Hepatology

Maddrey's Discriminant Function

Severity of alcoholic hepatitis.

Education and reference only. Not a substitute for clinical judgement, local policy or product labelling. Always verify before clinical use. Values are calculated in your browser and never stored.

When to use

Use in alcoholic hepatitis to assess severity and whether corticosteroid therapy may be indicated.

Why use

A value ≥32 identifies severe disease with high short-term mortality.

Background

Maddrey's discriminant function (mDF) grades the severity of alcohol-related (alcoholic) hepatitis and helps decide whether corticosteroid therapy should be considered. It was one of the first prognostic tools in this condition, derived from steroid trials by Maddrey and colleagues. The modified version is calculated as 4.6 × (patient prothrombin time − control prothrombin time) + serum bilirubin in mg/dL. The two components reflect impaired hepatic synthetic function and cholestasis, which together track disease severity and short-term mortality.

Interpreting the result

A value of 32 or more defines severe alcohol-related hepatitis and is associated with high short-term (around one-month) mortality without treatment. Below 32, disease is milder and corticosteroids are generally not indicated. A score at or above the threshold may prompt consideration of corticosteroids, balanced against infection, gastrointestinal bleeding and renal impairment. Response is often reassessed after about a week using the Lille score to decide whether to continue steroids.

Worked example

A patient with alcohol-related hepatitis has a prothrombin time of 22 s against a control of 13 s, and a bilirubin of 15 mg/dL. mDF = 4.6 × (22 − 13) + 15 = 4.6 × 9 + 15 = 56.4, well above 32, indicating severe disease and high short-term mortality.

Critical actions

Bilirubin in mg/dL (divide µmol/L by 17.1). A score ≥32 may prompt consideration of corticosteroids — a specialist decision weighing infection and other risks.

Pearls / pitfalls

  • Bilirubin must be entered in mg/dL — divide a µmol/L value by about 17.1 to convert, or the score will be badly inflated.
  • Prothrombin-time reagents vary between laboratories, so the control time matters; some clinicians prefer MELD, which uses the INR, for this reason.
  • A high score identifies severe disease but the decision to use corticosteroids is a specialist judgement that must exclude active sepsis and uncontrolled bleeding.
  • Reassess steroid responders with the Lille score after about seven days, as non-responders gain little benefit and continue to carry treatment risk.

Evidence & validation

Derived from Maddrey et al. (1978) and refined in later corticosteroid trials; it underpins severity assessment in guidelines from bodies such as EASL and the American College of Gastroenterology, often used together with MELD and the Lille score.

Frequently asked questions

What mDF value indicates severe disease?

A value of 32 or more defines severe alcohol-related hepatitis with high short-term mortality. This threshold is used to consider corticosteroid therapy in suitable patients.

Why is the bilirubin unit important?

The formula requires bilirubin in mg/dL, but many UK laboratories report µmol/L. Dividing the µmol/L value by about 17.1 gives mg/dL; using the wrong unit grossly overestimates the score.

How does mDF relate to the MELD score?

Both grade severity, but MELD uses the INR, bilirubin and creatinine and is less affected by laboratory reagent variation. Many units use MELD and the Lille score alongside, or instead of, mDF.

Does a high score mean steroids are always given?

No. A score of 32 or more raises the possibility of corticosteroids, but the decision is specialist and must weigh infection, bleeding and renal function. Some patients are unsuitable despite a high score.

How is treatment response judged?

The Lille score, calculated after about a week of corticosteroids, is commonly used to identify non-responders. Non-responders gain little benefit and treatment is usually stopped.

References

  1. Maddrey WC, Boitnott JK, Bedine MS, et al. Corticosteroid therapy of alcoholic hepatitis. Gastroenterology. 1978;75(2):193–199.
  2. European Association for the Study of the Liver. EASL Clinical Practice Guidelines: management of alcohol-related liver disease. J Hepatol. 2018;69(1):154–181.

Related calculators

Need a calculator we don't have — or a custom tool?

We build evidence-led clinical calculators, dashboards and reference tools for teams.

☎ Call Get a Proposal