Demineralized bone matrix as a biological scaffold for bone repair.

Experimental models were created in rat fibula to represent impaired bone healing so that biological deficiencies that cause bone repair to fail or to be delayed may be investigated. These models consist of a 4-mm-long segmental defect, created in rat fibula by osteotomy, and fitted with a 7-mm-long tubular specimen of demineralized bone matrix (DBM) over the cut ends of the fibula. The experiments in this study involved various modifications of the DBM scaffold designed to reduce its osteoinductive activity: steam sterilization (sDBM), ethylene oxide sterilization (eoDBM), trypsin digestion (tDBM), and guanidine hydrochloride extraction (gDBM). Bone healing was evaluated by bending rigidity of the fibula and mineral content of the repair site at 7 weeks post-surgery. The sDBM scaffolds resorbed completely by 7 weeks and hence this model was a nonhealing negative control. Rigidities in the unmodified DBM and tDBM groups were comparable, whereas in the gDBM and eoDBM groups it was significantly reduced. Histologically, in the 4-mm defects repaired with unmodified DBM, direct and endochondral bone formation in the scaffold and the defect resulted in a neocortex consisting of woven and lamellar bone uniting the broken bone by 7 weeks post-surgery. We conclude that the eoDBM and gDBM groups represent failure or delay of the bone repair process when compared with the unmodified DBM group in which the process is analogous to normal bone healing.

Repair of segmental bone defects in the rat: an experimental model of human fracture healing.

Bone repair models in animals may be considered relevant to human fracture healing to the extent that the sequence of events in the repair process in the model reflect the human fracture healing sequence. In the present study, the relevance of a recently developed segmental defect model in rat fibula to human fracture healing was investigated by evaluating temporal progression of rigidity of the fibula, mineral content of the repair site, and histological changes. In this model, a surgically created 2-mm-long defect was grafted with a 5-mm-long tubular specimen of demineralized bone matrix (DBM) by inserting it over the cut ends of the fibula. The temporal increase in rigidity of the healing fibula demonstrated a pattern similar to biomechanical healing curves measured in human fracture healing. This pattern was characterized by a short phase of rapidly rising rigidity during weeks 4-7 after surgery, associated with a sharp increase in the mineral content of the repair tissue. This was preceded by a phase of nearly zero rigidity and followed by a phase of slow rate of increase approaching a plateau. Histologically, chondroblastic and osteoblastic blastema originating from extraskeletal and subperiosteal (near fibula-graft junction) regions, infiltrated the DBM graft during the first 2 weeks. The DBM graft assumed the role of a "bridging callus." By weeks 6-8, most of the DBM was converted to new woven and trabecular bone with maximal osteoblastic activity and minimal endochondral ossification. Medullary callus formation started with direct new bone formation adjacent to the cortical and endosteal surfaces in the defect and undifferentiated cells in the center of the defect at 3 weeks. The usual bone repair process in rodents was altered by the presence of the DBM graft to recapitulate the sequential stages of human fracture healing, including the formation of a medullary callus, union with woven and lamellar bone, and recreation of the medullary canal.

Magnetic field induced inhibition of human osteosarcoma cells treated with adriamycin.

Morbidity resulting from the toxicity of chemotherapeutic drugs suggests that novel approaches are worthy of investigation. Development of the use of low intensity magnetic fields as an adjuvant to current treatment regimens to prevent metastatic disease may prove to be efficacious. Using a cell culture model, we have developed a magnetic field (MF) treatment that offers the possibility of lowering the therapeutic dose of these drugs and thereby reducing morbidity. Our studies have found that a low intensity (approximately 2 gauss) MF signal and a relatively low dose (0.1 microg/ml) of Adriamycin (ADR) inhibited proliferation of human osteosarcoma cells by 82%, whereas the MF and ADR acting individually caused only 19% and 44% inhibition, respectively.

Mineralization and pH relationships in healing skeletal defects grafted with demineralized bone matrix.

Early studies had indicated that tissue repair is initially associated with a lower than normal serum pH that later becomes more alkaline. To determine how tissue pH may affect skeletal healing and mineralization, we used a rat skeletal repair model consisting of a long bone segmental defect grafted with acid-demineralized bone matrix (DBM), a biomaterial possessing both osteoinductive and osteoconductive repair properties. In this study, femoral and tibial diaphyses from young adult Sprague Dawley rats were cut into cylinders approximately 0.5 cm in length, demineralized in acid, perforated to accommodate a needle-type combination pH microelectrode, and grafted around a 0.3-cm-long diaphyseal fibula defect. The pH of repair tissues was recorded at various time intervals up to 28 days postgrafting. Healing and mineralization were monitored histologically and by the ash and calcium content of repair tissues. During the early healing phase, tissue pH was lower than normal serum pH, presumably because of an accumulation of acidic metabolites in tissue fluids. Subsequent pH increases to more alkaline values were accompanied by a rapid mineral deposition phase and a later phase characterized by a slow, gradual increase in tissue calcium content. The results of this study support previous observations suggesting that the pH of repair tissue fluids may play a regulatory role in the healing and mineralization of bone.

