Doctoral Thesis

Under følger doktorgradsarbeidet (Doctoral Thesis, Dr. Med) til Eirik Solheim, Effects of Bioerodible Polyorthoester on Heterotopic and Orthotopic Bone Induction in Rats, utført ved UiO (Institutt for Kirurgisk forskning ved Rikshospitalet) og Kysthospitalet i Hagevik/UiB. Arbeidet ble forsvart ved disputas i Oslo 1993. Du kan også laste det ned som pdf fil Thesis. University of Oslo 1993 Solheim. I pdf filen er også artiklene inkludert. Her er det lagt til lenker til enkelt arbeidene (6 artikler).

Institute for Surgical Research, University of Oslo and Hagavik Orthopaedic Hospital, University of Bergen, Norway

Effects of Bioerodible Polyorthoester on Heterotopic and Orthotopic Bone Induction in Rats

Doctoral Thesis

by

Eirik Solheim
Oslo and Hagavik 1992

Contents

ACKNOWLEDGEMENTS. V

LIST OF PAPERS. VII

GENERAL INTRODUCTION. 1

Osteoinduction by demineralized bone. 1

Purified osteoinductors and growth factors. 3

Orthopedic absorbable implants. 4

Osteoinductor delivery systems. 4

Drug delivery systems. 5

Bioerodible polyorthoester. 6

Inhibition of bone formation. 7

Local hemostatics. 8

Aims of the present study. 9

SUMMARY OF RESULTS. 11

Paper I. 11

Paper II. 11

Paper III. 12

Paper IV. 12

Paper V. 13

Paper VI. 13

GENERAL DISCUSSION. 15

Material and methods. 15

Animals. 15

Implants. 15

Heterotopic model. 17

Orthotopic models. 19

Results. 21

Handling properties and local hemostasis. 21

Heterotopic osteoinduction. 21

Orthotopic osteogenesis. 23

CONCLUSIONS. 25

REFERENCES. 27

PAPERS I-VI. 43

Acknowledgements

The present work was carried out at the Institute for Surgical Research, Rikshospitalet, University of Oslo from 1988 through 1990 and at Hagavik Orthopaedic Hospital, University of Bergen from 1989.

I am deeply indebted to: Professor Egil Amundsen, then head of the Institute for Surgical Research, for the working facili­ties pro­vided and for constructive criticism; Professor Einar Sudmann for his expert advice as my senior supervisor; Professor Gisle Bang, Department of Oral Pathology and Forensic Odontology, Haukeland Hospital, University of Bergen for valuable advice in histology; Else Marie Pinholt, M.S., D.D.S., Ph.D. for perfect col­laboration throughout the work; Rune Andersen, M.D. for his help during the experiment presented in paper V; Research col­league Ole Kirkeby, M.D. for stimulating discussions; Katrine Hove and Gunnvor Øijordsbakken, Department of Oral Pathology and Forensic Odontology, Haukeland Hospital, University of Bergen and Inger Liv Nordli for excellent technical assistance; Professor Leif Schjerven and his staff at the Department for Experimental Animal Research, Rikshospitalet, University of Oslo for their assis­tance with the animals used in the experiments; Photographer Tor Christensen, Department of Pathology and the staff at the Depart­ment of Photography, Haukeland Hospital, University of Bergen for their services; Gerd Brøten, Pathology Building Library, Rikshospitalet, University of Oslo, the staff at the Medical Library and Information Center, Rikshospitalet and the staff at the University Library of Bergen, Medical Department, Haukeland Hospital for valuable library services; and my wife Janne and our children Siri, Kristin and Endre for their patience and support. The work has been financially aided by the J.E. Isberg Foundation and the Aslaug Andersen Foundation.

Hagavik, June 1992. Eirik Solheim

List of papers

This thesis is based on the following papers, to be referred to by their Roman numerals in the text:

I. Solheim E, Pinholt E M, Bang G, Sudmann E. Comparison of histomorphometry and 85Sr uptake in induced hetero­topic bone in rats. Acta Orthop Scand 1992;63(3):334-8.

II. Pinholt E M, Solheim E, Bang G, Sudmann E. Bone induction by composite of bioerodible polyorthoester and demineralized bone matrix in rats. Acta Orthop Scand 1991;62(5):476-80.

III.   Solheim E, Pinholt E M, Bang G, Sudmann E. Effect of local hemostatics on bone induction in rats. A comparative study of bioerodible polyorthoester with and without gen­tamicin, bone wax and fibrin-collagen paste. J Biomed Mater Res 1992;26(6):791-800.

IV. Solheim E, Pinholt E M, Bang G, Sudmann E. Regeneration of calvaria by composite of bioerodible polyorthoester and demineral­ized bone in rats. J Neurosurg, 1992;76(2):275-9.

V. Solheim E, Pinholt E M, Andersen R, Bang G, Sudmann E. The effect of a composite of polyorthoester and demineral­ized bone on the healing of large segmental defects of the radius in rats. J Bone Joint Surg (Am), 1992 Dec;74(10):1456-63.

VI. Solheim E, Pinholt E M, Bang G, Sudmann E. Inhibition of heterotopic osteogenesis in rats by a local bioerodible in­dometh­acin delivery system. J Bone Joint Surg (Am), 1992;74-A(5):705-12.

General introduction

OSTEOINDUCTION BY DEMINERALIZED BONE

Intramuscular implantation of fresh autogenous bone (Levander 1938, Chalmers 1959), HCl demineralized allogeneic bone (Urist 1965, Reddi and Huggins 1972) or purified osteoinductors (Urist et al. 1979) regularly evoke heterotopic osteogenesis in rodents. When heterotopic osteogenesis results from implantation of demin­eralized bone (DBM) or purified osteoinductors, void of living cells, the process is called osteoinduction.

The major phases of DBM-induced heterotopic osteogene­sis in rodents are chemotaxis of mesenchymal cells, mitosis, differ­en­tiation of cartilage, vascular invasion, bone differentiation and formation of an ossicle filled with bone marrow elements. Shortly after implantation of DBM there is a blood clot, platelet release and polymorphonuclear leukocytes arrive by chemotaxis. On day 1, osteo­progenitor cells are seen. By day 3, most leukocytes have disap­peared and the osteoprogenitor cells are proliferating. On day 5, the proliferating cells differentiate into chondroblasts. By day 7 to 8, there are many chondrocytes producing type II collagen and carti­lage specific proteoglycans. Then, there is a capillary invasion lead­ing to hypertrophy of chondrocytes and calcification of cartilage matrix on day 9. On day 10, osteoblasts appear close to the vascular endothelium and new bone is formed by appositional growth on the surface of the calcified matrix and the DBM. Remodeling of the bone by osteoclasts starts by day 12 and an ossicle forms by day 16 to 21 (Urist 1965, Reddi and Huggins 1972).

DBM-induced heterotopic osteogenesis in rodents has been used as an experimental model for investigation of the osteoinduc­tion itself (Urist 1965) as well as the effect of several factors on the osteoinduction: (1) age (Syftestad and Urist 1979, Reddi 1985) and species (Schwartz et al. 1989) of donor and recipient; (2) properties of the demineralized bone, e.g., demineralization procedure (Delloye et al. 1985, Marinak et al. 1989), particle size (Syftestad and Urist 1979), storage, and sterilization procedures (Munting et al. 1988); (3) dietary factors, e.g., magnesium, (Belanger et al. 1975, Schwartz and Reddi 1979), manganese and copper (Strause et al. 1987), fluoride (Turner et al. 1989), vitamin A (DeSimone and Reddi 1983) and vitamin D (Vandersteenhoven 1988); (4) hormones (Reddi and Sullivan 1980, Burnett and Reddi 1983, Kapur and Reddi 1989); (5) growth factors (Howes et al. 1988, Aspenberg and Lohmander 1989); (6) biphosphonates (Bauer et al. 1986); (7) indomethacin (Törnkvist et al. 1985); and (8) biomateri­als (Vandersteenhoven and Spector 1983, Uretzky et al. 1987, Schwarz et al. 1989, Ripamonti et al. 1989).

Experimentally in animals, DBM enhances (1) cranio-max­il­lofacial reconstructions (Senn 1889, Ray 1957, Mulliken and Glowacki 1980, Glowacki et al. 1981a, Kaban and Glowacki 1981, Pinholt et al. 1990), (2) healing of diaphyseal defects (Narang et al. 1979, Oikarinen and Korhonen 1979, Gupta and Tuli 1982, Einhorn et al. 1984, Aspenberg et al. 1986, Aspenberg et al. 1987, Gepstein et al. 1987, Köhler and Kreicbergs 1987, Hopp et al. 1989) and (3) spinal fusion (Lindholm et al. 1982, Ragni et al. 1987).

DBM have been used clinically for (1) cranio-maxillofacial reconstructions (Kümmell 1891, Glowacki et al. 1981b, Ousterhout 1985), (2) healing of diaphyseal defects (Senn 1889, Miller 1890, Urist 1968, Osepian et al. 1989), (3) pseudarthrosis operation (Urist 1968), (4) arthrodesis of hip (Urist 1968) and (5) spinal fusion (Sharrard and Collins 1961, Urist 1968, Urist and Dawson 1981)

Whereas it has been clearly documented that DBM induces heterotopic bone formation in rodents and enhances bone forma­tion in skeletal sites both in rodents and humans, the results regard­ing heterotopic bone formation by DBM in primates are conflict­ing. Aspenberg et al. (1988) found no bone and very little cartilage 40 days postoperatively in squirrel monkeys and concluded that the observed bone formation by DBM implantation in skeletal sites in humans is not necessarily caused by osteoinduction, but may be the result of the implant acting as a osteoconductive scaffold. Further, Aspenberg and Andolf (1989) found that human DBM implanted intramuscularly in athymic rats induced bone formation and they concluded that the possible decreased ability of such DBM to induce heterotopic bone formation in adult primates may be due to an inability by the recipients to respond to the inductive stimuli of the DBM (Aspenberg and Andolf 1989). However, Hosny and Sharawy (1985) found that intramuscularly implanted DBM in rhesus monkeys induced cartilage and bone although the process was delayed compared to that in rodents, and Ripamonti reported bone formation in 73 out of 96 DBM implants in 24 adult male ba­boons (1991).

