Mid-term clinical and echocardiographic results of the INSPIRIS RESILIA aortic valve: a retrospective comparison to the Magna Ease

Abstract OBJECTIVES The INSPIRIS aortic valve combines the RESILIA proprietary tissue preservation process and an expandable stent frame to benefit future transcatheter valve-in-valve procedures. As the INSPIRIS valve became commercially available in 2017, mid-term outcome reports are scarce. We aimed to evaluate mid-term safety and echocardiographic performance of the INSPIRIS valve in comparison to its predecessor, the Carpentier Edwards Perimount Magna Ease (ME). METHODS This study was a retrospective single-centre study. Clinical results included early postoperative outcomes, mid-term mortality and readmission for cardiovascular cause or stroke. Echocardiographic follow-up (FU) was performed at discharge and 1–3, 6, 12 and 24 months. Clinical end point analyses were accomplished with a propensity score matching analysis and FU echocardiographic data comparisons using pairwise analyses and linear mixed-effect models. RESULTS We included 953 patients who received an INSPIRIS (n = 488) or ME (n = 463) bioprosthesis between January 2018 and July 2021. In the matched population (n = 217 per group), no significant difference in short-term outcomes was observed, survival was similar at 30 months (INSPIRIS: 94% vs ME: 91%, P = 0.89), but freedom from readmission was higher in the INSPIRIS group (94% vs 86%, P = 0.014). INSPIRIS valves had a lower gradient at discharge (∼10 vs 14 mmHg, P < 0.001), 1–3 months (∼10 vs 12 mmHg, P < 0.001) and 24 months (∼11 vs 17 mmHg, P < 0.001) in paired analyses and significantly lower evolution of mean transvalvular gradients compared to ME. CONCLUSIONS This study represents the largest comparative evaluation of the INSPIRIS to the ME valves, which demonstrated safe clinical outcomes and favourable haemodynamic performance at 2 years. Long-term FU is underway.


INTRODUCTION
In patients undergoing surgical aortic valve replacement (SAVR), choosing a bioprosthetic tissue valve avoids the need for lifelong anticoagulation at the cost of potential structural valve deterioration (SVD) and reintervention. In some cases, valve-in-valve (VinV) transcatheter aortic valve replacement (TAVR) may provide a lower-risk alternative to redo-SAVR, further tipping the preference towards the choice for a bioprosthetic valve [1]. However, not all patients will become candidates for VinV therapy and the ideal strategy for lifetime management of aortic stenosis remains unknown. In young patients, risks of permanent pacemaker implantation, paravalvular leaks, surgical risk of TAVR explantation and limited percutaneous experience in patients with bicuspid aortic valve may favour a SAVR-first strategy [1][2][3][4].
To reduce the risks of reintervention and improve patient outcomes, there is a need for surgical bioprosthetic valves with improved durability and performance.
The INSPIRIS RESILIA Aortic Valve (INSPIRIS; Edwards Lifesciences Corporation, Irvine, CA, USA) is a stented tri-leaflet bovine pericardial tissue valve for SAVR. The valve was made commercially available in 2017 and integrates 2 technological advances: (i) an expandable stent frame to minimize gradients of future VinV and reduce potential prosthesis-patient mismatch (PPM) and (ii) the RESILIA Tissue Integrity Preservation anti-calcification process to reduce leaflet calcification and risk of SVD.
While initial clinical investigation of the RESILIA tissue has provided encouraging results, reports of the INSPIRIS valve are limited to observational studies [5][6][7]. Considering previous early failures of new bioprosthetic valves and the conceptual advantages of the INSPIRIS valve which favours its utilisation in younger patients, close clinical and haemodynamic performance follow-up (FU) is paramount [8]. We aimed to compare mid-term clinical and echocardiographic outcomes of patients who received an INSPIRIS valve to those of patients who were implanted with a Carpentier Edwards Perimount Magna Ease (ME; Edwards Lifesciences Corporation, Irvine, CA, USA) valve during the same period.

