SURTAVI: Transcatheter Aortic Valve Replacement With a Self-Expanding Bioprosthesis Compared With Surgical Aortic Valve Replacement in Patients at Intermediate Surgical Risk: First Results From the SURTAVI Clinical Trial (Presented by Michael J. Reardon, MD)

Key Question

What are the outcomes of patients with severe aortic stenosis (AS) undergoing transcatheter aortic valve replacement (TAVR) with a self-expanding bioprosthesis compared with surgical aortic valve replacement (SAVR) who are at intermediate surgical risk?

Characteristics

• 1,746 patients underwent randomization at 87 centers; 1,660 underwent either TAVR (CoreValve 84%, Evolut R 16%) or SAVR.
• Mean age: 79.8 ± 6.2 years.
• Mean STS score: 4.5 ± 1.6%.
• Primary outcome: composite of death from any cause or disabling stroke at 24 months.

Key Findings

• Primary outcome occurred in 12.6% in the TAVR group and 14.0% in the SAVR group (met non-inferiority).
• Death from any cause at 2 years was similar (11.4% in TAVR, 11.6% in SAVR).
• Disabling stroke at 2 years tended to occur more commonly after SAVR (2.6% in TAVR, 4.5% in SAVR; 95% confidence interval [CI] for difference -4.0 to 0.1).
• Echocardiographic findings: TAVR group compared with SAVR group had lower mean aortic valve gradients, larger aortic valve areas, and a higher prevalence of moderate to severe paravalvular aortic regurgitation (AR) at 1 year (5.3% in TAVR vs. 0.6% in SAVR).
• SAVR was associated with higher rates of acute kidney injury, atrial fibrillation (AF), and blood transfusion.
• TAVR was associated with a higher rate of post-procedural pacemaker implantation (25.9 vs. 6.6%).

Implications

• TAVR using a self-expanding bioprosthesis in patient at intermediate surgical risk with severe AS is non-inferior to SAVR over 2 years of follow-up.
• Post-procedural complications differ between TAVR and SAVR, most notably stroke (strong trend), and rates of atrial fillbrilation and pacemaker implantation; these may affect procedural choice and clinical management.
• The high pacemaker rate may be a barrier to widespread use of TAVR in younger and lower risk populations.
• Longer follow-up is needed to assess durability of TAVR self-expanding bioprosthesis.

STS/ACC TVT Registry: Comparison of In-Hospital and 1-Year Outcomes of Transcatheter Aortic Valve Replacement for Failed Surgical Bioprosthesis Versus Native Aortic Stenosis From the STS/ACC TVT Registry (Presented by E. Murat Tuzcu, MD)

Key Question

What are the 1-year outcomes of valve-in-valve (ViV) TAVR compared with native aortic valve (NV) TAVR?

Characteristics

• 757 ViV TAVR and 28,895 NV TAVR were included from the STS/ACC TVT Registry and linked to Medicare claims data for outcomes.
• 1:2 simple matching was done based on STS-PROM reoperative risk of 7%, inoperable or extreme risk designation, and hostile chest or porcelain aorta.
• Final cohort consisted of 750 ViV TAVR and 1,495 NV TAVR patients.
• ViV TAVR patients compared with NV TAVR were younger (79 vs. 84 years) and more often male (61 vs. 54%).
• ViV TAVR indications: 333 (48%) significant AS, 135 (17.8%) probable AS, 95 (12.5%) pure AR, 167 (22%) mixed AS and AR.
• Outcomes: All-cause mortality, stroke, and heart failure (HF) hospitalization at 30 days and 1 year.

Key Findings

• In-hospital mortality was 2% in ViV TAVR vs. 7% in NV TAVR.
• 1 year mortality was 13% in ViV TAVR vs. 23% in NV TAVR (hazard ratio (HR) 0.49, 95% CI 0.63-0.69). Results persisted with stratifying patients according to age <80 vs. ≥80.
• Stroke rate at 1 year was 2% in ViV TAVR vs. 4% in NV TAVR (HR 0.41, 95% CI 0.2-0.82).
• HF hospitalization rate at 1 year was 12.3% in ViV TAVR vs. 15.6% in NV TAVR (HR 0.71, 95% CI 0.54-0.95).
• On echocardiography, ViV TAVR had higher gradients across the aortic valve compared with NV TAVR (17 vs. 8 mmHg, p < 0.001).
• Moderate or severe AR occurred in 2.8% of ViV TAVR vs. 9.5% of NV TAVR (p = 0.05).

