Freedom Solo® versus Trifecta® bioprostheses: clinical and haemodynamic evaluation after propensity score matching

Freedom Solo® versus Trifecta® bioprostheses: clinical and haemodynamic evaluation after... Abstract OBJECTIVES The goal of this study was to compare the stentless Freedom Solo® (FS) and the stented Trifecta® (TF) aortic bioprostheses with regard to haemodynamic profile, left ventricular mass regression and early and late postoperative outcomes and survival. METHODS A longitudinal cohort study of consecutive patients undergoing aortic valve replacement (2009–16) with either the FS or TF at 1 centre was performed. Local databases and national records were queried. Prosthesis haemodynamics and left ventricular dimensions were obtained from postoperative echocardiograms. After propensity score matching (21 covariates), the Kaplan–Meier and competing risk analyses were performed for survival and the combined outcome of structural valve deterioration and endocarditis, respectively. Haemodynamics and mass regression were assessed by a mixed-effects model including propensity score as a covariate. RESULTS From a total sample of 397 patients with the FS and 525 TF bioprostheses with a median follow-up time of 4.0 (2.2–6.0) and 2.4 (1.4–3.7) years, respectively, a matched sample of 329 pairs was obtained. Matched groups showed no differences in survival (hazard ratio = 1.04, 95% confidence interval = 0.69–1.56) or cumulative hazards of combined outcome (subdistribution hazard ratio = 0.54, 95% confidence interval = 0.21–1.39). Although the TF showed an improved haemodynamic profile, no difference was found in mass regression. Patients with TF bioprostheses had higher rates of prolonged mechanical ventilation, whereas patients with the FS bioprosthesis showed higher rates of thrombocytopenia. CONCLUSIONS The TF showed an improved haemodynamic profile compared to the FS, but this did not translate into further reverse remodelling. Postoperative outcomes and survival rates were comparable for both bioprostheses. Long-term follow-up is needed for comparisons with classical bioprosthesis models. Aortic valve replacement, Bioprostheses, Haemodynamics, Durability INTRODUCTION The number of aortic valve procedures is increasing in the aging populations of the Western world. Bioprostheses are the mainstay of aortic valve replacement (AVR) in elderly persons, but their use is increasing in younger patients as well [1]. Although the impact on prognosis is still unclear, there is general agreement that prosthesis haemodynamics determines clinical success after AVR. Newly designed surgical valves were developed with a strong focus on haemodynamic performance. Sorin’s (now LivaNova) Freedom Solo® (FS) bioprosthesis is the latest evolution of pericardial stentless valves with a simplified, faster implantation procedure compared with other stentless valves [2–4]. Stentless prostheses are favoured precisely for their haemodynamic profile and their ability to mimic the native aortic valve. However, concerns have been raised about the risk of long-term structural valve deterioration (SVD) [5, 6]. St. Jude’s (now Abbott) Trifecta® (TF) is a new stented bioprosthesis with externally mounted pericardial leaflets that allow for theoretical optimization of effective orifice area (EOA). Its design has been proposed to compete with stentless prostheses while retaining the advantages of the stented frame. These 2 bovine pericardial bioprostheses have been widely adopted at most centres and we aimed to compare TF with FS, the most recently available models of their respective counterparts. The main goal of the present study was to compare the mid-term clinical outcomes and short-term haemodynamic performances of stentless FS and stented TF bioprostheses. MATERIALS AND METHODS Study design and patients In this longitudinal cohort study with prospectively collected data, we included consecutive patients undergoing AVR performed with either the FS or the TF (implanted at our centre since 2009 and 2011, respectively, until June 2016) at the Cardiothoracic Surgery Department of Centro Hospitalar São João, Porto, Portugal. Both cases of isolated AVR and combined procedures were included. Group allocation was the decision of the surgical team. Patients crossing-over at reoperation were excluded, as were those who had Bentall and aortic root enlargement procedures and a 29-mm TF prosthesis, because there was no counterpart for them in the FS group. Data sources Data were collected from the databases and clinical records of the Cardiothoracic Surgery Department. National records were consulted for clinical, echocardiographic follow-up and survival data. Data were censored on March 2017. This study was approved by the Centro Hospitalar São João ethics committee and was conducted in accordance with the 1964 Helsinki declaration and its later amendments. Informed consent was waived due to the retrospective nature of the study, and all data were anonymized for analysis. Surgical technique Bioprostheses were implanted in a supra-annular position under standard cardiopulmonary bypass (CPB) and hypothermic cardioplegic arrest with either transverse or longitudinal aortotomy. TF valves were sutured using interrupted polyester sutures, and FS valves were implanted using running polypropylene sutures. Postoperative vitamin K antagonists were prescribed routinely for 3 months in patients younger than 80 years, unless contraindicated. Outcomes Immediate postoperative events recorded were de novo atrial fibrillation (AF), permanent pacemaker implantation, renal impairment (duplication of serum creatinine values or need for renal replacement therapy), prolonged mechanical ventilation (>24 h), severe thrombocytopenia (platelet count <30 × 109/l), stroke, length of stay, early chest re-exploration for bleeding or cardiac tamponade and death. Prosthesis-related complications recorded were non-structural valve dysfunction, SVD and endocarditis. SVD during follow-up was defined according to the valve academic research consortium-2 (VARC-2) criteria: mean transprosthetic gradient (MTG) ≥20 mmHg, EOA ≤0.9–1.1 cm2 and/or Doppler velocity index ≤0.35 m/s, and/or moderate or severe prosthetic valve regurgitation [7]. Postoperative evaluation including clinical observation and transthoracic echocardiography was done at our centre’s outpatient clinic or, occasionally, at other centres. Left ventricular (LV) end-diastolic diameter (LVEDD), LV mass index, MTG, EOA and patient–prosthesis mismatch (PPM) were recorded. PPM was defined as moderate (0.65 cm2/m2 ≤ indexed EOA < 0.85 cm2/m2) or severe (indexed EOA <0.65 cm2/m2) [8]. LV mass was estimated using Devereux’s formula and indexed for body surface area. A composite outcome defined as SVD or prosthesis endocarditis was assessed [7]. Echocardiographic data for prosthesis haemodynamics and LV mass regression were obtained independently from non-fatal follow-up outcomes, and losses to follow-up were distinct for the various outcomes (Fig. 1). Figure 1: View largeDownload slide Study flowchart. AVR: aortic valve replacement; FS: Freedom Solo®; PPM: patient–prosthesis mismatch; SVD: structural valve deterioration; TF: Trifecta®. Figure 1: View largeDownload slide Study flowchart. AVR: aortic valve replacement; FS: Freedom Solo®; PPM: patient–prosthesis mismatch; SVD: structural valve deterioration; TF: Trifecta®. Statistical analysis Categorical and continuous variables are presented as count (valid percentage, excluding missing values) and mean (standard deviation) or median (interquartile range) according to their distribution. Incidence rates (hazard rates) are expressed as events per 100 person-years. Normality was assessed by the Shapiro–Wilk test. An unadjusted comparison between groups was carried out for preoperative characteristics, intraoperative findings and postoperative data. To account for measured confounding, we then performed a non-parsimonious 1:1 nearest-neighbour propensity score (PS) matching without replacement (calliper of 0.02) based on logistic regression including 21 clinically relevant preoperative independent covariates, regardless of differences between groups. The PS model was run on a complete case analysis framework due to the absence of missing data. For the remaining variables and outcomes, after auditing and double checking, missing data were found for 20 variables, and only 3 of these had missing values >2% (maximum 6%). Covariate balance was assessed by standardized bias. The Student’s t-test or the Mann–Whitney test was used for continuous variables and the χ2 or the Fisher’s exact test for categorical variables. The Bonferroni adjustment was used to correct for multiple testing. Kaplan–Meier survival analysis was used to evaluate all-cause mortality. For analysis of composite outcome, cumulative incidence function was modelled based on Fine and Gray’s proportional subdistribution hazards model considering death as a competing event. Because time to postoperative echocardiography differed between groups, the effect on LV mass index and LVEDD was estimated by applying a PS-adjusted mixed effects model, including the respective preoperative and postoperative values as the within-subject factor, group as the between-subject factor and time–group interactions. A P-value <0.05 was considered significant. Statistical analyses were run on Stata version 14.1 (StataCorp, College Station, TX, USA). RESULTS Sample characteristics A study flowchart is outlined in Fig. 1. From a total of 954 patients retrieved from our database, 32 were excluded. Sample characteristics for the remaining patients of the TF (n = 525) and the FS groups (n = 397) are presented in Table 1 (PS covariates) and Table 2 (other relevant variables), respectively. In the TF group, male gender was more prevalent as was active smoking and the Canadian Cardiovascular Society (CCS) Class III/IV, whereas the FS group showed a higher prevalence of AF. There was a non-significant trend for higher European System for Cardiac Operative Risk Evaluation II (EuroSCORE II) in the TF compared with the FS group (mean EuroSCORE II was 5.1 vs 4.3, respectively). Bioprostheses were implanted under various clinical scenarios with minor differences between groups. Patients in the FS and TF groups did not differ in LV dysfunction, aortic disease pathophysiology or aetiology, but both concomitant tricuspid valve intervention and AF ablation were performed more frequently in the FS group. Mean CPB and aortic cross-clamp times were similar between the groups. The 23-mm diameter was the most frequently implanted size for both prostheses (Fig. 2). After PS matching (329 pairs), groups were well balanced for all variables included in the PS model, as appraised by absolute standardized bias <10% (Table 1). From intraoperative findings, only concomitant AF ablation