Esthetic perception of gummy smile with the application of botulinum toxin, via eye-tracking

We aimed to evaluate laypeople’s visual perception of the gingival smile before and after botulinum -toxin application, using eye- tracking technology. Two frontal photographs of a female volunteer’s face with a gingival smile were used. One before and another 30 days after the toxin application in the upper lip elevator muscle region. The images were evaluated by 58 participants. Using the OGAMA software, visualization data were obtained in three areas of interest (AOI) through eye-tracking. A visual analog scale (VAS) was used to assess facial attractiveness, and a questionnaire to assess the perception of sympathy, satisfaction with the smile, dissatisfaction with dental and gingival exposure, and shyness. Heat and point maps showed that the participants visually concentrated on the mouth region, with a small transition to the nose and eyes in the pre-procedure image. The mouth region was the AOI where the first visualization occurred in a shorter time compared to the latter areas (p<0.05). VAS revealed that attractiveness increased in the image with less gingival exposure (p<0.05). After the toxin application, satisfaction with the smile increased (p<0.05), and dissatisfaction with gingival exposure decreased (p<0.05). However, dissatisfaction with dental exposure increased (p<0.05). We discovered that the mouth attracted greater attention from participants in less time compared to other AOI. The decrease in gingival exposure after the botulinum toxin application resulted in improved facial esthetics. There was a perceptible increase in “satisfaction with the smile” and a decrease in “dissatisfaction with gingival exposure.” Conversely, the “dissatisfaction with dental exposure” increased.


Introduction
An esthetic smile is the result of a combination of different components, such as soft tissue, dental, and gingival exposure. This is one of the goals for dental treatments, as these components must balance with each other (Wang et al., 2018).
The length of the upper lip, the size of the incisors, and excessive gingival exposure are factors that have the greatest impact on the smile (Albino et al., 1994). Specifically, the gingival smile, recognized as an esthetic problem, is influenced by all of these elements (Guo et al., 2011).
Lip movements control the amount of gingival exposure, dictating the composition of the smile. Thus, the visibility of the gingiva and the dental disposition are formed by the lip position and the height of the smile line (Flanary, 1992). A smile with more than 2 mm of gingival exposure can be seen as unsightly. Therefore, dissatisfaction with appearance is usually the main reason for seeking treatment (Kattimani et al., 2019), especially by women. This is because the prevalence of gingival smiles in women is twice as high as that of men (Tjan et al., 1984). This is often attributed to excessive contraction of the muscles of the upper lip. This is also a possible cause of the alteration of passive eruption pathways, vertical maxillary excess, anterior tooth-alveolar extrusion, and thin upper lip (Silberberg et al., 2009). A recommended treatment is botulinum toxin type A, which, when applied in small doses, limits the contraction of the upper lip elevating muscles and maintaining the ability to smile (Jaspers et al., 2011). This makes it a minimally invasive intervention option to be considered in the treatment plan. (Polo, 2008).
Studies concerning the esthetics of the smile and its interference in facial attractiveness contribute to determining the need for intervention (Gasparello et al., 2022). Thus, this can be performed using different methodologies and perspectives (Chang et al., 2011;Moore et al., 2005;Parrini et al., 2016). Eye-tracking for visual perception analysis consists of assessing the position and movement of the eyes using an eye-tracking device, verifying the path that the eye takes in a given image, providing data such as the path taken, time until the first fixation, and total fixation time in a given area (Itti, 2004). The location that receives the most attention from the viewer will be the area with the most fixations and where the focus occurs for the longest time; thus, allowing us to assess which characteristic holds the greatest interest (Berlyne, 1958).
Exploring this methodology, not yet used for this purpose, this study aimed to evaluate the esthetic perception of a gingival smile before and after the application of botulinum toxin type A, using eye-tracking technology, a visual analog scale, and a questionnaire.

