- Email: [email protected]
Stiffness and energy dissipation across the superficial and deeper third metacarpal subchondral bone in Thoroughbred racehorses under high-rate compression
Journal of the Mechanical Behavior of Biomedical Materials 85 (2018) 51–56
Contents lists available at ScienceDirect
Journal of the Mechanical Behavior of Biomedical Materials journal homepage: www.elsevier.com/locate/jmbbm
Stiffness and energy dissipation across the superficial and deeper third metacarpal subchondral bone in Thoroughbred racehorses under high-rate compression
T
⁎
Fatemeh Malekipoura, , Chris R. Whittonb, Peter Vee-Sin Leea a b
Department of Biomedical Engineering, University of Melbourne, Parkville, VIC 3010, Australia Equine Centre, Faculty of Veterinary and Agricultural Sciences, University of Melbourne, Werribee, VIC 3030, Australia
A B S T R A C T
Subchondral bone injury due to high magnitude and repetition of compressive loading is common in humans and athletic animals such as Thoroughbred racehorses. Repeated loading of the joint surface may alter the subchondral bone microstructure and initiate microdamage in the bone adjacent to the articular cartilage. Understanding the relationship between microdamage, microstructure and mechanical properties of the subchondral bone adjacent to the articular cartilage is, therefore, essential in understanding the mechanism of subchondral bone injury. In this study, we used high-resolution µCT scanning, a digital image-based strain measurement technique, and mechanical testing to evaluate the three-dimensional pre-existing microcracks, bone volume fraction (BVF) and bone mineral density (BMD), and mechanical properties (stiffness and hysteresis) of subchondral bone (n = 10) from the distopalmar aspect of the third metacarpal (MC3) condyles of Thoroughbred racehorses under high-rate compression. We specifically compared the properties of two regions of interest in the subchondral bone: the 2 mm superficial subchondral bone (SSB) and its underlying 2 mm deep subchondral bone (DSB). The DSB region was 3.0 ± 1.2 times stiffer than its overlying SSB, yet it dissipated much less energy compared to the SSB. There was no correlation between structural properties (BVF and BMD) and mechanical properties (stiffness and energy loss), except for BMD and energy loss in SSB. The lower stiffness of the most superficial subchondral bone in the distal metacarpal condyles may protect the overlying cartilage and the underlying subchondral bone from damage under the high-rate compression experienced during galloping. However, repeated high-rate loading over time has the potential to inhibit bone turnover and induce bone fatigue, consistent with the high prevalence of subchondral bone injury and fractures in athletic humans and racehorses.
1. Introduction Subchondral bone injury due to high magnitude and repetition of compressive loading is common in humans (Devas, 1958; Matcuk et al., 2016) and athletic animals such as Thoroughbred racehorses (Barr et al., 2009; Riggs et al., 1999). Repeated loading of the joint surface may alter the subchondral bone microstructure and initiate microdamage in the bone adjacent to the articular cartilage (Muir et al., 2008; Stepnik et al., 2004). Microdamage may stimulate both modelling and remodelling responses in subchondral bone and can lead to the development of stress fractures and/or osteoarthritis in the joint (Burr et al., 1985; Kawcak et al., 2001; Muir et al., 2006; Norrdin and Stover, 2006). Specifically, altered subchondral bone mechanics due to microdamage can influence how joint surface load is transferred to the
⁎
Corresponding author. E-mail address: [email protected] (F. Malekipour).
https://doi.org/10.1016/j.jmbbm.2018.05.031 Received 29 August 2017; Received in revised form 16 May 2018; Accepted 21 May 2018 Available online 22 May 2018 1751-6161/ © 2018 Elsevier Ltd. All rights reserved.
underlying trabecular bone, and thus trigger changes in bone modelling and remodelling which are highly load dependent. (Burr, 2004; Shirazi and Shirazi-Adl, 2009; Whitton et al., 2010). Understanding the relationship between microdamage, microstructure and mechanical properties of the subchondral bone adjacent to the articular cartilage is, therefore, essential in understanding the mechanism of microdamage initiation and propagation across the whole joint. The metacarpophalangeal joint in racehorses is subjected to extremely high compressive loads during galloping (Harrison et al., 2014). The palmar metacarpal subchondral bone, particularly the bone within 2–3 mm of the cartilage-calcified cartilage interface is a common site of fatigue-induced microdamage (Muir et al., 2006), which highlights the importance of investigation of subchondral bone mechanical/ structural properties at this site. Previous studies investigated the
Journal of the Mechanical Behavior of Biomedical Materials 85 (2018) 51–56
F. Malekipour et al.
2.2. Micro-CT scanning and bone microarchitecture
relationship between the microstructure and mechanical properties of equine subchondral bone from the metacarpal condyles under low-rate compression (Rubio-Martínez et al., 2008; Leahy et al., 2010). RubioMartínez et al. (2008) observed a gradient of stiffness across the subchondral bone depth; superficial 6 mm cubes of dense subchondral bone were less stiff than deeper 6 mm cubes of trabecular bone. Additionally, variations in the mechanical properties of the trabecular bone specimens from condylar regions were associated with structural measures derived from micro-computer tomography (µCT) but similar associations were not observed for the more superficial subchondral bone specimens. In the present study, we focused on the subchondral bone area closest to the joint surface, i.e. subchondral bone plate in RubioMartínez et al. (2008), to increase the resolution of the previous measurements by investigating two regions of interest (ROI) within the equine third metacarpal (MC3) subchondral bone: the superficial subchondral bone (SSB: 0.5–2.5 mm) and its underlying deep subchondral bone (DSB: 2.5–4.5 mm). We used high-resolution µCT and a digital image-based strain measurement system to evaluate the relative microstructure and mechanical properties (stiffness and energy absorption) of the two ROIS under high-rate compression. The image-based strain measurement technique allows for the investigation of tissue strain while avoiding artefacts associated with the free trabecular end of specimens, and their non-uniform thickness (Malekipour et al., 2013). The specific objectives of this study were to: 1) investigate the microstructure of the subchondral bone and identify pre-existing microcracks using µCT scanning, 2) determine the relative stiffness and shock-absorbing abilities (normalized hysteresis) of the SSB and DSB under experimental high-rate compression similar to one cycle of compression during galloping, and 3) investigate the relationship between the mechanical/structural properties of SSB and DSB. We hypothesized that the superficial 2 mm of subchondral bone would be less stiff, and dissipate more energy under high rate loading than the deeper 2 mm of bone, with associated differences in density and bone volume fraction gradient.
