Aesthetic perception of gingival display on smiling among laypeople seeking dental treatment

A close-up, smiling colour photograph of an 18-year-old female was selected as the subject. The dentitions were well-aligned, without crowding, rotations, nor spacing. Adobe^© Photoshop^© version 20.0.0 was used to digitally manipulate the gingival show to create displays ranging from +4 mm to -4 mm, each with a 2 mm incremental change and produce a total of 5 photographs (Figures 1–5). One of the photographs was duplicated to allow for Intra-class Coefficient measurements. The photographs were printed in A4 size, laminated, and mounted, in no particular sequence, in a booklet. Between each photograph, a laminated A4 sized picture of a waterfall was inserted. Photographs of a waterfall were shown to the subjects in between each smile photograph to reduce visual fatigue.

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

A 2-part questionnaire was constructed in English, of which the first part consisted of the subject’s demographic data and the second part consisted of a 10-point Likert scale to assist in rating the attractiveness of each photograph. The Likert scale used a numerical rating from one to ten, with the ”least attractive” and “most attractive” labelled on the left end and the right end, respectively. A Malay version was created by an expert panel using a forward and backward cross-cultural translation method from the English questionnaire.

Content validity was carried out as outlined by Rodrigues et al.²⁰ The questionnaire and photographs were validated by an expert panel of five orthodontists each with more than ten years of experience. The instructions, the overall suitability of photographs to test objectives, and different gingival exposure levels and lip positions were assessed based on their relevance, clarity, and necessity. All items were deemed necessary by the panel with a Content Validity Ratio (CVR) of 1.00. All items were also deemed relevant and clear as each item scored excellent on the Item-level Content Validity Index (I-CVI).

Face validity is a test conducted on simulators to determine if a questionnaire subjectively appears to measure the variable or construct that it is intended to measure. It was conducted by recruiting subjects to answer the questionnaire which was followed by a semi-structured interview. Based on the pilot results, no adjustment of the questionnaire was necessary. Subsequently, test-retest reliability was conducted by having the 10 subjects answer the same questionnaire again after 2-4 weeks. The intraclass correlation coefficient (ICC) for the test-retest reliability was 0.71 suggesting moderate reliability.²¹ The ICC of the duplicated photos was 0.70 for the 10 subjects. After validation and reliability testing, the questionnaires with the three photograph booklets were accordingly adjusted and printed to distribute to each centre.

Based on the inclusion and exclusion criteria, the participants were conveniently selected from patients who sought dental treatment in the primary dental care units of UKM KL Campus, UM, and UiTM. All participants were given a study information sheet and were asked to sign an informed consent letter.

The participants were offered the questionnaire in either English or the Malay language and requested to fill in their demographic information.

Subsequently, the image booklet was provided and each subject asked to appraise each individual smile, for up to 1 minute, and then rate the smile by circling the most appropriate number on the Likert scale. A comparison between photographs was not allowed.¹⁴ The participants received little information about the study design and were asked to judge the photographs based purely on their perception of attractiveness.

The subjects recruited were older than 18 years of age. To be categorised under either Malay, Chinese or Indian, each subject must have descended from at least two generations of either Malay, Chinese or Indian pure parentage. If these two conditions were not met, for example, the subject was of another ethnicity other than the three main ethnicities in Malaysia, or if the subject did not have at least two generations of pure parentage of the three main ethnicities, the subject was classified into an “Others” category.

Non-Malaysian citizens, dental students, dental professionals, and patients younger than 18 years of age were excluded. The study design targeted Malaysian laypeople, and deviations from these criteria would likely yield data that would not accurately represent the target group. Young subjects (those less than 18 years of age) were also excluded due to their inability to consent for their own treatment, as the results of the study were to aid in treatment planning. In addition, prostheses that involve gingival and smile aesthetics e.g., implants and crowns, are usually recommended after the gingiva has fully matured, which is usually after 18 years old.

The data were reported descriptively with tables as a median of the scores. The scores for different levels of gingival display were found to be not normally distributed using the Kolmogorov–Smirnov test. Therefore, non-parametric tests were applied. The Kruskal–Wallis test was used for intergroup comparisons between genders, the age groups and ethnicities. The Friedman test was used for overall and intragroup comparisons between the levels of gingival display. All tests were performed at a 0.05 significance level. To assess intra-rater agreement, the scores for the duplicated photographs were analysed using the ICC. All statistical tests were conducted using IBM SPSS© version 26.

There was a statistically significant difference in smile perception between male and female participants. Female participants regarded -2 mm, 0 mm, +2 mm, and +4 mm as most aesthetic, and -4 mm as the least aesthetic display. Females therefore found all levels of gingival display aesthetic but extreme coverage of the teeth intolerable. Males, however, preferred gingival displays of -2 mm and 2 mm over gingival displays of -4 mm, 0 mm, and +4 mm, demonstrating that they do not have a consistent preference pattern. Similar findings were demonstrated by Bolas–Colvee et al.,²⁸ in which males were found to be less critical in their perception of gingival display, although alternative studies^1,29,30 reported no gender difference. These findings might be related to ethnic, cultural, or social differences in smile preference. That the male participants had an inconsistent pattern may also be attributed to the lower intra-rater agreement measured using the ICC (0.643) compared to female participants (0.771). As previous literature did not report the ICC values of the genders, no comparison could be made.

