Introduction
Our objective was to study the perceptions of laypeople for digital alterations and the amount of nasolabial angle increase that is tolerable and how much chin-neck length increase is needed to achieve a desirable profile in Class II Division 1 women with mandibular retrognathia.
Methods
The profile image of a white woman with a Class II Division 1 mandibular rethrognatic profile was digitally modified to create 6 images: 3 with stepwise increased nasolabial angles of 113°, 121°, and 129°, and 3 with stepwise increased chin-neck distances of 51, 54, and 57 mm. These images were assessed and ranked by 155 white laypeople.
Results
The baseline profile was judged significantly as the least attractive. A nasolabial angle of 129° was judged as unattractive as the baseline profile. Profiles with a chin-neck length of 54 and 57 mm were equally judged as most attractive.
Conclusions
The untreated (baseline) profile was found to be least esthetic, as well as the profile with the largest nasolabial angle. Nasolabial angle increases up to 121° seem to be acceptable. Profiles simulating a chin-neck length increase as produced by surgery seem to be most favored.
Highlights
- •
The baseline profile was judged as least attractive.
- •
Increase in chin-neck length was more accepted than increase in nasolabial angle.
- •
The most acceptable nasolabial angle was 121° (4 SD).
- •
The most favorable range of chin-neck length increase was 54 (1.5 SD) to 57 mm (2 SD).
- •
There was no significant difference in the sexes in perceiving attractiveness.
Studies have often shown that attractive people are treated more positively in a variety of settings. They are considered smarter and more socially skilled than less attractive people. When judging beauty, the face is the most important item.
It is accepted that tooth movement can influence facial esthetics. Most people currently seek orthodontic treatment to improve their facial and dental esthetic appearance. Facial profile esthetics and the perception of the facial profile by laypeople and professionals have been studied extensively. According to the outcomes of many studies, an orthognathic profile is considered most attractive, whereas profiles with distinct prognathic and retrognathic mandibles are judged to be less attractive. Studies comparing judgments on pretreatment and posttreatment photographs of patients showed more favorable judgments of posttreatment esthetics. Other studies found no consensus when comparing the judgments of facial attractiveness by professionals and laypeople.
In adults, treatment of a dentoskeletal Class II Division 1 malocclusion can be managed in different ways. In camouflage treatment, the upper lip moves posteriorly, increasing the nasolabial angle. This approach could result in correct overbite and overjet, but the problem of a deficient mandible remains. Another treatment option is surgical advancement of the mandible, which can often improve the occlusion, dentofacial pattern, and profile esthetics. Surgical therapy increases the chin-neck length or, as defined in earlier studies, the throat length.
Soft tissue profile changes have been frequently described independently for camouflage treatment by extraction of the maxillary first premolars and retraction of the maxillary incisors and for mandibular advancement surgery. However, no study up to now has evaluated simultaneously esthetic changes produced by these 2 treatment options. Therefore, the aim of this study was to investigate the judgment of laypeople on the effect of different soft tissue alterations on the profile of a Class II Division 1 female subject with mandibular retrognathia, produced by simulated camouflage or mandibular advancement treatment, by showing the different alterations simultaneously: more specifically, evaluation of simulated digital alterations in the nasolabial angle and chin-neck length as perceived by laypeople, to investigate how much nasolabial angle increase is tolerable and how much chin-neck length increase is desired for successful profile improvement during treatment of a Class II Division 1 patient.
Material and methods
Because laypeople are more sensitive to changes in female images than to changes in male images, a profile image and a lateral cephalogram of a woman with an untreated skeletal Class II Division 1 relationship with normal face height and a normal mandibular plane angle were chosen. Both the profile image and the lateral cephalogram were scanned and imported into software (version 11.5; Dolphin Imaging and Management Solutions, Chatsworth, Calif) where the images were linked to each other to standardize the profile image with the given magnification of the lateral cephalogram.
The profile picture of the subject was first altered with Photoshop software (Adobe Systems, San Jose, Calif) to emphasize the mandibular retrusion. For the chin-neck length, the norm of 57 ± 6 mm as determined by Lehman minus 2 standard deviations, resulting in 45 mm, was used as the starting point for the alterations of this variable.
Using the Dolphin software, the profile image was further digitally altered to create 3 additional images simulating the treatment outcome of mandibular advancement surgery, increasing the chin-neck length by 1.0, 1.5, and 2.0 standard deviations (subsequently, 51, 54, and 57 mm) ( Fig 1 , E-G ).
