| Abstract: | During 1975-80, 908 patients admitted to Nottingham hospitals with gastrointestinal bleeding and found to have gastric or duodenal ulcers were analysed retrospectively for short-term outcome of treatment. Overall one-quarter of all patients underwent operation, but when the years 1975-7 were compared with 1978-80 the operation rate fell from one in three to just over one in five. Death rates were much lower in patients treated medically than in those who underwent operation, and the risks of operation were greater for patients with gastric ulcer. Less conventional operations were attended by greater mortality. Almost all patients who died during medical treatment and three-quarters of those who died after operation were over 65. No differences in age or clear variations in haemoglobin concentrations or transfusion requirements were found between the earlier and later periods. Reduction in operation rates had no appreciable effect on mortality, despite the accepted view that early operation is advisable. |
