Magnetic Resonance Imaging 33 (2015) 270–275
Contents lists available at ScienceDirect
Magnetic Resonance Imaging journal homepage: www.mrijournal.com
Dynamic MR imaging of temporomandibular joint: an initial assessment with fast imaging employing steady-state acquisition sequence Qi Sun a, Min-jun Dong a,⁎, Xiao-feng Tao a, Qiang Yu a, Kai-cheng Li a, Chi Yang b a
Department of Radiology,Ninth People's Hospital,School of Medicine,Shanghai Jiao Tong University, Shanghai, China, 200011 Department of Oral and Maxillofacial Surgery, Ninth People's Hospital, School of Stomatology, Shanghai Jiao Tong University School of Medicine, Shanghai Key Laboratory of Stomatology & Shanghai Research Institute of Stomatology, Shanghai, China, 200011
b
a r t i c l e
i n f o
Article history: Received 7 July 2014 Revised 23 October 2014 Accepted 31 October 2014 Keywords: TMJ FIESTA Dynamic technique MRI
a b s t r a c t Purpose: The study was aimed to determine and optimize the parameters for the MR fast imaging employing steady-state acquisition (FIESTA) sequence, which was to obtain an acceptable image to evaluate the value of the movement of the temporomandibular joint (TMJ). Methods: In this investigation, 20 volunteers were examined to determine and optimize the parameters of the FIESTA sequence. Then, 160 TMJs from 80 patients with temporomandibular joint disorders (TMD) of clinical suspicion were consecutively performed by both static MRI and dynamic FIESTA MRI on the oblique sagittal position. The FIESTA MR images of TMJs were obtained from a slow, consecutive, free and open–closed movement. Based on the cycles of TMJ movements during the process of FIESTA MRI (90 seconds), we classified all TMJs into 2 groups: cycles of open–closed mouths less than or equal to 3 (group 1) and more than 3 (group 2). Each image was marked level 1–3 by its quality. Meanwhile, the location of articular disc, mandibular condyle, motive artifact, “jumping sign” and the joint effusion in each TMJ were assessed respectively. Results: By dynamic FIESTA MRI among 160 TMJs, 92 TMJs (57.50%) were in group 1, and 68 TMJs were (42.50%) in group 2. There were statistically significant differences between group 1 and group 2 (p b 0.05). It was shown that the number of “level 3” in group 1 was greater than group 2, and the number of “level 1” in group 1 was less than group 2. The phenomenon of motion artifact and “jumping sign” were much significantly higher in group 2 than those in group 1 (p b 0.01). Furthermore, in all of the “jumping sign” cases, the phenomenon of “jumping sign” was significantly higher in group ADDwR than in group ADDw/oR (p b 0.01). There was a statistically significant correlation between disc-condyle complex in “jumping sign” phenomenon and group ADDwR (r = 0.621, p b 0.05). The data with the false matching rate of 31.52% showed that the maximum motion range on the dynamic imaging was greater than the static imaging. Among 160 TMJs, joint effusions of 37 TMJs (23.13%) were identified by dynamic FIESTA-MRI. Among 79 TMJs with ADDw/oR(anterior disc displacement without reduction), 42 sides were operated with Maxillofacial arthroscopy surgery. The surgical result was in agreement with the MR result. Conclusion: Most TMJs images with a slow free open–close movement (cycles≦3) could be successfully obtained by the dynamic FIESTA MRI. The FIESTA MRI might be considered as an additional method to evaluate the movement of the articular disk and the mandibular condyle. © 2015 Elsevier Inc. All rights reserved.
1. Introduction The temporomandibular joint (TMJ) is one of the most complex joints of the body, and magnetic resonance imaging (MRI) has been considered as a golden standard in diagnosing the temporomandibular joint disorders (TMD) defined as an assorted set of clinical ⁎ Corresponding author. E-mail address: [email protected] (M. Dong). http://dx.doi.org/10.1016/j.mri.2014.10.013 0730-725X/© 2015 Elsevier Inc. All rights reserved.
conditions, characterized by pain and dysfunction of the masticatory muscles, TMJ and associated hard and soft tissues [1–3]. MRI could assess the abnormal anatomic relationship between the articular disk and the mandibular condyle, owing to its no biological effect and good soft-tissue resolution [4]. However, the conventional static MRI fails to visualize the consecutive disc and mandibular condyle movement of the TMJ. Dynamic and pseudo-dynamic TMJ MRI techniques have been advocated by some investigators since 1987 [5–7]. With the improvements of MR technology, especially the
Q. Sun et al. / Magnetic Resonance Imaging 33 (2015) 270–275
A
271
in the clinical evaluation [5]. The study was aimed to determine and optimize the parameters for the MR FIESTA sequence and obtain an acceptable image to evaluate the value of the TMJ movement. 2. Materials and methods 2.1. Patients/Subjects
B
In this investigation, 20 volunteers (12 females and 8 males; range: 20 to 26 years; mean age: 23.15 ± 2.32 years) were examined to determine and optimize the parameters of FIESTA sequence. This study was approved by the ethics committee in our institution, and the informed consent was obtained from all subjects. Then, the study analyzed retrospectively the cases that included 160 TMJs from 80 patients who suffered from TMJ pain and dysfunction symptoms in our hospital. The patients, who were examined by using both static and dynamic FIESTA sequences, included 56 females and 24 males (range: 12 to 68 years mean age: 33.86 ± 15.62 years). 3. MRI All the patients were performed with a three-inch dual surface coil on a 1.5 Tesla Signa Twin-Speed Excite system (GE Healthcare, Milwaukee, Wis.). Two positions and three sequences were obtained in static MRI: (1) the closed-mouth position (maximum inter-cuspidation of teeth), oblique sagittal and coronal proton density- weighted imaging (PdWI); (2) the maximum opened position of the mandible, oblique sagittal T2-weighted imaging (T2WI). The scanning parameters were as follows respectively: (1) PdWI: TR/TE:1825.0/24.0,FOV: 10 cm × 10 cm,BW:15.63 kHz,SL/ SPACE:2.0/1.0 mm,MATRIX:320 × 192; (2) T2WI: TR/TE:3775.0/ 85.0,FOV: 10 cm × 10 cm,BW:15.63 kHz,SL/SPACE:1.5/0.5 mm, MATRIX:320 × 192. In the initial tests to determine optimal dynamic imaging, EPI, CINE and FIESTA pulse sequences were performed respectively for 20 volunteers. Dynamic oblique sagittal images were acquired separately during open–closed movement by 3-, 5-, and 7-mm slice thickness with varying spatial and temporal resolution. Then the optimal dynamic image by FIESTA sequence was obtained by the professors and radiologists in the department of radiology who have rich experience in clinical TMJ assessment. The FIESTA MRI was performed on a median oblique sagittal location that was perpendicular to the long axis of the mandibular condylar head. The
Table 1 The display comparison between TMJ disc and mandibular condyle classified on FIESTA MR imaging in both groups. Fig. 1. A. Static MR imaging showed anterior disc displacement without reduction (ADNR) and a small amount of effusion in upper cavity (black arrow). B. From the same patient, a series of sagittal images (16 images) including a full open–closed mouth movement were obtained from FIESTA-MR imaging. The ADDw/oR was shown, and a small amount of joint effusion in the upper cavity was identified as the high intensity signals.
