Introduction
Previous studies have looked at a variety of dental and facial asymmetries and compared their detection by dental professionals with those of laypersons. However, few studies have analyzed the diagnosis and perception of chin asymmetries. In this study, we assessed whether dental professionals can recognize and diagnose facial asymmetries of the chin better than laypeople.
Methods
Chin asymmetries were analyzed through a series of edited frontal photographs of 2 subjects (male and female). The transverse position of the chin was digitally altered from 0° (no alteration) to 6° (most severe alteration). Participant responses were collected from laypersons (n = 64), nonorthodontist dentists (n = 58), and orthodontists (n = 145). Participants graded the photographs according to esthetic appeal using a visual analog scale. Statistical analysis produced diagnostic threshold levels for identifying chin asymmetries.
Results
Nonorthodontist dentists and orthodontists were better able to diagnose transverse chin asymmetry in the female subject at a lower threshold level of 2° of deviation, compared with laypersons at a 3° deviation. Orthodontists could diagnose transverse chin asymmetry in the male subject at a lower threshold level of 1° of deviation, compared with laypersons and nonorthodontist dentists at a 2° deviation. All 3 groups of raters graded very small or no chin asymmetries (1° and 0°) as more attractive, whereas high degrees of chin asymmetries (5° and 6°) were graded as most unattractive by all 3 groups.
Conclusions
Transverse asymmetries of the chin influence the perception of facial attractiveness by laypersons, nonorthodontist dentists, and orthodontists. Subjects with small asymmetries are graded as most attractive, and subjects with large asymmetries are graded as least attractive. Orthodontists were the harshest graders, followed by the nonorthodontist dentists, whereas laypersons were the most lenient. The accuracy of diagnosing chin asymmetries varied among laypersons, nonorthodontist dentists, and orthodontists. Laypersons were the least accurate in diagnosing transverse chin asymmetries in a female subject, and they were just as accurate as nonorthodontist dentists when diagnosing this asymmetry in a male subject. Nonorthodontist dentists were just as accurate as orthodontists when diagnosing transverse chin asymmetries in a female subject, and orthodontists were the most accurate in diagnosing transverse chin asymmetries in a male subject.
Highlights
- •
Orthodontists had the highest accuracy in diagnosing male subjects.
- •
Orthodontists and dentists were equally accurate in diagnosing female subjects.
- •
All 3 groups graded very small and no asymmetries similarly.
- •
All 3 groups graded larger asymmetries dissimilarly but consistently low.
- •
Laypersons were the most lenient in their assessments and orthodontists the least.
Contemporary society places a strong emphasis on the importance of facial esthetics. This perception of esthetics is inherently related to facial symmetry. It has been shown that minor facial and dental asymmetries are normal and acceptable findings in adolescents and adults. Most people tend have a dominant side of their face that introduces a mild form of facial asymmetry, and this asymmetry tends to increase toward the lower and lateral aspects of the face.
The diagnosis and treatment planning of these minor discrepancies are routinely considered by the orthodontic practitioner before and during treatment. Today’s patients, however, are becoming more astute regarding their facial appearance, and thus orthodontists need to be more accurate in their assessments. For example, it has been reported that patients can become cognizant of their facial asymmetries during treatment, resulting in their eventual desire to pursue treatment for these conditions.
The etiology of a chin asymmetry is multifactorial. It can arise from developmental disturbances, such as asymmetric growth of craniofacial bones and congenital malformations such as hemifacial microsomia. Pathologic conditions of the temporomandibular joint such as ankyloses, condylar hyperplasia and hypoplasia, and arthritis also frequently contribute to chin asymmetries. Facial trauma, specifically condylar fractures in adolescents, and imbalances of muscular activity are also considered to be common causes of chin asymmetry. Occlusal interferences have been shown to create chin deviations, since they can cause a shift in the position of the mandible. Frequently, functional shifts can create chin asymmetries; with a transversely deficient maxilla, the mandible can deviate from its correct position.
Proper early diagnosis of a chin asymmetry can lead to early, nonsurgical, and often successful intervention, depending on the etiology. For a skeletal maxillary transverse discrepancy causing a mandibular shift, early treatment via palatal expansion therapy has been shown to eliminate the chin asymmetry. A late diagnosis in adults, however, lacks the benefit of maxillary suture patency and thus inability to provide true skeletal expansion. In addition, adults no longer possess the potential for growth modification and may require a combined orthodontic and surgical approach. If orthodontic intervention is not an option, only a surgical approach remains to address the facial asymmetry in the form of jaw surgery or cosmetic implant placement.
