https://orcid.org/0000-0002-9050-096X
Abstract
Maxillary lateral incisor agenesis is often managed with orthodontic space closure and canine substitution. Anatomic gingival margin heights associated with the maxillary anterior teeth are considered important for achieving aesthetic excellence with space closure, but evidence relating to the perceptions of lay people is poor.
This study investigated the influence of gingival margin height variation in the maxillary anterior teeth following orthodontic space closure and canine substitution in the absence of maxillary lateral incisor teeth on perceived aesthetics judged by a lay population.
Images of a case treated with space closure and bilateral canine substitution were digitally modified to create gingival margin height variation for the substituted lateral incisors and canines. Six variations were created, including one conforming to accepted norms for anatomic gingival margin heights. Lay people represented by parents of orthodontic patients ranked the images based on aesthetics, with data analysed statistically using linear models at 5%.
One hundred and twenty responders were included (median age 48.0 years), 50% male and 81.7% Caucasian, with 87% able to provide specific preferences. In absolute terms, the highest-ranking [image C] had substituted lateral incisor gingival margins below the central incisors (considered normal) but substituted canine margins symmetrically below the substituted lateral incisors (considered to be too low); followed by [image D] with symmetrically level gingival margins; and [image F] with asymmetric substituted lateral incisor margins and substituted canine margins symmetrically too low. The lowest-ranked [image E] had normal substituted lateral incisor margins but asymmetric substituted canine margins. Image C was most often ranked first (29.8% of responders) and image E last (22.1%). After adjusting for potential confounding effects, image E received the worse scores.
Anatomic norms for gingival margin height in absent maxillary lateral incisor-canine substitution do not correlate with the highest-ranking aesthetic choices of a lay population. Lateral incisor gingival margins symmetrically below the central incisors and substituted canine margins symmetrically below these ranked highest. Vertical asymmetry in the substituted canine position was considered the least aesthetic.
Introduction
Tooth agenesis is a relatively common condition in all populations and among those teeth that are most often absent, the maxillary lateral incisors feature prominently [1]. The management of maxillary lateral incisor agenesis often requires orthodontic treatment with fixed appliances and is associated with a relatively simple treatment algorithm: orthodontic space opening (or preservation) and prosthetic replacement of the absent maxillary lateral incisors; or orthodontic space closure and substitution of the absent maxillary lateral incisors with the maxillary canines (and the maxillary canines with the first premolars). The relative advantages and disadvantages of these approaches have been debated among orthodontists and restorative dentists for many years [2, 3]. In essence, space opening allows the replacement of all six maxillary anterior teeth but relies upon a fixed prosthesis for life, usually a single unit adhesive bridge or implant. Space closure negates the need for a lifetime restoration but can rely upon significant modification of the canine and first premolar teeth to achieve aesthetic excellence and permanent retention to prevent space opening is often essential [4–6]. Indeed, patients with absent maxillary lateral incisors often have microdontia of other teeth including the central incisors and first premolars, which can further increase the restorative burden in terms of obtaining aesthetic excellence in cases treated with space opening or space closure [7]. The overall evidence base relating to space opening and space closure remains at a low level, although the latter approach has been associated with better long-term periodontal outcomes [8–11]. More recently, retrospective observational data have suggested that in general, space closure is preferable to space opening and implant replacement in terms of clinical and aesthetic outcomes and total societal cost of treatment [12–14] although aesthetic improvements have been noted over time for both approaches [15].
When planning orthodontic space closure, multiple techniques have been described that can help to optimize aesthetic outcomes in terms of tooth position. In essence, the maxillary canine crown is often larger, more bulbous, and more pointed than the lateral incisor it will replace, and the root eminence is more prominent. These factors can be controlled to a certain extent by careful bracket selection and positioning [16], reshaping of the canine crown during orthodontic treatment [3–6] and if necessary, composite build-up or veneers at the end of treatment [17]. Indeed, larger, more pointed, mesial-tipped and darker canines are regarded as less attractive in simulated canine substitution [18]. In contrast, the maxillary first premolar crown is often smaller and narrower than the canine it will replace, lacks a prominent buccal root eminence and has an additional palatal cusp. Again, the orthodontist plays an important role in improving coronal aesthetics of the first premolar, including mesial rotation of the crown to increase width and the application of buccal root torque to increase the root eminence (and hide the palatal cusp). However, for both the canine and first premolar, the associated gingival aesthetics can be more challenging to manage, particularly when attempting to maximize coronal aesthetics.
