Dentomaxillofacial Radiology (2009) 38, 465469 2009 The British Institute of Radiology http://dmfr.birjournals.

org

RESEARCH

Diagnostic accuracy of microcomputed tomography for osseous abnormalities in the rat temporomandibular joint condyle
S Kameoka1, Y Kuroki2,3, K Honda*,1,4, N Kijima2, K Matsumoto1,4, M Asano5,6, Y Arai1 and T Shirakawa2,3
Department of Oral and Maxillofacial Radiology, Nihon University School of Dentistry, Tokyo, Japan; 2Department of Pediatric Dentistry, Nihon University School of Dentistry, Tokyo, Japan; 3Division of Oral and Craniomaxillofacial Research, Dental Research Center, Nihon University School of Dentistry, Tokyo, Japan; 4Division of Advanced Dental Treatment, Dental Research Center, Nihon University School of Dentistry, Tokyo, Japan; 5Department of Pathology, Nihon University School of Dentistry, Tokyo, Japan; 6Division of Immunology and Pathobiology, Dental Research Center, Nihon University School of Dentistry, Tokyo, Japan
1

Objectives: Our aim was to investigate the diagnostic accuracy of in vivo micro-CT for osseous abnormalities of the rat temporomandibular joint (TMJ) condyle, using macroscopic observations as the gold standard. Methods: A 30 TMJ arthritis model was prepared by injecting inflammatory complete Freunds adjuvant (CFA) into one side of the TMJ cavities of rats. The TMJ condyles were then imaged using micro-CT. The samples were macroscopically evaluated for osseous abnormalities, including erosions, osteophytes, flattening and concavity. The micro-CT images were independently assessed for abnormalities using the same criteria. Images in three planes were produced using the micro-XYZ technique with the micro-CT equipment. Results: According to the macroscopic observations, 26 of the 60 rat condyles showed osseous abnormalities. The micro-XYZ images detected abnormalities in 25 of the condyles. The condyle diagnostic accuracy of micro-CT was 0.98, the sensitivity was 0.96 and the specificity was 1.0. Conclusions: Good diagnostic results were obtained using micro-CT. It is therefore an effective technique for the evaluation of osseous abnormalities in the rat TMJ condyle. Dentomaxillofacial Radiology (2009) 38, 465469. doi: 10.1259/dmfr/24350438 Keywords: microcomputed tomography; osseous abnormalities; temporomandibular joint morphology

Introduction Temporomandibular joint (TMJ) disorders are frequently diagnosed and treated in adult humans. Several basic and clinical studies on TMJ morphology and pain have been reported previously.15 Clinical research has also employed cone beam CT (CBCT) using a 3DX (J Morita Mfg Co., Kyoto, Japan) to evaluate the bone changes associated with temporomandibular disorders (TMDs).69 However, the relationship between pain and inflammation accompanying changes in the bony component of the TMJ has so far remained unclear. Osteoarthritis (OA) is the most common form of
*Correspondence to: Kazuya Honda, Department of Oral and Maxillofacial Radiology, Nihon University School of Dentistry, 1-8-13 Kanda-Surugadai, Chiyoda-ku, Tokyo 101-8310, Japan; E-mail: honda-k@dent.nihon-u.ac.jp Received 13 June 2007; revised 20 June 2008; accepted 29 August 2008

arthritis, characterized by structural bony changes such as osteophytosis and progressive loss of articular cartilage in synovial joints.10 Patients with this disease often suffer from pain and stiffness of joints, which can result in physical disability. Clinical and laboratory data suggest that there is a role for inflammation in OA, at least in some patients and in some phases of the disease.11,12 OA in the TMJ presents as multiple joint noises during jaw movements that are accompanied by pain.13 Our understanding of the pathomorphological changes of the TMJ in individuals suffering from OA has been limited, however, by the lack of an appropriate animal model that expresses the bony changes in the TMJ and bears a close resemblance to human TMJ-OA.14,15 A recently developed micro-CT system, R_mCTH (Rigaku Co., Tokyo, Japan), has made it possible to

466

In vivo micro-CT of the rat TMJ S Kameoka et al

obtain images of anaesthetized animals positioned on a secure object stage.16,17 Clear bone images of small animals can be obtained with an exposure time of 17 s and the reconstruction time is approximately 150 s. To our knowledge, no published reports have evaluated the performance of micro-CT alongside macroscopic observations for the evaluation of rat TMJ condyle morphology. The purpose of the present study was to assess the ability of micro-CT to detect osseous abnormalities of the rat TMJ condyle, using macroscopic observations as the gold standard.