Ethical statement
The study was approved by the local Ethics Committee (Quebec Heart and Lung Institute, 31 August 2022, # 20831) and, due to its retrospective nature, patient consent was waived.

Study population
We retrospectively interrogated our institution's electronic medical records for patients who underwent SAVR with the INSPIRIS valve from the time of the first INSPIRIS implant in our centre in January 2018 until the end of July 2021. A comparison group was formed from patients who had received a ME valve in our institution during the same timeframe. Included were patients 18 years or older who underwent elective primary SAVR, isolated or in combination with coronary artery bypass graft, aortic root surgery or other valvular repair or replacement. Patients with active or recent (<3 months) endocarditis, redo-procedures, patients with prior transcatheter aortic valve replacement and urgent procedures were excluded.

Procedure and follow-up
The decision on bioprosthesis valve type was individualized and based on patient preference, and left completely at the discretion of the surgeon. SAVR was performed via median sternotomy in all patients except for 2 who underwent minimally invasive procedures (i.e. not included in matched analyses). All patients were maintained at least on a single antiplatelet therapy with a daily low-dose of aspirin initiated in the early postoperative period, depending on concomitant procedures and comorbidities. Patients did not receive a vitamin-K antagonist following SAVR unless they underwent concomitant mitral valve surgery or had other indication for anticoagulation with a vitamin-K antagonist. The choice of antithrombotic regimen was subject to change at the discretion of the surgeon or referring cardiologist during FU. Routine FU included clinical evaluation and transthoracic echocardiography at discharge, 1-3, 6 and 12 months and then yearly. Patients completed their FU in our centre or referring hospital based on geographic distribution and efforts were deployed to retrieve external echocardiographic results.

Clinical and echocardiographic parameters
Baseline patient characteristics including clinical, laboratory and echocardiography data were obtained, and operative risk was assessed using the European System for Cardiac Operative Risk Evaluation II criteria. Intraoperative data included valve type and size, concomitant procedures, bypass and cross-clamp time. Operative mortality and postoperative outcomes were documented up to 30 days. All-cause mortality, readmissions for cardiovascular causes or stroke/transient ischaemic attack and aortic valve reinterventions were collected for the duration of FU. Echocardiographic outcomes were defined per the Valve Academic Research Consortium-3 criteria [9]. Degree of measured PPM was evaluated at 1-3 months with the echocardiographic measured effective orifice area (EOA) and categorized as mild/no PPM for EOA indexed to body surface area (EOAi) >0.85 cm 2 /m 2 , moderate for EOAi >0.65 cm 2 /m 2 but < _0.85 cm 2 /m 2 and severe for EOAi < _0.65 cm 2 /m 2 . In patients with a body mass index >30 kg/m 2 , the cutoffs were: mild >0.70 cm 2 /m 2 , moderate >0.55 but < _0.70 cm 2 /m 2 and severe < _0.55 cm 2 /m 2 .

Study end points
Clinical end points included 30-day mortality and postoperative adverse events, aortic valve reintervention and mid-term incidence of readmission for cardiovascular causes or stroke/transient ischaemic attack and death. Short-term outcomes were assessed with our internal database and mid-term by an in-depth interrogation of the provincial institution (Institut de la statistique du Qué bec) central database. Echocardiographic end points were mean transvalvular gradients (MG), EOAi, incidence of SVD and degree of PPM.

