sign in sign up
<<
Home , Alcoholic hepatitis, Alcoholic liver disease, Alcohol abuse, Alcohol and health, Chronic liver disease, Hepatorenal syndrome

Diagnosis and Treatment of Alcoholic Disease and Its Complications

Luis S. Marsano, M.D., Christian Mendez, M.D., Daniell Hill, M.D., Shirish Barve, Ph.D., and Craig J. McClain, M.D.

Alcoholic (ALD) is a serious and potentially fatal consequence of use. The diagnosis of ALD is based on drinking history, physical , and laboratory tests. Treatment strategies for ALD include lifestyle changes to reduce alcohol consumption, cigarette smoking, and obesity; nutrition therapy; and pharmacological therapy. The diagnosis and management of the complications of ALD are important for alleviating the symptoms of the disease, improving quality of life, and decreasing mortality. KEY WORDS: alcoholic liver disorder; diagnosis; disease ; treatment method; lifestyle; nutritional deficiency; vitamin therapy; drug therapy; ; colchicines; corticosterone; alternative medical treatment; S-adenosylmethionine; ; ; disorder; esophageal varix; ;

he liver is one of the largest and and alcoholic often of ALD, and ALD can be difficult to most complex organs in the body. coexist and cause substantial morbidity diagnose because patients frequently TIt performs multiple functions, and mortality. For example, studies from minimize or deny . In including the production of proteins the Department of Veterans Affairs (VA) addition, there may be no evidence of and , detoxification, metabolic demonstrate that patients with both cir- ALD from the physical exam, and lab- functions, and the regulation of choles- rhosis and have a oratory abnormalities may not specifi­ terol and clotting. Because the rate of greater than 60 percent cally point to ALD. liver is primarily responsible for alcohol over a 4-year period, with most of the Ambulatory patients with alcoholic , it is especially vulnerable occurring in the first year (Chedid fatty liver often are asymptomatic. to alcohol-related . et al. 1991). Thus, the mortality rate for Patients with alcoholic hepatitis may be (ALD) is a ALD is greater than that of many com- asymptomatic, have only enlarged liver serious and potentially fatal consequence mon types of such as colon, breast, (i.e., ), or have full-blown of drinking alcohol. ALD encompasses and prostate. This article examines the three conditions: fatty liver, alcoholic issues of diagnosing and treating ALD LUIS S. MARSANO, M.D., is a professor of hepatitis, and cirrhosis (see figure 1). Fatty and the complications of this disease. internal medicine; CHRISTIAN MENDEZ, liver (i.e., ), the most common M.D., is a fellow; DANIELL alcohol-induced liver disorder, is marked HILL, M.D., is an associate professor of by the excessive accumulation of fat Diagnosis of Alcoholic internal medicine; SHIRISH BARVE, PH.D., inside the liver cells. Alcoholic hepatitis Liver Disease (ALD) is an associate professor of internal medicine; is and more severe injury and CRAIG J. MCCLAIN, M.D., is vice of the liver, in which the body’s immune The diagnosis of ALD is established by chair for research in the Department of system responds to and causes liver a history of habitual alcohol intake of Internal Medicine and professor of inter- damage. In cirrhosis, normal liver cells sufficient duration and quantity, together nal medicine and pharmacology and are replaced by scar tissue (i.e., ), with physical signs and laboratory evi- toxicology; all authors are associated with and consequently the liver is unable to dence of liver disease. the University of Louisville Medical perform many of its usual functions. is not a prerequisite for the development Center, Louisville, Kentucky.

Vol. 27, No. 3, 2003 247 alcoholic hepatitis with tender hep­ Other common signs are small star- Chronic alcohol consumption also atomegaly, , fever, accumulation shaped vessels (i.e., spider angiomata) may be associated with abnormally high of fluid in the (i.e., on the skin of the upper torso, blotchy triglyceride levels (i.e., hypertriglyceri­ ascites), nervous system effects such as redness on the palms (i.e., palmar ery­ demia), high blood levels of uric acid confusion and personality change (i.e., thema), and contracture of the palm (i.e., hyperuricemia), and low amounts ), anorexia, and tissue, causing the ring and pinky finger of potassium (i.e., hypokalemia) and . Other signs may include high to bend into the palm (i.e., Dupuytren’s magnesium, as well as an elevated index counts resembling palmar contracture). Enlargement of of size (i.e., mean corpus­ those seen in leukemia (i.e., leukemoid the parotid gland (one of the salivary cular erythrocyte volume [MCV]). reactions) and the rapid deterioration glands) and the lacrimal (tear) glands Hyperuricemia and hypertriglyc­ of kidney function (i.e., hepatorenal often is seen. Enlargement of the fin­ eridemia often normalize with absti­ syndrome). Even in the absence of cir­ gertips may be found in patients who nence, and hypokalemia normalizes rhosis, the main vein that brings blood develop a problem with the way blood with adequate potassium replacement. from the intestine and stomach into the passes through the lungs, resulting in Elevated MCV often is found in peo­ liver (i.e., the portal vein) may come blood not being properly oxygenated. ple who ingest more than 50 grams of under increased pressure because of Other physical signs, which may be alcohol per day,1 with sensitivity of 27 scarring of the liver, resulting in portal found during examination with a flexi­ to 52 percent and specificity of 85 to vein . ble fiberoptic instrument (i.e., endo­ Ten to 20 percent of patients with scopy), include changes in the stomach 1 In the , a drink is considered to be 0.5 alcoholic hepatitis develop cirrhosis, and lining that occur with portal hyperten­ ounces (oz) or 15 grams of alcohol, which is equivalent to up to 70 percent of alcoholic hepatitis sion, as well as engorged veins in the 12 oz (355 milliliters [ml]) of , 5 oz (148 ml) of , or patients go on to develop cirrhosis each , stomach, or another part of 1.5 oz (44 ml) of 80-proof distilled spirits (USDHHS 2000). year (Bird and Williams 1988). Women the , which expand are at higher risk for developing cirrho­ as a consequence of increased pressure sis, as are people who continue drinking in the blood flow of the venous system. or have severe alcoholic hepatitis (Pares Patients with hepatic encephalopathy et al. 1986). Some patients with alcoholic may have slow reaction times and mus­ hepatitis who abstain still may develop cle tremors causing involuntary jerking cirrhosis, but others will have complete of the hands. clinical and histologic recovery. ALD cannot be diagnosed based on Patients with early stage alcoholic cir­ any of the physical signs and symptoms rhosis with no complications (i.e., well- alone. Laboratory tests often assist in the Fatty Liver compensated) may be asymptomatic and diagnosis of ALD. Almost all patients have normal physical exams and normal will have elevated liver enzymes. The routine blood tests of liver function and level of the aspartate amino­ injury. In other patients, alcoholic fatty transferase (AST) will exceed that of liver or alcoholic hepatitis often coexist alanine aminotransferase (ALT), but and may be accompanied by hepatomegaly, both will be below 300 international an enlarged spleen (i.e., ), units per milliliter (IU/ml). When the or both. In cirrhotics with severe alco­ ratio of AST to ALT is greater than 2, holic hepatitis, hepatomegaly or spleno­ the most likely diagnosis is ALD. In Alcoholic Hepatitis megaly may be the dominant feature; some studies, more than 80 percent of in other patients, the signs and symp­ patients attain this ratio. toms of portal vein hypertension (e.g., Elevated blood levels of the liver ascites and engorged veins [varices] in the enzyme gamma glutamyltransferase esophagus) may predominate. As the (GGT) indicate heavy alcohol use and disease advances, the liver decreases in liver injury. This test has greater ability size, the left hepatic lobe becomes more to correctly test positive (i.e., sensitiv­ prominent, and the entire liver has a ity) but less ability to correctly test hard and nodular consistency. Spleno­ negative (i.e., specificity) than AST or Cirrhosis megaly of varying degrees is frequent. ALT tests. Of the three enzymes, GGT In later stage cirrhosis with compli­ is the best indicator of excessive alcohol cations (i.e., decompensated disease), consumption, but because GGT is Figure 1. Biopsies of alcoholic liver patients may have muscle wasting, present in many organs and because disease showing how a patient can ascites, and the adaptation of smaller some drugs raise GGT levels, high progress from fatty liver and alcoholic vessels to handle increased blood flow GGT levels are not necessarily an indi­ hepatitis to cirrhosis. (i.e., venous collateral circulation). cator of alcohol abuse.

