The aim of this study was to verify the accuracy of preformed wire shape templates on plaster models and those of customized digital arch form diagrams on digital models.
Twenty pairs of dental plaster models were randomly selected from the archives of the Department of Orthodontics of Federal Fluminense University, Niterói, Rio de Janeiro, Brazil. All plaster model samples were scanned in a plaster model scanner to create the respective digital models. Three examiners defined the arch form on the mandibular arch of these models by selecting the ideal preformed wire shape template on each plaster model or by making a customized digital arch form on the digital models using a digital arch form customization tool. These 2 arch forms were superimposed by the best-fit method. The greatest differences in the 6 regions on the superimposed arches were evaluated. Each examiner presented a descriptive analysis with the means, standard deviation, and minimum and maximum intervals of the differences on the superimpositions. Intraclass correlation coefficient and paired t tests were used to evaluate the accuracy of the superimpositions.
Among the 6 regions analyzed in the superimpositions, the largest differences in the anterior and premolar regions were considered clinically insignificant, whereas the largest differences in the right molar region, especially the second molar area, were considered clinically significant by all 3 examiners. The intraclass correlation coefficients showed a weak correlation in the premolar region and moderate correlations in the anterior and molar regions. The paired t test showed statistically significant differences in the left anterior and premolar regions.
The superimpositions between the arch forms on plaster and digital models were considered accurate, and the differences were not clinically significant, with the exception of the second molar area. Despite the favorable results, the requirement of correcting some software problems may hamper the transition from plaster to digital models.
The methods to define the arch form on plaster and digital models were accurate.
The largest differences between the methods were in the second molar region.
The digital method of arch form definition can substitute for the conventional method on plaster models.
It is necessary to correct digital arch form magnification in the software to perform the arch form definition on digital models.
The key to the success or failure of an orthodontic treatment is related to the correct positioning of the teeth in the apical base; the arch form must be preserved along with its transversal dimensions. It is also important to maintain a functional balance between the tongue and the circumoral muscle forces. Because of the immense variability in dental arch forms among patients, any arch form may not fit every dental arch. According to Lee et al, arch form types are influenced by tooth size, arch width, and inclination of the posterior teeth. Paranhos et al found that the most common shape of the mandibular dental arch was oval (41%), followed by square (39%) and tapered (20%).
Since the arch form is an important factor for the stability of the orthodontic treatment, several diagrams or wire shape templates were proposed to facilitate or make more didactic the representation of the mandibular arch shape. The plaster model is a traditionally used tool for diagnosis and treatment planning in orthodontics. It is often used to choose the best diagram that determines the shape of the mandibular arch. However, handling plaster models during wire shape definition might not always be practical; moreover, fractures are common. In such instances, the use of digital models may prove to be a good alternative.
Some studies have proposed arch form definitions with software programs on digital models and photocopied plaster models. The first attempts to draw a curve representing the arch forms from radiographs of plaster models using computer software programs were conducted in the late 1960s. However, within the next 2 decades, the use of software programs to define the arch form on photocopies of plaster models had gained popularity in clinical orthodontic practices.
Several studies have suggested different methods for the attainment of an optimum arch shape. Some standard forms such as semicircle, ellipse, parabola, catenary curve, and wire shape diagrams including tapered, ovoid, and square forms have been widely used to select prefabricated orthodontic archwires. The application of a Cartesian system onto the photocopies of the plaster models, identifying the x- and y-axes, facilitates the visual evaluation of arch morphology. Another option is the application of sixth-degree polynomials, establishing the 6 most preponderant arch configurations, thereby guiding the orthodontist to visually choose the one that best fits the patient. It was observed that, irrespective of the complexity of the methodology used to determine and choose the dental arch shape, the final choice is subjectively made by the orthodontist in a visual manner.
According to a study by Trivino et al, the arch curve morphology in the anterior region was divided into 8 groups with 3 sizes in each region. A wire shape diagram template for plaster models was created based on this study. Nowadays, customizing the designing of arch forms may provide an option for accurately describing the ideal orthodontic arch form for a particular patient.
