Seven-year outcomes after surgical aortic valve replacement with a stented bovine pericardial bioprosthesis in over 1100 patients: a prospective multicentre analysis

Abstract

Open in new tabDownload slide

INTRODUCTION

Bioprosthetic valves have been widely used for surgical aortic valve replacement (SAVR) because lifelong anticoagulation is not required, and they pose a much lower risk of bleeding and thrombotic complications than mechanical valves. However, durability is a consideration because the tissue used in bioprosthetic valves is prone to deterioration over time.

The intermediate-term follow-up period, defined as ∼5–7 years, is a critical period for evaluating the durability and haemodynamic stability of newly introduced bioprosthetic valves. Recently, the Trifecta valve was removed from commercial distribution due to concerns about durability, which became apparent predominantly between 5 and 8 years of follow-up [1]. Similarly, the Toronto SPV valve was also removed from commercial distribution because of a higher-than-expected rate of failure due to central regurgitation, a trend that became apparent at ∼7 years [2]. The PERIcardial SurGical AOrtic Valve ReplacemeNt (PERIGON) Pivotal Trial (Medtronic, Minneapolis, MN, USA) evaluated clinical outcomes and haemodynamic performance of the Avalus valve, a stented pericardial valve with an amino-oleic-acid anticalcification treatment. This analysis evaluated outcomes through 7 years.

PATIENTS AND METHODS

Ethical statement

The institutional review board or ethics committee (IRB/EC) at all sites approved the protocol (www.clinicaltrials.gov, NCT02088554; Supplementary Material, Table S1). All patients provided written informed consent. The pivotal trial had a planned duration of 5 years, which was extended to 12 years to continue data collection. Sites with 10 or more active patients at the time of evaluation were invited to continue in long-term follow-up (LTFU); 19 sites agreed to participate. An addendum to the original protocol for the pivotal study was sent to the IRB/EC at each participating site; additional approval numbers were issued if required (Supplementary Material, Table S1). All active patients at the LTFU sites were invited to continue in follow-up. Patients who accepted provided new written informed consent, and 576 patients were re-consented. At the time of analysis, 441 of 458 patients (96%) with an expected 7-year visit had completed the visit.

Study design

The PERIGON Pivotal Trial is a prospective, multicentre trial conducted at 39 sites in Europe, Canada and the United States. The trial, previously described in detail [3, 4], enrolled adults with moderate or severe aortic stenosis (AS) and/or chronic severe aortic regurgitation and an indication for SAVR. In the pivotal trial, in-person follow-up was performed at 3–6 months, 1 year and then annually through 5 years. For the LTFU study, follow-up will be conducted in person at years 7, 10 and 12 and by telephone at years 6, 8, 9 and 11. The following evaluations will be conducted annually: New York Heart Association (NYHA) classification, medication use (i.e. aspirin, other antiplatelet, anticoagulant and indication), vital status, and safety end-points as defined in the primary protocol. Transthoracic echocardiography and 12-lead electrocardiography will be performed at the 7-, 10- and 12-year visits.

End-points

The primary end-point of this analysis was freedom from surgical explant and/or percutaneous valve-in-valve reintervention due to structural valve deterioration (SVD) through 7 years of follow-up. Originally, in the PERIGON Pivotal Trial, SVD was defined as ‘Change to the function of a heart valve substitute resulting from an intrinsic abnormality that causes stenosis or regurgitation. The diagnosis should be confirmed by examination of the explanted or damaged valve. It includes intrinsic changes such as wear, fatigue failure, stress fracture, calcification, leaflet tear and stent creep. This definition excludes paravalvular leak (PVL), infection, pannus overgrowth or thrombosis of the heart valve substitute as determined by reoperation, autopsy, or in vivo investigation’ [5]. We also analyzed a composite end-point of SVD requiring reintervention and/or severe haemodynamic dysfunction (SHD) requiring reintervention to address the limitations of the protocol definition of SVD. The introduction of TAV-in-SAV as a reintervention method for valve failure has made it difficult to definitively determine the failure mode in some patients with significant haemodynamic dysfunction. The SHD end-point was used to categorize potential safety events that were characterized by severe AS, severe transvalvular regurgitation or progressive severe dysfunction that did not meet the protocol definitions of SVD or non-structural valve dysfunction (NSVD) [5]. We also report the Kaplan–Meier (KM) event rates for the individual components of the composite end-point.

