The patient’s perception of facial esthetics is not necessarily consistent with that of the practitioner. The aim of this study was to compare the perceptions of Persian orthodontists, oral surgeons, and laypersons with regard to facial profile attractiveness and the most favorable mandibular position.
Software (Dolphin Imaging and Management Systems, Chatsworth, Calif) was used to alter the mandibular position depicted on profile photographs of a young man and a young woman. Nine construction profile photos were produced with incremental changes of the G′-Sn-Pg′ angle at 2° intervals (6° to 22°). Thirty-two orthodontists, 32 maxillofacial surgeons, and 32 laypersons were asked to score all 18 profiles on a 1-to-10 visual analog scale. They also evaluated whether each profile needed orthognathic surgery for improvement of facial esthetics. Actual agreement and intraclass correlation coefficient tests were used to assess reliability. The data were analyzed using 2-way repeated measures analysis of variance and Wilcoxon signed-rank tests.
The intraparticipant reliability was acceptable (intraclass correlation coefficient >72%; actual agreement, >79%). The sex of the participants was not a significant factor influencing the scores, although the sex of the models was a discriminating factor for the most acceptable mandibular horizontal position. The scores given by the 3 groups were different, especially for the female model. The orthodontists had similar opinions and preferred a slightly more protrusive mandible (G′-Sn-Pg′, 12°-14°). Laypersons’ scores were the most inconsistent, and they generally preferred a retrusive profile (G′-Sn-Pg′, 14°-18°). There was no significant correlation with regard to the necessity of orthognathic surgery for G′-Sn-Pg′ angles over 20° in men and below 8° in women.
Although the laypersons’ perceptions were slightly different from those of the clinicians, most participants preferred a more protrusive mandible for male models compared with female models.
In Persians, the level of education was a factor in judgment of attractiveness.
Level of education was not a factor in advocating surgery.
Sex of the clinician did not influence judgment of profile or need for surgery.
Judges preferred a Class II profile for Persian women.
They preferred a slight Class III profile for Persian men.
Establishment of a proper occlusal relationship and the best possible facial esthetics are the goals of modern orthodontics. There are several occlusal indexes, such as the 6 keys of optimal occlusion by Andrews. However, achieving the best esthetic results and a well-balanced facial appearance are more challenging due to the subjectiveness of esthetics, leading to different facial evaluations. This has great importance when managing patients who seek cosmetic treatment. The patient’s perception of facial esthetics is not necessarily consistent with that of the orthodontist or the oral surgeon. This disparity between the patient and clinician could cause patients to be dissatisfied with treatment outcomes and make treatment procedures more complicated. On the other hand, the orthodontists’ and oral surgeons’ evaluations regarding the most pleasing profile for the patient affect both the orthodontic and surgical treatment plans. Thus, orthodontists and oral surgeons should reach consensus on a plan that addresses the patient’s facial esthetic concerns. Otherwise, even a highly standard and professional treatment may fail to achieve patient satisfaction. There is no definite index or gold standard for facial profile attractiveness because esthetics is highly affected by race and culture, and its related concepts are influenced by many factors such as ethnicity, nationality, sex, age, education, and profession. Research has shown that although science and technology have decreased intercultural variations in modern societies, the evaluations of facial attractiveness still vary. This can be attributed to the variations in skeletal and soft tissue patterns of various ethnic groups. Mejia-Maidl et al compared the concepts of 2 ethnic groups relative to profile attractiveness and found that Mexican Americans preferred more retruded lips, particularly in female profiles, compared with white subjects. Nomura et al reported that Hispanic Americans and Japanese also preferred retruded lip profiles compared with Africans.
Thus, before treatment planning, the clinician ought to be aware of the patient’s preference and his or her esthetic concerns. Most studies regarding esthetic perceptions have been conducted on white subjects; limited information is available in Asian and Middle Eastern populations. Furthermore, even in 1 ethnic population, the esthetic perception may be influenced by professional background and may vary between clinicians. Lines et al reported significant differences in perceptions of facial profile esthetic evaluations among orthodontists, oral surgeons, other dental professionals, and laypersons. Prahl-Anderson et al studied the perceptions of different professional groups regarding dentofacial morphology and found no significant difference between the perceptions of orthodontists and general dentists; however, there were significant differences between those of parents and professionals. Facial esthetics is multifactorial. Orthodontists typically base their judgments of facial esthetics on profile, full-face, and smile evaluations of the patient. The soft tissue profile affects the attractiveness of the whole face and has been studied extensively in various populations. Mandibular position as a contributing factor to profile esthetics has been largely studied using skeletal (eg, cephalometric technique) or soft tissue measures.
