Structural behaviour and strain distribution of the long bones of the human lower limbs

Quantifying the immediate post-implantation strain field of cadaveric tibiae implanted with cementless tibial trays: A time-elapsed micro-CT and digital volume correlation analysis during stair descent

Primary stability, the mechanical fixation between implant and bone prior to osseointegration, is crucial for the long-term success of cementless tibial trays. However, little is known about the mechanical interplay between the implant and bone internally, as experimental studies quantifying internal strain are limited. This study employed digital volume correlation (DVC) to quantify the immediate post-implantation strain field of five cadaveric tibiae implanted with a commercially available cementless titanium tibial tray (Attune, DePuy Synthes). The tibiae were subjected to a five-step loading sequence (0-2.5 bodyweight, BW) replicating stair descent, with concomitant time-elapsed micro-CT imaging. With progressive loads, increased compression of trabecular bone was quantified, with the highest strains directly under the posterior region of the tibial tray implant, dissipating with increasing distance from the bone-implant interface. After load removal of the last load step (2.5BW), residual strains were observed in all of the five tibiae, with residual strains confined within 3.14 mm from the bone-implant interface. The residual strain is reflective of the observed initial migration of cementless tibial trays reported in clinical studies. The presence of strains above the yield strain of bone accepted in literature suggests that inelastic properties should be included within finite element models of the initial mechanical environment. This study provides a means to experimentally quantify the internal strain distribution of human tibia with cementless trays, increasing the understanding of the mechanical interaction between bone and implant.

Development of a Bionic Tube with High Bending-Stiffness Properties Based on Human Tibiofibular Shapes

The human tibiofibular complex has undergone a long evolutionary process, giving its structure a high bearing-capacity. The distinct tibiofibular shape can be used in engineering to acquire excellent mechanical properties. In this paper, four types of bionic tubes were designed by extracting the dimensions of different cross-sections of human tibia-fibula. They had the same outer profiles, but different inner shapes. The concept of specific stiffness was introduced to evaluate the mechanical properties of the four tubes. Finite-element simulations and physical bending-tests using a universal testing machine were conducted, to compare their mechanical properties. The simulations showed that the type 2 bionic tube, i.e., the one closest to the human counterpart, obtained the largest specific-stiffness (ε = 6.46 × 10⁴), followed by the type 4 (ε = 6.40 × 10⁴) and the type 1 (ε = 6.39 × 10⁴). The type 3 had the largest mass but the least stiffness (ε = 6.07 × 10⁴). The specific stiffness of the type 2 bionic tube increased by approximately 25.8%, compared with that of the type 3. The physical tests depicted similar findings. This demonstrates that the bionic tube inspired by the human tibiofibular shape has excellent effectiveness and bending properties, and could be used in the fields of healthcare engineering, such as robotics and prosthetics.

Effect of fall direction on the lower hip fracture risk in athletes with different loading histories: A finite element modeling study in multiple sideways fall configurations

Physical loading makes bones stronger through structural adaptation. Finding effective modes of exercise to improve proximal femur strength has the potential to decrease hip fracture risk. Previous proximal femur finite element (FE) modeling studies have indicated that the loading history comprising impact exercises is associated with substantially higher fracture load. However, those results were limited only to one specified fall direction. It remains thus unclear whether exercise-induced higher fracture load depends on the fall direction. To address this, using magnetic resonance images of proximal femora from 91 female athletes (mean age 24.7 years with >8 years competitive career) and their 20 non-athletic but physically active controls (mean age 23.7 years), proximal femur FE models were created in 12 different sideways fall configurations. The athletes were divided into five groups by typical loading patterns of their sports: high-impact (H-I: 9 triple- and 10 high-jumpers), odd-impact (O-I: 9 soccer and 10 squash players), high-magnitude (H-M: 17 powerlifters), repetitive-impact (R-I: 18 endurance runners), and repetitive non-impact (R-NI: 18 swimmers). Compared to the controls, the FE models showed that the H-I and R-I groups had significantly (p < 0.05) higher fracture loads, 11-17% and 22-28% respectively, in all fall directions while the O-I group had significantly 10-11% higher fracture loads in four fall directions. The H-M and R-NI groups did not show significant benefit in any direction. Also, the analyses of the minimum fall strength (MFS) among these multiple fall configurations confirmed significantly 15%, 11%, and 14% higher MFSs in these impact groups, respectively, compared to the controls. These results suggest that the lower hip fracture risk indicated by higher fracture loads in athletes engaged in high impact or repetitive impact sports is independent of fall direction whereas the lower fracture risk attributed to odd-impact exercise is more modest and specific to the fall direction. Moreover, in concordance with the literature, the present study also confirmed that the fracture risk increases if the impact is imposed on the more posterolateral aspect of the hip. The present results highlight the importance of engaging in the impact exercises to prevent hip fractures and call for retrospective studies to investigate whether specific impact exercise history in adolescence and young adulthood is also associated with lower incidence of hip fractures in later life.

The effect of age and initial compression on the force relaxation response of the femur in elderly women

The effect of force amount, age, body weight and bone mineral density (BMD) on the femur's force relaxation response was analysed for 12 donors (age: 56-91 years). BMD and fracture load, F _L, were estimated from clinical CT images. The 30 min force relaxation was obtained using a constant compression generating an initial force F ₀ between 7% and 78% of F _L. The stretched decay function (F(t) = A × e ^(-t/τ)β ) proposed earlier for bone tissue was fitted to the data and analysed using robust linear regression. The relaxation function fitted well to all the recordings (R ² = 0.99). The relative initial force was bilinearly associated (R ² = 0.83) to the shape factor, β, and the characteristic time, τ, when F ₀/F _L was less than 0.4, although β was no longer associated with F ₀/F _L by pooling all the data. The characteristic time τ increased with age (R ² = 0.37, p = 0.03) explaining 35% of the variation of τ in the entire dataset. In conclusion, the relative initial force mostly determines the femur's force relaxation response, although the early relaxation response under subcritical loading is variable, possibly due to damage occurring at subcritical loading levels.

