Haematemesis and melaena at St. Vincent's Hospital, Melbourne.

Abstract

One hundred and fifty-three patients with haematemesis and melaena were studied retrospectively. The majority were managed in general medical units. Fifty-one per cent had a past history of peptic ulcer and/or bleeding, and only 9% had no identifiable predisposing factors. Endoscopy was the major diagnostic tool, with a diagnostic rate of 85%. Nineteen per cent of patients had no diagnostic investigations performed for varied reasons. The operative rate was 15% overall, with an operative mortality of 17%. The presence of other serious disease states concomitant with the haematemesis and melaena, and the presence of oesophageal varices as the aetiology, were both shown to be associated with statistically worse prognoses. Age of greater than 50 years and transfusion of four or more units of blood were also associated with a worse prognosis. Overall mortality was 14%, comparable to or less than that at several institutions, but more than double that of a Haematemesis and Melaena Unit in another Melbourne hospital [Br. Med. J. 1, 1238-40 (1979)] Patient population should be considered closely when comparing mortalities from different series. It is felt that results could be improved using a combined medical and surgical approach, and a set protocol of management, and that such changes should be monitored via a prospective study.