Alcoholic liver disease

When should I suspect my patient has alcoholic liver disease?

Alcoholic liver disease includes a spectrum of liver injury including steatosis, acute alcoholic hepatitis, and chronic hepatitis with progressive fibrosis eventually leading to cirrhosis. The diagnosis of alcoholic liver disease is based on a history of heavy alcohol use, the presence of signs or symptoms suggestive of liver disease, and derangements in liver biochemical tests.

The amount of alcohol consumption needed to cause liver injury remains unclear. The risk of chronic liver disease with progression to cirrhosis increases significantly in men who consume more than 60-80g of alcohol per day for more than 10 years. Women are more sensitive to alcohol and will develop liver related problems with lesser intake. As a point of reference, the NIH has defined a standard drink (12 oz regular beer, 5 oz wine, 1.5 oz shot of 80-proof spirit) as a drink that contains about 14g of alcohol.

Table I.

History	Qualifications; severity	Frequency of findings
Alcohol intake	>80 g/day; decades of heavy use	Always present
Jaundice	Present for days to weeks	Always present
Anorexia		Usually present

Physical examination
Fever	Usually low-grade	Often present
Tachycardia		Often present
Hypotension	Systolic BP <120 mmHg	Often present
Jaundice		Always present
Ascites	Mild to severe	Usually present
Firm liver edge		Often present
Bruit	Heard over the liver; due to increased blood flow	Occasionally present
Edema of legs	Absent to marked	Usually present
Confusion, encephalopathy	Absent to marked	Often present

Laboratory
WBC	>10,000/cm; PMN >6,400/cm	Usually present
Hemoglobin	<13g/dL	Usually present
Bilirubin	>5 mg/dL (up to 25-30 mg/dL); 50% direct	Always present
AST	Elevated, but <200 IU/mL	Almost always present
ALT	Normal or elevated; <AST	Almost always present
INR	>1.2	Usually present

A tabular or chart listing of features and signs and symptoms

Table I. Signs, symptoms, and routine lab tests in alcoholic hepatitis

How can I confirm the diagnosis of alcoholic hepatitis?

The diagnosis of alcoholic hepatitis is based on a long history of heavy alcohol consumption that has continued until recently (stopped less than 4 weeks prior to illness), the presence of jaundice and ascites on physical examination, and laboratory tests revealing elevated total bilirubin and elevated serum AST (which is higher than the serum ALT). The exclusion of other causes of liver disease, such as autoimmune hepatitis, hemochromatosis, hepatitis C, hepatitis B, etc., is required. (See Table II.)

Table II.

Exclude other liver disease	Test	Interpretation/comment
Hepatitis B	HBsAg	if positive, consider flare of hepatitis B or delta hepatitis
Hepatitis C	Anti-HCV. If positive, order HCV RNA	In the United States, approximately 20-25% of patients with alcoholic hepatitis are also infected with hepatitis C virus.
Autoimmune liver disease	ANA and smooth muscle antibody	If strongly positive (>/= 1:320), consider autoimmune hepatitis.
Hemochromatosis	Iron, TIBC, ferritin	These tests are often elevated in alcoholic hepatitis despite the absence of pathologic iron overload. It is often best to postpone testing until the patient has recovered. Hemochromatosis rarely presents with jaundice and ascites.
Cultures to exclude Infection
Urine	Required	Urinary tract infection is present in up to 10% of patients on admission.
Blood	Required	Blood cultures are positive in 5-10% of patients on admission.
Chest X-ray	Recommended	Pneumonia is present in 5-10% of patients on admission.
Paracentesis	Required	Spontaneous bacterial peritonitis is present in 10% of patients on admission.
Routine blood tests
CBC with diff	Required	Elevated WBC and PMN
Liver panel	Required	Elevated bilirubin, low albumin, elevated AST
Renal function panel	Required	May have hyponatremia; check for elevated creatinine.
INR/protime	Required	INR usually >1.2
Radiological Tests
Ultrasound of liver	Recommended	Evaluate for ascites, biliary duct dilation, and masses in liver.
Doppler US of liver	Recommended	Evaluate for hepatic vein obstruction, portal vein obstruction, and hepatic artery blood flow.
Liver spleen scan	Useful, but rarely performed	Posterior view shows marked redistribution of ratio-active tracer to bone marrow (spine) and spleen (i.e., marked reduction of uptake of tracer in liver).

