Magnetic resonance imaging in pediatric appendicitis: a systematic review

Magnetic resonance imaging for the evaluation of appendicitis in children has rapidly increased recently. This change has been primarily driven by the desire to avoid CT radiation dose. This meta-analysis reviews the diagnostic performance of MRI for pediatric appendicitis and discusses current knowledge of cost-effectiveness. We used a conservative Haldane correction statistical method and found pooled diagnostic parameters including a sensitivity of 96.5% (95% confidence interval [CI]: 94.3-97.8%), specificity of 96.1% (95% CI: 93.5-97.7%), positive predictive value of 92.0% (95% CI: 89.3-94.0%) and negative predictive value of 98.3% (95% CI: 97.3-99.0%), based on 11 studies. Assessment of patient outcomes associated with MRI use at two institutions indicates that time to antibiotics was 4.7 h and 8.2 h, time to appendectomy was 9.1 h and 13.9 h, and negative appendectomy rate was 3.1% and 1.4%, respectively. Alternative diagnoses were present in ~20% of cases, most commonly adnexal cysts and enteritis/colitis. Regarding technique, half-acquisition single-shot fast spin-echo (SSFSE) pulse sequences are crucial. While gadolinium-enhanced T1-weighted pulse sequences might be helpful, any benefit beyond non-contrast MRI has not been confirmed. Balanced steady-state free precession (SSFP) sequences are generally noncontributory. Protocols do not need to exceed five sequences; four-sequence protocols are commonly utilized. Sedation generally is not indicated; patients younger than 5 years might be attempted based on the child's ability to cooperate. A comprehensive pediatric cost-effectiveness analysis that includes both direct and indirect costs is needed.