498 - Properties of Ultrasound-Rapid MRI Clinical Diagnostic Pathway in Suspected Pediatric Appendicitis

Poster Number: 498
Publication Number: 498.311

Suzanne Schuh, The Hospital for Sick Children, Toronto, ON, Canada; Carina Man, The Hospital for Sick Children, Toronto, ON, Canada; Eman Marie, McMaster University Michael G. DeGroote School of Medicine, Ancaster, ON, Canada; Ghufran Alhashmi, King Abdulaziz University, Jeddah, Makkah, Saudi Arabia; Dan Halevy, The Hospital for Sick Children, Toronto, ON, Canada; Dana Singer-Harel, The Hospital for Sick Children, Toronto, ON, Canada; Aya Finkelstein, The Hospital for Sick Children, Toronto, ON, Canada; Judith M. Sweeney, The Hospital for Sick Children, Toronto, ON, Canada; Andrea S.. Doria, The Hospital for Sick Children, Toronto, ON, Canada

Background: Diagnosis of pediatric appendicitis remains challenging because no clinical feature, laboratory test or clinical score eliminate the need for imaging. Because screening ultrasound (US) frequently fails to visualize the appendix, the American College of Radiology recommends CT in children with non-diagnostic USs, with MRI as an option. However, there is paucity of information about the value of MRI for assessment of suspected appendicitis within the constraints of limited access to non-emergent MRI in tertiary-care pediatric centers.

Objective: To determine the clinical diagnostic accuracy of the screening US-ultra-fast MRI pathway in previously healthy children 4-17 years old with persistent clinical concern about appendicitis after a non-diagnostic screening US or with a discrepancy between a conclusive screening US and clinical re-examination in the emergency department (ED). Hypothesis: the US-MRI pathway would have a 95% probability of being at least as accurate as the previously evaluated serial US pathway (94% accurate), with 97% accuracy and 95% CI 94%-100%.

Design/Methods: Prospective tertiary-care pediatric ED cohort study of 624 previously healthy children 4-17 years old undergoing US for suspected appendicitis. Children with non-diagnostic USs and persistent appendicitis concern/conclusive US-reassessment discrepancies underwent ultra-rapid MRI (US-MRI pathway), interpreted as positive, negative or non-diagnostic. Cases with missed appendicitis, negative operations, and CT use were considered inaccurate. Primary outcome was the proportion of accurate diagnoses of appendicitis/lack thereof by the pathway.

Results: Of the 624 study children 150(24%) had appendicitis. 255 USs (40.9%) were non-diagnostic,169/255 (66%) had persistent concern about appendicitis and 117/169 underwent MRI. Of 139 US-MRI pathway children (after 117 non-diagnostic and 22 conclusive USs), 137 [98.6%; 95% CI 0.96-1.00] had clinically accurate outcomes (1 CT, 1 negative appendectomy): sensitivity 18/18 [100%], specificity 119/121 [98.3%], positive predictive value 18/20 [90.0%], negative predictive value 119/119 [100%]. MRI imaging accuracy was 134/139 (96.4%); 3 MRIs (1.4%) non-diagnostic (no appendicitis), 2 false-positive.

In the overall algorithm, 616/624 [98.7% (0.97-0.99)] patients had accurate outcomes: 147/150 (98.0%) appendicitis cases had confirmatory surgeries (3 CTs) and 469/474 (98.9%) appendicitis-negative children had no surgery/CT.

Conclusion(s): This study demonstrated good feasibility and high clinical accuracy of the US-rapid-MRI pathway in suspected pediatric appendicitis after non-diagnostic US.