Archives of Oral Biology (2003) 48, 299—308
Effect of distraction rate and consolidation period on bone density following mandibular osteodistraction in rats G.J. Kinga,*, Z.J. Liua, L.L. Wangd, I.Y. Chiub, M.F. Whelanc, G.J. Huanga a
Department of Orthodontics, University of Washington School of Dentistry, Box 357446, Seattle, WA 98195-3446, USA b Department of Dental Public Health Sciences, University of Washington School of Dentistry, Seattle, WA 98195-3446, USA c Division of Plastic Surgery, Children’s Hospital and Regional Medical Center and University of Washington School of Medicine, Seattle, WA 98195-3446, USA d Department of Orthodontics, College of Stomatology, Jilin University, Jilin, PR China Accepted 10 December 2002
KEYWORDS Mandibular distraction osteogenesis; Bone density; Microdensitometry; Rat
Summary The high cost of large animal protocols has limited the study of distraction osteogenesis (DO) in the craniofacial region. This study was designed to characterise a rat model for DO with regard to distraction rate and consolidation period. Unilateral mandibular distraction was performed on 129 male Sprague—Dawley rats using an osteotomy from the sigmoid notch to the inferior border of mandible. After a 3-day latency, 12 groups of 8—9 rats underwent distraction for 5 days at four different rates (0, 0.2, 0.4, 0.6 mm per day), with three different post-osteotomy sacrifice times (10, 24, and 38 days) and four final predicted distraction lengths (0, 1, 2, and 3 mm). Another four groups of rats (N ¼ 8 per group) were sacrificed 6 days post-osteotomy, resulting in distraction for 3 days with a predicted distraction length of 0, 0.6, 1.2, 1.8 mm. Changes in mandibular morphology were measured from radiographs of disarticluated hemimandibles. The bone density of the regenerate and control sites was measured using microdensitometry calibrated with an epoxy stepwedge. Distraction linearly increased mandibular length, distraction gap width and the area of the distraction gap (P < 0:00005). Mandibular length increased by 0.394 mm per distraction rate. Gap width and area increased by 0.67 and 5.8 mm2 per distraction rate, respectively. The increase in length represents only 39.4% of what was predicted, suggesting that compensatory alteration in condylar or mandibular morphology may have occurred. This speculation was further supported by the finding that mandibular length, measured without the condylar landmark, was 53.8% of predicted. During DO and early consolidation, the measures of bone density in the regenerates decreased compared to control for all groups. Thereafter, bone density in the regenerates generally increased in all groups until day 24 (P < 0:01), obtaining levels that were comparable to the unoperated side. At both rostral and caudal sites adjacent to the osteotomies, measures of bone density were enhanced over control in all groups, with the rostral site also showing significant increases over time in the sham and the highest distraction groups (P < 0:008 and P < 0:014). We conclude that this rat model for mandibular distraction osteogenesis provides bone density changes that are consistent with those reported using larger animal protocols. ß 2003 Elsevier Science Ltd. All rights reserved.
Corresponding author. Tel.: þ1-206-543-5788; fax: þ1-206-685-8163. E-mail address: [email protected]
0003–9969/03/$ — see front matter ß 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0003-9969(03)00004-9
Introduction Mandibular distraction is becoming a promising alternative for the treatment of retrognathia and asymmetry.1—4 Despite its growing popularity, distraction of the craniofacial skeleton presents with unique and complex problems that require further study in animal models. Dogs, pigs, sheep, monkeys and rabbits have been used to study craniofacial distraction.5—13 Although rat models have been described,14—16 none has yet been well characterised with respect to changes in mandibular morphology and bone remodelling under various distraction protocols. In this study, a recently described rat model for mandibular distraction14 is examined for the effects of distraction rate and consolidation period on the density of the bony regenerates and adjacent osteotomy sites, as well as on mandibular morphology.
Materials and methods Animal care One hundred twenty-nine 3-month male Sprague— Dawley rats weighing 340—380 g were used in this study. The housing, care, and experimental protocol were in accordance with the guidelines set by Animal Care and Use Committee at the University of Washington. Two days before surgery, both powdered diet and regular kibbles were supplied.