Materials and methods R_mCTH equipment Figure 1 shows an overview of the new in vivo microCT apparatus. The X-ray tube had a microfocus with a minimum size of 7 mm. The current study employed a 20 mm X-ray focus. The X-ray sensor comprises a flat panel detector with an effective matrix size of 4806480. The focus detector distance ranges from 212 mm to 496 mm, depending upon the object. A distance of 212 mm was used for the current study, and the voxel size was set at 50650650 mm. The voltage of the X-ray tube ranges from 43 kV to 90 kV, depending upon the object. The current study used a voltage of 90 kV. The tube current ranges from 70 mA to 300 mA. Our study used a current of 177 mA. The exposure time was 17 s. Projection data for a total of 512 frames were collected and reconstructed by CoreTM 2 Duo (Intel Co., Santa Clara, CA). Volume data for a cylindrical area, 2.4 cm in diameter and 2.4 cm in height, were also collected. The reconstructed image (micro-XYZ) voxel matrix size was set at 48064806480. I-View-R (Rigaku Co., Tokyo, Japan) was used for image processing.16,17

Study protocol Male SpragueDawley rats aged 89 weeks and weighing 250350 g were used for this study. A 30 TMJ arthritis model was prepared by injecting an inflammatory substance, 50 ml complete Freunds adjuvant (CFA), into one TMJ cavity of each rat. The CFA was suspended in an oil/saline (1:1) emulsion. The rats were anaesthetized with pentobarbital sodium (50 mg kg21, intraperitoneally). The CFA was injected through the carefully positioned tip of a 26gauge needle as follows. Radioscopy was performed from the horizontal direction and the vertical direction, while moving the object stage in the xyz direction, and the TMJ condyle was confirmed to be at the centre of the field of vision. We then verified that the tip of the needle was positioned on the TMJ (Figure 2). The animals were euthanized 14 days after CFA injection by an intra-abdominal injection of nembutal, and micro-CT was used to assess the osseous abnormalities of the TMJ immediately before the mandibles were incised and fixed in 10% neutral formalin. Images of the condyle were produced in the sagittal, axial and frontal planes using the micro-CT equipment. The same monitor was used to create uniform brightness at the time of the image evaluations. The reconstructed micro-XYZ images of 60 mandibular condyles were independently assessed by four dentists and one pathologist (S.K, Y.K, K.H, K.M and M.A) for osseous abnormalities (cortical erosion, osteophytes, flattening and concavity). If the evaluations differed, the images

Figure 1 Overview of micro-CT. The front doors were closed when the X-ray unit was activated
Dentomaxillofacial Radiology

Figure 2 Evidence of the needle tip in the temporomandibular joint (TMJ) cavity. The three-plane micro-CT image confirmed the presence of the needle tip in the TMJ cavity. (a) Coronal slice image, (b) horizontal slice image and (c) sagittal slice image

In vivo micro-CT of the rat TMJ S Kameoka et al

467

were re-checked by all five observers and only those findings upon which at least three observers agreed were recorded. The specimens were also macroscopically evaluated for the same abnormalities, using the same criteria and protocol employed for the micro-XYZ image evaluation, by the same observers. No information from the anatomical examination was available to the observers at the time of image evaluation. The sensitivity, specificity and accuracy for the detection and assessment of osseous abnormalities of the TMJ condyle were calculated. All experiments were approved by the Animal Experimentation Committee at the Nihon University School of Dentistry and by the Matsumoto Dental University Committee on Intramural Animal Use.

Table 1 Detection of osseous abnormalities in 60 mandibular condyles by macroscopic observation and micro-XYZ images. Macroscopic observation as the gold standard detected 26 condylar bony changes, whereas the micro-XYZ detected 25 osseous abnormalities of the condyles Macroscopic observation Mandibular condyle Osseous abnormalities Normal bone structure 60 26 34 Micro-XYZ images 60 25 35

Results According to the macroscopic observations, 26 of the 60 rat condyles showed osseous abnormalities. The micro-XYZ detected abnormalities in 25 of the condyles (Table 1). The micro-XYZ diagnoses were correct in 59 TMJ condyles (Table 2; Figures 35). The sensitivity, specificity and accuracy of the microXYZ approach were 0.96, 1.0 and 0.98, respectively (Table 2). Good agreement was found between the micro-XYZ and the macroscopic observations in terms of normal cortical bone (Figure 3), condylar osteophytosis and erosion (Figure 4).

