Association between the Temporomandibular Joint Morphology and Chewing Pattern

This study aimed to investigate whether the morphology of the temporomandibular joint (TMJ) is associated with chewing patterns while considering skeletal morphology, sex, age, and symptoms of temporomandibular disorder (TMD). A cross-sectional observational study of 102 TMJs of 80 patients (age 16–40 years) was performed using pretreatment records of cone-beam computed tomography imaging of the TMJ, mandibular kinesiographic records of gum chewing, lateral and posteroanterior cephalometric radiographs, patient history, and pretreatment questionnaires. To select appropriate TMJ measurements, linear regression analyses were performed using TMJ measurements as dependent variables and chewing patterns as the independent variable with adjustment for other covariates, including Nasion-B plane (SNB) angle, Frankfort-mandibular plane angle (FMA), amount of lateral mandibular shift, sex, age, and symptoms of TMD. In multiple linear regression models adjusted for other covariates, the length of the horizontal short axis of the condyle and radius of the condyle at 135° from the medial pole were significantly (p < 0.05) associated with the chewing patterns in the frontal plane on the working side. “Non-bilateral grinding” displayed a more rounded shape of the mandibular condyle. Conversely, “bilateral grinding” exhibited a flatter shape in the anteroposterior aspect. These findings suggest that the mandibular condyle morphology might be related to skeletal and masticatory function, including chewing patterns.


Introduction
The temporomandibular joint (TMJ) is the most unique joint in the body, as the two joints must move in synchronism. The TMJ can develop morphological bone remodeling as a result of various factors such as masticatory activities, aging, temporomandibular disorder (TMD), and occlusal changes [1][2][3][4]. Previous studies have found that the size and shape of the TMJ are associated with skeletal morphology, such as the size of the mandible and vertical growth patterns [5]. The concept that the TMJ is a component of an articular triad, with the TMJ providing two points of contact and dentition providing the third, regulating mandibular movement, has been widely broadcast. Consequently, changes in mandibular movement may be linked to alterations in the structure of the TMJs and occlusion [6].
Previous studies have demonstrated that different skeletal patterns show significant differences in condylar morphology, joint space, joint-fossa morphology, and condylar position [7,8]. Patients with a larger Frankfort-mandibular plane angle (FMA) had a smaller linear measurement of the mandibular condyle size, although the difference was not statistically significant [5,9]. Mendoza et al. [5] and Nakawaki et al. [10] discovered that In this study, we examined the association between TMJ morphology and bilateral chewing patterns. The independent variable, bilateral chewing patterns, and the dependent variable, the morphology of the TMJ, were adjusted for skeletal morphology, sex, age, and symptoms of TMD as covariates in the multiple linear regression models.
An a priori power analysis showed that at a 95% confidence level, a sample size of 103 would provide an 80% probability of demonstrating a medium effect (f = 0.15) for the association between TMJ morphology and chewing patterns using a multiple linear regression model after adjusting for seven independent variables in total. Patients with craniofacial anomalies, major head and neck injuries, or a history of head and neck surgery were excluded from the study. According to our TMJ side selection criteria, 58 TMJs on the deviated side of patients with mandibular shift and 22 on the left, and 22 on the right side of those without mandibular shift were selected. In total, 102 TMJs were used in this study.