A wide range of detailed anatomical parameters have been described in association with optimal gingival aesthetics for the maxillary anterior teeth [19, 20]. The most apical point of the free gingival margin is represented by the gingival zenith, which for the maxillary central incisor and canines should be vertically level with each other in relation to a horizontal gingival aesthetic line constructed perpendicular to the vertical midline, while the lateral incisor zenith should be 1–2 mm below this level [21]. A fundamental issue when managing gingival margin aesthetics in space closure cases with absent maxillary lateral incisors is that the pre-treatment canine gingival margin can be significantly higher than that of the central incisor, while the first premolar gingival margin is frequently lower than both. The orthodontist will often need to extrude the maxillary canine to facilitate coronal reduction, which will help to lower the gingival margin below that of the central incisor to give the most pleasing aesthetics [4, 18]; however, for the maxillary first premolar, achieving a gingival margin situated level with that of the central incisor will often require significant intrusion of this tooth, which can directly increase the restorative burden for these teeth through the necessity to increase crown height [4–6]. Indeed, with significant intrusion composite build-ups or veneers are often required to achieve the required coronal length and adequately compensate for the intrusion that is needed to correct the gingival margin and some burden of care is associated with this approach over the longer term [22].
The aim of this study was to investigate the influence of gingival margin height variation on smile aesthetics in the presence of maxillary lateral incisor-canine substitution using a lay population of observers. The objective was to further understand how a lay population perceives variation in gingival margin aesthetics to further inform treatment-planning decisions in cases of maxillary incisor agenesis treated with space closure and canine substitution.
Material and methods
Ethical approval for this observational study was granted by the research ethics review board at King’s College London LRU/DP-21/22-30017.
Images of a case with absent maxillary lateral incisor teeth previously treated with fixed appliances and space closure with canine substitution were manipulated in Adobe Photoshop 22.0 (Adobe Inc., California, USA) using the rectangular marquee and clone stamp tools, montaged onto a single A4 page and used for evaluation within a questionnaire provided to a lay population of individuals (Fig. 1A–F). For these manipulations, the region of interproximal contact and gingival margins of the maxillary central incisors were used as the midline vertical and horizontal reference lines, respectively, and the following six gingival margin contour variations were manipulated in relation to these references (Fig. 2A–F): (A) anatomically normal symmetrical gingival margin aesthetics: with the substituted lateral incisor (canine) gingival margins below the central incisor gingival margin line and the substituted canine (first premolar) and central incisor gingival margins level; (B) substituted lateral incisor gingival margins symmetrically above the central incisors [not ideal] and the substituted canines symmetrically below the central incisors [not ideal]; (C) substituted lateral incisor gingival margins symmetrically below the central incisors [ideal] and substituted canines symmetrically below the central incisors [not ideal]; (D) all gingival margins symmetrically level with the maxillary central incisors [not ideal for the substituted lateral incisors but ideal for the substituted canines]; (E) ideal gingival margin aesthetics represented on the patient’s left (substituted canine and central incisor margins are level and the substituted lateral incisor below this line); on the patient’s right, the substituted lateral incisor gingival margin is below the central incisor [ideal] while the substituted canine is below the central incisor [not ideal]; and (F) on the patient’s left, the substituted lateral incisor gingival margin is above the central incisor [not ideal] and the substituted canine is below this line [not ideal]; on the patient’s right, the substituted lateral incisor gingival margin level is below the central incisor [ideal] and the substituted canine is below the central incisor [not ideal]. There was no variation in any other parameter between images, including gingival shape, coronal morphology, or overall colour of the gingival or coronal tooth tissues.
Images used in the investigation derived from a treated case with maxillary lateral incisor agenesis managed using orthodontic space closure and canine substitution. Images were manipulated using Adobe Photoshop to show six (A–F) different combinations of gingival margin levels associated with the upper six maxillary anterior teeth. All manipulations were made using the maxillary central incisor interproximal contact and gingival margins as the midline (hatched black line) and vertical references (solid black line), respectively. The canines are substituting for the lateral incisors and the first premolars are substituting for the canines. The canine tips underwent enameloplasty during orthodontic treatment but no restorative work was carried out on any teeth.
Images from Figure 1 with horizontal black lines placed to illustrate the vertical height variation in manipulated gingival margin levels. For a detailed explanation see the Materials and Methods. Briefly, (A) represents normal gingival margin anatomy; (B) the substituted lateral incisor gingival margins are too high and the substituted canines are too low; (C) the substituted lateral incisor gingival margins are ideal and the substituted canines are too low; (D) all gingival margins are level (the substituted lateral incisors are therefore too high but the substituted canines are ideal); (E) gingival aesthetics are ideal on the patient’s left, but on the right, the substituted lateral incisor is ideal, while the substituted canine is too low; (F) on the patient’s left, the substituted lateral incisor is too high and the substituted canine is too low, but on the right, the substituted lateral incisor is ideal while the substituted canine is too low.