The aims of this study were to determine and compare the perceptions of orthodontists, oral surgeons, and laypersons with regard to facial profile attractiveness and the most favorable mandibular position in a Persian population.
Material and methods
In this study, a 17-year-old girl and a 21-year-old man were selected as role models. The criteria for choosing them included a well-balanced face; normal eyes, lips, and nose; Class I skeletal relationship; and normal vertical skeletal pattern according to, the Holdaway soft tissue analysis, the Z-angle, and the Ricketts’ esthetic planes. Most measurements of the profiles were in the normal ranges. All parameters were confirmed by 2 experienced orthodontists (E.S., E.A.N.). Consent forms were signed by the 2 models, and high-resolution standard color photographs with a white background and proper brightness were obtained. Lateral cephalograms based on the Frankfort horizontal plane were traced using software (Dolphin Imaging and Management Systems, Chatsworth, Calif). Fifty-one anatomic landmarks (14 soft tissue, 37 hard tissue) were identified. The 2 profile photographs were inserted into the Dolphin software and linked to their respective cephalograms via superimposition of 4 anatomic landmarks.
The profile of each model was traced, and the soft tissue convexity angle (G′-Sn-Pg′) based on the analysis of Legan and Burstone was defined. The facial profile angle was measured by drawing lines from the soft tissue glabella (G′) to subnasale (Sn) and to soft tissue pogonion (Pg′) ( Fig 1 ). The G′-Sn-Pg′ angle of the original picture was 12°. The images were digitally manipulated by changing the mandibular position by 2° increments of the G′-Sn-Pg′ angle (−8°, −6°, −4°, −2°, 0°, +2°, +4°, +6°, and +8°) using the Dolphin software; thus, a set of 9 profiles was created for each model.
The profile with the G′-Sn-Pg′ angle of 4° was then omitted, and a profile with the G′-Sn-Pg′ angle of 22° was added (this decision was made because the profile with the G′-Sn-Pg′ angle of 4° had a dramatic abnormal mandibular protrusion, whereas the profile with the G′-Sn-Pg′ angle of 22° was apparently more acceptable). Therefore, the middle image was set at a G′-Sn-Pg′ angle of 14°. The final 9 profile images (G′-Sn-Pg′, 6°, 8°, 10°, 12°, 14°, 16°, 18°, 20°, and 22°) accounted for the possible anteroposterior growth variations of the mandible. After reconstructing the profiles, the borders of the images were airbrushed out with PaintShop Pro (version 7.0; Corel, Ottawa, Ontario, Canada) to maintain the natural appearance of the images especially in the lip area. Finally, the 18 photographs were printed (4 × 6 in) and placed randomly in 2 albums with no special order (an album for the male model and another for the female model) ( Fig 2 ).
The participants consisted of 32 orthodontists and 32 oral and maxillofacial surgeons practicing in 4 dental schools in addition to 32 laypersons recruited from the patients in the waiting areas of 2 dental clinics with an age range of 18 to 35 years. They were well educated but not engaged in or trained for any job related to facial esthetics. A visual analog scale (VAS) was designed for each image, and the participants were asked to score the profiles in order from least to most attractive. They were also required to answer a yes-or-no question regarding whether they perceived that the patient needed surgery to improve the profile. So that we could evaluate interobserver reliability, each participant was asked to rate the profiles again 2 weeks later.
Intraclass correlation coefficients (ICC) and actual argument were used to determine the reliability of the VAS scores and the yes-or-no questions, respectively. Two-way repeated-measures analysis of variance (ANOVA) was applied to compare the means of the scores. Regarding the significant interaction between the profile and sex of the models, 1-way repeated measures ANOVA was used for each model independently. Since the sphericity assumption was violated (Mauchly’s sphericity test, P <0.0001), Greenhouse-Geisser adjusted P values and post-hoc paired comparisons with the Sidak test were used to identify significant differences between the profile angles. The Wilcoxon signed-rank test was performed to compare the percentages in the yes-or-no questions. Data analysis was carried out using SPSS software (version 16.0; SPSS, Chicago, Ill) and R software; the significance level was set at 0.05.