Intercalary reconstruction of long bones by massive allograft: Comparison of construct stability ensured by three different host-graft junctions and two types of fixations in a synthetic femur model

An intercalary segmental allograft is an option for limb salvage in bone tumours. Stable and congruent intercalary reconstructions are a prerequisite for achieving host-graft union. However, a too rigid fixation could increase the risk of late complications correlated with negative bone remodelling. This study compared the reconstruction stiffness achieved by three different host-graft junctions, namely, end-to-end, modified step-cut, and taper. A low-stiffness bone plate was used as the fixation method, except for the taper junction where a low-stiffness intramedullary nail was also used to investigate the effects of different types of fixation on construct stiffness. Composite femora were tested under four loading conditions to determine coronal and sagittal bending stiffness, as well as torsional stiffness in opposite directions. Stiffness values were expressed as a percentage of intact host bone stiffness (%IBS). While a reduction of coronal bending stiffness was found with taper junctions (76%IBS) compared with the high values ensured by end-to-end (96%IBS) and modified step-cut junctions (92%IBS), taper junctions significantly increased stiffness under sagittal bending and torsion in intra- and extra-direction: end-to-end 29%IBS, 7%IBS, 7%IBS, modified step-cut 38%IBS, 20%IBS, 21%IBS, and taper junction 52%IBS, 55%IBS, 56%IBS, respectively. Construct stiffness with taper junctions was decreased by 11-41%IBS by replacing the bone plate with an intramedullary nail. Taper junctions can be an alternative to achieve intercalary reconstructions with more homogeneous and, in three out of four loading conditions, significantly higher construct stability without increasing bone plate stiffness. The risk of instability under high torsional loads increases when taper junctions are associated with a low-stiffness intramedullary nail.

Which experimental procedures influence the apparent proximal femoral stiffness? A parametric study

Background

Experimental validation is the gold standard for the development of FE predictive models of bone. Employing multiple loading directions could improve this process. To capture the correct directional response of a sample, the effect of all influential parameters should be systematically considered. This study aims to determine the impact of common experimental parameters on the proximal femur’s apparent stiffness.

Methods

To that end, a parametric approach was taken to study the effects of: repetition, pre-loading, re-adjustment, re-fixation, storage, and μCT scanning as random sources of uncertainties, and loading direction as the controlled source of variation in both stand and side-fall configurations. Ten fresh-frozen proximal femoral specimens were prepared and tested with a novel setup in three consecutive sets of experiments. The neutral state and 15-degree abduction and adduction angles in both stance and fall configurations were tested for all samples and parameters. The apparent stiffness of the samples was measured using load-displacement data from the testing machine and validated against marker displacement data tracked by DIC cameras.

Results

Among the sources of uncertainties, only the storage cycle affected the proximal femoral apparent stiffness significantly. The random effects of setup manipulation and intermittent μCT scanning were negligible. The 15^∘ deviation in loading direction had a significant effect comparable in size to that of switching the loading configuration from neutral stance to neutral side-fall.

Conclusion

According to these results, comparisons between the stiffness of the samples under various loading scenarios can be made if there are no storage intervals between the different load cases on the same samples. These outcomes could be used as guidance in defining a highly repeatable and multi-directional experimental validation study protocol.

Real-time replication of three-dimensional and time-varying physiological loading cycles for bone and implant testing: A novel protocol demonstrated for the proximal human femur while walking

In vitro real-time replication of three-dimensional, time-varying load profiles acting on human bones during physical activity can advance bone and implant testing protocols. This study aimed to develop a novel protocol for applying the three-dimensional, time-varying hip contact force while walking to a human femur specimen. The target force profile was obtained from the literature. A proximal femur from an elderly female donor was instrumented using ten rosette strain gages and tested using a custom-made hexapod robot. A load-control algorithm determined the robot position generating the target force at low frequency (0.0004 Hz). Five cycles of the robot position were played back at five intermediate frequencies up to real-time (0.04, 0.08, 0.16, 0.4, and 0.8 Hz). The hip reaction force, the length of the actuators (position), and cortical strains were compared. The error in the load-control force was 0.3 ± 4.2 N (mean ± SD). The last three force, position, and strain cycles varied by less than 1.1% for every frequency analyzed. Across frequencies, the force increased by 28% at 0.8 Hz as a logarithmic function of frequency (R² = 0.98). The position and strain error linearly increased with frequency up to 0.4 Hz. The median position error and the interquartile range of the strain error reached 15% and 13% at 0.8 Hz. Changes of force and cortical strain at increasing frequencies were linearly related (R² = 0.99). Therefore, the protocol developed can provide repeatable three-dimensional time-varying load profiles, although the comparison of the specimen deformation obtained across frequencies should be considered with care, particularly in the higher frequency range. This information supports the design of dynamic tests of bone and implants.

Image-based finite-element modeling of the human femur

Fracture is considered a critical clinical endpoint in skeletal pathologies including osteoporosis and bone metastases. However, current clinical guidelines are limited with respect to identifying cases at high risk of fracture, as they do not account for many mechanical determinants that contribute to bone fracture. Improving fracture risk assessment is an important area of research with clear clinical relevance. Patient-specific numerical musculoskeletal models generated from diagnostic images are widely used in biomechanics research and may provide the foundation for clinical tools used to quantify fracture risk. However, prior to clinical translation, in vitro validation of predictions generated from such numerical models is necessary. Despite adopting radically different models, in vitro validation of image-based finite element (FE) models of the proximal femur (predicting strains and failure loads) have shown very similar, encouraging levels of accuracy. The accuracy of such in vitro models has motivated their application to clinical studies of osteoporotic and metastatic fractures. Such models have demonstrated promising but heterogeneous results, which may be explained by the lack of a uniform strategy with respect to FE modeling of the human femur. This review aims to critically discuss the state of the art of image-based femoral FE modeling strategies, highlighting principal features and differences among current approaches. Quantitative results are also reported with respect to the level of accuracy achieved from in vitro evaluations and clinical applications and are used to motivate the adoption of a standardized approach/workflow for image-based FE modeling of the femur.