The parents of patients undergoing orthodontic treatment with fixed appliances in university (Department of Orthodontics, Faculty of Dentistry, Oral & Craniofacial Sciences, King’s College London), secondary care (Royal Alexander Children’s Hospital, Brighton) and primary care (Petts Wood Orthodontics, Bromley) locations were included in the study. Exclusion criteria included any parents who were dentists or employed in clinical dentistry; those currently undertaking orthodontic treatment themselves; and those with children who had previously undertaken a course of orthodontic treatment, or currently being treated for absent teeth, trauma to the maxillary incisor teeth, or surgical correction of jaw discrepancies. Each participant was asked to evaluate the six images (A–F) and rank them in order of attractiveness (1 = most attractive to 6 = least attractive). They were asked to describe in a free text space their reason(s) for assigning number 1 (most attractive) and number 6 (least attractive). If they felt unable to rank the images, they were able to indicate this in the questionnaire (and therefore not rank them). According to participant rankings, n = 6 points (for the number 1 ranking) to n = 1 point (for the number 6 ranking) was allocated to each image. Respondents were also asked to report their gender, age, and ethnicity, and construct a numerical de-identification code based upon their birthday (day and month), the last three letters of their street address and the last three digits of their phone number. Each questionnaire contained an initial question asking whether the parent wished to participate or not; if the parent elected not to participate, the incomplete questionnaire could be posted into the sealed collection box in a de-identified manner. In addition, if any parent retrospectively decided they did not wish to participate, the de-identification code allowed their data to be withdrawn from the research at any time-point [23]. To mitigate any potential pressure to participate, parents were approached about participation in the waiting room (away from the treating clinician) prior to a routine appointment by an orthodontic nurse not directly involved in their child’s orthodontic care. The questionnaire was completed in the waiting room and then posted into a sealed box in reception.
Statistical analysis
Sample size calculation was based on a separate small pilot of the questionnaire (n = 10 participants who ranked the images). The average ranking score (according to the score of 6 points for #1 ranking to 1 point for #6) for the assessed images was 3.50 points with a standard deviation (SD) of 1.71. Setting α at 5% and power at 80%, a sample of n = 94 participants would be needed to identify a minimal difference of 0.5 points considered to be clinically significant, which was rounded up to n = 100 to account for data losses from non-responders. Participants were approached consecutively during routine appointments conducted over a 4-week time period across sites. Analysis of these initial data revealed that n = 16 participants were unable to give specific preferences in terms of ranking the images, leaving a sample size of n = 84. To achieve power, the sample was therefore extended over an additional 1-week time period across sites with the collection of a further (n = 20) questionnaires, which were all completed with specific preferences to give an overall sample size of n = 104 completed questionnaires.
Descriptive statistics were calculated, including median and interquartile range for continuous variables (after checking for normality visually and with the Shapiro-Wilk test) and absolute/relative frequencies for categorical variables. Ethnicity was categorized as Caucasian or non-Caucasian for the analyses. Differences in characteristics between participants able and not able to rank the images were assessed with Mann–Whitney or chi-squared tests. The average ranking score (and 95% Confidence Interval [CI]) was calculated for each image with generalized linear models adjusting for age, sex, ethnicity, and considering clustering within each participant. All analyses were run at an α of 5% in Stata (version 14.0; Stata Corp, College Station, Tex) with an openly provided dataset [doi: https://doi.org/10.5281/zenodo.13827785].
Results
A total of n = 120 participants were included in this study (Table 1) with a median age of 48.0 years, an equal distribution of males to females and a predilection for Caucasians (82%). No individuals refused to participate, and none requested to be withdrawn from the study. All questionnaires were completed correctly and not spoilt.
Characteristics of participants.
| Variable | Metric | Value |
|---|---|---|
| Age (n = 120) | Median (IQR) | 48.0 (40.5, 51.0) |
| Range | 27.0, 60.0 | |
| Sex (n = 120) | Male—n (%) | 60 (50.0%) |
| Female—n (%) | 60 (50.0%) | |
| Ethnicity (n = 120) | Caucasian—n (%) | 98 (81.7%) |
| Non-Caucasian—n (%) | 22 (18.3%) |
| Variable | Metric | Value |
|---|---|---|
| Age (n = 120) | Median (IQR) | 48.0 (40.5, 51.0) |
| Range | 27.0, 60.0 | |
| Sex (n = 120) | Male—n (%) | 60 (50.0%) |
| Female—n (%) | 60 (50.0%) | |
| Ethnicity (n = 120) | Caucasian—n (%) | 98 (81.7%) |
| Non-Caucasian—n (%) | 22 (18.3%) |
IQR = interquartile range.
