Michael D. Repplinger MD, Presenter: Nothing to Disclose

Perry J. Pickhardt MD, Abstract Co-Author: Co-founder, VirtuoCTC, LLC Stockholder, Cellectar Biosciences, Inc

Douglas Robert Kitchin MD, Abstract Co-Author: Nothing to Disclose

Jessica B. Robbins MD, Abstract Co-Author: Nothing to Disclose

Timothy J. Ziemlewicz MD, Abstract Co-Author: Nothing to Disclose

Scott Brian Reeder MD, PhD, Abstract Co-Author: Institutional research support, General Electric Company Institutional research support, Bracco Group

To determine the accuracy of an MRI protocol (with and without contrast plus DWI) when compared with a CE-CT protocol for the detection of acute appendicitis.

This is a HIPAA-compliant, IRB-approved prospective study of patients presenting to the emergency department with abdominal pain. Patients were eligible for enrollment if they were over 11 years old and had a CT ordered to evaluate for appendicitis. After consent was obtained, patients underwent CT and MR imaging in tandem. Three attending radiologists interpreted all MR and CT images independently. Image sets were de-identified. Multiple parameters were documented for each image set including characteristics of the appendix (size, location, etc), the likelihood of appendicitis, possible alternative diagnoses, and the time required to interpret the images. Follow-up consisted of a chart review for pathological/surgical findings or follow-up phone interview/chart review. Continuous variables were summarized with descriptive statistics using means and 95% confidence intervals. Receiver operating characteristic (ROC) curves for the likelihood of appendicitis were drawn. Pair-wise comparisons of AUCs were obtained. Cohen's kappa with quadratic weights was used to assess inter-reader agreement.

We enrolled 93 patients from 2/2012-7/2013, including 60 women (64.5%), with a mean age of 33.3 years (30.5, 36.2). The incidence of appendicitis was 37.6%. Sensitivity and specificity were 0.94 (0.79, 0.99) and 1 (0.91, 1) for unenhanced MRI/DWI, 0.94 (0.79, 0.99) and 0.92 (0.91, 0.98) for CE-MRI, and 1 (0.88, 1) and 0.98 (0.89, 1) for CT. The ROC curves had AUCs of 0.868 (0.784, 0.953), 0.885 (0.814, 0.956), and 0.903 (0.832, 0.973) for unenhanced MRI/DWI; 0.864 (0.782, 0.947), 0.867 (0.795, 0.938) and 0.9 (0.823, 0.976) for CE-MRI; and 0.947 (0.899, 0.996), 0.959 (0.915, 1), and 0.961 (0.915, 1) for CT. The mean time to read the MR images was 4.45 minutes (4.23, 4.67) compared with 2.04 minutes (1.91, 2.17) for CT. Kappa values were 0.643-0.805 for unenhanced MRI/DWI, 0.722-0.778 for CE-MRI, and 0.769-0.976 for CT.

The accuracy of this MRI protocol approached that of CT for the diagnosis of appendicitis, with substantial inter-rater agreement.

MRI may be a suitable first-line imaging test to diagnose appendicitis in the general population.

Repplinger, M, Pickhardt, P, Kitchin, D, Robbins, J, Ziemlewicz, T, Reeder, S, Direct Comparison of Contrast-Enhanced MRI with Contrast-Enhanced CT to Diagnose Appendicitis.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14016168.html