PURIFIED OSTEOINDUCTORS AND GROWTH FACTORS

The existence of osteoinductor(s) and growth factors in trans­planted bone and bone extracts has been postulated for a long time (Gallie and Robertson 1929, Levander 1938, Lacroix 1945, Urist 1965). In recent years several osteoinductors and growth factors have been isolated and characterized from bone matrix extracts: (1) bone morphogenetic protein (BMP) (Urist and Mikulski 1979); (2) os­teogenin (identical to BMP-3) (Sampath et al. 1987); (3) insulin-like growth factor I (IGF I, identical to somatomedin-C) (Canalis et al. 1988) and insulin-like growth factor II (IGF II, similar to SGF) (Farley and Baylink 1982); (4) transforming growth factor beta (TGF-b, identical to cartilage inducing factor, CIF) (Seyedin et al. 1985, Linkhart et al. 1986); (5) acidic and basic forms of fibroblast growth factor (FGF) (Hauschka et al. 1986); and (6) platelet-de­rived growth factor (PDGF) (Hauschka et al. 1986).

In the past few years, many osteoinductors and bone growth factors, including subgroups of the factors mentioned above, e.g., BMP 1-7, have become readily available by recombinant DNA technology and knowledge about their biologic effects is evolving (Hauschka et al. 1988, Gospodarowicz 1990, Marden et al. 1990, Ozkaynak et al. 1990, Pfeilschifter et al. 1990, Wozney et al. 1988, Wang et al. 1990, Mohan and Baylink 1991, Reddi and Cunningham 1991). The osteoinductors are necessary to initiate os­teogenesis, whereas growth factors may augment different stages of the process, e.g., attracting preosteoblasts, accelerating their proliferation, and stimulating angiogenesis (Kawamura and Urist 1988, Reddi et al. 1989, Marden et al. 1990).

ORTHOPEDIC ABSORBABLE IMPLANTS

Orthopedic absorbable implants have been developed for different purposes: (1) temporary internal fixation of skeletal parts (Getter et al. 1972, Rokkanen et al. 1985); (2) bone substitutes, i.e., polymers and ceramics, used alone or in combination with bone grafts (Driskell et al. 1972, Hollinger 1983, Higashi et al. 1986, Hollinger and Battistone 1986); (3) bone cement (Gerhart et al. 1989); (4) os­teoinductor delivery (Urist et al. 1984, Lovell et al. 1989, Lucas et al. 1990); (5) drug delivery, (Mackey et al. 1982, Tasslet and Inhoff 1988); and (6) local hemostasis (Geary and Frantz 1950, Harris and Capperauld 1980). Implants used for the latter three purposes are discussed below.

OSTEOINDUCTOR DELIVERY SYSTEMS

Demineralized bone matrix (DBM) used as chips or powder is technically difficult to apply for reconstructions and bridging of bone defects as the particles may displace during or after the opera­tion. Incorporating the DBM in a bioerodible carrier thus seems warranted (Glowacki and Mulliken 1985). Further, it has been claimed that effective osteoinductive stimulus in humans require supply of a proper combination of an osteoinductor, i.e., BMP-2 or osteogenin (BMP-3) to induce bone formation and growth factors to augment the process (Kawamura and Urist 1988). A biodegrad­able delivery system is needed for sustained release of such factors (Takagi and Urist 1982, Sato and Urist 1985, Marden et al. 1990).

The ideal delivery system should fulfill several criteria; it should (1) be biocompatible (Lyman and Searce 1974); (2) have the right physical properties (Yamazaki et al. 1988); (3) provide local hemostasis (Sudmann et al. 1990); (4) provide sustained, con­trolled release of the active substance (Heller et al. 1981, Rosen et al. 1983, Urist et al. 1984); (5) be resorbed and replaced by preosseous tissues within two weeks, cartilage within three weeks and bone within four to six weeks (Urist et al. 1987); and (6) not inhibit osteogenesis in any way, neither the different stages of os­teoinduction nor osteoconduction.

Several materials have been investigated as biodegradable vehicles such as b-tricalcium phosphate (Urist et al. 1984, Urist et al. 1987), collagen (Nathan et al. 1988, Takaoka et al. 1988, Nakahara et al. 1989), fibrin sealant (Schwarz et al. 1989), matrix gamma-carboxyglutamic acid rich protein (Sato and Urist 1985), plaster of paris (Yamazaki et al. 1988), polylactic acid (Lovell et al. 1989), copolymer of polylactide-polyglycolide (Schmitz and Hollinger 1988) and polyanhydride (Lucas et al. 1990).

DRUG DELIVERY SYSTEMS

Conventionally, bioactive agents are delivered by periodic inges­tion or injection. During this process drug levels reach a maximum and then fall to a minimum. If drug concentrations fall above or be­low the toxic level or minimum effective level, respectively, alternating periods of toxicity or inefficacy can result. The main goal of a controlled release system is to maintain the drug concen­tration between these two levels from a single dosage form (Langer 1986). Further, implanted drug delivery systems make possible local ap­plication of the drug in the effector tissue or organ. This principle may be especially useful in orthopedic surgery where such systems may be applied during bone transplantation, internal fixation and prosthesis surgery, e.g., to stimulate or inhibit new bone formation, provide infection prophylaxis or deliver other drugs with high local concentration without systemic effects.

In a controlled release system, a bioactive agent is incorpo­rated into a carrier, generally a polymeric material. The rate of release of the substance is determined by the properties of the poly­mer itself and is only weakly dependent on environmental factors. Controlled release systems can deliver substances slowly and con­tinuously for long periods. Polymers release drugs by four general mechanisms: diffusion, chemical control, solvent activation and magnetism.

The most commonly used biodegradable polymers for drug delivery systems are polylactic acid (Chang 1976), polyglycolic acid and their copolymers (Langer 1986). Other biodegradable polymers that have been used for controlled drug delivery systems include polymers based on caprolactone (Pitt et al. 1979), various copolymers of amino acids (Sidman et al. 1980), polyanhydrides (Rosen et al. 1983) and polyorthoesters (Heller et al. 1981).

Drug release from bioerodible matrix devices can be con­trolled by either diffusion or erosion. If erosion of the matrix is much slower than diffusion, the release kinetics are essentially those of a diffusion controlled matrix. If however the drug is immobilized in the matrix so that diffusional release is minimal com­pared to erosion, the rate of drug release will be erosion controlled (Heller 1986).

Heterogeneous erosion occurs when degradation takes place only at the surface of the polymer, whereas homogeneous erosion is the result of degradation occurring through the polymer matrix. Heterogeneous erosion can lead to zero-order drug release provid­ing diffusional release of the drug is minimal and the overall surface area of the device remains constant (Heller 1986). Hydropho­bic polymers are more likely to erode heterogeneously since water is excluded. Hydrophilic polymers such as polylactic acid and polyglycolic acid, in contrast, may absorb water and erode homo­geneously leading to progressive looseing of the matrix, changes in permeability and drug diffusion resulting in nonlinear drug release (Heller et al. 1981, Rosen et al. 1983). Polymers known to display hydrolytically controlled surface erosion and where the polymers break down to monomers are polyorthoesters when certain addi­tives are included (Heller et al. 1981) and polyanhydrides (Rosen 1983).

BIOERODIBLE POLYORTHOESTER

Poly(2,2-dioxy-cis,trans-1,4-cyclohex­ane dimethylene tetrahydro­furan) (Alzamer®, Alza Corporation, Palo Alto), a bioerodible polyorthoester, results from condensation of 2,2-di­ethoxytetra­hy­dro­furan and cis,trans-1,4-bis(hydroxymethyl)cyclo­hexane. The bio­degradation of the polyorthoester takes place by hydrolysis to the ultimate prod­ucts 4-hydroxybutyrate (4HB) and cis,trans-1,4-bis(hydroxymethyl)cyclohexane (CHDM). 4HB is further metabo­lized in the tricarboxylic acid cycle, with CO2 and H2O as the end products. CHDM is excreted in urine (Sendelbeck and Girdis 1985). The polyorthoester has been devel­oped to be used as the ve­hicle in a sustained drug release system and the drug added is re­leased at a constant rate (Capozza et al. 1978, Benagiano et al. 1979). The polymer can be formulated with different physical properties. Polyorthoester of the formulation used in this work is soft and moldable. The polyorthoester adheres to bone surfaces and probably provides local hemo­stasis by plugging of spongy bone and secondarily pro­moting concentration of plate­lets and coagu­lation factors (Sudmann et al. 1990, Solheim et al. 1991).

INHIBITION OF BONE FORMATION

Systemic administration of the nonsteroidal anti-inflammatory drugs (NSAIDs), indomethacin, acetyl salisylic acid or ibuprofen, has been shown experimentally in animals and clinically to inhibit both heterotopic and orthotopic osteogenesis (Dahl 1974, Sudmann 1975, Rø et al. 1976, Sudmann et al. 1979, Sudmann and Bang 1979, Allen et al. 1981, Törnkvist et al. 1983, Elmstedt et al. 1985, Solheim et al. 1986, Sodemann et al. 1988). Further, locally ad­min­istered indomethacin contained in bioerodible polyorthoester gel has been shown to inhibit the healing of closed mid-diaphyseal fe­mur fractures in rats (Engesæter et al. 1992).

