Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Prevention and Management of Alcoholic Liver Disease
Abbreviations
ALD
alcoholic liver disease
AH
alcoholic hepatitis
AUDIT
Alcohol Use Disorder Identification Test
G-CSF
granulocyte colony stimulating factor
HCV
hepatitis C virus
mDF
modified discriminant function
MELD
Model for End-Stage Liver Disease
NIAAA
National Institute of Alcoholism and Alcohol Abuse
STOPAH
Steroids or Pentoxifylline for Alcoholic Hepatitis
PTX
pentoxifylline
TNF
tumor necrosis factor
Prevention of Alcoholic Liver Disease
Alcohol is known to cause liver injury directly. The duration of alcohol consumption and the amount of alcohol consumed is the most important factor in predisposing an individual to the development of alcoholic liver disease (ALD). The National Institute of Alcoholism and Alcohol Abuse (NIAAA) defines harmful drinking above the safe limit as more than four drinks per day or more than 14 drinks per week for men and more than three drinks per day or more than seven drinks per week for women. In a meta-analysis of 15 observational studies, the risk of ALD increased exponentially with an average consumption of 50 g or more of alcohol per day for 5 to 10 years. Clearly, the most effective strategies for prevention of ALD should aim to reduce alcohol consumption among the population. Effective implementation of policies on alcohol sale in southern Europe has resulted in trends of decreasing mortality related to ALD. In contrast, similar trends have not been observed in eastern Europe and parts of northern Europe such as Ireland and the United Kingdom, where such policies on alcohol sale have not been implemented effectively.
Primary Prevention
This intervention refers to measures reducing the exposure to alcohol, the causative agent of ALD. The alcohol industry continues to grow; a network and lobbying in alcohol production, marketing and advertising, distribution and sale, and international trading has resulted in a continuous increase in alcohol consumption not only in developed countries but also in developing countries. Although much research is being done to understand the mechanisms and pharmacology of alcohol consumption behavior, there remains an urgent need to implement cost-effective policy measures ( Table 24-1 ) at federal, national, and state levels.
TABLE 24-1
Strategies and Policies to Reduce Alcohol Consumption
|
Secondary Prevention
Screening for Alcohol Consumption
This strategy refers to measures aiming to identify individuals who are exposed to the risk factor and implement measures to prevent development of ALD. A large survey from the NIAAA showed that approximately 25% of people in the United States drink above the safe limits. Screening for alcohol consumption should be performed at every face-to-face encounter of a person with the healthcare provider, including clinics, the emergency department, an inpatient setting, maternity centers, and ambulatory centers. Among individuals with self-report of alcohol consumption any time in their life, asking a single question such how many times in the past year they have had five drinks or more for men and four drinks or more for women is an efficient tool, as most heavy drinkers have engaged in this amount of drinking at some time point during the past year. CAGE ( Table 24-2 ) and Alcohol Use Disorder Identification Test (AUDIT) questionnaires are often used for formal screening. With 10 questions on the AUDIT tool, each answered on a scale from 0 to 4, a score of more than 6 for men and more than 4 for women identifies heavy drinkers with excellent reliability and validity. The best way to obtain accurate information on the AUDIT questionnaire is to let the patient complete the form.
TABLE 24-2
CAGE Questionnaire
|
Identifying at-risk drinkers provides an opportunity to
- 1.
provide brief counseling sessions for occasional heavy drinkers
- 2.
refer the patient for strategies on cessation of alcohol consumption for regular heavy drinkers (see the section entitled “ Alcohol Abstinence ”)
- 3.
evaluate the patient for any physical problems from alcohol consumption such as ALD, chronic pancreatitis, cardiomyopathy, and neuropathy
- 4.
evaluate the patent for psychosocial issues such as job performance, marital and domestic issues, depression, anxiety, and other substance use disorders
- 5.
promote a healthier lifestyle such as safe driving, smoking cessation, and exercise
Randomized studies have shown that brief counseling sessions for these individuals are beneficial and cost-effective, with reduction in alcohol consumption, improved physical and mental health, reduced hospitalizations and emergency department visits, and reduced number motor vehicle accidents.
Tertiary Prevention
This mode of prevention refers to strategies for maintaining abstinence and prevention of recidivism among patients with established ALD. Unfortunately, in patients with ALD, advanced liver disease with cirrhosis or its complications, or severe alcoholic hepatitis (AH) is frequently diagnosed at the time of initial clinical presentation. In this regard, it becomes critical to implement strategies for workup of liver disease among heavy drinkers identified during screening for alcohol consumption. However, the cost-efficacy of such an approach remains to be determined given that approximately 25% of the U.S. population consumes alcohol above the safe limit at some point in time.