Secondary end-points were the KM rates of survival (freedom from all-cause mortality), freedom from valve-related reintervention, freedom from valve-related death, NYHA functional classification status, and haemodynamic performance, including mean aortic gradient, dimensionless velocity index (DVI), effective orifice area (EOA), PVL, transvalvular leak (TVL) and prosthesis-patient mismatch (PPM). DVI is the ratio between the velocity time integral (VTI) of the left ventricular outflow track (LVOT) and of the aortic prosthesis (AV), i.e. LVOT VTI/AV VTI.

Valve-related safety end-points included valve-related thromboembolism, valve thrombosis, haemorrhage, PVL, endocarditis, hemolysis, NSVD, SVD, reintervention and explant [6, 7]. The safety end-points were also stratified by surgical procedure, i.e. isolated SAVR, SAVR + coronary artery bypass graft and SAVR + other concomitant procedures.

Deaths and valve-related safety events were adjudicated by an independent clinical events committee (Baim Clinical Research Institute, Boston, MA, USA), and explanted valves were evaluated by CV Path Institute (Gaithersburg, MD, USA). Echocardiograms obtained at the 3- to 6-month, 1-year, 5-year and 7-year visits were assessed by a core laboratory (Mayo Clinic, Rochester, MN, USA).

Statistical analysis

Categorical data are reported as frequencies and percentages, and continuous data are reported as means and standard deviations (SDs). KM analysis was performed to determine freedom from surgical explant and/or percutaneous valve-in-valve reintervention due to SVD/SHD requiring reintervention, freedom from valve-related reintervention, freedom from valve-related death and survival at 7 years. KM analysis was also used to calculate the event rates of valve-related safety end-points and the composite end-point. Haemodynamic performance was evaluated by analyzing the means and SDs of aortic gradient, DVI and EOA, as well as the severity of PVL and TVL.

Analyses were performed with SAS version 9.4 (SAS Institute, Cary, NC, USA) and R version 4.3.2 (The R Project for Statistical Computing, www.r-project.org).

RESULTS

Baseline and procedural characteristics

A total of 1312 patients were enrolled, and 1281 underwent baseline evaluation. In total, 1132 underwent successful implantation and are included in this analysis. Total follow-up for the full cohort was 6412 patient-years, and late follow-up (>30 days) was 6319 patient-years. Supplementary Material, Fig. S1 provides more information about patient follow-up and exits from the study.

The results reported in this manuscript are for the full cohort. Mean age at baseline was 70 years with 76% of patients >65 years of age (Supplementary Material, Fig. S2). Seventy-five percentage of patients were male. Fifty-eight % had a NYHA functional status of class I or II, and the mean Society of Thoracic Surgeons (STS) predicted risk of mortality was 2.0%. Hypertension, dyslipidaemia, and coronary artery disease were the most frequent comorbidities. Table 1 provides additional patient characteristics.

The primary indication for SAVR was AS in 84% of patients, aortic regurgitation in 6% of patients, mixed AS/aortic regurgitation in 10% of patients and a failed prosthesis in <1% of patients. Thirty % of patients had a congenital bicuspid valve. A median sternotomy was the most frequent surgical approach (80%), followed by hemisternotomy (13%), right thoracotomy (6%) and ‘other’ approach (1%). Coronary artery bypass graft was the most common concomitant procedure, performed in 32% of patients. The most common valve size was 23 mm, implanted in 35% of patients (Supplementary Material, Fig. S3). Table 2 provides additional procedural details.

