Interfacial Strength of Cement Lines in Human Cortical Bone

In human cortical bone, cement lines (or reversal lines) separate osteons from the interstitial bone tissue, which consists of remnants of primary lamellar bone or fragments of remodeled osteons. There have been experimental evidences of the cement line involvement in the failure process of bone such as fatigue and damage. However, there are almost no experimental data on interfacial properties of cement lines in human cortical bone. The objective of this study is to design and assemble a precision and computer controlled osteon pushout microtesting system, and to experimentally determine the interfacial strength of cement lines in human cortical bone by performing osteon pushout tests. Thirty specimens were prepared from humeral diaphyses of four human subjects. Twenty specimens were tested under the condition of a small hole in the supporting plate, in which the cement line debonding occurred. The cement line interfacial strength ranged from 5.38 MPa to 10.85 MPa with an average of 7.31±1.73 MPa. On the other hand, ten specimens were tested under the condition of a large hole in the supporting plate, in which the shear failure inside osteons was observed. The specimens tested under the condition of the large hole resulted in an average shear strength of 73.71±15.06 MPa, ranging from 45.97 MPa to 93.74 MPa. Therefore, our results suggest that the cement line interface between osteon and interstitial bone tissue is weaker than that between bone tissue lamellae.     

Geometric determinants to cement line debonding and osteonal lamellae failure in osteon pushout tests

Cement lines are the boundaries between secondary osteons and the surrounding interstitial bone matrix in cortical bone. The interfacial properties of cement lines have been determined by osteon pushout tests. However, distinctively different material properties were obtained when osteon pushout tests were performed under different test geometries. In the present study, an axisymmetric two-dimensional finite element model was used to simulate an osteon pushout test using the test geometry of actual experiments. The results indicated that shear failure within the osteonal lamellae would occur when the osteon pushout test was performed under the condition of a thick specimen and large supporting hole. On the other hand, cement line debonding occurred when the osteon pushout test was performed using a thin specimen and small supporting hole. The finite element results were consistent with previous experiments of osteon pushout tests under different test geometries. Furthermore, the finite-element results suggest that a smoothly curved punch would most likely cause debonding at the cement line instead of osteonal lamellae.     

Secondary osteon size and collagen/lamellar organization (“osteon morphotypes”) are not coupled,but potentially adapt independently for local strain mode or magnitude

In bone, matrix slippage that occurs at cement lines of secondary osteons during loading is an important toughening mechanism. Toughness can also be enhanced by modifications in osteon cross-sectional size (diameter) for specific load environments; for example, smaller osteons in more highly strained “compression” regions vs. larger osteons in less strained “tension” regions. Additional osteon characteristics that enhance toughness are distinctive variations in collagen/lamellar organization (i.e., “osteon morphotypes”). Interactions might exist between osteon diameter and morphotype that represent adaptations for resisting deleterious shear stresses that occur at the cement line. This may be why osteons often have a peripheral ring (or “hoop”) of highly oblique/transverse collagen. We hypothesized that well developed/distinct “hoops” are compensatory adaptations in cases where increased osteon diameter is mechanically advantageous (e.g., larger osteons in “tension” regions would have well developed/distinct “hoops” in order to resist deleterious consequences of co-existing localized shear stresses). We tested this hypothesis by determining if there are correlations between osteon diameters and strongly hooped morphotypes in “tension”, “compression”, and “neutral axis” regions of femora (chimpanzees, humans), radii (horse, sheep) and calcanei (horse, deer). The results reject the hypothesis—larger osteons are not associated with well developed/distinct “hoops”, even in “tension regions” where the effect was expected to be obvious. Although osteon diameter and morphotype are not coupled, osteon diameters seem to be associated with increased strain magnitudes in some cases, but this is inconsistent. By contrast, osteon morphotypes are more strongly correlated with the distribution of tension and compression.     