availability of the dynamic fast imaging employing steady-state acquisition (FIESTA) combined with the higher field strength, it offers the potential to re-establish the dynamic imaging component
Classified
Group 1 (92 side) cycles≦3 Group 2 (68 side) Cycles N 3 Total (side) 2 χ
Articular disc (side)
1 level 2 level 3 level 1 level 2 level 3 level – 1/1–2/1 1/2–2/2 1/3–2/3
Mandibular condyle (side)
Position Shape
Motion orbit
Position Shape
Motion orbit
2 26 64 6 46 16 160 P N 0.05 P b 0.01 P b 0.01
1 41 50 5 51 12 160 P N 0.05 P b 0.01 P b 0.01
1 28 63 7 42 19 160 P b 0.05 P b 0.01 P b 0.01
1 36 55 3 49 16 160 P N 0.05 P b 0.01 P b 0.01
2 39 51 8 45 15 160 P b 0.05 P b 0.01 P b 0.01
2 36 54 6 45 17 160 P N 0.05 P b 0.01 P b 0.01
272
Q. Sun et al. / Magnetic Resonance Imaging 33 (2015) 270–275
70 60 50 40
1LEVEL 2LEVEL
30
3LEVEL
20 10 0 DP1
DS1
DM1 CP1
CS1
CM1 DP2
DS2
DM2 CP2
CS2
CM2
D : Disc, P : Position, M : Motion orbit, C : Condyle, 1 : Group 1, 2 : Group 2 Chart 1. A line chart was showed the difference between two groups. D: Disc, P: Position, M: Motion orbit, C: Condyle, 1: Group 1, 2: Group 2.
Table 2 The data of motion artifact and “Jumping sign” of TMJ observed on FIESTA MR imaging in both groups. Group
“Jumping sign”
Motion artifact
Group 1 (92 side) Group 2 (68 side) Total (160 side) χ2
Disc
Condyle
Disc
Condyle
16 44 60 P b 0.01
13 50 63 P b 0.01
13 42 55 P b 0.01
10 46 56 P b 0.01
patients were instructed to move continuous slow free open–close their mouths during the process of FIESTA-MRI. The cycles of open– close movements for all the patients were unlimited. The FIESTAMRI parameters were listed as following: SL/SPACE: 5 mm/1 mm; FOV:18 cm × 18 cm; MATRIX:256 × 256. Sixty continuous movement images of the TMJ were obtained in the FIESTA MRI in 90 seconds.
3. By the MRI, the articular disc and mandibular condyle including the position, shape, motion orbit, motion artifact and “jumping sign” were independently assessed by a radiologist and an oral surgeon. The “jumping sign” is defined as a phenomenon that a rapid forward motion of the articular disc or mandibular condyle overcomes the movement resistance and the moving speed is faster than the imaging capture ability of dynamic FIESTA sequence. Meanwhile, the disc position, the bone change of condyle and the joint effusion in each TMJ were assessed. The results were compared with the static MRI 3 levels of image quality. Each image was marked level 1–3 by its quality: 1. unclear contours; 2. the probable recognition of the contour and available for the diagnosis; 3. definite recognition of the contour. The marking consensuses were reached by 2 observers, and a rating of level 2 or 3 was regarded as positive. If the motion artifact or “jumping sign” was observed, it was counted for “1”, otherwise for “0”. Some of the patients with the anterior disc displacement or the rotation displacement underwent arthroscopic surgeries. The comparison and analysis were made between MRI and operation results. The statistical analysis was made by approx chi-square test and correlation analysis. The statistical software used was SPSS19.0. The case with a P value less than 0.05 was considered as a statistically significant difference.
3.1. Criterion and statistical analysis 4. Results On the basis of the completed times of open–close mouth during the dynamic FIESTA-MRI, we divided all the TMJs into 2 groups. In group 1: the number of cycles of continued open–close mouth is less than or equal to 3, and in group 2: the number of cycles is more than
Table 3 The display comparison between the classified TMJ and “jumping sign” of disccondyle complex on FIESTA MR imaging in both groups. Group
Classify (side)
“Jumping sign” of disccondyle complex (side)
Display rate (%)
Group 1 (92 side)
ADDwR (47) ADDw/oR (34) Normal (11) ADDwR (32) ADDw/oR (30) Normal (6) 160
30 1 0 25 3 0 59
63.83% 2.94% 0% 78.13% 1.00% 0% 36.88%
Group 2 (68 side) Total (160 side)
All of images were obtained successfully from all patients, and the dynamic FIESTA imaging technique was rapidly feasible (Fig. 1A, B). In dynamic FIESTA MRI, 160 TMJs were composed of 92 TMJs (57.50%) in group 1 (cycles≦3) and 68 TMJs (42.50%) in group 2 (cycles N 3). The position, shape and whole motion orbit of TMJ discs were displayed with (level 2 or 3) rates at 95.00% (152/160 side), 93.75% (150/160 side) and 96.25% (154/160 side) respectively. Meanwhile, those of mandibular condyle were displayed with rates at 95.00% (152/160 side), 95.00% (152/160 side) and 97.50% (156/ 160 side) respectively. The retrospective statistical analysis showed that the number of anterior disc displacement with reduction (ADDwR) patients was 79 sides including 7 sides rotation displacement, the number of anterior disc displacement without reduction (ADDw/oR) was 64 sides including 5 sides rotation displacement, and the number of cases in normal position was 17 sides. The comparison between group 1 and group 2 was listed in the Table 1. There were statistically significant differences between group 1 and group 2, except for level 1 between them including position and motion orbit of disc, shape and motion orbit of condyle. Chart 1 showed that the number of “level 3” in group 1 was greater
Q. Sun et al. / Magnetic Resonance Imaging 33 (2015) 270–275
273
Fig. 2. Static and dynamic FIESTA MR of the same patient showed anterior disc displacement. Static MR imaging showed anterior and medial disc displacement with reduction (disc was showed by black arrow) and sharpening at the top of the condyle. And the disc was showed by white arrow in dynamic FIESTA MR imaging. The position and shape of articular disc-condyle complex were matched similarly between two sequences.