End-points

Freedom from surgical explant and/or percutaneous valve-in-valve reintervention due to SVD/SHD at 7 years was 98.8% (97.5–99.5%) (n = 7). All 7 cases were adjudicated as SHD requiring reintervention, and none were adjudicated as SVD. Because of the limitations of the protocol definition, the criteria for SVD were not met for these 7 cases, and it was not possible to determine the precise mode of failure. Thus, these cases were adjudicated as SHD. Supplementary Material, Table S2 provides additional details about these 7 patients. Freedom from valve-related reintervention was 94.3% (92.3–95.7%) at 7 years. Overall survival (freedom from all-cause mortality) was 82.6% (79.8–85.0%), and freedom from valve-related death was 97.4% (95.9–98.3%) (Fig. 1).

Figure 1:

Survival (KM event rates of freedom from all-cause, cardiac and valve-related mortality) through 7 years of follow-up. KM: Kaplan–Meier.

Open in new tabDownload slide

Table 3 shows the KM rates of valve-related safety events. At 7 years, the KM event rate was 6.3% (4.9–8.1%) for thromboembolism, 0.4% (0.1–1.1%) for valve thrombosis, 9.7% (8.1–11.6%) for all haemorrhage, 1.0% (0.5–1.8%) for PVL, 6.3% (4.9–8.3%) for endocarditis, 5.7% (4.3–7.7%) for reintervention and 4.4% (3.1–6.0%) for explant. Supplementary Material, Fig. S4 shows the rates of thromboembolism, endocarditis, haemorrhage, and reintervention when analyzed with death as a competing outcome. The KM event rate at 7 years was 2.8% (1.7–4.6%) for all SHD (requiring/not requiring reintervention) and 1.2% (0.5–2.5%) for SVD/SHD requiring reintervention. The rates for the subcategories were, respectively, 0% (0–0%) and 1.2 (0.5–2.5%). There were no adjudicated cases of SVD not requiring reintervention. At 7 years, the KM rates of valve-related safety events were similar among all three surgical procedure groups except for all haemorrhage, which was higher in the SAVR + ‘other’ concomitant procedures group (14.0% vs 7.9% vs 8.2%, P = 0.019; Supplementary Material, Table S3).

Forty-seven patients underwent a reintervention by the time of analysis. Of the 38 patients who had a surgical reintervention (explant), 33 had endocarditis, 1 had valve thrombosis, 1 had major PVL and 2 had NSVD (1 major PVL, 1 entrapment by pannus, tissue, or suture). Among the 9 patients who had a percutaneous reintervention, 7 had SHD, 1 had valve thrombosis, 1 had major PVL and 1 had NSVD (major PVL). Figure 2 shows the KM rates of reintervention stratified by cause. There were no statistical differences in the rates of valve-related events that led to a reintervention in patients ≤65 and >65 years of age (Supplementary Material, Table S4).

Figure 2:

KM rate of reintervention (surgical or percutaneous) stratified by cause through 7 years of follow-up. KM: Kaplan–Meier.

Open in new tabDownload slide

At 7 years, the mean aortic gradient was 13.8 ± 5.9 mmHg (n = 409), DVI was 0.42 ± 0.09 (n = 388), and EOA was 1.99 ± 0.53 (n = 367) (Supplementary Material, Table S5). Supplementary Material, Fig. S5 illustrates 99.0% and 98.3% of patients had none or trace PVL and TVL at 7 years, respectively. At 7 years, 76.8% of patients had no PPM, 19.6% had moderate PPM, and 3.5% had severe PPM (Supplementary Material, Fig. S6). The proportion of patients with an NYHA functional classification of I/II was ≥94% after valve implantation through 7 years of follow-up (Supplementary Material, Fig. S7).