Differences in the Cortical Structure of the Whole Fibula and Tibia Between Long-Distance Runners and Untrained Controls. Toward a Wider Conception of the Biomechanical Regulation of Cortical Bone Structure

The cortical structure of human fibula varies widely throughout the bone suggesting a more selective adaptation to different mechanical environments with respect to the adjacent tibia. To test this hypothesis, serial-pQCT scans of the dominant fibulae and tibiae of 15/15 men/women chronically trained in long-distance running were compared with those of 15/15 untrained controls. When compared to controls, the fibulae of trained individuals had similar (distally) or lower (proximally) cortical area, similar moments of inertia (MI) for anterior-posterior bending (xMI) and lower for lateral bending (yMI) with a lower "shape-index" (yMI/xMI ratio) throughout, and higher resistance to buckling distally. These group differences were more evident in men and independent of group differences in bone mass. These results contrast with those observed in the tibia, where, as expected, structural indicators of bone strength were greater in trained than untrained individuals. Proximally, the larger lateral flexibility of runners' fibulae could improve the ability to store energy, and thereby contribute to fast-running optimization. Distally, the greater lateral fibular flexibility could reduce bending strength. The latter appears to have been compensated by a higher buckling strength. Assuming that these differences could be ascribed to training effects, this suggests that usage-derived strains in some bones may modify their relative structural resistance to different kinds of deformation in different regions, not only regarding strength, but also concerning other physiological roles of the skeleton.

The Size of Simulated Lytic Metastases Affects the Strain Distribution on the Anterior Surface of the Vertebra

Metastatic lesions of the vertebra are associated with risk of fracture, which can be disabling and life-threatening. In the literature, attempts are found to identify the parameters that reduce the strength of a metastatic vertebra leading to spine instability. However, a number of controversial issues remain. Our aim was to quantify how the strain distribution in the vertebral body is affected by the presence and by the size of a simulated metastatic defect. Five cadaveric thoracic spine segments were subjected to non-destructive presso-flexion while intact, and after simulation of metastases of increasing size. For the largest defect, the specimens were eventually tested to failure. The full-field strain distribution in the elastic range was measured with digital image correlation (DIC) on the anterior surface of the vertebral body. The mean strain in the vertebra remained similar to the intact when the defects were smaller than 30% of the vertebral volume. The mean strains became significantly larger than in the intact for larger defects. The map of strain and its statistical distribution indicated a rather uniform condition in the intact vertebra and with defects smaller than 30%. Conversely, the strain distribution became significantly different from the intact for defects larger than 30%. A strain peak appeared in the region of the simulated metastasis, where fracture initiated during the final destructive test. This is a first step in understanding how the features of metastasis influence the vertebral strain and for the construction of a mechanistic model to predicted fracture.

Clinical hip fracture is accompanied by compression induced failure in the superior cortex of the femoral neck

Hip fractures pose a major health problem throughout the world due to their devastating impact. Current theories for why these injuries are so prevalent in the elderly point to an increased propensity to fall and decreases in bone mass with ageing. However, the fracture mechanisms, particularly the stress and strain conditions leading to bone failure at the hip remain unclear. Here, we directly examined the cortical bone from clinical intra-capsular hip fractures at a microscopic level, and found strong evidence of compression induced failure in the superior cortex. A total of 143 sections obtained from 24 femoral neck samples that were retrieved from 24 fracturing patients at surgery were examined using laser scanning confocal microscopy (LSCM) after fluorescein staining. The stained microcracks showed significantly higher density in the superior cortex than in the inferior cortex, indicating a greater magnitude of strain in the superior femoral neck during the failure-associated deformation and fracture process. The predominant stress state for each section was reconstructed based on the unique correlation between the microcrack pattern and the stress state. Specifically, we found clear evidence of longitudinal compression and buckling as the primary failure mechanisms in the superior cortex. These findings demonstrate the importance of microcrack analysis in studying clinical hip fractures, and point to the central role of the superior cortex failure as an important aspect of the failure initiation in clinical intra-capsular hip fractures.

Young's Modulus and Load Complexity: Modeling Their Effects on Proximal Femur Strain

Finite element analysis (FEA) is a powerful tool for evaluating questions of functional morphology, but the application of FEA to extant or extinct creatures is a non-trivial task. Three categories of input data are needed to appropriately implement FEA: geometry, material properties, and boundary conditions. Geometric data are relatively easily obtained from imaging techniques, but often material properties and boundary conditions must be estimated. Here we conduct sensitivity analyses of the effect of the choice of Young's Modulus for elements representing trabecular bone and muscle loading complexity on the proximal femur using a finite element mesh of a modern human femur. We found that finite element meshes that used a Young's Modulus between 500 and 1,500 MPa best matched experimental strains. Loading scenarios that approximated the insertion sites of hip musculature produced strain patterns in the region of the greater trochanter that were different from scenarios that grouped muscle forces to the superior greater trochanter, with changes in strain values of 40% or more for 20% of elements. The femoral head, neck, and proximal shaft were less affected (e.g. approximately 50% of elements changed by 10% or less) by changes in the location of application of muscle forces. From our sensitivity analysis, we recommend the use of a Young's Modulus for the trabecular elements of 1,000 MPa for the proximal femur (range 500-1,500 MPa) and that the muscular loading complexity be dependent on whether or not strains in the greater trochanter are the focus of the analytical question. Anat Rec, 301:1189-1202, 2018. © 2018 Wiley Periodicals, Inc.

