Kandler K.
      • Jensen M.E.
      • Nilsson J.C.
      • Moller C.H.
      • Steinbruchel D.A.
      Acute kidney injury is independently associated with higher mortality after cardiac surgery.
      Based on our hierarchical categories, procedural technical failures were the primary cause of early mortality in the TAVR patients. Of the 11 technical deaths, 4 patients experienced left ventricle perforation, and 6 required a second valve due to malposition or paravalvular leak. Vascular complications and valve related rehospitalizations were more frequent in patients who died early. The majority (84%) of TAVR patients in the SURTAVI trial were treated with an earlier generation CoreValve bioprosthesis. The second-generation Evolut R valve was added later during enrollment accounting for the remaining 16% of cases. This newer generation valve has a lower vascular profile (ie, can be implanted in vessels ≥5 mm [≥5.5 mm with the 34-mm valve] vs ≥6 mm with the CoreValve bioprosthesis), can be partially or fully recaptured to assist with valve positioning and has shown low mortality rates in high and extreme-risk patients of 8.6% at 1 year, and lower rates of moderate or severe paravalvular leak.
      • Popma J.J.
      • Reardon M.J.
      • Khabbaz K.
      • Harrison J.K.
      • Hughes G.C.
      • Kodali S.
      • et al.
      Early clinical outcomes after transcatheter aortic valve replacement using a novel self-expanding bioprosthesis in patients with severe aortic stenosis who are suboptimal for surgery: results of the Evolut R U.S. study.
      • Popma J.J.
      • Hughes G.C.
      • Kodali S.
      • George I.
      • Oh J.K.
      • Slater J.
      • et al.
      Transcatheter aortic valve replacement with a repositionable self-expanding bioprosthesis in patients with severe aortic stenosis at high risk for surgery: one-year results from the Evolut R US pivotal study [abstract TCT-37].
      The ability to reposition the Evolut R valve may allow more precise positioning and reduce the need for a second valve. Stroke following either TAVR or SAVR is associated with higher mortality,
      • Gleason T.G.
      • Schindler J.T.
      • Adams D.H.
      • Reardon M.J.
      • Kleiman N.S.
      • Caplan L.R.
      • et al.
      The risk and extent of neurologic events are equivalent for high-risk patients treated with transcatheter or surgical aortic valve replacement.
      which we also found in our study.
      The influence of AF on SAVR- and TAVR-related mortality is complex, due to cardioembolic risks associated with AF as well as potential bleeding complications related to antithrombotic therapy. Randomized AVR trials have shown new-onset AF occurs more often after SAVR than after TAVR.
      • Adams D.H.
      • Popma J.J.
      • Reardon M.J.
      • Yakubov S.J.
      • Coselli J.S.
      • Deeb G.M.
      • et al.
      Transcatheter aortic-valve replacement with a self-expanding prosthesis.
      • Leon M.B.
      • Smith C.R.
      • Mack M.J.
      • Makkar R.R.
      • Svensson L.G.
      • Kodali S.K.
      • et al.
      Transcatheter or surgical aortic-valve replacement in intermediate-risk patients.
      • Reardon M.J.
      • Van Mieghem N.M.
      • Popma J.J.
      • Kleiman M.S.
      • Sondergaard L.
      • Mumtaz M.
      • et al.
      Surgical or transcatheter aortic-valve replacement in intermediate-risk patients.
      The presence of AF at baseline was more common in patients undergoing SAVR.
      • Reardon M.J.
      • Van Mieghem N.M.
      • Popma J.J.
      • Kleiman M.S.
      • Sondergaard L.
      • Mumtaz M.
      • et al.
      Surgical or transcatheter aortic-valve replacement in intermediate-risk patients.
      In our study, the incidence of baseline AF did not differ between those who died and survivors, but patients who died early had a higher incidence of AF documented. Baseline and new-onset AF have been associated with increased mortality.
      • Tarantini G.
      • Mojoli M.
      • Windecker S.
      • Wendler O.
      • Lefevre T.
      • Saia F.
      • et al.
      Prevalence and impact of atrial fibrillation in patients with severe aortic stenosis undergoing transcatheter aortic valve replacement: an analysis from the SOURCE XT prospective multicenter registry.
      Bleeding events differ between patients undergoing SAVR and TAVR, because TAVR patients experience bleeding events often associated with vascular complications,