DISCUSSION

This analysis of the PERIGON trial demonstrates that the Avalus valve delivers excellent clinical outcomes. Over 7 years of follow-up, echocardiographic and clinical data reveal consistently stable haemodynamic performance and functional status. These findings underscore the valve's reliability and efficacy. Freedom from SVD/SHD requiring reintervention at 7 years was 98.8% (97.5–99.5%), indicating excellent intermediate durability of the study valve. Survival from all causes was 82.6% (79.8–85.0%). The mean aortic gradient was 13.8 ± 5.9 mmHg, EOA was 1.99 ± 0.53 cm² and DVI was 0.42 ± 0.09 at 7 years of follow-up. Rates of none/trace PVL and TVL were 99.0% and 98.3%, respectively. Functional status has been stable with ≥94% of patients in NYHA class I or II throughout follow-up.

The primary end-point for this analysis was defined as freedom from surgical explant and/or percutaneous valve-in-valve reintervention due to SVD. We also evaluated a composite end-point of SVD/SHD requiring reintervention to address the limitations of the protocol definition of SVD. While all 7 cases of this composite end-point were adjudicated as SHD by the clinical events committee, this does not definitively rule out the presence of SVD in the study cohort. Traditional surgical definitions of SVD have limitations, particularly with the increased use of valve-in-valve procedures, making it challenging to distinguish between thrombus, NSVD and SVD using in situ imaging for diagnosing valvular dysfunction. Although these cases might be considered SVD on the basis of haemodynamic criteria, such as those recommended by the Valve Academic Research Consortium 3 [8], a recent publication described the limitations of those definitions as well [9]. Reliable and standardized methods to identify SVD in both surgical and transcatheter valve clinical trials seem therefore warranted. We believe a reasonable approach is to remain agnostic regarding the mechanism of valvular dysfunction when unknown, but to conservatively group such events with SVD for comparison to historical and contemporary event rates from other trials.

The clinical and haemodynamic outcomes reported here are comparable to those reported by other studies of SAVR. Beaver et al. [10] reported a freedom from SVD rate of 99.3% at 7 years in the COMMENCE trial, whereas Tsui et al. [11] reported a freedom from reintervention due to an SVD rate of 93.6% at 8 years (median follow-up = 7.0 years). In contrast, studies of the Trifecta and Mitroflow prostheses have demonstrated that the incidence of SVD begins to increase substantially after 6 years [12, 13]. In this study, freedom from valve-related reintervention at 7 years was 94.3%, whereas Beaver et al. [10] reported a freedom from reoperation rate of 97.2%, and Tsui et al. [11] reported that freedom from reintervention at 8 years was 89.8%. Overall survival at 7 years in the PERIGON trial was 82.6%, compared with 85.4% and 80.7% reported by others [10, 11]. Freedom from valve-related death was 97.4% at 7 years in PERIGON, compared with 95.8% at 8 years in the Tsui et al. study [11].

Mean gradient and EOA, in addition to PVL and TVL, were similar to what has been reported for other surgical aortic valves [10, 11]. DVI was 0.42 ± 0.09 in our study. This index is useful for evaluating valve stenosis because it is not dependent on blood flow and does not require measurement of the left ventricular outflow tract. A DVI >0.30–0.35 is considered normal for a prosthetic valve [14, 15]. However, discharge DVIs ≤0.50 and ≤0.35 have been associated with worse clinical outcomes [16, 17].

Limitations

While the study does face some limitations (such as patient attrition after the 5-year visit), with 576 of the 1132 patients reconsenting to LTFU, it remains robust. The potential for selection bias is minimal, as only sites with 10 or more active patients at the 5-year evaluation were invited to participate. Although there is no control group to compare different valves, the study is the first to provide 7-year outcomes for the Avalus valve. Although echo follow-ups are not annual, with echocardiograms taken at 3- to 6-month, 1-, 5- and 7-year visits, these intervals still provide valuable insights. Finally, while the SVD end-point is based on the traditional surgical definition, it remains a valid comparison point alongside studies using longitudinal haemodynamic data criteria.

