Facial Soft Tissue Characteristics Among Sagittal and Vertical Skeletal Patterns: A Cone-Beam Computed Tomography Study

Background Facial esthetics depend on the skeletal and dental structures underlying variable facial soft tissue thickness. In this social context, determining the relationship between external soft tissue and underlying skeletal and dental hard tissue is essential for detailed orthodontic diagnosis and treatment planning. Objective This study aims to measure facial soft tissue thickness in different sagittal and vertical skeletal patterns. Methodology This is an observational study utilizing pre-existing cone-beam computed tomography (CBCT) images of 170 subjects (110 females and 60 males) with a mean age group of 37.45 ± 13.83 years. CBCT images were then classified sagittally based on the point A-Nasion-point B (ANB) angle from Steiner's analysis into skeletal Class I, Class II, and Class III. Furthermore, vertical patterns were grouped based on the Frankfort-mandibular plane angle (FMA) from Tweed's analysis into hyperdivergent, hypodivergent, and normodivergent facial types. One-way ANOVA was used to compare the means of facial soft tissue thickness between the skeletal groups, followed by Tukey's post-hoc test for individual comparison. Multinomial logistic regression analysis was used to test the association between gender, age, and skeletal groups. The significance level was 0.05. Results One-way ANOVA revealed statistically significant differences in both sagittal and vertical groups (p≤0.05). Tukey's post hoc analysis showed that the skeletal Class III group has increased soft tissue thickness in the subnasale, upper lip, and mention compared to Class I and Class II subjects. Moreover, the hypodivergent group demonstrated increased soft tissue thickness in gnathion and mentioned landmarks in relation to the other groups. Multinomial logistic regression analysis showed significant differences between groups according to both gender and sagittal skeleton patterns (p≤0.05), with males less likely to be in Class II. Conclusions Skeletal Class III and hypodivergent groups have thicker soft tissue in specific facial landmarks. Sexual dimorphism was marked in soft tissue measurements.


Introduction
The shape and size of the head and face, known as craniofacial morphology, is influenced by genetics and the environment.These factors work together to create the unique characteristics of an individual's face [1].The impact of genetics and the environment can lead to a wide range of facial profiles within a population [1].Facial harmony and attractiveness are associated with social acceptance, success, and psychosocial wellbeing [2,3].Yet, facial proportions and esthetics depend on the skeletal and dental structures underlying variable facial soft tissue thickness [4].For instance, increased lip thickness is perceived as preferable in females and might indicate youth in modern women, while less lip thickness is more desirable in males [5,6].In this social context, determining the relationship between external soft tissue and underlying skeletal

Materials And Methods
Our study was conducted using pre-existing CBCTs taken for dental or surgical purposes between 2017 and 2023.The radiographs were randomly selected from the archive of the King Abdulaziz Medical City in Riyadh (KAMC-RD) and the College of Dentistry (COD) in King Saud bin Abdulaziz University for Health Sciences (KSAU-HS).The study is a retrospective, cross-sectional, and comparative study that was approved by the institutional board of King Abdullah International Medical Research Center (KAIMRC) (protocol number: NRC22R/302/06, date of approval August 22nd, 2022).The inclusion criteria were that the CBCT images must be of decent quality and of subjects 18 years and above and Saudi Arabian.The variables were measured using CBCT imaging software (Romexis software version 4.5, Planmeca OY, Helsinki, Finland) with a volume size of 23.0 × 17.3 cm or above and a voxel size of 400 µm.Subjects with previous orthodontic treatment, a trauma in the head and neck region, congenital abnormalities, cleft lip and palate, previous plastic surgery, or previous orthognathic surgery were excluded, as were CBCT images with distortions.
Thereafter, the examiner (NA) was blinded to the groups through the creation of a new Excel spreadsheet (Microsoft, Redmond, Washington) by another examiner (FA) with subjects coded and lacking the skeletal groups.Then, the facial soft tissue thickness was assessed across eleven different landmarks as described by Gomes et al. [9] (Figure 1).The landmarks were defined in the sagittal, coronal, and axial views, as shown in Table 1.The total sample size was 170 subjects (110 females and 60 males) with a mean age of 37.45 ± 13.83 years.The sample distribution per skeletal group, as well as the gender distribution, is illustrated in Table 2.    [9].A minimum sample size of 11 subjects per sagittal skeletal group achieves 80% power to detect differences among the means versus the alternative of equal means using an F test with a 0.05 significance level.As for the vertical skeletal groups, a total sample of 60 subjects (20 subjects per group) was adequate to obtain 80% power and a type I error of 5%.
The Shapiro-Wilk test was used to assess the normality of the dataset.A Chi-squared test was performed to compare males and females in relation to facial soft tissue thickness.Furthermore, the one-way ANOVA test was used to investigate the mean differences in the soft tissue thickness between the skeletal sagittal and vertical groups, followed by Tukey's post hoc test for individual comparison.Moreover, multivariant logistic regression analysis was used to evaluate the relationship between age, gender, and soft tissue thickness, with p-values less than .05considered statistically significant.
For intra-examiner reliability, twenty CBCT images were chosen randomly one month after the first analysis and were re-measured by the same orthodontist (NA).The intraclass correlation coefficient (ICC) test was used for intra-examiner reliability.Statistical Analysis Software (SAS) (version 9.4, SAS Institute Inc., Cary, North Carolina) was used for data analysis.