Treatment of Alcoholic Liver Disease
The spectrum of ALD extends from steatosis and steatohepatitis to liver cirrhosis and its complications. Approximately 25% to 30% of individuals with ALD develop the unique syndrome of AH, presenting with acute on chronic liver failure, with a potential for high mortality of approximately 40% to 50% at 1 month from presentation. Mortality for the whole spectrum of ALD has significantly decreased in the last 3 decades. From the National Inpatient Sample database, inpatient mortality decreased from approximately 10% in 2002 to 5.8% in 2010 in one study, and in another study there was an approximately 7% annual decrease between 1998 and 2007. Similarly, there has been trend of improvement in survival of patients with alcoholic cirrhosis. For example, in one study using the National Vital Statistics data, age- and sex-adjusted incidence rates of mortality related to ALD decreased from 6.9 per 100,000 persons in 1980 to 4.4 per 100,000 persons in 2003. Although alcohol consumption in the United States and many parts of the world also decreased over the same time period, the reduction in mortality exceeds the reduction in consumption. This may reflect an improvement in the treatment of patients with ALD and its complications. This section describes the diagnosis and various treatment strategies for AH and for alcoholic cirrhosis, with focus on the current status of corticosteroid therapy, pentoxifylline (PTX) therapy, and liver transplant in the management of severe AH. We will also describe emerging pharmacologic therapies directed at the newer targets in the pathogenesis of AH, which are being evaluated by various U.S. research consortia, sponsored by the NIAAA.
Diagnosis of Alcoholic Hepatitis
AH presents with acute on chronic liver failure among patients with long-term and active alcohol consumption, and should be differentiated from decompensated ALD. The clinical features supporting this diagnosis are
- 1.
recent onset of jaundice (serum bilirubin level >3 mg/dL)
- 2.
history of heavy alcohol consumption on a long-term basis until at least 3 to 4 weeks before the onset of illness
- 3.
serum aminotransferases level greater than 50 U/L but less than 400 U/L, with a serum aspartate aminotransferase to alanine aminotransferase ratio greater than 1.5 : 1
- 4.
frequent presence of features of systemic inflammatory response syndrome such as fever, leukocytosis, tachycardia, and tachypnea ( Table 24-3 )
TABLE 24-3Initial Workup for a Patient With Decompensated Alcoholic Liver Disease and Diagnosis of Alcoholic Hepatitis- A.
Documentation of alcohol as the cause of liver disease
-
Detailed alcohol history on amount, duration, and date of last drink
-
Exclusion of other liver diseases
-
Evaluation of list of medications for potential drug-induced liver injury
-
- B.
Diagnosis of alcoholic hepatitis
-
Long-term and active heavy alcohol consumption until 3 wk before onset of illness
-
Recent onset of jaundice with bilirubin level >3mg/dL
-
Serum aminotransferase levels >50 U/L but <400 U/L, with AST/ALT ratio >1.5 : 1
-
Ultrasound examination to exclude biliary obstruction and cholangitis
-
Liver biopsy for uncertain diagnosis or with concomitant hepatotoxin/liver disease
-
- C.
Assessment for other comorbidities and complications
-
Liver disease complications (ascites, encephalopathy, and variceal bleeding)
-
Malnutrition
-
Alcohol withdrawal
-
Cardiomyopathy
-
Pancreatitis
-
Acute kidney injury and hepatorenal syndrome
-
- D.
Workup for infections *
-
Physical examination for any skin source, SBP, pneumonia, pyelonephritis
-
Blood cultures, including fungal cultures, when there is a high index of suspicion
-
Urine analysis and culture
-
Ascitic fluid examination and culture
-
Stool for Clostridium difficile infection
-
ALT, Alanine aminotransferase; AST, aspartate aminotransferase; SBP, spontaneous bacterial peritonitis.* Consider nosocomial source if infection onset during hospitalization and for patients hospitalized within the last 3 months or coming from a nursing home.
- A.