as a combined procedure remained statistically different between the groups (Table 2). Table 1: Sample characteristics before and after matching in the FS and TF groups     Percentage standardized bias and an absolute 10% interval cut-off used to assess balance between covariates are represented in the rightmost column. BMI: body mass index; CC: creatinine clearance; CKD: chronic kidney disease; COPD: chronic obstructive pulmonary disease; FS: Freedom Solo®; IQR: interquartile range; LV: left ventricle; NYHA: New York Heart Association; TF: Trifecta®. Table 1: Sample characteristics before and after matching in the FS and TF groups     Percentage standardized bias and an absolute 10% interval cut-off used to assess balance between covariates are represented in the rightmost column. BMI: body mass index; CC: creatinine clearance; CKD: chronic kidney disease; COPD: chronic obstructive pulmonary disease; FS: Freedom Solo®; IQR: interquartile range; LV: left ventricle; NYHA: New York Heart Association; TF: Trifecta®. Table 2: Other sample characteristics in unmatched and matched samples of the FS and TF groups   Unmatched   Matched   Variables  FS group (n = 397)  TF group (n = 525)  P-value  FS group (n = 329)  TF group (n = 329)  P-value  The EuroSCORE II, median (IQR)  2.7 (1.5–4.8)  2.9 (1.6–5.8)  0.81  2.6 (1.5–4.8)  2.9 (1.6–5.1)  1.00  CCS ≥III, n (%)  22 (6)  71 (14)  <0.001  20 (6)  34 (11)  0.54  Hypercholesterolemia, n (%)  266 (67)  335 (65)  1.00  231 (70)  211 (65)  1.00  Rheumatic aortic disease, n (%)  32 (8)  26 (5)  0.78  20 (6)  18 (6)  1.00  Degenerative aortic disease, n (%)  307 (78)  394 (75)  1.00  258 (79)  265 (80)  1.00  Aortic stenosis, n (%)  290 (73)  371 (71)  1.00  239 (73)  241 (73)  1.00   MTG (mmHg), median (IQR)  46 (38–59)  47 (38–57)  1.00  47 (38–60)  47 (39–56)  1.00   EOA (cm2), median (IQR)  0.8 (0.6–0.9)  0.80 (0.6–0.9)  1.00  0.8 (0.6–0.9)  0.8 (0.6–0.9)  1.00  Concomitant procedures    Mitral, n (%)  51 (13)  57 (11)  1.00  38 (12)  35 (11)  1.00    Tricuspid, n (%)  61 (15)  45 (9)  0.015  44 (13)  36 (11)  1.00    CABG, n (%)  115 (29)  166 (32)  1.00  94 (29)  107 (32)  1.00    AA replacement, n (%)  34 (9)  60 (11)  1.00  26 (8)  37 (11)  1.00    AF ablation, n (%)  38 (10)  11 (2)  <0.001  26 (8)  7 (2)  0.015  CPB time (min), median (IQR)  118 (94–160)  112 (87–154)  0.23  116 (94–158)  111 (85–156)  0.50  ACC time (min), median (IQR)  84 (66–112)  80 (61–106)  0.30  82 (66–112)  78 (60–108)  0.62    Unmatched   Matched   Variables  FS group (n = 397)  TF group (n = 525)  P-value  FS group (n = 329)  TF group (n = 329)  P-value  The EuroSCORE II, median (IQR)  2.7 (1.5–4.8)  2.9 (1.6–5.8)  0.81  2.6 (1.5–4.8)  2.9 (1.6–5.1)  1.00  CCS ≥III, n (%)  22 (6)  71 (14)  <0.001  20 (6)  34 (11)  0.54  Hypercholesterolemia, n (%)  266 (67)  335 (65)  1.00  231 (70)  211 (65)  1.00  Rheumatic aortic disease, n (%)  32 (8)  26 (5)  0.78  20 (6)  18 (6)  1.00  Degenerative aortic disease, n (%)  307 (78)  394 (75)  1.00  258 (79)  265 (80)  1.00  Aortic stenosis, n (%)  290 (73)  371 (71)  1.00  239 (73)  241 (73)  1.00   MTG (mmHg), median (IQR)  46 (38–59)  47 (38–57)  1.00  47 (38–60)  47 (39–56)  1.00   EOA (cm2), median (IQR)  0.8 (0.6–0.9)  0.80 (0.6–0.9)  1.00  0.8 (0.6–0.9)  0.8 (0.6–0.9)  1.00  Concomitant procedures    Mitral, n (%)  51 (13)  57 (11)  1.00  38 (12)  35 (11)  1.00    Tricuspid, n (%)  61 (15)  45 (9)  0.015  44 (13)  36 (11)  1.00    CABG, n (%)  115 (29)  166 (32)  1.00  94 (29)  107 (32)  1.00    AA replacement, n (%)  34 (9)  60 (11)  1.00  26 (8)  37 (11)  1.00    AF ablation, n (%)  38 (10)  11 (2)  <0.001  26 (8)  7 (2)  0.015  CPB time (min), median (IQR)  118 (94–160)  112 (87–154)  0.23  116 (94–158)  111 (85–156)  0.50  ACC time (min), median (IQR)  84 (66–112)  80 (61–106)  0.30  82 (66–112)  78 (60–108)  0.62  AA: ascending aorta; ACC: aortic cross-clamp; AF: atrial fibrillation; CABG: coronary artery bypass grafting; CCS: Canadian Cardiovascular Society; CPB: cardiopulmonary bypass; EOA: effective orifice area; EuroSCORE II: European System for Cardiac Operative Risk Evaluation II; FS: Freedom Solo®; IQR: interquartile range; MTG: mean transprosthetic gradient; TF: Trifecta®. Table 2: Other sample characteristics in unmatched and matched samples of the FS and TF groups   Unmatched   Matched   Variables  FS group (n = 397)  TF group (n = 525)  P-value  FS group (n = 329)  TF group (n = 329)  P-value  The EuroSCORE II, median (IQR)  2.7 (1.5–4.8)  2.9 (1.6–5.8)  0.81  2.6 (1.5–4.8)  2.9 (1.6–5.1)  1.00  CCS ≥III, n (%)  22 (6)  71 (14)  <0.001  20 (6)  34 (11)  0.54  Hypercholesterolemia, n (%)  266 (67)  335 (65)  1.00  231 (70)  211 (65)  1.00  Rheumatic aortic disease, n (%)  32 (8)  26 (5)  0.78  20 (6)  18 (6)  1.00  Degenerative aortic disease, n (%)  307 (78)  394 (75)  1.00  258 (79)  265 (80)  1.00  Aortic stenosis, n (%)  290 (73)  371 (71)  1.00  239 (73)  241 (73)  1.00   MTG (mmHg), median (IQR)  46 (38–59)  47 (38–57)  1.00  47 (38–60)  47 (39–56)  1.00   EOA (cm2), median (IQR)  0.8 (0.6–0.9)  0.80 (0.6–0.9)  1.00  0.8 (0.6–0.9)  0.8 (0.6–0.9)  1.00  Concomitant procedures    Mitral, n (%)  51 (13)  57 (11)  1.00  38 (12)  35 (11)  1.00    Tricuspid, n (%)  61 (15)  45 (9)  0.015  44 (13)  36 (11)  1.00    CABG, n (%)  115 (29)  166 (32)  1.00  94 (29)  107 (32)  1.00    AA replacement, n (%)  34 (9)  60 (11)  1.00  26 (8)  37 (11)  1.00    AF ablation, n (%)  38 (10)  11 (2)  <0.001  26 (8)  7 (2)  0.015  CPB time (min), median (IQR)  118 (94–160)  112 (87–154)  0.23  116 (94–158)  111 (85–156)  0.50  ACC time (min), median (IQR)  84 (66–112)  80 (61–106)  0.30  82 (66–112)  78 (60–108)  0.62    Unmatched   Matched   Variables  FS group (n = 397)  TF group (n = 525)  P-value  FS group (n = 329)  TF group (n = 329)  P-value  The EuroSCORE II, median (IQR)  2.7 (1.5–4.8)  2.9 (1.6–5.8)  0.81  2.6 (1.5–4.8)  2.9 (1.6–5.1)  1.00  CCS ≥III, n (%)  22 (6)  71 (14)  <0.001  20 (6)  34 (11)  0.54  Hypercholesterolemia, n (%)  266 (67)  335 (65)  1.00  231 (70)  211 (65)  1.00  Rheumatic aortic disease, n (%)  32 (8)  26 (5)  0.78  20 (6)  18 (6)  1.00  Degenerative aortic disease, n (%)  307 (78)  394 (75)  1.00  258 (79)  265 (80)  1.00  Aortic stenosis, n (%)  290 (73)  371 (71)  1.00  239 (73)  241 (73)  1.00   MTG (mmHg), median (IQR)  46 (38–59)  47 (38–57)  1.00  47 (38–60)  47 (39–56)  1.00   EOA (cm2), median (IQR)  0.8 (0.6–0.9)  0.80 (0.6–0.9)  1.00  0.8 (0.6–0.9)  0.8 (0.6–0.9)  1.00  Concomitant procedures    Mitral, n (%)  51 (13)  57 (11)  1.00  38 (12)  35 (11)  1.00    Tricuspid, n (%)  61 (15)  45 (9)  0.015  44 (13)  36 (11)  1.00    CABG, n (%)  115 (29)  166 (32)  1.00  94 (29)  107 (32)  1.00    AA replacement, n (%)  34 (9)  60 (11)  1.00  26 (8)  37 (11)  1.00    AF ablation, n (%)  38 (10)  11 (2)  <0.001  26 (8)  7 (2)  0.015  CPB time (min), median (IQR)  118 (94–160)  112 (87–154)  0.23  116 (94–158)  111 (85–156)  0.50  ACC time (min), median (IQR)  84 (66–112)  80 (61–106)  0.30  82 (66–112)  78 (60–108)  0.62  AA: ascending aorta; ACC: aortic cross-clamp; AF: atrial fibrillation; CABG: coronary artery bypass grafting; CCS: Canadian Cardiovascular Society; CPB: cardiopulmonary bypass; EOA: effective orifice area; EuroSCORE II: European System for Cardiac Operative Risk Evaluation II; FS: Freedom Solo®; IQR: interquartile range; MTG: mean transprosthetic gradient; TF: Trifecta®. Figure 2: View largeDownload slide Percent distribution of prosthesis size and corresponding MTGs and EOAs in the groups of patients who underwent aortic valve replacement with either the FS or the TF. To improve visualization, standard errors of the mean are presented. ME for EOA and MTG for group and AoPsize are given inline. AoPsize: aortic prosthesis size; EOA: effective orifice area; FS: Freedom Solo®; ME: main effects; MTG: mean transprosthetic gradient; TF: Trifecta®. Figure 2: View largeDownload slide Percent distribution of prosthesis size and corresponding MTGs and EOAs in the groups of patients who underwent aortic valve replacement with either the FS or the TF. To improve visualization, standard errors of the mean are presented. ME for EOA and MTG for group and AoPsize are given inline. AoPsize: aortic prosthesis size; EOA: effective orifice area; FS: Freedom Solo®; ME: main effects; MTG: mean transprosthetic gradient; TF: Trifecta®. Early postoperative outcomes In the unmatched sample, TF showed a higher 30-day mortality rate and more postoperative complications such as need for inotropes or intra-aortic balloon pump, prolonged mechanical ventilation and stroke. Cause of death was cardiac in 3 and 20 patients and infectious in 3 and 8 patients of the FS and TF groups, respectively. One patient in the TF group died of uncontrolled bleeding. Nevertheless, and despite the trend for more deaths, only prolonged mechanical ventilation remained statistically more frequent after matching (Table 3). The FS group showed a higher incidence of severe thrombocytopenia, though it did not translate into increased rate of reoperation. Hospital length of stay and worsening kidney function were very similar between the groups, as were other early postoperative outcomes (Table 3). One patient in each group was reoperated due to sternal wound complications/mediastinitis; 1 patient in the TF group needed early redo AVR due to stent distortion and 3 other patients in the TF group underwent urgent coronary artery bypass surgery due to coronary ostial obstruction. Table 3: Early and late postoperative outcomes in matched sample of the FS and TF groups Outcome  FS (n = 329)  TF (n = 329)  P-value  Early postoperative period (30 days)   Mortality, n (%)  5 (2)  16 (5)  0.21   Hospital length of stay (days), median (IQR)  7 (6–10)  8 (6–13)  0.46   Reoperation due to bleeding/tamponade, n (%)  4 (1)  11 (3)  0.95   Worsening kidney function, n (%)  17 (5)  12 (4)  1.00   Severe thrombocytopenia, n (%)  26 (8)  10 (3)  0.048   Infusion of ≥ 2 inotropes ± IABP, n (%)  87 (29)  119 (38)  0.28   Prolonged mechanical ventilation, n (%)  17 (5)  40 (12)  0.014   Stroke, n (%)  5 (2)  14 (4)  0.52   De novo atrial fibrillation, n (%)  112 (34)  91 (28)  1.00   Pacemaker implantation, n (%)  8 (2)  16 (5)  1.00  Late postoperative period (>30 days)   SVD, n (%)  17 (5)  3 (1)  0.098   Prosthetic endocarditis, n (%)  6 (2)  5 (1)  1.00   Reoperation due to SVD/endocarditis, n (%)  10 (3)  2 (0)  0.28   Moderate or severe PPM, n (%)  39 (14)  40 (14)  1.00  Outcome  FS (n = 329)  TF (n = 329)  P-value  Early postoperative period (30 days)   Mortality, n (%)  5 (2)  16 (5)  0.21   Hospital length of stay (days), median (IQR)  7 (6–10)  8 (6–13)  0.46   Reoperation due to bleeding/tamponade, n (%)  4 (1)  11 (3)  0.95   Worsening kidney function, n (%)  17 (5)  12 (4)  1.00   Severe thrombocytopenia, n (%)  26 (8)  10 (3)  0.048   Infusion of ≥ 2 inotropes ± IABP, n (%)  87 (29)  119 (38)  0.28   Prolonged mechanical ventilation, n (%)  17 (5)  40 (12)  0.014   Stroke, n (%)  5 (2)  14 (4)  0.52   De novo atrial fibrillation, n (%)  112 (34)  91 (28)  1.00   Pacemaker implantation, n (%)  8 (2)  16 (5)  1.00  Late postoperative period (>30 days)   SVD, n (%)  17 (5)  3 (1)  0.098   Prosthetic endocarditis, n (%)  6 (2)  5 (1)  1.00   Reoperation due to SVD/endocarditis, n (%)  10 (3)  2 (0)  0.28   Moderate or severe PPM, n (%)  39 (14)  40 (14)  1.00  FS: Freedom Solo®; IABP: intra-aortic balloon pump; IQR: interquartile range; PPM: patient–prosthesis mismatch; SVD: structural valve deterioration; TF: Trifecta®. Table 3: Early and late postoperative outcomes in matched sample of the FS and TF groups Outcome  FS (n = 329)  TF (n = 329)  P-value  Early postoperative period (30 days)   Mortality, n (%)  5 (2)  16 (5)  0.21   Hospital length of stay (days), median (IQR)  7 (6–10)  8 (6–13)  0.46   Reoperation due to bleeding/tamponade, n (%)  4 (1)  11 (3)  0.95   Worsening kidney function, n (%)  17 (5)  12 (4)  1.00   Severe thrombocytopenia, n (%)  26 (8)  10 (3)  0.048   Infusion of ≥ 2 inotropes ± IABP, n (%)  87 (29)  119 (38)  0.28   Prolonged mechanical ventilation, n (%)  17 (5)  40 (12)  0.014   Stroke, n (%)  5 (2)  14 (4)  0.52   De novo atrial fibrillation, n (%)  112 (34)  91 (28)  1.00   Pacemaker implantation, n (%)  8 (2)  16 (5)  1.00  Late postoperative period (>30 days)   SVD, n (%)  17 (5)  3 (1)  0.098   Prosthetic endocarditis, n (%)  6 (2)  5 (1)  1.00   Reoperation due to SVD/endocarditis, n (%)  10 (3)  2 (0)  0.28   Moderate or severe PPM, n (%)  39 (14)  40 (14)  1.00  Outcome  FS (n = 329)  TF (n = 329)  P-value  Early postoperative period (30 days)   Mortality, n (%)  5 (2)  16 (5)  0.21   Hospital length of stay (days), median (IQR)  7 (6–10)  8 (6–13)  0.46   Reoperation due to bleeding/tamponade, n (%)  4 (1)  11 (3)  0.95   Worsening kidney function, n (%)  17 (5)  12 (4)  1.00   Severe thrombocytopenia, n (%)  26 (8)  10 (3)  0.048   Infusion of ≥ 2 inotropes ± IABP, n (%)  87 (29)  119 (38)  0.28   Prolonged mechanical ventilation, n (%)  17 (5)  40 (12)  0.014   Stroke, n (%)  5 (2)  14 (4)  0.52   De novo atrial fibrillation, n (%)  112 (34)  91 (28)  1.00   Pacemaker implantation, n (%)  8 (2)  16 (5)  1.00  Late postoperative period (>30 days)   SVD, n (%)  17 (5)  3 (1)  0.098   Prosthetic endocarditis, n (%)  6 (2)  5 (1)  1.00   Reoperation due to SVD/endocarditis, n (%)  10 (3)  2 (0)  0.28   Moderate or severe PPM, n (%)  39 (14)  40 (14)  1.00  FS: Freedom Solo®; IABP: intra-aortic balloon pump; IQR: interquartile range; PPM: patient–prosthesis mismatch; SVD: structural valve deterioration; TF: Trifecta®. Prosthesis haemodynamics and ventricular mass regression During follow-up, 855 transthoracic echocardiography examinations were performed: 488 in the TF group and 367 in the FS group. Echocardiographic data were missing for 10% of patients due either to death (64%) or to loss to follow-up. The percentage lost to follow-up did not differ between the groups. Time to postoperative echocardiography differed between the FS and TF groups [141 (106–202) vs 121 (91–163) days, P < 0.001]. The TF group showed lower MTG and higher EOA (Fig. 2). Nevertheless, there was no difference in the incidence of PPM: severe PPM was observed in 2 patients with the FS and in 5 patients with the TF (Table 3). We found no statistically significant interaction between group and prosthesis size. Despite the differences in haemodynamic performance of the prostheses, reverse remodelling was comparable in both groups. The PS-adjusted model showed that LV mass index and LVEDD decreased to a similar extent at follow-up in the FS and TF groups (Fig. 3). Of note, LVEDD was higher in the TF group. Figure 3: View largeDownload slide Reverse remodelling after aortic valve replacement with either the FS or the TF bioprosthesis as assessed by the (A) LVMi and (B) LVEDD on follow-up echocardiography. To improve visualization, standard errors of mean are presented. Results of preoperative and follow-up echocardiographic examinations are compared; the ME of group and surgical intervention and their I (group*surgery) are presented. FS: Freedom Solo®; I: interaction; LVEDD: left ventricular end-diastolic dimensions; LVMi: left ventricular mass index; ME: main effects; TF: Trifecta®. Figure 3: View largeDownload slide Reverse remodelling after aortic valve replacement with either the FS or the TF bioprosthesis as assessed by the (A) LVMi and (B) LVEDD on follow-up echocardiography. To improve visualization, standard errors of mean are presented. Results of preoperative and follow-up echocardiographic examinations are compared; the ME of group and surgical intervention and their I (group*surgery) are presented. FS: Freedom Solo®; I: interaction; LVEDD: left ventricular end-diastolic dimensions; LVMi: left ventricular mass index; ME: main effects; TF: Trifecta®. Survival and late clinical outcomes Median follow-up time was 2.4 (1.4–3.7) and 4.0 (2.2–6.0) years in the TF and FS groups, respectively (P < 0.001), and maximum follow-up time was 5.8 and 7.9 years in the TF and FS groups. Death rates and reoperation status were available for all patients. Overall freedom from death at 6 years was 80% in the TF and 82% in the FS group, respectively. No difference was found in time to all-cause mortality (hazard ratio = 1.04, 95% confidence interval 0.69–1.56) in the matched sample (Fig. 4A). Follow-up data for the combined outcome of SVD and endocarditis were missing for 6% of patients, but in 88% of these due to death. Losses to follow-up were below 1% in both groups regardless of matching. In the matched sample, the cumulative incidence at 6 years for the composite outcome was 6% and 4% in patients with the FS and TF, respectively, and the corresponding incidence rates were 1.3/100 and 0.7/100 person-years. No differences were found in time to composite outcome (subdistribution hazard ratio = 0.54, 95% confidence interval 0.21–1.39) in the matched sample (Fig. 4B). The stacked cumulative incidence for death or the composite outcome showed that all-cause mortality was the dominant outcome (Fig. 4C). No differences were found in the incidence of endocarditis, but SVD was more common in patients with the FS (only before PS matching, Table 3). Most importantly, there were 9 cases with predominant stenotic SVD in the FS and none in the TF group. Ten patients with the FS and 2 with the TF were reoperated due to SVD or endocarditis. Most patients underwent reintervention due to prosthetic endocarditis, whereas SVD was responsible for redo surgery in 3 patients with the FS and 1 patient with the TF. A higher rate of paravalvular leak was found in the TF (9 cases, 1.7%) versus the FS (1 case, 0.3%) group (P = 0.003). Figure 4: View largeDownload slide Survival (A), cumulative incidence of the combined outcome of SVD or endocarditis (B) and stacked cumulative incidences of death and combined outcome (C) in the matched sample of patients who underwent aortic valve replacement either with the FS or the TF bioprosthesis. Kaplan–Meier curves and their corresponding 95% confidence intervals are plotted for both groups (A) and for the cumulative subdistribution hazards of the combined outcome (B). The number of patients at risk is presented. Stacked cumulative incidence plots for both outcomes and groups are represented by different shades of grey as indicated in the conforming legend (C). No differences were found between groups. FS: Freedom Solo®; SVD: structural valve deterioration; TF: Trifecta®. Figure 4: View largeDownload slide Survival (A), cumulative incidence of the combined outcome of SVD or endocarditis (B) and stacked cumulative incidences of death and combined outcome (C) in the matched sample of patients who underwent aortic valve replacement either with the FS or the TF bioprosthesis. Kaplan–Meier curves and their corresponding 95% confidence intervals are plotted for both groups (A) and for the cumulative subdistribution hazards of the combined outcome (B). The number of patients at risk is presented. Stacked cumulative incidence plots for both outcomes and groups are represented by different shades of grey as indicated in the conforming legend (C). No differences were found between groups. FS: Freedom Solo®; SVD: structural valve deterioration; TF: Trifecta®. DISCUSSION We found an excellent haemodynamic profile of the stentless FS and the stented TF pericardial bioprostheses with a slight advantage for the TF but with similar LV mass regression. SVD, endocarditis rates and all-cause mortality rates were similar between the groups. Haemodynamic performance of surgical aortic valves is of paramount importance. PPM carries a worse prognosis [8, 9], and small aortic prostheses with a residual gradient impair LV mass regression, which is predictive of poorer survival and more heart failure readmissions [9, 10]. Classical surgical bioprostheses show a high incidence of severe PPM, reaching almost one-third of the patients in some reports [9]. However, different valve designs might have a substantial impact on these parameters, and a relatively large dispersion of postoperative MTG can be found in the literature. In our cohort, we could demonstrate that an excellent haemodynamic profile can be obtained with newer bioprostheses designed specifically with this goal. Both the FS and TF bioprostheses have been shown previously to outperform classical pericardial valves, namely the Perimount Magna [10, 11]. But to our knowledge, only 1 small cohort study focusing on immediate/short-term outcomes has previously compared these 2 bioprostheses [12]. Our EOA and MTG results for the TF and FS are similar to those previously published [6, 13]. Interestingly, in our study, the stented design showed a small but significant haemodynamic advantage, even after adjustment for the PS and stratification for prosthesis size. Nevertheless, the clinical significance of this haemodynamic superiority does not seem to be relevant because no differences were found in LV mass regression, PPM, combined outcome of SVD and endocarditis or survival. These results bring into question the role of a stentless design in achieving optimal haemodynamics. We found a higher incidence of immediate postoperative complications in unmatched cases with TF compared with FS, including death and stroke, which may be partly explained by the trend towards higher surgical risk, as assessed by the EuroSCORE II. Methodological issues inherent in the study design may additionally explain this finding. However, we might speculate that other unmeasured factors may also play a role: severe root and annular calcification is a relative contraindication to FS; additionally, stentless valves are considered to be technically more demanding and