Background: The present study was undertaken to evaluate the accuracy of methods proposed by Tanaka and Johnston, Moyers, and Bernabé and Flores-Mir to forecast the mesiodistal dimensions of permanent canine and premolars in a North Indian population.
Methods: Mesiodistal tooth dimensions were measured from study models representing 68 male and 68 female subjects (aged 12-16 years) of North Indian descent. The mesiodistal dimensions of the teeth were measured using a digital Vernier caliper (providing measurements to 0.01 mm accuracy). The measurements of canine and premolars were summed up and compared with those derived from Tanaka and Johnston equations, Moyers probability tables (75 th percentile), and Bernabé and Flores-Mir equations.
Results: All the three methods exhibited overestimation of actual sum of permanent canine and premolars in both the arches and genders in this population. Because of the discrepancy observed, new regression analyses in the form of Y = A + B (X) were formulated similar to those proposed by Tanaka and Johnston originally, but separately for males and females. In males, the correlation coefficient between the sum of mesiodistal widths of mandibular incisors and the sum of widths of canine and premolars was 0.637 and 0.685 for maxilla and mandible, respectively. This value was 0.55 for maxilla and 0.64 for mandible in females. These values were found to be statistically significant.
Conclusion: All the methods evaluated in the study were found to be inaccurate and they overestimate the mesiodistal widths of permanent canine and premolars in North Indian population. Revised linear regression equations were derived using the data obtained from this population.