248 Alcohol Research & Diagnosing and Treating Alcoholic Liver Disease

90 percent. The blood protein known of identifying alcohol abuse, suscepti­ as carbohydrate-deficient transferrin bility to liver injury, and mechanisms of THERAPY FOR ALD frequently is used to detect current or liver injury, and of detecting and monitor­ recent alcohol abuse, especially con­ ing liver injury. • Lifestyle modification sumption in excess of 60 grams per day (decreased alcohol use, smok­ (Nilssen et al. 1992; Litten et al. 1995), ing, obesity) but there are no ideal tests to identify Treatment of ALD continuing alcohol intake. •Appropriate nutrition/ An increased number of white blood Treatment strategies for ALD include nutritional support cells (i.e., ) and decreased lifestyle changes to reduce alcohol con­ number of platelets (i.e., thrombocytope­ sumption, cigarette smoking, and obe­ •Use of pentoxifylline or pred­ nia) are common in alcoholic hepatitis. sity; nutrition therapy; pharmacological nisone for alcoholic hepatitis Thrombocytopenia may be transitory, therapy; and possibly liver transplanta­ but in patients with concomitant cir­ tion (see textbox). ( •Advice on complementary rhosis, it is persistent. Markers of severe is discussed in detail in the article by and alternative medicine alcoholic hepatitis or cirrhosis include Anantharaju and Van Thiel in this issue.) (e.g., silymarin or SAMe) for elevated levels of (a yellow-orange cirrhosis substance generated in the liver), pro­ Lifestyle Changes longed time required for a blood sample •Transplantation in selected to clot (i.e., prothrombin time [PT]), from alcohol is vital in order abstinent patients with severe and a low level of the main circulating to prevent further liver injury, scarring, (i.e., end-stage) disease. protein in the bloodstream (i.e., albumin), and possibly liver carcinoma; it appears which is synthesized by the liver (i.e., to benefit patients at every stage of the ). The most commonly disease. Fatty liver is reversible with , a disorder that is histo­ used prognostic index in alcoholic hep­ abstinence. Although evaluations of logically identical to alcoholic hepatitis. atitis is Maddrey’s Discriminant Function the effects of abstinence on the progres­ (DF), which is calculated by this equation: sion of ALD are few and have involved Body mass index has been shown to retrospective, nonrandomized trials, be an independent risk factor for the 4.6 [PT(patient) – PT(control)] + virtually all these studies have shown development of ALD (Raynard et al. total bilirubin (mg/dl). beneficial effects of abstinence (Powell 2002). An increasingly large subset of If this value exceeds 32, the mortality and Klatskin 1968; Merkel et al. 1996). ALD patients are obese, with alcohol rate during a current hospitalization may Patients with either compensated or intake as a source of excess and empty exceed 50 percent (Maddrey et al. 1978; decompensated cirrhosis benefit from calories (that is, having no nutritional Carithers et al. 1989). There also is abstinence. Thus, all patients with ALD value). Thus, as with many other gas­ evidence that blood concentrations of should be encouraged to abstain from trointestinal disorders (e.g., gastro­ proteins (i.e., ) that promote alcohol consumption. Newer medica­ esophageal reflux disease), the initial inflammation—such as tumor tions to facilitate abstinence, such as approach to treating ALD is lifestyle factor alpha (TNF–α), interleukin–6, naltrexone and , have been modification to reduce alcohol consump­ and interleukin–8—correlate with shown to be effective in some chronic tion, cigarette smoking, and obesity. mortality in patients with alcoholic alcoholics, but no large multicenter stud­ hepatitis (McClain et al. 1993), but ies have evaluated these medications in Nutrition Therapy levels of these cytokines are not patients with ALD. determined in routine clinical practice. Many people who drink alcohol also is prevalent in alcoholic mainly is used to clarify smoke cigarettes. In European studies, hepatitis and cirrhosis, especially in atypical cases, to better define the con­ fibrosis worsens more rapidly in ALD end-stage ALD, and can range from tribution of alcohol in patients with patients who smoke cigarettes (Klatsky deficiency in individual nutrients possible non-alcohol-related coexisting and Armstrong 1992; Corrao et al. 1994). (e.g., zinc, ) to global protein– conditions (e.g., , use of lipid- Patients with hepatitis C who drink calorie malnutrition. lowering medications), and to deter­ also deteriorate faster if they smoke Researchers at the VA Cooperative mine the severity of liver disease. Many cigarettes (Pessione et al. 2001). Cigarette Studies Program have conducted some laboratories are conducting research to smoking causes , a con­ of the most extensive studies of nutri­ evaluate biomarkers or identifier proteins dition that arises when an overabun­ tional status in patients with alcoholic for detecting ongoing alcohol abuse and dance of free radicals is present in the hepatitis (Mendenhall et al. 1995). The ALD. The importance of genetic varia­ body, which may be a factor leading to first of these studies (VA Cooperative tions in and ALD among accelerated liver disease in smokers. Study 119) demonstrated that virtually individuals also is under active investiga­ Obesity is associated with the devel­ every patient with alcoholic hepatitis tion. New tests may provide novel ways opment of fatty liver and nonalcoholic had some degree of malnutrition. Patients

