Abstract
Small aortic annuli are challenging in aortic valve replacement. Prosthesis-patient mismatch, accompanied by high transvalvular gradients and small orifice area, is an often faced problem impairing postoperative outcome. The new TRIBIO bioprosthesis aims to enable enhanced haemodynamics, being increasingly important with decreasing annular size. This study compares ex vivo hydrodynamics of TRIBIO with 2 established bioprostheses in small annuli at different stroke volumes, simulating ‘rest’ and ‘exercise’, and evaluates haemodynamics of TRIBIO in a sheep model.
Ex vivo: Porcine aortic roots were hardened with glutaraldehyde, approaching the stiffness of decalcified roots. Each bioprosthesis (TRIBIO, Trifecta™, Perimount® Magna Ease), size 19 mm, was implanted in each aortic root and inserted into a pulse duplicator, determining hydrodynamics and geometric orifice area at different stroke volumes (74, 90, 104 ml). Sheep model: Postoperative transvalvular gradients were assessed after implantation of 19 mm TRIBIO in aortic position.
Ex vivo: Mean transvalvular gradients were lowest in TRIBIO (7.3, 8.7, 10.2 mmHg at 74, 90, 104 ml) with significant difference to Perimount® Magna Ease. Geometric orifice area in TRIBIO was 2.7 cm2 and significantly larger compared to Perimount® Magna Ease and Trifecta™. Opening of TRIBIO was uniform and apparently complete, whereas leaflet’s outward movement was restricted particularly in Perimount® Magna Ease. Sheep model: TRIBIO presented with low transvalvular gradients (δpmean 4.1 ± 2 mmHg, δpmax 7.8 ± 4.6 mmHg), unimpaired leaflet motion and no central or paravalvular leakage.
Ex vivo, the TRIBIO achieved superior hydrodynamics compared to latest generation bioprostheses. These excellent data are supported by very low transvalvular gradients in a preliminary sheep model.
INTRODUCTION
Aortic valve stenosis remains one of the main pathologies requiring valve replacement, especially in elderly patients. Surgical aortic valve replacements (SAVR) with surgical aortic valve bioprostheses (SAVB) is still the standard of care, as it is recommended for patients older than 60–70 years, besides transcatheter aortic valve implantation (TAVI) in high-risk patients [1]. Nevertheless, there is an increasing number of younger patients desiring a bioprosthesis instead of a mechanical valve to avoid lifetime anticoagulation and valve noise, accepting an accelerated valve degeneration with a consequent need for re-intervention.
A special problem among the patients needing SAVR is characterized by a small size of the aortic annulus. In these patients, prosthesis patient mismatch (PPM) as defined by Rahimtoola [2] is one of the major problems, as severe PPM is known to increase the risk for perioperative mortality, heart failure and lower life expectancy [3–5]. This is supported by 2 publications reporting a significant reduction of late congestive heart failure associated with a decrease of only 5 mmHg of the mean transvalvular gradient and a relation of structural SAVB deterioration to the transvalvular gradient at implantation [6, 7].
Considering these aspects, it is important that the design of an SAVB allows for a maximum orifice area and a minimum transvalvular gradient in order to improve patients’ clinical outcome, particularly in small aortic annuli and young patients.
Following this urgent need for a novel SAVB with improved haemodynamics, the novel, supra-annular TRIBIO aortic valve bioprosthesis was designed, aiming to achieve superior valve characteristics compared with established supra-annular SAVB. Furthermore, the TRIBIO was constructed with integrated facilities for later TAVI valve-in-valve procedures.
This report presents the first ex vivo and in vivo results of the TRIBIO SAVB. The ex vivo model compares hydrodynamic performance as well as geometric orifice area (GOA) of the TRIBIO and 2 established supra-annular SAVB at 3 different cardiac output volumes, simulating resting conditions as well as mild to moderate exercise. Furthermore, we evaluated in vivo preclinical haemodynamic performance of the TRIBIO in a sheep model.
MATERIALS AND METHODS
Ex vivo model
Aortic root models
Because of its similar anatomy to human aortic roots, porcine aortic roots (n = 6) were chosen for this ex vivo model. In order to approach the stiffness of human aortic roots after surgical annular decalcification, the porcine aortic roots were hardened with glutaraldehyde (0.25%) for 24 h with a continuous pressure of 80 mmHg. Sizing was performed using Hegar dilators and the specific sizing devices for each SAVB type to simulate clinical conditions. Only roots with a diameter appropriate for the implantation of 19 mm bioprostheses as determined by the particular sizing devices were used.
