Performance of Cone-beam Computed Tomography (CT) Compared to Multidetector CT for Arthrography of the Wrist, Ankle, Elbow and Knee

ConeBeam scanner (CBCT) is a widely technique used for dental and maxillofacial imaging. Patients can be set up in a sitting or a laying position in the, Newtom 5G CBCT allowing for an analysis of the limbs distal joints. Our hypothesis is that the cone-beam scanner could detect and characterize chondral lesions the cartilage similarly to a multidetector CT (MDCT). 100 patients over 18 years requiring arthroscanner of the wrist, ankle, elbow or knee will be enrolled The primary endpoint based on the ability of cone-beam scanner to detect cartilage lesions. Secondary endpoints will concern the depiction of the joint lesions, quality of the images, tolerability of the exams and inter- intra-operator reproducibility. After arthrography, patients will have a MDCT considered as the reference exam, and a CBCT on a randomized order. Exams will be anonymized and read twice by two trained radiologists with 3-6 week intervals. The inter-observer agreement for the diagnosis of cartilage lesion type between the two techniques will be evaluated by a kappa coefficient. The evaluation of diagnostic performance will be achieved by studying the sensitivity, specificity, negative predictive value and positive predictive value, and its likelihood ratio. Kappa test will be performed to evaluate the consistency of the depiction of the lesions. A Student t test will be performed for paired data if distributions are normal, a Wilcoxon test if not. To assess patient's tolerance, the percentages of (painful, not painful) will be calculated for each technique and will be compared using the Chi2 test (or Fisher's exact test if the chi can not apply).