Development of a test protocol to study the biomechanical properties of iliac veins in comparison to the common iliac artery

Abstract

Despite development of open and endovascular treatments of the iliac veins, there is no reported research on the biomechanical properties of human iliac veins. The specific requirements for devices and prosthesis for venous interventions are primarily empirical adaptations of devices developed for treating arteries. Although a lot of research has gone into developing devices, the receiving organ has not yet been studied in detail. Using the new methods and testing protocols described in this article we aim to investigate and determine the biomechanical properties of the common and external iliac veins in comparison to the common iliac artery, with the future goal of formulating a new mathematical model for the pelvic veins to facilitate in silico modeling. After consulting literature and experts in the field of biomechanics in search of suitable testing protocols, porcine samples were used to review, develop and confirm methods of conservation, biomechanical testing and histological investigation.

Extension-inflation and planar biaxial extension testing for mechanical data, and histological and structural investigations were evaluated.

The protocol is specifically tailored to the mechanical and structural investigation of human iliac veins and arteries, to allow comprehensive comparison of these vessels, for a better understanding and improvements in biomechanical modeling.

Porcine tissue was used to show feasibility of the tests performed for ethical reasons and in anticipation of future research to evaluate the porcine model currently frequently used in endovascular research and training

Graphical abstract

Specifications table

Subject area	Engineering
More specific subject area	Biomechanics
Name of your method	Extension-Inflation testing, planar biaxial extension testing, multi photon microscopy
Name and reference of original method	Sommer, G.; Benedikt, C.; Niestrawska, J.A.; Hohenberger, G.; Viertler, C.; Regitnig, P.; Cohnert, T.U.; Holzapfel, G.A. Mechanical response of human subclavian and iliac arteries to extension, inflation and torsion. Acta Biomaterialia 2018, 75, 235–252. https://doi.org/10.1016/j.actbio.2018.05.043.
Resource availability	no data

Background

Endovascular treatment of arteries is well established, while interventional treatment of veins is a rapidly growing field [1,2]. Initially, surgical devices such as catheters, stents and prosthesis developed for arterial pathologies were used to treat venous pathologies [3]. The development of such dedicated venous devices has been largely empirical or focused on exploring the mechanical properties of the devices and adapting them using observational data, while the design requirements imposed by the biomechanics of veins have been largely based on assumptions rather than measurements [3,4]. To date, the biomechanical properties of human iliac veins appear to be unexplored. A literature review provides no results for research focused on the biomechanics of human iliac veins as the organ receiving treatment and very few in animal models [[5], [6], [7], [8]]. It is important to provide biomechanical baseline data for a better understanding of the tissue and to support the development of more realistic mathematical models. Since there are no biomechanical studies on iliac veins in the scientific literature and very limited data on iliac arteries, it was essential to develop a protocol to study and compare them mechanically and structurally. The mechanical behavior of the external iliac vein (EIV) and common iliac vein (CIV) under physiological and supra-physiological conditions is of particular interest, considering high wall stresses and deformations due to movement as well as mechanical manipulations during surgical procedures. Based on experimental findings, material models for arteries have been developed and improved continuously [9]. Contrary to human arteries, there are no established test methods and experimental data on human iliac veins. This initially required the development and evaluation of test protocols to study the biomechanical properties of these very delicate vessels in detail and maintain comparability to the common iliac arteries (CIA). The following presents the development of a suitable protocol for sample acquisition, conservation and mechanical testing by a highly interdisciplinary team from the fields of biomechanics, structural imaging, biosciences, bio-imaging, vascular research, vascular surgery, pathology and histo-pathology, for biomechanical, histological and structural multi-photon microscopy (MPM) investigation and analysis of these previously untested tissues.

For ethical reasons, porcine tissue was used instead of human tissue to develop and verify the technical feasibility of the adapted methods. The porcine model is a frequently used surrogate for human tissue and established as experimental animal in pre-clinical studies in the development of edovenous devices and procedures [[10], [11], [12]]. While literature on direct comparison of porcine and human iliac veins could not be found, empirical evidence and long established scientific practice suggest sufficient comparability for the use in the development of this protocol. The cellular and structural buildup of smooth muscle cells, elastic fibers and collagen in the three layers of the vessel wall are generally assumed to be similar in large mammals. However, the extraction of human pelvic vessels was also evaluated during the course of this development, as sample quality was expected to have a significant impact on the various investigations.