Implications

• ViV TAVR is safe and carried out with good outcomes.
• It is unclear why patients with ViV TAVR had significantly better outcomes than patients with NV TAVR, but it may have to do with differences in baseline characteristics, the indication for the procedure (AS vs. AR), or other factors.
• A registry-based study or a randomized trial comparing ViV TAVR vs. re-do SAVR is needed prior to changing clinical practice.

ADVICE Study: Real-World Comparative Effectiveness of Transcatheter Versus Surgical Aortic Valve Replacement in the United States: An Analysis from Two US Registries Linked to Medicare Data (Presented by J. Mathew Brennan, MD, MPH)

Key Question

What are the real-world outcomes of TAVR compared with SAVR for severe AS in the United States?

Characteristics

• 9,464 propensity-matched intermediate and high-risk (STS PROM ≥3%) patients enrolled in the STS/ACC TVT registry and linked to Medicare claims database.
• 4,732 underwent TAVR, and 4,732 underwent SAVR.
• Median age 82 years, 48% female, median STS-PROM 5.6%.
• 76.3% of TAVR patients underwent transfemoral access.
• Outcome at 1 year: discharge to home, death, stroke, days alive and out of the hospital.

Key Findings

• 3% of TAVR and 5% of SAVR patients had in-hospital death (p < 0.001).
• 2.5% of TAVR and 2.7% of SAVR patients had in-hospital stroke (p = 0.04).
• 70% of TAVR and 41% of SAVR patients were discharged home.
• 1 year mortality, stroke, and days alive and out of the hospital were not statistically different between TAVR and SAVR groups.

Implications

• TAVR is a safe and effective alternative treatment to SAVR in intermediate and high surgical risk patients with severe AS over 1 year of follow-up.
• Discharge to home is more common in TAVR group compared with the SAVR group, reflecting the less demanding early recovery.
• Longer term follow-up of such real-world data is needed because the clinical effects of differences in post-procedural complications (e.g., atrial fillbrilation, pacemaker implantation, paravalvular AR) may not become evident until later.

Procedural and Clinical Outcomes in Transcatheter Aortic Valve Replacement for Bicuspid Versus Tricuspid Aortic Valve Stenosis (Presented by Sung-Han Yoon, MD)

Key Question

What are the procedural and clinical outcomes in TAVR for bicuspid aortic valve stenosis (BAV) vs. tricuspid (TAV) aortic valve stenosis?

Characteristics

• A multi-center (33 centers) observational Bicuspid AS TAVR registry was utilized.
• Propensity-score matching design.
• 546 BAV patients were matched to 546 TAV patients. All patients underwent TAVR.
• Mean age in both groups was 77 ± 8 years. 63% males in BAV and 61% males in TAV.
• Mean STS score of 4.6 ± 4.6 in BAV vs. 4.3 ± 3.0% in TAV (p = 0.29).

Key Findings

• Procedural success occurred in 85.3% of BAV group vs. 91.4% in TAV group (p = 0.002). When only new-generation devices were included in the analysis (Sapien 3, Lotus, and Evolut R valves), procedural success was not statistically different between the groups.
• In BAV compared with TAV, there were more aortic root injury (1.6 vs. 0%, p = 0.004), second valve implantation (4.8 vs. 1.5%, p = 0.002), and paravalvular leak (10.4 vs. 6.8%, p = 0.04). When only new-generation devices were included in the analysis, there was no statistically significant difference in procedural outcomes between the groups.
• At 30 days, there was no statistically significant difference in mortality, stroke, bleeding, major vascular complications, or acute kidney injury.
• 1 year all-cause mortality occurred in 11.6% of BAV vs. 11.2% in TAV (p = 0.28).