Valves
The conventional SAVB were represented by 2 established valve types: the Trifecta™ bioprosthesis (St. Jude Medical Inc., St. Paul, MN, USA) and the Perimount® Magna Ease bioprosthesis (Edwards Lifesciences LLC, Irvine, CA, USA), both of 19 mm size.
These SAVB were compared with the novel TRIBIO bioprosthesis (TRIBIO, Heilbronn, Germany) (Fig. 1). This device consists of 3 different components: the flexible valve bearing crown, the force decoupled interface and the intra-annular base ring. The crown is built of a common-style nitinol stent frame with a conical configuration, allowing for expansion into the aortic sinuses after implantation and gaining an undisturbed cross-sectional area equal to the base ring area and thus the left ventricular outflow tract (LVOT). The stent frame is covered with Dacron fabric serving for the fixation of the valve leaflets, both mounted to a base structure incorporating the sewing ring. To achieve maximum opening capability, native aortic valve geometry was analysed in detail afore leading to a novel multiple curved leaflet shape, which allows for complete opening. Valve leaflets are made of commercially available glutaraldehyde-fixed bovine pericardium (Maverik Biomaterials Pty Ltd, Dubbo East, Australia).
Different aortic bioprostheses. Left: TRIBIO bioprosthesis; middle: Perimount® Magna Ease bioprosthesis; right: Trifecta™ bioprosthesis.
Pulse duplicator
The physiological conditions of circulation were imitated by an established pulse duplicator [8]. It enables the adjustment of different cardiac output volumes, afterload and stroke rates.
A high-speed camera MotionPro Y3 (Imaging Solutions GmbH, Eningen, Germany) mounted on top of the pulse duplicator allows for a visual observation during the experimental runs.
Experimental procedure
Per porcine aortic root, 1 measurement of each SAVB type was performed.
According to the particular test series, the TRIBIO, Trifecta™ or Perimount® Magna Ease was sewed into a porcine aortic root. The implantation of the TRIBIO was performed in the usual manner using interrupted sutures with felt pledgets as well as without felt pledgets. In this case, the base ring itself served as abutment and replaced the felt pledgets. The aortic conduits were inserted into the pulse duplicator, which allowed for the evaluation of mean and maximum transvalvular gradients (δpmean, δpmax). The evaluation of GOA was developed through photographs by the high-speed camera. Each conduit was tested at 3 different stroke volumes (74, 90, 104 ml/stroke) in order to investigate the characteristics of each SAVB under ‘rest’ and ‘exercise’ conditions. During all measurements, stroke rate was 64 beats per minute, diastolic pressure 80 mmHg and systolic 120 mmHg.
The test solution was represented by a physiological saline (0.9%) with a density of 1.0046 g/cm3 and a dynamic viscosity of 0.9 mPa·s at an ambient temperature of 20°C.
Technique of measurement
The left ventricular pressure (4 cm below the aortic valve) and aortic pressure (6 cm above the aortic valve) were measured with 2 capacitive pressure transducers (Envec Ceracore M, Endress + Hauser, Maulburg, Germany), calibrated to a measuring range of −20 to +160 mmHg and a resolution of 0.02 mmHg.
The sensor of an ultrasonic flowmeter (TS-410, Transonic System Inc., Ithaca, NY, USA) was mounted directly below the aortic valve to record the volume flow through the valve. The sensor works bidirectionally with a resolution of 2 ml/min and records flow rates up to 20 l/min.
A high-speed camera MotionPro Y3 above the conduit recorded the characteristics of motion of the aortic valves with 500 pictures per second. Video recordings and flow measurements were started simultaneously by using trigger signals.
Sheep model
Animals
The study was conducted in compliance with ISO 10993–2:2006 and with the internal SOP.30 ‘Ethics management’ of the testing facility IMMR (IMMR, Paris, France). The IMMR’s Animal Care and Use Committee is registered at the CNREEA (Création d'un comité national de réflexion éthique sur l'expérimentation animale, France) under the Ethics Committee number 37. The animal facilities were inspected by the ‘Direction Départementale de la Protection des Populations’ of the French government and were accredited. Humane care of all animals was provided according to the National Institutes of Health’s Guide for Care and use of Laboratory animals.
