Cambium cell stimulation from surgical release of the periosteum.

An autograft of periosteal tissue containing cambium cells has potential to become chondrogenic or osteogenic depending on the regeneration repair strategies. The potential number of harvestable cambium cells diminishes with age. Other factors may be associated with a reduction in the number or variable yields of cambium cells including harvest technique, harvest site location, and the time interval from harvest to implantation. Attempts to increase the number of cambium cells have included improvements in harvesting and handling technique, and expansion of the cells in tissue culture. An "in situ" stimulation and proliferation technique would offer the potential for increasing the number of cambium cells in a cost-effective manner for transplantation without the need for expansion in tissue culture.The hypothesis tested was that surgical release of the periosteum and its deep inner underlying cambium layer by sharply incising through the superficial periosteal fibrous layer down to and scoring the cortical bone surface would increase the number of cambium cells that could be harvested at a later time period. Two techniques for periosteal release were used to stimulate a proliferation of the underlying cambium layer and increase the cambium cells for harvest in skeletally mature goats: (1) sharply scoring all four-sides of the tissue test site perimeter, and (2) sharply scoring only two sides of the tissue test site.The two-sided and four-sided release scoring of the periosteum induced stimulatory responses in the number of cambium cells. In addition, a marked increase in mRNA expression for BMP-2 (p<0.001) was observed within 24 h and remained elevated over baseline values for up to 96 h after this stimulation to the cambium layer.     

Effect of periosteal stripping on cortical bone perfusion: A laser doppler study in sheep

The goals of internal fixation are an accurate reduction and stable fixation in the presence of adequate bony vascularity. This can be achieved by a variety of means including plate fixation. A certain amount of periosteal stripping is necessary for proper open reduction of a fracture and for proper plate application. With displaced diaphyseal fractures, cortical bone perfusion (CBP) is already compromised. Further damage, in terms of periosteal stripping for plate fixation, may not be acceptable. Little information is available as to what extent the periosteum contributes to cortical bone perfusion. The purpose of this study was to determine the acute effects of periosteal stripping on cortical bone perfusion in a sheep tibia model. Twenty-three sheep were operated on and had the medial aspect of their right tibia exposed. Cortical bone perfusion measurements were obtained using laser Doppler flowmetry prior to periosteal stripping and after periosteal stripping. The results of this study show that the cortical bone perfusion significantly decreased by 20% after periosteal stripping over the entire length of the tibia. We therefore conclude that the periosteum contributes to diaphyseal bone perfusion and that it is important to preserve this source with fractures where blood supply is already significantly compromised.     

去除外骨膜对引导性骨再生模型成骨过程的影响

目的研究去除外骨膜对引导性骨再生成骨过程的影响。方法对２４只新西兰兔双侧桡骨制作１０ｍｍ骨缺损,一侧保留外骨膜并用硅胶管连接,另一侧于缺损两端各去除外骨膜１０ｍｍ,其余手术方法相同,动物分别于术后３天和１、２、３、４、６、１０、１２周处死,标本行Ｘ线、非脱钙骨切片组织学检查。结果（１）去除外骨膜后对于膜管内、膜管外成骨过程均无影响;（２）外骨膜生发层中的成骨细胞与哈弗系统中央管内及骨表面的成骨细胞相互联系;（３）外骨膜的两层结构分别来自于不同的组织：纤维层组织来自于外周的软组织,生发层中的成骨细胞来自于骨表面及哈弗系统。结论去除外骨膜对引导性骨再生模型的成骨过程无影响     

