Diagnostic performance of magnetic resonance enterography and ultrasound in children with inflammatory bowel diseases: a diagnostic test accuracy meta-analysis.

ABSTRACT

OBJECTIVES: Magnetic resonance enterography (MRE) and ultrasound (US) can be used to diagnose inflammatory bowel diseases (IBD) in children. This meta-analysis aimed to determine the diagnostic performance of MRE and US in pediatric patients with IBD. METHODS: PubMed, Embase, and the Cochrane Library were searched for eligible studies published up to June 1, 2020. The outcomes were the performances of MRE and US at the segment and patient levels. Pooled sensitivity, specificity, positive and negative likelihood ratios (PLR and NLR), diagnostic odds ratio (DOR), and the area under the summary receiver operating characteristic curves value (SROC) were analyzed. RESULTS: Eight studies (340 children) were included. Compared with the reference standard, MRE showed pooled sensitivity of 93.0% (95% confidence interval (CI) 90.0-95.4%), specificity of 94.6% (95% CI 92.1-96.5%), PLR of 11.146 (95% CI 5.027-24.713), NLR of 0.094 (95% CI 0.057-0.155), and DOR of 134.21 (95% CI 40.72-442.29), with a SROC of 0.9721. Similar results were observed at the patient and segment levels. Compared with the reference standard, US had pooled sensitivity of 84.1% (95% CI 69.9-93.4%), specificity of 82.9% (95% CI 66.4-93.4%), PLR of 4.924 (95% CI 2.351-10.310), NLR of 0.207 (95% CI 0.103-0.413), and DOR of 25.919 (95% CI 7.63-88.07), but only two studies were included. US (reader 1) had a similar diagnostic value to US (reader 2). CONCLUSIONS: The present meta-analysis shows that MRE has good performance in detecting IBD in pediatric patients. Only two studies used US, and additional studies are necessary to confirm the diagnostic performance of US for IBD in children. KEY POINTS • MRE has good performance in the detection of IBD in pediatric patients. • Similar results were observed at the patient and segment levels for MRE. • Only two studies were included for US, without differentiating patient/segment.