RESUMO
The results of a global (general series of 3,270 episodes of upper gastrointestinal haemorrhage (UGH) admitted to our unit between the 15th of April 1983 and the 15th of April 1988 have been analyzed. All the patients entered a prospective protocol with 29 variables. Diagnostic and therapeutic approaches had previously been defined. The incidence of UGH in this area was 160 bleeding episodes/100,000 inhabitants/year. Mean age was 57 +/- 16.8 years and male/female ratio was 2.66/1. The percentage of patients older than 65 years was 33.85%. A history of non-steroidal anti-inflammatory drugs (NSAID) intake within 48 hours before the bleeding episode was obtained in 27.63%. Continued alcohol ingestion was observed in 25.96% and 34.37% of patients gave a history of a previous episode of bleeding. UGH presented with haematemesis and melena in 56% of cases, and 44% only with melena. On admission the bleeding was haemodynamically severe in 12.96% and a 19.69% of the patients had severe associated diseases. Early endoscopy in cases with UGH due to peptic ulcer revealed active bleeding in 16.35% (2.87% in jet and 13.48% oozing) and recent clot/visible vessel in 31.7%. The major causes of bleeding were peptic ulcer (54.31%), esophageal and gastric varices (10.73%) and acute lesions of the gastric mucosa (ALGM) (6.72%). Etiology of the haemorrhage could not be established in 8% of cases. Bleeding was persistent in 20.75% and limited in 79.25% of patients. Emergency surgery was needed in 14.43% of cases. The global mortality of the series was 7.65%.(ABSTRACT TRUNCATED AT 250 WORDS)