CONCLUSION

The 7-year data reported here demonstrate excellent durability of the Avalus valve with continued low rates of SVD/SHD requiring reintervention, mortality, and valve-related reinterventions. These results are important to guide clinical decision-making in the lifetime management of aortic valve disease.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

ACKNOWLEDGEMENTS

Julie A. Linick, of Medtronic, assisted in the drafting and revision of this manuscript under the direction of the authors. Shuzhen Li, of Medtronic, assisted with the statistical analysis.

FUNDING

Funded by Medtronic.

Conflict of interest: Joseph F. Sabik III: North American Principal Investigator of the PERIGON Pivotal Trial, Medtronic. Vivek Rao: Medtronic: Consultant, North American Surgical Advisory Board, Equity (<$25k). Abbott: Consultant. Gore: Consultant. Francois Dagenais: Honoraria from Medtronic, Abbott and Edwards for presentations and from Cook Medical for proctoring and presentation. Michael G. Moront: Trainer and consultant, Medtronic; trainer and speaker, Atricure; speaker and consultant, Haemonetics. Michael J. Reardon: Consultant to Medtronic; all payments go directly to his department. Himanshu J. Patel: Consultant and Surgical Advisory Board member for Medtronic. Jae K. Oh: Director of the Echo Core Lab for the PERIGON LTFU trial; consulting for valve diseases, Medtronic; Anumana equity for ECG AI for aortic stenosis. Shinichi Fukuhara: Consultant for Edwards Lifesciences, Medtronic, Artivion and Terumo Aortic. Tianhua Wu: Medtronic employee. Robert J.M. Klautz: European Principal Investigator of the PERIGON Pivotal Trial, research support and consultation fees from Medtronic. Louis Labrousse, Ralf Günzinger, Kamran Baig, and Saki Ito: none.

DATA AVAILABILITY

The data underlying this article are owned by the study sponsor and will not be made available to other investigators for research purposes.

Author contributions

Joseph F. Sabik III: Conceptualization; Investigation; Resources; Supervision; Visualization; Writing—original draft. Vivek Rao: Investigation; Resources; Writing—review & editing. Francois Dagenais: Investigation; Resources; Writing—review & editing. Michael G. Moront: Investigation; Resources; Writing—review & editing. Michael J. Reardon: Investigation; Resources; Writing—review & editing. Himanshu J. Patel: Investigation; Resources; Writing—review & editing. Jae K. Oh: Investigation; Writing—review & editing. Shinichi Fukuhara: Investigation; Resources; Writing—review & editing. Louis Labrousse: Investigation; Resources; Writing—review & editing. Ralf Günzinger: Investigation; Resources; Writing—review & editing. Kamran Baig: Investigation; Resources; Writing—review & editing. Saki Ito: Investigation; Writing—review & editing. Tianhua Wu: Formal analysis; Methodology; Writing—review & editing. Robert J.M. Klautz: Conceptualization; Investigation; Resources; Supervision; Writing—review & editing.

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks Mateo Marin-Cuartas and the other anonymous reviewers for their contribution to the peer review process of this article.

Presented at the 38th EACTS Annual Meeting, Lisbon, Portugal, 9–12 October 2024.

REFERENCES

ABBREVIATIONS

ABBREVIATIONS
 
• AS

  Aortic stenosis

 
• DVI

  Dimensionless velocity index

 
• EOA

  Effective orifice area

 
• LTFU

  Long-term follow-up

 
• NSVD

  Non-structural valve dysfunction

 
• NYHA

  New York Heart Association

 
• PERIGON

  PERIcardial SurGical AOrtic Valve ReplacemeNt clinical trial of the Avalus valve

 
• PVL

  Paravalvular leak

 
• SAVR

  Surgical aortic valve replacement

 
• SHD

  Severe hemodynamic dysfunction

 
• SVD

  Structural valve deterioration

 
• TVL

  Transvalvular leak