EP472 - A Massive GI Bleed that did not Resolve After a Total Abdominal Colectomy - A Case Report

Image or Table (Do Not add text here):

Purpose/Background: GI bleed can range from a minor self-resolving incident to a life threatening and sometimes a fatal one. Patients who experience passage of melanotic stool are presumed to have upper GI bleed due to the digestive action of the intestinal bacteria on the hemoglobin molecule. However, patients who experience bright red per rectum are most often presumed to have lower GI bleed.
Treatment of lower GI bleed can range from conservative treatment and outpatient colonoscopy for minor bleeding to inpatient colonoscopy for unresolving bleeding to emergent total colectomy for unstable bleeding so it’s important to promptly diagnose the source of the bleeding and establish a plan early on to avoid worse outcomes for the patient.

Methods/Interventions: A 40-year-old female with history of biliopancreatic diversion procedure for weight loss as well as cirrhosis secondary to NASH, alcohol use and chronic hepatitis C. The patient presented to the ER with abdominal pain secondary to acute cholecystitis. Initially admitted to the floor for observation but was quickly upgraded to the ICU for worsening hemodynamic instability. It was presumed that the patient was going through septic shock from her acute cholecystitis and the initial plan was to have interventional radiology perform a percutaneous cholecystostomy pending correction of her INR of 3.5 secondary to her liver disease. However, the patient started passing large amounts of blood from her rectum in the ICU and initially stabilized after blood transfusion which allowed time for a CTA of her abdomen. Her CT scan showed diffuse colitis, but no distinctive source of bleeding was identified. A few hours after, the patient started to become more hypotensive despite blood transfusions with more notable bleeding from her rectum. Decision was made to take the patient to the OR for a total abdominal colectomy given her uncorrectable hemodynamic instability.

Results/Outcomes: In the OR, the patient quickly underwent an exploratory laparotomy with total abdominal colectomy to control what was presumed to be a massive lower GI bleed. However, we were encountered with a biliopancreatic diversion anatomy that needed to be reversed to be able to create an end ileostomy with controlled output. Once we started to take down the ileo-ileal anastomosis, large amounts of blood were encountered in the small bowel. The entirety of the small bowel was palpated by no masses were found. At this time an EGD was done to rule out any gastric ulcer of esophageal variceal bleeding, but neither were found on endoscopy. As a last resort, we decided to open the duodenal stump to look at the major duodenal papilla and to our surprise, massive amounts of blood were pouring from the major duodenal papilla. The patient was then temporarily closed and transferred to the IR suite where she had an angiogram revealing an arterioportal fistula between the pancreaticoduodenal vessels which was embolized. The patient was left open and continued to deteriorate in the ICU for the subsequent days mostly from multiple infections and was later placed on comfort care by family and expired.

Conclusion/Discussion: Massive GI bleed can be challenging due to the limited amount of time available to make a diagnosis and establish a treatment plan. In our case the patient had a massive upper GI bleed from her pancreatic duct due to a fistulous connection between the pancreaticoduodenal vessels (arterioportal fistula) causing her bleeding to be mistaken for a lower GI bleed. To conclude, it is always important to consider so many etiologies of bleeding which it comes to massive GI bleeds.