Severe Alcoholic Hepatitis Admission Order Set
Order Set Overview
Severe alcoholic hepatitis (sAH) is a life-threatening syndrome of acute hepatic decompensation in patients with chronic heavy alcohol use, characterized by jaundice, coagulopathy, and a 28-day mortality near 30% even with optimal therapy. The diagnostic threshold is a Maddrey discriminant function \u2265 32 (or MELD > 20), and the therapeutic centerpiece since the STOPAH trial is prednisolone 40 mg PO daily for 28 days with day 7 Lille reassessment to identify non-responders for steroid discontinuation. Survival depends just as much on attention to the comorbid syndromes that consistently trigger death: infection, hepatorenal syndrome, GI bleeding, malnutrition, and alcohol withdrawal. This order set assembles the admission orders for a hepatology service caring for an sAH patient: scoring, infection screen, prednisolone with contraindication checklist, Lille reassessment workflow, AKI/HRS pathway with terlipressin, nutritional initiation including refeeding prevention, alcohol withdrawal protocol, and disposition planning including transplant evaluation triggers.
The framework integrates the AASLD 2018 alcoholic liver disease guideline, the EASL 2018 clinical practice guidelines on the management of alcohol-related liver disease, the ACG 2018 ALD guideline, the STOPAH trial (NEJM 2015) defining the negative result for pentoxifylline and the modest 28-day benefit of prednisolone, the CONFIRM trial (NEJM 2021) supporting terlipressin for HRS-AKI (FDA-approved 2022), and the early liver transplant for sAH literature (Mathurin 2011, ACCELERATE-AH 2018) reshaping the contraindication landscape. This is meant as an admission template; subsequent daily ICU and floor management decisions are made at the bedside as labs and clinical course evolve.
S1. Diagnosis, Scoring, and Admission Criteria
The presumptive clinical diagnosis of alcoholic hepatitis rests on heavy alcohol use within the prior 8 weeks (typically > 40 g/day in women and > 60 g/day in men for at least 6 months), recent onset of jaundice (< 8 weeks), and characteristic laboratory abnormalities (AST > ALT, both typically < 400 U/L; AST:ALT ratio > 1.5; total bilirubin > 3 mg/dL; INR elevated; neutrophilic leukocytosis common). Liver biopsy is not routinely required when the clinical and biochemical picture is convincing, but transjugular liver biopsy is recommended when the diagnosis is uncertain or when alternative etiologies (drug-induced liver injury, autoimmune hepatitis, ischemic hepatitis) are plausible.
Severity Scoring
| Score | Formula | Severe threshold | Use |
|---|---|---|---|
| Maddrey DF | 4.6 \u00d7 (patient PT \u2212 control PT) + total bilirubin (mg/dL) | \u2265 32 | Original cutoff for steroid initiation; depends on local control PT |
| MELD | 3.78 \u00d7 ln(bilirubin) + 11.2 \u00d7 ln(INR) + 9.57 \u00d7 ln(creatinine) + 6.43 | > 20 | Universally available; better than DF for transplant prognostication |
| GAHS (Glasgow Alcoholic Hepatitis Score) | 5 variables: age, WBC, urea, PT, bilirubin (each 1\u20133 points) | \u2265 9 | Predicts steroid responsiveness; less commonly used in US |
| ABIC (Age, Bilirubin, INR, Creatinine) | Age \u00d7 0.1 + bilirubin \u00d7 0.08 + creatinine \u00d7 0.3 + INR \u00d7 0.8 | > 6.71 (intermediate); > 9 (high) | 3-tier risk; less validated than DF/MELD |
| Lille (day 7) | 6 variables: age, albumin, baseline and day 7 bilirubin, PT, creatinine | > 0.45 (non-responder) | Predicts steroid response and informs discontinuation decision |
Differential Diagnosis
| Condition | Discriminating features | Workup |
|---|---|---|
| Drug-induced liver injury (DILI) | Recent medication initiation (acetaminophen, antibiotics, supplements); ALT often > AST | Detailed medication history; LiverTox.gov; acetaminophen level |
| Autoimmune hepatitis flare | Female predominant; other autoimmune disease; high IgG; ANA, SMA, anti-LKM positive | ANA, SMA, anti-LKM, IgG; liver biopsy if needed |
| Acute viral hepatitis (HAV, HBV, HCV, HEV) | Typical viral prodrome, exposure history; serology positive | HAV IgM, HBsAg, anti-HBc IgM, HCV Ab + RNA, HEV IgM |
| Ischemic hepatitis (shock liver) | Recent shock or hypotension; AST/ALT often > 1000; rapid resolution after hemodynamics improve | Hemodynamics review; echo if suspected |
| Wilson disease (acute presentation) | Younger patient; hemolytic anemia; low ceruloplasmin; Kayser-Fleischer rings | Ceruloplasmin, 24h urinary copper, slit lamp, liver biopsy with copper quantification |
