Mildly Elevated Levels in the Asymptomatic Patient PAUL T. GIBONEY, M.D., Keck School of Medicine, University of Southern California, Los Angeles, California

Mild elevations in liver chemistry tests such as transaminase and can reveal serious underlying conditions or have transient and benign etiologies. Potential causes of liver transaminase elevations include viral , alcohol use, medica- tion use, steatosis or steatohepatitis, and . The history should be thorough, with special attention given to the use of medications, vitamins, herbs, drugs, and alcohol; family history; and any history of blood-product transfusions. Other common health conditions, such as , disease, and thyroid disease, can cause or augment liver transaminase elevations. The recent American Gastroenterologi- cal Association guideline regarding the evaluation and management of abnormal liver chemistry tests proposes a practical, algorithmic approach when the history and physical examination do not reveal the cause. In addition to liver chemistries, an initial serologic evalu- ILLUSTRATION BY TODD BUCK ation includes a prothrombin time; albumin; complete blood count with platelets; , B, and C serologies; and iron studies. Depending on the etiology, management strategies may include cessation of alcohol use, attention to medications, control of diabetes, and modification of lifestyle fac- tors such as obesity. If elevations persist after an appropriate period of observation, further testing may include ultra- sonography and other studies. In some cases, biopsy may be indicated. (Am Fam Physician 2005;71:1105-10. Copyright© 2005 American Academy of Family Physicians.)

epatic transaminase tests such as who have more patients with obesity, diabe- (ALT) and tes, and hyperlipidemia will have to address aspartate transaminase (AST) this issue more often. often are part of standard labo- Given the frequency of this problem, physi- H ratory panels in asymptomatic outpatients, cians should develop an informed approach to similar to screening tests for blood donors the investigation of transaminase elevations. and for life insurance applicants. The evalu- An audit of primary care practices found that ation of an abnormal ALT or AST level in these abnormalities are not always investi- an asymptomatic patient therefore is a com- gated appropriately and that opportunities mon challenge encountered by primary care to intervene in treatable cases sometimes are physicians. missed.3 No controlled clinical trials have According to the American Gastroen- compared approaches to the management of terological Association (AGA), 1 to 4 percent abnormal transaminase levels. However, the of the asymptomatic population may have AGA recently published a technical review1 elevated serum liver chemistries.1 This is and a position statement4 on the evaluation consistent with the usual definition of an of liver chemistry tests. This article reviews elevated transaminase level of the interpretation of ALT and AST levels and the top 2.5 percent of the pop- summarizes the AGA recommendations on Up to 4 percent of the ulation range. Although one addressing reported elevations. asymptomatic population study2 of 19,877 asymptom- may have elevated serum atic young Air Force trainees Markers of Hepatic Injury and Necrosis liver chemistries. found that only 0.5 percent had ALT and AST are two of the most reliable elevated ALT levels, physicians markers of hepatocellular injury or necrosis.

March 15, 2005 ◆ Volume 71, Number 6 www.aafp.org/afp American Family Physician 1105 Downloaded from the American Family Physician Web site at www.aafp.org/afp. Copyright© 2005 American Academy of Family Physicians. For the private, noncommercial use of one individual user of the Web site. All other rights reserved. Contact [email protected] for copyright questions and/or permission requests. Strength of Recommendations TABLE 1 Etiology of ALT or AST Elevations Key clinical recommendation Label References When Less Than Five Times Normal

An algorithmic approach to evaluating mildly C 1 Common hepatic causes abnormal liver functions is recommended. Alcohol In the asymptomatic patient with negative serum C 1 Cirrhosis testing and mild transaminase elevations, a period of lifestyle modification can be tried. (chronic) If abnormalities persist at the six-month follow- C 1 (chronic) up visit, an ultrasonography of the liver is the Steatosis/steatohepatitis recommended imaging modality. Medications/toxins ALT and AST are not useful screening tests in an C 1, 10 Acute otherwise healthy population. Less common hepatic causes The AST/ALT ratio is only somewhat helpful in C 5, 7 Autoimmune hepatitis diagnosis. Hemochromatosis