Vol. 27, No. 3, 2003 249 had an average alcohol consumption in nutritional status and immune func­ with prednisone (an immunosuppressive of 228 grams per day, with almost 50 tion (Hirsch et al. 1999). medication), with its potential compli­ percent of energy intake coming from VA Cooperative Study 275 found cations (e.g., , , osteo­ alcohol. The severity of liver disease that the combination of an anabolic porosis), in hospitalized patients with generally correlated with the severity steroid and an oral nutritional supplement alcoholic hepatitis. of malnutrition. reduced the mortality rate of patients who Thus, traditional nutritional supple­ A followup VA study on alcoholic had moderate protein–energy malnu­ mentation clearly improves nutritional hepatitis (VA Cooperative Study 275) trition (Mendenhall et al. 1995). Those status and, in some instances, hepatic found similar results. In both of these with severe malnutrition did not signif­ function and other outcome indicators studies, patients were given a balanced icantly benefit from the therapy, possi­ in alcoholic hepatitis and cirrhosis. 2,500-kilocalorie hospital diet, moni­ bly because their malnutrition was so tored carefully by a dietitian, and were advanced that no intervention, includ­ Pharmacological Therapy encouraged to follow it. In the second ing nutrition, could help. study, patients in the treatment group Kearns and colleagues (1992) showed Although ALD remains a major cause also received a liquid nutritional sup­ that patients with ALD who were hos­ of morbidity and mortality in the plement high in three amino acids that pitalized for treatment and given an United States, there is no FDA-approved help to stimulate protein synthesis (which enteral nutritional supplement via tube therapy for either alcoholic cirrhosis or was administered as an oral food sup­ feeding had significantly improved serum alcoholic hepatitis. However, several plement), as well as the bilirubin levels and liver function. More­ drugs have been used “off label.” . In neither study were over, a major randomized study of enteral patients fed by tube if voluntary oral nutrition versus steroids in patients with Propylthiouracil (PTU). Orrego and intake was inadequate (probably a design alcoholic hepatitis showed similar over­ colleagues (1987) examined long-term flaw, in retrospect). Voluntary oral food all initial outcomes, as well as fewer PTU therapy in more than 300 patients intake correlated in a stepwise fashion long-term infections in the nutrition with various types of liver disease, includ­ with 6-month mortality data—that is, group (Cabre et al. 2000). This impor­ ing ALD. In this study, mortality was almost all patients who voluntarily con­ tant study suggests that aggressive nutri­ reduced by nearly 50 percent in patients sumed more than 3,000 kcal per day tional support is as effective as treatment receiving PTU. A recent review (Rambaldi still were living at the end of the 6-month period, whereas more than 80 percent of those consuming less than 1,000 kcal per day died within that time (see figure 2) (Mendenhall et al. 1995). Moreover, the degree of malnutrition correlated with the development of seri­ ous complications such as encephalopa­ thy, ascites, and hepatorenal syndrome (Mendenhall et al. 1995). Interest in nutrition therapy for cir­ rhosis was stimulated when Patek and colleagues (1948) demonstrated that a nutritious diet improved the 5-year out­ Mortality (%) come of patients with alcoholic cirrho­ sis compared with patients consuming an inadequate diet. Several recent stud­ ies have found improved outcomes in cirrhosis patients who were given nutri­ tional support. Hirsch and colleagues >3000 2500 3000 2000 2499 1500 1999 1000 1499 <1000 (1993) demonstrated that outpatients Energy Intake (kcal/day) receiving a nutritional support product (1,000 kcal, 34 grams protein) through a feeding tube (i.e., enteral nutritional Figure 2 Inadequate nutrition was directly related to mortality in the Veterans support) had significantly improved Health Administration studies in patients who had moderate to severe protein intake and significantly fewer alcoholic hepatitis. It is not known whether providing nutrients directly into the gastrointestinal tract (that is, enteral feeding) to those patients hospitalizations. These investigators with inadequate caloric intake would have improved their survival. subsequently gave enteral nutritional support to outpatients with alcoholic NOTE: Kcal/day = Kilocalories per day. cirrhosis and observed an improvement

