Multidetector CT and cone-beam CT have substantial agreement in detecting dental and sinus abnormalities in equine cadaver heads

The performance of cone-beam computed tomography (CBCT) systems compared to conventional helical multidetector CT (MDCT) imaging of the equine head is unknown. The aim of this study was to compare the ability of CBCT and MDCT to detect abnormalities in equine cadaver heads. In a prospective, method-comparison study design, eleven equine cadaver heads were scanned using a CBCT and also a 64-slice MDCT. Consensus evaluations for CBCT and MDCT scans were performed by three observers. Identified abnormalities were grouped into subcategories with a focus on dental abnormalities. Kappa agreement between detected abnormalities with CBCT and MDCT were calculated. Of 468 teeth evaluated, 122 (26.1%) were found to have abnormalities (including in 58 infundibula and 7 pulps) using MDCT and 105 (22.4%) were found to have abnormalities (including in 52 infundibula and 2 pulps) using CBCT. The agreement between CBCT and MDCT was almost perfect for overall detection of dental abnormalities (k=0.90) with k=1 for diastema k=0.95 for clinical crown abnormalities, and k=0.93 for infundibular abnormalities. However, the detection of pulp changes by CBCT was only moderate k=0.44. Increased scatter radiation, non-calibrated Hounsfield Unit and artefacts in CBCT images made accurate identification of the pulp density difficult. In conclusion, CBCT results were similar to conventional MDCT for the majority of dental abnormalities, however, pulp abnormalities were not reliably identified using CBCT, potentially limiting its clinical use for detecting endodontic disease in its current form. Further comparison with more cases with confirmed dental disease and studies in clinical cases are warranted.