J Oral Maxillofac Surg60:1170-1175, 2002

Induced Osteogenesis by PeriostealDistraction

Brian L. Schmidt, DDS, MD, PhD,* Laski Kung,†

Christopher Jones,‡ and Nardi Casap, DMD, MD§

Purpose: The purpose of this project was to evaluate a novel technique for inducing osteogenesisthrough periosteal distraction in a rabbit model.

Materials and Methods: A periosteal distraction device was rigidly fixed to the lateral surface of themandible in 10 adult rabbits. Periosteal distraction was started 7 days after placement of the periostealdistraction device. The periosteum was distracted 7 mm over 15 days. The unoperated, contralateral sideof the mandible served as the control. The animals were killed at postoperative days 28, 35, 42, and 56.The specimens were then fixed, decalcified, and stained with hematoxylin and eosin. Histologicexamination and histomorphometric analysis were performed on all specimens.

Results: Nine of 10 periosteal distraction devices remained rigidly fixed to the lateral surface of themandible. On postoperative day 28, the histologic specimen from the experimental side showedperiosteal proliferation and an increase in the number of osteoblasts. On postoperative days 35, 42, and56, the experimental side showed an increase in the number of osteocytes per unit area, collagen fibersparallel to the vector of distraction, islands of osteoblasts surrounded by newly formed bone, andmaturation of bone. An average of 2.86 � 0.56 mm of new bone height was formed.

Conclusion: We report on a novel technique for generating bone by periosteal distraction. Ourhistologic analysis showed proliferation of the periosteum, an increase in the number of osteoblasts andosteogenesis.© 2002 American Association of Oral and Maxillofacial SurgeonsJ Oral Maxillofac Surg 60:1170-1175, 2002

Reconstructive management of the atrophic, edentu-lous mandible and maxilla continues to pose a clinical

challenge for the oral and maxillofacial surgeon de-spite the description of numerous reconstructivetechniques. Endosseous implant placement in theatrophic alveolar ridge is often limited by the lack ofalveolar bone height and width. Reconstructive alter-natives for the atrophic jaw include autogenous bonegrafting and alveolar distraction osteogenesis. Autog-enous bone grafting requires a donor site and is oftencomplicated by resorption. Alveolar distraction osteo-genesis requires a corticotomy that is difficult to per-form in the thin alveolus and reduces available alveo-lar bone. In addition, although alveolar distraction hasbeen shown to increase alveolar bone height,1 thenative atrophic alveolar ridge is often knife-edged andthe final contour of newly generated bone is limitedby the width of the osteotomized segment.Distraction osteogenesis in the oral and maxillofa-

cial region is an adaptation of orthopedic distractionosteogenesis principles, and the mechanisms of osteo-genesis are the same in the facial bones as they are inthe long bones.2-5 One mechanism of osteogenesisafter distraction involves the periosteum, which iscomposed of cells that are capable of differentiatinginto osteoblasts.6-8 Distraction osteogenesis is success-ful because under appropriate levels of stimulation

*Assistant Clinical Professor, Department of Oral and Maxillofa-

cial Surgery, University of California, San Francisco, San Francisco,

CA.

†Dental Student, University of California, San Francisco, San

Francisco, CA.

‡Medical Student, New York University, New York, NY.

§Lecturer, Department of Oral and Maxillofacial Surgery, School

of Dental Medicine, Hebrew University-Hadassah Medical Center,

Jerusalem, Israel.

Presented at the 82nd Annual Meeting of the American Associ-

ation of Oral and Maxillofacial Surgeons, San Francisco, CA, Sep-

tember 20-23, 2000.

This work was supported by a grant from the Oral and Maxillo-

facial Surgery Foundation and by donated surgical devices from AO

North America Foundation-Synthes.

