Effect of Distraction Rate on Biomechanical, Mineralization, and Histologic Properties of an Ovine Mandible Model Ross D. Farhadieh, B.Sc.(Med.), Mark P. Gianoutsos, M.B.B.S., M.D., F.R.A.C.S., R. Dickinson, M.B.B.S., F.R.A.C.S., and William R. Walsh, B.Sc., Ph.D. Sydney, Australia

Craniofacial microsomia is a common congenital malformation. Ilizarov’s method of distraction osteogenesis applied to the mandible has yielded promising results both experimentally and clinically. Because the technique is used predominantly in a pediatric population, length of treatment and compliance may be problematic. To date, the limits of distraction rate in the craniofacial skeleton have not been defined. This study was designed to investigate the effects of distraction rate, in a large animal model, on the mineralization, biomechanical, and histologic properties of lengthened mandibles. Clinically faster distraction rates would decrease the overall treatment time. Twenty-four animals were divided into four groups, with varying rates of distraction (1, 2, 3, and 4 mm/day). A uniaxial distractor at the angle of the mandible was used. The mandibles were lengthened to 24 mm and fixed for a period of 5 weeks, when the animals were killed. The specimens were analyzed with respect to mineralization using dual energy x-ray absorptiometry, biomechanical strength, through a modified three-point bending test, and histologic properties with hematoxylin and eosin stains. Biomechanical, mineralization, and histologic analyses of the samples indicated that group 1 (1 mm/day) samples were significantly superior (p ! 0.05) to those of group 4 (4 mm/day). Although bone formation was achieved in all groups, group 1 (1 mm/day) demonstrated the strongest biomechanical and histologic properties. Bone mineral density obtained using dual energy x-ray absorptiometry may be clinically useful as a reliable, noninvasive, and relatively cheap predictor for removal time of the fixator. (Plast. Reconstr. Surg. 105: 889, 2000.)

ments by Ilizarov in the 1950s.2,3 Snyder et al.4 first applied distraction to the mandible in a dog model in 1973. Clinical use of distraction osteogenesis has only recently been applied to the membranous craniofacial bones,5 and interest in the technique has since grown significantly.6 –13 In contrast to distraction in the long bones, the limitations of distraction rate in the mandible have not been reported yet. In addition, no definitive guidelines for the removal of the fixation frame are available. A noninvasive assessment of bone mineral density, dual energy x-ray absorptiometry, may be a useful parameter to assess the quality of healing. The aims of this study were twofold: first, to establish the morphologic viability of faster distraction rates in the mandible and to evaluate the mineralization and biomechanical properties; second, to assess the feasibility of dual energy x-ray absorptiometry as a clinical predictor for removal of the external fixator during the fixation period. MATERIALS

AND

METHODS

Twenty-four 12-month-old sheep ("45 kg) were used in the study. Animals were randomly divided into four groups. Distraction rates of 1, 2, 3, and 4 mm/day were used, whereas all other parameters of distraction length and fixation were kept constant (Table I). The surgical procedure on the mandible began with external infiltration of the skin, masseter, and periosteum overlying the angle of

The first recorded case of bone elongation dates back to 1905 in the femur, by Codivilla of Italy.1 Clinical use of the technique was limited by a high complication rate until the develop-

From the Division of Surgery, Orthopaedic Research Laboratories, Prince of Wales Hospital, University of New South Wales, and the Division of Surgery at the Sydney Children’s Hospital. Received for publication December 1, 1998; revised July 6, 1999.

889

890

PLASTIC AND RECONSTRUCTIVE SURGERY,

TABLE I Study Design

Group

No.

1 2 3 4

6 6 6 6

Distraction Rate

1 2 3 4

mm/day mm/day mm/day mm/day

o.d. b.d. t.d. t.d.

Distraction Gap (mm)

Fixation Period (weeks)