Anaesthesia and surgical procedure The procedure was based on a previously reported method.14 Briefly, rats were anaesthetised with ketamine (70 mg/kg) and xylazine (13 mg/kg) intraperitonealy. Cefazolin (10 mg/kg) was also given preoperatively. At surgery, a transverse skin incision was made along the inferior border of the right hemimandible. The masseter muscle was incised and carefully stripped to expose the mandibular angle, sigmoid notch, and mandibular body posterior to the third molar. The lingual border of the mandibular angle was also freed of its muscular attachment, and a pre-fabricated methylmethacrylate block (7 mm 3 mm 1 mm) was placed against the lingual surface of the angle to give additional support for the posterior installation of the device because the bone is extremely thin in this area (Fig. 1A, insert, a). The osteotomy was performed using a round diamond bur (0.5 mm tip diameter, Axis Dental Co. Irving, TX). The distraction device was positioned on the external surface of the mandible near the inferior border. The device consisted of two Luhr L-shaped micro-plates (0.8 mm; five holes) and a
G.J. King et al.
Leone jackscrew (0.2 mm per 1/4 turn, Fig. 1A, insert, b and c, respectively). The anterior microplate was secured to the mandibular body with two self-tapping micro-screws (0:8 mm 3 mm) (Fig. 1A, white dots). The posterior plate was installed at the mandibular angle with two additional screws, which were also anchored to the methacrylate block. An osteotomy was then performed from the sigmoid notch down to the inferior border between the two plates using a diamond bur in a low speed handpiece with copious saline irrigation. The wound was then closed in layers. The terminology for the osteotomy used in this report is rostral and caudal. Rostral refers to the tooth-bearing fragment of the mandible and caudal refers to the ramal fragment. Lactated Ringer’s solution (20 ml) was administrated subcutaneously following surgery. Buprenorphin (0.1 mg/kg) was injected subcutaneously just before the rat became fully conscious. Rats received powdered diets until 10 days after the operation, followed by a mixture of powdered chow and regular kibbles for the next 3 days. Thereafter, only kibbles were provided until sacrifice. Cefazolin (10 mg/kg) was given intraperitoneally 3 and 7 days after the operation. Body weights were monitored every 2—3 days post-operatively.
Distraction procedure Following a 3-day latency period, 32—33 rats were randomised to one of four groups. Group I did not receive activation and served as sham group. Groups II, III and IV had the distraction devices activated at rates of 0.2, 0.4 and 0.6 mm once a day respectively from days 3 to 7 (i.e. the length of the distraction period was 5 days, but 3 days for rats sacrificed on the sixth post-osteotomy day). The activation rhythm of once per day was chosen as a convenience and for safety because it was necessary to anaesthetise the rats for this procedure. Eight to nine rats in each group were sacrificed at each of the following days: 6, 10, 24, and 38. The distraction rates are considered to be slow (0.2 mm per day), moderate (0.4 mm per day) and rapid (0.6 mm per day). The times were selected to provide for mid-distraction (day 6), early consolidation (day 10), mid-consolidation (day 24) and late consolidation (day 38) observations. After sacrifice, both hemimandibles were removed and fixed in 10% formalin.
Assessment methods Mandibular morphology Radiographs of all hemimandibles were taken on X-Omat TL 8 in: 10 in: diagnostic film (Kodak, Koyoto, Japan) using an enclosed X-ray unit (Picker
Mandibular osteodistraction in rats
Figure 1 (A) Schematic illustration of the distraction/osteotomy locations and regions of interest. The solid line (bc) was used to measure the mandibular length including the condyle. This line was constructed from point c, the most superior point on the lingual alveolar crest of the mandibular incisor, to point b, the bisecting point on the condyle from the angle (point a) formed by the intersection of two tangent lines passing the most superior and posterior points of condyle. The dashed line was used to measure the mandibular length excluding the condyle. This line was constructed from point d, most concave point of the posterior notch of the mandible to point c. The four white dots along the inferior border of the mandible indicate the approximate locations of micro-screws for securing the distraction device (insert) on the buccal surface of the mandible. The three shaded areas represent the distraction gap, the caudal and the rostral osteotomy edges. Three capital letters in the gap indicate the locations where the gap widths were measured. Insert: a, pre-fabricated methylmethacrylate block; b, two Luhr L-shaped micro-plates; c, Leone jackscrew. (B) Distractor in place during the consolidation period.
