Abstract
BACKGROUND: Half-ring sulcated external fixator is a bone external fixation device. Its structure is relatively simple, with multi-plane fixed, and small occupying space. So far, there are few experimental reports about construction of tibial defect model by it in large animals.
OBJECTIVE: To validate the practicality and repeatability of half-ring sulcated external fixator in establishing tibial defect models in goats.
DESIGN, TIME AND SETTING: The randomized controlled observation was performed at the Animal Laboratory of the Third Military Medical University from March 2005 to February 2007.
MATERIALS: Nine healthy adult (Animal Center of Third Military Medical University); Half-ring sulcated external fixator, made by Professor Li of the Third Military Medical University; Osseous pin, φ2.5 mm, Shanghai Medical Instruments Co., Ltd.
METHODS: After animals were anesthetized, two needles were threaded in the same plane of cancellous bone about 1.5-2.0 cm above the superior articular surface of the tibia, named group 1. The angle between the two needles was 40°. One needle was threaded at the place of 3.0 to 5.0 cm under the group 1 and parallel with optional needle of the group 1, named group 2. Needles of group 4 were threaded at the cancellous bone that was 2.0 cm from the joint surface of inferior extremity of the tibia. One needle was threaded at the place of 3.0 to 5.0 cm from group 4 and parallel with optional needle of the group 4. The included angle between these two needles of group 2 and group 3 was 60°. There were totally 6 needles. Tibia was fixed by half-ring sulcated external fixator. According to the length of the tibia from X-ray, tibia and periosteum were ampu-tated by wire saw between the two needles in the middle and inferior segment of the tibia to make segmental bone and periosteal defects of 20 percent of tibia.
MAIN OUTCOME MEASURES: ①Postoperative general state of the animals and effect of half-ring sulcated external fixator. ②Self-repair of bone defects by X-ray and Lane-Sandhu scoring standards.
RESULTS: Of the 9 selected goats, 1 died of postoperative infection, and 1 developed pinhole infection and supplemented. All the others survived and were included in final analysis. The goats were awake within 2-6 hours after surgery and able to stand up to eat. The loosening nuts were screwed up. No external fixation failure or loosening was found. No skin was necrotic. Injured limb could touch the ground in 2 days, but could not bear load until 2 weeks. The goats could walk in three weeks with a slight limp, then moved freely without claudication in four weeks after the operation. At the 5th week after operation, the radiographs showed little periosteal reaction. At the 10th week after operation, sclerosis was found in the fracture ends, and medullary cavity started to close. Fifteen weeks after operation, the medullary cavity closed and periosteal reaction did not aggravate. In addition, Lane score was 0 at any time point. No bone formation was found in the histological examination at 5, 10 and 15 weeks after operation, and Lane score was 0 at any time point.
CONCLUSION: The goat model established in goat tibia with 20% defects by the half-ring sulcated external fixator shows no bone healing after 15 weeks by X-ray or histological exanimation, indicating this is a practical and repetitive method to establish animal models of large segmental defect for bone tissue engineering.

INTRODUCTION

Application of tissue engineering technique in repairing long bone defect is developing. So far, it is still extremely difficult to repair long bone de-fect > 30 mm in clinic by tissue engineering ap-proach[1-3]. Therefore, to find a suitable tissue en-gineering method to repair large defects of long bone in animal models becomes an important as-pect in orthopedic surgery. In sheep, pigs, dogs and other large animals, the models of large segmental defects have been made in long bone, and treated by implanting composites of biological material and seed cells. The half ring sulcated external fixator has achieved good clinical effect in the treatment of bone defects, bone lengthening, bone nonunion and osteomyelitis. The aim of this study was to construct a tibial defect model by using half-ring sulcated external fixator and to observe the effectiveness.

MATERIALS AND METHODS

Materials
Animals
The experiment was performed at the Animal Labo-ratory of Third Military Medical University from March 2005 to February 2007. Nine goats, 6 to 12 months old, weighing 20-25 kg, irrespective of gen-der, were provided by the Animal Center of Third Military Medical University (No. SCXK2007-0004). The goats were fed separately, and the experimental procedure accorded with the Animal Ethical Stan-dards. The animals were randomly divided into 3 groups (n =3). Segmental bone defects of 20 percent (the length of lower tibia of osteotomy/total length of the tibia ×100%) were made in unilateral tibia of each goat. One animal in each group was killed re-spectively at the 5th, 10th, and 15th weeks postopera-tively for radiographic and histological examination.

Major equipment and reagents
Half-ring sulcated external fixator, made by Professor Li of the Third Military Medical University; Osseous pin, φ2.5 mm, Shanghai Medical Instruments Co., Ltd.