Prior to loading, cartilage-bone plugs were scanned using a µCT scanner (µCT50; Scanco Medical, Switzerland) at an isotropic resolution of 4 µm using 70 kVp tube voltage, 200 µA tube current, and 1050 ms integration time. A phantom was used to convert CT grayscale intensity values (Hounsfield units, HU) into equivalent bone mineral density (BMD, g/cm3 HA). A strong correlation has been demonstrated between quantitative CT and mineralisation and ash density in bone from the equine distal metacarpus (Drum et al., 2009). We used a voxel size of 4 µm, which was the highest resolution of the scanner that achieved good quality images for our specimens, to identify pre-existing microdamage in the subchondral bone. Following the examination for microdamage (ImageJ, 1.48 v, National Institute of Health, USA), images were down-sampled to 40 µm for further morphology processing (Simpleware, v5.1, UK). Damage was identified by visual examination of all µCT slices assessing for linear lucencies or linear areas of increased mineralisation (Williamson et al., 2017). Examination of profile lines and images confirmed that this down-sampling affected the density and bone volume fraction (BVF) measurements no more than 0.06% and 0.39%, respectively. Mineralised bone was segmented using a consistent set of threshold grey values. SSB and DSB regions corresponding to the experimental ROIs were identified based on their distances from the cartilage-bone interface and thicknesses in µCT images. BVF of each ROI and the entire bone was calculated by dividing the volume of bone tissue in that region by its apparent volume. BMD of each ROI and the entire bone were specified as the average BMD over that region. 2.3. Mechanical testing The experimental procedure was similar to that of a previous study (Malekipour et al., 2013). Cartilage-bone plugs were glued to a base plate using a thin layer (several micrometres) of cyanoacrylate glue on the proximal end of the bone (Burgin and Aspden, 2008). Vertical displacements were applied to the articular surface of cartilage via a stainless steel compression plate on a mechanical testing machine (Instron, 8874, UK). A time to peak strain of 0.05 s was chosen, similar to the time to peak bone strain during galloping of a racehorse (Davies and Mccarthy, 1994; Swanstrom et al., 2005). Displacements were applied to the cartilage surface using half-sine waveforms with magnitudes of 0.225 ± 0.018 mm at 5 Hz. The displacement magnitudes were calculated to apply 50–60% of equine subchondral bone yield strain (Rubio-Martínez et al., 2008). The stained flat surface of cartilage-bone plugs faced a microscope lens. A high-speed camera (MotionProY3, 1280 × 1024, USA) attached to a stereomicroscope (SZX7, Olympus, Japan) recorded the experimental deformation of each specimen at 1000 frames per second. Microscopic magnification was chosen that allowed imaging at a resolution of 170 pixels/mm (1024 pixels within 6 mm) in the vertical direction. Recorded images of each specimen were processed using a Matlab-based digital image correlation (DIC) code (Jones, 2013) to calculate the average displacement of bone at three different levels (Lines 1–3, Fig. 1B). A respective subset and grid size of 71 and 20 pixels was found to accurately calculate the displacement within ROIs. Using these parameters, the software measured the displacement of the Instron actuator with an accuracy of 1.89%. The ROIs were selected at a 0.5 mm distance from the cartilage-bone interface. Above this distance cartilage folding under loading hindered clear images, and an accurate displacement measurement of the bone. To remain unaffected by the cutting process and the presence of free edges, ROIs were specified away from the edges. In a preliminary study, the width of ROIs were reduced from the full width until the calculated stiffness and energy dissipation varied no more than 1.41 ± 1.35% and 1.18 ± 0.72%, respectively in the central 2 mm (Fig. 1B). The displacement data were then processed (MATLAB R2016a, MathWorks, USA) to calculate the overall strains of the two ROIs, i.e. SSB and DSB,
2. Methods 2.1. Specimen preparation Cartilage-bone plugs (Ø = 6.5 ± 0.2 mm) were extracted from the distopalmar aspect of fresh frozen (−20 °C) third metacarpal condyles of n = 10 racehorses using a diamond core drill (Starlite Industries, Rosemont, PA) under continuous irrigation. Metacarpal condyles were collected from Thoroughbred racehorses of median age of 3.5 years (range, 3–7 years) that died or were euthanatized on racetracks in Victoria, Australia between May 2011 and March 2014. Specimens were harvested from both medial (n = 7) and lateral (n = 3) condyles. During harvesting, the axis of the drill was oriented perpendicular to the cartilage surface. Bone was trimmed down on the proximal end of each specimen by using a self-irrigated diamond saw (Isomet, Buehler, Ltd., Lake Bluff, IL) to create a flat bone surface perpendicular to the longitudinal axis of the cartilage-bone plugs. The final trimmed bone thickness was 9.68 ± 0.79 mm with an average cartilage thickness of 0.46 ± 0.06 mm. To keep the integrity of the subchondral plate adjacent to the articular cartilage, we preserved the cartilage in situ. Subsequently, the cartilage-bone plugs were trimmed perpendicular to the cartilage surface to create a flat surface on the longitudinal side of each specimen (Fig. 1A). Prior to loading, this longitudinal surface was stained with fine graphite particles to create a random speckle pattern, which was necessary for the later image processing to calculate bone strains. The graphite particles were generated by rubbing pencil lead with sand paper, a technique which has been validated in a previous study (Malekipour et al., 2013).
52
Journal of the Mechanical Behavior of Biomedical Materials 85 (2018) 51–56
F. Malekipour et al.
Fig. 1. Schematic illustrations of (A) the topview of a cartilage-bone specimen indicating 1 mm trimmed, which was cut using a diamond saw to generate a flat surface for the staining, (B) the mechanical testing set-up indicating the side-view of a cartilage-bone plug, SSB and DSB regions of interest, the compression plate, and the microscope attached to a high-speed camera.
3.2. Mechanical testing
by dividing the relative displacements of the upper and lower bounds to the thickness of each ROI (2 mm). Overall compressive axial stress was calculated by dividing the compressive load measured on the Instron load cell ( ± 10 kN) by the horizontal cross-sectional area of each specimen. Stress data (Instron output) were synchronized to the strain data (calculated from image processing) by matching the time that the compression plate touched the articular surface, and stress-strain curves of each ROI for each specimen were determined. The slope of the linear portion (R2 > 0.96) of the loading stress-strain curves were calculated and specified as the experimental stiffness of SSB (KISB ) and DSB (KDSB ). The areas enclosed by each stress-strain curve (hysteresis) were divided by the area beneath the loading curve (absorbed energy) to calculate the shock-absorbing abilities (normalized hysteresis) of each ROI.
Applied overall strains of 2.33 ± 0.25% (mean ± SD) generated axial compressive stresses of 44.2 ± 10.7 MPa to cartilage-bone plugs. The two specimens with pre-existing microfractures failed during the compression. In one of these specimens, the induced microfracture remained distal to the SSB region. In the second one, the pre-existing microfracture propagated proximally into the SSB area, causing blurring of real-time images. Therefore, no post-yield and unloading strain data were available for this specimen. In all specimens, the DSB region was stiffer than its overlying SSB with an average DSB-to-SSB stiffness ratio of 3.0 ± 1.2 (Table 2). The normalized hysteresis of SSB was greater than that of DSB by a factor of 3.1 ± 0.8 (Table 2). Fig. 3 illustrates typical stress-strain curves associated with the SSB and DSB. Specimens that absorbed a higher strain in the SSB region under the applied overall strain exhibited greater energy dissipation at this region (Fig. 4A). Unlike the SSB, the DSB energy dissipation did not correlate with its maximum absorbed strain (Fig. 4B).
2.4. Statistical analysis A Grubbs' Test was applied to exclude the outliers, resulting in exclusion of one values of DSB stiffness. Paired t-tests were performed to compare the properties between SSB and DSB; where data was not normally distributed, Wilcoxon signed rank tests were used. Pearson's correlation coefficients were calculated to determine relationships between mechanical parameters. The statistical differences in BVF and BMD between lateral and medial condyles were also investigated using one-way ANOVA. The level of significance was set at p = 0.05. All statistical analyses were performed in Minitab (MINITAB Inc., version 17.0).
3.3. Relationship between bone microarchitecture, density and mechanical properties No significant correlations were detected between the structural (BVF , BMD ) and mechanical properties (stiffness and energy loss) of bone within either the SSB or the DSB (Table 3), except for the SSB energy loss which showed a strong correlation with its BMD (r = −0.83, p = 0.005). The relative stiffness ( KDSB ) also correlated K SSB
with the relative density ( BMDDSB ) (r = 0.63, p = 0.05), but not with the
3. Results
BMDSSB
relative BVF ( BVFDSB ) (r = 0.20, p = 0.58). BVFSSB
3.1. Micro-CT image analysis
4. Discussion
Based on the µCT images performed prior to mechanical testing, two specimens exhibited pre-existing microfractures in the subchondral bone immediately beneath the articular cartilage. Microfractures in one specimen appeared as irregular linear transverse lucencies 0.5 mm below the calcified cartilage/hyaline cartilage interface with an approximate width of 26 µm (Fig. 2B). In the same specimen, a site of hypermineralized material was observed projecting into the overlying uncalcified cartilage. In the other damaged specimen, a fine microcrack was observed just beneath the articular cartilage. The remaining specimens did not reveal any pre-existing microcracks in the µCT images (Fig. 2A). All specimens exhibited dense structure with a BVF > 0.9. There were no significant differences in BMD and BVF between lateral and medial specimens (p > 0.05). The SSB region had equivalent BVF, yet lower BMD when compared to its underlying DSB (Table 1).
We have shown that under loading rates comparable to those generated in the fetlock joint of a galloping horse, the dense subchondral bone of the distal palmar metacarpus is less stiff and able to dissipate more energy in its most superficial 2 mm (SSB) compared to the deeper 2 mm of subchondral bone (DSB), confirming our hypothesis. However, contrary to our hypothesis, we found a weak and insignificant association between the structural and mechanical properties of bone, except for the energy dissipation and BMD of the SSB region. In contrast, a significant correlation was revealed when the relative values of stiffness and BMD were compared. The stiffness of the entire thickness of bone specimens in our study (10 mm) was higher than that of the equivalent subchondral bone area reported previously (Rubio-Martínez et al., 2008). The difference can be attributed to the strain measurement technique we used here that 53
Journal of the Mechanical Behavior of Biomedical Materials 85 (2018) 51–56
F. Malekipour et al.
Fig. 2. Representative sagittal cross-sections µCT images of cartilage-bone plugs at 4 µm resolution (A) without any evidence of microfracture, and (B) with pre-existing microdamage including a microfracture ~0.5 mm beneath the tidemark with an approximate width of 26 µm (arrow), and a focal area of hypermineralisation (inside the circle) projecting into the unmineralised cartilage.
