Alcoholic Hepatitis

Alcoholic hepatitis (alcohol-associated hepatitis, AH) is an acute inflammatory hepatic syndrome caused by heavy, sustained alcohol use — the most florid, clinically dramatic presentation on the alcohol-associated liver disease (ALD) spectrum that runs from simple steatosis through steatohepatitis to fibrosis and cirrhosis. AH can develop in a liver that is already cirrhotic (roughly half to two-thirds of biopsy-confirmed cases have underlying cirrhosis at diagnosis) or, less commonly, in a liver with steatosis or early fibrosis but no cirrhosis yet — either way, the presentation is the same: rapid-onset jaundice in a patient with a recent history of heavy drinking.^[1]

The diagnostic criteria (from the NIAAA Alcoholic Hepatitis Consortia, adopted by current AASLD and ACG guidance) require jaundice onset within the preceding 8 weeks, in the setting of heavy alcohol use — conventionally more than 3 standard drinks/day (about 40 g/day) for women or more than 4 standard drinks/day (about 50-60 g/day) for men, sustained for at least 6 months, with no more than 8 weeks of abstinence before jaundice onset.^[2] The characteristic laboratory pattern — an AST:ALT ratio typically 1.5-2:1 or higher, with both transaminases usually staying under 300-400 U/L — is one of the most useful bedside clues distinguishing AH from viral, ischemic, or drug-induced hepatitis, where transaminases are typically far higher; an AST above roughly 500 U/L should prompt a search for an alternative or superimposed cause rather than being attributed to AH alone.^[1][2]

This chapter is the dedicated deep-dive on the acute hepatic complication itself — diagnosis, severity stratification, the Lille-score-guided corticosteroid decision, nutrition, and infection screening. It is deliberately complementary to three other chapters: Alcohol-Related Disorders covers the addiction-medicine side (screening, withdrawal management, thiamine/Wernicke prevention, and ongoing anti-craving pharmacotherapy) in full and is cross-referenced rather than repeated here; Cirrhosis covers the longitudinal, ambulatory management of established cirrhosis that many AH survivors go on to need; and Acute Liver Failure covers fulminant hepatic failure in a previously normal liver, a distinct entity from AH (though the most severe AH can itself evolve into acute-on-chronic liver failure with multi-organ dysfunction).

Typical presentation

The classic presentation is subacute jaundice developing over days to a few weeks, in a patient who has been drinking heavily and often continues to drink up to the time of presentation. Accompanying features commonly include low-grade fever (from the hepatitis itself — this does not automatically mean infection, but infection must always be actively excluded rather than assumed absent, see Assessment & Diagnosis), tender hepatomegaly, anorexia and rapid weight loss, and profound malaise. Many patients are markedly malnourished at presentation — temporal and proximal muscle wasting is common even in a patient who does not look obviously thin, because ascites and fluid retention can mask underlying sarcopenia (see Non-Pharmacologic Treatment).^[2]

If underlying cirrhosis is already present, the usual stigmata of chronic liver disease may also be found (spider angiomata, palmar erythema, gynecomastia, splenomegaly, ascites) — see the Cirrhosis chapter for the full differential of these findings. Encephalopathy, when present, is a marker of severe disease and should prompt full severity reassessment (see Assessment & Diagnosis) rather than being managed as an isolated finding.

Severity spectrum

Both scores are calculated the same way for every patient at presentation (see Assessment & Diagnosis) — severity is not something to estimate from the clinical picture alone, since the degree of jaundice or malaise does not reliably track with these objective scores.^[1]

