1. Introduction
The sensory inputs involved in human postural control are mainly visual, somatosensory, and vestibular [
1,
2,
3]. These sensory inputs are integrated into the central nervous system. Somatosensory input becomes a source of information about the relative positional relationship between the body and the basal plane, as well as the positional relationship between various parts of the body. Compared with cases where the support surface is unstable, somatic sensation is more important when a standing posture is maintained on a horizontal hard floor [
2]. Somatic sensations in the stomatognathic region involve organs such as the temporomandibular joint, tongue, teeth, and masticatory muscles and are presumably influenced by the jaw position, clenching, and occlusal contact state (i.e., occlusion) [
2,
4].
Occlusion may affect sensory input for postural control, and from a biomechanical perspective, it is assumed that it is influenced by the fascial chain. The fascia stabilizes bodily movements by transmitting tension from one unit to another, thereby contributing to maintaining posture [
5]. The masseter muscle belongs to the deep front line, and the sternocleidomastoid muscle belongs to the lateral line and superficial front line, so the activation of these muscles during clenching supports the stabilization of the body [
5,
6]. Although there are reports that occlusion does not affect postural control [
7], other studies have found that occlusion affects postural control only in difficult postural tasks and in environments where there is little contribution from other sensory inputs [
8], that the sensory information from occlusion exerts a large effect in unstable situations [
9], that postural control is affected by mandibular position whether the eyes are open or closed [
10], that individuals without malocclusion have higher balance test scores [
11], that clenching is influenced by neuromuscular co-contraction patterns [
12], and, finally, that changes in occlusion due to oral appliances affect dynamic stability [
13]. The studies that investigated the relationship between occlusion and postural control compared differences due to sagittal deviation (angle class) and/or transverse malocclusion (crossbite) [
11,
13] or between the mandibular rest position, intercuspal position, and different occlusal contacts on one or both sides, using, for example, oral appliances [
8], cotton rolls [
9,
11], and wax [
5]. However, regarding experimental changes in occlusion, there is no consensus regarding the influence of occlusion on postural control. One reason for the lack of consensus may be that the occlusal contact state of the subjects has not been investigated. In addition, when the occlusal position is changed using an oral device, it is difficult to distinguish whether the effect on postural control is due to a change in the occlusal height or due to a change in the occlusal contact state.
Postural stability is the ability to control the body’s center of gravity with respect to the base plane and is achieved through the sustained activity of antigravity muscles and postural reflexes [
14]. It is divided into static balance, which maintains a stable posture, and dynamic balance, which maintains dynamic equilibrium against changes in the base of support or center of gravity [
15]. Our previous work examined the relationship between the occlusal contact state and physical function [
6,
16]. The results revealed that individuals with good left–right balance of occlusal contacts have stable static balance and that static balance can be stabilized by correcting the occlusal contact condition using an intraoral appliance. However, these studies were limited by the small number of participants and the inability to compare groups with good and poor occlusal contact balance.
The purpose of the present study was to determine the stability of occlusal contacts based on the difference between the left and right occlusal contact areas, divide participants into well-balanced and unbalanced groups, and clarify the effect of light clenching (i.e., occlusion) on static and dynamic balance. The null hypothesis was that the effect of clenching on static and dynamic balance is not influenced by the stability of occlusal contacts.
2. Materials and Methods
This study was approved by the Ethics Committee of The Nippon Dental University School of Life Dentistry at Niigata and was conducted after the purpose of the study was fully explained to the participants and their informed consent was obtained (approval number: ECNG-R-443). This study was conducted according to the PRILE 2021 guidelines (
Figure 1) [
17].
2.1. Participants
Participants were 41 healthy men (age, 19.4 ± 1.1 years) with individual normal occlusion and no subjective or objective morphological or functional abnormalities in the stomatognathic system. The average participant height was 170.8 ± 4.6 cm, their average weight was 72.4 ± 5.1 kg, and their average body mass index was 23.6 ± 2.2. The exclusion criteria were individuals undergoing dental treatment, those with temporomandibular joint symptoms, those with missing teeth other than the third molar, those with otolaryngological disease, and those with a history of musculoskeletal pain, severe low back pain, or surgery on the lower limbs, spine, or pelvis within the past 12 months. Oral and temporomandibular joint examinations were performed by dentists, while others were based on participants’ self-reporting.
