Early detection and quantitative assessment of apical root resorption using crown-root ratio and tuned-aperture computed tomography (TACT)

Abstract

EARLY DETECTION AND QUANTITATIVE ASSESSMENT OF APICAL ROOT RESORPTION USING CROWNROOT RATIO AND TUNED-APERTURE COMPUTED TOMOGRAPHY (TACT). Andre Mol, Julie H. Mol, Onanong Chai-u-dom, and Donald A. Tyndall, University of North Carolina, Chapel Hill, NC, and Chulalongkorn University, Bangkok, Thailand Background. Apical root resorption (ARR) is a complication of orthodontic treatment without clear causative factors. The relationship between causative factors and ARR may be obscured by inaccuracies in current methods of detection and quantification. Twodimensional imaging, lack of image standardization, as well as orthodontic tooth movement and remodeling of the surrounding tissues all pose considerable challenges. Objective. The purpose of this study was to develop and test a tuned-aperture computed tomography (TACT) application and to assess the validity of the crown-root ratio (CRR) method for the detection and quantification of small oblique apical root resorption lesions. Methods and materials. A registered (R) and a nonregistered (NR) version of TACT were tested. In TACT-R, the detector moved with the tooth through mechanical standardization. TACT-NR allowed the tooth to move independently of the detector. Ten basis images with a maximum angular disparity of 20° were used. Oblique slicing of TACT-NR follow-up stacks was required for registration with baseline stacks. A slice shifting algorithm was applied to prevent drifting of cross-talk shadows. Twenty extracted maxillary incisors placed in a widened alveolus of a maxillary jaw segment served as test and control sites. Baseline images were acquired without lesions. Follow-up images were acquired with and without lesions. Lesions were created by cutting an oblique buccolingual segment of the root just below the apex. Teeth were moved between baseline and follow-up (rotation and torque), and a bone-putty mixture was used to simulate remodeling of the surrounding bone. CRR, TACT-R, and TACT-NR were tested by 6 radiologists using ROC-analysis. CRR and TACT-R were also tested for their quantitative ability. Results. Low Az-values were obtained for CRR (0.68; SD .09), TACT-R (0.64; SD .09), and TACT-NR (0.60; SD .09). Differences between modalities were not statistically significant (ANOVA: P .05). Lesion quantification was poor for both CRR (R .03) and TACT-R (R .13). Conclusions. Crown-root ratio and TACT are not valid for the detection and quantification of small oblique apical root resorption lesions. Imaging techniques with greater angular disparity between basis images reduce cross-talk effects and are likely to improve performance.