Measurements of TMJ
CBCT was used to examine the three-dimensional morphology of the TMJ. The patients were seated in a posture in which the Frankfort horizontal plane was horizontal and the jaws were in centric occlusion when CBCT images were taken. For image reconstruction, a single 360 • scan was used to collect projection data from a cylinder (height, 30 mm; diameter, 40 mm); 80 kV, 7 mA, and 17 s. The reconstructed slices were 1 mm thick. TMJ morphology was assessed using three-dimensional analysis software (AW server 3.2 Ext 4.0, GE Healthcare, Chicago, IL, USA). The measurements of TMJ were performed in the horizontal, coronal, and sagittal planes, as shown in Figure 1, which was modified from previous studies [2,7,19]. After constructing the three-dimensional rendered image, the measurement of the TMJ started by scrolling through the horizontal slices and then selecting the slice with the long axis of the condyle in the horizontal plane (HLC). Then, the short axis of the condyle in the horizontal plane (HSC) perpendicular to the HLC and the horizontal condylar angle (HCA), which was measured from a reference line perpendicular to the midsagittal plane, was measured ( Figure 1A). Subsequently, the slice in the center of the mandibular condyle was selected as coronal slices parallel to the HLC and sagittal slices perpendicular to the HLC. In the selected coronal slice, we measured the long axis of the condyle in the coronal plane from the medial pole to the lateral pole, the radius of the condyle at 45 • , 90 • (C90), and 135 • from the medial pole (C135), joint space at 45 • , joint space at 90 • , and the joint space at 135 • ( Figure 1B). Finally, we measured the height of the condyle in the sagittal plane (SHC), the depth of the condyle in the sagittal plane (SDC), anterior joint space, posterior joint space, and articular eminence inclination from the selected sagittal slice ( Figure 1D). To evaluate intraobserver reliability, the same observer repeated all measurements after six weeks, blinded to the previous measurements. A different observer evaluated the left and right TMJs of 30 patients and compared the results to the first observer's measurement to evaluate interobserver reliability. In this study, we examined the association between TMJ morphology and bilateral chewing patterns. The independent variable, bilateral chewing patterns, and the dependent variable, the morphology of the TMJ, were adjusted for skeletal morphology, sex, age, and symptoms of TMD as covariates in the multiple linear regression models.
An a priori power analysis showed that at a 95% confidence level, a sample size of 103 would provide an 80% probability of demonstrating a medium effect (f = 0.15) for the association between TMJ morphology and chewing patterns using a multiple linear regression model after adjusting for seven independent variables in total. Patients with craniofacial anomalies, major head and neck injuries, or a history of head and neck surgery were excluded from the study. According to our TMJ side selection criteria, 58 TMJs on the deviated side of patients with mandibular shift and 22 on the left, and 22 on the right side of those without mandibular shift were selected. In total, 102 TMJs were used in this study.

Measurements of TMJ
CBCT was used to examine the three-dimensional morphology of the TMJ. The patients were seated in a posture in which the Frankfort horizontal plane was horizontal and the jaws were in centric occlusion when CBCT images were taken. For image reconstruction, a single 360° scan was used to collect projection data from a cylinder (height, 30 mm; diameter, 40 mm); 80 kV, 7 mA, and 17 s. The reconstructed slices were 1 mm thick. TMJ morphology was assessed using three-dimensional analysis software (AW server 3.2 Ext 4.0, GE Healthcare, Chicago, IL, USA). The measurements of TMJ were performed in the horizontal, coronal, and sagittal planes, as shown in Figure 1, which was modified from previous studies [2,7,19]. After constructing the three-dimensional rendered image, the measurement of the TMJ started by scrolling through the horizontal slices and then selecting the slice with the long axis of the condyle in the horizontal plane (HLC). Then, the short axis of the condyle in the horizontal plane (HSC) perpendicular to the HLC and the horizontal condylar angle (HCA), which was measured from a reference line perpendicular to the midsagittal plane, was measured ( Figure 1A). Subsequently, the slice in the center of the mandibular condyle was selected as coronal slices parallel to the HLC and sagittal slices perpendicular to the HLC. In the selected coronal slice, we measured the long axis of the condyle in the coronal plane from the medial pole to the lateral pole, the radius of the condyle at 45°, 90° (C90), and 135° from the medial pole (C135), joint space at 45°, joint space at 90°, and the joint space at 135° ( Figure 1B). Finally, we measured the height of the condyle in the sagittal plane (SHC), the depth of the condyle in the sagittal plane (SDC), anterior joint space, posterior joint space, and articular eminence inclination from the selected sagittal slice ( Figure 1D). To evaluate intraobserver reliability, the same observer repeated all measurements after six weeks, blinded to the previous measurements. A different observer evaluated the left and right TMJs of 30 patients and compared the results to the first observer's measurement to evaluate interobserver reliability.