2.2. Measurement of Occlusal Contact State
The occlusal contact condition was measured using a pressure-sensitive film (Dental Prescale, 50H-R type; Fujifilm, Tokyo, Japan) and the manufacturer’s analyzer (Occluzer FPD-709; Fujifilm, Tokyo, Japan) according to previous reports [
6,
16,
18]. Participants with a left–right distribution of the occlusal contact area less than 10% were classified as the balanced occlusal contact (BOC) group, while those with a value of 10% or more were classified as the unbalanced occlusal contact (UOC) group (
Figure 2) [
18]. These groups comprised 22 and 19 participants, respectively.
2.3. Measurement of Balance Ability
Balance ability was measured using a center of gravity sway meter (GRAVICORDER GS-7, Anima Co., Ltd., Tokyo, Japan), following the manufacturer’s recommended measurement procedure. Participants with visual impairments wore glasses or contact lenses during the experiment.
Static balance was measured with the participants standing still with their eyes open, according to previous reports [
2,
6]. The outer peripheral area (ENV-area) and length of locus per unit area (LNG/E-area) calculated from the displacement of the center of foot pressure (COP) were recorded [
2,
6].
Dynamic balance was assessed using the cross test according to previous reports [
19,
20]. The participants were instructed to stand upright as a standard position and to move their upper bodies in the following order, taking 3 s for each: forward, standard position, backward, standard position, left, standard position, right, and back to standard position. The rectangular area (REC-area) calculated from the COP trajectory diagram was evaluated [
19,
20].
The measurement conditions were as follows: mandibular rest position (RP), in which the upper and lower lips were closed and the upper and lower teeth were not in contact; and intercuspal position, in which the jaw was lightly clenched in a position (LC) where the upper and lower teeth come into contact when the mouth is closed in the normal position. Measurements were performed three times for each condition. In advance, surface electromyography sensors (DSP wireless myoelectric sensor, Sports Sensing Co., Ltd., Fukuoka, Japan) were attached to the central part of both masseter muscles of the participants, and the participants were instructed to clench to the maximal voluntary contraction (100% MVC). Following previous research, light clenching was defined as 20% MVC [
21]. Participants were informed of the 20% MVC through visual feedback using a computer screen and were instructed to maintain this clenching strength throughout the measurement.
2.4. Statistical Analysis
Statistical analyses were performed with SPSS ver. 17.0 (SPSS Japan Inc., Tokyo, Japan). The Shapiro–Wilk test was used to test for normality while the Levene test was used to test for homogeneity of variance. Significance was set at p < 0.05.
Regarding the differences in balance ability between the BOC and UOC groups, Student’s t-test was used to compare two samples in which normality and homogeneity of variance were observed while the Mann–Whitney test was used to compare two samples including groups in which normality was not observed. Regarding the differences in balance ability due to clenching, a paired t-test was used to compare two variables in one sample where normality was observed while a Wilcoxon signed-rank test was used to compare one sample including variables where normality was not observed.
To examine the correlation between static balance and dynamic balance in the BOC and UOC groups, the correlations between the ENV-area and REC-area and between the LNG/E-area and REC-area were analyzed. When both variables showed normality, Pearson’s product moment correlation coefficient was used for analysis. When either variable did not show normality, Spearman’s rank correlation coefficient was used for analysis.
In each of the above analyses, post hoc tests were performed using the free software G-Power [
22]. In tests comparing two samples or one sample with two variables, the effect size was calculated from the mean and standard deviation. For the statistical tests, we selected [Difference between two independent means (two groups)] for Student’s
t-test, [Wilcoxon–Mann–Whitney test (two groups)] for the Mann–Whitney test, [Difference between two dependent means (matched pairs)] for the paired
t-test, and [Wilcoxon signed-rank test (matched pairs)] for the Wilcoxon signed-rank test. For each test, effect size, α error probability (0.05), sample size group or total sample size were entered, and power (1 − β error probability) was calculated. Effect size values for two-sample comparisons or one sample with two-variable comparisons were 0.2, 0.5, and 0.8 for small, medium, and large effect sizes, respectively [
23]. In the correlation analysis, the effect size was calculated from the coefficient of determination. The statistical test was selected as [Correlation: Point biserial model], and the α error probability was entered as 0.05 and the total sample size was entered to calculate the power (1 − β error probability). The effect sizes in correlation analyses were considered as 0.1, 0.3, and 0.5 for small, medium, and large effect sizes, respectively [
23]. The power of 0.8 or higher was determined to be a significant difference.
3. Results
Figure 3 shows a comparison of the ENV-area due to occlusal balance and clenching.
Table 1 shows the
p-value, effect size, and statistical power for each test. A significant difference was observed between the RP and LC conditions in the BOC group, with LC showing lower values (
p < 0.01, effect size; 0.690). No significant differences due to clenching and no differences due to occlusal balance were observed in the UOC group.