| Decompensated NAFLD/NASH | Metabolic syndrome features; less elevation in AST relative to ALT | Imaging features, FibroScan; liver biopsy if needed |
S2. Initial Workup and Infection Screen
Admission Labs (Day 0)
- CBC with differential, comprehensive metabolic panel (BUN, creatinine, electrolytes, LFTs)
- PT/INR, PTT (for Maddrey DF and MELD calculation)
- Total bilirubin, direct bilirubin, ALT, AST, ALP, GGT, albumin, total protein
- Lipase (rule out concurrent pancreatitis), magnesium, phosphate
- Lactate, ammonia (if encephalopathy present)
- Acetaminophen level (rule out concurrent ingestion)
- HBV serologies (HBsAg, anti-HBs, anti-HBc total + IgM); if HBsAg positive, add HBV DNA
- HCV antibody (reflex to HCV RNA if positive)
- HIV antigen/antibody combo
- HEV IgM (acute hepatitis E)
- HAV IgM (acute hepatitis A)
- ANA, SMA, anti-LKM, IgG (autoimmune hepatitis screen)
- Ceruloplasmin (if < 40 years old or other Wilson features)
- Iron studies, ferritin (hemochromatosis screen)
- Alpha-1 antitrypsin level
- Type and screen (anticipate transfusion need)
- Urinalysis with culture
- Toxicology screen including alcohol level, urine drug screen
Infection Screening (Mandatory Before Steroids)
- Blood cultures \u00d7 2 (separate sites)
- Urine culture
- Chest X-ray (PA + lateral)
- Diagnostic paracentesis if ascites: cell count + differential, gram stain, culture, total protein, glucose, LDH, amylase. SBP if ANC > 250 cells/mm\u00b3
- Sputum culture if pulmonary symptoms or infiltrate
- TB screening (history, CXR, IGRA if risk factors)
- Hold steroids until active infection excluded or treated for at least 48 hours
Imaging (Day 0\u20131)
- Right upper quadrant ultrasound with Doppler (assess for cirrhosis, hepatomegaly, portal vein patency, ascites, biliary tree dilation)
- Abdominal CT or MRI if focal lesions, malignancy concern, or HCC screening overdue
- Echo if suspected cardiomyopathy or shock physiology
- FibroScan or transient elastography (delayed; not reliable in acute hepatitis due to inflammation)
Critical Note
Sepsis is the most common cause of death in severe alcoholic hepatitis. Up to 25% of patients have an active infection at presentation, and another 25\u201340% develop a nosocomial infection during admission. Aggressive infection screening at baseline and surveillance during admission \u2014 including repeat paracentesis if ascites recurs and any clinical deterioration prompts repeat blood/urine cultures \u2014 is non-negotiable. Steroid therapy increases infection risk and the relative risk of fungal infections (Aspergillus, Candida).
S3. Pharmacotherapy: Prednisolone, NAC, and What Not to Use
Prednisolone (First-Line for Severe AH)
Prednisolone 40 mg PO daily for 28 days is the standard of care for severe alcoholic hepatitis (Maddrey DF \u2265 32 or MELD > 20) without contraindications. The STOPAH trial (NEJM 2015, n=1103) showed prednisolone reduced 28-day mortality (OR 0.72, p=0.06 borderline; subgroup with DF \u2265 32 OR 0.61) but with no effect on 90-day or 1-year mortality. The implication is that steroids buy time for the liver to recover but do not alter the long-term trajectory determined by abstinence and underlying cirrhosis.
Use prednisolone, not methylprednisolone, because oral prednisolone is reliably absorbed, while IV methylprednisolone has been associated with worse outcomes. After 28 days, discontinue abruptly without taper given the short duration and the need to minimize cumulative immunosuppression. If oral intake is impossible, methylprednisolone 32 mg IV daily is an acceptable substitute.
Absolute Contraindications to Steroid Therapy
- Active uncontrolled GI bleeding (control bleeding first; can initiate after 48\u201372 h hemostasis)
- Untreated active infection (treat for 48 h before initiating; recheck cultures)
- Active hepatitis B with high viral load (HBV DNA > 2000 IU/mL); start tenofovir or entecavir first
- Hepatorenal syndrome with multiorgan failure (steroids unlikely to help)
- Active TB without therapy
- Severe psychosis
N-Acetylcysteine (Adjunct, Controversial)
N-acetylcysteine (NAC) was tested in combination with prednisolone in a French trial (Nguyen-Khac NEJM 2011, n=174) showing 1-month mortality benefit (8% vs 24%) but no 6-month survival difference. AASLD 2018 acknowledges that the addition of NAC is not standard of care but is reasonable in selected centers. Dosing follows the acetaminophen overdose protocol: 150 mg/kg IV over 1 h, then 12.5 mg/kg/h \u00d7 4 h, then 6.25 mg/kg/h for additional 67 h, totaling approximately 5 days. Given the lack of long-term benefit and added cost, many centers omit NAC.