ALT = alanine transaminase; AST = aspartate transaminase. Alpha1-antitrypsin deficiency Wilson’s disease A = consistent, good-quality patient-oriented evidence; B = inconsistent or limited- quality patient-oriented evidence; C = consensus, disease-oriented evidence, usual Nonhepatic causes practice, opinion, or case series. See page 1046 for more information. Celiac disease Hemolysis Myopathy Their levels can be elevated in a variety of Hyperthyroidism hepatic disorders. Of the two, ALT is thought Strenuous exercise to be more specific for hepatic injury because Macro-AST

it is present mainly in the of the liver ALT = alanine transaminase; AST = aspartate trans- and in low concentrations elsewhere. AST aminase. has cytosolic and mitochondrial forms and Adapted with permission from Pratt DS, Kaplan MM. is present in tissues of the liver, heart, skel- Evaluation of abnormal liver- results in asymp- etal muscle, kidneys, brain, , and tomatic patients. N Engl J Med 2000;342:1267, with additional information from reference 5. lungs, and in white and red blood cells. AST is less commonly referred to as serum glu- tamic oxaloacetic transaminase and ALT as serum glutamic pyruvic transaminase. mary care medicine. The range of possible Although levels of ALT and AST can be etiologies at this level of transaminase eleva- extremely elevated (exceeding 2,000 U per L tion is broader (Table 15,6) and the tests less in cases of injury and necro- specific. It also is important to recall that sis related to drugs, toxins, ischemia, and patients with normal ALT and AST levels hepatitis), elevations less than five times the can have significant in the set- upper limit of normal (i.e., about 250 U per L ting of chronic hepatocyte injury (e.g., cir- and below) are much more common in pri- rhosis, hepatitis C). The ratio of AST to ALT has some clinical utility, but has important limitations. In many The Author forms of acute and chronic liver injury or ste- PAUL T. GIBONEY, M.D., is assistant professor of clinical family medicine at the atosis (fatty infiltration of the liver), the ratio Keck School of Medicine, University of Southern California, Los Angeles. He is less than or equal to 1. This is particularly received his medical degree from Northwestern University School of Medicine, true in patients with hepatitis C. However, an Chicago, and completed a residency in family medicine at John Peter Smith AST/ALT ratio greater than 2 characteristically Hospital, Fort Worth, Tex. is present in . A recent study7 Address correspondence to Paul T. Giboney, M.D., 123 S. Alvarado St., Los of 140 patients with nonalcoholic steatohepa- Angeles, CA 90057. Reprints are not available from the author. titis (NASH; confirmed by liver biopsy) or