Mechanical regulation of PTHrP expression in entheses

The PTHrP gene is expressed in the periosteum and in tendon and ligament insertion sites in a PTHrP-lacZ knockin reporter mouse. Here, we present a more detailed histological evaluation of PTHrP expression in these sites and study the effects of mechanical force on PTHrP expression in selected sites. We studied the periosteum and selected entheses by histological, histochemical, and in situ hybridization histochemical techniques, and tendons or ligaments were unloaded by tail suspension or surgical transection. In the periosteum, PTHrP is expressed in the fibrous layer and the type 1 PTH/PTHrP receptor (PTH1R) in the subjacent cambial layer. PTHrP has distinct temporospatial patterns of expression in the periosteum, one hot spot being the metaphyseal periosteum in growing animals. PTHrP is also strongly expressed in a number of fibrous insertion sites. In the tibia these include the insertions of the medial collateral ligament (MCL) and the semimembranosus (SM). In young animals, the MCL and SM sites display a combination of underlying osteoblastic and osteoclastic activities that may be associated with the migration of these entheses during linear growth. Unloading the MCL and SM by tail suspension or surgical transection leads to a marked decrease in PTHrP/lacZ expression and a rapid disappearance of the subjacent osteoblastic population. We have not been able to identify PTHrP-lacZ in any internal bone cell population in the PTHrP-lacZ knockin mouse in either a CD-1 or C57Bl/6 genetic background. In conclusion, we have identified PTHrP expression in surface structures that connect skeletal elements to each other and to surrounding muscle but not in intrinsic internal bone cell populations. In these surface sites, mechanical force seems to be an important regulator of PTHrP expression. In selected sites and/or at specific times, PTHrP may influence the recruitment and/or activities of underlying bone cell populations.     

Behaviour of cancellous bone graft with and without periosteal isolation in striated muscle. An experimental study

The capacity of the periosteum to inhibit resorption of cancellous bone grafts into muscle was investigated in 34 four- to six-week-old rabbits. In 17 experiments the periosteum was wrapped around the grafts with the cambium layer facing the bone, and in seven experiments with the cambium layer facing the muscle. In the control group of 10 experiments there was no periosteal wrapping around the bone grafts. In Series 1 with the cambium layer of the periosteum facing the bone, after 20 weeks a tubular bone with Haversian system and bone marrow was seen. The transplants were surrounded by normal-looking periosteum. Bone formation from the periosteum occurred through enchondral ossification. Inductive bone growth was observed from the cancellous graft. In Series 2 with the cambium layer facing the surrounding muscle tissue, after 20 weeks two laminar bone blocks with periosteum in between and surrounding each block was observed. In the control series without periosteal covering, after 20 weeks only fibrous tissue remained in the transplantation site. It is obvious that periosteal isolation of cancellous bone grafts inhibits their resorption when transplanted into muscle in young animals.     

Localization of chondrocyte precursors in periosteum

OBJECTIVE: Periosteal chondrogenesis is relevant to cartilage repair and fracture healing. Periosteum contains two distinct layers: a thick, outer fibrous layer and a thin, inner cambium layer which is adjacent to the bone. Specific chondrocyte precursors are known to exist in periosteum but have not yet been identified. In this study, the location of the chondrocyte precursors in periosteum was determined. METHOD: One hundred and twenty periosteal explants from 30 2-month-old NZ rabbits were cultured for up to 42 days. Histomorphological changes and spatio-temporal localization of Col. II mRNA and protein were analysed. RESULTS: On day 7, chondrocyte differentiation appeared in the most juxtaosseous region in the cambium layer. Col. II mRNA and protein were also evident in the same region. By day 14, chondrocyte differentiation progressed further into the juxtaosseous cambium layer, as did Col. II mRNA and protein. With growth of the neocartilage, the cambium layer gradually diminished to the extent that by 21-28 days it was no longer evident. Cartilage growth was significant and followed an appositional pattern, growing away from the fibrous layer. The fibrous layer remained essentially unchanged from 0-42 days, without evidence of hypertrophy or atrophy. Col. II mRNA expression was never seen in the fibrous layer. CONCLUSION: From these data, three conclusions can be drawn concerning chondrogenesis from periosteum: (1) the chondrocyte precursors are located in the cambium layer of periosteum; (2) chondrogenesis commences in the juxtaosseous area in the cambium layer and progresses from the juxtaosseous region to the juxtafibrous region of the cambium layer; (3) neocartilage growth is appositional, which displaces the fibrous layer away from the cartilage already formed, as new cartilage is formed between these two layers. These findings suggest that the least differentiated (stem or reserve) cells are located in the cambium layer furthest from the bone. CLINICAL RELEVANCE: These findings show that the chondrocyte precursors are located in the cambium layer of periosteum. Preservation of this layer is essential for chondrogenesis. As neocartilage growth is appositional, away from the fibrous layer, it can be expected that the new cartilage deposited in and adjacent to a periosteal graft would be expected to be located on the side of the cambium layer, rather than on the side of the fibrous layer of the graft.     