Experimentally in rats, Törnkvist (1985) found that in­do­methacin in doses of 2 or 3 mg/kg inhibited DBM-induced het­ero­topic osteogenesis, whereas rats treated with 6 mg/kg all died within one week of perforated gastric ulcer. DiCesare (1991) found a dose related inhibition on DBM-induced heterotopic osteogenesis by indomethacin over a range between 0.04 and 4 mg/kg body weight. In order to inhibit DBM-induced heterotopic osteogenesis in rats, indomethacin had to be present before or at the time of implan­tation of DBM (Törnkvist et al. 1985, Nilsson et al. 1986). A short period of indomethacin treatment at the time of implanta­tion of DBM is sufficient to reduce experimental bone formation, but the inhibitory effect slowly diminishes if the inductive process is continuous (Nilsson et al. 1986).

Inhibition of bone formation may be desirable in some clini­cal situations: (1) heterotopic ossification following total hip arthroplasty, (2) post-traumatic partial growth plate arrest, (3) cra­niosynostosis or (4) bone coalitions. Inhibition of bone formation may be accomplished by administration of NSAIDs or implanta­tion of interpositional materials. In arthroplasties, the NSAIDs in­domethacin or ibuprofen has been used either prophylactically at the time of the arthroplasty (Dahl 1974, Almåsbakk and Røysland 1977, Elmstedt et al. 1985, Sodemann et al. 1988) or after surgical removal of ectopic bone to prevent heterotopic ossification (Ritter and Gioe 1982, Kjærsgaard-Andersen and Schmidt 1986). Further, indomethacin inhibits osseous rebridging of the growth plate in rabbits (Sudmann et al. 1982).

Inhibition of osteogenesis by interpositional material has been accomplished clinically or experimentally in animals, by sur­gical removal of bone and interposition of (1) free fat transplant (Langenskiöld 1975, Merikanto et al. 1987, Olney and Asher 1987), (2) cartilage (Lennox et al. 1983), (3) muscle (Collins 1987), (4) bone wax (Friedenberg and Brashear 1956), (5) methyl methacrylate (Friedenberg 1957) or (6) silicone-rubber (Bright 1974). Implantation of nonresorbable materials may, however, induce a chronic inflammation, predispose for infections and require a second operation for removal of the implant. To prevent systemic adverse effects including inhibition of the osteogenesis in other parts of the skeleton than intended, a local bioerodible indometh­acin delivery system providing constant high local concentrations of the drug without systemic effects seems desirable.

LOCAL HEMOSTATICS

In bone surgery, nonabsorbable bone wax (Parker 1892, Horsley 1892) of 88% beeswax and 12% isopropylpalmitat, is being used for local hemostasis (Crenshaw 1987). Bone wax may, however, produce a chronic inflammation with foreign-body reaction (Geary and Frantz 1950), retard bone healing (Howard and Kelley 1969), predispose for infections (Culliford et al. 1976, Robicsek et al. 1978), impair bacterial clearance (Johnson and Fromm 1981) and cause wax embolization (Robicsek et al. 1981). These complica­tions has spurred the development of absorbable local hemostatics (Bergel 1909, Frantz et al. 1944, Correll and Wise 1945, Battista et al. 1967, Matras et al. 1972, Harris and Capperauld 1978, Silverstein et al. 1980, Sudmann et al. 1990), of which fibrin sealant (Matras et al. 1972), fibrin-collagen paste (Harris and Capperauld 1978) and bioerodible polyorthoester (Sudmann et al. 1990) may be the most suitable for bone surgery as they adhere readily to bleeding bone. The two former materials have been used in clinical orthopedic practice, while the latter has been shown in animal models to be suitable for such use (Sudmann et al. 1990).

AIMS OF THE PRESENT STUDY

The present work was designed to examine a new bioerodible polyorthoester as a DBM-carrier, a local hemostatic and a drug-delivery system in heterotopic and orthotopic models in rats. The specific objectives of the study are defined as listed.

1. To examine the relation between histomorphometry and 85Sr uptake of DBM-induced heterotopic bone (paper I), and thus, establish an accurate method for evaluating heterotopic bone forma­tion by 85Sr analyses.

2. To study the effect of bioerodible polyorthoester on DBM-induced heterotopic (paper II) or orthotopic (papers IV and V) osteogenesis.

3. To compare the host tissue response and the effect on osteoin­duction of three local hemostatics for osseous tissue, bone wax, fibrin-collagen paste and bioerodible polyorthoester with or without gentamicin (paper III).

4. To study the effect of a bioerodible local indomethacin delivery system on DBM-induced heterotopic osteogenesis (paper VI).

Summary of results

PAPER I

A method for assessing heterotopic bone formation by 85Sr analy­ses was established. Five, 10 or 15 mg DBM powder was implan­ted in 45 male Wistar rats and the relationship between the re­sults of histomorphometry and 85Sr uptake analysis of DBM-in­duced heterotopic bone was evaluated at 4 weeks postoperatively. Two indices of 85Sr uptake were calculated; the osteogenic index [(counts/min/mg implant)/(counts/min/mg os ilium)] and an index we have called the osteoquantum index in which the weight of the implant has been invalidated [(counts/min implant)/(counts/min/mg os ilium)]. The osteoquantum index was found to have a linear rela­tionship to the area of the induced bone with a correlation coeffi­cient of 0.90 (p<0.0001). Only weak linear relationships were found between the osteogenic index and the area of the new bone (r=0.32, p=0.03) and between the osteogenic index and the osteo­quantum index (r=0.33, p=0.03). The osteoquantum index and the area of the induced bone both increased proportionally to the quantity of implanted DBM, whereas the osteogenic index did not change sig­nificantly.

PAPER II

Bioerodible polyorthoester was evaluated as an osteoinductor de­livery system in 89 male Wistar rats by implantation of DBM, polyorthoester or a composite of the two in the abdominal muscle of the animals. Heterotopic bone formation and tissue reaction were evaluated by light microscopy in 44 rats at weeks 1, 2, 3, 4, 6 and 8 and strontium85 uptake in 45 rats at week 4. Compos­ite of polyorthoester and DBM induced cartilage and bone at the same rate as the DBM alone. The composite implant was technically easier to use than the DBM alone. Around the implants of the polyorthoester and the composite, inflammation with some giant cells was present until week two. The polyorthoester was clearly seen until 3-4 weeks but later only occasional traces could be identified.

PAPER III

The tissue response and effect on DBM-induced het­erotopic osteogenesis in the abdominal muscle of 120 male Wistar rats by different local hemostatics were evaluated by light micros­copy and 85Sr uptake analyses. Nonabsorbable bone wax of 88 % beeswax and absorbable bovine fibrin-collagen paste both significantly inhibited osteoinduction, whereas bioerodible poly­orthoester drug delivery system with or without 4% gentamicin caused no signifi­cant inhibition. Bone wax was not absorbed and induced a chronic foreign body reaction. Fibrin-collagen paste induced less inflam­mation with numerous monocytes and macrophages with engulfed material. Bioerodible polyorthoester caused a very moderate tissue reaction and was mostly resorbed at week 4.

PAPER IV

In 36 male Wistar rats, the healing of critical size calvarial defects without implant or filled with DBM, polyorthoester or a composite of the two former was studied. Defects filled with the composite of polyorthoester and DBM or the DBM alone were bridged by bone at 4 weeks histologically and radiographically, whereas unfilled defects or defects filled with the polyorthoester only, did not heal. The polyor­thoester caused a slight inflammation that subsided by 3 weeks. Only traces of the polyorthoester could be detected at 4 weeks. The polyorthoester provided local hemostasis when used either alone or in composites with DBM. The composite implant was moldable and easily contoured and technically easier to use than the DBM alone.

PAPER V

The healing of large osteoperiosteal defects of 50 % of the length of the radius was studied in 60 male Wistar rats. The defects were left with no implant or filled with DBM, polyorthoester or a compos­ite of the two. At 50 days the specimens were evaluated by light microscopy and measurement of bone formation within the original defect on radiographs by Sigma-Scan Measurement System includ­ing an electromagnetic digitizing tablet and a per­sonal computer with Sigma scan software. Defects filled with the composite of polyor­thoester and DBM or the DBM alone showed re­generation of bone cor­responding to respectively 93.6 per cent (SD 44) and 77.6 per cent (SD 30) of the area of the defect. Defects with no implant or de­fects filled with the polyorthoester only, showed significantly less re­generation. The DBM used alone was rapidly mixed with blood form­ing a grainy mixture that tended to be displaced, whereas the com­posite implant was moldable, easy to place in the defect and the tendency for displace­ment was less. The polyorthoester provided local hemostasis when used either alone or in composites with demineralized bone.

PAPER VI

The effect of a bioerodible system for local delivery of indomethacin on DBM-induced heterotopic bone formation in rat abdominal muscle was evaluated. Two separate series were conducted in a total of 48 Wistar rats. In both series two types of implants were used: (A) control group, polyorthoester and DBM; or (B) experimental group, polyorthoester with 5 % indomethacin and DBM. In the first series, host-tissue responses and os­teo­induction were evaluated at 2, 3 and 4 weeks histologically. In the second series, bone formation was quantified by 85Sr uptake at week 4. Blood samples were obtained under anesthesia from 3 random rats of each group in Series I at week 1 and of all rats of group B (polyorthoester with 5 % indomethacin and DBM) prior to death at week 2-4. Plasma was prepared and stored at -70°C until analyses. Indomethacin in plasma was assayed by a specific High Performance Liquid Chromatographic method with UV-detection at 320 nanometers. The detection limit was 25 nano­grams per milliliter. The bioerodible system for local delivery of indo­met­hacin significantly inhibited demineralized bone-induced het­ero­topic bone formation as evaluated by light mi­croscopy and 85Sr uptake. The polyorthoester, with or without drug, caused little tissue reaction and was mostly resorbed at week 4. Indomethacin could not be detected in any of the blood samples.