Discussion Injection of an adjuvant into, or in the vicinity of, the TMJ of animals has been reported to cause acute TMJ inflammation.1822 Injecting CFA into the TMJ of rats induced TMJ swelling and chromodacryorrhoea dosedependently, and a significant attenuation of food intake was also observed.20,22,23 The head withdrawal threshold was significantly decreased in rats injected with CFA into the TMJ,21 suggesting neuronal hyperexcitability in the trigeminal sensory system. Despite the evidence of acute inflammation in the TMJ, and subsequent malfunctioning of the jaw system caused by CFA, the temporal pattern of the possible morphological changes in the TMJ after CFA injection has not yet been investigated. Previously, conventional CT has been used to observe the bone structure in small animal experiments. In this approach, a sample is set on a turntable during scanning. This method is not suitable for examining live

animals because of the long exposure times, ranging from a few minutes to several hours. In 1997, Arai and colleagues developed limited CBCT systems for use in dentistry.24,25 Since then, several studies have demonstrated the effectiveness of such images for examining both dental and TMJ structures.2528 Arai and colleagues expanded upon this approach by developing a new micro-CT system for examining in vivo samples.16,17 The time required for imaging by the new micro-CT system is only 17 s and the operation time is much shorter than that required for conventional CT, thereby allowing animals to be observed safely and easily over a long experimental period. The voxel matrix of the new system is 51265126384 and the resolution ranges from 10610610 mm to 10061006100 mm. These functions have similar resolution to conventional CT, and an exposure time equal to that of 3DX. R_mCTH achieves a high resolution in a relatively short time-period, and therefore has the best basic performance. Arita and colleagues29 reported that micro-CT was an effective tool for evaluating the mouse gutter-shaped root as a quantitative trait. In addition, Osuga and colleagues30 observed rat dental pulp healing after pulpotomy using in vivo micro-CT, and concluded that it allowed the continuous monitoring of small experimental animals, such as rats, under anaesthesia, with adequate results. Indeed, this approach produced a high-quality image of the entire head region of the rat. It has been suggested that this approach could be used for the long-term continuous observation of changes in the condition of teeth after pulpotomy in experimental animals. However, no previous reports have assessed the efficacy of micro-CT observations of rat TMJ condyle morphology. In the present study, rats were anaesthetized and CFA was injected into the TMJ cavity through the facial skin. Several previous studies have used this method of inducing inflammation to investigate TMJ pain.1923,31,32 However, the relationship between pain and inflammation accompanying changes in the TMJ bone has remained unclear. Moreover, previous studies

Table 2 Diagnostic performance with micro-XYZ images. The sensitivity, specificity and accuracy obtained in this study were 0.96, 1.0 and 0.98, respectively Examination Micro-XYZ images True-positive 25 True-negative 34 False-positive 0 False-negative 1 Sensitivity 0.96 Specificity 1 Accuracy 0.98

Dentomaxillofacial Radiology

468

In vivo micro-CT of the rat TMJ S Kameoka et al

Figure 3 True negative results for a condyle. In this case, normal cortical bone structure was shown in both the macroscopic view and by micro-XYZ. (a) Macroscopic view, (b) coronal slice image, (c) horizontal slice image and (d) sagittal slice image

have not confirmed three-dimensionally that the CFA has been injected accurately into the TMJ cavity. In the current study, we verified that the CFA had been injected into the TMJ cavity and used micro-CT to evaluate the TMJ morphology of a rat with inflammation. The condyle diagnostic accuracy in the current study was 0.98 for the micro-XYZ images. The sensitivity and specificity were 0.96 and 1.0, respectively. Good diagnostic results were thus obtained using the microXYZ images. In a previous human study, we found that the mandibular condyle diagnostic accuracy was 0.90 for CBCT and 0.86 for helical-CT, the sensitivity was 0.80 for CBCT and 0.70 for helical-CT, and the specificity was 1.0 for both CBCT and helical-CT.6 We found CBCT to be superior to helical-CT in the diagnosis of single structural bone changes, whereas the techniques showed equivalent performance in the comprehensive diagnosis of TMJ disease. Therefore, CBCT seems to be a suitable alternative technique for the evaluation of osseous abnormalities of the TMJ. Notably, our current findings were similar to, or better than, those in the human study. The good diagnostic results obtained using the micro-XYZ images indicate that micro-CT is an effective technique for the evaluation of osseous abnormalities of the rat TMJ condyle. The current study allowed evaluation after bone transformation had occurred. The greatest advantage