Six female sheep with a weight of 74–85 kg (mean 79 kg) were implanted with the TRIBIO bioprosthesis, size 19 mm.
Valves
The TRIBIO bioprostheses used for the in vivo model were constructed as described above but with an additional thin layer of titanium (thickness 20 nm) on the pericardial leaflets as an anticalcification treatment (pfm medical, Cologne, Germany).
Surgical procedure
After premedication, all animals were operated on in general anaesthesia. Monitoring included electrocardiogram, blood pressure, end-tidal CO2, core body temperature and urinary output.
Surgical access was gained via left thoracotomy. After establishing cardiopulmonary bypass, cross-clamping of the aorta and performing an aortotomy, the native aortic valve was excised. The TRIBIO SAVB was implanted in the conventional manner using interrupted sutures and closing of the aorta was performed with a continuous suture. SAVR was performed during cardioplegic cardiac arrest using warm blood cardioplegia CP1B (Pharmacie Centrale des Hôpitaux de Paris, France). After removing cross-clamp and establishing stable haemodynamics, cardiopulmonary bypass was weaned and the cannulas removed. Chest tubes were positioned and the thoracotomy site closed.
One sheep completed 90-day follow-up and 5 of the 6 sheep died during wake up phase due to non-device related events. Necropsy revealed as cause of death in 4 sheep haemorrhage because of suture dehiscence at the aortic approach and in 1 sheep a massive left ventricular myocardial hypertrophy. The tissue of the ascending aortic wall in these sheep was very fragile and considerably retracted and the TRIBIO—although it was only size 19 mm—was difficult to implant because of a size mismatch between the considerably retracted ascending aorta and the TRIBIO.
Measurement of transvalvular gradients
After weaning of cardiopulmonary bypass and establishing stable haemodynamics, transvalvular gradients (δpmean, δpmax) were determined using transoesophageal echocardiography.
Analysis and statistics
All data were presented as mean ± standard deviation. Differences between the groups were calculated using univariate analysis of variance followed by pairwise Tukey’s honest significant difference post hoc comparison in parametric variables and Kruskal–Wallis test followed by Mann–Whitney U test post hoc in non-parametric variables. Analyses of longitudinal data were performed by univariate analysis of variance with repeated measures and Friedman test for parametric and non-parametric data, respectively. A P-value <0.0369 was considered as significant after adjusting for multiple comparisons by using the false discovery rate method by Benjamini and Hochberg [9]. All calculations were performed using IBM® SPSS® Statistics version 22.
RESULTS
Ex vivo model
Hydrodynamics
Considering δpmean and δpmax, we found a statistically significant difference between the TRIBIO and Perimount® Magna Ease SAVB with lower gradients in the TRIBIO group. Comparing the values for δpmax at different stroke volumes, the TRIBIO was the only SAVB that showed no significant increase with rising stroke volume (Table 1).
