CLINICAL STUDY

Disc Positions and Condylar Changes Induced byDifferent Stretching Forces in the Model for Anterior Disc

Displacement of Temporomandibular Joint

Hui Li, DDS, MD, PhD, Xieyi Cai, DDS, MD, PhD, Shaoyi Wang, DDS, MD, PhD,Chi Yang, DDS, MD, PhD, Hao Song, MDS, and Linjian Huang, MDS

Objective: The aim of this study was to compare the disc positionsand condylar changes induced by different stretching forces in themodified animal model for anterior disc displacement (ADD) ofthe temporomandibular joint.Methods: In the experimental group, 30 rabbits were equally di-vided into 3 subgroups and underwent surgical ADD via differentstretching forces: group Awith 0.5 N, group B with 1 N, and groupC with 2 N. In the sham group, 6 rabbits underwent the same sur-gery without the disc being pulled anteriorly. The diagnosis ofADD was made when the anterior band of the disc was locatedanteriorly to the articular eminence. Histologic and radiographicchanges of the condyles were observed under light microscopyand micro-computed tomography scanning 1 week after surgery.Results: The success rates of ADD were both 100% in groups Band C and 70% in group A. The correlations between the stretchingforce and severity of ADD, the stretching force and severity of car-tilage changes, and the severity of ADD and cartilage changes werestatistically significant (P < 0.01). The most advanced ADD and se-verest condylar changes were induced in group C. Condylar remod-eling and scleroses were found in micro-computed tomography scans.Conclusions: The rabbit model for ADD has been successfullyestablished in this study, which is feasible and minimally invasive.The stretching force of at least 1 N could induce the disc displacedsuccessfully. Larger stretching force would induce severer ADD andcondylar degenerative changes.

From the Department of Oral and Maxillofacial Surgery, Ninth People’s Hos-pital, Shanghai Jiao Tong University School of Medicine, and ShanghaiKey Laboratory of Stomatology, Shanghai, China.

Received January 29, 2014.Accepted for publication April 23, 2014.Drs. Xieyi Cai and Shaoyi Wang contributed equally to this work.Address correspondence and reprint requests to Dr Chi Yang, DDS,

Department of Oral and Maxillofacial Surgery, Ninth People’s Hospital,Shanghai Jiao Tong University School of Medicine, Zhi-Zao-Ju Road,No 639, Huangpu District, Shanghai 200011, China;E-mail: yangchi63@hotmail.com

Supported by the National Natural Science Foundation of China (Grant Nos.81200766, 81070848, and 81271114), Science and TechnologyCommission of Shanghai (Nos. 08DZ2271100, 08411967400, and13140902702), Program for Innovative Research Team of ShanghaiMunicipal Education Commission (No. 13XD1402300), and High-techResearch and Development Program of China (No. 2012AA030309).

The authors report no conflicts of interest.Copyright © 2014 by Mutaz B. Habal, MDISSN: 1049-2275DOI: 10.1097/SCS.0000000000001065

2112 The Journal of C

Copyright © 2014 Mutaz B. Habal, MD. Unauthor

Key Words: Temporomandibular joint, anterior disc displacement,animal model, stretching force, disc position, condylar change

(J Craniofac Surg 2014;25: 2112–2116)

Anterior disc displacement (ADD) of the temporomandibularjoint (TMJ) is one of the most common types of internal de-rangement (ID). The ADD induces pain, mandibular dysfunction,growth disturbance, and condylar osteoarthritis (OA), but its mech-anisms remain unclear.1,2 It is difficult to study detailed pathogene-sis in human bodies. Therefore, it is essential to establish an animalmodel of ADD and look into its effect on OA, condylar growth, andso on.

Because of the structural and functional similarity to humans,rabbit’s TMJ is mostly used to establish the ID models.3–9 Duringthe past 20 years, varied surgical-induced ADD models have beenestablished.

