How it's calculated
8 criteria — remembered by the mnemonic PANCREAS (PaO2, Age, Neutrophils/WBC, Calcium, Renal/urea, Enzymes, Albumin, Sugar/glucose) — are each scored 1 point if present, assessed at 48 hours from admission, and summed for a total of 0-8.
Score interpretation
| Score | Interpretation |
|---|---|
| 0–2 | Predicts mild pancreatitis |
| ≥3 | Predicts severe pancreatitis |
Clinical use
- Assessed at 48 hours from admission — unlike BISAP (within 24 hours), Glasgow-Imrie requires waiting for the full 48-hour data set before it's complete.
- A score of 3 or more predicts severe pancreatitis and a higher risk of complications and mortality, generally supporting consideration of higher-level (HDU/ICU) monitoring.
- Often used alongside Ranson's Criteria and/or BISAP — the three scores use different criteria and timepoints, and clinicians often use whichever is most practical given available labs at the time.
Frequently asked questions
What does the PANCREAS mnemonic stand for?
PaO2, Age >55, Neutrophils/WBC >15, Calcium <2 mmol/L, Renal/urea >16 mmol/L, Enzymes (LDH/AST), Albumin <32 g/L, and Sugar/glucose >10 mmol/L — the 8 criteria of the Modified Glasgow (Imrie) Score.
When is the Glasgow-Imrie score assessed?
At 48 hours from admission, using the worst values recorded during that window — unlike BISAP, which is designed for use within the first 24 hours.
How is Glasgow-Imrie different from Ranson's Criteria?
Both are pancreatitis severity scores assessed partly at 48 hours, but use somewhat different criteria and cutoffs — Ranson's also has an admission-time component with 5 separate criteria (11 total), while Glasgow-Imrie uses a single set of 8 criteria all assessed together at 48 hours. Both correlate with severity; the choice often comes down to institutional preference and which labs are available.
References
- Blamey SL, et al. Prognostic factors in acute pancreatitis. Gut. 1984;25:1340-1346.
- Corfield AP, et al. Prediction of severity in acute pancreatitis: prospective comparison of three prognostic indices. Lancet. 1985;2:403-407.