Accuracy of unenhanced, non-sedated MRI in the diagnosis of acute appendicitis in children

Background Children with suspected appendicitis generally undergo an ultrasound (most commonly) or a CT scan (rarely) as the first imaging study. At our hospital, patients with non-diagnostic ultrasound or CT scan results undergo a non-contrast (unenhanced), non-sedated MRI. We aimed to determine the accuracy of this study for this purpose in a large cohort of children. Methods A retrospective review of all unenhanced, non-sedated MRIs done for suspected appendicitis was performed from January 2014 to December 2018. MRI reports were correlated with pathology reports in cases that underwent appendectomy, and with clinical outcomes if no operation was done (clinical follow up: 30d). No patient was treated for appendicitis non-operatively. Results Three hundred fifty unenhanced, non-sedated MRIs were done and reviewed with median age: 12 (3 to 18) years. Sixty-five (18.6%) MRIs were positive for appendicitis, and 62 of those underwent appendectomy (3 excluded clinically). Pathology was positive in 59/62 cases. 256 (73.1%) MRIs were negative for appendicitis. Six cases underwent appendectomy (persistent symptoms). Pathology was positive in 2/6 cases. The overall diagnostic accuracy was: sensitivity 96.7% (95% CI: 88.6–99.6), specificity 97.7% (95% CI: 95.0–99.1), PPV: 90.8% (95% CI: 81.6–95.6; false positives 6/65), and NPV: 99.2% (95% CI: 97.0–99.8; false negatives 2/254). Twenty-nine (8.3%) MRIs were non-diagnostic. None of those 29 cases had appendicitis (4 negative pathology, 25 excluded clinically). Conclusions The unenhanced, non-sedated MRI is highly accurate for the diagnosis of appendicitis in children. It should be considered as an alternative to CT in the work-up of patients with suspected appendicitis to eliminate the risks associated with ionizing radiation. Level of Evidence Level IV.