Address correspondence and reprint requests to Dr Schmidt:

Department of Oral and Maxillofacial Surgery, C-522, University of

California, San Francisco, San Francisco, CA 94143-0440; e-mail:

briansc@itsa.ucsf.edu

© 2002 American Association of Oral and Maxillofacial Surgeons

0278-2391/02/6010-0012$35.00/0

doi:10.1053/joms.2002.34993

1170

periosteal mesenchymal stem cells differentiate intoosteoblasts and produce early subperiosteal calluswithin the osteotomized gap.9-11 The subperiostealcallus matures to form the peripheral part of thenewly generated bone.12 Although the application ofdistraction forces during distraction osteogenesisleads to subperiosteal bone, tension on the perios-teum alone is sufficient to produce bone.13 In the rat,tenting of the periosteum on the lateral ramus afterplacement of a subperiosteal capsule is sufficient toproduce significant amounts of subperiosteal bone.13

The generation of bone in situ through a mechanismof periosteal distraction without a corticotomy is the-oretically ideal for producing small amounts of bonein the oral and maxillofacial region. The purpose ofthis study was to determine whether controlled dis-traction of the periosteum alone, without corti-cotomy, could be used to produce bone in the max-illofacial region.

Materials and Methods

RABBIT MODEL AND SURGICAL PLACEMENT OF THEDISTRACTION DEVICE

Ten adult New Zealand White male rabbits (CharlesRiver Laboratories, Wilmington, MA) with a meanweight of 3.9 � 0.39 kg were used as the animalmodel. Experimental protocols were approved by theUniversity of California, San Francisco Committee on

Animal Research and conformed to National Institutesof Health guidelines for use of animals in research.The periosteal distraction device was a custom-

made device consisting of medical grade titanium(Synthes Maxillofacial, Paoli, PA) (Fig 1). The deviceconsisted of a U-shaped body with 2 legs that could berigidly fixed to the lateral aspect of the mandiblularramus using 1.3-mm screws. Threaded into the centerof the device was a center screw to which a perfo-rated titanium mesh could be snapped into place afterfixation of the device. Rotation of the center screwresulted in distraction of the titanium mesh and peri-osteum away from the bone surface.The rabbits were first sedated with a combination

of intramuscular midazolam (2 mg/kg) and ketamine(40 mg/kg). The animals were intubated, and generalanesthesia was maintained with isoflurane. The heartrate, respiratory rate, pulse oximetry, and tempera-ture of the animals were monitored. Local anesthesiaconsisting of lidocaine 0.5% with 1:200,000 epineph-rine was infiltrated into the lateral ramus and subman-dibular regions. The rabbits received a single dose ofenrofloxacin (5 mg/kg subcutaneous) before the op-eration. After preparation and draping of the animals,the lateral aspect of the ramus was exposed througha submandibular incision. The skin, subcutaneous tis-sue, muscle, and periosteum were reflected. Three5-mm linear incisions were made through the skin,muscle, and periosteum opposite the mandible with aNo. 11 scalpel blade. The 2 legs and center screw ofthe device were then placed through the tissue (Fig2). The legs were rigidly fixed with two 1.3-mmscrews, and the titanium mesh was snapped onto thecenter screw. The wound was closed in layers. Post-operative analgesics included ketorolac (0.5 mg/kg bymouth) and buprenorphine (0.3 mg intramuscular). Aperiosteal distraction device was placed on one sideof the mandible on each animal and the contralateral,

FIGURE 1. The custom-made periosteal distraction device. Activationof the center screw (white arrow) distracted the subperiosteal titaniummesh (black arrow) from the lateral ramus of the mandible.

FIGURE 2. Intraoperative photograph showing submandibular dis-section, rigid fixation of the periosteal distraction device to the lateralaspect of the ramus (black arrow), and subperiosteal placement of thetitanium mesh (white arrow). The titanium mesh was lowered to thebone (0 mm of distraction) before closure of the wound.

SCHMIDT ET AL 1171

unoperated side of the mandible served as the con-trol. Animals were weighed daily.

PERIOSTEAL DISTRACTION SCHEDULE

The same distraction schedule was performed in allanimals. The latency period was 7 days. On day 7 thecenter screw was turned so that the titanium meshwas distracted 2 mm from the surface of the mandi-ble. The mesh was further distracted 1 mm every 3days for the next 15 days. According to this schedule,by postoperative day 22 the periosteum would bedistracted a total of 7 mm.

