D0493 - Outcomes in Patients With Cirrhosis Undergoing Esophagogastroduodenoscopy for Upper Gastrointestinal Bleeding

Shabari M. Shenoy, MBBS¹, Mantej Sehmbhi, MBBS, MSc, MRCP², Suhail Haddadin, MD³, Maan El Halabi, MD³, Priya Simoes, MBBS^4
1Icahn School of Medicine at Mount Sinai Morningside-West, New York, NY; ²Mount Sinai Morningside and West Hospitals, New York, NY; ³Mount Sinai West and Morningside Hospitals, New York, NY; ⁴Mount Sinai West and Morningside Hospital, New York, NY

Introduction: Variceal hemorrhage is the most common cause of upper gastrointestinal bleeding (UGIB) in patients with liver cirrhosis. Endoscopic variceal ligation (EVL), along with pharmacotherapy, is the standard of care for treatment of bleeding gastroesophageal varices. Median re-bleeding rate in patients treated with EVL is about 7.6% at 2 weeks. We studied the rate of re-bleeding and difference in outcomes such as re-hospitalization, mortality in patients undergoing EVL compared to those that did not undergo EVL.






Methods: We identified adult patients with cirrhosis undergoing esophagogastroduodenoscopy (EGD) for UGIB, admitted over a 3-year period to a tertiary care center in New York City. Demographics, medical history, and rates of re-bleeding, rehospitalization and death at 60 days post-discharge were recorded. Patients undergoing EVL were compared to those who did not undergo EVL.






Results: 54 patients (median age 57 years, 44% male) underwent EGD for acute UGIB. Etiologies of UGIB were portal hypertensive gastropathy (34%), esophageal varices (34%), gastritis (32%) and gastric or duodenal ulcers (24.5%). 21 (38%) patients underwent EVL and 33 (61%) did not. Reasons for not undergoing EVL included other etiology of UGIB, varices less than 5mm or other. Age, gender, race, etiology of cirrhosis, MELD Na score, proton pump inhibitor (PPI) use and prior decompensating event, were similar between the 2 groups (Table 1). There was no difference in rates of re-bleeding, re-hospitalization due ascites, portosystemic encephalopathy, and infection and 60-day mortality between the 2 groups (Table 2). Interestingly, gastritis or gastroduodenal ulcers were the likely cause of UGIB in 23 patients (42%).14 patients were re-hospitalized within 60 days of initial admission. The leading cause of rehospitalization was rebleeding (41%).






Discussion: We found no differences in outcomes of cirrhotic patients hospitalized for UGIB based on etiology of UGIB and whether they underwent EVL. A significant proportion of patients rebled. Our study demonstrated that gastritis and gastric/duodenal ulcers were common etiologies of UGIB in cirrhotic patients. There is a higher risk of spontaneous bacterial peritonitis and liver-related mortality among cirrhotic patients on long term PPI. However, since a significant cause of UGIB in patients with cirrhosis may be peptic ulcer/gastritis related, future studies should focus on the length of PPI use after discharge in patients with cirrhosis who undergo EGD for UGIB.









Disclosures:


Shabari M. Shenoy, MBBS¹, Mantej Sehmbhi, MBBS, MSc, MRCP², Suhail Haddadin, MD³, Maan El Halabi, MD³, Priya Simoes, MBBS⁴. D0493 - Outcomes in Patients With Cirrhosis Undergoing Esophagogastroduodenoscopy for Upper Gastrointestinal Bleeding, ACG 2022 Annual Scientific Meeting Abstracts. Charlotte, NC: American College of Gastroenterology.