Keywords
esthetics - laypeople - photograph - silhouette - skeletal Class II
Introduction
Orthodontic practices have undergone significant transformations in clinical evaluation,
diagnosis, and treatment planning over time. Presently, a key focus lies in enhancing
facial esthetics as a primary treatment goal.[1]
[2]
[3] Patients pursue orthodontic treatment for various reasons, with psychosocial concerns
being a significant motivator. Particularly, young adults often seek treatment to
enhance their appearance or improve facial or dental imperfections because enhancement
of facial and dental appearance can have a major impact on improving the quality of
life.[1]
[4]
[5]
When addressing Class II patients, treatment options may involve growth modification,
orthognathic surgery, or utilizing dental compensation techniques to conceal skeletal
discrepancies. While all approaches lead to enhancements in the sagittal interlabial
step and as well as the harmonious beauty of the face, they target different facial
features and yield distinct outcomes, particularly in terms of facial esthetics.[1]
[6] The camouflage treatment in skeletal Class II typically involves retracting the
maxillary incisors, resulting in an increased nasolabial angle (NLA).[7]
[8] Conversely, surgical interventions aim to improve skeletal problem through methods
of mandibular advancement, maxillary setback, or both. These surgical corrections
also affect the facial profile by decreasing the facial contour angle (FCA).[9]
[10] Studies indicate that both of these changes are considered attractive.[7]
[8]
[9]
[10] Moreover, esthetic medicine plays an important role in today's treatments. Soft
tissue filler injections in the chin area enhance the likelihood of patient satisfaction,[11]
[12]
[13] which is crucial in facilitating camouflage treatments.
The studies that investigated the influence of assessors' personal profiles on esthetic
perception are small in number and inconsistent. Some of the studies found similarities
in facial esthetic perception with a slight difference among each facial profile group.[9]
[14]
[15] But there was a study that found differences in esthetic perception of adults with
different facial profiles.[16] Same researchers also found that adolescents (13–18 years) and adults with straight
profiles were more satisfied with their profiles than those with convex and concave
profiles. However, there was no study focusing on the esthetic perception of the Class
II corrections.
The tools commonly used for assessing esthetic preferences in the profile view are
the patient's photograph or the silhouette.[17] The patient's photograph was suggested to be used because the patient's facial appearance
can be illustrated more than the silhouette,[18]
[19] and using the silhouette could cause the clinician to select the profile differently
from the esthetic norm.[19] However, some researchers have used the silhouette to remove the influence of other
factors such as eye, skin complexion, and hair color that might bias the perception
of facial attractiveness.[20]
[21]
This study aims to investigate (1) the influences of assessors' personal profiles
on the esthetic perception of Class II facial profile corrections and (2) the agreement
of esthetic perception in selecting between the profile and silhouette images. The
results of the study can be used as part of the data to establish treatment planning
and decision-making processes, and can also be utilized in future studies.
Materials and Methods
Sample
The estimated sample size was calculated by using the n4Studies application (version
1.4.1),[22] using the formula for testing the infinite population proportion, n = (z
2
1 − α/2
p [1 − p])/d
2, with p = 0.571, d = 0.1, and α = 0.05 to detect the posttreatment preference of the laypeople.[8] The results indicated that a total of 95 participants were needed. The participants
consisted of 96 Thai laypeople randomly picked from Chulalongkorn university dental
clinic, shopping malls, and educational institutions in Bangkok, Thailand, categorized
into three groups based on their FCAs[23]: straight (FCA 5–13 degrees), convex (FCA >13 degrees), and concave (FCA <5 degrees).
