Developing a novel functional disc emulator to investigate the nucleus pulposus replacement.

We have developed a simple, inexpensive and innovative device for reproducing the global mechanical behavior of spinal motion segments and the local mechanical environment experienced by lumbar intervertebral discs. The device has several broad functions: (1) exploration of the basic mechanics underlying this complex skeletal system, (2) connecting changes in tissue characteristics with overall motion segment function, and (3) evaluation of strategies for repair and replacement of disc components. This "disc emulator" consists of three main parts: (1) an artificial annulus fibrosus (AAF), made out of silicone, with lumbar disc geometry and adjustable material properties, (2) a hydrogel nucleus pulposus (NP) also with lumbar disc geometry and adjustable material properties, and (3) simulated vertebral bodies 3D printed with trabecular bone simulated by a rigid polymer (Acrylonitrile Butadiene Styrene, ABS) and end plates crafted from a compliant polymer (Thermoplastic Polyurethane, TPU). Mechanical compression experiments have been conducted using the disc emulator under similar protocols to published studies of human cadaver samples. Bulging of the artificial annulus fibrosus was examined under axial compression loads using digital image correlation (DIC), and results show close agreement. We see this approach of using anatomical geometry and multiple adjustable components as a useful means of creating accurate local stress/strain environments for preliminary material evaluation, without the variability and difficulty inherent indirect testing of cadaveric materials.

In situ characterization of nanoscale strains in loaded whole joints via synchrotron X-ray tomography.

Imaging techniques for quantifying changes in the hierarchical structure of deforming joints are constrained by destructive sample treatments, sample-size restrictions and lengthy scan times. Here, we report the use of fast low-dose pink-beam synchrotron X-ray tomography in combination with mechanical loading at nanometric precision for in situ imaging, at resolutions below 100 nm, of the mechanical strain in intact untreated joints under physiologically realistic conditions. We show that in young, older and osteoarthritic mice, hierarchical changes in tissue structure and mechanical behaviour can be simultaneously visualized, and that the tissue structure at the cellular level correlates with the mechanical performance of the whole joint. We also use the tomographic approach to study the colocalization of tissue strains to specific chondrocyte lacunar organizations within intact loaded joints and to explore the role of calcified-cartilage stiffness on the biomechanics of healthy and pathological joints.

Quantifying Bulk Electrode Strain and Material Displacement within Lithium Batteries via High-Speed Operando Tomography and Digital Volume Correlation.

Tracking the dynamic morphology of active materials during operation of lithium batteries is essential for identifying causes of performance loss. Digital volume correlation (DVC) is applied to high-speed operando synchrotron X-ray computed tomography of a commercial Li/MnO2 primary battery during discharge. Real-time electrode material displacement is captured in 3D allowing degradation mechanisms such as delamination of the electrode from the current collector and electrode crack formation to be identified. Continuum DVC of consecutive images during discharge is used to quantify local displacements and strains in 3D throughout discharge, facilitating tracking of the progression of swelling due to lithiation within the electrode material in a commercial, spiral-wound battery during normal operation. Displacement of the rigid current collector and cell materials contribute to severe electrode detachment and crack formation during discharge, which is monitored by a separate DVC approach. Use of time-lapse X-ray computed tomography coupled with DVC is thus demonstrated as an effective diagnostic technique to identify causes of performance loss within commercial lithium batteries; this novel approach is expected to guide the development of more effective commercial cell designs.

Bone mineral density and donor age are not predictive of femoral ring allograft bone mechanical strength.

While metal or plastic interbody spinal fusion devices are manufactured to appropriate mechanical standards, mechanical properties of commercially prepared structural allograft bone remain relatively unassessed. Robust models predicting compressive load to failure of structural allograft bone based on easily measured variables would be useful. Three hundred twenty seven femoral rings from 34 cadaver femora were tested to failure in axial compression. Predictive variables included age, gender, bone mineral density (BMD), position along femoral shaft, maximum/minimum wall thickness, outer/inner diameter, and area. We used support vector regression and 10-fold cross-validation to develop robust nonlinear predictive models for load to failure. Model performance was measured by the root-mean-squared-deviation (RMSD) and correlation coefficients (r). A polynomial model using all variables had RMSD = 7.92, r = 0.84, indicating excellent performance. A model using all variables except BMD was essentially unchanged (RMSD = 8.12, r = 0.83). Eliminating both age and BMD produced a model with RMSD = 8.41, r = 0.82, again essentially unchanged. Compressive strength of structural allograft bone can be estimated using easily measured geometric parameters, without including BMD or age. As DEXA is costly and cumbersome, and setting upper age-limits for potential donors reduces the supply, our results may prove helpful to increase the quality and availability of structural allograft.