ASSESSMENT OF HARD AND SOFT TISSUES AFTERPERIOSTEAL DISTRACTION

The animals were then killed at 28, 35, 42, and 56days after placement of the distraction device. Therewere 3 animals in each of the 28- and 56-day groupsand 2 animals in each of the 35- and 42-day groups.

The rabbits were first sedated with a combination ofintramuscular midazolam (2 mg/kg) and ketamine (40mg/kg) and then killed with an intracardiac injectionof pentobarbital. The lateral mandiblular tissues in-cluding skin, muscle, periosteum, and bone with a1-cm margin around the device were removed andfixed for 10 days in 10% neutral buffered formalin. Anequal amount of tissue including skin, muscle, peri-osteum, and bone from the same lateral mandibularlocation on the contralateral side was removed. A silksuture was placed along the superior margin for ori-entation. The sections were decalcified with a so-dium-formate-formic acid solution. After decalcifica-tion, the device and titanium mesh were thencarefully removed. Sections were cut and stained withhematoxylin and eosin. The sections were examined,and the hard and soft tissues from the experimentaland control sides were compared. A pathologist whowas not part of the research team independentlyperformed the histomorphometric analysis. A 10 � 10ocular graticule outlining an area of 0.064 mm2 wasused at a magnification of �400 to assess sections.Histologic evidence of new bone consisted of a scal-loped appearance of the haversian system and anincreased amount of osteoid relative to the underlyingmature cortical bone. The distracted and control spec-imens from the same animal were serially examinedand compared to evaluate for new bone formation.The site of maximum new bone height within thedistracted specimen was measured.

Results

TECHNICAL CONSIDERATIONS OF PERIOSTEALDISTRACTION IN THE RABBIT MODEL

All animals survived placement of the periostealdistraction device. Nine of the 10 devices remainedrigidly fixed to the lateral ramus for the entire exper-

FIGURE 3. Photograph of postoperative day 28 animal showingcomplete healing around the device on the lateral aspect of themandible. For orientation, the nose of the rabbit is to the right, and theears are to the upper left. The center screw has been distracted the full7 mm.

FIGURE 4. A, Photomicrograph of control side from postoperative day 28 animal showing mature cortical bone, fibrous connective tissue on themedial and lateral aspect of the mandible representing periosteum (arrow), and muscle. B, Photomicrograph of distracted side from postoperativeday 28 animal showing periosteal proliferation (arrow), immature osteoblasts, and newly formed bone.

1172 OSTEOGENESIS BY PERIOSTEAL DISTRACTION

imental period. One device became loose 24 hoursafter surgery; this device was removed, and the ani-mal was excluded from the study. All animals re-sumed normal dietary habits during the first 24 hoursafter the operation, and none of the animals had aweight loss during the experimental study. In all ani-mals the titanium mesh was dislodged from the centerscrew after the distraction of approximately 4 to 5mm from the lateral ramus; therefore, the plannedperiosteal distraction height of 7 mm was notachieved. However, the distraction schedule wascompleted according to the protocol described here.At the completion of distraction, all operative siteshealed without evidence of infection (Fig 3).

HISTOLOGIC ANALYSIS OF PERIOSTEALDISTRACTION SITES

Postoperative Day 28Comparison of the histologic specimens from the

control and distracted sides from the postoperative

day 28 animal showed an increase in periosteal pro-liferation on the distracted side (Figs 4A, B). Periostealdistraction also resulted in the formation of osteo-blasts, early bone formation, and an increase in thenumber of osteocytes per unit area.

Postoperative Day 35Specimens from the distracted side of postopera-

tive day 35 animals continued to show periostealproliferation, an increase in osteoblasts, and newbone formation (Figs 5A, B). By postoperative day 35,collagen fibers were oriented parallel to the vector ofdistraction and perpendicular to the margin of newlyformed bone.