What NOT to Use
| Therapy | Trial | Verdict |
|---|---|---|
| Pentoxifylline 400 mg PO TID | STOPAH (NEJM 2015) | No mortality benefit; abandoned. Do not use as monotherapy or in combination. |
| Anti-TNF agents (infliximab, etanercept) | Naveau 2004, Boetticher 2008 | Increased infection-related mortality; do not use. |
| Propylthiouracil | Cochrane review | No benefit; do not use. |
| Anabolic steroids | Cochrane review | No benefit; do not use. |
| S-adenosylmethionine (SAMe) | Limited trials | Insufficient evidence; not recommended. |
| Insulin/glucagon infusion | Historical | Hypoglycemic risk; abandoned. |
Pearl
Use prednisolone and not methylprednisolone for the 28-day course. The pharmacokinetics differ: prednisolone is the active metabolite, while prednisone requires hepatic activation that can be impaired in severe liver dysfunction. Oral prednisolone bioavailability remains adequate even in severe AH. Reserve IV methylprednisolone 32 mg daily only for patients unable to take oral medications.
Pitfall
Skipping the infection screen because the patient looks "stable" is the most common preventable cause of steroid-related death. Up to a quarter of admitted sAH patients have an unrecognized infection at baseline. Blood cultures, urine culture, chest X-ray, and diagnostic paracentesis (if ascites) are non-negotiable; pause steroids if any culture grows or if ANC > 250 in ascites until antibiotics are started and clinical response is documented.
S4. Day 7 Lille Score and Steroid Continuation/Discontinuation
The Lille model (Louvet 2007) is the cornerstone of steroid stewardship in severe AH. After 7 days of corticosteroid therapy, calculate the Lille score using six variables: age, baseline albumin, baseline bilirubin, day 7 bilirubin, prothrombin time, and creatinine. The score ranges from 0 to 1 and identifies non-responders who derive no benefit from continued steroids and accumulate infection risk.
| Lille score | Response | 6-month mortality | Action |
|---|---|---|---|
| < 0.16 | Complete responder | ~ 8% | Continue prednisolone 40 mg daily for full 28-day course |
| 0.16\u20130.45 | Partial responder | ~ 30% | Continue prednisolone for full 28 days; intensified supportive care |
| > 0.45 | Non-responder | ~ 75% | Discontinue prednisolone; transplant evaluation if eligible; intensify supportive care |
Day 7 Reassessment Workflow
- Repeat CBC, CMP, PT/INR, total bilirubin on day 7 of steroid therapy
- Calculate Lille score (online calculator or spreadsheet)
- If Lille > 0.45: discontinue prednisolone (no taper needed); document discussion with patient and family about prognosis; activate liver transplant referral pathway if criteria met
- If Lille \u2264 0.45: continue prednisolone 40 mg daily through day 28; reassess for infection at each clinical change
- Repeat MELD score on day 7, 14, 21, 28 to track trajectory
- Consider day 4 partial Lille (using day 4 bilirubin) when faster decision-making is critical
S5. Hepatorenal Syndrome, Infection, and Bleeding Management
Hepatorenal Syndrome (HRS-AKI) Pathway
HRS-AKI (formerly HRS Type 1) develops in 10\u201320% of severe AH patients and carries a 90-day mortality near 90% without specific therapy. The 2015 International Club of Ascites criteria define HRS-AKI as: cirrhosis with ascites, AKI per AKIN/KDIGO (rise in SCr \u2265 0.3 mg/dL within 48 h or \u2265 1.5\u00d7 baseline within 7 days), no improvement after 48 h diuretic withdrawal and albumin challenge (1 g/kg up to 100 g/day), absence of shock, no current or recent nephrotoxin exposure, and absence of intrinsic renal disease (no proteinuria > 500 mg, no hematuria > 50 RBC/hpf, normal renal ultrasound).
