Bite and Sight: Is There a Correlation? Clinical Association between Dental Malocclusion and Visual Disturbances in Pediatric Patients

: Aim: The aim of this study was to evaluate the correlation between malocclusions and visual defects. This is a case-control study evaluating the prevalence of visual defects in patients with di ﬀ erent types of malocclusions. Methods: One-hundred and sixty patients aged from 5 to 14 were evaluated using the ROMA index to detect malocclusion; the ones with the lowest scores were used as the control group. They were also submitted to visual-capacity inspection for motility and refractive disorders. Results: Our work showed an enhanced prevalence of refractive defects or fusional vergence defects and alteration of eye movements (especially the saccades) in almost all dental malocclusions. Statistics: The Kappa test values for ROMA index were between 0.643 and 1.00 for the intraoperator agreement (0.00 < p < 0.002) and between 0.773 and 1.00 for the agreement between operators ( p = 0 < 0.001). The statistically signiﬁcance level for the correlation malocclusion / visual defects was set at p < 0.05. Statistical analyses were performed with the STATA software (version 15.0, Stata Corp LP, College Station, TX, USA). Conclusion: Considering the high level of the statistical analysis and the accuracy of the methodology used, these data allows the establishment of a huge correlation between sagittal, transversal and vertical malocclusions with ocular disorders (myopia, hyperopia, astigmatism and ocular motility defects).


Introduction
Anatomic systems are organized through a network of structural and functional relationships between their elements. This network of relationships generates a set of rules that guide and constrain morphogenetic processes. Decades of investigations have shown that craniofacial development is an intricate and complex series of events that results in the correct outgrowth, patterning and tissue integration, required to make a face [1]. The soft tissue matrix-in which skeletal elements are embedded-is the primary determinant of growth, while both the bone and cartilage are reactive secondary growth sites. This is the fundamental premise of the functional matrix theory [2].
The embryonic facial maxillary prominences develop numerous intramembranous ossification centers. One of them is the orbito-nasal center that starts formation around the seventh week. The eye, Hence, the interconnection between stomatognathic system and vision comes from the morphologic aspect to the functional one. Morphologic changes of orbit shape can make the individual more prone (acting as epigenetic factors) to other common visual problems like Myopia, Hyperopia and Astigmatism [5,15].
Myopia is a visual defect where the ocular globe is elongated and the image forms before the retina. Prevalence rates range from about 30% in middle-aged adults and from 35% to 37% in young adults. Malocclusions can enhance prevalence of Myopia [16].
Hypermetropia or Hyperopia is an eye condition in which incoming light rays reach the retina before converging into a targeted image. It can happen due to sagittal shortening of ocular globe. The prevalence of hypermetropia is not clear. It runs from 9.9%, to 28.9%. [17] Astigmatism is a common vision condition that causes blurred vision. It occurs when the cornea (the clear front cover of the eye) is irregularly shaped or sometimes because of the curvature of the lens inside the eye. Racial variations influence the prevalence and the degree of astigmatism. The alteration of orbit shape can lead to an enhancement of cornea curvature [18].
As shown above there are many studies discussing the prevalence of vision defects in the general population and it is also possible to find some works correlating vision defects with different types of alterations of maxilla/mandible relation. Malocclusion is a highly prevalent condition and it can disturb this relation in different directions; sagittal, vertical and transversal. Considering that, the objective of this study was to reinforce the concept of a correlation between visual defects and dentoskeletal malocclusions in a sample of growing patients, evaluating the maxilla/mandible relations of the three planes. In addition, since vision and occlusion are important afferences to establish a body posture, we also investigated body posture compensations of the sample.

