Variables associated with unfavorable evolution in the multivariate analysis were: systolic blood pressure ≤ 100 mmHg, heart rate ≥ 100 bpm, and high-risk lesions in Forrest classification. Guideline recommendations for immediate discharge were based on these variables. Patients were considered at a high risk if they had clinical
or endoscopic variables associated with unfavorable evolution (those were systolic blood pressure ≤ 100 mmHg, heart rate ≥ 100 bpm, and high-risk lesions in Forrest classification). Patients who did not meet these criteria were considered at low-risk. Physicians of the Gastroenterology Unit in our hospital were asked to adopt the guideline recommendations for immediate hospital discharge.
They made all patient care Doxorubicin decisions and were instructed to accept or reject the recommendations based on their best clinical Z-VAD-FMK manufacturer judgment. Guideline compliance was defined as patient discharge from the hospital immediately after being classified as low-risk patient. A prospective analysis was carried out on all UGIB episodes (hematemesis, coffee-ground emesis or melena) admitted to the emergency department at our hospital (a university hospital with a healthcare population of more than 500 000 people) from June 2006 to June 2007. Only patients with a gastroscopy and diagnosis of gastroduodenal ulcer or erosive gastritis/duodenitis were included. All patients underwent gastroscopy within 12 h of hospital admission as an outdoor emergency procedure, and the physician (a gastroenterologist) decided whether to discharge or admit the patient. Patients with other etiologies of UGIB and patients who did not undergo gastroscopy (due to refusal or contraindications for the examination) were excluded. The following clinical variables were considered: Age, sex, type of presentation of UGIB (coffee-ground emesis, melena or hematemesis), smoking habit (smokers or non-smokers), heavy alcohol intake (defined as more than
60 g/day), comorbidity (cardiac, respiratory, neurologic, hepatic or rheumatologic conditions, diabetes mellitus, active neoplasm, renal failure, and other diseases), gastric surgery for peptic ulcer, aspirin or other nonsteroidal N-acetylglucosamine-1-phosphate transferase anti-inflammatory drug use (defined as intake of these drugs, regardless of dose, during 7 days before the patient’s arrival at hospital), antiplatelet agent use, systolic blood pressure in the emergency department (less than or greater than 100 mmHg), heart rate in the emergency department (less than or greater than 100 bpm), and blood transfusion requirements. The following laboratory variables were recorded: hemoglobin, hematocrit, urea, and prothrombin time (the latter categorized as normal or abnormal). At discharge, hemoglobin and hematocrit were again analyzed.