A network meta-analysis of capsule endoscopy versus imaging modalities for diagnosing small bowel Crohn's disease.

INTRODUCTION

Clinicians have several imaging options to evaluate suspected or confirmed small bowel Crohn's disease (SB-CD), including computed tomography enterography (CTE), magnetic resonance enterography (MRE), intestinal ultrasound (IUS), and video capsule endoscopy (VCE).

METHODS

Direct head-to-head comparative analysis and network meta-analysis were conducted on all available modalities using a random effects model. Furthermore, each modality was ranked using the surface under the cumulative ranking curve approach (P-score).

RESULTS

The present review included 29 studies with a total population of 2609 individuals. The pooled sensitivity, specificity, and diagnostic accuracy for the detection of SB-CD were 89.6%, 86.2%, and 71.5% for VCE; 82%, 71.6%, and 67.9% for MRE; 79.6%, 82.7%, and 52.3% for CTE; and 89.3%, 72%, and 71% for IUS, respectively. The network meta-analysis found that VCE exhibited superior diagnostic accuracy compared to CTE and MRE, while demonstrating comparable performance between VCE and IUS, as well as among MRE, CTE, and IUS. Further, the ranking analysis positioned VCE (P-score .97) as the most effective diagnostic modality for SB-CD, followed by IUS, MRE, and CTE. Subgroup analysis showed that VCE had significantly better diagnostic accuracy than the other modalities for detecting proximal SB-CD. Regarding adverse events, VCE was associated with capsule retention in 3.3% of the cases in these studies.

CONCLUSIONS

VCE exhibited superior diagnostic performance for detecting established proximal SB-CD compared to other imaging modalities. Clinicians should weigh the benefits and risks, and incorporate other modalities, such as MRE and IUS to optimize diagnosis and management.