therefore more likely to be implanted by experienced surgeons. Because stentless FS implantation is more complex and time consuming [11], one would anticipate that the FS would have a lower number of combined procedures and extended periods of CPB compared with the TF. Yet, we found trends for precisely the opposite: a higher number of combined procedures with very similar CPB/aortic cross-clamp times. Except for the above-mentioned exclusion criteria, our analysis was carried out in an all-comers sample, reflecting our centre’s current surgical practice. Both prostheses were readily available to the surgical teams in diversified clinical scenarios, namely reoperations, multiple procedures and bicuspid aortic valves. Under real-world surgical practice settings, both prostheses performed well. Although our results show more than 5 years of follow-up with a reasonable number of at-risk patients (>200 at 4 years), future studies should address long-term durability. Older models of biological aortic valves, despite their questionable haemodynamics, have performed remarkably in this regard [14, 15]. Our follow-up data confirm a usually overlooked detail of post-AVR patients: the cumulative incidence of death is considerably higher than the cumulative incidence of SVD, endocarditis or reoperation [14, 15]. The competing risk of death is an important hurdle to non-fatal valve-related event analysis. We tried to address this issue by modelling the cumulative incidence function as accounting for the competing risk of death [16]. Although the low number of events precludes a definite conclusion, we found a trend towards an increased rate of SVD, mostly stenotic, in the FS group, which challenges the anticipated mechanism of stentless valve failure [5]. Two independent reports [6, 17] have shown that failure of the FS is mainly due to stenosis, and concerns have been raised about the risk of early SVD [18]. Unfortunately, the definition of SVD is not consistent between studies, and it should be noted that the anticipated mode of failure of stentless bioprostheses has not been frequently verified [5]. In our series, we defined SVD according to the VARC-2 criteria, which tend to overestimate prosthesis stenosis in non-reoperated patients [7, 17]. The TF bioprosthesis on the other hand was introduced more recently, so long-term results are scarce [13, 19, 20]. Questions have been raised about the risk of early SVD of TF valves [18, 21]. Our series reports 1 case of severe aortic regurgitation caused by stent distortion and 3 cases of need for urgent coronary artery bypass surgery due to coronary ostial obstruction in the early postoperative period. Others have speculated about the possible long-term impact of minor stent deformations that are not detected intraoperatively [18]. As may happen with the FS, if the implantation technique is not done correctly, the long-term durability of the TF might also be affected. Limitations PS analysis does not address unmeasured confounding. Follow-up time was limited and longer for the FS valve, which was available for surgical implantation earlier in our department, a bias which can only be partly controlled for. Time to postoperative transthoracic echocardiography was longer with the TF for logistic reasons (growing number of surgical cases and longer waiting time until follow-up echocardiography). This issue was partly considered by our statistical model. Many determinants of LV mass regression were not controlled for, namely postoperative systemic hypertension and AF. Therefore, our conclusions on mass regression deserve a word of caution. Power analysis for our sample size estimates shows an ability to detect 10% differences in survival on follow-up between groups (for 2-tailed significance set at 0.05) with a power above 80%. However, the low event rates of the remaining outcomes may preclude our ability to detect small differences. CONCLUSION In conclusion, TF and FS bioprostheses are good alternatives for AVR. Both achieve excellent haemodynamic performance and comparable degrees of LV mass regression. However, the acid test for these and other biological valves is necessarily durability, which will only be possible to determine with long-term studies. Funding This work was supported by the Project DOCnet (NORTE-01-0145-FEDER-000003), supported by Norte Portugal Regional Operational Programme (NORTE 2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF) and by the European Structural and Investment Funds (ESIF), under Lisbon Portugal Regional Operational Programme and National Funds through FCT—Foundation for Science and Technology under project (POCI-01-0145-FEDER-016385). Three of the authors received individual research support from the same programme. Conflict of interest: none declared. REFERENCES 1 Silaschi M, Conradi L, Treede H, Reiter B, Schaefer U, Blankenberg S et al.   Trends in Surgical aortic valve replacement in more than 3,000 consecutive cases in the era of transcatheter aortic valve implantations. Thorac Cardiovasc Surg  2016; 64: 382– 9. Google Scholar PubMed  2 Walther T, Falk V, Langebartels G, Kruger M, Bernhardt U, Diegeler A et al.   Prospectively randomized evaluation of stentless versus conventional biological aortic valves: impact on early regression of left ventricular hypertrophy. Circulation  1999; 100: II6– 10. Google Scholar PubMed  3 Ali A, Halstead JC, Cafferty F, Sharples L, Rose F, Coulden R et al.   Are stentless valves superior to modern stented valves? A prospective randomized trial. Circulation  2006; 114: I535– 40. Google Scholar CrossRef Search ADS PubMed  4 D'Onofrio A, Mazzucco A, Valfre C, Zussa C, Martinelli L, Casabona R et al.   Left ventricular remodeling, hemodynamics and early clinical outcomes after aortic valve replacement with the Pericarbon Freedom stentless bioprosthesis: results from the Italian Prospective Multicenter Trial. J Heart Valve Dis  2011; 20: 531– 9. Google Scholar PubMed  5 David TE. Invited commentary. Ann Thorac Surg  2016; 101: 109. Google Scholar CrossRef Search ADS PubMed  6 Sponga S, Barbera MD, Pavoni D, Lechiancole A, Mazzaro E, Valente M et al.   Ten-year results of the Freedom Solo stentless heart valve: excellent haemodynamics but progressive valve dysfunction in the long term. Interact CardioVasc Thorac Surg  2017; 24: 663– 9. Google Scholar CrossRef Search ADS PubMed  7 Kappetein AP, Head SJ, Genereux P, Piazza N, van Mieghem NM, Blackstone EH et al.   Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document (VARC-2). Eur J Cardiothorac Surg  2012; 42: S45– 60. Google Scholar CrossRef Search ADS PubMed  8 Pibarot P, Dumesnil JG. Hemodynamic and clinical impact of prosthesis-patient mismatch in the aortic valve position and its prevention. J Am Coll Cardiol  2000; 36: 1131– 41. Google Scholar CrossRef Search ADS PubMed  9 Pibarot P, Weissman NJ, Stewart WJ, Hahn RT, Lindman BR, McAndrew T et al.   Incidence and sequelae of prosthesis-patient mismatch in transcatheter versus surgical valve replacement in high-risk patients with severe aortic stenosis: a PARTNER trial cohort–a analysis. J Am Coll Cardiol  2014; 64: 1323– 34. Google Scholar CrossRef Search ADS PubMed  10 Rubens FD, Gee YY, Ngu JM, Chen L, Burwash IG. Effect of aortic pericardial valve choice on outcomes and left ventricular mass regression in patients with left ventricular hypertrophy. J Thorac Cardiovasc Surg  2016; 152: 1291– 8.e2. Google Scholar CrossRef Search ADS PubMed  11 van der Straaten EPJ, Rademakers LM, van Straten AHM, Houterman S, Tan MESH, Soliman Hamad MA. Mid-term haemodynamic and clinical results after aortic valve replacement using the Freedom Solo stentless bioprosthesis versus the Carpentier Edwards Perimount stented bioprosthesis. Eur J Cardiothorac Surg  2016; 49: 1174– 80. Google Scholar CrossRef Search ADS PubMed  12 Bujnoch DJ, Feyrer R, Harig F, Weyand M, Heim C. Comparison of the SJM Trifecta™ bioprosthetic valve with Sorin Freedom Solo™ regarding thrombocytopenia, hemolysis and hemodynamic performance after aortic valve replacement surgery. Thorac Cardiovasc Surg  2015; 63: OP207. Google Scholar CrossRef Search ADS   13 Goldman S, Cheung A, Bavaria JE, Petracek MR, Groh MA, Schaff HV. Midterm, multicenter clinical and hemodynamic results for the Trifecta aortic pericardial valve. J Thorac Cardiovasc Surg  2017; 153: 561– 9 e2. Google Scholar CrossRef Search ADS PubMed  14 Johnston DR, Soltesz EG, Vakil N, Rajeswaran J, Roselli EE, Sabik JF3rd et al.   Long-term durability of bioprosthetic aortic valves: implications from 12, 569 implants. Ann Thorac Surg  2015; 99: 1239– 47. Google Scholar CrossRef Search ADS PubMed  15 David TE, Armstrong S, Maganti M. Hancock II bioprosthesis for aortic valve replacement: the gold standard of bioprosthetic valves durability? Ann Thorac Surg  2010; 90: 775– 81. Google Scholar CrossRef Search ADS PubMed  16 Capodanno D, Petronio AS, Prendergast B, Eltchaninoff H, Vahanian A, Modine T et al.   Standardized definitions of structural deterioration and valve failure in assessing long-term durability of transcatheter and surgical aortic bioprosthetic valves: a consensus statement from the European Association of Percutaneous Cardiovascular Interventions (EAPCI) endorsed by the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS). Eur J Cardiothorac Surg  2017; 52: 408– 17. Google Scholar CrossRef Search ADS PubMed  17 Repossini A, Fischlein T, Santarpino G, Schafer C, Claus B, Passaretti B et al.   Pericardial stentless valve for aortic valve replacement: long-term results. Ann Thorac Surg  2016; 102: 1956– 65. Google Scholar CrossRef Search ADS PubMed  18 Santarpino G, Pfeiffer S. Early degeneration of the St Jude Medical Trifecta bioprosthetic aortic valve: a problem of the leaflets or of the stent? J Thorac Cardiovasc Surg  2017; 154: 820. Google Scholar CrossRef Search ADS PubMed  19 Lehmann S, Meyer A, Schroeter T, Uhlemann M, Fischer J, Leontyev S et al.   Midterm durability and hemodynamic performance of a third-generation bovine pericardial prosthetic aortic valve: the Leipzig experience. Ann Thorac Surg  2017; 103: 1933– 9. Google Scholar CrossRef Search ADS PubMed  20 Anselmi A, Ruggieri VG, Lelong B, Flecher E, Corbineau H, Langanay T et al.   Mid-term durability of the Trifecta bioprosthesis for aortic valve replacement. J Thorac Cardiovasc Surg  2017; 153: 21– 8.e1. Google Scholar CrossRef Search ADS PubMed  21 Kalra A, Rehman H, Ramchandani M, Barker CM, Lawrie GM, Reul RM et al.   Early Trifecta valve failure: report of a cluster of cases from a tertiary care referral center. J Thorac Cardiovasc Surg  2017; 154: 1235– 40. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Journal of Cardio-Thoracic Surgery Oxford University Press

Freedom Solo® versus Trifecta® bioprostheses: clinical and haemodynamic evaluation after propensity score matching

Loading next page...
