Durability of bioprosthetic aortic valve replacement in patients under the age of 60 years — 1-year follow-up from the prospective INDURE registry

Abstract OBJECTIVES We report 1-year safety and clinical outcomes in patients <60 years undergoing bioprosthetic surgical aortic valve intervention. METHODS The INSPIRIS RESILIA Durability Registry is a prospective, multicentre registry to assess clinical outcomes of patients <60 years. Patients with planned SAVR with or without concomitant replacement of the ascending aorta and/or coronary bypass surgery were included. Time-related valve safety, haemodynamic performance and quality of life (QoL) at 1 year were assessed. RESULTS A total of 421 patients were documented with a mean age of 53.5 years, 76.5% being male and 27.2% in NYHA class III/IV. Outcomes within 30 days included cardiovascular-related mortality (0.7%), time-related valve safety (VARC-2; 5.8%), thromboembolic events (1.7%), valve-related life-threatening bleeding (VARC-2; 4.3%) and permanent pacemaker implantation (3.8%). QoL was significantly increased at 6 months and sustained at 1 year. Freedom from all-cause mortality at 1 year was 98.3% (95% confidence interval 97.1; 99.6) and 81.8% were NYHA I versus 21.9% at baseline. No patient developed structural valve deterioration stage 3 (VARC-3). The mean aortic pressure gradient was 12.6 mmHg at 1 year and the effective orifice area was 1.9 cm2. CONCLUSIONS The 1-year data from the INSPIRIS RESILIA valve demonstrate good safety and excellent haemodynamic performance as well as an early QoL improvement. Clinical trial registration clinicaltrials.gov: NCT03666741.


INTRODUCTION
While mechanical valves have traditionally been preferred over bioprosthetic valves in younger patients, the use of bioprosthetic valves has expanded due to their durability, decreased risk of reoperation and the possibility of undergoing a transcatheter valve-in-valve procedure [1]. Retrospective observational studies have reported comparable long-term benefits in patients 50-69 years undergoing mechanical versus bioprosthetic valve replacement [2,3]. As a result, current American and European guidelines recommend lower age cut-offs (50-65 years of age) for the use of bioprostheses, emphasizing the importance of considering individual patient factors and informed shared decisionmaking [4,5].
The INSPIRIS RESILIA aortic valve (AV) (Edwards Lifesciences, Irvine, CA, USA) is a stented bioprosthetic, tri-leaflet valve comprised of bovine pericardial tissue. To date, 1 pre-clinical randomized controlled trial [6] and several clinical trials [7][8][9][10][11][12] involving the RESILIA tissue were performed. Flameng et al. [6] reported significantly improved haemodynamic and anticalcification properties of the RESILIA tissue compared with the standard Perimount valve in the juvenile sheep model. The findings from a single-arm registry and the COMMENCE trial have shown excellent safety and effectiveness at 5 years, with no structural valve deterioration (SVD) [7,11]. The INSPIRIS RESILIA valve has also demonstrated improved haemodynamic performance in early results of smaller cohorts [10,12].
Although data on safety and effectiveness of the RESILIA tissue are accumulating, studies focusing specifically on younger patients <60 years are lacking. The prospective INSPIRIS RESILIA Durability Registry (INDURE) aims to provide data on short-term clinical effectiveness, as well as on long-term haemodynamic and structural performance in patients <60 years. Here, we report 1year data of patients enrolled.

METHODS
INDURE is a prospective, open-label, multicentre, international registry to assess the clinical outcomes of patients younger than 60 years of age who undergo surgical aortic valve replacement (SAVR) with the INSPIRIS RESILIA AV [13]. Patients were enrolled at 21 sites across Austria, Belgium, France, Germany, Italy, Netherlands, Spain, the UK and Canada.

Ethics statement
The ethics committee responsible for each site granted approval, and written informed consent was obtained.

Patients
Adult patients 60 years of age or younger, undergoing SAVR and receiving the INSPIRIS RESILIA AV prosthesis were enrolled. In addition to the stipulations of the device Instructions for Use, inclusion criteria included a planned replacement of the native valve as indicated based on a preoperative evaluation. The aortic valve replacement (AVR) was either isolated or with concomitant replacement of the ascending aorta and/or coronary artery bypass graft. Patients undergoing pulmonary vein isolation were also allowed if it was not a full cox-maze procedure. Patients with (i) no possibility of valve implantation in accordance with the Instructions for Use; (ii) presence of active or within the last 3 months of the scheduled SAVR endocarditis/myocarditis; (iii) previous AVR; (iv) a Bentall (root) procedure or any surgery on other valves; or (v) life expectancy of <12 months were excluded.