Regarding intraparticipant reliability, the ICC for the attractiveness score of the participants was more than 72% for all participants. The actual agreement for the yes-or-no questions about the need for surgical treatment was more than 79%, which expressed acceptable reliability for participant perceptions ( Table I ).
|Facial profile angle (°)||Female||Male|
|Actual agreement||ICC||Actual agreement||ICC|
To assess the validity of the scores, each data group was put into a box plot to identify outliers and extreme outliers; this resulted in no data exclusion. For the VAS scores, 2-way repeated measures ANOVA showed a significant interaction between the facial angle of convexity and sex ( P <0.001); as a result, 1-way repeated measures ANOVA was performed for each group separately. Since no significant difference was observed between the scores for the perceptions of the male and female raters ( P >0.05), the data were pooled and analyzed together.
The interaction between the facial angle of convexity of the female model and the participants’ educational level was significant regarding the female model ( P = 0.02). Thus, the evaluation of the most favorable angle was done separately for each group ( Fig 3 ). The orthodontists gave the highest scores (9.06 ± 1.19) to the profile angle of 14°; this was significantly higher than the scores of 12° and 16° ( P <0.001) ( Table II ).
|Facial profile angle (°)||Orthodontists||Maxillofacial surgeons||Laypersons||Total|
|6||1.72 ± 0.81 a||1.82 ± 1.68 a||2.84 ± 2.68 ab||2.12 ± 1.93|
|8||2.34 ± 1.47 ab||1.91 ± 1.49 a||2.66 ± 2.21 a||2.3 ± 1.76|
|10||4.84 ± 2.02 d||4.27 ± 2.12 b||4 ± 2.56 abc||4.37 ± 2.25|
|12||6.81 ± 2.06 d||6.97 ± 2.23 de||5.91 ± 2.53 de||6.57 ± 2.3|
|14||9.06 ± 1.19 e||7.33 ± 2.15 de||7.53 ± 2.12 ef||7.97 ± 2.04|
|16||7.44 ± 1.48 d||7.88 ± 1.67 e||7.97 ± 1.99 f||7.76 ± 1.72|
|18||6.66 ± 2.32 dc||5.76 ± 2.02 cd||6.97 ± 2.22 ef||6.54 ± 2.23|
|20||4.44 ± 1.68 cd||4.61 ± 2.09 bc||5.47 ± 2.31 cd||4.84 ± 2.08|
|22||3.28 ± 1.42 bc||4.06 ± 1.89 b||4.78 ± 2.56 bcd||4.04 ± 2.08|
The oral surgeons assigned the highest scores to the profile angle of 16° (7.88 ± 1.99) showing a significant difference compared with the scores of 18° ( P <0.001) but not significantly higher than those of 14° ( P = 1) and 12° ( P = 0.91) ( Table II ).
According to the laypersons’ points of view, the profile angle of 16° had the most favorable appearance (7.97 ± 1.99); however, it was not significantly different from those with angles of 12°, 14°, and 18° ( Table II ).
The interaction between the facial angle of convexity of the male model and the participant’s educational level was not significant ( P = 0.17). Thus, the evaluation of the most favorable angle was done for all participants of the 3 groups simultaneously. Figure 3 shows that the highest scores were for the profile angle of 12° (8.31 ± 1.81), which was not significantly different from those of 10° ( P = 1), but significantly higher than those of 8° ( P <0.001) and 14° ( P <0.001) ( Table III ).
|Facial profile angle (°)||Orthodontists||Maxillofacial surgeons||Laypersons||Total|
|6||2.28 ± 1.28||1.73 ± 1.07||2.28 ± 1.8||2.09 ± 1.42 a|
|8||4.06 ± 2.36||3.03 ± 2.31||3.5 ± 2.33||3.53 ± 2.35 c|
|10||8.16 ± 2.3||8.61 ± 1.08||7.63 ± 2.68||8.13 ± 2.3 f|
|12||8.63 ± 1.34||8 ± 2.02||8.31 ± 1.99||8.31 ± 1.81 f|
|14||5.97 ± 1.69||5.39 ± 2.09||6.06 ± 2.17||5.8 ± 2 e|
|16||4.97 ± 2.02||4.58 ± 2||5.16 ± 2||4.9 ± 2 d|
|18||2.34 ± 1.31||3.03 ± 2.08||3.44 ± 2.34||2.94 ± 1.99 bc|
|20||2.13 ± 1.6||2.58 ± 1.52||3.03 ± 2.21||2.58 ± 1.82 ab|
|22||2.13 ± 1.43||2.24 ± 1.52||3.53 ± 2.3||2.63 ± 1.88 ab|