Studies that analyzed the perceptions of altered dental esthetics between dental professionals and laypersons have been conducted in the past. These studies have generated valuable findings in establishing important threshold levels of orthodontic diagnosis that ultimately aid in interdisciplinary treatment planning.
Pinho et al investigated altered dental asymmetries and their effect on the esthetic perception of the smile between laypersons and dental professionals. Computer imaging manipulation were made to either the gingival margins (height) of the maxillary central incisors, wear of the maxillary canine cusps, or the dental midline. The authors concluded that dental professionals and laypersons have different perceptions about dental esthetics when evaluating dental midline and gingival margin height discrepancies.
Kokich et al looked at the perceptions of dental professionals compared with those of laypersons for asymmetric esthetic alterations of dental esthetics. The types of computer imaging manipulation on that study included crown length, crown width, midline, papillary height, and gingiva to lip relationship (gummy smile). The authors concluded that dental asymmetries resulted in unesthetic perceptions of the smile by laypersons as well as dental professionals.
An et al similarly studied altered smile esthetics and the perceptions by laypersons who either did not have or had orthodontic treatment, as well as by dentists. Changes were made to the height of the gingival margin, crown length and width, canting of the incisal plane, or the maxillary dental midline. The authors found that orthodontists were the most accurate in their assessments of altered smile esthetics, followed by laypersons who had orthodontic treatment, whereas laypersons who never had orthodontic treatment were the least perceptive.
Both Pinho et al and Kokich et al, however, focused only on altered dental esthetics and not on asymmetries of the lower facial third, such as those of the chin Studies that analyzed the perceptions of altered facial esthetics between dental professionals and other groups of raters including laypersons have been conducted in the past; however, there are few, and most focus on full facial asymmetry, rather than strictly lower facial third asymmetry. A review of the literature showed various attempts at understanding how facial asymmetry is interpreted by groups as well as dental and nondental professionals.
The dearth of studies linking diagnosis of lower facial third asymmetries on extraoral photographs to underlying skeletal asymmetry is surprising because of the correlation already introduced in the literature. To date, few studies have focused on identifying a clinical threshold level for diagnosing underlying skeletal asymmetries of the face, specifically of the chin, by analyzing the severity of extraoral soft tissue asymmetry. In this study, we assessed whether dental professionals can better diagnose facial asymmetries of the chin compared with laypersons, when analyzing extraoral frontal photographs. The threshold of perception for the 3 categories of survey participants that included laypersons, dentists, and orthodontists was investigated. The following hypotheses were tested: (1) laypersons are less accurate in diagnosing chin asymmetries than are nonorthodontist dentists; (2) orthodontists are more accurate in diagnosing chin asymmetries than are nonorthodontist dentists.
Material and methods
Before the study, approval was granted by the Rutgers Institutional Review Board (study identification number Pro20140000655). Survey participants were procured from various sources. Laypersons were recruited and advised of the study by randomly selecting parents of patients at the Rutgers School of Dental Medicine orthodontic clinic. The parents were offered voluntary participation and verbally educated about the research by a study coinvestigator (K.F.J.). This investigator then provided the interested laypersons a link to the study on a hard copy document for them to participate at home.
Nonorthodontist dentists and orthodontists were invited to participate via e-mail. The New Jersey Dental Association and the American Association of Orthodontists contacted participants from these member groups, respectively. The New Jersey Dental Association contacted 3500 members via e-mail, and the American Association of Orthodontists contacted 2300 of its members. Due to the general nature of the New Jersey e-mail list, dentists from additional specialties were also invited to participate in the study. Responses from these specialties were collected for statistical analysis. For the purpose of brevity in this study, this large group of nonorthodontist dentists was called the “dentists.”
Two subjects (a coinvestigator, K.F.J., of this study and his wife) were invited to participate as models for the study. Both subjects had their photographs taken with 1 m between the subject and the camera. The subjects were seated with the Frankfort plane horizontal to the floor and in centric occlusion. They were asked to relax all facial expressions. The photographs were stored on the camera’s memory disk and then transferred to a password-protected computer at the university for digital alteration. Photoshop software (version CS6; Adobe Systems, San Jose, Calif) was used to alter the images.