HRS-AKI Treatment Orders
- Discontinue all diuretics, beta-blockers, and nephrotoxins (NSAIDs, IV contrast, aminoglycosides)
- Albumin challenge: 1 g/kg IV (max 100 g) on day 1, then 20\u201340 g IV daily
- If no response after 48 h: initiate terlipressin 1 mg IV q6h (max 12 mg/day); titrate up to 2 mg q6h by day 3 if no response (per CONFIRM trial)
- Where terlipressin unavailable: norepinephrine 0.5\u20133 mg/h IV titrated to MAP increase \u2265 10 mmHg + albumin OR midodrine 7.5\u201312.5 mg PO TID + octreotide 100\u2013200 mcg SC TID + albumin
- Continuous monitoring for ischemic complications (terlipressin: digital, mesenteric, cardiac)
- Daily creatinine; goal: SCr reduction by 0.3 mg/dL or normalization within 14 days
- Discontinue terlipressin if no response by day 14 or if SCr returns to baseline
- Liver transplant referral if MELD-Na \u2265 21 and patient meets criteria
Infection Management
| Infection | Empiric regimen | Adjunct |
|---|---|---|
| Spontaneous bacterial peritonitis (SBP) | Ceftriaxone 2 g IV q24h (or piperacillin-tazobactam if severe sepsis or recent antibiotic exposure) | Albumin 1.5 g/kg day 1, 1 g/kg day 3 (reduces HRS and mortality); norfloxacin or ciprofloxacin secondary prophylaxis |
| Health-care associated pneumonia | Piperacillin-tazobactam 4.5 g IV q8h (extended) + vancomycin 25\u201330 mg/kg IV load \u2192 AUC-guided | De-escalate based on cultures within 48\u201372 h |
| UTI / pyelonephritis | Ceftriaxone 1\u20132 g IV daily (or piperacillin-tazobactam if pyelonephritis with sepsis) | Adjust based on urine culture and susceptibility |
| Bacteremia (unknown source) | Piperacillin-tazobactam 4.5 g IV q8h \u00b1 vancomycin if MRSA risk | Source identification (CT, ECHO if endocarditis suspected) |
| Invasive fungal (Aspergillus, Candida) | Voriconazole 6 mg/kg IV q12h \u00d7 2 \u2192 4 mg/kg q12h for Aspergillus; echinocandin (caspofungin 70 mg load \u2192 50 mg IV daily) for Candida | ID consult; consider after \u2265 7 days of broad antibiotic exposure with persistent fever |
GI Bleeding Management
Acute variceal hemorrhage in severe AH triggers a coordinated response: airway protection (intubation if altered mental status or massive hematemesis), large-bore IV access (two 16\u201318G), aggressive resuscitation with packed red blood cells (transfusion threshold Hb < 7 g/dL per restrictive Villanueva strategy unless ongoing bleeding or cardiovascular instability), correction of coagulopathy with FFP and platelets only for active bleeding (do not over-correct given fluid overload risk), octreotide 50 mcg IV bolus then 50 mcg/h infusion \u00d7 3\u20135 days, ceftriaxone 1 g IV q24h \u00d7 7 days (mortality benefit per Bernard meta-analysis), and emergent EGD within 12 hours for endoscopic band ligation. TIPS within 72 hours is recommended for early hemodynamic instability or rebleeding (Garcia-Pagan NEJM 2010 reduced mortality and rebleeding compared with standard therapy in Child B with active bleeding or Child C without contraindication).
Critical Decision Point: Steroid Hold for GI Bleeding
Active GI bleeding is an absolute contraindication to initiating prednisolone. After hemostasis is achieved (endoscopic banding, no further bleeding for 48\u201372 hours), prednisolone may be initiated if other criteria are met. For patients already on prednisolone who develop variceal hemorrhage, hold the steroid until bleeding is controlled and resume after 48 hours of stability. Do not abruptly stop without a plan; coordinate with hepatology and the day 7 Lille reassessment must still be performed if therapy is resumed.
S6. Nutrition, Withdrawal, and Supportive Care
Nutritional Support
Severe alcoholic hepatitis patients are universally malnourished, with most having sarcopenia and protein-calorie deficits. Aggressive nutrition is associated with improved survival; the AASLD recommendation is 35\u201340 kcal/kg/day and 1.2\u20131.5 g/kg/day protein beginning within 24\u201348 hours of admission.
Nutrition Orders
- Calorie target: 35\u201340 kcal/kg/day; protein 1.2\u20131.5 g/kg/day (no protein restriction \u2014 prior dogma about protein worsening encephalopathy is disproven)
- Initiate enteral feeding within 24\u201348 hours; nasogastric tube placement when oral intake < 21.5 kcal/kg/day
- Frequent small meals plus late-evening snack to prevent overnight catabolism
- Branched-chain amino acid supplementation when standard protein is insufficient or in encephalopathy
- Pre-feeding workup: thiamine 100 mg IV daily \u00d7 3 days then PO
- Folate 1 mg PO daily
- Multivitamin daily (oral or IV per ability to tolerate)
- Vitamin K 10 mg SC \u00d7 3 days if INR > 1.5 (test responsiveness; if no decline, deficiency unlikely)
- Magnesium and phosphate replacement to maintain Mg > 1.8 mg/dL and Phos > 2.5 mg/dL
- Refeeding syndrome surveillance: check phosphate, K+, Mg2+ daily for first 5 days; provide IV thiamine before glucose load
Alcohol Withdrawal Management (CIWA-Ar Protocol)
Alcohol withdrawal in severe AH carries high risk because of altered drug metabolism and propensity to encephalopathy. Use a symptom-triggered protocol with frequent CIWA-Ar assessments (q1\u20132h initially) to avoid over-sedation while preventing severe withdrawal complications (seizures, delirium tremens).
