R.B. Martin

Toward a unifying theory of bone remodeling.

A theory is developed to resolve several inconsistencies between current concepts and observations about bone remodeling. For example, the observation that remodeling increases both when mechanical loading is excessively low, that is, in a disuse state, and when it is excessively high, producing substantial fatigue damage, is contrary to the widely held assumption that a signal generated by osteocytes in proportion to mechanical loading stimulates bone lining cells to activate remodeling. The new theory resolves this disparity by assuming that lining cells are inclined to activate remodeling unless restrained by an inhibitory signal, and that the mechanically provoked osteocytic signal serves this inhibitory function. Consequently, remodeling is elevated when signal generation declines due to reduced loading, or when signal generation or transmission is interrupted by damage due to excessive loading. Otherwise, remodeling is kept at a relatively low level by inhibitory signals produced through physiologic loading. Furthermore, the inhibitory signal is postulated to be identical to that proposed by Marotti as the mechanism for conversion of osteoblasts to osteocytes, and responsible for the diminishment of apposition rate during refilling of osteonal basic multicellular units. Consequently, a single, mechanically derived signal, produced in the osteocytic syncytium, may control osteoblast and bone lining cell functions, and thereby a variety of important phenomena in bone biology.

Bone ingrowth and mechanical properties of coralline hydroxyapatite 1 yr after implantation.

A previous study of coralline hydroxyapatite as a bone-graft substitute was extended from 4 to 12 months to determine better the relationships between implantation time, bone ingrowth and mechanical properties. The model consisted of a 10 x 30 mm window defect in the shaft of the canine radius (a cortical site), and a 10 mm diameter cylindrical defect in the head of the humerus (a cancellous site). In the new study, these two defects were made bilaterally in eight dogs, and filled with block-form coralline hydroxyapatite. The radius defects were supported by a metal fixation plate which was removed after 9 months. After 12 months, the dogs were killed and the left-side implants were analyzed histomorphometrically and mechanically. The right-side radius and humerus were reserved for structural analysis. The results were combined with those previously measured after 4, 8, 12 and 16 wk of implantation. In the cortical site, bone ingrowth increased from 52% at 16 wk to 74% at 1 yr. In the cancellous site, bone ingrowth was 38% after 4 wk, then fell monotonically, reaching 17% at 1 yr. Bending and compressive strength and stiffness of the radius implants increased throughout the post-implantation year, but compressive strength and stiffness of the humerus implants did not change after the first 2-4 months. Mechanical properties were strongly correlated to bone ingrowth in the cortical, but not the cancellous, site. The volume fraction of the coralline hydroxyapatite material diminished significantly with time in the cortical, but not the cancellous, site.

Effects of ovariectomy in beagle dogs

Beagle dogs 3-7 years old were ovariectomized (n = 9) or sham operated (n = 6) and followed for 48 weeks with measurements of body weight, tibial shaft bone mineral content (BMC), and serum biochemistry. Following killing, measurements were made of bone strength and histomorphometry. Ovariectomy (OX) significantly reduced serum estrone and estradiol concentrations and their variability from month to month. There was a transient decrease in cortical BMC of the OX dogs during the first 12 postoperative weeks but no difference between the groups after 48 weeks. Serum osteocalcin was elevated, but there was little effect on serum alkaline phosphatase, Ca, P, or calcitonin. OX increased the number of tetracycline-labeled osteons in cortical bone but reduced the percent trabecular surface labeled with tetracycline. OX produced no significant changes in the composition of the bones or loss of cortical area, but a statistically significant 15% trabecular bone loss occurred in the spine. However, bone strength had not been significantly affected at the time of sacrifice.

Relationships between marrow fat and bone turnover in ovariectomized and intact rats

It is known that marrow fat content increases in a variety of osteoporoses. This study sought to clarify this phenomenon by combining bone and marrow histomorphometry, and to determine (1) whether the marrow fat increase follows or precedes diminishment of bone volume, and (2) whether the increase in fat volume is due to adipocyte growth or proliferation. The relationship between marrow fat content and bone turnover was studied in the metaphysis and epiphysis of the proximal tibia in 20 ovariectomized and 20 intact 200 gm Sprague-Dawley rats. The results were examined after one and three months. The epiphysis had greater trabecular bone volume than the metaphysis. The metaphysis exhibited a decrease in bone volume with time; the epiphysis did not. Following ovariectomy, the bone volume diminished twice as much in the metaphysis as in the epiphysis. In the epiphysis, as bone volume fraction fell following ovariectomy, the vacated space was filled by both hemopoietic and adipose tissue. In the metaphysis, space-filling was time-dependent: hemopoietic tissue at one month and adipose tissue at three months. Marrow fat content increased with both time and ovariectomy in the metaphysis, but only with time in the epiphysis. Thus, marrow fat increased after bone volume began to decline in the metaphysis, and not at all in the epiphysis. Ovariectomy increased erosion surface in both the epiphysis and the metaphysis, but bone formation rate was increased only in the epiphysis. There was a reciprocal relationship between marrow fat content and bone formation rate.

