Bringing doctors together

or

A free, private network for verified physicians

Leading voices

Dr. Daniel Ambinder

Interventional & Structural Cards · UMSOM

View

Dr. Amit Goyal

Interventional & Structural Cards · UT Southwestern

View

Dr. David Cohen

Interventional & Structural Cards · Cardiovascular Research Foundation

View

Dr. Dorairaj Prabhakaran

General Cardiology

View

Dr. Pugazhendhi Vijayaraman

EP · Geisinger

View

Dr. Robert Schaller

EP · HUP

View

Dr. Trey Woods

General Cardiology · Emory Healthcare

View

Dr. Brahmajee Nallamothu

Interventional Cardiology · UMHS

View

Dr. Felix Yang

EP · NYU Langone

View

Dr. Jennifer Rymer

Interventional & Peripheral Cards · Duke University

View
  • Discover your community

  • Stay current

  • A profile that represents you

  • Find your people

About RoonPrivacyTerms

    Overview

    Hemodynamics and Mid-Term Clinical Outcomes Following Valve-in-Valve TAVR With Balloon-Expandable Valves.

    Feb 9, 2026DOI: 10.1161/circinterventions.125.015945Journal ArticleAHA

    Written by: Amr E AbbasTsuyoshi KanekoHouman KhaliliSamir R KapadiaVasilis C BabaliarosAdam B GreenbaumThomas A SchwannPradeep YadavIssam D MoussaGrant W ReedRoger J LahamMichael A MorsePedro VillablancaEvelio RodriguezJeremiah P DeptaJames M McCabeVinayak N BapatVinod H ThouraniAmar Krishnaswamy

    Find these authors on Roon

    HK
    Dr. Houman Khalili
    Interventional & Structural Cards · Memorial Healthcare System

    Methods

    The data used in this study are not publicly available, as it contains information that could compromise the privacy of participants. This retrospective analysis includes all patients undergoing transfemoral AViV- using balloon-expandable valves (SAPIEN 3, SAPIEN 3 Ultra, or SAPIEN 3 Ultra Resilia; Edwards Lifesciences, Irvine, CA) in the Society of Thoracic Surgeons (STS)/American College of Cardiology Transcatheter Valve Therapy (TVT) Registry and Centers for Medicare & Medicaid Services claims data from July 2015 to December 2023. The TVT Registry has been approved by a central institutional review board (Advarra) and granted a waiver of informed consent by the Duke University School of Medicine institutional review board under the Common Rule 45 CFR 46.3.Standardized definitions of adverse events and outcomes were based on the TVT Registry Data Coder Dictionary. Procedural, in-hospital, 30-day, and 1-year outcomes were derived from the TVT Registry and Centers for Medicare & Medicaid Services claims data. Outcomes beyond 1-year were analyzed solely through Centers for Medicare & Medicaid Services claims data. The TVT Registry was linked to Centers for Medicare & Medicaid Services claims data using probabilistic matching with patient birth date, sex, and TAVR procedure date. Patients eligible for linkage were all over the age of 65 years with Medicare coverage and enrolled in the Medicare Parts A and B fee-for-service program. The 5-year primary end points were all-cause mortality, a composite of all-cause mortality or stroke, and aortic valve reintervention. Sex-based outcomes were not studied.Continuous data were presented as mean±SD or median (interquartile range). Categorical data were presented as percentages and fractions of occurrences. Group comparisons were analyzed with the χ2 test or Fisher exact test, as appropriate, for categorical variables and the Student t test or Wilcoxon rank-sum test for continuous variables.Adjusted Cox regression models with restricted cubic splines were constructed to explore the relationship between post-AViV MG, at discharge, as a continuous variable, and 5-year all-cause mortality. Covariates used for the adjustment of the spline curve included age, aortic regurgitation, aortic valve mean gradient, atrial fibrillation, body mass index, cardiogenic shock within 24 hours, carotid stenosis, chronic lung disease, creatinine, currently on dialysis, diabetes, endocarditis, estimated glomerular filtration rate, hemoglobin level, home oxygen, hypertension, immunocompromised status, Kansas City Cardiomyopathy Questionnaire Overall Summary score, left ventricular EF <50%, male sex, mitral regurgitation, New York Heart Association functional class III/IV, pacemaker, peripheral arterial disease, porcelain aorta, previous implantable cardioverter defibrillator, prior coronary artery bypass graft surgery, prior percutaneous coronary intervention, prior stroke, prior transient ischemic attack, Society of Thoracic Surgeons score, and tricuspid regurgitation.
    A multivariable analysis was conducted to determine the relationship between discharge mean gradient and PPM on 5-year all-cause mortality following adjustment for clinically relevant baseline covariates. The baseline covariates included male sex, body mass index, infective endocarditis, pacemaker, implantable cardioverter defibrillator, prior percutaneous coronary intervention, prior CABG, prior stroke, prior transient ischemic attack, carotid stenosis, peripheral arterial disease, diabetes, currently on dialysis, chronic lung disease, immunocompromised status, prior myocardial infarction, heart failure within 2 weeks, cardiogenic shock, porcelain aorta, atrial fibrillation/flutter, hemoglobin, albumin, creatinine, Kansas City Cardiomyopathy Questionnaire Overall Summary, 5-meter walk test, left ventricular EF <50%, left main stenosis ≥50%, proximal left anterior descending coronary artery ≥70%, moderate or greater mitral regurgitation, moderate or greater tricuspid regurgitation, aortic stenosis, mitral stenosis, New York Heart Association class III/IV.
    Kaplan-Meier estimates and log-rank tests with hazard ratios were used to compare the occurrence of all-cause mortality and all-cause mortality or stroke over 5 years stratified according to the selected MG cutoffs (ie, 10 mm Hg and 20 mm Hg), left ventricular EF (ie, ≥50% or <50%), and measured PPM (ie, severe, moderate, or no PPM). Statistical analyses were performed using SAS, version 9.4. A 2-sided P<0.05 was considered statistically significant. There was no adjustment for multiplicity.

