Back to Annual Meeting Program

Background: Based on previous published data on small bowel obstruction (SBO) management, a three feature model has been adopted in our institution predicting the need for exploration. Obstipation combined with mesenteric edema and lack of the small bowel feces sign on computed tomography (CT) were associated with the need for exploration. Patients with 2 or less features were managed non-operatively and administered a Gastrografin (GG) challenge.
Hypothesis: We hypothesize that the (GG) challenge test, when used in combination with the predictive model, will decrease the rate of explorations in patients not meeting the criteria for immediate operation.
Methods: IRB approval was obtained to review patients admitted with SBO from November 2010 to September 2011. Presenting with signs of strangulation or all three features, and those who had an abdominal operation within 6 weeks of admission were excluded. All patients had an abdominal/ pelvic CT scan and GG challenge upon diagnosis. GG patients were compared to historic controls managed without the GG (July to December 2009). Successful GG challenge was considered as the presence of contrast in the colon after an 8 hour plain abdominal film or a bowel motion. Data is presented as medians or percentages. Significance was considered at p<0.05.
Results: One hundred and thirty three patients with a diagnosis of SBO were identified (47% male) with 54 receiving GG (study) and 79 historic control patients. There was no difference in age (71 vs 65 years), prior SBO (52% vs 47%), diabetes mellitus (20% vs 18%), history of malignancy (35% vs 41 %) or cardiac disease (29% vs 37%). Both groups had similar number of previous abdominal operations (2 vs 2). The presence of mesenteric edema (67% vs 76%), the lack of small bowel feces sign (50% vs 48%) and obstipation (24% vs 22%) were similar in both groups. Patients in the GG group had a lower rate of abdominal exploration (26% vs 43% p=0.044) and fewer complications (11% vs 33% p=0.004) compared to the control group. There was an equivalent number of strangulation obstructions (4% vs 10%), bowel perforation (4% vs 4%), length of hospital stay (4 vs 7 days), days from admission to operation (5 vs 3) and mortality (9% vs 6%). 46 patients had a successful GG challenge with 8 failures. There was a higher rate of exploration in patients with a failed challenge compared to those who passed (88% vs 15%, p<0.001).
Conclusion: The use of the GG challenge enhanced the SBO prediction model by decreasing the need for exploration in patients not meeting the criteria for immediate operation. Patients who failed the GG challenge test were much more likely to undergo exploration.