COLLAGEN FIBER ORGANIZATION IS RELATED TO MECHANICAL PROPERTIES AND REMODELING IN EQUINE BONE. A COMPARSION OF TWO METHODS

We studied birefringence as an indicator of collagen fiber orientation in the diaphysis of the equine third metacarpal bone. We had previously shown that tissue from the lateral cortex of this bone is stronger monotonically, but less fatigue resistant, than tissue from the medial and dorsal regions. To learn whether collagen fiber orientation might play a role in this regional specialization, we tested three hypotheses using the same specimens: (1) collagen fiber orientation is regionally dependent; (2) remodeling changes collagen fiber orientation; (3) longitudinal collagen fibers correlate positively with modulus and monotonic bending strength and negatively with flexural fatigue life. Beams (N = 36) cut parallel to the long axes of six pairs of bones had been tested to determine elastic modulus (N = 36), and fatigue life (N= 24) or monotonic strength (N = 12) in four-point bending. Subsequently, histologic cross-sections were prepared, and porosity, active remodeling and past remodeling were quantified. Birefringence was measured as an indicator of transverse collagen orientation using plane-polarized light (PPL), and again using circularly polarized light (CPL). The CPL measurement was less variable than the PPL measurement. Both birefringence measures indicated that collagen was more longitudinally oriented in the lateral cortex than in the other two cortices. Longitudinally disposed collagen correlated with greater modulus and monotonic strength, but did not correlate with fatigue life. Remodeling was associated with more transverse collagen. Neither measure of birefringence was significantly correlated with porosity. It was concluded that, in the equine cannon bone, longitudinal collage fiber orientation is regionally variable, contributes to increased modulus and strength but not fatigue life, and is reduced by osteonal remodeling.

CALCIUM BUFFERING IS REQUIRED TO MAINTAIN BONE STIFFNESS IN SALINE SOLUTION

This work determined whether mineral dissolution due to prolonged testing or storage of bone specimens in normal saline would alter their elastic modulus. In one experiment, small pieces of equine third metacarpal bone were soaked in normal saline supplemented with varying amounts of CaCl2. Changing Ca ion concentrations in the bath were monitored and the equilibrium concentration was determined. In a second experiment, the elastic moduli of twenty 4 x 10 x 100 mm equine third metacarpal beams were determined non-destructively in four-point bending. Half the beams were then soaked for 10 days in normal saline, and the other half in saline buffered to the bone mineral equilibrium point with Ca ions. Modulus measurements were repeated at 6 and 10 days. The equilibrium Ca ion concentration for bone specimens was found to be 57.5 mg l-1. The modulus of bone specimens soaked in normal saline significantly diminished 2.4%, whereas the modulus of those soaked in calcium-buffered saline did not change significantly.

Alendronate affects long bone length and growth plate morphology in the oim mouse model for Osteogenesis Imperfecta

Alendronate, a bisphosphonate drug, has shown promise in reducing remodeling and bone loss in postmenopausal osteoporosis. Alendronate acts directly on the osteoclast, inhibiting its resorption capability. This inhibition of osteoclast activity has led to the use of bisphosphonates in the treatment of the osteogenesis imperfecta condition. Treatment of osteogenesis imperfecta with bisphosphonates enhances bone strength, but the consequences on linear bone growth are not well defined. Using the oim mouse model for type III osteogenesis imperfecta, two doses of alendronate, low (0.125 mg/kg/wk) and high (2.5 mg/kg/wk) were administered weekly via intraperitoneal injection starting at 4 weeks of age and ending at 12 weeks of age to assess the effects of alendronate on humerus and ulna length. The higher dose of alendronate reduced humerus and ulna length in the oim/wt and wt/wt genotypes for both sexes (P < 0.05). The oim/oim humerus and ulna were not significantly affected by the higher dose of alendronate in females, but reduced bone length in males (P < 0.0085). Proximal humerus growth plate area was affected by both genotype and alendronate dose and growth plate diameter was increased at the chondro-osseous junction by both alendronate doses (P < 0.011). Genotype and alendronate dose affected growth plate height. The oim/oim genotype displayed taller growth plates. The high dosage of alendronate increased overall growth plate height, particularly within the hypertrophic zone, which suggests a failure of vascular invasion-induced apoptosis in the hypertrophic cells. In conclusion, these results indicate that high doses of alendronate (>2.5 mg/kg/wk) inhibit long bone length in mice through alteration of the growth plate and possibly reduced resorption at the chondro-osseous junction.