At first, ADD was induced by pulling the disc anteriorly afteropening the capsule and detaching the medial, anterior, and lateraldisc attachments.3,4 These models3–5 have been established at theexpense of damaging internal structures. The suture traction pulledthe disc forward abruptly, in a way which was not similar to that ofhumans.9 The stretching force generated by suture was hard to bequantified and kept consistent. Berteretche et al8 have made a greatinnovation for the ADD model. A nickel-titanium spring was ap-plied to pull the disc anteriorly in their study. The stretching forceof nickel-titanium spring was measured before surgery. Althoughthe success rate of ADD was reported as 100% in their study, thediscs were displaced in various positions. The possible explanationsare that the disc was displaced laterally and anteriorly and the springdid not function properly on the fat-filled orbital floor.

To keep the internal structure of TMJ intact, Gu et al9 pro-posed a novel method: the anterior extension of the disc was su-tured, and the disc was pulled forward by an elastic band withoutopening the capsule and detaching the attachments. However, thespace between the eyeball and the orbital floor is narrow and filledwith fat tissue. There is no anatomic feature or landmark for anteriorextension of the disc, and the tissue is thin and loose. Therefore, it isdifficult to put a suture in the anterior extension of the disc. In con-trast, the disc, which is fibrocartilaginous,10 is much more compact.Actually, the anterior part of disc is mostly chosen as the suturingposition.5–7 So, in a different way, Zhang et al6 sutured the elasticband to the anterior part of the disc in their report. However, partof the zygomatic branch was removed to expose the articular cap-sule, which would cause trauma and unsteadiness to the joint.6

The stretching force plays a crucial role in establishing anADD model. A suture was used to pull the disc forward in most ofprevious studies. However, the force generated by suture traction isuncontrollable and may decrease after the disc has been displacedabruptly. The elastic force is continuous and lasting and could bequantified by regulating the stretching length. Hence, more and more

raniofacial Surgery • Volume 25, Number 6, November 2014

ized reproduction of this article is prohibited.

The Journal of Craniofacial Surgery • Volume 25, Number 6, November 2014 Disc Positions and Condylar Changes

authors have been using elastic to draw the disc forward nowadays.6–9

The reported elastic forceswere various. The stretching forces of 1 N,8

1.20 N,6,7 and 85 g9 were applied in the previous studies, and thereported success rates of ADD were 100%,8 86.5%,6 100%,7 and93.8%,9 respectively. However, a single stretching force was used inall these studies, and no detailed reason was given for why the certainstretching force was chosen.

Hence, the aim of this study was to compare the disc posi-tions and condylar changes induced by different stretching forcesin the modified animal model for ADD of the TMJ.

FIGURE 1. A, A hole was drilled below the anterior process of thezygomatic arch. B, The elastic band was passed through the hole and thenfolded on the midpoint. C and E, The anterior part of the disc was sutured.The strands of suture were both led to the orbital floor. D and F, The suturewas knotted to the folded elastic band, and the elastic band wasstretched to the anticipant length. ZA, zygomatic arch; MR, mandibularramus; F, fossa; D, disc; C, condyle.

MATERIALS AND METHODS

Experimental DesignThirty-six New Zealand rabbits (Oryctolagus cuniculus),

aged 3 months and weighted from 1.80 to 2.05 kg, were providedby the Animal Experiment Laboratory of School of Medicine,Shanghai Jiao Tong University. The rabbits were randomly dividedinto an experimental group (n = 30) and a sham group (n = 6). An-esthesia was accomplished with intramuscular injection of ketamineof 0.2 mL and sulphamethoxazole of 0.8 mL. All the right jointsunderwent experimental ADD surgery or sham operation, andthe left joints were served as control. In the experimental group,30 rabbits were equally divided into 3 subgroups and underwentsurgical ADD via different stretching forces: group Awith the forceof 0.5 N, group B with the force of 1 N, and group C with the forceof 2 N. In the sham group, the rabbits underwent the same surgerywithout the disc being stretched anteriorly. One year of rabbitis equal to 13 years of human.11 All the rabbits were killed 1 weekafter surgery, corresponding to 13 weeks (about 3 months) in humansubjects. The disc position was identified by gross observation.The condyles were harvested and sent to histologic and radio-graphic examination under light microscopy and micro-computedtomography (CT) scanning.

This study was approved by the animal care and use ethicscommittee of Shanghai Jiao Tong University School of Medicine.

