TBA (16S138)

Glasgow Blatchford Score (GBS): Raising The Bar

Author(s)

Dr Etimbuk Umana, Dr Olayiwola Amoran, Dr Pardeep Maheshwari, Dr Carol Goulding.

Department(s)/Institutions

Galway University Hospital, Ireland Naas General Hospital, Ireland

Introduction

Upper GI bleed (UGB) is a common presentation to the emergency department (ED). It accounts for three percent of all hospital admission. The GBS has been found to be superior to both the Rockall and AIM65 score for risk stratification of UGB. A GBS score of ≤0 has been shown to have a sensitivity of >99% for stratifying low risk bleeds, however recent studies show a GBS of ≤1 is equally sensitive and therefore could be considered as an appropriate cut-off for stratifying patients as low risk.

Aims/Background

The aim of this study was to a) determine the frequency of use of the GBS score in the ED and b) to assess the relationship between GBS and esophagogastroduodenoscopy (OGDs) findings and need for intervention.

Method

Retrospectively, data was obtained from both ED and endoscopy unit from the 1st January to 31st of July 2015 of all patients presenting to University Hospital Galway with UGB. Patient’s demographics, GBS, OGD findings and intervention were recorded. Patients <14 years of age and those presenting with lower GI bleeding were excluded from this study.

Results

A total of 109 patients were identified but data was available for 86 patients who fulfilled our criteria. In this study 47(55%) were males and 39(45%) were females, mean age 59 and 65 respectively. Only 12(14%) had their GBS documented on admission. Sixty four patients had normal (non-active bleeding) OGD’s, six had active bleeding on OGD, while sixteen patients had no data available. Of the seventy patients who underwent OGD’s, nine patients required intervention. Seventeen patients(20%) had a GBS ≤1 who underwent OGD's, none had active bleeding or required intervention. A GBS≤1 had a sensitivity and specificity of 100% and 30% respectively. It also had a NPV of close to 100% in predicting active bleeding and need for intervention.

Conclusions

Using a GBS of ≤1 was shown to be sensitive for identifying low risk UGB which could therefore be offered OGD as out-patients, there by reducing admission for UGB by 20%. Education of first responders and appropriate referral system for outpatient management of patients presenting with a GBS of ≤1 should be facilitated.