Alcoholic Hepatitis

Abbreviations

ABIC

age, bilirubin, international normalized ratio, creatinine

ACLF

acute‐on‐chronic liver failure

ADH

alcohol dehydrogenase

AH

alcoholic hepatitis

Alk Phos

alkaline phosphatase

ALT

alanine aminotransferase

ANA

antinuclear antibody

ASMA

anti‐smooth muscle antibody

AST

aspartate aminotransferase

Cr

creatinine

CS

corticosteroid

CT

computed tomography

CY2E1

cytochrome P450 2E1

DAMP

damage‐associated molecular pattern

DILI

drug‐induced liver injury

EtOH

ethanol

FMT

fecal microbiota transplantation

GAHS

Glasgow Alcoholic Hepatitis Score

G‐CSF

granulocyte colony‐stimulating factor

GGT

gamma glutamyltransferase

GI

gastrointestinal

HBV

hepatitis B virus

HCC

hepatocellular carcinoma

HCV

hepatitis C virus

HIV

human immunodeficiency virus

IL

interleukin

INR

international normalized ratio

LPS

lipopolysaccharide

LT

liver transplantation

MDB

Mallory‐dense bodies

MDF

Maddrey’s discriminant function

MELD

Model for End‐Stage Liver Disease

MRCP

magnetic resonance cholangiopancreatography

MRI

magnetic resonance imaging

NAC

N‐acetylcysteine

NIAAA

National Institute on Alcohol Abuse and Alcoholism

NPP

negative predictive value

PAMP

pathogen‐associated molecular pattern

PT

prothrombin time

PTX

pentoxifylline

ROI

reactive oxygen intermediate

SBP

spontaneous bacterial peritonitis

SIRS

systemic inflammatory response syndrome

SOFA

sequential organ failure assessment

TB

tuberculosis

Tbil

total bilirubin

TGF‐β1

transcriptional growth factor beta

TLR

Toll‐like receptor

TNF‐α

tumor necrosis factor‐α

UTI

urinary tract infection

WBC

white blood cell

Alcoholic hepatitis (AH) is an entity characterized by a rapid onset of jaundice and liver‐related complications in patients with excessive alcohol use, and it has a high short‐term mortality. In the United States, the incidence of AH is on the rise with a major impact on young women. Although the diagnosis is based on a history of heavy alcohol use and its clinical features, a transjugular liver biopsy is recommended in patients with an inconsistent alcohol history or atypical findings. The Model for End‐Stage Liver Disease (MELD) score is the best laboratory‐based prognostic model to predict mortality and to define the severe forms. Pharmacotherapy is limited to corticosteroids (CSs), which improve only 28‐day survival but increase the risk for severe infections. Early liver transplantation (LT) is increasingly being accepted as a rescue therapy for nonresponders. Among survivors, alcohol abstinence is the main factor determining their long‐term transplant‐free mortality. Several ongoing clinical trials are testing novel pathophysiology‐based targeted therapies to improve survival in critically ill patients.

AH is characterized by a rapid onset of liver dysfunction, including jaundice, and portal hypertension in the setting of excessive alcohol use.¹ Undiagnosed cirrhosis exists in approximately 75% of patients with AH, and it is associated with a poor prognosis. Severe AH is associated with remarkably high mortality (up to 30% at 28 days and 50% at 6 months). Nonsevere AH cannot be considered a mild disease because it has a 12‐month mortality rate close to 15%.² In the United States, the incidence of AH is increasing among young women.³

Clinical Presentation and Diagnosis

The cardinal feature of AH is the development of recent jaundice after prolonged (usually >12 months) heavy alcohol drinking.⁴ Tender hepatomegaly and fever are common presenting signs. Many patients show signs of liver‐related complications, such as hepatic encephalopathy, ascites, or gastrointestinal (GI) bleeding. Common laboratory findings include elevated bilirubin (>3 mg/dL), a moderate elevation of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) (rarely exceeding 400 IU/mL), and an AST/ALT ratio >2.

The diagnosis of AH mainly relies on (1) the clinical features of AH; (2) a history of excessive alcohol use; and (3) excluding other alternative causative factors, such as viral hepatitis, ischemic hepatitis, drug‐induced liver injury (DILI), or autoimmune hepatitis. There is significant inconsistency in the necessity of histological confirmation between European and US societies. To establish consistency in the diagnosis of AH, a panel of experts convened by the National Institute on Alcohol Abuse and Alcoholism (NIAAA) proposed indications for liver biopsy and three diagnostic definitions for AH (Fig. 1).

FIG 1.

Diagnostic algorithm for the diagnosis of AH. In patients presenting with a clinical picture of AH, we recommend a diagnostic workup to rule out obstructive biliary disease, viral hepatitis, DILI, and autoimmune hepatitis. Then, in the absence of any confounding factors, a reasonable clinical diagnosis of AH can be made, defined by NIAAA as probable AH. If any of the confounding factors are present, a transjugular liver biopsy is recommended to confirm the diagnosis. If characteristic histological features of AH are found, a definite AH diagnosis is established. A diagnosis based on clinical suspicion even in the presence of confounding factors but without histological confirmation should be considered as possible AH.