Physiological joint line total knee arthroplasty designs are especially sensitive to rotational placement - A finite element analysis

Mechanical and kinematical aligning techniques are the usual positioning methods during total knee arthroplasty. However, alteration of the physiological joint line and unbalanced medio-lateral load distribution are considered disadvantages in the mechanical and kinematical techniques, respectively. The aim of this study was to analyse the influence of the joint line on the strain and stress distributions in an implanted knee and their sensitivity to rotational mal-alignment. Finite element calculations were conducted to analyse the stresses in the PE-Inlay and the mechanical strains at the bone side of the tibia component-tibia bone interface during normal positioning of the components and internal and external mal-rotation of the tibial component. Two designs were included, a horizontal and a physiological implant. The loading conditions are based on internal knee joint loads during walking. A medialization of the stresses on the PE-Inlay was observed in the physiological implant in a normal position, accompanied by higher stresses in the mal-rotated positions. Within the tibia component-tibia bone interface, similar strain distributions were observed in both implant geometries in the normal position. However, a medialization of the strains was observed in the physiological implant in both mal-rotated conditions with greater bone volume affected by higher strains. Although evident changes due to mal-rotation were observed, the stresses do not suggest a local plastic deformation of the PE-Inlay. The strains values within most of the tibia component-tibia bone interface were in the physiological strain zone and no significant bone changes would be expected. The physiological cut on the articular aspect showed no detrimental effect compared to the horizontal implant.

Effect of the In Vitro Boundary Conditions on the Surface Strain Experienced by the Vertebral Body in the Elastic Regime

The vertebral strength and strain can be assessed in vitro by both using isolated vertebrae and sets of three adjacent vertebrae (the central one is loaded through the disks). Our goal was to elucidate if testing single-vertebra-specimens in the elastic regime provides different surface strains to three-vertebrae-segments. Twelve three-vertebrae sets were extracted from thoracolumbar human spines. To measure the principal strains, the central vertebra of each segment was prepared with eight strain-gauges. The sets were tested mechanically, allowing comparison of the surface strains between the two boundary conditions: first when the same vertebra was loaded through the disks (three-vertebrae-segment) and then with the endplates embedded in cement (single-vertebra). They were all subjected to four nondestructive tests (compression, traction, torsion clockwise, and counterclockwise). The magnitude of principal strains differed significantly between the two boundary conditions. For axial loading, the largest principal strains (along vertebral axis) were significantly higher when the same vertebra was tested isolated compared to the three-vertebrae-segment. Conversely, circumferential strains decreased significantly in the single vertebrae compared to the three-vertebrae-segment, with some variations exceeding 100% of the strain magnitude, including changes from tension to compression. For torsion, the differences between boundary conditions were smaller. This study shows that, in the elastic regime, when the vertebra is loaded through a cement pot, the surface strains differ from when it is loaded through the disks. Therefore, when single vertebrae are tested, surface strain should be taken with caution.

THE EFFECT OF COMPUTED TOMOGRAPHY CURRENT REDUCTION ON PROXIMAL FEMUR SUBJECT-SPECIFIC FINITE ELEMENT MODELS

Many studies have addressed the modulation of computed tomography (CT) parameters, and particularly of tube current, to obtain a good compromise between the X-ray dose to the patient and the image quality for diagnostic applications. This study aimed at evaluating the influence of dose reduction by means of tube current reduction on the CT-based subject-specific finite element (FE) modeling. To this aim, CT scans at stepwise reduced values of tube current from 180[Formula: see text]mAs to 80[Formula: see text]mAs were performed on: (i) a densitometric phantom, to quantify the changes in the calibration equation; (ii) a fresh-frozen, water submersed, human cadaver femur, to quantify changes in geometry reconstruction and material mapping from CT, as well as strain prediction accuracy, based on the in vitro strain measurements available; (iii) a fresh-frozen human cadaver thigh with soft tissues attached, to quantify FE results changes in conditions similar to those found in vivo. The results showed that the tube current reduction does not affect the 3D modeling and the femur FE analysis. Our pilot study highlights the possibility of performing CT scans with reduced dose to generate biomechanical models, although a confirmation by performing larger studies with clinical CT data is needed.

Structural differences in cortical shell properties between upper and lower human fibula as described by pQCT serial scans. A biomechanical interpretation

This study describes the structural features of fibula cortical shell as allowed by serial pQCT scans in 10/10 healthy men and women aged 20-40years. Indicators of cortical mass (mineral content -BMC-, cross-sectional area -CSA-), mineralization (volumetric BMD, vBMD), design (perimeters, thickness, moments of inertia -MIs-) and strength (Bone Strength Indices, BSIs; polar Strength-Strain Index, pSSI) were determined. All cross-sectional shapes and geometrical or strength indicators suggested a sequence of five different regions along the bone, which would be successively adapted to 1. transmit loads from the articular surface to the cortical shell (near the proximal tibia-fibular joint), 2. favor lateral bending (central part of upper half), 3. resist lateral bending (mid-diaphysis), 4. favor lateral bending again (central part of the lower half), and 5. resist bending/torsion (distal end). Cortical BMC and the cortical/total CSA ratio were higher at the midshaft than at both bone ends (p<0.001). However, all MIs, BSIs and pSSI values and the endocortical perimeter/cortical CSA ratio (indicator of the mechanostat's ability to re-distribute the available cortical mass) showed a "W-shaped" distribution along the bone, with maximums at the mid-shaft and at both bone's ends (site effect, p<0.001). The correlation coefficient (r) of the relationship between MIs (y) and cortical vBMD (x) at each bone site ("distribution/quality" curve that describes the efficiency of distribution of the cortical tissue as a function of the local tissue stiffness) was higher at proximal than distal bone regions (p<0.001). The results from the study suggest that human fibula is primarily adapted to resist bending and torsion rather than compression stresses, and that fibula's bending strength is lower at the center of its proximal and distal halves and higher at the mid-shaft and at both bone's ends. This would favor, proximally, the elastic absorption of energy by the attached muscles that rotate or evert the foot, and distally, the widening of the heel joint and the resistance to excessive lateral bending. Results also suggest that biomechanical control of structural stiffness differs between proximal and distal fibula.