Methods
Preoperative preparation
Radiographs of bilateral tibia were taken before operation by measuring tibia shaft lengths and diameters of the middle segment using two aluminum rulers placing in transverse lon-gitudinal side of the tibia shaft. Tibial diameter was measured directly after operation.

Operation method
All the preparation of animal models of bone defect and ex-ternal fixation were made with the same procedure and com-pleted by the same group of researchers. Animals were anes-thetized by Sumianxin (0.08 mL/kg) and ketamine (2 mg/kg). During the operation, we supplied goat with ketamine continuously, and performed skin preparation, sterilization conventionally, and spreading of aseptic towel. Two needles were threaded in the same plane of cancellous bone about 1.5-2.0 cm above the superior articular surface of the tibia, named group 1. The angle between the two needles was 40° [4-6]. One needle was threaded at the place of 3.0 to 5.0 cm under the group 1 and parallel with optional needle of the group 1, named group 2. Needles of group 4 were threaded at the cancellous bone that was 2.0 cm from the joint surface of inferior extremity of the tibia. One needle was threaded at the place of 3.0 to 5.0 cm from group 4 and parallel with optional needle of the group 4. The included angle between these two needles of group 2 and group 3 was 60°. There were totally 6 needles. Tibia was fixed by half-ring sulcated external fixator, and all screw nuts were rigidified. Skin at anterolateral surface of tibia was cut. Blunt dissection was conducted to separate tibial muscle and tibial space. The anterior tibia vessels were protected. Then middle and inferior segment of the tibia was liberated. The periosteum was separated from the soft tissue but retained the surrounding tissue. According to the length of the tibia from X-ray, tibia and periosteum were amputated by wire saw between the two needles in the middle and inferior segment of the tibia to make segmental bone and periosteal defects of 20 percent of tibia. Then the tibia was closed layer by layer. The diameter of the sawed middle and inferior seg-ment of the tibia was measured. The nut was checked every-day to identify the tightness. According to the X-ray, we regu-larly observed the paratope and alignment conditions of the tibia at two ends of the defects and adjusted timely. Post-operation conditions were described in Figure 1. Goats were injected with penicillin 1.6 million U per day for one week. Animals were fed with conventional high-protein diet and allowed to live freely if no abnormity occurred.

Radiographic examination
After operation, lateral X-ray films of treated limbs were taken at the 5th, 10th, and 15th weeks. Animals were anesthe-tized by Sumianxin (0.08 mL/kg), fixed and shot under con-ventional conditions. All the X-ray films were evaluated ac-cording to Lane-Sandhu X-ray scoring standards of bone de-fects [7-9] (0-12 scores, high score represents strong self-repair ability) and were used for the quantitative analysis of self-repair ability of bone defects.

Histological examination
Histological examination was performed at postoperative 5th, 10th and 15th weeks. Bone defect specimens were harvested, fixed by 10% neutral formalin (v/v), decalcified by EDTA, dehydrated by alcohol by degrees, cleared by xylene and stained by HE. Bone formation of bone defects was observed according to Lane-Sandhu histological scoring standards [7-9].

RESULTS

Quantitative analysis of animals
Of the 9 selected goats, 1 died of postoperative infection, and 1 developed pinhole infection and supplemented. All the oth-ers were included in final analysis.

Tibial length, diameter and condition after surgery
There were 18 affected limbs. The lengths and diameters of the tibia were (138 ± 24) mm and (12 ± 1.8) mm. The goats were awake within 2-6 hours after surgery and able to stand up to eat. The loosening nuts were screwed up. No external fixation failure or loosening was found. No skin was necrotic. Injured limb could touch the ground in 2 days, but could not bear load until 2 weeks. The goats could walk in three weeks with a slight limp, then moved freely without claudication in four weeks after the operation.

Radiographic examination and Lane scoring
On the first day after operation, the radiographs showed flat and concinnous distal and proximal fracture ends (Figure 1), and all Lane scores were 0. At the 5th week after operation, bone density increase was found in bone defects of all animals, and distal and proximal fracture ends were fuzzy accompanied by a little periosteal reaction (Figure 2). In addition, all Lane scores were 0. At the 10th week after operation, no bone formed. Sclerosis was found in the fracture ends; Medullary cavity started to close, and no density increase in the bone defects appeared (Figure 3); Lane score was 0. At the 15th week after operation, density of bone defects did not increase, and fracture end density did not increase significantly com-pared with the 5th and 10th week, but medullary cavity closed and periosteal reaction did not aggravate (Figure 4). In addi-tion, Lane score remained 0.

