Long-term outcomes of surgical aortic valve replacement using a third-generation stented bovine pericardial bioprosthesis in 1907 patients Authors: Preetha Pamidighantam, Stephen Thorp, Jason M. Aubrey, Caleb Weissman, Lucas Leffler, Taylor Zimmer, Justin Fanning, Stephane Leung Wai Sang, David Ranney, Edward Murphy, Charles Willekes, Tomasz A. Timek Published: 2026 ABSTRACT ObjectivesThis large, single-center evaluation presents long-term survival, reintervention, and echocardiographic outcomes after implantation of a third-generation supra-annular stented bovine pericardial aortic valve with a proprietary anticalcification treatment.MethodsAfter excluding patients with ventricular assist devices, congenital repairs, and heart transplants, 1907 patients underwent aortic valve replacement with this bioprosthetic valve between August 2010 and September 2023. Primary outcomes were freedom from all-cause mortality and reintervention for structural valve deterioration. Secondary outcomes included mean transprosthetic gradient and Valve Academic Research Consortium 3–defined severe hemodynamic valve dysfunction.ResultsAmong 1907 patients included, 30-day mortality was 2.8% (N = 54), stroke occurred in 2.8% (N = 53), and a permanent pacemaker was implanted in 3.3% (N = 62). Echocardiographic follow-up was available in 90.46% of patients (mean follow-up: 4 years; total 6557 echocardiograms reviewed). Overall mean transprosthetic gradient was 11.76 ± 6.52 mm Hg, 12.87 ± 7.52 mm Hg, 16.10 ± 11.02 mm Hg, and 16.01 ± 8.31 mm Hg at 1, 5, 10, and 12 years, respectively. Overall survival was 93.25%, 77.62%, 47.99% and 37.84% at 1, 5, 10, and 12 years, respectively. Freedom from stage 3 hemodynamic valve dysfunction was 99.08%, 85.75%, and 64.32% at 5, 10, and 12 years, respectively. Structural valve deterioration–related reintervention occurred in 2.7% of patients (N = 51), with freedom from reintervention at 98.32%, 90.80%, and 85.13% at 5, 10, and 12 years, respectively.ConclusionsLong-term outcomes of this bioprosthetic valve are characterized by excellent overall survival, low structural valve deterioration rates, and sustained hemodynamic performance through 12 years of follow-up. The PMEAV exhibits excellent durability and patient survivability at 12-year follow-up. Central MessageThe PMEAV is a durable bioprosthetic valve that offers long-term advantages in survival, freedom from reintervention, and hemodynamic integrity on 12-year follow-up. PerspectiveWith the advent of third-generation bioprosthetic valves and shifting treatment guidelines, patients aged less than 65 years are increasingly receiving biologic prostheses despite limited long-term data. In 1907 PMEAV recipients, this comprehensive, 12-year study revealed excellent survival, reintervention, and hemodynamic outcomes, particularly in patients aged 50 to 65 years compared with patients aged less than 50 years. The PMEAV exhibits excellent durability and patient survivability at 12-year follow-up. Central Message The PMEAV is a durable bioprosthetic valve that offers long-term advantages in survival, freedom from reintervention, and hemodynamic integrity on 12-year follow-up. Perspective With the advent of third-generation bioprosthetic valves and shifting treatment guidelines, patients aged less than 65 years are increasingly receiving biologic prostheses despite limited long-term data. In 1907 PMEAV recipients, this comprehensive, 12-year study revealed excellent survival, reintervention, and hemodynamic outcomes, particularly in patients aged 50 to 65 years compared with patients aged less than 50 years. Over the past 2 decades, bioprosthetic valves have been more commonly used than their mechanical counterparts in surgical aortic valve replacement (SAVR) procedures,1 including among patients aged less than 65 years.2 Although first-generation bioprostheses were limited by structural valve deterioration (SVD) from leaflet degeneration or rupture, contemporary models exhibit improved long-term durability and hemodynamic performance. The Carpentier-Edwards Perimount Magna Ease Aortic Valve (PMEAV) (Edwards Lifesciences) is a third-generation, stented, supra-annular bovine pericardial valve that has evolved from the well-established Perimount and Magna models.3, 4, 5 Compared with its predecessors, the PMEAV features a lower valve profile and a flexible, scalloped sewing cuff to reduce the risk of valve dehiscence and a proprietary anticalcification fixation process to limit leaflet deterioration.6,7 These protective features have been associated with a favorable mid-term survival and freedom from reoperation for SVD, as reported by our center and other studies.8, 9, 10, 11 Although the PMEAV eliminates