Early endoscopic intervention in acute gastrointestinal bleeding may reduce the need for blood transfusion.

Abstract

Background

Acute gastrointestinal bleeding (GIB) is a common gastroenterological emergency worldwide with significant morbidity and mortality of 6 per cent-14 per cent. The main causes of death in patients with GIB include shock, aspiration, and therapeutic procedure carried out for the management of the GIB. Thus, the resuscitation strategy of blood transfusion plays a very important role in these patients before any other specific treatment. Currently, endoscopy is considered the mainstay of diagnosis and treatment for patients with GIB.

Aims

To assess the effect of an early endoscopic intervention on the need for blood transfusion in patients presented with GIB.

Methods

We retrospectively analysed the data for patients presented with hematemesis, melena, or hematochezia, from July 2015 to July 2016, in Ballarat Base Hospital (BHS) in Victoria, Australia. Data were extracted from the hospital coding system related to patient's demographic history, alcohol intake, comorbidity, procedure details including the timing and the type of procedure performed, and the number of units of blood transfused. Additionally, the laboratory blood test results for each patient were examined through the electronic records to assess the haemoglobin level before and after the blood transfusion.

Results

A total of 92 eligible patients with GIB during the 12 months study period, were included in this observational study. The median age of the study population was 67 years (range 2496) at the time of admission. A total of 67 patients (73 per cent) underwent inpatient endoscopic procedure with gastroscopy performed in 52 patients, colonoscopy in 5 patients, flexible sigmoidoscopy in 3 patients, and combined gastroscopy and colonoscopy in 7 patients. In the enrolled population (n=92), at time of presentation, 11 patients (12 per cent) had the haemoglobin level below 7grams per decilitre (g/dL), 17 patients (18 per cent) had haemoglobin level between 7 and 8g/dL, and 64 patients (70 per cent) had haemoglobin level greater than 8g/dL. Out of the 67 patients who had inpatient endoscopy, 12 patients underwent endoscopic procedure within 12 hours of admission (<12 hours group), including 5 patients who received blood transfusion; and 55 patients underwent endoscopic procedure greater than 12 hours after admission (>12 hours group), including 31 patients who received blood transfusion. Among participants who received a blood transfusion in the two groups, 1 out of 5 patients in the <12 hours group and 19 out of 31 patients in the >12 hours group had haemoglobin level below 8g/dL at the time of transfusion.

Conclusion

Trends of greater blood transfusion in patients with delayed (>12 hours) endoscopic procedure and administering blood transfusion at haemoglobin level >8g/dL in patients with early (<12 hours) endoscopic procedure were observed without achieving statistical significance. The results obtained from this study indicate that more saving in terms of cost of treatment from blood transfusion can be achieved by adopting an optimized restrictive transfusion strategy.