    Results

    The study population included 13 054 patients. At discharge, 2004 had a post-TAVR MG <10 mm Hg; 11 050, had MG ≥10 mm Hg; 8489, had MG < 20 mm Hg; and 4565 had ≥20 mm Hg.
    Baseline and procedural characteristics are reported in the Table for discharge MG < versus ≥10 mm Hg and in Table S1 for MG < versus ≥20 mm Hg. Other major clinical outcomes are in the accompanying article.
    Baseline and Procedural Characteristics for the <10 Versus ≥10 mm Hg Cohort
    Adjusted spline curves at 1 year (P<0.001 for nonlinearity) and 5 years (P<0.0001 for nonlinearity) demonstrated a nonlinear relationship between discharge MG and mortality (Figure 1A, 1B, 1C, and 1D). MG <10 mm Hg was associated with increased mortality compared with higher gradients. Adjusted spline curves at 1 year and at 5 years demonstrated no impact of elevated gradients at any point on mortality.
    1- and 5-year mortality according to mean discharge gradient postaortic valve-in-valve (AViV). Mean gradient (MG) <10 mm Hg was associated with increased mortality compared with higher gradients. A and B, Show adjusted spline curves for discharge MG relative to mortality at 1 year. B, Shows adjusted spline curves for discharge MG relative to mortality at 5 years. MG <10 mm Hg was associated with increased mortality compared with higher gradients.
    Transvalvular MG <10 mm Hg was associated with a significantly higher risk of death and the composite of death or stroke, but a lower reintervention risk at 5 years compared with MG ≥10 mm Hg (Figure 2A and 2B). When grouped by EF, MG <10 mm Hg with EF <50% was associated with a significantly higher risk of death and death or stroke (Figure 3A and 3B), and a similar reintervention at 5 years than MG <10 mm Hg with EF ≥50%.
    5-year all-cause mortality and mortality or stroke by discharge mean gradient (MG). A, Compares Kaplan-Meier survival probabilities for patients with discharge MGs <10 mm Hg and ≥10 mm Hg at discharge from transcatheter aortic valve-in-valve replacement. B, Compares estimates for mortality or stroke between the 2 cohorts. Estimates of mortality and mortality or stroke were significantly higher for the lower MG cohorts. HR indicates hazard ratio.
    5-year all-cause mortality and mortality or stroke by discharge mean gradient (MG) and left ventricular ejection fraction (LVEF). Results are stratified by echocardiographic MG ≥10 mm Hg and <10 mm Hg and LVEF at discharge from transcatheter aortic valve-in-valve replacement. A, Patients with lower MG and LVEF at discharge had a significantly higher risk of mortality. B, Patients with lower MG and LVEF at discharge had a significantly higher risk of mortality or stroke (4B) at 5 years. HR indicates hazard ratio.
    When the cutoff threshold was changed to an MG of 20 mm Hg, there was no difference in the risk of death and death or stroke, but a higher risk of reintervention at 5 years between patients with MG <20 mm Hg compared with MG ≥20 mm Hg. This was true for both the cohort of patients with an EF >50% and the cohort of patients with an EF <50% (Figure 4A and 4B). However, an MG <20 and EF <50% was associated with significantly higher risk of death and death or stroke, and similar reintervention at 5 years than MG <20 mm Hg and EF ≥50% (Figure S1A and S1B).
    5-year mortality or stroke by discharge mean gradient (MG) and left ventricular ejection fraction (LVEF). A, Kaplan-Meier survival probabilities for all-cause mortality or stroke stratified by an echocardiographic MG of ≥20 mm Hg and < 20 mm Hg at discharge following aortic valve-in-valve replacement. No significant differences between risk of mortality and mortality or stroke were seen between patients with MG ≥20 mm Hg and MG <20 mm Hg. B, Mortality risk by LVEF ≥50% or <50% at the same MG thresholds. LVEF cohorts saw no significant differences in survival. HR indicates hazard ratio.
    There was no difference in the risk of death at 5 years when comparing patients with severe, moderate, or no PPM. However, patients with severe PPM and EF <50% had a higher risk of death and death or stroke at 5 years than patients with severe PPM and EF ≥50% (Figure 5A and 5B). Discharge MG <10 mm Hg was a predictor of mortality at 5 years; however, severe PPM and MG ≥20 mm Hg at discharge were not.
    5-year all-cause mortality and mortality or stroke stratified by patient prosthesis mismatch (PPM). A and B compare Kaplan-Meier risks for death for patients with severe PPM and left ventricular ejection fraction (LVEF) <50% or ≥50%. Patients with severe PPM and LVEF below 50% had a higher risk of death and death or stroke than patients with severe PPM and LVEF ≥50%. B, Compares Kaplan-Meier risks for death or stroke for patients with severe PPM and LVEF <50% or ≥50%. HR indicates hazard ratio.
    The incidence of reintervention at 5 years was 6.3%. This was primarily secondary to stenosis and occurred most often by balloon valvuloplasty (39%) and surgical reintervention (44.5%) and was more common in patients with discharge MG ≥20 mm Hg (Table S2).

