Prediction of the combined right and left canine and premolar widths in both arches of the mixed dentition.

Presently available methods used to predict the mesiodistal widths of unerupted canines and premolars in the mixed dentition provide an estimate for the widths of one canine and two premolars which must be either doubled for the prediction of the six tooth widths on both sides of the arch, or which must be computed separately for each side of the arch. The purpose of this study was to develop, with multiple regression analysis, equations which would predict the sum of the mesiodistal widths of the right and left canines and premolars in each arch. Measurements were taken from plaster casts and periapical radiographs of the upper arch of 92 subjects (46 females and 46 males), and of the lower arch of 83 subjects (41 females and 42 males). Prediction equations were developed for sexes combined and for each sex. All equations performed satisfactorily when tested on data taken from orthodontic patients (upper arches of 43 patients and lower arches of 53 patients). Prediction equations developed from the sexes combined are recommended for clinical use. Pediction of the mesiodistal widths of unerupted canines and premolars in mixed dentition patients is important in the diagnosis of arch length deficiency problems. Prediction methods have been developed with simple regression analysis techniques, 1-5 multiple regression analysis, ~-8 and other approaches. 94° Presently available methods used to predict the mesiodistal widths of unerupted canines and premolars in the mixed dentition provide an estimate for the widths of one canine and two premolars which must be either doubled for the prediction of the six tooth widths on both sides of the arch, or which must be computed separately for each side of the arch. The purpose of this study was to develop, with multiple regression analysis, equations which would predict the sum of the mesiodistal widths of the right and left canines and premolars in each arch. Methods and Materials A sample of normal subjects of northwest European ancestry was selected from the longitudinal records of the Iowa Facial Growth Study’~ conducted between 1946 and 1960. Measurements were taken in the upper arches of 92 subjects, 46 females and 46 males, and in the lower arches of 83 subjects, 41 females and 42 males. A sample of Caucasian orthodontic patients, treated in the Department of Orthodontics at The University of Iowa between 1961 and 1974, was collected for the purpose of cross validating the prediction equations developed from data taken from the Growth Study subjects. In the cross validation sample, measurements were taken in the upper arches of 43 patients, 27 females and 16 males, and in the lower arches of 53 subjects, 30 females and 23 males. Measurements were taken with dial calipers a which measured distances to the nearest 0.05 mm. The following permanent teeth were measured on both sides of the arch on plaster casts made from alginate impressions: maxillary central incisors, canines, premolars, and first molars; mandibular incisors, canines, premolars, and first molars. Measurements were made as described by Seipe.1.12 Measurements were not made when a tooth had proximal surface restorations, was distorted, not fully erupted, or otherwise unsuitable for measurement. The mesiodistal crown widths of the permanent maxillary and mandibular canines, premolars, and molars were measured on periapical radiographs taken with a 16"-long cone paralleling or right-angle technique. Measurements were taken of teeth which were clearly visible, without distortion, and not rotated. One investigator, R.S., took two measurements of each dimension measured in the X3rowth Study subjects. The same investigator took one’ measurement of the dimensions measured in the orthodontic patient sample. Another investigator, J.H. took two measurements of each a Helios dial calipers, Fred V. Fowler Company, 315 Auburn St., Auburndale, Mass. 02166. PEDIATRIC DENTISTRY: Volume 5, Number 1 57 width measured in the maxillary arch of the Growth Study subjects. A third investigator, T.S., took two measurements of each width measured in the mandibular arch of the Growth Study subjects. The intraexaminer correlation coefficient for the measurements ranged from r = 0.94 to r = 0.99, and the intraexaminer correlation coefficients ranged from r = 0.97 to r = 0.99. The means of the measurements taken by R.S. were used in all further computations. The mean ages of the subjects for the radiograph measurements were for the maxilla: 8.8 years (Growth Study) and 9.9 years (patient sample); and for the mandible: 9.1 years (Growth Study) and 9.6 years (patient sample). The mean ages for the cast measurements were 13.8 years for both samples in the maxilla, and in the mandible, 13.1 years (Growth Study) and 13.5 years (patient sample). All computations were performed by a computer b located at the University of Iowa Computer Center. Ferguson 13 has described the basic aspects of multiple correlation. The universal dental numbering system was used to identify the measured teeth in this article (Figure 1). x preceding a tooth number denotes the mesiodistal width of that tooth as measured on a periapical radiograph. A tooth number standing alone denotes the width of the tooth as measured on a plaster cast.