INTRODUCTION

Mixed dentition analysis is an important aspect of orthodontic diagnosis and treatment planning. It is a valuable tool in determining whether the treatment plan may involve serial extraction, guidance of eruption, space maintenance, space regaining, or just periodic observation of the patient.

Three basic approaches for prediction of the size of the unerupted permanent teeth during the mixed dentition have been used. Measurement of the size of the unerupted teeth on radiographs, as recommended by Staley et al1 and de Paula et al2; estimation from proportionality tables, as reported by Moyers3 and Tanaka and Johnston4; and a combination of the radiographic and prediction table method, as recommended by Hixon and Oldfather5 and Bishara et al.6

Mixed dentition analysis using Moyers tables and Tanaka and Johnston have several advantages. No radiographs are required, tables can be used for both the maxillary and mandibular arch estimations, and there is a fairly good accuracy despite a tendency to overestimate the size of unerupted teeth.7 The development of these two methods, however, was based on data derived from a population of Northern European descent. Therefore, the accuracy of these prediction methods may be in question when applied to a population of different ethnic origin.

A review of the literature revealed that mixed dentition analyses were varied between different racial and population groups: Ferguson et al8 and Frankel and Benz9 for black Americans; Schirmer and Wiltshire10 for black Africans; Lee-Chan et al11 for Asian-Americans; Bishara et al6 for population samples from Egypt, Mexico, and the USA; Flores-Mir et al12 for Peruvians; Al-Khadra13 for Saudi Arabians; Nourallah et al14 for Syrians; Yuen et al15 for Hong Kong Chinese; Jaroontham and Godfrey16 for Thai population; Legovic et al17 for Croatians; Otuyemi and Noar18 for Nigerian; and Diagne et al19 for Senegalese population. Populations of different racial origins generally had average values that were significantly different from those reported for whites, but in most cases the clinical significance was questionable.20

Several recent studies investigating mixed dentition analyses in different populations derived their samples from subjects attending orthodontic clinics; this may have introduced a source of bias because they may not be representative of the total population.1011131417 In addition, there was a wide variation in the sample sizes within these studies ranging from 34 to 600 subjects.

There have also been questions about applying these methods, which are based on pooled male and female data, rather than considering the sexes separately. In addition, there is some evidence of secular trends of changing dimension of teeth, which may require progressive modification of mixed dentition analysis for different populations.20

There are few published data on mixed dentition analysis for Arabic populations.1314 The purposes of this study were to: (1) examine the applicability of the Tanaka and Johnston method of prediction in a Jordanian population; and (2) develop a new prediction method for this specific population if it proved to be necessary.

MATERIALS AND METHODS

RESULTS

Table 1 shows the descriptive statistics for the sums of the mesiodistal widths of the four mandibular incisors and the average of the right and left maxillary and mandibular canine and premolar segments divided according to gender. A gender discrepancy was seen in this study with boys showing significantly greater size of mandibular incisors and maxillary and mandibular canine and premolar segments. Statistical analysis of measurements at an individual tooth level revealed that there were significant differences between genders for all teeth measured (P < .05) except for the mandibular right central incisor and the maxillary right and left second premolars. There were no significant differences between measurements of contralateral teeth (P < .05) except for the maxillary and mandibular second premolars.

Table 1.  Descriptive Statistics for Summations of Mesiodistal Widths of Mandibular Incisors, and Maxillary and Mandibular Canine and Premolar Segments^a

View Fullscreen

Table 2 shows the regression parameters for prediction of summations of mesiodistal widths of the canine and premolars (dependent variable) using the mesiodistal widths of the mandibular incisors (independent variable). The correlation coefficients ranged from 0.57 to 0.68 with coefficients higher in female subjects. The r² values ranged from 33% to 47% with the power of the regression model greater in female subjects.

Table 2.  Regression Parameters for Prediction of Summations of Mesiodistal Widths of the Canine and Two Premolars in One Quadrant ^a

View Fullscreen

Table 3 shows the descriptive statistics for the prediction of mesiodistal widths of the canine and two premolars using regression equations derived from the present study. Standard deviation and standard errors were relatively low indicating a reduced variance about the mean for the predictions.