    Discussion

    The main findings of this study are as follows: (1) the relationship between all-cause mortality and echocardiography-derived MG after AViV replacement is nonlinear, wherein discharge low MG <10 mm Hg is associated with increased mortality, while elevated discharge gradients (≥20 mm Hg) are not associated with increased mortality at 5 years in patients with an EF ≥50% and patients with an EF <50%; (2) there is no relationship between severe measured PPM at discharge and 5-year clinical outcomes; and (3) reintervention occurs primarily in patients with increased MG ≥20 mm Hg at discharge is treated primarily by surgical aortic valve replacement and balloon valvuloplasty, and leads to a lower number of patients with MG >20 mm Hg at 1 year.
    These findings have significant implications for understanding the impact of echocardiographic hemodynamics on mid-term clinical outcomes and are consistent with the previously reported literature. The findings of the current analysis suggest that low gradients per se, or the absence of severe PPM, is not necessarily a marker of good outcomes following AViV. Rather, the EF and comorbidities need to be considered. Moreover, while reintervention occurred more commonly in patients with MG ≥20 mm Hg, it remains unclear, given the findings of this study, whether patients with MG ≥20 mm Hg exhibit worse clinical outcomes and require any reintervention based solely on the MG. This is particularly relevant because there was no difference in clinical outcomes between patients with gradients ≥ and <20 mm Hg in both the preserved (>50%) and reduced (<50%) left ventricular EF cohorts.
    We have previously demonstrated that after TAVR, echocardiographic gradients demonstrated a nonlinear and complex relationship with 2-year mortality. Compared with low (<10 mm Hg) echocardiographic gradients, intermediate gradients (10–20 mm Hg) were associated with improved 2-year survival and higher echocardiographic stroke volume index and EF.9,10 Our study aligns with previous studies that have also failed to demonstrate an association between an elevated echocardiographic MG ≥20 mm Hg and mortality following native TAVR, including the PARTNER 3 trial13 and the Society of Thoracic Surgeons/American College of Cardiology TVT balloon-expandable valve Registry6 extending out to 5 years in over 300 000 patients.9
    Following AViV, the PARTNER 2 AViV Registry,14 the VIVID (Valve-in-Valve International Data) Registry,7 and a prior analysis of the Society of Thoracic Surgeons/American College of Cardiology TVT Registry of patients undergoing AViV with the SAPIEN 3 valve at 1 year8 all failed to demonstrate an association of echocardiographic gradient ≥20 mm Hg with mortality. Recently, a study of AViV patients from the TVT Registry reported a similar nonlinear relationship between MG and mortality in all TAVR valves, with worse clinical outcomes reported at 1 year in patients with MG <10 mm Hg.15 Our study demonstrates the absence of clinical impacts at 5 years and argues against the impact of severe PPM on mid-term clinical outcomes. However, the impact of severe PPM on long-term structural valve degeneration, mortality, and functional status requires longer-term follow-up and remains unknown following TAVR.
    In native or prosthetic valve stenosis, the primary cause of an increase in blood flow velocity is a reduced valve area. A quadratic relationship exists between transaortic velocity and pressure gradient, by the Bernoulli equation, known as convective acceleration. Hence, the transvalvular velocity and echocardiographic-derived gradient are primarily a function of valvular hemodynamics and, to a lesser extent, flow and fluid dynamics. However, in the absence of valve stenosis, as in normal prosthetic valves, the transvalvular velocity and echocardiographic gradient largely reflect flow and fluid dynamics rather than valvular hemodynamics from a reduced valve area.16–18 Thus, the presence of low flow and EF, which have been associated with worse outcomes following TAVR, is also associated with lower MG.10 We have also demonstrated that with low flow, a higher incidence of severe PPM increases. Low-flow severe PPM has a worse prognosis following TAVR and surgical aortic valve replacement compared with normal-flow severe PPM, despite exhibiting a lower gradient.12