Statistical analysis
This manuscript follows the European Journal of Cardio-Thoracic Surgery and the Interactive CardioVascular and Thoracic Surgery guidelines for statistical and data reporting [10]. Results are expressed as standard descriptive statistics: absolute and relative frequencies for categorical variables and mean (standard deviation) or median and interquartile range (IQR) for continuous data. Unmatched data were compared between bioprosthesis groups using chi-squared or the Fisher exact test for categorical variables and the Student's t-test or Mann-Whitney test for continuous variables, as appropriate. Matched data were compared with McNemar, paired t-test or Wilcoxon signed rank test, as appropriate. Standardized mean differences were calculated to examine the characteristic distributions between groups after 1:1 propensity score matching (PSM). Standardized mean differences >0.15 or <-0.15 was considered imbalance. Early and mid-term clinical end points were assessed in the PSM cohort. Mid-term outcomes were depicted in Kaplan-Meier curves and compared with the stratified log-rank analyses. Univariate/multivariate Cox analyses were used to test the association of the ME prosthesis with the study end points after PSM. Evolution of MG and EOAi were assessed in the PSM cohort with pairwise analysis to weight for incomplete data. Linear mixed-effect models were also used to show and compare the evolution of both main haemodynamic performance variables (i.e. EOAi, MG) during FU [11]. The PSM has already been described elsewhere [12]. Ours included valve size, age, sex, body surface area, body mass index and left ventricle ejection fraction at baseline. Incidence of SVD and PPM was assessed in the PSM cohort with available echocardiographic FU. Changes between FUs were assessed using paired analysis. Data were analysed using SPSS software (V.28, IBM, Chicago, IL, USA) and Prism (GraphPad Software, San Diego, CA, USA). All measures were tested at a significance level of 0.05.

Patient characteristics
The study design is presented in Supplementary Table 1. In the unmatched population, patients in the INSPIRIS group were younger, with a mean age of 67 (7) vs 74 (7) years old in the ME group, and 83 (17%) vs 24 (5%) aged under 60, respectively (P < 0.001). Approximately 73% of all patients were male. Coronary artery disease and atrial fibrillation were more common in the ME group (coronary artery disease: 55 vs 48%, P = 0.03, atrial fibrillation: 24 vs 14%, P < 0.001). Patients in the INSPIRIS group were at lower operative risk (European System for Cardiac Operative Risk Evaluation II: 1.

Procedure data and early postoperative outcomes
Operative data of both unmatched and matched populations are presented in Table 2. Patients in the ME group were more likely to undergo a combined procedure (75 vs 63%, P < 0.001) of which coronary artery bypass graft was the most frequent (49% overall). There were 13 and 11% of concomitant ascending aorta replacement in the ME and INSPIRIS groups, respectively (P = 0.33). Concomitant mitral valve replacement was more common in patients receiving an ME (12 vs 6%, P = 0.002  Table 3. There were no significant differences in early postoperative outcomes between groups in the matched population (Table 3). Early adverse events stratified by isolated AVR are presented in Supplementary Material, Table S1.

Echocardiographic parameters
Both bioprostheses demonstrated good performance and stability over time. In the matched cohort, the INSPIRIS valve demonstrated superior haemodynamics to the ME valve at discharge, 1-3 months and 24 months ( Fig. 2A and B, see also Supplementary Material, Table S3). The mean gradients at 24 months in the INSPIRIS group were 11.4 (3.6) mmHg compared to 17.3 (6.6) mmHg in the ME group (P < 0.001). Linear mixed-effect analyses are shown in Fig. 3A and B and demonstrate similar results as paired analyses, with a significant difference in MG at 1-3 months and a strong tendency at 24 months between valves. Although there were numerically more severe PPM with the ME valve in the matched population (16% vs 7%), degree of PPM was not statistically different overall between both valves (P = 0.27; Fig. 4). There were 2 events of moderate SVD with the INSPIRIS and none with the ME valve (P = 0.87). Both prostheses were 21 mm and SVD was detected at 24 months. These patients did not require reintervention.

DISCUSSION
The aim of this study was to evaluate the mid-term safety and performance of the INSPIRIS aortic bioprosthesis and benchmark them against those of the ME. The main findings of the current study are as follows: (i) this study represents the largest 'real-life' comparative echocardiographic evaluation of the INSPIRIS bioprosthesis; (ii) the INSPIRIS valve demonstrated excellent early postoperative and mid-term clinical outcomes; and (iii) haemodynamics of the INSPIRIS valve are stable through 2-year FU and superior to those of the ME in a propensity score matched cohort.