Characteristics of participants.
| Variable | Metric | Value |
|---|---|---|
| Age (n = 120) | Median (IQR) | 48.0 (40.5, 51.0) |
| Range | 27.0, 60.0 | |
| Sex (n = 120) | Male—n (%) | 60 (50.0%) |
| Female—n (%) | 60 (50.0%) | |
| Ethnicity (n = 120) | Caucasian—n (%) | 98 (81.7%) |
| Non-Caucasian—n (%) | 22 (18.3%) |
| Variable | Metric | Value |
|---|---|---|
| Age (n = 120) | Median (IQR) | 48.0 (40.5, 51.0) |
| Range | 27.0, 60.0 | |
| Sex (n = 120) | Male—n (%) | 60 (50.0%) |
| Female—n (%) | 60 (50.0%) | |
| Ethnicity (n = 120) | Caucasian—n (%) | 98 (81.7%) |
| Non-Caucasian—n (%) | 22 (18.3%) |
IQR = interquartile range.
Among the participants, n = 104 out of 120 (87%) were able to give specific preferences in terms of ranking the images, while n = 16 (13%) could not. There were no systematic differences between these two groups of participants, in terms of age, sex, or ethnicity (P > .05 in all instances; Table 2), indicating a lack of non-response bias.
Participants able or unable to give a preference between the images and the demographic differences between these groups.
| Variable | Metric | Value | *Age (Mann–Whitney) | *Sex (χ2) | *Ethnicity (χ2) |
|---|---|---|---|---|---|
| Gave a preference (n = 120) | Yes—n (%) | 104 (86.7%) | 0.94 | 0.59 | 0.14 |
| No—n (%) | 16 (13.3%) |
| Variable | Metric | Value | *Age (Mann–Whitney) | *Sex (χ2) | *Ethnicity (χ2) |
|---|---|---|---|---|---|
| Gave a preference (n = 120) | Yes—n (%) | 104 (86.7%) | 0.94 | 0.59 | 0.14 |
| No—n (%) | 16 (13.3%) |
Participants able or unable to give a preference between the images and the demographic differences between these groups.
| Variable | Metric | Value | *Age (Mann–Whitney) | *Sex (χ2) | *Ethnicity (χ2) |
|---|---|---|---|---|---|
| Gave a preference (n = 120) | Yes—n (%) | 104 (86.7%) | 0.94 | 0.59 | 0.14 |
| No—n (%) | 16 (13.3%) |
| Variable | Metric | Value | *Age (Mann–Whitney) | *Sex (χ2) | *Ethnicity (χ2) |
|---|---|---|---|---|---|
| Gave a preference (n = 120) | Yes—n (%) | 104 (86.7%) | 0.94 | 0.59 | 0.14 |
| No—n (%) | 16 (13.3%) |
According to the participant responses (Table 3), the number 1 (#1) ranking was most often allocated to image C (29.8%), followed by image B (20.2%), image D (18.3%) and images F, A, and E (15.8, 8.7, and 4.8%, respectively). Image A (with normal gingival margin height anatomy) was most often ranked number 2 (#2) (22.1%), while image F was most often allocated to rank number 3 (#3) and #5 (23.8 and 20.8%, respectively) and image E to rank number 6 (#6) (22.1%).
Rank given to each image by the (n = 104) participants who were able to rank them.
