When and how to screen for liver disease.

Time is of the essence for patients with liver disease because early intervention can often prevent life-threatening complications. Some patients with liver disease are asymptomatic, but their histories may reveal risk factors to justify liver analyte testing. What are these risk factors? What tests are essential for a basic liver panel? What tests can differentiate various types of liver disease?

Chronic liver disease is the eighth leading cause of death in the United States. Important causes of chronic liver disease include alcohol consumption and the viruses for hepatitis B (HBV) and hepatitis C (HCV). (Congenital, metabolic, autoimmune, and drug-induced conditions are also important contributors.)

Today's healthcare environment demands that any diagnosis be accomplished as efficiently and cost-effectively as possible. However, diagnosing liver disease can be a challenge, especially when patients are asymptomatic. In its early stages, liver disease is often insidious; it silently and progressively destroys the organ for months or even years before symptoms appear. Nonetheless, much of the morbidity and mortality associated with hepatic dysfunction can be prevented if liver disease is recognized before irreversible damage occurs. Diagnosing liver disease caused by viral hepatitis is especially urgent because of the risk of transmitting this infectious disease to others.

Potential high-risk indicators

Hepatologists recommend a liver panel for all patients at their periodic physical examinations. This approach permits the physician to obtain baseline measurements and to screen for common liver diseases, including hepatitis C and hemochromatosis. However, cost-cutting efforts have caused some insurers (including Medicare) not to reimburse providers for these tests.

In reality, many patients seek medical care only for treatment of symptoms or for preemployment physical examinations, but even these office visits present good opportunities to identify patients with undetected liver problems. Although liver disease may be present in patients who have no apparent increased risk, a patient's history should be reviewed for risk factors that might predispose to liver disease.

Exposure to contaminated blood or body fluids. Improvements in laboratory tests to screen blood donors have made the blood supply dramatically safer. Although donor-blood screening has significantly reduced the risk of post-transfusion infection with HBV and HCV, patients who have received blood transfusions are at increased risk for both viruses. In 1-2% of otherwise healthy adults with acute HBV infection and in more than 75% of patients with acute HCV infection, some form of chronic viremia develops, often with evidence of liver disease that can eventually cause cirrhosis, liver failure, or hepatocellular carcinoma. [1,2]

Until recently, transfused patients were more likely to acquire HCV than HBV. In the 1960s, the risk of contracting HCV from blood transfusion was about 1 in 10 units of blood. Since then, the risk has dropped dramatically to about 1 in 100,000 units of blood (according to Dr. George Nemo, scientific research group leader of the Transfusion Medicine Scientific Research Group, National Heart, Blood, and Lung Institute, Bethesda, MD). In most cases, HCV infection is initially silent, and some affected patients can remain asymptomatic for more than 20 years after infection. To date, approximately 4 million people are chronically infected with HCV in the US. [2] Less common causes of transfusion-associated liver disease, including contracting cytomegalovirus post-transfusion, can also occur.

Other risk factors for viral hepatitis include:

* hemodialysis

* organ transplants

* hemophilia

* occupational exposure to blood or body fluids for healthcare professionals, including doctors, dentists, nurses, and lab technologists

* domestic exposure to blood or body fluids when living with someone who has chronic HBV or is an HBV carrier (HCV familial transmission is rare)

* tattoos or body piercing

* intravenous or intranasal drug use

* sexual contact with an infected person.

Sexual activity. Hepatic experts recommend screening members of all groups with a high prevalence of infection. As a general rule, a high index of suspicion should be maintained in adolescent patients because of the prevalence of unprotected sex in this age group. At any age, multiple sexual partners and/or contact with prostitutes is particularly risky. Any sexually active individual who lives in a community with high rates of sexually transmitted diseases is also at greater risk for HBV.

In addition, because of the risk for congenital transmission, all pregnant women should be tested for hepatitis B surface antigen (HBsAg). Many states require vaccination of children for HBV. This should dramatically reduce the incidence of HBV infection as these children become teenagers and adults, which are also the ages at highest risk for the disease.

The US Public Health Service estimates that the risk of sexually transmitting HCV is less than 5%, which is relatively low. Predisposing factors that increase the likelihood of infection have not been identified. However, some evidence suggests that contact with multiple partners may increase the risk and that the carrier's virus level might be important.