The LNG/E-area due to occlusal balance and clenching is compared in
Figure 4 while the
p-value, effect size, and statistical power for each test are shown in
Table 2. A significant difference was observed between the RP and LC conditions in the BOC group, with higher values for LC (
p < 0.01, effect size; 0.774). No significant differences due to clenching and no differences due to occlusal balance were seen in the UOC group.
A comparison of the REC-area due to occlusal balance and clenching is illustrated in
Figure 5.
Table 3 details the
p-value, effect size, and statistical power for each test. The RP and LC conditions were significantly different in the BOC group, with LC exhibiting lower values (
p < 0.01, effect size; 0.934). In addition, the RP condition values were significantly higher in the BOC group than in the UOC group (
p < 0.01, effect size; 1.409).
For the BOC group,
Figure 6 demonstrates the results of the correlation analysis between the static and dynamic balance while
Table 4 reports the
p-value, effect size, and statistical power for each test. Under the RP and LC conditions, significant negative correlations were found between the ENV-area and REC-area (
p < 0.01, effect size; 0.490, 0.495), whereas significant positive correlations were found between the LNG/E-area and REC-area (
p < 0.05, effect size; 0.449, 0.431).
The results of the correlation analysis between the static and dynamic balance in the UOC group are illustrated in
Figure 7, with
Table 5 showing the
p-value, effect size, and statistical power for each test. Under the RP condition, a significant negative correlation was identified between the ENV-area and REC-area (
p < 0.01, effect size; 0.469) while a significant positive correlation was found between the LNG/E-area and REC-area (
p < 0.05, effect size; 0.518). Under the LC condition, no significant correlation was found between the ENV-area and REC-area or between the LNG/E-area and REC-area.
4. Discussion
The results of this study showed that the effect of clenching on the static and dynamic balance was limited to the group with stable occlusal contact. Furthermore, a significant correlation was observed between the static balance and dynamic balance, except under the condition of clenching in the group with unstable occlusal contacts. Therefore, the null hypothesis that the effect of clenching on static and dynamic balance is not influenced by the stability of occlusal contacts was rejected.
Occlusal contact conditions depend on the teeth alignment, chewing habits, and lifestyle habits such as bruxism. It has been reported that the global sleep bruxism prevalence is 21% and awake prevalence is 23% [
24]. In addition, when participating in sports or work that requires the habitual exertion of muscle strength, repeated excessive clenching of the teeth in specific postures can cause excessive wear (attrition) of the teeth [
25]. Due to these factors, the occlusal contact state may exhibit a left–right imbalance. Our previous work focused on the left–right balance of occlusal contact and investigated its relationship with postural control [
6,
26]. The results revealed a correlation between the left–right balance of occlusal contacts and static balance and suggested that the static balance can be improved by correcting the occlusal contact condition using an intraoral appliance. However, the sample sizes of these studies were small, and it was not possible to directly compare the individuals with good and poor occlusal contact conditions. In the present study, the stability of occlusal contact was determined by the difference in the occlusal contact area between the left and right sides and the participants were divided into groups with good and poor occlusal balance. The influence of differences in the stability of the occlusal contact state on the balance ability during clenching was investigated.
In this study, the static and dynamic balance was evaluated by measuring the COP displacement using a center of gravity sway meter. In accordance with previous studies, the static balance was evaluated using the ENV-area and LNG/E-area in a stationary standing position. The ENV-area indicates the amount of movement from the center of gravity and is an index used in many reports [
2,
7,
27]. The LNG/E-area tends to be less influenced by vision and more easily influenced by deep senses and is used as a parameter that indicates the fineness of postural control [
27,
28,
29]. The smaller the ENV-area and the larger the LNG/E-area, the better the static balance tends to be. Our results indicated no differences in any of the indices between the participant groups due to occlusal balance. Significant differences due to clenching were observed only in the BOC group, with the ENV-area smaller and the LNG/E-area larger under the LC condition than under the RP condition, and the effect size was medium at 0.690 and 0.774, respectively. In other words, the static balance of the BOC group improved with clenching. This is consistent with previous studies showing that the static balance improved when an intraoral device was worn for occlusion correction [
6,
26]. Occlusion may have contributed to the postural control through deep sensory input, because the occlusal contact state was stable in the BOC group. The reason why occlusion did not affect the postural control in the UOC group may be that unstable occlusal contacts do not have an advantageous effect on postural adjustment.