General discussion

MATERIAL AND METHODS

Animals

Young male Wistar rats were chosen as the experimental animals in this study. Rats are well standardized, easily available and their short life span permits experimental intervention for a significant frac­tion of their lifetime. Rats retain a basically non-Haversian lamellar bone structure throughout their life, however, the physio­logical mechanisms of bone remodeling are similar to that of humans (Simmons 1976). Models for evaluating both heterotopic (Urist 1965, Elves 1974) or orthotopic (Mulliken and Glowacki 1980, Schmitz and Hollinger 1986, Gepstein et al. 1987) osteo­genesis have been established in rats.

The animals were fed standard laboratory food and water ad libitum. Anesthesia was induced by intramuscular injec­tion of a combination of 0.075 mg fentanyl, 3.75 mg fluanisone and 1.875 mg midazolam (Hypnorm®-Dormi­cum®) per kilogram of body weight (Flecknell and Mitchell 1984). The research protocol was approved by The Norwegian State Commission for the Regulation of Animal Experiments. The national guidelines for the care and use of laboratory animals were observed.

Implants

Demineralized bone matrix (DBM) was prepared by sterile tech­nique from the femur, tibia and fibula of male Wistar rats of the same age and weight as the recipients of the corresponding series, except in paper IV where the donor rats were older than the recipients. Dis­sected diaphyses were crushed and the marrow was removed. The cortical bone was cut into chips and demineralized in 0.2 N HCl for 48 hours at 4°C and flushed in saline (Bang 1973). The demineralized bone chips were suspended in liquid nitrogen and lyophilized for 22 hours. Several protocols for preparing deminer­alized bone have been proposed (Urist et al. 1975, Reddi and Huggins 1972, Bang 1973). The more extensive processing of the bone in the antigen-extracted autodigested alloimplant (AAA) pro­tocol (Urist et al. 1975) may lead to some loss of inductive poten­tial compared to demineralization and lyophiliazation only, as used in the present work (Delloye et al. 1985, Marinak et al. 1989).

Whereas non-sterile demineralized bone may be used in small laboratory animals without inconvenience (Schwarz et al. 1988), the clinical success of this implant depends on its sterility (Harakas 1984). Sterility may be difficult to maintain during har­vesting of cadaver bone and the subsequent processing of the bone and demineralization in HCl does not seem to assure sterility (Dahners and Hoyle 1989). Thus, ethylene oxide sterilization has been used in many studies by different investigators, both experi­mentally (Bang 1973, Nilsen 1977, Syftestad and Urist 1982, Schmitz and Hollinger 1988, Pinholt et al. 1990) and clinically (Urist 1968, Ousterhout 1985), apparently without deleterious effects on the osteoinduction. Thus, in the first part of the present study, the DBM was sterilized in ethylene oxide gas (Alcon Universal Ltd., Fort Worth, Texas, USA) for 3 hours (Bang 1973). The normal histologic sequence of induction was observed in ethylene oxide-sterilized DBM with or without polyorthoester car­rier. However, as recent studies has indicated that such sterilization may reduce the osteoinductive potency of the implant (Munting et al. 1988, Aspenberg et al. 1990), no sterilization of the DBM was used in the latter part of the work.

DBM was used either as chips (0.5 x 2.0 x 2.0 mm; weight 0.7 mg) or as a coarse powder (0.1-2.0 mm2). The size was assessed by area measurements on photomicrographs of random samples. The osteoinductive potential may increase with decreas­ing particle size (Syftestad and Urist 1979, Glowacki et al. 1981a, Sampath and Reddi 1984). Pulverization to particle size less than approximately 0.1 mm may, however, cause reduction of the osteo­inductive potential, probably by denaturation of contained induc­tive proteins (Syftestad and Urist 1979, Sampath and Reddi 1984).

The demineralized bone was kept at 4°C and implanted within 48 hours. When stored at this temperature, DBM has been shown to retains its osteoinductivity for a period of up to six months (Hosny et al. 1987).

The total mass of the implanted DBM ranged from 2.8 mg to 15 mg. Whereas Muthukumaran et al. (1988) found a threshold for bone induction at 10 mg DBM, in the present work, typical histologic signs of induction was found in all experiments regardless of the mass of the DBM; and in paper I, we found no qualitative dif­ferences of the bone induced by 5, 10 or 15 mg DBM. In the same paper, both the osteoquantum index and the area of the induced bone increased proportionally to the quantity of implanted DBM. These findings agree with those of earlier studies in which the area of induced bone on radiographs was directly proportional to the weight of implanted osteoinductor, bone morphogenetic protein (Kawai and Urist 1988, Mahy and Urist 1988).

In our studies of heterotopic osteogenesis by polyorthoester and DBM, spherical composite implants were made by mixing approximately equal volumes of DBM as chips or pow­der and poly­orthoester manually at room temperature under sterile condi­tions immediately before implantation. The DBM particles were partly exposed and partly embedded in the polyorthoester.

Heterotopic model

DBM-induced heterotopic osteogenesis in rat abdominal muscle was used as the experimental model for evaluating biocompatibility and effect on osteoinduction by polyorthoester (paper II), local hemostatics (paper III) and polyorthoester indomethacin delivery system (paper VI). This model has several advantages. First, DBM regularly induces heterotopic bone in rodents and the process has been well described (Urist 1965, Reddi and Huggins 1972). Sec­ondly, composites of biomaterials and osteoinductor should first be evaluated by their ability to induce heterotopic bone, as new bone formation in orthotopic sites may be due to osteoblastic activity by the bone proper, i.e., osteoconduction, making quantification of the osteoinductive potential of the composite difficult. Finally, to evaluate the effect of the biomaterial on osteoinduction, the composite of biomaterial and osteoinductor should be compared with the osteoinductor alone. This is, however, difficult to accomplish when purified, soluble osteoinductors are used as the osteoinductor is rapidly absorbed before bone is induced (Urist et al. 1984).

Heterotopic osteoinduction has been evaluated by both (1) qualitative methods: light microscopy (Urist 1965, Reddi and Huggins 1972), radiography (Urist 1965) and electron microscopy (Nilsen 1977) and (2) quantitative methods: ash weight (Urist et al. 1970), calcium content (Urist et al. 1970), alkaline phosphatase activity of implant (Firschein and Urist 1972), 85Sr uptake analyses (Elves 1974), 45Ca uptake analyses (Reddi 1975), histomorphome­try (Hosny and Sharawy 1985, Marinak et al. 1989) and computer­ized image analyses of the area of induced bone on radiographs (Kawai and Urist 1988).

In the present work, qualitative histologic examination (papers II, III and VI), computer-assisted histometry (paper I) and 85Sr uptake (papers I, II, III and VI) were used to evaluate os­teoin­duction. In paper II, qualitative histologic examination was per­formed at 1, 2, 3, 4, 6 and 8 weeks to examine the effect of the carrier on different stages of osteoinduction. It was shown (paper II), in concordance with previous studies (Reddi and Huggins 1972), that mature ossicles were formed by 4 weeks. Thus, 4 weeks postoperatively was chosen as an appropriate point of time for his­to­logic evaluation in pa­per III. In paper VI, qualitative histologic examination was per­formed at 2, 3, and 4 weeks to examine the effect of the indometh­acin deliv­ery system on both cartilage and bone formation.

85Sr uptake at 4 weeks postoperatively was chosen for quanti­tative evaluation of osteoinduction in papers II, III and VI. 85Sr has been used for evaluation of bone grafts both as (1) 85Sr uptake in grafts, calculated as the osteogenic index [(counts/min/mg implant)/(counts/min/mg os ilium)] (Elves 1974, Elves 1975, Delloye et al. 1985, Munting et al. 1988) or the total content (cpm/implant) (Yoshikawa et al. 1988) and (2) loss of 85Sr from prelabeled grafts (Rønningen et al. 1985, Solheim et al. 1986).

In paper I, the 85Sr uptake expressed as the novel osteo­quan­tum index [(counts/min implant)/(counts/min/mg os ilium)] showed a linear relationship to the histometric area of the induced bone with a correlation coefficient (r) of 0.90, whereas only weak linear relationships were found between the osteogenic index and the histometric area of the bone (r=0.32) and between the osteogenic index and the osteoquantum index (r=0.33). Thus, the os­teoquantum index was used in papers II, III and VI; in paper II in addition to the osteogenic index. As the weight of the implant is disregarded in the osteoquantum index, this index, in contrast to the osteogenic index, permits evaluation of the effect of biomateri­als on osteoinduction in composites without influence of density and biodegradation of the biomaterial as long as the same amount of osteoinductor is used.

Orthotopic models

Both cranial defects and radial defects in rodents are commonly used models for evaluation of autogenous bone substitutes. The po­tential for regeneration of cranial defects is low, especially in adults. Defects that do not heal during the animals’ lifetime are termed critical size defects and it has been proposed that evaluation of cranio-maxillofacial bone repair materials should be initiated in critical size defects of the rat calvaria (Schmitz and Hollinger 1986). Nor do large radial diaphyseal defects in rats heal spontane­ously (Gepstein et al. 1987, Alper et al. 1989). The absence of supination and the radio-ulnar synostosis contribute to the stability of the bones and to the retention of the implanted material (Alper et al. 1989).

When DBM is used for promoting healing of bone defects, the mode of action of the implant is more difficult to establish than in the heterotopic model as osteoclasts and osteoblasts are already present and might be stimulated by the surgical trauma.

Regeneration of bone defects by DBM is a multistep process. In paper IV, we wanted to evaluate the effect of the polyor­thoester on the different phases of the healing of calvarial defects. Thus, a positive treatment response was defined as osteoinduction, i.e., bone formation within the defect and not in contact with the bone edges at weeks 2 and 3 histologically and bone bridging of the defects at week 4 histologically and radiographically. The treatment response of each rat was recorded as 0 (no induction or bone bridging), 1 (induction or bone bridging in one defect) or 2 (induction or bone bridging in both defects).

