Unexpected Findings in Magnetic Resonance Enterography and Their Clinical Significance

The clinical significance of biliary findings in magnetic resonance enterography of patients with inflammatory bowel disease

PURPOSE

Given the association of inflammatory bowel disease (IBD) and primary sclerosing cholangitis (PSC), we aimed to investigate the clinical relevance of abnormal hepatobiliary findings on magnetic resonance enterography (MRE) of IBD patients considering the risk of over- or underestimation of PSC at MRE.

MATERIAL AND METHODS

Using the MRE dataset of patients referring to a tertiary hospital and the National Registry of Crohn's and Colitis, 69 MREs, including 23 IBD-PSC, 23 IBD-without PSC, and 23 healthy controls (HC), were retrospectively reviewed by 2 experienced radiologists blinded to the clinical data, to evaluate hepatobiliary abnormalities. Sensitivity, specificity, and likelihood ratios were calculated.

RESULTS

Bile duct irregularities were the most common finding in the IBD-PSC group, with a frequency of 91%. Intra- and extrahepatic bile duct (IHBD and EHBD) irregularities were observed in 87% and 78% of PSC patients, respectively. Higher frequency of IHBD and EHBD wall thickening, bile duct dilation, EHBD stricture, and periportal oedema were observed in the IBD-PSC group. Peribiliary T2-weighted hyperintensities and contrast-enhancement were significantly more common in the IBD-PSC group than in the IBD and HC groups (48% and 35%, respectively) (p < 0.001). Detection of biliary irregularities on MRE had a specificity of 94% (95% CI: 82-99%), a sensitivity of 91% (95% CI: 72-99%), and a positive likelihood ratio of 14.0 (95% CI: 4.7-42.1) for the diagnosis of PSC.

CONCLUSIONS

This study emphasizes the importance of assessing and reporting hepatobiliary abnormalities visible in the MRE of patients with IBD to avoid a delayed diagnosis of PSC.

Frequency and significance of incidental liver lesions on MR enterography in adult patients

PURPOSE

To evaluate the frequency and significance of incidental liver lesions identified on MR enterography (MRE) examinations to determine if dedicated sequences for liver evaluation are necessary in the routine MRE protocol.

METHODS

A retrospective departmental database search identified 353 adult (212 women and 141 men; mean [±SD] age, 41.4 [17.0] years; range 18.1-91.9 years) MRE examinations performed in 2017. Radiology reports were reviewed for the presence and characterization of liver lesions, follow-up recommendation, and known malignancy. Follow-up cross-sectional imaging reports were reviewed for liver lesion stability. A senior abdominal radiologist with expertise in liver imaging categorized liver lesions as benign, indeterminate, or malignant and re-characterized indeterminate lesions using follow-up imaging as benign or malignant.

RESULTS

Seventy-nine MRE (22.4%) described liver lesions and follow-up imaging was recommended in 4/79 (5.1%). Seventy-six liver lesions (96.2%) were characterized as benign (cysts/hemangiomas) on routine interpretation and expert review. One of these was recommended for follow-up imaging on initial report, which was characterized as hemangioma by expert review. The remaining 3 lesions (3.8%) were characterized as indeterminate both by initial report and expert radiologist review but re-characterized as benign after reviewing follow-up examinations.

CONCLUSION

All incidental liver lesions identified on MRE in our cohort were benign. Therefore, additional sequences evaluating the liver are unnecessary for routine MRE.