Method details

Porcine common iliac vein (PCIV), porcine common iliac artery (PCIA) and porcine caudal caval vein (PCCV) samples harvested in full length of approximately 4 to 6 cm from freshly slaughtered, healthy, 8-month-old pigs of 100–110 kg were used to establish and verify conservation methods and mechanical tests. As these were primarily intended to develop a coherent testing protocol and proof of concept, sample sizes were kept low with two to three porcine vessels per test. Samples were taken from a total of 7 pigs.

Sample conservation

After extraction samples were immediately transported to the laboratory in phosphate-buffered saline (PBS) (pH 7.2–7.4) at 4 °C. Anticipating possible delays before testing due to availability of laboratory resources and personnel, various sample conservation methods were considered to minimize tissue degradation before testing. In our inquiry of relevant literature, cryopreservation was found to be a suitable method to ensure viable samples for mechanical and histological testing. Adequate conservation for studying biomechanical properties may be achieved for at least 6 months by storing the tissue in PBS at −20 °C according to the literature [13]. Previous data investigating porcine aorta by members of this research group have fond no relevant changes in biomechanical behavior and structure of the relevant constituents collagen and elastin induced by sample freezing. Similar results were found by Stemper et al. [14]. Immediately after harvesting and transport samples were therefore rapidly frozen in small vials of PBS at −20 °C and carefully thawed at 4 °C 24 h before testing. This limited the total time of samples outside of cryopreservation at −20 °C to <30 h.

Human pelvic vessels were sourced through the Diagnostic and Research Institute of Pathology, Medical University of Graz, Austria. During sample harvesting in autopsies, receiving intact iliac veins through conventional extraction by lab technicians was rarely possible, due to adhesions of short segments to surrounding tissue and venous confluences. This often resulted in leakage, leading to exclusion of such samples from mechanical testing. Many venous samples that appeared suitable after harvesting showed superficial but devastating damage when inflated. The tissue acquisition protocol was therefore adjusted so that future samples are harvested by a vascular surgeon to minimize damage. Even so, approximately one in five vessels harvested was not in a suitable condition for mechanical and thereof especially extension-inflation testing.

Mechanical testing

In the laboratory of the Institute of Biomechanics at Graz University of Technology, two different mechanical test methods were evaluated to later investigate and determine the biomechanical differences and the vessel-specific behavior of the human common iliac artery (CIA), the common iliac vein (CIV) and the external iliac vein (EIV). The tests selected for this research were extension-inflation testing (EIT), which simulates physiological conditions like axial pre-stretch of vessels in connective tissue, as described by Kamenskiy et al. [15,16] and planar biaxial extension testing (PBET), in which a load is applied along two orthogonal material axes of a sample, i.e., the axial and circumferential directions of the vessel.

Extension-inflation testing

To validate the method for its feasibility with venous tissue, mechanical tests were performed on porcine samples. After careful preparation, the sample’s wall thickness $H_{mean}$ was measured using a video-extensometer. Samples were then attached to custom 3D-printed fixation nozzles mounted on the test apparatus and immersed in PBS at 37 °C. To avoid sample torsion, the nozzles at the base mount of the testing setup consisted of two separate components (see Fig. 1) One, a cannula designed for insertion into the vessel, produced in different gauges to fit different vessel diameters and featuring a groove at the front to secure the sample using braided 2–0 Vicryl threads, ensuring uniform pressure distribution and increased friction. The second component was designed for connection to the pump, equipped with a tight thread for secure attachment.

Fig. 1.

Midline section of the fixation nozzle (a) Technical drawing and (b) 3-D model of the nozzle. 1 .. cannulation component, 2 .. fixation component, A .. groove for sample fixation, B .. channel for fluid inflation, C .. plane for tight sealing, D .. thread to fixate and press the entire nozzle against the inflow adapter.

This nut and bolt like configuration applied pressure at the conical conjunction point creating a watertight connection. The unloaded (reference) state for both veins and arteries subjected to extension-inflation was defined at an axial force of 0 N and an intraluminal pressure of 0 kPa. A constant axial length was maintained over the whole circumference to evenly distribute the stress and avoid torsion of the sample during test preparation. The vessel walls surface was then marked with tissue dye to create a tracking pattern for video-extensometric measurement. Inflation pressure was increased over several cycles (described below) until tissue rupture occurred.