than in group 2, and the number of “level 1” in group 1 less than in group 2. Moreover, the comparison about the motion artifacts and “jumping sign” between 2 groups was listed in the Table 2. A statistically significant difference between group 1 and group 2 was noticed (p b 0.01). The phenomenon of motion artifact and “jumping sign” was much higher in group 2 than in group 1 (p b 0.01). There was a statistically significant correlation between motion artifact (60 TMJs) and “jumping sign” of disc (55TMJ) (r = 0.798, p b 0.01). Also, a statistically high correlation between motion artifact (63TMJ) and “jumping sign” of mandibular condyle (56TMJ) was found (r = 0.869, p b 0.01). Furthermore, in all of the “jumping sign” cases (Table 3), the appearance of “jumping sign” was significantly more frequent in group ADDwR than in group ADDw/oR (p b 0.01).There was a statistically significant correlation between disc-condyle complex in “jumping sign” phenomenon and group ADDwR (r = 0.621, p b 0.05). Among 160 TMJs, the joint effusion of 37 TMJs (23.13%) was identified by dynamic FIESTA-MRI. The results agreed with the static MRI results. Furthermore, the position and shape of TMJs were matched between static and FIESTA sequences in the close position of group 1 (Fig. 2), with a coincidence rate of 97.83%. Statistical test was not finished in the group 2 because of a mount of the motion artifact. In the group 1, Fig. 3 showed that the maximum motion range on the dynamic imaging was greater than on the static imaging. The false matching rate was 31.52% (29/92 side). Meanwhile, 12 TMJs rotational displacement of disc could not be exactly observed on dynamic FIESTA MRI. In 79 TMJs with ADDw/oR, 42 sides were operated with the Maxillofacial arthroscopy surgery. The results of surgical operation were in close agreement with the MR results. The arthroscopy and dynamic FIESTA MRI of the same patient showed the anterior disc displacement without reduction (Fig. 4A,B).
5. Discussion FIESTA MRI is a gradient-echo technique using steady-state free precession with the fully refocused transverse magnetization [8]. This MRI sequence generates T2/T1 contrast because of its short repetition time and the symmetrical and balanced gradient around the echo time [9,10]. Therefore, the sequence gives very strong signals from tissues with large T2/T1 ratios, such as fluid, blood, and fat. In addition, the flow-compensating gradients, the short TR and TE in the FIESTA sequence, enabled the thinner slice imaging and significantly reduced scanning time and fluid flow artifacts [11]. In this study, about one fourth TMJs with joint effusion shown by FIESTA-MRI proved that this sequence might offer a better tissue contrast, especially between the joint liquid and the articular disc. The dynamic and pseudo-dynamic TMJ MRI is able to reveal the morphologic changes and movement relationships between the TMJ disc and the mandibular condyle [12,13]. Pseudo-dynamic TMJ MRI, which obtained TMJ images from a serial of multiple static MRI, not only required a longer examination time [14], but also provided the movement images of the TMJ disc and the mandibular condyle in the non-physiological situation. Chen et al. [15] studied the dynamic TMJ with echo-planar MRI (EPI), and Wang et al. [15] described the disc adequately by using turbo-fast low-angle shot (HASTE) MRI pulse sequence [15–17]. However, their visualization of the disc was limited due to the relatively low SNR and spatial resolution. Shimazaki et al. [12] reported adequate visualization of the disc by using the true FISP sequence with a reported temporal resolution ≦ 255 ms. In this study, we found that a consecutive movement relationship of the TMJ disc and the mandibular condyle was successfully shown in most subjects with the dynamic FIESTA-MRI. It indicates that the FIESTA MRI can be used to evaluate the dynamic TMJ imaging. In the process of FIESTA MRI, no additional device was utilized for open–close mouth due to the voluntary TMJ motion from the subjects. Nevertheless, the voluntary TMJ motion was limited by
274
Q. Sun et al. / Magnetic Resonance Imaging 33 (2015) 270–275
Fig. 3. This case was the ADDwR; it showed that the maximum motion range on the dynamic imaging (upper right corner) was greater than on the static imaging (lower right corner). White arrows indicated the articular disc, and the articular cavity effusion was indicated by black arrow, which was matched between the dynamic and static imaging.
the velocity of capturing images from the sequence when FIESTAMRI was employed in this study. In our investigation, we also got the same result: the imaging quality in group 1 (cycles≦3) was better than in group 2 (cycles N 3). In fact, the motion artifact of the TMJ disc and mandibular condyle was considered as the most important factor affecting the quality of the dynamic FIESTA-MRI. Our results indicated that the motion artifact and “jumping sign” of the mandibular condyle increased along with the acceleration of the TMJ movements. Furthermore, a statistically significant correlation between the motion artifact and “jumping sign” was found in the movement of the mandibular condyle. We can specifically identify
4
the exact anatomic condylar position relative to the glenoid fossa and eminence at which the disc was reduced. In addition, the disc morphology just before reduction was also well depicted. In another subject with unilateral joint clicking, the reduction phase was not depicted as well as the above one, but the reciprocal displacement phase was optimally visualized. However, the lateral and rotational displacement of the disc could not be exactly diagnosed by dynamic FIESTA MRI. The “jumping sign” of the mandibular condyle, which was infrequently described in the previous reports in regard as the dynamic and pseudo-dynamic MRI of the TMJ, was noticed in this study. A statistically significant difference of the “jumping sign”
4
Fig. 4. Arthroscopy and FIESTA MR of the same patient showed the anterior disc displacement without reduction.
