Background: The main objectives of this study were to review the clinic and endoscopic variables of patients with upper gastrointestinal bleeding (UGB) due to peptic gastroduodenal lesions who were evaluated in our hospital during one year, to identify the characteristics associated with a negative evolution of the UGB and to analyse the characteristics of those patients who were discharged immediately after the endoscopy was performed.
Patients and method: A one-year retrospective analysis of all UGB episodes was performed. Patients having gastroduodenal ulcer or erosive gastritis/duodenitis at endoscopy were included. The prognostic value of several clinic, laboratory and endoscopic variables was evaluated. Persistence or recurrence of bleeding, surgery and mortality were considered as outcome variables (the variable evolution was categorized as negative when any of these was observed).
Results: 156 patients were identified, with a mean (SD) age of 61 (17) years. Melena was the most frequent UGB presentation (79%). 46% patients had associated diseases and 50% were taking gastroerosive drugs. Duodenal ulcer was the commonest cause of UGB (52%), followed by gastric ulcer (30%). The evolution of UGB was negative in 7% cases. Variables associated with a negative evolution in the multivariate analysis were: age, red hematemesis, systolic blood pressure >= 100 mmHg, heart rate >= 100 b.p.m. and a more severe Forrest endoscopic classification. 11% patients were discharged immediately, without complications afterwards. If predictive variables obtained in the multivariate analysis would have been applied, hospitalization would have been avoided in 59 patients (38%) without subsequent complications.
Conclusions: A number of clinic and endoscopic variables (UGB presentation form, blood pressure, heart rate, and Forrest endoscopic classification) with prognostic value have been identified in this study. These variables are easy to obtain and apply in clinical practice, allowing a precise estimate of the UGB evolution. Thus, a relatively high number of patients with UGB susceptible of ambulatory treatment and management can be reliably identified.