042700 evaluation of abnormal liver-enzyme results

The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne recommend adjusting aminotransferase values for sexand body-mass index,2 but these adjustments are rare-ly made. Aspartate aminotransferase is found, in de-creasing order of concentration, in the liver, cardiac EVALUATION OF ABNORMAL
muscle, skeletal muscle, kidneys, brain, pancreas, lungs, LIVER-ENZYME RESULTS IN
leukocytes, and erythrocytes. The highest level of ASYMPTOMATIC PATIENTS
alanine aminotransferase is in the liver, and levels ofthis enzyme are accordingly more specific indicatorsof liver injury. Both enzymes are released into the DANIEL S. PRATT, M.D., AND MARSHALL M. KAPLAN, M.D.
blood in increasing amounts when the liver cell mem-brane is damaged. Necrosis of liver cells is not re-quired for the release of the aminotransferases. In OW that routine laboratory testing is auto- fact, there is poor correlation between the degree of mated and is frequently part of an annual liver-cell damage and the level of the aminotransfer- checkup, physicians are often faced with the ases. If the aminotransferase levels are normal on re- problem of a patient with one abnormal result on testing, no further evaluation is necessary. If the results measurement of serum aminotransferases or alkaline of repeated tests remain abnormal, further evaluation phosphatase but no symptoms. Many batteries of screening tests now include measurement of serum The first step in the evaluation is to obtain a com- alanine aminotransferase, aspartate aminotransferase, plete history in an effort to identify the most com- alkaline phosphatase, and g-glutamyltransferase. Al- mon causes of elevated aminotransferase levels: alco- though these enzymes are present in tissues through- hol-related liver injury, chronic hepatitis B and C, out the body, they are most often elevated in pa- autoimmune hepatitis, hepatic steatosis (fatty infiltra- tients with liver disease and may reflect liver injury.
tion of the liver), nonalcoholic steatohepatitis, hemo- The first step in the evaluation of a patient with chromatosis, Wilson’s disease, alpha -antitrypsin de- elevated liver-enzyme levels but no symptoms is to ficiency, and a recently recognized cause, celiac sprue repeat the test to confirm the result. If the result is (Table 1). Table 2 lists the blood tests that can be still abnormal, the physician should evaluate the de- used to identify many of these disorders. It is more gree of the elevation. A minor elevation (less than efficient to order all the blood tests in the first group twice the normal value) may be of no clinical impor- initially, unless the history strongly suggests a defi- tance if the disorders listed in Table 1 have been nite diagnosis, such as alcohol abuse. The cause of ruled out and, in fact, may not even be abnormal.
the aminotransferase elevation can usually be identi- The normal range for any laboratory test is the mean fied on assessment of the pattern of the results of liv- value in a group of healthy persons ±2 SD. Thus, er-enzyme tests and additional testing.
5 percent of the results obtained from these normal The cause of an elevated alanine aminotransferase persons fall outside the defined normal range, 2.5 level varies greatly depending on the population stud- percent of which may be above the upper limit of ied. Among 19,877 Air Force trainees who volun- normal. There are also circumstances in which ele- teered to donate blood, 99 (0.5 percent) had elevat- vations in liver-enzyme levels are physiologic; for ex- ed alanine aminotransferase levels. A cause for the ample, alkaline phosphatase levels are increased in elevation was found in only 12: 4 had hepatitis B, healthy women during the third trimester of preg- 4 had hepatitis C, 2 had autoimmune hepatitis, 1 had nancy. The evaluation of the patient with an isolated cholelithiasis, and 1 had acute appendicitis. In a group elevation of an aminotransferase differs from that for of 100 consecutive blood donors with elevated ala- a patient with an isolated elevation of alkaline phos- nine aminotransferase levels, 48 percent had changes phatase or g-glutamyltransferase.
related to alcohol use, 22 percent had fatty liver, 17percent had hepatitis C, 4 percent had another iden- AMINOTRANSFERASE LEVELS
tified problem, and in the remaining 9 percent, no Aminotransferase levels are sensitive indicators of specific diagnosis was made. In another study of liver-cell injury and are helpful in recognizing hepa- 149 asymptomatic patients with elevated alanine ami- tocellular diseases such as hepatitis.1 Both aminotrans- notransferase levels who underwent liver biopsy, 56 ferases are normally present in serum at low levels, usu- percent had fatty liver, 20 percent had non-A, non- ally less than 30 to 40 U per liter. The normal range B hepatitis, 11 percent had changes related to alco- varies widely among laboratories. Some researchers hol use, 3 percent had hepatitis B, 8 percent hadother causes, and in 2 percent, no cause was identi-fied.5 A recent study assessed 1124 consecutive pa-tients who were referred for chronic elevations in ami- From New England Medical Center, Box 217, 750 Washington St., Bos- notransferase levels.6 Eighty-one of these patients had ton, MA 02111, where reprint requests should be addressed to Dr. Pratt.