      • Conte J.V.
      • Hermiller Jr., J.
      • Resar J.R.
      • Deeb G.M.
      • Gleason T.G.
      • Adams D.H.
      • et al.
      Complications after self-expanding transcatheter or surgical aortic valve replacement.
      • Genereux P.
      • Webb J.G.
      • Svensson L.G.
      • Kodali S.K.
      • Satler L.F.
      • Fearon W.F.
      • et al.
      Vascular complications after transcatheter aortic valve replacement: insights from the PARTNER (Placement of AoRTic TraNscathetER Valve) trial.
      which has been associated with higher risk of 30-day and 1-year mortality.
      • Genereux P.
      • Webb J.G.
      • Svensson L.G.
      • Kodali S.K.
      • Satler L.F.
      • Fearon W.F.
      • et al.
      Vascular complications after transcatheter aortic valve replacement: insights from the PARTNER (Placement of AoRTic TraNscathetER Valve) trial.
      Within the TAVR group, transfusions in the early period were significantly higher in those who died. The reason for the transfusion in the early period for TAVR is usually related to a procedure complication such as vascular injury or ventricular perforation. For SAVR patients in the early period, the need for transfusion was higher than for TAVR patients during this period. Major bleeding within 30-days postprocedure was a positive predictor of 1-year mortality following TAVR.
      • Généreux P.
      • Cohen D.J.
      • Williams M.R.
      • Mack M.
      • Kodali S.K.
      • Svensson L.G.
      • et al.
      Bleeding complications after surgical aortic valve replacement compared with transcatheter aortic valve replacement: insights from the PARTNER I trial (Placement of Aortic Transcatheter Valve).
      Reducing early vascular complications and bleeding in TAVR may have positive longer-term consequences.
      After 120 days, most of the deaths in both groups were categorized as other using our system, and were likely related more to the patients underlying comorbid condition, than the procedural insult.
      Patient prosthetic mismatch (PPM) has previously been studied,
      • Zorn III, G.L.
      • Little S.H.
      • Tadros P.
      • Deeb G.M.
      • Gleason T.G.
      • Heiser J.
      • et al.
      Prosthesis–patient mismatch in high-risk patients with severe aortic stenosis: a randomized trial of a self-expanding prosthesis.
      with severe mismatch more common in surgery patients. Severe mismatch was associated with increased mortality at 1 year for TAVR and SAVR in that study of high-risk patients.
      • Zorn III, G.L.
      • Little S.H.
      • Tadros P.
      • Deeb G.M.
      • Gleason T.G.
      • Heiser J.
      • et al.
      Prosthesis–patient mismatch in high-risk patients with severe aortic stenosis: a randomized trial of a self-expanding prosthesis.
      For our intermediate-risk study, severe PPM was more common in surgery than TAVR but did not influence early or 1-year survival. A manuscript with a more detailed analysis of PPM in this trial is under review.
      Although the failure to recover from the procedure was an important driver for mortality in high-risk SAVR patients,
      • Gaudiani V.
      • Deeb G.M.
      • Popma J.J.
      • Adams D.H.
      • Gleason T.G.
      • Conte J.V.
      • et al.
      Causes of death from the randomized CoreValve US pivotal high-risk trial.
      this was not the case in this intermediate-risk cohort with complications shown to be a major cause of death, especially during the early and recovery phases. This likely reflects this group's lower risk and ability to withstand and recover from the physiologic insult of surgery. In SURTAVI, more patients in both groups were discharged home than in the CoreValve US Pivotal High-Risk Trial.
      • Gaudiani V.
      • Deeb G.M.
      • Popma J.J.
      • Adams D.H.
      • Gleason T.G.
      • Conte J.V.
      • et al.
      Causes of death from the randomized CoreValve US pivotal high-risk trial.
      Whereas nearly all TAVR patients were discharged home, about 50% of SAVR patients were discharged to a step-down facility.
      Overall, our results suggest that prevention of technical, procedure-related complications during TAVR may further reduce the incidence of fatal outcome during the early postprocedural period while reducing complications may further improve early survival after SAVR and overall survival for both TAVR and SAVR.