Table 1 Mean ± SD of bone volume fraction (BVF), and bone tissue mineral density (BMD) determined via µCT image analysis. P-values calculated from a paired ttest between SSB and DSB regions.
BV/TV BMD (mg/ cm3 HA)
Total bone (~10 mm)
SSB (2 mm)
DSB (2 mm)
P-value
0.936 ± 0.036 883.4 ± 15.4
0.984 ± 0.012 852.9 ± 18.7
0.970 ± 0.018 890.7 ± 13.6
0.069 0.000
The detailed results for n = 10 specimens are provided in Supplementary Materials Table S1.
eliminated end-cap artefacts. Moreover, Rubio-Martínez et al. (2008) used samples from horses with mild and severe subchondral bone changes, whereas in the current study all specimens were from horses with mild subchondral bone disease. The subchondral bone of the distal metacarpus of equine athletes experiences high magnitude cyclic loading when horses are in race training, resulting in densification of the bone when compared to the subchondral bone of young or non-race horses (Boyde and Firth, 2005). The BVF of the subchondral bone from this area is in a range similar to that of cortical bone (Cardoso et al., 2013). However, unlike cortical bone compressive stiffness which was predicted by its porosity (Schaffler and Burr, 1988), the subchondral bone stiffness in our study did not show any relationship with its porosity (1-BVF). Furthermore, differences in microstructure such as the orientation of collagen fibres in the SSB and DSB regions may play a role in differences in stiffness of the SSB and DSB. Martin and Boardman (1993) showed that collagen fibre orientation was the best predictor of bending properties of bovine cortical bone. Future studies are required to evaluate such effects on the equine subchondral bone. Similar to what has previously been observed in equine metacarpal subchondral bone from Thoroughbred racehorses to a depth of 6 mm, we were unable to identify any significant relationships between mechanical and structural properties despite a higher resolution of µCT imaging and a more accurate strain measurement technique (Rubio-
Fig. 3. Representative experimental stress-strain curves associated with SSB and DSB regions of a typical cartilage-bone specimen.
Martínez et al., 2008). Similarly, Ding et al. (2001) found that human subchondral bone apparent density failed to explain the reduced mechanical strength of the subchondral bone at early stages of OA, whereas the mechanical strength of healthy subchondral bone did correlate with its apparent bone mineral density. Such a lack of correlation between the subchondral bone structural and mechanical factors may be due to a differing microstructure such as the orientation of collagen fibres in the SSB and DSB, or deteriorated bone quality at the tissue-level, and/or the accumulation of microcracks due to fatigue loading. In addition to the bone microstructure, microdamage has been shown to have substantial effects on material properties of bone. For example, in trabecular bone, a damage volume fraction (damage volume/bone volume) of 1.5% caused a 50% reduction in the compressive stiffness (Hernandez et al., 2014). Furthermore, cortical bone subjected to fatigue loading showed a 15% reduction in stiffness even before microscopical evidence of microcracks (Burr et al., 1998). In the present study, except for two specimens, the µCT images at 4 µm resolution did
Table 2 Mean ± SD of the overall strains, compressive stiffness and normalized hysteresis of the mechanical parameters of bone specimens from distopalmar aspect of MC3 of Thoroughbred racehorses (N = 10) measured via mechanical testing for the entire thickness of bone, SSB and DSB regions. P-values are from a paired t-test between SSB and DSB measurements.
Apparent strain (%) Stiffness (MPa) Normalized hysteresis (%)
Total bone (~10 mm)
SSB (2 mm)
DSB (2 mm)
P-value
0.80 ± 0.26 5608.3 ± 2407.6a 31.8 ± 11.4
1.71 ± 0.88 2446.4 ± 1132.8a 47.4 ± 11.9b
0.60 ± 0.18 6890.4 ± 1741.1a 12.8 ± 8.0
0.001 0.000 0.000
a
One specimen was excluded because it was an outlier with extremely large stiffness. One specimen failed at the SSB region while loading, and thus the post-yield and unloading strain, and thus hysteresis of the specimen in the SSB region was not available from the real-time images. b
54
Journal of the Mechanical Behavior of Biomedical Materials 85 (2018) 51–56
F. Malekipour et al.
Fig. 4. Relationship between normalized hysteresis and the maximum strain absorbed by SSB (A) and DSB (B) areas (n = 9).
dissipation is most likely associated with microdamage (including preexisting microdamage and that induced by the testing procedure) within this region as opposed to the viscoelasticity of the material.
Table 3 Pearson correlation coefficient and p-values from the linear regression analyses between the mechanical and structural properties of SSB and DSB regions for equine third metacarpal subchondral bone (n = 9). SSB
BMD
BVF
DSB
5. Limitations
K
Eloss
K
Eloss
r = 0.571 p = 0.108 r = 0.448 p = 0.226
r = −0.835 p = 0.005 r = −0.588 p = 0.096
r = 0.247 p = 0.521 r = −0.342 p = 0.368
r = −0.146 p = 0.688 r = −0.314 p = 0.376
We included specimens from both lateral and medial condyles. These two regions may experience different loading conditions in the joint; however, we found no significant differences in BMD and BVF between lateral and medial specimens. The unconfined compression boundary conditions varied between SSB and DSB mainly due to the presence of cartilage at the distal end of the bone. We preserved the cartilage intact to keep the integrity of the most superficial bone distally, which was of particular interest in this study. However, for each ROI we analysed only the central 2 mm of the bone core to limit the effect of cartilage lateral deformation on bone compressive deformation. In addition, since ROIs were selected at a 0.5 mm distance from the interface, we do not believe the stiffness gradient was an artefact of the boundary conditions of the mechanical testing setup. Despite a high-resolution, µCT imaging can be limited in detecting pre-existing diffuse microdamage which is common in MC3 specimens from racehorses (Muir et al., 2008). However, it high-resolution µCT imaging is a non-destructive method of assessing bone microcracks that allows subsequent mechanical testing, as opposed to histological analysis. We measured the strain based on in-plane deformation of bone, which may have affected on the absolute values of strains. However, we believe the relative stiffness of SSB and DSB in each specimen was unaffected. Lastly, larger number of specimens may allow the detection of significant correlations between mechanical and structural properties of bone at each ROI.
not reveal any microcracks or microfractures. However, non-mineralised cracks with minimal separation between the crack surfaces, as well as areas of diffuse microdamage in dense bones, similar to those in this study, are difficult to observe using µCT imaging (Laverty et al., 2015; Williamson et al., 2017). Considering that the specimens in this study were collected from racehorses with varying histories of intensive fatigue loading, it is likely that they have accumulated some microdamage during training/racing (Bani Hassan et al., 2016). Thus, it may be possible that the low stiffness of SSB in our specimens were due to pre-existing microdamage in this area. The presence of microdamage in the SSB may also explain the contradictory finding of a significant association between relative stiffness and BMD for the two regions, in that accelerated bone remodelling associated with microdamage over time would result in lower BMD in the SSB. Energy dissipation in materials can be due to either viscoelasticity (rate-dependent) or the presence of microdamage. In this study, the SSB strain-rate was not greater than three times that of the DSB strain-rate, and thus is unlikely to have caused the difference in the stiffness and energy dissipation observed between the two ROIs (McElhaney, 1966). In addition, the SSB energy dissipation was greater in specimens that absorbed a greater strain. Under the applied strain in this study, the absorbed SSB strain remained below the apparent-level failure strain of equine subchondral bone (5% reported by Rubio-Martínez et al., 2008), yet were close to its tissue-level failure strain under impact loading (1.99%, Malekipour et al., 2016). These strains could have generated microcracks at the tissue-level prior to the apparent failure (Nagaraja et al., 2005), and led to a high energy dissipation in SSB. DSB axial strains, however, remained well below the estimated compressive failure threshold of subchondral bone (Fig. 4B, Table 2). At this level of strain, the DSB region may have remained in the elastic region with negligible energy dissipation due to the hysteresis of the mechanical testing system, which may also explain the weak relationship between DSB hysteresis and strain. These findings suggest that the energy
6. Conclusions A gradient of increasing BMD was observed with increasing distance from the tidemark and although this gradient may have contributed to the difference in bone stiffness and energy dissipation between superficial and deeper bone, it did not explain the magnitude of this difference. The lower stiffness of the most superficial subchondral bone in the distal metacarpal condyles may protect the overlying cartilage as well as the underlying subchondral bone from damage under the high-rate compression experienced during galloping. However, repeated highrate loading over time has the potential to inhibit bone turnover and induce fatigue damage, consistent with the high prevalence of subchondral bone injury and fractures at this site in racehorses. Future studies are required to investigate the effect of micro-damage on the mechanical properties of this region. 55
