Reliability and agreement of root length measurements during orthodontic treatment in images from different CBCT machines using multiplanar reconstruction

Abstract

Objectives: To assess inter- and intrarater reliability and agreement for measurements of root lengths using multiplanar reconstruction (MPR) in cone beam computed tomography (CBCT) examinations. Furthermore, to determine whether using MPR from different CBCT machines was a reliable and reproducible method for assessment of root length during orthodontic treatment of adolescents.  Materials and methods: A total of 40 CBCT examinations obtained before, during and after orthodontic treatment of 14 adolescents, with fixed appliances from a multicentre randomised controlled trial, were used. All roots from the incisors to the first molars were measured by two independent raters and in accordance with a protocol preceded by a multi-step calibration. Reliability was assessed by intra class correlation (ICC). Agreement was assessed by measurement error according to the Dahlberg formula and Bland–Altman plot. Results: The number of repeated measurements varied from 436 to 474 for the different timepoints. Good to excellent inter- and intrarater reliability for different tooth groups and timepoints were shown. Measurement error for inter- and intrarater agreement varied between 0.41 mm and 0.77 mm. The Bland–Altman plot with 95% limits of agreement varied between +1.43 mm and −2.01 mm for different tooth groups and timepoints.  Conclusions: The results of this study indicate that CBCT using MPR from different machines is a reproducible method for measuring root length during different phases of orthodontic treatment. When interpreting root shortening measurements in CBCT using MPR for clinical or research purposes, values below 2 mm should be approached with caution, as they may contain measurement errors.