Implications

• In older patients with BAV, TAVR is feasible and safe compared with TAVR in a TAV, with improved procedural success and outcomes using newer generation devices.
• A trial of patients with BAV randomized to TAVR vs. SAVR is needed to guide clinical practice.

RESOLVE and SAVORY: Subclinical Leaflet Thrombosis in Transcatheter and Surgical Bioprosthetic Aortic Valves: Results From RESOLVE and SAVORY Registries (Presented by Raj R. Makkar, MD)

Key Questions

What are the outcomes of subclinical leaflet thrombosis (SLT) in patients who underwent TAVR vs. SAVR? What is the impact of novel oral anticoagulants (NOACs) and warfarin on SLT and its clinical outcomes?

Characteristics

• 931 patients were enrolled (RESOLVE: 657 patients; SAVORY: 274 patients).
• 890 patients had interpretable computed tomography scans (CTs) (median time to CT was 83 days), of which 138 underwent SAVR and 752 underwent TAVR.
• Patients with SLT were older (82 vs. 78.9 years, p < 0.001) and had less AF (16% vs. 29.9%, p = 0.003).

Key Findings

• SLT occurred in 4% of SAVR patients vs. 13% of TAVR patients (p < 0.01).
• In the TAVR group, SLT occurred in 7.2% of CoreValve, 13.8% of Sapien, 13.3% of Lotus, and 30.8% in Portico.
• SLT occurred more frequently in patients on dual antiplatelet therapy (15%) compared with patients on oral anticoagulants (4%) (p < 0.0001).
• NOACs (3%) and warfarin (4%) had similar rates of SLT (p = 0.72).
• In a subgroup of patients who underwent a follow-up echocardiogram, 14% (12/88) with SLT vs. 1% (7/632) without SLT had mean AV gradients >20 mmHg and an increase in AV mean gradient >10 mmHg (p < 0.0001).
• Stroke rates were not statistically different between SLT group (4.12 strokes per 100 person-years) and no SLT group (1.92 strokes per 100 person-years). However, transient ischemic attack rates were higher (4.18 vs. 0.6 transient ischemic attacks per 100 person-years; HR 5.89, 95% CI 1.87-18.62).
• In a subgroup of patients who underwent a follow-up CT scan; SLT resolved in 36/36 patients receiving NOACs (n = 12) or warfarin (n = 24) but persisted in 20/22 patients not on oral anticoagulants (p < 0.0001).

Implications

• SLT is more frequent in TAVR patient than in SAVR patients. SLT is associated with hemodynamic deterioration of the implanted valve. Oral anticoagulation could help in treating SLT. Stroke rates were not statistically different between both groups; however, transient ischemic attack rates were higher in SLT group.
• Further studies are needed to establish whether SLT can be prevented by short-term anticoagulation after TAVR, whether SLT leads to further valve deterioration over longer follow-up, whether it has a causal relationship to stroke, and whether treating with oral anticoagulants leads to better clinical outcomes.

Silent Cerebral Microbleeds During TAVR: Insight From a Prospective Cerebral MRI Cohort (Presented by Eric Van Belle, MD, PhD)

Key Questions

Are silent cerebral microbleeds (CMB) associated with the TAVR procedures? What are the clinical and procedural predictors of CMB occurring during TAVR procedures?

Characteristics

• 84 patients enrolled and completed the imaging protocol.
• Detailed neurological assessments was done pre-TAVR and at 6 months post-TAVR.
• Cerebral magnetic resonance imaging (MRI) was performed before and after the TAVR procedure.
• Outcomes: incidence of CMB, post-procedural events, and mortality at 1 year.

Key Findings

• On pre-procedural MRI, 26% of patients had at least one CMB.
• Post-TAVR, CMB developed in 19 (23%) asymptomatic patients.
• History of bleeding, longer fluoroscopy time, prolonged exposure to anticoagulation, balloon post-dilation, and post-procedural acquired Von Willebrand disease were all predictors of new post-TAVR CMB.
• The presence of CMB at baseline and at 6 months follow-up was associated with impaired cognitive function. No relationship with mortality was found.