 
/lp/ou_press/freedom-solo-versus-trifecta-bioprostheses-clinical-and-haemodynamic-l7wjlzCbWX
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
ISSN
1010-7940
eISSN
1873-734X
D.O.I.
10.1093/ejcts/ezx472
Publisher site
See Article on Publisher Site

Abstract

Abstract OBJECTIVES The goal of this study was to compare the stentless Freedom Solo® (FS) and the stented Trifecta® (TF) aortic bioprostheses with regard to haemodynamic profile, left ventricular mass regression and early and late postoperative outcomes and survival. METHODS A longitudinal cohort study of consecutive patients undergoing aortic valve replacement (2009–16) with either the FS or TF at 1 centre was performed. Local databases and national records were queried. Prosthesis haemodynamics and left ventricular dimensions were obtained from postoperative echocardiograms. After propensity score matching (21 covariates), the Kaplan–Meier and competing risk analyses were performed for survival and the combined outcome of structural valve deterioration and endocarditis, respectively. Haemodynamics and mass regression were assessed by a mixed-effects model including propensity score as a covariate. RESULTS From a total sample of 397 patients with the FS and 525 TF bioprostheses with a median follow-up time of 4.0 (2.2–6.0) and 2.4 (1.4–3.7) years, respectively, a matched sample of 329 pairs was obtained. Matched groups showed no differences in survival (hazard ratio = 1.04, 95% confidence interval = 0.69–1.56) or cumulative hazards of combined outcome (subdistribution hazard ratio = 0.54, 95% confidence interval = 0.21–1.39). Although the TF showed an improved haemodynamic profile, no difference was found in mass regression. Patients with TF bioprostheses had higher rates of prolonged mechanical ventilation, whereas patients with the FS bioprosthesis showed higher rates of thrombocytopenia. CONCLUSIONS The TF showed an improved haemodynamic profile compared to the FS, but this did not translate into further reverse remodelling. Postoperative outcomes and survival rates were comparable for both bioprostheses. Long-term follow-up is needed for comparisons with classical bioprosthesis models. Aortic valve replacement, Bioprostheses, Haemodynamics, Durability INTRODUCTION The number of aortic valve procedures is increasing in the aging populations of the Western world. Bioprostheses are the mainstay of aortic valve replacement (AVR) in elderly persons, but their use is increasing in younger patients as well [1]. Although the impact on prognosis is still unclear, there is general agreement that prosthesis haemodynamics determines clinical success after AVR. Newly designed surgical valves were developed with a strong focus on haemodynamic performance. Sorin’s (now LivaNova) Freedom Solo® (FS) bioprosthesis is the latest evolution of pericardial stentless valves with a simplified, faster implantation procedure compared with other stentless valves [2–4]. Stentless prostheses are favoured precisely for their haemodynamic profile and their ability to mimic the native aortic valve. However, concerns have been raised about the risk of long-term structural valve deterioration (SVD) [5, 6]. St. Jude’s (now Abbott) Trifecta® (TF) is a new stented bioprosthesis with externally mounted pericardial leaflets that allow for theoretical optimization of effective orifice area (EOA). Its design has been proposed to compete with stentless prostheses while retaining the advantages of the stented frame. These 2 bovine pericardial bioprostheses have been widely adopted at most centres and we aimed to compare TF with FS, the most recently available models of their respective counterparts. The main goal of the present study was to compare the mid-term clinical outcomes and short-term haemodynamic performances of stentless FS and stented TF bioprostheses. MATERIALS AND METHODS Study design and patients In this longitudinal cohort study with prospectively collected data, we included consecutive patients undergoing AVR performed with either the FS or the TF (implanted at our centre since 2009 and 2011, respectively, until June 2016) at the Cardiothoracic Surgery Department of Centro Hospitalar São João, Porto, Portugal. Both cases of isolated AVR and combined procedures were included. Group allocation was the decision of the surgical team. Patients crossing-over at reoperation were excluded, as were those who had Bentall and aortic root enlargement procedures and a 29-mm TF prosthesis, because there was no counterpart for them in the FS group. Data sources Data were collected from the databases and clinical records of the Cardiothoracic Surgery Department. National records were consulted for clinical, echocardiographic follow-up and survival data. Data were censored on March 2017. This study was approved by the Centro Hospitalar São João ethics committee and was conducted in accordance with the 1964 Helsinki declaration and its later amendments. Informed consent was waived due to the retrospective nature of the study, and all data were anonymized for analysis. Surgical technique Bioprostheses were implanted in a supra-annular position under standard cardiopulmonary bypass (CPB) and hypothermic cardioplegic arrest with either transverse or longitudinal aortotomy. TF valves were sutured using interrupted polyester sutures, and FS valves were implanted using running polypropylene sutures. Postoperative vitamin K antagonists were prescribed routinely for 3 months in patients younger than 80 years, unless contraindicated. Outcomes Immediate postoperative events recorded were de novo atrial fibrillation (AF), permanent pacemaker implantation, renal impairment (duplication of serum creatinine values or need for renal replacement therapy), prolonged mechanical ventilation (>24 h), severe thrombocytopenia (platelet count <30 × 109/l), stroke, length of stay, early chest re-exploration for bleeding or cardiac tamponade and death. Prosthesis-related complications recorded were non-structural valve dysfunction, SVD and endocarditis. SVD during follow-up was defined according to the valve academic research consortium-2 (VARC-2) criteria: mean transprosthetic gradient (MTG) ≥20 mmHg, EOA ≤0.9–1.1 cm2 and/or Doppler velocity index ≤0.35 m/s, and/or moderate or severe prosthetic valve regurgitation [7]. Postoperative evaluation including clinical observation and transthoracic echocardiography was done at our centre’s outpatient clinic or, occasionally, at other centres. Left ventricular (LV) end-diastolic diameter (LVEDD), LV mass index, MTG, EOA and patient–prosthesis mismatch (PPM) were recorded. PPM was defined as moderate (0.65 cm2/m2 ≤ indexed EOA < 0.85 cm2/m2) or severe (indexed EOA <0.65 cm2/m2) [8]. LV mass was estimated using Devereux’s formula and indexed for body surface area. A composite outcome defined as SVD or prosthesis endocarditis was assessed [7]. Echocardiographic data for prosthesis haemodynamics and LV mass regression were obtained independently from non-fatal follow-up outcomes, and losses to follow-up were distinct for the various outcomes (Fig. 1). Figure 1: View largeDownload slide Study flowchart. AVR: aortic valve replacement; FS: Freedom Solo®; PPM: patient–prosthesis mismatch; SVD: structural valve deterioration; TF: Trifecta®. Figure 1: View largeDownload slide Study flowchart. AVR: aortic valve replacement; FS: Freedom Solo®; PPM: patient–prosthesis mismatch; SVD: structural valve deterioration; TF: Trifecta®. Statistical analysis Categorical and continuous variables are presented as count (valid percentage, excluding missing values) and mean (standard deviation) or median (interquartile range) according to their distribution. Incidence rates (hazard rates) are expressed as events per 100 person-years. Normality was assessed by the Shapiro–Wilk test. An unadjusted comparison between groups was carried out for preoperative characteristics, intraoperative findings and postoperative data. To account for measured confounding, we then performed a non-parsimonious 1:1 nearest-neighbour propensity score (PS) matching without replacement (calliper of 0.02) based on logistic regression including 21 clinically relevant preoperative independent covariates, regardless of differences between groups. The PS model was run on a complete case analysis framework due to the absence of missing data. For the remaining variables and outcomes, after auditing and double checking, missing data were found for 20 variables, and only 3 of these had missing values >2% (maximum 6%). Covariate balance was assessed by standardized bias. The Student’s t-test or the Mann–Whitney test was used for continuous variables and the χ2 or the Fisher’s exact test for categorical variables. The Bonferroni adjustment was used to correct for multiple testing. Kaplan–Meier survival analysis was used to evaluate all-cause mortality. For analysis of composite outcome, cumulative incidence function was modelled based on Fine and Gray’s proportional subdistribution hazards model considering death as a competing event. Because time to postoperative echocardiography differed between groups, the effect on LV mass index and LVEDD was estimated by applying a PS-adjusted mixed effects model, including the respective preoperative and postoperative values as the within-subject factor, group as the between-subject factor and time–group interactions. A P-value <0.05 was considered significant. Statistical analyses were run on Stata version 14.1 (StataCorp, College Station, TX, USA). RESULTS Sample characteristics A study flowchart is outlined in Fig. 1. From a total of 954 patients retrieved from our database, 32 were excluded. Sample characteristics for the remaining patients of the TF (n = 525) and the FS groups (n = 397) are presented in Table 1 (PS covariates) and Table 2 (other relevant variables), respectively. In the TF group, male gender was more prevalent as was active smoking and the Canadian Cardiovascular Society (CCS) Class III/IV, whereas the FS group showed a higher prevalence of AF. There was a non-significant trend for higher European System for Cardiac Operative Risk Evaluation II (EuroSCORE II) in the TF compared with the FS group (mean EuroSCORE II was 5.1 vs 4.3, respectively). Bioprostheses were implanted under various clinical scenarios with minor differences between groups. Patients in the FS and TF groups did not differ in LV dysfunction, aortic disease pathophysiology or aetiology, but both concomitant tricuspid valve intervention and AF ablation were performed more frequently in the FS group. Mean CPB and aortic cross-clamp times were similar between the groups. The 23-mm diameter was the most frequently implanted size for both prostheses (Fig. 2). After PS matching (329 pairs), groups were well balanced for all variables included in the PS model, as appraised by absolute standardized bias <10% (Table 1). From