- Email: [email protected]
A new approach to repairing cleft palate and acquired palatal defects with distraction osteogenesis
Int. J. Oral Maxillofac. Surg. 2006; 35: 718–726 doi:10.1016/j.ijom.2006.03.010, available online at http://www.sciencedirect.com
Leading Research Paper Congenital Craniofacial Deformities
A new approach to repairing cleft palate and acquired palatal defects with distraction § osteogenesis
D.-Z. Wang1, G. Chen1,2, Y.-M. Liao1, S.-G. Liu1, Z.-W. Gao1, J. Hu1, J.-H. Li1, C.-H. Liao1 1 Department Oral & Maxillofacial Surgery, West China College of Stomatology, Sichuan University, 14#, 3rd section, Renminnan Rd., Chengdu 610041, PR China; 2Department Oral & Maxillofacial Surgery, College of Stomatology, Tianjin Medical University, 12# Qixiangtai Rd, Heping district, Tianjin 300070, PR China
D.-Z.Wang, G. Chen, Y.-M. Liao, S.-G. Liu, Z.-W. Gao, J. Hu, J.-H. Li, C.-H. Liao: A new approach to repairing cleft palate and acquired palatal defects with distraction osteogenesis. Int. J. Oral Maxillofac. Surg. 2006; 35: 718–726. # 2006 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved. Abstract. Cleft palate (CP) is one of the most common human congenital deformities, and acquired palate defects after trauma or tumour resection are also common. In this study, distraction osteogenesis (DO) for CP and other palatal bone defects was evaluated. Twenty cats were assigned randomly to 3 groups of (1) 15, (2) 3 and (3) 2 cats. In groups 1 and 2, a rectangular ostectomy, in the posterior of the palatal bone shelf, was performed in the sagittal axis to establish the CP defect model. At the same time, a pure titanium intraoral distractor was fixed to molar teeth with brackets and to the palatal bone shelf across the defect with titanium miniscrews bilaterally. Four weeks later, a secondary transport disc (TD) osteotomy was performed, and gradual DO treatment started at 0.4 mm twice a day, after 6 days of latency. DO was performed until the TD reached the opposite margin over the gap in 5–6 days. Three cats each of group 1 were killed at 2, 4, 6, 8 and 12 weeks after completion of DO. In group 2, the bone and soft-tissue defects were untreated until death 6 weeks later. Group 3 cats (control) were killed after 6 weeks. The TD successfully recombined with the opposite palatal bone stump, and proportional expansion of the overlay mucoperiosteal flap was achieved. Intramembranous bone formation was revealed: parallel collagen bundles gradually deposited on new bone trabeculae while the proliferative osteoblasts produced bone matrix. The bone defect was finally reconstructed by de novo osteogenesis. The control group was observed to have no spontaneous repairing. These results suggest that the CP defect was reconstructed by osteogenesis in situ, and the soft tissues expanded simultaneously to achieve functional correction. The intraoral distractor provided both effective distraction and stability.
§
Key words: distraction osteogenesis; cleft lip/ palate; animal model; intraoral distractor. Accepted for publication 3 March 2006 Available online 9 May 2006
The preliminary report was presented at the 82nd general session and exhibition of the IADR, 10–13 March 2004, Honolulu, Hawaii, USA.
0901-5027/080718 + 09 $30.00/0
# 2006 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
Repairing CP and acquired palatal defects with DO Cleft palate (CP), with or without cleft lip, is one of the most common human congenital deformities. Chewing, swallowing, talking and breathing are severely affected, and the condition also causes some psychological problems. Acquired palate defects after trauma or tumour resection are also very common. Anatomical disfigurement is of both the soft tissues and bone shelf, especially in the posterior part of the hard palate and the soft palate. Traditional surgical repair of CP depends on transferring adjacent palatal mucoperiosteal flaps to close the gap, with or without lengthening the soft palate, and to narrow the pharyngeal cavity. It is still the case that surgeons cannot truly reconstruct the CP bone defect and correct the abnormal muscular attachment to physiologically normal positions, so called to ‘put normal to normal’. Also, scarring after traditional surgical procedures may gradually cause a series of secondary deformities, such as decreased maxillary growth and impeded arch expansion, as the patients grow up. This may have highly detrimental psychosocial ramifications. Distraction osteogenesis (DO) is a technique involving the application of an endosseous appliance combined with cortical osteotomies and the use of tensional distraction to lengthen existing bone and soft tissue. This technique has been widely used in the field of oral maxillofacial surgery in recent decades. Although there are certain difficulties, this technique is effective because, compared to long bones, the blood supply associated with the maxillary and palatal region is richer. Since the hard palatal bone defect of CP is more complex than at any other site, because of the thinner and structurally weaker bones, complicated by interrelationships with dental arches, and oral and nasal cavities, an individually manufactured intraoral distractor was designed to practice DO correction of CP deformities.
long incision from the canine cusp level and 1 cm into the soft palate was made, parallel to the central palatal fissure but 5 mm to the side, and then the mucoperiosteal flap was lifted to expose the bone shelf. Two sagittal incisions were made on the bone shelf 2 and 10 mm to the side of the central hard palatal fissure unilaterally, from the level of the anterior canine cusps to the posterior palatal mar-
719
gin. Another 8-mm incision at the level of the canine cusps frontally was made to connect the 2 sagittal incisions to form an 8 15 mm2 longitudinal rectangular ostectomy (Fig. 1). The overlaying soft tissues were cropped correspondingly. The margins of the oral mucoperiosteal flap and nasal mucosa were sutured together to ensure penetration into the oral nasal defect.
Materials and methods The CP animal model
Twenty cats (0.5–0.8 years old, 3.5–4.5 kg weight) were used in this study, assigned randomly into three groups of (1) 15 cats, (2) 3 cats and (3) 2 cats. The 18 animals in groups 1 and 2 were used to perform the CP experimental modelling procedures, with group 3 acting as control. Under general anaesthesia combined with local infiltration of lidocaine 2% and epinephrine 1:100,000, an 18-mm-
Fig. 1. The intraoral distractor and the CP animal model. (a) Schematic drawing of the intraoral CP distractor. A: immobile unit with jackscrew bolt; B: mobile unit; C: two parallel poles. (b) Oronasal penetration defect in CP experimental animal model with the pure titanium distractor mounted in position. B: molar bracket; S: pure titanium miniscrew; TDv: transporting parts; TD: incision line of transport disc. (c) View of the CP animal model in group 2 (6 weeks) shows the hard palate bone defect unrepaired. TD: incision line of transport disc.
720
Wang et al.
Design of intraoral distractor
The pure titanium intraoral distraction appliance (Chinese patent ZL 01 2 06830.6) consists of an immobile unit with a jackscrew bolt (0.4 mm for each revolution) impelling a mobile unit to move in an axial direction along 2 parallel poles. Both of the units are fixed, by 2 endosseous pure titanium miniscrews each, into the palatal bone shelf across the CP bone defect gap. Another molar bracket is made by plaster cast and then welded to the distractor to supply perpendicular anchorage to the titanium miniscrews (Fig. 2). DO protocol
As soon as the surgical procedures of CP modelling were completed, the pre-made intraoral distractor welded with molar brackets was tested and then mounted. Prophylactic i.m. antibiotic was administered at the time of surgery and for 5 postoperative days. A secondary osteotomy for transport disc (TD) formation was performed 4 weeks later so that early bone integrity of the titanium screw to the palatal shelf could be achieved. During the operation of TD formation, bone cutting was performed 2 mm adjacent to the alveolar ridge on the CP side to insure inclusion of the major palatal artery in the segment, so that the TD would be transported with an adequate blood supply. The CP animals were fed with a fluid diet for 5 days postoperatively, and most of them returned to normal food in 10 days. After 6 days of latency, the jackscrew shaft was activated twice a day (1 revolution each time, to a total distance of 0.8 mm) to impel the intraoral distractor with the TD segment. Once the TD had reached the opposite bone edge over the CP defect gap, the distractor was left in place for an additional 6 weeks before removal. Specimen retrieval and preparation
The 15 CP model animals assigned to group 1 underwent the DO procedure. Three cats each of this group were killed with excessive pentobarbital at 2, 4, 6, 8 and 12 weeks after completion of DO. The 3 other CP animals (group 2) and the 2 control animals (group 3) were killed after six weeks. Tetrachloride fluorescent labeling procedures (30 mg/kg) were administered 6 days before death. After gross observations and photographs were taken, X-ray examinations were made in situ. The specimens were
Fig. 2. Gross view after 4 weeks of DO correction on soft tissue before (a) and after (b) removal of the distractor, and view of the bone shelf (c). Only thin trace of original defect (arrow). B: molar bracket; S: pure titanium miniscrew; TDv: transporting parts; M: molar on normal side; M0 : molar on distraction side; T: hole of removal of titanium screw; P: posterior margin of hard palate; A and B: sutures; O: palatal foramen on normal side; O0 : palatal foramen on distraction side; S-S0 : anteroposterior borders of new bone formation; C-C0 : bilateral borders of new bone formation; NB: DO newly formed bone; TD: transport disc.