| A | B | C | D | E | F |
|---|---|---|---|---|---|
| #2 (22.1%) | #1 (20.2%) | #1 (29.8%) | #2 (20.2%) | #6 (22.1%) | #3 (23.8%) |
| #6 (20.2%) | #5 (18.3%) | #4 (20.2%) | #3 (19.2%) | #4 (20.2%) | #5 (20.8%) |
| #4 / #5 (17.3%) | #4 / #6 (17.3%) | #5 (13.5%) | #1 (18.3%) | #3 / #5 (19.2%) | #2 (16.8%) |
| #3 (14.4%) | #2 (14.4%) | #2 / #6 (12.5%) | #4 (16.4%) | #2 (14.4%) | #1 (15.8%) |
| #1 (8.7%) | #3 (12.5%) | #3 (11.5%) | #6 (14.4%) | #1 (4.8%) | #6 (13.9%) |
| #5 (11.5%) | #4 (8.9%) |
| A | B | C | D | E | F |
|---|---|---|---|---|---|
| #2 (22.1%) | #1 (20.2%) | #1 (29.8%) | #2 (20.2%) | #6 (22.1%) | #3 (23.8%) |
| #6 (20.2%) | #5 (18.3%) | #4 (20.2%) | #3 (19.2%) | #4 (20.2%) | #5 (20.8%) |
| #4 / #5 (17.3%) | #4 / #6 (17.3%) | #5 (13.5%) | #1 (18.3%) | #3 / #5 (19.2%) | #2 (16.8%) |
| #3 (14.4%) | #2 (14.4%) | #2 / #6 (12.5%) | #4 (16.4%) | #2 (14.4%) | #1 (15.8%) |
| #1 (8.7%) | #3 (12.5%) | #3 (11.5%) | #6 (14.4%) | #1 (4.8%) | #6 (13.9%) |
| #5 (11.5%) | #4 (8.9%) |
# = ranking number.
Rank given to each image by the (n = 104) participants who were able to rank them.
| A | B | C | D | E | F |
|---|---|---|---|---|---|
| #2 (22.1%) | #1 (20.2%) | #1 (29.8%) | #2 (20.2%) | #6 (22.1%) | #3 (23.8%) |
| #6 (20.2%) | #5 (18.3%) | #4 (20.2%) | #3 (19.2%) | #4 (20.2%) | #5 (20.8%) |
| #4 / #5 (17.3%) | #4 / #6 (17.3%) | #5 (13.5%) | #1 (18.3%) | #3 / #5 (19.2%) | #2 (16.8%) |
| #3 (14.4%) | #2 (14.4%) | #2 / #6 (12.5%) | #4 (16.4%) | #2 (14.4%) | #1 (15.8%) |
| #1 (8.7%) | #3 (12.5%) | #3 (11.5%) | #6 (14.4%) | #1 (4.8%) | #6 (13.9%) |
| #5 (11.5%) | #4 (8.9%) |
| A | B | C | D | E | F |
|---|---|---|---|---|---|
| #2 (22.1%) | #1 (20.2%) | #1 (29.8%) | #2 (20.2%) | #6 (22.1%) | #3 (23.8%) |
| #6 (20.2%) | #5 (18.3%) | #4 (20.2%) | #3 (19.2%) | #4 (20.2%) | #5 (20.8%) |
| #4 / #5 (17.3%) | #4 / #6 (17.3%) | #5 (13.5%) | #1 (18.3%) | #3 / #5 (19.2%) | #2 (16.8%) |
| #3 (14.4%) | #2 (14.4%) | #2 / #6 (12.5%) | #4 (16.4%) | #2 (14.4%) | #1 (15.8%) |
| #1 (8.7%) | #3 (12.5%) | #3 (11.5%) | #6 (14.4%) | #1 (4.8%) | #6 (13.9%) |
| #5 (11.5%) | #4 (8.9%) |
# = ranking number.
In absolute terms (Table 4), the highest ranked image was C (with a mean 3.88 points), followed by image D (3.74 points), F (3.64 points), B (3.49 points), A (3.27 points), and finally image E (2.99 points) as the lowest-ranked image. After adjusting for potential different responses according to participant age, sex, or ethnicity, image E was significantly worse than images C, D, and F (P < .05) and while also worse than images A and B, these differences were non-significant (Table 5).
Response-based ranking of images.
| Ranking | Image | Mean (95% CI) |
|---|---|---|
| 1st | C | 3.88 (3.53, 4.22) |
| 2nd | D | 3.74 (3.41, 4.07) |
| 3rd | F | 3.64 (3.30, 3.97) |
| 4th | B | 3.49 (3.14, 3.84) |
| 5th | A | 3.27 (2.95, 3.59) |
| 6th | E | 2.99 (2.70, 3.29) |
| Ranking | Image | Mean (95% CI) |
|---|---|---|
| 1st | C | 3.88 (3.53, 4.22) |
| 2nd | D | 3.74 (3.41, 4.07) |
| 3rd | F | 3.64 (3.30, 3.97) |
| 4th | B | 3.49 (3.14, 3.84) |
| 5th | A | 3.27 (2.95, 3.59) |
| 6th | E | 2.99 (2.70, 3.29) |
CI = confidence interval.