intraoperative findings, only concomitant AF ablation as a combined procedure remained statistically different between the groups (Table 2). Table 1: Sample characteristics before and after matching in the FS and TF groups     Percentage standardized bias and an absolute 10% interval cut-off used to assess balance between covariates are represented in the rightmost column. BMI: body mass index; CC: creatinine clearance; CKD: chronic kidney disease; COPD: chronic obstructive pulmonary disease; FS: Freedom Solo®; IQR: interquartile range; LV: left ventricle; NYHA: New York Heart Association; TF: Trifecta®. Table 1: Sample characteristics before and after matching in the FS and TF groups     Percentage standardized bias and an absolute 10% interval cut-off used to assess balance between covariates are represented in the rightmost column. BMI: body mass index; CC: creatinine clearance; CKD: chronic kidney disease; COPD: chronic obstructive pulmonary disease; FS: Freedom Solo®; IQR: interquartile range; LV: left ventricle; NYHA: New York Heart Association; TF: Trifecta®. Table 2: Other sample characteristics in unmatched and matched samples of the FS and TF groups   Unmatched   Matched   Variables  FS group (n = 397)  TF group (n = 525)  P-value  FS group (n = 329)  TF group (n = 329)  P-value  The EuroSCORE II, median (IQR)  2.7 (1.5–4.8)  2.9 (1.6–5.8)  0.81  2.6 (1.5–4.8)  2.9 (1.6–5.1)  1.00  CCS ≥III, n (%)  22 (6)  71 (14)  <0.001  20 (6)  34 (11)  0.54  Hypercholesterolemia, n (%)  266 (67)  335 (65)  1.00  231 (70)  211 (65)  1.00  Rheumatic aortic disease, n (%)  32 (8)  26 (5)  0.78  20 (6)  18 (6)  1.00  Degenerative aortic disease, n (%)  307 (78)  394 (75)  1.00  258 (79)  265 (80)  1.00  Aortic stenosis, n (%)  290 (73)  371 (71)  1.00  239 (73)  241 (73)  1.00   MTG (mmHg), median (IQR)  46 (38–59)  47 (38–57)  1.00  47 (38–60)  47 (39–56)  1.00   EOA (cm2), median (IQR)  0.8 (0.6–0.9)  0.80 (0.6–0.9)  1.00  0.8 (0.6–0.9)  0.8 (0.6–0.9)  1.00  Concomitant procedures    Mitral, n (%)  51 (13)  57 (11)  1.00  38 (12)  35 (11)  1.00    Tricuspid, n (%)  61 (15)  45 (9)  0.015  44 (13)  36 (11)  1.00    CABG, n (%)  115 (29)  166 (32)  1.00  94 (29)  107 (32)  1.00    AA replacement, n (%)  34 (9)  60 (11)  1.00  26 (8)  37 (11)  1.00    AF ablation, n (%)  38 (10)  11 (2)  <0.001  26 (8)  7 (2)  0.015  CPB time (min), median (IQR)  118 (94–160)  112 (87–154)  0.23  116 (94–158)  111 (85–156)  0.50  ACC time (min), median (IQR)  84 (66–112)  80 (61–106)  0.30  82 (66–112)  78 (60–108)  0.62    Unmatched   Matched   Variables  FS group (n = 397)  TF group (n = 525)  P-value  FS group (n = 329)  TF group (n = 329)  P-value  The EuroSCORE II, median (IQR)  2.7 (1.5–4.8)  2.9 (1.6–5.8)  0.81  2.6 (1.5–4.8)  2.9 (1.6–5.1)  1.00  CCS ≥III, n (%)  22 (6)  71 (14)  <0.001  20 (6)  34 (11)  0.54  Hypercholesterolemia, n (%)  266 (67)  335 (65)  1.00  231 (70)  211 (65)  1.00  Rheumatic aortic disease, n (%)  32 (8)  26 (5)  0.78  20 (6)  18 (6)  1.00  Degenerative aortic disease, n (%)  307 (78)  394 (75)  1.00  258 (79)  265 (80)  1.00  Aortic stenosis, n (%)  290 (73)  371 (71)  1.00  239 (73)  241 (73)  1.00   MTG (mmHg), median (IQR)  46 (38–59)  47 (38–57)  1.00  47 (38–60)  47 (39–56)  1.00   EOA (cm2), median (IQR)  0.8 (0.6–0.9)  0.80 (0.6–0.9)  1.00  0.8 (0.6–0.9)  0.8 (0.6–0.9)  1.00  Concomitant procedures    Mitral, n (%)  51 (13)  57 (11)  1.00  38 (12)  35 (11)  1.00    Tricuspid, n (%)  61 (15)  45 (9)  0.015  44 (13)  36 (11)  1.00    CABG, n (%)  115 (29)  166 (32)  1.00  94 (29)  107 (32)  1.00    AA replacement, n (%)  34 (9)  60 (11)  1.00  26 (8)  37 (11)  1.00    AF ablation, n (%)  38 (10)  11 (2)  <0.001  26 (8)  7 (2)  0.015  CPB time (min), median (IQR)  118 (94–160)  112 (87–154)  0.23  116 (94–158)  111 (85–156)  0.50  ACC time (min), median (IQR)  84 (66–112)  80 (61–106)  0.30  82 (66–112)  78 (60–108)  0.62  AA: ascending aorta; ACC: aortic cross-clamp; AF: atrial fibrillation; CABG: coronary artery bypass grafting; CCS: Canadian Cardiovascular Society; CPB: cardiopulmonary bypass; EOA: effective orifice area; EuroSCORE II: European System for Cardiac Operative Risk Evaluation II; FS: Freedom Solo®; IQR: interquartile range; MTG: mean transprosthetic gradient; TF: Trifecta®. Table 2: Other sample characteristics in unmatched and matched samples of the FS and TF groups   Unmatched   Matched   Variables  FS group (n = 397)  TF group (n = 525)  P-value  FS group (n = 329)  TF group (n = 329)  P-value  The EuroSCORE II, median (IQR)  2.7 (1.5–4.8)  2.9 (1.6–5.8)  0.81  2.6 (1.5–4.8)  2.9 (1.6–5.1)  1.00  CCS ≥III, n (%)  22 (6)  71 (14)  <0.001  20 (6)  34 (11)  0.54  Hypercholesterolemia, n (%)  266 (67)  335 (65)  1.00  231 (70)  211 (65)  1.00  Rheumatic aortic disease, n (%)  32 (8)  26 (5)  0.78  20 (6)  18 (6)  1.00  Degenerative aortic disease, n (%)  307 (78)  394 (75)  1.00  258 (79)  265 (80)  1.00  Aortic stenosis, n (%)  290 (73)  371 (71)  1.00  239 (73)  241 (73)  1.00   MTG (mmHg), median (IQR)  46 (38–59)  47 (38–57)  1.00  47 (38–60)  47 (39–56)  1.00   EOA (cm2), median (IQR)  0.8 (0.6–0.9)  0.80 (0.6–0.9)  1.00  0.8 (0.6–0.9)  0.8 (0.6–0.9)  1.00  Concomitant procedures    Mitral, n (%)  51 (13)  57 (11)  1.00  38 (12)  35 (11)  1.00    Tricuspid, n (%)  61 (15)  45 (9)  0.015  44 (13)  36 (11)  1.00    CABG, n (%)  115 (29)  166 (32)  1.00  94 (29)  107 (32)  1.00    AA replacement, n (%)  34 (9)  60 (11)  1.00  26 (8)  37 (11)  1.00    AF ablation, n (%)  38 (10)  11 (2)  <0.001  26 (8)  7 (2)  0.015  CPB time (min), median (IQR)  118 (94–160)  112 (87–154)  0.23  116 (94–158)  111 (85–156)  0.50  ACC time (min), median (IQR)  84 (66–112)  80 (61–106)  0.30  82 (66–112)  78 (60–108)  0.62    Unmatched   Matched   Variables  FS group (n = 397)  TF group (n = 525)  P-value  FS group (n = 329)  TF group (n = 329)  P-value  The EuroSCORE II, median (IQR)  2.7 (1.5–4.8)  2.9 (1.6–5.8)  0.81  2.6 (1.5–4.8)  2.9 (1.6–5.1)  1.00  CCS ≥III, n (%)  22 (6)  71 (14)  <0.001  20 (6)  34 (11)  0.54  Hypercholesterolemia, n (%)  266 (67)  335 (65)  1.00  231 (70)  211 (65)  1.00  Rheumatic aortic disease, n (%)  32 (8)  26 (5)  0.78  20 (6)  18 (6)  1.00  Degenerative aortic disease, n (%)  307 (78)  394 (75)  1.00  258 (79)  265 (80)  1.00  Aortic stenosis, n (%)  290 (73)  371 (71)  1.00  239 (73)  241 (73)  1.00   MTG (mmHg), median (IQR)  46 (38–59)  47 (38–57)  1.00  47 (38–60)  47 (39–56)  1.00   EOA (cm2), median (IQR)  0.8 (0.6–0.9)  0.80 (0.6–0.9)  1.00  0.8 (0.6–0.9)  0.8 (0.6–0.9)  1.00  Concomitant procedures    Mitral, n (%)  51 (13)  57 (11)  1.00  38 (12)  35 (11)  1.00    Tricuspid, n (%)  61 (15)  45 (9)  0.015  44 (13)  36 (11)  1.00    CABG, n (%)  115 (29)  166 (32)  1.00  94 (29)  107 (32)  1.00    AA replacement, n (%)  34 (9)  60 (11)  1.00  26 (8)  37 (11)  1.00    AF ablation, n (%)  38 (10)  11 (2)  <0.001  26 (8)  7 (2)  0.015  CPB time (min), median (IQR)  118 (94–160)  112 (87–154)  0.23  116 (94–158)  111 (85–156)  0.50  ACC time (min), median (IQR)  84 (66–112)  80 (61–106)  0.30  82 (66–112)  78 (60–108)  0.62  AA: ascending aorta; ACC: aortic cross-clamp; AF: atrial fibrillation; CABG: coronary artery bypass grafting; CCS: Canadian Cardiovascular Society; CPB: cardiopulmonary bypass; EOA: effective orifice area; EuroSCORE II: European System for Cardiac Operative Risk Evaluation II; FS: Freedom Solo®; IQR: interquartile range; MTG: mean transprosthetic gradient; TF: Trifecta®. Figure 2: View largeDownload slide Percent distribution of prosthesis size and corresponding MTGs and EOAs in the groups of patients who underwent aortic valve replacement with either the FS or the TF. To improve visualization, standard errors of the mean are presented. ME for EOA and MTG for group and AoPsize are given inline. AoPsize: aortic prosthesis size; EOA: effective orifice area; FS: Freedom Solo®; ME: main effects; MTG: mean transprosthetic gradient; TF: Trifecta®. Figure 2: View largeDownload slide Percent distribution of prosthesis size and corresponding MTGs and EOAs in the groups of patients who underwent aortic valve replacement with either the FS or the TF. To improve visualization, standard errors of the mean are presented. ME for EOA and MTG for group and AoPsize are given inline. AoPsize: aortic prosthesis size; EOA: effective orifice area; FS: Freedom Solo®; ME: main effects; MTG: mean transprosthetic gradient; TF: Trifecta®. Early postoperative outcomes In the unmatched sample, TF showed a higher 30-day mortality rate and more postoperative complications such as need for inotropes or intra-aortic balloon pump, prolonged mechanical ventilation and stroke. Cause of death was cardiac in 3 and 20 patients and infectious in 3 and 8 patients of the FS and TF groups, respectively. One patient in the TF group died of uncontrolled bleeding. Nevertheless, and despite the trend for more deaths, only prolonged mechanical ventilation remained statistically more frequent after matching (Table 3). The FS group showed a higher incidence of severe thrombocytopenia, though it did not translate into increased rate of reoperation. Hospital length of stay and worsening kidney function were very similar between the groups, as were other early postoperative outcomes (Table 3). One patient in each group was reoperated due to sternal wound complications/mediastinitis; 1 patient in the TF group needed early redo AVR due to stent distortion and 3 other patients in the TF group underwent urgent coronary artery bypass surgery due to coronary ostial obstruction. Table 3: Early and late postoperative outcomes in matched sample of the FS and TF groups Outcome  FS (n = 329)  TF (n = 329)  P-value  Early postoperative period (30 days)   Mortality, n (%)  5 (2)  16 (5)  0.21   Hospital length of stay (days), median (IQR)  7 (6–10)  8 (6–13)  0.46   Reoperation due to bleeding/tamponade, n (%)  4 (1)  11 (3)  0.95   Worsening kidney function, n (%)  17 (5)  12 (4)  1.00   Severe thrombocytopenia, n (%)  26 (8)  10 (3)  0.048   Infusion of ≥ 2 inotropes ± IABP, n (%)  87 (29)  119 (38)  0.28   Prolonged mechanical ventilation, n (%)  17 (5)  40 (12)  0.014   Stroke, n (%)  5 (2)  14 (4)  0.52   De novo atrial fibrillation, n (%)  112 (34)  91 (28)  1.00   Pacemaker implantation, n (%)  8 (2)  16 (5)  1.00  Late postoperative period (>30 days)   SVD, n (%)  17 (5)  3 (1)  0.098   Prosthetic endocarditis, n (%)  6 (2)  5 (1)  1.00   Reoperation due to SVD/endocarditis, n (%)  10 (3)  2 (0)  0.28   Moderate or severe PPM, n (%)  39 (14)  40 (14)  1.00  Outcome  FS (n = 329)  TF (n = 329)  P-value  Early postoperative period (30 days)   Mortality, n (%)  5 (2)  16 (5)  0.21   Hospital length of stay (days), median (IQR)  7 (6–10)  8 (6–13)  0.46   Reoperation due to bleeding/tamponade, n (%)  4 (1)  11 (3)  0.95   Worsening kidney function, n (%)  17 (5)  12 (4)  1.00   Severe thrombocytopenia, n (%)  