Transcatheter Aortic Valve-in-Valve Implantation in Failed Stentless Valves: A Single-Center Experience

© 2025 HMP Global. All Rights Reserved.
Any views and opinions expressed are those of the author(s) and/or participants and do not necessarily reflect the views, policy, or position of the Journal of Invasive Cardiology or HMP Global, their employees, and affiliates.

J INVASIVE CARDIOL 2025. doi:10.25270/jic/25.00006. Epub February 20, 2025.

Objectives. Valve-in-valve transcatheter aortic valve implantation (ViV-TAVI) offers an alternative to repeat cardiac surgery in selected patients. However, ViV-TAVI in previously implanted stentless valves is more challenging than in a stented valve. This single-center study aimed to analyze the outcomes of ViV-TAVI in stentless valves.

Methods. Data were retrospectively collected from 1476 patients undergoing TAVI between January 2018 and December 2023. The primary study endpoint was 30-day and follow-up mortality. Secondary outcomes included Valve Academic Research Consortium (VARC)-3 technical success, VARC-3 device success, and VARC-3 early safety.

Results. A total of 15 patients underwent ViV-TAVI within previously implanted stentless valves. The mean age of the patients was 78.1 ± 9.3 years, with a median STS-PROM score of 4.30 (3.05, 6.75). The mean time between SAVR and TAVI was 14.7 ± 3.9 years. Eight (53.3%) patients presented with severe aortic regurgitation, while 5 (33.3%) patients had severe aortic stenosis. There were no surgical conversions, but a single case of transcatheter heart valve malposition required the deployment of a second valve. One patient died within 30 days. VARC-3 technical success was achieved in 87.6% (n = 13) of patients, device success in 80% (n = 12), and early safety in 73.3% (n = 11). The median follow-up period was 2.9 (range, 1.3-4.3) years, during which 4 patients died.

Conclusions. Careful patient selection is critical for ViV-TAVI procedures because of the high risk of complications, which is increased with stentless valves in addition to the standard risks associated with the ViV procedure. Larger studies are warranted to confirm these findings.

In recent years, biological valves have been used for surgical aortic valve replacement (SAVR) far more frequently than mechanical valves.¹ This shift is partly driven by the aging population, which has led to a greater number of older patients requiring aortic valve replacement. Additionally, the age threshold for implanting biological valves has been progressively lowered, contributing to their increased use.² Currently, the most frequent type of bioprostheses employed for SAVR are stented valves, with only a minority of patients receiving stentless valves. Stentless valves can be effectively employed in the surgical management of endocarditis, aortic aneurysms, aortic dissections, and deteriorated aortic valves.^3,4 These alternatives provide optimal hemodynamics and enhance left ventricular function, while eliminating the need for lifelong anticoagulation and preventing patient-prosthesis mismatch.^5,6

The growing use of stentless valves over the past decades has resulted in an increase in the number of reoperations for these valves, a trend expected to continue because all bioprosthetic valves suffer from structural valve deterioration (SVD) over time with comparable failure rates between stented and stentless valves.⁷ The bioprosthetic tissue deteriorates at a rate estimated between 3.4% at 5 years and up to 50% by 8 years.⁸ Indeed, stentless valves also suffer from long-term durability concerns, as they often require reoperation within 7 to 15 years.^9-11 These valves are prone to extensive calcification over time, which can complicate surgical interventions, increasing the need for complex aortic root reconstructions.¹²

In this context, valve-in-valve (ViV) transcatheter aortic valve implantation (TAVI) has emerged as a less invasive alternative to redo surgery for selected high-risk patients. However, ViV-TAVI in stentless valves is technically more challenging than stented valves because of the absence of fluoroscopic markers, the variability of root anatomy, the lack of a stented frame for valve anchoring, and higher rates of aortic regurgitation rather than aortic stenosis.¹² These challenges lead to a higher rate of complications, including device malposition, paravalvular leak (PVL), and coronary obstruction.¹³ For this purpose, this study sought to describe the outcomes of ViV-TAVI procedures in stentless valves from a single-center experience.

This study included patients who underwent ViV-TAVI due to SVD of a previously implanted stentless valve at the Lankenau Heart Institute in Wynnewood, Pennsylvania, between January 2018 and December 2023. The study aimed to assess the outcomes of this subset of patients. The institutional review board (IRB) of Main Line Health Hospitals approved the study protocol (IRB 45CFR164.512), and informed consent was waived due to the retrospective nature of the study.