Histological examination
Five weeks after operation, little irregular new bone callus was found in the sections of distal and proximal ends; a lot of connective tissues were observed in the sections of bone de-fects, but there was no bone-like tissue formation. Lane score was 0. Ten weeks after operation, the amount of callus in-creased and bone trabecula formed in callus. Bone defects were still filled by a lot of connective tissues, and there was still no bone tissue formation. Lane score was 0. At the 15th week after operation, fracture end showed reduction of the bone trabecula and formation of compact bone. The connec-tive tissue was compact in the bone defects, but bone tissue still did not form. Lane score was 0.

DISCUSSION

Rabbits, dogs, pigs, monkeys, sheep, and other animals are frequently used as the animal models for tissue engineering to study the repair of long bone defects. Compared with other animals, goats have some advantages. For example, they are gentle, easy to train and feed with large figure and long limbs. In addition, they are easy to obtain, cheaper than monkey. Both femur and tibia of goat can be used to establish bone defect model, especially the tibia.
In clinic, the bone defect is usually accompanied by soft tissue injury and inadequate blood supply, indicating it is not appropriate to bone defect model in femur with rich blood supply. 1/3 of goat tibia inferior segment is not surrounded by muscle and directly below the skin. The middle and inferior segment occupies 1/2 to 2/3 of the whole tibia [10], which is a suitable site to establish model. Sexual maturity of goat needs six months, and adequate breeding-age period is at one and a half age. So 6 to 12 months aged goats are consisted with clinical cases of tibial defects conditions using young adults with high morbidity. Younger goat is not in accordance with clinical cases, and older one needs longer fed and bone defect grows slowly. Jugular vein and veins of lower extremity of goat are superficial and thick, easy to draw blood and conven-ient in serological diagnosis and immunology testing [10]. After a comprehensive comparison, goat is a good model for tibial defect research.
In clinic treatment of bone defects, nonunion, and osteo-myelitis, interior or external fixation is a common method. Except for complications such as infection, implant rupture, intraoperative fracture and fat embolism, locking intramarrow nail fixation results in implant filling in bone defects, and less implanted bone tissue engineered complex. AO (fur Arbeit Osteosyntes) fixation easily creats stress shielding effect, which depletes normal physiological stimulation in the end of fracture and causes osteoporosis, bone atrophy and refracture. Unilateral external fixator often leads to joint dysfunction of bending and extension, angular deformities and pinhole infec-tions. Half-ring sulcated external fixator, one kind of bone external fixators, has a relatively simple structure, with a multi-plane and stable fixation. Compared with internal fixa-tion devices, half-ring sulcated external fixator separates less periosteum, destroys less soft tissue and eases recovery after fixation. According to different objectives, compressed fixa-tion and elastic fixation could be used for bone defect sites.
For construction of animal model of bone tissue engineer-ing, the amount of bone grafting or filling of composite of biological materials and seed cells is larger. External fixators do not occupy large space of tibia defect model, which re-sulted in more direct detection the formation and amount of bone. Therefore, half-ring sulcated external fixator displays unique advantages in making bone defect model in goats [11-13].
Compared with clinical applications of half-ring sulcated external fixator, features of goat tibial defect fixation are de-scribed as follows: The upper and lower end of defects is fixed by double half-ring, but not single-ring. The goats are more difficult to manage than people, which contributs to lateral or angular displacement easily.
The distance between two half-rings of proximal defect and two half-rings of distal end should keep long, while the dis-tance between two half-ring of proximal and distal end should be close. Half-ring gap should face to posterior surface in fix-ing the defects to avoid half-ring caused skin injury of the contralateral leg. In the same plane, angle between the two needles should be larger to ensure to put the ring easily, and avoid the poor X-ray results for the coincidence of screw, nut, half-ring imaging and defect sites after operation. Lateral or oblique films are taken. The duration of intramuscular injec-tion of antibiotics should be relatively long.
Excluding systemic diseases, nutrition, local infection and other reasons, whether bone defect can heal by itself is de-pendent on the length of defects.
Schmitz et al [14-16] suggested the critical value for experi-mental animal that could not heal by itself was the length of 1/10 and above by summarizing the model with unhealed bone defects. Many studies have reported different lengths for the preparation of goat tibial bone defects ranging from 30 to 50 mm according to sheep species and different research pur-poses besides age, size of goats and fixed factors.
Some goats are larger, with heavier weight and longer tibia. The length of the tibia is (138 ± 24) mm and the diameter is (12 ± 1.8) mm. In some studies the goats between 6 to 12 months age, weighing 20 to 25 kg were selected, and made 20% of bone and periosteum defects in the middle of tibia, 57-62.5 mm from the tibial tubercle. In 15 weeks after opera-tion, the X-ray and histological examination confirmed that 20% of the defect could not heal by itself in goat. After nearly 15 weeks, the medullary cavity was almost closed, which meant no possibility of healing. Currently, no studies report the length of tibial bone defects of sheep that could not heal by itself. Larger length could be performed, but the increase in defect length would bring a series of problems such as diffi-culties in fixation.
In this study, we used 20% of the length of the tibia, instead of 15 mm or 30 mm, which avoided the differences of defect percentage of the tibia caused by individual differences in goat. The proportion of the same length of defect of the tibia in the larger goat is smaller and the healing is easier, and vice versa.
On the other hand, the length of defect of the tibia was de-termined by comparing the average length of the tibia, distal limb blood flow impact of different defect length of goat tibia and choice of external fixation method by Yang et al [17]. The length of the tibia was (138 ± 24) mm, which was signifi-cantly higher than the standards that the critical value that bone defect could not self-heal was 1/10 of the length [14-16]. Gener-ally, most animal experiments have the disadvantages such as more individual differences, and more influent factors. In ad-dition, because in the middle of sheep tibia, there are no ob-vious bone signs, and fibula is absent, the rotation and separa-tion, shortening, dislocation, and difficulties of the external fixator easily occur.
In this experiment, the external fixator was first installed in the intact tibia, and the nuts were not completely tightened until osteotomy fully. Infection rate after operation was higher, likely related to needle position exposure and susceptible en-vironment of feeding after operation, which was consisted with higher incidence rate of pinhole infection complications of bone external fixation [18-20].
Our experimental model identifies the general conditions and details of operation of half-ring sulcated external fixator in goat model of tibial defect, and minimizes the error with simple manipulation and repeatability.
It is a better model to test the osteogenic ability of compos-ite of biological material and seed cells in bone tissue engi-neering, and worth widely applying in the experimental study of long bone defect in dogs, apes, pigs or other large animals.