Journal of the Mechanical Behavior of Biomedical Materials 85 (2018) 51–56
F. Malekipour et al.
Acknowledgement
subchondral bone in joint disease: a review. Equine Vet. J. 33, 120–126. Laverty, S., Lacourt, M., Gao, C., Henderson, J.E., Boyde, A., 2015. High density infill in cracks and protrusions from the articular calcified cartilage in osteoarthritis in standardbred horse carpal bones. Int. J. Mol. Sci. 16, 9600–9611. Leahy, P.D., Smith, B.S., Easton, K.L., Kawcak, C.E., Eickhoff, J.C., Shetye, S.S., Puttlitz, C.M., 2010. Correlation of mechanical properties within the equine third metacarpal with trabecular bending and multi-density micro-computed tomography data. Bone 46, 1108–1113. Malekipour, F., Oetomo, D., Lee, P.V.-S., 2016. Equine subchondral bone failure threshold under impact compression applied through articular cartilage. J. Biomech. 49, 2053–2059. Malekipour, F., Whitton, C., Oetomo, D., Lee, P.V.S., 2013. Shock absorbing ability of articular cartilage and subchondral bone under impact compression. J. Mech. Behav. Biomed. Mater. 26, 127–135. Martin, R.B., Boardman, D.L., 1993. The effects of collagen fiber orientation, porosity, density, and mineralization on bovine cortical bone bending properties. J. Biomech. 26, 1047–1054. Matcuk, G.R., Mahanty, S.R., Skalski, M.R., Patel, D.B., White, E.A., Gottsegen, C.J., 2016. Stress fractures: pathophysiology, clinical presentation, imaging features, and treatment options. Emerg. Radiol. 1–11. McElhaney, J.H., 1966. Dynamic response of bone and muscle tissue. J. Appl. Physiol. 21, 1231–1236. Muir, P., McCarthy, J., Radtke, C.L., Markel, M.D., Santschi, E.M., Scollay, M.C., Kalscheur, V.L., 2006. Role of endochondral ossification of articular cartilage and functional adaptation of the subchondral plate in the development of fatigue microcracking of joints. Bone 38, 342–349. Muir, P., Peterson, A.L., Sample, S.J., Scollay, M.C., Markel, M.D., Kalscheur, V.L., 2008. Exercise-induced metacarpophalangeal joint adaptation in the Thoroughbred racehorse. J. Anat. 213, 706–717. Nagaraja, S., Couse, T.L., Guldberg, R.E., 2005. Trabecular bone microdamage and microstructural stresses under uniaxial compression. J. Biomech. 38, 707–716. Norrdin, R.W., Stover, S.M., 2006. Subchondral bone failure in overload arthrosis: a scanning electron microscopic study in horses. J. Musculoskelet. Neuron. Interact. 6, 251–257. Riggs, C.M., Whitehouse, G.H., Boyde, A., 1999. Pathology of the distal condyles of the third metacarpal and third metatarsal bones of the horse. Equine Vet. J. 31, 140–148. Rubio-Martínez, L.M., Cruz, A.M., Gordon, K., Hurtig, M.B., 2008. Mechanical properties of subchondral bone in the distal aspect of third metacarpal bones from Thoroughbred racehorses. Am. J. Vet. Res. 69, 1423–1433. Schaffler, M.B., Burr, D.B., 1988. Stiffness of compact bone: effects of porosity and density. J. Biomech. 21, 13–16. Shirazi, R., Shirazi-Adl, A., 2009. Computational biomechanics of articular cartilage of human knee joint: effect of osteochondral defects. J. Biomech. 42, 2458–2465. Stepnik, M.W., Radtke, C.L., Scollay, M.C., Oshel, P.E., Albrecht, R.M., Santschi, E.M., Markel, M.D., Muir, P., 2004. Scanning electron microscopic examination of third metacarpal/third metatarsal bone failure surfaces in Thoroughbred racehorses with condylar fracture. Vet. Surg. 33, 2–10. Swanstrom, M.D., Zarucco, L., Hubbard, M., Stover, S.M., Hawkins, D.A., 2005. Musculoskeletal modeling and dynamic simulation of the thoroughbred equine forelimb during stance phase of the gallop. J. Biomech. Eng. 127, 318. Whitton, R.C., Trope, G.D., Ghasem-Zadeh, A., Anderson, G.A., Parkin, T.D.H., Mackie, E.J., Seeman, E., 2010. Third metacarpal condylar fatigue fractures in equine athletes occur within previously modelled subchondral bone. Bone 47, 826–831. Williamson, A., Sims, N.A., Thomas, C.D.L., Lee, P.V.S., Stevenson, M., Whitton, R.C., 2017. Biomechanical testing of the calcified metacarpal articular surface and its association with subchondral bone microstructure in Thoroughbred racehorses. Equine Vet. J. 1–6.
This study was funded by Racing Victoria Limited and the Victorian Racing Industry Fund of the Victorian State Government. Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.jmbbm.2018.05.031. References Bani Hassan, E., Mirams, M., Ghasem-Zadeh, A., Mackie, E.J., Whitton, R.C., 2016. Role of subchondral bone remodelling in collapse of the articular surface of Thoroughbred racehorses with palmar osteochondral disease. Equine Vet. J. 48, 228–233. Barr, E.D., Pinchbeck, G.L., Clegg, P.D., Boyde, A., Riggs, C.M., 2009. Post mortem evaluation of palmar osteochondral disease (traumatic osteochondrosis) of the metacarpo/metatarsophalangeal joint in Thoroughbred racehorses. Equine Vet. J. 41, 366–371. Boyde, a., Firth, E.C., 2005. Musculoskeletal responses of 2-year-old Thoroughbred horses to early training. 8. Quantitative back-scattered electron scanning electron microscopy and confocal fluorescence microscopy of the epiphysis of the third metacarpal bone. N. Z. Vet. J. 53, 123–132. Burgin, L.V., Aspden, R.M., 2008. Impact testing to determine the mechanical properties of articular cartilage in isolation and on bone. J. Mater. Sci. Mater. Med. 19, 703–711. Burr, D.B., 2004. The Importance of Subchondral Bone in the Progression of Osteoarthritis 31, 77–80. Burr, D.B., Martin, R.B., Schaffler, M.B., Radin, E.L., 1985. Bone remodeling in response to in vivo fatigue microdamage. J. Biomech. 18, 189–200. Burr, D.B., Turner, C.H., Naick, P., Forwood, M.R., Ambrosius, W., Sayeed Hasan, M., Pidaparti, R., 1998. Does microdamage accumulation affect the mechanical properties of bone? J. Biomech. 31, 337–345. Cardoso, L., Fritton, S.P., Gailani, G., Benalla, M., Cowin, S.C., 2013. Advances in assessment of bone porosity, permeability and interstitial fluid flow. J. Biomech. 46, 253–265. Davies, H.M.S., Mccarthy, R.N., 1994. Surface strain on the dorsomedial metacarpus of yearling Thoroughbreds at different speeds and gaits on a treadmill 17, 25–28. Devas, M.B., 1958. Stress fractures of the tibia in athletes or shin soreness. J. Bone Jt. Surg. Br. 40–B, 227–239. Ding, M., Danielsen, C.C., Hvid, I., 2001. Bone density does not reflect mechanical properties in early-stage arthrosis. Acta Orthop. Scand. 72, 181–185. Drum, M.G., Les, C.M., Park, R.D., Norrdin, R.W., McIlwraith, C.W., Kawcak, C.E., 2009. Correlation of quantitative computed tomographic subchondral bone density and ash density in horses. Bone 44, 316–319. Harrison, S.M., Whitton, R.C., Kawcak, C.E., Stover, S.M., Pandy, M.G., 2014. Evaluation of a subject-specific finite-element model of the equine metacarpophalangeal joint under physiological load. J. Biomech. 47, 65–73. Hernandez, C.J., Lambers, F.M., Widjaja, J., Chapa, C., Rimnac, C.M., 2014. Quantitative relationships between microdamage and cancellous bone strength and stiffness. Bone 66, 205–213. Jones, E.M.C., 2013. Improved Digital Image Correlation (DIC) [WWW Document]. URL 〈http://www.mathworks.com/matlabcentral/fileexchange/43073-improved-digitalimage-correlation–dic-〉. Kawcak, C.E., McIlwraith, C.W., Norrdin, R.W., Park, R.D., James, S.P., 2001. The role of
56
Contents lists available at ScienceDirect
Journal of the Mechanical Behavior of Biomedical Materials journal homepage: www.elsevier.com/locate/jmbbm
Stiffness and energy dissipation across the superficial and deeper third metacarpal subchondral bone in Thoroughbred racehorses under high-rate compression
T
⁎
Fatemeh Malekipoura, , Chris R. Whittonb, Peter Vee-Sin Leea a b
Department of Biomedical Engineering, University of Melbourne, Parkville, VIC 3010, Australia Equine Centre, Faculty of Veterinary and Agricultural Sciences, University of Melbourne, Werribee, VIC 3030, Australia
A B S T R A C T
Subchondral bone injury due to high magnitude and repetition of compressive loading is common in humans and athletic animals such as Thoroughbred racehorses. Repeated loading of the joint surface may alter the subchondral bone microstructure and initiate microdamage in the bone adjacent to the articular cartilage. Understanding the relationship between microdamage, microstructure and mechanical properties of the subchondral bone adjacent to the articular cartilage is, therefore, essential in understanding the mechanism of subchondral bone injury. In this study, we used high-resolution µCT scanning, a digital image-based strain measurement technique, and mechanical testing to evaluate the three-dimensional pre-existing microcracks, bone volume fraction (BVF) and bone mineral density (BMD), and mechanical properties (stiffness and hysteresis) of subchondral bone (n = 10) from the distopalmar aspect of the third metacarpal (MC3) condyles of Thoroughbred racehorses under high-rate compression. We specifically compared the properties of two regions of interest in the subchondral bone: the 2 mm superficial subchondral bone (SSB) and its underlying 2 mm deep subchondral bone (DSB). The DSB region was 3.0 ± 1.2 times stiffer than its overlying SSB, yet it dissipated much less energy compared to the SSB. There was no correlation between structural properties (BVF and BMD) and mechanical properties (stiffness and energy loss), except for BMD and energy loss in SSB. The lower stiffness of the most superficial subchondral bone in the distal metacarpal condyles may protect the overlying cartilage and the underlying subchondral bone from damage under the high-rate compression experienced during galloping. However, repeated high-rate loading over time has the potential to inhibit bone turnover and induce bone fatigue, consistent with the high prevalence of subchondral bone injury and fractures in athletic humans and racehorses.