lifelong anticoagulation burden and offers the potential for future transcatheter valve-in-valve replacements in younger patients,12 data on the extended durability and echocardiographic outcomes of the PMEAV stratified by age remain unexplored.10,13, 14, 15 This current study aims to evaluate long-term outcomes in patients undergoing SAVR with PMEAV with or without concomitant procedures at a single high-volume center, focusing on overall survival, freedom from SVD, and other valve-related clinical outcomes. Patients and Methods This retrospective analysis was approved by our Institutional Review Board on November 20, 2023 (IRB 2019-116), with patient consent waived per institutional guidelines. Between January 2010 and August 2023, 1931 consecutive patients underwent SAVR with the Perimount Magna Ease aortic valve at our institution. After excluding patients with a history of cardiac transplantation, immunosuppression, ventricular assist device implantation, or congenital heart disease other than bicuspid aortic valve disease, our cohort included a total of 1907 patients. The cohort included patients undergoing isolated and combined SAVR and both emergency and reoperations. Postoperatively, all patients received lifelong aspirin; warfarin was additionally prescribed for those with concomitant mitral or tricuspid valve procedure or elevated thromboembolic risk. Baseline demographic and clinical characteristics were evaluated. The primary outcomes were overall survival and freedom from aortic valve reintervention. Secondary outcomes included echocardiography-derived mean transprosthetic gradient (mTPG) and left ventricular mass index (LVMI). Patient characteristics and perioperative outcomes were obtained from the institutional Society of Thoracic Surgeons database and electronic medical records. Mortality data were queried from the State of Michigan Death Index. Valve-specific outcomes included patient-prosthesis mismatch (PPM), defined using manufacturer-reported effective orifice area (EOA), and aortic valve reintervention categorized by reintervention for SVD per Valve Academic Research Consortium-3 (VARC-3) criteria, prosthetic valve endocarditis, and nonstructural valve dysfunction. Valve-specific reinterventions included re-do SAVR, transcatheter valve-in-valve replacement, and balloon valvuloplasty (including for palliative or salvage indications). Postoperative transthoracic echocardiography (TTE) was largely performed through an Intersocietal Accreditation Commission–accredited laboratory and reviewed for ejection fraction (EF), left ventricular (LV) dimensions, and valvular gradients. Postoperative TTEs were grouped into the following intervals: 3 months (0-6 months), 1 year (6-24 months), 3 years (24-48 months), 5 years (48-72 months), 8 years (7-9 years), 10 years (9-11 years), and 12 years (12-14 years). When multiple studies were available within a timeframe, the study closest to the target interval was selected. LVMI was calculated through the Devereux equation and indexed to body surface area (BSA).16 Per VARC-3 criteria,17 severe hemodynamic valve deterioration (HVD stage 3) was defined as follows:a.increase in mean gradient 20 mm Hg or more resulting in a final gradient 30 mm Hg or more with concomitant 0.6 cm2 or more reduction in EOAb.50% or more decrease in Doppler velocity index 0.2 or more or 40% or more compared with the reference echocardiographic examination performed 1 to 3 months postprocedurec.new or worsening intra-prosthetic aortic regurgitation by 2 or more grades, resulting in severe AR a.increase in mean gradient 20 mm Hg or more resulting in a final gradient 30 mm Hg or more with concomitant 0.6 cm2 or more reduction in EOA increase in mean gradient 20 mm Hg or more resulting in a final gradient 30 mm Hg or more with concomitant 0.6 cm2 or more reduction in EOA b.50% or more decrease in Doppler velocity index 0.2 or more or 40% or more compared with the reference echocardiographic examination performed 1 to 3 months postprocedure 50% or more decrease in Doppler velocity index 0.2 or more or 40% or more compared with the reference echocardiographic examination performed 1 to 3 months postprocedure c.new or worsening intra-prosthetic aortic regurgitation by 2 or more grades, resulting in severe AR new or worsening intra-prosthetic aortic regurgitation by 2 or more grades, resulting in severe AR PPM was categorized as follows:a.severe (EOA index ≤0.65 cm2/m2)b.moderate (0.66-0.85 cm2/m2)c.nonsignificant (>0.85 cm2/m2) a.severe (EOA index ≤0.65 cm2/m2) severe (EOA index ≤0.65 cm2/m2) b.moderate (0.66-0.85 cm2/m2) moderate (0.66-0.85 cm2/m2) c.nonsignificant (>0.85 cm2/m2) nonsignificant (>0.85 cm2/m2) Descriptive statistics were