26 (8)  10 (3)  0.048   Infusion of ≥ 2 inotropes ± IABP, n (%)  87 (29)  119 (38)  0.28   Prolonged mechanical ventilation, n (%)  17 (5)  40 (12)  0.014   Stroke, n (%)  5 (2)  14 (4)  0.52   De novo atrial fibrillation, n (%)  112 (34)  91 (28)  1.00   Pacemaker implantation, n (%)  8 (2)  16 (5)  1.00  Late postoperative period (>30 days)   SVD, n (%)  17 (5)  3 (1)  0.098   Prosthetic endocarditis, n (%)  6 (2)  5 (1)  1.00   Reoperation due to SVD/endocarditis, n (%)  10 (3)  2 (0)  0.28   Moderate or severe PPM, n (%)  39 (14)  40 (14)  1.00  FS: Freedom Solo®; IABP: intra-aortic balloon pump; IQR: interquartile range; PPM: patient–prosthesis mismatch; SVD: structural valve deterioration; TF: Trifecta®. Table 3: Early and late postoperative outcomes in matched sample of the FS and TF groups Outcome  FS (n = 329)  TF (n = 329)  P-value  Early postoperative period (30 days)   Mortality, n (%)  5 (2)  16 (5)  0.21   Hospital length of stay (days), median (IQR)  7 (6–10)  8 (6–13)  0.46   Reoperation due to bleeding/tamponade, n (%)  4 (1)  11 (3)  0.95   Worsening kidney function, n (%)  17 (5)  12 (4)  1.00   Severe thrombocytopenia, n (%)  26 (8)  10 (3)  0.048   Infusion of ≥ 2 inotropes ± IABP, n (%)  87 (29)  119 (38)  0.28   Prolonged mechanical ventilation, n (%)  17 (5)  40 (12)  0.014   Stroke, n (%)  5 (2)  14 (4)  0.52   De novo atrial fibrillation, n (%)  112 (34)  91 (28)  1.00   Pacemaker implantation, n (%)  8 (2)  16 (5)  1.00  Late postoperative period (>30 days)   SVD, n (%)  17 (5)  3 (1)  0.098   Prosthetic endocarditis, n (%)  6 (2)  5 (1)  1.00   Reoperation due to SVD/endocarditis, n (%)  10 (3)  2 (0)  0.28   Moderate or severe PPM, n (%)  39 (14)  40 (14)  1.00  Outcome  FS (n = 329)  TF (n = 329)  P-value  Early postoperative period (30 days)   Mortality, n (%)  5 (2)  16 (5)  0.21   Hospital length of stay (days), median (IQR)  7 (6–10)  8 (6–13)  0.46   Reoperation due to bleeding/tamponade, n (%)  4 (1)  11 (3)  0.95   Worsening kidney function, n (%)  17 (5)  12 (4)  1.00   Severe thrombocytopenia, n (%)  26 (8)  10 (3)  0.048   Infusion of ≥ 2 inotropes ± IABP, n (%)  87 (29)  119 (38)  0.28   Prolonged mechanical ventilation, n (%)  17 (5)  40 (12)  0.014   Stroke, n (%)  5 (2)  14 (4)  0.52   De novo atrial fibrillation, n (%)  112 (34)  91 (28)  1.00   Pacemaker implantation, n (%)  8 (2)  16 (5)  1.00  Late postoperative period (>30 days)   SVD, n (%)  17 (5)  3 (1)  0.098   Prosthetic endocarditis, n (%)  6 (2)  5 (1)  1.00   Reoperation due to SVD/endocarditis, n (%)  10 (3)  2 (0)  0.28   Moderate or severe PPM, n (%)  39 (14)  40 (14)  1.00  FS: Freedom Solo®; IABP: intra-aortic balloon pump; IQR: interquartile range; PPM: patient–prosthesis mismatch; SVD: structural valve deterioration; TF: Trifecta®. Prosthesis haemodynamics and ventricular mass regression During follow-up, 855 transthoracic echocardiography examinations were performed: 488 in the TF group and 367 in the FS group. Echocardiographic data were missing for 10% of patients due either to death (64%) or to loss to follow-up. The percentage lost to follow-up did not differ between the groups. Time to postoperative echocardiography differed between the FS and TF groups [141 (106–202) vs 121 (91–163) days, P < 0.001]. The TF group showed lower MTG and higher EOA (Fig. 2). Nevertheless, there was no difference in the incidence of PPM: severe PPM was observed in 2 patients with the FS and in 5 patients with the TF (Table 3). We found no statistically significant interaction between group and prosthesis size. Despite the differences in haemodynamic performance of the prostheses, reverse remodelling was comparable in both groups. The PS-adjusted model showed that LV mass index and LVEDD decreased to a similar extent at follow-up in the FS and TF groups (Fig. 3). Of note, LVEDD was higher in the TF group. Figure 3: View largeDownload slide Reverse remodelling after aortic valve replacement with either the FS or the TF bioprosthesis as assessed by the (A) LVMi and (B) LVEDD on follow-up echocardiography. To improve visualization, standard errors of mean are presented. Results of preoperative and follow-up echocardiographic examinations are compared; the ME of group and surgical intervention and their I (group*surgery) are presented. FS: Freedom Solo®; I: interaction; LVEDD: left ventricular end-diastolic dimensions; LVMi: left ventricular mass index; ME: main effects; TF: Trifecta®. Figure 3: View largeDownload slide Reverse remodelling after aortic valve replacement with either the FS or the TF bioprosthesis as assessed by the (A) LVMi and (B) LVEDD on follow-up echocardiography. To improve visualization, standard errors of mean are presented. Results of preoperative and follow-up echocardiographic examinations are compared; the ME of group and surgical intervention and their I (group*surgery) are presented. FS: Freedom Solo®; I: interaction; LVEDD: left ventricular end-diastolic dimensions; LVMi: left ventricular mass index; ME: main effects; TF: Trifecta®. Survival and late clinical outcomes Median follow-up time was 2.4 (1.4–3.7) and 4.0 (2.2–6.0) years in the TF and FS groups, respectively (P < 0.001), and maximum follow-up time was 5.8 and 7.9 years in the TF and FS groups. Death rates and reoperation status were available for all patients. Overall freedom from death at 6 years was 80% in the TF and 82% in the FS group, respectively. No difference was found in time to all-cause mortality (hazard ratio = 1.04, 95% confidence interval 0.69–1.56) in the matched sample (Fig. 4A). Follow-up data for the combined outcome of SVD and endocarditis were missing for 6% of patients, but in 88% of these due to death. Losses to follow-up were below 1% in both groups regardless of matching. In the matched sample, the cumulative incidence at 6 years for the composite outcome was 6% and 4% in patients with the FS and TF, respectively, and the corresponding incidence rates were 1.3/100 and 0.7/100 person-years. No differences were found in time to composite outcome (subdistribution hazard ratio = 0.54, 95% confidence interval 0.21–1.39) in the matched sample (Fig. 4B). The stacked cumulative incidence for death or the composite outcome showed that all-cause mortality was the dominant outcome (Fig. 4C). No differences were found in the incidence of endocarditis, but SVD was more common in patients with the FS (only before PS matching, Table 3). Most importantly, there were 9 cases with predominant stenotic SVD in the FS and none in the TF group. Ten patients with the FS and 2 with the TF were reoperated due to SVD or endocarditis. Most patients underwent reintervention due to prosthetic endocarditis, whereas SVD was responsible for redo surgery in 3 patients with the FS and 1 patient with the TF. A higher rate of paravalvular leak was found in the TF (9 cases, 1.7%) versus the FS (1 case, 0.3%) group (P = 0.003). Figure 4: View largeDownload slide Survival (A), cumulative incidence of the combined outcome of SVD or endocarditis (B) and stacked cumulative incidences of death and combined outcome (C) in the matched sample of patients who underwent aortic valve replacement either with the FS or the TF bioprosthesis. Kaplan–Meier curves and their corresponding 95% confidence intervals are plotted for both groups (A) and for the cumulative subdistribution hazards of the combined outcome (B). The number of patients at risk is presented. Stacked cumulative incidence plots for both outcomes and groups are represented by different shades of grey as indicated in the conforming legend (C). No differences were found between groups. FS: Freedom Solo®; SVD: structural valve deterioration; TF: Trifecta®. Figure 4: View largeDownload slide Survival (A), cumulative incidence of the combined outcome of SVD or endocarditis (B) and stacked cumulative incidences of death and combined outcome (C) in the matched sample of patients who underwent aortic valve replacement either with the FS or the TF bioprosthesis. Kaplan–Meier curves and their corresponding 95% confidence intervals are plotted for both groups (A) and for the cumulative subdistribution hazards of the combined outcome (B). The number of patients at risk is presented. Stacked cumulative incidence plots for both outcomes and groups are represented by different shades of grey as indicated in the conforming legend (C). No differences were found between groups. FS: Freedom Solo®; SVD: structural valve deterioration; TF: Trifecta®. DISCUSSION We found an excellent haemodynamic profile of the stentless FS and the stented TF pericardial bioprostheses with a slight advantage for the TF but with similar LV mass regression. SVD, endocarditis rates and all-cause mortality rates were similar between the groups. Haemodynamic performance of surgical aortic valves is of paramount importance. PPM carries a worse prognosis [8, 9], and small aortic prostheses with a residual gradient impair LV mass regression, which is predictive of poorer survival and more heart failure readmissions [9, 10]. Classical surgical bioprostheses show a high incidence of severe PPM, reaching almost one-third of the patients in some reports [9]. However, different valve designs might have a substantial impact on these parameters, and a relatively large dispersion of postoperative MTG can be found in the literature. In our cohort, we could demonstrate that an excellent haemodynamic profile can be obtained with newer bioprostheses designed specifically with this goal. Both the FS and TF bioprostheses have been shown previously to outperform classical pericardial valves, namely the Perimount Magna [10, 11]. But to our knowledge, only 1 small cohort study focusing on immediate/short-term outcomes has previously compared these 2 bioprostheses [12]. Our EOA and MTG results for the TF and FS are similar to those previously published [6, 13]. Interestingly, in our study, the stented design showed a small but significant haemodynamic advantage, even after adjustment for the PS and stratification for prosthesis size. Nevertheless, the clinical significance of this haemodynamic superiority does not seem to be relevant because no differences were found in LV mass regression, PPM, combined outcome of SVD and endocarditis or survival. These results bring into question the role of a stentless design in achieving optimal haemodynamics. We found a higher incidence of immediate postoperative complications in unmatched cases with TF compared with FS, including death and stroke, which may be partly explained by the trend towards higher surgical risk, as assessed by the EuroSCORE II. Methodological issues inherent in the study design may additionally explain this finding. However, we might speculate that other unmeasured factors may also play a role: severe root and annular calcification is a relative contraindication to FS; additionally, stentless valves