Repairing CP and acquired palatal defects with DO
721
retrieved en bloc, including the palatal region 5 mm either side of the distraction gap, and then were divided into 2 parts for microscopic histological and scanning electron microscopic (SEM) ultrastructural study. For histological study, the specimens were immersed in 10% neutral buffered formalin solution, after being rinsed with saline, for 7 days of fixation. After being decalcified in sodium formate-formic acid, the specimens were embedded in paraffin. Frontal 6-mm sections were cut with a microtome, including the whole distraction gap and bilateral original bone, and then stained with haematoxylin & eosin and modified Masson trichrome. For ultrastructural study, the specimens were trimmed and rinsed with saline and 0.2 mol/l phosphate-buffered saline, and then immersed in 2.5% glutaraldehyde at 4 8C for 24 h fixation. Frontal splitting was performed with the intermediate agent of liquid nitrogen, followed by dehydration in increasing concentrations of acetone. Every specimen was manipulated by vacuum-sputtering of gold membrane to insure electroconductibility and analysed with a SEM (AMRAY, 1000B, USA). Results Gross observations
All animals tolerated the implantation of appliances without infection or impairment in respiratory or visual functions. After the surgical procedure of CP modelling, the animals manifested dramatic symptoms as in clinical human cases, including excessive nasal secretion, sneezing, nasal flowing of food crumbs and water, and so on. Intraorally, an obvious sagittal rectangular oral–nasal penetration defect was observed. At 12 weeks after the DO of group 1, the original defect had disappeared (Fig. 3), and measurement showed that the relationship between the maxilla and mandible was normal, i.e. as before DO. In group 2 (without DO), after 6 weeks, the defects in the hard palate were observed to be unhealed but a little narrower than the original size. At the same time, a mild cusp-to-cusp bite of molar teeth was observed, implying a slight collapse of the bilateral palatal bone into the defect area, because of the discontinuity of the hard bone shelf support. During the whole distraction period, no dehiscence of the incision was seen but the palatal plicate was observed to be gradually flattening. Interestingly, inosculation between the overlying soft tissue of the TD and the opposite CP defect bone stump was also observed.
Fig. 3. Gross view after 12 weeks of DO on soft tissue (upper) and bone shelf (lower). The original defect has disappeared (arrow). M0 : molar on the distraction side; NB: DO newly formed bone; P: posterior margin of hard palate; TD: transport disc; T: hole of removal of titanium screw; T0 : cured screw holes on TD.
The distraction gap had filled with tough fibrous tissues in continuity with the medial and lateral palatal bone margins after 2 weeks. The oral and nasal cavities were separated again. There was only a central fibrous zone left after around 4–6 weeks (Fig. 4). After 8 weeks, the primary defect was corrected by TD transportation across the distraction gap and bone tissue regeneration had occurred. Hard palatal bone shelf continuity was recovered. The overlying soft tissue looked much like that in group 3 (control), without signs of alteration. There was in particular no necrosis or scar formation.
Roentgenographic findings
Although the hard palatal bone is thin and structurally weak, a fine-scale contrast gradient of anatomical structures could be identified in groups 2 and 3. The half-elliptical dental arch, middle palate fissure with nasal septum, posterior margin of the hard palate, pterygoid plates and major palatal foramen were important landmarks, with additionally the CP defect area (Fig. 5a). In group 1, the radiograph made after 2 weeks showed a rectangular radiolucent area between the cut bone edges of the secondary osteotomy for TD formation,
722
Wang et al.
Fig. 4. Roentgenographic view of hard palate in group 2 (a, arrow indicating CP bone defect area), and 2 weeks (b, distractor in place), 6 weeks (c, arrow indicating new bone formation in distraction gap) after DO correction (group 1). TD: transport disc; NB: newly formed bone.
indicating the distraction gap filled with collagenous fibres (Fig. 5b). For this reason, the distractors were removed after 6 weeks; the landmark of the major palatal foramen and the miniscrew holes were recognizable in radiographs made after more than 6 weeks. Obviously, the major palatal foramen on the DO side was medially closer to the middle palatal fissure since it was within the TD and being transported medially (Fig. 5c). The radiographs made at different time intervals demonstrate a tendency of the radiodensity of the distraction gap to increase, extending gradually from the bilateral bone edges to the centre, until eventually the gap became indistinguishable from the adjacent bone after 12 weeks. Histological and fluorescent labeling evaluation
Group 3
In the normal palatal mucoperiosteal soft tissue, multi-layered structures characterized by long keratose epithelial tags into subcutaneous connective tissue, containing perpendicularly interweaved collagen fibre networks, were observed. Hard palatal bone tissue consisted of well-calcified lamellar bone and mature Haversian system. No tetrachloride fluorescent labelling could be found. Group 1
Histological study of the group 1 specimens revealed that there was a uniform gradient of mineralization from the centre of the distraction gap bilaterally to original bone tissue. The characteristic histological view of newly formed tissue within the distraction gap was of a triple-zone structure from the centre to the original bone edges bilaterally (Fig. 5a). The three zones are: (1)
fibrous zone: plenty of collagen fibre bundles were arranged in parallel in the direction of the distraction vector, and fibroblast cells and mesenchymal precursor cells were scattered throughout the matrix; (2) new bone formation zone: newly formed bone trabeculae oriented in the direction of the distraction vector, and osteoblasts congregating on the surfaces of those trabeculae; (3) remodelling and mature zone: a honeycomb structure of bone trabeculae woven in different directions with abundant cellular components, and gradual transition to more mature bone through remodelling activities. After 2 weeks of retention, the collagenous fibres oriented along the distraction vector were predominant components. A few fine needle-like newly formed spiculae were viewed near the cut bone edges bilaterally pointing to the centre of the distraction zone. Fibroblasts and undifferentiated mesenchymal precursor cells were viewed in direct continuity with osteoblasts on the surfaces of early bone spicules (Fig. 5b). Fluorochrome labeling of tetrachloride was observed on the surface of the osteoid layer of new trabeculae in high density. In the 4- and 6-week subgroups, proliferative osteoblasts adjacent to osteoid on the surfaces of newly formed trabeculae were observed. Collagenous fibres deposited and calcified on thick trabeculae. These bone columns did not contain Haversian systems, but instead were surrounded by large, thin-walled blood vessels and loose connective tissues in medullary cavities. Near the original cut bone edges bilaterally, active bone absorption and remodelling were observed and osteoclasts could be identified (Fig. 5c). High-density tetrachloride fluorescent labelling on new trabeculae was distinct; meanwhile, discontinuity of the labelling
strips was observed, caused by immature woven bone near the cut bone edges (Fig. 5d). After 8 weeks, more mature and calcified bone with recognizable lamellar structure and newly formed Haversian systems could be seen (Fig. 5e). Bone formation and remodelling destruction were observed simultaneously. Fluorochrome labelling on bone trabeculae surfaces and bone lacunae was broad with discontinuity of the lining in common (Fig. 5f). After 12 weeks, the border between newly formed bone and original bone was still distinct but the structure of the new bone was more mature. There was no evidence of tetrachloride fluorescent labelling. Chondral tissues and cartilage were not found in any of the subgroups. As for the soft tissue extending over the distraction gap, no histological changes were found evident, but the epithelial tags became shorter, and the orientation of the collagenous fibre network changed slightly to that of the distraction vector. No inflammatory infiltration into soft tissue was observed (Fig. 5g). Group 2
There was no evidence of spontaneous repair in the original cut bone stumps of the CP defect gap, but only some appositional covering of amorphous structure. There was no tetrachloride fluorochrome labelling (Fig. 5h). Ultrastructural evaluations
Group 3
Under the SEM, normal bones demonstrated a well-calcified uniform structure, with mature spindle-shaped osteocytes located in small, flat bone lacunae.
