Digital archive of theses discussed at the University of Pisa


Thesis etd-06132017-125026

Thesis type
Tesi di specializzazione (5 anni)
email address
Thesis title
Adhesive Small Bowel Occlusion: which CT signs predict surgery?
Course of study
relatore Prof. Caramella, Davide
correlatore Dott. Tonerini, Michele
  • Surgery due to ischemia
  • Gastrografin® test
  • Non Operative Management
  • Emergency surgical treatment.
  • Adhesive Small Bowel Occlusion
Graduation session start date
- Objectives: To determine in adhesive small bowel occlusion (ASBO), which MDCT findings are predictive for the failure of a non-operative management (NOM), of Gastrografin® test outcome and finally for the need of surgery.

- Methods and Materials: From January 2015 to April 2017 we examined 137 admissions for ASBO: 71 Females, 66 Males, mean age 69 years (10-97), at our first level Emergency Department (ED). We excluded the patients without an adherential occlusion and those patients with spontaneous resolution of the bowel occlusion before the Gastrografin® test. This test is based on administration of the water soluble contrast (per os or via NGT) and to follow the progression of the contrast by means of seriated abdominal X-rays. It is considered successful if the cecum is opacified within eight hours.
The MDCT parameters, taken in consideration, were twelve: maximum caliber of the bowel; wall thickness greater than 5mm; parietal pneumatosis; presence or absence of peritoneal free abdominal fluid (FAF), peritoneal fluid density (measured in Hounsfield Units-H.U.); whirl sign; number of transition points; closed loop obstruction; small bowel feces sign; reduced bowel wall enhancement (RBE); mesenteric fluid congestion and fat notch sign.
The peritoneal fluid density has been considered measurable in the 84 admissions in which FAF, identified on MDCT, resulted to be sized >3cm2. The HU was measured by a round region of interest (ROI) positioned in the largest and lower pool of FAF (to include eventual blood stratifications) avoiding adjacent structures. A sensitivity analysis was performed to determine a high-density HU threshold. The FAF density in patients who underwent therapeutic laparotomy due to ischemia was compared with those successfully discharged without surgical approach.
We evaluated the previously mentioned MDCT parameters: a) in all the patients that underwent surgery (n=86), b) in the group of patients treated surgically owing to ischemia (n=22), c) in the patients treated successfully with NOM implemented by Gastrografin® test (n=51) and d) in patients that underwent failed NOM implemented by Gastrografin® test (n=56) to find out any MDCT sign predictive of NOM failure.

- Results: The RBE resulted to be the more significant MDCT parameter (p value: <0,025) in predicting the surgical approach in general. Wall thickness greater than 5mm (p value: <0,0001); presence of peritoneal fluid (p value: 0,013); closed loop obstruction (p value: 0,044); RBE (p value: <0,0001) resulted to be significant in the prediction of surgery due to ischemia. Peritoneal fluid density (measured in HU) resulted to be significant for ischemia too and a best cut off value of 13,5-14,5 HU has been found with an AUC of 0,69 and a p value of 0,012. In fact, all patients with FAF density >14,5 UH resulted to need surgery due to ischemia with a sensitivity of 79% and specificity of 55%, PPV of 34% and NPV 90% an accuracy of 60% and a Youden Index of 0,34. The other MDCT parameters did not show any significant correlation.
In our study no MDCT sign resulted to be significant in prediction of a failure of NOM performed with Gastrografin® test.

- Conclusions: In ASBO, MDCT is fundamental for the management of the patients. Among the MDCT findings, RBE resulted significant in predicting the need of a surgical approach in general; while thickness greater than 5mm, presence of peritoneal fluid, closed loop obstruction, RBE and an increased peritoneal fluid density ( >14,5 UH) are useful to elaborate a model to predict surgery due to ischemic complications. No MDCT parameter resulted to be significant in prediction of a failure of NOM implemented with Gastrografin® test.