reported for the full cohort. Continuous variables were summarized as means ± SD or medians with interquartile ranges (IQRs), depending on normal distribution. Categorical variables were presented as counts and percentages. Categorical outcomes were compared with chi-square analysis or Fisher exact test if more than 20% of expected cell counts were less than 5. Postoperative outcomes by procedure type were evaluated with Kruskal–Wallis analysis; if significant, a follow-up pairwise Wilcoxon rank-sum with Bonferroni correction was completed. Group differences in freedom from mortality and reintervention were analyzed using Kaplan-Meier curves with log-rank tests. Univariate and multivariate Cox proportional hazard models estimated hazard ratios (HRs) for mortality. A Fine-Gray competing risk model was used to estimate risk of reintervention, treating death as a competing event. Beta regression analysis was used to identify factors associated with unit change in LVMI. Candidate variables were selected on clinical variance and prior literature. To maximize cohort inclusion, only variables with less than 10% missingness were entered into multivariable models, and records with missing data for included variables were excluded from the analysis. All statistical tests were 2-sided. Analyses were conducted using SAS Enterprise Guide v7.1 (SAS Institute Inc). Results Baseline Patient Characteristics and Operative Details Baseline characteristics and operative details are displayed in Table 1. Approximately one-third of the cohort was categorized as New York Heart Association (NYHA) Class III or IV. The mean predicted Society of Thoracic Surgeons risk score was 2% ± 1%. Most operations were elective in patients without prior cardiac surgery. Aortic stenosis (AS) was the most common indication (70.4%, N = 1342), and 5.5% (N = 105) underwent surgery for endocarditis. More than half of the cohort underwent concomitant procedures, with coronary artery bypass grafting (CABG) being completed in more than one-third of the cohort. Among patients with endocarditis, 13 (12.4%) required annular or aortic root patch reconstruction. Of the 104 patients undergoing a Bentall procedure, 57 (54.8%) required concomitant arch repair with circulatory arrest. Survival stratified by concomitant mitral or tricuspid valve repair is displayed in Figure E1. The median size of implanted valve was 25 mm (IQR, 23-27 mm). Annular/root enlargement procedures were performed in 58 of 1907 patients (3.04%); operative mortality in this subgroup was 3 of 58 patients (5.17%). Moderate PPM was present at implant in 0.18% of patients (N = 1) with no severe PPM noted in this cohort.Table 1Patient and operative detailsVariableMissingN = 1907Patient characteristics Age, y67.30 ± 11.28 Gender (male)1403 (73.57%) Body mass index (kg/m2)5530.67 ± 10.89 Body surface area (m2)732.08 ± 0.27 Diabetes592 (31.04%) Dialysis35 (1.84%) Lung disease467 (24.49%) Cerebrovascular disease333 (17.46%) Peripheral arterial disease256 (13.42%) Prior myocardial infarction260 (13.63%) NYHA heart failure Class I115 (6.03%) Class II421 (22.08%) Class III489 (25.64%) Class IV112 (5.87%) None770 (40.38%) Hypertension1571 (82.38%) Dyslipidemia1124 (58.94%) Liver disease52 (2.73%) Family history of premature CAD304 (15.94%) Arrhythmia Persistent/permanent112 (5.87%) Paroxysmal201 (10.54%) None1594 (83.59%) Smoking history Current smoker312 (16.36%) Former smoker704 (36.92%) Never smoker891 (46.72%) Endocarditis105 (5.51%) Mediastinal radiation37 (1.94%) Cardiogenic shock45 (2.36%) Total albumin1063.66 ± 0.52 Ejection fraction (%)4755.86 ± 12.04 Hemoglobin42213.36 ± 2.06 Platelets (×103/μL)215217.65 ± 71.15 Degree of aortic insufficiency None466 (24.44%) Trivial/trace260 (13.63%) Mild529 (27.74%) Moderate309 (16.20%) Severe343 (17.99%) Society of Thoracic Surgeons Score5810.02 [0.01-0.03] Indication Aortic insufficiency310 (16.26%) Aortic stenosis1342 (70.37%) Endocarditis105 (5.51%) Root/aorta150 (7.87%) Redo operations193 (10.12%)Operative characteristics Operative status Elective1481 (77.66%) Urgent377 (19.77%) Emergency49 (2.57%) Operative approach Sternotomy1660 (87.05%) Ministernotomy219 (11.48%) Right thoracotomy28 (1.47%) Concomitant procedure Coronary artery bypass grafting715 (37.49%) Mitral or tricuspid valve procedure166 (8.70%) Antiarrhythmic procedure110 (5.77%) Root/aorta156 (8.18%) Crossclamp time (min)59134.04 ± 63.03 Cardiopulmonary bypass time (min)59166.38 ± 75.74 Valve size (mm) 1915 (0.79%) 21178 (9.33%) 23533 (27.95%) 25604 (31.67%) 27417 (21.87%) 29160 (8.39%) PPM BMI <30 Insignificant: >0.85754 (100%) Moderate: 0.66-0.850 (0%) Severe: ≤0.550 (0%) PPM BMI ≥30 Insignificant: >0.70560 (99.82%) Moderate: 0.56-0.701 (0.18%) Severe: ≤0.550 (0%)Categorical variables are represented as n (%). Normally distributed data expressed as mean ± SD. NYHA, New York Heart Association; CAD, coronary artery disease; PPM, patient-prosthesis mismatch; BMI, body mass index. Table 1Patient and operative detailsVariableMissingN = 1907Patient characteristics Age, y67.30 ± 11.28 Gender (male)1403 (73.57%) Body mass index (kg/m2)5530.67 ± 10.89 Body surface area (m2)732.08 ± 0.27 Diabetes592 (31.04%) Dialysis35 (1.84%) Lung disease467 (24.49%) Cerebrovascular disease333 (17.46%) Peripheral arterial disease256 (13.42%) Prior myocardial infarction260 (13.63%) NYHA heart failure Class I115 (6.03%) Class II421 (22.08%) Class III489 (25.64%) Class IV112 (5.87%) None770 (40.38%) Hypertension1571 (82.38%) Dyslipidemia1124 (58.94%) Liver disease52 (2.73%) Family history of premature CAD304 (15.94%) Arrhythmia Persistent/permanent112 (5.87%) Paroxysmal201 (10.54%) None1594 (83.59%) Smoking history Current smoker312 (16.36%) Former smoker704 (36.92%) Never smoker891 (46.72%) Endocarditis105 (5.51%) Mediastinal radiation37 (1.94%) Cardiogenic shock45 (2.36%) Total albumin1063.66 ± 0.52 Ejection fraction (%)4755.86 ± 12.04 Hemoglobin42213.36 ± 2.06 Platelets (×103/μL)215217.65 ± 71.15 Degree of aortic insufficiency None466 (24.44%) Trivial/trace260 (13.63%) Mild529 (27.74%) Moderate309 (16.20%) Severe343 (17.99%) Society of Thoracic Surgeons Score5810.02 [0.01-0.03] Indication Aortic insufficiency310 (16.26%) Aortic stenosis1342 (70.37%) Endocarditis105 (5.51%) Root/aorta150 (7.87%) Redo operations193 (10.12%)Operative characteristics Operative status Elective1481 (77.66%) Urgent377 (19.77%) Emergency49 (2.57%) Operative approach Sternotomy1660 (87.05%) Ministernotomy219 (11.48%) Right thoracotomy28 (1.47%) Concomitant procedure Coronary artery bypass grafting715 (37.49%) Mitral or tricuspid valve procedure166 (8.70%) Antiarrhythmic procedure110 (5.77%) Root/aorta156 (8.18%) Crossclamp time (min)59134.04 ± 63.03 Cardiopulmonary bypass time (min)59166.38 ± 75.74 Valve size (mm) 1915 (0.79%) 21178 (9.33%) 23533 (27.95%) 25604 (31.67%) 27417 (21.87%) 29160 (8.39%) PPM BMI <30 Insignificant: >0.85754 (100%) Moderate: 0.66-0.850 (0%) Severe: ≤0.550 (0%) PPM BMI ≥30 Insignificant: >0.70560 (99.82%) Moderate: 0.56-0.701 (0.18%) Severe: ≤0.550 (0%)Categorical variables are represented as n (%). Normally distributed data expressed as mean ± SD. NYHA, New York Heart Association; CAD, coronary artery disease; PPM, patient-prosthesis mismatch; BMI, body mass index. Patient and operative details Categorical variables are represented as n (%). Normally distributed data expressed as mean ± SD. NYHA, New York Heart Association; CAD, coronary artery disease; PPM, patient-prosthesis mismatch; BMI, body mass index. Perioperative Outcomes Postoperative outcomes categorized by overall cohort and operative subset are presented in Table 2. Thirty-day all-cause mortality was 2.8% (N = 54) for the overall study population and 1.9% (N = 14) for isolated SAVR. Operative cause of mortality for the overall population is shown in Table E1; none of these deaths were due to early structural valve failure, thrombosis, or hemodynamic mismatch. The most common postoperative complication was atrial fibrillation, occurring in 33.8% of patients (N = 645); rates of stroke, renal failure, and sternal infection were low in this cohort. Permanent pacemaker implantation rates were similar across operative indications and procedural complexity, occurring in 5.2% of patients with aortic insufficiency versus 3.7% with AS (P = .218) and in 4.0% of isolated AVR versus 4.2% of multi-valve procedures (P = .89). Multi-valve procedures were associated with higher in-hospital mortality (P = .017), increased reoperation for bleeding (P = .002), and prolonged ventilation (P < .001). Median length of stay was shortest for isolated SAVR (6 [IQR, 5-9] days) and longest for multi-valve cases (12 [IQR, 8-20] days).Table 2Postoperative outcomesCharacteristicOverallN = 1907Isolated AVRN = 728AVR + CABGN = 715AVR + MV/TVN = 166AVR + antiarrhythmic procedureN = 110Root/aorta N = 156AVR + otherN = 32P valueIn-hospital mortality42 (2.2%)11 (1.5%)15 (2.1%)8 (4.8%)0 (0%)7 (4.5%)1 (3.1%).01730-d mortality54 (2.8%)14 (1.9%)19 (2.7%)11 (6.6%)3 (2.7%)6 (3.8%)1 (3.1%).052Reoperation134 (6.9%)42 (5.8%)46 (6.4%)22 (13.3%)3 (2.7%)19 (12.2%)2 (6.3%).002Prolonged ventilation211 (11.0%)38 (5.2%)91 (12.7%)43 (25.9%)6 (5.4%)30 (19.2%)3 (9.4%)<.001Pneumonia137 (7.2%)44 (6.0%)54 (7.6%)17 (10.2%)12 (10.9%)8 (5.1%)2 (6.3%).212Stroke53 (2.8%)15 (2.1%)28 (3.9%)4 (2.4%)3 (2.7%)3 (1.9%)0 (0%).689Renal failure96 (5.0%)17 (2.3%)38 (5.3%)17 (10.2%)3 (2.7%)17 (10.9%)4 (12.5%)<.001Atrial fibrillation645 (33.8%)204 (28.0%)292 (40.8%)56 (36.7%)29 (26.4%)57 (36.5%)7 (21.9%).004Pacemaker62 (3.3%)29 (4.0%)15 (2.1%)7 (4.2%)4 (3.6%)3 (1.9%)4 (12.5%).023Sternal infection5 (0.3%)1 (0.1%)1 (0.1%)1 (0.6%)0 (0%)1 (0.6%)1 (3.1%).055Sepsis29 (1.5%)8 (1.1%)12 (1.7%)4 (2.4%)0 (0%)4 (2.6%)1 (3.1%).490LOS (d)7 [5.5-12]6 [5-9]9 [6-13]12 [8-20]8 [6-13]7 [6-12]8 [6-17]<.001Categorical data represented as n (%). Non-normal numerical data expressed as median [IQR 25th- 75th quartile]. AVR, Aortic valve replacement; CABG, coronary artery bypass grafting; MV, mitral valve; TV, tricuspid valve; LOS, length of stay. Table 2Postoperative outcomesCharacteristicOverallN = 1907Isolated AVRN = 728AVR + CABGN = 715AVR + MV/TVN = 166AVR + antiarrhythmic procedureN = 110Root/aorta N = 156AVR + otherN = 32P valueIn-hospital mortality42 (2.2%)11 (1.5%)15 (2.1%)8 (4.8%)0 (0%)7 (4.5%)1 (3.1%).01730-d mortality54 (2.8%)14 (1.9%)19 (2.7%)11 (6.6%)3 (2.7%)6 (3.8%)1 (3.1%).052Reoperation134 (6.9%)42 (5.8%)46 (6.4%)22 (13.3%)3 (2.7%)19 (12.2%)2 (6.3%).002Prolonged ventilation211 (11.0%)38 (5.2%)91 (12.7%)43 (25.9%)6 (5.4%)30 (19.2%)3 (9.4%)<.001Pneumonia137 (7.2%)44 (6.0%)54 (7.6%)17 (10.2%)12 (10.9%)8 (5.1%)2 (6.3%).212Stroke53 (2.8%)15 (2.1%)28 (3.9%)4 (2.4%)3 (2.7%)3 (1.9%)0 (0%).689Renal failure96 (5.0%)17 (2.3%)38 (5.3%)17 (10.2%)3 (2.7%)17 (10.9%)4 (12.5%)<.001Atrial fibrillation645 (33.8%)204 (28.0%)292 (40.8%)56 (36.7%)29 (26.4%)57 (36.5%)7 (21.9%).004Pacemaker62 (3.3%)29 (4.0%)15 (2.1%)7 (4.2%)4 (3.6%)3 (1.9%)4 (12.5%).023Sternal infection5 (0.3%)1 (0.1%)1 (0.1%)1 (0.6%)0 (0%)1 (0.6%)1 (3.1%).055Sepsis29 (1.5%)8 (1.1%)12 (1.7%)4 (2.4%)0 (0%)4 (2.6%)1 (3.1%).490LOS (d)7 [5.5-12]6 [5-9]9 [6-13]12 [8-20]8 [6-13]7 [6-12]8 [6-17]<.001Categorical data represented as n (%). Non-normal numerical data expressed as median [IQR 25th- 75th quartile]. AVR, Aortic valve replacement; CABG, coronary artery bypass grafting; MV, mitral valve; TV, tricuspid valve; LOS, length of stay. Postoperative outcomes Categorical data represented as n (%). Non-normal numerical data expressed as median [IQR 25th- 75th quartile]. AVR, Aortic valve replacement; CABG, coronary artery bypass grafting; MV, mitral valve; TV, tricuspid valve; LOS, length of stay. Long-Term Outcomes The median clinical follow-up for mortality was 3.79 years (IQR, 1.02-6.76). Kaplan–Meier estimates of overall survival, reintervention for SVD, and HVD are presented in Figure 1. Freedom from all-cause mortality at 1, 5, 7, 10, and 12 years was 93.3%, 77.6%, 66.0%, 48.0%, and 37.8%, respectively. Survival did not differ by operative indication but was significantly lower in patients aged more than 65 years compared with those aged less than 50 years (P = .0011) and those aged 50 to 65 years (P < .0001) (Figure 2). At 12 years, isolated SAVR was associated with higher survival than nonisolated SAVR patients (47.16% vs 32.82%, P < .001). Among both isolated and nonisolated SAVRs, patients aged more than 65 years had a higher mortality risk than patients aged 50 to 65 years and less than 50 years (P < .001) (Figure E2 and Table E2). Cox regression analysis identified patients aged more than 65 years, NYHA Class IV, lower postoperative mean EF, and higher mean gradient as predictors for 12-year mortality (Table E3).Figure 1Kaplan–Meier curve displaying overall freedom from mortality, reintervention for SVD, and HVD. Shaded areas indicate 95% CI.Figure 2Kaplan–Meier curve displaying freedom from mortality stratified by (A) operative indication and (B) age group. AI, Aortic insufficiency; AS, aortic stenosis. Shaded areas indicate 95% CI. Kaplan–Meier curve displaying overall freedom from mortality, reintervention for SVD, and HVD. Shaded areas indicate 95% CI. Kaplan–Meier curve displaying freedom from mortality stratified by (A) operative indication and (B) age group. AI, Aortic insufficiency; AS, aortic stenosis. Shaded areas indicate 95% CI. Re-do surgery was performed in 4.8% of patients (N = 91), with a median time to reintervention of 4.2 years (IQR, 1.80-7.33). At 12 years, 80.5% of the cohort was free from reoperation. Patients operated on for endocarditis underwent earlier reintervention compared with those with AS (Figure E3). The most common reason for reintervention was SVD (2.67%, N = 51) (Figure E4). Freedom from SVD reintervention at 1, 5, 10, and 12 years was 99.8%, 98.4%, 90.7%, and 84.8%, respectively (Figure 1). Those aged less than 50 years experienced the shortest time to reintervention, followed by those aged 50 to 65 years and those aged more than 65 years (Figure 3). In the competing risks regression analysis, age more than 65 years (HR, 0.24, P = .003) and larger valves (25-29 mm) (HR, 0.38, P = .043) were protective against reintervention (Table E3), whereas body surface area (HR, 3.03, P = .054) was associated with high mortality risk, but this did not reach statistical significance.Figure 3Kaplan–Meier curve displaying freedom from reintervention for SVD in Perimount Magna-Ease aortic valve recipients stratified by age group. Shaded areas indicate 95% CI. Kaplan–Meier curve displaying freedom from reintervention for SVD in Perimount Magna-Ease aortic valve recipients stratified by age group. Shaded areas indicate 95% CI. Echocardiography Data Functional assessment of the implanted valve through postoperative TTE was available on 90.5% of patients, with a total of 6557 echocardiograms reviewed. Echocardiographic follow-up completeness among patients with at least 1 postoperative study was 51.0% at 1 year, 42.1% at 5 years, 20.6% at 8 years, 9.8% at 10 years, and 4.0% at 12 years; median follow-up time was 3.22 years (IQR, 0.55-6.05 years). Mean TPG was reported in 76.7% of studies. At 3 months, overall postoperative mTPG was 11.2 ± 5.9 mm Hg, decreased from preoperative mTPG of 35.4 ± 17.4 mm Hg (Table E4). For patients with AS alone, preoperative mTPG averaged 40.0 ± 15.0 mm Hg. Overall mTPG remained low over the study period, with a gradual increase to 16.0 ± 8.3 mm Hg at 12 years (Table E4). During the first 5 years, larger valves (25-29 mm) had lower mTPGs than smaller valves (19-23 mm) (P ≤ .001), although this difference diminished beyond this period (Figure 4 and Table E4). Patients aged more than 65 years had lower gradients than patients 50 to 65 years of age throughout the study period (Figure 4). VARC-3 defined stage 3 HVD occurred in 0.79% (N = 15) of patients. Freedom from stage 3 HVD was 100.0%, 99.1%, and 96.4% at 1, 5, and 7 years, respectively (Figure 1).Figure 4Mean gradients over 12 years stratified by (A) valve size group and (B) age group. Boxplots display median (line), IQR (bottom and top of box), error bars (1.5 times IQR), and outliers (circles).Figure 5LVMI in AS patients over 12 years stratified by valve size group. Boxplots display median (line), IQR (bottom and top of box), error bars (1.5 times IQR), and outliers (circles). Mean gradients over 12 years stratified by (A) valve size group and (B) age group. Boxplots display median (line), IQR (bottom and top of box), error bars (1.5 times IQR), and outliers (circles). LVMI in AS patients over 12 years stratified by valve size group. Boxplots display median (line), IQR (bottom and top of box), error bars (1.5 times IQR), and outliers (circles). In PMEAV recipients with severe preoperative AS as the indication for aortic valve replacement (AVR), LVMI declined significantly at all follow-up intervals (Table E5). Beta regression showed that higher LV EF was associated with LVMI reduction (−0.82 per unit), whereas higher postoperative gradient correlated with increased LVMI (+0.86 per unit) (Table E6). Although valve size was not statistically associated with LVMI change, both size groups demonstrated incremental LVMI regression over 12 years (Figure 5 and Table E7). Discussion The optimal choice of valve implant for SAVR has become complicated by an expanding range of mechanical and bioprosthetic prostheses, evolving evidence-based recommendations,12,18 and the well-established role of transcatheter devices.1 The PMEAV is the third-generation Carpentier-Edwards stented, supra-annular aortic valve bioprosthesis with promising mid-term clinical and hemodynamic outcomes8,9,11,19 yet limited information on its long-term performance. To our knowledge, the current study is the largest single-center series to expand on the safety profile of the PMEAV 3300 TFX implant in with 1907 patients, enabling robust longitudinal assessment of valve-related outcomes and age-stratified durability particularly in patients aged 50 to 65 years. Inclusion of concomitant procedures, reoperations, and complex endocarditis and aortic root pathology reflects contemporary SAVR practice, in which isolated AVR is increasingly less common. Although this heterogeneity may influence overall survival, we examined valve-related outcomes such as SVD, reintervention, and echocardiographic performance of the prosthesis that are less dependent on operative complexity. Given the sustained structural and hemodynamic durability of the PMEAV and favorable long-term survival, our data support the continued use of this bioprosthesis in clinical practice. Short-term outcomes in our cohort parallel contemporary PMEAV reports. One-year survival reached 93.3%, within the published range of 94.5 to 97.6%, despite our larger, higher risk population. Five-year survival was 77.6%, similar to reported series that range from 80.9 to 91.5%.9,10,15 Notably, 14% of our