Liver biopsy is usually not required but should be performed if the diagnosis of alcoholic hepatitis is uncertain. Histology findings may include steatosis, ballooning degeneration, inflammation, and presence of Mallory bodies. None of these findings are diagnostic but rather supportive of the diagnosis. Patients may also have evidence of fibrosis or cirrhosis depending on extent of alcohol abuse. The Alcoholic Hepatitis Histologic Score (AHHS) is a validated scoring system to determine the risk of death in patients presenting with alcoholic hepatitis. Four parameters have been identified that are independently associated with increased 90-day mortality. These include presence of fibrosis, bilirubinostasis, neutrophil infiltration, and megamitochrondria.

What other diseases, conditions, or complications should I look for in patients with alcoholic liver disease?

Infections

Approximately 25% of patients with severe alcoholic hepatitis are infected on admission to hospital (11% with spontaneous bacterial peritonitis [SBP] or bacteremia, 8% with urinary tract infection, 3% with lung infections, and 3% with other sites of infection). Consequently, all patients should be evaluated for infection with blood cultures, paracentesis (for albumin, cell count with differential, and culture), urinalysis, and chest X-ray at admission.

The presence of SIRS on presentation is associated with a higher 90-day mortality and development of multi-organ failure, independent of the severity of alcoholic hepatitis (based on discriminant function) or degree of liver dysfunction.

Malnutrition

Malnutrition is present in almost all patients with alcoholic hepatitis, although it may not be clinically obvious. The diagnosis of clinical malnutrition is often based on the appearance of muscle wasting on inspection. More detailed examinations that may show malnutrition include evidence of muscle loss on anthropomorphic analysis, muscle weakness on strength testing, decreased creatinine-height index (requires 24-hour urine testing for creatinine), anergy to skin testing, and blood tests for lymphocyte count.

Common vitamin deficiencies include vitamin A, vitamin D, thiamine, folate, pyridoxine, and zinc. Protein calorie malnutrition increases the risk of mortality in patients with alcoholic liver disease.

Altered Mental Status

Any patient with alcoholic liver disease, including those with alcoholic hepatitis or alcohol-related cirrhosis, may present with altered mental status. The clinician must rule out direct effects of acute alcohol intoxication, alcoholic withdrawal i.e. delirium tremens, or Wernicke-Korsakoff syndrome due to thiamine deficiency which can occur in alcoholics. Furthermore, portosystemic encephalopathy (also termed hepatic encephalopathy) can occur in the setting of impaired liver function. Hepatic encephalopathy may be triggered or exacerbated in the setting of infections, excessive nitrogen (i.e. GI bleed, renal failure, constipation), metabolic alterations (i.e. hyponatremia, hypokalemia, hypoxia, dehydration), or medications (i.e. narcotics, sedatives, anticholinergics, alcohol, anesthetics). These underlying causes should be excluded in all patients presenting with altered mental status. Treatment includes administration of lactulose (available in oral and rectal formulations) titrated for 3-4 bowel movements daily and/or rifaximin 500 mg twice daily.

Gastrointestinal Bleeding

Patients with alcoholic liver disease and concomitant portal hypertension are at risk for development of gastroesophageal varices. The risk of bleeding from varices increase as the hepatic venous pressure gradient to 10-12 mmHg or greater. As with any cause of cirrhosis, patient who present with acute gastrointestinal hemorrhage should be given prophylactic antibiotics, continuous IV proton pump inhibitor and somatostatin, and upper endoscopy performed within 12 hours or once hemodynamically stabilized.

Additionally, consideration should be given to beta blockers long term if the patient can tolerate them.

Renal Dysfunction

Patients with acute alcoholic hepatitis who present with SIRS have an increased risk of developing renal failure and as a result, higher 90-day mortality. Thus, renal function should be monitored closely. As with any cause of cirrhosis, those with alcohol-related cirrhosis may also develop hepatorenal syndrome.