Methods
This observational cross-sectional prospective study was approved by the Research Ethics Committee of the University.
The 58 participants were recruited via advertisement at the university. The inclusion criteria were: (1) aged between 18 and 75 years regardless of gender, (2) with no visual impairment or psychologic problems, and (3) lay in dentistry. All participants signed an informed consent form after learning the details and conditions of the study but were blinded to the study's purpose.
The images viewed by the participants were created through two facial photographs of a female volunteer model, 24 years old, with a gingival smile, one before and another 30 days after the botulinum toxin type A application ( Figure 1). The site of injection of the toxin was at the Yonsei point, (Hwang et al., 2009)  Photographs were taken with a Canon XT digital camera (Canon Inc., Tokyo, Japan), 50 mm Sigma macro lens, and a Sigma flash (Sigma Corp., Kawasaki, Japan) at a standard distance of 90 cm, with high-resolution quality, in an illuminated studio and with a black background. Her head was oriented with the Frankfurt plane parallel to the ground and with the median sagittal plane perpendicular to the horizontal plane, smiling, with limited makeup, without earrings and piercings. The photographs were standardized using Photoshop® (Adobe®, San Jose, CA, USA), removing imperfections that could distract the viewer's attention and focus from the objective (such as props and skin blemishes). Additionally, one side was mirrored so that the face was symmetrical.
For the analysis of eye movements, three areas of interest (AOIs) were determined in the images to be viewed (the participants were unaware of these delimitations at the time when the information was collected), to define regions on the face, such as the mouth, nose, and eyes ( Figure 2).

Source: Authors.
To start the eye-tracking test, the participants were invited to sit on a chair in a quiet room, 60 cm away from the high resolution (768 × 1366 pixels) monitor (Dell P2317H; Dell Inc., Round Rock, TX, USA) positioned vertically and from the  hardware eye-tracker. The calibration of the eye movement was performed, and only participants with software-designated admissible results were considered.
The hardware used for eye-tracking was TheEyeTribe® (The Eye Tribe Aps, Copenhagen, Denmark), in conjunction with OGAMA 5.0 software (Freie Universität, Berlin, Germany) to obtain the eye-tracking data collection. Based on the results, the number of fixations, the time until the first fixation, and the total fixation time on each AOI were analyzed, and the heat maps and map of visualization points were generated through the software. In the heat map, the averages between the visualizations were compared, progressing from the cold color (green) to the warm colors (red). The more reddish the color, the more visualizations a given area received. In the map of the visualization point, a dot was inserted into the image each time a visualization was detected.
The observers visualized a random sequence of 21 images, including the two images of the current study's volunteer model. One before and the other after the toxin application. The 19 extra images that were visualized were not included in this study. These consisted of facial images of children, youth, adults, and the elderly, all smiling, both male and female. Additional images were used to avoid comparisons between the images taken before and after the toxin application due to the proximity of the display. Each image was displayed for 5 seconds. To avoid interference from the final connection point to the previous slide, the first connection point on the next slide was programmed to be ignored, at a time interval of 200 ms, in the OGAMA 5.0.
A Visual Analogue Scale was used to analyze the attractiveness of facial esthetics after the eye-tracking session. The scale was presented on the screen, where a score should be established with a gradation of 0 to 100. A value closer to 0 indicates that the evaluator considered the image to be less attractive, and closer to 100, more attractive (Howells & Shaw, 1985). The twenty-one images used previously were also used to assess the VAS.
In the last stage of the research, the observers complete a survey on a Dell Inspiron 7375 touch 2 -1 computer, opened in the Qualtrics® application (Qualtrics Labs Inc., Provo, UT. They could choose to respond on their mobile device, with the premise that the responses would be sent in the presence of the person responsible for the research. The questionnaire contained the following questions: "Do you consider this person to be friendly?," "Do you think that person is satisfied with their smile?," "Do you consider this person dissatisfied with their dental exposure?," "Do you consider this person dissatisfied with their gum exposure?," and "Do you consider this person to be shy?." The questions appeared in this sequence on the computer screen, below the image of the voluntary model, for the images before and after the application of botulinum toxin separately. The answers were arranged in the form of YES, if the person agreed with the question, and NO if they disagreed.

