The experience of a haematemesis and melaena unit: a review of the first 513 consecutive admissions.

This report is of the results of management in a haematemesis and melaena unit at Prince Henry's Hospital, Melbourne. The unit was established in October, 1972, in response to unpublished data for the decades 1950 to 1959 and 1960 to 1969, which showed a mortality of about 15%. In the 39 months to December, 1975, 513 patients were received into a semi-intensive care setting. The unit staff consisted of a group of four surgeons and four physicians working a weekly roster. Primary care and liaison were the responsibility of the gastroenterology registrar. The basic diagnostic measure taken was the routine early use of fibreoptic duodenoscopy. The unit was set up with an agreed policy of management of the common causes of haematemesis and melaena, and data were prospectively recorded in a form suitable for computer analysis in every case. Of the 513 admissions, 378 were of males and 135 were of females. Forty-five patients died, giving an overall admission mortality of 8-8. There were 143 admissions for bleeding duodenal ulceration, 102 for acute peptic ulceration, 47 for chronic gastriculceration, 43 for oesophageal varices, 33 for Mallory-Weiss syndrome and 45 for less common causes of upper gastrintestinal bleeding. In 100 cases the source of bleeding was not discovered. Of the 143 patients admitted for chronic duodenal ulcer, either patients died, giving a mortality of 5-6%; 72 patients underwent operation, with an operative mortality of 9-7%. Of the 47 admitted with bleeding gastric ulcer, nine died (19-1%), while 26 came to operation; the operative mortality was 26-9%. There were 102 admissions for acute peptic ulceration, with an overall mortality of 11-7% (12 patients); 16 patients came to operation, with an operative mortality of 43-7%. Eleven deaths occurred in the 43 patients admitted for bleeding oesophageal varices (25-6%), with 10 patients coming to operation; the operative mortality was 30-0%. An age of greater than 50 years and shock on admission were the most significant factors for poor prognosis in this group of patients.