procedures were urgent or emergency compared with 4.9% in the study by Anselmi and colleagues, whereas 62% involved concomitant operations, compared with 34.2 to 55.3% in other cohorts.9,10,13,15 Additionally, one-third of patients presented in NYHA class III or IV, in contrast to the predominantly class II cohort described by Tsui and colleagues.9 These findings demonstrate that early and intermediate-term PMEAV outcomes remain robust even in patients with greater comorbidity and surgical complexity. Beyond 5 years, survival in our cohort diverged from other contemporary PMEAV series. Ten- and 12-year survival declined to 48.0% and 37.8%, respectively, lower than the 56.8% reported by Anselmi and colleagues15 and 66.7% by Kermen and colleagues10 at 10 years, and 54.0% at 12 years by Piperata and colleagues.13 Although mean age was similar across studies, our patients had a heavier comorbidity burden, underwent urgent or emergency surgery more frequently, and included more than 5% treated for active endocarditis, subgroups under-represented in comparator cohorts. Taken together, these factors likely reflect greater biological age and competing nonvalve mortality risks and are a possible explanation for the observed lower long-term survival. Notably, patients aged 50 to 65 years and less than 50 years displayed higher long-term survival than those aged more than 65 years after both isolated and nonisolated SAVR. One prior evaluation of the PMEAV13 has reported long-term survival outcomes in patients aged 65 years or less, although these analyses did not further stratify outcomes within the population aged less than 65 years and included fewer patients with 12-year follow-up compared with our cohort. After several landmark trials found no difference in long-term mortality between mechanical and bioprosthetic valves,20, 21, 22 current recommendations18 now permit use of either type of prosthesis in SAVR patients 50 to 65 years of age. Our findings align with survival data reported for earlier-generation bioprostheses and mechanical valves,23 further supporting PMEAV implantation in this age group. However, in addition to age more than 65 years, NYHA Class IV and lower postoperative EF were also found to be independent mortality predictors on regression analysis. Diabetes demonstrated an associated mortality risk but did not reach statistical significance. These findings suggest that younger patients with significant comorbidities may have a prognosis similar to or worse than older, lower-risk individuals. These nuances underscore the importance of tailoring valve selection beyond age alone and by accounting for individual clinical risk factors and patient-centered decision-making. An expected consequence of longer survival is an increased burden of reintervention and SVD. Freedom from reintervention at 12 years was 80.5%, with more than half of reoperations attributed to SVD, followed by endocarditis and nonaortic valve pathology. Our SVD-related reintervention rates, 90.8% at 10 years and 85.13% at 12 years by Kaplan–Meier estimates, are slightly higher than those in prior PMEAV series,13,15 likely reflecting our longer follow-up, comprehensive echocardiographic surveillance in more than 90% patients, and application of standardized VARC-3 definitions. Stage 3 HVD was consistent with previously reported values by Kermen and colleagues.10 Although some studies reported endocarditis as the sole driver of reintervention in the PMEAV compared with other bioprostheses,24 our data are consistent with most studies demonstrating late failure of the PMEAV by SVD in addition to other mechanisms. Because bioprostheses are more frequently offered to patients aged less than 65 years due to declining reoperation risk and increasing favorability of valve-in-valve TAVR, the current study provides clarity on age-stratified durability of the PMEAV in patients aged less than 50 years and patients aged 50 to 65 years. The group aged less than 50 years has a higher cumulative risk of SVD-related reintervention than the group aged 50 to 65 years, with an earlier decline noted before year 6. In contrast, the group aged 50 to 65 years demonstrated 10-year reintervention outcomes comparable to those aged more than 65 years, although only the latter group was associated with a 71% lower hazard of reintervention compared with age less than 50 years on competing risk regression. Body surface area (BSA) was associated with increased reintervention risk; however, this did not reach statistical significance in multivariable analysis. Ochi and colleagues25 similarly identified younger age and BSA, along with PPM and smoking, as significant risk factors to affect long-term valve performance. Although