Clinical and procedural data were extracted from patient medical records. All participants underwent pre-procedural transthoracic and/or transesophageal echocardiography, as well as multi-detector computed tomography (MDCT) for TAVI assessment. The primary study endpoint was 30-day and follow-up mortality. Secondary outcomes included Valve Academic Research Consortium (VARC)-3 technical success, VARC-3 device success, and VARC-3 early safety.

Definitions

Clinical information followed standardized definitions of data elements of the STS/ACC TVT (Society of Thoracic Surgeons [STS]/American College of Cardiology [ACC] Transcatheter Valve Therapy) Registry.¹⁴ The estimated glomerular filtration rate (eGFR) was estimated with the 2021 CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) creatinine equation.¹⁵ SVD was defined according to modified VARC-3 criteria.¹⁶ Severe SVD was identified by an increase of at least 20 mm Hg in the mean transvalvular gradient compared with the echocardiographic assessment conducted 1 to 3 months after the procedure, resulting in a mean gradient of 30 mm Hg or more. Additionally, severe SVD included the new onset or a rise of at least 2 grades in intraprosthetic aortic regurgitation, leading to severe regurgitation.

Patient-prosthesis mismatch (PPM) was evaluated using the indexed effective orifice area (iEOA) of the transcatheter heart valve (THV), calculated by dividing the EOA by the patient’s body surface area.¹⁷ Based on VARC-3 guidelines, PPM is diagnosed in patients with a body mass index (BMI) greater than or equal to 30 kg/m² when the iEOA is less than or equal to 0.70 cm²/m², and severe PPM when the iEOA is less than or equal to 0.55 cm²/m². In patients with a BMI less than 30 kg/m², PPM is defined as an iEOA below 0.85 cm²/m², with severe PPM occurring when the iEOA is under 0.65 cm²/m².¹⁶

Statistical methods 

Categorical data were expressed as frequencies and percentages. The normality of continuous variables was assessed using the Kolmogorov-Smirnov test. Normally distributed variables were reported as means and SDs, while non-normally distributed variables were presented as medians and IQR.

All statistical tests were 2-sided, and results were considered significant at a P-value of less than 0.05. Data were analyzed using R software version 4.4.1 (R Project for Statistical Computing) through RStudio. Upon request and subject to institutional approval, data supporting the conclusions of the study can be made available. The study followed the STROBE guidelines (STrengthening the Reporting of Observational Studies in Epidemiology) for observational research.¹⁸

Between 2018 and 2023, our center performed a total of 1476 TAVI procedures. Of these, 15 patients underwent ViV-TAVI in a previously implanted stentless valve. The mean age of the patients was 78.1 ± 9.3 years, and 6 (40%) were female. The median STS-PROM score was 4.30 (3.05, 6.75). Three (20%) patients had a history of coronary artery bypass grafting (CABG), and 2 (13.3%) had undergone prior percutaneous coronary intervention (PCI). The mean time between SAVR and TAVI was 14.7 ± 3.9 years. Eight (53.3%) patients presented with severe aortic regurgitation and 5 (33.3%) had severe aortic stenosis, while 2 (13.3%) a mixed etiology. A summary of all baseline characteristics can be found in Table 1.

Most patients (n = 9, 60%) had previously received a Freestyle valve (Medtronic), followed by 3 patients (20%) with a Toronto Root (St. Jude Medical), 2 patients (13.3%) with a Prima/Prima Plus valve (Edwards Lifesciences), and 1 patient with a 3F Enable valve (ATS Medical).

A balloon-expandable valve was used in 9 (60%) patients, while a self-expandable valve was deployed in the remaining 6 (40%) patients. Three high-risk patients for coronary obstruction underwent concomitant PCI with the snorkel technique to protect the coronary ostia. There were no cases of surgical conversion. However, 1 patient experienced THV malposition, requiring the deployment of a second valve; the THV migrated to the ascending aorta after post-dilatation and was snared to the ascending aorta to avoid further migration. A second THV was then placed across the previous stentless valve. During the postoperative period, this patient experienced a stroke and required the implantation of a new permanent pacemaker for a third-degree atrioventricular block. She was discharged and remains alive as of February 2025.

Perioperative complications were rare and are detailed in Table 2. There was 1 case of postoperative stroke, 1 case of need for new pacemaker implantation, and 1 death at 30 days. VARC-3 technical success was achieved in 87.6% (n = 13) of patients, device success in 80% (n = 12), and early safety in 73.3% (n = 11).