2000, Massachusetts Medical Society.
no definable cause of the elevation and underwent a Downloaded from www.nejm.org by TIONG KIAT TEO MD on August 5, 2009 . Copyright 2000 Massachusetts Medical Society. All rights reserved. P R I M A RY C A R E
CAUSES OF ELEVATED
TABLE 1. CAUSES OF CHRONICALLY ELEVATED
AMINOTRANSFERASE LEVELS
Alcohol Abuse
Hepatic causes
The diagnosis of alcohol abuse can be difficult be- cause many patients conceal information about their alcohol use. The diagnosis is supported by the find- Steatosis and nonalcoholic steatohepatitis ing of a ratio of aspartate aminotransferase to alanine aminotransferase of at least 2:1. In a study of hun- Wilson’s disease (in patients «40 years old) dreds of patients who had histologically confirmed liver disorders, more than 90 percent of the patients Nonhepatic causes
who had an aspartate aminotransferase:alanine ami- notransferase ratio of at least 2:1 had alcoholic liver Inherited disorders of muscle metabolismAcquired muscle diseases disease.7 The percentage increased to more than 96 percent when the ratio was greater than 3:1. The in-creased ratio reflects primarily the low serum activityof alanine aminotransferase in patients with alcohol-ic liver disease. This decrease is due to an alcohol-related deficiency of pyridoxal 5-phosphate.8 Measurement of g-glutamyltransferase may also be helpful in diagnosing alcohol abuse. A g-glutamyl- TABLE 2. LABORATORY TESTS THAT MAY IDENTIFY THE CAUSE
transferase level that is twice the normal level in pa- OF ELEVATED AMINOTRANSFERASE LEVELS IN A PATIENT tients with an aspartate aminotransferase:alanine ami-notransferase ratio of at least 2:1 strongly suggeststhe diagnosis of alcohol abuse. However, the lack of DIAGNOSIS
specificity of the g-glutamyltransferase level precludes Initial tests
its use as a single test to diagnose alcohol abuse.
The degree of elevation of aminotransferase levels may also be helpful in identifying alcohol abuse. It is rare for the aspartate aminotransferase level to be more than eight times the normal value in patients with alcohol abuse, and it is even less common for the alanine aminotransferase level to be more than five times the normal value in such patients.7 In fact, the alanine aminotransferase level may be normal, even in patients with severe alcoholic liver disease.
Marked decrease in a-globulin bands Medication
Additional tests*
Reverse-transcriptase polymer-
A careful history-taking and meticulous review of laboratory data are critical for identifying a medica- tion as the cause of elevated aminotransferase levels.
A drug effect is a possibility if the increase in liver- enzyme levels was associated with the initiation of a medication. Almost any medication can cause an el- evation in liver-enzyme levels. Common ones includenonsteroidal antiinflammatory drugs, antibiotics, an- *If the results of the initial set of tests are normal, these additional tests tiepileptic drugs, inhibitors of hydroxymethylglutar-yl–coenzyme A reductase, and antituberculosis drugs(Table 3). In addition to medications, herbal prepa-rations and illicit drugs or substances may cause el-evations in liver-enzyme levels (Table 3).
liver biopsy. Of these 81 patients, 41 had steatosis, The easiest way to determine whether a medica- 26 had steatohepatitis, 4 had fibrosis, 2 had cirrho- tion is responsible for the elevation is to stop treat- sis, and 8 had normal histologic findings. The pa- ment and see whether the test results return to nor- tients with histologic evidence of fibrosis and cirrho- mal. If the identified medication is essential to the sis also had evidence of fatty metamorphosis. None patient’s well-being and no suitable substitute is avail- of the biopsies yielded a specific diagnosis except those able, the physician needs to make a risk–benefit analy- showing steatosis and steatohepatitis.
sis to determine whether the drug should be contin- Downloaded from www.nejm.org by TIONG KIAT TEO MD on August 5, 2009 . Copyright 2000 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne tors for infection is sufficient to make the diagnosis, TABLE 3. MEDICATIONS, HERBS, AND DRUGS OR
but the diagnosis is usually confirmed by measure- ment of serum levels of hepatitis C virus RNA with use of the reverse-transcriptase polymerase chain re- Medications
action. This approach is currently the gold standard for detecting hepatitis C infection.10 A positive test should prompt consideration of a liver biopsy to as- sess the severity of damage. Patients with chronic hepatitis C and evidence of fibrosis are usually treated.