24 24 24 24

5 5 5 5

the right side of the mandible with a solution of 0.25% bupivacaine with 1:400,000 adrenaline. An external incision over the angle of the mandible was made. The masseter was reflected posteriorly in a subperiosteal plane. The osteotomy was sufficiently posterior to the last mandibular molar tooth to enable two 2-mm-threaded half pins to be placed 4 mm apart and 10 mm from the osteotomy site itself. Two bicortical 1.5-mm drill holes were made on either side of the proposed osteotomy site at the distances indicated. These holes were collinear and perpendicular to the proposed osteotomy site, standardized through our jig with respect to the inclination angle on the body of the mandible. The osteotomy was made using a nitrogen gas-powered oscillating saw for the posterior half and a burr drill for the anterior half. The osteotomy was completed using two 5-mm osteotomes placed at either end of the osteotomy gap and twisted until the lingual surface of the mandible—not completely cut through by the burr drill— fractured. The pins were inserted through incisions made in the skin. The wound was closed in two layers with 3-0 Vicryl to the platysma and 3-0 nylon to the skin, and the uniaxial distractor (Howmedica Ltd., Rutherford, N.J.) was put in place. After a 7-day latency period, distraction (as outlined in Table I) was commenced. The pins were monitored for sepsis and loosening throughout the distraction and the fixation periods. After distraction of 24 mm was achieved, the distractor was removed and replaced by a modified mini-Hoffman frame. Fixation was maintained for a period of 5 weeks. The animals were killed with an overdose of sodium pentabarbitone administered intravenously. The mandibles were dissected out and analyzed for bone mineral density and biomechanical and histologic properties. Bone mineral densities were noninvasively assessed using two commercially available dual energy x-ray absorptiometry (DEXA) systems

March 2000

to correlate with the mechanical data and to determine whether any differences existed between the two systems. A fan beam (LunarExpert) and a pencil beam (Lunar-DPXL) dual energy x-ray absorptiometry were used to assess the bone mineral density (g/cm2) of regions of interest, which were defined in two separate ways. Region “A” was defined as the region between the site of insertion of the inner most 2-mm half pins, whereas region “B” was measured to correspond precisely with the distraction length (Fig. 1). Bone mineral density values were measured on the control and the lengthened sides. The harvested mandibles were placed in the DEXA scanners underneath a soft-tissue phantom solution of potassium chloride (Figs. 2 and 3). The Lunar DPXL was set on small animal software 1.0 C, high-resolution mode, tube current at 150 mA, and sampling at 0.6 # 1.2 ml; the scan width was set at 90 mm. The Expert system used Lunar-Expert software (version 1.70) and was set up for a right-hand scan, the tube current at 1 mA, fast mode, and scanning length and width at 23.0 # 14.4 cm. An analysis of variance test was applied to compare the bone mineral density values obtained from the

FIG. 1. Diagram demonstrating regions “A” and “B” measured in the harvested mandibles using DEXA systems.

Vol. 105, No. 3 /

891

OVINE MANDIBLE MODEL

FIG. 2. Bone mineralization density studies of harvested hemimandble using a DPXL (pencil beam) with phantom solution in place.

bles with the mechanical data were performed within each group. Biomechanical properties of the specimen were examined using a modified three-point bending test.14 A material testing system Mini Bionix 858 machine was used to measure the following parameters: the stiffness of the specimen (N/mm), the failure load (N), energy of failure (Nmm), and the length of displacement at point of break (mm). The biomechanical parameters of the lengthened mandibles from the four groups were compared using an analysis of variance followed by a Duncan’s post hoc multiple comparison when appropriate. Sections from the hemimandibles were evaluated for routine morphologic properties under light microscopy applying Harris’s hematoxylin and eosin. RESULTS

The bone mineral density values obtained with the DPXL system (pencil beam) were greater and significantly different from those in the Expert system (fan beam). A comparison of these values in different groups and areas (Tables II and III) demonstrated distraction at 4 mm per day; group 4 was inferior to all others (p ! 0.05). The biomechanical data are summarized in Table IV. The mechanical properties of the control sides were greater than those of the lengthened sides in all groups. In comparing the groups’ mechanical properties, the mandibles distracted at 1 mm per day (group 1) were superior (p ! 0.05) to all other groups, being distracted at faster rates. Although the mandibles distracted at 2 and 3 mm per day (groups 2 and 3) did not differ significantly, they were both considerably (p ! 0.05) superior to those distracted at 4 mm per day (group 4). A scatter plot between ultimate load and the two DEXA machines (DPXL and Expert) and two regions (A and B) of the mandible meaTABLE II Bone Mineral Density Values of the Lengthened Mandible in Area A FIG. 3. Bone mineralization density studies of harvested hemimandble using an Expert (fan beam) with phantom solution in place.

two different machines and different distraction rates. Pearson’s correlations of the DEXA data from the lengthened and control mandi-

Groups

Pencil Beam DPXL Bone Mineral Density (g/cm2)

1 2 3 4

0.579 0.531 0.531 0.460

SD

Fan Beam Expert Bone Mineral Density (g/cm2)