Repairing CP and acquired palatal defects with DO
723
Fig. 5. Micrographs and fluorochrome labeling views. (a) Triple-zone structure within the distraction gap. OB, original bone; NB, newly formed bone; COL, collagenous fibrous tissue (H&E, 100). (b) At 2 weeks, collagenous fibers oriented in the distraction vector direction with proliferated osteoblasts on surfaces of early bone spicules (H&E, 400). (c) At 4 weeks, newly formed bone trabeculae oriented in the direction of distraction (H&E, 100). (d) Fluorochrome labeling at 4 weeks, strong tetrachloride strip on surfaces of parallel newly formed trabeculae (100). (e) Newly formed Haversian system at 8 weeks (Masson, 400). (f) At 8 weeks, discontinuity of fluorochrome strips on trabeculae surfaces indicating bone formation and remodeling simultaneously (100). (g) Micrograph of elongated soft tissue over new bone of distraction gap (H&E, 100). (h) In group 2, only thin layer of amorphous substance on cut bone edge without repair (arrow) (H&E, 100).
Group 1
SEM viewing of specimens after 2 weeks revealed many collagenous fibre bundles oriented parallel to the direction of the distraction vector. Abundant polymorphous cellular components including fibroblasts and undifferentiated mesenchymal precursor cells were scattered all over the matrix. Minimally calcified osteoid was on the surfaces of early bone spicules (Fig. 6a).
After 4–6 weeks, active osteoblast proliferation adjacent to osteoid accumulation on surfaces of newly formed trabeculae was observed. Collagenous fibre deposition calcified on thick trabeculae. Newly generated osteoblasts were gradually embedded in the bone matrix and became osteocytes, with relatively large, round lacunae compared to those in the original bone; in this regard, the newly formed bone was shown to be immature. Honeycomb-like newly formed woven bone was
observed near the original bone stumps (Fig. 6b). After 8–12 weeks, the bone structure was observed to be more mature with strong trabeculae, a more calcified matrix, a lamellar bone structure, and so on. Osteocytes in round bone lacunae looked more like what was found in the original bone. Newly formed Haversian systems were also found in some fields of vision. Nevertheless, the interface between original bone and new bone was still distinguishable (Fig. 6c).
724
Wang et al.
Fig. 6. SEM ultrastructural views. (a) At 2 weeks, many collagenous fiber bundles oriented parallel to the direction of distraction (1000). (b) Honeycomb-like structure of cellular new bone at 4 weeks (1000; OB, original bone; NB, new formed bone). (c) Newly forming Haversian system (arrow) in new bone tissue at 6 weeks (500; OB, original bone; NB, newly formed bone). (d) In group 2, no sign of repair of original cut bone stumps (600; OB, original bone).
Group 2
The original cut bone stumps showed an amorphous substance instead of osteogenesis occurring (Fig. 6d). Discussion
Using surgically established CP animal models, the method of hard palatal DO, with newly designed intraoral distractors, was shown to be effective for repairing CP bone and soft-tissue defects for the first time in this study. The TD segment and overlying soft tissue were transported over the defect area followed by de novo osteogenesis in the distraction gap, and consequently palatal integrity was successfully reconstructed by endogenous tissue engineering. This new method minimized surgical trauma to local structures and avoided scar formation, to which a series of secondary developing deformities have been primarily attributed. No reliable and repeatable drug induction method for a congenital CP model in large animal studies is available yet. This is a future target, but for now a CP experimental model must be established surgically. The CP experimental animal model used here simulated the clinical demonstration of CP effectively and responded positively to the DO procedure for bone regeneration. Although in all the dimensions investigated so far, ‘short-faced’
non-human primates have shown the most similar growth patterns to humans, several other non-primate animal models have been identified for modelling specific regions at various ages1–3. The results of modelling CP changes suggested that domestic cats might be a good alternative to non-human primate such as chimpanzees for the study. In addition, being a carnassial species with wide mouth opening, cats are a good model for repeating intraoral manipulations of distraction, and are less aggressive and dangerous than dogs. The cats used in this study were all juveniles aged 0.5–0.8 years old; the longest time interval followed after active distraction was 12 weeks, after which the cats would have reached maturity. It could therefore be observed that there was no influence of the DO technique on development of the oral and maxillofacial area3,4. In addition, comparative measurement of the jaws using experimental casts implied that the DO procedure might not retard the normal developmental tendency of the midface, and could maintain a normal occlusal relation in these cats. In group 2, the horizontal distance of the maxilla became narrower and presented a slight crossbite relation. Other investigators’ results have shown that defects in long bones equalling 1 to 2 times the size of the diaphyseal diameter routinely produce a permanent non-union
termed a ‘critical size defect’ (CSD); hence, only osteogenesis factors can consistently repair CSDs5–7. In this study, the size of the CP defect was designed to be far larger than that of the thin palatal bone shelf. The amount of healing that will occur in a bony defect is to a great extent dependent upon wound size; therefore, spontaneous healing was precluded. Interestingly, it was observed in this study that when the transport disc met the contralateral hard palatal defect margin at the end of active distraction, the thin fissure between the 2 parts disappeared automatically within 2 weeks. The mechanism of this healing process is unclear as yet. If this or a similar technique could be used clinically in human CP patients in the future, it would involve a simple secondary surgical trimming procedure to heal the fissure. In the present study, application of the DO principle was carried out according to a new design to suit the thin and weak palatal bone structure. The ultimate goals were: (1) regenerating new bone and overlying soft tissue and reconstructing continuity of the CP defect area and distraction gap; (2) rehabilitating osculation between the bone transport disc and opposite bone stump; (3) maintaining functional and structural stability. These are quite different aims to that of lengthening the posterior part of the hard palate and the soft palate8.
