Systematic evaluation of the features influencing the accuracy of third order measurements

Abstract

Lateral cephalograms and corresponding dental casts were obtained from 39 untreated Caucasians (12 males, 27 females; mean age 19.5 years; standard deviation 3.7 years) with occlusal relationships considered to be 'normal'. The upper (U1) and lower (L1) incisors were assessed with reference to the occlusal plane perpendicular which was established on the lateral radiographs, including third order angles (U1TR, L1TR) which were also derived from direct dental cast measurements (U1TA, L1TA). Both single regression and multiple linear regression analysis with stepwise variable selection were performed using third order measurements on casts as the dependent variable and crown axis (U1C, L1C), root axis (U1R, L1R), tip-apex connecting line (U1E, L1E), and radiographic third order measurements as independent variables. Single regression analysis indicated an overall difference of 0.02 degrees between radiographic third order inclination and cast assessment in the maxilla (mandible: -2.83 degrees). A change of 1 degree in radiographic third order inclination would produce a change of 0.65 degrees for U1TA and 0.86 degrees for L1TA assessments. Third order measurements on dental casts can best be explained by a linear combination of U1TR and U1E (maxilla) and of L1TR and L1C (mandible) measurements. This study demonstrates the functional enmeshment between two different third order assessments and the most common incisor features on lateral radiographs. Both methods of third order evaluation show sufficient reliability and are appropriate for routine orthodontic practice.