| Sign In to gain access to subscriptions and/or personal tools. |
Dynamic Intra-articular Space Variation in Clicking TMJsClinic for Masticatory Disorders and Complete Dentures, Center for Oral Medicine, Dental and Maxillo-Facial Surgery, University of Zürich, Plattenstrasse 11, CH-8028 Zürich, Switzerland; Correspondence: * corresponding author, luigi{at}zui.unizh.ch
During mandibular movement, the geometric relationships of the articular surfaces in the temporomandibular joint (TMJ) change, so that the disc undergoes different stress concentrations with respect to time and position. In this study, we compared the intra-articular space variations of 13 clicking and 15 asymptomatic TMJs for jaw opening/closing. Magnetic resonance imaging and jaw tracking were combined to display the motion of the whole condyle within the fossa. In clicking TMJs, the mediolateral spread s of the stress-field trajectories was 2.4 ± 1.0 mm (smax = 4.9 ± 2.1 mm) with an aspect ratio a/h of 2.5 ± 1.6, both significantly greater than in controls (p < 0.05). The stress-field trajectories of the controls coincided during opening/closing (s = 0.9 ± 0.2 mm, smax = 1.8 ± 0.8 mm, a/h = 1.6 ± 0.3). Clicking TMJs showed much less coincident stress-field paths and much "flatter" stress-fields than controls during jaw opening/closing.
Key Words: biomechanics • clicking • congruence • kinetics • MRI • temporomandibular disorders • temporomandibular joint
The articular disc is the primary mechanism of stress distribution and lubrication in the TMJ. Thus, it has been speculated that altered mechanical properties of this tissue might reduce its functional effectiveness, increasing the likelihood of osteoarthritis (OA) (Nickel et al., 2001; Nickel and McLachlan, 1994). Also, animal models have shown that operative TMJ disc displacement or perforation may induce OA of the articular surfaces (Narinobou et al., 2000; Tominaga et al., 2002; Sharawy et al., 2003). TMJ disc failure seems to be related to trauma or higher concentrations of strain energy, but the mechanical etiology of this disease remains unclear. An unanswered issue is why lesions and perforations are more often localized in the lateral portion of the disc (Öberg et al., 1971; Stratmann et al., 1996). Similar to other joints such as the knee, TMJ articular surfaces are highly incongruent, and the disc compensates for their different curvatures. During mandibular movement, the geometric relationships of the TMJ articular surfaces vary, so that the disc undergoes stress concentrations that change with time and location. Recently, we demonstrated that mediolateral stress-field translation occurs in healthy human TMJs during jaw opening/closing (Gallo et al., 2000), protrusion, and laterotrusion (Chiaravalloti, 2002). Mediolateral shear stresses to the disc may result and, in turn, may compromise disc integrity (Waldman and Bryant, 1997), since this tissue is weaker in the mediolateral direction (Beatty et al., 2001). The analysis of mediolateral stress-field translation was possible with the use of a novel system that combines reconstructed TMJ anatomies with their kinematic data recorded with six degrees of freedom (Krebs et al., 1995). The system yields an accurate description of the geometry of TMJ articular surfaces and of the dynamic variation of the intra-articular space. Paths and aspect ratios (a/h) of minimum joint space areas (a = mean radius and h = mean intra-articular distance of the minimum joint space area) characterize the changes in stress concentrations with respect to time and location. For higher stress concentrations to occur, the aspect ratio of the stress-field has to be reduced, thus increasing localized compression and pressurization of the cartilage (Setton et al., 1993; Macirowski et al., 1994; Soltz and Ateshian, 1998). The study of mediolateral stress-field translation was performed on a group of subjects with asymptomatic TMJs, where we visually found the stress-field paths to coincide between jaw-opening and -closing phases. To date, little is known about the three-dimensional (3D) variation of the intra-articular space in symptomatic TMJs, where altered condylar kinematics might modify stress-field translation patterns and/or remodeling effects might occur. TMJ clicking is a typical TMD symptom with uncertain evolution (Elfving et al., 2002; Sato et al., 2003). This type of sound is generally related to a sudden acceleration of condylar and internally displaced disc tissues. It is unclear to what extent the hard structures undergo changes in clicking TMJs. The current study investigated the hypothesis that mediolateral stress-field translation through the cartilage matrix also occurs in clicking TMJs, and that the stress-field characteristics differ from those of asymptomatic joints. Specific aims of the current study were: (1) to test whether mediolateral stress-field translation occurs in clicking TMJs during jaw opening/closing; and (2) to compare the characteristics of the stress-field paths, particularly the aspect ratio between these TMJs and asymptomatic ones.