Equine cortical bone exhibits rising R-curve fracture mechanics

Previous studies of the fracture properties of cortical bone have suggested that the fracture toughness increases with crack length, which is indicative of rising R-curve behavior. Based on this indirect evidence and the similarity of bone to ceramic matrix composites, we hypothesized that bone would exhibit rising R-curve behavior in the transverse orientation and that the characteristics of the R-curves would be regionally dependent within the cortex due to variations in bone microstructure and toughening mechanisms. To test these hypotheses, we conducted R-curve experiments on specimens from equine third metacarpal bones using standard fracture mechanics testing methods. Compact type specimens from the dorsal and lateral regions in the middle of the diaphysis were oriented for crack propagation transverse to the longitudinal axis of the bone.The test results demonstrate that equine cortical bone exhibits rising R-curve behavior during transverse crack propagation as hypothesized. Statistical analyses of the crack growth initiation toughness, K0, the peak toughness, Kpeak, and the crack extension at peak toughness, deltaa, revealed significant regional differences in these characteristics. Specifically, the lateral cortex displayed higher crack growth initiation and peak toughnesses. The dorsal cortex exhibited greater crack extension at the peak of crack growth resistance. Scanning electron microscopy revealed osteon pullout on fracture surfaces from the dorsal cortex and but not in the lateral cortex. Taken together, the significant differences in R-curves and the SEM fractography indicate that the fracture mechanisms acting in equine cortical bone are regionally dependent.     

Osteonal effects on elastic modulus and fatigue life in equine bone

We hypothesized that recently formed, incompletely mineralized, and thus, relatively deformable osteons in the equine third metacarpus enhance in vitro load-controlled fatigue life in two ways. Macroscopically, there is a compliance effect, because reduced tissue elastic modulus diminishes the stress required to reach a given strain. Microscopically, there is a cement line effect, in which new osteons and their cement lines more effectively serve as barriers to crack propagation. We studied 18 4 x 10 x 100 mm beams from the medial, lateral, and dorsal cortices of metacarpal bones from 6 thoroughbred racehorses. Following load-controlled fatigue testing to fracture in 4 point bending, a transverse, 100 microm thick, basic fuchsin-stained cross-section was taken from the load-bearing region. The number and diameter of all intact (and thus recently formed/compliant) secondary osteons in a 3.8 x 3.8 mm region in the center of the section were determined. The associated area fraction and cement line length of intact osteons were calculated, and the relationships between these variables, elastic modulus (E), and the logarithm of fatigue life (logN(F)) were analyzed. As expected, logN(F) was negatively correlated with E, which was in turn negatively correlated with intact osteon area fraction and density. (LogN(F))/E increased in proportion to intact osteon density and nonlinearly with cement line density (mm/mm(2)). These results support the hypothesis that remodeling extends load-controlled fatigue life both through the creation of osteonal barriers to microdamage propagation and modulus reduction.     

Three-dimensional reconstruction of Haversian systems in ovine compact bone

Haversian systems or secondary osteons are an integral component of compact bone. However, as their exact shape is debatable, this study describes a technique to view their morphology in three dimensions. Bone remodeling in adult ovine long bones was labelled at intervals using a series of chelating fluorochromes. A series of longitudinal sections were cut at 25 microm intervals through blocks of the distal radius embedded in methylmethacrylate using a sledge macrotome. The chelating agents were used as markers of bone formation in the study of bone growth and osteon morphology. The two-dimensional image of each section was examined using an epifluorescence microscope. Images were transferred to a PC via a CCD low light colour video camera. Surface reference points were noted on each of the sections and, using computer software, a three-dimensional image of a refilling labelled osteon was reconstructed and its dimensions measured. Haversian systems may have a gentle spiral course along the longitudinal axis of the bone. They intertwine with adjacent osteons and give multiple branches along their course producing a complex pattern of organization. The mean labelled length and diameter of the osteons was 1.4 + 1 mm and 145 + 0.42 microm [Mean + S.D], respectively.     

