Medicine
Alcoholic Hepatitis Scoring Systems and Therapeutic Options
Quick fact
The Maddrey discriminant function (DF) was the first widely used score: a DF above 32 identifies severe alcoholic hepatitis and is associated with a 30-day mortality of about 30-40% without treatment, and it is the threshold used to consider corticosteroids.
Why this is interesting
A patient with alcoholic hepatitis can appear stable, yet their liver may be failing rapidly. How do clinicians decide who needs urgent treatment and who might recover on their own?
Read the full explanation
Understanding Alcoholic Hepatitis Scoring Systems and Therapeutic Options
Alcoholic hepatitis is an acute inflammation of the liver caused by heavy alcohol use, often with a high risk of death. Because clinical evaluation alone is unreliable, doctors use scoring systems to quantify severity. The most common are the Maddrey discriminant function (DF) and the Model for End-Stage Liver Disease (MELD). The DF uses prothrombin time (a measure of blood clotting function) and bilirubin (a yellow waste product that builds up when the liver is damaged). When DF is 32, the disease is considered severe, and corticosteroids (like prednisolone) may be the main treatment. The MELD score incorporates creatinine (kidney function), bilirubin, and INR (another clotting measure) and is also used to predict survival. For therapies, the options include: abstinence from alcohol (essential), nutritional support, and for severe cases, corticosteroids or pentoxifylline (though evidence for the latter is now weaker). Liver transplantation is an option for a highly selected group of patients who do not respond to steroids and meet strict criteria, including a commitment to sobriety.
A deeper explanation
The core of these scoring systems is that they capture the severity of liver dysfunction and its systemic consequences. The Maddrey DF is calculated as 4.6 × (prothrombin time – control time) + serum bilirubin (mg/dL). The threshold of 32 was derived from early trials showing a dramatic increase in mortality above this level. The DF is a continuous variable, but the binary cut-off simplifies clinical decisions. Corticosteroids work by suppressing the intense inflammatory response that drives liver injury. In a landmark study (Carithers et al., 1989), severe AH patients treated with prednisolone had a 28-day mortality of 6% versus 35% in placebo. However, corticosteroids are not universally effective; patients who are septic or have GI bleeding may not benefit and could be harmed. The Lille Model, a dynamic score that incorporates the change in bilirubin after 7 days of steroids, is used to identify non-responders early. For non-responders, continuing steroids is futile, and alternative treatments like early liver transplantation are considered. Therefore, scoring systems are not just for prognostication; they directly guide whether to start, continue, or stop treatment. The therapeutic options are anchored in the underlying pathophysiology: inflammation, oxidative stress, and the potential for liver regeneration. Abstinence is the cornerstone, as continued drinking worsens prognosis and may negate any benefit from therapy.