Comparison of growth factor treatments on the fibrochondrogenic potential of canine fibroblast-like synoviocytes for meniscal tissue engineering.

OBJECTIVE: To determine the in vitro effects of differing growth factor treatments on the fibrochondrogenic potential of fibroblast-like synoviocytes from cruciate ligament deficient femorotibial joints of dogs. STUDY DESIGN: In vitro study. SAMPLE POPULATION: Synoviocytes from dogs (n = 8) with naturally occurring cruciate ligament insufficiency. METHODS: Synoviocytes were cultured in monolayer and synthesized into tensioned synoviocyte bioscaffolds (TSB) suspended in media containing TGF-ß3, or FGF-2, TGF-ß1, and IGF-I. The 1,9-dimethylmethylene blue (DMMB) assay and toluidine blue stain assessed glycosaminoglycan content; hydroxyproline assay, and collagen I and II immunohistochemistry assessed collagen content. Biomechanical properties were determined by materials testing/force-deformation curves. RESULTS: All tissue cultures formed tensioned fibrous tissue-like constructs. Mean tissue cellularity and cellular viability was significantly greater in the triple growth factor-treated TSB by 0.09% and 44%, respectively. Percentage collagen content, and relative gene expression for collagen I, II, and aggrecan was not significantly different between groups. Median percentage of GAG content was significantly greater in triple growth factor-treated TSB by 1.6%. Biomechanical properties were not different in compression. Triple growth factor-treated TSB were significantly stronger in toughness, peak load to failure, and stiffness in tension. CONCLUSIONS: TGF-ß3 cultured bioscaffolds failed to outperform triple growth factor-treated TSB. Architectural extracellular matrix (ECM) organization and cellularity likely explained the differences between groups. TGF-ß3 alone cannot be recommended at this time for in vitro formation of autologous fibrocartilage bioscaffolds for meniscal deficiency.

Culture of canine synoviocytes on porcine intestinal submucosa scaffolds as a strategy for meniscal tissue engineering for treatment of meniscal injury in dogs.

Meniscal injury is a common cause of canine lameness. Tissue engineered bioscaffolds may be a treatment option for dogs suffering from meniscal damage. The aim of this study was to compare in vitro meniscal-like matrix formation and biomechanical properties of porcine intestinal submucosa sheets (SIS), used in canine meniscal regenerative medicine, to synoviocyte-seeded SIS bioscaffold (SSB), cultured with fetal bovine serum (SSBfbs) or chondrogenic growth factors (SSBgf). Synoviocytes from nine dogs were seeded on SIS and cultured for 30days with 17.7% fetal bovine serum or recombinant chondrogenic growth factors (IGF-1, TGFß1 and bFGF). The effect on fibrochondrogenesis was determined by comparing mRNA expression of collagen types Iα and IIα, aggrecan, and Sry-type homeobox protein-9 (SOX9) as well as protein expression of collagens I and II, glycosaminoglycan (GAG), and hydroxyproline. The effect of synoviocyte seeding and culture conditions on biochemical properties was determined by measuring peak load, tensile stiffness, resilience, and toughness of bioscaffolds. Pre-culture SIS contained 13.6% collagen and 2.9% double-stranded DNA. Chondrogenic growth factor treatment significantly increased SOX9, collagens I and IIα, aggrecan gene expression (P<0.05), and histological deposition of fibrocartilage extracellular matrix (GAG and collagen II). Culture with synoviocytes increased SIS tensile peak load at failure, resilience, and toughness of bioscaffolds (P<0.05). In conclusion, culturing SIS with synoviocytes prior to implantation might provide biomechanical benefits, and chondrogenic growth factor treatment of cultured synoviocytes improves in vitro axial meniscal matrix formation.

Evaluation of skull and tooth morphology and mineralization using high-resolution X-ray tomography.