• Cumulative dose and pattern of alcohol use — both the total lifetime quantity and a binge-drinking pattern (rather than steady daily consumption of the same total amount) increase risk; most patients have been drinking heavily for a decade or more before presenting with AH.
• Female sex — women develop AH and progress to cirrhosis at a lower cumulative alcohol dose than men, likely reflecting differences in gastric alcohol dehydrogenase activity, body composition, and first-pass metabolism.
• Obesity and metabolic syndrome — act synergistically with alcohol to accelerate hepatic injury; a patient with both heavy alcohol use and metabolic risk factors can have a mixed steatohepatitis picture (see the Fatty Liver Disease chapter for the MASLD/MetALD overlap terminology).
• Concurrent viral hepatitis, particularly chronic hepatitis C, accelerates fibrosis progression and worsens outcomes when combined with heavy alcohol use.
• Genetic susceptibility — polymorphisms in PNPLA3 and TM6SF2 are associated with increased risk of alcohol-associated liver injury, though genetic testing does not currently change management.
• Malnutrition — both a consequence of AH and an independent risk factor for more severe disease and worse outcomes, creating a self-reinforcing cycle that nutritional support is specifically intended to interrupt (see Non-Pharmacologic Treatment).
• Smoking — commonly co-occurs with heavy alcohol use and is independently associated with faster fibrosis progression.
• Pre-existing cirrhosis — a patient with known alcohol-associated cirrhosis who continues drinking is at ongoing risk of a superimposed AH episode, which can present as an acute decompensation event layered on chronic disease.

Confirming the diagnosis

AH is a clinical diagnosis in the great majority of cases; liver biopsy is not routinely required and is reserved for atypical presentations (uncertain alcohol history, an atypical laboratory pattern, or genuine diagnostic uncertainty about a competing cause) given the bleeding risk from the coagulopathy that usually accompanies AH.^[2] A confident clinical ("probable") diagnosis requires all of the following:

• Heavy, sustained alcohol use — conventionally more than 3 standard drinks/day (~40 g/day) for women or more than 4 standard drinks/day (~50-60 g/day) for men, for at least 6 months, with no more than 8 weeks of abstinence before symptom onset.^[2]
• Onset of jaundice within the preceding 8 weeks.
• Serum bilirubin greater than 3 mg/dL (>51 micromol/L).
• AST:ALT ratio typically 1.5-2:1 or higher, with both AST and ALT usually under 300-400 U/L — a genuinely useful bedside distinguishing feature, since viral, ischemic, and most drug-induced causes of acute hepatitis produce far higher transaminase elevations; an AST over roughly 500 U/L should prompt reconsideration of the diagnosis or a search for a superimposed cause.^[1]
• Exclusion of other causes of acute jaundice/hepatitis — viral hepatitis serologies (hepatitis A/B/C at minimum), a directed medication and supplement history for drug-induced liver injury, and imaging (ultrasound, at minimum) to exclude biliary obstruction and characterize the liver.^[2]

Infection screening — before any severity-driven treatment decision

Because infection changes both the diagnostic picture (fever and leukocytosis can be produced by AH itself, not only by infection) and the treatment plan (active infection affects corticosteroid candidacy — see Choose Initial Treatment), every patient with suspected AH should be screened for infection at presentation: blood cultures, urinalysis and urine culture, chest x-ray, and a diagnostic paracentesis with ascitic fluid analysis and culture in any patient with ascites, using the same spontaneous-bacterial-peritonitis criteria as the Cirrhosis chapter.^[2][1] A low threshold for repeating this work-up applies to any subsequent clinical deterioration, particularly once corticosteroid therapy is underway (see Special Populations and Monitoring & Follow-Up).

Severity stratification — Maddrey Discriminant Function

The Maddrey Discriminant Function (mDF), the original and still most widely used AH severity score, was derived specifically to identify patients whose short-term mortality was high enough, and whose potential benefit from corticosteroid therapy large enough, to justify treatment.^[3]

Severity stratification — MELD

The MELD score — calculated the same way as in the Cirrhosis chapter (3.78 × ln[bilirubin, mg/dL] + 11.2 × ln[INR] + 9.57 × ln[creatinine, mg/dL] + 6.43) — is increasingly used alongside, or in some centres instead of, mDF: it performs at least as well at predicting short-term mortality and has the practical advantage of being calculated automatically by most laboratory information systems.^[1] A MELD score over 20 should also prompt consideration of corticosteroid therapy, whether or not mDF has been calculated.^[1]

