Evaluation of 3T Dedicated Dental Magnetic Resonance Imaging: Are Endodontic Measurements Reliable and Accurate? An Ex Vivo Feasibility Study

Objectives Current 3T dedicated dental magnetic resonance imaging (3TddMRI) techniques cannot compete with cone-beam computed tomography (CBCT) for endodontic imaging because of limitations in spatial resolution and/or long examination times. The aim of this ex vivo study was to evaluate the increase in diagnostic performance of 3TddMRI using novel intraoral coils and optimised protocols for high-resolution endodontic measurements under clinical conditions. Methods In a realistic human tooth model (3 canines and 3 molars), 3TddMRI performance with intraoral coils was evaluated for 5 different sequences (true fast imaging with steady-state precession [trueFISP], fast low-angle shot [FLASH], double echo steady state [DESS], constructive interference in steady state [CISS], and research prototype ultrashort time of echo [UTE]) for established endodontic parameters (working length, pulp chamber dimensions, apical root canal diameter, and apical dentine thickness) and compared with CBCT as the reference standard. Using intraoral coils, 3TddMRI data with an isotropic resolution of 0.21 mm 3 could be acquired in 6 to 8 minutes. We assessed reliability using intraclass correlation coefficients (ICCs) and accuracy using a linear mixed-effects model. Results Intra- and interrater reliability was excellent for all MRI sequences (ICC ≥0.98). In general, 3TddMRI revealed a slight overestimation of measurements compared with CBCT, with a mean signed difference of –0.11 mm (95% limits of agreement, 0.85-0.65; CBCT-MRI). Accuracy did not differ by measurement type, nor was one sequence superior to all the others. However, CISS/UTE sequences performed significantly worse in 1 measurement category each versus all other sequences ( P value range: <.001-.03). Conclusion A novel 3TddMRI setup with an intraoral coil and endodontically optimised high-resolution 3-dimensional sequences (trueFISP, DESS, or FLASH) enables accurate endodontic measurements comparable to CBCT within clinically feasible acquisition times.