X-ray Corporation, Cleveland, USA). Hemimandibles were placed flat on the film with the lingual side facing the film. The settings were as follows: distance 65 cm, voltage 35 kVP, current, 35 mA, and exposure time 25 s. Exposures were made with an epoxy stepwedge (Gammex Inc., WI, USA) that allowed standardisation of radiographic densities into epoxy equivalent thicknesses. These radiographs were scanned (ScanJet 4c/T, Hewlett Packard Co.) and analysed by means of NIH Image software (version 1.62, NIH, USA). Two lengths of the operated mandibles were measured along the molar occlusal plane. The first was measured from the most posterior—superior point of the condyle (i.e. the intersection of the bisector of the angle formed by tangents to the posterior and superior
points of the condyle) to the tip of the incisor lingual alveolar crest (Fig. 1A, solid line bc). The second was measured form the most concave point of the posterior notch to the tip of the incisor lingual alveolar crest with the condyle excluded (Fig. 1A, dashed line dc). In addition, the widths of the distraction gaps were also measured in dorsal (D), middle (M) and ventral (V) locations and the mean of these three measurement was calculated (Fig. 1A). The area of the distraction gap was also measured as described below. Microdensitometry Radiographic density measurements were collected from both operated and unoperated hemimandibles. In order to locate a position on the unoper-
ated hemimandible that was comparable to the operated site, a reference line was drawn along the occlusal plane between the cusp tips of the first and third molars. The image of the unoperated side was flipped horizontally to keep its orientation similar to that of operated side. Both hemimandibular images were then superimposed on the reference lines. The distraction gap and its adjacent regions (at a width of 1 mm) were outlined on the operated side (Fig. 1A). Using this outline, the identical site on the unoperated side was also located. The radiographic density (pixel) measured from the outlined regions was converted into the equivalent bone density using the algorithm generated from the image of an epoxy stepwedge which was put on the each film together with harvested hemimandibles during radiography. The percentage value (percentage bone density), operated side versus unoperated side times 100%, was calculated. Percentages below 100 represented a decreased radiographic bone density on the operated side compared with the unoperated side. To test for both inter- and intra-investigator reliability regarding the measurements, 10 images were randomly selected and measured twice with a 5-day interval by the same two investigators (L.L.W. and Z.J.L.). There were no significant differences between investigators in any measurement using a paired t-test. The method error was 0.28 and 0.45 for measurements of morphology and density, respectively. Statistical analysis Quantitative comparisons of outcome measurements are reported by percentage change. These comparisons include the same outcome measurements from the operated side compared to the control side, as well as comparisons between different distraction groups or post-osteotomy times. The sample distribution of each measurement (the mandibular length and the ratio of bone density of the operated side to the control side) was examined by using histograms. Two-way ANOVA was used to test for distraction rate or time effects. The interaction between these two factors was also examined and found to be insignificant and diagnostics for the ANOVA model were done to check the adequacy of the model (data not shown). Tukey’s method for simultaneous multiple comparisons was applied to examine whether or not measurements in each pair of different subgroups were significantly different at an overall 0.05 type I error level. The relationships between mandibular length versus distraction rate and gap area versus distraction rate were examined using linear regression. Difference levels were set at P < 0:05 or less.
G.J. King et al.
Results General examination Generally, the experimental procedures were well tolerated by rats, and the distraction device was rigid enough to produce and maintain the distraction gap (Fig. 1B). Body weight changes and complication rates have been previously reported.14 There were no wound infections, but localised and encapsulated abscesses were found in some animals during the middle and late consolidation periods. Most of these were in the location of rostral micro-plate with various degrees of osseous hyperplasia. In addition, the bone underlying the methylmethacrylate block showed a tendency to resorb at the later times. The devices were stable throughout the whole experimental period with no visible screw migration, tipping or mobility. By gross observation, the unions were fibrous (i.e. fully bendable) for all 6-day specimens and most of the day 10. For most of day 24 and 38 specimens, the unions were either fibrous/osseous (i.e. partially bendable) or osseous (i.e. rigid). Radiographs of hemimandibles revealed that despite the two segments being separated nearly parallel to each other by a distraction gap, uneven gaps and minor offsets supero-inferiorly were found in a few cases. By 10 post-operative days, regenerate bone could be detected radiographically in the area of the distraction site adjacent to the mandibular foramen. In some samples, it was apparent that the condyle was deformed and its size decreased, especially during late-consolidation period. These effects will be reported elsewhere.