Response-based ranking of images.
| Ranking | Image | Mean (95% CI) |
|---|---|---|
| 1st | C | 3.88 (3.53, 4.22) |
| 2nd | D | 3.74 (3.41, 4.07) |
| 3rd | F | 3.64 (3.30, 3.97) |
| 4th | B | 3.49 (3.14, 3.84) |
| 5th | A | 3.27 (2.95, 3.59) |
| 6th | E | 2.99 (2.70, 3.29) |
| Ranking | Image | Mean (95% CI) |
|---|---|---|
| 1st | C | 3.88 (3.53, 4.22) |
| 2nd | D | 3.74 (3.41, 4.07) |
| 3rd | F | 3.64 (3.30, 3.97) |
| 4th | B | 3.49 (3.14, 3.84) |
| 5th | A | 3.27 (2.95, 3.59) |
| 6th | E | 2.99 (2.70, 3.29) |
CI = confidence interval.
Pairwise comparisons of average ranking scores adjusting for the effect of patient age, sex, ethnicity and taking patient-clustering into account
| Reference group | ||||||
|---|---|---|---|---|---|---|
| Image | A | B | C | D | E | F |
| A | −0.22 P = .39 | −0.61 P = .007 | −0.47 P = .07 | +0.28 P = .26 | −0.37 P = .19 | |
| B | +0.22 P = .39 | −0.38 P = .19 | −0.25 P = .28 | +0.50 P = .06 | −0.14 P = .60 | |
| C | +0.61 P = .007 | +0.38 P = .19 | +0.13 P = .65 | +0.88 P < .001 | +0.24 P = .35 | |
| D | +0.47 P = .07 | +0.25 P = .28 | −0.13 P = .65 | +0.75 P = .001 | +0.11 P = .68 | |
| E | −0.28 P = .26 | −0.50 P = .06 | −0.88 P < .001 | −0.75 P = .001 | −0.64 P = .004 | |
| F | +0.37 P = .19 | +0.14 P = .60 | −0.24 P = .35 | −0.11 P = .68 | +0.64 P = .004 | |
| Reference group | ||||||
|---|---|---|---|---|---|---|
| Image | A | B | C | D | E | F |
| A | −0.22 P = .39 | −0.61 P = .007 | −0.47 P = .07 | +0.28 P = .26 | −0.37 P = .19 | |
| B | +0.22 P = .39 | −0.38 P = .19 | −0.25 P = .28 | +0.50 P = .06 | −0.14 P = .60 | |
| C | +0.61 P = .007 | +0.38 P = .19 | +0.13 P = .65 | +0.88 P < .001 | +0.24 P = .35 | |
| D | +0.47 P = .07 | +0.25 P = .28 | −0.13 P = .65 | +0.75 P = .001 | +0.11 P = .68 | |
| E | −0.28 P = .26 | −0.50 P = .06 | −0.88 P < .001 | −0.75 P = .001 | −0.64 P = .004 | |
| F | +0.37 P = .19 | +0.14 P = .60 | −0.24 P = .35 | −0.11 P = .68 | +0.64 P = .004 | |
Pairwise comparisons of average ranking scores adjusting for the effect of patient age, sex, ethnicity and taking patient-clustering into account
| Reference group | ||||||
|---|---|---|---|---|---|---|
| Image | A | B | C | D | E | F |
| A | −0.22 P = .39 | −0.61 P = .007 | −0.47 P = .07 | +0.28 P = .26 | −0.37 P = .19 | |
| B | +0.22 P = .39 | −0.38 P = .19 | −0.25 P = .28 | +0.50 P = .06 | −0.14 P = .60 | |
| C | +0.61 P = .007 | +0.38 P = .19 | +0.13 P = .65 | +0.88 P < .001 | +0.24 P = .35 | |
| D | +0.47 P = .07 | +0.25 P = .28 | −0.13 P = .65 | +0.75 P = .001 | +0.11 P = .68 | |
| E | −0.28 P = .26 | −0.50 P = .06 | −0.88 P < .001 | −0.75 P = .001 | −0.64 P = .004 | |
| F | +0.37 P = .19 | +0.14 P = .60 | −0.24 P = .35 | −0.11 P = .68 | +0.64 P = .004 | |
| Reference group | ||||||
|---|---|---|---|---|---|---|
| Image | A | B | C | D | E | F |
| A | −0.22 P = .39 | −0.61 P = .007 | −0.47 P = .07 | +0.28 P = .26 | −0.37 P = .19 | |
| B | +0.22 P = .39 | −0.38 P = .19 | −0.25 P = .28 | +0.50 P = .06 | −0.14 P = .60 | |
| C | +0.61 P = .007 | +0.38 P = .19 | +0.13 P = .65 | +0.88 P < .001 | +0.24 P = .35 | |
| D | +0.47 P = .07 | +0.25 P = .28 | −0.13 P = .65 | +0.75 P = .001 | +0.11 P = .68 | |
| E | −0.28 P = .26 | −0.50 P = .06 | −0.88 P < .001 | −0.75 P = .001 | −0.64 P = .004 | |
| F | +0.37 P = .19 | +0.14 P = .60 | −0.24 P = .35 | −0.11 P = .68 | +0.64 P = .004 | |
Free text comments by responders most often suggested difficulty in detecting differences between some of the images.