Unique characteristics of the INSPIRIS valve and current state of evidence
The RESILIA tissue was first introduced in the Edwards Model 11000A valve (ME with RESILIA tissue). The tissue was initially tested in randomized study on juvenile sheep [13]. Forty-five animals received a ME valve with either standard or treated leaflets in the mitral position. At 8 months, the treated valves demonstrated lower gradients and less calcium content. Model 11000A was then evaluated in 2 large prospective nonrandomized single-arm reglementary approval trials. The EU RESILIA feasibility study recruited 133 patients between 2011 and 2013 at 2 sites in Poland [6]. The mean patient age was 65 (13.5) years, including 25% of patients under 60. The mean FU was 4.2 years. About 44% of patients received a 19-or 21-mm valve, and no SVD events were reported. The mean gradient and EOA at 5 years was 14.8 (7.6) mmHg and 1.4 (0.5) cm 2 , respectively. The larger COMMENCE trial was a multinational study conducted in 27 centres [5,7]. From 2013 to 2016, 689 patients were recruited, with a median FU of 4 years. The mean age was 66.9 (11.6) years with 21% of patients under 60 years old. All-cause early and 1-and 4-year mortality were 1.2%, 2.3% and 7.1%, respectively. At 4 years, EOA was 1.5 (0.5) cm 2 and the mean transvalvular gradient was 11 (5.6) mmHg (22% of patients received a 19-or 21-mm valve). There were also no reported cases of SVD.
The INSPIRIS (Model 11500A) combines the RESILIA tissue and the VFit Technology expandable stent frame technological advances. Conceptually, the latter could benefit younger patients, who risk outliving their SAVR, by allowing larger valves/lower gradients in future VinV therapy. However, there is a lack of publicly available data to support this feature. In contrast to Model11000A (ME with RESILIA tissue), there is little evidence in the safety and performance of the INSPIRIS valve. Only a few (4) centres have shared their experience [14][15][16][17]. The largest series regroups 100 patients with a 1-year echocardiographic FU [15]. However, most patients were aged <60 years, and 17% received a 19-or 21-mm valve. At 1 year, the mean EOA was 1.8 (0.1) cm 2 and the mean gradient was 11.5 (2.3) mmHg. The only comparative report that was similar to our study, in that it also compared the INSPIRIS to the ME valve, is the study by Shala et al. who presented discharge results in a series of 125 consecutive patients, 59 of whom underwent SAVR with an INSPIRIS valve [16]. In accordance with our findings, postoperative transthoracic echocardiography at 1-week demonstrated lower gradients in favour of the INSPIRIS valve (mean gradient of 10 vs 12 mmHg, P < 0.001).

Clinical outcomes
The current study demonstrated the safety of the INSPIRIS valve at 2-year FU. The valve demonstrated excellent postoperative outcomes with a 30-day mortality rate of 1% and 0% stroke rate for isolated AVR procedures. Postoperative outcomes were similar between both valves in the matched population. All-cause mortality at 1 year and 30 months of the ME valve was similar to that of the INSPIRIS valve, which was expected. It should be noted that due to the cost-sharing nature of our healthcare system, and as a general practice, the more expensive INSIPIRIS valve was generally proposed to patients <70 years old. Thus, patients in the INSPIRIS group were generally younger, underwent less combined procedures, and were overall at lower operative risk compared to patients in the ME group. Lower readmission rates with the INSPIRIS valve observed in the matched population may be attributable to the superior haemodynamic performance of the valve, but this hypothesis needs to be validated in further studies. Interaction analyses revealed no significant association with isolated versus non-isolated AVR regarding mid-term outcomes. There were also no reinterventions on the INSPIRIS valve. These results support the mid-term safety of the INSPIRIS valve.   studies with larger FU may demonstrate sustained better performance at long term. There were 2 events of moderate SVD in the INSPIRIS group at 2 years and none in the ME group. The difference was not statistically significant and is most likely attributable to the younger patient age in the INSPIRIS group. Interestingly, there were no SVD events reported in the 5-year results of the EU RESILIA and COMMENCE trials [5,6]. This is most likely attributable to differences in definitions of SVD, long-term FU of the INSPIRIS valve is necessary to better assess this end point. The degree of PPM was similar between both valve types and correlated with what is previously known in the literature [18].

