Non-emergency small bowel obstruction: Assessment of CT findings that predict need for surgery

Swati D. Deshmukh, David S. Shin, Juergen K. Willmann, Jarrett Rosenberg, Lewis Shin, R. Brooke Jeffrey

Research output: Contribution to journalArticle

31 Scopus citations

Abstract

Objective: To identify CT findings predictive of surgical management in non-emergency small bowel obstruction (SBO). Methods: Contrast-enhanced abdominal CT of 129 patients with non-emergency SBO were evaluated for small bowel luminal diameter, wall thickness, presence of the small bowel faeces sign (intraluminal particulate matter in a dilated small bowel) and length, transition point, submucosal oedema, mesenteric stranding, ascites and degree of obstruction (low grade partial, high grade partial and complete obstruction). Medical records were reviewed for age, gender, management and history of abdominal surgery, abdominal malignancy, or SBO. Statistical analyses were performed with Stata Release 9.2. Results: Degree of obstruction was the only predictor of need for surgery. Whereas 18.0% of patients with low-grade partial obstruction (n=50) underwent surgery, 32.5% of patients with high-grade partial obstruction (n=77) and 100% of patients with complete obstruction (n=2) required surgery (P=0.004). The small bowel faeces sign was inversely predictive of surgery (P=0.018). Conclusion: In non-emergency SBO patients with contrast-enhanced CT imaging, grade of obstruction predicts surgery, while the small bowel faeces sign inversely predicts need for surgery.

Original languageEnglish (US)
Pages (from-to)982-986
Number of pages5
JournalEuropean Radiology
Volume21
Issue number5
DOIs
StatePublished - May 1 2011

Keywords

  • CT
  • Grade of obstruction
  • Small bowel feces sign
  • Small bowel obstruction
  • Surgery

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Non-emergency small bowel obstruction: Assessment of CT findings that predict need for surgery'. Together they form a unique fingerprint.

  • Cite this