1. Introduction Subchondral bone injury due to high magnitude and repetition of compressive loading is common in humans (Devas, 1958; Matcuk et al., 2016) and athletic animals such as Thoroughbred racehorses (Barr et al., 2009; Riggs et al., 1999). Repeated loading of the joint surface may alter the subchondral bone microstructure and initiate microdamage in the bone adjacent to the articular cartilage (Muir et al., 2008; Stepnik et al., 2004). Microdamage may stimulate both modelling and remodelling responses in subchondral bone and can lead to the development of stress fractures and/or osteoarthritis in the joint (Burr et al., 1985; Kawcak et al., 2001; Muir et al., 2006; Norrdin and Stover, 2006). Specifically, altered subchondral bone mechanics due to microdamage can influence how joint surface load is transferred to the
⁎
Corresponding author. E-mail address: [email protected] (F. Malekipour).
https://doi.org/10.1016/j.jmbbm.2018.05.031 Received 29 August 2017; Received in revised form 16 May 2018; Accepted 21 May 2018 Available online 22 May 2018 1751-6161/ © 2018 Elsevier Ltd. All rights reserved.
underlying trabecular bone, and thus trigger changes in bone modelling and remodelling which are highly load dependent. (Burr, 2004; Shirazi and Shirazi-Adl, 2009; Whitton et al., 2010). Understanding the relationship between microdamage, microstructure and mechanical properties of the subchondral bone adjacent to the articular cartilage is, therefore, essential in understanding the mechanism of microdamage initiation and propagation across the whole joint. The metacarpophalangeal joint in racehorses is subjected to extremely high compressive loads during galloping (Harrison et al., 2014). The palmar metacarpal subchondral bone, particularly the bone within 2–3 mm of the cartilage-calcified cartilage interface is a common site of fatigue-induced microdamage (Muir et al., 2006), which highlights the importance of investigation of subchondral bone mechanical/ structural properties at this site. Previous studies investigated the
Journal of the Mechanical Behavior of Biomedical Materials 85 (2018) 51–56
F. Malekipour et al.
2.2. Micro-CT scanning and bone microarchitecture
relationship between the microstructure and mechanical properties of equine subchondral bone from the metacarpal condyles under low-rate compression (Rubio-Martínez et al., 2008; Leahy et al., 2010). RubioMartínez et al. (2008) observed a gradient of stiffness across the subchondral bone depth; superficial 6 mm cubes of dense subchondral bone were less stiff than deeper 6 mm cubes of trabecular bone. Additionally, variations in the mechanical properties of the trabecular bone specimens from condylar regions were associated with structural measures derived from micro-computer tomography (µCT) but similar associations were not observed for the more superficial subchondral bone specimens. In the present study, we focused on the subchondral bone area closest to the joint surface, i.e. subchondral bone plate in RubioMartínez et al. (2008), to increase the resolution of the previous measurements by investigating two regions of interest (ROI) within the equine third metacarpal (MC3) subchondral bone: the superficial subchondral bone (SSB: 0.5–2.5 mm) and its underlying deep subchondral bone (DSB: 2.5–4.5 mm). We used high-resolution µCT and a digital image-based strain measurement system to evaluate the relative microstructure and mechanical properties (stiffness and energy absorption) of the two ROIS under high-rate compression. The image-based strain measurement technique allows for the investigation of tissue strain while avoiding artefacts associated with the free trabecular end of specimens, and their non-uniform thickness (Malekipour et al., 2013). The specific objectives of this study were to: 1) investigate the microstructure of the subchondral bone and identify pre-existing microcracks using µCT scanning, 2) determine the relative stiffness and shock-absorbing abilities (normalized hysteresis) of the SSB and DSB under experimental high-rate compression similar to one cycle of compression during galloping, and 3) investigate the relationship between the mechanical/structural properties of SSB and DSB. We hypothesized that the superficial 2 mm of subchondral bone would be less stiff, and dissipate more energy under high rate loading than the deeper 2 mm of bone, with associated differences in density and bone volume fraction gradient.