Objectives
The primary objective was to determine the time-related valve safety at 1 year depicted as freedom from events in patients undergoing SAVR and receiving the INSPIRIS RESILIA AV prosthesis. Time-related valve safety was defined as composite end point according to the valve academic research consortium (VARC)-2 criteria [requiring of repeat procedure; prosthetic valve endocarditis, prosthetic valve thrombosis, thromboembolic events (e.g. stroke) and life-threatening bleeding] [14]; however, due to more precise definitions compared to VARC-2, SVD stage 3 was presented according to VARC-3 criteria comparing 1 year versus discharge echo [increase in mean AV pressure gradient (PG) > _20 mmHg resulting in mean AV PG > _30 mmHg with a concomitant decrease in effective orifice area (EOA) > _0.6 cm 2 or > _50% and/or a decrease in doppler velocity index (DVI) > _0.2 or > _40%, OR new occurrence, or increase of > _2 grades, of intraprosthetic AR resulting in severe AR] [15].
The secondary objective was the assessment of haemodynamic performance of the INSPIRIS RESILIA AV and further durability parameters, clinical outcomes and quality of life (QoL). Further clinical outcomes of interest were all-cause, cardiovascular and valverelated mortality [15], valve-related dysfunction, requirement of repeat procedure due to any cause, permanent pacemaker implantation, Acute Kidney Injury Network stage 2/3 and New York Heart Association (NYHA) functional class compared to baseline. QoL was assessed using the Kansas City Cardiomyopathy Questionnaire (KCCQ) and Short Form-12 Health Survey Version 2 (SF-12v2).
Outcomes according to the VARC-2 criteria were adjudicated by an independent clinical event committee. Digital imaging and communication in medicine files of echocardiograms generated at 1-year follow-up were collected for analysis by the Echo Core Laboratory to ensure unbiased and consistent analysis of the diagnostic data.

Statistical analysis
Data were analysed using descriptive statistics, with categorical variables presented as absolute values and frequencies (%) and the continuous variables presented as means [standard deviation (SD)] and/or median [interquartile range (IQR)]. Test for normal distribution was carried using the Kolmogorov-Smirnov test. Wilcoxon signed ranks test for paired data was used for comparing QoL scores between baseline and follow-up visits. For outcome reporting Kaplan-Meier estimates were provided. A P-value of <0.05 was considered statistically significant. Statistical analysis was performed using SPSS Version 28.0 (Armonk, NY, IBM Corp.) [16].

RESULTS
A total of 457 patients were enrolled between April 2019 and May 2021. For the present analysis, 36 patients with a Bentall procedure and mitral/pulmonary valve replacement were excluded, resulting in a total of 421 patients. Within the first-year post-SAVR, 17 patients were lost to follow-up (4.0%). Of the remaining 404 patients, 7 (1.7%) patients died, which resulted in a total of 397 (94.3%) patients alive with available data at 1 year (Fig. 1).

Patient characteristics
Patients had a mean age of 53.5 (SD: 6

Procedural details
The prevalence of bicuspid valves was 73.2% (Table 2). A total of 346 (82.4%) patients in the overall population had AS of any severity and 277 (66.0%) had AR. AS was dominating in 304 (72.4%) patients while AR was dominating in 98 (23.3%) patients. Pure forms of AS and AR were present in 142 (33.8%) and 73 (17.4%) patients. The aetiology of valve pathology in the overall population was as follows: 73.6% were congenital, 22.8% were degenerative, 1.0% were rheumatic, 0.7% were endocarditic and 1.9% (n = 8) were other/unknown (n = 5 unknown, n = 2 prolaps/pocket rupture and n = 1 dilation of aortic root).

Procedural and in-hospital outcomes
Lost to FU (n=2)  Compared to discharge, mean aortic PG was only slightly higher at 1 year while EOA was lower in patients (Fig. 2B). In addition, 81.8% of patients were in NYHA class I at 1 year compared to 21.9% at baseline and only 3.6% were in NYHA class III/IV compared to 27.2% at baseline (Fig. 3). There were no cases of mild/severe PVL at 1 year.