Q. Sun et al. / Magnetic Resonance Imaging 33 (2015) 270–275
between both groups indicated that the sign was associated with the motion velocity of mandibualr condyle. As a consecutive movement of the TMJ disc and the mandibular condyle in FIESTA-MRI, the “jumping sign” usually takes place in the process of forward movement of the mandibular condyle from a slow to a fast velocity. The slow velocity of capturing images by the FIESTA sequence may lose some images of the mandibular condyle motion and lead to the phenomenon of “jumping” condyle. We speculated the “jumping sign” was related to the following situations. Firstly, there was some kind of resistance in the process of forward movements of the mandibular condyle (open mouth). Once the resistance was overcome by the mandibular condyle, the jerking movement from the mandibular condyle might form the “jumping sign”. In our research, we found that 69.62% (55/79 sides) cases of “jumping sign” were obtained in group ADDwR (group 1 + group 2). There was a statistically significant correlation between the disc-condyle complex in the “jumping sign” phenomenon and the group ADDwR. One of the most important evidence was that the “jumping sign” pictures were obtained when the disc-condyle complex recovered from the normal anatomy relationship in the process of open-mouth motion and a “click” sound was also noticed by the clinician. But, only 3 TMJs obtained the “jumping sign” phenomenon in group ADDw/oR because of the slight partial reduction of the discs. Therefore, we believed that there was a close relationship between the “jumping sign” phenomenon and the movement of the disc-condyle complex in the group of ADDwR. Secondly, the moving velocity of the mandibular condyle obviously exceeded the maximum speed of capturing images by the FIESTA sequence. A relative perfect dynamic MRI image obtained by the FIESTA sequence required a good cooperation from the subjects. Based on our study, most of subjects were able to comply with the instructions of slow open–close mouth. In this study, we also noticed that some TMJ discs in group 2 were not clearly demonstrated by FIESTA-MRI. Abolmaali et al [5] pointed out the phenomenon might be resulted from the thinning of the disk due to degenerative changes and rotation or lateral displacements of the disc during the mouth opening in their near-real-time MRI. On the other hand, the medio-lateral displacement of the TMJ disc was difficult to be estimated because of the mid-sagittal images of mandibular condyle in FIESTA-MRI. The similar situation was also mentioned by a previous report [18]. While we did not observe this form of internal derangement, we intended to further evaluate the efficacy of this protocol in detecting the rotational or sideways displacement. In conclusion, our initial study indicates that FIESTA-MRI may be used to demonstrate the consecutive movement of the TMJ disc and the mandibular condyle, and the motion relationship between the TMJ disc and the mandibular condyle in the process of slow open– close mouths on the sagittal plane. The FIESTA-MRI is able to provide a better contrast between the TMJ disc and the joint effusion.
275
Specifically, the technique could be applied in clicking joints to rule out additional causes of joint noises including those caused by ADDwR. As a dynamic imaging method, the FIESTA sequence may be considered as the complementarity to the static MRI, and offer additional information for clinic diagnosis and management.
References [1] Emshoff R, Brandlmaier I, Bertram S, et al. Comparing methods for diagnosing temporomandibular joint disk displacement without reduction. J Am Dent Assoc 2002;133(4):442–51. [2] Larheim TA. Role of magnetic resonance imaging in the clinical diagnosis of the temporomandibular joint. Cells Tissues Organs 2005;180(1):6–21. [3] Griffiths RH. Report of the president’s conference on the examination, diagnosis, and management or temporomandibular disorders. J Am Dent Assoc 1983;106: 75–7. [4] Drace JE, Young SW, Enzmann DR. TMJ meniscus and bilaminar zone MR imaging of the substructure-diagnostic landmarks and pitfalls of interpretation. Radiology 1990;177(1):73–6. [5] Abolmaali ND, Schmitt J, Schwarz W, Toll DE, Hinterwimmer S, Voql TJ, et al. Visualization of the articular disk of the temporomandibular joint in near-realtime MRI: feasibility study. Eur Radiol 2004;14(10):1889–94. [6] Burnett KR, Davis CL, Read J. Dynamic display of the temporomandibular joint meniscus by using “fast-scan” MR imaging. Am J Roentgenol 1987;149:959–62. [7] Conway WF, Hayes CW, Campbell RL. Dynamic magnetic resonance imaging of the temporomandibular joint using FLASH sequences. J Oral Maxillofac Surg 1988;46:930–8. [8] Nitz WR. Fast and ultrafast non-echo-planar MR imaging techniques. Eur Radiol 2002;12:2866–82. [9] Mikami T, Minamida Y, Yamaki T, Koyanaqi I , Nonaka T, Houkin K, et al. Cranial nerve assessment in posterior fossa tumors with fast imaging employing steadystate acquisition (FIESTA). Neurosurg Rev 2005;28:261–6. [10] Fuchs F, Laub G, Othomo K. TrueFISP:technical considerations and cardiovascular applications. Eur J Radiol 2003;46:28–32. [11] Carr JC, Simonetti O, Bundy J, Li D, Pereles S, Finn JP, et al. Cine MR angiography of the heart with segmented true fast imaging with steady-state precession. Radiology 2001;219:828–34. [12] Shimazaki Yoko, Saito Kazuhiro, Matsukawa Satoshi, Onizawa R, Kotake F, Nishio R, et al. Image quality using dynamic MR imaging of the temporomandibular joint with true-FISP sequence. Magn Reson Med Sci 2007;6:15–20. [13] Tomas X, Pomes J, Berenguer J, Quinto L, Nicolau C, Mercader JM, et al. MR imaging of temporomandibular joint dysfunction:a pictorial review. Radiographics 2006;26:765–81. [14] Burnett KR, Davis CL, Read J. Dynamic display of the temporomandibular joint meniscus by using fast-scan MR imaging. AJR Am J Roentgenol 1987;149: 959–62. [15] Chen YJ, Gallo LM, Meier D, et al. Dynamic magnetic resonance imaging technique for the study of the temporomandibular joint. J Orofac Pain 2000;14: 65–73. [16] Wang EY, Mulholland TP, Pramanik BK, Nusbaum AO, Babb J, Pavone AG, et al. Dynamic sagittal half-Fourier acquired single-shot turbo spin-echoMRimaging of the temporomandibular joint: initial experience and comparison with sagittal oblique proton-attenuation images. AJNR Am J Neuroradiol 2007;28:1126–32. [17] Maniere-Ezvan A, Havet T, Franconi JM, et al. Cinematic study of temporomandibular joint motion using ultrafast magnetic resonance imaging. Cranio 1999; 17:262–7. [18] Ren YF, Westesson PL, Isberg A. Magnetic resonance imaging of the temporomandibular joint:value of pseudodynamic images. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1996;81(1):110–23.
