Maxillary canines are generally considered important both cosmetically and functionally. Most claims on the importance of maxillary canines, however, have been based on expert opinions and clinician-based studies. There are no scientific studies in the literature reporting on their cosmetic importance or how laypeople perceive a smile treated by maxillary canine extractions. Our objective was to investigate whether there is any difference in the perceptions of patients’ smiles treated by extracting either maxillary canines or first premolars, as judged by orthodontists, dentists, and laypeople.
This retrospective study included 24 participants who had unilateral or bilateral extraction of maxillary permanent canines and fixed appliances in the maxillary and mandibular arches to comprehensively correct the malocclusion, selected from orthodontic patients treated at Chesterfield Royal Hospital NHS trust in the United Kingdom over the last 20 years. The control group of patients had extraction of maxillary first premolars followed by fixed appliances and finished to an extremely high standard judged by the requirement that they had been submitted for the Membership in Orthodontics examination. The finished Peer Assessment Rating scores for this group were less than 5. The end-of-treatment frontal extraoral smiling and frontal intraoral views were presented for both groups. The photographs were blinded for extraction choice and standardized for size and brightness using computer software (Adobe Photoshop CC version 14.0; Adobe Systems, San Jose, Calif). The work file was converted to an editable pdf file and e-mailed to the assessors. The assessor panel consisted of 30 members (10 orthodontists, 10 dentists, and 10 laypeople), who were purposely selected. The measures were rated on a 10-point Likert scale.
The attractiveness ratings were not statistically significantly different between the canine extraction and premolar extraction groups, with a mean difference of 0.33 (SD, 0.29) points. A 1-way repeated-measures analysis of variance to test the difference in scores among the laypeople, orthodontists, and dentists (n = 30) showed no statistically significant difference (Wilks lambda = 0.835; P = 0.138), and the Bonferroni test indicated that no pair-wise difference was statistically significant.
No statistically significant difference was found in the smile attractiveness between canine extraction and premolar extraction patients as assessed by general dentists, laypeople, and orthodontists. Further high-quality studies are required to evaluate the effect of canine extraction and premolar substitution on functional occlusion.
Little evidence supports the cosmetic and functional importance of maxillary canines.
Raters found little difference in attractiveness between canine and premolar extractions.
Canine extraction should be a treatment option when it has other advantages.
Effect of canine extraction and premolar substitution on functional occlusion should be studied.
Maxillary canines are infrequently extracted for orthodontic treatment, since these teeth are considered important both cosmetically and functionally. The concept of canine-guided occlusion is often claimed to be the optimal type of functional occlusion for the natural dentition. It is also firmly believed by some that orthodontists who do not establish a canine-guided occlusion may predispose patients to temporomandibular disorders. The evidence for this claim, however, is scarce. Weinberg showed that 81% of a sample with an untreated natural dentition had a group function, whereas only 5% had a canine-protected occlusion. This has been reinforced by other researchers. It has also been shown that canine-guided occlusion is a popular choice for orthodontic and prosthodontic rehabilitation. Interestingly, a recent study on congenitally missing lateral incisors and space closure concluded that substitution of first premolars for canines does not incur any risk for temporomandibular disorders in the long term.
Canines are also considered extremely important for cosmetic appearance. Wheeler described canines as “a foundation that insures normal facial expression.” Furthermore, Dewel emphasized the importance of canines as “indispensable for maintenance of harmony and symmetry of occlusal relationships.” The evidence for these claims, however, is scarce. In a study of 56 cases, Senty et al reported that the first premolar can serve as an adequate substitute for the canine, both functionally and esthetically. With the lack of convincing evidence for both functional and cosmetic importance of canines, the philosophy of canine “sanctity” may indeed have been an exaggeration of the real situation.
There are undeniably many situations when maxillary canines must at least be considered candidates for extraction: eg, a patient with an ectopically placed maxillary permanent canine with a severe arch-length discrepancy where the first premolar has almost completely replaced the erupted canine. Even though canine surgical removal is by far the quickest and simplest approach, and often in the best interest of patient, parent, and orthodontist, we still opt for extraction of first premolars to heroically align the impacted canines. This clinical scenario creates something of a dilemma for a clinician practicing in the present climate of evidence-based dentistry.
Most of the claims on the importance of the maxillary canines have been based on expert opinions and clinician-based studies. There are no scientific studies in the literature reporting on the cosmetic importance of canines or how laypeople perceive a smile treated by maxillary canine extractions.
Our aim in this study was to investigate whether there is any difference in the perceptions of patients’ smiles treated by extracting either maxillary canines or premolars, as judged by orthodontists, dentists, and laypeople.