Surgical TechniqueThe fur in the right maxillofacial and periobital region was

shaved, and the skin was sterilized. A 4-cm anteroposterior incisionwas made from the skin overlying zygomatic arch to bone contact.Zygomatic arch was exposed from its root to the orbital floor. Twosmall processes could be found at the superior border of the anteriorzygomatic arch. A hole was drilled below the anterior process of thezygomatic arch, which was about 40 mm to the root of the zygo-matic arch (Fig. 1A). Elastic bands (Φ4.6, Unitek Elastics, 3M)were used for drawing the disc forward. This elastic band couldbe stretched to 10 mm with a tension force of 0.5 N, to 22 mm with1 N, and to 38 mm with 2 N. The elastic band was passed throughthe hole and then folded on the midpoint (Fig. 1B). The periosteumon the posterior border of the root of the zygomatic arch and thearticular fossa was peeled. We were careful not to open the TMJcapsule. The mandible was pushed backward, and the disc was seenthrough the translucent periosteum. The anterior part of the disc wassutured (2/0 suture), as shown in Fig. 1C. The strands of the suturewere both led to the orbital floor (Fig. 1E). The suture was knottedto the folded elastic band, and the elastic band was stretched to theanticipant length (Figs. 1D, F). The measured length of the bandwas half of the calculated length mentioned previously, for theelastic band had been folded. The disc was checked to see whetherit was drawn forward by the elastic band. After surgery, the woundwas washed and sutured in layers.

© 2014 Mutaz B. Habal, MD

Copyright © 2014 Mutaz B. Habal, MD. Unautho

Evaluation of ADDAll the rabbits were killed by an intravenous overdose injec-

tion of pentobarbital 1 week after surgery. The ADD was judged bygross observation of the relationship between the disc and the ar-ticular eminence in jaw-closed state. No ADD indicates that theanterior band of the disc was right beneath the crest of articular em-inence. Mild ADD indicates that the disc was anteriorly displacedwith the anterior band of the disc located anteriorly to the crest ofarticular eminence and posterior band located between the crestsof condyle and articular eminence. Severe ADD indicates that theposterior band of the disc was located anteriorly to the crest of thearticular eminence.

Specimen PreparationRabbits were killed by an intravenous overdose injection of

pentobarbital 1 week after surgery. The rabbits’ mandibular con-dyles were fixed in 4% of phosphate-buffered formalin solution for24 hours for histology observation and micro-CT scanning.

Histologic AnalysisThe condyles were decalcified in ethlenediamine-tetra-acetic

acid at pH 7.2–7.4 at 4°C for 4 weeks after fixation. The decalcifiedcondyles were initially sectioned on the sagittal plane and then em-bedded in paraffin. Moreover, 5-μm-thick sections were made andstained with hematoxylin and eosin for histomorphometric observation.

Micro-CT ScanningThe condyles in experimental groups were scanned by a

micro-CT system (mCT-80; Scanco Medical AG, Switzerland) witha spot size of 7 μm and maximum voltage of 36 kV. The slices were

2113

rized reproduction of this article is prohibited.

TABLE 1. The Distribution of Various Stretching Forces and the Severity of ADD

P = 0.001† No ADD Mild ADD Severe ADD Total

0.5 N 3 7 0 10

1 N 0 6 4 10

2 N 0 1 9 10

Total 3 14 13 30

†P < 0.01.

Li et al The Journal of Craniofacial Surgery • Volume 25, Number 6, November 2014

scanned parallel to the condylar sagittal plane. A high-resolutionprotocol (pixel matrix, 1024 � 1024; voxel size, 36 μm; slice thick-ness, 36 μm) was applied.

Statistical AnalysisThe statistical analysis was carried out by using SPSS (ver-

sion 16.0; SPSS, Chicago, IL). Chi-squared test was applied to in-vestigate the relationship among ADD, cartilage changes, andstretching forces. The Spearman method was used to investigatethe correlations between stretching force and severity of ADD/car-tilage changes. An alpha value of less than 0.05 was consideredto be statistically significant.