Effect of distraction rate on mandibular morphology The relationships between mandibular morphology and incremental distraction rates were linear and highly significant (P < 0:00005). Mandibular length measured from the condyle increased at a rate of 0.394 mm per distraction increment (i.e. 1:97 0:2 mm ¼ 0:394 mm). It increased at a rate of 0.538 mm per distraction increment (i.e. 2:69 0:2 mm ¼ 0:538 mm) when the measurement excluded the condyle (Fig. 2A and B). Therefore, the actual lengthening of the mandible was much less than the planned distraction amount (i.e. 0:2 mm 5 days ¼ 1 mm). The width and area of the distraction gap increased 0.670 mm and 5.836 mm2, respectively per distraction increment (i.e. 3:35 0:2 mm ¼ 0:670 mm and 29:18 mm 0:2 mm ¼ 5:836 mm2; Fig. 2C and D). It should be mentioned that the actual width of the gap
Mandibular osteodistraction in rats
Figure 2 Comparison of effects of distraction rates on mandibular morphology on the operated side: (A) mandibular length including the condyle; (B) mandibular length excluding the condyle; (C) width of distraction gap; (D) area of the distraction gap. The results of linear regression analyses are shown by the superimposed lines and on each panel. LR: linear regression.
(0.825 mm) in sham distraction was much greater than the predicted value from linear regression (0 mm). This was partially due to the width of the bur used for the osteotomy (tip dimension, 0.5 mm). This feature contributed most to the under-estimation of the gap width and mandibular length. If the data were corrected for this (i.e. subtracting the gap width from mandibular lengths in the sham), the incremental rates in the gap and mandibular lengths, with and without the condyle, were 0.932, 0.653 and 0.799 mm, respectively.
Microdensitometry At the distraction gap, the time points selected for mid-distraction (day 6), and early consolidation (day 10) had significant reductions in bone density compared to a similar site on the unoperated side in all distraction rates (56—68%; P < 0:001), but there were no differences between these two time
points. The times chosen for the mid- (day 24) and late-consolidation (day 38) had slightly increased bone densities compared to the unoperated side (13—25%, P ¼ 0:089 to P < 0:01, respectively, Fig. 3A and B) for sham and slow distraction rates, with no difference between these two consolidation time points (Table 1). There were highly significant differences between the two earlier times and the two consolidation times in all distraction rates (P < 0:001 to P < 0:048; Table 1). The caudal edge of the osteotomies had more dense bone than a similar site on the unoperated side at all time points in all groups (30—93%, P < 0:01 to P < 0:001). However, there were no significant temporal changes in radiographic density at this site. At the rostral edge, the equivalent bone density was also elevated compared to the unoperated side at all time points in all groups, but less than what was found at the caudal edge (28—66%, all,
G.J. King et al.
Figure 3 Comparison of time-course effects on the equivalent bone densities of mandibles at regions of interest in each distraction group: (A) sham group; (B) slow distraction group (0.2 mm per day); (C) moderate distraction group (0.4 mm per day); (D) rapid distraction group (0.6 mm per day). Results of the one-way ANOVA are shown for each region of interest on each panel. See Table 1 for the P-values of multiple comparisons.
P-values of multiple comparisons. Day 6 vs. day 10
Equivalent bone density Sham Gap 1.000 Caudal 0.668 Rostral 0.529
Day 6 vs. day 24
Day 6 vs. day 38
Day 10 vs. day 24
Day 10 vs. day 38
Day 24 vs. day 38
0.000 0.084 0.003
0.000 0.318 0.258
0.000 0.542 0.078
0.000 0.929 0.955
0.995 0.881 0.209
0.2 mm Gap Caudal Rostral
1.000 0.774 0.853
0.005 0.937 0.145
0.001 0.995 0.572
0.007 0.983 0.530
0.001 0.899 0.963
0.902 0.988 0.813
0.4 mm Gap Caudal Rostral
0.127 0.427 0.836
0.000 0.086 0.128
0.000 0.154 0.128
0.048 0.786 0.488
0.001 0.919 0.487
0.327 0.991 1.000
0.6 mm Gap Caudal Rostral
0.819 0.447 0.332
0.000 0.697 0.008
0.000 0.901 0.014
0.000 0.975 0.308
0.000 0.845 0.421
0.983 0.977 0.996
Values in bold indicate significant differences.
