Enameloplasty of maxillary canines is often needed for aesthetic substitution in patients with congenitally missing lateral incisors. The exact enamel thicknesses for the various canine surfaces are unknown because previous studies failed to employ accurate measurement tools to report and compare detailed enamel thicknesses for each surface at various crown heights.
Thirty-two extracted maxillary canines were collected and scanned in a microcomputed tomography scanner. The scans were imported into a custom-written MATLAB software (version 9.2; MathWorks, Natick, Mass) and the enamel thickness on the mesial, distal, labial, fossa, cingulum, and incisal edge of each tooth was computed, obtaining the mean value from slices at 0.1 mm intervals. The overall mean enamel thickness for each surface was also calculated, and these values were compared using paired t tests. Incisal wear stage and incisal enamel thickness that was measured were compared using Spearman rank correlation coefficient.
The mean enamel thickness was significantly thinner at the gingival level when compared with the incisal for all surfaces that were analyzed (1-tailed, P <0.001). The mean enamel coverage at the mesial was significantly thinner than the distal when measured gingival to the widest mesiodistal area. The mean enamel coverage of the cingulum was particularly thin and therefore requires extreme care in reshaping it. Incisal edge enamel thickness was highly negatively correlated with the wear stage of the scoring system that was used (1-tailed, P <0.001).
The enamel coverage of the maxillary canine varies depending on the tooth surface and the incisogingival measurement location.
Enamel became thinner from incisal to gingival at all surfaces.
Enamel at the mesial was thinner than the distal at the gingival level.
Mean enamel coverage of the cingulum was particularly thin.
Smith’s scoring system of incisal wear is a reliable predictor for enamel coverage.
Patients with congenitally missing maxillary lateral incisors are commonly seen by orthodontists. According to the literature, the prevalence of missing maxillary lateral incisors has been reported to be 1%-3%, and it occurs bilaterally more commonly than unilaterally. , In these patients, the missing tooth can be replaced with an implant or substituted with the adjacent canine. ,
The decision is case-sensitive and depends on factors such as the patient’s age, the occlusion, the periodontium, and the skeletal growth pattern. Both treatment options have similar outcomes concerning the patient’s periodontal health and temporomandibular joint function. , The evolution of implants and their placement techniques have also improved their esthetics; however, in a study conducted by Schneider et al, laypeople preferred the intraoral images of a canine substitution case over those of a case treated with implant restorations.
Replacing the missing tooth with an implant is technique-sensitive and requires a multispecialty team that includes an oral surgeon or a periodontist, an orthodontist, and a restorative dentist. The substitution option can be performed by the orthodontist alone or with assistance from a restorative dentist, but enameloplasty will likely need to be performed on the canine, possible bleaching and/or a composite or veneer restoration. , ,
Enameloplasty usually needs to be performed on all surfaces of the maxillary canine. The canine is on average 1.2 mm wider than the maxillary lateral incisor, its facial surface is more convex, its incisal edge is sharper, and its lingual surface has prominent mesial and distal marginal ridges and a cingulum that might interfere with the mandibular incisors. , ,
Thin remaining enamel or exposed dentin can result in tooth sensitivity, or the tooth may remain asymptomatic. Interproximal reduction has not been correlated with increased caries, and induced tooth sensitivity is highly uncommon. Currently, air-rotor stripping guidelines suggest that up to 50% of enamel thickness can be removed without compromising dental and periodontal health. Others mention that up to 0.5 mm enamel tissue can be safely removed from each anterior contact area and up to 1 mm in the posterior region.
Another potential side effect of enamel removal is a change in the color esthetics of the tooth surface. When Oguro et al removed layers of enamel in stages from the facial surface of extracted central incisors, they showed that color, chroma, lightness, and the blue-yellow coordinate of value changed substantially as layers of enamel were removed.
The enamel thickness of maxillary canines has been previously studied in the literature. , Sectioning the teeth to measure the enamel directly from the sections is an accurate method, but it can introduce problems in the specimen orientation and destroy part of the enamel tissue causing loss of useful information. , , ,
Nondestructive methods to visualize the enamel have been obtained with bitewing radiographs, computed tomography, ultrasound and lately microcomputed tomography (microCT). However, most noninvasive methods lack the accuracy that is needed to measure enamel thickness. , , Of the noninvasive methods, microCT is the “gold standard” to measure dental enamel thickness because of its high accuracy. , ,
The purpose of this study was to employ microCT technology to measure the thickness of the enamel on the mesial, distal, labial, cingulum, fossa, and incisal surfaces of maxillary canines; compare the measurements; and provide clinical guidance for lateralization in patients who require canine substitution.
Material and methods
For this study, 32 extracted permanent maxillary canines were collected from several periodontal and oral surgery clinics. These teeth came from patients in which extractions were indicated because of impaction, ankyloses, or poor periodontal health. The teeth were visually assessed by 1 investigator (E.A.), and only integral tooth surfaces, without severe decay or severe attrition, abrasion, and/or abfraction, were accepted. The selected sample was evaluated for incisal wear by the same investigator based on the index by Smith et al. Although Smith et al suggested evaluating the incisal edge as a whole, the teeth in the sample were scored from their most prominent incisal point, which also coincided with the first most incisal slice of the microCT scan.