Periosteum: biology, regulation, and response to osteoporosis therapies

Periosteum contains osteogenic cells that regulate the outer shape of bone and work in coordination with inner cortical endosteum to regulate cortical thickness and the size and position of a bone in space. Induction of periosteal expansion, especially at sites such as the lumbar spine and femoral neck, reduces fracture risk by modifying bone dimensions to increase bone strength. The cell and molecular mechanisms that selectively and specifically activate periosteal expansion, as well as the mechanisms by which osteoporosis drugs regulate periosteum, remain poorly understood. We speculate that an alternate strategy to protect human bones from fracture may be through targeting of the periosteum, either using current or novel agents. In this review, we highlight current concepts of periosteal cell biology, including their apparent differences from endosteal osteogenic cells, discuss the limited data regarding how the periosteal surface is regulated by currently approved osteoporosis drugs, and suggest one potential means through which targeting periosteum may be achieved. Improving our understanding of mechanisms controlling periosteal expansion will likely provide insights necessary to enhance current and develop novel interventions to further reduce the risk of osteoporotic fractures.     

Bone formation by vascularized periosteal and osteoperiosteal grafts

The osteogenic capacity of vascularized periosteal and osteoperiosteal grafts was investigated in 82 Wistar rats about 8 weeks old. The periosteal flaps, pedicled on the descending genicular artery, were taken by stripping the lower third of the femur. In the right hindleg, the grafts were made with periosteum only, while in the left hindleg, the periosteal flaps were associated with cancellous bone. The animals were divided into two groups of 41. In group I, both the periosteal and osteoperiosteal grafts were placed in contact with cortical bone, and in group II, the grafts were buried in muscle. Subgroups of 8 animals were killed after 1, 2, 4, 8, and 16 weeks postoperatively. The grafted region was evaluated radiographically, macroscopically, and histologically. Membranous ossification was the main source of bone formation. Osteoperiosteal grafts produced a greater amount of new bone than periosteal ones. There was evidence that the contact of the graft with living cortical bone favored bone formation.     

Osteogenic potential of primed periosteum graft in the rat calvarial model

Repair of bone defects remains a major concern in plastic and maxillofacial surgery. Based on modern concepts of tissue engineering, periosteum has gained attention as a suitable osteogenic material. We tested the hypothesis that surgically released and immediately repositioned periosteum would exhibit high osteogenic capacity upon grafting in a rat calvarial defect. Seven days after periosteum was released from the tibia and immediately repositioned, the "primed periosteum graft" (PPG; n = 15) was placed into a critical-sized defect of rat calvaria and the process of bone formation was evaluated histologically, immunohistologically, and radiographically at 7, 14, and 21 days after grafting. Findings were compared with a nonprimed periosteal graft (NPG; n = 15).Endochondral ossification was observed in both the PPG and NPG. The PPG showed higher expression of proliferative cell nuclear antigen, bone morphogenetic protein, and vascular endothelial growth factor than the NPG. Three-dimensional radiographic examination revealed significantly increased bone formation in the PPG than in the NPG (P < 0.01). These findings suggested that surgical stimulation of the periosteum enhanced the osteogenic potential of periosteal cells. This method may be suitable for the clinical repair of bone defects.     

Osteogenic ability of free periosteal autografts in tibial fractures with severe soft tissue damage: an experimental study

