Modeling by using regression analysis of teleroentgenographic individual indicators used in the method of Charles J. Burstone

Abstract

In spite of the existence of numerous developed methods of cephalometric analysis, which should help to choose the right direction of orthodontic treatment, usually the doctor has to act intuitively, based on his experience, because their development didn’t take into account numerous factors (ethnicity, age, gender, etc.). Improving these techniques, considering the above-mentioned factors, would significantly increase their effectiveness, and hence the quality of providing dental care to the population. The purpose of the work is to construct and analyze a regression model of teleroentgenographic indicators used in the method of C. J. Burstone in young men and women with normal occlusion close to orthognathic bite and harmonic face. Primary side teleroentgenograms of 38 young men (aged 17-21 years) and 55 young women (aged 16-20 years) with normal occlusion close to orthognathic bite and harmonic face, obtained from the Veraviewepocs 3D device, Morita (Japan), taken from the bank data of research center of National Pirogov Memorial Medical University, Vinnytsya. Cephalometric measurements were performed according to the recommendations of C. J. Burstone. All indicators were divided into three groups: 1 - metric characteristics of the skull, which usually do not change during surgical and orthodontic treatment; 2 - indicators of the tooth-jaw system, the definitions of which most often need to be guided by the orthodontic treatment of growing patients and orthodontic surgery, which allows people with already formed bone skeleton to change the width, length, angles and position of the bones of the upper and lower jaws; 3 - indicators that actually characterize the position of each individual tooth relative to each other, to the bony cranial structures and face profile. Regression models of individual teleroentgenographic indicators used in the method of C. J. Burstone, built using the licensed package “Statistica 6.0”. Constructed all 6 reliable models of indicators included in the second group (anterior lower facial height ANS-Gn/Me, maxillary length ANS-PNS, ramus length Ar-Go, mandibular length Go-Pog, anterior upper facial height N-ANS and posterior upper facial height PNS-N) depending on the indicators of the first group (posterior section of cranial base Ar-Pt, anterior skull base length N-CC, angle of the cranial tilt POr-NBa, anterior section of cranial base Pt-N and distance P-PTV); as well as all 7 reliable models of indicators included in the third group (distances 1u-NF, 1l-MP, 6u-NF, 6l-MP and angles OP-HP, Max1-NF/Max1-SpP, Mand1-Mp/Mand1-MeGo) depending on the indicators of the first and second (distance A-B, A-NPog, Gо-CF, Max-Mand, N-A, N-B, N-Pog and Xi-Pm and angles MeGo-NPog, MP-HP, NAPog, N-ANS-Pog, N-CF-A, NPog-POr, POr-CFXi і POr-ANSPNS) groups. It was established that in young men the model of telerentgenographic indices included in the second group depending on the indicators of the first group and included in the third group, depending on the indicators of the first and second groups, have a higher determination coefficient than in young women (R2 from 0.806 to 0.918 in young men and from 0.510 to 0.768 for young women, and from 0.750 to 0.993 for young men and from 0.510 to 0.986 for young women). In the analysis of entering into the regression models of the relevant predictors found that in young men among the teleroentgenographic indicators of the first group included in the models of indicators of the second group most often included - distances P-PTV (33.3%), Pt-N and N-CC (by 25.0%); and in young women– distances N-CC (38.5%) and P-PTV (30.8%). It was also found that among young men among the teleroentgenographic indicators of the first and second groups that were included in the models of the third group of indicators most often included - distance ANS-Gn/Me (12.8%), the magnitude of the angles NAPog, POr-CFXi and POr-ANSPNS (by 9.4%); and in young women– distance ANS-Gn/Me (13.2%), distances А-В and PNS-N and the magnitude of the angle NAPog (by 7.9%).