TABLE 2. The Severity of ADD and Cartilage Changes for Each Joint in theExperimental Group

No. Stretching Force, N ADD* Cartilage Changes†

1 0.5 0 0

2 0.5 1 1

3 0.5 1 1

4 0.5 0 0

5 0.5 1 1

6 0.5 0 0

7 0.5 1 0

8 0.5 1 1

9 0.5 1 1

10 0.5 1 2

11 1 1 1

12 1 1 2

13 1 2 2

14 1 2 2

15 1 1 1

16 1 1 2

17 1 2 2

18 1 1 2

19 1 2 2

20 1 1 1

21 2 1 3

22 2 2 4

23 2 2 3

24 2 2 3

25 2 2 4

26 2 2 3

27 2 2 4

28 2 2 4

29 2 2 4

30 2 2 4

*0, no ADD; 1, mild ADD; 2, severe ADD.

†0, no changes; 1, minor changes; 2, moderate changes; 3, severe changes; 4, de-structive changes.

2114

Copyright © 2014 Mutaz B. Habal, MD. Unauthor

RESULTS

Evaluation of ADDThe rabbits were all alive before killed, with good heal-

ing incision and normally eating. The severity of ADD in eachsubgroup was shown in Table 1. The success rate of ADD was70% in group A and 100% in both of groups B and C. The severityof ADD was different with various stretching forces (P < 0.01)(Table 1). Ninety percent were severe ADD in group C. The detailsof stretching forces and the severity of ADD for each rabbit wereshown in Table 2.

Histopathologic ResultsThe condylar cartilage was evaluated according to the classi-

fication of Bryndahl et al.5 The term “control” was used to refer tononoperated and sham-operated joints because there was no obvi-ous difference between them. The cartilage on the control condylewas intact, and its surface was smooth. The 5 layers in cartilage,which were composed of fibrous, proliferative, upper hypertrophic,lower hypertrophic, and endochondral ossified layers, were distinctlyrecognized under microscope (Fig. 2A). The severity of cartilagechanges in each subgroup was shown in Table 2. The condylar carti-lage changes with experimental ADD were shown in Fig. 2. The dis-tribution of stretching forces and the severity of condylar cartilagechanges were shown in Table 3 (P < 0.001). In group A, the condyle

FIGURE 2. A, No changes (n = 4): normal condylar cartilage with 5well-organized basic layers was distinctly recognized. B, Minor changes (n = 8):layers were intact with thinner thickness of cartilage, and slight cellular atrophywas observed. C, Moderate changes (n = 8): absence of specific layers,coexisting of moderate cartilage hypoplasia and hyperplasia, cellular atrophy,and vertical fibrous bundles between the fibrous layer and the subchondralbone were found. D, Severe changes (n = 4): loss of layer organization,pronounced cartilage hypoplasia, and severe cellular atrophy were observed.E, Destructive changes (n = 6): absence of condylar cartilage, exposure ofsubchondral bone, and fibrous tissue covered on were observed (hematoxylinand eosin stain, �20).

© 2014 Mutaz B. Habal, MD

ized reproduction of this article is prohibited.

TABLE 3. The Distribution of Stretching Forces and Condylar Cartilage Changes

P < 0.001†No

ChangesMinorChanges

ModerateChanges

SevereChanges

DestructiveChanges Total

0.5 N 4 5 1 0 0 10

1 N 0 3 7 0 0 10

2 N 0 0 0 4 6 10

Total 4 8 8 4 6 30

†P < 0.01.

The Journal of Craniofacial Surgery • Volume 25, Number 6, November 2014 Disc Positions and Condylar Changes

cartilage had no, minor, and moderate changes. In group B, the carti-lage changes were minor and moderate. In group C, they were severeand destructive.

Correlations Between Stretching Force and theSeverity of ADD and Cartilage Changes

The correlations between stretching force and the severityof ADD (r = 0.767, P < 0.001), stretching force and the severityof cartilage changes (r = 0.900, P < 0.001), and the severity ofADD and cartilage changes (r = 0.829, P < 0.001) were all statisti-cally significant.

Micro-CT EvaluationMicro-CT for the condyle showed that the condylar cartil-

age in those with experimental ADD was lost or altered and thecontours of the condyles were remodeled and were more flattenedin the joints with increasing stretching force (Fig. 3).