Repairing CP and acquired palatal defects with DO The results of the histological study revealed a significant intramembranous osteogenetic process in the distraction gap8–10,11–13,15. The new bone was less organized with larger vascular channels than the original bone nearby14. Proliferated osteoblasts were gradually embedded in the bone matrix and became osteocytes with relatively larger osteocytic lacunae than those in the pre-existing bone. In this regard, the newly formed bone was shown to be immature; active resorption of unfavourable structures and remodelling eventually made it more suitable for normal functional loads8,11,14,16–18. The cancellous structure of newly regenerated bone, consisting of bony trabeculae surrounding large spaces occupied by soft tissue and blood vessels, assures sufficient blood supply9. During the distraction phase, the blood supply to the distraction zone reached a maximum level of seven times the normal contralateral side, and could remain at approximately 3 times the normal level for the remodelling phase thereafter11,19,20. There was no chondral phase observed at any time interval with any of the specimens. The double perpendicular anchorages of the distractor supplied stable fixation, which was essential for adequate formation of microcolumns of bone during DO. It is obvious that bending or shear stress could induce fractures of the microcolumns with local haemorrhage and resultant histological cartilage interposition11,18. Most investigators attribute the presence of cartilage to imperfect fixation that has resulted in some motion between the distracted bone edges18. Fibroblasts and undifferentiated mesenchymal precursor cells were found in direct continuity with osteoblasts on the surfaces of early bone spicules. This implies that the bone-forming cells are created by transformation of these precursor cells throughout the matrix. In the ultrastructural study, SEM views corroborated the results of the histological evaluation, indicating that, although at a thin and structurally weak site, osteogenetic processes under tensional forces were active, dependent upon sufficient blood supply9. A lamellar bone structure and Haversian systems were observed in various phases of formation. The new bone structure was gradually remodelled to be suitable for functional loads11–13,16,21. It is interesting that, throughout the retention phase, the radiographs showed an increase in radiodensity in the distraction gap from the bilateral margins to the centre, indicating that osteogenesis and calcification were in progress. After 6 weeks, radiographs showed
good calcification with only a thin fissurelike radiolucent trail in the centre. This was the case until removal of the distractors after 6 weeks according to the protocol, and indicates that, in the palatal region, 6 weeks of retention provides a good balance between fixation for osteogenesis and early exposure to functional loads for remodeling13,22,23. Since the palatal bone shelf in the radiographs of this region was not shown in strong contrast, the newly formed trabeculae that oriented in parallel to the distraction vector could be easily identified like in other sites12,13,24,25. The results of tetrachloride fluorescent labelling suggest that there are three main stages in the DO procedure, consisting of 2 weeks as the early stage, 4–6 weeks as the middle stage, and 8–12 weeks as the late stage. The fluorochrome strip deposited on trabeculae indicated active osteogenesis while discontinuity of the strip indicated active resorption of new bone remodelling. These results reveal a continuously dynamic changing rules of new bone formation and remodelling in one aspect. In this study, the newly designed intraoral distractor proved effective in steadily transporting the TD segment to close the defect area, and simultaneously maintaining osteogenesis and remodelling in the retention phase. The rate of 0.4 mm twice a day worked better than other rates of 0.5 mm/day, 1 mm/day and 2 mm/day that had been tried before. The best outcomes were achieved with regard to osteogenesis efficiency, remodelling activity and calcification. The 6 days of latency before the DO procedure were crucial for both early bone callus formation and bony fusion avoidance at the TD osteotomy incision. There are some problems with this technique. The vectors of distraction will be limited by the placement of internal devices as well as their finite mechanics, and implantation of internal devices will likely interfere with the design of the transverse maxillary osteotomy. In this preliminary study, evaluation of a new DO internal tissue-engineering method to reconstruct the ‘normal’ structure of the hard-palate bone shelf and mucoperiosteal soft tissue was the main aim. When the transport disc meets the contralateral hard palatal defect margin at the end of active distraction, the soft-palate tissues on opposite sides become closer, which could make secondary functional repair of the soft-palate musculature easier and more efficient. To sum up, the results of this study suggest that the DO technique should be considered for repairing
725
acquired palate defects, such as that might occur after trauma and tumour resection, as well as cleft and other congenital cases. Acknowledgment. This study was supported by the Sichuan Science and Technology Foundation for Key Research Project (Grant No. G9906). References 1. Aronson J. Temporal and spatial increases in blood flow during distraction osteogenesis. Clin Orthop 1993: 301: 124–131. 2. Aronson J, Good B, Stewart C, Harrison B, Harp J. Preliminary studies of mineralization during distraction osteogenesis. Clin Orthop 1990: 250: 43–51. 3. Ascherman JA, Marin VP, Rogers L, Prisant N. Palatal distraction in a canine cleft palate model. Plast Reconstr Surg 2000: 105: 1687–1694. 4. Block MS, Daire J, Stover J, Matthews M. Changes in the inferior alveolar nerve following mandibular lengthening in the dog using distraction osteogenesis. J Oral Maxillofac Surg 1993: 51: 652–660. 5. Carls FR, Jackson IT, Topf JS. Distraction osteogenesis for lengthening of the hard palate: part I. A possible new treatment concept for velopharyngeal incompetence. Experimental study in dogs. Plast Reconstr Surg 1997: 100: 1635– 1647. 6. Cook SD, Baffes GC, Wolfe MW, Sampath TK, Rueger DC, Whitecloud TS. The effect of recombinant human osteogenic protein-1 on healing of large segmental bone defects. J Bone Joint Surg 1994: 76-A: 827–837. 7. Cook SD, Wolfe MW, Salkeld SL, Rueger DC. Effect of recombinant human osteogenic protein-1 on healing of segmental defects in non-human primates. J Bone Joint Surg 1995: 77-A: 734–749. 8. Costantino MPD, Friedman CD, Shindo ML, Houston CG, Sisson GA. Experimantal mandibular regrowth by distraction osteogenesis. Arch Otolaryngol Head Neck Surg 1993: 119: 511– 516. 9. Delloye C, Delefortrie G, Coutelier L, Vincent A. Bone regenerate formation in cortical bone during distraction osteogenesis. Clin Orthop 1990: 250: 34–42. 10. Dixit UB, Kelly KM, Squier CA, Bardach J. Periosteum in regeneration of palatal defects. Cleft Palate Craniofac J 1995: 32: 228–234. 11. Juenger TH, Klingmueller V, Howaldt HP. Application of ultrasound in callus distraction of the hypoplastic mandible: an additional method for the follow-up. J Craniomaxillofac Surg 1999: 27: 160–167.
726
Wang et al.
12. Karp NS, McCarthy JG, Schreiber JS, Sissons HA, Thorne CHM. Membranous bone lengthening: a serial histological study. Ann Plast Surg 1992: 29: 2–7. 13. Karp NS, Thorne CHM, McCarthy JG, Sissons HA. Bone lengthening in the craniofacial skeleton. Ann Plast Surg 1990: 24: 231–237. 14. Kobayashi ET, Hashimoto F, Kobayashi Y, Sakai E, Miyazaki Y, Kamiya T, Kobayashi K, Kato Y, Sakai H. Forceinduced rapid changes in cell fate at midpalatal suture cartilage of growing rats. J Dent Res 1999: 78: 1495–1504. 15. Liu C, Song R, Song Y. Sutural expansion osteogenesis for management of the bony-tissue defect in cleft palate repair: experimental studies in dogs. Plast Reconstr Surg 2000: 105: 2012–2025 discussion 2026–2027. 16. Losken A, Mooney MP, Siegel MI. Comparative cephalometric study of nasal cavity growth patterns in seven animal models. Cleft Palate J 1994: 31: 17–23.
17. McCarthy JG, Schreiber J, Karp N, Thorne CH, Grayson BH. Lengthening the human mandible by gradual distraction. Plast Reconstr Surg 1992: 89: 1–8. 18. Molina F, Monasterio FO. Mandibular elongation and remodelling by distraction: a farewell to major osteotomies. Plast Reconstr Surg 1995: 96: 825–834. 19. Oda T, Sawaki Y, Fukuta K, Ueda M. Segmental mandibular reconstruction by distraction osteogenesis under skin flaps. Int J Oral Maxillofac Surg 1998: 27: 9– 13. 20. Paley D. Current techniques of limb lengthening. J Pediatr Orthop 1988: 8: 73–92. 21. Paley D, Rumley TO, Kovelman H. The Ilizarov technique: a method to regenerate bone and soft tissue. Adv Plast Reconstr Surg 1991: 7: 1–11. 22. Rachmiel A, Levy M, Laufer D, Clayman L, Jackson IT. Multiple segmental gradual distraction of facial skeleton: an experimental study. Ann Plast Surg 1996: 36: 52–59.