1106 American Family Physician www.aafp.org/afp Volume 71, Number 6 ◆ March 15, 2005 Elevated Transaminase Levels

found a mean AST/ALT and in male ALT is thought to be more ratio of 0.9 in patients with NASH and 2.6 in bank employees and found the specific than AST for patients with alcoholic liver disease. Within positive predictive value of the hepatic injury because it is the population studied, 87 percent of patients test to be low. Only 3.9 percent present mainly in the cyto- with an AST/ALT ratio of 1.3 or less had NASH of the men with an abnormal sol of the liver and in low (87 percent sensitivity, 84 percent specific- ALT level had hepatitis C; 8 concentrations elsewhere. ity). The severity of NASH as measured by percent were excessive users of the degree of fibrosis increased, as did the alcohol; and 35.7 percent had AST/ALT ratio. A mean ratio of 1.4 was found fatty liver. in patients with cirrhosis related to NASH. Wilson’s disease, a rare problem, can cause the Management AST/ALT ratio to exceed 4.5 While these ratios A thorough medical history and physi- are suggestive of certain conditions, there is cal examination are the cornerstone of the too much overlap between groups to rely on evaluation of patients with mildly elevated them exclusively when making a diagnosis. liver transaminase levels.1 The history should (LDH) is a less spe- attempt to identify risk factors for disease, cific marker of hepatocellular necrosis and with special attention directed toward fam- usually does not add diagnostic information ily history, medications, vitamins, herbal to that obtained with ALT and AST testing. supplements, drug use, alcohol use, abnor- An exception to this is the transient but mas- mal liver testing, blood-product transfusions, sive rise of LDH in cases of ischemic hepatitis and symptoms of liver disease. Table 26 lists and its sustained elevation that, along with selected medications and herbal supplements elevated levels, suggests that may cause elevated transaminase lev- malignant infiltration of the liver.5 els. Physicians should ask patients directly Elevations of ALT and AST are not exclu- about their use of illicit drugs, herbal supple- sive to liver pathology. Hyperthyroidism has ments, and other alternative “supplements” been found in several studies to increase serum levels of liver including ALT TABLE 2 and AST.8 Genetic influences on the level of 9 Common Agents That Can Cause Liver Transaminase ALT also are possible. A study of Danish Elevations twins showed that genetic factors accounted for 33 to 66 percent of the variation in ALT, Medications Herbal supplements/vitamins gamma glutamyl transpeptidase, LDH, and Acetaminophen Chaparral leaf in patients 73 to 94 years of age. Amiodarone (Cordarone) Ephedra The AGA technical review states that serum Amoxicillin- clavulanic acid Gentian ALT has diurnal variation, may vary day to Carbamazepine (Tegretol) Germander day, and may be affected by exercise. It also Fluconazole (Diflucan) Jin bu huan notes that serum AST may be 15 percent Glyburide (Micronase) Kava 1 higher in black men than white men. Scutellaria (skullcap) Another cause of elevated liver trans- Isoniazid (INH) Senna aminase levels is muscle injury. Strenuous Ketoconazole (Nizoral) Shark cartilage exercise or myopathy can cause elevations Labetalol (Normodyne) Vitamin A (especially of AST) without causing any Nitrofurantoin (Furadantin) other symptoms. A kinase or other Nonsteroidal anti-inflammatory drugs muscle marker can be obtained to confirm Phenytoin (Dilantin) or exclude such a process. Protease inhibitors Annual screening of healthy, asymptom- Sulfonamides atic patients for liver disease using ALT and Trazodone (Desyrel) AST levels is not useful. A Japanese study10 assessed the accuracy of ALT and AST for Information from reference 6. detecting hepatitis C, excess alcohol use,

March 15, 2005 ◆ Volume 71, Number 6 www.aafp.org/afp American Family Physician 1107 cination, hepatitis B remains a common TABLE 3 cause of chronic liver disease in adults. Test- Clues in the Evaluation of Mildly Elevated Liver ing for hepatitis C is essential because its Transaminase Levels incidence has increased in the past decade, and new treatment strategies have been Clinical clue Suggested diagnosis developed that can address this frequently 11 Longstanding alcohol abuse Cirrhosis missed problem. Intravenous drug use, history of blood Hepatitis B or C A prothrombin time (PT) and serum product transfusions, nonsterile needle albumin should be ordered to identify exposure, AST/ALT ratio < 1.0 patients with abnormalities of protein Obesity, diabetes, hyperlipidemia, AST/ALT Steatosis/steatohepatitis synthesis and liver function. Evaluation ratio < 1.0 should be accelerated for patients with AST/ALT ratio > 2.0 Alcoholic liver disease, impaired hepatic synthetic function. A Wilson’s disease complete blood count with platelets also Increased iron levels Hemochromatosis should be ordered. In addition to ruling Polypharmacy, illicit drug use, or certain Substance/ herbal supplement use medication-induced out infection, neutropenia or thrombo- Frequent, strenuous exercise Exercise-induced cytopenia can, along with an elevated Intestinal bloating; oily, bulky stools Celiac sprue PT, suggest advanced liver disease. An Hypergammaglobulinemia Autoimmune hepatitis elevated mean red cell volume suggests Reduced ceruloplasmin levels, Kayser- Wilson’s disease heavy alcohol intake. Alkaline phospha- Fleischer ring tase and bilirubin are markers for hepatic Depressed thyroid-stimulating hormone levels Hyperthyroidism and should be ordered as part of the initial laboratory evaluation. While ALT = alanine transaminase; AST = aspartate transaminase. sometimes useful, they often are normal in the presence of hepatic injury.

