Understanding Alcoholic Hepatitis

Alcoholic hepatitis represents acute liver inflammation triggered by heavy alcohol consumption, distinct from alcoholic cirrhosis or fatty liver disease. It develops over days to weeks and carries mortality rates of 20–50% in severe cases, particularly without intervention. Presentations range from asymptomatic laboratory abnormalities to fulminant liver failure.

Clinical features often include jaundice (yellowing of skin and eyes), hepatomegaly (enlarged liver), ascites (fluid accumulation in the abdomen), and systemic signs like fever and leukocytosis. Patients may report fatigue, anorexia, and abdominal discomfort. Laboratory findings characteristically show elevated transaminases (especially AST over ALT), hyperbilirubinemia, and prolonged prothrombin time—reflecting impaired hepatic synthetic function.

Management centres on alcohol cessation, nutritional support, and consideration of pharmacotherapy (corticosteroids, pentoxifylline, or N-acetylcysteine) based on prognostic scoring. Early identification of high-risk patients enables timely intervention and transfer to specialist centres when indicated.

Maddrey's Discriminant Function Formula

Maddrey and colleagues developed this formula in 1978 specifically to predict 28-day mortality and steroid responsiveness. It combines coagulation (prothrombin time prolongation) and synthetic liver dysfunction (bilirubin) into a single numeric score. An MDF score above 32 traditionally indicated severe disease and potential benefit from corticosteroid therapy, though modern management practices continue to evolve.

MDF = 4.6 × (PT − Reference PT) + Bilirubin

  • PT — Patient's prothrombin time, measured in seconds. Normal range is typically 11–13 seconds.
  • Reference PT — Laboratory reference (control) prothrombin time, usually 11–13 seconds depending on the testing method used.
  • Bilirubin — Total serum bilirubin level in mg/dL. Normal values range from 0.1 to 1.2 mg/dL.

Interpreting Your Discriminant Function Score

The MDF score stratifies patients into prognostic groups:

  • Score < 32: Lower risk of short-term mortality; prognosis generally favourable with supportive care.
  • Score 32–40: Intermediate risk; closer monitoring and consideration of steroid therapy warranted.
  • Score > 40: High risk; significant mortality risk within 28 days; corticosteroid therapy and intensive monitoring strongly recommended.

However, the MDF score should never be used in isolation. Clinical judgment, additional prognostic models (such as MELD or Glasgow score), and individual patient factors must all inform treatment decisions. A single elevated score does not diagnose alcoholic hepatitis—histology via liver biopsy remains the gold standard, though rarely performed in clinical practice due to coagulopathy and clinical urgency.

Key Considerations When Using This Calculator

Several practical points will help you apply this score correctly in clinical contexts.

  1. Timing of Laboratory Tests — PT and bilirubin values fluctuate during acute illness. Ensure samples are collected within a 24-hour window and obtained before any transfusions or coagulation factor replacement. Values obtained during or shortly after transfusion may not reflect the patient's true synthetic function and can mislead scoring.
  2. Method Matters for PT Measurement — Different laboratories use different reagents and methods (rabbit brain vs. human placenta thromboplastin), which affect absolute PT values. Always use the specific reference PT from your institution for the same analytical batch—never substitute a generic '12 seconds' without verification.
  3. Score Does Not Equal Diagnosis — A high MDF score in a patient with chronic liver disease or on anticoagulation may reflect background coagulopathy rather than acute alcoholic hepatitis. Always confirm the clinical syndrome (recent heavy drinking, acute jaundice, neutrophilic response) and exclude competing diagnoses like viral hepatitis or autoimmune liver disease.
  4. Complement with Other Prognostic Tools — The MDF score has moderate predictive power (AUROC ~0.80 for 28-day mortality). Pairing it with the Glasgow Alcoholic Hepatitis Score, MELD score, or serum procalcitonin often improves risk stratification and may influence decisions about escalation of care or ICU referral.