A testing protocol with constant axial stretch as has usually been applied in extension-inflation tests of arterial samples [17] was not applicable to both arterial and venous tissue due to the high compliance of the latter [18]. With the aim of comparing tissue behavior at supra-physiological conditions as they frequently occur during medical interventions, the inflation pressure during testing was deliberately set to exceed physiological central venous pressure many times over. This lead to severe vessel wall buckling, which made optical tracking and reliable measurements impossible (see Fig. 2). An increasing axial stretch to mitigate the buckling effect of veins would result in rupture of the less compliant arterial samples. Therefore, the protocol was changed from an isometric test to an isotonic test with the application of a constant axial force F_ax during vessel inflation.

Fig. 2.

Buckling of the vessel wall of a porcine inferior vena cava with severe buckling that makes video-extensometric tracking of the length and diameter measurements of the vessel very difficult during inflation.

Ideally, the so-called inversion stretch is achieved to allow axial length change independent of transmural pressure [19]. This was validated during protocol development using PCIV and PCCV samples.

Considering the differences in vessel wall geometry, the individual axial force can be calculated to ensure that all vessels are exposed to the same axial engineering stress. Vessels should be pre-stretched during the extension phase and then inflated with constant axial force, i.e. $F_{ax}$ [19,20]. $F_{ax}$ is calculated based on the definition of the first Piola-Kirchhoff stress $P_{ax}$ in the axial direction [21], i.e.

$$F_{ax}=P_{ax}A,$$ (1)

where $A$ is the cross-sectional area of the unloaded vessel, which is calculated using the mean outer diameter $D_{o,mean}$ and the mean wall thickness $H_{mean}$ according to

$$A=\pi {\left(\frac{D_{o,mean}}{2}\right)}^2-\pi {\left(\frac{D_{o,mean}-2H_{mean}}{2}\right)}^2.$$ (2)

The determination of $P_{ax}$ required to estimate the inversion stretch during EIT was carried out by a combination of two different approaches. First, the axial force values obtained from PCCV and PCIV, where no buckling occurred during inflation, can be used to calculate an approximate stress by rearranging Eq. (1). Second, to validate this for arterial tissue, the work of Schulze-Bauer et al. [19], in which the inversion line for human iliac arteries is found at an axial force $F_{ax}$ between 0.5 and 1 N, was consulted. When the data from the arteries studied in [19] are inserted into Eq. (2), the mean cross-sectional area is 35.2 mm². For a mean $F_{ax}$ of 0.75 N, the corresponding $P_{ax}$ is 21.3 kPa. Since these calculated axial stresses and the empirical data from porcine veins gave similar values, a five-phase protocol was defined in terms of axial stresses and inflation pressures (Table 1).

Table 1.

Summary of inflation and extension magnitudes for each phase of the testing protocol.

Phase	Inflation pressure p (mmHg)	Axial stress Pax (kPa)	Inflation rate kPa/min
Phase 1	10	10
	30	10	4
	60	10
Phase 2	100	15	4
Phase 3	150	20	6
Phase 4	200	30	6
Phase 5	700	30	6

Phase 5 can be considered as the “rupture phase”. It is aimed at determining the maximum inflation pressure that arteries and veins can withstand. Each phase includes four extension cycles, starting from zero pre-stretch up to the calculated maximum axial force at rate of 0.01 N/s, followed by four inflation cycles during which $F_{ax}$ is held constant (see Fig. 3). The dynamic software control allows force-controlled holding during inflation, adjusted to the inflation pressure of the respective phase.

Fig. 3.

Phase 1 extension-inflation test cycle: Extension preconditioning (blue) followed by inflation (orange) at constant axial forces.

Changes in axial length and outer diameter were determined using a video-extensometer (videoXtens, version 6.4.0.0, Zwick Roell Group) (see Fig. 4). From these axial length measurements and outer diameter measurements the axial stretch l_ax=l/L (l .. current axial length, L .. reference axial length) and the circumferential stretch along the mean radius l_circ= (r_i + h/2) / (R_i + H_mean/2) (r_i .. current inner radius, R_i .. reference inner radius, h .. current wall thickness) were calculated. This allowed for evaluation of the inversion stretch in the axial direction and correlation of circumferential stretch with the circumferential stress ${\sigma }_{circ}=\frac{p{r}_i}{h}$ ($p$ .. inflation pressure). The calculated stretch and stress can then be plotted against each other to visualize the behavior of the sample tissue during testing.

Fig. 4.

Measuring vessel dimensions during extension-inflation testing (EIT) using black tissue dye as contrast for tracking with a video-extensometer.