Chewing Movement
The pretreatment path of the mandibular incisors of the patients was recorded using a mandibular kinesiograph (MKG) system (K7-I craniomandibular evaluation system, Myotronics-Noro Med, Seattle, WA, USA). MKG recordings of gum chewing on the left and right sides in the frontal plane were utilized using a modified method described in previous studies to determine the main chewing patterns as grinding, chopping, reversed, or crossover, as shown in Figure 2 [14,15]. Grinding chewing patterns have a more lateral form than other chewing patterns, with an opening toward the nonworking side and a closing from the working side, and the opening and closing paths are wide apart in the frontal plane. Chopping, reversed, or crossover chewing patterns are classified as abnormal chewing patterns that have a mixed or ambiguous sense of direction and end with a form that has reversed sequencing.
Those who showed grinding patterns on both sides were defined as the "Bilateral grinding" group, whereas those who showed chopping, reversing, or crossing patterns on at least one side were defined as the "Non-bilateral grinding" group, as shown in Figure 3. The main chewing patterns are determined as grinding, chopping, reversed, or crossover chewing patterns after each chewing stroke is evaluated. Chewing pattern group "Bilateral grinding": grinding patterns on both sides. Chewing pattern group "Non-bilateral grinding": with chopping, reversed, or crossover patterns on at least one side.

Figure 2.
Grinding chewing patterns have a more lateral form than other chewing patterns, with an opening toward the nonworking side and a closing from the working side, and the opening and closing paths are wide apart in the frontal plane. Chopping, reversed, or crossover chewing patterns are classified as abnormal chewing patterns that have a mixed or ambiguous sense of direction and end with a form that has reversed sequencing.

Chewing Movement
The pretreatment path of the mandibular incisors of the patients was recorded using a mandibular kinesiograph (MKG) system (K7-I craniomandibular evaluation system, Myotronics-Noro Med, Seattle, WA, USA). MKG recordings of gum chewing on the left and right sides in the frontal plane were utilized using a modified method described in previous studies to determine the main chewing patterns as grinding, chopping, reversed, or crossover, as shown in Figure 2 [14,15].

Figure 2.
Grinding chewing patterns have a more lateral form than other chewing patterns, with an opening toward the nonworking side and a closing from the working side, and the opening and closing paths are wide apart in the frontal plane. Chopping, reversed, or crossover chewing patterns are classified as abnormal chewing patterns that have a mixed or ambiguous sense of direction and end with a form that has reversed sequencing.
Those who showed grinding patterns on both sides were defined as the "Bilateral grinding" group, whereas those who showed chopping, reversing, or crossing patterns on at least one side were defined as the "Non-bilateral grinding" group, as shown in Figure 3. The main chewing patterns are determined as grinding, chopping, reversed, or crossover chewing patterns after each chewing stroke is evaluated. Chewing pattern group "Bilateral grinding": grinding patterns on both sides. Chewing pattern group "Non-bilateral grinding": with chopping, reversed, or crossover patterns on at least one side. The main chewing patterns are determined as grinding, chopping, reversed, or crossover chewing patterns after each chewing stroke is evaluated. Chewing pattern group "Bilateral grinding": grinding patterns on both sides. Chewing pattern group "Non-bilateral grinding": with chopping, reversed, or crossover patterns on at least one side.

Skeletal Morphology, Age, Sex, and Symptoms of TMD
Lateral and posteroanterior (PA) cephalometric radiographs were obtained using a cephalostat (Axiom Aristos VX, Siemens, Munich, Germany). Tracings of lateral cephalograms were used to evaluate the angle between the Sella-Nasion plane and the Nasion-B plane (the SNB angle), which assesses the anteroposterior position of the mandible relative to the upper cranial structures and the FMA ( Figure 4A). The landmarks of the PA cephalogram were identified using the methods proposed by Sassouni to evaluate the lateral mandibular shift ( Figure 4B) [20].