Day-7 response assessment — the Lille score

For a patient who starts corticosteroid therapy, the Lille score is calculated after 7 days of treatment to decide whether to continue or stop — a genuinely important, guideline-directed decision point rather than an optional refinement.^[4] It combines 6 variables: age, serum albumin, serum creatinine (or a term for renal insufficiency), prothrombin time, bilirubin at treatment initiation (day 0), and the evolution of bilirubin by day 7. Because it is a continuous logistic-regression model rather than a simple point-sum score, it is not reproduced here as a calculator — the same reasoning the Acute Liver Failure chapter applies to King's College Criteria — see Choose Initial Treatment for how the result is used.

Additional baseline investigations

• CBC — neutrophilic leukocytosis is common in AH itself (a reactive, hepatitis-driven finding) but must never be assumed to exclude infection; thrombocytopenia may reflect underlying portal hypertension if cirrhosis is also present.
• Renal function and electrolytes, including magnesium and phosphate — malnourished patients are at risk of refeeding-related electrolyte shifts once nutritional support begins (see Non-Pharmacologic Treatment).
• INR/coagulation studies — required for both mDF and MELD calculation.
• Formal nutritional assessment — malnutrition is the rule rather than the exception in AH and should be actively assessed (dietitian referral, functional/strength assessment) rather than inferred from body weight alone, since ascites and fluid retention mask true muscle loss.^[2]

• Confirm the diagnosis and actively exclude infection and competing causes of acute hepatitis before any severity-driven treatment decision.
• Stratify severity with the Maddrey Discriminant Function and MELD score in every patient at presentation.
• In severe AH without contraindications, initiate corticosteroid therapy and reassess response with the Lille score at day 7 — continuing therapy only in genuine responders.
• Provide aggressive nutritional support from day 1, regardless of corticosteroid candidacy.
• Screen for infection before starting corticosteroids, and maintain a low threshold to re-screen during therapy.
• Support complete alcohol abstinence — the only intervention that changes long-term survival, independent of whether corticosteroids are used or effective.
• Identify corticosteroid non-responders early and reassess goals of care, including selective early liver-transplant evaluation where appropriate.
• Prevent and manage complications: hepatorenal syndrome, hepatic encephalopathy, GI bleeding, and infection.

• Complete alcohol abstinence — the only intervention that reliably changes long-term survival after an AH episode, independent of corticosteroid response; refer every patient to addiction medicine and psychosocial support before discharge, not only those who "seem motivated." See the Alcohol-Related Disorders chapter for the full screening, withdrawal-management, and anti-craving pharmacotherapy pathway — it is not repeated here, but every patient with AH should be routed into it.^[2]
• Aggressive nutrition support, started immediately — target roughly 35-40 kcal/kg/day and 1.2-1.5 g/kg/day of protein; do not restrict protein out of a mistaken concern about hepatic encephalopathy, the same outdated and disproven practice flagged in the Cirrhosis chapter.^[2][6] Use oral nutritional supplements first; escalate to enteral tube feeding if the patient cannot meet targets by mouth (a common scenario given anorexia and encephalopathy) — parenteral nutrition is reserved for patients who cannot tolerate enteral feeding.^[1]
• Frequent small meals with a late-evening snack (roughly 50 g complex carbohydrate plus 25 g protein) to offset the accelerated overnight catabolic state seen in advanced liver disease, using the same principle as the Cirrhosis chapter.^[6]
• Give thiamine before or together with any glucose-containing fluid, feeding, or refeeding — the same Wernicke-encephalopathy-prevention rule detailed in the Alcohol-Related Disorders chapter applies with particular force here, since AH patients are both heavily alcohol-exposed and often severely malnourished at the point nutritional repletion begins.^[1]
• Screen for and correct other common micronutrient deficiencies — folate, zinc, magnesium, phosphate, and vitamin D — alongside thiamine.
• Monitor closely for refeeding syndrome (falling phosphate, magnesium, and potassium) once aggressive nutritional support begins in a severely malnourished patient.
• Vaccination status review — hepatitis A and B (if non-immune) and standard age-based adult vaccines, as in the Cirrhosis chapter — once the patient is medically stable enough for this to be addressed.
• Avoid unregulated herbal or dietary supplements and uncapped acetaminophen use, both disproportionately risky in an already-injured liver (see Drug Tables).
• Address the psychosocial and socioeconomic barriers to abstinence directly — unstable housing, lack of a support system, and untreated psychiatric comorbidity are all common in this population and materially affect the success of any abstinence plan.