High-resolution X-ray tomography (microCT) is increasingly available in research settings, and is a valuable tool in the study of mineralized tissue development. At resolutions of 2-20 µm, achievable for typical murine scale samples, it provides nondestructive visualization of three-dimensional tissue morphology and a limited ability for quantitative measurement of developmental parameters. Sample preparation is simple and can be tailored for compatibility with other biological assays. Here, we describe the application of microCT to the investigation of lower incisor development in the context of overall skull morphology.

Preliminary biomechanical proof of concept for a hybrid locking plate/variable pitch screw construct for anterior fixation of type II odontoid fractures.

STUDY DESIGN: A human cadaveric biomechanical proof-of-concept study. OBJECTIVE: To test whether adding a locking plate to the anterior surface of C2 attaching directly to the interfragmentary screw may reduce potential for anterior screw cutout and improve construct strength. SUMMARY OF BACKGROUND DATA: The most common mode of failure for screw fixation of dens fractures is via cutout at the anterior body of C2. METHODS: A human, cadaveric model of type II dens fractures was created and fixed using either a headless, fully threaded variable pitch screw (FTVPS) or a screw with an attachable locking plate construct (LPC). Following quasistatic loading to failure, stiffness and load to failure were compared using t tests. Mode of failure was determined from radiographical and gross inspection. RESULTS: Load to failure was greater for the LPC than for the FTVPS alone (498 N vs. 362 N, P = 0.04). The LPC consistently failed via compression of cancellous bone posterior to the lag screw, whereas the FTVPS constructs failed via cutout of the screw from the anterior C2 body. CONCLUSION: Locking plate supplementation of anterior screw fixation of type II odontoid fractures improves construct strength and changes the failure mechanism from anterior screw cutout to posterior displacement of the screw. An attachable locking plate/interfragmentary screw construct may improve clinical outcomes for these fractures.

Relationship of donor variables and graft dimension on biomechanical performance of femoral ring allograft.

Structural femoral ring allograft is commonly used in interbody spinal arthrodesis. Fractures of implanted femoral ring allograft have been reported. Data to guide donor screening and tissue processing by allograft tissue banks for factors that affect graft strength are incomplete. Fresh frozen human femora from 34 cadaveric donors were sectioned into ten 20-mm thick specimens. Bone mineral density (BMD), donor age, and graft dimensions were recorded for each specimen. Three hundred twenty-seven specimens were tested in quasi-static axial compression. Linear regression models compared load to failure with BMD, sex-specific donor age, minimum/maximum cortical wall thickness, and minimum/maximum outer ring diameter. Correlations between minimum and maximum cortical wall thickness and load to failure were significant (r = 0.73, p < 0.001 and r = 0.74, p < 0.001, respectively). BMD showed a weaker negative correlation with load to failure (r = -0.11, p = 0.05). Correlations between load to failure and minimum and maximum outer ring diameter and age (r = 0.06, p = 0.31) were not significant. We found that the minimum and maximum cortical wall thicknesses of femoral ring allograft are strongly correlated with the axial compressive load to failure of the graft. Other tested parameters did not prove to be effective predictors of resistance to axial loading.

Towards a modified consumer haptic device for robotic-assisted fine-motor repetitive motion training.

PURPOSE: To develop, test and evaluate affordable haptic technology to provide robotic-assisted repetitive motion fine-motor training. METHODS: A haptic computer/user interface was modified by adding a pantograph to hold a pen and to increase the haptic workspace. Custom software moves a pen attached to the device through prescribed three-dimensional (3D) stroke sequences to create two-dimensional glyphs. The pen's position is recorded in 3D coordinates at 1 kHz. Twenty-one healthy child volunteers were taught a standard handwriting curriculum in a group setting, two times per week for 45-60 min each session over 8 wks. The curriculum was supplemented by the device under the supervision of occupational therapy students. Outcomes were measured using the Evaluation Tool of Children's Handwriting (ETCH), and the Beery-Buktenica Developmental Test of visual-motor integration. RESULTS: Word legibility made significant gains on near point copying task (p=0.04; effect size=0.95). Letter legibility made no significant improvement. One healthy volunteer with illegible handwriting improved significantly on 8 of 14 ETCH measures. The children found the device engaging, but made several recommendations to redesign the pantograph and scribing movements. CONCLUSIONS: A consumer haptic device can be modified for robotic-assisted repetitive motion training for children. The device is affordable, portable, and engaging. It is safe for healthy volunteers. Objective time-stamped data offer the potential for telerehabilitation between a remote therapist and the school or home.