Results for hydrodynamics and geometric orifice area
| Stroke volume (ml) | Trifecta | Magna Ease | TRIBIO | P-value | P-valuea | P-valueb | P-valuec | |
|---|---|---|---|---|---|---|---|---|
| δpmean (mmHg) | 74 | 8.2 ± 1 | 11.9 ± 1.5 | 7.3 ± 1.1 | <0.001 | <0.001 | <0.001 | 0.436 |
| 90 | 10 ± 1.2 | 14.2 ± 1.6 | 8.7 ± 1.4 | <0.001 | <0.001 | <0.001 | 0.248 | |
| 104 | 11.4 ± 1.3 | 14.5 ± 1 | 10.2 ± 1.5 | <0.001 | 0.002 | <0.001 | 0.323 | |
| P longitudinal | <0.001 | 0.007 | <0.001 | |||||
| δpmax (mmHg) | 74 | 19.2 ± 3 | 22.3 ± 3.5 | 18.2 ± 3.8 | 0.138 | 0.309 | 0.134 | 0.858 |
| 90 | 21.1 ± 2.9 | 26.4 ± 3.2 | 18.6 ± 4.3 | 0.005 | 0.048 | 0.004 | 0.470 | |
| 104 | 23.7 ± 1.6 | 29.3 ± 2.9 | 19.6 ± 3.7 | <0.001 | 0.011 | <0.001 | 0.063 | |
| P longitudinal | 0.003 | <0.001 | 0.162 | |||||
| GOA (cm2) | 74 | 2.1 ± 0.1 | 1.5 ± 0.1 | 2.7 ± 0.1 | 0.001 | 0.002 | 0.002 | 0.002 |
| 90 | 2.2 ± 0.1 | 1.6 ± 0 | 2.7 ± 0.1 | 0.001 | 0.004 | 0.002 | 0.004 | |
| 104 | 2.1 ± 0.1 | 1.6 ± 0 | 2.7 ± 0.1 | 0.001 | 0.004 | 0.002 | 0.004 | |
| P longitudinal | 0.819 | 0.115 | 0.513 |
| Stroke volume (ml) | Trifecta | Magna Ease | TRIBIO | P-value | P-valuea | P-valueb | P-valuec | |
|---|---|---|---|---|---|---|---|---|
| δpmean (mmHg) | 74 | 8.2 ± 1 | 11.9 ± 1.5 | 7.3 ± 1.1 | <0.001 | <0.001 | <0.001 | 0.436 |
| 90 | 10 ± 1.2 | 14.2 ± 1.6 | 8.7 ± 1.4 | <0.001 | <0.001 | <0.001 | 0.248 | |
| 104 | 11.4 ± 1.3 | 14.5 ± 1 | 10.2 ± 1.5 | <0.001 | 0.002 | <0.001 | 0.323 | |
| P longitudinal | <0.001 | 0.007 | <0.001 | |||||
| δpmax (mmHg) | 74 | 19.2 ± 3 | 22.3 ± 3.5 | 18.2 ± 3.8 | 0.138 | 0.309 | 0.134 | 0.858 |
| 90 | 21.1 ± 2.9 | 26.4 ± 3.2 | 18.6 ± 4.3 | 0.005 | 0.048 | 0.004 | 0.470 | |
| 104 | 23.7 ± 1.6 | 29.3 ± 2.9 | 19.6 ± 3.7 | <0.001 | 0.011 | <0.001 | 0.063 | |
| P longitudinal | 0.003 | <0.001 | 0.162 | |||||
| GOA (cm2) | 74 | 2.1 ± 0.1 | 1.5 ± 0.1 | 2.7 ± 0.1 | 0.001 | 0.002 | 0.002 | 0.002 |
| 90 | 2.2 ± 0.1 | 1.6 ± 0 | 2.7 ± 0.1 | 0.001 | 0.004 | 0.002 | 0.004 | |
| 104 | 2.1 ± 0.1 | 1.6 ± 0 | 2.7 ± 0.1 | 0.001 | 0.004 | 0.002 | 0.004 | |
| P longitudinal | 0.819 | 0.115 | 0.513 |
Trifecta versus Magna.
Magna versus TRIBIO.
Trifecta versus TRIBIO.
δpmean: mean transvalvular gradient; δpmax: maximum transvalvular gradient; GOA: geometric orifice area.
Results for hydrodynamics and geometric orifice area
| Stroke volume (ml) | Trifecta | Magna Ease | TRIBIO | P-value | P-valuea | P-valueb | P-valuec | |
|---|---|---|---|---|---|---|---|---|
| δpmean (mmHg) | 74 | 8.2 ± 1 | 11.9 ± 1.5 | 7.3 ± 1.1 | <0.001 | <0.001 | <0.001 | 0.436 |
| 90 | 10 ± 1.2 | 14.2 ± 1.6 | 8.7 ± 1.4 | <0.001 | <0.001 | <0.001 | 0.248 | |
| 104 | 11.4 ± 1.3 | 14.5 ± 1 | 10.2 ± 1.5 | <0.001 | 0.002 | <0.001 | 0.323 | |
| P longitudinal | <0.001 | 0.007 | <0.001 | |||||
| δpmax (mmHg) | 74 | 19.2 ± 3 | 22.3 ± 3.5 | 18.2 ± 3.8 | 0.138 | 0.309 | 0.134 | 0.858 |
| 90 | 21.1 ± 2.9 | 26.4 ± 3.2 | 18.6 ± 4.3 | 0.005 | 0.048 | 0.004 | 0.470 | |
| 104 | 23.7 ± 1.6 | 29.3 ± 2.9 | 19.6 ± 3.7 | <0.001 | 0.011 | <0.001 | 0.063 | |
| P longitudinal | 0.003 | <0.001 | 0.162 | |||||