250 Alcohol Research & Health Diagnosing and Treating Alcoholic Liver Disease

and Gluud 2002) evaluated PTU therapy and the proinflammatory of ALD reviewed here are: ascites (accu­ for ALD, including alcoholic fatty liver, response. Most randomized studies have mulations of fluid in the abdominal alcoholic fibrosis, alcoholic hepatitis, supported the use of in cavity), infections in this fluid that and cirrhosis. Combining the results of moderate to severe alcoholic hepatitis develop without any apparent cause six randomized clinical trials (710 patients), (Carithers et al. 1989), but a large multi­ (i.e., spontaneous bacterial peritonitis these researchers did not find any sig­ center VA study yielded negative results [SPB]), hepatorenal syndrome, and nificant effects of PTU versus placebo (Mendenhall et al. 1984). Steroids have . on mortality from all causes or liver- been found to be effective against severe related mortality, complications of liver acute alcoholic hepatitis by most meta- Ascites disease, or liver histology. The negative analyses, including the most recent result of this review limits enthusiasm study by Mathurin and colleagues (2002). Ascites is one of the most common for PTU as a treatment for ALD. These investigators reported significantly complications of advanced liver disease improved survival at 28 days (85 per­ and generally indicates a poor progno­ Colchicine. Because of its antifibrotic cent vs. 65 percent) in severely ill alco­ sis and a high likelihood of death. effects, colchicine has been suggested holic hepatitis patients having a DF Approximately 8 of every 10 patients as a treatment for ALD. Initial positive greater than 32. This survival advantage who have ascites in the United States studies by Kershenobich and colleagues may extend to 1 year but not 2. Inde­ have it as a consequence of cirrhosis of (1988) led to a large VA Cooperative pendent prognostic factors associated the liver. It is estimated that 3 of every Study evaluating colchicine therapy in with survival at 28 days in this meta- 10 patients who have cirrhosis without patients with alcoholic cirrhosis. Results analysis were steroid treatment, age, and complications will develop ascites within showed no beneficial effect on either serum levels. Patients with the next 5 years. In patients with ascites, overall mortality or liver-related mor­ infections, gastrointestinal bleeding, and the likelihood of death in the following tality (Morgan et al. 2002). A recent many other common complications year is approximately 50 percent, com­ smaller study from Europe also showed were excluded from these studies. pared with 10 percent for cirrhosis no beneficial effects of colchicine (Cortez- Most investigators agree that if cor­ patients without complications. Ascites Pinto et al. 2002). Thus, despite initial ticosteroids are to be used, they should cases can be classified as easily treatable enthusiasm and biochemical rationale be reserved for patients with severe liver or refractory and difficult to control. for use of this drug, it does not appear disease (i.e., DF greater than 32), and Patients with the latter type are likely to be effective in ALD treatment. possibly those with hepatic encephalop­ to develop hepatorenal syndrome athy. Steroids have well-documented (Garcia-Tsao 2001). Pentoxifylline (PTX). Akriviadis and side effects, including enhancing risk Ascites leads not only to aesthetic colleagues (2000) evaluated PTX in a of , which already is substan­ changes in body shape but, more impor­ prospective, randomized, double-blind tial in patients with alcoholic hepatitis. tantly, to: clinical trial in patients with severe Thus, a major disadvantage to cortico­ alcoholic hepatitis. Forty-nine patients steroids is that they cannot be used by •Increased risk of spontaneous infec­ received PTX and 52 received placebo many patients with alcoholic hepatitis. tion of the ascitic fluid. (vitamin B12) for 4 weeks. PTX treatment improved survival: 12 PTX patients Complementary and Alternative •Development of abdominal died (24.5 percent), compared with Medicine (CAM) Agents. CAM agents where the abdominal wall muscle 24 placebo patients (46 percent). PTX have had some success in patients with becomes weakened and part of the also decreased the percentage of deaths liver disease and are widely used. Research abdomen protrudes (sometimes to caused by hepatorenal syndrome. Renal demonstrates a strong rationale for using the extreme of spontaneous rupture) failure was the cause of death of 6 of many of these CAM agents, such as S­ (see figure 3). the 12 PTX-treated patients who died adenosylmethionine, but shows that (50 percent), compared with 22 of the others may have no efficacy or may •Difficulty breathing because of 24 control patients who died (92 percent). even cause harm, including liver injury. pressure of the abdomen on the Multivariate analysis revealed that age, respiratory muscles. kidney test results for the waste prod­ uct creatinine at randomization, and Diagnosis and Treatment •Decreased food intake, with pro­ treatment with PTX were independent of the Complications of gressive malnutrition. factors associated with survival. ALD •Decreased physical activity with Corticosteroids. Although corticosteroids Proper diagnosis and management of consequent loss of muscle mass. are the most extensively studied form of the complications of ALD are vital to therapy for alcoholic hepatitis, their role decreasing the deterioration of illness, Treating ascites does not seem to remains limited. The rationale for steroid improving quality of life, and possibly prolong life in cirrhotic patients but use is to decrease the immune response decreasing mortality. The complications improves the quality of life and protects