OBJECTIVE: The present study was undertaken to assess whether free nonvascularized autologous periosteum transplants enhance bone healing in a rabbit fracture model designed to resemble a tibial fracture with severe soft tissue damage. DESIGN: Transplantation of free autologous periosteal grafts on the anteromedial site of the tibia (experimental group) was compared with nontransplantation on the contralateral tibia (control group). We produced a standardized transverse osteotomy of both tibial diaphyses in white male adult New Zealand rabbits. The endomedullary cavity was reamed and nailed, and then a one-centimeter segment of periosteum was excised from either side of the osteotomy. To prevent periosteal and extraosseous ingrowth at the osteotomy site, a silastic sheet was wrapped around two-thirds of the circumference of the tibia. In the first group, on the silastic-free bone window, we then spanned the osteotomy with a free, nonvascularized, longitudinally oriented autologous periosteum and sewed it to the adjacent periosteum both proximally and distally. In the second group, the periosteum was placed transversely, leaving a gap between it and the adjacent periosteum proximally and distally. Revascularization of the graft was determined with the colored microsphere technique. MAIN OUTCOME MEASUREMENTS: Histomorphometric analysis of the periosteal callus was done on a transparent grid superimposed on enlarged photographs of the histologic sections. RESULTS: Free, nonvascularized, longitudinally placed autologous periosteum in contact with intact periosteum produced significantly more periosteal callus than was seen in the control group, in which no periosteal graft was used. However, when transversely placed periosteal grafts were set in the silastic-free bone window and there was no contact with surrounding remnants of intact periosteum, no significant difference in callus production was noted when compared with the control. Revascularization of these grafts was seen within one week after transplantation. Bone healing occurred mainly through endochondral ossification. CONCLUSION: Our data suggest that orthotopically placed autologous nonvascularized periosteum retains its osteogenic potential in a poorly vascularized environment such as a tibial fracture with severe soft tissue damage. The effect is enhanced if the graft is in contact with intact periosteum. Histologically, callus formation after periosteal grafting resembles endochondral and intramembranous ossification.     

Normal and heterotopic periosteum

Normal periosteum is an osteoprogenitor cell-containing bone envelope, which can be activated to proliferate by trauma, retroviruses, tumors, and lymphocyte mitogens. Activated periosteum produces cartilage and bone, and is colonized by bone-resorbing cells. The osteogenic activity of periosteum is maintained in heterotopic sites and in vitro. Ectopic bone, however, is colonized by bone marrow precursor cells but does not develop a true periosteum. The absence of true periosteal envelope in the heterotopically induced bone may be the major, if not the only, difference between heterotopic and orthotopic bone deposits.     

Human femoral neck has less cellular periosteum, and more mineralized periosteum, than femoral diaphyseal bone

Periosteal expansion enhances bone strength and is controlled by osteogenic cells of the periosteum. The extent of cellular periosteum at the human femoral neck, a clinically relevant site, is unclear. This study was designed to histologically evaluate the human femoral neck periosteal surface. Femoral neck samples from 11 male and female cadavers (ages 34-88) were histologically assessed and four periosteal surface classifications (cellular periosteum, mineralizing periosteum, cartilage, and mineralizing cartilage) were quantified. Femoral mid-diaphysis samples from the same cadavers were used as within-specimen controls. The femoral neck surface had significantly less (P<0.05) cellular periosteum (18.4+/-9.7%) compared to the femoral diaphysis (59.2+/-13.8%). A significant amount of the femoral neck surface was covered by mineralizing periosteal tissue (20-70%). These data may provide an alternate explanation for the apparent femoral neck periosteal expansion with age and suggest the efficiency of interventions that stimulate periosteal expansion may be reduced, albeit still possible, at the femoral neck of humans.     

Structural and cellular differences between metaphyseal and diaphyseal periosteum in different aged rats

In both physiological and pathological processes, periosteum plays a determinant role in bone formation and fracture healing. However, no specific report is available so far focusing on the detailed structural and major cellular differences between the periostea covering different bone surface in relation to ageing. The aim of this study is to compare the structural and cellular differences in diaphyseal and metaphyseal periostea in different aged rats using histological and immunohistochemical methods. Four female Lewis rats from each group of juvenile (7 weeks old), mature (7 months old) and aged groups (2 years old) were sacrificed and the right femur of each rat was retrieved, fixed, decalcified and embedded. Five-micrometer thick serial sagittal sections were cut and stained with Hematoxylin and Eosin, Stro-1 (stem cell marker), F4/80 (macrophage marker), TRAP (osteoclast marker) and vWF (endothelial cell marker). One-millimeter lengths of middle diaphyseal and metaphyseal periosteum were selected for observation. The thickness, total cell number and positive cell number for each antibody were measured and compared in each periosteal area and different aged groups. The results were subjected to two-way ANOVA and SNK tests. The results showed that the thickness and cell number in diaphyseal periosteum decreased with age (p<0.001). In comparison with diaphyseal area, the thickness and cell number in metaphyseal periosteum were much higher (p<0.001). There were no significant differences between the juvenile and aged groups in the thickness and cell number in the cambial layer of metaphyseal periosteum (p>0.05). However, the juvenile rats had more Stro1(+), F4/80(+) cells and blood vessels and fewer TRAP(+) cells in different periosteal areas compared with other groups (p<0.001). The aged rats showed much fewer Stro1(+) cells, but more F4/80(+), TRAP(+) cells and blood vessels in the cambial layer of metaphyseal periosteum (p<0.001). In conclusion, structure and cell population of periosteum appear to be both age-related and site-specific. The metaphyseal periosteum of aged rats seems more destructive than diaphyseal part and other age groups. Macrophages in the periosteum may play a dual important role in osteogenesis and osteoclastogenesis.     