| GOA (cm2) | 74 | 2.1 ± 0.1 | 1.5 ± 0.1 | 2.7 ± 0.1 | 0.001 | 0.002 | 0.002 | 0.002 |
| 90 | 2.2 ± 0.1 | 1.6 ± 0 | 2.7 ± 0.1 | 0.001 | 0.004 | 0.002 | 0.004 | |
| 104 | 2.1 ± 0.1 | 1.6 ± 0 | 2.7 ± 0.1 | 0.001 | 0.004 | 0.002 | 0.004 | |
| P longitudinal | 0.819 | 0.115 | 0.513 |
| Stroke volume (ml) | Trifecta | Magna Ease | TRIBIO | P-value | P-valuea | P-valueb | P-valuec | |
|---|---|---|---|---|---|---|---|---|
| δpmean (mmHg) | 74 | 8.2 ± 1 | 11.9 ± 1.5 | 7.3 ± 1.1 | <0.001 | <0.001 | <0.001 | 0.436 |
| 90 | 10 ± 1.2 | 14.2 ± 1.6 | 8.7 ± 1.4 | <0.001 | <0.001 | <0.001 | 0.248 | |
| 104 | 11.4 ± 1.3 | 14.5 ± 1 | 10.2 ± 1.5 | <0.001 | 0.002 | <0.001 | 0.323 | |
| P longitudinal | <0.001 | 0.007 | <0.001 | |||||
| δpmax (mmHg) | 74 | 19.2 ± 3 | 22.3 ± 3.5 | 18.2 ± 3.8 | 0.138 | 0.309 | 0.134 | 0.858 |
| 90 | 21.1 ± 2.9 | 26.4 ± 3.2 | 18.6 ± 4.3 | 0.005 | 0.048 | 0.004 | 0.470 | |
| 104 | 23.7 ± 1.6 | 29.3 ± 2.9 | 19.6 ± 3.7 | <0.001 | 0.011 | <0.001 | 0.063 | |
| P longitudinal | 0.003 | <0.001 | 0.162 | |||||
| GOA (cm2) | 74 | 2.1 ± 0.1 | 1.5 ± 0.1 | 2.7 ± 0.1 | 0.001 | 0.002 | 0.002 | 0.002 |
| 90 | 2.2 ± 0.1 | 1.6 ± 0 | 2.7 ± 0.1 | 0.001 | 0.004 | 0.002 | 0.004 | |
| 104 | 2.1 ± 0.1 | 1.6 ± 0 | 2.7 ± 0.1 | 0.001 | 0.004 | 0.002 | 0.004 | |
| P longitudinal | 0.819 | 0.115 | 0.513 |
Trifecta versus Magna.
Magna versus TRIBIO.
Trifecta versus TRIBIO.
δpmean: mean transvalvular gradient; δpmax: maximum transvalvular gradient; GOA: geometric orifice area.
Geometric orifice area
At all different stroke volumes, the TRIBIO presented with a significantly larger GOA when compared with Trifecta™ and Perimount® Magna Ease (Table 1).
Because of the visual observation with the high-speed camera, a complete and uniform opening behaviour of the TRIBIO became apparent, whereas in particular the Perimount® Magna Ease showed a non-uniform and restricted opening (Fig. 2).
Maximum valve opening. Photographs by the high-speed camera during systole depicting the different opening behaviour of the surgical aortic valve bioprostheses. Left: TRIBIO; middle: Trifecta™; right: Perimount® Magna Ease.
Sheep model
Postoperative assessment revealed a δpmean of 4.1 ± 2 mmHg and a δpmax of 7.8 ± 4.6 mmHg. Leaflet motion of the TRIBIO was unimpaired in every sheep with absence of relevant central or paravalvular leakage.
DISCUSSION
This combined ex vivo and in vivo study introduces the novel TRIBIO surgical aortic valve bioprosthesis including a conical stent configuration with a leaflet design resembling native aortic leaflet shape, a flexible interface and a base ring, resulting in an improved orifice area and minor transvalvular gradients.
Up to now, a small aortic annulus of <21 mm is still a challenge in SAVR [10] potentially causing prosthesis-patient mismatch with a negative effect at rest and further worsening at exercise. Severe PPM is known to impair left ventricular function and life expectancy [3–5, 11]. These results underline the need for a further development of improved, more physiological bioprostheses, tested under exercise conditions.