      Limitations

      Figure thumbnail fx2
      Video 1Recorded presentation from the American Association for Thoracic Surgery Conference May 1, 2018. Video available at: https://www.jtcvs.org/article/S0022-5223(18)33290-2/fulltext.
      The analysis addressed all-cause mortality at 1-year postprocedure in patients at intermediate risk for surgery. As such patients may be of younger age than typical high-risk patients, longer-term follow-up is particularly important for this population and warrants further research. Other aspects not explicitly assessed in this analysis may influence longer-term survival after TAVR and SAVR, such as PPM and paravalvular leak. These phenomena may have different incidences and develop differently at longer-term in TAVR versus SAVR patients.

      Conclusions

      All-cause mortality at 1 year was similar for intermediate-risk patients undergoing TAVR or SAVR, with no differences in mortality during the early, recover, or late time periods. The primary cause of early death in the TAVR patients was related to the procedure, and in the SAVR patients, the primary cause of early death was postoperative complications. The greatest opportunity for decreasing mortality for both TAVR and SAVR is in the procedural period, when the greatest instantaneous risk of death occurred. There is a potential for further reduction of all-cause mortality after TAVR by prevention of technical, procedure-related problems. Further reduction of complications may improve survival of intermediate-risk patients after SAVR and TAVR.

      Conflict of Interest Statement

      Dr Amrane serves as a consultant to Abbott and has received research grants from Medtronic and Abbott . Dr Deeb serves as on an advisory board and as a proctor for Medtronic, as a consultant and research investigator for Edwards Lifesciences, as a consultant and proctor for Terumo, and as a research investigator for Gore Medical. He receives no personal remunerations. Dr Williams serves as a consultant for Edwards Lifesciences and Medtronic, as a speaker for Abbott Laboratories, and has received research grants from Medtronic . Dr Popma has received institutional research grants from Medtronic , Boston Scientific , and Direct Flow Medical , and has served on a Medical Advisory Board for Boston Scientific, Cordis Corporation, and Edwards Lifesciences. Dr Mumtaz has served as a proctor, served as a consultant, and received honoraria, consulting fees, and travel reimbursements from Medtronic, Abbott, Edwards Lifesciences, and Atricure. Dr Kappetein is an employee of Medtronic. Dr Yakubov has received institutional research grants from Boston Scientific , Direct Flow , and Medtronic . Dr Van Mieghem has received grant support from Abbott Vascular , Boston Scientific , Claret Medical , and Medtronic . Dr Gleason receives institutional grant support from Medtronic but receives no personal income. Dr Serruys receives personal fees from Abbott Laboratories, AstraZeneca, Biotronik, Cardialysis, GLG research, Medtronic, Sino Medical Sciences Technology, and Societe Europa Digital Publishing, Stentys Fran, Svellte Medical Systems, Philips/Volcano, St Jude Medical and Qualimed. Dr Tadros discloses a financial relationship with Medtronic and St. Jude Medical . Dr Zorn has a financial relationship with Medtronic and Edwards Lifesciences . Dr Li is an employee and shareholder of Medtronic. Dr Reardon has received fees from Medtronic for providing educational services. All other authors have nothing to disclose with regard to commercial support.
      Jane Moore, MS, ELS, an employee of the sponsor, created all figures and tables, and drafted the introduction, methods, and results sections under the direction of Drs Amrane and Reardon, and ensured the technical accuracy of the information presented.
      Members of the SURTAVI Trial Causes of Death Working Group include Mathew R. Williams, MD (New York University-Langone Medical Center, New York, NY); Mubashir Mumtaz, MD (University of Pittsburg Medical Center Pinnacle, Wormsleyburg, Pa); Arie Pieter Kappetein, MD, PhD (Erasmus Medical Center, Rotterdam, the Netherlands); Patrick W. Serruys, MD, PhD (International Centre for Circulatory Health, National Heart Lung Institute, Imperial College London, London, United Kingdom); Peter Tadros, MD, and George L. Zorn III, MD (University of Kansas, Kansas City, Kan); Piet W. Boonstra, MD, and Ad van Boven, MD (Medisch Centrum Leeuwarden, the Netherlands); and Shuzhen Li, PhD (Statistical Services, Medtronic, Minneapolis, Minn).