Discussion
This observational study has investigated the influence of gingival margin height variation on smile aesthetics in the presence of maxillary lateral incisor-canine substitution using a lay population of observers.
A number of previous studies have investigated the aesthetics of maxillary anterior teeth in relation to space opening or space closure for absent maxillary lateral incisors using images of the dentition. Although the precise methodology varies, there is some evidence from these investigations that the aesthetics of space closure are generally favoured by the lay person (including patients) [15, 24–26]; although orthodontists and dentists do prefer an intact unrestored dentition, even when compared with cases treated with space opening or closing, chosen on the basis of excellence in outcome [26]. This is perhaps not surprising because most clinicians would be expected to be able to differentiate an intact dentition from one involving restorations or space closure in the upper labial segment. However, there is less data relating to variation in gingival margin aesthetics in association with maxillary lateral incisor-canine substitution cases [9] although gingival height for the substituted canine 0.5 mm below the gingival margin of the central incisors has been associated with the most favourable aesthetics among orthodontists, dentists, and lay people when asked to judge constructed images [18].
In this study, image C most frequently attracted the number 1 rank (with almost one third of respondents ranking it first) and was the highest-scoring image in absolute terms. This is clinically useful data because image C represents a common scenario in the management of maxillary lateral incisor-canine substitution cases. Extrusion and coronal reduction of the canine is often readily achievable by the orthodontist and is helpful in achieving the goal of substituting this tooth for a lateral incisor: lowering the gingival margin below that of the central incisor and facilitating occlusal coronal reduction of the canine crown to make it smaller. However, obtaining an ideal gingival margin contour for the first premolar substituting as a canine (ideally, at the same level or slightly higher than that of the central incisor) is more difficult to achieve, and it can be technically easier to leave the gingival margin of this tooth below that of the central incisors and avoid significant coronal intrusion of this tooth. This can also be advantageous because it reduces the reliance on restorative build-up of the premolar crown, and there is some association between intrusion of this tooth and gingival recession over the longer-term [22]. Our investigation would suggest that in the presence of ideal gingival margin aesthetics associated with canines substituting for lateral incisors, the gingival margins of the premolar teeth substituting for the canines can be left below those of the canines substituting for the lateral incisors, and this will not detract from good aesthetics for a significant number of lay persons. However, these judgements are quite subjective, images B and D were the second and third-most often assigned the number 1 rank, respectively; and these were associated with gingival margins for the substituted lateral incisors that were above [image B] and level with [image D] the central incisors. Image D was also second-highest ranked in absolute terms although image B was only the fourth highest. These data suggest that some variation around the established ideal for substituted lateral incisors can also result in pleasing aesthetics. This is also clinically useful information because in the presence of larger canines, achieving a gingival margin below that of the central incisor can result in a requirement for significant coronal reduction, which can potentially compromise this tooth.
The variations most often highest ranked [images C, B, D] represented symmetrical variation in gingival margin heights. Image D had all the anterior teeth with gingival margins level with the central incisors, essentially representing the opposite of image C in that, the substituted lateral incisor margins (canine teeth) were now too high, and the substituted canines (premolar teeth) were ideal. The most common second-ranked image was image B, which also represents a common clinical scenario, where the substituted lateral incisor gingival margins are high, and the substituted canines are low. Again, this was not seen to be associated with poor aesthetics by many of the lay observers. Interestingly, image A representing normal gingival aesthetics was most commonly ranked second-highest (although in absolute terms it was the fifth out of six ranked image), indicating that ideal aesthetics can resonate with the lay person, but other gingival margin contours can be regarded as more aesthetic. Overall, lay people often found it difficult to differentiate between the aesthetic manipulations that were carried out, particularly in relation to images A, B, C, and D, treating these smiles similarly in terms of preference. Indeed, a number of free comments within the questionnaire described some difficulty in identifying any differences between the images for some observers.