because these sometimes are omitted from LIFESTYLE MODIFICATION the patient’s initial response to questions. If the patient is asymptomatic and the initial The presence of other significant health serum testing is negative, a period of lifestyle conditions that can cause or augment liver modification can be attempted. Effective life- transaminase elevations also should be style modification includes complete absti- noted; examples are diabetes, heart disease nence from alcohol, control of diabetes and (including congestive ), thyroid hyperlipidemia, weight loss in overweight disease, muscle disease, and cancer. Physical patients, and stopping or changing poten- findings and sequelae of liver dysfunction tially hepatotoxic medications and supple- are given in Table 3. ments. Such lifestyle changes directly impact Once the history and physical examina- several of the causes of mild transaminase tion are completed, additional testing can elevation (Table 1).5,6 These seemingly small help discern the etiology of the transami- modifications may be all that is needed to nase elevation (Figure 1).4 correct the abnormalities.

INITIAL LABORATORY EVALUATION FOLLOW-UP AND IMAGING STUDIES Additional laboratory tests should be A repeat set of liver chemistries should be obtained when the history and physical obtained after six months. If the patient’s examination show no obvious etiology for presentation changes or the physician has ALT and AST elevations. , total iron- concern for an evolving process, shorter binding capacity, and can be intervals can be used. If abnormalities per- used to look for hemochromatosis, while sist at the six-month follow-up visit, ultra- hepatitis A, B, and C serologies are used to sonography of the liver is recommended. rule out acute or chronic hepatitis. Computed tomography of the abdomen also Despite the emergence of widespread vac- is used in this setting, although clinical trials

1108 American Family Physician www.aafp.org/afp Volume 71, Number 6 ◆ March 15, 2005 Elevated Transaminase Levels

Management of Mild ALT and AST Abnormalities

Elevated ALT and AST less than five times normal

History and physical examination; discontinue hepatotoxic medications and alcohol.

Confirm abnormality if an error is suspected.

Liver chemistries; prothrombin time; albumin; complete blood count with platelets; hepatitis A, B, and C serologies; serum iron; total iron-binding capacity; ferritin

Negative If abnormal, proceed Evidence of chronic liver with directed Expedited evaluation disease or hepatic evaluation/treatment decompensation

Negative serology, asymptomatic patient without hepatic decompensation

Lifestyle modification: discontinue alcohol use, stop hepatotoxic medications, weight loss, diabetes control

Six-month follow-up

Repeat liver chemistries Normal Observation

Abnormal

Ultrasonography and serologic evaluation: antinuclear antibodies, anti–smooth-muscle

antibody, ceruloplasmin, alpha1-antitrypsin, antigliadin, and antiendomysial antibody

Liver biopsy

Figure 1. Algorithm to manage mild ALT and AST abnormalities. (ALT = alanine transaminase; AST = aspartate transaminase.)

Adapted with permission from American Gastroenterological Association. Medical position statement: evaluation of liver chemistry tests. Gastroenterology 2002;123:1365. have not demonstrated an advantage of this mild liver chemistry elevations and is espe- more expensive modality. cially common in patients who are obese, and Steatohepatitis (or nonalcoholic fatty liver those who have diabetes or hyperlipidemia. disease) often is discovered by imaging. This One study12 of patients referred to a hospi- condition may be the most frequent cause of tal-based gastroenterology practice found

March 15, 2005 ◆ Volume 71, Number 6 www.aafp.org/afp American Family Physician 1109 Elevated Transaminase Levels