Effects of growth on residual stress distribution along the radial depth of cortical cylinders from bovine femurs

Residual stress is defined as the stress that remains in bone tissue without any external forces. This study investigated the effects of growth on residual stress distributions from the surface to deeper regions of cortical cylinders obtained from less-than-one-month-old (Group Y) and two-year-old (Group M) bovine femurs. In these experiments, five diaphysis specimens from each group were used. Residual stress was measured using a high-energy synchrotron white X-ray beam to penetrate X-rays into the deeper region of the bone specimens. The measurements in the cortical cylinders from Groups Y and M were performed at 0.5- and 1-mm intervals, respectively, from the outer surface to the deeper region of the diaphysis specimens at four positions: anterior, posterior, lateral, and medial. The residual stress was calculated on the basis of variation in the interplanar spacing of hydroxyapatite crystals in the bone tissue. According to the results, the diaphysis specimens from Group Y were not subjected to large residual stresses (average −1.2 MPa and 2.4 MPa at the surface region and 1.5 mm depth, respectively). In Group M, the surface region of the diaphysis specimens was subjected to tensile residual stresses (average 6.7 MPa) and the deeper region was subjected to compressive stresses (average −8.2 MPa at 3 mm depth). There was a strong significant difference between both these regions. The value of residual stresses at the surface region of the diaphysis specimens in both the groups had a positive statistical correlation with the cortical thickness at the measured locations.     

Scaling of Haversian canal surface area to secondary osteon bone volume in ribs and limb bones

Studies of secondary osteons in ribs have provided a great deal of what is known about remodeling dynamics. Compared with limb bones, ribs are metabolically more active and sensitive to hormonal changes, and receive frequent low‐strain loading. Optimization for calcium exchange in rib osteons might be achieved without incurring a significant reduction in safety factor by disproportionally increasing central canal size with increased osteon size (positive allometry). By contrast, greater mechanical loads on limb bones might favor reducing deleterious consequences of intracortical porosity by decreasing osteon canal size with increased osteon size (negative allometry). Evidence of this metabolic/mechanical dichotomy between ribs and limb bones was sought by examining relationships between Haversian canal surface area (BS, osteon Haversian canal perimeter, HC.Pm) and bone volume (BV, osteonal wall area, B.Ar) in a broad size range of mature (quiescent) osteons from adult human limb bones and ribs (modern and medieval) and various adult and subadult non‐human limb bones and ribs. Reduced major axis (RMA) and least‐squares (LS) regressions of HC.Pm/B.Ar data show that rib and limb osteons cannot be distinguished by dimensional allometry of these parameters. Although four of the five rib groups showed positive allometry in terms of the RMA slopes, nearly 50% of the adult limb bone groups also showed positive allometry when negative allometry was expected. Consequently, our results fail to provide clear evidence that BS/BV scaling reflects a rib versus limb bone dichotomy whereby calcium exchange might be preferentially enhanced in rib osteons. Am J Phys Anthropol 151:230–244, 2013. © 2013 Wiley Periodicals, Inc.     

Mechanical hysteresis loops from single osteons: technical devices and preliminary results

The purpose of this paper is to describe an original apparatus for recording hysteresis loops from single osteons and a special microgrinding lathe for preparing osteonic samples for testing. The results obtained by testing isolated osteons, and composite bone samples ground to the same size as an osteon sample justify one to draw the following main conclusions: at an equal degree of calcification, longitudinal osteons show larger hysteresis loops under compression and alternate osteons show larger hysteresis loops under tension; there seems to be little chance of acquiring detailed information on the mechanical effects of osteon calcification recording hysteresis loops; collagen orientation in lamellae is the main factor determining the kind of hysteresis loops displayed by a composite bone sample.     

Orientation of collagen at the osteocyte lacunae in human secondary osteons

This work characterizes an aspect of human bone micro-structure, pertinent to fracture initiation and arrest. It addresses how the orientation of elementary components proximate to osteocyte lacunae influences secondary osteon micro-biomechanics. New data at the perilacunar region concerning orientation of collagen-apatite, and prior data on collagen orientation outside the perilacunar region, are incorporated in a novel simulation of osteons to investigate how orientation relates to strains and stresses during mechanical testing. The perilacunar region was observed by confocal microscopy within single lamellar specimens, isolated from osteons. The specimens were separated by extinct or bright appearance in transverse section under circularly polarizing light. This is because synchrotron diffraction and confocal microscopy had established that each type, away from the perilacunar region, corresponds to specific dominant collagen orientation (extinct lamellae's dominant collagen forming small angles with the original osteon axis, while the bright lamellae's forms larger angles). Morphometry of serial confocal images of each perilacunar region showed collagen orientation generally following the orientation of canaliculi, circumambiently-perpendicular to the lacuna. The lacunae tilted relative to the lamellar walls were more numerous in extinct than in bright lamella. Their apices were less likely in extinct than bright lamella to show collagen following the canalicular orientation. The simulation of osteocyte lacunae in osteons, under tension or compression loading, supports the hypothesis that collagen orientation affects strains and stresses at the equatorial perilacunar region in conjunction with the presence of the lacuna. We further conjecture that collagen orientation diverts propagation of micro-cracks initiating from apices.     