The Osteogenic Capacity of Free Periosteal and Osteoperiosteal Grafts: A Comparative Study in Growing Rabbits

The behaviour of free periosteal and 200 micron thick osteoperiosteal grafts was studied histologically in 40 six-week-old rabbits. The grafts were taken from the tibia and fixed on either side of the same lumbar vertebra between the spinous and mamillary processes. The free stripped periosteum had better osteogenic activity than the 200 micron thick osteoperiosteum. The new bone was formed by the osteogenic cells of the cambium layer in both types of graft.     

The osteogenic capacity of free periosteal and osteoperiosteal grafts. A comparative study in growing rabbits.

The behaviour of free periosteal and 200 micron thick osteoperiosteal grafts was studied histologically in 40 six-week-old rabbits. The grafts were taken from the tibia and fixed on either side of the same lumbar vertebra between the spinous and mamillary processes. The free stripped periosteum had better osteogenic activity than the 200 micron thick osteoperiosteum. The new bone was formed by the osteogenic cells of the cambium layer in both types of graft.     

Periosteal PTHrP Regulates Cortical Bone Remodeling During Fracture Healing

Parathyroid hormone-related protein (PTHrP) is widely expressed in the fibrous outer layer of the periosteum (PO), and the PTH/PTHrP type I receptor (PTHR1) is expressed in the inner PO cambial layer. The cambial layer gives rise to the PO osteoblasts (OBs) and osteoclasts (OCs) that model/remodel the cortical bone surface during development as well as during fracture healing. PTHrP has been implicated in the regulation of PO modeling during development, but nothing is known as regards a role of PTHrP in this location during fracture healing.We propose that PTHrP in the fibrous layer of the PO may be a key regulatory factor in remodeling bone formation during fracture repair. We first assessed whether PTHrP expression in the fibrous PO is associated with PO osteoblast induction in the subjacent cambial PO using a tibial fracture model in PTHrP-lacZ mice. Our results revealed that both PTHrP expression and osteoblast induction in PO were induced 3 days post-fracture. We then investigated a potential functional role of PO PTHrP during fracture repair by performing tibial fracture surgery in 10-week-old CD1 control and PTHrP conditional knockout (PTHrP cKO) mice that lack PO PTHrP. We found that callus size and formation as well as woven bone mineralization in PTHrP cKO mice were impaired compared to that in CD1 mice. Concordant with these findings, functional enzyme staining revealed impaired OB formation and OC activity in the cKO mice.We conclude that deleting PO PTHrP impairs cartilaginous callus formation, maturation and ossification as well as remodeling during fracture healing. These data are the initial genetic evidence suggesting that PO PTHrP may induce osteoblastic activity and regulate fracture healing on the cortical bone surface.     

Testing of a new one-stage bone-transport surgical procedure exploiting the periosteum for the repair of long-bone defects