In addition to the ex vivo model and in order to provide in vivo preclinical haemodynamic parameters, the TRIBIO was evaluated in a sheep model, being an established concept for the preclinical evaluation of cardiovascular devices and closely mimicking human anatomy and physiology.
Geometric orifice area
When analysing the GOA under dynamic conditions, the TRIBIO presented with a significantly larger GOA at all stroke volumes when compared with Trifecta™ and Perimount® Magna Ease (all P < 0.005) (Fig. 3).
Geometric orifice area.
As described by Sievers and Scharfschwerdt [12], an ideal GOA of a substitute, e.g. an SAVB, should be at least of the same size as the implantation area. Referring this thesis to the present study with a given aortic annulus of 19 mm diameter, the GOA should be at least 2.8 cm2. The 19 mm TRIBIO presented with a GOA of 2.7 cm2 and thus nearly achieved the minimum ideal GOA, whereas the 19 mm Trifecta™ presented with 2.1 cm2 and the Perimount® Magna Ease with only 1.6 cm2, meaning a smaller GOA compared to the minimum ideal GOA of 33% for Trifecta™ and 75% for Perimount® Magna Ease, respectively. The superior opening behaviour of the TRIBIO due to its special construction with a conical shape and novel, multiple curved leaflets, allowing a complete opening of the leaflets, is additionally shown by visual observations through the high-speed camera (Fig. 2).
However, when analysing each SAVB separately under the aspect of a possible physiological increase of GOA at higher stroke volumes, neither the TRIBIO nor the Trifecta™ and Perimount® Magna Ease showed a physiological behaviour. GOA remained at the same level at all stroke volumes (all P-values non-significant). Most likely, this is caused by the stent frame of the SAVB, which is more rigid in all different SAVB types when compared to a native aortic valve complex.
Considering the entirety of the results of this ex vivo model, the aim of increasing the GOA by an innovative design is achieved, nearly reaching a minimum ideal GOA.
Hydrodynamics
The TRIBIO showed minor transvalvular gradients at all different stroke volumes compared with both conventional SAVB with significant difference to the Perimount® Magna Ease (Figs 4 and 5).
Mean transvalvular gradients.
Maximal transvalvular gradients.
Considering the results for δpmax at the different stroke volumes for each SAVB type separately, it has to be highlighted that only the TRIBIO showed no significant increase with rising stroke volumes, whereas δpmax in Trifecta™ as well as Perimount® Magna Ease increased significantly. The underlying mechanism of this stability of δpmax is probably the increased GOA to an approximately ideal degree, as well as the flexibility and interaction between the 3 components of the TRIBIO.
Nevertheless, not only the GOA and opening behaviour of the SAVB determine transvalvular gradients but also LVOT obstruction is known to increase δp. The design of conventional stented SAVB owns a sewing ring that is inevitably positioned within the blood flow after surgical implantation and therefore causes a relative blood flow obstruction in the LVOT [13]. Additionally, conventional stented SAVB are implanted by the use of interrupted sutures with single felt pledgets. These pledgets may protrude partially into the LVOT, potentially causing obstruction. The novel design of the TRIBIO with its base ring prevents the interference of a protruding sewing ring and felt pledgets. Furthermore, the base ring itself can be used as a replacement of the pledgets. Additional space is gained to fully use and prevent obstruction of the LVOT.
According to Johnston et al. [6], as explant of SAVB due to structural valve degeneration is related with increased δp at implantation, the minor δp in the novel TRIBIO SAVB might increase long-term durability, which is especially important in younger patients.
Haemodynamics
The evaluation of haemodynamic performance of the TRIBIO bioprosthesis in the sheep model resulted in low transvalvular gradients, emphasizing the promising hydrodynamic results of the ex vivo model.
A research group led by Modi et al. [14] investigated early postoperative haemodynamics of 19 mm Trifecta™ and Perimount® Magna Ease. Transvalvular gradients in the Trifecta™ group were 12.7 ± 4.4 mmHg (δpmean) and 24.7 ± 10 mmHg (δpmax) and in the Perimount® Magna Ease group 17.4 ± 6.5 mmHg (δpmean) and 33.5 ± 16 mmHg (δpmax). Considering the transvalvular gradients of the 19 mm TRIBIO, the in vivo tests in sheep resulted in a δpmean of 4.1 ± 2 mmHg and a δpmax of 7.8 ± 4.6 mmHg. Although haemodynamics in sheep are not strictly comparable to humans, the in vivo results of this study, underlined by the excellent ex vivo hydrodynamics, may implicate a haemodynamic superiority of the TRIBIO compared with the established SAVB Trifecta™ and Perimount® Magna Ease. However, it has to be highlighted that only early haemodynamics in still anaesthetized sheep were evaluated.