    In this study, higher echocardiographic gradients were not associated with worse mortality. This may be related to the presence of discordance between echocardiographic and invasive gradients post-TAVR. We previously demonstrated that the higher the echocardiographic gradient, the higher the discordance, and that small valve size and AViV were associated with discordance >10 mm Hg between echocardiographic and invasive gradients.19
    Echocardiography remains the primary tool to assess prosthetic valve function post-AViV and provides complementary information regarding the left ventricular function, aortic regurgitation, and temporal changes in valve function. However, this study questions the utility of a single echocardiographic measurement of an elevated gradient or severe PPM post-AViV in evaluating prosthetic valve hemodynamics and function. The change in gradients overtime, accompanied by a change in aortic calve area and dimensionless index as highlighted in Valve Academic Research Consortium-3, appears more clinically meaningful than a single post-TAVR measurement.
    Given the nonlinear relation of echocardiographic gradient to outcomes and the significant impact of left ventricular EF on clinical outcomes, assessing procedural success based on postoperative echocardiographic gradient alone, without including the left ventricular function, may not be appropriate. Thus, postprocedural echocardiographic gradients should not be used as a standalone target for additional procedures to optimize the valve, such as valvuloplasty or balloon fracture, but rather considered together with cardiac function and comorbidities. The 20 mm Hg echocardiographic gradient limit for assessing procedural success, especially in the AViV population, needs to be revisited as it is not associated with clinical outcomes in this population.
    Before aortic valve reintervention, it may be prudent to confirm the echocardiographic gradients with invasive evaluation.20 DISCORDANCE TAVR demonstrated that patients who develop an increase in echocardiographic gradient ≥20 mm Hg or exhibit ≥ Valve Academic Research Consortium-3 stage 2 structural valve deterioration ≥1 month after TAVR, with no evidence of valve thrombus on CTA, were reclassified to no structural valve deterioration in >50% and <20 mm Hg gradient in 25% of cases by invasive evaluation. Future studies should integrate the temporal changes in echocardiography-derived valve hemodynamics as demonstrated in Valve Academic Research Consortium-3 and their impact on valve durability and long-term mortality (presented at TCT 2024).
    Limitations of our study include those of a retrospective and observational analysis. Echocardiographic hemodynamics were determined at each institution, were not core-laboratory adjudicated. The current study, therefore, provides no conclusions based on changing/rising gradients or the need for further invasive and anatomic evaluation of the same. Other limitations include the fact that no flow or Doppler velocity index data are reported by echocardiography (only EF), no invasive corroboration of MG was available, and there is no independent adjudication of adverse events, with a potential for underreporting. In January 2021, the TVT Registry began capturing bioprosthetic valve fracture during the procedure. Therefore, our analysis included this outcome only from that time forward.
    Similar to TAVR, the relationship between all-cause mortality and echocardiographic MG post-AViV is nonlinear. Discharge MG <10 mm Hg following AViV is associated with higher 5-year all-cause mortality and all-cause mortality or stroke compared with higher gradients and there is no impact of severe PPM or MG ≥20 mm Hg on 5-year clinical outcomes. Echocardiographic MG and severe PPM post-AViV primarily reflect myocardial function and valve flow as determined by the EF and not optimum valve hemodynamics or clinical outcomes. Thus, postprocedural echocardiographic gradients should not be used as a standalone target for additional procedures to optimize the valve, and the 20 mm Hg echocardiographic gradient, as a marker of procedural success or call for additional reintervention, should be revisited.