Vol. 27, No. 3, 2003 251 patients from spontaneous infections of 78 mEq sodium per day in . To SBP if bacteria are found in the ascitic the fluid, which are associated with be able to eliminate this amount of fluid, but the concentration of bacteria high death rates. sodium, most patients need to take in the fluid is extremely low—an esti­ Patients who develop ascites for the . mated 1 bacterium per milliliter of first time, those with ascites who are Patients who have no swelling of the fluid—so that the bacteria cannot be admitted to the hospital because of illness, extremities should not lose more than seen by examining the fluid under the those who have difficult-to-control ascites, half a kilogram of weight per day in order microscope. In order to have a reasonable and those who develop symptoms because to control ascites. Those who have chance of getting a positive culture, the of tense ascites (i.e., the abdomen is tight swelling in the extremities may be able fluid should be injected at the bedside with ascites)—all should have the fluid to lose up to 1 kilogram per day. Fluid into blood culture bottles specially removed for diagnostic evaluation and restriction is not required except in designed to recover small amounts of to lessen discomfort in the abdomen. patients with very low sodium concen­ bacteria. This technique will detect the Because these patients are at high risk tration in the blood. So that patients vast majority of infections. If improper of infection in the fluid and because will not become dehydrated, the dose techniques are used (e.g., sending the the fluid accumulation may be a conse­ of diuretics is adjusted based on the fluid to the laboratory to be placed on quence not only of liver disease but patients’ response to the medication a culture plate), chances of proper diag­ also of other associated disorders, this (i.e., it is titrated) to obtain a zero balance nosis decrease to 4 out of 10. fluid should be thoroughly evaluated. of sodium once the ascites has been Patients with SBP may not have Evaluation should include a cell count controlled (Runyon 1997, 1998; Andy symptoms, or manifestations of the (Runyon 1997, 1998) as well as deter­ and Ke-Qin 2001). Once a patient has infection may appear to be unrelated to minations of total protein and albu­ developed ascites, he or she is at high risk min. At around the same time, a blood of death. For that reason, liver trans­ sample should be checked to measure plantation should be contemplated if the amount of albumin in order to the patient is a suitable candidate. A facilitate interpretation of the findings on the ascitic fluid. For patients who Spontaneous Bacterial Peritonitis have ascites because of portal vein (SBP) hypertension, determining serum albu­ min concentration–ascitic albumin Of patients hospitalized with ascites, concentration (SAAG) is recommended. between 10 percent and 30 percent will SAAG is calculated by subtracting the develop SBP. This infection is thought albumin concentration of the ascitic to occur either by spontaneous passage fluid from the albumin concentration of normal bacteria that reside in the of a serum specimen obtained on the gut into the ascitic fluid, or by seeding same day. Patients with SAAGs of 1.1 of bacteria into the blood from a distant grams per deciliter (g/dl) or higher may source (e.g., a urinary infection or lung have cirrhosis or alcoholic hepatitis, infection), leading to growth of this among other conditions. bacteria in the ascitic fluid (O’Grady B To control the formation of ascites, et al. 2000). patients must eliminate more sodium Patients who have low protein con­ than they acquire through diet. Patients centrations in the ascites (less than 1.5 with cirrhosis and ascites tend to retain g/dl) are at higher risk of developing sodium very efficiently, and most patients SBP. Once a patient has had an episode need to have dietary sodium restricted of SBP, there is a 7 in 10 chance that a to less than 2 grams per day (88 mEq new episode will occur within the next [milliequivalent, or the number of grams year. With every episode, 2 to 3 of every of solute dissolved in one milliliter of 10 patients will die from complications solution]). To prevent additional ascites, of the infection, and only 3 of every a patient who is following a 2-gram 10 are expected to survive for 2 years sodium diet needs to lose at least 78 (Garcia-Tsao 2001). mEq sodium per day through the urine The diagnosis of SBP is made when Figure 3. (A) Patient with alcoholic in addition to the 10 mEq that are lost high numbers of a type of white blood cirrhosis who shows ascites, an regularly through the skin. Research cell that is especially protective against umbilical , and wasting of has shown that if a person is eliminat­ bacterial infections (i.e., polymorphonu­ muscle. (B) After 2 years of absti­ nence and appropriate nutrition, the ing a ratio of sodium to potassium clear cells [PMN]) are found in the patient gained back muscle mass through urine that is greater than 1, ascitic fluid (i.e., in excess of 250 per and his ascites improved. that person will be eliminating at least ml). Patients are diagnosed as having

252 Alcohol Research & Health Diagnosing and Treating Alcoholic Liver Disease

the abdominal cavity. For example, SBP Hepatorenal Syndrome artery that feeds the kidney, which may patients may have confusion, changes occur as a response to excessive relax­ Hepatorenal syndrome is the deteriora­ in kidney function, poorly controlled ation of the vessels in the rest of the tion of kidney function in patients who ascites, or overall progressively deteriorat­ body and a relatively low volume of ing health. Despite the fact that more have acute or chronic but blood inside the vascular system. Thus, than half of the patients with SBP com­ otherwise healthy kidneys. This disorder the first approach to treatment of this plain of some degree of abdominal pain may occur spontaneously but more disorder is to try to expand vascular or discomfort, the physical exam of the often is the result of an infection, an volume and then to increase the degree abdomen usually is completely benign. episode of gastrointestinal bleeding, or of contraction of vessels other than the Usually only one type of bacteria overly aggressive use of diuretics. There kidney artery. appears in the culture of patients who are two types of hepatorenal syndrome: Several strategies have been used to have SBP. Clinicians should suspect the Type I conveys the highest risk of death reverse hepatorenal syndrome. The first possibility of a secondary peritonitis and develops rapidly over a couple of involves intravenous infusions of albu­ (e.g., some intra-abdominal perforation weeks; Type II progresses more slowly, min (to expand vascular volume), fol­ or abscess formation in the abdomen) usually over several months. The diag­ lowed by administration of ornipressin, if (1) multiple kinds of bacteria are nosis of hepatorenal syndrome requires a medication that increases the contrac­ recovered from the culture, or (2) the fulfillment of all of the following criteria: tion of most of the vessels in the body patient develops an infection consistent (Guevara et al. 1998a). Similar effects with SBP but in the presence of total •Deterioration of kidney function with fewer complications have been protein concentration in the ascitic fluid demonstrated by a concentration of obtained using intravenous albumin of more than 1.5 g/dl, or (3) the creatinine in the blood of more than with , which causes vessel patient fails to respond promptly to 1.5 mg/dl or a decrease in creatinine contraction more safely (Gines et al. proper antibiotic therapy. A secondary clearance to less than 40 milliliters 2001; Moreau et al. 2002). infection also should be suspected if per minute. In the United States, neither orni­ direct examination of fluid under the pressin nor terlipressin is available, and microscope shows bacteria because, as •No evidence of , no the most popular intervention for hep­ mentioned, the concentration of bacteria exposure to a drug or infection that atorenal syndrome is to expand blood in SBP is so low that bacteria should causes kidney , and absence volume using intravenous albumin, not be detectable by microscopic exam­ of severe low (). then to administer and ination (Andy and Ke-Qin 2001; to regulate vascular contrac­ Garcia-Tsao 1992). •Failure to respond to discontinua­ tion (Angeli et al. 1999). For patients Patients who develop SBP are at tion of diuretics or to treatment who cannot tolerate medication by extremely high risk of developing kidney with 1.5 liters of solution to mouth, intravenous albumin can be dysfunction and hepatorenal syndrome. expand blood volume. used, followed by continuous infusion Expansion of the volume within the of norepinephrine (Duvoux et al. 2002). blood vessels (i.e., the intravascular vol­ •No evidence of obstruction of urine With the latter two regimens, benefi­ ume) using intravenous albumin infu­ flow or primary kidney disease, as cial effects usually can be seen by day sions has been shown to decrease the demonstrated by ultrasound exam. 10 of therapy and, if not successful, the frequency of hepatorenal syndrome treatment usually is discontinued after and, for that reason, improves survival • Little or no protein in the urine. 15 days. Patients who do respond can rates in patients who develop SBP (Sort complete at least 2 weeks of therapy, et al. 1999). Because of concerns about The majority of patients who will and after that, therapy can be discon­ kidney toxicity, it is important to avoid develop hepatorenal syndrome will first tinued, usually without deterioration antibiotics or other medications that may develop -resistant ascites, usually of kidney function. exacerbate kidney damage. Current with very low sodium in the urine (i.e., Other therapies are being used, therapy for SBP includes use of the less than 10 mEq/L). In general, among including the transjugular intrahepatic antibacterial drug cefotaxime. patients who have ascites, the risk of portal systemic shunt (TIPS), in which Patients at special risk for SBP are developing hepatorenal syndrome is a long is inserted via the jugu­ those who are hospitalized and have a 2 in 10 during the first year and 4 in lar vein in the neck, advanced into a total protein concentration in the ascitic 10 during the first 5 years. Among hepatic vein and then into a large branch fluid that is less than 1.5 g/dl, those who patients with Type I hepatorenal syn­ of the portal vein in the liver. Using an have gastrointestinal bleeding from any drome, the estimated mortality without inflatable balloon-tipped catheter tube, source, and those who have had previous appropriate therapy is 80 percent after the section between the portal vein episodes of SBP (Andy and Ke-Qin 2 weeks and 90 percent after 10 weeks branch and the hepatic vein is widened 2001). Prophylactic therapy (i.e., anti­ (Gines et al. 1993). and then kept open (stented) with a biotic treatment) is indicated for all Hepatorenal syndrome is thought to cylindrical wire-mesh stent (Guevara these groups of patients. result from a severe contraction of the et al. 1998b). This helps to lower the