Excessive alcohol consumption. High alcohol intake causes a broad spectrum of liver disease, ranging from alcoholic fatty liver (which may be benign) to cirrhosis. In some patients, the only signs of hepatic dysfunction are minimal abnormalities in liver analytes. Excessive alcohol consumption also increases the risk for acetaminophen-induced liver damage from even small amounts of this drug. Research suggests that chronic HBV and HCV infections might be important cofactors for transforming alcoholic liver disease into hepato-cellular carcinoma.

Although alcohol consumption is associated with an increased risk of liver disease, the exact quantity of alcohol and the precise duration of consumption required to cause this are unknown. Generally, however, daily consumption of 70-80 g is believed to increase the risk for men, and 35-40 g daily increases the risk for women. Some experts believe that the level of alcohol likely to cause liver damage is even lower (40 g daily for men and 20-30 g daily for women). A standard drink--12 oz of beer, 5 oz of wine, or 1.5 oz of 80-proof distilled spirits--contains approximately 12 g of absolute alcohol.

Hepatotoxic drug use. Various prescription and over-the-counter medications have been associated with liver toxicity. In most cases, hepatotoxicity associated with drug use is recognized in its acute stages before chronic liver damage develops. However, drugs such as methyldopa and methotrexate can cause a more indolent form of liver damage that might not be detected until cirrhosis and its complications occur. [3]

Because drug-induced chronic liver injury is uncommon, routine screening of aminotransferase levels in all patients taking medications is unnecessary and not fiscally prudent. Nonetheless, patients taking drugs that are known or believed to be hepatotoxic should be more frequently monitored, with the frequency of testing dependent on the type of drug used. [3-6]

Vitamins can also cause liver problems. For example, daily consumption of as little as 25,000 IU of vitamin A (5 times the recommended daily allowance) over a 6-year period has led to cirrhosis. [6] Patients may also use large quantities of what they perceive to be harmless herbs and other alternative health products that are potentially hepatotoxic. [3,7]

Occupation. Besides the increased risk of viral hepatitis for healthcare professionals who are exposed to blood or body fluids, people in other occupations might be at risk for liver damage as well. For example, exposure to organic solvents used in dry cleaning agents can damage the liver (see Table 1). [8] Hepatotoxic substances may be a hazard for people working in a number of occupations, including textile and dye manufacturing and painting; patients who are known to be exposed to these hepatotoxins should be screened. [8]

Family history. Liver disease in other family members might be important for a patient. Both common and rare hereditary hepatic diseases, such as hereditary hemochromatosis, Gilbert syndrome (benign hyperbilirubinemia), Wilson's disease (a copper metabolism abnormality), and [alpha] l-antitrypsin deficiency syndrome may be found in families. Genetic factors may also play a role in the pathogenesis of autoimmune chronic active hepatitis. [9]

Systemic diseases. Various extrahepatic illnesses can affect the liver. For example, a broad spectrum of liver disease has been noted in patients with sickle cell disease. Nonalcoholic steatohepatitis is common in patients with diabetes and may also be associated with type IV hyperlipidemia.

Which tests for the basic panel?

Although the tests included in a liver panel vary between laboratories, hepatic experts believe that the following tests are the most important for screening and diagnosis. (Sample reference intervals for each analyte discussed are provided in Table 2.)

Aminotransferases. Elevated levels of the enzymes aspartate aminotransferase (AST) and alanine aminotransferase (ALT) often reflect ongoing hepatic injury. Because ALT is found predominantly in the liver, its increase in serum more specifically reflects hepatocellular injury than AST, which can also be released from damage to muscles, kidneys, and brain. Nonetheless, a measurable increase in either enzyme suggests possible damage or even hepatocellular necrosis with a severity that varies directly with these enzymes (see Table 2).

In an asymptomatic patient, aminotransferase elevations can suggest a variety of conditions. For example, an elevated ALT could be the first sign of chronic hepatitis C. Asymptomatic increases (often slight) in ALT and AST are also common early signs of the progressive iron overload associated with hemochromatosis. Mild or moderate elevations (as defined by individual laboratory parameters) of aminotransferases have been noted in asymptomatic patients with nonalcoholic steatohepatitis; mildly high serum aminotransferase levels occur in asymptomatic patients with Wilson's disease.

In most primary liver diseases, the ALT and AST levels are elevated in roughly a 1:1 (DeRitis) ratio. The AST:ALT ratio is generally highest in alcoholic liver disease and lowest in acute and chronic viral hepatitis when the ratio may be [less than] 1:1. For patients with chronic alcohol-induced liver damage, the ratio is often [greater than] 2:1.. However, differences in laboratory methods limit the usefulness of the ratio. Another rule of thumb: If the ALT is high ([greater than] 400 U/L), alcoholic liver disease is unlikely, regardless of the aminotransferase ratio.