Ctip2/Bcl11b controls ameloblast formation during mammalian odontogenesis.

The transcription factor Ctip2/Bcl11b plays essential roles in developmental processes of the immune and central nervous systems and skin. Here we show that Ctip2 also plays a key role in tooth development. Ctip2 is highly expressed in the ectodermal components of the developing tooth, including inner and outer enamel epithelia, stellate reticulum, stratum intermedium, and the ameloblast cell lineage. In Ctip2(-/-) mice, tooth morphogenesis appeared to proceed normally through the cap stage but developed multiple defects at the bell stage. Mutant incisors and molars were reduced in size and exhibited hypoplasticity of the stellate reticulum. An ameloblast-like cell population developed ectopically on the lingual aspect of mutant lower incisors, and the morphology, polarization, and adhesion properties of ameloblasts on the labial side of these teeth were severely disrupted. Perturbations of gene expression were also observed in the mandible of Ctip2(-/-) mice: expression of the ameloblast markers amelogenin, ameloblastin, and enamelin was down-regulated, as was expression of Msx2 and epiprofin, transcription factors implicated in the tooth development and ameloblast differentiation. These results suggest that Ctip2 functions as a critical regulator of epithelial cell fate and differentiation during tooth morphogenesis.

Biomechanical comparison of a fully threaded, variable pitch screw and a partially threaded lag screw for internal fixation of Type II dens fractures.

STUDY DESIGN: Stiffness and load to failure were studied in a human cadaver model of Type II odontoid fractures stabilized with either a single partially threaded lag screw and washer or a headless fully threaded variable pitch screw. OBJECTIVE: To determine whether a headless fully threaded variable pitch screw provides biomechanically superior fixation of Type II odontoid fractures in comparison with a partially threaded, cannulated lag screw and washer. SUMMARY OF BACKGROUND DATA: Surgical treatment of Anderson and D'Alonzo Type II odontoid fractures is often performed using a partially threaded cannulated screw and washer. Reported clinical failure rates of this construct are as high as 20%. This technique requires perforation of the cortex of the tip of the dens, placing the brainstem and vertebrobasilar circulation at risk. A headless fully threaded variable pitch screw has not been described for this application. METHODS: A transverse osteotomy was created at the base of the dens in 16 human cadaver C2 specimens and stabilized using either a headless fully threaded variable pitch screw or a partially threaded cannulated lag screw and washer. Specimens were loaded to failure under a static, posteriorly directed force applied to the surface of the dens. Stiffness and load to failure were measured and the mode of failure for each specimen was determined. RESULTS: Stiffness and load to failure were greater for the headless, fully threaded variable pitch screw compared with the partially threaded lag screw and washer. The mode of failure for all specimens was via anterior screw cut-out at the C2 vertebral body. CONCLUSION: A headless, fully threaded variable pitch screw was biomechanically favorable in comparison with a partially threaded lag screw and washer in this cadaver model of Type II dens fractures. The mode of failure at the C2 vertebral body may have important implications for further improvements in construct strength.

Bone contact forces on the distal tibia during the stance phase of running.

Although the tibia is a common site of stress fractures in runners, the loading of the tibia during running is not well understood. An integrated experimental and modeling approach was therefore used to estimate the bone contact forces acting on the distal end of the tibia during the stance phase of running, and the contributions of external and internal sources to these forces. Motion capture and force plate data were recorded for 10 male runners as they ran at 3.5-4 m/s. From these data, the joint reaction force (JRF), muscle forces, and bone contact force on the tibia were computed at the ankle using inverse dynamics and optimization methods. The distal end of the tibia was compressed and sheared posteriorly throughout most of stance, with respective peak forces of 9.00+/-1.13 and 0.57+/-0.18 body weights occurring during mid stance. Internal muscle forces were the primary source of tibial compression, whereas the JRF was the primary source of tibial shear due to the forward inclination of the leg relative to the external ground reaction force. The muscle forces and JRF both acted to compress the tibia, but induced tibial shear forces in opposing directions during stance, magnifying tibial compression and reducing tibial shear. The superposition of the peak compressive and posterior shear forces at mid stance may contribute to stress fractures in the posterior face of the tibia. The implications are that changes in running technique could potentially reduce stress fracture risk.

The behavior of thoracic trabecular bone during flexion.