Initial tests for hepatitis B infection include sero- logic tests for hepatitis B surface antigen, hepatitis B surface antibody, and hepatitis B core antibody. A pos- Inhibitors of hydroxymethylglutaryl–coenzyme A reductase itive test for hepatitis B surface antibody and core antibody indicates the presence of immunity to hep- atitis B, and another cause for the elevated amino- transferase levels should be sought. A positive test for hepatitis B surface antigen and core antibody indicates the presence of infection. Tests to determine wheth- Herbs and homeopathic treatments
er there is viral replication, including serologic tests for hepatitis B e antigen, hepatitis B e antibody, and hepatitis B virus DNA, should be undertaken. In pa- tients with positive tests for hepatitis B virus DNA and hepatitis B e antigen, liver biopsy and treatment Autoimmune Hepatitis
Drugs and substances of abuse
Autoimmune hepatitis occurs primarily in young- to-middle-aged women.11 The ratio of female pa- tients to male patients is 4:1.12 The diagnosis is based 5-Methoxy-3,4-methylenedioxymethamphetamine on the presence of elevated aminotransferase levels, the absence of other causes of chronic hepatitis, and serologic and pathological features suggestive of the Glues containing tolueneTrichloroethylene, chloroform disease.12 A useful screening test is serum protein elec-trophoresis. More than 80 percent of patients withautoimmune hepatitis have hypergammaglobuline-mia.13 However, a finding of more than twice thenormal level of polyclonal immunoglobulins is most ued despite the elevation in aminotransferase levels.
suggestive of the diagnosis. Additional tests that are Often, consultation with a hepatologist is necessary.
commonly ordered include serologic tests for anti- Occasionally, a liver biopsy is necessary to determine nuclear antibodies, antibodies against smooth mus- the nature and severity of liver injury.
cle, and liver–kidney microsomal antibodies. The firsttwo tests have reported sensitivities of 28 percent Chronic Hepatitis
and 40 percent, respectively.13 The third test is rarely Chronic hepatitis C is very common in the United positive among patients in the United States, Austral- States. Approximately 3.9 million Americans are pos- ia, and Japan.12 We do not recommend the routine itive for antibodies against hepatitis C, and an estimat- use of these three tests for the diagnosis of autoim- ed 2.7 million people are considered to be chronical- mune hepatitis. A liver biopsy is essential to confirm ly infected on the basis of the presence of hepatitis C virus RNA in serum.9 The risk of chronic infec-tion is highest in patients with a history of parenteral Hepatic Steatosis and Nonalcoholic Steatohepatitis
exposure to the virus (e.g., because of blood trans- The only clinical evidence of hepatic steatosis and fusions, intravenous drug use, or work-related duties), a condition that may be associated with it, nonalco- cocaine use, tattoos, body piercing, and high-risk sex- holic steatohepatitis, may be mild elevations in ami- notransferase levels. The levels are usually less than The initial test for hepatitis C infection is serolog- four times the normal value.14,15 In contrast to pa- ic testing for the hepatitis C antibody. The testing tients with alcohol-related liver disease, patients with has a sensitivity of 92 to 97 percent, depending on nonalcoholic steatohepatitis usually have an aspartate the assay.10 A positive test in a patient with risk fac- aminotransferase:alanine aminotransferase ratio that Downloaded from www.nejm.org by TIONG KIAT TEO MD on August 5, 2009 . Copyright 2000 Massachusetts Medical Society. All rights reserved. P R I M A RY C A R E
is less than 1:1.15,16 Fatty infiltration of the liver can tial screening test for Wilson’s disease is measurement be identified by ultrasonography or computed to- of serum ceruloplasmin (Table 2). The levels will be mography. Ultrasonography should be part of the reduced in approximately 85 percent of affected pa- evaluation of patients with chronically elevated ami- tients. Patients should also be examined by an oph- notransferase levels. The diagnosis of nonalcoholic thalmologist for Kayser–Fleischer rings.