Each group consisted of 32 participants with a 1:1 male-to-female ratio. Exclusion
criteria were participants who are dental professionals (consisting of dentists, dental
students, dental assistants, and dental hygienists), younger than 16 years, older
than 40 years, with a history of facial trauma, conditions with syndromes, or serious
medical conditions. The measurement of participants' FCA was done by taking a photo
of each participant in the right nonsmiling profile view with the teeth in centric
occlusion and lips relaxation. The participants were positioned 5 feet from the camera
with their heads in a natural posture. Measurement of the FCA was done by a single
researcher (W.T.), using Adobe Photoshop 2020 (Adobe Systems Inc., San Jose, California,
United States). The FCA is defined by the points: soft tissue glabella (G′), subnasale
(Sn), and soft tissue pogonion (Pog′) as illustrated in [Fig. 1]. Twenty percent of the participants' FCA (19 participants) were measured again at
least 4 weeks after the initial assessment to determine the reliability of the examiner.
Fig. 1 Facial contour angle: an angle defined by soft tissue glabella (G′), subnasale (Sn),
and soft tissue pogonion (Pog′).
Photo Album
A photograph capturing the right nonsmiling profile view was taken of a female subject
presenting the following characteristics: an untreated skeletal Class II relationship
with orthognathic maxilla, retrognathic mandible, straight forehead, straight nose
dorsum, normal vertical proportion, and a normal mandibular plane angle.[23] The subject was positioned at a distance of 5 feet from the camera, maintaining
a natural head posture. Permission was granted by the subject for the photograph to
be taken, the image to be adapted, and for online publication.
The profile picture was initially modified with Adobe Photoshop 2020 (Adobe Systems
Inc.) to accentuate the mandibular retrusion. This was achieved by increasing the
FCA by 2 standard deviations (SD), resulting in a 17-degree FCA, using the Thai FCA
norm of 9 ± 4 degrees as indicated by Sorathesn.[23] Additionally, the NLA was adjusted according to the Thai NLA norm of 91 ± 8 degrees,[23] resulting in a 91-degree NLA. Hence, the “base image” used for further modifications
depicted a profile with a 17-degree FCA and a 91-degree NLA, representing the largest
sagittal interlabial step.
Two additional alteration images were created, using the “base image” in Photoshop,
focusing on changes in the anteroposterior plane while maintaining vertical proportion.
In the first image, the soft tissue Pog′ point was advanced to reduce the FCA by 1.5
SD, resulting in an 11-degree FCA.[23]
[24]
[25] Additionally, the mentolabial sulcus was adjusted in a 1:1 ratio following the movement
of the soft tissue Pog′. In the second image, the labrale superius point (Ls point)
was retruded to increase the NLA by 2.0 SD, resulting in a 107-degree NLA.[7]
[23]
[26]
In summary, three altered profiles ([Fig. 2]; images Ap, Bp, and Cp) were created. One depicted the most pronounced Class II
division 1 characteristic, with a 17-degree FCA and 91-degree NLA which was an untreated-simulating
profile ([Fig. 2]; image Bp). Another simulated a more retruded upper lip position which was a camouflage-simulating
profile ([Fig. 2]; image Ap). The third simulated a more protruded mandibular position which was a
mandibular advancement-simulating profile ([Fig. 2]; image Cp). Subsequently, all three altered profile images were converted to black
and white ([Fig. 2]; images Ap, Bp, and Cp) and modified to produce the silhouette images ([Fig. 2]; images As, Bs, and Cs).
Fig. 2 Altered profile and silhouette images; most pronounced Class II division 1 characteristic
(Bp and Bs), more retruded upper lip position (Ap and As), and more protruded mandibular
position (Cp and Cs). FCA, facial contour angle; NLA, nasolabial angle; SD, standard
deviation. (Adapted with permission from Tipyanggul et al.[31])
All three altered profile images were simultaneously presented to the participants
using the photos application on iPad Pro 10.5-inch (Apple Inc., Cupertino, California,
United States). Three profile images were placed alongside each other. The base image
with the most pronounced Class II division 1 characteristic ([Fig. 2]; image Bp) was placed in the center, the image with an increased NLA ([Fig. 2]; image Ap) was on the left, and the image with a decreased FCA ([Fig. 2]; image Cp) was on the right.