SD

0.023 0.047 0.081 0.033

0.376 0.330 0.321 0.299

0.024 0.042 0.072 0.065

892

PLASTIC AND RECONSTRUCTIVE SURGERY,

TABLE III Bone Mineral Density Values of the Lengthened Mandible in Area B

and (3) determine if a noninvasive assessment of bone mineral density could serve as a predictor of biomechanical strength for the newly formed bone. Dual energy x-ray absorptiometry data have long been used as a clinical tool to assess bone density in postmenopausal osteoporosis as a predictor of fracture risk16 and, more recently, as a method to determine periprosthetic bone loss17 and to monitor the healing of long bone.18 The method also allows for precise measurements of bone mineral in other regions of interest of the skeleton.19 The basic mechanism of the technique provides two distinct peaks of energy supplied by the x-ray tubes coupled to a sensitive detector providing the bone mineral content and density.20 A study comparing the relative accuracy of the fan beam and a second-generation pencil beam system in the lower limb demonstrated the latter to give a more accurate reading.21 Our study demonstrated and confirmed this difference in a distracted mandible model. Fan beam scanners operate at much faster rates than the pencil beam scanners, with the advantage of practicality in clinical practice and with the drawback of increased radiation dose. Other noninvasive imaging modalities, such as ultrasound, may be useful to predict the properties of the distracted zone.16 In the long bones, where distraction was first applied, the optimal rate of distraction has been established at 1 mm/day in four equal increments of 0.25 mm.3 Ilizarov3 suggested that the medullary and periosteal blood flow were essential for bone formation in the lengthened region. It has since been shown that although the medullary blood flow may contribute to the process of bone formation, it is not vital to its success.22–24 In fact, the periosteal blood flow has been deemed the paramount influence in the success of bone regeneration in the lengthened region.24 In a series conducted to examine the feasibility of faster distraction rates in the tibia, animals distracted at 2.8 mm/day demonstrated a reduction of bone formation and malunions.25 In the same series, microangiographic studies revealed compromised periosteal blood flow across the distraction gap in that group. The intramembranous bones of the craniofacial skeleton afford a more generous blood flow. Morphologic examination of all of the lengthened sections obtained in our series demonstrated bone formation. However, the biomechanical, mineral-

Groups

Pencil Beam DPXL Bone Mineral Density (g/cm2)

1 2 3 4

0.512 0.533 0.531 0.434

SD

Fan Beam Expert Bone Mineral Density (g/cm2)

SD

0.017 0.068 0.099 0.026

0.332 0.304 0.333 0.264

0.029 0.029 0.080 0.042

TABLE IV Biomechanical Properties of the Lengthened Mandibles

Groups

Failure Load (N)

SD

Energy of Failure (N)

SD

1 2 3 4

689.28 505.90 472.97 384.5

120.79 59.36 176.16 83.08

3080.1 1819.2 1818.8 2088.8

1552.60 544.44 789.37 574.61

sured is presented in Figure 4 for the control (above) and lengthened (below) sides for group 1. Light microscopic examination of the hematoxylin and eosin sections demonstrated new bone formation in all of the lengthened sections. Cartilaginous tissue was not observed in any of the sections. Samples distracted at 1 mm/day (group 1) revealed more woven bone in the distraction site compared with other groups. Increasing the rate of distraction was associated with more disorganization in the bone matrix (Fig. 5). Immunohistochemical studies of the sections for different growth factors are reported elsewhere.15 DISCUSSION

Distraction osteogenesis in the mandible is primarily applied to a pediatric population, where compliance may be an issue. A reduction in the overall treatment time would go a long way in alleviating this problem. Distraction consists of three phases: latency, distraction, and fixation periods. A reduction in any of these phases would benefit the patients. The present study is limited in that a reduction in the distraction phase was examined. Biomechanical measurements, mineralization, and histology were used to (1) determine whether any differences existed between pencil beam (DPXL) and fan beam (Expert DEXA) scanners; (2) compare the biomechanical properties, bone mineral density, and histology of the different groups distracted at a variable rate;

March 2000

Vol. 105, No. 3 /

893

OVINE MANDIBLE MODEL

FIG. 4. Scatter plots of ultimate load versus bone mineral density obtained in regions A and B using the Lunar DPXL or Lunar Expert DEXA systems for the lengthened side (above) and the control side (below). These data reveal a similar trend for both regions with either machine. The values are different for the two DEXA systems.