steatohepatitis requires a liver biopsy. In addition to If the ceruloplasmin level is normal and Kayser– fatty infiltration, the histologic findings in patients Fleischer rings are absent, but the physician still sus- with nonalcoholic steatohepatitis include pericentral pects that Wilson’s disease may be present, the next fibrosis, inflammation, liver-cell necrosis, and hyaline test is a 24-hour urine collection for a quantitative cytoplasmic inclusions in hepatocytes that are iden- assessment of copper excretion. Excretion of more tical to Mallory’s bodies, which are characteristic of than 100 µg of copper per day is suggestive of Wil- son’s disease. The diagnosis is usually confirmed by The two conditions have different natural histories: liver biopsy to measure hepatic copper levels. Patients steatosis appears to have a benign course, whereas with Wilson’s disease have hepatic copper levels of nonalcoholic steatohepatitis can progress to cirrhosis.17 more than 250 µg per gram of liver, dry weight. Al- Liver failure as a result of nonalcoholic steatohepati- though the gene responsible for Wilson’s disease has tis is uncommon. Weight loss is the cornerstone of been identified, the number of disease-specific mu- treatment in patients who are obese.18 Other treat- tations is so great that molecular diagnosis is not yet ments for nonalcoholic steatohepatitis that are being studied include vitamin E and ursodiol. Vitamin Ewas associated with decreases in alanine aminotrans- Alpha -Antitrypsin Deficiency
ferase and aspartate aminotransferase levels and in Alpha -antitrypsin deficiency is an uncommon cause histologic abnormalities in two pilot studies.19,20 Ur- of chronic liver disease in adults. Decreased levels of sodiol decreased alanine aminotransferase and aspar- alpha -antitrypsin can be detected either by direct tate aminotransferase levels but not the histologic ab- measurement of serum levels or by the lack of a peak in a-globulin bands on serum protein electrophoresis.
In affected patients, however, serum levels of alpha - Hemochromatosis
antitrypsin may be increased in response to inflam- Hereditary hemochromatosis is a common genet- mation, causing a false negative result. The diagnosis ic disorder.22 Cost-effective screening starts with the is best established by phenotype determination.
measurement of serum iron and total iron-bindingcapacity (Table 2). A transferrin-saturation value (ob- Nonhepatic Causes
tained by dividing the serum iron level by the total In a recent study, occult celiac sprue was the cause iron-binding capacity) of more than 45 percent is of chronically elevated aminotransferase levels in 13 suggestive of hemochromatosis.22 Measurement of se- of 140 asymptomatic patients who were referred for rum ferritin provides less specific information, because this reason to a liver clinic.23 The diagnosis was made by measuring serum levels of antigliadin and antien- If screening tests suggest the presence of iron over- domysial antibodies. None of these patients had pri- load, a liver biopsy should be performed to assess mary biliary cirrhosis, a liver disease that is occasion- hepatic iron levels and the severity of liver damage.
ally found in patients with celiac sprue. On the basis A hepatic iron index (the hepatic iron level in micro- of this study, we recommend testing for occult celiac moles per gram of dry weight divided by the pa- sprue if other, more common causes of elevated ami- tient’s age) of more than 1.9 is consistent with the notransferase levels have been ruled out (Table 2).
presence of homozygous hereditary hemochromato- Elevated serum aminotransferase levels, especially sis.22 Genetic testing is now available to identify the aspartate aminotransferase levels, may be caused by mutation in the hemochromatosis (HFE ) gene that disorders that affect organs or tissues other than the causes the majority of cases. A liver biopsy is not liver, with the most common being striated muscle.
necessary for patients with hereditary hemochroma- Conditions or activities that can cause such elevations tosis who are younger than 40 years of age and who include subclinical inborn errors of muscle metabo- lism; acquired muscle disorders, such as polymyosi-tis; and strenuous exercise, such as long-distance run- Wilson’s Disease
ning. If striated muscle is the source of increased Wilson’s disease, a genetic disorder of biliary cop- aminotransferase levels, serum levels of creatine ki- per excretion, may cause elevated aminotransferase nase and aldolase will be elevated to the same degree levels in patients with no other symptoms of the dis- or to an even higher degree. Creatine kinase or aldo- ease. The clinical onset is usually between the ages lase levels should be measured if other, more common of 5 and 25 years, but the diagnosis should be con- hepatic conditions have been ruled out (Table 2).