The alkaline phosphatases (ALPs). Multiple forms of ALP (some of which are true isoenzymes) are produced in the liver, intestine, kidney, placenta, and bone. Therefore, an increase in total ALP is not diagnostically specific for liver disease. Determination of the different isoenzymes might be diagnostically useful for a wide variety of diseases affecting bone or liver, but the differences in properties among the ALP isoenzymes are quite small, especially for bone and liver ALP. Methods for identifying ALP isoenzymes are not reliable.

An increase in the liver isoenzyme for ALP is often a useful indicator of impaired bile flow. [10] In adults ages 25 and younger and in children, the liver ALP level is normally high (see Table 2). However, a marked elevation in liver ALP levels, especially when the ALT and AST are normal or only modestly increased, suggests bile duct obstruction or disease of the bile ducts, such as primary biliary cirrhosis or primary sclerosing cholangitis. A marked increase in liver ALP with normal or mildly increased bilirubin and AST and ALT levels occurs when primary or metastatic cancer, fungi, or other pathogens infiltrate the liver.

Bilirubin. The bilirubin assay is important for evaluating liver function because it reflects the liver's functional ability to absorb, conjugate, and excrete bilirubin into the bile. Serum bilirubin might be elevated for any of several reasons, including the following:

* increased bilirubin production in the presence of active erythrocyte hemolysis 

* decreased uptake by the liver

* decreased conjugation in the liver

* decreased secretion from the liver

* bile duct blockage.

Normally, the total bilirubin level is [less than] 1.1 mg/dL, and approximately 70% is unconjugated. Elevated levels of unconjugated bilirubin can be associated with increased erythrocyte hemolysis, decreased liver uptake of bilirubin, or decreased bilirubin conjugation. When [greater than] 80% of the total bilirubin is unconjugated, either hemolysis or Gilbert syndrome is the likely cause. In Gilbert syndrome, the unconjugated bilirubin elevation is usually not [greater than] 3 mg/dL.

An increase in conjugated bilirubin suggests decreased secretion or bile duct obstruction. If [greater than] 50% of the total bilirubin is conjugated, either hepatocellular dysfunction or cholestasis may exist. In patients with bile duct obstruction or diseases, the ALP level also is often elevated. In patients with common bile duct obstruction caused by gallstones, it's unusual for the bilirubin level to exceed 15 mg/dL because obstruction is usually incomplete. In fact, the level is usually [less than] 6.0 mg/dL. With complete obstruction, bilirubin levels of 25-30 mg/dL may be seen. Higher levels suggest that extrahepatic cholestasis is an unlikely diagnosis. [11]

Iron status. In addition to these routine liver tests, a serum assessment of iron status should be obtained at the initial screening exam to rule out iron overload associated with hemochromatosis. This disease and the hepatic damage it causes are reversible, but only if recognized and treated early. Most patients are asymptomatic, and liver analytes may not be markedly elevated.

Viral testing. Every patient with an elevated AST or ALT level or risk factors for viral hepatitis should also be tested for anti-HCV and hepatitis B core antibody (anti-HBc). Anti-HBc is detected within a few weeks after the appearance of HBsAg and is then followed weeks to months later by the appearance of antibodies to HBsAg (anti-HBs). If the patient is acutely infected with HBV, evidence of HBsAg will be apparent 1-10 weeks after exposure and will remain detectable for a few weeks during the acute phase of the illness. However, for patients with chronic HBV, HBsAg can be detected for years. [1] Most HCV antibody tests become positive within 8-12 weeks after exposure. More than 75% of those with positive test results are chronic HCV carriers.

It is now possible to test for HBV-DNA and HCV-RNA, and some hepatologists use these tests to confirm initial results. However, the use of such tests in the initial screen would be costly and is not warranted.

Confirming abnormal results

Abnormal results should be repeated with a second specimen, and a second abnormal result should trigger an additional workup. If the second result is normal, the test should be performed a third time for confirmation. Confirming the abnormal results helps rule out the possibility that transient non-hepatic illnesses, undetected preanalytical or technical error, or some other factor caused the abnormal result.