This is, of course, a rather crude semi-quantitative evalu­ation. For statistical analyses, the results of the radiographic analy­ses should preferably be continuous data of high accuracy. For this purpose, measurements of the area of new bone formation on radio­graphs by digitizer and area analyses software have been used in heterotopic models (Kawai and Urist 1988, Mahy and Urist 1988). Further, we were primarily interested in the effect of the polyor­thoester on the final result of the process. Thus, in paper V, we measured the area of bone formed within the original defect on radio­graphs by Sigma-Scan Measurement System including an electro­magnetic digitizing tablet and a personal computer with Sigma scan software at 50 days postoperatively. In both papers IV and V the host-tissue response and osteoinduction were evaluated by qualitative his­tologic examination.

RESULTS

Handling properties and local hemostasis

Both in the heterotopic model (paper II) and in the two orthotopic models (papers IV and V) the DBM was rapidly mixed with blood forming a grainy mass that tended to be displaced. In contrast, the composite of DBM and polyorthoester was easy to place in the muscle pouch or bone defect and the tendency for displacement was much less. The polyorthoester and the composite had a plastic con­sistency and could easily be molded peroperatively according to needs. The bioerodible polyorthoester adhered to bone surfaces and provided local hemostasis, probably by plugging of spongy bone and secondarily promoting concentration of platelets and coagula­tion factors (Sudmann et al. 1990, Solheim et al. 1991). The evalu­ation of both handling properties and local hemostatic effect was purely qualitative.

Heterotopic osteoinduction

A requirement for an osteoinductor delivery system is that the carrier does not inhibit osteoinduction. In composites of carrier and osteoinductor, the osteoinduction may be inhibited by the carrier by two mechanisms. First, unabsorbed carrier will physically in­hi­bit bone growth and bony bridging of a defect. According to Urist et al. (1987) the ideal delivery system should be resorbed and replaced by preosseous tissues within two weeks, cartilage within three weeks and bone within four to six weeks. Some existing de­livery systems do not fulfill these requirements as they are resorbed more slowly than it takes for the DBM-induced regeneration of the defect to take place, thus, inhibiting or delaying the bone formation (Urist et al. 1987, Schmitz and Hollinger 1988, Lovell et al. 1989). Secondly, osteoinduction may be inhibited by a bioincompatible carrier that interferes physiologically with some part of the mult­istep cascade of bone induction, e.g., by inducing a chronic inflam­mation (Sela et al. 1986, Alper et al. 1989).

By qualitative microscopy, no difference in bone induction could be identified between the composite of polyorthoester and DBM (papers II, III and IV). At week one, osteoprogenitor cells encapsulated the DBM and some chondrocytes were seen. At week two, cartilage and incipient bone formation were present. At week three, more bone was seen and by week four, ossicles with bone marrow had formed. Around the implants of polyorthoester and composite an in­flammation with some giant cells was present until week two. The polyorthoester was seen until week 3-4 but there­after only traces could be identified in some sections. The composite of DBM and polyorthoester showed highly significant (p<0.0001) increased 85Sr uptake compared with that of bioero­dible polyorthoester alone (paper II), whereas no significant differ­ence in 85Sr uptake was found between the composite and the DBM (papers II and III). Thus, it seems unlikely that bioerodible polyorthoester inhibited bone formation by either of the two above-mentioned mechanisms.

In contrast to the polyorthoester, two local hemostatics in clini­cal use; ordinary nonabsorbable bone wax of 88% beeswax and fi­brin-collagen paste were incompletely absorbed, induced a chronic inflammation and inhibited osteoinduction (paper III). The bone wax caused an inflammation with numerous multinuclear giant cells while fibrin-collagen paste caused an inflammation with ac­cumulation of macrophages and proliferation of fibroblasts. Con­nective tissue and inflammatory cells were seen uniformly distributed through all parts of the composite implants of DBM and bone wax or fibrin-colla­gen paste, interspersed between the partly resorbed DBM and the local hemostatic. Thus, the lack of osteo­induction cannot be explained solely as an effect of a physical barrier of the local he­mostatic. It seems likely that the chronic inflammation induced may be the cause of the inhibition of osteo­induction. Clinically, the re­sult of the study (paper III) suggest that bone wax and fibrin-colla­gen paste should be used sparingly, whereas polyorthoester with or without gentamicin seems promis­ing as a local hemostatic for use in bone surgery.

Polyorthoester with 5% indomethacin also significantly inhib­ited DBM-induced heterotopic osteogenesis (paper VI). However, as opposed to bone wax and fibrin-collagen paste, the drug delivery system was absorbed and induced no persistent inflamma­tion. Further, no indomethacin was detectable in blood samples. The results indicate that the inhibition was caused by a specific local drug effect, e.g., inhibition of local prostaglandin synthesis.

Orthotopic osteogenesis

The critical size defect of the calvaria varies according to species and age. It has been shown that 2 mm parietal defects in 500 g Wistar rats (Freeman and Turnbull 1973), 4 mm parietal defects in 28 day old CD strain, Charles River Breeding Laboratories (Mulliken and Glowacki 1980) and 8 mm parietal defects in 6 months old Sprague-Dawley rats (Tagaki and Urist 1982) do not heal in 12 weeks, 6 months and 12 weeks respectively. We used 4 mm parietal de­fects in 8 weeks old Wistar male rats and no defect without implant healed during the observation period, while all defects filled with the composite or DBM were completely bridged by bone histologically and radiographically by 4 weeks post­opera­tively (paper IV).

In the defects filled with the composite or polyorthoester, the polyor­thoester was present and slight inflammation with mono­cytes and some giant cells was seen at week two (paper IV). The inflammation subsided by week three and only traces of the polyorthoester could be de­tected at week four. In the defects filled with DBM or without implant, the in­flammation was milder and no giant cells were de­tected at week two.

In the radial defects, no significant difference was found between the area of new bone by DBM alone and the composite of DBM and polyorthoester radiographically (paper V). The area of new bone was significantly less in the defects with no implant or filled with polyor­thoester only. Only a few of the defects filled with DBM or the com­posite were fully bridged by new bone. Typically, union of the new bone with the proximal fragment of the radius had occurred, while a small defect remained distally. In the defects with no implant or filled with polyorthoester only some bone formation was typically seen at the bone edges.

Conclusions

1. The novel osteoquantum index was found to have a linear rela­tionship to the area of the induced bone with a correlation coeffi­cient (r) of 0.90 and high statistical significance (p<0.0001). Only weak linear relationships were found between the osteogenic index and the area of the bone (r=0.32, p=0.03) and between the os­teogenic index and the osteoquantum index (r=0.33, p=0.03) (paper I). The osteoquantum index and the area of the induced bone both increased with increasing mass of implanted DBM, whereas the osteogenic index did not change significantly (paper I).

2. Bioerodible polyorthoester did not inhibit heterotopic osteoin­duction, it caused only a slight inflammation that subsided within 3 weeks postoperatively and it was mostly absorbed by week 4 (papers II, III and VI). In contrast, nonabsorbable bone wax and fibrin-collagen paste both were incompletely absorbed, induced a chronic inflammation and inhibited osteoinduction (paper III).

3. The composite of polyorthoester and DBM induced bony heal­ing of large calvarial (paper IV) and radial defects (paper V) as DBM alone. Moreover, the composite was moldable and easily contoured, and technically easier to use than DBM alone. Finally, the polyorthoester provided local hemostasis when used either alone or in composites with demineralized bone (papers IV and V).

4. Polyorthoester with 5% indomethacin significantly inhibited DBM-induced heterotopic osteogenesis, probably by local mecha­nisms as indomethacin could not be detected in any blood sample (paper VI).

References

Allen HA, Wase A, Bear WT. Indomethacin and aspirin: Effect of nonsteroidal anti-inflammatory agents on the rate of fracture re¬pair in the rat. Acta Orthop Scand 1981;51:595-600.

Almåsbakk K, Røysland P. Does indomethacin prevent postopera¬tive ectopicossification in total hip replacement. Acta Orthop Scand 1977;48:556.

Alper G, Bernick S, Yazdi M, Nimni ME. Osteogenesis in bone defects in rats: the effects of hydroxyapatite and demineralized bone matrix. Am J Med Sci 1989;298(6):371-6.

Aspenberg P, Andolf E. Bone induction by fetal and adult human bone matrix in athymic rats. Acta Orthop Scand 1989;60(2):195-9.

Aspenberg P, Johnsson E, Thorngren K-G. Dose dependent reduc¬tion of bone inductive properties by ethylene oxide. J Bone Joint Surg 1990;72-B(6):1036-7.

Aspenberg P, Lohmander LS. Fibroblast growth factor stimulates bone formation. Bone induction studied in rats. Acta Orthop Scand 1989;60(4):473-6.

Aspenberg P, Lohmander LS, Thorngren K-G. Failure of bone in¬duction by bone matrix in adult monkeys. J Bone Joint Surg 1988;70-B(4):625-7.

Aspenberg P, Wittbjer J, Thorngren KG. Bone matrix and marrow versus cancellous bone in rabbit radial defects. Arch Orthop Trauma Surg 1987;106(6):335-40.

Aspenberg P, Wittbjer J, Thorngren KG. Pulverized bone matrix as an injectable bone graft in rabbit radius defects. Clin Orthop 1986;(206):261-9.

Bang G. Induction of heterotopic bone formation by demineralized dentin: an experimental model in guinea pigs. Scand J Dent Res 1973;81:240-50.

Battista OA, Erdi NZ, Ferraso CF. Colloidal macromolecular phe¬nomena: II. Novel microcrystals of polymer. J Appl Polymer Sci 1967;11:481-98.

Bauer HC, Törnkvist H, Nilsson OS. Effects of bisphosphonates on the incorporation of calcium-45 and 3H-proline in orthotopic and in demineralized matrix-induced bone in rats. Bone 1986;7(2):129-35.

Belanger LF, Robichon J, Ontario O, Urist MR. The effects of magnesium deficiency on the host response to intramuscular bone matrix implanted in the rat. J Bone Joint Surg 1975;57-A(4):522-6.