Materials and Methods
The study was conducted in accordance with the Declaration of Helsinki and the protocol was approved by the Ethics Committee of the University of L'Aquila (Document DR206/2013, 16 July 2013). This retrospective clinical study was conducted on 216 patients (104 male subjects and 112 female subjects) of the Ophthalmology Clinic of the Head-Neck Department, between September 2018 and June 2019. A total of 55 patients were excluded for presenting other visual defects, history of orthodontic treatment and past extraction of permanent teeth. The remaining 160 subjects (76 male and 84 female) were included in the study. The age between 5 and 14 years, average age 10 years (SD 3). The patients were all submitted to an ophthalmologic and orthoptic examination for evaluation of: Myopia, Hypermetropia (Hyperopia), Astigmatism, Strabismus, Latent strabismus, Manifested strabismus, fusional vergence, saccadic eye movements and smooth pursuit movements. The orthoptic examination was performed through the following tests: Cover/uncover test; cover test with prism; Convergence; fusional amplitude, Northeastern State University College of Optometry (NSUCO) test. It gives precise values to the observations made by the clinician. (See in-Appendix A). The orthodontic conditions were classified as follows: sagittal, vertical and transversal malocclusions, according to the Risk Of Malocclusion Assessment (ROMA) Index [19] that distinguishes the problems of the three level of malocclusions, crowding and spoiled habits (nonnutritive suction, bruxism and oral breathing). (Appendix B- Table A1). Dental exam was performed by two dentists (CR and FA). The ROMA index was evaluated by operators who have previously undergone a one-month training period following the instructions in a specific manual, in order to apply the index with the same standard of judgment and to minimize errors. Furthermore, the index has already been validated and its intraoperator and interoperator reproducibility was verified. The orthoptic exam and the ophthalmic examination was executed by two ortoptist (VC and EDN), in order to eliminate inter-examiner differences and examiner bias.

Statistic Analysis
The vision situation of the sample, stratified for orthodontic condition, was compared to a control group (no malocclusions). Descriptive statistics, such as frequencies with percentages, was used to examine the characteristics of the sample. Statistical significance of differences between each orthodontic group and the control group was analyzed using χ 2 test.
To verify the potential association between each of the analyzed vision defects, chosen as dependent variables, and the various categories of malocclusion, univariate logistic regression models were used, with associations reported as odds ratios (ORs) and 95% confidence intervals (CIs). The absence of each specific malocclusion was chosen as reference category (e.g., the reference category for the Class II malocclusion in each regression model was the absence of this orthodontic condition). Factors with a p-value of less than 0.05 in univariate models were included in the multivariate logistic regression models, to identify independent associations, with ORs adjusted for the other factors in the models. The statistically significant level was set at p < 0.05.
The statistical analysis was performed with the STATA software (version 15.0, Stata Corp LP, College Station, TX, USA).
The power of the study was evaluated using Pearson's chi-squared test comparing two independent proportions: presence of astigmatism in the control group (n = 52) and in subjects with malocclusions (n = 108) (11% vs. 31%). The estimated power (1-β) was 82%.

Results
A total of 160 subjects (76 male and 84 female) were included in the study: -52 (32.50%, 52/160) patients had no malocclusions (Group "0"-control group We observed the results obtained by the following groups (Table 1). •

Sagittal malocclusions
-Class II. The presence of sight defects was detected in 40.38% of the subjects. Higher percentages were found for almost all visual defects compared to the control group.
In particular, children with myopia were more than twice (61.54% vs. 23.53%, p < 0.001) that of the control group, and children with latent or manifested strabismus were almost four or three times more represented, respectively (40.38% vs. 11.54%, p < 0.01; 36.54% vs. 11.54%, p < 0.01, respectively). A tracking defect was found in 42.31% of children compared to 21.57% of the control group (p < 0.001) and an alteration in the tracking precision was detected in 40.38% (vs. 17.31%, p < 0.001). The percentage of children with strong compensation of the body and head was greater than the control group (40.38% vs. 15.38%, p < 0.01). -Class III. The presence of vision defects was detected in 44.44% of the subjects. Children with astigmatism, latent or manifested strabismus were almost five times more represented than in the control group (55.56% vs. 11.54%, p < 0.01). A tracking defect and an alteration in the tracking precision were found in 55.56% of children compared to 21.57% and 17.31%, respectively, of the control group (p < 0.001 and p < 0.01, respectively). The percentage of children with strong compensation of the body and head was greater than the control group (55.56% vs. 15.38%, p < 0.01). •