TABLE 5: Soft tissue thickness measurements in different vertical skeletal groups
One-way ANOVA revealed differences between the three vertical groups and showed that there are significant differences in hypodivergent subjects in different landmarks: a) significant differences in hypodivergent vs. hyperdivergent (p-value≤0.05)(Tukey's post hoc test), b) significant differences in normodivergent vs. hypordivergent (p-value≤0.05)(Tukey's post hoc test), c) significant differences in normodivergent vs. hyperdivergent (p-value≤0.05)(Tukey's post hoc test).

Sagittal groups Independent variables Adjusted odd ratio (AOR) 95% confidence interval for AOR p-value
Class  a -multinomial logistic regression analysis showed no significant differences between age and sagittal groups (p-value>0.05).However, the model revealed an association between gender and sagittal groups (p-value≤0.05).Our reference was the Class I group and female participants, and we found that there is a significant difference between gender in Class II where male participants were less likely to be in Class II.In Class III, however, there is no significant difference between males and females.
b -multinomial logistic regression analysis showed no significant differences between gender and vertical groups (p-value>0.05).The model did, however, reveal an association between age and vertical groups (p-value≤0.05).Our reference was the normodivergent group and female participants, and the analysis showed that there is a significant difference between age and hypodivergent group.
Based on Intraclass Correlation Coefficient (ICC) guidelines [20], the intra-examiner reliability of all soft tissue measurements was good (ICC values were between 0.75 and 0.90).

