Upper GI Bleed Risk Synthesiser: Glasgow-Blatchford, AIMS65 and Rockall
Three scores from one set of admission bloods and vitals. GBS of 0 is the discharge signal · v1.01. Demographics and Vitals
2. Laboratory
Enter blood urea, not BUN. Most Indian lab slips already report blood urea in mg/dL. If your report says BUN, multiply by 2.14 to get blood urea in mg/dL first, or switch the unit here if the report is already in mmol/L.
3. Comorbidity and Mental Status
4. After Endoscopy (optional, completes the full Rockall score)
Clinical Application & Nuances
1. The Glasgow-Blatchford Score Is Built to Answer One Question
Not "how sick is this bleed" in general, but "does this patient need a hospital-based intervention at all." It contains no endoscopic information because it is meant to be used before endoscopy, often before the decision to admit. NICE CG141 supports considering a patient with a pre-endoscopy Rockall or Blatchford score of 0 for early discharge, and the Glasgow-Blatchford Score of 0 is the more extensively prospectively validated of the two for that specific decision.
2. Blood Urea, Not BUN, and Not Renal Function Either
The urea in this score is not a kidney function test. A brisk upper GI bleed digests blood into the gut, and the nitrogen load raises blood urea within hours through gut absorption, well before creatinine moves. A rising urea with a stable creatinine after a bleed is expected physiology, not acute kidney injury. As with the CURB-65 module, enter blood urea directly: Indian lab reports already give this in mg/dL, and a BUN value needs multiplying by 2.14 first.
3. AIMS65 Predicts Mortality, Not the Need for Endoscopic Therapy
Two of its five inputs, albumin and INR, are markers of hepatic synthetic function and chronic illness rather than of the bleed itself, which is why AIMS65 tracks in-hospital mortality well without saying anything about whether a lesion will need clipping or injection. Use the Glasgow-Blatchford Score for the intervention question and AIMS65 alongside it for the mortality question; they are not measuring the same thing and one does not replace the other.
4. The Rockall Score Changes Meaning Before and After Endoscopy
The clinical (pre-endoscopy) Rockall score uses only age, shock and comorbidity, and a score of 0 carries the same NICE-endorsed discharge consideration as a Glasgow-Blatchford Score of 0. The full score adds the endoscopic diagnosis and the stigmata of recent haemorrhage, and is a rebleeding and mortality tool rather than a discharge tool: a low pre-endoscopy score can rise once a visible vessel or active bleeding is found at scope, and the two numbers should not be confused with each other in the notes.
5. Exact Outcome Percentages by Score Should Be Checked Against the Original Papers
This tool reports AIMS65's mortality figures as published in the original derivation and validation cohorts. The Rockall score's original paper reports detailed rebleeding and mortality rates by exact score, but the point of this pearl is honesty about sourcing: before quoting a specific percentage from the Rockall paper to a patient or writing it into a discharge summary, check it against Rockall TA, Logan RF, Devlin HB, Northfield TC. Gut. 1996;38(3):316-321 directly rather than a secondary table, this one included. What is not in doubt is the ordinal message: a pre-endoscopy or full Rockall score of 2 or less is the widely used low-risk threshold.
6. What Indian Comparative Studies Have Found
Two prospective Indian cohorts comparing all three scores side by side found that no single score wins on every outcome. In a Mumbai cohort, the Rockall score performed best for predicting mortality while the Glasgow-Blatchford Score performed best for identifying genuinely low-risk patients. In a larger Delhi cohort spanning both variceal and non-variceal bleeding, every score performed markedly worse in variceal bleeding than in non-variceal bleeding, AIMS65 included, which is the more clinically important finding for an Indian caseload with a substantial variceal-bleed burden: none of these three scores was derived in, or should be leaned on heavily for, a patient bleeding from varices.
Algorithm References & Evidence Base
- Blatchford O, Murray WR, Blatchford M. A risk score to predict need for treatment for upper-gastrointestinal haemorrhage. Lancet. 2000;356(9238):1318-1321.
- Stanley AJ, Ashley D, Dalton HR, et al. Outpatient management of patients with low-risk upper-gastrointestinal haemorrhage: multicentre validation and prospective evaluation. Lancet. 2009;373(9657):42-47.
- National Institute for Health and Care Excellence. Acute upper gastrointestinal bleeding in over 16s: management. NICE guideline CG141. 2012 (updated 2016).
- Saltzman JR, Tabak YP, Hyett BH, Sun X, Travis AC, Johannes RS. A simple risk score accurately predicts in-hospital mortality, length of stay, and cost in acute upper GI bleeding. Gastrointest Endosc. 2011;74(6):1215-1224.
- Rockall TA, Logan RF, Devlin HB, Northfield TC. Risk assessment after acute upper gastrointestinal haemorrhage. Gut. 1996;38(3):316-321.
- Chandnani S, Rathi P, Sonthalia N, et al. Comparison of risk scores in upper gastrointestinal bleeding in western India: a prospective analysis. Indian J Gastroenterol. 2019;38(2):117-127.
- Rout G, Sharma S, Gunjan D, Kedia S, Nayak B, Shalimar. Comparison of various prognostic scores in variceal and non-variceal upper gastrointestinal bleeding: a prospective cohort study. Indian J Gastroenterol. 2019;38(2):158-166.
- Dhir V, Shah R. Scoring systems for upper gastrointestinal bleeding: which one scores better? [editorial]. Indian J Gastroenterol. 2019;38(2):95-97.
How to Cite This Tool
AMA Style:
Umakanth S. Upper GI Bleed Risk Synthesiser: Glasgow-Blatchford, AIMS65 and Rockall. MEDiscuss. Published 2026. Accessed .
Vancouver Style:
Umakanth S. Upper GI Bleed Risk Synthesiser: Glasgow-Blatchford, AIMS65 and Rockall [Internet]. MEDiscuss.org; 2026 [cited ]. Available from:
Save this calculation
Whatever you type here stays on this device and is not sent anywhere. The server receives only a scrambled code made from it, so nobody with access to the server can tell which patient a saved calculation belongs to. Enter the same nickname the next time to see that patient's earlier values. What is stored