23. Schmitz JP, Hollinger JO. The critical size defect as an experimental model for craniomandibulofacial nonunions. Clin Orthop 1986: 205: 299–307. 24. Siegel MI, Mooney MP, Eichberg JW, Gest T, Lee DR. Septopremaxillary ligament resection and midfacial growth in a chimpanzee model. J Craniofac Surg 1990: 1: 182–186. 25. Siegel MI, Mooney MP. Appropriate animal models for craniofacial biology. Cleft Palate J 1990: 27: 18–25. Address: Gang Chen Department Oral & Maxillofacial Surgery College of Stomatology Tianjin Medical University 12# Qixiangtai Rd Heping district Tianjin 300070 PR China Tel: +86 13920095115 Fax: +86 22 23332122 E-mail: [email protected]
Leading Research Paper Congenital Craniofacial Deformities
A new approach to repairing cleft palate and acquired palatal defects with distraction § osteogenesis
D.-Z. Wang1, G. Chen1,2, Y.-M. Liao1, S.-G. Liu1, Z.-W. Gao1, J. Hu1, J.-H. Li1, C.-H. Liao1 1 Department Oral & Maxillofacial Surgery, West China College of Stomatology, Sichuan University, 14#, 3rd section, Renminnan Rd., Chengdu 610041, PR China; 2Department Oral & Maxillofacial Surgery, College of Stomatology, Tianjin Medical University, 12# Qixiangtai Rd, Heping district, Tianjin 300070, PR China
D.-Z.Wang, G. Chen, Y.-M. Liao, S.-G. Liu, Z.-W. Gao, J. Hu, J.-H. Li, C.-H. Liao: A new approach to repairing cleft palate and acquired palatal defects with distraction osteogenesis. Int. J. Oral Maxillofac. Surg. 2006; 35: 718–726. # 2006 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved. Abstract. Cleft palate (CP) is one of the most common human congenital deformities, and acquired palate defects after trauma or tumour resection are also common. In this study, distraction osteogenesis (DO) for CP and other palatal bone defects was evaluated. Twenty cats were assigned randomly to 3 groups of (1) 15, (2) 3 and (3) 2 cats. In groups 1 and 2, a rectangular ostectomy, in the posterior of the palatal bone shelf, was performed in the sagittal axis to establish the CP defect model. At the same time, a pure titanium intraoral distractor was fixed to molar teeth with brackets and to the palatal bone shelf across the defect with titanium miniscrews bilaterally. Four weeks later, a secondary transport disc (TD) osteotomy was performed, and gradual DO treatment started at 0.4 mm twice a day, after 6 days of latency. DO was performed until the TD reached the opposite margin over the gap in 5–6 days. Three cats each of group 1 were killed at 2, 4, 6, 8 and 12 weeks after completion of DO. In group 2, the bone and soft-tissue defects were untreated until death 6 weeks later. Group 3 cats (control) were killed after 6 weeks. The TD successfully recombined with the opposite palatal bone stump, and proportional expansion of the overlay mucoperiosteal flap was achieved. Intramembranous bone formation was revealed: parallel collagen bundles gradually deposited on new bone trabeculae while the proliferative osteoblasts produced bone matrix. The bone defect was finally reconstructed by de novo osteogenesis. The control group was observed to have no spontaneous repairing. These results suggest that the CP defect was reconstructed by osteogenesis in situ, and the soft tissues expanded simultaneously to achieve functional correction. The intraoral distractor provided both effective distraction and stability.
§
Key words: distraction osteogenesis; cleft lip/ palate; animal model; intraoral distractor. Accepted for publication 3 March 2006 Available online 9 May 2006
The preliminary report was presented at the 82nd general session and exhibition of the IADR, 10–13 March 2004, Honolulu, Hawaii, USA.
0901-5027/080718 + 09 $30.00/0
# 2006 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
Repairing CP and acquired palatal defects with DO Cleft palate (CP), with or without cleft lip, is one of the most common human congenital deformities. Chewing, swallowing, talking and breathing are severely affected, and the condition also causes some psychological problems. Acquired palate defects after trauma or tumour resection are also very common. Anatomical disfigurement is of both the soft tissues and bone shelf, especially in the posterior part of the hard palate and the soft palate. Traditional surgical repair of CP depends on transferring adjacent palatal mucoperiosteal flaps to close the gap, with or without lengthening the soft palate, and to narrow the pharyngeal cavity. It is still the case that surgeons cannot truly reconstruct the CP bone defect and correct the abnormal muscular attachment to physiologically normal positions, so called to ‘put normal to normal’. Also, scarring after traditional surgical procedures may gradually cause a series of secondary deformities, such as decreased maxillary growth and impeded arch expansion, as the patients grow up. This may have highly detrimental psychosocial ramifications. Distraction osteogenesis (DO) is a technique involving the application of an endosseous appliance combined with cortical osteotomies and the use of tensional distraction to lengthen existing bone and soft tissue. This technique has been widely used in the field of oral maxillofacial surgery in recent decades. Although there are certain difficulties, this technique is effective because, compared to long bones, the blood supply associated with the maxillary and palatal region is richer. Since the hard palatal bone defect of CP is more complex than at any other site, because of the thinner and structurally weaker bones, complicated by interrelationships with dental arches, and oral and nasal cavities, an individually manufactured intraoral distractor was designed to practice DO correction of CP deformities.
long incision from the canine cusp level and 1 cm into the soft palate was made, parallel to the central palatal fissure but 5 mm to the side, and then the mucoperiosteal flap was lifted to expose the bone shelf. Two sagittal incisions were made on the bone shelf 2 and 10 mm to the side of the central hard palatal fissure unilaterally, from the level of the anterior canine cusps to the posterior palatal mar-
719
gin. Another 8-mm incision at the level of the canine cusps frontally was made to connect the 2 sagittal incisions to form an 8 15 mm2 longitudinal rectangular ostectomy (Fig. 1). The overlaying soft tissues were cropped correspondingly. The margins of the oral mucoperiosteal flap and nasal mucosa were sutured together to ensure penetration into the oral nasal defect.
Materials and methods The CP animal model
Twenty cats (0.5–0.8 years old, 3.5–4.5 kg weight) were used in this study, assigned randomly into three groups of (1) 15 cats, (2) 3 cats and (3) 2 cats. The 18 animals in groups 1 and 2 were used to perform the CP experimental modelling procedures, with group 3 acting as control. Under general anaesthesia combined with local infiltration of lidocaine 2% and epinephrine 1:100,000, an 18-mm-
Fig. 1. The intraoral distractor and the CP animal model. (a) Schematic drawing of the intraoral CP distractor. A: immobile unit with jackscrew bolt; B: mobile unit; C: two parallel poles. (b) Oronasal penetration defect in CP experimental animal model with the pure titanium distractor mounted in position. B: molar bracket; S: pure titanium miniscrew; TDv: transporting parts; TD: incision line of transport disc. (c) View of the CP animal model in group 2 (6 weeks) shows the hard palate bone defect unrepaired. TD: incision line of transport disc.
720
Wang et al.
Design of intraoral distractor
The pure titanium intraoral distraction appliance (Chinese patent ZL 01 2 06830.6) consists of an immobile unit with a jackscrew bolt (0.4 mm for each revolution) impelling a mobile unit to move in an axial direction along 2 parallel poles. Both of the units are fixed, by 2 endosseous pure titanium miniscrews each, into the palatal bone shelf across the CP bone defect gap. Another molar bracket is made by plaster cast and then welded to the distractor to supply perpendicular anchorage to the titanium miniscrews (Fig. 2). DO protocol
As soon as the surgical procedures of CP modelling were completed, the pre-made intraoral distractor welded with molar brackets was tested and then mounted. Prophylactic i.m. antibiotic was administered at the time of surgery and for 5 postoperative days. A secondary osteotomy for transport disc (TD) formation was performed 4 weeks later so that early bone integrity of the titanium screw to the palatal shelf could be achieved. During the operation of TD formation, bone cutting was performed 2 mm adjacent to the alveolar ridge on the CP side to insure inclusion of the major palatal artery in the segment, so that the TD would be transported with an adequate blood supply. The CP animals were fed with a fluid diet for 5 days postoperatively, and most of them returned to normal food in 10 days. After 6 days of latency, the jackscrew shaft was activated twice a day (1 revolution each time, to a total distance of 0.8 mm) to impel the intraoral distractor with the TD segment. Once the TD had reached the opposite bone edge over the CP defect gap, the distractor was left in place for an additional 6 weeks before removal. Specimen retrieval and preparation
The 15 CP model animals assigned to group 1 underwent the DO procedure. Three cats each of this group were killed with excessive pentobarbital at 2, 4, 6, 8 and 12 weeks after completion of DO. The 3 other CP animals (group 2) and the 2 control animals (group 3) were killed after six weeks. Tetrachloride fluorescent labeling procedures (30 mg/kg) were administered 6 days before death. After gross observations and photographs were taken, X-ray examinations were made in situ. The specimens were
Fig. 2. Gross view after 4 weeks of DO correction on soft tissue before (a) and after (b) removal of the distractor, and view of the bone shelf (c). Only thin trace of original defect (arrow). B: molar bracket; S: pure titanium miniscrew; TDv: transporting parts; M: molar on normal side; M0 : molar on distraction side; T: hole of removal of titanium screw; P: posterior margin of hard palate; A and B: sutures; O: palatal foramen on normal side; O0 : palatal foramen on distraction side; S-S0 : anteroposterior borders of new bone formation; C-C0 : bilateral borders of new bone formation; NB: DO newly formed bone; TD: transport disc.