Contents lists available at ScienceDirect
Magnetic Resonance Imaging journal homepage: www.mrijournal.com
Dynamic MR imaging of temporomandibular joint: an initial assessment with fast imaging employing steady-state acquisition sequence Qi Sun a, Min-jun Dong a,⁎, Xiao-feng Tao a, Qiang Yu a, Kai-cheng Li a, Chi Yang b a
Department of Radiology,Ninth People's Hospital,School of Medicine,Shanghai Jiao Tong University, Shanghai, China, 200011 Department of Oral and Maxillofacial Surgery, Ninth People's Hospital, School of Stomatology, Shanghai Jiao Tong University School of Medicine, Shanghai Key Laboratory of Stomatology & Shanghai Research Institute of Stomatology, Shanghai, China, 200011
b
a r t i c l e
i n f o
Article history: Received 7 July 2014 Revised 23 October 2014 Accepted 31 October 2014 Keywords: TMJ FIESTA Dynamic technique MRI
a b s t r a c t Purpose: The study was aimed to determine and optimize the parameters for the MR fast imaging employing steady-state acquisition (FIESTA) sequence, which was to obtain an acceptable image to evaluate the value of the movement of the temporomandibular joint (TMJ). Methods: In this investigation, 20 volunteers were examined to determine and optimize the parameters of the FIESTA sequence. Then, 160 TMJs from 80 patients with temporomandibular joint disorders (TMD) of clinical suspicion were consecutively performed by both static MRI and dynamic FIESTA MRI on the oblique sagittal position. The FIESTA MR images of TMJs were obtained from a slow, consecutive, free and open–closed movement. Based on the cycles of TMJ movements during the process of FIESTA MRI (90 seconds), we classified all TMJs into 2 groups: cycles of open–closed mouths less than or equal to 3 (group 1) and more than 3 (group 2). Each image was marked level 1–3 by its quality. Meanwhile, the location of articular disc, mandibular condyle, motive artifact, “jumping sign” and the joint effusion in each TMJ were assessed respectively. Results: By dynamic FIESTA MRI among 160 TMJs, 92 TMJs (57.50%) were in group 1, and 68 TMJs were (42.50%) in group 2. There were statistically significant differences between group 1 and group 2 (p b 0.05). It was shown that the number of “level 3” in group 1 was greater than group 2, and the number of “level 1” in group 1 was less than group 2. The phenomenon of motion artifact and “jumping sign” were much significantly higher in group 2 than those in group 1 (p b 0.01). Furthermore, in all of the “jumping sign” cases, the phenomenon of “jumping sign” was significantly higher in group ADDwR than in group ADDw/oR (p b 0.01). There was a statistically significant correlation between disc-condyle complex in “jumping sign” phenomenon and group ADDwR (r = 0.621, p b 0.05). The data with the false matching rate of 31.52% showed that the maximum motion range on the dynamic imaging was greater than the static imaging. Among 160 TMJs, joint effusions of 37 TMJs (23.13%) were identified by dynamic FIESTA-MRI. Among 79 TMJs with ADDw/oR(anterior disc displacement without reduction), 42 sides were operated with Maxillofacial arthroscopy surgery. The surgical result was in agreement with the MR result. Conclusion: Most TMJs images with a slow free open–close movement (cycles≦3) could be successfully obtained by the dynamic FIESTA MRI. The FIESTA MRI might be considered as an additional method to evaluate the movement of the articular disk and the mandibular condyle. © 2015 Elsevier Inc. All rights reserved.
1. Introduction The temporomandibular joint (TMJ) is one of the most complex joints of the body, and magnetic resonance imaging (MRI) has been considered as a golden standard in diagnosing the temporomandibular joint disorders (TMD) defined as an assorted set of clinical ⁎ Corresponding author. E-mail address: [email protected] (M. Dong). http://dx.doi.org/10.1016/j.mri.2014.10.013 0730-725X/© 2015 Elsevier Inc. All rights reserved.
conditions, characterized by pain and dysfunction of the masticatory muscles, TMJ and associated hard and soft tissues [1–3]. MRI could assess the abnormal anatomic relationship between the articular disk and the mandibular condyle, owing to its no biological effect and good soft-tissue resolution [4]. However, the conventional static MRI fails to visualize the consecutive disc and mandibular condyle movement of the TMJ. Dynamic and pseudo-dynamic TMJ MRI techniques have been advocated by some investigators since 1987 [5–7]. With the improvements of MR technology, especially the
Q. Sun et al. / Magnetic Resonance Imaging 33 (2015) 270–275
A
271
in the clinical evaluation [5]. The study was aimed to determine and optimize the parameters for the MR FIESTA sequence and obtain an acceptable image to evaluate the value of the TMJ movement. 2. Materials and methods 2.1. Patients/Subjects
B
In this investigation, 20 volunteers (12 females and 8 males; range: 20 to 26 years; mean age: 23.15 ± 2.32 years) were examined to determine and optimize the parameters of FIESTA sequence. This study was approved by the ethics committee in our institution, and the informed consent was obtained from all subjects. Then, the study analyzed retrospectively the cases that included 160 TMJs from 80 patients who suffered from TMJ pain and dysfunction symptoms in our hospital. The patients, who were examined by using both static and dynamic FIESTA sequences, included 56 females and 24 males (range: 12 to 68 years mean age: 33.86 ± 15.62 years). 3. MRI All the patients were performed with a three-inch dual surface coil on a 1.5 Tesla Signa Twin-Speed Excite system (GE Healthcare, Milwaukee, Wis.). Two positions and three sequences were obtained in static MRI: (1) the closed-mouth position (maximum inter-cuspidation of teeth), oblique sagittal and coronal proton density- weighted imaging (PdWI); (2) the maximum opened position of the mandible, oblique sagittal T2-weighted imaging (T2WI). The scanning parameters were as follows respectively: (1) PdWI: TR/TE:1825.0/24.0,FOV: 10 cm × 10 cm,BW:15.63 kHz,SL/ SPACE:2.0/1.0 mm,MATRIX:320 × 192; (2) T2WI: TR/TE:3775.0/ 85.0,FOV: 10 cm × 10 cm,BW:15.63 kHz,SL/SPACE:1.5/0.5 mm, MATRIX:320 × 192. In the initial tests to determine optimal dynamic imaging, EPI, CINE and FIESTA pulse sequences were performed respectively for 20 volunteers. Dynamic oblique sagittal images were acquired separately during open–closed movement by 3-, 5-, and 7-mm slice thickness with varying spatial and temporal resolution. Then the optimal dynamic image by FIESTA sequence was obtained by the professors and radiologists in the department of radiology who have rich experience in clinical TMJ assessment. The FIESTA MRI was performed on a median oblique sagittal location that was perpendicular to the long axis of the mandibular condylar head. The
Table 1 The display comparison between TMJ disc and mandibular condyle classified on FIESTA MR imaging in both groups. Fig. 1. A. Static MR imaging showed anterior disc displacement without reduction (ADNR) and a small amount of effusion in upper cavity (black arrow). B. From the same patient, a series of sagittal images (16 images) including a full open–closed mouth movement were obtained from FIESTA-MR imaging. The ADDw/oR was shown, and a small amount of joint effusion in the upper cavity was identified as the high intensity signals.