Table 3.  Descriptive Statistics for the Prediction of Mesiodistal Widths of the Canine and Two Premolars Using Regression Equa tions Derived From the Present Study^a

View Fullscreen

The equations of Tanaka and Johnston4 were applied to the present sample to allow comparison between the predicted mesiodistal widths of maxillary and mandibular buccal segments and the actual measured widths. Statistical comparisons using t-tests showed significant differences between predicted and measured widths of maxillary and mandibular buccal segments for boys and girls and the pooled total sample (P < .05). Figure 1 illustrates an example of the highly significant differences between the predicted mesiodistal widths of the mandibular buccal segments using Tanaka and Johnston's equation and actual measured widths.

View Figure

• Download as Powerpoint
• Download Figure

Table 4 shows the difference between mesiodistal widths predicted using regression equations derived from the present study and those derived by Tanaka and Johnston. Mean differences ranged from 0.5 to 1.3 mm. A paired sample t-test showed significant differences between the two prediction methods for boys, girls, and the pooled total sample (P < .05)

Table 4.  A Comparison Between Predicted Mesiodistal Widths of the Canine and Two Premolars Using Regression Equations From the Present Study and Tanaka and Johnston's Equations

View Fullscreen

DISCUSSION

There have been a few studies investigating a mixed dentition analysis in school-aged children.121516 The age of the sample was relatively young in order to eliminate and minimize the influence of tooth wear and loss. The study sample in this investigation was a random selection of 367 Jordanian 10th grade school children. The sample represents 0.1% of the 10th grade subjects in the Amman governance. The sample size in the present study is one of the largest assessing mixed dentition analyses (Table 5). This was evident from the power calculation performed.

Table 5.  Summation of Mesiodistal Widths of Mandibular Incisors and Maxillary and Mandibular Canine Premolar Segments from Various Studies^a

View Fullscreen

A gender discrepancy was seen in this study, with boys showing significantly greater size of mandibular incisors, maxillary and mandibular canines, and premolar segments. This sexual dimorphism has been seen in other studies.15–1719 Division of subjects according to sex when performing mixed dentition analysis was therefore necessary.

Table 5 shows mesiodistal crown dimensions for male and female subjects published by various authors. The combined mesiodistal diameters of the canine-premolar segments and the mandibular incisors were relatively smaller than those of black South Africans,10 but comparable to those of black Americans,89 Hong Kong Chinese,15 Thai,16 and Senegalese groups.19

Findings from the present study showed that Tanaka and Johnston's method of prediction overestimates the mesiodistal widths of both maxillary and mandibular buccal segments compared with actual measurement on study casts (P < .05). This finding was similar to that of Asian Americans,11 Senegalese,19 black South Africans,10 and Saudi Arabians.13 Buwembo and Luboga25 conducted a meta-analysis on the applicability of Moyer's method in different ethnic groups and concluded that it cannot be universally applied to different populations and it is safer to develop prediction tables for specific populations.

There have been no studies in the literature that attempted to investigate the error of tooth size prediction and its clinical significance. The threshold for clinical significance, however, was investigated by few studies when estimating the Bolton discrepancy.26–28 Othman and Harradine28 recommended a threshold of 2 mm in expressing a tooth size discrepancy. Bernabè et al,27 however, chose 1.5 mm as a threshold based on Proffit7 who stated that tooth size discrepancies less than this amount are rarely significant. But there was no mention of any investigation to support this. The amount of the threshold for clinical significance in the prediction of tooth size in the mixed dentition needs further investigations.

CONCLUSIONS

• There are limitations in the application of Tanaka and Johnston's prediction method to a Jordanian population.

• A gender discrepancy was seen in the present study with male subjects having significantly wider mandibular incisors and maxillary and mandibular canine and premolar segments.

• Jordanian subjects should be divided according to gender prior to carrying out a mixed dentition analysis and the corresponding equation applied.