      Appendix

      Figure thumbnail fx4
      Figure E1TAVR, Transcatheter aortic valve replacement; SAVR, surgical aortic valve replacement.
      Table E1Baseline characteristics for survivors and nonsurvivors for the transcatheter aortic valve replacement (TAVR) and surgical aortic valve replacement (SAVR) groups during the 3 time intervals studied
      0 to 30 d31 to 120 d121 to 365 d
      TAVRSAVRTAVRSAVRTAVRSAVR
      Dead (n = 18)Alive (n = 846)Dead (n = 13)Alive (n = 778)Dead (n = 10)Alive (n = 833)Dead (n = 16)Alive (n = 753)Dead (n = 28)Alive (n = 777)Dead (n = 23)Alive (n = 691)
      Age (y)79.7 ± 7.679.9 ± 6.281.0 ± 3.579.7 ± 6.181.1 ± 6.479.9 ± 6.281.3 ± 6.979.7 ± 6.180.5 ± 7.479.9 ± 6.281.3 ± 5.979.7 ± 6.1
      Men9 (50.0)489 (57.8)5 (38.5)431 (55.4)7 (70.0)482 (57.9)12 (75.0)418 (55.5)19 (67.9)448 (57.7)13 (56.5)379 (54.8)
      STS score (%)5.0 ± 1.64.4 ± 1.54.7 ± 1.54.5 ± 1.65.5 ± 2.0
      P < .05 for dead versus alive.
      4.4 ± 1.5
      P < .05 for dead versus alive.
      5.2 ± 2.04.5 ± 1.65.1 ± 1.6
      P < .05 for dead versus alive.
      4.4 ± 1.5
      P < .05 for dead versus alive.
      5.1 ± 2.14.5 ± 1.6
      SCr >2 mg/dL1 (5.6)13 (1.5)0 (0.0)17 (2.2)0 (0.0)13 (1.6)2 (12.5)15 (2.0)1 (3.6)12 (1.5)0 (0.0)14 (2.0)
      Peripheral vascular disease6 (33.3)260 (30.7)3 (23.1)232 (29.8)5 (50.0)253 (30.4)7 (43.8)222 (29.5)9 (32.1)238 (30.6)5 (21.7)203 (29.4)
      Cerebrovascular disease4 (22.2)147 (17.4)6 (46.2)
      P < .05 for dead versus alive.
      123 (15.8)
      P < .05 for dead versus alive.
      2 (20.0)145 (17.4)4 (25.0)119 (15.8)5 (17.9)136 (17.5)4 (17.4)110 (15.9)
      Chronic lung disease6 (33.3)299 (35.4)4 (30.8)263 (33.8)4 (40.0)292 (35.1)8 (50.0)251 (33.3)11 (39.3)271 (35.0)6 (26.1)233 (33.7)
      Coronary artery disease13 (72.2)528 (62.4)7 (53.8)500 (64.3)6 (60.0)520 (62.4)12 (75.0)483 (64.1)21 (75.0)482 (62.0)13 (56.5)442 (64.0)
      Pre-existing PPI/ICD2 (11.1)85 (10.0)0 (0.0)79 (10.2)1 (10.0)84 (10.1)2 (12.5)77 (10.2)5 (17.9)75 (9.7)5 (21.7)70 (10.1)
      Atrial fibrillation/flutter6 (33.3)237 (28.0)4 (30.8)206 (26.5)4 (40.0)232 (27.9)7 (43.8)198 (26.3)12 (42.9)210 (27.0)13 (56.5)
      P < .05 for dead versus alive.
      177 (25.6)
      P < .05 for dead versus alive.