Hysteretic pinching of human secondary osteons subjected to torsion

The mechanical behavior of bone tissue's ultra- and micro- structure is fundamental to assessment of macroscopic bone mechanics. This paper explores the ultra-structural characteristics of human femoral tissue responsible for energy absorption of secondary osteons under mechanical loading. A novel mathematical interpretation of single osteon mechanics elucidates the behavior of the collagen-apatite interface. Fully calcified single osteon specimens were mechanically tested quasi-statically under cyclic torsional loading about their longitudinal axis. On each hysteretic diagram, all cycles after the initial monotonic cycle appear pinched and share two points. Stiffness degradation and pinching degradation were investigated on the torque versus deflection-angle-per-unit-length diagrams as the number of cycles increases, in relation to the appearance of osteons in cross-section under circularly polarized light microscopy. Material science's Bauschinger effect, originally defined for metals and later extended to structures reinforced with metal bars, is adapted to describe pinching. Material science's prying effect, defined as amplification of eccentric tensile load through lever action, is employed to explain pinching. The presence of the two points shared by all complete cycles is analyzed in terms of the mathematical fixed point theorem. The results allow formulation of the following conjectures: (1) the prying of carbonated apatite crystallites at the interface with the 40 nm long bands of non-calcified collagen fibrils causes pinching; (2) the prying effect increases with the increasing percentage of collagen-apatite elements that form a larger angle with the osteon axis; and (3) micro-cracks increase more in number than in length as the number of cycles increases.     

Load transfer along the bone–dental implant interface

In this paper the variation of normal and shear stresses along a path defined on the bone–dental implant interface is investigated. In particular, the effects of implant diameter, collar length and slope, body length, and the effects of four different types of external threads on the interfacial stress distribution are studied. The geometry of the bone is digitized from a CT scan of a mandibular incisor and the surrounding bone. The bone and the implant are assumed to be perfectly bonded. The finite element method with 2D plane strain assumption is used to compute interfacial stresses. Highest continuous interfacial stresses are encountered in the region where the implant collar engages the cortical region, and near the apex of the implant in the subcortical region. Stress concentrations in the interfacial stresses occur near the geometric discontinuities on the implant contour, and jumps in stress values occur where the elastic modulus of the bone transitions between the cortical and trabecular bone values. Among the six contour parameters, the slope and the length of the implant collar, and the implant diameter influence the interfacial stress levels the most, and the effects of changing these parameters are significantly noticed only in the cortical bone (alveolar ridge) area. External threads cause significant stress concentrations in interfacial stresses in otherwise smoothly varying regions. This work shows that the presence of external threads could cause significant variations in both normal and shear stresses along the bone–implant interface, but not reduction in shear stress as previously thought.     

Effects of the basic multicellular unit and lamellar thickness on osteonal fatigue life

A remodeling cycle sets the size of the osteon and associated lamellae in the basic multicellular unit. Treatments and aging affect these micro-structural features. We previously demonstrated decreased fatigue life with an unexplained mechanism and decreased osteon size in cortical bone treated with high-dose bisphosphonate. Here, three finite element models were examined: type-1: a single osteon, as a homogeneous unit and with heterogeneous lamellae and interlamellae, type-2: a control, interstitial-only tissue and type-3: the osteon with cement line, set within the interstitial tissue. Models were loaded in simulated, sinusoidal bending fatigue. As osteon size was decreased, lamellar number and lamellar thickness were incrementally adjusted for each model. As hypothesized, lamellae within the larger type-1 models attained greater cycles to failure and the addition of an osteon to type-2 models (generating a type-3 model set) yielded increased fatigue life. However, as the osteon size was decreased, the potential for compressive damage nucleation was increased within the lamellae of the osteons versus the interstitium. Also, osteons with fewer, thicker lamellae displayed increased fatigue life. Osteonal microstructure plays a role in damage initiation location, especially when BMU size is smaller. Previous findings by us and others could partially be explained by this further understanding of increased probability for damage nucleation in smaller osteons.     