Thoracic compression fractures are often described as anterior wedge fractures. Although the radiographic signs of these fractures are easily identified, the mechanism of the trabecular failure is not well understood. The current study addressed this mechanism in the lower thoracic spine by measuring the trabecular strain. Trabecular strain was measured in six human thoracic cadaver spines during 1) compressive and 2) flexural loading. The strains were measured at incremental loads using a texture correlation. They were analyzed by global contour plots and regional analysis of the T11 vertebrae. Specimens loaded under only compression exhibited uniform strains in the vertebral body. During flexion, however, the strains were concentrated in the anterosuperior margin of the vertebral body and the compressive and shear strain magnitudes in this region were significantly increased. These results demonstrate that the flexural position places the lower thoracic spine at greater risk of anterior compression fracture as seen clinically.

Development of an animal model of acetabular fractures.

A closed intraarticular fracture is a complex injury that consists of a physical disruption of the subchondral bone and articular surface, and an impaction injury to the articular surface that occurs at the time of the fracture. Few experimental models have been able to incorporate the elements of displaced articular fractures and blunt impaction injury to the articular surface. This work details the initial stages of an attempt to develop such model. Using a model of a dorsal wall fracture of the acetabulum in a goat, we have developed a bench-top method for assessing articular contact stress. Additionally, preliminary in vivo survival data are presented.

Biomechanical comparison of posterior pelvic ring fixation.

OBJECTIVE: To determine relative stiffness of various methods of posterior pelvic ring internal fixation. DESIGN: Simulated single leg stance loading of OTA 61-Cl.2, a2 fracture model (unilateral sacroiliac joint disruption and pubic symphysis diastasis). SETTING: Orthopaedic biomechanic laboratory. OUTCOME VARIABLES: Pubic symphysis gapping, sacroiliac joint gapping, hemipelvis coronal plane rotation. METHODS: Nine different posterior pelvic ring fixation methods were tested on each of six hard plastic pelvic models. Pubic symphysis was plated. The pelvic ring was loaded to 1000N. RESULTS: All data were normalized to values obtained with posterior fixation with a single iliosacral screw. The types of fixation could be grouped into three categories based on relative stiffness of fixation: For sacroiliac joint gapping, group 1-fixation stiffness 0.8 and above (least stiff) includes a single iliosacral screw (conditions A and J), an isolated tension band plate (condition F), and two sacral bars (condition H); group 2-fixation stiffness 0.6 to 0.8 (intermediate stiffness) includes a tension band plate and an iliosacral screw (condition E), one or two sacral bars in combination with an iliosacral screw (conditions G and I); group 3-fixation stiffness 0.6 and below (greatest stiffness) includes two anterior sacroiliac plates (condition D), two iliosacral screws (condition B), and two anterior sacroiliac plates and an iliosacral screw (condition C). For sacroiliac joint rotation, group 1-fixation stiffness 0.8 and above includes a single iliosacral screw (conditions A and J), two anterior sacroiliac plates (condition D), a tension band plate in isolation or in combination with an iliosacral screw (conditions E and F), and two sacral bars (condition H); group 2-fixation stiffness 0.6 to 0.8 (intermediate level of instability) includes either one or two sacral bars in combination with an iliosacral screw (conditions G and I); group 3-fixation stiffness 0.6 and below (stiffest fixation) consists of two iliosacral screws (condition B) and two anterior sacroiliac plates and an iliosacral screw (condition C). DISCUSSION: Under conditions of maximal instability with similar material properties between specimens, differences in stiffness of posterior pelvic ring fixation can be demonstrated. The choice of which method to use is multifactorial.

Trabecular bone strain changes resulting from partial and complete meniscectomy.

Previous studies have documented how partial and complete meniscectomy affect articular contact pressure, but changes in load transfer through the complete osteochondral structure of the proximal tibia after partial and complete meniscectomy are not well known. The current study measured trabecular bone strain changes in the medial tibial plateau resulting from partial and complete medial meniscectomy. Midcoronal sections were prepared from knees from cadavers. High quality digital images, made from contact radiographs of loaded samples, were compared with digital images of unloaded samples using in-house software to measure trabecular bone strain. Measurements were made on specimens with an intact medial meniscus, after removal of the inner (2/3) of the meniscus, and after complete meniscectomy. Partial meniscectomy caused minimal increases in trabecular bone strain throughout the proximal tibia. However specimens with complete meniscectomy had significant trabecular bone strain increases. Many patients sustaining meniscus tears are young, therefore, it is important to understand mechanical changes associated with partial meniscectomy. The data suggest partial meniscectomy causes little change in load transfer through the proximal tibia, supporting partial meniscectomy as a good surgical option for patients with meniscus tears.

