Introduction
Patients with lower gastrointestinal bleeding form a significant part of the emergency surgical workload and many of them occupy hospital beds for extended periods. There are currently no guidelines regarding the admission to hospital of these patients.
Patients and Methods
All patients attending a dedicated surgical assessment unit over the period of a year whose primary problem was lower gastrointestinal bleeding were included in the study. Various relevant data sets were collected for each patient and these were compared to the binary outcome measure of blood transfusion. We aimed to define more clearly those high-risk patients that require admission to hospital for observation or treatment.
Results
A total of 90 patients were assessed in the unit during the study period and only 5 were discharged (admission rate of 94%). Of those that were admitted, over half (52%) received no in-patient investigation or treatment other than observation. Independent factors that made a blood transfusion more likely were: anticoagulation, a past history of diverticular disease, passage of dark blood and initial haemoglobin of < 11.5 g/dl for men and < 10.5 g/dl for women. Using these indicators, the admission rate could have been reduced from 94% to 61%. In addition, knowing that patients without these factors bleed for a mean duration of 2 days, a minimum saving of 62 bed-occupancy days per year is possible.
Discussion
Blood transfusion is an appropriate outcome measure in this study because it requires admission in its own right and is an excellent marker for in-patient investigation and treatment. We have defined a group of high-risk patients that should always be admitted for observation after an acute lower gastrointestinal bleed. Those patients that fall outside this group can be discharged after initial assessment with the understanding that persistent (> 48 h) or further heavy bleeding will require admission and that appropriate out-patient investigation will be organised.
