Condylar volume and surface in Caucasian young adult subjects

The 3-D CBCT scans of 300 temporomandibular joints (TMJ) in 150 Caucasian young adults subjects (mean age 19.2 years; range: 15-29; 74 males and 76 females), who did not show pain or dysfunction of TMJ, and referred to the public clinic of head and facial medicine for orthodontic evaluation, were examined. The selection of subjects was made among those who did not show pain or dysfunction of TMJ [10] at the first orthodontic visit.

From a clinical point of view, we evaluated whether there was spontaneously evoked pain, localized in the muscles of mastication, the preauricular area, and/or the TMJs, which may be related to trauma (such as a blow to the face), interanal derangement, inflammatory or degenerative arthritis, or by the mandible being pushed back towards the ears whenever the patient chews or swallows. The muscles around the TMJ used for chewing were palpated, and head and neck pain were evaluated. In addition, difficulty opening the mouth normally was evaluated. Then, earache, headache, and facial pain were evaluated. Finally, limited or asymmetric jaw movement and joint sounds (i.e., clicking, popping, grating, crepitus) in the temporomandibular joint were evaluated to exclude TMJ dysfunction signs.

The CBCT scans were taken as orthodontic diagnostic tools. In our orthodontic department, patients who come for evaluation are asked to take the traditional radiographs (i.e., lateral skull radiographs, postero-anterior skull radiographs, and orthopantomographs) or, alternatively, to make a CBCT acquisition, avoiding the three radiographs. The sample included in this retrospective study was selected among those patients who chose to have CBCT scans.

The protocol was evaluated by the Ethical Committee of the University of Chiati, that stated no ethical approval of the protocol is necessary for this retrospective study, as patients have gone spontaneously to the Department for their medical evaluation of the orthodontic problem. Only a signed informed agreement was obtained from each subject, by the radiological study, to know and approve the medical procedure.

The evaluation of CBCT data included:

Cone Beam Computerized Tomography (CBCT) data sets were acquired using ILUMA™ (IMTEC, 3 M Company, Ardmore, Oklahoma, USA), with a reconstructed layer thickness of 0.1 mm and a 512 × 512 matrix. The device was operated at 120 kVp and 3-8 mA using a high-frequency generator with a fixed anode and a 0.5-mm focal spot. A single 40-s (because the complete volume of the head was taken) high-resolution scan of each skull was made with a voxel size (mm³) set at 0.25 mm³, and a 17.0 mm diameter and 13.2 cm field of view.

All CBCT images were taken with the subjects sitting in an upright position, with their backs as nearly perpendicular to the floor as possible. The head was always stabilized with ear rods in the external auditory meatus. The subjects were instructed to look into their own eyes in a mirror 1,5 m in front of them to obtain the natural head position.

Image segmentation of the anatomic structures of interest based on 2-D Digital Imaging and Communications in Medicine (DICOM) formatted data provided different planes of view as well as three-dimensionally reconstructed volumes using Mimics™ 9.0 software (Materialise NV Technologielaan, Leuven, Belgium) (Figure 1).

The 3-D reconstruction of the condyle requires the mandibular condyle to be separated in all the three planes of space from all the other structures, mostly the soft tissues.

Each condyle was visualized in the recommended range of bone density (range of gray scale from -1350 to 1650) and segmented using an adaptive threshold, which was visually checked prior to making 3-D and volumetric measurements.

Specifically, after enlargement of the TMJ area, the remaining surrounding structures were progressively removed using various sculpting tools for the upper, lower, and side condylar contours (cortical bone), as shown in Figure 1. The segmentation was made on coronal views, and the superior, inferior, and lateral limits of the condyle were standardized (Table 1 and Figure 2a-b). On the coronal views, the superior contour of the condyle was defined where the first radiopaque point was viewed in the image depicting the synovia; the lateral contours for each section were easily identified through clear visualization of the cortical bone (Figure 2a). The inferior contour of the condyle was traced where its section passed from an "elipsoidal" shape (owing to the presence of anterior crest of the condylar head) to a more "circular" shape (suggesting that the view was at the level of the condylar neck) (Figure 2b). Accordingly, the condyle CBCT data sets were segmented with a dedicated Mimics™ tool to construct a new mask that included only the mandibular condyle (Figures 1 and 3). After the condylar segmentation, 3-D multiplanar reconstructions were produced (Figure 1), and volumetric (mm³) and surface measurements (mm²) were made for each condyle through the Mimics™ automatic function (Figure 1)

To assess the intra-operator error, the CBCT data of 10 patients were processed by the same operator (S.T.) twice (one week apart [11]) and the differences in the condylar volumes and condylar surfaces were evaluated using the Wilcoxon Signed Rank test, because of the small number of subjects included in the study of intra-operator method error. The intra-operator method error was expressed with the formula:

Where X1 and X2 are the two measurements and n is the number of double measurements.