The first page of the photo album showed images as depicted in [Fig. 2]; images Ap, Bp, and Cp. Pages 2 and 3 featured the same image set but with positions
randomly arranged to create a washout effect. Page 4 displayed the silhouette images
as in [Fig. 2]; images As, Bs, and Cs, also in randomly arranged positions. These pages, from 1
to 4, were utilized to fulfill Part 2 of the questionnaire.
Twelve participants (four per group) were asked to reassess the altered profile images
at least 4 weeks after the initial assessment to determine the reliability of the
test.
Questionnaire
In Part 1 of the questionnaire, participants provided demographic details including
age, sex, ethnicity, and level of education. Part 2 required participants to view
each page of a photo album. The participants were also asked, in the form of a closed-ended
question, to select the profile image they found most attractive in terms of facial
appearance. Participants were given 60 seconds to complete the assessment for each
page and were asked not to go back to the page they had already assessed. The questionnaire
utilized in this study is available in [Supplementary Materials S1] (available in the online version only).
Statistical Analysis
The statistical analysis was conducted using SPSS version 22.0 for Mac (IBM, Chicago,
Illinois, United States). Intraparticipant and intra-examiner reliabilities were evaluated
by calculating Cohen's kappa coefficient and intraclass correlation coefficients,
respectively. The profile preference among laypeople of different facial profiles
(between-group preference) was tested using chi-square tests. The relationship of
the profile preference with other factors (sex and level of education) was analyzed
using chi-square tests. The agreement in selecting the profile and silhouette images
was assessed with Cohen's kappa coefficient. A significance level of 0.05 was set
for all tests.
Results
Demographic Data
Ninety-six participants were stratified into three groups (n = 32): straight, convex, and concave. Each group had an equal male-to-female ratio
of 1:1. [Table 1] presents the baseline data for the three groups, including varying levels of education
within each group.
Table 1
Baseline data of the three profile groups
|
Group
|
|
Straight
|
Convex
|
Concave
|
|
Age median (Q1, Q3)
|
21.89 (18.23, 25.49)
|
22.31 (18.87, 25.91)
|
22.39 (20.29, 26.11)
|
|
Level of education (n)
|
|
Nongraduate
|
14 (43.8%)
|
13 (40.6%)
|
8 (25%)
|
|
University graduate
|
18 (56.2%)
|
19 (59.4%)
|
24 (75%)
|
Reliability Coefficients of Intraparticipants and Intra-examiner
The intrarater reliability of the participants in the profile and silhouette images
was substantial (0.692 and 0.667, respectively). The reliability of the examiner was
excellent (0.999).
Profile Preference among Laypeople of Different Facial Profiles
The chi-square test did not show a significant difference in the profile preference
among laypeople of different facial profiles (p = 0.649) ([Table 2]). Image Ap was chosen by 25% of the straight group, 12.5% of the convex group, and
21.9% of the concave group. Image Bp was chosen by 9.4% of the straight group, 18.8%
of the convex group, and 12.5% of the concave group. Image Cp was chosen by 65.6%
of the straight group, 68.8% of the convex group, and 65.6% of the concave group.
Table 2
Profile preference among laypeople of different facial profiles (between-group difference)
|
Preference
|
Group
|
|
N (%)
|
Straight
|
Convex
|
Concave
|
|
Ap
|
8 (25%)
|
4 (12.5%)
|
7 (21.9%)
|
|
Bp
|
3 (9.4%)
|
6 (18.8%)
|
4 (12.5%)
|
|
Cp
|
21 (65.6%)
|
22 (68.8%)
|
21 (65.6%)
|
|
Pearson's chi-square value = 2.477
p-Value = 0.649
|
Profile Preference among Laypeople within the Same Facial Profile
The most chosen profile for all 96 participants was a more protruded mandibular position
profile (image Cp) (straight group: 65.6%, convex group: 68.8%, and concave group:
65.6%), followed by a more retruded upper lip position profile (image Ap) in the straight
(25%) and concave (21.9%) groups. While the convex group preferred the most pronounced
Class II division 1 profile (image Bp) (18.8%) more than image Ap. The significant
differences were found between images Cp–Ap and images Cp–Bp ([Fig. 3]).