Figure 4 True positive results for a condyle. In this case, both the macroscopic view and micro-XYZ clearly showed condylar osteophytosis (arrowheads) and erosion (arrows). (a) Macroscopic view, (b) coronal slice image, (c) horizontal slice image and (d) sagittal slice image
Dentomaxillofacial Radiology

Figure 5 False negative results for a condyle. In this case, microXYZ produced negative results, but the macroscopic view produced positive results. The arrow indicates the osteophyte of the condyle in the macroscopic view. (a) Macroscopic view, (b) coronal slice image (c) horizontal slice image and (d) sagittal slice image

In vivo micro-CT of the rat TMJ S Kameoka et al

469

of micro-CT is the ability to observe animals under mild anaesthesia without killing them. R_mCTH makes it possible to perform longitudinal studies that examine changes in bone structure without considering individual differences. Our findings suggest that micro-CT has high diagnostic precision. This technology therefore has the potential both to evaluate TMJ morphology and to help develop treatment plans for TMDs.
References
1. Westesson PL. Reliability and validity of imaging diagnosis of temporomandibular joint disorder. Adv Dent Res 1993; 7: 137151. 2. Solberg WK, Hansson TL, Nordstro m B. The temporomandibular joint in young adults at autopsy: a morphologic classification and evaluation. J Oral Rehab 1985; 12: 303321. 3. McNeill C. History and evolution of TMD concepts. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1997; 83: 5160. 4. Alamoudi N, Farsi N, Salako NO, Feteih R. Signs and symptoms of temporomandibular disorder (TMD) and craniofacial form. Am J Orthod Dentofacial Orthop 1996; 110: 7378. 5. Najlaa A, Najat F, Nathanael OS, Rabab F. Temporomandibular disorders among school children. J Clin Pediatr Dent 1998; 22: 323328. 6. Honda K, Larheim TA, Maruhashi K, Matsumoto K, Iwai K. Osseous abnormalities of the mandibular condyle: diagnostic reliability of cone beam computed tomography compared with helical computed tomography based on an autopsy material. Dentomaxillofac Radiol 2006; 35: 152157. 7. Honda K, Arai Y, Kashima M, Takano Y, Sawada K, Ejima K, et al. Evaluation of the usefulness of the limited cone-beam CT (3DX) in the assessment of the thickness of the roof of the glenoid fossa of the temporomandibular joint. Dentomaxillofac Radiol 2004; 33: 391395. 8. Matsumoto K, Honda K, Sawada K, Tomita T, Araki M, Kakehashi Y. The thickness of the roof of the glenoid fossa in the temporomandibular joint: relationship to the MRI findings. Dentomaxillofac Radiol 2006; 35: 357364. 9. Tsiklakis K, Syriopoulos K, Stamatakis HC. Radiographic examination of the temporomandibular joint using cone beam computed tomography. Dentomaxillofac Radiol 2004; 33: 196201. 10. Dieppe PA, Lohmander LS. Pathogenesis and management of pain in osteoarthritis. Lancet 2005; 365: 965973. 11. Abramson SB. Inflammation in osteoarthritis. J Rheumatol Suppl 2004; 70: 7076. 12. Brooks P. Inflammation as an important feature of osteoarthritis. Bull World Health Organ 2003; 81: 689690. 13. Stegenga B, de Bont LG, de Leeuw R, Boering G. Assessment of mandibular function impairment associated with temporomandibular joint osteoarthrosis and internal derangement. J Orofac Pain 1993; 7: 183195. 14. Lundeberg T, Alstergren P, Appelgren A, Appelgren B, Carleson J, Kopp S, et al. A model for experimentally induced temporomandibular joint arthritis in rats: effects of carrageenan on neuropeptide-like immunoreactivity. Neuropeptides 1996; 30: 3741. 15. Lai YC, Shaftel SS, Miller JN, Tallents RH, Chang Y, Pinkert CA, et al. Intraarticular induction of interleukin-1b expression in the adult mouse, with resultant temporomandibular joint pathologic changes, dysfunction, and pain. Arthritis Rheum 2006; 54: 11841197. 16. Arai Y. Development of a new mCT(R_mCT) for examining in vivo animals. Rigaku J 2005; 36: 2834. 17. Arai Y, Yamada A, Ninomiya T, Kato T, Masuda Y. Microcomputed tomography newly developed for in vivo small animal imaging. Oral Radiol 2002; 21: 1418.