are considered to be technically more demanding and therefore more likely to be implanted by experienced surgeons. Because stentless FS implantation is more complex and time consuming [11], one would anticipate that the FS would have a lower number of combined procedures and extended periods of CPB compared with the TF. Yet, we found trends for precisely the opposite: a higher number of combined procedures with very similar CPB/aortic cross-clamp times. Except for the above-mentioned exclusion criteria, our analysis was carried out in an all-comers sample, reflecting our centre’s current surgical practice. Both prostheses were readily available to the surgical teams in diversified clinical scenarios, namely reoperations, multiple procedures and bicuspid aortic valves. Under real-world surgical practice settings, both prostheses performed well. Although our results show more than 5 years of follow-up with a reasonable number of at-risk patients (>200 at 4 years), future studies should address long-term durability. Older models of biological aortic valves, despite their questionable haemodynamics, have performed remarkably in this regard [14, 15]. Our follow-up data confirm a usually overlooked detail of post-AVR patients: the cumulative incidence of death is considerably higher than the cumulative incidence of SVD, endocarditis or reoperation [14, 15]. The competing risk of death is an important hurdle to non-fatal valve-related event analysis. We tried to address this issue by modelling the cumulative incidence function as accounting for the competing risk of death [16]. Although the low number of events precludes a definite conclusion, we found a trend towards an increased rate of SVD, mostly stenotic, in the FS group, which challenges the anticipated mechanism of stentless valve failure [5]. Two independent reports [6, 17] have shown that failure of the FS is mainly due to stenosis, and concerns have been raised about the risk of early SVD [18]. Unfortunately, the definition of SVD is not consistent between studies, and it should be noted that the anticipated mode of failure of stentless bioprostheses has not been frequently verified [5]. In our series, we defined SVD according to the VARC-2 criteria, which tend to overestimate prosthesis stenosis in non-reoperated patients [7, 17]. The TF bioprosthesis on the other hand was introduced more recently, so long-term results are scarce [13, 19, 20]. Questions have been raised about the risk of early SVD of TF valves [18, 21]. Our series reports 1 case of severe aortic regurgitation caused by stent distortion and 3 cases of need for urgent coronary artery bypass surgery due to coronary ostial obstruction in the early postoperative period. Others have speculated about the possible long-term impact of minor stent deformations that are not detected intraoperatively [18]. As may happen with the FS, if the implantation technique is not done correctly, the long-term durability of the TF might also be affected. Limitations PS analysis does not address unmeasured confounding. Follow-up time was limited and longer for the FS valve, which was available for surgical implantation earlier in our department, a bias which can only be partly controlled for. Time to postoperative transthoracic echocardiography was longer with the TF for logistic reasons (growing number of surgical cases and longer waiting time until follow-up echocardiography). This issue was partly considered by our statistical model. Many determinants of LV mass regression were not controlled for, namely postoperative systemic hypertension and AF. Therefore, our conclusions on mass regression deserve a word of caution. Power analysis for our sample size estimates shows an ability to detect 10% differences in survival on follow-up between groups (for 2-tailed significance set at 0.05) with a power above 80%. However, the low event rates of the remaining outcomes may preclude our ability to detect small differences. CONCLUSION In conclusion, TF and FS bioprostheses are good alternatives for AVR. Both achieve excellent haemodynamic performance and comparable degrees of LV mass regression. However, the acid test for these and other biological valves is necessarily durability, which will only be possible to determine with long-term studies. Funding This work was supported by the Project DOCnet (NORTE-01-0145-FEDER-000003), supported by Norte Portugal Regional Operational Programme (NORTE 2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF) and by the European Structural and Investment Funds (ESIF), under Lisbon Portugal Regional Operational Programme and National Funds through FCT—Foundation for Science and Technology under project (POCI-01-0145-FEDER-016385). Three of the authors received individual research support from the same programme. Conflict of interest: none declared. REFERENCES 1 Silaschi M, Conradi L, Treede H, Reiter B, Schaefer U, Blankenberg S et al.   Trends in Surgical aortic valve replacement in more than 3,000 consecutive cases in the era of transcatheter aortic valve implantations. Thorac Cardiovasc Surg  2016; 64: 382– 9. Google Scholar PubMed  2 Walther T, Falk V, Langebartels G, Kruger M, Bernhardt U, Diegeler A et al.   Prospectively randomized evaluation of stentless versus conventional biological aortic valves: impact on early regression of left ventricular hypertrophy. Circulation  1999; 100: II6– 10. Google Scholar PubMed  3 Ali A, Halstead JC, Cafferty F, Sharples L, Rose F, Coulden R et al.   Are stentless valves superior to modern stented valves? A prospective randomized trial. Circulation  2006; 114: I535– 40. Google Scholar CrossRef Search ADS PubMed  4 D'Onofrio A, Mazzucco A, Valfre C, Zussa C, Martinelli L, Casabona R et al.   Left ventricular remodeling, hemodynamics and early clinical outcomes after aortic valve replacement with the Pericarbon Freedom stentless bioprosthesis: results from the Italian Prospective Multicenter Trial. J Heart Valve Dis  2011; 20: 531– 9. Google Scholar PubMed  5 David TE. Invited commentary. Ann Thorac Surg  2016; 101: 109. Google Scholar CrossRef Search ADS PubMed  6 Sponga S, Barbera MD, Pavoni D, Lechiancole A, Mazzaro E, Valente M et al.   Ten-year results of the Freedom Solo stentless heart valve: excellent haemodynamics but progressive valve dysfunction in the long term. Interact CardioVasc Thorac Surg  2017; 24: 663– 9. Google Scholar CrossRef Search ADS PubMed  7 Kappetein AP, Head SJ, Genereux P, Piazza N, van Mieghem NM, Blackstone EH et al.   Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document (VARC-2). Eur J Cardiothorac Surg  2012; 42: S45– 60. Google Scholar CrossRef Search ADS PubMed  8 Pibarot P, Dumesnil JG. Hemodynamic and clinical impact of prosthesis-patient mismatch in the aortic valve position and its prevention. J Am Coll Cardiol  2000; 36: 1131– 41. Google Scholar CrossRef Search ADS PubMed  9 Pibarot P, Weissman NJ, Stewart WJ, Hahn RT, Lindman BR, McAndrew T et al.   Incidence and sequelae of prosthesis-patient mismatch in transcatheter versus surgical valve replacement in high-risk patients with severe aortic stenosis: a PARTNER trial cohort–a analysis. J Am Coll Cardiol  2014; 64: 1323– 34. Google Scholar CrossRef Search ADS PubMed  10 Rubens FD, Gee YY, Ngu JM, Chen L, Burwash IG. Effect of aortic pericardial valve choice on outcomes and left ventricular mass regression in patients with left ventricular hypertrophy. J Thorac Cardiovasc Surg  2016; 152: 1291– 8.e2. Google Scholar CrossRef Search ADS PubMed  11 van der Straaten EPJ, Rademakers LM, van Straten AHM, Houterman S, Tan MESH, Soliman Hamad MA. Mid-term haemodynamic and clinical results after aortic valve replacement using the Freedom Solo stentless bioprosthesis versus the Carpentier Edwards Perimount stented bioprosthesis. Eur J Cardiothorac Surg  2016; 49: 1174– 80. Google Scholar CrossRef Search ADS PubMed  12 Bujnoch DJ, Feyrer R, Harig F, Weyand M, Heim C. Comparison of the SJM Trifecta™ bioprosthetic valve with Sorin Freedom Solo™ regarding thrombocytopenia, hemolysis and hemodynamic performance after aortic valve replacement surgery. Thorac Cardiovasc Surg  2015; 63: OP207. Google Scholar CrossRef Search ADS   13 Goldman S, Cheung A, Bavaria JE, Petracek MR, Groh MA, Schaff HV. Midterm, multicenter clinical and hemodynamic results for the Trifecta aortic pericardial valve. J Thorac Cardiovasc Surg  2017; 153: 561– 9 e2. Google Scholar CrossRef Search ADS PubMed  14 Johnston DR, Soltesz EG, Vakil N, Rajeswaran J, Roselli EE, Sabik JF3rd et al.   Long-term durability of bioprosthetic aortic valves: implications from 12, 569 implants. Ann Thorac Surg  2015; 99: 1239– 47. Google Scholar CrossRef Search ADS PubMed  15 David TE, Armstrong S, Maganti M. Hancock II bioprosthesis for aortic valve replacement: the gold standard of bioprosthetic valves durability? Ann Thorac Surg  2010; 90: 775– 81. Google Scholar CrossRef Search ADS PubMed  16 Capodanno D, Petronio AS, Prendergast B, Eltchaninoff H, Vahanian A, Modine T et al.   Standardized definitions of structural deterioration and valve failure in assessing long-term durability of transcatheter and surgical aortic bioprosthetic valves: a consensus statement from the European Association of Percutaneous Cardiovascular Interventions (EAPCI) endorsed by the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS). Eur J Cardiothorac Surg  2017; 52: 408– 17. Google Scholar CrossRef Search ADS PubMed  17 Repossini A, Fischlein T, Santarpino G, Schafer C, Claus B, Passaretti B et al.   Pericardial stentless valve for aortic valve replacement: long-term results. Ann Thorac Surg  2016; 102: 1956– 65. Google Scholar CrossRef Search ADS PubMed  18 Santarpino G, Pfeiffer S. Early degeneration of the St Jude Medical Trifecta bioprosthetic aortic valve: a problem of the leaflets or of the stent? J Thorac Cardiovasc Surg  2017; 154: 820. Google Scholar CrossRef Search ADS PubMed  19 Lehmann S, Meyer A, Schroeter T, Uhlemann M, Fischer J, Leontyev S et al.   Midterm durability and hemodynamic performance of a third-generation bovine pericardial prosthetic aortic valve: the Leipzig experience. Ann Thorac Surg  2017; 103: 1933– 9. Google Scholar CrossRef Search ADS PubMed  20 Anselmi A, Ruggieri VG, Lelong B, Flecher E, Corbineau H, Langanay T et al.   Mid-term durability of the Trifecta bioprosthesis for aortic valve replacement. J Thorac Cardiovasc Surg  2017; 153: 21– 8.e1. Google Scholar CrossRef Search ADS PubMed  21 Kalra A, Rehman H, Ramchandani M, Barker CM, Lawrie GM, Reul RM et al.   Early Trifecta valve failure: report of a cluster of cases from a tertiary care referral center. J Thorac Cardiovasc Surg  2017; 154: 1235– 40. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

European Journal of Cardio-Thoracic SurgeryOxford University Press

Published: Jan 16, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off