Planar biaxial extension testing

The second method evaluated was a planar biaxial extension test (PBET), in which a load is applied along two orthogonal material axes of a sample, i.e., the axial and circumferential directions of the vessel. While EIT is more comprehensive compared to PBET, the latter is more widely accessible and has been performed in research on various human and animal vessels [15,[22], [23], [24]]. Fresh and thawed cryopreserved porcine samples were also tested. A 20 × 20 mm sample was punched out of the vessels and the axial direction was marked. The mean wall thickness $H_{mean}$ was calculated from optical thickness measurements on all four sides using a video-extensometer. Briefly, the sample was placed on an objective slide of known thickness (1 mm) and one lateral side of the samples was oriented toward and perpendicular to a CCD camera. The thickness of the contour sample was then measured from all sides using a calibrated video-extensometer. The samples were mounted in the testing device with five barbless fishhooks strung on a braided Kevlar fishing line on each side and then submerged in PBS at 37 °C. Marking the sample with black tissue dye resulted in strongly contrasting irregular patterns on the sample surface, allowing visual tracking of specifically selected areas of these patterns using a video-extensometer (laserXtens version 3.4.0.0, Zwick/Roell Group Fuerstenfeld, Austria) during testing (see Fig. 5). From these video-extensometer measurements the stretches in the axial direction l_ax = l_ax/L_ax (l_ax .. current axial gage length, L_ax .. reference axial gage length) and the circumferential direction l_circ = l_circ /L_circ (l_circ .. current circumferential gage length, L_circ .. reference circumferential gage length) are calculated. In repeated cycles of stretching and relaxing of the vessel walls, equibiaxial loading is applied with actuator speeds of v = 3 mm/min, with stretch gradually increased by 5 % increments in every cycle until tears occur, leading to complete tissue failure. At each increment a set of cycles consisting of four rounds of stretching and relaxing the sample for preconditioning and subsequent single testing cycles with different loading ratios are applied (Table 2).

Fig. 5.

A sample of the common iliac vein during a planar biaxial extension test (PBET) just prior to tissue failure. Black tissue dye specs for optical tracking, blue line to mark the axial direction.

Table 2.

Planar biaxial extension testing protocol. With each increased stretch increment, the vessel is subjected to varying ratios of stretch in the circumferential and axial directions.

Stretch increment	Loading ratio λax: λcirc
1.05	1: 1	1: 0.75	0.75: 1	1: 0.5	0.5: 1
1.10	1: 1	1: 0.75	0.75: 1	1: 0.5	0.5: 1
:	:	:	:	:	:
:	:	:	:	:	:

To calculate the Cauchy stress $\sigma$ in the respective direction, the axial and circumferential forces measured during testing $f_{ax}$ and $f_{circ}$ and the calculated stretches ${\lambda }_{ax}$ and ${\lambda }_{circ}$ were used [25], i.e.

$${\sigma }_{ax}=\frac{f_{ax}{\lambda }_{ax}}{H_{mean}{L}_{0,circ}}{\sigma }_{circ}=\frac{f_{circ}{\lambda }_{circ}}{H_{mean}{L}_{0,ax}},$$ (3)

where $L_{0,ax}$ and $L_{0,circ}$ are the initial dimensions of the samples in the unloaded configuration.

Light microscopy

Histological analysis is carried out at the Diagnostic and Research Institute of Pathology at the Medical University of Graz. To better understand the structural composition at both the cellular and extracellular level, as well as the effects of mechanical stress, each sample is examined histologically before and after biomechanical testing. A 3 mm-wide circumferential strip is cut from the vessel proximal to the biaxial test sample. Following PBET, the sample is halved along the circumferential axis, resulting in two rectangular histological samples with a defined orientation and a side-length ratio of 2:1. Samples are fixed in 4 % formaldehyde, embedded in paraffin and sectioned at 5 μm thickness. Hematoxylin and eosin (HE) and Elastica van Gieson (EVG) stains are prepared to show collagen, elastin, smooth muscle cells, and the overall tissue structure. This will allow not only for matching of the sample structure and mechanical behavior, but also comparison of untested and tested matched tissue pairs.

Multi-photon microscopy

In addition to conventional light microscopy the MPM techniques of second harmonic generation (SHG) for collagen and two-photon excited fluorescence (TPF) imaging for elastin are used to visualize the three-dimensional architecture of mechanically relevant constituents in the vessel walls. MPM enables high-resolution imaging of extracellular structures relevant to biomechanical behavior [26]. SHG is a type of coherent nonlinear microscopy that uses scattered photons to precisely image structural elements [27]. In TPF, excitation leads to auto-fluorescence of materials when excited electrons emit photons while returning to their ground state with defined wavelengths for specific molecules.