Prognostic Scores

Several prognostic scores are used to select candidates for pharmacological therapy (in particular, prednisolone) and to predict the clinical outcomes (Table 1). Historically, Maddrey’s discriminant function (MDF) >32 was used to select patients for CS therapy; however, recent studies have shown that MELD, ABIC (age, bilirubin, international normalized ratio [INR], creatinine [Cr]), and Glasgow Alcoholic Hepatitis Score (GAHS) are better predictors of 90‐day mortality.⁵ The Lille score is useful to assess the response to CSs⁶ (Table 1). Patients with severe AH, especially those with a systemic inflammatory response syndrome (SIRS), can experience multiorgan failure, a condition referred to as acute‐on‐chronic liver failure (ACLF).⁷ The ACLF scoring system, such as sequential organ failure assessment (SOFA), chronic liver failure‐SOFA, and ACLF grading, was recently studied as a strong predictor of short‐term mortality in AH.⁸

TABLE 1.

Prognostic Models in AH

Prognostic Models	Variables	Cutoff	Studied Clinical Use	Limitation
Static
MDF	PT, bilirubin	Severe: >32	Most commonly studied score to initiate CSs (score >32)	Not including renal function
				PT variability between centers
				Low specificity, overtreatment with CSs
MELD	Bilirubin, INR, Cr	Severe: >21	Superior to MDF to predict mortality at 28 days and 6 months	Cutoff to initiate CSs not well studied
GAHS	Age, WBC, urea, INR, bilirubin	Severe: >9	When combined with MDF, increases specificity to select candidates for CSs	Not well studied in United States
ABIC	Age, bilirubin, INR, Cr	Severe: >9	High NPP to detect patients with low risk	Not studied to initiate CSs
		Low risk: <6.7		Low accuracy for mortality prediction in severe group
Dynamic
Lille	Age, albumin, Cr, PT, bilirubin day 0, bilirubin day 4 or 7	>0.45	To determine response to CSs on day 4 or 7	Complex to calculate
			Most accurate test to determine 6‐month mortality
			Optimizes prognostication when used in combination with static scores

Pathophysiology

Recent translational studies have demonstrated that a combination of pathways is involved in the pathogenesis of hepatic failure after excessive alcohol intake. These include but are not limited to: (1) the direct toxic effect of alcohol metabolites triggering cascades of an unbalanced systemic inflammatory response, (2) intestinal microbiome dysbiosis and an altered intestinal barrier resulting in translocation of pathogen‐associated molecular patterns (PAMPs), and (3) inefficient liver regeneration and hepatocyte dedifferentiation (Fig. 2).

FIG 2.

Pathogenesis of AH. (A) Alcohol directly affects hepatocytes by metabolizing to acetaldehyde and ROIs, which subsequently results in hepatocyte necrosis. Injury to hepatocytes, in turn, releases DAMPs, which trigger acute inflammatory response in the liver by activating the caspase‐1 complex in Kupffer cells. Caspase‐1 generates a cascade of proinflammatory cytokines, such as IL‐1b, TNF‐α, and IL‐10. (B) Alcohol can directly compromise the intestinal barrier by downregulating tight junctions in the intestinal epithelial cells. This, in turn, increases intestinal permeability and facilitates translocation of bacterial degrades products, so called PAMPs including LPS. PAMPs can trigger an acute inflammatory response in a similar way as DAMPs. (C) Impairment of hepatocyte regeneration and hepatocyte dedifferentiation have been shown in patients with AH. Failure to differentiate will result in liver progenitor cell expansion (ductular reaction), which is mainly secondary to lack of liver‐specific RNA transcription factors, such as TGF‐β1. (D) Gut microbiota can be altered by alcohol. Bacterial dysbiosis secondary to excessive alcohol use can result in an altered bile acid profile, which, in turn, can further enhance bacterial dysbiosis. Furthermore, alcohol can shift the intestinal microbiota toward more pathogenic organisms and decreased biodiversity. Gut dysbiosis can affect immune response and deteriorate the inflammatory response in AH. Dysbiosis is not limited to bacterial organisms, and intestinal fungal dysbiosis has been shown in patients with AH. A significant decrease in fungal biodiversity in the gut and an increase in fungi such as Candida have been shown in cases of severe AH.

Pharmacological Therapy

CSs

Prednisolone (40 mg/day for 4 weeks) modestly increases 1‐month survival.⁹ However, it can favor severe bacterial and fungal infections.¹⁰ Although it is recommended in most guidelines for patients with MELD >21, many experts are reluctant to administer steroids due to concerns about potential adverse reactions (particularly infection) and their unclear benefit. Active sepsis and GI bleeding are considered contraindications. The response to therapy should be assessed at 4 or 7 days using the Lille score (Table 2).⁶

TABLE 2.