the PMEAV demonstrates exceptional mid-term and long-term durability,8,9,19,24,26 younger patients and those with higher BSA may benefit from closer longitudinal follow-up and consideration of either mechanical AVR or future valve-in-valve compatibility at the index operation. Smaller valves (19-23 mm) were also found to be independent risk factors for reintervention in our analysis. Few studies have observed a similar effect in smaller valves,13,19 likely a surrogate for higher gradients and PPM. In our cohort, smaller valves exhibited higher gradients than larger valves during the first 5 years, although both groups remained within physiologic ranges and the difference became negligible beyond 8 years. Overall, PMEAV demonstrated favorable long-term hemodynamic performance, with mTPG increasing modestly to 16.0 ± 8.3 at 12 years. These sustained low gradients likely reduced wall stress and promoted LV reverse remodeling,27 supported by consistent LVMI regression across all valve sizes over 12 years postoperatively. Beta regression analysis further supported this relationship, demonstrating increasing mean gradients and reduced ventricular function associated with decreased LVMI regression, regardless of valve size. Clinically, sustained LV mass regression reflects effective and durable relief of afterload and has been associated with improved functional recovery and long-term outcomes after SAVR.28 This provides a practical functional marker of prosthesis performance beyond survival and reintervention. Collectively, these findings reinforce the need for individualized prosthesis selection and long-term surveillance tailored to age, body habitus, and valve size. The current study extends evidence for durable long-term hemodynamic performance of the PMEAV; however, recent studies have begun reporting mid-term comparison between PMEAV and newer-generation bioprostheses, including the Avalus valve, with similar transvalvular gradients. Prior studies demonstrate comparable hemodynamic performance between the PMEAV and Avalus valve at mid-term comparison with similar gradients.29 The current study extends the evidence for durable long-term hemodynamic profile of the PMEAV. Long-term data on comparable aortic valves are required to determine clinical significance. The INSPIRIS RESILIA valve represents the next evolution of the Perimount platform and has shown encouraging mid-term outcomes30,31; however, true long-term durability data are not yet available. Because both prostheses share the same fundamental stented pericardial tissue design, the present long-term experience provides an important benchmark against which the performance of newer-generation valves will be interpreted. Limitations The retrospective, single-center design of our study limits its generalizability to other institutions. Although our center serves as the major regional referral hub, follow-up and reintervention data may have been missed if patients sought care at other centers. Aside from the initial preoperative study, patients only received routine echocardiography beyond 30 days if symptomatic or hospitalized, potentially overestimating mTPG in the follow-up cohort. Furthermore, the absence of standardized echocardiogram protocols resulted in grouping echocardiograms by time periods, and regurgitation or stenosis severity, peak velocities, and gradients were inconsistently available for longitudinal analysis of prosthetic function. Ideally, standardized annual TTE surveillance would provide better assessment of the evolution of the valve's hemodynamic properties; however, this would be costly to the healthcare system. Although the median follow-up reflects mid-term observation, this study provides valuable longitudinal clinical and echocardiographic benchmarks on a large cohort with real-world operative complexity for comparison with newer surgical and transcatheter valve therapies in contemporary practice. Conclusions Our 12-year evaluation of the largest single-institution PMEAV recipient cohort to date found that this valve confers acceptable rates of long-term mortality and excellent freedom from reoperation for SVD and HVD in the long term. The echocardiographic data obtained in this study reveal the PMEAV has low mean transprosthetic gradients at 12 years regardless of valve size with evidence of sustained reverse ventricular remodeling, previously unreported in the literature. Although reintervention rates were higher in patients aged less than 50 years, this study adds to the robust existing data supporting the use of the PMEAV in current practice. Conflict of Interest Statement The authors reported no conflicts of interest. The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.