Image E was the least popular overall, with almost a quarter of respondents ranking it in last place. After adjusting for potential different responses according to participant age, sex, or ethnicity, image E was significantly worse than images C, D, and F (P < .05), and worse than image B (although marginally non-significant; P = .06). Image E was associated with bilateral asymmetry in gingival heights of the substituted lateral incisors and canines, and it would seem from these findings that asymmetrical gingival margin heights, particularly affecting the substituted canines are something to avoid in association with the upper six maxillary anterior teeth.
There is a body of evidence to suggest that the judgements of patients or lay people in terms of pleasing oro-facial and dental aesthetics do not always overlap with those of clinicians [27–29]. Indeed, orthodontic training and clinical adherence to strict norms of facial beauty can result in more critical judgment of female attractiveness among orthodontists compared to the lay person [30] and significant differences in how facial harmony is assessed after certain types of treatment, including mandibular advancement surgery [31]. It is interesting to note that image A representing normal gingival margin anatomy, did not rank highly among lay observers (being ranked highest by only 8.7% and being rated fifth out of six in absolute terms), despite being ranked at number 2 most often (25%).
Limitations of the study
There are a number of potential limitations associated with a subjective study of this nature. The observational cross-sectional design provides data from only a single point in time, which makes generalisability of the results poor. The lay population were parents of patients undergoing orthodontic treatment with fixed appliances, and while the study criteria excluded those working in dentistry or with direct or indirect experience of orthodontic treatment, or treatment involving the maxillary incisor teeth, there is some risk of sample bias within this population. In addition, the morphology, colour, and width ratios of the maxillary teeth were not altered as part of this study because it was felt that this would introduce too many additional variables, but in reality, these are often issues when managing these patients clinically [7]. Overall, it was felt to be more relevant if a real canine substitution case was manipulated, increasing realism of the images but also meaning that the canine coronal morphology and shade were not necessarily perfect. In addition, the use of professional images of the anterior teeth, taken in colour using dental retractors represents a clinical scenario that lay people would be unlikely to experience. The use of a social smile with visible lips, ideal upper lip curvature and smile arc might have been more appropriate; however, this also introduces additional variables and necessitates the use of a high smile line to allow visibility of the gingival margins, which in itself might influence aesthetic judgements. It was felt more appropriate to focus on the gingival margins using professional imaging, which is a common methodology in these types of observational study [18, 24, 26]. A final consideration is tooth size variation because cases with maxillary lateral incisor agenesis are often accompanied by reduced maxillary central incisor crown width and indeed, canine crown width [32]. This can also affect aesthetics of the maxillary labial segment and often requires restorative intervention. However, no attempt was made to manipulate coronal widths in this study, the case was selected on the basis of having maxillary central incisors and substituting canine and first premolar teeth with coronal widths within normal limits. This was primarily because any variation would have introduced an additional variable and the case was selected on the basis of pleasing dental aesthetics following space closure and canine substitution to allow for gingival height variation manipulation in isolation.
Conclusions
We find that in the presence of maxillary lateral incisor-canine substitution, lay people regard a range of gingival margin heights as having pleasing aesthetics. Normal anatomical gingival margin heights were not the highest-ranking aesthetic choice. Lateral incisor gingival margins symmetrically below the central incisors and substituted canine margins symmetrically below ranked highest, while normal lateral incisor gingival margins and asymmetric substituted canine margins ranked lowest. Vertical asymmetry in substituted canine position was considered the least aesthetic, while vertical asymmetry of substituted lateral incisors was not considered to worsen the aesthetics.
Author contributions
Catherine Hershaw (Conceptualization [equal], Formal analysis [equal], Methodology [equal], Writing—original draft [equal], Writing—review & editing [equal]), Jadbinder Seehra (Formal analysis [equal], Writing - original draft [equal], Writing - review & editing [equal], Nicola Johnson (Project administration [equal], Resources [equal]), Carmel Slipper (Project administration [equal], Resources [equal]), Spyridon Papageorgiou (Formal analysis [equal], Methodology [equal], Writing original draft [equal], Writing - review & editing [equal], Koula Asimakopoulou (Conceptualization [equal], Investigation [equal], Methodology [equal], Writing—original draft [equal], Writing—review & editing [equal]), and Martyn Cobourne (Conceptualization [equal], Project administration [equal], Resources [equal], Formal analysis [equal], Methodology [equal], Writing - original draft [equal], Writing - review & editing [equal].
Conflict of interest
The authors declare that there are no conflicts of interest.
Funding
None declared.
Data availability
The source data are freely available at doi: 10.5281/zenodo.8394975.