Alcohol Withdrawal Orders
- CIWA-Ar assessment q1\u20132h while patient is in active withdrawal (q4h once stable for 24 h)
- For CIWA-Ar 8\u201315: lorazepam 1 mg PO/IV q4\u20136h PRN
- For CIWA-Ar 16\u201320: lorazepam 2 mg PO/IV q4h PRN
- For CIWA-Ar > 20 (severe withdrawal/delirium tremens): lorazepam 4 mg IV q15min until CIWA-Ar < 20, then transition to scheduled dosing; consider ICU and phenobarbital if refractory
- Lorazepam is preferred over chlordiazepoxide and diazepam in cirrhosis because hepatic glucuronidation (intact in cirrhosis) avoids CYP-mediated metabolism and prolonged sedation
- Phenobarbital 10 mg/kg IV load \u2192 130 mg IV q15\u201330 min for refractory withdrawal (avoids GABA receptor downregulation issues)
- Avoid haloperidol unless agitation persists despite benzos (lowers seizure threshold)
- Thiamine 100 mg IV daily \u00d7 3 days BEFORE any glucose-containing fluid (Wernicke prevention)
- Continuous cardiac monitoring during active withdrawal; seizure precautions
- Withdrawal management is symptom-triggered, NOT prophylactic-scheduled, in liver failure
Coagulopathy and Bleeding Risk
Coagulopathy in severe AH reflects impaired hepatic synthesis of clotting factors and is best managed with targeted intervention rather than empiric correction. Routine FFP transfusion does not prevent bleeding and contributes to volume overload; reserve FFP and platelets for active bleeding or planned invasive procedures. Vitamin K 10 mg SC \u00d7 3 days is reasonable to test for vitamin K deficiency component (response within 24\u201348 h indicates deficiency; no response indicates pure synthetic failure). Consider DDAVP for procedural bleeding risk reduction; cryoprecipitate for fibrinogen < 100 mg/dL with active bleeding.
Encephalopathy Management
Hepatic encephalopathy is graded by West Haven criteria (Grade 1\u20134). Treat any precipitant first (infection, GI bleed, electrolyte derangement, medication, dehydration, constipation). Lactulose 15\u201345 mL PO/PR q4\u20136h titrated to 2\u20133 soft stools per day is first-line. Add rifaximin 550 mg PO BID for breakthrough or recurrent episodes. Avoid sedatives (benzodiazepines, opioids) when possible during encephalopathy; if benzodiazepines required for withdrawal, titrate carefully and monitor mental status closely.
S7. Disposition: Transplant Evaluation and Long-Term AUD
Liver Transplant Evaluation
Historically, liver transplantation for alcoholic hepatitis was withheld until 6 months of documented abstinence ("the 6-month rule"). The Mathurin 2011 NEJM paper demonstrating early transplant for steroid non-responders with severe AH challenged this dogma; ACCELERATE-AH (Lee 2018) confirmed that highly selected patients (first decompensation, supportive social environment, no prior alcohol-related sequelae) had 3-year survival of 79% with sustained alcohol abstinence in approximately 75%. Major transplant centers now offer early transplantation for sAH patients meeting strict selection criteria.
Early Liver Transplant Eligibility (Selected Centers)
- First episode of liver decompensation
- Strong social support and family commitment
- No prior failed alcohol treatment attempts
- Absence of severe psychiatric comorbidity
- Patient and family commitment to lifelong abstinence
- Multidisciplinary committee approval (hepatology, psychiatry, social work, transplant surgery)
- Refer all Lille non-responders (> 0.45) early; do not wait until terminal
Long-Term Alcohol Use Disorder Treatment
Sustained abstinence is the single strongest determinant of long-term survival in alcoholic hepatitis survivors. AUD pharmacotherapy is generally deferred until liver function recovers (typically 4\u20138 weeks post-discharge), then initiated with appropriate liver-specific dosing.