At discharge and at the 1-year follow-up, echocardiograms demonstrated satisfactory hemodynamic outcomes, with no instances of PVL greater than 2, as shown in Table 3. The median follow-up period was 2.9 years (range, 1.3-4.3 years), during which 4 patients died. All other follow-up complications are summarized in Table 4.

We aimed to assess the outcomes of patients with stentless bioprosthesis failure undergoing ViV-TAVI using VARC-3 criteria. The key findings of this series can be summarized as follows: (1) coronary obstruction protection using the snorkel technique was employed in 3 patients with low-lying coronary ostia; (2) 1 case of THV dislocation occurred, requiring the deployment of a second THV; and (3) despite these challenges, satisfactory VARC-3 outcomes were achieved.

Failure and degeneration of surgically implanted stentless valves are unavoidable over time. In this series, an average of almost 15 years elapsed between SAVR and TAVI, in line with recent publications on this topic.^4,13 Although current guidelines recommend redo surgical valve replacement, this approach carries significant risks, with morbidity and mortality rates often underrepresented by the surgical risk scores. For patients at higher surgical risk, ViV-TAVI offers an alternative in selected patients, potentially reducing the need for multiple surgeries over a lifetime. In fact, ViV-TAVI is increasingly utilized to treat SVD following SAVR.¹⁹ However, it presents unique technical challenges, such as a higher risk of coronary obstruction compared with TAVI for native aortic stenosis.²⁰ This occurs because the leaflets of the surgical aortic valve extend beyond the aortic root and above the sinotubular junction. During ViV-TAVI, the valve leaflets are pushed outward, forming a cylinder that can trap the sinuses and block blood flow to the coronary arteries. This is further enhanced in narrow aortic roots and non-coaxial implant of the previous bioprosthesis. Duncan et al reported from the Valve-in-Valve International Data (VIVID) registry that coronary obstruction was 4 times more likely in failing stentless ViV-TAVI than in stented valves, and 8 times more common in non-homograft stentless cases than in failing homograft valves.⁹ Previous analyses from the VIVID registry also highlighted anatomical risk factors for coronary obstruction in ViV-TAVI, including stentless bioprostheses and stented bioprostheses with externally mounted leaflets.²⁰

To evaluate the risk of coronary obstruction in ViV-TAVI, a multidetector computed tomography scan is recommended to measure the valve-to-coronary and valve-to-sinotubular junction distances. A measurement of less than 3 mm indicates a high risk for coronary obstruction.²¹ Several strategies have been developed to mitigate this risk, including the BASILICA (Bioprosthetic or native Aortic Scallop Intentional Laceration to prevent Iatrogenic Coronary Artery obstruction) technique, which involves intentionally lacerating the valve leaflets to prevent them from obstructing the coronary arteries during TAVI.^22-24 Another strategy considers coronary protection with guidewires and stents, which involves placing guidewires and undeployed stents in the coronary arteries before TAVI to address any obstruction that might occur, or deploy the coronary stent as prevention with the chimney or snorkel techniques.²⁵ The latter was deemed necessary in 3 patients and no cases of coronary obstruction were reported.

The use of stented bioprostheses facilitates THV deployment in ViV-TAVI, as most come with radiopaque markers that assist in precise placement. This benefit is diminished in stentless or homograft valves, where such markers are absent (Figure). Besides, there may be a lack of structural support for valve anchoring, particularly in cases of aortic regurgitation and no calcification.¹² Data from the VIVD registry showed that stentless ViV-TAVI had a nearly doubled malposition rate and more than a double need for a second prosthesis compared with stented cases.⁹ Device oversizing may have played a role in malpositioning, especially given the lack of established sizing guidelines for failing stentless bioprostheses.

Despite the complexity of the procedure, there was only 1 instance of THV migration intraoperatively, while postoperative complications were minimal. These included 1 major and 1 minor vascular complication, 1 case of postoperative stroke, and 1 need for new pacemaker implantation. Increased clinical experience has demonstrated similarly favorable outcomes in line with recent studies.^4,13 However, earlier publications reported higher perioperative complication rates and lower success according to VARC criteria.^26,27 Huczek et al found a 15% rate of incorrect THV positioning and 5% procedural mortality, and reported a VARC-2 device success rate of 50% with 74% VARC-2 early safety and 72% VARC-2 clinical efficacy.²⁷ Choi et al reported a 7.55% rate of THV procedural embolization.²⁶