Vol. 27, No. 3, 2003 253 increased pressure in the portal vein. In addition, hepatorenal syndrome has been treated using high doses of antioxidants (Holt et al. 1999). More recently, a system known as MARS (molecular absorbent recirculating system) has been used (Mitzner et al. 2000); with this treatment, a patient’s blood is transported to a fil­ ter, where it is mixed with albumin, which carries the out of the blood. Although these treatments are avail­ able, hepatorenal syndrome will very likely recur, and patients should be moved quickly in the direction of pos­ sible liver transplantation.

Esophageal Varices Figure 4 View of varices in the esophagus, a consequence of liver disease, using endoscopy. These dilated veins are being banded to prevent gastro­ Cirrhosis causes an increase in pressure intestinal bleeding. in the fine net of blood vessels inside the liver. This pressure is transmitted are developing, endoscopic evaluation proven to be effective but usually is back to the portal vein and onto the of the esophagus and stomach is rec­ reserved for patients who have large veins that form it. Because of the high pressure on the veins forming the por­ ommended at 1- to 2-year intervals in varices and cannot tolerate beta block­ tal vein and the portal vein itself, blood patients with cirrhosis to determine ers (see figure 4). As mentioned, patients tries to escape, forming new collateral whether varices have formed and if with esophageal varices may bleed veins (i.e., the increased pressure causes they are large enough to require treat­ spontaneously. Approximately half of very small varices to grow larger). Many ment to prevent bleeding. these patients will stop bleeding spon­ of these veins are localized superficially Varices in the esophagus that are large taneously. The frequency of death for inside the esophagus and the upper or show evidence of high risk for bleed­ each of these episodes is 3 in 10. It is part of the stomach. When these veins ing should be considered for treatment estimated that bleeding will recur engorge because of increased pressure, to prevent bleeding (i.e., primary pro­ within 6 weeks in 4 of 10 patients who they are called varices. phylaxis), such as the use of a class of have esophageal variceal bleeding. Varices in the esophagus and stomach drugs that block the action of the invol­ Thus, treatment of bleeding from are present in about half of all cirrhosis untarynervous system on the heart, esophageal varices has two components: patients. Each year, 5 percent to 20 and therefore relieve stress on the heart immediate treatment to control the percent of patients with cirrhosis expe­ (i.e., nonselective beta blockers). An present bleed and additional treatment rience the formation of varices. Once analysis of 11 prospective studies evalu­ to prevent consequent bleeding (i.e., varices have been formed, 5 percent to ating the effectiveness of nonselective secondary prophylaxis). 15 percent become large varices (Garcia beta blockers at preventing initial variceal In the United States the initial med­ and Sanyal 2001). Large varices are bleeding showed that beta blockers ical intervention for suspected variceal highly likely to rupture. Patients with decreased the risk of first bleed by at bleeding is to administer intravenous more severe liver disease tend to have least 40 percent (25 percent in control octreotide, which has been shown to more varices, and their varices tend to subjects vs. 15 percent in the treatment safely and effectively control variceal be larger. Spontaneous rupture of these group) after a median of 2 years (Garcia- bleeding by regulating vascular contrac­ varices and severe bleeding occur when Tsao 2001). Isosorbide mononitrate tion and decreasing the pressure in the the pressure on the portal vein, expressed and isosorbide dinitrate, which dilate portal venous system (Garcia and Sanyal as the difference between the pressure both the veins and the arteries, have been 2001; D’Amico et al. 1999). The method in the portal vein and the pressure in the evaluated for this purpose but have that has proven to be the best in con­ vein that drains out of the liver (i.e., proven to be ineffective as single agents. trolling acute variceal hemorrhage is the portal pressure gradient), is greater Adding low doses of these medications, endoscopic banding of the esophageal than 12 mm of mercury. Large varices however, to a regimen of nonselective varices. When banding is not possible, have a 2-year risk of bleeding sponta­ beta blockers may be of some help. injecting the varices with scarring sub­ neously in approximately 3 of every 10 Treatment in the form of ligation of stances also can be used. Because cir­ cases. Small varices have a risk of only 1 varices, in which elastic bands are placed rhotic patients with gastrointestinal in 10 (Garcia and Sanyal 2001). Because around varices using a device attached bleeding are at high risk of developing it is difficult to determine that varices to the end of the endoscope, also has serious infections, a prophylactic 7-day