ization, and histologic properties of the slowest distracted group (1 mm/day) yielded the most favorable results. Biomechanical properties, bone mineral density, and histology of the specimens confirmed distraction at 1 mm/day (group 1) to be superior to all others. In comparing the bone mineral density values of the lengthened specimen across the groups in both areas A and B, only mandibles distracted at 4 mm/day (group 4) demonstrated a significant difference to all other groups (p ! 0.05). Groups 1 through 3 did not reveal any significant differ-

ence when compared. The mean failure load value, the most reliable indicator of bone strength, for group I (distracted at 1 mm/day) was measured at 689.28 N and was recorded to be the lowest in specimen from group 4 (distracted at 4 mm/day) at 384.5 N. A significant difference was established between mandibles distracted at 1 mm/day (group 1) and all other specimen (p ! 0.05). Morphologic comparison of the regenerate regions in the different groups demonstrated a more mature bone formation in group 1. There was a gradual shift to immature tissue composition in the faster dis-

894

PLASTIC AND RECONSTRUCTIVE SURGERY,

March 2000

FIG. 5. (Above, left) Hematoxylin and eosin section of mandible distracted at 1 mm/day (76.5#). A distraction rate of 1 mm/day produced mature, thick regenerate trabeculae in the center of the distraction site. Haversian systems are shown (an indication of remodeling is noted). (Above, right) Hematoxylin and eosin section of mandible distracted at 2 mm/day (76.5#). A distraction rate of 2 mm/day is less organized compared with 1 mm/day; however, Haversian remodeling is still noted. Thick trabeculae are still noted, but the bone is less mature than with the distraction rate at 1 mm/day. (Below, left) Hematoxylin and eosin section of mandible distracted at 3 mm/day (76.5#). A distraction rate of 3 mm/day reveals a further decrease in maturity of the regenerative bone with thinner trabeculae. (Below, right) Hematoxylin and eosin section of mandible distracted at 4 mm/day (76.5#). A distraction rate of 4 mm/day revealed the most immature bone in the regenerative region with a lack of unity among the trabeculae, which may, in part, have contributed to the decrease in the mechanical properties.

tracted groups, indicating more extensive remodeling of the newly formed bone. Bone formation was noted in all groups through intramembranous ossification pathway. No cartilaginous tissue was noted, which confirms the stability of the bone fragments during the course of distraction and fixation.2 This evidence supports a more stable distraction at the slower rates of 1 mm/day. During the period of fixation, conventional wisdom for removal of the fixation frame has been x-ray evidence of bone formation in the lengthened region.3,26 This has led to some refracture in the regenerate bone, particularly in the weight-bearing long bones.27 In a recent study, single photon absorptiometry, quantitative computed tomography, and dual emission x-ray absorptiometry were correlated to the torsional properties of the canine tibia, lengthened during healing. Dual emission x-ray absorptiometry was found to correlate well with

the torsional biomechanical properties of the healing tibia. However, both of the other methods showed a better correlation. Furthermore, clinical bone mineral density evaluations of the proximal neck of the femur and lumbar spine have been correlated to the risk of developing fractures.16,20 Clinically, dual emission x-ray absorptiometry could be applied as a predictor to determine the most appropriate time for frame removal, possibly reduce treatment time, and decrease risk of refracture of the lengthened mandible. CONCLUSIONS

Although morphologic viability of bone regeneration was demonstrated across all groups, the biomechanical and mineralization properties of the slowest distracted group (distracted at 1 mm/day) proved to be biomechanically superior. Clinical application of faster distraction rates may become viable using bioactive

Vol. 105, No. 3 /

OVINE MANDIBLE MODEL

supplements. Developing the optimal clinical protocol, however, requires further studies into the latency and fixation periods. The clinical use of DEXA scanners would prove an effective, relatively cheap method for assessment of bone formation in the lengthened region. It could be used as an accurate and reliable predictor for removal of the external frame during the fixation phase. Clinically, application of the fan beam scanner, where available, would be more time efficient. William R. Walsh Level 2 North, Edmund Blackett Building Prince of Wales Hospital Division of Surgery High Street, Randwick 2031 Sydney, Australia [email protected] REFERENCES 1. Codivillla, A. On the means of lengthening, in the lower limbs, the muscles and tissue which are shortened through deformity. Am. J. Orthop. Surg. 2: 353, 1905. 2. Ilizarov, G. A. The tension-stress effect on the genesis and growth of tissues: Part I. The influence of stability of fixation and soft-tissue preservation. Clin. Orthop. 238: 249, 1989. 3. Ilizarov, G. A. The tension-stress effect on the genesis and growth of tissues: Part II. The influence of the rate and frequency of distraction. Clin. Orthop. 239: 263, 1989. 4. Snyder, C. C., Levine, G. A., Swanson, H. M., and Browne, E. Z. Mandibular lengthening by gradual distraction. Plast. Reconstr. Surg. 51: 506, 1973. 5. McCarthy, J. G., Schreiber, J., Karp, N., Thorne, C. H., and Grayson, B. H. Lengthening the human mandible by gradual distraction. Plast. Reconstr. Surg. 89: 1, 1992. 6. Aronson, J. Experimental and clinical experience with distraction osteogenesis. Cleft Palate Craniofac. J. 31: 473, 1994. 7. Costantino, P. D., Friedman, C. D., Shindo, M. L., et al. Experimental mandibular regrowth by distraction osteogenesis: Long-term results. Arch. Otolaryngol. Head Neck Surg. 119: 511, 1993. 8. Costantino, P. D., and Friedman, C. D. Distraction osteogenesis: Applications for mandibular regrowth. Otolaryngol. Clin. North Am. 24: 1433, 1991. 9. Costantino, P. D., Shybut, G., Friedman, C. D., et al. Segmental mandibular regeneration by distraction osteogenesis: An experimental study. Arch. Otolaryngol. Head Neck Surg. 166: 535, 1990. 10. Holbein, O., Neidlinger-Wilke, C., Suger, G., et al. Ilizarov callus distraction produces systemic bone cell mitogens. J. Orthop. Res. 13: 629, 1995.