The volume and surface measurements were processed and analyzed using SPSS 14.0 (SPSS Inc, Rainbow Technologies, Chicago, Ill). For the three variables, the mean, standard deviation, range, minimum, maximum, 95% confidence interval for mean and the results for normality distribution (Kolmorgorov-Smirnov Z test) were calculated, based on gender and TMJ side (Table 2 and Table 3).

As all the data were normally distributed (Table 2), the t-test for independent samples was employed to calculate the statistically significant differences between the males and the females (Table 4) and the paired samples t-test to calculate the statistically significant differences between the right and left sides (Table 4).

Finally, a multiple regression analysis using condylar volume as the dependent variable was performed to assess the influence of side, gender, and condylar surface, as a multiple factor. For each test P was set at 0.05 value.

The mean difference between the means of condylar volume in males and females is 21.61 mm³, which is about the 3.3% of the mean condylar volume in the whole sample. This difference was statistically significant (Table 4).

The difference between the means of condylar surface in males and females is 11.25 mm², which is about 2.8% the condylar surface of the whole sample), but not statistically significant (Table 4).

These differences expressed as percentages respect to the means found in the whole sample, are in accordance with those of a recent study that investigated the female-to-male proportions in head and facial linear dimensions, and found a mean difference of 3-5% in the frontal and lateral views in young and adult patients, between males and females [20].

The differences in volume or surface suggest high variability among the subjects. But, it is not clear the clinical relevance, as no subject included in this report had any pain or dysfunction of TMJ, but only malocclusions.

For the variable MI that indicates the ratio between volume and surface, the difference between females and males is 0.018, which corresponds to about 1% of the mean MI in the whole sample. There is not a statistically significant difference between males and females in the MI values (Table 4).

These data confirm that sexual dimorphism concerns mostly with condylar volume, rather than with condylar surface or the Morphological Index, almost in subjects with malocclusions.

The mean difference between the means of condylar volume in the right and the left TMJ is 25.39 mm³, which is about the 3.9% of the mean condylar volume in the whole sample (p < 0.01)

The difference between the means of condylar surface in the right and the left TMJ is 17.15 mm², which is about 5.45% the condylar surface of the whole sample (p < 0.01).

The clinical relevance of this high variability in the mean differences is yet to be understood, considering that no subject included in the sample had any pain or dysfunction of TMD, but only malocclusions. Consequently, the observed anatomical differences between the two condyles cannot be related to the subject's predisposition to develop any specific pathology on one side, rather than on the other. However, these types of associations are better evaluated through longitudinal studies.

In conclusion, it is hard to confirm whether the dimensions measured on the images correspond to the real dimensions of the structures, because the data obtained here cannot be compared with the anatomical truth. Nevertheless, it is attempted to interpret these data on the basis of the asymmetry that is normal to all human-body structures.

A comparison between the results of this study and those of a recent study (based on 30 subjects with class I malocclusion [aged 15-30 years]), reveals differences, though not statistically significant, in the size of the mandibular condyle between the right and left sides: [21] mean of medio-lateral condylar diameter is higher on the right side than that on the left side. However, the contrary is true for the antero-posterior condylar diameter.

Besides, other variables concerning TMJ morphology also show significant difference between the two sides. The depths of the mandibular fossae and the mean anterior and posterior joint spaces are higher for the right TMJs than those of the left TMJs; however, the contrary is true for the mean superior joint space. As regards the posterior joint space, the difference between the two sides is significant. These observations corroborate the present finding of a physiologically significant asymmetry of condylar size, in almost all subjects with malocclusion.