When evaluating liver panels, any laboratory value higher than the upper limit of the reference interval should be carefully investigated. Even slight increases might be important. For example, if the upper limit of the reference interval for AST is 36 U/L and a patient's value is 37 or 39, the physician can review the patient history to identify possible risk factors, such as:

* the use of drugs that can cause a mild elevation of liver enzymes

* the presence of a systemic disease

* a family history of liver disease.

Liver test results can also be above the upper limit of the reference interval for reasons other than liver disease. For example, a bacterial or nonhepatic viral infection, increased alcohol intake, or medication that affects the liver can all cause a spurious elevation of AST or ALT. In addition, AST is found in muscle and can also be elevated in patients who have engaged in strenuous workouts, such as running a marathon or weight lifting.

If the elevation is spurious, attempts should be made to eliminate the possible cause before repeating the test. Clinical and laboratory abnormalities associated with drug-induced hepatitis usually resolve within 2-4 weeks after discontinuing the offending medication. Abstinence from alcohol for 2-4 weeks will often substantially lower the AST level, thus helping to confirm or rule out alcohol use as the cause of the AST elevation. If the reason for an increase is suspected to be an acute problem or if the elevation is marked, the test should be repeated without delay.

When results are either slightly elevated or within the upper limits of the reference interval, the test should be repeated. If the result is still abnormal or even borderline above the reference interval, further testing may be warranted to identify some of the less common liver diseases, such as metabolic or autoimmune.

Although no increase should be ignored, some patients have slight elevations in various liver analytes with no demonstrable cause. If the patient's results for aminotransferase ALT/AST, ALP, and bilirubin are indeed normal, it's usually safe for the physician to wait until the next scheduled physical examination to retest for liver function. However, if the patient is engaged in high-risk behavior, more frequent follow-up may be advisable.

Other tests

Other liver function tests are best reserved for follow-up after initial abnormal results have been confirmed. Results from more specialized tests, such as ceruloplasmin and antimitochondrial antibodies, can identify specific etiologies of acute or chronic liver disease but have no role in routine liver testing for asymptomatic patients.

Gamma-glutamyltransferase (GGT).

The enzyme GGT is found in the liver, pancreas, and kidney. Measurements of GGT are extremely sensitive,, and elevated levels are associated with most liver disease. Sometimes GGT is found in patients without liver disease. Elevations are induced by many drugs, including acetaminophen and alcohol. Levels may be elevated in modest drinkers, even in the absence of liver damage or inflammation. However, this enzyme is not elevated in all patients with chronic alcoholism. Thus, the GGT determination is not a useful screening test because of its extreme sensitivity, and it is no longer in Medicare's basic liver panel.

Albumin and prothrombin time. Most plasma proteins, including albumin and many coagulation proteins, are produced in the liver. Serum albumin levels and the prothrombin time (PT) (which measures coagulation factors I, II, V, VII, and X) can reflect the functional status of the liver, although changes in these proteins are not specific for liver disease. [12] PT results are especially abnormal when hepatic dysfunction is fulminant and usually of an infectious or toxic origin. [13] It's a good prognostic sign to see the PT return to normal and a poor prognostic sign when serial PT results continue to rise or coagulation factor assay results decrease.

The prothrombin time and a platelet count are important tests to perform before a liver biopsy. In many institutions, a liver biopsy is postponed until PT results are [less than] 2 seconds (an arbitrary number that can vary from 2-4 seconds) above the reference interval of approximately 10-12 seconds and the platelet count is [greater than] 50,000/mL (normal is approximately 150,000-400,000/mL). [13,14] In asymptomatic patients, these tests can be reserved for follow-up when aminotransferase, ALP, or bilirubin levels are elevated.

Lactate dehydrogenase. Lactate dehydrogenase is a relatively poor analyte because it is highly nonspecific. This supplementary liver test can be used as a marketer for hemolysis. Lactate dehydrogenase (LD) is present in many tissues, including liver, cardiac tissue, kidneys, and erythrocytes. In addition, many of these tissues exhibit different LD isoenzyme concentrations. When only a specific tissue (such as liver) is damaged, measurement of total LD can be informative. However, when multiple organs are damaged, measurement of LD isoenzymes is more informative. LD5 and LD4 are found primarily in the liver and are increased in viral or toxic hepatitis, extrahepatic biliary obstruction, acute necrosis of the liver, and cirrhosis. [10]

Conclusion

The diagnosis of liver disease will remain difficult in asymptomatic patients unless risk factors are considered and evaluated by appropriate screening tests for liver function. Follow-up of abnormal results should include repeat testing, further investigation of patient history, and further diagnostic testing, when appropriate.