Vertical malocclusions
-Deep bite. The presence of vision defects was detected in 51.43% of the subjects. Significantly higher percentages than those in the control group were found for hyperopia (40.00% vs. 17.65%, p < 0.001) and manifested strabismus (40.00% vs. 17.65%, p < 0.01); -Open bite. The presence of vision defects was detected in 53.33% of the subjects. Children with astigmatism were four times more represented than in the control group (46.67% vs. 11.54%, p < 0.01). •

Transversal malocclusions
-Cross bite. The presence of vision defects was detected in 45.45% of the subjects. Children with myopia or hyperopia were significantly less numerous than those in the control group (2.95% vs. 23.53%, p < 0.001 and 2.94% vs. 17.65%, p < 0.001, respectively). On the contrary, children with astigmatism (50.00% vs. 11.54%, p < 0.05) or latent and manifested strabismus (38.24% vs. 11.54%, p < 0.01) were more represented than in the control group. The percentage of subjects with strong compensation of the body and head was greater than the control group (38.24% vs. 15.38%, p < 0.01). Statistical significance of frequency distributions compared to the control group using χ 2 test: * p < 0.001, ** p < 0.01, *** p < 0.05, • p ≥ 0.05.

Discussion
The purpose of this study was to highlight the relationship between the visual system and the stomatognathic system. Some authors have published interesting works about this subject but have generally focused on one malocclusion condition or another [5,15]. Our work brings a comprehensive correlation of different ocular disorders and malocclusions presented in the sagittal, transversal and vertical planes. We tried to clarify for the health community the importance of understanding the human being not in a compartmentalized state, but as an integrated group of systems that communicate each other. Problems that occur in the visual system can result in consequences of imbalances in the stomatognathic system or vice versa. It was also emphasized the relevance of including oculist-orthoptic exams of the protocol of stomatognathic system evaluation, to support preventive interventions in a pediatric population. Our work showed an enhanced prevalence of refractive defects, fusional vergence defects and alteration saccadic eye movements and pursuit movements, in almost all dental malocclusions. These findings can be reinforced by the results presented in a study that demonstrate the improvement of symmetry in the orbital and maxillary regions in patients after rapid maxillary expansion [20]. In sagittal malocclusions, refractive defect myopia were found in 61.54% of the patients presenting Class II malocclusion, while control group showed 23.53% of prevalence. They also presented alterations of the ability of saccadic movements two times bigger than control group. Caruso had also demonstrated a statistically significant association between Class II and the occurrence of exodeviations, associating occlusal molar Class II to fusional vergence defects [21]. In relation to refractive vices we are in agreement with Monaco et al. as they found a greater prevalence of refractive defects in patients with sagittal malocclusions too [22]. However, the higher prevalence of strabismus in the Class II group found in our sample, where children with latent or manifested strabismus were almost four or three times more represented, (40.38% vs. 11.54%, p < 0.01; 36.54% vs. 11.54%, p < 0.01, respectively), is not in accord to Monaco work, since they did not find an enhancement in the vergence capacity. It indicates that further investigations are necessary, expanding the sample examined, to be able to demonstrate and better understand this divergence of results. For Class III patients we found a huge statistical difference (5 times more) in the prevalence of astigmatism and strabismus. It was interesting to note from our results that the values of a fusional amplitude below the normal range and an alteration of the abilities of saccadic movements presented themselves with very similar percentages in almost all malocclusions. It is also important to highlight that body compensations presented by patients with sagittal malocclusions were three times bigger when compared to the control group. Many works had demonstrated that vision influences body posture and also there is a correlation between mandibular position and body posture [23][24][25][26][27]. In our sample is impossible to determine if the malocclusions generated a body compensation, if the visual problem was the cause of it or if both were related to the observed increase in posture alteration. Specific design of investigation must be done to clarify this finding.