that in 83 percent of patients with elevated REFERENCES transaminase levels whose serum evaluation was otherwise negative, liver biopsy revealed 1. Green RM, Flamm S. AGA technical review on the evaluation of liver chemistry tests. Gastroenterology steatosis or steatohepatitis. In 10 percent 2002;123:1367-84. of the patients, however, liver biopsy was 2. Kundrotas LW, Clement DJ. Serum alanine aminotrans- normal—a reminder that, at times, mildly ferase (ALT) elevation in asymptomatic US Air Force basic elevated transaminase levels do not repre- trainee blood donors. Dig Dis Sci 1993;38:2145-50. 3. Sherwood P, Lyburn I, Brown S, Ryder S. How are sent any underlying pathology. Excellent abnormal results for dealt with reviews of the management of nonalcoholic in primary care? Audit of yield and impact. BMJ fatty liver disease have been published.13,14 2001;322:276-8. If the diagnosis is not appar- 4. American Gastroenterological Association. Medical position statement: evaluation of liver chemistry tests. ent from the ultrasound Gastroenterology 2002;123:1364-6. Patients with impaired examination, further testing is 5. Davern TJ, Scharschmidt BF. Biochemical liver tests. hepatic synthetic func- suggested for alpha1-antitryp- In: Feldman M, Friedman LS, Sleisenger MH, eds. tion (manifested by an Sleisenger & Fordtran’s Gastrointestinal and liver dis- sin deficiency (alpha1-antitryp- ease: , diagnosis, management. 7th increased prothrombin sin levels), Wilson’s disease ed. Philadelphia: Saunders, 2002:1227-38. time or decreased serum (serum ceruloplasmin), celiac 6. Pratt DS, Kaplan MM. Evaluation of abnormal liver- albumin) should have a disease (antigliadin and anti- enzyme results in asymptomatic patients. N Engl J Med 2000;342:1266-71. more accelerated evalu- endomysial antibody), and 7. Sorbi D, Boynton J, Lindor KD. The ratio of aspartate ation of their abnormal autoimmune hepatitis (antinu- aminotransferase to alanine aminotransferase: poten- . clear antibody, anti–smooth- tial value in differentiating nonalcoholic steatohepa- muscle antibody), as well as for titis from alcoholic liver disease. Am J Gastroenterol 1999;94:1018-22. nonhepatic causes of transami- 8. Bayraktar M, Van Thiel DH. Abnormalities in measures nase elevation. According to the AGA, the of liver function and injury in thyroid disorders. Hepa- decision to perform a liver biopsy needs to togastroenterology 1997;44:1614-8. be made on an individual basis, taking into 9. Bathum L, Petersen HC, Rosholm JU, Hyltoft Petersen P, Vaupel J, Christensen K. Evidence for a substantial consideration the patient’s age, lifestyle, liver genetic influence on biochemical liver function tests: chemistry abnormalities, desire for prognos- results from a population-based Danish twin study. Clin tic information, and associated comorbid Chem 2001;47:81-7. conditions.1 Only with chronic mild trans- 10. Yano E, Tagawa K, Yamaoka K, Mori M. Test validity of periodic liver function tests in a population of Japanese aminase elevations would an asymptomatic male bank employees. J Clin Epidemiol 2001;54:945-51. patient be considered a possible candidate 11. Kaur S, Rybicki L, Bacon BR, Gollan JL, Rustgi VK, Carey for biopsy. WD. Performance characteristics and results of a large- scale screening program for viral hepatitis and risk fac- tors associated with exposure to viral hepatitis B and C: The author indicates that he does not have any conflicts results from the National Hepatitis Screening Survey. of interest. Sources of funding: none reported. Hepatology 1996;24:979-86. This article is one in a series on problem-oriented diag- 12. Daniel S, Ben-Menachem T, Vasudevan G, Ma CK, nosis coordinated by the Department of Family Medicine Blumenkehl M. Prospective evaluation of unexplained at the University of Southern California, Los Angeles. chronic liver transaminase abnormalities in asymptom- atic and symptomatic patients. Am J Gastroenterol 1999;94:3010-4. 13. Angulo P. Nonalcoholic fatty liver disease. N Engl J Med 2002;346:1221-31. 14. Sheth SG, Gordon FD, Chopra S. Nonalcoholic steatohepatitis [published correction appears in Ann Intern Med 1997;127(8 pt 1):658]. Ann Intern Med 1997;126:137-45.

1110 American Family Physician www.aafp.org/afp Volume 71, Number 6 ◆ March 15, 2005