895 11. Karaharju, E. O., Alto, K. A., Kahri, A., et al. Distraction bone healing. Clin. Orthop. 297: 38, 1993. 12. Karaharju-Suvanto, T., Karaharju, E. O., and Ranta, R. Mandibular distraction: An experimental study on sheep. J. Craniomaxillofac. Surg. 18: 280, 1993. 13. Karp, N. S., McCarthy, J. G., Schreiber, J. S., Sissons, H. A., and Thorne, C. H. Membranous bone lengthening: A serial histological study. Ann. Plast. Surg. 29: 2, 1992. 14. Elovic, R. P., Hipp, J. A., and Hayes, W. C. A method for measuring the structural properties of the rat mandible. Arch. Oral Biol. 39: 1029, 1994. 15. Farhadieh, R. D., Yu, Y., Dickinson, R., Gianoutsos, M. P., and Walsh, W. R. The role of transforming growth factor-!, insulin-like growth factor-1 and basic fibroblast growth factor in distraction osteogenesis of the mandible. J. Craniofac. Surg. 10: 80, 1999. 16. Rizzoli, R., Slosman, D., and Bonjour, J. P. The role of dual energy x-ray absorptiometry of lumbar spine and proximal femur in the diagnosis and follow-up of osteoporosis. Am. J. Med. 98: 33, 1995. 17. Wixson, R. L., Stulberg, S. D., Van Flandern, G. J., and Puri, L. Maintenance of the proximal bone mass with an uncemented femoral stem analysis with dualemission x-ray absorptiometry. J. Arthroplasty 12: 365, 1997. 18. Petersen, M. M., Gehrchen, P. M., Nielsen, P. K., and Lund, B. Loss of bone mineral of the hip assessed by DEXA following tibial shaft fractures. Bone 20: 491, 1997. 19. Sievanen, H., Kannus, P., Oja, P., and Vuori, I. Precision of dual-energy x-ray absorptiometry in the upper extremities. Bone Miner. 20: 235,1993. 20. Mazess, R. B. Bone densiometry using dual energy x-ray absorptiometry. Curr. Opin. Orthop. 7: 5, 1996. 21. Blake, G. M., Parker, J. C., Buxton, F. M., and Fogelman, I. Dual x-ray absorptiometry: A comparison between fan beam and pencil beam scans. Br. J. Radiol. 66: 902, 1993. 22. Delloye, C., Delefortrie, G., Coutelier, L., and Vincent, A. Bone regenerate formation in cortical bone during distraction lengthening: An experimental study. Clin. Orthop. 250: 34, 1990. 23. Hyodo, A., Kotschi, H., Kambic, H., and Muschler, G. Bone transport using intramedullary fixation and a single flexible traction cable. Clin. Orthop. 325: 256, 1996. 24. Kojimoto, H., Yasui, N., Goto, T., Matsuda, S., and Shimomura, Y. Bone lengthening in rabbits by callus distraction: The role of periosteum and endosteum. J. Bone Joint Surg. Br. 70: 543, 1988. 25. Yasui, N., Kojimoto, H., Sasaki, K., et al. Factors affecting callus distraction in limb lengthening. Clin. Orthop. 293: 55, 1993. 26. Hughes, T. H., Maffuli, N., Green, V., and Fixsen, J. A. Imaging in bone lengthening: A review. Clin. Orthop. 308: 50, 1994. 27. Dahl, M. T., Gulli, B., and Berg, T. Complications of limb lengthening. Clin. Orthop. 301: 10, 1994.