Underlying cirrhosis

A large proportion of AH patients have underlying cirrhosis at diagnosis. The same severity thresholds and corticosteroid decision apply, but nutritional support is even more critical given the higher baseline malnutrition burden, and renal function should be monitored particularly closely given the added hepatorenal-syndrome risk described in the Cirrhosis chapter.^[1]

Infection and sepsis

Infection must be actively excluded before corticosteroids are started, and treated first if identified (see Choose Initial Treatment). A patient who develops infection while already on corticosteroids needs aggressive treatment of the infection; the decision to continue or stop the corticosteroid course in that setting is individualized with the treating hepatology team rather than governed by a fixed rule.^[2]

Renal impairment / hepatorenal syndrome

Significant renal failure or hepatorenal syndrome is a relative contraindication to corticosteroid therapy and independently worsens prognosis. Evaluate and manage per the hepatorenal-syndrome pathway in the Cirrhosis chapter; the decision to proceed with corticosteroids in a patient with renal dysfunction should be individualized rather than reflexively withheld or given.^[1]

Diabetes

Corticosteroid therapy predictably worsens glycemic control. Diabetes is not, by itself, a reason to withhold indicated corticosteroid therapy in severe AH, but glucose should be monitored closely and antihyperglycemic therapy adjusted proactively (including a temporary increase in insulin dose) rather than reactively once hyperglycemia is already marked.

Pregnancy

AH in pregnancy is rare but carries added risk to both mother and fetus; manage jointly with maternal-fetal medicine and hepatology. Alcohol use in pregnancy is itself an independent priority requiring urgent, dedicated counselling and referral regardless of AH severity — see the Alcohol-Related Disorders chapter for fetal alcohol spectrum disorder risk and pregnancy-specific withdrawal management. Corticosteroid safety data specific to AH treatment in pregnancy are essentially absent, so any decision to treat requires specialist input weighing the severity of the AH against the uncertainty of fetal risk.

Older adults

Older patients are at higher risk of corticosteroid-related complications — infection, hyperglycemia, and GI bleeding — and often have more advanced malnutrition at presentation. Apply the same severity-based decision framework, but monitor more closely and involve the care team early if complications emerge.

• Daily clinical assessment during hospitalization: vital signs (a new or worsening fever is infection until proven otherwise, not simply "the hepatitis"), mental status for evolving encephalopathy, and renal function/electrolytes.
• Track bilirubin daily from day 0 through day 7 in every patient on corticosteroids — this trend is what the Lille score at day 7 is built from (see Assessment & Diagnosis and Choose Initial Treatment).^[4]
• Glucose monitoring at least daily while on corticosteroid therapy, more often if hyperglycemia develops.
• Maintain a low threshold to repeat infection work-up (cultures, imaging) for any clinical deterioration, particularly during and immediately after a corticosteroid course.^[2]
• Reassess nutritional status regularly — weight (interpreted cautiously given ascites/edema), intake tracking, and functional/strength measures where available — and escalate the nutrition plan if targets are not being met.
• After discharge: arrange hepatology follow-up to track resolution or progression to cirrhosis, and confirm the patient has an active connection to addiction medicine and psychosocial support, not just a referral on paper.^[2]