    ARTICLE INFORMATION

    The statistical analysis was conducted by Edwards Life Sciences.
    Dr Abbas has received research grants and consulting fees from Edwards Lifesciences. Dr Kaneko is on an advisory board for Edwards Lifesciences and Abbott, and is a consultant for Medtronic, Johnson and Johnson, Anteris, and 4C Medical. Dr Khalili has received a research grant from Edwards Lifesciences and is a speaker for Abbott Vascular. Dr Babaliaros has received consulting fees from Abbott Vascular, Edwards Lifesciences, Medtronic, and Transmural Systems; has an equity interest in Transmural Systems; and has received institutional research support from Abbott Vascular, Ancora Heart, Boston Scientific, Edwards Lifesciences, Gore Medical, JenaValve, Medtronic, Polares Medical, St. Jude Medical, Transmural Systems, and 4C Medical. Dr Greenbaum has received consulting fees from Abbott Vascular, Edwards Lifesciences, and Medtronic; has an equity interest in Transmural Systems; and has received institutional research support from Abbott Vascular, Ancora Heart, Edwards Lifesciences, Gore Medical, JenaValve, Medtronic, Polares Medical, Transmural Systems, and 4C Medical. Dr Yadav has received consultant and speaker fees from Abbott Vascular, Dasi Simulations, Edwards Lifesciences, Medtronic, and Shockwave Medical. Dr Moussa has received speaker fees from Abbott Vascular and Janssen Pharmaceuticals. Dr Laham has received consulting fees from Abbott Vascular, Edwards Lifesciences, and Medtronic; and has received research grants from Abbott Vascular, Boston Scientific, Edwards Lifesciences, and Medtronic. Dr Morse has received consulting fees from Edwards Lifesciences. Dr Villablanca is a consultant for Edwards Lifesciences and Teleflex. Dr Rodriguez has received honoraria or consulting fees from Abbott, AtriCure, Boston Scientific, CardioMech, Edwards Lifesciences, NeoChord, Philips, and Teleflex; and has received grants and research support from Abbott, AtriCure, Boston Scientific, CardioMech, and Edwards Lifesciences. Dr Depta has received consulting or advisory board fees from Abbott, Boston Scientific, Edwards Lifesciences, W.L. Gore & Associates, Medtronic, and V-Wave. Dr McCabe has received honoraria or consulting fees from Abbott Vascular, Boston Scientific, Cardiovascular Systems Inc, Edwards Lifesciences, Medtronic, and Teleflex; and holds equity in Excision Medical. Dr Bapat has received consulting fees from Abbott Structural, Anteris, Boston Scientific, Edwards Lifesciences, Medtronic, Meril Lifesciences, and 4C Medical. Dr Thourani has received consulting or advisory fees or research grants from Abbott Vascular, Artivion, Atricure, Boston Scientific, Cryolife, Edwards Lifesciences, JenaValve, Medtronic, and Shockwave; and is a stockholder of DasiSimulations, LLC. The other authors report no conflicts. No grants, contracts, or other forms of financial support were utilized for this study. As this analysis was restricted to the Edwards Lifesciences SAPIEN 3, SAPIEN 3 Ultra, and SAPIEN 3 Ultra Resilia transcatheter heart valves, statistical analyses were performed by Edwards Lifesciences. The views or opinions presented here do not represent those of the American College of Cardiology (ACC), the Society of Thoracic Surgeons (STS), or the STS/ACC Transcatheter Valve Therapy (TVT) Registry.
    Tables S1–S2
    Figure S1
    HKDr. Houman Khalili

    Dr. Houman Khalili

    Cardiology · Interventional Cardiology & Structural
    Memorial Healthcare System
    houman