Percent osteonal bone versus osteon counts: the variable of choice for estimating age at death.

Ahlqvist and Damsten's (1969) modification of the Kerley (1965) method for histological age estimation uses percent osteonal bone, rather than actual osteon counts, in order to eliminate the difficulty of distinguishing between intact and fragmentary osteons. Since their method has been developed for the femur only, and several more recent methods have been proposed that utilize percent osteonal bone, a study was undertaken to ascertain the relative value of percent osteonal bone compared with osteon counts to estimate age at death for the radius, tibia, and fibula. First the question of how much of the cross-section of a bone should be sampled was addressed by comparing the results of regression against age for percent osteonal bone derived from sampling only four fields with those derived from the entire cross-section of the radius. A significant age association was found only when the entire cross-section was sampled. In order to evaluate the relative merit of using either percent osteonal bone, or osteon counts to estimate age, each variable was regressed against age. Significant correlation coefficients were found for all three bones when the independent variable was osteon counts. When percent osteonal bone was employed, a significant correlation was found only for the radius. Stepwise linear regression found osteon counts for the fibula alone to be the best age predictor. Finally, a repeated measures analysis of variance revealed that percent osteonal bone and osteon counts both differ among the three bones within an individual. Based upon these results, osteon counts, rather than percent osteonal bone, should be the variable of choice when developing histological age predicting methods.     

Residual stress distribution in rabbit limb bones

The presence of the residual stresses in bone tissue has been noted and the authors have reported that there are residual stresses in bone tissue. The aim of our study is to measure the residual stress distribution in the cortical bone of the extremities of vertebrates and to describe the relationships with the osteon population density. The study used the rabbit limb bones (femur, tibia/fibula, humerus, and radius/ulna) and measured the residual stresses in the bone axial direction at anterior and posterior positions on the cortical surface. The osteons at the sections at the measurement positions were observed by microscopy. As a result, the average stresses at the hindlimb bones and the forelimb bones were 210 and 149 MPa, respectively. In the femur, humerus, and radius/ulna, the residual stresses at the anterior position were larger than those at the posterior position, while in the tibia, the stress at the posterior position was larger than that at the anterior position. Further, in the femur and humerus, the osteon population densities in the anterior positions were larger than those in the posterior positions. In the tibia, the osteon population density in the posterior position was larger than that in the anterior position. Therefore, tensile residual stresses were observed at every measurement position in the rabbit limb bones and the value of residual stress correlated with the osteon population density (r=0.55, P<0.01).     

Rib remodeling dynamics in a skeletal population from Kulubnarti, Nubia

Bone remodeling variables in the rib were analyzed for a skeletal population of medieval antiquity (ca. A.D. 550-1450) from Kulubnarti, in Sudanese Nubia. The skeletal remains are naturally mummified and in an excellent state of preservation. The study sample consists of thin sections from the ribs of 80 individuals, ranging in age from 15-50+ years. Ribs were examined using a standard microscope and image analysis software. Numbers of intact osteons, fragmentary osteons, forming osteons, and resorption spaces were counted, osteon and Haversian canal areas were measured, and several variables were calculated to assess morphometric and remodeling status in the rib. Variables calculated included mean annual activation frequency, mean bone formation rate, and net osteonal remodeling. Results indicate that age changes are consistent with those observed for other archaeological and modern samples. High numbers of resorption spaces in young males may reflect slower skeletal development in boys compared to girls. Comparisons of rib data with results of a previous study on patterns of femoral bone remodeling in the same population indicate that ribs have more osteons and higher bone formation rates compared to the femur. Also, sexual differences in osteon size observed in the femur were not observed in the rib. Activation frequency and bone formation rate are low in the Kulubnarti population compared to previously published data for a modern sample, a finding consistent with reported results from other archaeological samples. Genetic factors influencing the minimum effective strain setpoint and duration of skeletal maturation, in addition to repetitive high strains at Kulubnarti, may contribute to observed differences.     