Prior to loading, cartilage-bone plugs were scanned using a µCT scanner (µCT50; Scanco Medical, Switzerland) at an isotropic resolution of 4 µm using 70 kVp tube voltage, 200 µA tube current, and 1050 ms integration time. A phantom was used to convert CT grayscale intensity values (Hounsfield units, HU) into equivalent bone mineral density (BMD, g/cm3 HA). A strong correlation has been demonstrated between quantitative CT and mineralisation and ash density in bone from the equine distal metacarpus (Drum et al., 2009). We used a voxel size of 4 µm, which was the highest resolution of the scanner that achieved good quality images for our specimens, to identify pre-existing microdamage in the subchondral bone. Following the examination for microdamage (ImageJ, 1.48 v, National Institute of Health, USA), images were down-sampled to 40 µm for further morphology processing (Simpleware, v5.1, UK). Damage was identified by visual examination of all µCT slices assessing for linear lucencies or linear areas of increased mineralisation (Williamson et al., 2017). Examination of profile lines and images confirmed that this down-sampling affected the density and bone volume fraction (BVF) measurements no more than 0.06% and 0.39%, respectively. Mineralised bone was segmented using a consistent set of threshold grey values. SSB and DSB regions corresponding to the experimental ROIs were identified based on their distances from the cartilage-bone interface and thicknesses in µCT images. BVF of each ROI and the entire bone was calculated by dividing the volume of bone tissue in that region by its apparent volume. BMD of each ROI and the entire bone were specified as the average BMD over that region. 2.3. Mechanical testing The experimental procedure was similar to that of a previous study (Malekipour et al., 2013). Cartilage-bone plugs were glued to a base plate using a thin layer (several micrometres) of cyanoacrylate glue on the proximal end of the bone (Burgin and Aspden, 2008). Vertical displacements were applied to the articular surface of cartilage via a stainless steel compression plate on a mechanical testing machine (Instron, 8874, UK). A time to peak strain of 0.05 s was chosen, similar to the time to peak bone strain during galloping of a racehorse (Davies and Mccarthy, 1994; Swanstrom et al., 2005). Displacements were applied to the cartilage surface using half-sine waveforms with magnitudes of 0.225 ± 0.018 mm at 5 Hz. The displacement magnitudes were calculated to apply 50–60% of equine subchondral bone yield strain (Rubio-Martínez et al., 2008). The stained flat surface of cartilage-bone plugs faced a microscope lens. A high-speed camera (MotionProY3, 1280 × 1024, USA) attached to a stereomicroscope (SZX7, Olympus, Japan) recorded the experimental deformation of each specimen at 1000 frames per second. Microscopic magnification was chosen that allowed imaging at a resolution of 170 pixels/mm (1024 pixels within 6 mm) in the vertical direction. Recorded images of each specimen were processed using a Matlab-based digital image correlation (DIC) code (Jones, 2013) to calculate the average displacement of bone at three different levels (Lines 1–3, Fig. 1B). A respective subset and grid size of 71 and 20 pixels was found to accurately calculate the displacement within ROIs. Using these parameters, the software measured the displacement of the Instron actuator with an accuracy of 1.89%. The ROIs were selected at a 0.5 mm distance from the cartilage-bone interface. Above this distance cartilage folding under loading hindered clear images, and an accurate displacement measurement of the bone. To remain unaffected by the cutting process and the presence of free edges, ROIs were specified away from the edges. In a preliminary study, the width of ROIs were reduced from the full width until the calculated stiffness and energy dissipation varied no more than 1.41 ± 1.35% and 1.18 ± 0.72%, respectively in the central 2 mm (Fig. 1B). The displacement data were then processed (MATLAB R2016a, MathWorks, USA) to calculate the overall strains of the two ROIs, i.e. SSB and DSB,
2. Methods 2.1. Specimen preparation Cartilage-bone plugs (Ø = 6.5 ± 0.2 mm) were extracted from the distopalmar aspect of fresh frozen (−20 °C) third metacarpal condyles of n = 10 racehorses using a diamond core drill (Starlite Industries, Rosemont, PA) under continuous irrigation. Metacarpal condyles were collected from Thoroughbred racehorses of median age of 3.5 years (range, 3–7 years) that died or were euthanatized on racetracks in Victoria, Australia between May 2011 and March 2014. Specimens were harvested from both medial (n = 7) and lateral (n = 3) condyles. During harvesting, the axis of the drill was oriented perpendicular to the cartilage surface. Bone was trimmed down on the proximal end of each specimen by using a self-irrigated diamond saw (Isomet, Buehler, Ltd., Lake Bluff, IL) to create a flat bone surface perpendicular to the longitudinal axis of the cartilage-bone plugs. The final trimmed bone thickness was 9.68 ± 0.79 mm with an average cartilage thickness of 0.46 ± 0.06 mm. To keep the integrity of the subchondral plate adjacent to the articular cartilage, we preserved the cartilage in situ. Subsequently, the cartilage-bone plugs were trimmed perpendicular to the cartilage surface to create a flat surface on the longitudinal side of each specimen (Fig. 1A). Prior to loading, this longitudinal surface was stained with fine graphite particles to create a random speckle pattern, which was necessary for the later image processing to calculate bone strains. The graphite particles were generated by rubbing pencil lead with sand paper, a technique which has been validated in a previous study (Malekipour et al., 2013).
52
Journal of the Mechanical Behavior of Biomedical Materials 85 (2018) 51–56
F. Malekipour et al.
Fig. 1. Schematic illustrations of (A) the topview of a cartilage-bone specimen indicating 1 mm trimmed, which was cut using a diamond saw to generate a flat surface for the staining, (B) the mechanical testing set-up indicating the side-view of a cartilage-bone plug, SSB and DSB regions of interest, the compression plate, and the microscope attached to a high-speed camera.
3.2. Mechanical testing
by dividing the relative displacements of the upper and lower bounds to the thickness of each ROI (2 mm). Overall compressive axial stress was calculated by dividing the compressive load measured on the Instron load cell ( ± 10 kN) by the horizontal cross-sectional area of each specimen. Stress data (Instron output) were synchronized to the strain data (calculated from image processing) by matching the time that the compression plate touched the articular surface, and stress-strain curves of each ROI for each specimen were determined. The slope of the linear portion (R2 > 0.96) of the loading stress-strain curves were calculated and specified as the experimental stiffness of SSB (KISB ) and DSB (KDSB ). The areas enclosed by each stress-strain curve (hysteresis) were divided by the area beneath the loading curve (absorbed energy) to calculate the shock-absorbing abilities (normalized hysteresis) of each ROI.
Applied overall strains of 2.33 ± 0.25% (mean ± SD) generated axial compressive stresses of 44.2 ± 10.7 MPa to cartilage-bone plugs. The two specimens with pre-existing microfractures failed during the compression. In one of these specimens, the induced microfracture remained distal to the SSB region. In the second one, the pre-existing microfracture propagated proximally into the SSB area, causing blurring of real-time images. Therefore, no post-yield and unloading strain data were available for this specimen. In all specimens, the DSB region was stiffer than its overlying SSB with an average DSB-to-SSB stiffness ratio of 3.0 ± 1.2 (Table 2). The normalized hysteresis of SSB was greater than that of DSB by a factor of 3.1 ± 0.8 (Table 2). Fig. 3 illustrates typical stress-strain curves associated with the SSB and DSB. Specimens that absorbed a higher strain in the SSB region under the applied overall strain exhibited greater energy dissipation at this region (Fig. 4A). Unlike the SSB, the DSB energy dissipation did not correlate with its maximum absorbed strain (Fig. 4B).
2.4. Statistical analysis A Grubbs' Test was applied to exclude the outliers, resulting in exclusion of one values of DSB stiffness. Paired t-tests were performed to compare the properties between SSB and DSB; where data was not normally distributed, Wilcoxon signed rank tests were used. Pearson's correlation coefficients were calculated to determine relationships between mechanical parameters. The statistical differences in BVF and BMD between lateral and medial condyles were also investigated using one-way ANOVA. The level of significance was set at p = 0.05. All statistical analyses were performed in Minitab (MINITAB Inc., version 17.0).
3.3. Relationship between bone microarchitecture, density and mechanical properties No significant correlations were detected between the structural (BVF , BMD ) and mechanical properties (stiffness and energy loss) of bone within either the SSB or the DSB (Table 3), except for the SSB energy loss which showed a strong correlation with its BMD (r = −0.83, p = 0.005). The relative stiffness ( KDSB ) also correlated K SSB
with the relative density ( BMDDSB ) (r = 0.63, p = 0.05), but not with the
3. Results
BMDSSB
relative BVF ( BVFDSB ) (r = 0.20, p = 0.58). BVFSSB
3.1. Micro-CT image analysis
4. Discussion
Based on the µCT images performed prior to mechanical testing, two specimens exhibited pre-existing microfractures in the subchondral bone immediately beneath the articular cartilage. Microfractures in one specimen appeared as irregular linear transverse lucencies 0.5 mm below the calcified cartilage/hyaline cartilage interface with an approximate width of 26 µm (Fig. 2B). In the same specimen, a site of hypermineralized material was observed projecting into the overlying uncalcified cartilage. In the other damaged specimen, a fine microcrack was observed just beneath the articular cartilage. The remaining specimens did not reveal any pre-existing microcracks in the µCT images (Fig. 2A). All specimens exhibited dense structure with a BVF > 0.9. There were no significant differences in BMD and BVF between lateral and medial specimens (p > 0.05). The SSB region had equivalent BVF, yet lower BMD when compared to its underlying DSB (Table 1).
We have shown that under loading rates comparable to those generated in the fetlock joint of a galloping horse, the dense subchondral bone of the distal palmar metacarpus is less stiff and able to dissipate more energy in its most superficial 2 mm (SSB) compared to the deeper 2 mm of subchondral bone (DSB), confirming our hypothesis. However, contrary to our hypothesis, we found a weak and insignificant association between the structural and mechanical properties of bone, except for the energy dissipation and BMD of the SSB region. In contrast, a significant correlation was revealed when the relative values of stiffness and BMD were compared. The stiffness of the entire thickness of bone specimens in our study (10 mm) was higher than that of the equivalent subchondral bone area reported previously (Rubio-Martínez et al., 2008). The difference can be attributed to the strain measurement technique we used here that 53
Journal of the Mechanical Behavior of Biomedical Materials 85 (2018) 51–56
F. Malekipour et al.
Fig. 2. Representative sagittal cross-sections µCT images of cartilage-bone plugs at 4 µm resolution (A) without any evidence of microfracture, and (B) with pre-existing microdamage including a microfracture ~0.5 mm beneath the tidemark with an approximate width of 26 µm (arrow), and a focal area of hypermineralisation (inside the circle) projecting into the unmineralised cartilage.