Statistical analysis
The results obtained from eye-tracking, VAS, and the questionnaire was tabulated in Microsoft Excel software and analyzed using SPSS version 25 (Statistical Package for Social Sciences Software, SPSS Inc., Chicago, IL, USA). The Student's t-test for independent samples was applied to analyze significant differences between the images concerning the number of fixations, time until the first fixation, total fixation time, VAS, and differences between male and female participants. The Kruskal-Wallis test was also applied between the variables of gaze tracking comparing the three AOIs. Pearson's Chi-square test was applied between the images in contrast to the variables sympathy, satisfaction with the smile, dissatisfaction with dental exposure, dissatisfaction with gingival exposure, and shyness.

Results
The study was conducted with 58 participants (60.3% men and 39.7% women). The average age was 35.81 years, with 37 years old being the average age for men and 34 years for women. Qualitative results were generated in the form of a heat map and dot map, and quantitative results were generated in the form of statistical analysis.
The results of the heat map before and after the toxin application are shown in Figure 3. For the pre-application image, the greatest focus was observed to be registered on the mouth, around the region of central incisors and gingival exposure, with little or no deviation to other areas. For the post-application image, the greater focus remained on the mouth, in the region of the incisors and lower lip. Furthermore, a greater deviation to the nose and ocular region was noticed, which was corroborated by the dot map analysis. (Figure 4).
There was no statistically significant difference in the AOIs before and after toxin application (Table 1). Concerning the AOI (eyes, mouth, and nose) ( Table 2), it was noted that the region of the mouth took less time to attract attention (p <0.05) when compared to the eyes and nose.  Research, Society and Development, v. 11, n. 6, e56111629518, 2022 (CC BY 4. Figure 5 represents this statistical analysis, allowing verification of the difference between the time until the first fixation ( Fig. 5B) before and after the application of the botulinum toxin.
The VAS scores showed a statistically significant difference between the two research images (p = 0.000) ( Table 3).
The pre-treatment image received lower average scores than the post-treatment image, which was deemed more attractive than the image with gingival exposure.
The five categories: "sympathy," "satisfaction with the smile," "dissatisfaction with dental exposure," "dissatisfaction with gingival exposure," and "shyness" correlated with the images (Table 4). The variables "satisfied with the smile" (p = 0.000), "dissatisfied with the exposure of the gums" (p = 0.000), and "dissatisfied with the exposure of the teeth" (p <0.05) were statistically different. According to the observers, the model would be more dissatisfied with the excessive gum exposure before the application than after, and the latter would be satisfied with her smile and dissatisfied with her dental exposure.
A comparison of the questionnaire items between the sexes is shown in Table 5. There was a statistical difference between the sexes (p <0.05) in the item "sympathy" of the questionnaire for the image after application of the botulinum toxin.
For women, the person in the image with the least gingival exposure appeared to be more "sympathetic". For men, this decreased after the application of botulinum toxin. In the image before the toxin application, the response rate was similar between men and women.   There were no statistically significant differences in the other items of the questionnaire. Both men and women considered that there was an increase in satisfaction with the smile and a decrease in dissatisfaction with gingival exposure in the image after the toxin application.
Both men and women took less time for the first visualization in the mouth, compared with the eyes and nose; in both, the image before and after the toxin application (Table 6).  There was a statistically significant difference (p <0.05) in the time until the first fixation on the nose, in the image without a gingival smile, where men took considerably longer to look at this region when compared to women. The total time of fixation on the mouth by male viewers was greater than in the AOIs "eye" and "nose." For women, there was a balance between the views of the three areas of interest. Both sexes exhibited increased VAS scores in the image with less gingival exposure, considering the image to be more attractive after the toxin application.