Subjects Twelve subjects (13 joints, three females, nine males, ages 16–38 yrs) with clicking TMJs and eight controls (five females, three males, ages 22–38 yrs) with asymptomatic TMJs participated in the study (Table 1  ). Initially, all potential subjects had a medical history recorded. They also responded to a questionnaire to uncover any history of TMJ pain or sounds, masticatory muscle pain or fatigue, before being examined by an expert clinician. The symptomatic subjects had to have pain-free TMJ clicking and no other signs and symptoms of TMD. Controls had to be totally free from past or present myoarthropathies of the masticatory system (Palla, 1986), and disc position had to be normal on MRI. All subjects were in good general health and gave informed consent before participating. The ethics committee of the Dental School approved the study protocol.
The protocol required acquisition of the TMJ anatomy of each subject by MRI. Subsequently, jaw movement was recorded for each subject while performing 10 cycles of symmetric jaw opening and closing at a deliberate rate, because of the impaired mandibular mobility of the symptomatic subjects.
Data Acquisition A set of 14 sagittal image slices perpendicular to the condylar long axis was made through each TMJ with 1.5 T MRI equipment. During imaging, each subject bit on a custom occlusal splint that was attached to a face bow with a reference system consisting of 3 non-collinear plastic spheres surrounded by magnetic contrast liquid. An additional set of 12 image slices was made through the reference system. For the 3D reconstruction of the MR images, the object contours were first input manually. After contour approximation by polylines and triangulation, the surface was described by a set of cubic parametric patches and rendered for visual inspection. The centers of the reference spheres were determined automatically. Mandibular movements were recorded by means of an opto-electronic jaw tracker (Mesqui et al., 1986; Airoldi et al., 1994). Two triangular target frames (TTFs), carrying 3 light-emitting diodes (LED) each, were fixed to the dental arches by means of customized splints that did not interfere with the occlusion. The TTFs defined a maxillary and a mandibular coordinate system. The LEDs were pulsed sequentially at 70 Hz each, and three linear cameras with fixed geometry recorded their positions. The temporospatial changes of the mandibular coordinate system relative to the maxillary coordinate system were determined by means of coordinate transforms. The positions of the three reference spheres were measurable with both the MRI and jaw-tracking devices and acted as an intermediate frame for the expression of MRI coordinates in the jaw-tracker coordinate system. The motion information obtained from the jaw-tracking system was thus applied to the 3D TMJ anatomy by means of a further set of coordinate transforms. The occurrence of TMJ clicking was recorded with an accelerometer positioned on the skin over the palpated condylar pole.
Data Analysis At each time step of mandibular motion, the stress-field area was defined as the area comprised of a set of 30 minimal distances between the polygon vertices approximating the fossa and condyle surfaces. The mean of this set described the minimum condyle-fossa distance (h). The standard deviation around the centroid of the stress-field was defined as the mean radius of the stress-field area (a). We determined the stress-field path by connecting the centroids and geometrically low-pass-filtering the 3D curve obtained, as in a previous study (Gallo et al., 2000). For every point on this curve, we also calculated the instantaneous velocity (V), the stress-field area (A), and the aspect ratio (a/h).
The stress-field paths were divided into segments by planes parallel to the coronal plane parallel to the condylar main axis, each separated by a distance of 1 mm (Fig. 1
Statistical Analysis Peak velocities (Vp) of the stress-field centroid were calculated for each jaw-opening and -closing phase. The mediolateral stress-field translation  D was calculated only if the velocity V was within 95% of Vp, excluding the time steps where the mandible was virtually motionless. The path spread (s) was averaged within each movement for all opening/closing cycles. The values of D, of smax, and of the maximum mediolateral component Vp, as well as the data describing aspect ratios (a/h) and areas (A), were averaged over the 10 opening/closing cycles. Given the non-normal distributions for all parameters, we performed Mann-Whitney U-tests at the significance level of p = 0.05 to compare the results between symptomatic and control groups.
We recorded 7 joints with opening clicking and 6 joints with reciprocal clicking (Table 1  ). Typical examples of reconstructed joints and stress-field paths from two clicking and one asymptomatic TMJ during jaw opening/closing are shown in the Figs. In clicking TMJ #03, the stress-field paths differed markedly between opening and closing, moving mainly in the medial joint portion during jaw opening and mainly in the lateral joint portion during closing (mean s = 3.8 ± 0.9 mm, smax = 6.9 ± 1.8 mm) (Fig. 1). The clicking TMJ #11 had mean s = 2.0 ± 0.5 mm and smax = 4.3 ± 0.3 mm (Fig. 2A). For comparison, in asymptomatic TMJ #17 the stress-field paths were almost perfectly coincident and moved in the lateral part of the joint (mean s = 0.6 ± 0.2 mm, smax = 1.1 ± 0.4 mm) (Fig. 2B).
The stress-field translation ( D) and the aspect ratio (a/h) were significantly greater in clicking TMJs than in controls (p < 0.05); a higher level of significance was found between the overall mean and maximum spread (s and smax) of the clicking TMJs and the controls (p < 0.001). Finally, the average peak mediolateral velocity of the stress-field centroids (Vp) and the overall stress-field area (A) did not differ statistically between the two groups (Table 2). Neither the subgroups with partial or total anterior displacement nor the subgroups with opening and reciprocal clicking showed statistical differences in any parameters. The parameter values of the single joints are given in the Appendix.