Repairing CP and acquired palatal defects with DO
721
retrieved en bloc, including the palatal region 5 mm either side of the distraction gap, and then were divided into 2 parts for microscopic histological and scanning electron microscopic (SEM) ultrastructural study. For histological study, the specimens were immersed in 10% neutral buffered formalin solution, after being rinsed with saline, for 7 days of fixation. After being decalcified in sodium formate-formic acid, the specimens were embedded in paraffin. Frontal 6-mm sections were cut with a microtome, including the whole distraction gap and bilateral original bone, and then stained with haematoxylin & eosin and modified Masson trichrome. For ultrastructural study, the specimens were trimmed and rinsed with saline and 0.2 mol/l phosphate-buffered saline, and then immersed in 2.5% glutaraldehyde at 4 8C for 24 h fixation. Frontal splitting was performed with the intermediate agent of liquid nitrogen, followed by dehydration in increasing concentrations of acetone. Every specimen was manipulated by vacuum-sputtering of gold membrane to insure electroconductibility and analysed with a SEM (AMRAY, 1000B, USA). Results Gross observations
All animals tolerated the implantation of appliances without infection or impairment in respiratory or visual functions. After the surgical procedure of CP modelling, the animals manifested dramatic symptoms as in clinical human cases, including excessive nasal secretion, sneezing, nasal flowing of food crumbs and water, and so on. Intraorally, an obvious sagittal rectangular oral–nasal penetration defect was observed. At 12 weeks after the DO of group 1, the original defect had disappeared (Fig. 3), and measurement showed that the relationship between the maxilla and mandible was normal, i.e. as before DO. In group 2 (without DO), after 6 weeks, the defects in the hard palate were observed to be unhealed but a little narrower than the original size. At the same time, a mild cusp-to-cusp bite of molar teeth was observed, implying a slight collapse of the bilateral palatal bone into the defect area, because of the discontinuity of the hard bone shelf support. During the whole distraction period, no dehiscence of the incision was seen but the palatal plicate was observed to be gradually flattening. Interestingly, inosculation between the overlying soft tissue of the TD and the opposite CP defect bone stump was also observed.
Fig. 3. Gross view after 12 weeks of DO on soft tissue (upper) and bone shelf (lower). The original defect has disappeared (arrow). M0 : molar on the distraction side; NB: DO newly formed bone; P: posterior margin of hard palate; TD: transport disc; T: hole of removal of titanium screw; T0 : cured screw holes on TD.
The distraction gap had filled with tough fibrous tissues in continuity with the medial and lateral palatal bone margins after 2 weeks. The oral and nasal cavities were separated again. There was only a central fibrous zone left after around 4–6 weeks (Fig. 4). After 8 weeks, the primary defect was corrected by TD transportation across the distraction gap and bone tissue regeneration had occurred. Hard palatal bone shelf continuity was recovered. The overlying soft tissue looked much like that in group 3 (control), without signs of alteration. There was in particular no necrosis or scar formation.
Roentgenographic findings
Although the hard palatal bone is thin and structurally weak, a fine-scale contrast gradient of anatomical structures could be identified in groups 2 and 3. The half-elliptical dental arch, middle palate fissure with nasal septum, posterior margin of the hard palate, pterygoid plates and major palatal foramen were important landmarks, with additionally the CP defect area (Fig. 5a). In group 1, the radiograph made after 2 weeks showed a rectangular radiolucent area between the cut bone edges of the secondary osteotomy for TD formation,
722
Wang et al.
Fig. 4. Roentgenographic view of hard palate in group 2 (a, arrow indicating CP bone defect area), and 2 weeks (b, distractor in place), 6 weeks (c, arrow indicating new bone formation in distraction gap) after DO correction (group 1). TD: transport disc; NB: newly formed bone.
indicating the distraction gap filled with collagenous fibres (Fig. 5b). For this reason, the distractors were removed after 6 weeks; the landmark of the major palatal foramen and the miniscrew holes were recognizable in radiographs made after more than 6 weeks. Obviously, the major palatal foramen on the DO side was medially closer to the middle palatal fissure since it was within the TD and being transported medially (Fig. 5c). The radiographs made at different time intervals demonstrate a tendency of the radiodensity of the distraction gap to increase, extending gradually from the bilateral bone edges to the centre, until eventually the gap became indistinguishable from the adjacent bone after 12 weeks. Histological and fluorescent labeling evaluation
Group 3
In the normal palatal mucoperiosteal soft tissue, multi-layered structures characterized by long keratose epithelial tags into subcutaneous connective tissue, containing perpendicularly interweaved collagen fibre networks, were observed. Hard palatal bone tissue consisted of well-calcified lamellar bone and mature Haversian system. No tetrachloride fluorescent labelling could be found. Group 1
Histological study of the group 1 specimens revealed that there was a uniform gradient of mineralization from the centre of the distraction gap bilaterally to original bone tissue. The characteristic histological view of newly formed tissue within the distraction gap was of a triple-zone structure from the centre to the original bone edges bilaterally (Fig. 5a). The three zones are: (1)
fibrous zone: plenty of collagen fibre bundles were arranged in parallel in the direction of the distraction vector, and fibroblast cells and mesenchymal precursor cells were scattered throughout the matrix; (2) new bone formation zone: newly formed bone trabeculae oriented in the direction of the distraction vector, and osteoblasts congregating on the surfaces of those trabeculae; (3) remodelling and mature zone: a honeycomb structure of bone trabeculae woven in different directions with abundant cellular components, and gradual transition to more mature bone through remodelling activities. After 2 weeks of retention, the collagenous fibres oriented along the distraction vector were predominant components. A few fine needle-like newly formed spiculae were viewed near the cut bone edges bilaterally pointing to the centre of the distraction zone. Fibroblasts and undifferentiated mesenchymal precursor cells were viewed in direct continuity with osteoblasts on the surfaces of early bone spicules (Fig. 5b). Fluorochrome labeling of tetrachloride was observed on the surface of the osteoid layer of new trabeculae in high density. In the 4- and 6-week subgroups, proliferative osteoblasts adjacent to osteoid on the surfaces of newly formed trabeculae were observed. Collagenous fibres deposited and calcified on thick trabeculae. These bone columns did not contain Haversian systems, but instead were surrounded by large, thin-walled blood vessels and loose connective tissues in medullary cavities. Near the original cut bone edges bilaterally, active bone absorption and remodelling were observed and osteoclasts could be identified (Fig. 5c). High-density tetrachloride fluorescent labelling on new trabeculae was distinct; meanwhile, discontinuity of the labelling
strips was observed, caused by immature woven bone near the cut bone edges (Fig. 5d). After 8 weeks, more mature and calcified bone with recognizable lamellar structure and newly formed Haversian systems could be seen (Fig. 5e). Bone formation and remodelling destruction were observed simultaneously. Fluorochrome labelling on bone trabeculae surfaces and bone lacunae was broad with discontinuity of the lining in common (Fig. 5f). After 12 weeks, the border between newly formed bone and original bone was still distinct but the structure of the new bone was more mature. There was no evidence of tetrachloride fluorescent labelling. Chondral tissues and cartilage were not found in any of the subgroups. As for the soft tissue extending over the distraction gap, no histological changes were found evident, but the epithelial tags became shorter, and the orientation of the collagenous fibre network changed slightly to that of the distraction vector. No inflammatory infiltration into soft tissue was observed (Fig. 5g). Group 2
There was no evidence of spontaneous repair in the original cut bone stumps of the CP defect gap, but only some appositional covering of amorphous structure. There was no tetrachloride fluorochrome labelling (Fig. 5h). Ultrastructural evaluations
Group 3
Under the SEM, normal bones demonstrated a well-calcified uniform structure, with mature spindle-shaped osteocytes located in small, flat bone lacunae.