| Drug | Dose | Considerations in liver disease |
|---|---|---|
| Naltrexone (oral) | 50 mg PO daily | Hepatotoxicity risk in active hepatitis; defer until LFTs normalize; avoid if severe ongoing hepatitis |
| Naltrexone (IM extended-release) | 380 mg IM monthly | Same hepatotoxicity considerations; benefit of monthly dosing for adherence |
| Acamprosate | 666 mg PO TID | Renally cleared; safe in liver disease; preferred when hepatic concerns |
| Disulfiram | 250 mg PO daily | Avoid in active liver disease (hepatotoxic); reserve for highly motivated patients with monitoring |
| Baclofen | 5 mg PO TID titrated to 10\u201320 mg TID | Off-label; safest in cirrhosis (renally cleared); ALPADIR trial evidence in heavy drinkers |
| Topiramate | 25 mg PO daily titrated to 200\u2013300 mg/day | Off-label; CNS side effects; useful in some patients |
Discharge Planning
Discharge Criteria and Orders
- Hemodynamic stability without vasopressor support
- No active GI bleeding for \u2265 48 hours
- Tolerating oral nutrition with adequate caloric intake
- Stable mental status (no active encephalopathy or controlled with lactulose/rifaximin)
- Reliable outpatient follow-up arranged within 1\u20132 weeks (hepatology)
- Discharge medications: continue prednisolone if responder (complete 28-day course), lactulose for encephalopathy, rifaximin if recurrent encephalopathy, propranolol or carvedilol for variceal prophylaxis if applicable, ceftriaxone-completed and switched to ciprofloxacin for SBP secondary prophylaxis if SBP occurred
- Vaccinations: influenza, pneumococcal (PCV13 + PPSV23), hepatitis A and B if non-immune
- Outpatient AUD treatment referral (counseling, mutual aid group, psychiatry)
- Social work and case management for housing, financial, transportation barriers to follow-up
- HCC surveillance: every 6 months ultrasound + AFP if cirrhotic
- EGD for variceal screening within 6\u201312 weeks of recovery
When to Refer (Hepatology and Transplant Team)
- Same-day: All severe alcoholic hepatitis (Maddrey DF \u2265 32 or MELD > 20) on admission; HRS-AKI; uncontrolled GI bleeding; hepatic encephalopathy Grade 3\u20134
- Within 24 hours: Lille non-responders (day 7 score > 0.45) for early transplant evaluation
- Within 72 hours: All sAH patients should have hepatology consultation for management coordination and disposition planning
- Outpatient (1\u20132 weeks post-discharge): Hepatology follow-up, AUD treatment program, primary care
Protocol Experience: Severe Alcoholic Hepatitis After Lille-Score Stewardship and Early Transplant Listing
(The content below synthesizes published clinical patterns and expert consensus rather than first-person case reports.)The six-month abstinence rule is quoted to patients who will not survive six weeks, and severe alcoholic hepatitis runs on decisions made within the first week: (1) Maddrey discriminant function ≥ 32 and MELD-Na > 20 trigger same-day prednisolone initiation, not "wait and see" while ruling out infection; (2) Lille score at day 7 drives a hard go/no-go decision on continuing corticosteroids — non-responders get steroids stopped, not extended; (3) early liver transplant evaluation in selected non-responders (Mathurin protocol) has saved patients experienced clinicians would have lost a decade ago. The "must demonstrate 6 months of sobriety before listing" rule has been progressively relaxed for severe steroid-non-responder phenotypes, and outcomes have improved.
Same-day prednisolone for Maddrey ≥ 32 or MELD-Na > 20
Severe alcoholic hepatitis (Maddrey discriminant function ≥ 32 OR MELD-Na > 20) has 28-day mortality of 30-50% untreated. Seasoned clinicians initiate prednisolone 40 mg PO/IV daily on the day of severity assessment — same-day, not after a delayed infection workup. Active infection (treated bacterial sepsis controlled, GI bleeding actively bleeding, untreated viral hepatitis) is the only contraindication that delays steroid initiation. The STOPAH trial confirmed corticosteroid mortality reduction at 28 days, although the survival difference attenuates by 90 days — early steroids on day 1 maximize the window of benefit.
Lille score at day 7 — stop steroids in non-responders
Day-7 Lille score is the single most useful prognostic and decisional tool experienced practitioners have for alcoholic hepatitis. Lille < 0.45 = responder, continue prednisolone for full 28-day course. Lille ≥ 0.45 = non-responder, STOP corticosteroids (no further mortality benefit, continued infection/bleed risk). They have learned to act on day-7 Lille with the same conviction they use for surgical hemostasis — extending steroids in non-responders is harm without benefit. The Lille calculator uses age, albumin, bilirubin at admission, bilirubin at day 7, and renal function (creatinine or PT).
Early liver transplant evaluation for severe non-responders (Mathurin protocol)
For Lille-non-responder patients with severe alcoholic hepatitis, early (within 90 days of admission) liver transplant evaluation was pioneered by the 2011 Mathurin NEJM study showing 6-month survival of 77% in transplanted vs 23% in standard care. The selection criteria are strict: first liver decompensation episode, no prior listing, strong family support, abstinence pledge, psychiatric clearance, no recurrent substance use. Experienced clinicians initiate transplant team consultation in any Lille-non-responder while continuing supportive care. The "6 months of sobriety required" dogma has been progressively relaxed for this phenotype in many centers — selected patients now go to transplant within weeks of admission.
