We retrospectively reviewed data of 401 patients aged ?#xA0;60 years presenting with upper gastrointestinal bleeding. Information collected included history, physical examination findings, laboratory data, endoscopic findings and length of hospital stay. Patients were divided into two groups: group A (65?9 years) and group B (>79 years).
Group A included 315 patients and group B 86 patients. There was a male preponderance in both groups. Co-morbidity (p < 0.01) and use of non-steroidal anti-inflammatory drugs (NSAIDs) or anti-platelet drugs (p < 0.01) were more common in group B. Oesophagitis was the cause of bleeding in 38.37 % in group B, as compared with 19 % in group A. The main cause of bleeding in group A was peptic ulcer. Rebleeding (6/86) and emergency surgery (1/86) were rare in group B and not different from those in group A. However, the bleeding-related mortality in the very elderly group was higher (13.9 % vs. 4.76 % ; p = 0.02).
In multivariate analysis, only shock on admission was independently related to mortality (p = 0.02).
Oesophagitis is the major cause of upper gastrointestinal haemorrhage in the very elderly patients. While rebleeding and emergency surgery rates are relatively low, the bleeding-related mortality was higher in the very elderly group.