In clinical orthodontic practice, the selection of preformed archwires is estimated by visual examination or with the aid of arch form templates. The choice of diagrams or wire shape templates in plaster models is a routine procedure used by orthodontists. However, there are doubts about the accuracy of diagrams in digital models when compared with plaster models because of the lack of scientific evidence. Furthermore, since it is a new procedure, some orthodontists are not familiar with the use of diagrams in digital models either in the form of digitized arch form templates or by creating customized digital diagrams using specific software programs.
In this study, we aimed to verify the accuracy of the use of wire shape diagrams on plaster models and customized digital arch forms on digital models based on evaluations by 3 examiners.
Material and methods
From the archives of the Department of Orthodontics of Federal Fluminense University, Niterói, Rio de Janeiro, Brazil a sample containing 20 pairs of dental plaster models was randomly selected. The following inclusion criteria were used in this study: presence of all maxillary and mandibular permanent teeth up to the second molars, malocclusions with different levels of severity, various arch shapes, and treatments without dental extractions. Exclusion criteria were models of surgical patients and those with severe growth abnormalities. The local ethics committee of our university approved this study on July 22, 2016 (process number 57075116.0.0000.5243).
The following 3 examiners were included in this study: an undergraduate student of dentistry (examiner 1), a postgraduate student of orthodontics (examiner 2), and an orthodontist with more than 10 years of experience (examiner 3). Mucha’s arch form individualized diagram, a wire shape diagram template used in the Orthodontics Department of Federal Fluminense University, Niterói, Rio de Janeiro, Brazil, presents 20 arch types printed on transparent acetate that is superimposed on the patient’s original plaster model. These arch forms are divided into 5 shapes (1, tapered; 2, flattened; 3, rounded; 4, ovoid; and 5, squared). Each shape has 4 sizes ranging from small to large ( Fig 1 ). This wire shape diagram template was used by the 3 examiners in this study.
All examiners selected the ideal wire shape diagram on each plaster model on the mandibular arch according to the guidelines of Trivino et al. Markings made from visual inspection were used to identify the points corresponding to the mandibular midline, the position of the bracket slots on the labial face of the mandibular canines, and the position of the bracket slots or tubes on the labial surface of the mandibular first molars. After calibration, each examiner chose the diagram that best fit the mandibular arch shape on the plaster models of the sample ( Fig 2 ). Two weeks later, all examiners made a new arch form selection on the same plaster models to evaluate the reproducibility of the method.
Samples of all 20 pairs of plaster models were scanned in a plaster model scanner (R700; 3Shape, Copenhagen, Denmark) to create the respective digital models. Each examiner made a digital arch form diagram on the mandibular arch of each digital model using the digital arch form customization tool in the Ortho Analyzer software (version 184.108.40.206, updated October 30, 2015; 3Shape) according to the same references used to define the arch form diagram for the plaster models. Each digital arch form diagram, superimposed onto the mandibular arch, was individually exported as a report generated in PDF format by the software. The arch form figure was cropped from the report using the software program Photoshop CS6 (Adobe Systems, San Jose, Calif). A difference was noticed in magnification between the arch form size in the PDF report and the actual size of the models. On average, the arch sizes of the samples in the reports were 39.52% larger (range, 39.10%-40.22%) than the real dimensions of the digital models. This magnification was corrected in each digital arch form to standardize a real proportion of 1:1 to enable a comparison by superimposition onto the arch forms selected on the plaster models ( Fig 3 ).
The arch form of each digital model created in the Ortho Analyzer software was superimposed onto the respective arch form diagram selected on the plaster model by each examiner in the first set ( Fig 4 ). The best-fit method, selecting the central region as a reference, was used to superimpose both arch forms using the Photoshop software. Differences between the superimposed arch forms were evaluated by splitting the diagrams into 6 segments (molar, premolar, and anterior regions on the left and right sides; Fig 5 ). The wire shape diagram selected for each plaster model was used as the reference. The largest difference between the superimposed arches in each region was calculated using the Photoshop software. An expansion of the customized digital arch form when compared with the wire shape diagram for the plaster model was considered to be a positive value, whereas a contraction of the customized digital arch form was considered to be a negative value.