Future directions
Our single-centre experience with the INSPIRIS valve is growing and longer echocardiographic FU is on-going. We project to share our 5-year clinical outcomes and echocardiographic results in the near future. Within the next 5 years, results from large multicentre studies of the INSPIRIS valve/RESILIA tissue with a focus on younger patients should also be available. The Indure Resilia Durability Registry is an open-label, multicentre, European registry with a 5year FU to assess clinical outcomes in 400 patients under 60 undergoing SAVR with the INSPIRIS [18]. The study was initiated in mid-2019 and is expected to be completed mid-2025. The RESILIENCE trial (NCT03680040) is a US-based study investigating time to valve failure due to valve deterioration requiring reintervention and exploring early potential predictors of valve durability in patients under 65 who underwent SAVR with a RESILIA tissue valve. The study aims to recruit 250 patients which will be followed up to 11 years with an expected completion end in 2027. Interestingly, this study will include valve calcification assessment by multi-slice computerized tomography. The IMPACT registry will recruit 500 patients across 25 EU sites and will have a particular focus on the impact of pre-existing comorbidities on patient outcomes and bioprosthetic valve performance, more specifically, performance of the INSPIRIS valve at a target FU of 5 years [19]. Finally, the INSPIRIS, Valve-in-Valve Surveillance Study (INVIVITY; NCT04902053) should provide initial clues to the potential advantages of the VFit technology.

Limitations
This study represents the largest echocardiographic comparative analysis of the INSPIRIS and ME bioprosthesis. Some limitations of the current study should be acknowledged. First, this study was retrospective in nature and is thus subject to its inherent limitations. Due to our cost-sharing healthcare system, younger patients were more likely to receive an INSPIRIS valve. The choice between the type of bioprosthesis was not standardized, but operating surgeons had large experience with both valves. The use of PSM has excluded several patients of this initial sample which could limit the generalization of the current findings to the general population. Due to multicentre FU echocardiographic assessment, there may be interobserver variations in assessing EOAi, but the potential variability should be equally distributed in both groups. Further prospective studies should use a standardized assessment of prosthesis measurement and should follow the Valve Labelling Task Force Position Statement [20]. As PPM was evaluated with echocardiographic data, there may be overestimations of the PPM rates, but we encourage the manufacturer to provide standardized charts for the INSPIRIS valve to allow the PPM assessment using the predicted method [20,21]. In analysing clinical outcomes, the matched population may not have accounted for all potential disparities between the 2 groups due to very different baseline presentation of our patients (e.g. Figure 3: Evolution of haemodynamic performance parameters in follow-up between INSPIRIS and ME valves. In linear mixed-effect models, the INSPIRIS demonstrated significantly better mean gradient (A) at 1-3 months and a strong tendency towards significant difference (P = 0.06) at 2 years compared to Magna Ease, but no significant differences were observed in EOAi (B). Abbreviations as in Figs 1 and 2. *P < 0.05 versus ME valve; † P < 0.10 versus ME. Figure 4: Degree of PPM with the INSPIRIS and ME valves. In the matched cohort, there were numerically more severe PPM in the ME group, although rates were not statistically significant overall (P = 0.27). PPM: prosthesis-patient mismatch. Abbreviations as in  concomitant procedures, as expected), and thus results should be interpreted accordingly. However, multivarible analyses after PSM with the remaining differences confirmed our observations. While clinical FU was complete, there was some missing echocardiographic data that could not be retrieved from referring centres. We however attempted to offer an accurate view of the clinical and echocardiographic comparison of the 2 valves by providing the reader with results from a PSM population of 'real-life' patients treated during the same period of time.