Both techniques have established themselves as effective methods for structural imaging in vessel walls [26,27]. Data extracted from these images will be used as structural parameters in modeling the biomechanical properties of the vessels. Using tissue adjacent to that taken for PBET may allow a direct correlation of MPM data and mechanical data.

Since MPM is limited to about 0.12 mm by the laser penetration into the tissue, it was decided to increase this to 1.5 mm by an optical clearing technique, thus enabling imaging through the vessel wall thickness [28]. For this, tissue fixed in formaldehyde must be rinsed in PBS and then dehydrated in a graded series of ethanol. Thereafter samples are cleared in a mixture of ethanol and benzyl alcohol benzyl benzoate (BABB) 1:1 followed by pure BABB for a minimum of 12 h. Samples remain in this medium until imaging. Image composition and filters are similar to those of Pukaluk and colleagues [27]. A picosecond laser and an optical parametric oscillator are used to induce SHG and TPF with a wavelength of 880 nm. Detection and imaging is performed with a band-pass emission filter and a water immersion objective. Images will be acquired from the circumferential-axial plane (across wall thickness, in-plane stacking) and circumferential-radial plane (out-of-plane tiling) to analyze fiber orientation, isotropic and anisotropic regions in the corresponding x-, y- and z-axes.

Method validation

As mentioned above, to validate the novel approaches of mechanical testing and adaptation for venous tissue, porcine samples were used. The following preliminary results show data collected from PCIV, PCIA and PCCV.

Cryopreservation

The decision to store samples frozen at −20 °C was made under the consideration of possible alterations of tissue behavior induced through freezing and thawing. The relevant vessel wall constituents responsible for passive mechanical behavior are collagen and elastin. The scientific literature and prior research done in arterial tissue at the institute of biomechanics consulted on this matter showed no significant effect of freezing and thawing on these nor the mechanical behavior of arterial vessel walls [14]. A comparison of PBET data of a frozen and fresh sample can be found below. (see Fig. 8) The slight variance in the stress-strain curves shown here fall well within the variability observed in tissue testing as described in previous findings in porcine aortas [14]. Histological investigation also did not show freezing induced microstructural damage to elastin nor collagen fibers as described below.

Fig. 8.

Cauchy stress vs. stretch behavior of fresh and cryopreserved PCCV samples obtained from PBET. Red area indicates the formed hysteresis area observed in these samples during PBET. Preliminary data of arterial and venous tissue show reliable results in applicability of the test protocol with sufficient tissue compliance to acquire data from multiple stretch cycles exceeding 15 % stretch before tissue failure in both sample types while exhibiting seemingly different behavior under increased stress. (see Fig. 9).

Extension-inflation testing

While PCIA samples were successfully tested using the initial protocol [19], testing using PCIV and PCCV proved to be infeasible. The samples showed large deformations in the circumferential and axial direction, which resulted in considerable elongation requiring a substantial increase in axial stretch, to the point where the measurements were no longer comparable to those of arteries due to the buckling effect described above (see Fig. 2). Adapting the novel approach of using a constant axial first Piola-Kirchhoff stress (isotonic) rather than a constant axial stretch (isometric) as a common reference resolved these issues. The implementation of this approach showed promising and consisting results in EIT using venous and arterial samples, with the plotted data approaching the inversion stretch (see Fig. 6, Fig. 7). This preliminary data in combination with the results of Schulze-Bauer et al. [19] resulted in a testing protocol applicable for human common iliac arteries and veins, providing adequate conditions to maintain the same core testing procedures (sample preparation, mounting, preconditioning, loading protocol, and data acquisition) for arterial and venous specimens. First results of EIT testing on human CIA and CIV samples are given (see Fig. 7), which validates the feasibility of a direct comparison between arteries and veins, since the testing protocol is applicable on both vessel types in the same manner, which ensures comparability and standardization.

Fig. 6.

Inflation pressure vs. axial stretch behavior of a porcine common iliac vein at different inflation pressures.

Fig. 7.

Inflation pressure vs. axial stretch behavior of a human CIA and CIV sample at given inflation pressures. With the newly developed testing protocol, the phenomenon of inversion stretch to minimize axial length elongation during inflation testing [19] was achieved.