White blood cell and platelet count may be helpful in improving the accuracy of clinical diagnosis. Patients with an uncertain clinical diagnosis and those with the presence of concomitant other hepatotoxin or liver disease should undergo liver biopsy to confirm the diagnosis of AH. As these patients often have coagulopathy, a transjugular route is often used to obtain the liver tissue. Histologic findings supporting the diagnosis of AH include macrovesicular steatosis, neutrophilic lobular infiltration, hepatocyte ballooning, and Mallory-Denk bodies. In more severe cases intrahepatic cholestasis (hepatocytes, bile ductules, or bile canaliculi) and zone 3 perivenular injury with pericellular fibrosis or a “chicken-wire” pattern of fibrosis may develop ( Fig. 24-1 ). Liver biopsy may also be useful in predicting disease severity and outcome of AH with use of fibrosis stage, degree of lobular neutrophilic infiltration, location of cholestasis, and presence of megamitochondria ( Table 24-4 ). Ultrasonography should be performed to rule out bile duct obstruction and hepatocellular carcinoma. Detailed assessment of patients is crucial for evaluation of comorbidities, liver disease complications, infections, and/or sepsis (see Table 24-3 ).
TABLE 24-4
Scoring Systems to Stratify the Severity of Alcoholic Hepatitis Episodes
| Score | Severe Disease | Advantages | Limitations |
|---|---|---|---|
| mDF | ≥32 | Simple and widely used for treatment initiation | Variations in prothrombin time across laboratories, does not guide treatment, and potential for inaccuracy in mild alcoholic hepatitis |
| MELD | ≥21 | Use of international normalized ratio, widely used | Emerging consensus on cutoff for severe disease and not used widely in clinical trials |
| CTP * | ≥7 | Simple and widely used | Includes two subjective variables and not widely used |
| GAHS † | ≥9 | Simple to use | Not validated outside the United Kingdom |
| Lille | ≥0.45 | Validated in many French studies | Not validated outside France, complex formula to use, and cannot guide treatment initiation |
| ABIC | ≥9 | Simple | Not validated and no data on guiding treatment |
| AHHS ‡ | ≥5 | Accurate | Not validated and invasive with need for liver biopsy |
ABIC, Age, bilirubin concentration, international normalized ratio, creatinine concentration; AHHS, alcoholic hepatitis histologic score; CTP, Child-Turcotte Pugh; GAHS, Glasgow Alcoholic Hepatitis Score; mDF, modified discriminant function; MELD, Model for End-Stage Liver Disease.
* Serum bilirubin concentration scored from 1 to 3: less than 2, 2 to 3, and greater than 3; prothrombin time scored from 1 to 3: less than 4 seconds, 4 seconds to 6 seconds, more than 6 seconds above control; serum albumin concentration scored from 1 to 3: greater than 3.5, 2.8 to 3.5, less than 2.8; ascites scored from 1 to 3: absent, slight, tense; and encephalopathy scored from 1 to 3: none, grade I or II, grade III or IV. Total score range 5 to 15.
† Age scored as 1 or 2: younger than 50 years or 50 years or older; white blood cell count scored as 1 or 2: less than 15 or 15 or greater; blood urea nitrogen concentration scored as 1 or 2: less than 14 or 14 or greater; serum bilirubin concentration scored from 1 to 3: less than 7.3, 7.3 to 14.6, and greater than 14.6; international normalized ratio scored from 1 to 3: less than 1.5, 1.5 to 2.0, greater than 2.0. Total score range 5 to 12.
‡ Bridging fibrosis or cirrhosis scored as 3 otherwise as 0, neutrophilic infiltration absent scored as 2 and present scored as 0, bilirubinostasis scored as 0 to 2: absent or hepatocellular 0, ductular or canalicular 1, canalicular/ductular and hepatocellular 2; megamitochondria scored as 2 if absent and 0 if present. Total score range 0 to 9.
Assessment of Disease Severity
After AH has been diagnosed, the disease severity should be assessed to identify patients with a severe episode of AH and to administer specific therapy to these patients. Many scoring systems are currently available to stratify the disease severity of AH patients (see Table 24-4 ).
Modified Discriminant Function Score
Initially described by Maddrey et al. in 1989 as a discriminant function index, this score was modified with prolongation of prothrombin time over the control value instead of the use of the absolute value: 4.6 × (prothrombin time in seconds − control value) + serum bilirubin concentration in milligrams per deciliter. Patients with a modified discriminant function (mDF) score of 32 or greater are categorized as having severe AH and are considered for corticosteroid therapy. Use of this score is limited as patients identified as having a milder episode and not considered candidates for corticosteroid therapy have an approximately 7% risk of dying within 28 days.