Laboratory Markers in Alcoholic Hepatitis

Prothrombin time (PT) measures the activity of the vitamin K-dependent coagulation factors (II, VII, IX, X)—all synthesised by the liver. Prolongation in alcoholic hepatitis signals impaired hepatic synthetic capacity and worse prognosis. PT typically normalises more rapidly than other markers following abstinence, making it a useful indicator of short-term recovery.

Serum bilirubin accumulates when hepatocytes cannot conjugate and excrete it efficiently. Hyperbilirubinemia in alcoholic hepatitis often correlates with cholestasis and hepatocellular damage. Severe hyperbilirubinemia (above 10 mg/dL) carries particular prognostic weight and may indicate acute decompensation requiring intensive care.

Together, these two parameters capture both dynamic coagulopathy and cholestatic dysfunction, making them sensitive markers of acute liver failure risk. Neither is perfect alone; the weighted combination (MDF formula) performs better than either variable independently.

Frequently Asked Questions

What MDF score indicates a patient needs corticosteroid treatment?

Historically, an MDF score ≥32 suggested potential benefit from corticosteroids in controlled trials. However, current practice is more nuanced. Patients with MDF >32 and absence of contraindications (active infection, GI bleeding, renal failure) are commonly offered corticosteroids based on supportive trial data. That said, recent meta-analyses suggest the benefit is modest (~5% absolute mortality reduction at 28 days), and individualised decision-making—incorporating patient factors, comorbidities, and institutional protocols—remains standard. Always discuss risks and benefits with the patient.

Can I use this calculator if the patient is on warfarin or other anticoagulants?

Anticoagulation significantly prolongs PT independently of liver synthetic function, which confounds MDF interpretation. If possible, obtain PT values before anticoagulation initiation, or suspend warfarin temporarily (with careful risk assessment) to obtain a baseline PT. Alternatively, consider the INR (International Normalised Ratio) as a more standardised measure, though INR has its own limitations in acute liver disease. Discuss the clinical scenario with your laboratory and liver specialist to select the most appropriate coagulation metric.

How accurate is the Maddrey discriminant function at predicting outcomes?

The MDF score has an AUROC of approximately 0.75–0.80 for 28-day mortality prediction, meaning it correctly classifies roughly 75–80% of patients. This is useful for risk stratification but not diagnostic on its own. Performance varies across populations and treatment eras. Modern prognostic tools like the Glasgow score or MELD often perform comparably or better. Use MDF as one input in a broader clinical assessment, not as a definitive predictor.

What is the difference between MDF and MELD score in alcoholic hepatitis?

MDF combines PT and bilirubin; MELD (Model for End-Stage Liver Disease) incorporates PT, bilirubin, and serum creatinine. MELD was originally developed for predicting mortality in cirrhosis across all causes and is now widely used as a rationing tool for organ transplant allocation. For acute alcoholic hepatitis specifically, MDF was designed with this condition in mind. Neither is perfect; MELD may better reflect renal dysfunction (hepatorenal syndrome), while MDF may capture the acute hepatic necrosis phenotype more sensitively. Many centres report both scores.

Should I repeat the MDF calculation during the hospital stay?

Yes, serial MDF calculations (e.g., at day 3, day 7, day 14) can track trajectory and help identify non-responders to corticosteroids. A persistently high or worsening score despite therapy suggests poor prognosis and may trigger escalation to intensive care or transplant evaluation. A falling MDF score within days of abstinence and treatment is reassuring and associated with improved survival.

What if the reference PT value differs from 12 seconds at my laboratory?

Always use your institution's specific reference PT value for the reagent and analyser in use. PT reference ranges typically span 11–13 seconds, but precision matters for MDF calculation. For example, using 11 seconds instead of 13 seconds could shift the MDF score by ~9 points. Double-check the reference value with your haematology or chemistry laboratory before calculating the score, or ask them to calculate MDF directly if your lab system supports it.

More health calculators (see all)