Skeletal Morphology, Age, Sex, and Symptoms of TMD
Lateral and posteroanterior (PA) cephalometric radiographs were obtained using cephalostat (Axiom Aristos VX, Siemens, Munich, Germany). Tracings of lateral cephalo grams were used to evaluate the angle between the Sella-Nasion plane and the Nasion-B plane (the SNB angle), which assesses the anteroposterior position of the mandible rela tive to the upper cranial structures and the FMA ( Figure 4A). The landmarks of the PA cephalogram were identified using the methods proposed by Sassouni to evaluate the lat eral mandibular shift ( Figure 4B) [20]. Data on age, sex, and history of symptoms of TMD were collected from the patients histories and pretreatment questionnaires. The sex was listed as either male or female, and the age of each individual was determined using the number of complete years. The his tory of TMD included sounds, pain, and trismus. Patients were categorized as having o not having a history of TMD.

Statistical Analysis
Statistical analysis was based on the first measurement of the TMJ. The collected dat were coded and analyzed using STATA software (Stata Statistical Software version 17.0 Stata Corp LP, College Station, TX, USA). The intra-and interobserver reliabilities of al TMJ measurement parameters were evaluated using a one-sample t-test. Univariate anal ysis was conducted between different measurements of the TMJ and chewing patterns t identify variables (p < 0.2) [21], which were included in the multiple linear regression analysis. In the multiple linear regression models, the dependent variable was TMJ mor phology, whereas the independent variables were bilateral chewing patterns, the SNB an gle, the FMA angle, the amount of lateral mandibular shift, sex, age, and symptoms o Data on age, sex, and history of symptoms of TMD were collected from the patients' histories and pretreatment questionnaires. The sex was listed as either male or female, and the age of each individual was determined using the number of complete years. The history of TMD included sounds, pain, and trismus. Patients were categorized as having or not having a history of TMD.

Statistical Analysis
Statistical analysis was based on the first measurement of the TMJ. The collected data were coded and analyzed using STATA software (Stata Statistical Software version 17.0, Stata Corp LP, College Station, TX, USA). The intra-and interobserver reliabilities of all TMJ measurement parameters were evaluated using a one-sample t-test. Univariate analysis was conducted between different measurements of the TMJ and chewing patterns to identify variables (p < 0.2) [21], which were included in the multiple linear regression analysis. In the multiple linear regression models, the dependent variable was TMJ morphology, whereas the independent variables were bilateral chewing patterns, the SNB angle, the FMA angle, the amount of lateral mandibular shift, sex, age, and symptoms of TMD. The variables whose p-value was less than 0.05, from multiple linear regressions, were declared as statistically significant.

Three-Dimensional Reconstruction of TMJ
The CBCT images of TMJ from groups "Bilateral grinding" and "Non-bilateral grinding" were three-dimensionally reconstructed and observed as shown in Figure 5A,B. TMD. The variables whose p-value was less than 0.05, from multiple linear regressions, were declared as statistically significant.

Three-Dimensional Reconstruction of TMJ
The CBCT images of TMJ from groups "Bilateral grinding" and "Non-bilateral grinding" were three-dimensionally reconstructed and observed as shown in Figure 5A,B.

Statistical Analysis
Based on the criteria we established for selecting TMJ sides, a total of 58 TMJs located on the deviated side were chosen from patients exhibiting mandibular shift. Additionally, 22 TMJs were selected from the left side and 22 from the right side of individuals without mandibular shift. In total, the study utilized 102 TMJs to gather data and conduct analyses. The total sample size was 102 TMJs from 80 patients; the characteristics of each variable used in this study are listed in Table 1. All measurements of the TMJ showed no statistically significant difference between the intra-and interobserver reliabilities at p > 0.05, as shown in Table 2.