The present study also indicated that the age of a person influenced smile perception. The youngest age group regarded smiles with the smile line closer to the gingival zenith (-2 mm, 0 mm, +2 mm) to be the most aesthetic, whereas smiles with gingival underexposure (-4 mm) and overexposure (+4 mm) were regarded as the least aesthetic. Accordingly, Pausch and Katsoulis found that subjects aged between 22 and 35 years old were less tolerant of smiles with over- or under-exposed gingiva.¹⁰ Similarly, Sriphadungporn and Chamnannidiadha reported that subjects aged between 15 and 29 years of age rated smiles with -4 mm and +6 mm of gingival display to be the least attractive.⁸ It was also suggested that excessive lip coverage was unfavourable among young subjects as it is a feature associated with ageing.

The middle age group preferred smiles with -2 mm gingival display but disliked smiles with -4 mm gingival display. This is in agreement with Sriphadungporn and Chamnannidiadha who revealed that subjects aged 36 to 50 years old perceived smiles with -2 mm smile line to be attractive.⁸ However, the study reported smiles showing a +6 mm smile line to be the least attractive among the middle age group.

The oldest age group deemed the smile with +2 mm gingival display to be the most attractive, and the smile with -4 mm, 0 mm and +4 mm gingival display as the least attractive smiles. This finding is in contrast to Sriphadungporn and Chamnannidiadha,⁸ in which it was found that the older age groups were more accepting of incisal coverage, as it was likely to occur at their age. Interestingly, the oldest age group in the present study had the lowest ICC of 0.576 of all the age groups, reflecting that the elderly might find it difficult to differentiate between smiles of different gingival exposures. There were also no significant differences between the different gingival exposures in this age group. This could explain the inconsistency of the findings between the present study and the existing literature. Nevertheless, the difficulty in differentiating smiles with varying gingival displays among the elderly could be attributed to their increased satisfaction with their dental appearance compared to younger people, as demonstrated by Alkhatib, Holt and Bedi.³¹ This unique finding should be considered during orthodontic treatment of the elderly, when more attention could be directed to other more important parts of the elderly smile.

The present study also showed that smile perception was influenced by ethnicity. This is in concordance with Nimbalkar et al., who demonstrated that Malaysia’s three main ethnic groups (Malay, Chinese and Indian) perceived smiles differently.³²

Malay participants regarded a gingival exposure of -2 mm as the most aesthetic and -4 mm as the least aesthetic. These findings were consistent with a study conducted of Indonesian laypersons.¹⁶ An additional study conducted among dental students in Indonesia found that the most aesthetic smile was when the smile line covered the gingival zenith and only allowed visibility of the gingival embrasure.³³ However, participants in the study rated the smile with excessive gingival exposure as the least aesthetic and the smile line covering the gingival zenith and embrasure as the second least aesthetic. A direct comparison between the past study and the present could not be made as no measurement of the gingival display was conducted in millimeters.

Chinese subjects were the most tolerant towards smiles displaying all types of gingival exposure as they rated all smiles (-4 mm, -2 mm, 0 mm, +2 mm, +4 mm) the same score in the present study. A study conducted in China indicated that Chinese laypeople preferred smiles with the smile line revealing less than 75% to 100% of the maxillary anterior teeth and interproximal gingiva, over smiles with the gingival display revealing 100% of the maxillary anterior teeth and a contiguous band of gingiva.¹⁷ In the present study, subjects from the Chinese ethnic group had no intra-rater reliability as the value of the reported ICC was 0, suggesting that Chinese subjects were not able to differentiate between the various gingival displays.

In the present study, no significant differences were found between gingival exposures within the Indian ethnic group. A gingival display of +2 mm was regarded as the most aesthetic and smiles with a gingival display of -4 mm, -2 mm and +4 mm to be the least aesthetic. Studies conducted in India reported the most aesthetic gingival display to be -2 mm¹⁸ and 0 mm.¹⁵

Subjects who were outside of the three main ethnic groups in Malaysia (Malay, Chinese and Indian) were classified into the “Others” category which consisted of a heterogeneous group of indigenous tribes in Malaysia, for example, Dayak, Iban, Kadazan and Orang Asli. People from the “Others” group favoured all smiles except gingival overexposure (+4 mm). The present study was the first to investigate smile perception in the minority ethnic group in Malaysia.

Overall, when comparing the different ethnic groups, Malays seem to be the most specific when evaluating smiles, as they only had a single most and least aesthetic gingival display value. Additionally, their ICC was the highest of the four groups, at 0.881, which suggested that they were critical in differentiating smiles of various gingival displays. However, Chinese subjects were accepting of all the smiles, as the difference in various levels of gingival display may be largely imperceptible to them. Subjects from the “Others” group disliked gummy smiles.

The discrepancies between the findings of the current study and past literature about the perception of smile aesthetics between different ethnic groups could be attributed to the cultural differences, varying lifestyles, and environmental factors between Malaysia and other countries. Furthermore, due to Malaysia’s multi-ethnic nature, smile perception differences exist between the different ethnicities within Malaysia and must be considered during treatment planning.

As subjects in the present study were recruited from patients attending primary care units in the three institutions, these subjects may not represent the whole Malaysian population. However, data can be assumed as homogenous as they were sampled from an area of mixed population. The sample size calculation was carried out to achieve the minimum sample size except for the different ethnicities. Therefore, the findings for each ethnic group should be interpreted with caution. Future studies are required with a larger sample size to better determine the effect of ethnicity on the perception of smile aesthetics in the Malaysian population.

An additional limitation of this study was that the assessment of smile perception was only on a female smile and not a male smile. As smiles of different gender may be evaluated differently,¹⁴ future studies on the smile perception in both male and female subjects appear to be warranted. The present study investigated the effect of varying gingival display only on a relaxed smile. Since the shape of the lips and visibility of teeth might affect smile perception, the authors recommend future studies to assess the effect of a broad, full smile compared to a relaxed smile.