Fig. 3 Profile preference among laypeople within the same facial profile (within-group difference).
(Ap—17-degree FCA and 107-degree NLA, Bp—17-degree FCA and 91-degree NLA, and Cp—11-degree
FCA and 91-degree NLA). FCA, facial contour angle; NLA, nasolabial angle.
The Relationship of the Profile Preference with Other Factors
According to the chi-square test, the profile preference did not differ by sex (p = 0.198) and level of education (p = 0.105).
The Agreement in Selecting the Profile and Silhouette Images
The profile preferences for profile and silhouette images are shown in [Table 3]. The percentage of agreement between the methods in the total sample was 67.71%,
which could be considered as a fair agreement (kappa = 0.386). A more protruded mandibular
position profile (image C) was the most chosen in both profile and silhouette images.
Table 3
The agreement in selecting the profile and silhouette images
|
Preference
|
Images
|
|
N (%)
|
Profile
|
Silhouette
|
|
A
|
19 (19.8%)
|
27 (28.1%)
|
|
B
|
13 (13.5%)
|
11 (11.5%)
|
|
C
|
58 (60.4%)
|
|
64 (66.7%)
|
|
Kappa = 0.386
Percentage of agreement = 67.71%
|
Discussion
The present study investigated the influence of the assessors' facial profile on the
esthetic perception of Class II facial profile corrections. The results demonstrated
that the esthetic preference among laypeople with different facial profiles was similar.
There was no significant difference in the profile preference among laypeople of different
facial profiles. Volpato et al[15] reported that the pleasantness scores assigned by the patients of three facial profile
types were not different, although the patients with straight profile assigned slightly
greater scores than patients with concave or convex profiles. Suphatheerawatr and
Chamnannidiadha[9] also found that participants with different facial profiles had similar facial profile
preferences.
Image Cp with normal NLA (11-degree FCA and 91-degree NLA) was the most chosen profile
for all three groups of laypeople, followed by image Ap with obtuse NLA (17-degree
FCA and 107-degree NLA) in the straight and concave groups. The least chosen profile
for the straight and concave groups was image Bp with the most pronounced Class II
division 1 profile (17-degree FCA and 91-degree NLA). While the convex group preferred
image Bp more than image Ap. These results indicated that the convex group of laypeople
preferred the convex profile more than other groups. Jarungidanan and Sorathesn[14] reported that the subject with a convex profile accepted convex profiles equally
or more than any other profile subjects. Suphatheerawatr and Chamnannidiadha[9] also reported that an extremely convex profile was preferred by the convex group
of the assessors more than the concave group of the assessors.
Moreover, the images Ap and Cp were the profile images that were intended to simulate
the Class II treatment which was a camouflage-simulating profile and mandibular advancement-simulating
profile, respectively. These two profile images were chosen more than image Bp which
was the most pronounced Class II division 1 characteristic (untreated-simulating profile).
These results indicated that treating Class II patients with dental compensation or
orthognathic surgery has different effects on attractiveness according to laypeople.
Camouflage treatment for skeletal Class II typically involves retracting the maxillary
incisors, which increase the NLA.[7]
[8] On the other hand, surgical interventions aim to correct skeletal problems through
methods such as mandibular advancement, maxillary setback, or both. These surgical
corrections also influence the facial profile by reducing the FCA.[9]
[10] Yüksel et al[27] also reported that the untreated profile was found to be least preferred and the
mandibular advancement and camouflage treatment were considered more attractive than
the untreated profile with the mandibular advancement profiles more attractive than
the camouflage treatment profiles.
This study found the fair agreement in selecting the profile and silhouette images.
Participants tended to choose the profile and silhouette images in similar trend.
The most chosen profile was image Cs, followed by image As, and the least chosen was
image Bs. However, the number of the participants that chose image Cs was less in
silhouette images with the increasing number of the participants that chose image
As. These results showed that the profile flatter than the esthetic norm was more
preferred in silhouette than in photograph. Hockley et al[19] also found that using the photographs in the profile esthetic assessment, the participants
preferred the photographs that closer the esthetic norm more than using silhouettes.