Repairing CP and acquired palatal defects with DO
723
Fig. 5. Micrographs and fluorochrome labeling views. (a) Triple-zone structure within the distraction gap. OB, original bone; NB, newly formed bone; COL, collagenous fibrous tissue (H&E, 100). (b) At 2 weeks, collagenous fibers oriented in the distraction vector direction with proliferated osteoblasts on surfaces of early bone spicules (H&E, 400). (c) At 4 weeks, newly formed bone trabeculae oriented in the direction of distraction (H&E, 100). (d) Fluorochrome labeling at 4 weeks, strong tetrachloride strip on surfaces of parallel newly formed trabeculae (100). (e) Newly formed Haversian system at 8 weeks (Masson, 400). (f) At 8 weeks, discontinuity of fluorochrome strips on trabeculae surfaces indicating bone formation and remodeling simultaneously (100). (g) Micrograph of elongated soft tissue over new bone of distraction gap (H&E, 100). (h) In group 2, only thin layer of amorphous substance on cut bone edge without repair (arrow) (H&E, 100).
Group 1
SEM viewing of specimens after 2 weeks revealed many collagenous fibre bundles oriented parallel to the direction of the distraction vector. Abundant polymorphous cellular components including fibroblasts and undifferentiated mesenchymal precursor cells were scattered all over the matrix. Minimally calcified osteoid was on the surfaces of early bone spicules (Fig. 6a).
After 4–6 weeks, active osteoblast proliferation adjacent to osteoid accumulation on surfaces of newly formed trabeculae was observed. Collagenous fibre deposition calcified on thick trabeculae. Newly generated osteoblasts were gradually embedded in the bone matrix and became osteocytes, with relatively large, round lacunae compared to those in the original bone; in this regard, the newly formed bone was shown to be immature. Honeycomb-like newly formed woven bone was
observed near the original bone stumps (Fig. 6b). After 8–12 weeks, the bone structure was observed to be more mature with strong trabeculae, a more calcified matrix, a lamellar bone structure, and so on. Osteocytes in round bone lacunae looked more like what was found in the original bone. Newly formed Haversian systems were also found in some fields of vision. Nevertheless, the interface between original bone and new bone was still distinguishable (Fig. 6c).
724
Wang et al.
Fig. 6. SEM ultrastructural views. (a) At 2 weeks, many collagenous fiber bundles oriented parallel to the direction of distraction (1000). (b) Honeycomb-like structure of cellular new bone at 4 weeks (1000; OB, original bone; NB, new formed bone). (c) Newly forming Haversian system (arrow) in new bone tissue at 6 weeks (500; OB, original bone; NB, newly formed bone). (d) In group 2, no sign of repair of original cut bone stumps (600; OB, original bone).
Group 2
The original cut bone stumps showed an amorphous substance instead of osteogenesis occurring (Fig. 6d). Discussion
Using surgically established CP animal models, the method of hard palatal DO, with newly designed intraoral distractors, was shown to be effective for repairing CP bone and soft-tissue defects for the first time in this study. The TD segment and overlying soft tissue were transported over the defect area followed by de novo osteogenesis in the distraction gap, and consequently palatal integrity was successfully reconstructed by endogenous tissue engineering. This new method minimized surgical trauma to local structures and avoided scar formation, to which a series of secondary developing deformities have been primarily attributed. No reliable and repeatable drug induction method for a congenital CP model in large animal studies is available yet. This is a future target, but for now a CP experimental model must be established surgically. The CP experimental animal model used here simulated the clinical demonstration of CP effectively and responded positively to the DO procedure for bone regeneration. Although in all the dimensions investigated so far, ‘short-faced’
non-human primates have shown the most similar growth patterns to humans, several other non-primate animal models have been identified for modelling specific regions at various ages1–3. The results of modelling CP changes suggested that domestic cats might be a good alternative to non-human primate such as chimpanzees for the study. In addition, being a carnassial species with wide mouth opening, cats are a good model for repeating intraoral manipulations of distraction, and are less aggressive and dangerous than dogs. The cats used in this study were all juveniles aged 0.5–0.8 years old; the longest time interval followed after active distraction was 12 weeks, after which the cats would have reached maturity. It could therefore be observed that there was no influence of the DO technique on development of the oral and maxillofacial area3,4. In addition, comparative measurement of the jaws using experimental casts implied that the DO procedure might not retard the normal developmental tendency of the midface, and could maintain a normal occlusal relation in these cats. In group 2, the horizontal distance of the maxilla became narrower and presented a slight crossbite relation. Other investigators’ results have shown that defects in long bones equalling 1 to 2 times the size of the diaphyseal diameter routinely produce a permanent non-union
termed a ‘critical size defect’ (CSD); hence, only osteogenesis factors can consistently repair CSDs5–7. In this study, the size of the CP defect was designed to be far larger than that of the thin palatal bone shelf. The amount of healing that will occur in a bony defect is to a great extent dependent upon wound size; therefore, spontaneous healing was precluded. Interestingly, it was observed in this study that when the transport disc met the contralateral hard palatal defect margin at the end of active distraction, the thin fissure between the 2 parts disappeared automatically within 2 weeks. The mechanism of this healing process is unclear as yet. If this or a similar technique could be used clinically in human CP patients in the future, it would involve a simple secondary surgical trimming procedure to heal the fissure. In the present study, application of the DO principle was carried out according to a new design to suit the thin and weak palatal bone structure. The ultimate goals were: (1) regenerating new bone and overlying soft tissue and reconstructing continuity of the CP defect area and distraction gap; (2) rehabilitating osculation between the bone transport disc and opposite bone stump; (3) maintaining functional and structural stability. These are quite different aims to that of lengthening the posterior part of the hard palate and the soft palate8.