Statistical analysis was performed with SPSS software for Windows (version 20.0; IBM, Armonk, NY). The agreement between the 2 sets of wire shape diagrams selected on the plaster models by each examiner was evaluated using the kappa statistical test, at the 5% significance level. Kappa values range from −1 to +1, and according to the literature, +1 establishes perfect agreement; from 0.99 to 0.81 is excellent agreement; from 0.80 to 0.61 is good agreement; from 0.60 to 0.41 is regular agreement; from 0.40 to 0.21 is fair agreement; from 0.20 to 0.00 is poor agreement; and <0.00 is no agreement. The interexaminer level of agreement on the first set of wire shape diagrams selected on the plaster models was also tested by the kappa statistical test at a significance level of 5%. Both intraexaminer and interexaminer agreements for each chosen diagram were evaluated according to the individual arch form and considering only the selected shape (1, 2, 3, 4, or 5).
A descriptive analysis was presented to report the means, standard deviations, and minimum and maximum intervals of the superimpositions of the diagrams of each examiner. The largest differences between the superimpositions of the customized digital arch form on the digital models and the selected arch shape diagram for the plaster model in the 6 selected regions were compared among the 3 examiners using the intraclass correlation coefficient and paired t tests to evaluate the accuracy. P values <0.05 were considered statistically significant.
Table I presents the intraexaminer and interexaminer agreements of the selected wire shape diagrams on the plaster models using the kappa statistical test. The diagrams selected were compared both individually and considering only the selected arch shape. In the case of arch shape selection, intraexaminer tests showed perfect agreement for examiner 3, excellent agreement for examiner 2, and good agreement for examiner 1, whereas interexaminer tests showed perfect agreement between examiners 1 and 3, and excellent agreements between examiners 1 and 2 and examiners 2 and 3. In the case of the individual arch shape diagram, all intraexaminer and interexaminer comparisons had good agreement, with the exception of the intraexaminer agreement for examiner 1, which was considered to be regular.
|Parameter||Arch form diagram (considering only shape)||Arch form diagram|
|Examiners 1 × 2||0.846||0.611|
|Examiners 1 × 3||1.000||0.716|
|Examiners 2 × 3||0.846||0.719|
Table II shows the descriptive analysis of the largest differences between the arch form diagrams selected for the plaster models superimposed onto the customized digital arch forms on the digital models. The thickness of the line in both diagrams was 0.50 mm. Differences were calculated in the 6 regions, but the molar region on both sides was further divided into first and second molar regions. The differences were evaluated in 2 rankings according to the clinically perceptible level, since a difference of less than 1 mm is compatible with the accuracy of the human eye. Differences of 0 to 1.00 mm were considered clinically insignificant, and those larger than 1.00 mm were considered clinically significant.
|Parameter||Examiner 1||Examiner 2||Examiner 3|
|Left first molar||0.08||0.60||−1.40||1.30||0.22||0.51||−0.90||1.00||0.17||0.60||−0.90||1.30|
|Left second molar||0.33||1.33||−1.40||3.30||1.06||1.01||−0.70||3.10||0.82||1.02||−1.30||3.00|
|Right first molar||0.79||0.64||−0.70||2.00||0.61||0.60||−0.30||1.90||0.76||0.51||0.00||2.20|
|Right second molar||1.33||1.17||−1.20||3.30||1.32||1.08||−0.30||4.40||1.20||0.67||0.00||2.40|
The largest differences between the diagram superimpositions in the anterior and premolar regions were considered clinically insignificant by all examiners. The largest differences in the right molar region were considered clinically significant by all examiners, whereas those in the left molar region were considered clinically insignificant by examiners 1 and 3, and clinically significant by examiner 2. Considering only the molar regions on the left and right sides, the largest differences in the first molar for both sides were not deemed to be clinically significant by the examiners. However, for the second molar, clinical significance was noted by all examiners on the right side and only by examiner 2 on the left side.
Tables III and IV present the intraclass correlation coefficients and paired t test results, respectively, for the largest differences in the superimpositions of the selected arch shape diagrams for the plaster models and the customized digital arch forms for the digital models according to the different arch regions among the 3 examiners. The results showed a weak correlation in the premolar region and moderate correlations in the anterior and molar regions. Considering only the molar regions on both sides, the second molars had a better correlation compared with the first molars.
|Parameter||ICC||95% CI lower bound||95% CI upper bound|
|Left first molar||0.404||0.128||0.671|
|Left second molar||0.712||0.499||0.862|
|Right first molar||0.366||0.090||0.643|
|Right second molar||0.698||0.479||0.854|