Benagiano G, Schmitt E, Wise D, Goodman M. Sustained release hormonal preparations for the delivery of fertility-regulaing agents. J Polymer Sci 1979;Polymer Symp 66:129-48.

Bentz H, Nathan RM, Rosen DM, Armstrong RM, Thompson AY, Segarini PR et al. Purification and characterization of a unique os¬teoinductive factor from bovine bone. J Biol Chem 1989;264(34):20805-10.

Bergel S. Ueber Wirkungen des Fibrins. Dtsch Med Wochenschr 1909;35:663-5.

Bright RW. Operative correction of partial epiphyseal plate closure by osseous-bridge resection and silicone-rubber implant:an ex¬perimental study in dogs. J Bone Joint Surg 1974;56-A(4):655-64.

Burnett CC, Reddi AH. Influence of estrogen and progesterone on matrix-induced endochondral bone formation. Calcif Tissue Int 1983;35(4-5):609-14.

Canalis E, McCarthy T, Centrella M. Isolation of growth factors from adult bovine bone. Calcif Tissue Int 1988;43(6):346-51.

Capozza R, Sendelbeck L, Balkenhol W. Preparation and evalu¬ation of a bioerodible naltrexone delivery system. Kostelnick RJ, Editor. Polymeric delivery systems. New York: Gordon and Breach Science Publishers; 1978:59-73.

Chalmers J. Transplantation immunity in bone homografting. J Bone Joint Surg 1959;41-B(1):160-79.

Chang TMS. Biodegradable semipermeable microcapsules contain¬ing enzymes, hormones, vaccines, and other biologicals. J Bio¬eng 1976;1:24-32.

Collins B. Tarsal coalitions. A new surgical procedure. Clin Podiatr Med Surg 1987 Jan;4(1):75-98.

Correll JT, Wise EC. Certain properties of a new physiologically absorbable sponge. Proc Soc Exp Biol Med 1945;58:233-5.

Culliford AT, Cunningham JN, Zeff RH, Isom OW, Teiko P, Spencer FC. Sternal and costochondral infections following open-heart surgery. A review of 2,594 cases. J Thorac Car-diovasc Surg 1976;72(5):714-26.

Cutright DE, Hunsuck EE, Beasley JD. Fracture reduction using a biodegradable material, polylactic acid. J Oral Maxillofac Surg 1971;29:393-7.
Dahl, H. K. Kliniske obser¬vasjoner, in Symposium om artrose. Drammen: Merck Sharp & Dome 1975, pp 37-46.

Dahners LE, Hoyle M. Chemical sterilization of bacterially con¬taminated bone without destruction of osteogenic potential. J Orthop Trauma 1989;3(3):241-4.

Delloye C, Hebrant A, Munting E, Piret L, Coutelier L. The os¬teoinductive capacity of differently HCl-decalcified bone allo¬im¬plants. Acta Orthop Scand 1985;56:318-22.

DeSimone DP, Reddi AH. Influence of vitamin A on matrix-in¬duced endochondral bone formation. Calcif Tissue Int 1983;35(6):732-9.

DiCesare PE, Nimni ME, Peng L, Yazdi M, Cheung DT. Effects of indomethacin on demineralized bone-induced heterotopic ossifi¬ca¬tion in the rat. J Orthop Res 1991;9(6):855-61.

Driskell TD, O’Hara MJ, Sheets HD Jr, Greene GW Jr, Natiella JR, Armitage J. Development of ceramic and ceramic composite de¬vices for maxillofacial applications. J Biomed Mater Res 1972;6(1):345-61

Einhorn TA, Lane JM, Burstein AH, Kopman CR, Vigorita VJ. The healing of segmental bone defects induced by demineral¬ized bone matrix. A radiographic and biomechanical study. J Bone Joint Surg 1984;66-A(2):274-9.

Elmstedt E, Lidholm TS, Nilsson OS, Törnkvist H. Effect of ibup¬rofen on heterotopic ossification after hip replacement. Acta Orthop Scand 1985;56:25-7.

Engesæter LB, Sudmann B, Sudmann E. Fracture healing in rats in¬hibited by locally administered indomethacin. Acta Orthop Scand 1992;63(3):330-33.

Farley JR, Baylink DJ. Purification of a skeletal growth factor from human bone. Biochemistry 1982;21(14):3502-07.

Flecknell PA, Mitchell M. Midazolam and fentanyl-fluanisone: as¬sessment of anaesthetic effects in laboratory rodents and rabbits. Lab Anim 1984;18(2):143-6.

Frantz VK, Clarke HT, Lattes R. Hemostasis with absorbable gauze (oxidized cellulose). Ann Surg 1944;120(2):181-98.

Freeman E, Turnbull RS. The value of osseous coagulum as a graft material. J Periodont Res 1973;8:229-36.
Friedenberg ZB, Brashear R. Bone growth following partial resec¬tion of the epiphyseal cartilage. Am J Surg 1956;91:362-4.

Friedenberg ZB. Reaction of the epiphysis to partial surgical resec¬tion. J Bone Joint Surg 1957;39-A(2):332-40.

Gallie WE, Robertson DE. The repair of bone. Br J Surg 1929;7:211-23.

Geary JR, Frantz VK. New absorbable bone wax surgery. Ann Surg 1950;132:1128-37.

Gepstein R, Weiss RE, Hallel T. Bridging large defects in bone by demineralized bone matrix in the form of a powder. A radio¬graphic, histological, and radioisotope-uptake study in rats. J Bone Joint Surg 1987;69-A(7):984-92.

Gerhart TN, Renshaw AA, Miller RL, Noecker RJ, Hayes WC. In vivo histologic and biomechanical characterization of a biode¬grad¬able particulate composite bone cement. J Biomed Mater Res 1989;23(1):1-16.

Getter L, Cutright DE, Baskar SN, Augsburg JK. A biodegradable intraosseous appliance in the treatment of mandibular fractures. J Oral Surg 1972;30:344-8.

Glowacki J, Altobelli D, Mulliken JB. Fate of mineralized and demineralized osseous implants in cranial defects. Calcif Tissue Int 1981a;33(1):71-6.

Glowacki J, Kaban LB, Murray JE, Folkman J, Mulliken JB. Application of the biological principle of induced osteogenesis for craniofacial defects. Lancet 1981b;1(8227):959-62.

Glowacki J, Mulliken JB. Demineralized bone implants. Clin Plast Surg 1985;12(2):233-41.

Gospodarowicz D. Fibroblast growth factor. Chemical structure and biological function. Clin Orthop 1990;(257):231-48.

Gupta D, Tuli SM. Osteoinductivity of partially decalcified alloim¬plants in healing of large osteoperiosteal defects. Acta Orthop Scand 1982;53(6):857-65.

Harakas N. Demineralized bone-matrix-induced osteogenesis. Clin Orthop 1984;(188):239-51.

Harris P, Capperauld I. A new absorbable haemostatic bone sealent. J R Coll Surg Edinb 1978;23:285-91.

Harris P, Capperauld I. Clinical experience in neurosurgery with Absele: A new absorbable haemostatic bone sealent. Surg Neu¬rol 1980;13:231-5.

Hauschka PV, Chen TL, Mavrakos AE. Polypeptide growth factors in bone matrix. Ciba Found Symp 1988;136:207-25.

Hauschka PV, Mavrakos AE, Iafrati MD, Doleman SE, Klagsbrun M. Growth factors in bone matrix. Isolation of multiple types by affinity chromatography on heparin-Sepharose. J Biol Chem 1986;261(27):12665-74.

Heller J, Penhale DWH, Helwing RF, Fritzinger BK. Release of norethindrone from poly(ortho esters). Polym Eng Sci 1981;21(11):727-31.

Higashi S, Yamamuro T, Nakamura T, Ikada Y, Hyon S-H, Jamshidi K. Polymer-hydroxyapatite composites for biodegrad¬able bone fillers. Biomaterials 1986;7:183-7.

Hollinger JO, Battistone GC. Biodegradable bone repair materials. Synthetic polymers and ceramics. Clin Orthop 1986;(207):290-305.

Hollinger JO. Preliminary report on the osteogenic potential of a biodegradable copolymer of polylactide (PLA) and polyglycol¬ide (PGA). J Biomed Mater Res 1983;17:71-82.

Hopp SG, Dahners LE, Gilbert JA. A study of the mechanical strength of long bone defects treated with various bone autograft substitutes: an experimental investigation in the rabbit. J Orthop Res 1989;7(4):579-84.

Horsley V. Antiseptic wax. Br Med J 1892;1:1165.

Hosny M, Arcidi C, Sharawy M. Effects of preservation on the osteoinductive capacity of demineralized bone powder allo¬grafts. J Oral Maxillofac Surg 1987;45(12):1051-4.

Hosny M, Sharawy M. Osteoinduction in rhesus monkeys using demineralized bone powder allografts. J Oral Maxillofac Surg 1985;43:837-44.

Howard TC, Kelley RR. The effect of bone wax on the healing of experimental rat tibial lesions. Clin Orthop 1969;(63):226-31.

Howes R, Bowness JM, Grotendorst GR, Martin GR, Reddi AH. Platelet-derived growth factor enhances demineralized bone matrix-induced cartilage and bone formation. Calcif Tissue Int 1988;42(1):34-8.

Johnson P, Fromm D. Effects of bone wax on bacterial clearance. Surgery 1981;89:206-9.

Kaban LB, Glowacki J. Induced osteogenesis in the repair of experimental mandibular defects in rats. J Dent Res 1981;60(7):1356-64.

Kapur SP, Reddi AH. Influence of testosterone and dihydrotestos¬terone on bone-matrix induced endochondral bone formation. Calcif Tissue Int 1989;44(2):108-13.

Kawai T, Urist MR. Quantitative computation of induced hetero¬topic bone formation by an image analysis system. Clin Orthop 1988;(233):262-7.

Kawamura M, Urist MR. Growth factors, mitogens, cytokines, and bone morphogenetic protein in induced chondrogenesis in tissue culture. Dev Biol 1988;130(2):435-42.