      Home oxygen1 (5.6)17 (2.0)2 (15.4)
      P < .05 for dead versus alive.
      19 (2.4)
      P < .05 for dead versus alive.
      0 (0.0)17 (2.0)1 (6.3)17 (2.3)2 (7.1)13 (1.7)1 (4.3)14 (2.0)
      5-m gait speed >6 s8 (53.3)420 (51.7)6 (50.0)396 (53.2)5 (50.0)413 (51.7)7 (46.7)383 (53.0)14 (51.9)387 (52.0)9 (42.9)352 (53.0)
      Falls in past 6 mo2 (11.1)100 (11.8)2 (15.4)99 (12.7)1 (10.0)99 (11.9)3 (18.8)96 (12.8)3 (10.7)93 (12.0)1 (4.3)91 (13.2)
      ≥ 2 KATZ ADL deficits1 (5.6)8 (0.9)0 (0.0)7 (0.9)0 (0.0)8 (1.0)0 (0.0)7 (0.9)0 (0.0)8 (1.0)1 (4.3)6 (0.9)
      Procedural
       Procedure time (min)93.8 ± 62.3
      P < .05 for dead versus alive.
      51.6 ± 31.5
      P < .05 for dead versus alive.
      219.5 ± 110.9203.6 ± 68.381.3 ± 72.551.1 ± 30.5211.6 ± 57.4203.7 ± 68.754.7 ± 25.051.2 ± 30.7228.9 ± 66.6201.3 ± 67.5
       Concomitant CABG or PCI
      Patients in the surgical group had concomitant coronary artery bypass grafting (CABG) and patients in the transcatheter group underwent percutaneous coronary intervention (PCI).
      6 (35.3)
      P < .05 for dead versus alive.
      122 (14.4)
      P < .05 for dead versus alive.
      4 (30.8)169 (21.8)1 (10.0)120 (14.4)3 (18.8)166 (22.1)7 (25.0)108 (13.9)8 (34.8)147 (21.3)
       TAV delivery time (min)30.3 ± 35.414.7 ± 15.218.3 ± 16.814.6 ± 15.120.3 ± 24.714.4 ± 14.6
       CPS time (min)127.1 ± 106.797.4 ± 37.2105.8 ± 40.097.4 ± 37.2110.5 ± 43.796.4 ± 36.8
       Crossclamp time (min)74.2 ± 52.174.3 ± 30.084.8 ± 32.874.3 ± 30.075.4 ± 32.673.8 ± 29.8
      More than 1 valve implanted4 (23.5)
      P < .05 for dead versus alive.
      54 (6.4)
      P < .05 for dead versus alive.
      2 (20.0)51 (6.1)5 (17.9)
      P < .05 for dead versus alive.
      45 (5.8)
      P < .05 for dead versus alive.
      Values are presented as mean ± standard deviation or n (%). STS, Society of Thoracic Surgeons; SCr, serum creatinine; PPI, permanent pacemaker implantation; ICD, implantable cardioverter defibrillator; ADL, activities of daily living; CPS, cardiopulmonary bypass support.
      P < .05 for dead versus alive.
      Patients in the surgical group had concomitant coronary artery bypass grafting (CABG) and patients in the transcatheter group underwent percutaneous coronary intervention (PCI).

