MRI for pediatric appendicitis in an adult-focused general hospital: A clinical effectiveness study-challenges and lessons learned

OBJECTIVE. The objective of our study was to determine the feasibility and accuracy of MRI for pediatric appendicitis in an adult-predominant general hospital setting where non- pediatric-trained radiologists routinely interpret the studies. MATERIALS AND METHODS. MRI was performed in pediatric patients with equivocal ultrasound (US) findings and persistent clinical concern for appendicitis. Neither IV contrast material nor a sedative was administered. Our MRI protocol evolved early during the study period, quickly settling on three sequences (total scanning time, 11 minutes). The clinical reference standard for statistical analysis was appendicitis diagnosed on operative or pathology report; 95% Clopper-Pearson CIs were calculated. RESULTS. Between 2012 and 2016, 528 pediatric patients (mean age, 9.9 years; age range, 1-17 years) underwent MRI after US evaluation yielded equivocal findings: 10.4% (55/528) of patients were found to have surgically proven or pathologically proven appendicitis. Sensitivity and specificity of MRI for appendicitis were 96.4% and 98.9%, respectively. Positive and negative predictive values were 91.2% and 99.6%. A normal appendix and abnormalities not involving the appendix were identified on MRI in 21.6% (114/528) of patients. CONCLUSION. Our data show that unenhanced MRI for suspected appendicitis in pediatric patients is clinically effective when performed in a nonpediatric hospital setting with nonpediatric radiologists, emergency physicians, and surgeons.