Statistical Analysis
Based on the criteria we established for selecting TMJ sides, a total of 58 TMJs located on the deviated side were chosen from patients exhibiting mandibular shift. Additionally, 22 TMJs were selected from the left side and 22 from the right side of individuals without mandibular shift. In total, the study utilized 102 TMJs to gather data and conduct analyses. The total sample size was 102 TMJs from 80 patients; the characteristics of each variable used in this study are listed in Table 1. All measurements of the TMJ showed no statistically significant difference between the intra-and interobserver reliabilities at p > 0.05, as shown in Table 2. "Bilateral grinding" denotes grinding patterns on both sides, while "Non-bilateral grinding" denotes chopping, reversed, or crossover patterns on at least one side. Abbreviations: SNB, the SNB angle; FMA, the Frankfortmandibular plane angle; TMD, temporomandibular disorders; SD, standard deviation. From the univariate analysis (Table 3), HSC, HCA, C90, C135, SHC, and SDC were the measurement items of TMJ that were eligible for the multiple linear regression models (p < 0.2). The details of each multiple linear regression model are listed in Table 4. The p-values of chewing patterns in each of the multiple linear regression models, including the averages and standard deviations of the TMJ measurements, are shown in Table 5.   There was a significant association among chewing patterns, amount of lateral mandibular shift, sex, age, and the FMA angle on HSC and C135 (p < 0.05) in multiple linear regression models adjusted for covariates. HSC was positively associated with the chewing pattern group "Non-bilateral grinding". Accordingly, when adjusted with the same covariates, C135 also showed a significant positive association with the chewing pattern group "Non-bilateral grinding" (p < 0.05). The association between the chewing patterns group and SDC was p = 0.066. The results of SDC with the chewing patterns group were not statistically significant, although they did coincide with the results of HSC and C135 ( Figure 6).
According to the multiple linear regression models, the size of the mandibular condyle was negatively associated with the FMA angle, the amount of lateral mandibular shift, and the female sex. The vertical dimensions of the mandibular condyle (C90, C135, and SHC) were positively associated with the SNB angle (Table 4). Measurement of the sagittal depth of condyle (SDC) in a patient from the group "Bilateral grinding" (E) Measurement of SDC in a patient from the group "Non-bilateral grinding" (F) Distribution of SDC of chewing pattern group "Bilateral grinding" and group "Non-bilateral grinding". Chewing pattern group "Bilateral grinding": grinding patterns on both sides. Chewing pattern group "Nonbilateral grinding": with chopping, reversed, or crossover patterns on at least one side. This figure illustrates the relationship between the anteroposterior aspects of the mandibular condyle (HSC and SDC) and chewing patterns when compared to the mediolateral aspect (HLC). The chewing pattern group "Non-bilateral grinding" had a notably rounder mandibular condyle, particularly in the horizontal slice, whereas the chewing pattern group "Bilateral grinding" had a flatter shape in the anteroposterior aspect.  Measurement of the sagittal depth of condyle (SDC) in a patient from the group "Bilateral grinding" (E) Measurement of SDC in a patient from the group "Non-bilateral grinding" (F) Distribution of SDC of chewing pattern group "Bilateral grinding" and group "Non-bilateral grinding". Chewing pattern group "Bilateral grinding": grinding patterns on both sides. Chewing pattern group "Nonbilateral grinding": with chopping, reversed, or crossover patterns on at least one side. This figure illustrates the relationship between the anteroposterior aspects of the mandibular condyle (HSC and SDC) and chewing patterns when compared to the mediolateral aspect (HLC). The chewing pattern group "Non-bilateral grinding" had a notably rounder mandibular condyle, particularly in the horizontal slice, whereas the chewing pattern group "Bilateral grinding" had a flatter shape in the anteroposterior aspect.

