One hundred and twenty-six of 157 consecutive patients (80%) admitted
for a bleeding peptic ulcer were treated conservatively and retrospect
ively analysed. There were 52% duodenal, 41% gastric and 7% combined u
lcers. The initial shock index (pulse/ systolic blood pressure) was in
excess of one in 10%. For 22% of the patients no transfusion was requ
ired but 10% had more than 6 units of blood during their hospital stay
. Forty-nine per cent were on nonsteroidal anti-inflammatory drugs and
83% had at least one coexisting systemic disease. Six patients (5%) h
ad a further haemorrhage, four of whom died. A total of 10 patients (8
%) died. Five of them were related to the peptic ulcer disease but als
o had terminal or multiple systemic diseases precluding any surgery. T
heir poor short-term prognosis shows how difficult it will be to effec
tively reduce the mortality in this particular group of conservatively
treated patients, even with the recent advent of endoscopic haemostas
is, and stresses the importance of carefully identifying high risk pat
ients in trials mounted to improve on the current mortality figures.