Model for End-Stage Liver Disease Score
The Model for End-Stage Liver Disease (MELD) score is worldwide to allocate livers for patients listed for liver transplant. Many studies have shown the utility of the MELD score in predicting the outcome of AH patients, and a patient with a score of 20 or greater is considered to have severe AH. The score can be calculated on the website http://www.mayoclinic.org/meld/mayomodel7.html with use of the serum bilirubin concentration, international normalized ratio, and serum creatinine concentration. In another study, the MELD modified with the serum sodium concentration was no better than the MELD among AH patients with ascites.
Child-Turcotte-Pugh Score
Using three objective (serum bilirubin concentration, serum albumin concentration, and prothrombin time) and two subjective (hepatic encephalopathy and ascites) variables, this score categorizes a patient a having stage A, B, or C disease with a total score of 5 or 6, 7 to 9, or more than 9, respectively (see Table 24-4 ). The Child-Turcotte-Pugh score is not widely used for assessing severity of AH.
Glasgow Alcoholic Hepatitis Score
With a range from 5 to 12, a Glasgow AH score of 9 or more was shown to accurately predict survival of severe AH patients at 28 days and 84 days (see Table 24-4 ), and also predict response to corticosteroid treatment. In spite of its availability for almost a decade, this score is not widely used in routine clinical practice.
Age, Bilirubin Concentration, International Normalized Ratio, Creatinine Concentration
With use of the cutoff values of 6.71 and 9.0 for this score, AH patients can be categorized as having low, intermediate, and high risk of death at 90 days and at 1 year.
Lille Score
Derived from a prospective study of 320 biopsy-proven severe AH patients, the Lille score identified AH patients not responding to corticosteroid therapy at 1 week of therapy (see Table 24-4 ). Although not useful to guide initiation of therapy, this score is helpful in guiding treatment with corticosteroids beyond 1 week of therapy. Patients with a Lille score greater than 0.45 will likely not receive further benefit from corticosteroids and are recommended to discontinue therapy. The score can be calculated on the website http://www.lillemodel.com/score.asp with use of the patient's age, serum bilirubin concentration at day 0 and day 7 of corticosteroid therapy, serum creatinine concentration at day 0, serum albumin concentration at day 0, and prothrombin time at day 0. The website allows one to enter bilirubin and creatinine values in micromoles per liter as well as in milligrams per deciliter. In another study from the same center, the Lille score could be further categorized for high, intermediate, and low risk of death with a score of less than 0.16, 0.16 to 0.56, or more than 0.56, with survival rates of 87%, 70%, and 21%, respectively, at 6 months.
Comparison of Scoring Systems
Many studies have compared MELD with mDF, MELD with CPT score, and Glasgow AH score with mDF. The comparison studies have shown conflicting data as to which score is best, likely due to heterogeneity of the study population or variation in the time at which survival is estimated across different studies. Some of the other potential factors explaining these differences are the use of liver biopsy in AH diagnosis, the presence of underlying cirrhosis, inclusion and exclusion criteria, and the use of specific treatment for AH. All the scoring systems have some limitation, and an ideal score does not exist (see Table 24-4 ). Currently, in routine practice an mDF score of 32 or greater and a MELD score of 20 or greater are used for initiation of treatment and the Lille score is used for guiding treatment response. Recently, a dynamic model using the MELD score at the baseline and the Lille score at 1 week of corticosteroid therapy performed better than either score alone in predicting outcome of AH.
Liver Histology
In a prospective biopsy-proven cohort of AH patients, fibrosis stage, neutrophilic infiltration, type of bilirubinostasis, and megamitochondria independently predicted survival of AH patients at 90 days (see Table 24-4 ). With a total score ranging between 0 and 9, the AH histologic score can accurately categorize patients as having low, intermediate, and high risk of dying at 90 days.
Emerging Scoring Systems
Neutrophilic leukocytosis, portal blood flow changes, hepatic venous pressure gradient, serum ammonia levels, organ failure score, cytokine levels, degree of steatosis on biopsy, and AH histologic score correlate with outcome and response to corticosteroids. Data are also emerging on defects in mitochondrial bioenergetics and oxidative burst, with reduced levels of NADPH oxidase in monocytes isolated from peripheral blood predicting disease severity and response to corticosteroids.
Treatment of Severe Episodes of Alcoholic Hepatitis
Patients with severe AH are at high risk of short-term death from progressive liver failure. Many pharmacologic options have been examined ( Table 24-5 ).
