3127 Blakemore? Wait More!

Abstract

INTRODUCTION: This is a case of a decompensated cirrhotic patient with a non-variceal upper GI bleed that was nearly treated with placement of a Sengstaken-Blakemore tube, which would likely have perforated her acutely ischemic esophagus. CASE DESCRIPTION/METHODS: A 62-year-old woman with decompensated alcoholic cirrhosis and a history of esophageal varices was brought to the ED for “coffee ground” emesis and lethargy. She relapsed into drinking alcohol two weeks before presentation and shortly thereafter developed abdominal pain. Three days before admission, she developed non-bloody emesis. On the day of presentation she had at least 10 episodes of clear emesis that eventually turned black. She had a history of esophageal varices requiring banding, most recently 5 years prior to admission. Her most recent EGD was 3 years ago, which was significant for portal hypertensive gastropathy and small varices that did not require banding. Initial vital signs were significant for a HR of 110 and BP of 77/37. Abdominal exam demonstrated normoactive bowel sounds, diffuse tenderness to palpation, no guarding, and no rebound tenderness. In the ED, patient had at least 2000 cc of “coffee ground” emesis and she was intubated for airway protection. Placement of a Blakemore tube was considered but ultimately decided against given that observed emesis was not bright red and emergent endoscopy would soon be performed. Initial labs were significant for: Hgb 14.4 platelets 106 lactate 24 pH 6.9 INR 1.5 creatinine 3.4 total bilirubin 2.8 AST 281 ALT 168. Emergent EGD in the MICU revealed no esophageal or gastric varices, however severe erosive esophagitis and acute esophageal necrosis with a “black esophagus.” Patient underwent a chest CT and cardiothoracic surgery consultation. After a brief stay in the MICU with supportive care, patient was ultimately extubated, her diet cautiously advanced and she was discharged to home. DISCUSSION: This case illustrates the need to think critically about the etiology of a brisk GI bleed, as not all significant GI bleeds in cirrhotic patients are related to varices. A clinical clue indicating that this may not have been an active variceal bleed was that the large-volume emesis was coagulated rather than fresh blood. Furthermore, the objective data reflected a state of severe hypovolemia and ischemia but no anemia in spite of recurrent hematemesis. Placement of a Blakemore tube in the setting of acute esophageal necrosis could have resulted in a catastrophic outcome of esophageal rupture.