Perspectives
The Trifecta™ provides hydrodynamics close to the TRIBIO, but the principle of construction is different, especially considering the fixing of leaflet tissue to the stent. In contrast to the TRIBIO and other SAVB with excellent long-term data, such as the Perimount® Magna Ease [15], the Trifecta™ bears its leaflets outside the valve’s stent. This construction principle may lead to increased tension on the commissures during valve motion. Whether this has an effect on degeneration, as it was seen in the comparably constructed Mitroflow (Sorin, Vancouver, Canada), remains speculative [16–18].
Even though the results for haemodynamics, hydrodynamics and GOA of the TRIBIO are promising in this combined ex vivo and in vivo study and may lead to increased durability of this novel SAVB, also the TRIBIO will degenerate over time with the subsequent need for reintervention. Nowadays, TAVI valve-in-valve is an established treatment concept for patients with failing SAVB being a minimal-invasive alternative to redo cardiac surgery. However, the size of the transcatheter heart valve is determined by the size of the SAVB implanted before, which is a problem in SAVB sizes of ≤ 21 mm. To overcome this shortcoming, the base ring of the TRIBIO can be dilated by a balloon to increase the size for the implantation of a larger transcatheter heart valve. In addition, the base ring might serve as an anchoring for later TAVI valve-in-valve.
Limitations
Ex vivo model
This study presents the results of preclinical testing inhering a limited ability to reproduce in vivo conditions, thus the limitations of this study are the consequences of using an ex vivo model: The test series were conducted using physiological saline instead of blood or a glycerine/water mixture providing a more blood-like viscosity. The lower viscosity of physiological saline may lead to lower pressure gradients compared to in vivo conditions. The primary goal of this study, however, was to compare 3 different bioprostheses. Additionally, the SAVB could only be tested at stroke volumes up to 104 ml, which does not reflect maximum exercise. Furthermore, the artificial stiffening of the porcine aortic roots using glutaraldehyde is only an approximation to surgically decalcified human aortic roots during SAVR and has not been validated yet. Whether this causes differences in haemodynamic performance compared with fresh porcine aortic roots remains speculative. Considering the implantation technique in the different SAVB, it needs to be recognized that the TRIBIO was implanted without pledgets in some cases using the base ring itself as abutment. However, this difference in implantation technique is unlikely to cause haemodynamic changes because the base ring of the TRIBIO defines LVOT geometry and excludes pledgets from blood flow.
Sheep model
Sheep are an established model for evaluating the preclinical in vivo safety and haemodynamic performance of cardiovascular devices, as they are known to closely mimic human anatomy and physiology. Nevertheless, an animal model is always merely an approximation to the human anatomy and physiology and therefore, the results of this in vivo model cannot be transferred strictly to clinical conditions. Additionally, it has to be highlighted that only early haemodynamics after weaning of cardiopulmonary bypass but still in anaesthetized sheep were evaluated. Because of lower blood pressure compared with unsedated individuals, transvalvular gradients might be underestimated.
The TRIBIO was not used in men yet, thus the clinical durability is unknown.
Statistical interpretability
Because of the small sample size, statistical results should be interpreted carefully.
CONCLUSIONS
This combined ex vivo and in vivo study shows that the novel TRIBIO bioprosthesis provides superior hydrodynamics and a larger orifice area compared with established SAVB. The TRIBIO was the only SAVB reaching approximately an ideal GOA [12] and provided minor transvalvular gradients during ‘rest’ and ‘exercise’. These ex vivo results are supported by low transvalvular gradients in the preliminary sheep model. Thus, the novel design of the TRIBIO has the potential to create a new generation of SAVB and to improve haemodynamic results in men compared to conventional SAVB, which is increasingly important in small sized aortic annuli in the regard of potential PPM.
ACKNOWLEDGEMENTS
We would like to thank Michael Diwoky for his excellent data management and analyses and for his assistance in preparing this manuscript for publication.
Conflict of interest: Hans-Hinrich Sievers is a shareholder of the TRIBIO Company. None of the remaining authors has any conflicts of interests to disclose.