Discussion
Facial esthetics is the principal goal in orthodontic practice and the main motivation for patients seeking orthodontic treatment [5,21].Orthodontists can identify skeletal patterns associated with unpleasant soft tissue thickness and transform it into attractive facial thickness by means of orthodontic treatment, orthognathic surgery, or plastic surgery.This study used CBCT images to assess the facial soft tissue thickness in different sagittal and vertical skeletal patterns.A combined assessment of both vertical and sagittal skeletal patterns has not previously been undertaken in the Saudi Arabian population.A previous study conducted on Saudi Arabian subjects assessed facial soft tissue differences arising from different dental malocclusions instead of skeletal discrepancies [22].
Earlier studies have shown that changes in the skeletal structures are reflected in the overlying soft tissue.For example, maxillary and dental protrusion affects the soft tissue protrusion and upper lip position [23,24].Moreover, alterations in facial soft tissue have been reported after orthognathic surgery [25].Hence, the present study excluded subjects with previous orthodontic treatment and orthognathic surgeries.In addition, subjects with previous plastic surgeries were excluded.Furthermore, CBCT images of patients with difficulty in lip closure due to severe skeletal discrepancy were excluded from the study.
Our findings demonstrated significantly greater soft tissue thickness in all landmarks in male subjects compared to female subjects, except for the Sm h -Sm s landmark.This result is in line with previous studies that were conducted in different populations [7,9,17,[26][27][28], suggesting the presence of sexual dimorphism in facial soft tissue thickness.Males have thicker facial bones and muscles compared to females, and that could explain the gender differences in facial soft tissue [29].In addition, males' skin is thicker since the testosterone hormone promotes collagen synthesis, unlike in females and in whom the estrogen hormone promotes hyaluronic acid synthesis [30].
With respect to the skeletal sagittal groups, the initial hypothesis was rejected, indicating that there were some statistically significant differences between the groups.Specifically, the soft tissue thickness was increased in GL h -GL s , N h -N s , Rhi h -Rhi s , Ss h -Sn s , U1-LS, Inc h -Sto s , and Me h -Me s landmarks in the skeletal Class III group compared to Class I and Class II.In agreement with our results, previous studies reported thicker soft tissue in the maxilla in class III subjects [9,26,27].The thicker soft tissue in skeletal class III in our studied population can be explained by soft tissue compensation in skeletal regions with deficient skeletal development or positioned posteriorly.In addition, the upper incisors were proclined, and lower incisors were retroclined in the majority of class III individuals, and this could push the upper lip forward, leading to greater soft tissue thickness [9].
The influence of the vertical skeletal pattern on facial soft tissue thickness has been reported previously in other populations [31][32][33].In agreement with our results, a study showed that the hypodivergent group had thicker soft tissue in Me h -Me s and Gn h -Gn s [34].Previous studies found that hyperdivergent subjects had reduced soft tissue thickness in Rhi h -Rhi s , Ss h -Sn s , and U1-LS, compared to hypodivergent subjects [32,33].
In concordance with previous literature, our findings highlight that the hyperdivergent group had reduced soft tissue thickness in GL h -GL s , Rhi h -Rhi s , U1-LS, Gn h -Gn s , and Me h -Me s compared to the hypodivergent group.This could be explained by soft tissue compensation for the skeletal vertical disharmony [22].Another reason could be due to the increased stretching of the soft tissue in the hyperdivergent group, resulting in reduced thickness.On the other hand, the hypodivergent group had increased soft tissue thickness, probably due to the presence of hypertrophy of perioral musculature, which tends to occur as a way of compensating for vertical deficiency and maintaining lip competence [35].
Although the results of this study highlight that male subjects are more likely to be in skeletal class III in relation to female subjects, our analysis showed no significant differences between gender and vertical groups.This can be explained by the late onset of the pubertal growth spurt and increased pubertal growth duration in males compared to females, resulting in continued mandibular size growth leading towards a skeletal class III relationship [36].In agreement with our findings, a study reported class III tendency in male subjects.However, this study reported a significant difference between vertical groups and gender.Male subjects were more likely to be hypodivergent, and hyperdivergence was found in female subjects [37].It is worth noting that the study used different measurement tools (anteroposterior dysplasia indicator (APDI), and sum total of nasion-sella-articulaire angle, sella-articulaire-gonion angle, and articulaire-gonionmention angle (sum angle)) for grouping the subjects into sagittal and vertical patterns instead of ANB and FMA angles.Thus, the disagreement in findings could be the different measurement tools and different ethnic backgrounds.
Our results support previous studies [38][39][40] that age-related changes of the face are associated with hypodivergence of the face.The mandibular plane decreases with age, and the posterior-to-anterior facial height ratio increases with age, possibly due to the counterclockwise rotation of the mandible with age.In addition, similar to the results of previous studies, our data reported high intra-examiner reliability, suggesting that CBCT scans are reliable for facial soft tissue analysis [9,41].
The clinical implication of the present study includes providing a diagnostic tool offering a reliable assessment for the diagnosis and treatment of different skeletal patterns.Orthodontists and surgeons should be familiar with sex-specific variations in certain populations during therapeutic interventions.
Considering the studied population, the presence of facial soft tissue sexual dimorphism in adults has important implications in orthodontic treatment planning and outcome assessment.A study found that patients with thick lips displayed no correlation between incisor retraction and lip retraction, whereas patients with reduced lip thickness showed a significant correlation [42].Based on this information and our study findings, orthodontists should be mindful that extraction therapy and incisor retraction might have a greater effect on facial profiles in female patients compared to male patients.Moreover, the present study indicates different clinical considerations for different skeletal patterns.Patients with skeletal class III and hypodivergent patterns might be less affected by orthodontic extraction treatment and incisor retraction

TABLE 2 : Distribution of subjects in different skeletal patterns
The calculation of the sample size was performed using Number Cruncher Statistical System (NCSS) software 2022 version (NCSS LLC company, Kaysville, Utah) and the Power Analysis and Sample Size software (PASS) version 15 (NCSS LLC company, Kaysville, Utah), based on the effect size stated inGomes et al. study ). Tukey's post hoc analysis showed that, compared to Class I and Class II, the skeletal Class III group had increased facial soft tissue thickness in GL h -GL s (p-value=0.0268),Nh-N s (p-value=0.0029),Rhih-Rhi s (p-value=0.0078),Ssh-Sn s (p-value=0.0001),U1-LS(p-value≤0.0001),Inch -Sto s (p-value=0.0037) and Me h -Me s (p-value=0.0024).Table4revealed more details about the differences between the three classes.