FIGURE 3. A, Control condyle: cartilage was intact, and subchondral boneand trabecular were apparent. B, Group A: the thickness of cartilage wasaltered. The condylar contour was slightly remodeled. C, Group B: thethickness of cartilage was obviously thinner. The condylar contour wasflattened. The subchondral bone was denser than that in the control condyle.The trabecular was obviously thicker than that in the control and group A.D, Group C: the cartilage loss and flattened contour were obviously observed.The subchondral bone was also denser than that in the control condyle.The trabecular was thicker than that in the condyles of control and group A.

DISCUSSIONThe mechanisms of OA and growth disturbance secondary

to ADD are still not well understood. Despite the studies on pa-tients and cadavers, the most important method is to study on ani-mal models. The previous reported models were still not perfectto simulate ADD in humans. A good animal model of ADD shouldbe minimally traumatic, feasible, easy to operate, and repeatable,with consistent results and high success rate.

In this study, the modified ADD model has been successfullyestablished. The operation is safe and minimally invasive to the in-ternal structure of TMJ. The joint capsule is kept intact, without sur-rounding bones removed or disc attachments broken. The operationis easy to manipulate. Compared with the method of Gu et al,9 theanatomic landmark of the root of the zygomatic arch is good to beaccessed, and the disc is easy to expose. These have made the oper-ation more feasible in this study. The tension of elastic band couldbe easily quantified and controlled by regulating the stretchedlength of the band. The disc is drawn forward with the elastic band,which is sutured to the anterior part of the disc, and it makes the ten-sion acting steadily.

The success rates of ADD induced by the stretching forces of1 and 2 N were both 100% in the present study, and the percentagesof severe ADD were 40% and 90%, respectively. The success ratewith the force of 0.5 N was 70%, which were all mild ADD in-duced. Moreover, 1 N is approximately equal to 102 g. In the modelof Berteretche et al,8 a force of 1 N generated by nickel-titaniumspring was chosen and induced small and large partial displace-ments. Considering the difficulty to control tension pulling bynickel-titanium spring, Gu et al9 used elastic band with a pullingstrength of 85 g instead, in which 15 of the 16 discs were success-fully anteriorly displaced but one failed for the band was broken.

© 2014 Mutaz B. Habal, MD

Copyright © 2014 Mutaz B. Habal, MD. Unautho

Zhang et al6 reported the success rates of 86.5% and 100%7 in theirstudies with an elastic pulling force of 1.20 N. Imai et al12 reportedanother method with the success rates of 68.8%: an ID and OAmodel was established by tracting the mandibular ramus with astretching force of 120 g for 4 weeks. The results of the presentand previous studies have indicated that a stretching force of ap-proximately 1 N would induce ADD with high success rate and thatlarger stretching force (2 N) could induce much severer ADD. Italso has demonstrated that the stretching force of 0.5 N could notmake all the discs be displaced successfully.

The condylar cartilage is considered as a biologically uniquestructure with a special ability for adaptive remodeling in responseto external stimuli.5,13 If the functional biomechanics applied on thecondyle exceed its adaptive ability, it could lead to tissue degenera-tion and joint disease.14 The disc buffers the stress on the condyle.Finite element models for ADD based on magnetic resonance im-ages15 indicated that the stress on posterior region of the condylewas higher, which may lead to breakdown of the condylar cartil-age. In this study, the condylar contour and cartilage features dif-fered significantly with various stretching forces and differentseverity of disc displacements. Larger stretching force induced se-verer ADD and cartilage changes. Berteretche et al8 reported thatOA changes were observed in most advanced ADD 96 days aftersurgery. In this study, the ADD induced by 2 N can also lead to con-dylar OA changes 1 week after surgery. Moreover, condylar erosionand sclerosis, which are typical radiographic findings of TMJ OA,16

were observed by micro-CT scanning. These results indicate thatsevere ADD would lead to OA in short term. Compared with groupC, minor and moderate cartilage changes were observed in groupsA and B. The severity of cartilage changes was correlated to thestretching force and the severity of ADD. On the basis of thesefindings, the model with stretching force of 2 N would be suitablefor the study of ADD with OA, and 1 N-induced ADD would be fa-vorable for the researches on its effects on mandibular dysfunc-tion and growth disturbance.