254 Alcohol Research & Health Diagnosing and Treating Alcoholic Liver Disease

course of the antibiotic norfloxacin is and, in some cases, even reducing mor­ Variceal hemorrhage, ascites, and spontaneous bacterial recommended. A patient who has had tality rates. ■ peritonitis. Gastroenterology 120(3):726–748, 2001. the first episode of variceal bleeding faces GINES, A.; ESCORSELL, A.; GINES, P.; ET AL. an 8-in-10 chance of bleeding again in Incidence, predictive factors, and prognosis of the References hepatorenal syndrome in cirrhosis with ascites. the next 3 years. In addition, treatment Gastroenterology 105(1):229–236, 1993. with nonselective beta blockers, in a AKRIVIADIS, E.; BOTLA, R.; BRIGGS, W.; ET AL. manner similar to primary prophylaxis, GINES, P.; ORTEGA, R.; URIZ, J.; ET AL. Effects of Pentoxifylline improves short-term survival in terlipressin administration with and without albu­ should be given to patients who are not severe acute alcoholic hepatitis: A double-blind, min in hepatorenal syndrome (HRS). A Phase II already receiving these drugs. Patients placebo-controlled trial. Gastroenterology Study. 34:186A, 2001. who cannot tolerate the medication, 119(6):1637–1648, 2000. GUEVARA, M.; GINES, P.; FERNANDEZ-ESPARRACH, ANDY, S.Y., AND KE-QIN, H. Management of or for whom beta blockers are contra­ G.; ET AL. Reversibility of hepatorenal syndrome by indicated, can be treated with repeated ascites. Clinics in Liver Disease 5:541–568, 2001. prolonged administration of ornipressin and plasma banding of the varices at 10- to 14-day ANGELI, P.; VOLPIN, R.; GERUNDA, G.; ET AL. volume expansion. Hepatology 27(1):35–41, 1998a. Reversal of type 1 hepatorenal syndrome with the intervals. GUEVARA, M.; GINES, P.; BANDI, J.C.; ET AL. administration of midodrine and octreotide. Transjugular intrahepatic portosystemic shunt in hep­ Patients who rebleed despite therapy Hepatology 29(6):1690–1697, 1999. with beta blockers and endoscopy atorenal syndrome: Effects on renal function and BIRD, G.L., AND WILLIAMS, R. Factors determining vasoactive systems. Hepatology 28(2):416–422, 1998b. should be considered for the TIPS pro­ cirrhosis in alcoholic liver disease. Molecular Aspects HIRSCH, S.; BUNOUT, D.; DE LA MAZA, P.; ET AL. cedure or surgery performing a distal of Medicine 10(2):97–105, 1988. spleno-renal shunt, thereby decreasing Controlled trial on nutrition supplementation in CABRE, E., RODRIGUEZ-IGLESIAS, P.; CABALLERIA, outpatients with symptomatic alcoholic cirrhosis. the high pressure of the veins by con­ J.; ET AL. Short- and long-term outcome of severe JPEN: Journal of Parenteral and Enteral Nutrition necting the high-pressure vessels to the alcohol-induced hepatitis treated with steroids or 17(2):119–124, 1993. enteral nutrition: A multicenter randomized trial. inferior vena cava system, which is a low- HIRSCH, S.; DE LA MAZA, M.P.; GATTAS, V.; ET AL. Hepatology 32(1):36–42, 2000. pressure system (i.e., it carries oxygen- Nutritional support in alcoholic cirrhotic patients poor blood to the heart from the lower CARITHERS, R.L., JR.; HERLONG, H.F.; DIEHL, improves host defenses. Journal of the American half of the body). Despite the fact that A.M.; ET AL. therapy in College of Nutrition 18(5):434–441, 1999. patients with severe alcoholic hepatitis. A random­ HOLT, S.; GOODIER, D.; MARLEY, R.; ET AL. TIPS is more effective than endoscopic ized multicenter trial. Annals of Internal Medicine Improvement in renal function in hepatorenal syn­ therapy in decreasing rebleeding from 110(9):685–690, 1989. esophageal varices (19 percent rebleed­ drome with N-. Lancet 353(9149): CHEDID, A.; MENDENHALL, C.L.; GARTSIDE, P.; ET 294–295, 1999. ing with TIPS vs. 47 percent with endo­ AL. Prognostic factors in alcoholic liver disease. VA KEARNS, P.J.; YOUNG, H.; GARCIA, H.; ET AL. scopic therapy), 1 in 3 of these patients Cooperative Study Group. American Journal of Accelerated improvement of alcoholic liver disease Gastroenterology 86(2):210–216, 1991. will develop hepatic encephalopathy with enteral nutrition. Gastroenterology 102(1): after TIPS, and this intervention does CORRAO, G.; LEPORE, A.R.; TORCHIO, P.; ET AL. 200–205, 1992. The effect of drinking and smoking cigarettes not affect survival rates. Because of the KERSHENOBICH, D.; VARGAS, F.; GARCIA-TSAO, G.; on the risk of cirrhosis associated with alcohol con­ problems with encephalopathy and the ET AL. Colchicine in the treatment of cirrhosis of sumption. A case-control study. Provincial Group cost of TIPS, this approach usually is for the Study of . European the liver. New England Journal of Medicine 318(26): reserved as rescue therapy. Liver trans­ Journal of 10(6):657–664, 1994. 1709–1713, 1988. plantation is, of course, definitive ther­ KLATSKY, A.L., AND ARMSTRONG, M.A. Alcohol, CORTEZ-PINTO, H.; ALEXANDRINO, P.; CAMILO, smoking, coffee, and cirrhosis. American Journal of apy and should be considered for all of M.P.; ET AL. Lack of effect of colchicine in alco­ these patients. holic cirrhosis: Final results of a double blind ran­ Epidemiology 136(10):1248–1257, 1992. domized trial. European Journal of Gastroenterology LITTEN, R.Z.; ALLEN, J.P.; AND FERTIG, J.B. and Hepatology 14(4):377–381, 2002. Gamma-glutamyltranspeptidase and carbohydrate deficient transferrin: Alternative measures of exces­ Conclusions D’AMICO, G.; PAGLIARO, L.; AND BOSCH, J. Pharma­ cological treatment of : An sive alcohol consumption. Alcoholism: Clinical and Experimental Research 19(6):1541–1546, 1995. In summary, it now is generally possible evidence-based approach. Seminars in Liver Disease 19 (4):475–505, 1999. MADDREY, W.C.; BOITNOTT, J.K.; BEDINE, M.S.; to accurately diagnose ALD, and new ET AL. therapy of alcoholic hepatitis. DUVOUX, C.; ZANDITENAS, D.; HÉZODE, C.; ET AL. biomarkers or identifier proteins for Effects of noradrenalin and albumin in patients with Gastroenterology 75(2):193–199, 1978. detecting ongoing alcohol abuse and type I hepatorenal syndrome: A pilot study. Hepatology MATHURIN, P.; MENDENHALL, C.L.; CARITHERS, ALD are being investigated, as is the 36(2):374–380, 2002. R.L., JR.; ET AL. Corticosteroids improve short-term survival in patients with severe alcoholic hepatitis role of in ALD. Although there GARCIA, N., JR., AND SANYAL, A.J. Portal hyperten­ are no FDA-approved therapies for sion. Clinics in Liver Disease 5(2):509–540, 2001. (AH): Individual data analysis of the last three ran­ alcoholic liver disease, lifestyle changes, domized placebo controlled double blind trials of GARCIA-TSAO, G. Spontaneous bacterial peritonitis. corticosteroids in severe AH. Journal of Hepatology nutritional support, and “off-label” Gastroenterology Clinics of North America 21(1): 36(4):480–487, 2002. therapies such as PTX can improve out­ 257–275, 1992. MCCLAIN, C.; HILL, D.; SCHMIDT, J.; ET AL. come. Similarly, new therapies for com­ GARCIA-TSAO, G. Current management of the Cytokines and alcoholic liver disease. Seminars in plications are improving quality of life complications of cirrhosis and portal hypertension: Liver Disease 13(2):170–182, 1993.