Quality of life outcomes
SF-12v2 and KCCQ were used for assessing QoL in patients (Table 3 and Supplementary Material, Table S1). The mean SF-12v2 physical summary score at 3-6 months was 47.7 points (P < 0.001) compared to baseline (41.5 points) with a further increase at 1 year (49.2 points; P < 0.001). The mean SF-12v2 mental summary score at 3-6 months was 50.0 points (P < 0.001) from baseline (45.6 points) and remained relatively stable at 1 year (49.9 points; P < 0.001). Both physical and mental summary scores at 1 year were near the general population mean (50.0 points). Overall, changes in PCS and MCS at 1 year compared to baseline were classified as follows: 39.6% and 29.8% of patients had a large improvement in QoL. 1.9% for both died (Fig. 4).
The KCCQ total symptom score and clinical summary score significantly increased both at 3-6 months (87.6 points; P < 0.001 and 88.4 points; P < 0.001) and at 1 year (90.0 points; P < 0.001 and 89.2 points; P < 0.001) compared to baseline (74.6 and 75.1 points) (Table 3). Similarly, there was an increase in mean overall summary score already at 3-6 months (85.2 points; P < 0.001) and further at 1 year (87.1 points; P < 0.001) in comparison to baseline (66.1 points). Overall, 1-year changes in patient overall summary score from baseline were classified as follows: 43.7% of patients had a large improvement in QoL and 1.9% died (Fig. 4).
Total valve-related bleeding (categorized in minor, major and life threatening) occurred in 45 (10.7%) patients as early outcome and there were no further cases at 1 year. Sixteen (3.8%) patients required a permanent pacemaker implantation as early outcome and further 3 (0.7%/vy) patients required it as late outcome. Six (1.4%) patients developed Acute Kidney Injury Network stage 2/3 at 30 days with no further incidence at 1 year.

DISCUSSION
The 1-year results of the INDURE demonstrate (i) high hospital and 1-year survival rates with an absence of valve-related mortality; (ii) satisfactory and stable performance of the INSPIRIS RESILIA with complete freedom from stage 3 SVD based on a standardized CoreLab adjudicated assessment; and (iii) an improvement in the patients' QoL early after the intervention which was sustained at 1 year.

Hospital and 1-year survival rates
In-hospital all-cause mortality rate in our study (0.7%) was lower than the rates reported in the trials by Useini (2.5%) and Fukunaga (3.4%), which both evaluated hospital outcomes after AVR using the INSPIRIS RESILIA bioprosthesis in smaller cohorts  [ 10,12]. The reported in-hospital mortality rates for the Carpentier-Edwards Perimount Magna Ease bioprosthesis with RESILIA tissue range between 1.2% and 2.3% [7,8].
We report excellent survival rates at 1 year: overall survival was 98.3% and valve-related survival was 100%. Although survival may vary depending on patient characteristics, the survival rates in our patient cohort are comparable or potentially even slightly better to those reported in previous trials using bioprostheses with RESILIA tissue [7,8]. Puskas et al. [7] reported a 1-year overall survival of 97.6% and valve-related survival of 98.8% with the Carpentier-Edwards Perimount Magna Ease bioprosthesis (Model 11000A) in the COMMENCE trial. Bartus et al. [9] reported an overall mortality rate of 6.8% for the same bioprosthesis. Furthermore, Didier et al. [17] reported higher mortality rates after transcatheter aortic valve replacement (TAVR) with balloonexpandable transcatheter heart valves at 1 year (23.2%).