Postoperative Day 42Periosteal proliferation with further maturation

of bone was observed in the postoperative day 42specimens (Figs 6A, B). The newly formed collagenfibers observed in the postoperative day 35 speci-

FIGURE 5. A, Photomicrograph of control side from postoperative day 35 animal showing cortical bone, a thin layer of periosteum (arrow), andmuscle. B, High-power photomicrograph of distracted side from postoperative day 35 animal showing periosteal proliferation and immatureosteoblasts at the margin of newly formed bone surrounded by collagen fibers (white arrow). The cleft (black arrow) represents the previous site ofthe titanium mesh. The collagen fibers are running parallel to the vector of distraction and perpendicular to the margin of newly formed bone.

FIGURE 6. A, Photomicrograph of control side from postoperative day 42 animal showing cortical bone, periosteum, and muscle. B, Low-powerphotomicrograph of distracted side from postoperative day 42 animal showing periosteal proliferation, maturation of the bone, and a significantincrease in the number of osteocytes per unit area relative to the control.

SCHMIDT ET AL 1173

mens were also observed in postoperative day 42specimens.

Postoperative Day 56In the postoperative day 56 specimens the collagen

fibers were adjacent to immature osteoblasts and ex-tended from the periosteum into the margin of newlyformed bone; such collagen fibers were not present inthe control, nondistracted specimens (Figs 7A, B).The newly formed bone was characterized by both anincrease in the number of osteocytes per unit area andan increase in trabeculation (Fig 7C).

HISTOMORPHOMETRIC ANALYSIS OF PERIOSTEALDISTRACTION SITES

Table 1 presents the increase in bone height ineach of the distracted specimens. The average heightof newly formed bone on the distracted sides was

2.86 � 0.56 mm. There was no evidence of new boneformation on the unoperated, control specimens.

Discussion

In this study we show that controlled distraction ofthe periosteum without corticotomy induces osteo-genesis in the rabbit model. The rabbit model waschosen because the size of the device was appropri-ate for the lateral aspect of the mandible. In addition,our pilot studies showed that the rabbit could toleratethe device and resume normal dietary habits within24 hours after placement of the device. A drawback ofthis animal model is that the osteogenic response ofthe rabbit is robust and may not accurately representthe response that would be observed in the human.Because it was not known what the effects of place-

FIGURE 7. A, Photomicrograph of control side from postoperativeday 56 animal. B, High-power photomicrograph of distracted sidefrom postoperative day 56 animal showing newly formed collagenfibers surrounding immature osteoblasts. C, Low-power view of spec-imen from distracted side on postoperative day 56 animal. The in-creased trabeculation representing newly formed bone is obviousadjacent to mature bone.

Table 1. THE INCREASE IN BONE HEIGHT AFTER PERIOSTEAL DISTRACTION BASED ON HISTOMORPHOMETRICANALYSIS

Postoperative Day

Overall (n � 9)28 (n � 3) 35 (n � 2) 42 (n � 2) 56 (n � 2)

Bone height increase (mm) 2.61 � 0.58 2.94 � 0.14 2.38 � 0.14 2.80 � 0.84 2.86 � 0.56

1174 OSTEOGENESIS BY PERIOSTEAL DISTRACTION

ment of the device for up to 56 days would be onfeeding and drinking behavior, only 1 device peranimal was placed. In future studies, the placement of2 periosteal distraction devices should be performedwith distraction of the periosteum on one side only toassess the effect of periosteal distraction on boneformation. Despite these limitations, our study didconfirm that controlled distraction of the periosteumis technically feasible. Technical problems with thedevice included loosening of 1 device and disengage-ment of the titanium mesh from the center screw onall devices. Further studies investigating periostealdistraction would require modification of the deviceto ensure that the tension of the distracted tissues(periosteum, muscle, and skin) does not lead to dis-lodgement of the titanium mesh.In this study, the histologic analysis showed that peri-

osteal distraction leads to the formation of collagen fi-bers oriented parallel to the vector of distraction. Colla-gen bundle formation after distraction osteogenesis hasbeen previously shown.14,15 Distraction osteogenesis inboth the long bones and the craniofacial skeleton leadsto the formation of such collagen fibers oriented longi-tudinally to the vector of the distraction force.14,15 Earlysubperiosteal osteogenesis and woven bone formationoccur along these collagen bundles.16 Our current find-ings suggest a possible functional role for the collagenfibers. Immature osteoblasts were surrounded by thelongitudinally oriented collagen fibers that extendedfrom the periosteum to bone. Tension on the perios-teum might induce differentiation of mesenchymal cellsto osteoblasts. Ultimately, these fibers could transmit thetensile forces to the immature osteoblasts and providethe mechanism to induce osteogenesis.Histomorphometric analysis showed that periosteal