Nutrition + addiction medicine consultation are co-equal admission orders
High-protein, high-calorie nutrition (35-40 kcal/kg/day, 1.5 g/kg/day protein) via oral or NG enteral feeding is core therapy — malnutrition in alcoholic hepatitis carries independent mortality risk. Seasoned clinicians order nutrition consultation on admission, NOT after stabilization. Addiction medicine consultation is also same-day for naltrexone or acamprosate initiation and AA/SMART recovery referral. Thiamine 100 mg IV BID for 3 days, folic acid 1 mg PO daily, multivitamin daily. Naltrexone 50 mg PO daily is safe in compensated liver disease but contraindicated in decompensated cirrhosis with high transaminases — experienced practitioners default to acamprosate 666 mg PO TID for decompensated patients.
Anonymized Case: Severe Alcoholic Hepatitis, Lille Responder, Successful 28-Day Course
A woman in her late thirties with 15+ years of heavy alcohol use (8-12 drinks/day for past year) presented with progressive jaundice over 2 weeks, ascites, mild hepatic encephalopathy (grade 2 — disorientation), and decreased oral intake. Labs: bilirubin 22 mg/dL, INR 2.4, AST 165, ALT 78 (AST/ALT > 2 — alcoholic pattern), albumin 2.4, creatinine 0.9. Maddrey discriminant function 58 (severe; threshold 32). MELD-Na 28. Cultures negative, no GI bleeding, hepatitis B and C negative. The treating clinicians initiated prednisolone 40 mg PO daily on hospital day 1, IV NAC bridge for first 5 days, lactulose titrated to 3 BM/day for encephalopathy, thiamine 100 mg IV BID x 3 days, folic acid, multivitamin, and high-protein enteral nutrition via NG (35 kcal/kg/day, 1.5 g/kg protein). Day-7 Lille score: 0.32 (responder). Continued prednisolone for the full 28 days. By week 4, MELD-Na had dropped to 16, encephalopathy resolved, jaundice clearing. Discharged with addiction medicine follow-up, naltrexone 50 mg PO daily, weekly AA, hepatology follow-up at 2 weeks. At 90-day follow-up: MELD-Na 14, sustained abstinence, no recurrent decompensation. Liver transplant evaluation deferred pending 6-month sustained sobriety per local center policy.
Four Protocol Deviations Experienced Clinicians Now Default To
- Same-day prednisolone 40 mg PO/IV for Maddrey DF ≥ 32 OR MELD-Na > 20 — no delay for infection rule-out. Contraindications are active untreated bacterial sepsis, active GI bleeding, or untreated viral hepatitis (concurrent treatment OK). Day 1 initiation maximizes 28-day mortality benefit (STOPAH).
- Day-7 Lille score is the go/no-go decision for continuing corticosteroids. Lille < 0.45 = responder, complete 28-day course; Lille ≥ 0.45 = non-responder, STOP steroids (continued harm without benefit). Calculator inputs: age, admission albumin, admission bilirubin, day-7 bilirubin, renal function.
- Early liver transplant evaluation (within 90 days) for Lille-non-responders meeting Mathurin selection criteria. First decompensation, no prior listing, family support, abstinence pledge, psychiatric clearance. 6-month survival 77% vs 23% standard care. The "6 months sobriety required" dogma is being relaxed for this phenotype.
- Nutrition + addiction medicine consults are co-equal admission orders. 35-40 kcal/kg/d + 1.5 g/kg/d protein via oral or NG enteral. Naltrexone 50 mg PO daily for compensated; acamprosate 666 mg TID for decompensated cirrhosis. Thiamine 100 mg IV BID x 3 days, folate, multivitamin. NAC IV adjunct (combined NAC + prednisolone) emerging data for 1-month survival benefit.
S8. Frequently Asked Questions
Severe alcoholic hepatitis is defined by Maddrey discriminant function (DF) \u2265 32, calculated as DF = 4.6 \u00d7 (patient PT \u2212 control PT) + total bilirubin (mg/dL). Equivalent thresholds include MELD score > 20 and Glasgow Alcoholic Hepatitis Score (GAHS) \u2265 9. Use MELD when PT control is unavailable or unreliable. All three scores identify patients with 28-day mortality of approximately 30% who may benefit from corticosteroid therapy. Calculate on admission, day 7 (Lille score), and at discharge for prognosis.
