Computed tomography with clinical scoring to differentiate phytobezoar from feces in childhood small bowel obstruction.

Abstract

Identification of phytobezoar in childhood small bowel obstruction (SBO) characterized by smallbowel feces sign (SBFS) is still challenging. The aim of our study was to assess the diagnostic performance of quantitative computed tomography (CT) analysis combined with the Acute General Emergency Surgical Severity-Small Bowel Obstruction (AGESS-SBO) scoring system in determining phytobezoar-related SBO. Sixteen phytobezoar-related SBO were categorized as the phytobezoar group and the other 19 SBFSpositive SBO was categorized as the control group. Demographic data, clinical presentation, and laboratory and CT findings were collected and analyzed. Each patient`s AGESS-SBO score was determined according to the individual medical record. Multivariate logistic regression analyses were used to identify significant variables associated with phytobezoar-related SBO. Diagnostic performance of key variables was assessed using receiver operating characteristic (ROC) curve analysis. Compared to the control group, the phytobezoar group showed a significantly shorter debris maximal length (3.0 ± 0.5 cm vs. 3.5 ± 0.7 cm, P < 0.05), stronger attenuation (12.6 ± 5.9 HU vs. 8.2 ± 4.0 HU, P < 0.05) in CT, and higher AGESS-SBO scores (4.5 [interquartile (IQR): 4-5]) vs. (2 [IQR: 1-4]). With the combination of debris attenuation (with a cut-off of > 9.0 HU) and AGESS-SBO score (with a cut-off of > 3 points), the positive predictive value (PPV) and negative predictive value (NPV) to diagnose phytobezoar-related SBO were 80% (12/15) and 84% (16/19), respectively. The diagnostic method of integrating quantitative CT analysis and the AGESS-SBO scoring system can improve the identification accuracy of phytobezoar in SBFS-positive childhood SBO.