Discussion
To the best of our knowledge, this is the first study to demonstrate a significant association between TMJ morphology and bilateral chewing patterns in multiple linear regression models adjusted for the SNB angle, the FMA angle, amount of lateral mandibular shift, sex, age, and symptoms of TMD as covariates. The anteroposterior aspects of the mandibular condyle were the HSC and SDC, whereas the superolateral aspect was described by C135. The lengths of HSC and C135 were positively associated with the chewing pattern group "Non-bilateral grinding" when compared to the chewing pattern group "Bilateral grinding". Although the results of the SDC multiple linear regression model with bilateral chewing patterns and other covariates were not statistically significant (p = 0.066), they did correspond with those of HSC and C135 ( Figure 6). Figure 6 describes the relationship between the anteroposterior aspect of the mandibular condyle (HSC and SDC) and chewing patterns compared with the mediolateral aspect (HLC). The differences in the anteroposterior aspects were much greater. From the horizontal slice, the chewing pattern of the group "Non-bilateral grinding" presented a much rounder shape of the mandibular condyle, while the chewing pattern of the group "Bilateral grinding" presented a flatter shape in the anteroposterior aspect and the sagittal slice; the chewing pattern of the group "Non-bilateral grinding" exhibited a larger SDC compared to the chewing pattern of the group "Bilateral grinding". Multiple linear regression models demonstrated significant correlations between specific TMJ morphology measurement items and chewing patterns (p < 0.05). Furthermore, in the present study, significant correlations were reported between the size of the mandibular condyle and the SNB angle, the FMA angle, the amount of lateral mandibular shift, and sex.
The previous studies suggested that the structures of the TMJ on the nondeviated side of the jaw were larger than those on the deviated side in patients with lateral mandibular shifts [11,12]. The deviated side of the TMJ was selected in our study and adjusted with the amount of lateral mandibular shift in the statistical model. Our findings are consistent with those of previous studies [11,12]. The nondeviated sides, which were larger than the deviated sides, were not included because this would cause the characteristics of the TMJ to become less distinctive. Preliminary studies revealed that most cases of large lateral mandibular deviation in our sample were skeletal Class III cases and that the condylar morphology of the nondeviated sides was large due to overgrowth. Therefore, we thought that including the mandibular condyles on the nondeviated sides would mask the morphological characteristics of the mandibular condyles on the deviated sides. One of the purposes of this study was to clarify the relationship between mandibular condyle morphology and gum chewing patterns on the deviated sides. Therefore, we decided not to include the mandibular condyles on the nondeviated sides in cases where mandibular deviation was evident.
The size of the TMJ on the deviated side was found to have a significant negative relationship with the amount of lateral mandibular shift (p < 0.05). The age range of the study population was confined to 16-40 years, and age was included as an independent variable in the multiple linear regression models. Correlations between morphological changes in the TMJ, sex, age, and symptoms of TMD are a major topic of discussion in various studies [3,4]. Sex, age, and symptoms of TMD are inevitably correlated independent variables that must be included in the statistical model. The prevalence of TMD was reported to be much higher in women, and symptoms of TMD have been reported to be related to age [4]. Thus, the coefficients of the independent variables may have been undermined.
According to Negishi et al. [22], chewing patterns can be altered through chewing exercises. Chewing gum was initiated 3 months after sagittal split ramus osteotomy and was performed for 5 min twice a day for 3 months. Following the exercise, the masticatory width increased significantly, suggesting a natural adaptation from narrow chewing patterns in the frontal plane to grinding chewing patterns as a result of morphological changes in the TMJ.
Patients with internal derangements of the TMJ had a significantly restricted range of chewing movement in both the lateral and vertical dimensions, as well as movement deceleration, compared to individuals without internal derangements of the TMJ [23,24]. Furthermore, different stages of internal derangement show varying degrees of aberrant chewing movement, according to Kuwahara et al. [23]: Patients with TMJ disc displacement without reduction were found to have the most chewing impairment, followed by TMJ disc displacement without reduction with perforation and TMJ disc displacement with a late reduction. However, no study has specified a standard criterion for diagnosing internal derangement of the TMJ.