      Supplementary Data

      References

        • Cribier A.
        • Eltchaninoff H.
        • Bash A.
        • Borenstein N.
        • Tron C.
        • Bauer F.
        • et al.
        Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description.
        Circulation. 2002; 106: 3006-3008
        • Adams D.H.
        • Popma J.J.
        • Reardon M.J.
        • Yakubov S.J.
        • Coselli J.S.
        • Deeb G.M.
        • et al.
        Transcatheter aortic-valve replacement with a self-expanding prosthesis.
        N Engl J Med. 2014; 370: 1790-1798
        • Leon M.B.
        • Smith C.R.
        • Mack M.
        • Miller D.C.
        • Moses J.W.
        • Svensson L.G.
        • et al.
        Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery.
        N Engl J Med. 2010; 363: 1597-1607
        • Popma J.J.
        • Adams D.H.
        • Reardon M.J.
        • Yakubov S.J.
        • Kleiman N.S.
        • Heimansohn D.
        • et al.
        Transcatheter aortic valve replacement using a self-expanding bioprosthesis in patients with severe aortic stenosis at extreme risk for surgery.
        J Am Coll Cardiol. 2014; 63: 1972-1981
        • Leon M.B.
        • Smith C.R.
        • Mack M.J.
        • Makkar R.R.
        • Svensson L.G.
        • Kodali S.K.
        • et al.
        Transcatheter or surgical aortic-valve replacement in intermediate-risk patients.
        N Engl J Med. 2016; 374: 1609-1620
        • Reardon M.J.
        • Van Mieghem N.M.
        • Popma J.J.
        • Kleiman M.S.
        • Sondergaard L.
        • Mumtaz M.
        • et al.
        Surgical or transcatheter aortic-valve replacement in intermediate-risk patients.
        New Engl J Med. 2017; 376: 1321-1331
        • Gaudiani V.
        • Deeb G.M.
        • Popma J.J.
        • Adams D.H.
        • Gleason T.G.
        • Conte J.V.
        • et al.
        Causes of death from the randomized CoreValve US pivotal high-risk trial.
        J Thorac Cardiovasc Surg. 2017; 153: 1293-12301.e1
        • Database SACS
        Executive summary.
        (Available at:)
        • Eggebrecht H.
        • Mehta R.H.
        Transcatheter aortic valve implantation (TAVI) in Germany 2008-2014: on its way to standard therapy for aortic valve stenosis in the elderly?.
        EuroIntervention. 2016; 11: 1029-1033
        • Van Mieghem N.
        • Reardon M.J.
        • Popma J.J.
        • Windecker S.
        • Gleason T.
        • LEe J.
        • et al.
        Transcatheter aortic valve replacement with a self-expanding prosthesis or surgical aortic valve replacement in intermediate-risk patients: complete 1-year outcome from the SURTAVI trial [abstract TCT-107].
        J Am Coll Cardiol. 2017; 70: B47
        • Kappetein A.P.
        • Head S.J.
        • Genereux P.
        • Piazza N.
        • van Mieghem N.M.
        • Blackstone E.H.
        • et al.
        Updated standardized endpoint definitions for transcatheter aortic valve implantation: the valve academic research consortium-2 consensus document (VARC-2).
        Eur J Cardiothorac Surg. 2012; 42: S45-S60
        • Arsalan M.
        • Squiers J.J.
        • Farkas R.
        • Worley C.
        • Herbert M.
        • Stewart W.
        • et al.
        Prognostic usefulness of acute kidney injury after transcatheter aortic valve replacement.
        Am J Cardiol. 2016; 117: 1327-1331
        • Kandler K.
        • Jensen M.E.
        • Nilsson J.C.
        • Moller C.H.
        • Steinbruchel D.A.
        Acute kidney injury is independently associated with higher mortality after cardiac surgery.
        J Cardiothorac Vasc Anesth. 2014; 28: 1448-1452
        • Popma J.J.
        • Reardon M.J.
        • Khabbaz K.