The dynamic balance was evaluated in this study using a cross test method [
19,
20]. The cross test records the COP displacement when the upper body is slowly moved over 3 s without a change in the foot position. The cross test trajectory diagram will show a clear cross if the center of gravity can be moved smoothly, and the REC-area tends to become larger when the center of gravity significantly moves forward, backward, or side to side; therefore, it is also a way to evaluate the effectiveness of physical therapy on the dynamic balance ability. The cross test is considered to use the ankle strategy and hip strategy, two of the three postural strategies for maintaining balance [
30]. It is inferred that the participants used the ankle strategy to maintain their balance by moving their ankles when they lost their balance slightly because the participants in this study were young healthy individuals, and that the hip strategy works to control balance by moving the hip joint more when the individual loses their balance significantly or when the support surface is narrow. It is speculated that the ankle strategy was primarily responsible for the small range of center of gravity movement at the beginning of the movement from the reference position because the range of upper body movement over the 3 s in the cross test was left to the participant’s discretion, and that the hip strategy was responsible for maintaining balance during the later part of the movement when the center of gravity shifted more significantly.
The results of this study indicated that the difference in the REC-area due to clenching was present only in the BOC group and that it was significantly lower in the LC than in the RP condition, and the effect size was large at 0.934. On the other hand, in the UOC group, no significant difference was observed in the REC-area due to clenching. Additionally, under the RP condition, the REC-area was significantly larger in the BOC group than in the UOC group, and the effect size was large at 1.409. Accordingly, among the participants in this study, the BOC participants may have been a group with excellent dynamic balance ability. However, because significant differences due to occlusion were observed in the BOC group, clenching may be a factor that impedes center of gravity movement in individuals with good occlusal balance. In other words, clenching with stable occlusal contact may support the stability of the trunk and limit the center of gravity movement. Previous studies have investigated the effect of clenching on spinal alignment [
31] and found that, in groups with stable occlusal contacts, clenching reduced the range of motion of the lumbar spine and hip joint. The hip joint range of motion during trunk flexion in this report could be considered the hip strategy in this study. Therefore, it is possible that clenching had no significant effect in the UOC group with unstable occlusal contact. The R/E value, which reflects the clarity of the trajectory diagram, is an index that can be used to comparatively evaluate cases that have the same REC-area. In this study, because a significant difference was found between the BOC and UOC groups under the RP condition, the R/E value could not be examined.
There is no consensus on the relationship between the static and dynamic balance because various assessment methods and measurement environments have been used in the field [
15,
32,
33]. This may be due in part to the variety of factors that influence the balance ability: muscle strength, flexibility, joint and spinal mobility, neuromuscular mechanisms, fascial chains, and sensory input for postural control. In other words, this is probably because our experiments evaluated physical functions that are likely to be influenced by the environment and condition settings at the time of measurement. In this study, a center of gravity sway meter was used to evaluate the static and dynamic balance and to determine the effects of clenching in an environment with a stable base of support. As a result, the effect sizes were medium to large in the range of 0.449 to 0.518, and a significant correlation was observed between the static and dynamic balance in both the BOC and UOC groups. In contrast, only the BOC group showed a significant correlation with a medium effect size under the LC condition, revealing that the effect of clenching on the postural control depends on the stability of the participants’ occlusal contact state. In the BOC group, the ENV-area, LNG/E-area, and REC-area were all affected by clenching, suggesting that the occlusal intervention influenced the postural control and, as a result, a significant correlation was observed between the two groups. However, in the UOC group, no variables were affected by clenching, so the occlusal intervention did not affect the postural control, and, thus, no correlation was observed between the static and dynamic balance.
This study has four main limitations. First, the participants were limited to healthy men. The center of gravity sway may be affected by the menstrual cycle [
34,
35], and it is also assumed that it is influenced by age and health conditions and dependent on social background such as work environment and lifestyle habits, so we targeted healthy university students. Second, the DC/TMD or RDC/TMD test [
36,
37] was not performed to examine the temporomandibular joint. Since the light clenching (LC) measured in this study was a relatively weak force of 20% of the MVC, the participants were selected based on examinations by dentists for masticatory muscle pain, temporomandibular joint pain, clicking, and trismus. In future studies, when measurement conditions involve a high clenching strength or jaw movement, it will be essential to select participants based on these tests. Thirdly, a single method was used to assess the static and dynamic balance, which means that the effect of occlusion on the postural control may show different tendencies when other methods are used to assess the balance ability. Fourth, the R/E value could not be examined because there was a significant difference in the REC-area between the BOC and UOC groups. Multiple physical factors are reflected in the dynamic balance, and it is expected that various results will be shown depending on the evaluation method. In terms of future work, one of our objectives is to use multiple evaluation methods to clarify the relationship between occlusion and postural control, including differences among participant groups.