While using the silhouettes, the participants tended to select the flatter profile
than the esthetic norm.
This study aimed to isolate the factors influencing esthetic perception. Thus, all
three profiles were adjusted to achieve normal vertical proportions[21] and a straight nose dorsum. Prior research has shown that a straight nose dorsum
is perceived as more esthetically pleasing, as opposed to differing nose shapes in
Class II profiles.[28] Therefore, variations in vertical proportions and nose shapes among Class II patients
could yield different esthetic preferences compared with those found in this study.
Treatment planning in orthodontics involves considering various factors such as incisor
display, gingival display, tooth proportion, gingival shape and contour, and tooth
shade. The findings of this study should be regarded as one component among many others
in aiding the treatment planning process.
The purpose of altering the profile images in this study was to establish different
degrees of NLA and FCA, corresponding to mandibular advancement or camouflage treatment.
These adjustments also affected soft tissue in other facial areas, particularly the
sagittal interlabial step. Additional images altered from the “base image” displayed
varying degrees of sagittal interlabial steps, with less prominence compared with
the base image. This was achieved through retrusion of Ls point (image Ap) and the
advancement of Pg′ point which led to the advancement of the lower lip (image Cp).
Interestingly, the least favored profile image among the straight and concave groups
of laypeople was the “base image” Bp, which had the largest sagittal interlabial step.
These findings align with Yüksel et al's study,[27] which concluded that reducing the sagittal interlabial step through changes in NLA
or FCA increased assessor satisfaction compared with profiles with larger interlabial
steps. Moreover, prior research has shown that postoperative changes in the Pg′ point
and mentolabial sulcus correlate with changes in underlying hard tissue at a 1:1 ratio.[24]
[29]
[30] Thus, in altering image Cp, adjustments to the mentolabial sulcus were made in accordance
with these previous findings.
This study was primarily concerned with the personal profiles of participants, with
a specific focus on their FCA. Consequently, data collection was centered around assessing
this parameter. Acknowledging the influence of age on preferences as the results found
in the previous study,[31] participants recruited for this study ranged in age from 16 to 40 years. Varatharaju
et al[32] observed that self-recognition of facial profiles tends to improve with age, with
individuals older than 15 years showing significantly better recognition compared
with younger subjects. Additionally, young adults between the ages of 20 and 39 years
often seek orthodontic treatment for esthetic improvement or correction of perceived
defects.[1] However, other participant factors were not controlled, making it challenging to
regulate sample size across these variables. Our findings revealed no statistically
significant differences in profile preference related to sex and level of education,
aligning with the results reported by Pithon et al,[17] who found no significant differences based on race, sex, or educational background.
To explore these factors further, we recommend future studies collect larger, more
evenly distributed samples across these variables. Additionally, this study focused
on participants of the same ethnicity to control for this contributing factor, but
future research could include participants from diverse ethnic backgrounds for a broader
understanding of profile preference trends.
Based on the factors we attempted to control, there are limitations in this study
concerning variables such as ethnicity, participant age, vertical image proportion,
and nasal shape. As a result, the study's data are applicable clinically as only a
part of the decision-making process, rather than as the sole determining factor. Thus,
we advocate for future research endeavors to further investigate the controlled factors
in this study to enhance clinical knowledge and applicability.
Conclusion
A mandibular advancement-simulating profile was the most preferred profile for all
three groups of participants. There was no significant difference in the profile preference
among laypeople of different facial profiles, but the convex group tended to prefer
an untreated-simulating profile more than the straight and concave group. The profile
preference did not differ by sex and level of education. Using the photographs or
silhouettes to assess the esthetic preference resulted in a similar trend with the
profile flatter than the esthetic norm was more preferred in silhouette than in the
photograph.
Orthodontists and patients can use these findings as part of the data to establish
their treatment planning and decision-making processes. However, successful treatment
planning in each case requires consideration of numerous factors, such as correct
diagnosis, appropriate timing of treatment, and the patient's treatment goals.