Mandibular osteodistraction in rats
F- and P-values of two-way ANOVA. Without interaction
Mandibular length 12.15
Equivalent bone density Gap 34.69 Caudal 3.24 Rostral 7.45
0.0013 0.1195 0.3069
<0.0000 0.0057 <0.0001
14.06 1.64 3.36
0.0015 0.1187 0.2993
<0.0000 0.0059 <0.0001
0.6704 0.7644 0.6740
Values in bold indicate significant effects. Rates: distraction rates; days: post-operative days; R D: interaction term between rates and days.
P < 0:001). At this site, the mid-consolidation time point (day 24) was significantly denser than day 6 (P ¼ 0:003; Table 1). In the greatest distraction group, both mid- and late-consolidation time points were significantly more dense than day 6 (P ¼ 0:008 and P ¼ 0:014, respectively; Table 1), but neither was different from each other.
Combined effects of distraction rates and post-operative days The overall experimental design was a two-way analysis with four distraction rates and four postoperative days as main factors. The two-way ANOVA analysis without interaction demonstrated that there was a strong effect of both rates and days on the mandibular length (P < 0:0001). While time had a strong effect on bone density at all sites, the rate only had a significant effect on the bone density in the gap. The two-way ANOVA analysis with interaction further indicated that these two main factors did not affect each other significantly (P > 0:67) (Table 2).
Discussion Experimental research on mandibular distraction osteogenesis (DO) has been conducted using animal models such as the dog, monkey, pig, sheep, and rabbit. In this study, a recently described rat model for mandibular DO14 was examined for mandibular morphology and bone healing in the regenerate and at the adjacent osteotomy sites as functions of distraction rate and time following osteotomy. Using this model, rats maintain normal activity and gained weight. Furthermore, improved stability of distraction devices, low mortality, and low infection rates allow investigating DO with a sample large enough to protect statistical power. The use of
rodent models for osteodistraction should also facilitate studies on cellular and molecular mechanisms.17,18 The rat models for mandibular DO described to date have been varied. Some use larger devices with extensive dissection, while others use smaller devices and less undermining of tissues.15,16 One model creates an osteotomy between two molars and secures the device to the body of the mandible.16 This region has been the preferred site of device attachment, as opposed to the ramus, because the ramus of the rat mandible is very thin, resulting in fractures and high rates of device dislodgement. However, the inter-molar technique puts teeth in close proximity to the distraction site and increases the risk of infection due to communication with the oral cavity. Our findings were consistent with those from large animal and clinical models. Bone mineralisation during and after limb lengthening procedures in humans has been monitored quantitatively using dual energy X-ray absorptiometry (DEXA).19,20 As lengthening proceeds, the bone density of the gap first falls, reaching minimum values at the time of maximal distraction. This is consistent with the reports of active bone resorption at the cut bone ends.21 Bone density then increases during consolidation, but this seems to be site specific and recovery is only partial in some bones, even after subsequent weight-bearing. Our study suggests that full recovery occurs in the regenerate and cut bone ends following mandibular distraction. In fact, the cut bone ends have a tendency to be denser than unoperated controls. There are reports that lengthened leg bones are as strong as control bones.22 However, stress testing of distracted mandibles report that the average ultimate strength of the regenerate segment remains at 77% of normal.7 This study does not suggest that rapid distraction causes a greater loss of bone density or that this
does not completely recover during the consolidation period. Moderately slow distraction rates have been reported to lead to bone formation with the strongest biomechanical and histological properties, despite having similar bone density readings to more rapidly distracted animals.23 Acute lengthening also seems to lead to an increased risk of fibrous unions compared to gradual distraction.16,24,25 Several studies have examined the effects of other biomechanical modifications to the standard slow distraction rate to lengthen bones. Some have reported equivalent healing in a moderate range of latency periods.13 However, others have found that delayed distraction, compared with immediate, improved the quality of the callus with quicker, denser, and more homogeneous bone formation.26 Alternating compression and lengthening periods during distraction also has no substantial