Patients with elevated serum creatinine should have diuretics discontinued, urinalysis performed, urine electrolytes and creatinine checked, and cultures of blood and urine taken. Patients with worsening renal function should be given intravenous albumin (up to 100 g/24 hours) and monitored closely for progressive worsening in serum creatinine, which usually signifies hepatorenal syndrome. (See Table III.)

Table III.

Malnutrition	Usually present	Low serum albumin, decreased muscle strength, decreased lymphocyte count
Infections		Can occur during treatment.
Spontaneous bacterial peritonitis	~10% on admission	Paracentesis with cell count and differential; serum and ascites albumin concentration; ascites cultures
Bacteriuria	~10% on admission	Urinalysis and urine culture
Pneumonia	~5% on admission	Chest X-ray if patient is febrile or has cough
Sepsis	~5% on admission	Blood cultures on admission
Variceal bleeding		Endoscopy if patient has evidence of upper intestinal bleeding. Elective endoscopy to evaluate for varices is often postponed until patient has improved and is able to tolerate endoscopy.
Hepatorenal syndrome		Consider HRS if serum creatinine >1.5 mg/dL. Patients need to be evaluated for infection (cultures) and intravascular volume depletion (IV fluids). Consider treating HRS with midodrine, IV albumin, and SQ octreotide. Survival after the development of HRS is unlikely.

Hepatorenal syndrome occurs in approximately 10% to 25% of patients with severe alcoholic hepatitis, and is a common cause of death. Patients with a serum creatinine of more than 1.0 mg/dL on admission, or who develop a progressive increase in serum creatinine to more than 1.4 mg/dL after admission, are at high risk of developing hepatorenal syndrome.

What is the right therapy for the patient with alcoholic liver disease?

Numerous models have been developed to prognosticate patients with alcoholic hepatitis. The most commonly referenced static, or baseline, models to determine severity include the Maddrey discriminant function (DF), Model for End-stage Liver Disease (MELD), Glasgow alcoholic hepatitis (GAH), and ABIC scores, all of which are validated models. The most commonly used dynamic model to determine response to treatment is the Lille model. A recent study has shown that a joint model using the dynamic Lille model in conjunction with one of the static models have the greatest accuracy in predicting mortality independent regardless of treatment strategy.

Maddrey’s discriminate function was the first model to assess severity of alcoholic hepatitis and is still widely used to determine whether treatment is indicated (Table IV). Maddrey’s discriminant function (DF) is calculated as: bilirubin (mg/dL) + (4.6 x prothrombin time, in seconds prolonged). A score of more than 32 signifies severe alcoholic hepatitis with a one-month mortality of 30-50%.

Table IV.

Variable	Calculation of DF	Interpretation of DF
Bilirubin (mg/dL) (To convert from mmol/L to mg/dL, divide by 17.1)	Bilirubin (mg/dL) + (4.6 x prothrombin time, in sec prolonged)	>32 signifies serious alcoholic hepatitis and recommendation to start specific treatment.
Prothrombin time (in sec). Also requires control prothrombin time (in sec).

The Model for End-Stage Liver Disease (MELD) is also used to calculate severity of alcoholic hepatitis, with a score of more than 20 signifying disease severe enough to warrant specific treatment (Table V). The calculation of MELD uses three laboratory tests (serum bilirubin, INR, and creatinine) and is best performed using an online MELD calculator or a PDA with a MELD calculator option. An accurate online calculator is at http://optn.transplant.hrsa.gov/resources/MeldPeldCalculator.asp?index=98. An increase in MELD of 2 or more points in the first week is predictive of a greater in-hospital mortality.

Table V.

Variables	Calculation of MELD	Interpretation
Bilirubin, creatinine, INR	http://optn.transplant.hrsa.gov/resources/MeldPeldCalculator.asp?index=98	Scores >22-26 signify severe alcoholic hepatitis and the recommendation to start specific drug treatment.

