Impact of gadolinium-based contrast agent in the assessment of Crohn's disease activity: Is contrast agent injection necessary?

Purpose To determine whether magnetic resonance enterography (MRE) performed without intravenous contrast injection is diagnostically noninferior to conventional contrast-enhanced MRE (CE-MRE) in patients with Crohn's disease (CD). Materials and Methods This was an Institutional Review Board (IRB)-approved retrospective study. Ninety-six patients (52 male and 44 female; 47.18 years ± 13.6) with a diagnosis of CD underwent MRE at 1.5T including T₂-weighted single-shot turbo-spin-echo, T₂-weighted spectral fat presaturation with inversion recovery (SPAIR), T₁-weighted balanced fast-field-echo MR sequences, and CE-MRE consisting in T₁-weighted breath-hold THRIVE 3D MRI sequences after administration of gadobenate dimeglumine (0.2 mL/kg of body weight). Unenhanced MRE, CE-MRE, and unenhanced MRE plus CE-MRE were reviewed in separate sessions with blinding by two readers in consensus, and subsequently by two other readers independently considering a subgroup of 20 patients. Crohn's Disease Endoscopic Index of Severity (CDEIS) and/or histologic analysis of the surgical specimen were considered as reference standards for the assessment of inflammatory activity. Results Patients revealed prevalently active (n = 55 patients) or quiescent CD (n = 41 patients). The agreement between unenhanced MRE vs. CE-MRE in interpreting active bowel inflammation was 96% (123/128 bowel segments; one-sided 95% confidence interval [CI], >94.4%). Unenhanced MRE vs. CE-MRE vs. unenhanced MRE plus CE-MRE revealed a diagnostic accuracy of 93% [90/96] vs. 92% [88/96] vs. 97% [93/96] (P > 0.05) in the diagnosis of active CD. Interreader agreement was very good for all variables (κ value = 0.8-0.9) except for the measurement of the length of disease (κ value = 0.45). Conclusion Unenhanced MRE was noninferior to CE-MRE in diagnosing active inflammation in patients with CD. J. MAGN. RESON. IMAGING 2016;43:688-697.