Acknowledgments This study was supported by Grant-in-Aids for scientific research (B) 17390504, 2005, 2006 and research (C) 18592069, 2006, 2007. A grant for the promotion of multidisciplinary research projects entitled Translational research network on orofacial neurological disorders from the Japanese Ministry of Education, Culture, Sports, Science, and Technology to KH.

18. Mazzier AG, Laskin DM, Catchpole RC. Adjuvant induced arthritis in the temporomandibular joint of the rat. Arch Path 1967; 83: 543549. 19. Iwata K, Tashiro A, Tsuboi Y, Imai T, Sumino R, Morimoto T, et al. Medullary dorsal horn neuronal activity in rats with persistent temporomandibular joint and perioral inflammation. J Neurophysiol 1999; 82: 12441253. 20. Harper RP, Kerins CA, McIntosh JE, Spears R, Bellinger LL. Modulation of the inflammatory response in the rat TMJ with increasing doses of complete Freunds adjuvant. Osteoarthr Cartil 2001; 9: 619624. 21. Sato T, Kitagawa J, Ren K, Tanaka H, Tanabe A, Watanabe T, et al. Activation of trigeminal intranuclear pathway in rats with temporomandibular joint inflammation. J Oral Sci 2005; 47: 6569. 22. Guan G, Kerins CC, Bellinger LL, Kramer PR. Estrogenic effect on swelling and monocytic receptor expression in an arthritic temporomandibular joint model. J Steroid Biochem Mol Biol 2005; 97: 241250. 23. Kerins CA, Carlson DS, McIntosh JE, Bellinger LL. Meal pattern changes associated with temporomandibular joint inflammation/pain in rats; analgesic effects. Pharmacol Biochem Behav 2003; 75: 181189. 24. Arai Y, Tammisalo E, Iwai K, Hashimoto K, Shinoda K. Development of a compact computed tomographic apparatus for dental use. Dentomaxillofac Radiol 1999; 28: 245248. 25. Arai Y, Honda K, Iwai K, Shinoda K. Practical model 3DX of limited cone-beam X-ray CT for dental use. In: Lemke HU, Vannier MW, Inamura K, Farman AG, Doi K (eds). Proceedings of CARS (Computer Assisted Radiology and Surgery) 2001. Amsterdam: Elsevier, 2001, pp 671675. 26. Terakado M, Hashimoto K, Arai Y, Honda M, Sekiwa T. Diagnostic imaging with newly developed ortho cubic super high resolution CT (ortho-CT). Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2000; 89: 509518. 27. Nakajima A, Sameshima GT, Arai Y, Homme Y, Shimizu N, Dougherty H. Two- and three-dimensional orthodontic imaging using limited cone beam-computed tomography. Angle Orthod 2005; 75: 895903. 28. Katsumata A, Hirukawa A, Noujeim M, Okumura S, Naitoh M, Fujishita M, et al. Image artifact in dental cone-beam CT. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2006; 101: 652657. 29. Arita K, Saito I, Arai Y. Evaluation of mouse gutter shaped root(s) as a quantitative trait using micro-CT. Pediatr Dent J 2006; 16: 2327. 30. Osuga N, Yang J, Yamakawa Y, Ninomiya T, Arai Y, Raorao W, et al. Micro-CT observation of rat dental pulp healing after pulpotomy in in vivo study. Pediatr Dent J 2006; 16: 132137. 31. Hutchins B, Spears R, Hinton RJ, Harper RP. Calcitonin generelated peptide and substance P immunoreactivity in rat trigeminal ganglia and brainstem following adjuvant-induced inflammation of the temporomandibular joint. Arch Oral Biol 2000; 45: 335345. 32. Takeda M, Tanimoto T, Ikeda M, Nasu M, Kadoi J, Shima Y, et al. Temporomandibular joint inflammation potentiates the excitability of trigeminal root ganglion neurons innervating the facial skin in rats. J Neurophysiol 2005; 93: 27232738.

Dentomaxillofacial Radiology