Introduction Colonoscopy is the gold standard for evaluation of mucosal inflammation. However, CTE and MRE are now being considered as the initial investigation for small bowel Crohn’s disease (SBCD) to assess beyond the terminal ileum. This study aimed to assess the accuracy of diagnosing and staging SBCD with cross-sectional imaging and correlating the findings with faecal calprotectin (FCP).
Method We analysed patients with suspected SBCD at Hereford County Hospital from 2013–2016. Inclusion criteria were patients who had a CTE or MRE with a recent FCP. Patient’s notes, laboratory results and clinical imaging were reviewed and analysed. An FCP >50 ug/g was considered as a positive result.
Results 48 patients underwent FCP and cross-sectional small bowel imaging. CTE was used in 34 patients versus 14 for MRE. Patients who underwent CTE had a PPV and NPV of 50% for SBCD. Patients who underwent an MRE had a PPV of 70% and a NPV of 100% for SBCD. The use of FCP and MRE gave a sensitivity of 100% and a specificity of 57%. See tables 1 and 2 for more detailed results.
We further analysed patients with positive FCP but negative imaging results by investigating their recent colonoscopy findings. In the 6 patients who had a positive colonoscopy but normal CTE, 2 patients did not have contrast given thus they were poor quality images. 2 patients had a CTE after treatment with steroids and infliximab had started. 1 patient had mucopurulent exudate on colonoscopy, however biopsies were normal and a repeat FCP was normal. The last patient had a CTE which showed narrowing of anastomosis at TI which was shown to be a stricture on colonoscopy. Of those patients with a positive FCP but a normal scan and colonoscopy, 2 patients were diagnosed with bile salt malabsorption, 1 amoebic dysentery, 1 recurrent pharyngitis, 1 non-ulcer dyspepsia and 1 patient had a CTE after steroids were started.
The patient with a positive FCP and normal MRE was found to have mild erythema on colonoscopy. The other two patients with a positive FCP but normal findings thus far are still being investigated for their repeatedly elevated FCP. The patient with negative FCP and normal MRE had a colonoscopy showing patchy granularity and mucopurulent exudate in TI, however biopsies were normal.
Conclusion In our study, MRE gave a higher sensitivity for FCP and active SBCD. CTE gave a low PPV with a high false negative rate, which was subsequently correctly identified on colonoscopy. These results suggest that MRE is the cross-sectional imaging modality of choice for SBCD.
Disclosure of Interest None Declared
- Crohn’s disease
- FAECAL CALPROTECTIN