1. Lille score ≥0.45 at day 7 — stop corticosteroids immediately (continuing gives no further survival benefit while still carrying infection risk); re-screen for infection, intensify nutritional support, and reassess goals of care.^[4] In carefully selected patients — favourable psychosocial profile, no response to medical therapy, low predicted risk of a return to harmful drinking — some specialized transplant centres now consider early liver transplantation without requiring the traditional fixed 6-month sobriety interval; current AASLD and ACG guidance both support individualized, centre-specific selection over a rigid time-based rule, so raise this option with the transplant/hepatology team for a genuine non-responder rather than assuming ineligibility by default.^[1][2] For a patient who is not a transplant candidate, shift the conversation toward comfort-focused, goals-of-care-directed supportive management.
2. Infection identified before corticosteroids were started — treat the infection first; initiate or resume corticosteroid consideration once it is controlled and severity criteria are still met.^[2]
3. Worsening renal function or evolving hepatorenal syndrome during treatment — evaluate per the hepatorenal-syndrome pathway in the Cirrhosis chapter and reassess corticosteroid continuation individually with the treating team.
4. GI bleeding during a corticosteroid course — hold corticosteroids, manage the bleed per standard acute GI-bleed/variceal-hemorrhage protocols, and reassess restarting individually once controlled.
5. Relapse to drinking after recovery from an AH episode — treat as an expected part of a chronic, relapsing-remitting disease rather than a treatment failure or a reason to withhold future medical care; re-engage addiction medicine promptly (see the Alcohol-Related Disorders chapter).

• Prescribing prednisone instead of prednisolone for severe AH — the unnecessary hepatic activation step should be avoided in a patient whose hepatic function is, by definition, already impaired.^[1]
• Starting corticosteroids without first actively excluding infection.
• Continuing corticosteroids past day 7 in a confirmed Lille-score non-responder (≥0.45) — this provides no further benefit and adds ongoing infection risk.^[4]
• Reflexively using pentoxifylline as if it remained an evidence-based option — current guidance recommends against it, and it is not a substitute for corticosteroids in an eligible candidate.^[5][2]
• Treating mild-moderate AH (mDF under 32 and MELD 20 or under) with corticosteroids — no benefit has been demonstrated at this severity, and the therapy carries unnecessary risk.^[1]
• Restricting protein "to protect the liver" — outdated, does not reliably help, and worsens the malnutrition and sarcopenia that independently predict poor outcomes (the same error flagged in the Cirrhosis chapter).^[6]
• Missing or underestimating malnutrition because the patient doesn't look obviously thin — ascites and fluid retention routinely mask significant underlying muscle wasting.
• Failing to give thiamine before glucose or refeeding in a malnourished, alcohol-using patient.
• Treating an AH episode purely as an acute medical event without connecting the patient to addiction medicine — abstinence is the only intervention that changes long-term outcome, independent of the corticosteroid decision.^[2]
• Assuming a fixed 6-month sobriety interval automatically rules out liver transplantation for a genuine corticosteroid non-responder, rather than raising selective early transplantation with a specialized centre.^[1][2]

How to use this table: Doses are adult doses and require patient-specific confirmation (renal/hepatic function, weight, allergy history). Pediatric dosing is weight-based per the current product monograph unless stated. Drug pricing is intentionally omitted. Select any drug name to open its complete Pharmasist Drug Database page.

Corticosteroid therapy — severe AH, Lille-score-guided

Confirm infection has been excluded (or is now controlled) before the first dose. Reassess with the Lille score at day 7 and stop immediately in a non-responder (see Choose Initial Treatment).

Adjuvant therapy

Historical / not currently recommended

Included for teaching context and to prevent reflexive reuse of an outdated protocol — this is not a current first-line or alternative option.

Nutritional and micronutrient support

Give before or together with any glucose-containing fluid, feeding, or refeeding — see the Alcohol-Related Disorders chapter for the full Wernicke-encephalopathy-prevention protocol.

Medications to avoid or use cautiously in alcoholic hepatitis

Impaired hepatic clearance, coagulopathy, and — once corticosteroids are started — added infection and hyperglycemia risk all argue for caution with these agents.