sidered in patients up to the age of 40 years. The ini- If, despite comprehensive testing as outlined in Ta- Downloaded from www.nejm.org by TIONG KIAT TEO MD on August 5, 2009 . Copyright 2000 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne ble 2, the cause of the elevation in aminotransferase sitive and specific way of doing this, neither method levels remains unidentified, then a percutaneous liver is widely available.1 If gel electrophoresis is not avail- biopsy may be indicated. If the alanine aminotrans- able, measurement of either serum 5'-nucleotidase ferase and aspartate aminotransferase levels are less or g-glutamyltransferase should be performed. Lev- than twice the normal value and no chronic liver con- els of these enzymes are usually elevated in parallel dition has been identified, we recommend observa- with the elevation in the alkaline phosphatase level tion alone. Supporting this position are the results in patients with liver disorders, but they are not in- of two recent studies. The first study suggested that creased in patients with bone disorders. The finding close clinical follow-up is the most cost-effective strat- of an elevated serum alkaline phosphatase level but a egy for asymptomatic patients with negative tests for normal 5'-nucleotidase or g-glutamyltransferase lev- viral, metabolic, and autoimmune markers of liver dis- el should prompt an evaluation for bone diseases. Tests ease and chronically elevated aminotransferase lev- involving heat and urea denaturation of serum alka- els.24 The second study examined 36 patients with line phosphatase are still used by many laboratories a chronic elevation (at least 50 percent above nor- but are neither sensitive nor specific.
mal) of alanine aminotransferase, aspartate aminotrans- If the excess alkaline phosphatase is determined to ferase, or alkaline phosphatase levels.25 Patients with be of liver origin and persists over time, the patient strong evidence of a particular liver disease were ex- should be evaluated for chronic cholestatic or infil- cluded. All patients underwent liver biopsy. The re- trative liver diseases. Cholestatic diseases or condi- sults of liver biopsy led to a change in the diagnosis tions include partial obstruction of bile ducts, pri- in only five patients and to a change in treatment in mary biliary cirrhosis, primary sclerosing cholangitis, adult bile ductopenia, and cholestasis induced by the If the alanine aminotransferase and aspartate ami- use of drugs such as anabolic steroids. Infiltrative notransferase levels are persistently more than twice diseases include sarcoidosis, other types of granulo- the normal value, we recommend a biopsy. Although matous diseases, and less often, unsuspected metas- the results of the biopsy are unlikely to lead to a di- tasis of cancer to the liver. The appropriate initial tests agnosis or to changes in management, they often are ultrasonography of the right upper quadrant to provide reassurance to the patient and the physician assess the hepatic parenchyma and bile ducts and se- that no serious disorder is present.
rologic tests for antimitochondrial antibodies. Thepresence of antimitochondrial antibodies is highly CAUSES OF ELEVATED ALKALINE
suggestive of the presence of primary biliary cirrho- PHOSPHATASE LEVELS
sis. A finding of biliary dilatation suggests the pres- Elevations in serum alkaline phosphatase levels orig- ence of obstruction of the biliary tree. This finding is inate predominantly from two sources, liver and bone.1 unlikely in the absence of hyperbilirubinemia. Should Women in the third trimester of pregnancy have el- biliary dilatation or choledocholithiasis be present, evated serum alkaline phosphatase levels because of endoscopic retrograde cholangiopancreatography is an influx of placental alkaline phosphatase into their necessary to identify the cause of obstruction and blood. In persons with blood type O or B, serum can also be used to remove a stone or place a stent alkaline phosphatase levels may increase after the in- if required. Patients with serum antimitochondrial an- gestion of a fatty meal, because of an influx of intes- tibodies should undergo liver biopsy to confirm the tinal alkaline phosphatase. There are also reports of diagnosis of primary biliary cirrhosis.
a benign familial elevation in serum alkaline phos- If the serologic test for antimitochondrial anti- phatase levels because of increased levels of intestinal bodies is negative and ultrasonography reveals no ab- alkaline phosphatase. Alkaline phosphatase levels also normality, but the alkaline phosphatase level remains vary with age. Rapidly growing adolescents can have more than 50 percent above the normal level, we sug- serum alkaline phosphatase levels that are twice those gest a liver biopsy and either endoscopic retrograde of healthy adults as a result of the leakage of bone cholangiopancreatography or magnetic resonance cho- alkaline phosphatase into blood. Also, serum alka- langiopancreatography. If the increase in the alkaline line phosphatase levels normally increase gradually phosphatase level is less than this, the results of all between the ages of 40 and 65 years, particularly in the other liver-enzyme tests are normal, and the pa- women. The normal alkaline phosphatase level in an tient has no symptoms, we suggest observation alone.
