Premium
Small intestinal contrast ultrasonography for the detection of small bowel complications in C rohn's disease: Correlation with intraoperative findings and magnetic resonance enterography
Author(s) -
Kumar Shankar,
Hakim Aishah,
Alexakis Christopher,
Chhaya Vivek,
Tzias Demitrios,
Pilcher James,
Vlahos Johnny,
Pollok Richard
Publication year - 2015
Publication title -
journal of gastroenterology and hepatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.214
H-Index - 130
eISSN - 1440-1746
pISSN - 0815-9319
DOI - 10.1111/jgh.12724
Subject(s) - medicine , crohn's disease , ultrasonography , magnetic resonance imaging , inflammatory bowel disease , concordance , gastroenterology , radiology , disease
Background and Aims In evaluating small bowel C rohn's disease ( CD ), small intestine contrast‐enhanced ultrasonography ( SICUS ) is emerging as an alternative to magnetic resonance enterography ( MRE ). This retrospective study compared the diagnostic accuracy of SICUS and MRE with surgical findings, and their level of agreement. Methods We identified a cohort of CD patients investigated by either SICUS and/or MRE that subsequently required resective bowel surgery within 6 months. The accuracy and agreement of SICUS and MRE to detect small bowel complications were compared with intraoperative findings using kappa coefficient (κ). Agreement between SICUS and MRE in those undergoing both modalities was also assessed. Results A total of 67 patients were evaluated; 25 underwent SICUS and 17 underwent MRE prior to surgery. Another 25 patients underwent both SICUS and MRE . When compared with intraoperative findings, the sensitivity of SICUS and MRE was 87.5% and 100%, respectively, in detecting strictures, 87.7% and 66.7% for fistulae, 100% for both in identifying abscesses, 100% and 66.7% for bowel dilatation, and 94.7% and 81.8% in defining bowel wall thickening. When correlating SICUS and MRE with surgery, there was a high level of agreement in localizing strictures (κ = 0.75, 0.88, respectively), fistulae (κ = 0.82, 0.79) and abscesses (κ = 0.87, 0.77). Concordance between SICUS and MRE was substantial or almost complete in identifying stricturing disease (κ = 0.84), their number and location (κ = 0.85), fistulae (κ = 0.65), and mucosal thickening (κ = 0.61). Conclusion SICUS accurately identified small bowel complications and correlated well with MRE and intraoperative findings. SICUS offers an alternative in the preoperative assessment of CD .