Electrophysiology of direct current stimulation of fracture healing in canine radius.

Electrophysiological mechanisms involved in the electrical stimulation of fracture healing remain largely unknown. The purpose of the present study was to establish relationships between osteogenetic response and intraosseous measures of electrical dose in experimental fractures (osteotomies) of canine radii stimulated by direct currents. The response was determined postmortem at seven weeks after osteotomy by measuring the bending rigidity and four physicochemical properties: tissue density, mineral density, matrix density, and mineral-to-matrix ratio. The currents measured in bone ranged from 0.1 to 17 microA. Three regions of enhanced osteogenetic response were observed at approximately 1, 7, and 13 microA, separated by regions of unstimulated response. Evidence presented in this paper suggests that enhanced response resulted mainly from electrical modulation of early events in the fracture repair sequence.

Mineral and matrix contributions to rigidity in fracture healing.

The purpose of this study was to investigate the relationships among selected properties of fracture callus: bending rigidity, tissue density, mineral density, matrix density and mineral-to-matrix ratio. The experimental model was an osteotomized canine radius in which the development of the fracture callus was modified by electrical stimulation with various levels of direct current. This resulted in a range of values for the selected properties of the callus, determined post mortem at 7 weeks after osteotomy. We found that the rigidity (R) of the bone-callus combination obeyed relationships of the form R = axb, where x is the tissue density, mineral density, matrix density or the mineral-to-matrix ratio of the repair tissue. These are analogous to power-law relationships found in studies of compact and cancellous bone. The results suggest that fracture callus at 7 weeks after osteotomy in canine radius behaves more like immature compact bone than cancellous bone in its mechanical and physicochemical properties. The present study demonstrates the feasibility of developing non-invasive in vivo densitometric methods to monitor fracture healing, since models may be developed that can predict mechanical properties from densitometric data. Further studies are needed to develop a refined model based on experimental data on the mechanical and physicochemical properties and microstructure of fracture callus at different stages of healing.

Altered tissue oxygen tension responsiveness in diabetes.

Subcutaneous oxygen tension (tissue PO2) was measured by a polarographic method in the legs of insulin-dependent diabetics (IDDM) and controls. Current flow was measured continuously using a five-stage protocol: baseline; 4 min of complete arterial occlusion; during recovery from ischemia; baseline approximately re-established; induction of hyperemia by local application of heat. Eleven patients with IDDM of 4-32 years of duration, without peripheral arterial disease, were studied and compared with 10 controls. The mean baseline subcutaneous PO2 in diabetics was less than controls; however, the difference was not statistically significant. At the end of arterial occlusion the mean decrease in tissue PO2 was less (P less than 0.025) in diabetics (4.7 +/- 0.9 mm Hg, SEM) compared to controls (10.2 +/- 1.6 mm Hg). With induction of hyperemia the increase in tissue PO2 was lower (P less than 0.001) in diabetics (7.4 +/- 0.4 mm Hg) than in controls (18.6 +/- 1.7 mm Hg). The observed differences provide for the first time direct evidence of altered tissue PO2 responses in diabetes.

Fluid flow in bone in vitro.

Recent dielectric measurements suggest that the electromechanical effect in wet (fluid saturated) bone is not due to a piezoelectric effect. This would imply that the electromechanical effect observed in wet bone is due to a streaming potential and is therefore dependent on fluid flow in stressed bone. A model for fluid flow in bone in vitro is presented. This model predicts a rapid decay (of the order of a millisecond or less) for the fluid flow in Haversian systems. The implications of this result for the interpretation of the electromechanical effect in wet bone are discussed.

Electrical properties of compact bone.

Dielectric properties of compact bone tissue have been measured in the wet, i.e., fluid-saturated, state. Comparison of these with other measurements at high relative humidity (RH) shows that the dc conductivity of wet bone is about 100 times larger than that of the high RH sample. Thus, the extrapolation of the high RH results to in vivo situations is not valid. In addition, the results of electrical measurements on dry bone samples cannot be extrapolated to the in vivo state because of the dominance of the fluid-filled pores. The difference in the results for longitudinal, tangential, and radial samples, both in dc resistivity and relaxation time, reflects the difference in connectivity of the pores on bone in these three orientations. Quantitative estimates of the cross-sectional area of connected pores are obtained from measurements on photomicrographs and correlated with dc conductivity of the samples. Further evidence for the dominance of the fluid-filled pores in determining the properties of the tissue comes from the results for bone conductivity g measured as a function of saline conductivity g0. The ratio g/g0 is approximately constant with respect to changes in g0 over a range corresponding to the conductivities of various body fluids. The influence of the dielectric properties in all but destroying the piezoelectrically generated voltage in going from the dry to the wet state is discussed. It is suggested that some mechanism other than the piezoelectric effect (e.g., streaming potentials) must be considered to account for the magnitude and decay time of the electromechanical voltage measured in wet bone. Our studies suggest that fluid transport plays a significant role not only in various aspects of bone metabolism such as mineralization, but also in the electrical, mechanical, and electromechanical properties of bone.