A Glasgow alcoholic hepatitis score (GAHS) of >9 signifies severe alcoholic hepatitis and recommendation to start specific treatment. The Glasgow score requires five variables: age, WBC, blood urea nitrogen, protime or INR, and bilirubin (Table VI).

Table VI.

Variable	1	2	3
Age (years)	<50	>=50
Leukocytes (10^9/L)	<15	>=15
Urea (mmol/L)	<5	>=5
Protime ratio or INR	<1.5	1.5 – 2.0	>2.0
Bilirubin (mmol/L) (To convert to mg/dL, divide by 17.1)	<125	125 – 250	>250

What is the most effective initial therapy?

Treatment is recommended for patients with severe alcoholic hepatitis, defined as a Maddrey’s discriminant function score of more than 32, a MELD of more than 20, or a Glasgow score of 9 or higher, or the presence of hepatic encephalopathy.

No drug is approved by the FDA for the treatment of alcoholic hepatitis.

Glucocorticosteroids (prednisolone) and pentoxifylline are the most common treatments. Experts disagree about which of these two treatments is preferable.

All patients should receive an adequate diet (at least 35-40 calories and 1.5 g of protein per kg of ideal body weight).

Abstinence is essential for long-term survival. Therefore, patients should be evaluated by a substance abuse specialist.

Listing of usual initial therapeutic options, including guidelines for use, along with expected result of therapy.

The most important therapy for patients with alcoholic liver disease is complete abstinence from alcohol as this has been shown to improve outcomes, decrease portal pressure, and reduce progression to cirrhosis if patient has not already reached that stage. Pharmacological support for maintaining abstinence has been investigated. The two most commonly used medications, acamprosate and naltrexone, are associated with a reduction in return to drinking in a recent meta-analysis that included 95 randomized controlled trials. Disulfiram has been found to decrease the number of drinking days but did not achieve long-term alcohol abstinence. Referral to an addictions specialist is typically recommended to solidify abstinence.

Patients with mild or moderate alcoholic hepatitis (i.e., defined as those with a DF <32, a MELD <20 or a Glasgow score <9) have a relatively low 6-month mortality related to liver disease and do not require specific treatment.

Patients with severe alcoholic hepatitis (i.e., defined as mDF >32, MELD >22 or Glasgow score >9) have a relatively high 30-day mortality (up to 30%) and warrant consideration for specific treatment for alcoholic hepatitis. (See Table VII and Table VIII.)

Table VII.

Drug/dose	Duration	Contraindications
Prednisolone, 30 mg, PO, per day	28 days, with or without a 14-day taper	Uncontrolled infection, hepatitis B, uncontrolled diabetes, uncontrolled peptic ulcer, hepatorenal syndrome
Pentoxifylline, 400 mg, PO, TID	28 days	Hepatorenal syndrome

Table VIII.

Nutritional support	Confirm consumption of more than 3000 KCal/day and at least 1.5 g of protein per kg ideal body weight. Consider placing a nasogastric feeding tube if the patient does not spontaneously eat a sufficient diet.
Abstinence	Abstinence is essential for long-term survival. Consider consultation with substance abuse expert. Medications to decrease alcohol craving are not routinely prescribed.

Prednisolone (40 mg daily) and pentoxifylline (400 mg three times daily) are the two most widely studied medications for treatment of severe alcoholic hepatitis and controversy remains on the use of both medications. The most recent STOPAH trial was a multicenter, randomized, double blind trial to evaluate the use of prednisolone and/or pentoxifylline in a 2×2 study design with a total 1103 patients. Patients treated with prednisolone had lower 28-day mortality in the multivariate model that controlled for prognostic variables; however, this benefit did not persist to 90-day or 12-month outcomes. Pentoxifylline had no statistically significant benefit over placebo at any of the three endpoints studied. In a meta-analysis of 22 randomized control trials, including the aforementioned STOPAH trial, corticosteroids improved short-term mortality with moderate quality evidence as did pentoxifylline but with low quality evidence. The use of corticosteroids with pentoxifylline did not provide additional mortality benefit to use of corticosteroid monotherapy. There was no mortality benefit seen beyond the initial one month. The theoretical concern that treatment with prednisolone increases the patient’s risk for developing bacterial infections or gastrointestinal bleeding has not been substantiated by published studies when compared to placebo.