Prednisolone 40 mg PO daily for 28 days is recommended (AASLD 2018, EASL 2018) for severe alcoholic hepatitis (Maddrey DF \u2265 32 or MELD > 20) without contraindications. The STOPAH trial (NEJM 2015) showed prednisolone reduced 28-day mortality (OR 0.72) without effect on 90-day or 1-year mortality, and pentoxifylline was ineffective. Use prednisolone (not methylprednisolone) because IV methylprednisolone has worse outcomes due to drug delivery differences. Contraindications include uncontrolled GI bleeding, untreated sepsis, active hepatitis B with high HBV DNA, hepatorenal syndrome with multiorgan failure, and active TB.
Lille score is calculated on day 7 of corticosteroid therapy and predicts response. Variables: age, albumin, baseline bilirubin, day 7 bilirubin, PT, and creatinine. Score range 0\u20131. A Lille score > 0.45 identifies non-responders with 6-month mortality of 75%; corticosteroids should be discontinued because continued therapy increases infection risk without survival benefit. Score \u2264 0.45 indicates response and steroids should be continued for the full 28-day course followed by abrupt discontinuation (no taper required given short course). The day 4 'partial Lille' (using day 4 bilirubin) is acceptable when faster decision-making is needed.
Infection screening is mandatory because severe AH patients have impaired immune function and steroids increase infection risk. Required workup: blood cultures \u00d7 2, urine culture, chest X-ray, diagnostic paracentesis if ascites present (cell count, gram stain, culture; SBP defined by ANC > 250 cells/mm\u00b3), HBV serologies (HBsAg, anti-HBc, HBV DNA if positive), HCV antibody, HIV, TB screening (history, chest X-ray, IGRA if risk factors). Treat any active infection before initiating steroids. SBP requires ceftriaxone 2 g IV q24h plus albumin 1.5 g/kg on day 1 and 1 g/kg on day 3.
HRS-AKI (formerly HRS Type 1) is diagnosed by AKIN/KDIGO AKI criteria (rise in creatinine \u2265 0.3 mg/dL within 48 h, or \u2265 1.5\u00d7 baseline within 7 days) plus cirrhosis with ascites, no improvement after 48 h albumin challenge (1 g/kg up to 100 g/day), absence of shock, no nephrotoxic drugs, no parenchymal renal disease (no proteinuria > 500 mg, no hematuria > 50 RBC/hpf, normal renal ultrasound). Treatment: terlipressin 1 mg IV q4\u20136h (max 12 mg/day) titrated to creatinine response per CONFIRM trial (FDA-approved 2022) plus albumin 20\u201340 g/day. Norepinephrine plus midodrine + octreotide are alternatives where terlipressin is unavailable. Liver transplant evaluation if MELD-Na high enough.
Aggressive nutrition is essential because severe AH patients are universally malnourished. Target: 35\u201340 kcal/kg/day and 1.2\u20131.5 g/kg/day protein (do NOT restrict protein \u2014 prior dogma about protein worsening encephalopathy is disproven). Initiate enteral feeding within 24\u201348 hours; nasogastric tube when oral intake < 21.5 kcal/kg/day. Pre-feeding workup: thiamine 100 mg IV daily \u00d7 3 days then PO, folate 1 mg PO daily, multivitamin daily, magnesium and phosphate replacement. Watch for refeeding syndrome (hypophosphatemia, hypokalemia, hypomagnesemia, thiamine deficiency) within 72 h of starting feeds; check phosphate, K+, Mg2+ daily for first 5 days.
Alcohol withdrawal in severe AH carries high risk because of altered drug metabolism in cirrhosis. Use lorazepam 1\u20132 mg PO/IV q4\u20136h PRN per CIWA-Ar protocol (Score \u2265 8 triggers dose); lorazepam preferred over chlordiazepoxide and diazepam because hepatic glucuronidation (intact in cirrhosis) avoids CYP-mediated metabolism. Avoid scheduled benzodiazepines unless severe withdrawal documented; symptom-triggered approach reduces over-sedation and encephalopathy. Phenobarbital is an alternative for refractory withdrawal. Avoid haloperidol unless agitation persists despite benzos. Initiate AUD treatment medications (acamprosate, naltrexone) only after liver recovery and not during acute hepatitis.
- Severe Ulcerative Colitis Flare Admission Order Set
- Severe Community-Acquired Pneumonia (CAP) Admission Order Set: ICU Triage, Antibiotics, Adjuncts, and Day-by-Day Workflow
- Hepatic Encephalopathy Admission Order Set: AASLD Lactulose Rifaximin Ammonia Workup and Precipitant Reversal
- Alcohol Withdrawal Admission Order Set
- Caustic Ingestion Admission Order Set: Initial Resuscitation, Endoscopy Timing, Zargar Grading, and Stricture Surveillance
- STEMI Admission Order Set: Door-to-Balloon Pathway and Post-PCI Care
Primary guideline documents this article draws on. Links resolve via DOI to the publisher; access may require a subscription.
- AASLD 2019 Practice Guidance: Alcohol-Associated Liver Disease (Hepatology · 2020)