Patients with transverse malocclusion or crossbite have refractive defects almost exclusively represented by astigmatism. The prevalence was almost five times bigger than control. The group also showed a prevalence of strabismus three times bigger than control, showing alteration in the saccades ability. Referring to astigmatism, we are in accord to with Monaco [22]. In their study they have specifically reported statistically significant correlations between astigmatism and crossbite. In our data we also find body compensation in a high prevalence (more than two times compared to the control) in patients presenting crossbites. The relationship between Strabismus and malocclusion was investigated by Giuca et al. [28]. They found a slightly increase of cross bite in the group of subjects with strabismus, but not statistically significant. Malocclusion is a multifactorial condition of multiple features for which is difficult to demonstrate a cause-effect relationship. In our results we found a significance of the presence of cross bite in children with strabismus. Therefore, this relationship deserves to be investigated further [28].
In patients with vertical malocclusions, refractive errors were found in 75% of patients with a deep bite and in 68% in patients with an open bite, thus presenting the highest incidence compared to the other groups of malocclusions; Deep bite patients showed a prevalence of hyperopia and strabismus two times bigger than control, showing alterations in saccadic movements and a fusional amplitude below the normal range. On the other hand, open bite patients showed the astigmatism prevalence four times bigger than control, showing also a low fusional amplitude and difficulty in the abilities of saccade movements. The value of this research is given by having investigated an aspect of the relationship between visual problems and malocclusion so far still little known. The results encourage a deepening of the investigations, to improve the approach to the prevention and treatment of patients in developmental age. This point of view is shared by Silvestrini et al. [29,30] that evaluated the dominant eye and the presence of vertical malocclusions. Most subjects had a right dominant eye.
Subjects with an open bite showed a slightly different distribution in eye dominance, with a smaller difference between right or left preferences. Patients with a deep bite were similar to the control. Hence, it is necessary to create an investigation design constructed to evaluate the correlation of vertical malocclusions and ocular disorders.
Analyzing our sample, in patients with functional asymmetries we found a greater incidence of refractive vices (67%), of fusional vergence (52%) and an alteration of the saccades (56%); in patients with flawed habits, refractive defects occur at a rate of 61%, the amplitude was poor with 52%; while in patients with misalignment/crowding, refractive vices occur with 61%, the amplitude fused poorly with 57% and the alteration of the saccades with 57%.
A major limitation of studies of the causes of malocclusion is the multiplicity of its clinical manifestations and the multifactor character of the causes. For this reason, a study that observes more clinical and functional aspects that can interact with the craniofacial development cannot lead to evident results. However, there are many interesting aspects in this work that suggest continuing to investigate the relationship between vision problems and jaw development.
In particular, this research has the limitation of having separately considered myopia, hyperopia and astigmatism. Some children in the sample analyzed had astigmatism at the same time as myopia or hyperopia. However, in these aspects it was possible to compare with literature data. Instead, as regards the defects of ocular motility, this work provides results that have little precedent in the scientific literature.

Conclusions
Based on our results, it seems that exist a correlation between sagittal, transversal and vertical malocclusions with ocular disorders (myopia, hyperopia, astigmatism and ocular motility defects). For this reason, it would be advisable that visual exams should be added to the evaluation protocols used to analyze the stomatognathic system, since there is an interrelationship among malocclusions, visual problems and body posture. It has been shown that body compensation is present in a great prevalence when a malocclusion or visual defect is present. It is necessary to establish very well structured protocols of investigation to check the cause-effect relation and to establish preventive approaches when dealing with pediatric population.
The ROMA index was evaluated by operators who have previously undergone a one-month training period following the instructions in a specific manual, in order to apply the index with the same standard of judgment and to minimize errors.
To assess intraoperator reproducibility, the intraoperator error was calculated by repeating the examination for 20 children one month after.
A second operator independently collected a third table index for each of the 20 children to assess the interoperator error.
The Kappa values were between 0.643 and 1.00 for the intraoperator agreement (0.00 < p < 0.002) and between 0.773 and 1.00 for the agreement between operators (p = 0 < 0.001).

GRADE 1
None of the problems listed above N