Discussion
This study aimed to evaluate the esthetic perception of the face following a reduction of excessive gingival exposure by application of botulinum toxin type A in a female volunteer, using eye-tracking technology, a visual analog scale, and a questionnaire. The intervention in this study sought to obtain an authentic result, which may not be credible with the use of digitally manipulated images. The alteration of normality in the visualizations in this study, with the mouth receiving more attention than the nose and mouth, was similar to other eye-tracking studies where the attention was highlighted (Richards et al., 2015) , (Johnson et al., 2017) in the modified region, as opposed to the studies where the attention pattern is on the eyes in images considered normal (Mertens et al., 1993) , (Hickman et al., 2010).
The gingival smile, considered an unattractive feature, (Celikdelen & Bicakci, 2020;Pithon et al., 2013) may have influenced this greater concentration of visualizations on the region of the mouth in the image displaying greater gingival exposure, before the toxin application, as verified by the heat and dot maps. Although there were differences in the eye region before and after the application of botulinum toxin in the present study, this did not affect the results since this region did not receive as much visualization as the AOIs.  (Gobel, 2015) and Indians, (Bagepally, 2015) where the reading pattern occurred from left to right. In studies carried out in Asian countries, for example in China, where the reading pattern is from right to left, the result may be reversed, with the right side receiving more views. (X. Wang et al., 2016) For this reason, it is advisable to mirror one side of the image, as performed in the present study.
Regarding the benefit of botulinum toxin, the results suggest a statistically significant increase in facial attractiveness after the application of the toxin. These results are consistent with the esthetic improvement of the gingival smile after the toxin application in other studies, which also used the visual analog scale as an evaluation measure. In these surveys, the scales varied from 1 to 5 (Sriphadungporn & Chamnannidiadha, 2017), from 1 to 10 (Cengiz et al., 2020), and, as in the present study, from 1 to 100 (Cengiz et al., 2020;Sucupira & Abramovitz, 2012). It was found that, regardless of the gradation used in the VAS in these analyses, the average of the grades that the evaluators gave the images before the toxin were smaller, increasing in the images after the toxin application. The same result was obtained in this study, with an average visual analog scale score of 48.76 before and 69.57% after the toxin application, and an 42.67% increase in facial attractiveness.
In the present study, there was no statistically significant difference in perception between men and women when evaluating the face as a whole, both in the tracking of the visualization and in the evaluation of the VAS. The same result was found in other studies that used the same tools (Chang et al., 2011;Ker et al., 1999). Conversely, one study found that the female sex evaluates the gingival smile with higher marks than does the male sex (Geron & Atalia, 2005). Forty-six participants used only images of the lower third of the model, where a complete facial evaluation was not possible. By using the image of the entire face of the voluntary model, we allowed the gingival smile to be analyzed not as an isolated component, but in conjunction with other characteristics for a more complete assessment of its influence on facial attractiveness.
The results suggest that the possibility of correcting the gingival smile proposed in this study can be presented to patients as an alternative treatment since laypeople considered the face of the voluntary model more attractive after decreasing excessive gingival exposure. Although botulinum toxin has a short duration of action (from 12 to 24 weeks) (Chagas et al., 2018), this minimally invasive treatment (Polo, 2008) may be preferred by patients, especially considering its reversibility. This is because there is a physiological decrease in the exposure of the upper incisors and an increase in exposure of the lower incisors with aging, compared to other existing irreversible treatments, such as orthodontic treatment, orthognathic surgery, and lip repositioning (Abdullah et al., 2014). Therefore, it can be used as an adjunct treatment in orthodontics and periodontics. Also, the perception of esthetic attractiveness after improving teeth morphology was higher compared to before (Martins et al., 2021).

Conclusions
There was a variation in the perception of facial esthetics after the application of botulinum toxin. A decrease in gingival exposure resulted in improved attractiveness. There was also an increase in "satisfaction with the smile" and "dissatisfaction with dental exposure," as well as a decrease in "dissatisfaction with gingival exposure." This study has clinical implications and is relevant because showed a treatment alternative for the patient with dental exposition. As a future perspective, it is suggested that should be carried in elder patients.