Finally, a direct comparison between the 2 THV was not feasible in this analysis. However, the low rates of postoperative complications suggest favorable outcomes with either valve type when the appropriate valve is selected for the right patient using an individual optimized strategy. Currently, there is no definitive evidence to guide THV selection for redo-TAVR. The choice depends on the index THV type, the failure mechanism, and anatomical relationships, including left ventricular outflow tract dimensions, native annulus, sinuses of Valsalva, sino-tubular junction, and ascending aorta. Coronary access impairment is a key consideration, particularly in cases with a narrow sinus of Valsalva or a short distance from the coronary ostia to the index THV frame, where balloon-expandable valves (BEV) may provide an advantage.¹⁹ On the other hand, self-expandable valves (SEV) may be beneficial for patient-prosthesis mismatch at index TAVR or for procedures with high risk for elevated residual gradients.¹⁹ The design of the original THV affects the neoskirt height, with supra-annular leaflet positioning in a tall stent frame increasing the obstruction risk.^28,29 This height varies across THV sizes and generations. THV selection and implantation technique for redo-TAVR can mitigate risk, such as by implanting a short-frame THV at a lower position within a tall-frame THV. Conversely, using a tall-frame THV in a failed tall-frame THV may result in a higher neoskirt and greater obstruction risk.¹⁹

Of note, Moubarak et al previously compared SEV and BEV THVs in ViV-TAVI with stentless valves.³⁰ The authors showed that ViV-TAVI for failed stentless bioprostheses using SEV had higher rates of periprocedural complications, including device malposition, need for post-deployment balloon dilation, and requirement of a second transcatheter device. However, both BEV and SEV demonstrated improved aortic valve function at 30-day and 1-year follow-up. Additionally, ViV-TAVI for degenerative stentless bioprostheses was linked to acceptable survival rates at 1 and 3 years, with no significant survival difference between BEV and SEV groups.

Limitations

The main limitation of this study arises from its retrospective design, which may not account for all potential confounding factors. Additionally, the single-center setting may limit the broader applicability of these findings to a wider population. This limitation also highlights the need for long-term follow-up, as it may reveal important differences in late survival outcomes. The study's small sample size and the selected cohort of patients further limit the generalizability of its findings, as the outcomes observed may not fully represent broader patient populations.

Careful patient selection is critical for ViV-TAVI procedures because of the high risks of complications, increased by the presence of a stentless valve in addition to the standard risks associated with the ViV procedure. This cohort of patients was highly comorbid and symptomatic but was associated with acceptable early VARC-3 outcomes; however, significant complications were reported at mid-term follow-up. Larger studies are warranted to confirm these findings.

Massimo Baudo, MD¹; Serge Sicouri, MD¹; Yoshiyuki Yamashita, MD, PhD^1,2; Dimitrios Magouliotis, MD, PhD, MSc¹; Francesco Cabrucci, MD¹; Scott M. Goldman, MD²; Roberto Rodriguez, MD²; Eric M. Gnall, DO³; Paul M. Coady, MD³; William A. Gray, MD^1,3; Sandro Gelsomino, MD, PhD⁴; Basel Ramlawi, MD^1,2

Dr Gelsomino and Dr Ramlawi share equal senior authorship.

From the ¹Department of Cardiac Surgery Research, Lankenau Institute for Medical Research, Main Line Health, Wynnewood, Pennsylvania; ²Department of Cardiac Surgery, Lankenau Heart Institute, Main Line Health, Wynnewood, Pennsylvania; ³Department of Interventional Cardiology, Lankenau Heart Institute, Main Line Health, Wynnewood, Pennsylvania; ⁴Cardiovascular Research Institute Maastricht - CARIM, Maastricht University Medical Centre, Maastricht, the Netherlands.

Disclosures: Dr Gray is a consultant for, and is on the advisory board of Philips, Medtronic, Boston Scientific, Surmodics, Encompass Vascular, eFemoral, Cagent, Sonovascular, Conformal Medical, Contego Medical, ReValve, Alucent, and Edwards LifeSciences; and holds stock with eFemoral, Encompass Vascular, Xenter, and Reflow Medical. Dr Ramlawi is a consultant for, and is on the advisory board of Medtronic Inc., Boston Scientific Corporation, AtriCure Inc., and Corcym Inc.

Address for correspondence: Massimo Baudo, MD, Department of Cardiac Surgery Research, Lankenau Institute for Medical Research, Main Line Health, 100 E Lancaster Avenue, Wynnewood, PA 19096, USA. Email: massimo.baudo@icloud.com; X: @Takotsubo91