Vol. 27, No. 3, 2003 255 MENDENHALL, C.L.; ANDERSON, S.; GARCIA-PONT, VA trial (Csp 352). Gastroenterology 122(Suppl.): Influence of alcohol withdrawal, and possible P.; ET AL. Short-term and long-term survival in 641A, 2002. effects of recent changes in general management of patients with alcoholic hepatitis treated with oxan­ the disease. American Journal of Medicine 44(3): drolone and . New England Journal of NILSSEN, O.; HUSEBY, N.E.; HOYER, G.; ET AL. 406–420, 1968. Medicine 311(23):1464–1470, 1984. New alcohol markers—how useful are they in pop­ ulation studies: The Svalbard Study 1988–89. RAMBALDI, A., AND GLUUD, C. Propylthiouracil for MENDENHALL, C.; ROSELLE, G.A.; GARTSIDE, P.; Alcoholism: Clinical and Experimental Research alcoholic liver disease. Cochrane Database System ET AL. Relationship of protein calorie malnutrition 16(1):82–86, 1992. Review (2)CD002800, 2002. to alcoholic liver disease: A reexamination of data from two Veterans Administration Cooperative O’GRADY, J.; LAKE, J.; AND HOWDLE, P.D. RAYNARD, B.; BALIAN, A.; FALLIK, D.; ET AL. Risk Studies. Alcoholism: Clinical and Experimental Comprehensive Clinical Hepatology. Orlando, FL: factors of fibrosis in alcohol-induced liver disease. Research 19(3):635–641, 1995. Mosby, 2000. Hepatology 35(3):635–638, 2002.

MERKEL, C.; MARCHESINI, G.; FABBRI, A.; ET AL. ORREGO, H.; BLAKE, J.E.; BLENDIS, L.M.; ET AL. RUNYON, B.A. Treatment of patients with cirrhosis The course of galactose elimination capacity in Long-term treatment of alcoholic liver disease with and ascites. Seminars in Liver Disease 17(3):249– patients with alcoholic cirrhosis: Possible use as a propylthiouracil. New England Journal of Medicine 260, 1997. surrogate marker for death. Hepatology 24(4):820– 317(23):1421–1427, 1987. 823, 1996. RUNYON, B.A. Management of adult patients with PARES, A.; CABALLERIA, J.; BRUGUERA, M.; ET AL. ascites caused by cirrhosis. Hepatology 27(1):264– MITZNER, S.R.; STANGE, J.; KLAMMT, S.; ET AL. Histological course of alcoholic hepatitis. Influence 272, 1998. Improvement of hepatorenal syndrome with extra­ of abstinence, sex and extent of hepatic damage. corporeal albumin MARS: Results of a Journal of Hepatology 2(1):33–42, 1986. SORT, P.; NAVASA, M.; ARROYO, V.; ET AL. Effect prospective, randomized, controlled clinical trial. of intravenous albumin on renal impairment and PATEK, A.J.; POST, J.; AND RALNOFF, O.D. Dietary Liver Transplantation 6(3):277–286, 2000. mortality in patients with cirrhosis and spontaneous treatment of cirrhosis of the liver. JAMA: Journal of bacterial peritonitis. New England Journal of MOREAU, R.; DURAND, F.; POYNARD, T.; ET AL. the American Medical Association 139:543–549, 1948. Terlipressin in patients with cirrhosis and type 1 Medicine 341(6):403–409, 1999. PESSIONE, F.; RAMOND, M.J.; NJAPOUM, C.; ET AL. hepatorenal syndrome: A retrospective multicenter U.S. Department of Health and Human Services study. Gastroenterology 122(4):923–930, 2002. Cigarette smoking and hepatic lesions in patients (USDHHS). 10th Special Report to the U.S. Congress with chronic hepatitis C. Hepatology 34(1):121– MORGAN, T.; NEMCHAUSKY, B.; SCHIFF, E.; ET AL. 125, 2001. on : Highlights from the Current Colchicine does not prolong life in patients with Research. Rockville, MD: DHHS, Public Health advanced alcoholic cirrhosis: Results of a prospec­ POWELL, W.J., JR., AND KLATSKIN, G. Duration of Service, National Institutes of Health, National tive, randomized, placebo-controlled, multicenter survival in patients with Laennec’s cirrhosis. Institute on Alcohol Abuse and Alcoholism, 2000.

256 Alcohol Research & Health