To perform the scans, we sectioned the crowns of the teeth from their roots without affecting the enamel margins in the cervical region. The crowns were then suspended in 2% agarose in a 20 mm diameter tube with their anatomic tooth axis set vertically, as visually assessed, to be scanned by a microCT scanner.
A Scanco μ40 scan (μCT 40; Scanco Medical, Bassersdorf, Switzerland) was selected to be used for this study as it produces comparable results with other machines on the market. The scanner had undergone weekly calibrations compared to a known phantom. The scanner was set at 70 kVp, 115 μA, and at high resolution with isometric voxels, the size of which was 10 μm 3 for all scans. Each slice that was generated consisted of 2048 × 2048 pixels, 16 bits per pixel. The slice thickness was kept constant at 0.01 mm.
Because the teeth were already extracted, their contact areas were assumed to be near their widest mesiodistal slice. This slice was located digitally and chosen as a “reference slice.” The enamel thickness was measured on the mesial and distal surfaces starting from the “reference slice” and at each slice within 2 mm incisally and 4 mm gingivally ( Fig 1 ).
The “reference slice” that was chosen for the labial measurements was consistent with the midpoint between the most incisal slice and the most gingival slice that contained enamel on its labial surface. The “reference slice” was taken at 0.25 mm and 0.5 mm more incisally from that midpoint in the case of teeth with incisal wear stages 1 and 2 respectively to account for the estimated incisal wear of 0.5 mm and 1 mm. Measurements were performed at each slice within 2 mm incisally and 4mm gingivally from that “reference slice” to contain the surface that clinicians will need for lateralization enameloplasty ( Fig 1 ).
The lingual surface was separated into the area of the lingual fossa and the area of the cingulum. The “reference slices” for the fossa and cingulum were chosen to be the slice consistent with the midpoint between the most incisal and most gingival slice where the anatomy of the fossa was discerned and the midpoint between the most incisal and most gingival slice where the anatomy of the cingulum was discerned, respectively. The enamel thickness of each slice within 1.5 mm incisally and 1.5 mm gingivally from each of the 2 “reference slices” was measured ( Fig 1 ).
The enamel thickness on the incisal edge was determined by measuring the distance from the most incisal slice of the tooth to the most incisal slice that contained dentin ( Fig 1 ).
To perform the measurements, we used custom-written MATLAB software (version 9.2; MathWorks, Natick, Mass) was used. The digital imaging and communications in medicine files of the scans were uploaded in the software, and the images were filtered with a 3-dimensional-median filter using a pixel neighborhood with radius 5 to account for some speckle noise. After that, manual segmentation was performed to isolate each tooth. Each tooth was then reoriented so that its y-axis would coincide with the anatomic long axis. A threshold of 17,000 gray scale units was empirically determined and applied in each tooth to highlight the enamel from the surrounding structures.
A mask was drawn to isolate an area of interest at each slice to measure the enamel thickness. Eight landmarks were used to define the areas and draw the masks—4 enamel and 4 corresponding dentinal landmarks. The enamel and dentinal landmarks that were used are shown in Figure 2 . The enamel landmarks were connected with their corresponding dentinal landmarks to draw the mask for the area of interest ( Fig 3 ).
The surfaces of canines that contained caries or other obvious defects were removed from the sample and were not measured. Each surface was isolated with the mask, and the mean thickness and standard deviation (SD) of the highlighted enamel area were measured for each slice at 0.1 mm intervals using the Hildebrand method. In addition, the overall mean thickness and SD of all teeth were computed for each slice and each surface.
Statistical analysis was performed with SPSS software (version 24.0; SPSS, Chicago, Ill). Paired t test analysis was used to compare various mean enamel thicknesses at the mesial, distal, labial, fossa, and cingulum. Spearman rank correlation coefficient was used to investigate for possible correlation between the incisal wear stage index that was used to classify the teeth and the mean enamel thickness of the incisal edge that was found for each group. Finally, 10% of the sample was randomly selected to be remeasured by the same person 1 week after the initial analysis. In addition, the incisal wear of the entire sample was scored again. The results were then compared with the initial results with the intraclass correlation coefficient to evaluate the reliability of the methods.
The mean enamel thickness for all the teeth at each slice with 0.1 mm intervals is shown in Figure 4 . The mean enamel thickness values for all the teeth at the “reference slices” and the extremes of each analyzed surface are shown in Table I . The mesial, distal, and labial surfaces that were analyzed over a 6 mm area were divided into 3 segments every 2 mm, and the means and SDs of the segments are reported in Table II . The enamel thickness for 1 tooth that was representative of the mean observed measurements is shown in spectral colors ( Fig 5 ).