Nancy Bach, MD, is assistant professor of medicine, Mount Sinai School of Medicine of the City University of New York, NY; Raymond S. Koff, MD, is professor of medicine, University of Massachusetts Medical School, Worchester, MA, and chairman, Department of Medicine, Columbia MetroWest Medical Center, Framingham, MA; and Willis Maddrey, MD, is professor of internal medicine and executive vice president for clinical affairs, The University of Texas Southwestern Medical Center at Dallas, TX, and IM Internal Medicine board member emeritus.

References

(1.) Koff RS. Hepatitis B and D, in Bone RC (series ed). Current Practice of Medicine. Vol 4. Philadelphia, PA: Current Medicine, Inc; 1996; 3.1-3.7.

(2.) Gross, JB Jr. Clinician's guide to hepatitis C. Mayo Clin Proc. 1998;73(4):355.

(3.) Lee WM. Drug-induced hepatotoxicity. N Engl J Med. 1995;333(17):1118.

(4.) Schenker S, Mazloum B. Drug-induced liver disease, in Bone RC (series ed). Current Practice of Medicine. Vol 4. Philadelphia, PA: Current Medicine, Inc; 1996; 8.1-8.7.

(5.) Watkins PB, Whitcomb RW. Hepatic dysfunction associated with troglitazone. N Engl J Med. 1998;338(13):916.

(6.) Moseley RH. Evaluation of abnormal liver function tests. Med Clin North Am. 1996; 80(5):887.

(7.) Ernst E. Harmless herbs? A review of the recent literature. Am J Med. 1998;104(2):170.

(8.) Redlich C, Brodkin CA. Liver diseases. In: Rosenstock L, Cullen MR, eds. Textbook of Clinical Occupational and Environmental Medicine. Philadelphia, PA: WB Saunders Co; 1994; 423-436.

(9.) Krawitt EL. Idiopathic autoimmune chronic active hepatitis. In Bone RC (series ed). Current Practice of Medicine. Vol 4. Philadelphia, PA: Current Medicine, Inc; 1996; 6.1-6.6.

(10.) Moss DW, Henderson RH. Enzymes. In: Burtis CA, Ashwood ER, eds. Tietz Textbook of Clinical Chemistry. Philadelphia, PA: WB Saunders Company; 1994;735-896.

(11.) Kamath PS. Clinical approach to the patient with abnormal liver test results. Mayo Clin Proc. 1996;71(11):1089.

(12.) Sherwin JE, Sobenes JR. Liver function. In: Kaplan LA, Peace AJ, Kazmierczak SC, eds. Clinical Chemistry: Theory, Analysis, and Correlation. St. Louis, MO: Mosby; 1996;505-527.

(13.) Balistreri WF, Rej R. Liver Function. In: Burtis CA, Ashwood ER, eds. Tietz Textbook of Clinical Chemistry. Philadelphia, PA: WB Saunders Company; 1994;1449-1512.

(14.) Bruns C, Metz J. Laboratory methods in hematology. In: McKenzie SB. Textbook of Hematology. Baltimore, MD: Williams & Wilkins; 1996;601-622.

Selected human hepatotoxins

Alcohol

Ethyl alcohol

Aromatic hydrocarbons

Styrene

Toluene 

Xylene 

Chlorinated aromatic compounds

Chloronaphthalenes

Polychlorinated biphenyls

Environmental agents

Aflatoxin

Amanita phalloides (death cap mush-room) toxin

Halogenated hydrocarbons

Anesthetic gases

Carbon tetrachloride

Chloroform

Halothane 

Methoxyflurane

Tetrachloroethane

Tetrachlorethylene 

Trichloroethylene 

Vinyl chloride

Metallic compounds

Arsenic

Beryllium

Cadmium

Copper

Lead

Phosphorus

Nitro compounds

N,N-dimethylformamide

Dinitrobenzene 

2-nitropropane

Trinitrotoluene 

Pesticides

Dioxin (TCDD; 2,3,7,8 tetrachlorodibenzo-p-dioxin)

Fungicide

Hexachlorobenzene

Insecticide

Paraquat 

Oarganochlorine pesticide

Chlordecone

Source: Adapted with permission from Redlich C, Brodkin CA. Liver diseases. In: Rosenstock L, Cullen MR, eds. Textbook of Clinical Occupational and Environmental Medicine, Philadelphia, PA: WB Saunders Co; 1994:423-434.