Before the alterations of the digital photographs, soft tissue landmarks previously defined by Lee et al were identified on the photos. This was done to delineate the patient’s facial midline and to standardize the selection of the lower facial third to be altered. The decision on which soft tissue points were chosen for the photograph area to be digitally altered was based on previous literature that discussed and demonstrated significant skeletal mandibular asymmetries, as well as on the clinical experience of 2 coinvestigators (J.A.B. and R.B.). From an anatomic perspective, an asymmetric chin has deviated bony anatomy and is associated with a corresponding altered soft tissue appearance, where the entire lower facial third frequently appears to be rotated. For the purposes of standardization, 4 soft tissue points were selected and used to outline the portion of the lower facial third to be modified with such a rotation. These points included subnasale, left and right; soft tissue gonion; and soft tissue menton ( Fig 1 ).
The patient portion of the original image ( Fig 2 , A ) was cropped out of the background, and color corrected to compensate for unideal lighting. To eliminate bias from judging symmetry in an asymmetric face, a perfectly symmetrical face was constructed ( Fig 2 , B ). This procedure was accomplished by using a mirror image of 1 side of the original face, as depicted by Proffit et al. The facial midline was constructed by identifying the following midsagittal points on the original frontal photograph: glabella, soft tissue nasion, pronasale, subnasale, and soft tissue menton. A line of best fit was then drawn through these points, and a half image was mirrored horizontally to create the perfect facial symmetry ( Fig 2 , B ). The same methodology was repeated for the female subject. Linear measurements of each subject’s face were standardized by using a clear ruler (Trubyte millimeter ruler; Dentsply Sirona, York, Pa) that was held against the subject’s skin in the infraorbital region ( Fig 3 ). This measurement was used to calibrate the Photoshop ruler tool.
The resultant lateral shift of the chin was defined as the distance in millimeters from the original soft tissue menton to the altered soft tissue menton. The distance between subnasale and the original soft tissue menton was calculated by using the calibrated ruler tool in Photoshop. The male subject had a slightly larger subnasale to soft tissue menton distance than did the woman: 65.7 vs 63.5 mm, respectively. This translated into larger lateral shifts of the chin with corresponding increases in rotation. By rotating the lower facial third in increments of degrees, the amount of horizontal linear discrepancy of the chin (soft tissue menton) could be easily calculated using basic trigonometry ( Fig 4 ). This formula was applied to all rotations to quantify the corresponding lateral chin asymmetries ( Table I ).
Male subject | Female subject | |
---|---|---|
Subnasale-soft tissue menton distance (mm) | 65.7 | 63.5 |
Rotation (°) | ||
1 | 1.15 | 1.11 |
2 | 2.29 | 2.22 |
3 | 3.44 | 3.32 |
4 | 4.58 | 4.43 |
5 | 5.73 | 5.53 |
6 | 6.87 | 6.64 |
To reduce bias, 1 lower facial third segment was rotated to the right (male subject) and the other to the left (female subject). Seven images per subject were produced, by rotating the segment in increments of 1°: 0°, 1°, 2°, 3°, 4°, 5°, and 6° ( Figs 5 and 6 ).
The survey was administered via online software (SurveyMonkey, San Mateo, Calif) and taken on the participants’ personal computers or devices. After agreeing to the study’s informed consent, the participants graded the photographs in order of esthetic quality on a visual analog scale from 0 to 10. Along with the altered photographs, 1 photograph (0° rotation) from each subject was shown twice to the participants as a control to examine reliability. The order of the photographs in the questionnaire was randomized for both male and female subjects. This randomized order was the same for all study participants. The largest asymmetry was shown near the end of the survey sample ( Table II ). All photographs of the female subject were shown first, followed by the male subject. The participants were not told what facial characteristics were altered; rather, they were asked to rate facial appearance on the visual analog scale.
