Retrospective evaluation of patients with gastrointestinal bleeding who applied to the emergency department

Objective: In our study, we aimed to evaluate the management and outcomes of patients diagnosed with gastrointestinal bleeding in our hospital's emergency department and to highlight similarities and differences with the literature. Material and Method: Our retrospective, cross-sectional study was conducted at the Adult Emergency Department of Çukurova University Faculty of Medicine Hospital. Patients' demographic, clinical, endoscopic, and laboratory data, Glasgow-Blatchford Bleeding and AIMS65 scores, and outcomes were evaluated, and patients were divided into two groups: those who died within 15 days and those who survived, and compared. Results: A total of 161 patients (99 males) were included in the study, with a median age of 72 (IQR=20), and 87.6% had comorbidities, with a high frequency of cirrhosis (p=0.025). Mortality was 1.2%, and no association was found between 15-day mortality and medication use, presenting complaints, physical examination findings, GB and AIMS65 scores (p>0.05). Pulse rates (p=0.012), shock indices (p=0.001), INR (p=0.038), AST (p=<0.001), ALT (p=<0.001), and lactate levels (p=0.008) were higher in deceased patients compared to survivors. Peptic/duodenal ulcers and erosive gastritis were the most common findings. Proton pump inhibitors, tranexamic acid, and erythrocyte suspension were the most commonly used drugs/treatments. Deceased patients were found to have received more tranexamic acid (p=0.037), noradrenaline (p=0.001), somatostatin (p=0.021), and dopamine (p=<0.001). The most common outcome was admission to the intensive care unit (ICU) (51%), followed by discharge (37.9%), with 13% of patients being monitored in the emergency department for more than 24 hours, and very few being admitted to the ward (4.3%). Result: We believe that tranexamic acid reduces mortality in patients with gastrointestinal bleeding. Pulse rate and shock index may be used instead of GB and AIMS65 scores to predict mortality. We suggest that arrangements should be made for patients to be monitored in clinics rather than in the emergency department.