otherwise healthy 65-year-old woman is more than This position is supported by the results of a recent 50 percent higher than the level in a healthy 30- The first step in the evaluation of an elevated al- CAUSES OF ELEVATED
kaline phosphatase level in a patient with no other g-GLUTAMYLTRANSFERASE LEVELS
symptoms is to identify the source of the elevation.
g-Glutamyltransferase is found in hepatocytes and Although electrophoretic separation on either poly- biliary epithelial cells. Measurement of serum g-glu- acrylamide gel or Sepharose columns is the most sen- tamyltransferase provides a very sensitive indicator of Downloaded from www.nejm.org by TIONG KIAT TEO MD on August 5, 2009 . Copyright 2000 Massachusetts Medical Society. All rights reserved. P R I M A RY C A R E
the presence or absence of hepatobiliary disease, but Relationship between pyridoxal 5'-phosphate deficiency and aminotransfer- the usefulness of this test is limited by its lack of ase levels in alcoholic hepatitis. Gastroenterology 1984;86:632-6.
9. Alter MJ, Kruszon-Moran D, Nainan OV, et al. The prevalence of hep-
specificity. Elevated levels of g-glutamyltransferase atitis C virus infection in the United States, 1988 through 1994. N Engl J have been reported in a wide variety of clinical con- Med 1999;341:556-62.
10. Schiff ER , de Medina M, Kahn RS. New perspectives in the diagnosis
ditions, including pancreatic disease, myocardial in- of hepatitis C. Semin Liver Dis 1999;19:Suppl 1:3-15.
farction, renal failure, chronic obstructive pulmo- 11. Krawitt EL. Autoimmune hepatitis. N Engl J Med 1996;334:897-903.
nary disease, diabetes, and alcoholism.27 High serum 12. Manns MP. Autoimmune hepatitis. In: Schiff ER , Sorrell MF, Mad-
drey WC, eds. Schiff ’s diseases of the liver. 8th ed. Vol. 2. Philadelphia:
g-glutamyltransferase levels are also found in patients who are taking medications such as phenytoin and 13. Czaja AJ. Natural history, clinical features, and treatment of autoim-
mune hepatitis. Semin Liver Dis 1984;4:1-12.
14. Diehl AM, Goodman Z, Ishak KG. Alcohollike liver disease in nonal-
Some have advocated the use of serum g-glutamyl- coholics: a clinical and histologic comparison with alcohol-induced liver in- transferase measurements to identify patients with un- jury. Gastroenterology 1988;95:1056-62.
15. Bacon BR , Farahvash MJ, Janney CG, Neuschwander-Tetri BA. Non-
reported alcohol use. The reported sensitivity of an alcoholic steatohepatitis: an expanded clinical entity. Gastroenterology elevated g-glutamyltransferase level for detecting al- cohol ingestion has ranged from 52 to 94 percent.29,30 16. Sorbi D, Boynton J, Lindor KD. The ratio of aspartate aminotransfer-
ase to alanine aminotransferase: potential value in differentiating nonalco-
Its lack of specificity makes the use of this test for holic steatohepatitis from alcoholic liver disease. Am J Gastroenterol 1999; this purpose questionable. In our opinion, measure- ment of serum g-glutamyltransferase is best used as 17. Matteoni CA, Younossi ZM, Gramlich T, Boparai N, Liu YC, Mc-
Cullough AJ. Nonalcoholic fatty liver disease: a spectrum of clinical and
a way of evaluating the meaning of elevations in oth- pathological severity. Gastroenterology 1999;116:1413-9.
er serum enzyme levels. For instance, it can be used 18. Eriksson S, Eriksson KF, Bondesson L. Nonalcoholic steatohepatitis in
obesity: a reversible condition. Acta Med Scand 1986;220:83-8.
to confirm the hepatic origin of elevated alkaline phos- 19. Hasegawa T, Yoneda M, Nakamura K, et al. Diagnostic significance of
phatase levels or to support the diagnosis of alcohol measurement of serum transforming growth factor b1 (TGF-b1) level and abuse in a patient with an elevated aspartate amino- effect of a-tocopherol in patients with nonalcoholic steatohepatitis (NASH). Gastroenterology 1997;112:Suppl 1:A1278. abstract.
transferase level and an aspartate aminotransferase:ala- 20. Lavine JE. Treatment of obesity-induced steatohepatitis with vitamin
nine aminotransferase ratio of at least 2:1.