BACKGROUND: A recently proposed one-stage bone-transport surgical procedure exploits the intrinsic osteogenic potential of the periosteum while providing mechanical stability through intramedullary nailing. The objective of this study was to assess the efficacy of this technique to bridge massive long-bone defects in a single stage. METHODS: With use of an ovine femoral model, an in situ periosteal sleeve was elevated circumferentially from healthy diaphyseal bone, which was osteotomized and transported over an intramedullary nail into a 2.54-cm (1-in) critical-sized diaphyseal defect. The defect-bridging and bone-regenerating capacity of the procedure were tested in five groups of seven animals each, which were defined by the absence (Group 1; control) or presence of the periosteal sleeve alone (Group 2), bone graft within the periosteal sleeve (Groups 3 and 5), as well as retention of adherent, vascularized cortical bone chips on the periosteal sleeve with or without bone graft (Groups 4 and 5). The efficacy of the procedure was assessed qualitatively and quantitatively. RESULTS: At sixteen weeks, osseous bridging of the defect was observed in all twenty-eight experimental sheep in which the periosteal sleeve was retained; the defect persisted in the remaining seven control sheep. Among the experimental groups 2 through 5, significant differences were observed in the density of the regenerated bone tissue; the two groups in which vascularized bone chips adhered to the inner surface of the periosteal sleeve (Groups 4 and 5) showed a higher mean bone density in the defect zone (p < 0.02) than did the other groups. In these two groups with the highest bone density, the addition of bone graft was associated with a significantly lower callus density than that observed without bone graft (p < 0.05). The volume of regenerate bone (p < 0.02) was significantly greater in the groups in which the periosteal sleeve was retained than it was in the control group. Among the experimental groups (groups 2 through 5), however, with the numbers studied, no significant differences in the volume of regenerate bone could be attributed to the inclusion of bone graft within the sleeve or to vascularized bone chips remaining adherent to the periosteum. CONCLUSIONS: The novel surgical procedure was shown to be effective in bridging a critical-sized defect in an ovine femoral model. Vascularized bone chips adherent to the inner surface of the periosteal sleeve, without the addition of morselized cancellous bone graft within the sleeve, provide not only a comparable volume of regenerate bone and composite tissue (callus and bone) but also a superior density of regenerate bone compared with that after the addition of bone graft.     

Periosteal augmentation of a tendon graft improves tendon healing in the bone tunnel

Secure fixation of tendon or ligament to bone has been a challenging problem. The periosteum is an osteogenic organ that regulates bone growth and remodeling at the outer surface of cortical bone and also is known to play an important role in forming a tendon insertion site to bone. Therefore, we hypothesized that a freshly harvested periosteum can be used as a stimulative scaffold to biologically reinforce the attachment of tendon graft to bone. Using a rabbit hallucis longus tendon and calcaneus process model, we found that a periosteal augmentation of a tendon graft could enhance the structural integrity of the tendon-bone interface, when the periosteum is placed between the tendon and bone interface with the cambium layer facing toward the bone. Clinically, the use of an autogenous periosteum patch would be an optimal choice for biologic augmentation of the tendon graft in the bone tunnel, because the tissue is readily available for harvest from the patient's body.     

In vivo osteochondrogenic potential of cultured cells derived from the periosteum

Periosteal cells were isolated from young chicks, introduced into cell culture, subcultured, and then inoculated into athymic, nude mice to test the in vivo osteochondrogenic potential of cultured periosteal cells. In monolayer cultures, the adherent periosteal cells showed fibroblastlike morphology and overtly expressed neither osteogenic nor chondrogenic phenotypes. Cultured cells inoculated heterotopically into nude mice eventually gave rise to bone tissue at the subcutaneous injection site. The process of bone formation occurred through two different mechanisms: intramembranous bone formation at the peripheral portion of the inoculum early and endochondral bone formation in the central portion later. Frozen, preserved periosteal cells also formed bone after introduction into nude mice in the same temporal histologic sequence as the unfrozen cells. Cultured chick muscle fibroblasts from donors that were the same age as controls did not form bone or cartilage when inoculated under identical conditions to those of cultured periosteal cells. These results suggest that periosteum of young chicks contains subsets of progenitor cells that possess the potential to differentiate directly into osteoblasts or chondrocytes when inoculated in vivo. Importantly, this potential is retained after enzymatic isolation, cell culture, subculturing, and freeze preservation.     

Callus formation and microvascular regeneration during the fracture healing process in the rat femur]

Correlation between the callus formation and its microvascular regeneration during the fracture healing process in the rat femur was examined under SEM and TEM utilizing the plastic injection method. Stabilization of bony fragments was provided by a miniature external fixator. Normal periosteal microvasculature consisted of 2 layers. In the outer layer, arterioles and venules formed the course network. In the inner layer, capillaries were situated in contact with the compact bone, forming polygonal meshes. In the medulla, the central longitudinal artery gave rise to numerous arterioles, which communicated with sinusoidal capillaries. The proliferation of the internal layer of the periosteal capillary network was observed in the periosteal callus. It is revealed that the trabecular structure of the periosteal and medullary calluses depended on their microvascular architecture. The anastomoses of newly-formed capillaries at the fracture site started first from the outer layer of the periosteum extended to the medulla then finally to the inner layer of the periosteum.