Recent studies have shown that steepening of the condylar pathway or deepening of the articular eminence inclination is associated with the preferred chewing side, which has flatter lateral guidance or can be described as a grinding pattern compared with the opposite side [17,18,25]. We did not find a significant association between articular eminence inclination and chewing patterns in our study. The different results could be due to different focus points in bilateral chewing patterns or the preferred chewing side. According to our findings, an increase in the lateral width of the chewing pattern was significantly associated with a decrease in the anteroposterior and superolateral aspects of the mandibular condyle. This suggests that mandibular condyle remodeling may occur primarily in the anteroposterior and superolateral aspects, as the chewing patterns may shift to a grinding pattern with greater lateral width, either as a physiological adaptation or as a result of pathological bone resorption.
Balcioglu et al. [26] discovered that in patients with the preferred chewing side, the volumes of both the inferior and superior heads of the lateral pterygoid muscle on the affected side were significantly greater than those on the unaffected side. The lateral pterygoid muscle, which is essential for facilitating mastication and mandibular movement, is attached to the mandibular condyle. Recent anatomical investigations have focused on the insertion area of the lateral pterygoid muscle, not only the anterior aspect of the pterygoid fovea but also the medial aspect of the mandibular condyle [27,28]. While the present study's findings suggest that mandibular condyle remodeling occurs largely in the anteroposterior and superolateral aspects, since the muscle force vector of the lateral pterygoid muscle in mandibular movements is defined by its origin and insertion point, this circumstance may affect the function of the lateral pterygoid muscle.
Another challenge is constructing multivariable statistical models and selecting independent variables. A careful and critical selection of variables is crucial to avoid multicollinearity among the independent variables [29]. The independent variables were chewing patterns and the dependent variable was the morphology of the TMJ while adjusting for the SNB angle, the FMA angle, the amount of lateral mandibular shift, sex, age, and symptoms of TMD as other covariates in the statistical models. These independent variables were biologically relevant factors that were known or based on a scientific hypothesis. The models included the SNB angle to represent the anteroposterior position of the mandible in relation to the upper cranial structures, and the FMA angle to represent the vertical growth pattern. Since the SNB angle, as one of the independent variables, has a high association with measurements related to dental occlusion such as Angle's classification, overjet, and overbite, we did not include direct dental occlusion parameters in our statistical model. Table 2 presents the results of the analysis conducted to assess the intra-and interobserver reliabilities of the TMJ measurements. The findings reveal that there were no statistically significant differences observed between the levels of reliability for both intra-and interobserver assessments, as indicated by p-values greater than 0.05. This implies that the agreement or consistency among observers and the consistency over time for the same observer were comparable for all TMJ measurements evaluated in the study. Therefore, the data in Table 2 suggest that the TMJ variables examined exhibited similar levels of reliability regardless of whether they were assessed by the same observer on multiple occasions or by different observers independently. The findings of this study do not represent the general population because the samples were collected from individuals seeking orthodontic treatment. The study population and the estimated sample size from the power analysis were comparable.
The new findings demonstrated that there was a significant correlation between TMJ morphology and chewing patterns, even when considered with other covariates in multiple linear regression models, contributing to a better understanding of the TMJ in the fields of orthodontics and dentistry. Considering the previously reported association between occlusion and masticatory movement patterns [13][14][15][16], this study suggests that orthodontic treatment to improve occlusion may have some effect on TMJ morphology.
Further research is needed to clarify the possible mechanism, as well as whether this is a physiological adaptation, pathological condition, or merely a random coincidence [30]. Prospective studies with larger sample sizes and with an improvement in the studypopulation selection criteria are needed to clarify the cause-and-effect scenario of these correlations.

Conclusions
The findings of this study indicate that the anteroposterior and superolateral aspects of the mandibular condyle were significantly associated with chewing patterns in our statistical models. The findings of this study suggest that the morphology of the mandibular condyle may be associated with not only skeletal morphology but also masticatory function, such as chewing patterns that are related to occlusion and orthodontic treatment. Informed Consent Statement: Patient consent was waived due to a retrospective study of the registered database under the regulations of the Ethics Committee.

Data Availability Statement:
The data that support the findings of this study are available upon request from the corresponding author. The data are not publicly available due to privacy and ethical restrictions.