TABLE 24-5
Specific Therapeutic Options in the Management of Alcoholic Hepatitis
|
Corticosteroids
These drugs have remained controversial in the management of AH ever since they were evaluated in a randomized controlled study for AH treatment.
Efficacy of Corticosteroids in Alcoholic Hepatitis Treatment
In the last 4 decades, 14 randomized controlled studies have assessed the use of corticosteroids in the treatment of AH. Of these, 12 studies compared corticosteroids with placebo, with survival benefit in 5 studies, of no benefit in 6 studies, and a trend for benefit in the recently reported Steroids or Pentoxifylline for Alcoholic Hepatitis (STOPAH) study. One study comparing corticosteroids with nutritional supplementation showed similar survival at 28 days, and higher mortality in the prednisolone arm than in the enteral tube feeding arm. In another study, corticosteroids compared with an antioxidant cocktail showed survival benefit at 30 days (70% vs. 54%, p = 0.05). The differences in efficacy of corticosteroids across studies are likely due to variations in study inclusion/exclusion criteria such as the proportion with underlying cirrhosis, the proportion of biopsy-proven AH, and variation in disease severity. For example, in the STOPAH study, patients with less severe AH were included, with an overall 28-day mortality in placebo group of only 16%, which is much lower than the historical 28-day mortality among untreated severe AH patients. Further, most patients enrolled in the STOPAH study were recruited with a clinical diagnosis of AH, and it is possible that patients with decompensated alcoholic cirrhosis may received a diagnosis of AH.
In a meta-analysis, pooled analysis of individual patient data on 418 patients (221 corticosteroid-treated patients) from the five largest randomized controlled studies, corticosteroids provided approximately 50% survival benefit at 28 days (80% vs. 66%, p < 0.0001). In another meta-analysis of all randomized controlled trials, in spite of heterogeneous data, use of corticosteroids was associated with survival benefit in a subgroup of patients with severe AH.
Current Status of Corticosteroid Therapy in the Management of Alcoholic Hepatitis
Current guidelines propose that severe AH patients (an mDF score ≥32 and/or presence of hepatic encephalopathy) be treated with corticosteroids, provided there are no contraindications for their use. However, these guidelines do not take into account the findings of the recently reported STOPAH study. Using a 2 × 2 factorial design, this study recruited 1103 severe AH patients (mean age 49 years, 63% males, 27% with hepatic encephalopathy, 10% with treated sepsis, mean mDF score 63, mean MELD score 21) from 65 centers in the United Kingdom. Patients with active infection/sepsis, gastrointestinal bleeding, renal failure, and jaundice for more than 3 months were excluded. Analysis of 1053 patients showed a trend for mortality benefit at 28 days with prednisolone compared with placebo (13.8% vs. 18%, p = 0.06). Prednisolone therapy, however, reduced mortality by 40% at 28 days on multivariate analysis after patient demographics, AH severity, liver disease complications, and receipt of treatment had been controlled for. Other predictors of mortality at 28 days were the patient's age, disease severity, renal function, and hepatic encephalopathy.
In a recently reported network meta-analysis of pharmacologic therapies in AH, data were pooled from 22 randomized studies on 2621 AH patients, including the STOPAH study. There was moderate-quality evidence for efficacy of corticosteroids in reducing short-term mortality (30 days) when used alone, 0.54 (range 0.39 to 0.73), combined with pentoxifylline, 0.53 (range 0.36 to 0.78), or combined with N -acetylcysteine, 0.15 (range 0.05 to 0.39). Currently, prednisolone with or without N -acetylcysteine may be considered for severe AH patients, given the lack of availability of other effective options (see Fig. 24-1 ).
Mechanism of Action
Corticosteroids have antiinflammatory effects and inhibit T-cell immunity. Inflammation is a major component in the pathogenesis of AH ( Fig. 24-2 ), with increased levels of tumor necrosis factor (TNF)-α and other cytokines. In one study, corticosteroids reduced the levels of soluble intercellular adhesion molecule 1 in hepatic venous blood and on hepatocyte membrane, reduced levels of TNF-α in hepatic vein and peripheral blood, and increased lobular infiltration of neutrophils. In another study, corticosteroids induced activation of neutrophils, with a consequent decrease in the level of the preinflammatory cytokine interleukin-8, and an increase in the level of the antiinflammatory cytokine interleukin-10.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here