The bending properties of single osteons

The bending properties of two fully calcified osteon types (longitudinal and alternate) have been investigated in 62 cylindrical samples by applying the technique of three-point bending loading. The bending of each sample was recorded using a microwave micrometer based on the cavity and pulse technique. It has been shown that alternate osteons are better able to withstand bending stress than longitudinal ones. This result offers a definitive explanation for the high concentration of transverse lamellae in pathologically bowed bone shaft.     

Composition of the cement line and its possible mechanical role as a local interface in human compact bone

Human compact bone may be viewed as a fiber reinforced composite material in which the secondary osteons act as the fiber reinforcements. The cement line, which is the interface between the 'fibers' (osteons) and extraosteonal bone matrix, may impart important mechanical properties to compact bone. The nature of these properties is not known partly because the composition of the cement line is unknown. This analysis examines the constituents of the osteon cement line using scanning electron microscopy and X-ray microprobe analysis to address its biomechanical functions as a local interface. The analysis suggests that the cement line is a region of reduced mineralization which may contain sulfated mucosubstances. This composition is consistent with the hypothesis that the cement line provides a relatively ductile interface with surrounding bone matrix, and that it provides the point specific stiffness differences, poor 'fiber'-matrix bonding and energy transfer qualities required to promote crack initiation but slow crack growth in compact bone.     

The system of bony canals as the basis for the angioarchitectonics of bones]

The spatial interrelationships of osteon canals were studied in corrosion preparations with the help of rastral electron microscopy. The structure of microvessels, their belonging to a definite link of the microcirculatory bed, the interaction of vessels and their position with respect to the osseous matrix were studied in bone sections impregnated with silver nitrate after V. V. Kuprijanov. Haversian canals in the compact substance of the bone are longitudinally oriented, can duplicate and form a single system of canals. The neighbouring canals of osteons might be bound by means of Volkmann's canals. The investigation of the Haversian canals in serial sections has shown that the diameter of the same canals of osteons can change at different levels, the diameter of the osteons themselves remaining unchanged. This seems to speak of uneven development of osteons in their different parts. In the Haversian canal there are one-two or occasionally three vessels having all three links of the morphocirculatory bed. The course and ramification of the vessel are identical to the shape of the osteon canal which includes them. The vessels are closely connected with the bony matrix by means of connective tissue bundles directed from the canal wall to the vessel wall. These bundles appear to serve as a peculiar anchor or amortizing apparatus and its elasticity might be a factor of a change of the shape and direction of the canal vessels in the bone development process.     

Regional structural and viscoelastic properties of fibrocartilage upon dynamic nanoindentation of the articular condyle

Fibrocartilage,a tissue with macromaterial properties between dense fibrous tissue and hyaline cartilage, is not well understood in its ultrastructure and regional viscoelastic properties. Here nanoindentation with atomic force microscopy was performed on fresh fibrocartilage samples of rabbit jaw joint condyles. Each sample was divided into anteromedial, anterolateral, posteromedial, and posterolateral regions for probing and topographic imaging in 2 x 2 microm and 10 x 10 microm scan sizes. Young's moduli differed significantly among these regions in a descending gradient from the anteromedial (2.34 +/- 0.26 MPa) to the posterolateral (0.95 +/- 0.06 MPa). The Poisson ratio, defined as lateral strain over axial strain, had the same gradient distribution: highest for the anteromedial region (0.46 +/- 0.05) and lowest for the posterolateral region (0.31 +/- 0.05). The same four regions showed a descending gradient of surface roughness: highest for the anteromedial (321.6 +/- 13.8 nm) and lowest for the posterolateral (155.6 +/- 12.6 nm). Thus, the regional ultrastructural and viscoelastic properties of fibrocartilage appear to be coregulated. Based on these region-specific gradient distributions, fibrocartilage is constructed to withstand tissue-borne shear stresses, which likely propagate across its different regions. A model of shear gradient and concentric gradient is proposed to describe the region-specific capacity of fibrocartilage to sustain shear stresses in tendons, ligaments, joints, and the healing bone across species.