effect on the radiological or histological appearance of the regenerate.27 Excessively slow rates of distraction may not maximally stimulate angiogenesis in the central fibrous zone, whereas higher rates may impair this response. Neovascularization in the central fibrous zone seems to be maximally stimulated by moderate rates of distraction.17 Most research efforts at reducing the time required for osteodistraction have focused on the consolidation period because this represents the longest part of the procedure. Physical modifications including ultrasound28 and electrical stimulation29 have proven effective. TGF-beta 1 treatment has no detectable effect,30 whereas several other substances with osteogenic activities, including IGF-1,31 DMB, Vitamin D332 and FGF-233 seem to have positive influences on consolidation. Implantation of osteogenic cells also seems to promote maturity of the distracted callus.34,35 Cigarette smoking may be of particular concern for adult mandibular distraction patients as it lengthens the consolidation time and decreases the mechanical strength of the regenerate.36 Obviously much more work can be done on reducing the time required for osteodistraction and the quality of bone obtained. This model can facilitate these types of studies. There are no previous reports on the changes in mandibular morphology following mandibular DO. Our data show that the mandibles increased at about 39.4 to 53.8% of the predicted amount (1 mm) at all of the distraction rates examined, and measurement from the posterior notch showed 15% greater increment than that from the condyle (Compare Fig. 2A and B). Significant deformities of the mandibular condyles were evident in some of the distracted mandibles. Furthermore, only 67% of predicted amount was obtained using direct mea-
G.J. King et al.
surements of the width of the distraction gap (Fig. 2C). The following may account for these discrepancies. (1) The larger gap width in the sham group (predicted: 0 mm, actual: 0.872 mm, Fig. 1C) minimised the increments calculated by linear regression. (2) A portion of the mandibular increase may be lost due to muscle traction and subsequent mandibular remodelling (e.g. at the condyle). Clearly posteriorly directed traction to the mandible can stimulate changes in the condyle and glenoid fossa that can lead to mandibular shortening in growing individuals.37 (3) Some devices might have been slack at the time of installation. (4) The 6-day group had smaller final distraction lengths than the other groups (0.6, 1.2 and 1.8 mm, instead of 1— 3 mm). Their inclusion in the linear regressions tend to minimise the measured increments. (5) There were some fragments that were offset supero-inferiorly. (6) The border between the original and regenerate bone in some late specimen was difficult to identify because of remodelling changes. In any case, the current analysis of the morphology of the distracted mandible clearly suggest that negative alteration in the condyle is a complication following mandibular distraction. The condylar changes contribute about 15% to the disparity between measured and predicted mandibular lengths. Another possibility may be that some relapse occurs during consolidation because the gap widths did achieve the predicted values in all three distraction groups (Fig. 2C). It is also noteworthy that if the gap width is corrected for an over-estimation in the sham group due to bur width, an incremental rate of 0.932 mm per distraction rate is achieved. This is very close to what is predicted (1.0 mm) based on the amount of screw advancement done per group. It is noteworthy that the first appearance of bone formation in our study was near the mandibular foramen. Nerves and vessels richly supply this area. An intense vascular response associated with mandibular DO has been reported primarily during the early stages of distraction.18 Callus distraction also leads to moderate degenerative changes in nerves, followed by repair, almost complete recovery, and some nerve fibre growth.38 In conclusion, this report characterises the remodelling responses in a rat model for mandibular distraction osteogenesis. This small animal model behaves similarly to the large animal models with regard to bone density changes in the distracted callus and surrounding bone. This combined with improved distractor stability and lessened animal morbidity will facilitate studies on the molecular and cellular changes associated with mandibular DO and possible biological means to improve the procedure.
Mandibular osteodistraction in rats
Acknowledgements This research was supported by NIDCR grant DE13061. The authors would like to thank StrykerLeibinger for donating surgical plates and screws, Leone for donating the distraction devices, and Takahiro Chino for helping with the surgeries.
