The appearance of a smile has direct implications for esthetics and facial attractiveness. Among the parts of the face, the eyes play the most important role in esthetics, followed by the mouth—especially when smiling.

Orthodontic literature has given greater emphasis to studies about the dentoskeletal relationship, in spite of the appearance of the smile’s having gained much clinical importance, because frequently it is from this point of view of self-perception that patients also seek to evaluate their treatment. Therefore, the orthodontist’s perception of beauty might differ from the concepts of laypersons about what is attractive and satisfactory.

Quite often, the consequence of orthodontic treatment is the creation of black spaces in the region of the crenations of the maxillary central incisors. These can be the result of an accentuated inclination of the incisors in the mesial direction, bone loss during treatment, or the triangular shape of the maxillary central incisors. These black spaces might cause esthetic alterations, depending on the patient’s self-evaluation.

Pertinent questions arise when black spaces occur at the end of orthodontic treatment. Will these spaces affect the esthetics of the smile? Does age affect the perception of these black spaces? Because the prevalence of black spaces increases with aging as a result of periodontal disease and aging of the periodontium, one could deduce that the older the person, the more tolerant he or she would be of these black spaces.

Based on this premise, we proposed in this study to evaluate the degrees of perception of laypersons in various age groups about the influence of black spaces between the maxillary central incisors on the esthetics of the smile. Subjects with age ranges of 15 to 19, 35 to 44, and 65 to 74 years were used, as recommended by the World Health Organization and International Dental Federation.

Material and methods

An extraoral photograph of a 30-year-old patient, in normal occlusion, was obtained with a digital camera (Rebel XTI; Canon, Tokyo, Japan). The photograph was cropped to show the lips, teeth, and gingiva to focus attention on the smile. The photograph was manipulated with the Photoshop (CS3; Adobe Systems, San Jose, Calif) computer program to make black spaces of different sizes in the region of the crenations between the maxillary central incisors; the mandibular arch was left intact. This resulted in 8 photographs: 1 had no black spaces, and the others had increasing sizes of black spaces (0.5, 1, 1.5, 2, 2.5, 3, and 3.5 mm). No image showed a deviation of the midline ( Fig ).

Images showing black spaces between the incisors. Image 1 ( left ): A, without space; B, space of 0.5 mm; C, space of 1.0 mm; D, space of 1.5 mm; E, space of 2.0 mm; F, space of 2.5 mm; G, space of 3.0 mm; H, space of 3.5 mm. Image 2 ( right ): A, space of 3.5 mm; B, space of 3.0 mm; C, space of 2.5 mm; D, space of 2.0 mm; E, space of 1.5 mm; F, space of 1.0 mm; G, space of 0.5 mm; H, without space.

The photographs were evaluated by 150 laypersons in 3 age groups: 15 to 19, 35 to 44, and 65 to 74 years. None had received orthodontic treatment. They were asked to complete questionnaires about the images. First, they analyzed a series of miniatures (10 × 6 cm) and chose the most and the least pleasing images. Then a sequence of the images was made, starting with the most pleasing, going to the least. The process was repeated to evaluate the reliability of the evaluation. After this, the evaluators analyzed larger (20 × 12 cm) versions of each image and rated them according to a visual analog scale: 0, not very attractive; 5, attractive; and 10, very attractive. The evaluators were not allowed to make comparisons among the images. All images were placed at a distance of 30 cm from the evaluators’ eyes. One factor for exclusion of an evaluator was lack of visual acuity.

The frequencies of replies by the participants in each age group (15-19, 35-44, and 65-74 years) were compared with the chi-square test. When the expected frequency was less than 5, the Fisher exact test was used. The scores awarded to each photograph were compared with the Kruskal-Wallis test, and comparisons between pairs were performed with the Mann-Whitney test. The means of scores awarded to each photograph were calculated in each group to determine the Spearman correlation coefficients, to evaluate the similarity between perceptions according to the age groups. The level of significance adopted was 5% ( P <0.05). The data were tabulated and analyzed in the statistical program BioEstat (version 5.0; Bioestat, Belém, Pará, Brazil).