Response to prednisolone treatment should be assessed after 7 days of treatment by measuring serum bilirubin and calculating the Lille score (www:lillemodel.com). Prednisolone should be discontinued in patients with a Lille score higher than 0.45, although recent data suggests that patients with a Lille score of less than 0.55 might benefit from prednisolone treatment. Patients who fail to respond to prednisolone are unlikely to respond to pentoxifylline. Prednisolone should be discontinued after 28 days (with or without a 2-week taper).

In general, pentoxifylline is continued for 28 days, without an early assessment for efficacy.

Adequate nutritional intake (protein of 1.0-1.5 g/kg ideal body weight and 35-40 calories/kg ideal body weight/day) should be assured in all patients with severe alcoholic hepatitis. Oral intake is preferred and may be administered by fine bore nasogastric feeding tube if patients cannot voluntarily consume sufficient calories. Parenteral hyperalimentation can be administered if enteral feeding is not sufficient.

Vitamin and mineral supplementation (e.g., multivitamins) may be considered to ensure sufficient vitamin intake. Sodium intake should be restricted in patients with ascites (i.e., 2 g sodium diet). Most experts believe that a standard diet (sodium restricted) or standard nutritional supplements can be used as first-line nutritional supplementation in patients with alcoholic hepatitis. Most patients with hepatic encephalopathy do not require protein restriction.

A listing of a subset of second-line therapies, including guidelines for choosing and using these salvage therapies

Liver transplantation remains an option but is controversial in acute alcoholic hepatitis. In the United States, patients are required to prove their ability to remain sober for at least 6 months and participate in a formal alcohol rehabilitation program prior to approval of liver transplant candidacy. This 6-month requirement, however, is problematic as patients with severe alcoholic hepatitis have a 6-month mortality of 70%. Both a recent North American trials and prior European trials have shown that very carefully selected patients with favorable psychosocial profiles have dramatically improved survival benefit to patients who did not receive transplant and no increase in rate of complications. The rate of recidivism in these patients was no different than patients who waited the aforementioned 6 months. The social stigma and ethical concerns still remain so early liver transplantation in alcoholics still remain controversial and is not standard of care in the United States.

Numerous medical therapies have been studied for long-term management in alcoholic liver disease but failed to show statistically significant benefit when compared to placebo. Prophylthiouracil (PTU) was used to decrease the hypermetabolic state in alcoholics but failed to show a mortality benefit, improvement in liver histology, or decrease in subsequent liver disease complications. S-adenosyl L-methionine, similarly was tested but in a Cochrane review failed to show benefit.Colchicine theoretically has anti-inflammatory and antifibrotic properties but did not show mortality benefit and furthermore had an increased risk for adverse events. Silmyarin (found in milk thistle) was shown to have potential for survival benefit and improvement in liver biochemical test and histology although a systematic review found the quality of evidence of be low. Anti-TNF agents i.e. infliximab or etanercept were used to decrease the apoptosis that occurs in acute injury but studies did not show clinical effectiveness.

Listing of these, including any guidelines for monitoring side effects.

There are limited studies on the long-term treatment of alcoholic hepatitis. Abstinence is essential; patients who return to heavy drinking can develop a second episode of alcoholic hepatitis. Some physicians prescribe pentoxifylline for 6 months or longer, although evidence that pentoxifylline is beneficial is limited. Zinc (200-400 mg/day) should be continued for at least 6 months. An evening snack may reduce muscle loss in patients with cirrhosis.

Patients with cirrhosis are candidates for endoscopic screening for esophageal varices and ultrasound surveillance for hepatocellular carcinoma.

How should I monitor the patient with alcoholic liver disease?

Patients should be monitored for hepatic encephalopathy, hepatorenal syndrome, infection (pneumonia, SBP, urinary tract infection, bacteremia), and poor nutritional intake. Adequate nutritional intake is essential and should be monitored several times per week (qualitative).