Köhler P, Kreicbergs A. Incorporation of autoclaved autogeneic bone supplemented with allogeneic demineralized bone matrix. An experimental study in the rabbit. Clin Orthop 1987;(218):247-58.

Kjærsgaard-Andersen P, Schmidt SA. Indomethacin for prevention of ectopic ossification after hip arthroplasty. Acta Orthop Scand 1986;57:12-4.

Kümmell H. Ueber Knochenimplantation. Deutsch med Wochen¬schr 1891;17:389-92.
Lacroix P. Recent investigations on the growth of bone. Nature (London) 1945;156:576.

Langenskiöld A. The possibilities of eliminating premature partial closure of an epiphyseal plate caused by trauma or disease. Acta Orthop Scand 1967;38:267-79.

Langer R. Biopolymers in controlled release systems, in Piskin E, Hoffman AS (eds.) Polymeric Biomaterials. Dordrecht: Martinus Nijhof Publishers 1986, pp 161-9.

Lennox DW, Goldner RD, Sussman MD. Cartilage as an interposi¬tion material to prevent transphyseal bone bridge formation: an ex¬perimental model. J Pediatr Orthop 1983;3(2):207-10.

Levander G. A study of bone regeneration. Surg Gynec Obster 1938;67:705-14.

Lindholm TS, Nilsson OS, Lindholm TC. Extraskeletal and in¬traskeletal new bone formation induced by demineralized bone matrix combined with bone marrow cells. Clin Orthop 1982;(171):251-5.

Linkhart TA, Jennings JC, Mohan S, Wakley GK, Baylink DJ. Characterization of mitogenic activities extracted from bovine bone matrix. Bone 1986;7(6):479-87.

Linkhart TA, Jennings JC, Mohan S, Wakley GK, Baylink DJ. Characterization of mitogenic activities extracted from bovine bone matrix. Bone 1986;7(6):479-87.

Lovell TP, Dawson EG, Nilsson OS, Urist MR. Augmentation of spinal fusion with bone morphogenetic protein in dogs. Clin Orthop 1989;(243):266-74.

Lucas PA, Laurencin C, Syftestad GT, Domb A, Goldberg VM, Caplan AI et al. Ectopic induction of cartilage and bone by water-soluble proteins from bovine bone using a polyanhydride delivery vehicle. J Biomed Mater Res 1990;24(7):901-11.

Lymann DJ, Seare WJ. Biomedical materials in surgery. Ann Rev Mater 1974;4:415-34.

Mackey D, Varlet A, Debeaumont D. Antibiotic loaded plaster of paris pellets:An in vitro study of a possible method of local antibiotic therapy in bone infection. Clin Orthop 1982;(167):263-8.

Mahy PR, Urist MR. Experimental heterotopic bone formation in¬duced by bone morphogenetic protein and recombinant human interleukin-1B. Clin Orthop 1988;(237):236-44.

Marden LJ, Reddi AH, Hollinger JO. Growth and differentiation factors: role in bone induction and potential application in craniofacial surgery. J Craniofac Surg 1990;1(3):154-60.

Marinak KW, Mellonig JT, Towle HJ. The osteogenic potential of two human demineralized bone preparations using a xenogeneic model. J Periodontol 1989;60(1):12-8.

Matras H, Dinges HP, Lassmann H, Mamoli B. Zur nathlosen interfaszikulären Nerventransplantation im Tierexperiment. Wien Med Wochenschr 1972;122(37):517-23.

Merikanto JE, Alhopuro S, Ritsil:a VA. Free fat transplant prevents osseous reunion of skull defects. A new approach in the treat¬ment of craniosynostosis. Scand J Plast Reconstr Surg Hand Surg 1987;21(2):183-8.

Miller AG. A case of bone grafting with decalcified bone chips; Remarks. Lancet 1890;2:618-9.

Mohan S, Baylink DJ. Bone growth factors. Clin Orthop 1991;(263):30-48.

Mulliken JB, Glowacki J. Induced osteogenesis for repair and construction in the craniofacial region. Plast Reconstr Surg 1980;65(5):553-60.

Munting E, Wilmart J-F, Wijne A, Hennebert P, Deloye C. Effect of sterilization on osteoinduction. Comparison of five methods in demineralized rat bone. Acta Orthop Scand 1988;59(1):34-8.

Muthukumaran N, Ma S, Reddi A. Dose-dependence of and threshold for optimal bone induction by collagenous bone ma¬trix and osteogenin-enriched fraction. Coll Relat Res 1988;8(5):433-41.

Nakahara H, Takaoka K, Koezuka M, Sugamoto K, Tsuda T, Ono K. Periosteal bone formation elicited by partially purified bone morphogenetic protein. Clin Orthop 1989;(239):299-305.

Narang R, Lloyd W, Wells H. Grafts of decalcified allogeneic bone matrix promote the healing of fibular fracture gaps in rats. Clin Orthop 1971;(80):174-80.

Nathan RM, Bentz H, Armstrong RM, Piez KA, Smestad TL, Ellingsworth LR et al. Osteogenesis in rats with an inductive bo¬vine composite. J Orthop Res 1988;6(3):324-34.

Nilsen R. Electron microscopy of induced heterotopic bone formation in guinea pigs. Arch Oral Biol 1977;22:485-93.

Nilsson OS, Bauer HC, Brosjö O, Törnkvist H. Influence of in¬domethacin on induced heterotopic bone formation in rats. Im¬por¬tance of length of treatment and of age. Clin Orthop 1986;(207):239-45.

Oikarinen J, Korhonen LK. Repair of bone defects by bone induc¬tive material. Acta Orthop Scand 1979;50(1):21-6.

Olney BW, Asher MA. Excision of symptomatic coalition of the middle facet of the talocalcaneal joint. J Bone Joint Surg 1987;69-A(4):539-44.

Osepian IA, Vardevanian GG, A:ivazian VP. :Treatment of post-osteomyelitic circular defects in shin bones by the method of com¬pression-distraction osteosynthesis with the use of allo-genic bone matrix. Ortop Travmatol Protez 1989;(3):21-3.

Ousterhout DK. Clinical experience in cranial and facial recon¬struction with demineralized bone. Ann Plast Surg 1985;15:367-73.

Ozkaynak E, Rueger D, Drier E, Corbett C, Ridge R, Sampath T et al. OP-1 cDNA encodes an osteogenic protein in the TGF-beta family. EMBO J 1990;9(7):2085-93.

Parker R. Aural pyaemia successfully treated by removing putrid thrombu of jugular vein and lateral sinus. Br Med J 1892;1:1076-7.

Pfeilschifter J, Bonewald L, Mundy GR. Characterization of the latent transforming growth factor beta complex in bone. J Bone Miner Res 1990;5(1):49-58.

Pinholt EM, Bang G, Haanaes HR. Alveolar ridge augmentation by osteoinduction in rats. Scand J Dent Res 1990;98(5):434-41.

Pitt CG, Jeffcoat AR, Zweidinger RA, Schindler A. Sustained drug delivery systems . I. The permiability of poly(e-caprolactone), poly(DL-lactic acid), and their copolymers. J Biomed Mater Res 1979;13:497-507.

Ragni P, Lindholm TS, Lindholm TC. Vertebral fusion dynamics in the thoracic and lumbar spine induced by allogenic demin¬er¬al¬ized bone matrix combined with autogenous bone marrow. An ex¬perimental study in rabbits. Ital J Orthop Traumatol 1987;13(2):241-51.

Ray RD, Holloway JA. Bone implants. Preliminary report of an experimental study. J Bone Joint Surg 1957;39-A:1119-28.

Reddi AH. Age-dependent decline in extracellular matrix-induced local bone differentiation. Isr J Med Sci 1985;21(3):312-3.

Reddi AH, Cunningham NS. Recent progress in bone induction by osteogenin and bone morphogenetic proteins: challenges for biomechanical and tissue engineering. J Biomech Eng 1991;113(2):189-90.

Reddi AH, Huggins C. Biochemical sequences in the transforma¬tion of normal fibroblasts in adolescent rats. Proc Natl Acad Sci U S A 1972 Jun;69(6):1601-5.

Reddi AH, Muthukumaran N, Ma S, Carrington JL, Luyten FP, Paralkar VM et al. Initiation of bone development by osteogenin and promotion by growth factors. Connect Tissue Res 1989;20(1-4):303-12.

Reddi AH, Sullivan NE. Matrix-induced endochondral bone differ¬entiation: influence of hypophysectomy, growth hormone, and thyroid-stimulating hormone. Endocrinology 1980;107(5):1291-9.

Ripamonti U. Bone induction in nonhuman primates. An experi¬mental study on the baboon. Clin Orthop 1991;(269):284-94.

Ripamonti U, Schnitzler CM, Cleaton-Jones PC. Bone induction in a composite allogeneic bone/alloplastic implant. J Oral Maxillo¬fac Surg 1989;47(9):963-9.

Ritter MA, Gioe TJ. The effect of indomethacin on para-articular ectopic ossification following total hip arthroplasty. Clin Orthop 1982;(167):113-7.

Robicsek F, Daugherty HK, Cook JW, Selle JG, Masters TN, O’Bar PR et al. Mycobacterium fortuitum epidemics after open-heart surgery. Thorac Cardiovasc Surg 1978;1:91-6.

Robicsek F, Masters TN, Littman L, Born GV. The embolization of bone wax from sternotomy incisions. Ann Thorac Surg 1981;31(4):357-9.

Rokkanen P, Böstman O, Vainionpää S, Vihtonen K, Törmälä P, Laiho J, Kilpikari J, Tamminmäki M. Biodegradable implants in fracture fixation: early results of treatment of fractures of the ankle. Lancet 1985;1(8443):1422-4

Rosen HB, Chang HB, Wnek GE, Linhardt RJ, Langer R. Bioerod¬ible polyanhydrides for controlled drug delivery. Biomaterials 1983;4:131-33.