Also, in a study conducted on eight cadaveric heads (16 dissected mandibular condyles) to investigate the position of auriculo-temporal nerve in relation to the mandibular condyle, the nerve was observed to have a closed anatomic relationship with the condyle; however, although it was in direct contact with the condylar neck in all the cases, there was no correlation between the positions of the nerves in the right and left sides, suggesting marked asymmetry in the right and left condylar anatomy [22].

The asymmetry observed in the present study between the right and left sides can also be related to the presence of a preference side for mastication in subjects with malocclusion.

According to literature, the most significant morphologic alterations and positioning asymmetries of TMJ structures are related to the absence of teeth, dental abrasion, premature occlusal contact points, functional mandibular deviations, unilateral posterior crossbites, and dentoskeletal asymmetries. Specifically, the rule of articular cartilage - a growth center - has been demonstrated to respond to the degenerative changes and nonphysiological strain in the joint areas (application of soft diet or extractions), through changes in the thicknesses of single cartilage layers and total layer thickness, causing a change in the vertical dimensions and width, which is manifested by changes in the maturation processes of centrally unloaded cartilage sections in rats [23]. With regard to the condylar volume, an earlier study demonstrated that hyperplasia of mandibular condyle is histologically characterized by the presence of an uninterrupted layer of undifferentiated germinative mesenchymal cells, a layer of hypertrophic cartilage, and the presence of islands of chondrocytes in the subchondral trabecular bone [24]. The present study suggests that about 4-6% of difference in volume and surface can be observed between the two sides of subjects with malocclusion and without pain or dysfunction in TMJs, along with a wide range of percentages in physiological asymmetry. It has also been shown that the shape and linear dimensions of the mandibular condyle are highly variable [25].

Thus, in the light of the wide range of percentages of asymmetry between the right and left sides, these clinical variables do not seem to be accurate or clinically useful to determine the existence of an abnormal condition in a subject with malocclusion and without pain of dysfunction in TMJs.

Instead, MI could be a better indicator, because no statistically significant difference is found in the MI values between the right and the left sides, or between males and females, in young adult Caucasian subjects with malocclusion and without pain or dysfunction in TMJs.

In young adult Caucasian subjects with malocclusion and without pain or dysfunction in TMJs, its range of values seem to be (both in males and females; both in the right and left sides) 1.72 ± 0.17 mm, with a total range from 1.15 to 2.15 mm.

One can interpret the data on volume and surface of condyle from a clinical point of view. But, before doing so the clinician must know that (i) the functional loads applied to the TMJ might influence TMJ's morphology, (ii) the shape and function are intimately related, although this concept is given due importance only in studies on class II and class III skeletal patterns, [26] and (iii) both the volume and the surface of a condyle differ between the right and left sides, in subjects with malocclusion and without pain or dysfunction in TMJs.

With the advent of 3-D CBCT scan, the clinician can request the radiologist to directly evaluate or calculate condylar volume and surface, as also the MI, using dedicated software.

If the standards of MI are established, then the clinician can use them to directly establish the presence or absence of an abnormal shape.

It would also be useful to evaluate the abnormality in the TMJ morphology of a specific patient, based mostly on the fact that no difference exists in the MI score between males and females or between right and left TMJs of almost all Caucasian young adult subjects, suggesting that it must be between 1.15 and 2.15 in subjects with malocclusion and TMJ without pain or dysfunction.

Numerous factors should be considered in applying the results of this investigation to clinical situations. The 3-D volumetric depiction depends on the appropriateness of segmentation, the thresholding of bone pixel values, and the accurate suppression of the surrounding tissue values to enhance the structure of interest. The depiction is dependent on the software algorithm, the spatial and contrast resolution of the scan, the thickness and degree of calcification or cortication of bone structure, and the technical skill of the operator. The Mimics software used in this study enables semi-manual segmentation by interaction of the operator with the data to produce a visually acceptable 3-D rendering. According to Periago (2008), [27] these limitations cause deficiencies or voids in the surface of the image, which occur in regions represented by few voxels or have gray values still representing the bone, but outside the threshold. These areas include the cortical bone of the mandibular condyle, and thus may lead to greater identification error (e.g., for condylar contours) and consequently to measurement error. However, no significant difference is found between the intra-observer method errors, thus suggesting that an accurate procedure of segmentation can occurr.