Planar biaxial extension testing

There was no obvious difference between fresh and slowly thawed cryopreserved PCCV samples (see Fig. 8). PBET of porcine samples yielded valuable data, with stretch ranges between 20 % and 25 %, comparable to arterial data reported in previous studies [15]. This led to the adaptation of a modified multi-ratio stretch-controlled testing protocol based on the work of Niestrawska et al. [24] for all further investigation of human samples.

Fig. 9.

Cauchy stress vs. stretch behavior of CIA and CIV at 10 % and 15 % stretch in PBET. Red area indicates the formed hysteresis area. Initially flat curves of CIV with steep incline at higher stretch may indicate high initial compliance with prompt tissue stiffness due to differences in wall structure.

Histology

Cell nuclei of frozen porcine samples showed partial damage. However, no relevant damage occurred at a cellular and structural (extracellular matrix) level. Histological investigation of the samples showed no damage or changes in overall elastin and collagen fiber structure in fresh versus cryopreserved tissue in the specific histological analysis relevant to this study (see Fig. 10). The content of elastin and collagen in the samples – relevant for the passive mechanical behavior – and smooth muscle cells as well as their morphological changes in the tissue after PBET could be adequately assessed by histological investigations with HE and EVG staining.

Fig. 10.

Representative PCCV sections stained with HE (left) and EVG (right). Fresh samples (top row) and frozen, slowly thawed samples (bottom row) are shown. Blue scale bars indicate 200 µm.

Digital pathology was considered a well-suited tool for future qualitative and quantitative analysis of the structure of collagen, elastin, and smooth muscle cells in human samples.

Multi-photon microscopy

MPM imaging was initially performed in the laboratory of the Institute of Biosciences at the University of Graz. The results indicate that the microstructure of arteries, and especially veins, can be effectively analyzed in both in-plane and out-of-plane orientations. The TPF signal is calibrated to show elastin. It highlights differences in fiber density, continuity, and lamellar organization of elastin throughout the vessel wall, visualized as red color, whereas the SHG signal emphasizes variations in fiber orientation, packing, and network organization of collagen. The visualization of the structural build up and analysis of these components will allow for comparison between arterial and venous tissues. All images were acquired using identical acquisition as well as identical filter settings in further processing. The stark difference in vessel thickness and general structure might already be appreciated in the sample overview. (see Fig. 11)

Fig. 11.

Overview of out-of-plane tile of a representative CIA in the left column and CIV in the right column obtained by multi-photon microscopy (MPM), showing the two-photon excited fluorescence (TPF) signal (elastin fiber density, continuity, and lamellar organization) in the top row, the second harmonic generation (SHG) signal (collagen fiber orientation, packing, and network organization) in the middle row, and both signals indicating collagen and elastin in the third row. White scale bar indicates 200 µm.

Limitations

Difficulty during extraction and exclusion of damaged samples might lead to a selection bias in tissue. This could exclude less biomechanically stable material from the investigation. It is therefore advisable to stratify samples according to factors possibly detrimental to tissue stability and keeping thorough records of age, sex and comorbidities as well as cause of death when dealing with larger sample sizes. While damage during extraction may be a sign of pathological alteration of the tissue, it must also be stated that handling pelvic veins is a distinctly delicate procedure even for experienced vascular surgeons because of their known fragility and anatomical variance of confluences.

While sample preservation in PBS is commonly used in vascular biomechanics studies due to its advantage of tissue hydration at physiological pH and osmolarity, the authors are aware, that long term storage of vascular tissue in saline solutions may lead to saline induced wall damage this has only been shown in the case of multiple day refrigerated storage. This research is conducted using cryopreservation at −20 °C with efforts undertaken to minimize the time of unfrozen tissue exposure.

Ethics statements

Approval was sought from the ethics board of the medical university Graz for the use of human (vote 32–161) and porcine (vote 33–385) samples. The need for patient consent was waived as human samples were extracted during routine autopsies. This research was conducted in accordance with the Declaration of Helsinki.

CRediT authorship contribution statement

Gregor Siegl: Conceptualization, Methodology, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Supervision, Project administration. Katharina Rampitsch: Methodology, Formal analysis, Investigation, Resources, Data curation, Writing – original draft. Selda Sherifova: Methodology, Formal analysis, Data curation, Writing – review & editing. Gerhard A. Holzapfel: Conceptualization, Writing – review & editing. Gerhard Sommer: Conceptualization, Methodology, Formal analysis, Investigation, Resources, Data curation, Writing – review & editing, Supervision. Tina Cohnert: Conceptualization, Writing – review & editing, Supervision.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.