A PRELIMINARY IN VITRO BIOMECHANICAL EVALUATION OF PROPHYLACTIC CEMENT AUGMENTATION OF THE THORACOLUMBAR VERTEBRAE

In this paper, the biomechanical effectiveness of prophylactic augmentation in preventing fracture was investigated. In vitro biomechanical tests were performed to assess which factors make prophylactic augmentation effective/ineffective in reducing fracture risk. Nondestructive and destructive in vitro tests were performed on isolated osteoporotic vertebrae. Five sets of three-adjacent-vertebrae were tested. The central vertebra of each triplet was tested in the natural condition (control) non-destructively (axial-compression, torsion) and destructively (axial-compression). The two adjacent vertebrae were first tested nondestructively (axial-compression, torsion) pre-augmentation; prophylactic augmentation (uni- or bi-pedicular access) was then performed delivering 5.04[Formula: see text]mL to 8.44[Formula: see text]mL of acrylic cement by means of a customized device; quality of augmentation was CT-assessed; the augmented vertebrae were re-tested nondestructively (axial-compression, torsion), and eventually loaded to failure (axial-compression). Vertebral stiffness was correlated with the first-failure, but not with ultimate failure. The force and work to ultimate failure in prophylactic-augmented vertebrae was consistently larger than in the controls. However, in some cases the first-failure force and work in the augmented vertebrae were lower than for the controls. To investigate the reasons for such unpredictable results, the correlation with augmentation quality was analyzed. Some augmentation parameters seemed more correlated with mechanical outcome (statistically not-significant due to the limited sample size): uni-pedicular access resulted in a single cement mass, which tended to increase the force and work to first- and ultimate failure. The specimens with the highest strength and toughness also had: at least 25% cement filling, cement mass shifted anteriorly, and cement-endplate contact. These findings seem to confirm that prophylactic augmentation may aid reducing the risk of fracture. However, inadequate augmentation may have detrimental consequences. This study suggests that, to improve the strength of the augmented vertebrae, more attention should be dedicated to the quality of augmentation in terms of amount and position of the injected cement.

USE OF DIGITAL IMAGE CORRELATION TO INVESTIGATE THE BIOMECHANICS OF THE VERTEBRA

Digital image correlation (DIC) is being introduced to the biomechanical field. However, as DIC relies on a number of major assumptions, it requires a careful optimization in order to obtain accurate and precise results. The first step was the preparation of the speckle pattern by an airbrush spray gun following a factorial design to explore the different settings: the different speckle patterns created were analyzed to achieve the optimal speckle size, with minimal dispersion of speckle sizes. A benchmark test, with an aluminum specimen prepared with the speckle pattern, was conducted in which the errors affecting the computed strain were measured in a zero-displacement, zero-strain condition. The software parameters (facet size, step, and local regression) were singularly analyzed in order to understand their behavior on the final output. Moreover, the hardware parameters (camera gain, exposure, lens distortion) were analyzed. The output showed that a careful optimization allowed the reducing the systematic and random errors, respectively, from 150 to 10 microstrain and from 600 to 110 microstrain. Finally, the acquired know-how was applied to a biological specimen (human vertebra).

Strain distribution in the proximal Human femur during in vitro simulated sideways fall

This study assessed: (i) how the magnitude and direction of principal strains vary for different sideways fall loading directions; (ii) how the principal strains for a sideways fall differ from physiological loading directions; (iii) the fracture mechanism during a sideways fall. Eleven human femurs were instrumented with 16 triaxial strain gauges each. The femurs were non-destructively subjected to: (a) six loading configurations covering the range of physiological loading directions; (b) 12 configurations simulating sideways falls. The femurs were eventually fractured in a sideways fall configuration while high-speed cameras recorded the event. When the same force magnitude was applied, strains were significantly larger in a sideways fall than for physiological loading directions (principal compressive strain was 70% larger in a sideways fall). Also the compressive-to-tensile strain ratio was different: for physiological loading the largest compressive strain was only 30% larger than the largest tensile strain; but for the sideways fall, compressive strains were twice as large as the tensile strains. Principal strains during a sideways fall were nearly perpendicular to the direction of principal strains for physiological loading. In the most critical regions (medial part of the head-neck) the direction of principal strain varied by less than 9° between the different physiological loading conditions, whereas it varied by up to 17° between the sideways fall loading conditions. This was associated with a specific fracture mechanism during sideways fall, where failure initiated on the superior-lateral side (compression) followed by later failure of the medially (tension), often exhibiting a two-peak force-displacement curve.

QCT-based failure analysis of proximal femurs under various loading orientations

In this paper, the variations of the failure strength and pattern of human proximal femur with loading orientation were analysed using a novel quantitative computed tomography (QCT)-based linear finite element (FE) method. The QCT images of 4 fresh-frozen femurs were directly converted into voxel-based finite element models for the analyses of the failure loads and patterns. A new geometrical reference system was used for the alignment of the mechanical loads on the femoral head. A new method was used for recognition and assortment of the high-risk elements using a strain energy-based measure. The FE results were validated with the experimental results of the same specimens and the results of similar case studies reported in the literature. The validated models were used for the computational investigation of the failure loads and patterns under 15 different loading conditions. A consistent variation of the failure loads and patterns was found for the 60 different analysed cases. Finally, it was shown that the proposed procedure can be used as a reliable tool for the failure analysis of proximal femurs, e.g. identification of the relevant loading directions for specific failure patterns, or determination of the loading conditions under which the proximal femurs are failure-prone.