Effects of bone ingrowth on the strength and non-invasive assessment of a coralline hydroxyapatite material.

The dependence of strength on the amount of bone growth into a hydroxyapatite material made from coral was investigated. Block and granular forms of the material were implanted into cortical and trabecular regions of the skeletons of 16 dogs. The results were examined after 4, 8, 12 and 16 wk, with four dogs in each experimental group. When implanted into cortical bone, the bending strength of the implant material was found to be highly correlated with the amount of pore space which had become occupied by bone (r = 0.92, P less than 0.005 for the block form; r = 0.84, P less than 0.005 for the granular form). Multiple regression analysis showed that six histomorphometric measures of ingrowth accounted for 96% of the variability in bending strength of the block material, and there were no significant differences between block and granular forms of the material. On the other hand, when implanted into trabecular bone, the block form of the material achieved greater compressive strength than the granular form. While both strength and ingrowth increased with time, there were poor correlations between these two variables. Finally, when the material is implanted into trabecular bone, it becomes stronger in compression than the surrounding bone; when implanted in cortical bone, linear modelling suggests that resorption and replacement of the implant would be required to approximate the bending strength of the surrounding bone.

Osteon pullout in the equine third metacarpal bone: effects of ex vivo fatigue

An important concept in bone mechanics is that osteons influence mechanical properties in several ways, including contributing to toughness and fatigue strength by debonding from the interstitial matrix so as to „bridge”︁ developing cracks. Observations of „pulled out„ osteons on fracture surfaces are thought to be indicative of such behavior. We tested the hypothesis that osteon pullout varies with mode of loading (fatigue vs. monotonic), cortical region, elastic modulus, and fatigue life. Mid‐diaphseal beams from the dorsal, medial, and lateral regions of the equine third metacarpal bone were fractured in four point bending by monotonic loading to failure under deflection control, with or without 105 cycles of previous fatigue loading producing 5000 microstrain (15–20% of the expected failure strain) on the first cycle; or sinusoidal fatigue loading to failure, under load or deflection control, with the initial cycle producing 10,000 microstrain (30–40% of the expected failure strain). Using scanning electron microscopy, percent fracture surface area exhibiting osteon pullout (%OP.Ar) was measured. Monotonically loaded specimens and the compression side of fatigue fracture surfaces exhibited no osteon pullout. In load‐controlled fatigue, pullout was present on the tension side of fracture surfaces, was regionally dependent (occurring to a greater amount dorsally), and was correlated negatively with elastic modulus and positively with fatigue life. Regional variation in %OP.Ar was also significant for the pooled (load and deflection controlled) fatigue specimens. %OP.Ar was nearly significantly greater in deflection controlled fatigue specimens than in load‐controlled specimens (p < 0.059). The data suggest that tensile fatigue loading of cortical bone eventually introduces damage that results in osteonal debonding and pullout, which is also associated with increased fatigue life via mechanisms that are not yet clear.

Does osteocyte formation cause the nonlinear refilling of osteons

Marotti (Ital J Anat Embryol 1 01:25-79, 1996) described a theory of osteocyte differentiation from osteoblasts during bone formation. This theory postulates that, when a previously formed osteocyte is sufficiently covered by new bone and osteoid, it sends an inhibitory signal through its dendritic processes to the neighboring osteoblasts that reduces their individual apposition rates. The osteoblast most affected by this inhibition becomes buried by its neighbors, and becomes one of the next layer of osteocytes. By pursuing this concept, the present study develops a mathematical theory that predicts another observation about bone remodeling: the diminishment of the apposition rate during the refilling of basic multicellular units (BMUs). This decrease in apposition rate is different in osteonal and surface (e.g., trabecular) BMUs, and the theory shows that this result is consistent with the accrual of osteocyte inhibition throughout the refilling period, with the different ratios of bone volume to surface area in these two types of BMUs.