Photo order | Sex and degree of rotation | Photo order | Sex and degree of rotation |
---|---|---|---|
1 | F 1° | 9 | M 2° |
2 | F 0° | 1° | M 0° |
3 | F 2° | 11 | M 1° |
4 | F control (°) | 12 | M control (°) |
5 | F 3° | 13 | M 6° |
6 | F 5° | 14 | M 4° |
7 | F 6° | 15 | M 5° |
8 | F 4° | 16 | M 3° |
Statistical analysis
Because of a limitation of the online software and to simplify the grading process for the survey takers, the visual analog scale was modified to a sliding scale from 0 to 10. Thus, a total of 11 grades were possible for the rater to select from. Repeated measures analysis of variance (ANOVA) was used to identify the diagnostic threshold levels of clinically significant chin asymmetry. This was independently calculated for both subjects across all 3 groups of survey takers.
Repeated contrasts were used to test for differences among the levels within subjects. Contrasts (like post hoc) are used to test for differences among the levels of a factor in a general linear model. Repeated contrasts (like a method of post hoc) is 1 contrast type. It compares the mean of each level (except for the last) with the mean of the subsequent level. In the SPSS software (version 24; IBM, Armonk, NY), a contrast other than none must be selected for within-subject factors (different degrees of chin asymmetry) in GLM repeated measures. We selected “repeated” as the contrast type.
The degree of incremental chin asymmetry was presented with the corresponding mean grade assigned by the survey participant. The mean grades were then compared with each other as the degree of asymmetry increased; thus, the first comparison, where a statistically significant difference in mean grade of esthetic appearance was produced, was the diagnostic threshold level for this group of survey takers. An alpha level of 0.05 was selected as significant. The corresponding 1-way ANOVA tests were used to determine whether there was a statistically significant difference in ratings of each photo among the 3 groups. Scheffé post hoc tests were used for the pairwise comparisons if the ANOVA test was significant. To test for reliability of the grades assigned to the photographs by the survey takers, identical control photographs (0°) of both subjects were presented twice in their separate categories for grading in random order in the series of photos to reduce grading bias.
Results
The diagnosis of asymmetries of the chin varied across the 3 groups of participants in this study, as well as between the female and male subjects. The diagnostic threshold levels across the 3 groups of raters are summarized in Table III . The findings showed that for the female subject the threshold for diagnosis of chin asymmetry was 3° (3° of rotation) for the laypersons and 2° for the dentists and orthodontists. For the male subject, the threshold for diagnosis of chin asymmetry was 2° for the layperson, dentists, and orthodontists.
Raters | Female subject Degrees ( P value) |
Male subject Degrees ( P value) |
---|---|---|
Laypersons | 3° (0.004) | 2° (0.007) |
Dentists | 2° (0.020) | 2° (0.041) |
Orthodontists | 2° (0.000) | 1° (0.001) |
The intrasurvey-taker reliability for the male and female subjects was good, indicating that the method chosen to verify whether the survey takers would equally recognize the same picture twice was adequate. This was concluded, since for both subjects, the 0° alterations and the 0° control received essentially identical mean scores by the survey takers ( Tables IV and V ).
Degree of chin asymmetry (°) | Survey participant group | Mean grade | SD | P value | Sheffé post hoc |
---|---|---|---|---|---|
0 | L D O |
7.73 7.00 7.30 |
1.683 1.578 1.672 |
0.048 ∗ | A———— ——B—— A—-B—— |
0 (control) | L D O |
7.80 6.91 7.24 |
1.535 1.699 1.655 |
0.010 ∗ | A———— ——B—— A—-B—— |
1 | L D O |
7.67 7.24 7.32 |
1.564 1.582 1.703 |
0.276 | A———— A———— A———— |
2 | L D O |
7.77 6.90 6.86 |
1.466 1.541 1.683 |
0.001 ∗ | A———— ——B—— ——B—— |
3 | L D O |
7.45 6.40 5.87 |
1.501 1.891 1.834 |
0.000 ∗ | A———— ——B—— ——B—— |
4 | L D O |
6.34 5.29 4.55 |
1.929 1.920 1.837 |
0.000 ∗ | A———— ——B—— ——B—— |
5 | L D O |
6.28 5.10 4.28 |
1.759 2.092 1.710 |
0.000 ∗ | A———— ——B—— ————C |
6 | L D O |
5.81 4.64 3.46 |
2.054 1.944 1.852 |
0.000 ∗ | A———— ——B—— ————C |