availability of the dynamic fast imaging employing steady-state acquisition (FIESTA) combined with the higher field strength, it offers the potential to re-establish the dynamic imaging component
Classified
Group 1 (92 side) cycles≦3 Group 2 (68 side) Cycles N 3 Total (side) 2 χ
Articular disc (side)
1 level 2 level 3 level 1 level 2 level 3 level – 1/1–2/1 1/2–2/2 1/3–2/3
Mandibular condyle (side)
Position Shape
Motion orbit
Position Shape
Motion orbit
2 26 64 6 46 16 160 P N 0.05 P b 0.01 P b 0.01
1 41 50 5 51 12 160 P N 0.05 P b 0.01 P b 0.01
1 28 63 7 42 19 160 P b 0.05 P b 0.01 P b 0.01
1 36 55 3 49 16 160 P N 0.05 P b 0.01 P b 0.01
2 39 51 8 45 15 160 P b 0.05 P b 0.01 P b 0.01
2 36 54 6 45 17 160 P N 0.05 P b 0.01 P b 0.01
272
Q. Sun et al. / Magnetic Resonance Imaging 33 (2015) 270–275
70 60 50 40
1LEVEL 2LEVEL
30
3LEVEL
20 10 0 DP1
DS1
DM1 CP1
CS1
CM1 DP2
DS2
DM2 CP2
CS2
CM2
D : Disc, P : Position, M : Motion orbit, C : Condyle, 1 : Group 1, 2 : Group 2 Chart 1. A line chart was showed the difference between two groups. D: Disc, P: Position, M: Motion orbit, C: Condyle, 1: Group 1, 2: Group 2.
Table 2 The data of motion artifact and “Jumping sign” of TMJ observed on FIESTA MR imaging in both groups. Group
“Jumping sign”
Motion artifact
Group 1 (92 side) Group 2 (68 side) Total (160 side) χ2
Disc
Condyle
Disc
Condyle
16 44 60 P b 0.01
13 50 63 P b 0.01
13 42 55 P b 0.01
10 46 56 P b 0.01
patients were instructed to move continuous slow free open–close their mouths during the process of FIESTA-MRI. The cycles of open– close movements for all the patients were unlimited. The FIESTAMRI parameters were listed as following: SL/SPACE: 5 mm/1 mm; FOV:18 cm × 18 cm; MATRIX:256 × 256. Sixty continuous movement images of the TMJ were obtained in the FIESTA MRI in 90 seconds.
3. By the MRI, the articular disc and mandibular condyle including the position, shape, motion orbit, motion artifact and “jumping sign” were independently assessed by a radiologist and an oral surgeon. The “jumping sign” is defined as a phenomenon that a rapid forward motion of the articular disc or mandibular condyle overcomes the movement resistance and the moving speed is faster than the imaging capture ability of dynamic FIESTA sequence. Meanwhile, the disc position, the bone change of condyle and the joint effusion in each TMJ were assessed. The results were compared with the static MRI 3 levels of image quality. Each image was marked level 1–3 by its quality: 1. unclear contours; 2. the probable recognition of the contour and available for the diagnosis; 3. definite recognition of the contour. The marking consensuses were reached by 2 observers, and a rating of level 2 or 3 was regarded as positive. If the motion artifact or “jumping sign” was observed, it was counted for “1”, otherwise for “0”. Some of the patients with the anterior disc displacement or the rotation displacement underwent arthroscopic surgeries. The comparison and analysis were made between MRI and operation results. The statistical analysis was made by approx chi-square test and correlation analysis. The statistical software used was SPSS19.0. The case with a P value less than 0.05 was considered as a statistically significant difference.
3.1. Criterion and statistical analysis 4. Results On the basis of the completed times of open–close mouth during the dynamic FIESTA-MRI, we divided all the TMJs into 2 groups. In group 1: the number of cycles of continued open–close mouth is less than or equal to 3, and in group 2: the number of cycles is more than
Table 3 The display comparison between the classified TMJ and “jumping sign” of disccondyle complex on FIESTA MR imaging in both groups. Group
Classify (side)
“Jumping sign” of disccondyle complex (side)
Display rate (%)
Group 1 (92 side)
ADDwR (47) ADDw/oR (34) Normal (11) ADDwR (32) ADDw/oR (30) Normal (6) 160
30 1 0 25 3 0 59
63.83% 2.94% 0% 78.13% 1.00% 0% 36.88%
Group 2 (68 side) Total (160 side)
All of images were obtained successfully from all patients, and the dynamic FIESTA imaging technique was rapidly feasible (Fig. 1A, B). In dynamic FIESTA MRI, 160 TMJs were composed of 92 TMJs (57.50%) in group 1 (cycles≦3) and 68 TMJs (42.50%) in group 2 (cycles N 3). The position, shape and whole motion orbit of TMJ discs were displayed with (level 2 or 3) rates at 95.00% (152/160 side), 93.75% (150/160 side) and 96.25% (154/160 side) respectively. Meanwhile, those of mandibular condyle were displayed with rates at 95.00% (152/160 side), 95.00% (152/160 side) and 97.50% (156/ 160 side) respectively. The retrospective statistical analysis showed that the number of anterior disc displacement with reduction (ADDwR) patients was 79 sides including 7 sides rotation displacement, the number of anterior disc displacement without reduction (ADDw/oR) was 64 sides including 5 sides rotation displacement, and the number of cases in normal position was 17 sides. The comparison between group 1 and group 2 was listed in the Table 1. There were statistically significant differences between group 1 and group 2, except for level 1 between them including position and motion orbit of disc, shape and motion orbit of condyle. Chart 1 showed that the number of “level 3” in group 1 was greater
Q. Sun et al. / Magnetic Resonance Imaging 33 (2015) 270–275
273
Fig. 2. Static and dynamic FIESTA MR of the same patient showed anterior disc displacement. Static MR imaging showed anterior and medial disc displacement with reduction (disc was showed by black arrow) and sharpening at the top of the condyle. And the disc was showed by white arrow in dynamic FIESTA MR imaging. The position and shape of articular disc-condyle complex were matched similarly between two sequences.