Workup Before Initiation of CS Therapy in AH

Workup before initiation of CSs

Blood cultures Chest X‐ray Urinalysis and urine culture Abdominal imaging Diagnostic paracentesis in case of presence of ascites (ascitic fluid culture)

Contraindications to consider

Active infection (SBP, pneumonia, cellulitis, UTI) Severe kidney injury Active GI bleeding Concomitant HBV, HCV, DILI, acute pancreatitis, HIV, TB Multiorgan failure or shock

Pentoxifylline

Pentoxifylline (PTX) monotherapy or as adjunctive therapy to CSs did not provide any improvement in either short‐term or long‐term mortality.

N‐acetylcysteine

The addition of N‐acetylcysteine (NAC) to prednisolone seems to reduce 2‐month mortality by preventing severe infections and renal failure.¹¹ Additional trials are warranted to further investigate the benefits of NAC add‐on therapy.

Early LT

Historically, a “6‐month sobriety rule” has been required before listing patients with alcoholic liver disease for LT.¹² Patients with severe AH with unsuccessful medical therapy have an up to 70% mortality rate within 6 months of diagnosis; thus, the 6‐month rule is not applicable in these patients.¹³ In a multicenter study in Europe, 26 patients with severe AH with unsuccessful CS therapy were selected for early LT with excellent 6‐month survival after early LT compared with conservative therapy (77% versus 23%).¹⁴ Notably, only 2% of the patients with severe AH underwent LT because of stringent eligibility criteria implemented in an attempt to minimize the risk for an alcohol use relapse. A retrospective review of patients with severe AH who underwent early LT from 12 US centers showed a survival rate of 94% at 1 year and 84% at 3 years.¹⁵ The incidence rate of sustained alcohol use 1 year after LT in this cohort was approximately 10%. Despite excellent clinical outcomes, there is an ongoing debate on the appropriateness of allocating organs to patients with AH and the precise criteria for patient selection.

Abstinence

Among patients who survive an episode of AH, alcohol consumption is the main driver of long‐term outcomes. Hence maintaining abstinence is the single most important factor that improves long‐term transplant‐free survival.¹⁶ Early referral to an addiction counselor can improve survival.¹⁷ Among pharmacological agents to reduce craving, baclofen reduces the risk for relapse in patients with alcohol‐associated cirrhosis. Acamprosate is probably safe but has not been studied in AH and is contraindicated in patients with severe renal injury.¹⁸

Nutritional Support

Protein‐calorie malnutrition and deficiencies in certain micronutrients, such as folate, thiamine, vitamin D, and zinc, have been reported in patients with AH. In a recent randomized clinical trial, intensive nutrition with enteral feeding did not increase survival compared with oral feeding.¹⁹ Regardless of feeding route, low daily calorie intake (less than 21.5 kcal/kg/day) with decreased protein and lipid content was associated with a higher mortality. In patients with AH, a high‐protein, high‐calorie diet and supplementation with vitamin B complex and vitamin D are recommended.

Novel Therapies and Future Direction

Clinical trials testing novel targeted therapies for AH are underway. Table 3 summarizes clinical trials testing new therapeutic approaches of AH. The most promising ones include interleukin‐22 (IL‐22), granulocyte colony‐stimulating factor (G‐CSF),²⁰ and fecal transplantation.

TABLE 3.

Novel Targeted Therapies Under Investigation for AH

Mechanism	Agent‐target	Study Arms	Target Population	Preliminary Results	ClinicalTrials.gov
Inflammatory mediators	Anakinra (IL‐1R antagonist)‐targeting lL‐1b	CS vs anakinra + zinc + PTX	MELD 20‐35	Pilot study suggests improved 3‐month survival	NCT04072822
	Canakinumab (monoclonal antibody anti‐IL‐1)	Canakinumab versus placebo	MDF > 32 and MELD < 27	Phase II, results not available	NCT03775109
Fecal dysbiosis	FMT: altered microbiome	FMT versus standard	In CS‐ineligible patients	Improved survival at 90 days	NCT03827772
Intestinal decontamination	Amoxicillin‐clavulanic acid: PAMPs	Antibiotic versus placebo	MELD > 21	Trial completed Awaiting results	NCT02281929
Hepatocyte regeneration	G‐CSF: mobilization of stem cells	G‐CSF + CSs in responders and G‐CSF without CSs in nonresponders	MDF > 32	Improvement in survival at 90 days	NCT04066179
	Peg filgrastim	Peg filgrastim + CSs or PTX versus CSs or PTX	MDF > 32	Results not available	NCT02776059
	IL‐22 (antiapoptotic)	Safety dose study	MELD 11‐28	F‐652 is safe, preliminary results show decrease in MELD, and Lille	NCT02655510