distraction produced an increase of approximately2.9 mm of bone height in our model. In situ osteo-genesis with the periosteal distraction techniquewould avoid a corticotomy as required by distractionosteogenesis. In addition, bone fill within the gaposteotomy would not be required. By the earliest timepoint, postoperative day 28, there was clear evidenceof newly formed bone, which went on to mature bylater time points. The ultimate goal of periosteal dis-traction would be adaptation of the technique for usein the oral cavity. The generation of a few millimetersof alveolar bone height and width would allow for

endosseous implant placement in the case of an atro-phic, edentulous jaw. Although the most obvious useof the technique would be reconstruction of the atro-phic, edentulous ridge, other possible uses might in-clude in situ generation of bone within an alveolardefect such as an alveolar cleft. Future investigationsinto the technique of periosteal distraction shouldinvolve a higher animal model and evaluation of thesurgical technique for intraoral use.

References1. Block MS, Chang A, Crawford C: Mandibular alveolar ridge

augmentation in the dog using distraction osteogenesis. J OralMaxillofac Surg 54:309, 1996

2. Alho A, Bang G, Karaharju E, et al: Filling of a bone defectduring experimental osteotaxis distraction. Acta Orthop Scand53:29, 1982

3. Ilizarov GA: The tension-stress effect on the genesis and growthof tissues: Part II. The influence of the rate and frequency ofdistraction. Clin Orthop 276:263, 1989

4. Ilizarov GA: The tension-stress effect on the genesis and growthof tissues: Part I. The influence of stability of fixation andsoft-tissue preservation. Clin Orthop 276:249, 1989

5. Karp NS, McCarthy JG, Schreiber JS, et al: Membranous bonelengthening: A serial histological study. Ann Plast Surg 29:2,1992

6. Bassett CAL, Herrmann I: Influence of oxygen concentrationand mechanical factors in differentiation of connective tissuesin vivo. Nature 190:460, 1961

7. Matthews JL: Bone Structure and Ultrastructure. Philadelphia,PA, JB Lippincott, 1980, pp 4-44

8. Simmons DJ: Fracture Healing. Philadelphia, PA, JB Lippincott,1980, pp 280-330

9. Ozerdem OR, Kivanc O, Tuncer I, et al: Callotasis in nonvas-cularized periosteal bone grafts and the role of periosteum: Anew contribution to the concept of distraction osteogenesis.Ann Plast Surg 41:148, 1998

10. Choi IH, Ahn JH, Chung CY, et al: Vascular proliferation andblood supply during distraction osteogenesis: A scanning elec-tron microscopic observation. J Orthop Res 18:698, 2000

11. Hikiji H, Takato T, Matsumoto S, et al: Experimental study ofreconstruction of the temporomandibular joint using a bonetransport technique. J Oral Maxillofac Surg 58:1270, 2000

12. Delloye C, Delefortrie G, Coutelier L, et al: Bone regenerateformation in cortical bone during distraction lengthening: Anexperimental study. Clin Orthop 775:34, 1990

13. Kostopoulos L, Karring T: Role of periosteum in the formationof jaw bone: An experiment in the rat. J Clin Periodontol22:247, 1995

14. Karp NS, Thorne CH, McCarthy JG, et al: Bone lengthening inthe craniofacial skeleton. Ann Plast Surg 24:231, 1990

15. Jazrawi LM, Majeska RJ, Klein ML, et al: Bone and cartilageformation in an experimental model of distraction osteogene-sis. J Orthop Trauma 12:111, 1998

16. Peltonen J: Bone formation and remodeling after symmetricand asymmetric physeal distraction. J Pediatr Orthop 9:191,1989

SCHMIDT ET AL 1175