Full-Field Strain Measurement During Mechanical Testing of the Human Femur at Physiologically Relevant Strain Rates

Understanding the mechanical properties of human femora is of great importance for the development of a reliable fracture criterion aimed at assessing fracture risk. Earlier ex vivo studies have been conducted by measuring strains on a limited set of locations using strain gauges (SGs). Digital image correlation (DIC) could instead be used to reconstruct the full-field strain pattern over the surface of the femur. The objective of this study was to measure the full-field strain response of cadaver femora tested at a physiological strain rate up to fracture in a configuration resembling single stance. The three cadaver femora were cleaned from soft tissues, and a white background paint was applied with a random black speckle pattern over the anterior surface. The mechanical tests were conducted up to fracture at a constant displacement rate of 15 mm/s, and two cameras recorded the event at 3000 frames per second. DIC was performed to retrieve the full-field displacement map, from which strains were derived. A low-pass filter was applied over the measured displacements before the crack opened in order to reduce the noise level. The noise levels were assessed using a dedicated control plate. Conversely, no filtering was applied at the frames close to fracture to get the maximum resolution. The specimens showed a linear behavior of the principal strains with respect to the applied force up to fracture. The strain rate was comparable to the values available in literature from in vivo measurements during daily activities. The cracks opened and fully propagated in less than 1 ms, and small regions with high values of the major principal strains could be spotted just a few frames before the crack opened. This corroborates the hypothesis of a strain-driven fracture mechanism in human bone. The data represent a comprehensive collection of full-field strains, both at physiological load levels and up to fracture. About 10,000 points were tracked on each bone, providing superior spatial resolution compared to ∼15 measurements typically collected using SGs. These experimental data collection can be further used for validation of numerical models, and for experimental verification of bone constitutive laws and fracture criteria.

Changes in strain patterns after implantation of a short stem with metaphyseal anchorage compared to a standard stem: an experimental study in synthetic bone

Short stem hip arthroplasties with predominantly metaphyseal fixation, such as the METHA® stem (Aesculap, Tuttlingen, Germany), are recommended because they are presumed to allow a more physiologic load transfer and thus a reduction of stress-shielding. However, the hypothesized metaphyseal anchorage associated with the aforementioned benefits still needs to be verified. Therefore, the METHA short stem and the Bicontact® standard stem (Aesculap, Tuttlingen, Germany) were tested biomechanically in synthetic femora while strain gauges monitored their corresponding strain patterns. For the METHA stem, the strains in all tested locations including the region of the calcar (87% of the non-implanted femur) were similar to conditions of synthetic bone without implanted stem. The Bicontact stem showed approximately the level of strain of the non-implanted femur on the lateral and medial aspect in the proximal diaphysis of the femur. On the anterior and posterior aspect of the proximal metaphysis the strains reached averages of 78% and 87% of the non-implanted femur, respectively. This study revealed primary metaphyseal anchorage of the METHA short stem, as opposed to a metaphyseal-diaphyseal anchorage of the Bicontact stem.

DIFFERENCES BETWEEN CONTRALATERAL BONES OF THE HUMAN LOWER LIMBS: A MULTISCALE INVESTIGATION

This study addressed side asymmetry between human lower limb long bones.

A multiscale approach was taken to investigate differences between contralateral femurs, tibias and fibulas, at body-level (total-body CT-scans, anatomical dissection), organ-level (volume and moments of areas; structural stiffness and strain distribution in bending and torsions) and tissue-level (mineral density, elastic modulus, hardness).

Because of the large amount of measurements taken, the study was limited to two donors. However, high statistical power within the same donor was achieved thanks to a large number of highly-repeatable measurements. Muscle cross-sections suggested that both donors were right-legged. The right bones had higher structural stiffness (up to +115%, statistically significant, except for the tibia). The right bones also experienced generally lower strain than the contralateral ones (up to -25%, statistically significant). The right bones had larger volume (up to +16%) and moments of area (up to +116%, statistically significant in most cases) than the left ones. Difference in tissue density between contralateral bones (< 7%) was not statistically significant in most cases. Also the differences found in elastic modulus of the femur cortical tissue (2–5%) were not statistically significant. Similarly, tissue hardness in the right bones was only marginally higher than in the contralateral ones (+1% to +4%, not statistically significant). Therefore, it seems that structural differences between contralateral bones associated with laterality are mainly explained by differences in bone quantity (volume) and organization (area moments). Bone tissue quality (density, hardness) seems to give a marginal contribution to structural side asymmetry.

A NEW PARADIGM FOR THE IN VITRO SIMULATION OF SIDEWAYS FALL LOADING OF THE PROXIMAL HUMAN FEMUR

Although the direction of loads applied to the proximal human femur is unpredictable during sideways fall, most in vitro and numerical simulations refer to a single loading condition (15° internal rotation; 10° adduction), which has been anecdotally suggested in the 1950s. The aim of the present study was to improve in vitro simulations of sideways falls on the proximal femur.

An in vitro setup was developed that allowed exploring a range of loading directions +/-90° internal–external rotation; 0°–50° adduction). To enable accurate control of the loading conditions (direction and magnitude of all load components applied to the femur), the setup included a number of low-friction linear and rotary bearings. The setup was instrumented with an axial and a torsional load cell, three displacement transducers and a rotation transducer to monitor the most significant components of load/displacement during testing. The strain distribution was measured on the bone surface (16 triaxial strain gauges, 2,000 Hz). Fracture was recorded with a high-speed camera.