The same was carried out for the nasolabial angle; First, Photoshop software was used to adjust this angle to the norm value of 104.9° ± 4° according to the study of Sinno et al. This norm was used as the starting point for further alterations. Using the Dolphin software, the nasolabial angle was digitally altered to create 3 additional profile types, simulating increases of the nasolabial angle by 2.0, 4.0, and 6.0 standard deviations (subsequently, 113°, 121°, and 129°:) ( Fig 1 , C, B, and A ).
The starting point for all alterations was the pretreatment profile ( Fig 1 , D ) with a chin-neck length of 45 mm and a nasolabial angle of 104.9°. This was the profile with the most pronounced Class II Division 1 features, displaying the greatest sagittal interlabial step.
After creating these profile images, we inserted them into a booklet that also included a questionnaire. This questionnaire collected information about age, sex, nationality, background, and educational level of the observers. Also, orthodontic treatment in the past was noted. This questionnaire provided information on factors that could influence the judgment of the laypeople.
White laypeople were approached on the streets and in public places. Those who had completed at least high school were included in this study. Approval from the ethics committee of the University of Sydney was obtained before starting this research (number X14-0355).
The profile images to be assessed were printed alongside each other on an A4 page as shown in Figure 1 . It was intended that participants should judge a set of profiles offering clear options by placing the most pronounced Class II Division 1 profile in the center ( Fig 1 , D ). The maximum achievable compensation with camouflage treatment was on the far left ( Fig 1 , A ), and the profile with the maximal mandibular skeletal correction was on the far right ( Fig 1 , G ), assuming a distinct selection option for the judges for the profile they might prefer.
Participants were asked to rank the profiles from 7, for most attractive, to 1, for least attractive. They were also asked whether they considered the profile images “acceptable” or “unacceptable.” Finally, all images were scored on a visual analog scale (VAS). This scale was numbered from 1 to 10; on the left of this scale, the description was “not attractive,” and on the right the description was very attractive.” The reliability test was carried out by 20 undergraduate dental students, who assessed the set of profile images with the VAS score twice with a washout period of 2 weeks.
Statistical analysis
Facial attractiveness rankings were analyzed using Friedman and Wilcoxon tests. To analyze the influence of background factors on rank, the Kruskal-Wallis test and Mann Whitney U test were used.
The VAS scores were analyzed using analysis of variance and paired t tests. For the analysis of background factors influencing the VAS scores, independent t tests and analysis of variance with post hoc analysis were used.
The Pearson correlation was also performed to analyze correlations between ranking and VAS scores. The intraclass correlation coefficient was used in assessing intrarater reliability for the VAS scores.
Results
A total of 155 laypeople assessed the profile images: 90 women and 65 men. Approximately 55% of the sample was between 18 and 30 years of age. A third of the sample had previous orthodontic treatment, and 90% of those had full fixed appliances.
The estimated means of the ranking were calculated for the entire group as shown in Table I . The judges considered profile A to be as unattractive as profile D. Profiles E, F, and G were judged more favorably than the other profiles. Profiles F and G were judged significantly better than the other profiles.
| A | (2) 2.81 B,C,E,F,G |
| B | (4) 3.46 A,D,E,F,G |
| C | (3) 3.41 A,D,E,F,G |
| D | (1) 2.54 B,C-E,F,G |
| E | (5) 4.79 A,B,C,D,F,G |
| F | (6) 5.47 A,B,C,D,E |
| G | (7) 5.53 A,B,C,D,E |
According to the VAS score, profile D was considered the least attractive by all judges. Profiles A, B, and C were considered successively and significantly more attractive than profile D. However, no significant difference was found between the mean scores of profiles A, B, and C. Similar to the ranking, profiles E, F, and G were found to be most attractive; profile E was significantly less attractive than profiles F and G ( Table II ).
| A | (2) 4.68 D,E,F,G | .19 |
| B | (3) 4.92 D,E,F,G | .17 |
| C | (4) 4.96 D,E,F,G | .15 |
| D | (1) 3.99 A,B,C,E,F,G | .18 |
| E | (5) 6.67 A,B,C,D,F,G | .14 |
| F | (6) 7.21 A,B,C,D,E | .14 |
| G | (7) 7.31 A,B,C,D,E | .18 |
The consistency of the responses of the judges was tested by intraclass correlation coefficients, calculated for the VAS scores, resulting in .623 with a confidence interval of 95%.
Positive Pearson correlations were found between the rankings and the VAS scores, ranging from r = .362 to r = .607 ( Table III ). A relatively large discrepancy between ranking and VAS score was found for profile E.
