Maddrey Discriminant Function Calculator — Alcoholic Hepatitis Severity

How to Use This Calculator

1. Enter the patient's prothrombin time (PT) in seconds.
2. Enter the laboratory control PT (typically 12 seconds — use your lab's value).
3. Enter total bilirubin in mg/dL.
4. Maddrey DF and steroid indication display instantly.
5. Use the STOPAH tab for evidence-based management guidance.

Formula

Example

Frequently Asked Questions

• What is the Maddrey discriminant function? ▼
  The Maddrey discriminant function (DF) is a clinical scoring system used to assess the severity of alcoholic hepatitis and identify patients who may benefit from corticosteroid therapy. It was first described by Maddrey et al. in a landmark paper published in Gastroenterology in 1978, in which the formula was derived from a randomised controlled trial of prednisolone in alcoholic hepatitis. The formula is: DF = 4.6 × (patient prothrombin time − control prothrombin time in seconds) + total bilirubin in mg/dL. The prothrombin time component reflects impaired hepatic synthesis of coagulation factors, while bilirubin reflects hepatic excretory and synthetic failure. A DF of 32 or above defines severe alcoholic hepatitis and is associated with a 30-day spontaneous mortality of approximately 35–50% without treatment. This threshold was chosen because it was the cutpoint below which corticosteroid therapy showed no survival benefit in the original trial. Scores above 54 are associated with very high mortality. The DF is straightforward to calculate bedside and requires only a PT and bilirubin, making it practical in clinical settings.
• When are corticosteroids indicated for alcoholic hepatitis? ▼
  Corticosteroids — specifically prednisolone 40 mg orally once daily for 28 days — are indicated in patients with severe alcoholic hepatitis defined by Maddrey DF ≥32, provided there are no contraindications. The pathophysiological rationale is that alcoholic hepatitis involves intense hepatic inflammation mediated by TNF-alpha, IL-1, IL-6, and neutrophil infiltration; corticosteroids suppress this inflammatory cascade. Absolute contraindications to corticosteroid use include: active bacterial or fungal infection or uncontrolled sepsis (steroids will worsen the infectious process); active gastrointestinal bleeding; active hepatitis B virus co-infection (risk of HBV reactivation); and severe uncontrolled diabetes mellitus. Renal failure (creatinine >2.3 mg/dL) is a relative contraindication — some guidelines recommend avoiding steroids due to risk of further renal deterioration. Before starting steroids, infections must be actively screened for and treated. Response to steroids is assessed at day 7 using the Lille score: a Lille score ≥0.45 indicates non-response, and steroids should be discontinued to avoid the adverse effects without benefit.
• How does Maddrey compare to MELD for alcoholic hepatitis prognosis? ▼
  Both Maddrey DF and MELD score predict short-term mortality in alcoholic hepatitis, but they have different derivation contexts and clinical applications. Maddrey DF was specifically developed in alcoholic hepatitis and uses PT directly, while MELD uses INR (derived from PT). Multiple studies have compared their prognostic accuracy in alcoholic hepatitis: MELD generally shows slightly better discriminative ability (AUROC approximately 0.77–0.82 for MELD vs 0.70–0.78 for Maddrey in alcoholic hepatitis cohorts). A MELD score of 18–20 in alcoholic hepatitis corresponds approximately to a Maddrey DF of 32 in terms of mortality risk. MELD has the advantage of being continuous (providing more granular risk stratification) and being widely used outside of alcoholic hepatitis, facilitating transplant assessment. However, Maddrey DF retains clinical importance as the validated threshold for corticosteroid initiation — the ≥32 cutpoint for prednisolone comes from Maddrey's original trial, not from MELD, and guidelines continue to use DF ≥32 as the treatment trigger despite MELD's superior prognostic performance.
• What is the Lille score and when is it used? ▼
  The Lille score is a validated model developed by Mathurin et al. and published in Gastroenterology in 2007 to predict corticosteroid response in patients with severe alcoholic hepatitis. It is calculated on day 7 of prednisolone therapy using the following formula involving: age, albumin at day 0, PT at day 0, creatinine at day 0, day-0 bilirubin, and day-7 bilirubin. The resulting score ranges from 0 to 1: a Lille score below 0.16 indicates a complete responder with excellent prognosis; scores 0.16–0.45 represent partial responders with intermediate prognosis; scores above 0.45 define non-responders with a 6-month survival of approximately 25%. For non-responders (Lille ≥0.45), the STOPAH and European guidelines recommend stopping prednisolone at day 7 to prevent the adverse effects of prolonged immunosuppression (infection, renal failure, GI bleeding) without mortality benefit. For responders (Lille <0.45), the full 28-day course of prednisolone 40 mg/day is completed. The Lille score effectively answers the clinical question "is this patient responding to steroids?" at the single most informative timepoint.
• Are there alternatives to corticosteroids in alcoholic hepatitis? ▼
  Corticosteroids (prednisolone) remain the only treatment with evidence of short-term mortality benefit in severe alcoholic hepatitis, but several alternatives and adjuncts have been evaluated. Pentoxifylline, a non-selective phosphodiesterase inhibitor that inhibits TNF-alpha synthesis, was used widely before the STOPAH trial (Thursz et al., NEJM 2015) demonstrated no mortality benefit at 28 days, 90 days, or 1 year when compared to placebo or prednisolone. Current AASLD and EASL guidelines do not recommend pentoxifylline as primary therapy. N-acetylcysteine (NAC) combined with prednisolone showed a survival benefit at 30 days but not 90 days in a randomised trial (Nguyen-Khac 2011) — some centres use it as an adjunct in severe disease but evidence is insufficient for routine recommendation. Liver transplantation is increasingly considered for highly selected patients with severe alcoholic hepatitis not responding to medical therapy — the experience from centres in France, the UK, and the United States shows excellent short-term post-transplant outcomes in carefully selected patients without prior alcohol treatment failure, though acceptance is variable. Abstinence from alcohol and adequate nutritional support (at least 1.2–1.5 g protein/kg/day) are universally recommended and improve both short and long-term outcomes.

Related Calculators