Rø J, Sudmann E, Marton PF. Effect of indomethacin on fracture healing in rats. Acta Orthop Scand 1976;47:588-99.

Rønningen H, Solheim LF, Langeland N. Bone formation en¬hanced by induction. Bone growth in titanium implants in rats. Acta Orthop Scand 1985;56:67-71.

Sampath TK, Muthukumaran N, Reddi AH. Isolation of osteogen¬in, an extracellular matrix-associated, bone-inductive protein, by heparin affinity chromatography. Proc Natl Acad Sci U S A 1987;84(20):7109-13.

Sampath TK, Reddi AH. Importance of geometry of the extracellu¬lar matrix in endochondral bone differentiation. J Cell Biol 1984;98(6):2192-7.

Sato K, Urist MR. Induced regeneration of calvaria by bone mor¬phogenetic protein (BMP) in dogs. Clin Orthop 1985;(197):301-11.

Schmitz JP, Hollinger JO. A preliminary study of the osteogenic potential of a biodegradable alloplastic-osteoinductive alloim¬plant. Clin Orthop 1988;(237):245-55.

Schmitz JP, Hollinger JO. The critical size defect as an experimen¬tal model for craniomandibulofacial nonunions. Clin Orthop 1986;(205):299-308.

Schwartz Z, Sela J, Ramirez V, Amir D, Boyan BD. Changes in extracellular matrix vesicles during healing of rat tibial bone: a morphometric and biochemical study. Bone 1989;10(1):53-60.

Schwarz N, Dinges HP, Schiesser A, Redl H, Schlag G. Dog bone less osteogenetic than rat bone. Bone-matrix transplants in nude rats. Acta Orthop Scand 1989;60(6):693-5.

Schwarz N, Redl H, Schiesser A, Schlag G, Thurnher M, Lintner F et al. Irradiation-sterilization of rat bone matrix gelatin. Acta Orthop Scand 1988;59(2):165-7.

Schwarz N, Redl H, Schlag G, Schiesser A, Lintner F, Dinges H et al. The influence of fibrin sealant on demineralized bone ma¬trix-dependent osteoinduction. A quantitative and qualitative study in rats. Clin Orthop 1989;(238):282-7.

Sela J, Applebaum J, Uretzky G. Osteogenesis induced by bone matrix is inhibited by inflammation. Biomater Med Devices Ar¬tif Organs 1986;14(3-4):227-37.

Sendelbeck SL, Girdis CL. Disposition of a 14C-labeled bioerod¬ible polyorthoester and its hydrolysis products, 4-hydroxybutu¬rate and cis,trans-1,4-bis(hydroxymethyl)cyclohexane, in rats. Drug Metab Dispos 1985;13:291-5.

Senn N. On the healing of aseptic bone cavities by implantation of aseptic decalcified bone. Am J Med Sci 1889;18:219-43.

Seyedin SM, Thomas TC, Thompson AY, Rosen DM, Piez KA. Purification and characterization of two cartilage-inducing fac¬tors from bovine demineralized bone. Proc Natl Acad Sci U S A 1985;82(8):2267-71.

Sharrard WJ W, Collins DH. The fate of human decalcified bone grafts. Proc Roy Soc Med 1961;54:1101-2.

Sidman KR, Schwope AD, Steber WD, Rudolph SE, Poulin SB. Biodegradable, implantable sustained release systems based on glu¬tamic acid copolymers. J Memb Sci 1980;7:277-291.

Silverstein ME, Keown K, Owen JA, Chvapil M. Collagen fibers as a fleece hemostatic agent. J Trauma 1980;20(8):688-94.

Simmons DJ. Comparative physiology of bone, in Bourne GH (ed.) The biochemistry and physiology of bone. Volume IV. Calcifi¬ca¬tion and physiology. New York: Academic Press 1976, pp 445-515.

Sodemann B, Persson P-E, Nilsson OS. Prevention of heterotopic ossification by nonsteroid antiinflammatory drugs after total hip ar¬throplasty. Clin Orthop 1988;(237):158-63.

Solheim E, Anfinsen OG, Holmsen H, Sudmann E. Effect of local hemostatics on platelet aggregation. Eur Surg Res 1991;23(1):45-50.

Solheim LF, Rönningen H, Langeland N. Effects of acetylsalicylic acid on heterotopic bone resorption and formation in rats. Arch Orthop Trauma Surg 1986;105(3):142-5.

Strause L, Saltman P, Glowacki J. The effect of deficiencies of manganese and copper on osteoinduction and on resorption of bone particles in rats. Calcif Tissue Int 1987;41(3):145-50.

Sudmann B, Anfinsen O-G, Rait M, Bang G, Sudmann E. Use of a new hemostatic, bioerodible polymer versus bone wax made of beeswax – a clinical and experimental study. Acta Orthop Scand 1990;61(Suppl 237):63-4.

Sudmann E. Effect of indomethacin on bone remodelling in rabbit ear chambers. Acta Orthop Scand Suppl 1975;(160):91-115

Sudmann E, Bang G. Indomethacin-induced inhibition of haversian remodelling in rabbits. Acta Orthop Scand 1979;50(6 Pt 1):621-7.

Sudmann E, Dregelid E, Bessesen A, Mørland J. Inhibition of frac¬ture healing by indomethacin in rats. Eur J Clin Invest 1979;9(5):333-9.

Sudmann E, Husby OS, Bang G. Inhibition of partial closure of epiphyseal plate in rabbits by indomethacin. Acta Orthop Scand 1982;53(4):507-11.

Syftestad G, Urist MR. Degradation of bone matrix morphogenetic activity by pulverization. Clin Orthop 1979;(141):281-5.

Syftestad GT, Urist MR. Bone aging. Clin Orthop 1982;(162):288-97.

Takagi K, Urist M. The reaction of the dura to bone morphogenetic protein (BMP) in repair of skull defects. Ann Surg 1982;196(1):100-9.

Takagi K, Urist MR. The role of bone marrow in bone morphoge¬netic protein-induced repair of femoral massive diaphyseal de¬fects. Clin Orthop 1982;(171):224-31.

Takaoka K, Nakahara H, Yoshikawa H, Masuhara K, Tsuda T, Ono K. Ectopic bone induction on and in porous hydroxyapatite combined with collagen and bone morphogenetic protein. Clin Orthop 1988;(234):250-4.

Tassler H, Imhoff M. Gentamicin-haltiges Kollagenvlies. Ein erfahrungsbericht. Fortschr Med 1988;106:43-7.

Törnkvist H, Bauer FC H, Nilsson OS. Influence of indomethacin on experimental bone metabolism in rats. Clin Orthop 1985;(193):264-70.

Törnkvist H, Nilsson OS, Bauer FC, Lindholm TS. Experimentally induced heterotopic ossification in rats influenced by anti-in¬flam¬matory drugs. Scand J Rheumatol 1983;12(2):177-80.

Turner RT, Francis R, Brown D, Garand J, Hannon KS, Bell NH. The effects of fluoride on bone and implant histomorphometry in growing rats. J Bone Miner Res 1989;4(4):477-84.

Uretzky G, Appelbaum J, Sela J. Inhibition of the inductive activ¬ity of demineralized bone matrix by different percutaneous im¬plants. Biomaterials 1988;9(2):195-7.

Urist MR. Bone: formation by autoinduction. Science 1965;150(698):893-9.

Urist MR, Dawson E. Intertransverse process fusion with the aid of chemosterilized autolyzed antigen-extracted allogeneic (AAA) bone. Clin Orthop 1981;(154):97-113.

Urist MR, Lietze A, Dawson E. b-tricalcium phosphate delivery system for bone morphogenetic protein. Clin Orthop 1984;(187):277-80.

Urist MR, Mikulski A, Boyd SD. A chemosterilized antigen-ex¬tracted autodigested alloimplant for bone banks. Arch Surg 1975;110(4):416-28. Eng.

Urist MR, Mikulski A, Lietze A. Solubilized and insolubilized bone morphogenetic protein. Proc Natl Acad Sci USA 1979;76:1828-32.

Urist MR, Mikulski AJ. A soluble bone morphogenetic protein ex¬tracted from bone matrix with a mixed aqueous and nonaqueous solvent. Proc Soc Exp Biol Med 1979;162(1):48-53.

Urist MR, Nilsson O, Rasmussen J, Hirota W, Lovell T, Schmalzreid T et al. Bone regeneration under the influence of a bone morphogenetic protein (BMP) beta tricalcium phosphate (TCP) composite in skull trephine defects in dogs. Clin Orthop 1987;(214):295-304.

Urist MR. Surface-decalcified allogeneic bone (SDAB) implants. A preliminary report of 10 cases and 25 comparable operations with undecalcified lyophilized bone implants. Clin Orthop 1968;(56):37-50.

Vandersteenhoven JJ, DeLustro FA, Bell NH, Turner RT. Osteoin¬duction by implants of demineralized allogeneic bone matrix is di¬minished in vitamin D-deficient rats. Calcif Tissue Int 1988;42(1):39-45.

Vandersteenhoven JJ, Spector M. Osteoinduction within porous polysulfone implants at extraosseous sites using demineralized al¬logeneic bone matrix. J Biomed Mater Res 1983;17(5):793-806.

Wang EA, Rosen V, D’Alessandro JS, Bauduy M, Cordes P, Harada T et al. Recombinant human bone morphogenetic pro¬tein induces bone formation. Proc Natl Acad Sci U S A 1990;87(6):2220-4.

Wozney JM, Rosen V, Celeste AJ, Mitsock LM, Whitters MJ, Kriz RW et al. Novel regulators of bone formation: molecular clones and activities. Science 1988;242(4885):1528-34.

Yamazaki Y, Oida S, Akimoto Y, Shioda S. Response of the mouse femoral muscle to an implant of a composite of bone mor¬phogenetic protein and plaster of Paris. Clin Orthop 1988;(234):240-9.