Duodenal diverticuli are acquired false diverticuli of unknown etiology. Although mostly asymptomatic, they can occasionally cause upper gastrointestinal hemorrhage, rarely with massive bleeding. In this report we present a case of severe duodenal diverticular bleeding causing abdominal compartment syndrome.

A 50-year-old man with Child-Pugh class A alcoholic cirrhosis and recent increased ibuprofen use was brought to our Emergency Department after a syncopal episode at work. Shortly after arriving, he developed brisk hematemesis and hematochezia and had a full cardiac arrest. His hemoglobin dropped from 13.1 g / dL (on presentation) to 4 g / dL. The patient was intubated, and approximately four minutes of cardiopulmonary resuscitation (CPR) were required. He was resuscitated with six units of packed red blood cells and two units of fresh frozen plasma. During his resuscitation, he continued to pass a large amount of blood by mouth and per rectum. Once he was stabilized, a CT scan of the abdomen and pelvis was obtained, which showed mildly distended loops of the jejunum and ileum, with no free intra-abdominal air or fluid. No other pathological findings were identified. During the ongoing resuscitation, the patient's abdomen became distended. His peak airway pressures quickly increased (exceeding 40 mmHg). Consequently, pressure-controlled mechanical ventilation was required. His bladder pressures were 25 mmHg. Based on these findings, the patient was diagnosed with an abdominal compartment syndrome and was taken emergently to the Operating Room for a decompressive laparotomy and further assessment of his gastrointestinal hemorrhage.

Immediately after opening the patient's abdomen in the operating room, his peak airway pressures returned to normal. His stomach and small bowel were distended with blood, but otherwise appeared normal. Intraoperative upper gastrointestinal endoscopy was performed. There were no esophageal varices. The stomach was full of clotted blood and had no other abnormalities. An ulcer with an adherent clot on the posterior wall of the duodenal bulb was noted, although there was no active bleeding. Based on these findings, it was felt that the source of his hemorrhage was a duodenal ulcer that had eroded into the proximal gastroduodenal artery (GDA). The pylorus was opened using a longitudinal incision. The duodenal ulcer was identified, and the vessel was oversewn. The small bowel contents were milked retrograde from the ileocecal valve to the ligament of Treitz, yielding approximately 4500 mL of old blood. Shortly thereafter, the patient developed active arterial hemorrhage seen through the gastroduodenotomy, but the origin was not visible. Using manual occlusion of the stomach (proximally) and proximal jejeunum at the ligament of Treitz, we noted ongoing bleeding from the gastroduodenotomy, thereby localizing the source between the second and fourth portions of the duodenum. We kocherized the duodenum and noted a diverticulum on the medial aspect of the second portion of the duodenum, but we could not identify a clear bleeding source. With the abdominal compartment syndrome relieved and uncontrolled, unlocalized hemorrhage; it was felt that the patient would be best served by an angiography. Consequently, a pyloroplasty and temporary abdominal closure were performed, and the patient was taken for emergent angiography.

The CT scan of the abdomen and pelvis that was obtained upon admission was reviewed again; the duodenal diverticulum that we visualized in the Operating Room was noted [Figure 1]. Selective arteriography of the celiac artery demonstrated frank contrast extravasation from the mid to the distal gastroduodenal artery (GDA) (adjacent to the duodenal diverticulum) [Figure 2]. The pseudoaneurysm was successfully embolized with coils. Thereafter, the patient had no evidence of ongoing hemorrhage. His abdomen was closed the following day.

Throughout his stay in the Surgical Intensive Care Unit, the patient remained hemodynamically stable. He was extubated on postoperative day 3, and his diet was advanced. On postoperative day 5, the patient developed deep vein thromboses in his bilateral lower extremities and right upper extremity despite prophylactic enoxaparin (30 mg subcutaneously twice a day). Thus, his enoxaparin was changed to a therapeutic dose (1 mg/kg subcutaneously twice a day).

The patient was transferred to the floor on postoperative day 8. However, on the floor, the patient developed right upper quadrant pain and leukocytosis (peaked at 16.4 k / μl). A CT scan of the abdomen on postoperative day 14 demonstrated a new loculated abscess in the gastrohepatic ligament deep to the left hemidiaphragm, not amenable to CT-guided drainage. The patient was treated with 10 days of vancomycin, piperacillin-tazobactam, and fluconazole, and his white blood cell count normalized. He completed an additional two-week course of oral antibiotics (amoxicillin-clavulanate, ciprofloxacin, and metronidazole). A follow-up CT scan on postoperative day 20 showed no significant improvement in fluid collection. He was discharged home on postoperative day 25.

Duodenal diverticuli are acquired false diverticuli; the etiology is unknown. Most are periampullary. They have been identified in 2.5% of the upper gastrointestinal examinations and up to 22% of the endoscopic retrograde cholangiopancreaticographies and autopsies.[^{1, 2}] However, because fewer than 5% of the duodenal diverticuli are symptomatic (due to bleeding, perforation or obstruction), the true prevalence is unknown.

Duodenal diverticuli are a rare cause of upper gastrointestinal hemorrhage.[¹] To our knowledge, there are only 55 reports of bleeding from duodenal diverticuli in the literature, 12 out of which have been associated with massive bleeding.[³–¹⁴] All the cases of massive bleeding from duodenal diverticula reviewed by us were handled surgically, which was virtually always diverticulectomy. One case was treated with a Whipple procedure to manage a fistulous tract that formed a communication between the diverticulum and the pancreas.[¹²] As the bleeding source was not clearly delineated at the time of laparotomy, and as the patient was unstable, we chose to manage our patient with angiography rather than with diverticulectomy.

To our knowledge, our patient is the first reported case of abdominal compartment syndrome (evidenced by abdominal distention, elevated peak airway, and elevated bladder pressures, with prompt resolution upon opening the abdomen) resulting from duodenal diverticular hemorrhage (evidenced by the angiographic findings). The hemorrhage was brisk enough to cause massive bowel distention with approximately 5 L of blood. On account of his thin body habitus, that degree of intraluminal blood was sufficient to cause an abdominal compartment syndrome. There was no edema or other abnormal findings at the time of laparotomy, to offer an alternative explanation as the cause of the compartment syndrome.

As with any abdominal compartment syndrome, surgical decompression is the cornerstone of treatment. Like many cases (including ours), a multidisciplinary approach to complex cases of gastrointestinal hemorrhage is critical. Interventional gastroenterologists, surgeons, and interventional radiologists, all played a critical role in the care of this patient.

To the best of our knowledge, we are the first to present the case of bleeding from a duodenal diverticulum resulting in an abdominal compartment syndrome. Regardless of the cause, abdominal compartment syndromes require prompt identification and surgical decompression. As with our case, a multidisciplinary approach to managing patients with massive gastrointestinal hemorrhage is crucial.

Source of Support: Nil

Conflict of Interest: Authors have no competing interest to disclose.