Repairing CP and acquired palatal defects with DO The results of the histological study revealed a significant intramembranous osteogenetic process in the distraction gap8–10,11–13,15. The new bone was less organized with larger vascular channels than the original bone nearby14. Proliferated osteoblasts were gradually embedded in the bone matrix and became osteocytes with relatively larger osteocytic lacunae than those in the pre-existing bone. In this regard, the newly formed bone was shown to be immature; active resorption of unfavourable structures and remodelling eventually made it more suitable for normal functional loads8,11,14,16–18. The cancellous structure of newly regenerated bone, consisting of bony trabeculae surrounding large spaces occupied by soft tissue and blood vessels, assures sufficient blood supply9. During the distraction phase, the blood supply to the distraction zone reached a maximum level of seven times the normal contralateral side, and could remain at approximately 3 times the normal level for the remodelling phase thereafter11,19,20. There was no chondral phase observed at any time interval with any of the specimens. The double perpendicular anchorages of the distractor supplied stable fixation, which was essential for adequate formation of microcolumns of bone during DO. It is obvious that bending or shear stress could induce fractures of the microcolumns with local haemorrhage and resultant histological cartilage interposition11,18. Most investigators attribute the presence of cartilage to imperfect fixation that has resulted in some motion between the distracted bone edges18. Fibroblasts and undifferentiated mesenchymal precursor cells were found in direct continuity with osteoblasts on the surfaces of early bone spicules. This implies that the bone-forming cells are created by transformation of these precursor cells throughout the matrix. In the ultrastructural study, SEM views corroborated the results of the histological evaluation, indicating that, although at a thin and structurally weak site, osteogenetic processes under tensional forces were active, dependent upon sufficient blood supply9. A lamellar bone structure and Haversian systems were observed in various phases of formation. The new bone structure was gradually remodelled to be suitable for functional loads11–13,16,21. It is interesting that, throughout the retention phase, the radiographs showed an increase in radiodensity in the distraction gap from the bilateral margins to the centre, indicating that osteogenesis and calcification were in progress. After 6 weeks, radiographs showed
good calcification with only a thin fissurelike radiolucent trail in the centre. This was the case until removal of the distractors after 6 weeks according to the protocol, and indicates that, in the palatal region, 6 weeks of retention provides a good balance between fixation for osteogenesis and early exposure to functional loads for remodeling13,22,23. Since the palatal bone shelf in the radiographs of this region was not shown in strong contrast, the newly formed trabeculae that oriented in parallel to the distraction vector could be easily identified like in other sites12,13,24,25. The results of tetrachloride fluorescent labelling suggest that there are three main stages in the DO procedure, consisting of 2 weeks as the early stage, 4–6 weeks as the middle stage, and 8–12 weeks as the late stage. The fluorochrome strip deposited on trabeculae indicated active osteogenesis while discontinuity of the strip indicated active resorption of new bone remodelling. These results reveal a continuously dynamic changing rules of new bone formation and remodelling in one aspect. In this study, the newly designed intraoral distractor proved effective in steadily transporting the TD segment to close the defect area, and simultaneously maintaining osteogenesis and remodelling in the retention phase. The rate of 0.4 mm twice a day worked better than other rates of 0.5 mm/day, 1 mm/day and 2 mm/day that had been tried before. The best outcomes were achieved with regard to osteogenesis efficiency, remodelling activity and calcification. The 6 days of latency before the DO procedure were crucial for both early bone callus formation and bony fusion avoidance at the TD osteotomy incision. There are some problems with this technique. The vectors of distraction will be limited by the placement of internal devices as well as their finite mechanics, and implantation of internal devices will likely interfere with the design of the transverse maxillary osteotomy. In this preliminary study, evaluation of a new DO internal tissue-engineering method to reconstruct the ‘normal’ structure of the hard-palate bone shelf and mucoperiosteal soft tissue was the main aim. When the transport disc meets the contralateral hard palatal defect margin at the end of active distraction, the soft-palate tissues on opposite sides become closer, which could make secondary functional repair of the soft-palate musculature easier and more efficient. To sum up, the results of this study suggest that the DO technique should be considered for repairing
725
acquired palate defects, such as that might occur after trauma and tumour resection, as well as cleft and other congenital cases. Acknowledgment. This study was supported by the Sichuan Science and Technology Foundation for Key Research Project (Grant No. G9906). References 1. Aronson J. Temporal and spatial increases in blood flow during distraction osteogenesis. Clin Orthop 1993: 301: 124–131. 2. Aronson J, Good B, Stewart C, Harrison B, Harp J. Preliminary studies of mineralization during distraction osteogenesis. Clin Orthop 1990: 250: 43–51. 3. Ascherman JA, Marin VP, Rogers L, Prisant N. Palatal distraction in a canine cleft palate model. Plast Reconstr Surg 2000: 105: 1687–1694. 4. Block MS, Daire J, Stover J, Matthews M. Changes in the inferior alveolar nerve following mandibular lengthening in the dog using distraction osteogenesis. J Oral Maxillofac Surg 1993: 51: 652–660. 5. Carls FR, Jackson IT, Topf JS. Distraction osteogenesis for lengthening of the hard palate: part I. A possible new treatment concept for velopharyngeal incompetence. Experimental study in dogs. Plast Reconstr Surg 1997: 100: 1635– 1647. 6. Cook SD, Baffes GC, Wolfe MW, Sampath TK, Rueger DC, Whitecloud TS. The effect of recombinant human osteogenic protein-1 on healing of large segmental bone defects. J Bone Joint Surg 1994: 76-A: 827–837. 7. Cook SD, Wolfe MW, Salkeld SL, Rueger DC. Effect of recombinant human osteogenic protein-1 on healing of segmental defects in non-human primates. J Bone Joint Surg 1995: 77-A: 734–749. 8. Costantino MPD, Friedman CD, Shindo ML, Houston CG, Sisson GA. Experimantal mandibular regrowth by distraction osteogenesis. Arch Otolaryngol Head Neck Surg 1993: 119: 511– 516. 9. Delloye C, Delefortrie G, Coutelier L, Vincent A. Bone regenerate formation in cortical bone during distraction osteogenesis. Clin Orthop 1990: 250: 34–42. 10. Dixit UB, Kelly KM, Squier CA, Bardach J. Periosteum in regeneration of palatal defects. Cleft Palate Craniofac J 1995: 32: 228–234. 11. Juenger TH, Klingmueller V, Howaldt HP. Application of ultrasound in callus distraction of the hypoplastic mandible: an additional method for the follow-up. J Craniomaxillofac Surg 1999: 27: 160–167.
726
Wang et al.
12. Karp NS, McCarthy JG, Schreiber JS, Sissons HA, Thorne CHM. Membranous bone lengthening: a serial histological study. Ann Plast Surg 1992: 29: 2–7. 13. Karp NS, Thorne CHM, McCarthy JG, Sissons HA. Bone lengthening in the craniofacial skeleton. Ann Plast Surg 1990: 24: 231–237. 14. Kobayashi ET, Hashimoto F, Kobayashi Y, Sakai E, Miyazaki Y, Kamiya T, Kobayashi K, Kato Y, Sakai H. Forceinduced rapid changes in cell fate at midpalatal suture cartilage of growing rats. J Dent Res 1999: 78: 1495–1504. 15. Liu C, Song R, Song Y. Sutural expansion osteogenesis for management of the bony-tissue defect in cleft palate repair: experimental studies in dogs. Plast Reconstr Surg 2000: 105: 2012–2025 discussion 2026–2027. 16. Losken A, Mooney MP, Siegel MI. Comparative cephalometric study of nasal cavity growth patterns in seven animal models. Cleft Palate J 1994: 31: 17–23.
17. McCarthy JG, Schreiber J, Karp N, Thorne CH, Grayson BH. Lengthening the human mandible by gradual distraction. Plast Reconstr Surg 1992: 89: 1–8. 18. Molina F, Monasterio FO. Mandibular elongation and remodelling by distraction: a farewell to major osteotomies. Plast Reconstr Surg 1995: 96: 825–834. 19. Oda T, Sawaki Y, Fukuta K, Ueda M. Segmental mandibular reconstruction by distraction osteogenesis under skin flaps. Int J Oral Maxillofac Surg 1998: 27: 9– 13. 20. Paley D. Current techniques of limb lengthening. J Pediatr Orthop 1988: 8: 73–92. 21. Paley D, Rumley TO, Kovelman H. The Ilizarov technique: a method to regenerate bone and soft tissue. Adv Plast Reconstr Surg 1991: 7: 1–11. 22. Rachmiel A, Levy M, Laufer D, Clayman L, Jackson IT. Multiple segmental gradual distraction of facial skeleton: an experimental study. Ann Plast Surg 1996: 36: 52–59.
23. Schmitz JP, Hollinger JO. The critical size defect as an experimental model for craniomandibulofacial nonunions. Clin Orthop 1986: 205: 299–307. 24. Siegel MI, Mooney MP, Eichberg JW, Gest T, Lee DR. Septopremaxillary ligament resection and midfacial growth in a chimpanzee model. J Craniofac Surg 1990: 1: 182–186. 25. Siegel MI, Mooney MP. Appropriate animal models for craniofacial biology. Cleft Palate J 1990: 27: 18–25. Address: Gang Chen Department Oral & Maxillofacial Surgery College of Stomatology Tianjin Medical University 12# Qixiangtai Rd Heping district Tianjin 300070 PR China Tel: +86 13920095115 Fax: +86 22 23332122 E-mail: [email protected]