Because of the design of the study, the cartilage changeswere observed only 1 week after surgical ADD in the model of

2115

rized reproduction of this article is prohibited.

Li et al The Journal of Craniofacial Surgery • Volume 25, Number 6, November 2014

different stretching forces. A further study has been designed to ex-plore the mechanisms of OA and growth disturbance secondary toADD based on this model.

CONCLUSIONSA modified rabbit model for ADD has been successfully

established, and the trauma to joint structure and function is mini-mal. The stretching force of at least 1 N could induce the discdisplaced successfully. The stretching force of 2 N could induceADD and OA rapidly. The larger stretching force induced, the se-verer ADD and condylar degenerative changes would be found.

REFERENCES1. Dolwick MF. Intra-articular disc displacement. Part I: its questionable

role in temporomandibular joint pathology. J Oral Maxillofac Surg1995;53:1069–1072

2. Hall HD. Intra-articular disc displacement. Part II: its significant rolein temporomandibular joint pathology. J Oral Maxillofac Surg1995;53:1073–1079

3. Ali AM, Sharawy M, O’Dell NL, et al. Morphological alterations inthe elastic fibers of the rabbit craniomandibular joint followingexperimentally induced anterior disk displacement. Acta Anat1993;147:159–167

4. Mills DK, Daniel JC, Herzog S, et al. An animal model for studyingmechanisms in human temporomandibular joint disc derangement.J Oral Maxillofac Surg 1994;52:1279–1292

5. Bryndahl F, Warfvinge G, Eriksson L, et al. Cartilage changes linkretrognathic mandibular growth to TMJ disc displacement in a rabbitmodel. Int J Oral Maxillofac Surg 2011;40:621–627

2116

Copyright © 2014 Mutaz B. Habal, MD. Unauthor

6. Zhang S, Cao W, Wei K, et al. Expression of VEGF-receptors in TMJsynovium of rabbits with experimentally induced internal derangement.Br J Oral Maxillofac Surg 2013;51:69–73

7. Shen P, Zhang S, Yang C, et al. Synovial tissue in internalderangement of temporomandibular joint. J Craniofac Surg2013;24:748–752

8. Berteretche MV, Foucart JM, Meunier A, et al. Histologic changesassociated with experimental partial anterior disc displacement in therabbit temporomandibular joint. J Orofac Pain 2001;15:306–319

9. Gu Z, Zhou Y, Zhang Y, et al. An animal model for inducing anteriordisc displacement of the temporomandibular joint. J Orofac Pain2006;20:166–173

10. Tanaka E, Iwabuchi Y, Rego EB, et al. Dynamic shear behavior ofmandibular condylar cartilage is dependent on testing direction.J Biomech 2008;41:1119–1123

11. Souba WW, Wilmore WD. Surgical Research. San Diego, CA:Academic Press; 2001:359

12. Imai H, Sakamoto I, Yoda T, et al. A model for internal derangement andosteoarthritis of the temporomandibular joint with experimentaltraction of the mandibular ramus in rabbit. Oral Dis 2001;7:185–191

13. Shen G, Darendeliler MA. The adaptive remodeling of condylarcartilage—a transition from chondrogenesis to osteogenesis.J Dent Res 2005;84:691–699

14. Milam SB. Pathogenesis of degenerative temporomandibular jointarthritides. Odontology 2005;93:7–15

15. Nishio C, Tanimoto K, Hirose M, et al. Stress analysis in the mandibularcondyle during prolonged clenching: a theoretical approach with thefinite element method. Proc Inst Mech Eng [H] 2009;223:739–748

16. Israel HA, Diamond B, Saed-Nejad F, et al. Osteoarthritis andsynovitis as major pathoses of the temporomandibular joint:comparison of clinical diagnosis with arthroscopic morphology.J Oral Maxillofac Surg 1998;56:1023–1027

© 2014 Mutaz B. Habal, MD

ized reproduction of this article is prohibited.