Comparative Validation of Abdominal CT Models that Predict Need for Surgery in Adhesion-Related Small-Bowel...

Area of Science:

• Radiology
• Gastroenterology
• Surgical Oncology

Background:

• Adhesion-related small-bowel obstruction (ASBO) is a common surgical complication.
• Non-operative management is feasible for selected ASBO patients.
• Accurate prediction of surgical necessity is crucial for optimal patient management.

Purpose of the Study:

• To validate three previously published computed tomography (CT) models for predicting the need for surgery in patients with ASBO.
• To assess the diagnostic performance of these models in a real-world clinical setting.

Main Methods:

• Retrospective analysis of 233 patients with ASBO admitted between November 2009 and April 2015.
• CT findings were assessed by a blinded radiologist.
• Validation of three established predictive models for surgical intervention.

Main Results:

• One validated model, combining obstipation, free intraperitoneal fluid, and high-grade obstruction, demonstrated a positive likelihood ratio (LR+) of 2.9 and a negative likelihood ratio (LR-) of 0.9.
• Another model using mesenteric edema, free fluid, and absence of fecalization showed a sensitivity of 38% and specificity of 88%.
• The performance of two models was not reproducible, potentially due to measurement bias.

Conclusions:

• The predictive performance of previously published CT models for ASBO surgery varied significantly in this validation cohort.
• Standardized and reproducible predictors are essential for developing reliable future predictive models.
• Further research is needed to refine CT-based prediction of surgical necessity in ASBO.