Glasgow-Blatchford Score: Validated tool for pre-endoscopic triage of suspected UGI bleeding. Score of 0 identifies patients at very low risk for intervention.
Guidelines & Evidence
Clinical Details
Section 1
When to Use
When to Use
Pre-endoscopic triage of patients presenting with suspected acute upper GI bleeding (UGIB).
Identifying low-risk patients (score 0) suitable for safe outpatient management.
Applied at first clinical contact — requires no endoscopy results.
Appropriate for ED, acute medical units, and out-of-hours assessment.
Clinical Setting
The GBS is the preferred pre-endoscopic tool per BSG and ESGE guidelines. Score 0 has Se 99% for identifying patients who require no intervention.
Section 2
Formula & Logic
Scoring Variables
Variable
Criterion
Points
BUN (mmol/L)
6.5–7.9
2
BUN (mmol/L)
8.0–9.9
3
BUN (mmol/L)
10.0–24.9
4
BUN (mmol/L)
≥ 25
6
Hgb Male (g/dL)
12–13
1
Hgb Male (g/dL)
10–12
3
Hgb Male (g/dL)
< 10
6
Systolic BP
100–109 mmHg
1
Systolic BP
90–99 mmHg
2
Systolic BP
< 90 mmHg
3
Pulse ≥ 100 bpm
Yes
1
Melena
Yes
1
Syncope
Yes
2
Hepatic disease
Yes
2
Cardiac failure
Yes
2
Interpretation
Score = 0: Very low risk — suitable for outpatient management.
Score 1–5: Moderate risk — outpatient vs. inpatient decision based on full clinical picture.
Score ≥ 6: High risk — requires inpatient care and endoscopy.
Section 3
Pearls/Pitfalls
Key Clinical Pearls
Score 0 rule-out: AUC 0.90 — identifies ~16% of UGIB patients who need no endoscopic or blood transfusion intervention.
Superior to Rockall pre-endoscopy for need-for-intervention prediction.
Assess syncope as a proxy for haemodynamic instability.
Important Limitation
GBS does not predict rebleeding or mortality as well as post-endoscopic Rockall score. Use GBS pre-endoscopy, then transition to Rockall after endoscopy.
Section 4
Next Steps
Clinical Actions
01
Score 0: Discharge with urgent outpatient endoscopy; return precautions and written advice.
02
Score 1–5: Admit for observation; endoscopy within 24 hours.
03
Score ≥ 6: Urgent endoscopy (within 24 hours or sooner if haemodynamically unstable); IV access, resuscitation.
04
All patients: Stop NSAIDs and anticoagulants if safe to do so; document H. pylori status and test if endoscopy performed.
Section 5
Evidence Appraisal
Validation Metrics
Metric
Value
AUC (need for intervention)
0.90
Sensitivity (score 0)
~99%
Specificity (score 0)
~32%
Primary Reference
A risk score to predict need for treatment for upper-gastrointestinal haemorrhage
Blatchford O et al. • Lancet. 2000;356(9238): 1318–1321
Outpatient management of patients with low-risk upper-gastrointestinal haemorrhage
Stanley AJ et al. • Lancet. 2009;373(9657): 42–47
Section 6
Literature
Development
Developed by Oliver Blatchford and colleagues in Glasgow, Scotland, and published in The Lancet in 2000. The score was derived from a cohort of 1,748 patients presenting to four hospitals in the West of Scotland with acute upper GI bleeding. The primary objective was to identify patients at very low risk who could be safely managed as outpatients — a major innovation at a time when universal admission was standard.
Validation & Adoption
Stanley et al. (Lancet 2009) performed the key validation study showing that a score of 0 predicted no need for intervention (blood transfusion, endoscopy, or surgery) with exceptional accuracy. BSG and ESGE guidelines now mandate GBS as the triage tool of choice for pre-endoscopic UGIB risk stratification, with score 0 enabling safe outpatient management and acute endoscopy avoidance.