        • Harrison J.K.
        • Hughes G.C.
        • Kodali S.
        • et al.
        Early clinical outcomes after transcatheter aortic valve replacement using a novel self-expanding bioprosthesis in patients with severe aortic stenosis who are suboptimal for surgery: results of the Evolut R U.S. study.
        J Am Coll Cardiol Intv. 2017; 10: 268-275
        • Popma J.J.
        • Hughes G.C.
        • Kodali S.
        • George I.
        • Oh J.K.
        • Slater J.
        • et al.
        Transcatheter aortic valve replacement with a repositionable self-expanding bioprosthesis in patients with severe aortic stenosis at high risk for surgery: one-year results from the Evolut R US pivotal study [abstract TCT-37].
        J Am Coll Cardiol. 2016; 68: B16
        • Gleason T.G.
        • Schindler J.T.
        • Adams D.H.
        • Reardon M.J.
        • Kleiman N.S.
        • Caplan L.R.
        • et al.
        The risk and extent of neurologic events are equivalent for high-risk patients treated with transcatheter or surgical aortic valve replacement.
        J Thorac Cardiovasc Surg. 2016; 152: 85-96
        • Tarantini G.
        • Mojoli M.
        • Windecker S.
        • Wendler O.
        • Lefevre T.
        • Saia F.
        • et al.
        Prevalence and impact of atrial fibrillation in patients with severe aortic stenosis undergoing transcatheter aortic valve replacement: an analysis from the SOURCE XT prospective multicenter registry.
        JACC Cardiovasc Interv. 2016; 9: 937-946
        • Conte J.V.
        • Hermiller Jr., J.
        • Resar J.R.
        • Deeb G.M.
        • Gleason T.G.
        • Adams D.H.
        • et al.
        Complications after self-expanding transcatheter or surgical aortic valve replacement.
        Semin Thorac Cardiovasc Surg. 2017; 29: 321-330
        • Genereux P.
        • Webb J.G.
        • Svensson L.G.
        • Kodali S.K.
        • Satler L.F.
        • Fearon W.F.
        • et al.
        Vascular complications after transcatheter aortic valve replacement: insights from the PARTNER (Placement of AoRTic TraNscathetER Valve) trial.
        J Am Coll Cardiol. 2012; 60: 1043-1052
        • Généreux P.
        • Cohen D.J.
        • Williams M.R.
        • Mack M.
        • Kodali S.K.
        • Svensson L.G.
        • et al.
        Bleeding complications after surgical aortic valve replacement compared with transcatheter aortic valve replacement: insights from the PARTNER I trial (Placement of Aortic Transcatheter Valve).
        J Am Coll Cardiol. 2014; 63: 1100-1109
        • Zorn III, G.L.
        • Little S.H.
        • Tadros P.
        • Deeb G.M.
        • Gleason T.G.
        • Heiser J.
        • et al.
        Prosthesis–patient mismatch in high-risk patients with severe aortic stenosis: a randomized trial of a self-expanding prosthesis.
        J Thorac Cardiovasc Surg. 2016; 151: 1014-1023.e3

      Linked Article

      • Commentary: Thank you, sir, may I TAVI another: Procedural problems and postoperative complications in SURTAVI
        The Journal of Thoracic and Cardiovascular SurgeryVol. 158Issue 3
        • Preview
          We congratulate the Surgical Replacement and Transcatheter Aortic Valve Implantation (SURTAVI) Trial Causes of Death Working Group on a fantastic presentation of the first year of data after randomization, presented by Amrane and colleagues1 in this issue of the Journal. Readers will recall that the industry-funded (Medtronic, Minneapolis, Minn) SURTAVI trial enrolled 1660 intermediate-risk patients from June 2012 to June 2016 to receive TAVR with self-expanding valves or surgical aortic valve replacement (SAVR).
        • Full-Text
        • PDF
        Open Archive