Abstract
Objectives
To validate the use of a modified Glasgow-Blatchford Score (mGBS) for risk stratification of patients with suspected upper gastrointestinal bleeding (UGIB) in an accident and emergency department in Hong Kong.
Methods
This was a retrospective cohort study of patients who attended the emergency department of the study centre from January 2014 to June 2014 who were subsequently admitted to surgical wards with suspected UGIB. High risk patients were considered to be those who required in-patient clinical interventions (blood transfusion, therapeutic endoscopy, angiographic embolisation, or surgery). The mGBS was calculated for each patient. The sensitivity, specificity, and area under the receiver-operating characteristic curve (AUC) of the score were calculated.
Results
A total of 372 patients were included in the study. With an mGBS of 0 (low risk) for detecting the primary outcome, the sensitivity was 99.2% (95% CI, 95.6100%), and the specificity was 25.91 (95% CI 20.6-31.8%). The negative likelihood ratio was 0.031 (95% CI 0.004-0.2). The AUC was 0.90 (95% CI 0.87 to 0.93).
Conclusion
The modified Glasgow-Blatchford Score is a clinically useful tool for emergency physician to identify UGIB patients at low-risk of requiring in-hospital clinical interventions. (Hong Kong j.emerg.med. 2016;23:3-11)