E. Gastroenterology 1998;114:A1284. abstract.
21. Laurin J, Lindor KD, Crippin JS, et al. Ursodeoxycholic acid or clofi-
brate in the treatment of non-alcohol-induced steatohepatitis: a pilot study.
REFERENCES
Hepatology 1996;23:1464-7.
22. Powell LW, George DK, McDonnell SM, Kowdley KV. Diagnosis of
1. Pratt DS, Kaplan MM. Laboratory tests. In: Schiff ER , Sorrell MF,
hemochromatosis. Ann Intern Med 1998;129:925-31.
Maddrey WC, eds. Schiff ’s diseases of the liver. 8th ed. Vol. 1. Philadelphia: 23. Bardella MT, Vecchi M, Conte D, et al. Chronic unexplained hyper-
transaminasemia may be caused by occult celiac disease. Hepatology 1999; 2. Piton A, Poynard T, Imbert-Bismut F, et al. Factors associated with se-
rum alanine transaminase activity in healthy subjects: consequences for the 24. Das A, Post AB. Should liver biopsy be done in asymptomatic patients
definition of normal values, for selection of blood donors, and for patients with chronically elevated transaminases: a cost-utility analysis. Gastroenter- with chronic hepatitis C. Hepatology 1998;27:1213-9.
3. Kundrotas LW, Clement DJ. Serum alanine aminotransferase (ALT) el-
25. Sorbi D, McGill DB, Thistle JL, Henry JJ, Therneau TM, Lindor KD.
evation in asymptomatic US Air Force basic trainee blood donors. Dig Dis An assessment of the role of liver biopsies in asymptomatic patients with chronic liver test abnormalities. Hepatology 1999;30:Suppl:477A. abstract.
4. Katkov WN, Friedman LS, Cody H, et al. Elevated serum alanine ami-
26. Wolf PL. Clinical significance of an increased or decreased serum al-
notransferase levels in blood donors: the contribution of hepatitis C virus. kaline phosphatase level. Arch Pathol Lab Med 1978;102:497-501.
27. Goldberg DM, Martin JV. Role of gamma-glutamyl transpeptidase ac-
5. Hultcrantz R , Glaumann H, Lindberg G, Nilsson LH. Liver investiga-
tivity in the diagnosis of hepatobiliary disease. Digestion 1975;12:232-46.
tion in 149 asymptomatic patients with moderately elevated activities of se- 28. Rosalki SB, Tarlow D, Rau D. Plasma gamma-glutamyl transpeptidase
rum aminotransferases. Scand J Gastroenterol 1986;21:109-13.
elevation in patients receiving enzyme-inducing drugs. Lancet 1971;2:376- 6. Daniel S, Ben-Menachem T, Vasudevan G, Ma CK, Blumenkehl M.
Prospective evaluation of unexplained chronic liver transaminase abnormal- 29. Moussavian SN, Becker RC, Piepmeyer JL, Mezey E, Bozian RC. Se-
ities in asymptomatic and symptomatic patients. Am J Gastroenterol 1999; rum gamma-glutamyl transpeptidase and chronic alcoholism: influence of alcohol ingestion and liver disease. Dig Dis Sci 1985;30:211-4.
7. Cohen JA, Kaplan MM. The SGOT/SGPT ratio — an indicator of al-
30. Orrego H, Blake JE, Israel Y. Relationship between gamma-glutamyl
coholic liver disease. Dig Dis Sci 1979;24:835-8.
transpeptidase and mean urinary alcohol levels in alcoholics while drinking 8. Diehl AM, Potter J, Boitnott J, Van Duyn MA, Herlong HF, Mezey E.
and after alcohol withdrawal. Alcohol Clin Exp Res 1985;9:10-3.
Downloaded from www.nejm.org by TIONG KIAT TEO MD on August 5, 2009 . Copyright 2000 Massachusetts Medical Society. All rights reserved.

Source: http://www.sohs.com.sg/pdf/LFTs.pdf