Implications

• There is a high incidence of new CMB after TAVR procedures. Although its presence is associated with impaired cognitive function, its effect on clinical outcomes and how the incidence can be modulated remain to be defined.

Clinical Outcomes at 1-Year After Commercial Transcatheter Mitral Valve Repair in the United States (Presented by Paul Sorajja, MD)

Key Question

What are the acute, 30-day, and 1-year outcomes of commercial transcatheter mitral valve repair with the MitraClip system in the United States?

Characteristics

• 2,952 patients were enrolled in the STS/ACC TVT registry. Of those, 1,867 (63%) were linked to Medicare claims data.
• Acute procedural success was defined as achieving mitral regurgitation (MR) ≤2, without surgery or death.
• Median age was 82, 56% males; 86% had degenerative MR only.
• Median STS-PROM was 6.1% for MV repair and 9.2% for MV replacement.

Key Findings

• Acute procedural success was achieved in 91.8% of all patients enrolled in the STS/ACC TVT registry. In-hospital mortality was 2.7%, and 85.9% were discharged home.
• Of the 1,867 patients who were linked to Medicare data, 25.9% died at 1 year, and 20.2% had HF rehospitalization; 37.9% either died or were rehospitalized for HF by 1 year.
• At 1 year, 6.2% required repeat MitraClip procedure, and 2.1% underwent MV surgery.
• Predictors of all-cause mortality at 1 year: age, dialysis, moderate to severe lung disease, left ventricular ejection fraction, severe tricuspid regurgitation, and residual grade III and IV MR.

Implications

• The real-world data in the United States show the acute safety and effectiveness of the MitraClip system for MR.
• Further studies are needed to enhance patient selection given the high all-cause mortality rate at 1 year.

Comparative Effectiveness of Left Atrial Appendage Occlusion Among Atrial Fibrillation Patients Undergoing Cardiac Surgery: A Report From the Society of Thoracic Surgeons Adult Cardiac Surgery Database (Presented by Daniel J. Friedman, MD)

Key Question

What is the effectiveness of surgical left atrial appendage occlusion (S-LAAO) in the reduction of thromboembolism (TE) among patients with AF?

Characteristics

• 10,524 patients (37% had S-LAAO) were enrolled from STS registry and linked to Medicare claims data.
• Median age: 75 years.
• Surgeries: 35% coronary artery bypass grafting (CABG), 35% aortic procedures ± CABG, and 30% mitral procedures ± CABG.
• Primary endpoint: 1 year readmission for TE.
• Secondary endpoint: hemorrhagic stroke, death, and a composite endpoint.
• Inverse probability weighted analysis was done.

Key Findings

• 1.6% of patients who underwent S-LAAO had an admission for TE at 1 year compared with 2.5% of patients without S-LAAO; adjusted HR of 0.62 (95% CI 0.46-0.83).
• S-LAAO was associated with a reduction in all-cause mortality at 1 year (HR 0.85, CI 0.74-0.97). No reduction in the risk of hemorrhagic stroke was observed.
• When patients were stratified according to discharge anticoagulation status, and after the adjustment of concomitant surgical AF ablation, S-LAAO was associated with less TE among patients discharged without anticoagulation (HR 0.29, CI 0.14-0.60) but not among patients discharged on anticoagulation.

Implications

• In this retrospective evaluation of a heterogeneous population, after adjustment for surgical AF ablation, S-LAAO is associated with a reduced incidence of TE, but this benefit is confined to those patients discharged without anticoagulation.
• Although S-LAAO appears to be effective, a randomized trial is needed prior to large-scale adoption in clinical practice.

Keywords: ACC Annual Scientific Session, ACC17, Aortic Valve, Aortic Valve Stenosis, Atrial Appendage, Atrial Fibrillation, Bicuspid, Bioprosthesis, Heart Valve Prosthesis, Thrombosis, Transcatheter Aortic Valve Replacement, Tricuspid Valve Stenosis

< Back to Listings