Harvesting and sample preparation for mechanical testing
A total of nine complexes comprising of the human AT, adjacent calcaneus and PF (subsequently referred to as AT-calcaneus-PF complex) were retrieved post mortem (mean age 74, age range 28 to 93). One of the complexes was used for histological assessment (28-year-old male), two for plastination (87-year-old female, corresponding to one left and one right complex) and six for mechanical testing (four phenoxyethanol-based embalmed cadavers12, left and right complex was used of two cadavers, two right complexes of two other cadavers, mean age 82 years, 3 males, 1 female). The histologically processed AT-calcaneus-PF complex was retrieved at the Institute of Legal Medicine, University of Leipzig, Germany during leg preparation and the remaining ones were acquired from bequeathed cadavers for medical education and research purposes at the Department of Anatomy of the University of Otago in Dunedin, New Zealand. The University of Otago Ethics Committee approved this study (approval number H17/20), in conjunction with Maori consultation being sought from the Ngai Tāhu Research Consultation Committee and the Ethics committee of the University of Leipzig, Germany Study approved the tissue sampling (protocol number 486/16-ek).
Dissection, decalcification and sample retrieval
Following the retrieval, the six AT-calcaneus-PF complexes of the mechanical testing group were prepared for decalcification as follows: First, the AT was exposed carefully until the calcaneus was reached distally, consecutively removing all surrounding tissues except for the periosteum. Second, the AT was proximally transsected, distally to the macroscopically last visible muscle fibres of the soleus muscle. Thereafter, the PF was exposed and mobilized from the heads of the metatarsal bones distally to the calcaneus proximally with a sharp scalpel. Following this, the lower half of the calcaneus was cut with an ultrasound knife (PIEZOSURGERY®, Mectron, Saline, MI, USA) with subsequent blunt removal of trabecular bone parts by means of the handle of a scalpel to facilitate the following decalcification of the superficial calcaneal trabecular layer, delineating from the surrounding trabeculae. Decalcification of the samples removed the hydroxyapatite leaving behind the collagen scaffold that connects the AT and the PF. This allowed to investigate the mechanical properties of the collagenous scaffold that connects the AT and the PF and compare it to the two structures. Following the hydroxyapatite removal, the collagen scaffold mechanically behaves in a soft tissue-like manner. For the decalcification of calcaneal samples, the resulting complexes of AT and PF connected by the most superficial inferior and posterior calcaneal trabeculae were submerged into a 10wt% ethylene-diamine-tetra-acetate (EDTA) solution for four weeks with a weekly renewal. Then, dog bone-shaped samples adapted from an ISO-527–2 template13 for biomechanical testing were cut from the AT-calcaneus-PF complex according to Fig. 1A. Samples for biomechanical testing were retrieved along the AT, calcaneus and PF along a medial and a lateral line based on the following morphological observations: the median fibrous septum of the soleus muscle (separating the AT fibre bundles into a medial and lateral line), median fibrous septae proximal to the most proximal insertion of the AT into the calcaneus (symmetrically separating the AT into medial and lateral bundles shown in histology sections) and finally the division of the PF into two larger bundles (central and lateral, the medial bundle was neglected here due to its smaller size). Along the medial and lateral lines, the following corresponding pairs were retrieved: MPAT (medial proximal AT) and LPAT (lateral proximal AT), MDAT (medial distal AT) and LDAT (lateral distal AT); MB (medial bone) and LB (lateral bone); MCP (medial calcaneal periosteum) and LCP (lateral calcaneal periosteum); MPPF (medial proximal PF) and LPPF (lateral proximal PF) and MIPF (medial intermediate PF) and LDPF (lateral distal PF). The MDPF (medial distal PF) of the medial line did not have a lateral line counterpart.
Uniaxial quasi-static mechanical testing
The samples’ cross-sectional areas were determined by creating polysiloxane (medium-bodied, Exahiflex; GC Corporation, Tokyo, Japan) cross section casts, which were subsequently scanned (Perfection 7V750Pro; Seiko Epson Corporation, Suwa, Japan) and computed by means of the Measure 2.1d software (DatInf, Tübingen, Germany). Self-printed clamps with sharp pyramids to prevent specimen slippage during tensile testing were used (Fig. 1B)14. Before tensile testing was conducted, the samples were speckled using a black pencil to create a randomly distributed pattern, necessary for the digital image correlation (DIC) (Fig. 1C). A uniaxial testing machine (Allround Table Top Z020; Zwick Roell, Ulm, Germany) equipped with an Xforce P load cell (2.5 kN; Zwick Roell, Ulm, Germany) was used to conduct the tensile tests at room temperature. The testControl II software (Zwick Roell, Ulm, Germany) was used. Twenty load-unload preconditioning cycles with a force range of 0.5 to 2.0 N were applied before the tissues were stretched until failure (Fig. 1C–E). All tissues were strained in the longitudinal axis according to the samples’ predominant collagen orientation. The displacement rate was 20 mm/min and the sample reading rate was 100 Hz. A single-charge coupled camera with a resolution of 2.8 Megapixels (Q400; Limess, Krefeld, Germany) and the ISTRA 4D software (VRS 220.127.116.114; Dantec Dynamics, Ulm, Germany) were used for strain data evaluation of the mechanical tests.
Data processing and statistical analysis
Synchronized force readings by MATLAB R2017b software (Mathworks, Natick, USA) and DIC data were used to calculate mechanical properties. Elastic modulus (Emod), strain at maximum force (SFmax) and ultimate tensile strength (UTS) were evaluated (Fig. 2). For statistical evaluation, Excel version 16.15 (Microsoft Corporation, Redmond, USA) was chosen. Firstly, the biomechanical parameters of the subsamples were averaged according to their respective location (e.g., all MB values of the different tested samples were averaged) within the AT-calcaneus-PF complex. Following this, the averaged values of the subsamples were compared among the 13 different sites (e.g., MB vs. LB) and classified into the following groups according to being part of the percentile fractions shown in parentheses: low (0 to 24th percentile), below median (25th to 49th percentile), median (50th percentile), above median (51 to 74th percentile) and high (75th to 100th percentile).
E12-sheet plastination and histology
The two E12-plastinated AT-calcaneus-PF complexes used in this study were taken from one entirely plastinated cadaver. The cadaver was embalmed according to Xu et al.15. The plastinates were scanned at 1200 dpi (Epson Perfection V750 Pro Scanner, Epson, Jakarta, Indonesia). For histological analyses, one fresh AT-calcaneus-PF complex was embedded into paraffin and horizontally sectioned with a section thickness of 20 µm using the same orientation as for the plastinates. Subsequently, the selected sections that corresponded to the plastinated slices were stained with hematoxylin eosin (H&E) and Masson–Goldner as trichromic staining (all consumables by Dr. Hollborn GmbH & Co KG, Leipzig, Germany) and photographed using a Zeiss Axioskop 40 (Carl Zeiss AG, Oberkochen, Germany) combined with an Olympus DP22 camera system (Olympus K.K., Shinjuku, Japan).