Mixed dentition analysis aims to calculate the difference between the amount of the dental arch space and the amount of tooth material that should be accommodated in perfect alignment. This analysis was performed when the 4 mandibular permanent incisors and permanent first molars were already erupted. Predicting the mesiodistal widths of unerupted permanent canines and premolars is important to determine whether the available space in the arch is greater than, smaller than, or equal to the sum of these unerupted teeth; it is fundamental in determining whether the treatment plan will involve serial extractions, tooth eruption guidance, space maintenance, space gain, or just monitoring the oclusion. Because of the arch length decrease during the dental transition, particularly in the mandible, the mixed dentition is usually analyzed in the mandibular arch.

Tooth size prediction methods can be grouped into 3 categories: those that use linear regression equations (eg, tables of prediction), those based on radiographs, and a combination of both. The first method is the most widely used, especially the tables of Moyers and the equations of Tanaka and Johnston. However, these equations and tables can either overestimate or underestimate the actual widths of the permanent canines and premolars. In these cases, overestimating is better to prevent further lack of space in the arch, and to avoid anterior or posterior crowding.

The first attempts to predict mesiodistal width of unerupted tooth were proposed by Black, who developed tables based on the average sizes of the teeth. Later, periapical radiographs were used, but the images overestimated dental sizes. Now, mixed dentition analysis based on 45° cephalometric radiography and computed tomography scans are the most reliable and accurate methods. Lima and Monnerat used a 45° cephalometric radiograph to predict the widths of unerupted teeth and proposed an index (0.928) to correct the image magnification (7.2%). The methods based on cone-beam computed tomography provide even more reliable and precise information; when compared with conventional tomographies, cone-beam computed tomography uses lower doses of radiation. However, these techniques are also more time-consuming and expensive and require sophisticated equipment. To avoid these complicating factors, statistical correlation methods such as tables and regression equations have been proposed and are most often applied. Melgaço et al postulated that the sum of the mesiodistal widths of the mandibular permanent first molars with the mesiodistal widths of the 4 mandibular permanent incisors is a good predictor for the sum of the mesiodistal widths of permanent canines and premolars. They proposed the following regression equations: Y = 0.975X (male patients), Y = 0.971X (female patients), and Y = 0.973X (both sexes).

In these equations, Y represents the sum of the mesiodistal widths of the permanent canines and premolars in millimeters (both sides), and X is the sum of the mandibular permanent first molars and incisors in millimeters (both sides). In the original study, Melgaço et al found small differences between the predicted and actual values of the sum of the mesiodistal widths of the permanent canines and premolars (0.02 mm, SD of 1.49 mm, for female subjects; 0.04 mm, SD of 1.36 mm, for male subjects; and 0 mm, SD of 1.44 mm, for both sexes).

Dental size can vary considerably between different ethnic groups. Sex variations have also been observed in important studies; some teeth are statistically larger in male patients. Genetics and environment are important determinants also. Therefore, evaluating the consistency of the methods used to predict unerupted dental widths is fundamental for testing their applicability.

In this study, we aimed to test the following null hypothesis: the method of Melgaço et al is consistent for predicting the sum of the mesiodistal widths of the permanent canines and premolars in a different population from the original study. The methods of Moyers and Tanaka and Johnston were also tested.

Material and methods

The sample consisted of the initial plaster casts of patients coming for orthodontic treatment at the dental school of the Federal University of Minas Gerais, Belo Horizonte, in Brazil. The first 200 consecutive study casts that fulfilled all inclusion criteria were selected for the study: permanent dentition in both arches (except third molars), no previous orthodontic treatment, all teeth measured (from left permanent first molar to right permanent first molar) totally erupted, and no interproximal restorations, distortions, fractures, or cavities. No proximal or occlusal attrition, or alterations in tooth shape, size, or number were present. The total sample resulted in 123 white female patients and 77 white male patients.

An analog caliper (Mitutoyo, Tokyo, Japan) with an accuracy of 0.1 mm was used to obtain all measurements. To facilitate access to the interdental spaces, the tips of the caliper were narrowed and positioned to achieve the greatest mesiodistal diameter (contact points of the tooth), parallel to the occlusal surface and perpendicular to the long axes of the teeth, as described in other studies ( Fig ).

Positioning of the caliper for the measurements.

To determine the consistency of measurements, 2 researchers (F.C.B. and V.C.N.) measured 60 teeth separately 3 times, with intervals of 10 days between the measurements (total of 180 measurements). The intraclass correlation coefficient was used to determine the consistency of the measurements. Paired Student t tests with a significance level of 5% were used to compare the differences between the actual and predicted sums of the mesiodistal widths of the permanent canines and premolars determined by the method of Melgaço et al. The actual values were also compared with those predicted by the charts of Moyers at the 75th percentile level and the method proposed by Tanaka and Johnston.