Performance of the INSPIRIS RESILIA
One-year haemodynamic performance of INSPIRIS RESILIA, evaluated by an independent CoreLab, was favourable. Mean aortic PG [12.5 (SD: 5.3) mmHg] and EOA [1.9 (SD: 0.6) cm 2 ] at 1 year were within the ranges reported in other studies. In the COMMENCE trial, the mean aortic PG and EOA at 1 year were 10.4 (SD: 4.9) mmHg and 1.7 (SD: 0.5) cm 2 [7]. Bartu s et al. reported mean aortic PG and EOA at 1 year to be 13.9 (SD: 6.1) mmHg and 1.8 (SD: 0.6) cm 2 [8]. The mean aortic PG and EOA reported in a Japanese cohort undergoing AVR with INSPIRIS RESILIA were 11.2 (SD: 3.2) mmHg and 1.8 (SD: 0.4) cm 2 . In the recently published early results after INSPIRIS RESILIA AVR (including only discharge data), the mean aortic PG was 10.2 (SD: 4.1) mmHg, which is slightly lower than the mean aortic PG at discharge in our cohort [    noted, however, that patients receiving 19-mm valves (n = 5; all female) in our study exhibited elevated mean aortic PG at both discharge (21.3 mmHg) and 1 year (21.8 mmHg). Increased aortic PG may lead to risks associated with patient-prosthesis mismatch as well as accelerated degeneration of the implanted valve. Therefore, it is important to provide reduced gradients to patients requiring smaller valves, particularly those requiring a 19-mm valve.
It is known that the use of bioprosthetic valves is associated with higher rates of SVD, particularly in younger patients. Although high freedom from SVD at 1 year in the current study further highlights the durability of RESILIA tissue reported in previous studies [7][8][9], it is important to note that the rates of SVD in the first years are generally very low and the incidence rises in later years. However, SVD is caused by degenerative calcification over time in the majority of cases, which can be permanently reduced by the novel integrity preservation technology applied during the preparation of RESILIA tissue [6]. This preservation technology is described as a capping process, which permanently blocks residual aldehyde content known to bind with calcium. Further glycerolization preserves the tissue in dry storage, which provides a persistent protection of collagen. The rates of prosthetic valve endocarditis and prosthetic valve thrombosis at 1 year were 0.2% and 1.0%, respectively. A recent meta-analysis concluded that bioprosthetic valves may be associated with a higher risk of endocarditis than mechanical valves [18]. However, freedom from prosthetic valve endocarditis in our patient cohort was still very high (99.8%) and comparable to the rates previously reported for RESILIA tissue [7][8][9]. The prevalence of prosthetic valve thrombosis in our study (1.0%) is comparable with that reported in the literature (0.6-0.7%), although the authors state that their prevalence is currently underestimated since routine prospective follow-up imaging is frequently not performed in the absence of symptoms or haemodynamic changes noted by echocardiography [19]. It has also been reported that the risk of thrombosis is higher with stented bioprosthetis, such as INSPIRIS RESILIA, compared to stentless valves [20]. In addition, we did not differentiate between clinical valve thrombosis and subclinical leaflet thrombosis characterized by hypoattenuated leaflet thickening, which is defined as an incidental finding of an increase in the thickness of the prosthetic valve leaflets without associated symptoms. Hypoattenuated leaflet thickening may be an early indicator of valve thrombosis, although its relationship to clinical events is still not clear [21,22]. Therefore, we feel that the prevalence of prosthetic valve thrombosis after valve implantation in our study is acceptable.

Quality of life
We assessed QoL in patients in this study, which has not been reported in previous trials on valves with the RESILIA tissue [7][8][9]. Myken et al. assessed the differences between patients receiving mechanical and bioprosthetic valves for heart valve surgery and found no differences [23]. Repack et al. [24] also compared postoperative QoL in patients undergoing aortic root replacement with mechanical versus bioprosthetic valves and reported similar outcomes in QoL between 2 patient groups. In our study, there was a significant improvement in QoL already at 6 months postsurgery, with further improvement at 1 year, suggesting that SAVR with INSPIRIS RESILIA improves QoL in young patients.

Limitations
The INDURE provides real-world data of a large patient cohort with the applicability of findings to clinical practice across Europe and Canada. However, as we did not include an active control group, different bioprosthetic valves or valve generations could not be compared and selection bias cannot be excluded. Furthermore, there is no comparison of the bioprosthetic valve data with the outcomes and performance of mechanical valves. Lastly, although the results presented here are limited to 1-year data and may not reflect the ultimate safety outcomes and performance of the valve prosthesis, the present cohort will be followed up for 5 years, indicating the reporting of long-term outcomes in the future. a 406 valve years. b Within the first year, it was unknown in a total of 4 patients whether the death was valve related. c SVD stage 3 according to VARC-3 comparing 1 year versus discharge echo (increase in mean AV PG > _20 mmHg resulting in mean AV PG > _30 mmHg with a concomitant decrease in EOA > _0.6 cm 2 or > _50% and/or a decrease in doppler velocity index (DVI) > _0.2 or > _40%, OR new occurrence, or increase of > _2 grades, of intraprosthetic aortic regurgitation (AR) resulting in severe AR). d Requiring repeat procedure of the prosthetic valve due to prosthetic endocarditis. e Valve-related bleeding reported as minor, major and life threatening according to VARC-2.