Our hypothesis was that there is no difference in the perceptions of patients’ smiles treated by extracting either maxillary canines or premolars, as judged by orthodontists, dentists, and laypeople.
Material and methods
This was a retrospective study in which participants were recruited from orthodontic patients treated at Chesterfield Royal Hospital in the United Kingdom over the last 20 years. The inclusion criteria for the treatment group were patients who had unilateral or bilateral extraction of maxillary permanent canines followed by fixed appliances in the maxillary and mandibular arches to comprehensively correct the malocclusion.
Exclusion criteria were syndromic patients, patients with single-arch treatment, and patients who failed to complete the full course of orthodontic treatment.
Patients in the canine extraction group were matched with patients who had unilateral or bilateral extraction of maxillary first premolars followed by maxillary and mandibular fixed appliances for the treatment of their malocclusion. We selected a control group of patients who had been finished to an extremely high standard. The finished Peer Assessment Rating scores were less than 5 for this group. No cosmetic restorative work was carried out for any patient. Institutional approval was obtained from Chesterfield Royal Hospital NHS trust.
End-of-treatment photographic records were obtained for both groups. The photographs were standardized using computer software (Adobe Photoshop CC version 14.0; Adobe Systems, San Jose, Calif) for size and brightness. For each patient, frontal extraoral smiling and frontal intraoral views were presented. The images were cropped, rescaled, and pasted into a word file. The photographs were blinded as to the extraction choice, and a random number was assigned for blinded analysis by a researcher (B.T.). Samples of photographs of patients in the canine extraction group and the premolar extraction group are shown in Figures 1 and 2 , respectively. The Word file was converted to an editable pdf file and e-mailed to the assessors.
The 30 members of the rating panel, who were purposely selected, comprised 10 orthodontists, 10 dentists, and 10 laypeople. The laypeople were adult volunteers from Chesterfield Royal Hospital NHS trusts. None of the assessors were involved in the care of the patients included in the study. The assessors were asked to rate the photographs according to the attractiveness of the patient’s smile. The measures were rated on a 10-point Likert scale.
The Likert scale from 1 to 10 was selected over the visual analog scale for logistical reasons. Jaeschke et al showed, however, that both methods of presenting the response data have the same levels of validity and responsiveness. In this study, high scores indicated an unattractive smile. A reliability assessment was carried out for 12 assessors (4 in each group) with at least a 4-week interlude to minimize recall bias and fatigue.
Data were entered into an Excel spreadsheet (2007; Microsoft, Redmond, Wash), and SPSS software (version 18; SPSS, Chicago, Ill) was used for the statistical tests. Because the 10-point Likert scale was used with the intervals between points considered approximately equal (the underlying concept was continuous), parametric tests were used in this study. The distribution of the data was examined and found to be normally distributed. The reliability was assessed using a paired t test for systematic error and intraclass correlation coefficients for random errors.
The mean scores for the assessors’ judgments of the 2 groups were generated from the Likert scale for smile attractiveness. The following rater groupings were made: professionals and laypeople. The professionals were subdivided into dentists and orthodontists. One-way analysis of variance (ANOVA) was used to compare the groups’ scores, with P values of 0.05 or less considered statistically significant.
Forty-eight patients (32 female, 16 male) were included in this study: 24 in the treatment group and 24 in the control group. The canine extraction group included 14 with unilateral extraction and 10 with bilateral extraction. The photographs were assessed by 6 specialist orthodontists, 6 general dentists, and 6 laypeople on 1 occasion, and by 4 specialist orthodontists, 4 general dentists, and 4 laypeople on 2 occasions. The level of agreement for the 12 judges on the smile attractiveness between the canine extraction and premolar extraction groups is shown in Table I . The mean difference was 0.25, suggesting low systematic error. The random error showed substantial agreement (intraclass correlation coefficient, 0.728).
|Judges (n)||Mean difference||SD of difference||95% CI||P value||ICC|
|Lower bound||Upper bound|
The mean scores for the esthetic evaluation by the assessors of the canine extraction and premolar extraction groups are shown in Figure 3 . The attractiveness ratings were not statistically significantly different between the 2 groups; hence, the null hypothesis was accepted ( Table II ).
|Groups||Mean difference||SE of difference||95% CI||P value|
|Lower bound||Upper bound|
|Canine vs premolar extraction||0.33||0.29||−0.26||0.91||0.268|
A 1-way ANOVA repeated measures was conducted to evaluate the null hypothesis that there was no difference in the scores of laypeople, orthodontists, and dentist (n = 30). The results of the ANOVA indicated no significant difference in scores (Wilks lambda = 0.835; F 92, 22 = 2.17; P = 0.138; λ2) ( Table III ).
|Assessors||Mean difference||SE of difference||95% CI||P value|
|Lower bound||Upper bound|