References 1. Klein C, Howaldt HP. Lengthening of the hypoplastic mandible by gradual distraction in childhood–—a preliminary report. J Craniomaxillofac Surg 1995;23(2):68—74. 2. McCarthy JG, Schreiber J, Karp N, Thorne CH, Grayson BH. Lengthening the human mandible by gradual distraction. Plast Reconstr Surg 1992;89(1):1—8. 3. Molina F, Ortiz Monasterio F. Mandibular elongation and remodelling by distraction: a farewell to major osteotomies. Plast Reconstr Surg 1995;96(4):825—40. 4. Rachmiel A, Levy M, Laufer D. Lengthening of the mandible by distraction osteogenesis: report of cases. J Oral Maxillofac Surg 1995;53(7):838—46. 5. Aronson J. Experimental and clinical experience with distraction osteogenesis. Cleft Palate Craniofac J 1994; 31(6):473—81. 6. Aronson J, Good B, Stewart C, Harrison B, Harp J. Preliminary studies of mineralization during distraction osteogenesis. Clin Orthop 1990;(250):43—9. 7. Costantino PD, Friedman CD, Shindo ML, Houston G, Sisson Sr GA. Experimental mandibular regrowth by distraction osteogenesis. Long-term results. Arch Otolaryngol HeadNeck Surg 1993;119(5):511—6. 8. Costantino PD, Shybut G, Friedman CD, Pelzer HJ, Masini M, Shindo ML, et al. Segmental mandibular regeneration by distraction osteogenesis. An experimental study. Arch Otolaryngol Head-Neck Surg 1990;116(5):535—45. 9. Karaharju-Suvanto T, Karaharju EO, Ranta R. Mandibular distraction. An experimental study on sheep. J Craniomaxillofac Surg 1990;18(6):280—3. 10. Karp NS, McCarthy JG, Schreiber JS, Sissons HA, Thorne CH. Membranous bone lengthening: a serial histological study. Ann Plast Surg 1992;29(1):2—7. 11. Komuro Y, Takato T, Harii K, Yonemara Y. The histologic analysis of distraction osteogenesis of the mandible in rabbits. Plast Reconstr Surg 1994;94(1):152—9. 12. Snyder CC, Levine GA, Swanson HM, Browne Jr EZ. Mandibular lengthening by gradual distraction. Preliminary report. Plast Reconstr Surg 1973;51(5):506—8. 13. Troulis MJ, Glowacki J, Perrott DH, Kaban LB. Effects of latency and rate on bone formation in a porcine mandibular distraction model. J Oral Maxillofac Surg 2000;58(5): 507—13. 14. Connolly J, Liu Z, Wang L, Whelan M, Huang G, Williams J, et al. A custom mandibular distraction device for the rat. J Craniofac Surg 2002;13(3):445—52. 15. Ignelzi M, Buchman S, Goldstein S, Radu C, Wilensky J, Rosenthal A. A rat model of mandibular distraction osteogenesis. In: McNamara JA, editor. Craniofacial Growth Series. Ann Arbor, MI: Center for Human Growth and Development; 1998. p. 141—152. 16. Rowe NM, Mehrara BJ, Dudziak ME, Steinbreck DS, Mackool RJ, Gittes GK, et al. Rat mandibular distraction osteogenesis. Part I. Histologic and radiographic analysis. Plast Reconstr Surg 1998;102(6):2022—32.