The setup was successfully tested on a cadaveric femur non-destructively (12 loading configurations) and destructively (15° internal rotation; 10° adduction). All measurements were highly repeatable (the displacements of the femoral head varied by < 2% between repetitions; the tilt in the frontal plane by < 0.05°; and strain varied on average 0.34% between repetitions). The displacement of the femoral head varied by over 50% when the same force was applied in different directions. Principal strains at the same location varied by over 70%, depending on the direction of the applied force. The high-speed video enabled the identification of the point of fracture initiation.

This study has shown that a new paradigm for testing the proximal femur (including improved testing conditions and a variety of loading configurations) can provide more accurate and more extensive information about the state of strain.

EXPERIMENTAL METHODS FOR THE BIOMECHANICAL INVESTIGATION OF THE HUMAN SPINE: A REVIEW

In vitro mechanical testing of spinal specimens is extremely important to better understand the biomechanics of the healthy and diseased spine, fracture, and to test/optimize surgical treatment. While spinal testing has extensively been carried out in the past four decades, testing methods are quite diverse. This paper aims to provide a critical overview of the in vitro methods for mechanical testing the human spine at different scales. Specimens of different type are used, according to the aim of the study: spine segments (two or more adjacent vertebrae) are used both to investigate the spine kinematics, and the mechanical properties of the spine components (vertebrae, ligaments, discs); single vertebrae (whole vertebra, isolated vertebral body, or vertebral body without endplates) are used to investigate the structural properties of the vertebra itself; core specimens are extracted to test the mechanical properties of the trabecular bone at the tissue-level; mechanical properties of spine soft tissue (discs, ligaments, spinal cord) are measured on isolated elements, or on tissue specimens. Identification of consistent reference frames is still a debated issue. Testing conditions feature different pre-conditioning and loading rates, depending on the simulated action. Tissue specimen preservation is a very critical issue, affecting test results. Animal models are often used as a surrogate. However, because of different structure and anatomy, extreme caution is required when extrapolating to the human spine. In vitro loading conditions should be based on reliable in vivo data. Because of the high complexity of the spine, such information (either through instrumented implants or through numerical modeling) is currently unsatisfactory. Because of the increasing ability of computational models in predicting biomechanical properties of musculoskeletal structures, a synergy is possible (and desirable) between in vitro experiments and numerical modeling. Future perspectives in spine testing include integration of mechanical and structural properties at different dimensional scales (from the whole-body-level down to the tissue-level) so that organ-level models (which are used to predict the most relevant phenomena such as fracture) include information from all dimensional scales.

Strain distribution in the lumbar vertebrae under different loading configurations

BACKGROUND CONTEXT

The stress/strain distribution in the human vertebrae has seldom been measured, and only for a limited number of loading scenarios, at few locations on the bone surface.

PURPOSE

This in vitro study aimed at measuring how strain varies on the surface of the lumbar vertebral body and how such strain pattern depends on the loading conditions.

METHODS

Eight cadaveric specimens were instrumented with eight triaxial strain gauges each to measure the magnitude and direction of principal strains in the vertebral body. Each vertebra was tested in a three adjacent vertebrae segment fashion. The loading configurations included a compressive force aligned with the vertebral body but also tilted (15°) in each direction in the frontal and sagittal planes, a traction force, and torsion (both directions). Each loading configuration was tested six times on each specimen.

RESULTS

The strain magnitude varied significantly between strain measurement locations. The strain distribution varied significantly when different loading conditions were applied (compression vs. torsion vs. traction). The strain distribution when the compressive force was tilted by 15° was also significantly different from the axial compression. Strains were minimal when the compressive force was applied coaxial with the vertebral body, compared with all other loading configurations. Also, strain was significantly more uniform for the axial compression, compared with all other loading configurations. Principal strains were aligned within 19° to the axis of the vertebral body for axial-compression and axial-traction. Conversely, when the applied force was tilted by 15°, the direction of principal strain varied by a much larger angle (15° to 28°).

CONCLUSIONS

This is the first time, to our knowledge, that the strain distribution in the vertebral body is measured for such a variety of loading configurations and a large number of strain sensors. The present findings suggest that the structure of the vertebral body is optimized to sustain compressive forces, whereas even a small tilt angle makes the vertebral structure work under suboptimal conditions.

Simulated transversus abdominis muscle force does not increase stiffness of the pubic symphysis and innominate bone: an in vitro study

BACKGROUND

The transversus abdominis muscle is thought to exert a stiffening effect on the sacroiliac joints. However, it is unknown whether this muscle is capable of increasing pubic symphysis and innominate bone stiffness during load exerted on the pelvis. The objective of this study is to investigate whether in vitro simulated force of transversely oriented fibres of the transversus abdominis increases stiffness of the pubic symphysis and innominate bone.

METHODS

In 15 embalmed specimens an incremental moment was applied in the sagittal plane to one innominate with respect to the fixated contralateral innominate. For pubic symphysis motion and innominate bone deformation load-deformation curves were plotted and slopes of adjusted linear regression lines were calculated. The slopes are considered to be a measure of pubic symphysis and innominate bone stiffness. Slopes were tested for significant differences before and after simulation of the transversus abdominis force.

FINDINGS

Stiffness of pubic symphysis and innominate bone does not change under influence of simulated force of the transversus abdominis. For pubic symphysis, the slope of the regression line hardly changes, from 0.0341mm/Nm (SD 0.0277) before transversus abdominis force simulation to 0.0342mm/Nm (SD 0.0273) during simulation. For innominate bone, the mean slope increases minimally, from 0.0368mm/Nm (SD 0.0369) to 0.0413mm/Nm (SD 0.0395), respectively.