Table 1 Mean ± SD of bone volume fraction (BVF), and bone tissue mineral density (BMD) determined via µCT image analysis. P-values calculated from a paired ttest between SSB and DSB regions.
BV/TV BMD (mg/ cm3 HA)
Total bone (~10 mm)
SSB (2 mm)
DSB (2 mm)
P-value
0.936 ± 0.036 883.4 ± 15.4
0.984 ± 0.012 852.9 ± 18.7
0.970 ± 0.018 890.7 ± 13.6
0.069 0.000
The detailed results for n = 10 specimens are provided in Supplementary Materials Table S1.
eliminated end-cap artefacts. Moreover, Rubio-Martínez et al. (2008) used samples from horses with mild and severe subchondral bone changes, whereas in the current study all specimens were from horses with mild subchondral bone disease. The subchondral bone of the distal metacarpus of equine athletes experiences high magnitude cyclic loading when horses are in race training, resulting in densification of the bone when compared to the subchondral bone of young or non-race horses (Boyde and Firth, 2005). The BVF of the subchondral bone from this area is in a range similar to that of cortical bone (Cardoso et al., 2013). However, unlike cortical bone compressive stiffness which was predicted by its porosity (Schaffler and Burr, 1988), the subchondral bone stiffness in our study did not show any relationship with its porosity (1-BVF). Furthermore, differences in microstructure such as the orientation of collagen fibres in the SSB and DSB regions may play a role in differences in stiffness of the SSB and DSB. Martin and Boardman (1993) showed that collagen fibre orientation was the best predictor of bending properties of bovine cortical bone. Future studies are required to evaluate such effects on the equine subchondral bone. Similar to what has previously been observed in equine metacarpal subchondral bone from Thoroughbred racehorses to a depth of 6 mm, we were unable to identify any significant relationships between mechanical and structural properties despite a higher resolution of µCT imaging and a more accurate strain measurement technique (Rubio-
Fig. 3. Representative experimental stress-strain curves associated with SSB and DSB regions of a typical cartilage-bone specimen.
Martínez et al., 2008). Similarly, Ding et al. (2001) found that human subchondral bone apparent density failed to explain the reduced mechanical strength of the subchondral bone at early stages of OA, whereas the mechanical strength of healthy subchondral bone did correlate with its apparent bone mineral density. Such a lack of correlation between the subchondral bone structural and mechanical factors may be due to a differing microstructure such as the orientation of collagen fibres in the SSB and DSB, or deteriorated bone quality at the tissue-level, and/or the accumulation of microcracks due to fatigue loading. In addition to the bone microstructure, microdamage has been shown to have substantial effects on material properties of bone. For example, in trabecular bone, a damage volume fraction (damage volume/bone volume) of 1.5% caused a 50% reduction in the compressive stiffness (Hernandez et al., 2014). Furthermore, cortical bone subjected to fatigue loading showed a 15% reduction in stiffness even before microscopical evidence of microcracks (Burr et al., 1998). In the present study, except for two specimens, the µCT images at 4 µm resolution did
Table 2 Mean ± SD of the overall strains, compressive stiffness and normalized hysteresis of the mechanical parameters of bone specimens from distopalmar aspect of MC3 of Thoroughbred racehorses (N = 10) measured via mechanical testing for the entire thickness of bone, SSB and DSB regions. P-values are from a paired t-test between SSB and DSB measurements.
Apparent strain (%) Stiffness (MPa) Normalized hysteresis (%)
Total bone (~10 mm)
SSB (2 mm)
DSB (2 mm)
P-value
0.80 ± 0.26 5608.3 ± 2407.6a 31.8 ± 11.4
1.71 ± 0.88 2446.4 ± 1132.8a 47.4 ± 11.9b
0.60 ± 0.18 6890.4 ± 1741.1a 12.8 ± 8.0
0.001 0.000 0.000
a
One specimen was excluded because it was an outlier with extremely large stiffness. One specimen failed at the SSB region while loading, and thus the post-yield and unloading strain, and thus hysteresis of the specimen in the SSB region was not available from the real-time images. b
54
Journal of the Mechanical Behavior of Biomedical Materials 85 (2018) 51–56
F. Malekipour et al.
Fig. 4. Relationship between normalized hysteresis and the maximum strain absorbed by SSB (A) and DSB (B) areas (n = 9).
dissipation is most likely associated with microdamage (including preexisting microdamage and that induced by the testing procedure) within this region as opposed to the viscoelasticity of the material.
Table 3 Pearson correlation coefficient and p-values from the linear regression analyses between the mechanical and structural properties of SSB and DSB regions for equine third metacarpal subchondral bone (n = 9). SSB
BMD
BVF
DSB
5. Limitations
K
Eloss
K
Eloss
r = 0.571 p = 0.108 r = 0.448 p = 0.226
r = −0.835 p = 0.005 r = −0.588 p = 0.096
r = 0.247 p = 0.521 r = −0.342 p = 0.368
r = −0.146 p = 0.688 r = −0.314 p = 0.376
We included specimens from both lateral and medial condyles. These two regions may experience different loading conditions in the joint; however, we found no significant differences in BMD and BVF between lateral and medial specimens. The unconfined compression boundary conditions varied between SSB and DSB mainly due to the presence of cartilage at the distal end of the bone. We preserved the cartilage intact to keep the integrity of the most superficial bone distally, which was of particular interest in this study. However, for each ROI we analysed only the central 2 mm of the bone core to limit the effect of cartilage lateral deformation on bone compressive deformation. In addition, since ROIs were selected at a 0.5 mm distance from the interface, we do not believe the stiffness gradient was an artefact of the boundary conditions of the mechanical testing setup. Despite a high-resolution, µCT imaging can be limited in detecting pre-existing diffuse microdamage which is common in MC3 specimens from racehorses (Muir et al., 2008). However, it high-resolution µCT imaging is a non-destructive method of assessing bone microcracks that allows subsequent mechanical testing, as opposed to histological analysis. We measured the strain based on in-plane deformation of bone, which may have affected on the absolute values of strains. However, we believe the relative stiffness of SSB and DSB in each specimen was unaffected. Lastly, larger number of specimens may allow the detection of significant correlations between mechanical and structural properties of bone at each ROI.
not reveal any microcracks or microfractures. However, non-mineralised cracks with minimal separation between the crack surfaces, as well as areas of diffuse microdamage in dense bones, similar to those in this study, are difficult to observe using µCT imaging (Laverty et al., 2015; Williamson et al., 2017). Considering that the specimens in this study were collected from racehorses with varying histories of intensive fatigue loading, it is likely that they have accumulated some microdamage during training/racing (Bani Hassan et al., 2016). Thus, it may be possible that the low stiffness of SSB in our specimens were due to pre-existing microdamage in this area. The presence of microdamage in the SSB may also explain the contradictory finding of a significant association between relative stiffness and BMD for the two regions, in that accelerated bone remodelling associated with microdamage over time would result in lower BMD in the SSB. Energy dissipation in materials can be due to either viscoelasticity (rate-dependent) or the presence of microdamage. In this study, the SSB strain-rate was not greater than three times that of the DSB strain-rate, and thus is unlikely to have caused the difference in the stiffness and energy dissipation observed between the two ROIs (McElhaney, 1966). In addition, the SSB energy dissipation was greater in specimens that absorbed a greater strain. Under the applied strain in this study, the absorbed SSB strain remained below the apparent-level failure strain of equine subchondral bone (5% reported by Rubio-Martínez et al., 2008), yet were close to its tissue-level failure strain under impact loading (1.99%, Malekipour et al., 2016). These strains could have generated microcracks at the tissue-level prior to the apparent failure (Nagaraja et al., 2005), and led to a high energy dissipation in SSB. DSB axial strains, however, remained well below the estimated compressive failure threshold of subchondral bone (Fig. 4B, Table 2). At this level of strain, the DSB region may have remained in the elastic region with negligible energy dissipation due to the hysteresis of the mechanical testing system, which may also explain the weak relationship between DSB hysteresis and strain. These findings suggest that the energy
6. Conclusions A gradient of increasing BMD was observed with increasing distance from the tidemark and although this gradient may have contributed to the difference in bone stiffness and energy dissipation between superficial and deeper bone, it did not explain the magnitude of this difference. The lower stiffness of the most superficial subchondral bone in the distal metacarpal condyles may protect the overlying cartilage as well as the underlying subchondral bone from damage under the high-rate compression experienced during galloping. However, repeated highrate loading over time has the potential to inhibit bone turnover and induce fatigue damage, consistent with the high prevalence of subchondral bone injury and fractures at this site in racehorses. Future studies are required to investigate the effect of micro-damage on the mechanical properties of this region. 55
