King’s College Criteria Calculator
King's College Criteria Calculator
Apply O’Grady’s transplant-referral criteria in acute liver failure, on the correct arm for the aetiology, with Bernal’s lactate addition.
King's College criteria
Aetiology → arm → verdictParacetamol-induced. pH 7.34, INR 7.2, creatinine 340 µmol/L, grade III encephalopathy, lactate 2.4 mmol/L
The two arms
Non-paracetamol: INR > 6.5 OR any three of five minor criteria
- pH < 7.30
- arterial, after fluid resuscitation, more than 24 hours after ingestion — paracetamol arm only
- The triad
- INR > 6.5 (PT > 100 s) with creatinine > 300 µmol/L and grade III/IV encephalopathy — all three, not two
- Lactate > 3.0
- mmol/L, arterial, after full fluid resuscitation at about 12 hours (Bernal 2002) — not one of O’Grady’s original criteria
- INR > 6.5
- PT > 100 s — satisfies the non-paracetamol arm alone, irrespective of encephalopathy grade
- The five minor criteria
- age < 10 or > 40; seronegative hepatitis or idiosyncratic drug reaction; jaundice to encephalopathy > 7 days; INR > 3.5 (PT > 50 s); bilirubin > 300 µmol/L
Worked example
Paracetamol-induced. pH 7.34, INR 7.2, creatinine 340 µmol/L, grade III encephalopathy, lactate 2.4 mmol/L
pH 7.34 is not below 7.30 → the acidosis criterion is not met
INR 7.2 > 6.5, creatinine 340 > 300 µmol/L, encephalopathy grade III → all three of the triad are present
Lactate 2.4 mmol/L is below 3.0 → Bernal's criterion adds nothing here
→ criteria met on the triad alone, with a normal pH and a normal lactate
The criteria, arm by arm
| Arm | Satisfied by | Threshold |
|---|---|---|
| Paracetamol | Arterial pH, alone | < 7.30 after fluid resuscitation, > 24 h post-ingestion |
| Paracetamol | Triad, all three together | INR > 6.5 (PT > 100 s) + creatinine > 300 µmol/L + grade III/IV encephalopathy |
| Paracetamol | Arterial lactate, alone | > 3.0 mmol/L after full resuscitation (Bernal 2002); > 3.5 mmol/L at 4 h |
| Non-paracetamol | INR, alone | > 6.5 (PT > 100 s), irrespective of encephalopathy grade |
| Non-paracetamol | Any three of five minor criteria | Age < 10 or > 40; seronegative hepatitis or idiosyncratic drug reaction; jaundice to encephalopathy > 7 days; INR > 3.5 (PT > 50 s); bilirubin > 300 µmol/L |
How well they perform
| Source | Sensitivity | Specificity |
|---|---|---|
| Bailey et al, systematic review (via the NHS DARE abstract) | 58.2% | 94.6% |
| Deranged Physiology, quoting a separate series | 68% | 88% |
| Bernal 2002, lactate > 3.5 mmol/L at 4 hours | 67% | 95% |
| Bernal 2002, pH < 7.30 or lactate > 3.0 mmol/L at 12 hours | 76% | 97% |
| Bernal 2002, criteria plus post-resuscitation lactate | 91% | not stated |
Referral criteria, not a prognosis
O’Grady’s criteria were derived to answer one operational question: which patient with acute liver failure should be discussed with a transplant centre today. They were never designed to tell a patient how likely they are to survive, and they behave badly when used that way.
The single most important property is the asymmetry. Across the published series, specificity runs between 88% and 95% while sensitivity runs between 58% and 68%. A patient who meets the criteria almost certainly has a liver that will not recover without transplantation. A patient who does not meet them has been told nothing reassuring: roughly two in five of those who go on to die never satisfy the criteria at any point in their admission. This is why the criteria are a floor for referral rather than a gate, and why they are recounted at every blood gas rather than applied once.
The two arms are separate systems and blending them is the commonest error. The paracetamol arm turns on acidosis — pH below 7.30 alone is sufficient — and otherwise requires a complete triad of coagulopathy, renal failure and advanced encephalopathy. The non-paracetamol arm has no pH criterion at all, allows coagulopathy alone at a higher threshold, and otherwise counts three of five minor criteria in a patient who is already encephalopathic. An INR of 4.0 scores in the non-paracetamol arm and is worth nothing in the paracetamol arm. The selector above exists so that cannot happen.
Bernal’s 2002 lactate work is the one addition that has stuck. An arterial lactate above 3.0 mmol/L after full fluid resuscitation carried a sensitivity of 76% and a specificity of 97% on its own, and added to the criteria it lifted sensitivity from 76% to 91%. Critically, it identified non-survivors a median of six hours earlier than the criteria did — which in a disease where the window for transplantation closes over hours is the entire point. The resuscitation qualifier is not optional: an unresuscitated lactate is a circulatory measurement, not a hepatic one.
A serum phosphate above 1.2 mmol/L at 48 to 96 hours has been proposed as a further criterion and is reproduced on several calculators. It rests on a single small series and is not part of O’Grady’s criteria, so it is not built into the calculator above.
Most transplant units now run these criteria alongside dynamic assessment rather than instead of it — serial lactate and INR, the direction of the encephalopathy, the response to N-acetylcysteine — and alongside the generic severity scores. For chronic liver disease the equivalent questions are answered by MELD 3.0 and the Child-Pugh score, neither of which was derived in acute liver failure and neither of which replaces these criteria.
Frequently asked questions
What is the King’s College criteria threshold for pH?
Arterial pH below 7.30, in the paracetamol arm only, measured after fluid resuscitation and more than 24 hours after ingestion. It satisfies the arm on its own, whatever the grade of encephalopathy. The non-paracetamol arm has no pH criterion.
Can a patient who does not meet the criteria still die?
Yes, and often. Pooled sensitivity is around 58%, so roughly two in five patients who die of acute liver failure never meet the criteria. They are highly specific and poorly sensitive: a positive result is close to conclusive, a negative result is close to uninformative.
Do I need the whole triad in the paracetamol arm?
Yes. INR above 6.5, creatinine above 300 µmol/L and grade III or IV encephalopathy must all three be present. Two of the three does not satisfy the arm — although a pH below 7.30, or a post-resuscitation lactate above 3.0 mmol/L, satisfies it independently.
Which bilirubin threshold is correct in the non-paracetamol arm?
300 µmol/L, which is 17.5 mg/dL. Some reproductions print 18 mg/dL and then convert it back to 308 µmol/L; that is a rounding artefact of the conversion, not a different criterion. This calculator uses 300 µmol/L.
Is the lactate criterion part of the original criteria?
No. O’Grady’s 1989 criteria contain no lactate term. Bernal and colleagues added it in 2002 after showing it identified non-survivors about six hours earlier and raised sensitivity from 76% to 91%. Most units now use it, but it is a modification and the page labels it as one.
Related calculators
References
- O’Grady JG, Alexander GJ, Hayllar KM, Williams R. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology. 1989;97(2):439–45.
- Bernal W, Donaldson N, Wyncoll D, Wendon J. Blood lactate as an early predictor of outcome in paracetamol-induced acute liver failure: a cohort study. Lancet. 2002;359(9306):558–63.
- Bailey B, Amre DK, Gaudreault P. Fulminant hepatic failure secondary to acetaminophen poisoning: a systematic review and meta-analysis of prognostic criteria determining the need for liver transplantation. Abstracted in the NHS Database of Abstracts of Reviews of Effects (NCBI Bookshelf NBK79544).
Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.