REFERENCES

1 Li XM, Feng QL, Liu XH, el at. Collagen-based implants reinforced by chitin fibres in a goat shank bone defect model. Biomaterials 2006; 9(27):1917-1923
2 Borhane H F, Olivier G, Pierre W. Osteogenicity of biphasic calcium phosphate ceramics and bone autograft in a goat model. Biomaterials 2008; 29(9):1177-1188
3 Cui L, Liu B, Liu GP. Repair of cranial bone defects with adipose derived stem cells and coral scaffold in a canine model. Biomaterials 2007;28(36):5477-5486
4 Li QH, Zen XZ, O BP, et al. Half-ring external fixators research and clinical applications. Zhonghua Guke Zazhi 1984; 4(16):332-334
5 Xu JZ, Wang AM, Li QH, et al. Research of animal experiment to stability of half ring external fixation. Disan unyi Daxue Xuebao 1992;14(3):274-277
6 Wang HL, Li J, Lin ZF, et al. Advanced development and application on adjust radius canoula external fixation. Zhonghua XiaoEr Waike Zazhi 1994;15(3):187-189
7 Lane JM, Sandhu HS. Current approaches to experimental bone grafting s. Orthop Clin North AM 1987;18(2):213
8 Ranieri C, Paolo G, Maddalena M, et al. A tissue engineering approach to bone repair in large animal models and in clinical practice. Biomaterials 2007;28(29):4240-4250
9 Moyo CK, Wouter JA, Huipin YC,et al. Bone tissue engineering in a critical size defect compared to ectopic implantations in the goat. J Orthop Res 2004;22(3):544-551
10 Wei H. Medical Experimental Zoology. Chengdu: Sichuan Science and Technology Publisher 1998:203
11 Qing H, Xu JZ, Wang XQ, et al. Preparation of animal model of long bone defect used in tissue engineering. Zhongguo Linchuang Kangfu 2004;8(4):3974-3975
12 Li XY, Yang ZH, Zhang LC, et al. Blast injury and establishment of its protective mechanism. Zhongguo Linchuang Kangfu 2004; 8(20):4028-4029
13 Wang K, Pei GX, Chen B, et al. The preparation of the bone defect model of goat for application in tissue engineering. Zhongguo Jiaoxing Waike Zazhi 2002;9(7):664-666
14 Schmitz JP, Hollinger JO. The critical size defect as an experimental model for craniomandibul of caial nonunions. Clin orthop Relat Res 1986;6(205):299-308
15 Teixeira CR, Rahal SC, Volpi RS, et al. Tibial segmental bone defect treated with bone plate and cage filled with either xenogeneic composite or autologous cortical bone graft. An experimental study in sheep. Vet Comp Orthop Traumatol 2007;20(4):269-276
16 Avery CM, Patterson AB, Rolton J, et al. Biomechanical study of prophylactic internal fixation of the radial osteocutaneous donor site using the sheep tibia model. Br J Oral Maxillofac Surg 2007; 45(6):441-446
17 Yang L, Li QH, Zhou ZA, et al. The study of the effect of limb blood circulation bone fixation for the treatment of long bone defect. Zhonghua Waike Zazhi 2000;38(2):145-147
18 Li QH. Questions of external fixation of bone in clinical application. Zhonghua Guke Zazhi 1996;16(10):60
19 Li QH. Theorem and Application in External Fixation of Bone. Chengdu: Sichuan Science and Technology Publisher 1992;12-13
20 Xu JZ, Li QH, Yang L, et al. Treatment of complex bone nonunion with external skeletal fixation. Zhonghua Guke Zazhi 2002; 40(4): 280-283