Patients should be assessed for response 7 days after starting corticosteroid treatment by calculating the Lille Score (www.lillemodel.com/score.asp), which requires patient age, lab tests at baseline (bilirubin, creatinine, IRN, albumin), and bilirubin at day 7 of corticosteroid treatment (scores are between 0-1.0). Prednisolone should be continued if the Lille Score is lower than 0.45 and stopped if the Lille Score is higher than 0.55. The effectiveness of corticosteroids for patients with a Lille Score of between 0.45 and 0.55 is uncertain, but continued corticosteroids with close monitoring is reasonable.

The short-term prognosis of severe alcoholic hepatitis is poor, with a 30-day mortality untreated up to 30%, which decreases to 15% with corticosteroid treatment. Six-month mortality with corticosteroid treatment is approximately 30%. Some patients who stop drinking can survive long term and may have a significant improvement in liver histology. However, many patients with alcoholic hepatitis have underlying cirrhosis, which may limit their ability to improve, despite abstinence.

What's the evidence?

Altamirano, J, Miquel, R, Katoonizadeh, A. “A histologic scoring system for prognosis of patients with alcoholic hepatitis”. Gastroenterology. vol. 146. 2014. pp. 1231-1239. (A validated scoring system for patients with alcoholic hepatitis. For patients who received a liver biopsy, the 90-day mortality was higher in those with fibrosis, bilirubinostasis, neutrophil infiltration, and megamitochondria.)

Michelena, J, Altamirano, J, Abraldes, JG. “Systemic inflammatory response and serum lipopolysaccharide levels predict multiple organ failure and death in alcoholic hepatitis”. Hepatology. vol. 62. 2015. pp. 762-772. (Study showing that SIRS, even in the absence of infection, if independently associated with development of multiorgan failure and greater 90 day mortality.)

Papastergiou, V, Tsochatzis, EA, Pieri, G. “Nine scoring models for short-term mortality in alcoholic hepatitis: cross-validation in a biopsy-proven cohort”. Aliment Pharmacol Ther. vol. 39. 2014. pp. 721-732. (Study that validated use of the Maddrey DF, GAHS, MELD, ABIC, and Lille scores in alcoholic hepatitis to predict 30- and 90-day outcomes. All scoring systems were found to have negative predictive value >90%.)

Louvet, A, Labreuche, J, Artu, F. “Combining data from liver disease scoring systems better predicts outcomes of patients with alcoholic hepatitis”. Gastroenterology. vol. 149. 2015. pp. 398-406. (Study to investigate the use of the different scoring systems. The best predictor of mortality was determined by joint static and dynamic models.)

Jonas, DE, Amick, HR, Feltner, C. “Pharmacotherapy for adults with alcohol use disorders in outpatient settings: a systematic review and meta-analysis”. JAMA. vol. 311. 2014. pp. 1889-1900. (Systematic review and meta-analysis to compare the use of acamprosate and naltrexone in achieving alcohol abstinence.)

Thursz, MR, Richardson, P, Allison, M. “Prednisolone or pentoxifylline for alcoholic hepatitis”. N Engl J Med. vol. 372. 2015. pp. 1619-1628. (Largest study to date. Multicenter, double-blind, randomized trial with a 2×2 design to compare prednisolone and pentoxifylline. Decreased 28-day mortality with prednisolone but no improvement in 90-day or 1-year outcomes. Pentoxifylline had no survival benefit.)

Singh, S, Murad, MH, Chandar, AK. “Comparative effectiveness of pharamacological interventions for severe alcoholic hepatitis: a systematic review and network meta-analysis”. Gastroenterology. vol. 149. 2015. pp. 958-970. (Most recently published and largest systemic review and meta-analysis comparing treatment options for alcoholic hepatitis.)

Im, GY, Kim-Schluger, L, Shenoy, A. “Early liver transplantation for severe alcoholic hepatitis in the united states – A single-center experience”. Am J Transplant. 2015. (First North American study looking at early liver transplantation in patients with severe alcoholic hepatitis who are not responding to therapy and pass a thorough psychosocial evaluation. Compared to European studies previously published. Showed favorable outcomes with improved mortality benefit but no increase in complications or rates of recidivism.)

No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.

Jump to Section