Journal of the Mechanical Behavior of Biomedical Materials 85 (2018) 51–56
F. Malekipour et al.
Acknowledgement
subchondral bone in joint disease: a review. Equine Vet. J. 33, 120–126. Laverty, S., Lacourt, M., Gao, C., Henderson, J.E., Boyde, A., 2015. High density infill in cracks and protrusions from the articular calcified cartilage in osteoarthritis in standardbred horse carpal bones. Int. J. Mol. Sci. 16, 9600–9611. Leahy, P.D., Smith, B.S., Easton, K.L., Kawcak, C.E., Eickhoff, J.C., Shetye, S.S., Puttlitz, C.M., 2010. Correlation of mechanical properties within the equine third metacarpal with trabecular bending and multi-density micro-computed tomography data. Bone 46, 1108–1113. Malekipour, F., Oetomo, D., Lee, P.V.-S., 2016. Equine subchondral bone failure threshold under impact compression applied through articular cartilage. J. Biomech. 49, 2053–2059. Malekipour, F., Whitton, C., Oetomo, D., Lee, P.V.S., 2013. Shock absorbing ability of articular cartilage and subchondral bone under impact compression. J. Mech. Behav. Biomed. Mater. 26, 127–135. Martin, R.B., Boardman, D.L., 1993. The effects of collagen fiber orientation, porosity, density, and mineralization on bovine cortical bone bending properties. J. Biomech. 26, 1047–1054. Matcuk, G.R., Mahanty, S.R., Skalski, M.R., Patel, D.B., White, E.A., Gottsegen, C.J., 2016. Stress fractures: pathophysiology, clinical presentation, imaging features, and treatment options. Emerg. Radiol. 1–11. McElhaney, J.H., 1966. Dynamic response of bone and muscle tissue. J. Appl. Physiol. 21, 1231–1236. Muir, P., McCarthy, J., Radtke, C.L., Markel, M.D., Santschi, E.M., Scollay, M.C., Kalscheur, V.L., 2006. Role of endochondral ossification of articular cartilage and functional adaptation of the subchondral plate in the development of fatigue microcracking of joints. Bone 38, 342–349. Muir, P., Peterson, A.L., Sample, S.J., Scollay, M.C., Markel, M.D., Kalscheur, V.L., 2008. Exercise-induced metacarpophalangeal joint adaptation in the Thoroughbred racehorse. J. Anat. 213, 706–717. Nagaraja, S., Couse, T.L., Guldberg, R.E., 2005. Trabecular bone microdamage and microstructural stresses under uniaxial compression. J. Biomech. 38, 707–716. Norrdin, R.W., Stover, S.M., 2006. Subchondral bone failure in overload arthrosis: a scanning electron microscopic study in horses. J. Musculoskelet. Neuron. Interact. 6, 251–257. Riggs, C.M., Whitehouse, G.H., Boyde, A., 1999. Pathology of the distal condyles of the third metacarpal and third metatarsal bones of the horse. Equine Vet. J. 31, 140–148. Rubio-Martínez, L.M., Cruz, A.M., Gordon, K., Hurtig, M.B., 2008. Mechanical properties of subchondral bone in the distal aspect of third metacarpal bones from Thoroughbred racehorses. Am. J. Vet. Res. 69, 1423–1433. Schaffler, M.B., Burr, D.B., 1988. Stiffness of compact bone: effects of porosity and density. J. Biomech. 21, 13–16. Shirazi, R., Shirazi-Adl, A., 2009. Computational biomechanics of articular cartilage of human knee joint: effect of osteochondral defects. J. Biomech. 42, 2458–2465. Stepnik, M.W., Radtke, C.L., Scollay, M.C., Oshel, P.E., Albrecht, R.M., Santschi, E.M., Markel, M.D., Muir, P., 2004. Scanning electron microscopic examination of third metacarpal/third metatarsal bone failure surfaces in Thoroughbred racehorses with condylar fracture. Vet. Surg. 33, 2–10. Swanstrom, M.D., Zarucco, L., Hubbard, M., Stover, S.M., Hawkins, D.A., 2005. Musculoskeletal modeling and dynamic simulation of the thoroughbred equine forelimb during stance phase of the gallop. J. Biomech. Eng. 127, 318. Whitton, R.C., Trope, G.D., Ghasem-Zadeh, A., Anderson, G.A., Parkin, T.D.H., Mackie, E.J., Seeman, E., 2010. Third metacarpal condylar fatigue fractures in equine athletes occur within previously modelled subchondral bone. Bone 47, 826–831. Williamson, A., Sims, N.A., Thomas, C.D.L., Lee, P.V.S., Stevenson, M., Whitton, R.C., 2017. Biomechanical testing of the calcified metacarpal articular surface and its association with subchondral bone microstructure in Thoroughbred racehorses. Equine Vet. J. 1–6.
This study was funded by Racing Victoria Limited and the Victorian Racing Industry Fund of the Victorian State Government. Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.jmbbm.2018.05.031. References Bani Hassan, E., Mirams, M., Ghasem-Zadeh, A., Mackie, E.J., Whitton, R.C., 2016. Role of subchondral bone remodelling in collapse of the articular surface of Thoroughbred racehorses with palmar osteochondral disease. Equine Vet. J. 48, 228–233. Barr, E.D., Pinchbeck, G.L., Clegg, P.D., Boyde, A., Riggs, C.M., 2009. Post mortem evaluation of palmar osteochondral disease (traumatic osteochondrosis) of the metacarpo/metatarsophalangeal joint in Thoroughbred racehorses. Equine Vet. J. 41, 366–371. Boyde, a., Firth, E.C., 2005. Musculoskeletal responses of 2-year-old Thoroughbred horses to early training. 8. Quantitative back-scattered electron scanning electron microscopy and confocal fluorescence microscopy of the epiphysis of the third metacarpal bone. N. Z. Vet. J. 53, 123–132. Burgin, L.V., Aspden, R.M., 2008. Impact testing to determine the mechanical properties of articular cartilage in isolation and on bone. J. Mater. Sci. Mater. Med. 19, 703–711. Burr, D.B., 2004. The Importance of Subchondral Bone in the Progression of Osteoarthritis 31, 77–80. Burr, D.B., Martin, R.B., Schaffler, M.B., Radin, E.L., 1985. Bone remodeling in response to in vivo fatigue microdamage. J. Biomech. 18, 189–200. Burr, D.B., Turner, C.H., Naick, P., Forwood, M.R., Ambrosius, W., Sayeed Hasan, M., Pidaparti, R., 1998. Does microdamage accumulation affect the mechanical properties of bone? J. Biomech. 31, 337–345. Cardoso, L., Fritton, S.P., Gailani, G., Benalla, M., Cowin, S.C., 2013. Advances in assessment of bone porosity, permeability and interstitial fluid flow. J. Biomech. 46, 253–265. Davies, H.M.S., Mccarthy, R.N., 1994. Surface strain on the dorsomedial metacarpus of yearling Thoroughbreds at different speeds and gaits on a treadmill 17, 25–28. Devas, M.B., 1958. Stress fractures of the tibia in athletes or shin soreness. J. Bone Jt. Surg. Br. 40–B, 227–239. Ding, M., Danielsen, C.C., Hvid, I., 2001. Bone density does not reflect mechanical properties in early-stage arthrosis. Acta Orthop. Scand. 72, 181–185. Drum, M.G., Les, C.M., Park, R.D., Norrdin, R.W., McIlwraith, C.W., Kawcak, C.E., 2009. Correlation of quantitative computed tomographic subchondral bone density and ash density in horses. Bone 44, 316–319. Harrison, S.M., Whitton, R.C., Kawcak, C.E., Stover, S.M., Pandy, M.G., 2014. Evaluation of a subject-specific finite-element model of the equine metacarpophalangeal joint under physiological load. J. Biomech. 47, 65–73. Hernandez, C.J., Lambers, F.M., Widjaja, J., Chapa, C., Rimnac, C.M., 2014. Quantitative relationships between microdamage and cancellous bone strength and stiffness. Bone 66, 205–213. Jones, E.M.C., 2013. Improved Digital Image Correlation (DIC) [WWW Document]. URL 〈http://www.mathworks.com/matlabcentral/fileexchange/43073-improved-digitalimage-correlation–dic-〉. Kawcak, C.E., McIlwraith, C.W., Norrdin, R.W., Park, R.D., James, S.P., 2001. The role of
56