This study showed that mediolateral stress-field translation through the cartilage matrix also occurs in clicking TMJs during jaw/opening closing and is significantly larger than in asymptomatic joints. Moreover, further stress-field characteristics differed from those of asymptomatic joints. In particular, the stress-field paths in clicking joints were mostly non-coincident during the opening and closing phases, whereas they coincided in controls, which was reflected by the significantly larger average and maximum spread of the paths in the clicking TMJs. Also, the aspect ratios in clicking joints were significantly larger than in controls, which indicates a smaller penetration of the stress-field in the cartilage. The only non-differing parameters were stress-field areas and peak mediolateral velocities. In clicking TMJs, selected on the basis of clinical examinations, we found, from MRI, an almost equal occurrence of partially anteriorly displaced and totally anteriorly displaced discs, the only normal disc position being in TMJ #12, which conversely had a variation of the fossa shape. Almost all condyles exhibited some type of concavity. There was also an almost even occurrence of opening and reciprocally clicking joints. The stress-field parameters, however, were not sensitive to any of these subdivisions. It was not possible to analyze the stress-field characteristics related to gender, because of the difficulty in recruiting subjects for the whole examination protocol. In discussing the results, one must consider that our method locates TMJ stress-fields from the spatial relationship between the osseous surfaces, using a novel combination of accurate static and dynamic in vivo measurements not commonly available. Similar information is still unavailable for other synovial joints, although congruity variation and contact analysis between articular surfaces in other human and animal joints have been studied (Oberlander, 1978; Scherrer and Hillberry, 1979; Eckstein et al., 1994). Nonetheless, most of these previous investigations were performed on cadavers (Fujikawa et al., 1983; Warner et al., 1998) or only under static conditions (Reichle and Snaps, 1999), and therefore lack the active and spontaneous kinematic component peculiar to our method. An exact true 3D dynamic analysis of the soft-tissue deformation in the TMJ with in vivo data is still unachievable, since technical hurdles arise, especially from transient effects such as disc reduction. However, our findings seem to indicate that, in addition to functioning less regularly, these symptomatic TMJs—selected only for the presence of the clicking symptom—might be undergoing or have also undergone some sort of structural change. As a matter of fact, the greater mediolateral translation and the the non-reciprocal traces of the stress-fields suggest that greater areas of the joint soft tissues are subject to stresses than in asymptomatic joints. These larger areas, plowed by the stress-fields, could be explained either by the non-reciprocal movement of a displaced disc or by muscular discoordination that would cause unstable or different opening and closing condylar paths. More interesting is the significantly greater aspect ratio a/h in clicking TMJs, which at first seems only to imply smaller plowing forces than in healthy joints. If the average stress-field area is also considered, however, the increase in the aspect ratio suggests a general decrease in the intra-articular distance in clicking joints, and possibly a thinning of the disc, since stress-field areas do not differ between asymptomatic and symptomatic joints. This finding is in concordance with the observation that anteriorly displaced discs often appear deformed and partially mediolaterally stretched, as seen in MRI (Chen et al., 2002). The increased aspect ratio also suggests a possible flattening of the bony surfaces that could be a consequence of the discal and condylar remodeling already observed in static MR images (Rao et al., 1990). Of course, a limitation of this analysis is the fact that the aspect ratio was averaged over the total opening/closing movements. Since the paths of clicking joints are less regular than in normal joints, it is likely that the aspect ratio varies more within the opening/closing path in these than in normal joints. An analysis with a system with higher time and spatial resolution, which is in preparation in our laboratory, is necessary for a more detailed analysis of the variation of the aspect ratio. The current study was a basic investigation on the dynamic variations of the intra-articular space in clicking TMJs. The results suggest that the stress-field areas in these joints were influenced by dislocated and deformed disc tissue, and the congruity of the articular surfaces changed with respect to normal joints. Future investigations must attempt to characterize the stress-field traces relative to the anatomical variability, i.e., condyle and fossa shape, possibly detecting potential risk situations that could arise, e.g., from extreme shear stresses and strains through the disc. For this purpose, a deeper insight could be gained by biphasic finite element analysis, already successfully implemented with the use of data provided by our method (Donzelli et al., 2004).
This work was entirely supported by the standard financial plan of the University of Zürich. The authors thank Ms. Diana Dembski, medical technical assistant, Dept. KFS, Dental School, University of Zürich, for her invaluable help in preparing the splints and collecting the data.
A supplemental appendix to this article is published electronically only at http://www.dentalresearch.org. Received for publication June 19, 2003. Revision received March 8, 2004. Accepted for publication March 16, 2004.
Journal of Dental Research, Vol. 83, No. 6,
480-484 (2004)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
