Subscribe - the global online resource for all aspects of gastroenterology, hepatology and endoscopy

 18 June 2018

Advanced search - the global online resource for all aspects of gastroenterology, hepatology and endoscopy Profile of Roy Pounder


Review Articles
Slide Atlas
Video Clips
Online Books
Advanced Digestive Endoscopy
Classical Cases  
Conference Diary
International GH Links
USA GH Links
National GH Links
National GI Societies
Other Useful Links

Emails on Gastroenterology and Hepatology
the National AIDS Treatment Advocacy Project
Visit the gastroenterology section of the EUMS

Classical Case Studies

GastroenterologySmall intestine


Click a heading in the right-hand column to see the selected articles in that subject area

See any comments for this case

Charles Chaya A bleeding catastrophe: Aortoenteric fistula
Charles Chaya, 06 November 2003


We report a case of a bleeding catastrophe in the form of an aortoenteric fistula. This unfortunate and uncommon event took place in a patient who had received elective surgery with graft repair for a history of claudication.

On this admission, a stent was placed through the graft to seal off the leak and serve as a bridge towards surgery. This case emphasizes several points in the management of massive gastrointestinal bleeding, particulary when an aortoenteric fistula is suspected.

Case report


An 86-year-old male presented with a history of fever, chills, weakness and productive cough. The patient was a poor historian and denied any other complaints.

Prior to this illness, his white count was elevated with a shift in immature cells which prompted a cancer workup. In the clinic, he underwent a biopsy of an enlarged lymph node in the post cervical chain to rule out lymphoma.

His past medical history included in 2000 an aortobifemoral graft placement for claudication. A colonoscopy performed in 2001 showed extensive diverticular disease.

The patient took no medications and denied any alcohol or illicit drug use.

On hospital day 8, we were asked to see this man when he began having bright red blood per rectum. The gastroenterology team was consulted for a colonoscopy. The patient was transferred to the intensive care unit.


In the intensive care unit, the patient's blood pressure was 76/37. The patient lay in a pool of bright red blood extending to his ankles.

Physical examination findings were that his nasogastric lavage was clear; there was no palpable pulsatile mass overlying his abdominal surgical scar which was clean, closed and dry.


Laboratory findings included white count of 39 K/uL with a differential showing neutrophils 60%, bands 9%, lymphocytes 1% and monocytes 15% immature lymphocytes 2%, metamyelocytes 3% and myelocytes 9%; hemoglobin of 2.1 g/dl (baseline 8.7); platelet count of 162; and liver chemistries and PT/PTT within normal limits.

The patient was given 8 units of packed red blood cells and 6 units of fresh frozen plasma over a 12-hour period. Repeat hemoglobin was 7 g/dl after transfusions.

An endoscopic procedure was not feasible or advisable. Our recommendations as the consulting service were to obtain a CT scan of the abdomen/pelvis to rule out an aortoenteric fistula, a tagged red blood scan to localize the site of bleeding, and angiogram for a possible therapeutic intervention.

The patient had a tagged red blood cell scan but did not have a CT scan. Bleeding was localized in the right lower quadrant (Figure 1).

An angiogram was obtained which was negative. Despite the negative angiogram, the patient continued to bleed and was taken for surgery. He received a right hemicolectomy. The pathology specimen report read, "a single diverticulum showing ulceration and severe acute inflammation, and hemorrhage".

However, on postoperative day 10 the patient bled again; passing not only bright red blood per rectum but also hematemesis.

Again, the patient was taken for an angiogram. The celiac, superior mesenteric and inferior mesenteric arteries showed no bleeding though it was obvious he was vomiting blood.

At that point, the decision was made to infuse contrast through the aortobifemoral graft. In the words of the radiologist, "there was extravasation of the contrast filling the ileum" (Figure 2).

The bleeding originated from the right iliac artery at the distal end of the graft where blood leaked into the ileum.

Figure 1
Tagged red blood cell scan: bleeding in right lower quadrant

Figure 2
Angiogram: "extravasation of contrast filling the ileum"

Management and outcome

At the time when the fistula was identified, surgery was not an option because of the patient's unstable condition. A stent was placed through the graft to seal the leak - serving as a bridge towards surgery. The bleeding stopped immediately.

The patient's condition remained critical with the persistently elevated white count and labile blood pressure.

Flow cytometry studies and bone marrow biopsies demonstrated that the patient's white count was due to a leukomoid reaction. The patient was found to be bacteremic. His blood cultures grew an unidentified Streptococcus species. The source of the infection was presumed to be his graft. The patient's moribund state prevented timely surgery and he ultimately died.


Aortoenteric fistulas are a rare event of gastrointestinal bleeding. However, their prompt recognition can potentially alter a fatal course.

This case underscored the importance of several key points in managing aortoenteric fistulas.

A CT-scan has a sensitivity and specificity of 94% and 85% respectively in diagnosing aortoenteric fistulas [1]. Findings suggestive of an aortoenteric fistula on imaging include ectopic gas, perigraft fluid, focal bowel wall thickening or pseudoaneurysm [1].

All patients with a prior history of vascular surgery with graft repair who present with a gastrointestinal bleed should receive a CT-scan to rule out an aortoenteric fistula. Prior to the advent of CT-scan imaging, aortoenteric fistulas were diagnosed preoperatively only 24% of the time [2].

However, knowledge of the location of this patient's prior surgery with graft repair proved to be important. The 2nd angiogram may have missed the aortoenteric fistula if it was limited to only the celiac, superior mesenteric and inferior mesenteric arteries.

The history of surgery was made known to the interventional radiologist who was able to identify the leak from the distal end of the graft in the right iliac artery. From this case, we conclude that an angiogram, since it is both diagnostic and therapeutic, may hold equal value in importance as a CT scan.

This article was first published on on 6 November 2003.


Dr Charles Chaya MD
2nd year Gastroenterology Fellow
Division of Gastroenterology, Tulane University, New Orleans, Louisiana, USA

Contact information

Dr Charles Chaya
Division of Gastroenterology, Tulane University, New Orleans, Louisiana, USA


  1. Low RN, Wall SD, Jeffrey RB Jr et al. Aortoenteric fistula and perigraft infection: evaluation with CT. Radiology 1990; 175(1): 157-62.
  2. O'Mara CS, Williams GM, Ernst CB. Secondary aortoenteric fistula. A 20 year experience. Am J Surg 1981; 142(2): 203-9.

Go to top of page


Go to top of page Email this page Email this page to a colleague

Open folder Gastroenterology
Close folder Hepatology
Biliary tract
Viral hepatitis
Liver diseases
Cirrhosis and portal hypertension
Liver transplantation
Liver & other diseases
Pediatric hepatology
Basic science
Close folder Endoscopy
Upper endoscopy
Endoscopic ultrasound
Practice issues

Blackwell Publishing is a Blackwell Publishing registered trademark
© 2018 Wiley-Blackwell and and contributors
Privacy Statement
About Us