S0611 Gastrointestinal Bleeding in Patients With COVID-19: Outcomes With and Without Endoscopic Evaluation

INTRODUCTION: With the onset of the COVID-19 pandemic, hospitals have restructured criteria for endoscopic procedures in order to reduce the number of aerosol-generating procedures and to minimize exposure of consultation services and staff to the virus and to preserve personal protective equipment We aimed to evaluate outcomes of patients who were hospitalized for SARS-CoV-2 and experienced gastrointestinal bleeding (GIB) METHODS: We performed a retrospective review of patient charts with confirmed COVID-19 between 03/12 and 05/11/2020 across four hospitals for whom gastroenterology was consulted for GIB We evaluated demographics, past medical history, anticoagulation use, number of units transfused, and mortality of patients who had GIB during hospital stay Patients who underwent endoscopy and managed medically were compared with Wilcoxon Rank Sum, Chi-Square and Fisher's exact test as appropriate RESULTS: Of the total 68 patients with GIB, 14 underwent endoscopic evaluation (Group A) and 54 were managed medically (Group B) Only two patients underwent endoscopic hemostatic therapy Age, gender and ethnicity were not significantly different between the two groups ICU admissions accounted for 5 (35 7%) from Group A and 15 (27 8%) from the Group B In terms of respiratory status, 21% in Group A compared to 48 7% in Group B required mechanical ventilation, BiPAP or high flow nasal canula Patients who were not receiving any anticoagulation were more likely to undergo endoscopy, however this was not statistically significant (P = 0 069) In the medically managed group, 35 patients (64 8%) received packed red blood cell transfusion with a mean of 2 8 units whereas 11 (78 6%) patients in the endoscopically managed group received a mean of 4 2 units Patients who underwent endoscopy had a mean hospital stay of 13 9 days whereas those treated conservatively had a mean of 12 6 days Mortality was 21 4% (3/14) in the group receiving endoscopy and 42 6% (23/54) in the medically managed group (P = 0 22) CONCLUSION: Although not statistically significant, GIB patients who were medically managed had higher mortality rate than patients who underwent endoscopies This is likely due to the severity of COVID-19, which was indicated by the patient's oxygenation requirement, and less likely from blood loss due to the lower mean units of pack red blood cell transfusion