应用半环槽外固定器建立山羊胫骨缺损模型☆

简月奎1，田晓滨1，赵筑川1，安荣泽2，李 波1，邱 冰1，杨 震1
1贵州省人民医院，贵州省贵阳市 550002;2遵义医学院第五附属医院，广东省珠海市 519100
简月奎☆，1971年生，男，贵州省遵义市人，汉族，2007年第三军医大学毕业，博士，主治医师。主要从事组织工程和脊柱外科研究。
摘要
背景：方便成骨及其成骨量的检测、占用空间相对较小，半环槽式外固定器属于骨外固定器，结构相对简单，多平面式固定，作者尚未查到用其建立胫骨缺损大动物模型可信性的较多实验报告。
目的：验证半环槽外固定器构建山羊胫骨段缺损模型的可操作性及可重复性。
设计、时间、地点：随机对照观察。实验于2005-03/2007-02在第三军医大学动物实验室完成。
材料：9只健康成年山羊由第三军医大学动物中心提供；半环槽外固定器为第三军医大学李起鸿教授研制、直径2.5 mm骨元针为上海医疗器械公司产品。
方法：动物经麻醉后，在距胫骨上关节面1.5~2.0 cm的松质骨同一平面各穿两针为第1组，两针夹角为40°。在第1组下3.0~5.0 cm处与第1组任一针平行穿一针为第2组。第4组穿针处距胫骨下端关节面2.0 cm松质骨处。第3组与第4组任一针平行，与第2组交叉成40°角，两组相距3.5~5.0 cm，共6针。上半环槽外固定器, 游离胫骨中段，根据X线片得到的胫骨长度，在胫骨中下段中间两穿针处之间用线锯锯断胫骨和骨膜, 造成20%骨与骨膜缺损。
主要观察指标：①术后动物一般情况及半环槽外固定器应用状况。②采用X线及Lane-Sandhu评分观察骨缺损自行修复成骨情况。
结果：纳入实验山羊9只，1只因术后感染死亡，1只出现针道感染，已予补充,其余均存活并进入结果分析。①实验动物术后2~6 h苏醒,并能站立进食。螺母松动后及时旋紧，无外固定松动失败,皮肤无坏死。术后2 d内伤肢能落地,但不能负重,2周后可以部分负重,3周有轻微跛行,4周后能自由活动,无跛行。②术后5周X-侧位片胫骨缺损处可见少许骨膜反应，术后10周断端有硬化,髓腔开始有闭合迹象,术后15周残端髓腔闭合,残端骨膜反应未加重，各时间点Lane-Sandhu X线评分均为0分。术后5，10，15周组织学检查未见骨形成，各时间点Lane-Sandhu 组织学评分均为0分。
结论：应用半环槽外固定器在造成山羊胫骨20%缺损的条件下构建的骨缺损动物模型，术后15周经检X线及组织学检测未出现骨愈合现象，证明该方法具有可操作性及可重复性，适用于组织工程用大段骨缺损动物模型的建立。
关键词: 半环槽外固定器；山羊；骨缺损；模型
中图分类号: R681 文献标识码: A 文章编号: 1673-8225(2008)17-03393-05
简月奎，田晓滨，赵筑川，安荣泽，李波，邱冰，杨震. 应用半环槽外固定器建立山羊胫骨缺损模型[J].中国组织工程研究与临床康复，2008，12(17):3393-3397
[www.zglckf.com/zglckf/ejournal/upfiles/08-17/17k-3393(ps).pdf]
（Edited by Tang LL/Su LL/Wang L）