than in group 2, and the number of “level 1” in group 1 less than in group 2. Moreover, the comparison about the motion artifacts and “jumping sign” between 2 groups was listed in the Table 2. A statistically significant difference between group 1 and group 2 was noticed (p b 0.01). The phenomenon of motion artifact and “jumping sign” was much higher in group 2 than in group 1 (p b 0.01). There was a statistically significant correlation between motion artifact (60 TMJs) and “jumping sign” of disc (55TMJ) (r = 0.798, p b 0.01). Also, a statistically high correlation between motion artifact (63TMJ) and “jumping sign” of mandibular condyle (56TMJ) was found (r = 0.869, p b 0.01). Furthermore, in all of the “jumping sign” cases (Table 3), the appearance of “jumping sign” was significantly more frequent in group ADDwR than in group ADDw/oR (p b 0.01).There was a statistically significant correlation between disc-condyle complex in “jumping sign” phenomenon and group ADDwR (r = 0.621, p b 0.05). Among 160 TMJs, the joint effusion of 37 TMJs (23.13%) was identified by dynamic FIESTA-MRI. The results agreed with the static MRI results. Furthermore, the position and shape of TMJs were matched between static and FIESTA sequences in the close position of group 1 (Fig. 2), with a coincidence rate of 97.83%. Statistical test was not finished in the group 2 because of a mount of the motion artifact. In the group 1, Fig. 3 showed that the maximum motion range on the dynamic imaging was greater than on the static imaging. The false matching rate was 31.52% (29/92 side). Meanwhile, 12 TMJs rotational displacement of disc could not be exactly observed on dynamic FIESTA MRI. In 79 TMJs with ADDw/oR, 42 sides were operated with the Maxillofacial arthroscopy surgery. The results of surgical operation were in close agreement with the MR results. The arthroscopy and dynamic FIESTA MRI of the same patient showed the anterior disc displacement without reduction (Fig. 4A,B).
5. Discussion FIESTA MRI is a gradient-echo technique using steady-state free precession with the fully refocused transverse magnetization [8]. This MRI sequence generates T2/T1 contrast because of its short repetition time and the symmetrical and balanced gradient around the echo time [9,10]. Therefore, the sequence gives very strong signals from tissues with large T2/T1 ratios, such as fluid, blood, and fat. In addition, the flow-compensating gradients, the short TR and TE in the FIESTA sequence, enabled the thinner slice imaging and significantly reduced scanning time and fluid flow artifacts [11]. In this study, about one fourth TMJs with joint effusion shown by FIESTA-MRI proved that this sequence might offer a better tissue contrast, especially between the joint liquid and the articular disc. The dynamic and pseudo-dynamic TMJ MRI is able to reveal the morphologic changes and movement relationships between the TMJ disc and the mandibular condyle [12,13]. Pseudo-dynamic TMJ MRI, which obtained TMJ images from a serial of multiple static MRI, not only required a longer examination time [14], but also provided the movement images of the TMJ disc and the mandibular condyle in the non-physiological situation. Chen et al. [15] studied the dynamic TMJ with echo-planar MRI (EPI), and Wang et al. [15] described the disc adequately by using turbo-fast low-angle shot (HASTE) MRI pulse sequence [15–17]. However, their visualization of the disc was limited due to the relatively low SNR and spatial resolution. Shimazaki et al. [12] reported adequate visualization of the disc by using the true FISP sequence with a reported temporal resolution ≦ 255 ms. In this study, we found that a consecutive movement relationship of the TMJ disc and the mandibular condyle was successfully shown in most subjects with the dynamic FIESTA-MRI. It indicates that the FIESTA MRI can be used to evaluate the dynamic TMJ imaging. In the process of FIESTA MRI, no additional device was utilized for open–close mouth due to the voluntary TMJ motion from the subjects. Nevertheless, the voluntary TMJ motion was limited by
274
Q. Sun et al. / Magnetic Resonance Imaging 33 (2015) 270–275
Fig. 3. This case was the ADDwR; it showed that the maximum motion range on the dynamic imaging (upper right corner) was greater than on the static imaging (lower right corner). White arrows indicated the articular disc, and the articular cavity effusion was indicated by black arrow, which was matched between the dynamic and static imaging.
the velocity of capturing images from the sequence when FIESTAMRI was employed in this study. In our investigation, we also got the same result: the imaging quality in group 1 (cycles≦3) was better than in group 2 (cycles N 3). In fact, the motion artifact of the TMJ disc and mandibular condyle was considered as the most important factor affecting the quality of the dynamic FIESTA-MRI. Our results indicated that the motion artifact and “jumping sign” of the mandibular condyle increased along with the acceleration of the TMJ movements. Furthermore, a statistically significant correlation between the motion artifact and “jumping sign” was found in the movement of the mandibular condyle. We can specifically identify
4
the exact anatomic condylar position relative to the glenoid fossa and eminence at which the disc was reduced. In addition, the disc morphology just before reduction was also well depicted. In another subject with unilateral joint clicking, the reduction phase was not depicted as well as the above one, but the reciprocal displacement phase was optimally visualized. However, the lateral and rotational displacement of the disc could not be exactly diagnosed by dynamic FIESTA MRI. The “jumping sign” of the mandibular condyle, which was infrequently described in the previous reports in regard as the dynamic and pseudo-dynamic MRI of the TMJ, was noticed in this study. A statistically significant difference of the “jumping sign”
4
Fig. 4. Arthroscopy and FIESTA MR of the same patient showed the anterior disc displacement without reduction.