Trabecular bone strain changes associated with subchondral stiffening of the proximal tibia.

Subchondral stiffening is a hallmark pathologic feature of osteoarthritis but its mechanical and temporal relationship to the initiation or the progression of osteoarthritis is not established. The mechanical effect of subchondral stiffening on the surrounding trabecular bone is poorly understood. This study employs a relatively new application of digital image correlation to measure strain in the trabecular region of the proximal medial tibia in normal specimens and in specimens with simulated subchondral bone stiffening. Coronal sections from eight normal human cadaveric proximal tibiae were loaded in static compression and high resolution contact radiographs were made. Repeat contact radiographs were collected after the subchondral bone near the jointline was stiffened using polymethylmethacrylate. Digital images, made from loaded and unloaded contact radiographs, were compared using in-house software to measure trabecular displacement and calculate trabecular bone strain. Overall strain was higher in the stiffened specimens suggesting experimental artifiact significantly affected our results. Consistent increases in median maximum shear strain, median maximum principal strain, median minimum principal strain, and peak shear strain were measured near the inner and outer edges of the stiffened segment. Our experiment provides direct experimental measurement of increases in trabecular bone strain caused by subchondral stiffening, however, the clinical and biologic importance of strain increases is unknown.

Trabecular bone strain changes associated with subchondral comminution of the distal tibia.

OBJECTIVE: To measure trabecular bone strain changes resulting from three increasing subchondral bone defects in the distal tibia. DESIGN: Cadaveric biomechanical model. SETTING: Contact radiographs were made from sagittal sections of human cadaveric distal tibia under no load and loaded to 400 N. Digital images, made from contact radiographs of unloaded specimens, were compared to corresponding digital images of loaded specimens using custom software that measures trabecular deformation and calculates trabecular bone strain. INTERVENTION: Twelve specimens were initially loaded intact in compression. Testing was repeated after creating three increasing circular subchondral bone defects in the center of a sagittal cross-section of the distal tibia. Defects were 10%, 20%, and 30% of the sagittal diameter of the distal tibia. MAIN OUTCOME MEASURES: Maximum shear strain, maximum principal strain, and minimum principal strain were measured in six discrete regions in the trabecular bone in the distal tibia. RESULTS: Small defects (10%) caused minimal strain elevations. Significant increases in trabecular bone strain were measured with medium (20%) and large (30%) defects. Compressive strain increases as high as 1400 microstrain (10 strain) were measured adjacent to and proximal to the defects with medium and large defects. CONCLUSIONS: Subchondral defects cause size-dependent elevations in trabecular bone strain in the distal tibia. Medium and large defects caused rapidly increasing trabecular bone deformation under load.

Trabecular bone strain changes associated with subchondral bone defects of the tibial plateau.

OBJECTIVE: To measure trabecular bone strain changes resulting from three increasing subchondral bone defects in the medial tibial plateau. DESIGN: Cadaveric biomechanical model. SETTING: Contact radiographs were made from coronal sections of human cadaveric proximal tibia under no load and loaded to 400 newtons (N). Digital images made from contact radiographs of unloaded specimens were compared to corresponding digital images of loaded specimens using in-house software that detects trabecular deformation and measure trabecular bone strain. INTERVENTION: Ten specimens were loaded intact and with three increasing circular subchondral bone defects and centered under the subchondral plates in the medial tibial plateau that were 10%, 20%, and 30% of the coronal width of the medial plateau. MAIN OUTCOME MEASURE: Maximum shear strain and minimum principal strain were measured at approximately 2,600 discrete points in the trabecular bone in the medial tibial plateau. RESULTS: Trabecular strain increased most dramatically as defects increased from the medium (20%) to the large (30%) defect. The regions of greatest strain elevation were between the physeal scar and joint line near the medial cortex. Small (10%) and medium (20%) defects resulted in modest strain elevations. CONCLUSIONS: Subchondral defects cause size-dependent elevations in trabecular bone strain in the medial tibial plateau. A size threshold may exist, above which the trabecular bone is subjected to rapidly increasing deformation under load.