INTERPRETATION

Simulation of the force of a single muscle - transversus abdominis - does not increase stiffness of the pubic symphysis and innominate bone. The hypothesized stiffening influence of the transversus abdominis on the pelvic ring was not confirmed in vitro.

MECHANICAL PROPERTIES OF THE HUMAN METATARSAL BONES

Despite the incidence of metatarsal fractures and the associated risk of significant disability, little is known about the biomechanical properties (strength and stiffness) of metatarsal bones. In most cases a single metatarsal bone (first, second and fifth) has been investigated. An extensive investigation of the biomechanical properties of the metatarsal bones is essential in the understanding and prevention of metatarsal injuries. Entire sets of metatarsal bones from four feet were tested. The first foot was used to fine-tune the testing set-ups. To measure the stiffness, each metatarsal bone was subjected to non-destructive four-point-bending in the sagittal and transverse planes, axial compression and torsion. Strain was measured at two locations. To measure the strength, each metatarsal bone was tested to failure in torsion. Significant differences (p < 0.0001) existed among the stiffness of the five metatarsal bones: (i) in torsion the first metatarsal bone was 2–3 times stiffer than the others; (ii) in four-point-bending and axial compression this difference was less pronounced than in torsion; (iii) differences were smaller among the other metatarsal bones; (iv) the second metatarsal bone was less stiff than the third and fourth in bending. The second, third and fourth metatarsal bones were stiffer in the sagittal than in the transverse plane (p < 0.0001). Conversely, there was no significant difference between the two planes of bending for the first and fifth bones. During destructive testing, all metatarsal bones exhibited a linear elastic behavior and brittle failure. The torsional strength at failure ranged between 1.9 Nm and 6.9 Nm. The first metatarsal bone was stronger than all the others. Stiffness in different loading conditions and failure were measured and compared for all metatarsal bones. These data corroborate previous biomechanical studies concerning the role and load sharing of the different metatarsal bones.

Human bone hardness seems to depend on tissue type but not on anatomical site in the long bones of an old subject

It has been hypothesised that among different human subjects, the bone tissue quality varies as a function of the bone segment morphology. The aim of this study was to assess and compare the quality, evaluated in terms of hardness of packages of lamellae, of cortical and trabecular bones, at different anatomical sites within the human skeleton. The contralateral six long bones of an old human subject were indented at different levels along the diaphysis and at both epiphyses of each bone. Hardness value, which is correlated to the degree of mineralisation, of both cortical and trabecular bone tissues was calculated for each indentation location. It was found that the cortical bone tissue was harder (+18%) than the trabecular one. In general, the bone hardness was found to be locally highly heterogeneous. In fact, considering one single slice obtained for a bone segment, the coefficient of variation of the hardness values was up to 12% for cortical bone and up to 17% for trabecular bone. However, the tissue hardness was on average quite homogeneous within and among the long bones of the studied donor, although differences up to 9% among levels and up to 7% among bone segments were found. These findings seem not to support the mentioned hypothesis, at least not for the long bones of an old subject.

Is bone loss the reversal of bone accrual? Evidence from a cross-sectional study in daughter-mother-grandmother trios

Bone adapts to mechanical loads applied on it. During aging, loads decrease to a greater extent at those skeletal sites where loads increase most in earlier life. Thus, the loss of bone may occur preferentially at sites where most bone has been deposited previously; ie, bone loss could be the directional reversal of accrual. To test this hypothesis, we compared the bone mass distribution at weight-bearing (tibia) and non-weight-bearing (radius) bones among 18-year-old girls, their premenopausal mothers, and their postmenopausal maternal grandmothers. Bone and muscle properties were measured by pQCT, and polar distribution of bone mass was obtained in 55 girl-mother-maternal grandmother trios. Site-matched differences in bone mass were compared among three generations. The differences between girls and mothers and between mothers and grandmothers were used to represent the patterns of bone mass accrual from early adulthood to middle age and bone loss from middle to old age, respectively. Compared to the mothers, 18-year old girls had less bone mass in the anterior and medial-posterior regions of the tibial shaft, while the grandmothers had less bone in the anterior and posterior regions. In contrast, the bone mass differences in the radial shaft between girls and mothers and mothers and grandmothers were relatively uniform. We conclude that both bone accrual and loss are direction-specific in weight-bearing bones but relatively uniform in non-weight-bearing bones. Bone loss in old age is largely, but not completely, a reversal of the preferential deposition of bone in the most highly loaded regions during early life.

Mechanical testing of bones: the positive synergy of finite-element models and in vitro experiments

Bone biomechanics have been extensively investigated in the past both with in vitro experiments and numerical models. In most cases either approach is chosen, without exploiting synergies. Both experiments and numerical models suffer from limitations relative to their accuracy and their respective fields of application. In vitro experiments can improve numerical models by: (i) preliminarily identifying the most relevant failure scenarios; (ii) improving the model identification with experimentally measured material properties; (iii) improving the model identification with accurately measured actual boundary conditions; and (iv) providing quantitative validation based on mechanical properties (strain, displacements) directly measured from physical specimens being tested in parallel with the modelling activity. Likewise, numerical models can improve in vitro experiments by: (i) identifying the most relevant loading configurations among a number of motor tasks that cannot be replicated in vitro; (ii) identifying acceptable simplifications for the in vitro simulation; (iii) optimizing the use of transducers to minimize errors and provide measurements at the most relevant locations; and (iv) exploring a variety of different conditions (material properties, interface, etc.) that would require enormous experimental effort. By reporting an example of successful investigation of the femur, we show how a combination of numerical modelling and controlled experiments within the same research team can be designed to create a virtuous circle where models are used to improve experiments, experiments are used to improve models and their combination synergistically provides more detailed and more reliable results than can be achieved with either approach singularly.