Q. Sun et al. / Magnetic Resonance Imaging 33 (2015) 270–275
between both groups indicated that the sign was associated with the motion velocity of mandibualr condyle. As a consecutive movement of the TMJ disc and the mandibular condyle in FIESTA-MRI, the “jumping sign” usually takes place in the process of forward movement of the mandibular condyle from a slow to a fast velocity. The slow velocity of capturing images by the FIESTA sequence may lose some images of the mandibular condyle motion and lead to the phenomenon of “jumping” condyle. We speculated the “jumping sign” was related to the following situations. Firstly, there was some kind of resistance in the process of forward movements of the mandibular condyle (open mouth). Once the resistance was overcome by the mandibular condyle, the jerking movement from the mandibular condyle might form the “jumping sign”. In our research, we found that 69.62% (55/79 sides) cases of “jumping sign” were obtained in group ADDwR (group 1 + group 2). There was a statistically significant correlation between the disc-condyle complex in the “jumping sign” phenomenon and the group ADDwR. One of the most important evidence was that the “jumping sign” pictures were obtained when the disc-condyle complex recovered from the normal anatomy relationship in the process of open-mouth motion and a “click” sound was also noticed by the clinician. But, only 3 TMJs obtained the “jumping sign” phenomenon in group ADDw/oR because of the slight partial reduction of the discs. Therefore, we believed that there was a close relationship between the “jumping sign” phenomenon and the movement of the disc-condyle complex in the group of ADDwR. Secondly, the moving velocity of the mandibular condyle obviously exceeded the maximum speed of capturing images by the FIESTA sequence. A relative perfect dynamic MRI image obtained by the FIESTA sequence required a good cooperation from the subjects. Based on our study, most of subjects were able to comply with the instructions of slow open–close mouth. In this study, we also noticed that some TMJ discs in group 2 were not clearly demonstrated by FIESTA-MRI. Abolmaali et al [5] pointed out the phenomenon might be resulted from the thinning of the disk due to degenerative changes and rotation or lateral displacements of the disc during the mouth opening in their near-real-time MRI. On the other hand, the medio-lateral displacement of the TMJ disc was difficult to be estimated because of the mid-sagittal images of mandibular condyle in FIESTA-MRI. The similar situation was also mentioned by a previous report [18]. While we did not observe this form of internal derangement, we intended to further evaluate the efficacy of this protocol in detecting the rotational or sideways displacement. In conclusion, our initial study indicates that FIESTA-MRI may be used to demonstrate the consecutive movement of the TMJ disc and the mandibular condyle, and the motion relationship between the TMJ disc and the mandibular condyle in the process of slow open– close mouths on the sagittal plane. The FIESTA-MRI is able to provide a better contrast between the TMJ disc and the joint effusion.
275
Specifically, the technique could be applied in clicking joints to rule out additional causes of joint noises including those caused by ADDwR. As a dynamic imaging method, the FIESTA sequence may be considered as the complementarity to the static MRI, and offer additional information for clinic diagnosis and management.
References [1] Emshoff R, Brandlmaier I, Bertram S, et al. Comparing methods for diagnosing temporomandibular joint disk displacement without reduction. J Am Dent Assoc 2002;133(4):442–51. [2] Larheim TA. Role of magnetic resonance imaging in the clinical diagnosis of the temporomandibular joint. Cells Tissues Organs 2005;180(1):6–21. [3] Griffiths RH. Report of the president’s conference on the examination, diagnosis, and management or temporomandibular disorders. J Am Dent Assoc 1983;106: 75–7. [4] Drace JE, Young SW, Enzmann DR. TMJ meniscus and bilaminar zone MR imaging of the substructure-diagnostic landmarks and pitfalls of interpretation. Radiology 1990;177(1):73–6. [5] Abolmaali ND, Schmitt J, Schwarz W, Toll DE, Hinterwimmer S, Voql TJ, et al. Visualization of the articular disk of the temporomandibular joint in near-realtime MRI: feasibility study. Eur Radiol 2004;14(10):1889–94. [6] Burnett KR, Davis CL, Read J. Dynamic display of the temporomandibular joint meniscus by using “fast-scan” MR imaging. Am J Roentgenol 1987;149:959–62. [7] Conway WF, Hayes CW, Campbell RL. Dynamic magnetic resonance imaging of the temporomandibular joint using FLASH sequences. J Oral Maxillofac Surg 1988;46:930–8. [8] Nitz WR. Fast and ultrafast non-echo-planar MR imaging techniques. Eur Radiol 2002;12:2866–82. [9] Mikami T, Minamida Y, Yamaki T, Koyanaqi I , Nonaka T, Houkin K, et al. Cranial nerve assessment in posterior fossa tumors with fast imaging employing steadystate acquisition (FIESTA). Neurosurg Rev 2005;28:261–6. [10] Fuchs F, Laub G, Othomo K. TrueFISP:technical considerations and cardiovascular applications. Eur J Radiol 2003;46:28–32. [11] Carr JC, Simonetti O, Bundy J, Li D, Pereles S, Finn JP, et al. Cine MR angiography of the heart with segmented true fast imaging with steady-state precession. Radiology 2001;219:828–34. [12] Shimazaki Yoko, Saito Kazuhiro, Matsukawa Satoshi, Onizawa R, Kotake F, Nishio R, et al. Image quality using dynamic MR imaging of the temporomandibular joint with true-FISP sequence. Magn Reson Med Sci 2007;6:15–20. [13] Tomas X, Pomes J, Berenguer J, Quinto L, Nicolau C, Mercader JM, et al. MR imaging of temporomandibular joint dysfunction:a pictorial review. Radiographics 2006;26:765–81. [14] Burnett KR, Davis CL, Read J. Dynamic display of the temporomandibular joint meniscus by using fast-scan MR imaging. AJR Am J Roentgenol 1987;149: 959–62. [15] Chen YJ, Gallo LM, Meier D, et al. Dynamic magnetic resonance imaging technique for the study of the temporomandibular joint. J Orofac Pain 2000;14: 65–73. [16] Wang EY, Mulholland TP, Pramanik BK, Nusbaum AO, Babb J, Pavone AG, et al. Dynamic sagittal half-Fourier acquired single-shot turbo spin-echoMRimaging of the temporomandibular joint: initial experience and comparison with sagittal oblique proton-attenuation images. AJNR Am J Neuroradiol 2007;28:1126–32. [17] Maniere-Ezvan A, Havet T, Franconi JM, et al. Cinematic study of temporomandibular joint motion using ultrafast magnetic resonance imaging. Cranio 1999; 17:262–7. [18] Ren YF, Westesson PL, Isberg A. Magnetic resonance imaging of the temporomandibular joint:value of pseudodynamic images. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1996;81(1):110–23.