17. Eingartner C, Coerper S, Fritz J, Gaissmaier C, Koveker G, Weise K. Growth factors in distraction osteogenesis. Immuno-histological pattern of TGF-beta1 and IGF-I in human callus induced by distraction osteogenesis. Int Orthop 1999;23(5):253—9. 18. Mehrara BJ, Rowe NM, Steinbrech DS, Dudziak ME, Saadeh PB, McCarthy JG, et al. Rat mandibular distraction osteogenesis. II. Molecular analysis of transforming growth factor beta-1 and osteocalcin gene expression. Plast Reconstr Surg 1999;103(2):536—47. 19. Eyres KS, Bell MJ, Kanis JA. Methods of assessing new bone formation during limb lengthening. Ultrasonography, dual energy X-ray absorptiometry and radiography compared. J Bone Joint Surg Br 1993;75(3):358—64. 20. Reiter A, Sabo D, Pfeil J, Cotta H. Quantitative assessment of callus distraction using dual energy X-ray absorptiometry. Int Orthop 1997;21(1):35—40. 21. Welch R, Birch J, Makarov M, Samchukov M. Histomorphometry of distraction osteogenesis in a caprine tibial lengthening model. J Bone Miner Res 1998;13(1):1—9. 22. Schickendantz MS, Watson JT, Sferra JJ, Kambic HE. A model for evaluating the strength of bones lengthened by distraction osteogenesis. Clin Orthop 1992;(275):248— 52. 23. Farhadieh RD, Gianoutsos MP, Dickinson R, Walsh WR. Effect of distraction rate on biomechanical, mineralization, and histologic properties of an ovine mandible model. Plast Reconstr Surg 2000;105(3):889—95. 24. Paccione MF, Mehrara BJ, Warren SM, Greenwald JA, Spector JA, Luchs JS, et al. Rat mandibular distraction osteogenesis: latency, rate, and rhythm determine the adaptive response. J Craniofac Surg 2001;12(2):175—82. 25. Warren SM, Mehrara BJ, Steinbrech DS, Paccione MF, Greenwald JA, Spector JA, et al. Rat mandibular distraction osteogenesis. Part III. Gradual distraction versus acute lengthening. Plast Reconstr Surg 2001;107(2):441—53. 26. Gil-Albarova J, de Pablos J, Franzeb M, Canadell J. Delayed distraction in bone lengthening. Improved healing in lambs. Acta Orthop Scand 1992;63(6):604—6. 27. Greenwald JA, Luchs JS, Mehrara BJ, Spector JA, Mackool RJ, McCarthy JG, et al. ‘‘Pumping the regenerate’’: an evaluation of oscillating distraction osteogenesis in the rodent mandible. Ann Plast Surg 2000;44(5):516—21. 28. Shimazaki A, Inui K, Azuma Y, Nishimura N, Yamano Y. Lowintensity pulsed ultrasound accelerates bone maturation in distraction osteogenesis in rabbits. J Bone Joint Surg Br 2000;82(7):1077—82. 29. Hagiwara T, Bell WH. Effect of electrical stimulation on mandibular distraction osteogenesis. J Craniomaxillofac Surg 2000;28(1):12—9. 30. Rauch F, Lauzier D, Travers R, Glorieux F, Hamdy R. Effects of locally applied transforming growth factor-beta1 on distraction osteogenesis in a rabbit limb-lengthening model. Bone 2000;26(6):619—24. 31. Stewart KJ, Weyand B, van’t Hof RJ, White SA, Lvoff GO, Maffulli N, et al. A quantitative analysis of the effect of insulin-like growth factor-1 infusion during mandibular distraction osteogenesis in rabbits. Br J Plast Surg 1999; 52(5):343—50. 32. Hagino T, Hamada Y. Accelerating bone formation and earlier healing after using demineralized bone matrix for limb lengthening in rabbits. J Orthop Res 1999;17(2): 232—7. 33. Okazaki H, Kurokawa T, Nakamura K, Matsushita T, Mamada K, Kawaguchi H. Stimulation of bone formation by recombinant fibroblast growth factor-2 in callotasis bone lengthening of rabbits. Calcif Tissue Int 1999;64(6):542—6.
34. Takushima A, Kitano Y, Harii K. Osteogenic potential of cultured periosteal cells in a distracted bone gap in rabbits. J Surg Res 1998;78(1):68—77. 35. Tsubota S, Tsuchiya H, Shinokawa Y, Tomita K, Minato H. Transplantation of osteoblast-like cells to the distracted callus in rabbits. J Bone Joint Surg Br 1999;81(1):125—9. 36. Ueng SW, Lin SS, Wang CR, Liu SJ, Tai CL, Shih CH. Bone healing of tibial lengthening is delayed by cigarette
G.J. King et al.
smoking: study of bone mineral density and torsional strength on rabbits. J Trauma 1999;46(1):110—5. 37. Wendell PD, Nanda R, Sakamoto T, Nakamura S. The effects of chin cup therapy on the mandible: a longitudinal study. Am J Orthod 1985;87(4):265—74. 38. Fink B, Neuen-Jacob E, Lehmann J, Francke A, Ruther W. Changes in canine peripheral nerves during experimental callus distraction. Clin Orthop 2000;376:252—67.