Identifying the cause of a persistent, asymptomatic aminotransferase elevation can be challenging. The possible diagnoses are many and varied. To narrow the differential, begin with a detailed history.
Liver function test (LFT) abnormalities are common incidental findings in patients who show no other signs of hepatobiliary disease. They may also be found in patients who present with vague complaints, such as fatigue. The differential diagnosis of LFT elevations is extensive and ranges from fairly benign conditions to life-threatening disease in which early diagnosis is critical.
A focused and detailed medical history can usually help narrow the differential considerably. In this article, we discuss what to cover when taking the history of a patient with a persistent aminotransferase elevation. We also indicate when additional testing is needed-and which tests to order-to nail down a diagnosis. In a second article in a coming issue, we will describe appropriate workups for patients with a persistent, asymptomatic alkaline phosphatase elevation and for pregnant patients with abnormal LFT results.
WHEN TO EVALUATE ALT/AST ELEVATIONS
When you first notice an alanine aminotransferase (ALT) or aspartate aminotransferase (AST) level greater than the upper limit of normal in an asymptomatic patient or one with only vague symptoms, always repeat the test before proceeding with evaluation of any kind. The most common nonspecific symptom in a patient with chronic liver disease is fatigue.
If a second test establishes that the elevation is persistent, and if the value is greater than 3 times normal, evaluation is warranted. If the value is less than this, monitor the patient; if it remains elevated, no matter how slightly, initiate evaluation.
COMMON CAUSES OF AMINOTRANSFERASE ELEVATIONS
Because aminotransferase elevations have many and diverse possible causes, a useful history taking (Table 1) must cover considerable ground. A good history will narrow the differential diagnosis significantly and point you toward the most appropriate follow-up tests (Table 2).
Occupational exposures. The most common occupational exposure associated with aminotransferase elevations is a needle-stick injury. Consider hepatitis B or C in any health care worker with an unexplained aminotransferase elevation.
Organic solvents are another-although less common-occupational exposure associated with elevated aminotransferase levels. Find out whether the patient works with organic solvents (for example, someone who works in an airport using degreasing agents to clean jet engines). When an exposure to organic solvents is identified, a special toxicology screen to measure parameters of such solvents is required.
Drugs. Many drugs affect the liver, and drug-induced injury can mimic almost any kind of liver disease. Some of the more commonly used agents that can produce chronic liver disease include methotrexate, nitrofurantoin, amiodarone, isoniazid, and statins. However, many others are also potential culprits.
Typically, a drug-induced aminotransferase elevation will appear a few weeks or months after the medication is started. (An important exception is methotrexate, which usually causes liver injury only after years of regular use for conditions such as psoriasis and rheumatoid arthritis.) Thus, be sure to ask when therapy was started with each medication the patient takes. Keep in mind that drug-induced aminotransferase elevations are generally mild-around 3 times the upper limit of normal.
When you suspect that a recently started drug or increase in dosage is the culprit, you need not necessarily stop the drug right away. If the elevation is less than 3 times the upper limit of normal, you may opt to monitor the patient's liver enzyme levels.
Stabilization at the initial (slightly elevated) level or lower-or a gradual decline-in a patient with a drug-related liver enzyme elevation is fairly common and is typically a sign that the patient's body has adapted to the drug. Scenarios in which aminotransferase levels continue to rise suggest a more serious problem, and stopping the culprit drug is wisest in such cases.
Statins. These are a good example of a class of drugs that fairly frequently produce mild aminotransferase elevations yet are rarely a cause of significant liver injury. In most cases, patients' liver enzyme levels normalize over time or stabilize at a slightly elevated level and do not cause trouble. In fact, randomized controlled trials of statin therapy in patients with preexisting liver disease have shown that those participants who received statin therapy wound up with lower aminotransferase levels than those who received placebo.1
Acetaminophen. Although millions of persons take acetaminophen safely and without incident, this medication is a common cause of aminotransferase elevations. Because many people are not aware that common over-the-counter (OTC) cough and cold preparations frequently contain acetaminophen, it is easy for those who use multiple OTC medications to unknowingly take more than the recommended maximum daily dose. Moreover, a study has shown that even the maximum daily dose (4 g) can produce mild aminotransferase elevations in otherwise healthy persons when taken for several days in succession.2,3 If you suspect that chronic use of acetaminophen at or above the maximum recommended dosage is the cause of a patient's aminotransferase elevation, order a measurement of his or her acetaminophen level.
Herbal remedies. Ask the patient about herbal products as well as OTC agents and prescription drugs. It is often helpful to have patients bring in any herbal or vitamin supplements that they take; many of these contain ingredients that patients are unaware of. For example, ma huang derivatives can cause severe liver disease in some people; however, patients who take herbal preparations for weight loss seldom know that these usually contain ma huang or one of its derivatives. Other herbs that can produce liver toxicity include (but are not limited to!) kava kava, jim-buan, Chinese herbal mixtures, mistletoe, and valerian.
Alcohol. Persistent, alcohol-related aminotransferase elevations are associated with long-term and substantial alcohol use: usually the regular consumption of 4 or more drinks a day for women or 6 or more drinks a day for men. Occasional bouts of binge drinking generally do not affect liver enzyme levels; in fact, not all persons who drink heavily have abnormal LFTs. Women who drink are more likely to have aminotransferase elevations than men are.
Because people often are not candid about actual consumption, you may need to rely on clues other than those from the history to detect a possible alcohol-related aminotransferase elevation. One such clue is the ratio of AST to ALT. Usually, in alcohol-related liver disease, the AST level is more elevated than the ALT-typically more than twice as high. In general, the elevations are not dramatic; even in patients with advanced alcohol-related liver disease, it is unusual for the AST level to be greater than 300 U/L or the ALT greater than 150 U/L. If you suspect heavy drinking despite patient denial, it is occasionally helpful to order a urine alcohol screen.
Excess weight or weight gain. As the percentage of Americans who are overweight or obese increases, nonalcoholic fatty liver disease, or NASH (nonalcoholic steatohepatitis), is becoming an increasingly common cause of liver disease in this country. In fact, it may now be the most common cause.
In the absence of signs of another cause, suspect NASH in any patient with a persistent aminotransferase elevation whose body mass index (BMI) is 25 or greater, particularly those with a BMI of 30 or more and those who carry their excess weight disproportionately in their abdomen. However, NASH may also occur in patients whose BMI is less than 25; this is particularly likely in persons who have a small potbelly and who have recently gained weight. Thus, be sure to ask patients about any changes in weight over the past few years.
If you suspect NASH, measure the patient's lipid levels (Algorithm). In most affected patients, the triglyceride level is elevated; any value that is above normal increases the suspicion of the diagnosis. Although NASH is seen in patients of both sexes and any age, in younger patients-and NASH can occur even in children-it is likely to be associated with a family history of diabetes or glucose intolerance.
If the triglyceride level is elevated, the next step is to order ultrasonography or CT scanning of the liver. If an ultrasonogram shows coarseness consistent with fatty infiltration or a CT scan shows radiolucency consistent with fatty liver disease, the diagnosis is highly likely. Neither study is specific enough to confirm a diagnosis of NASH; however, biopsy is not always necessary. A positive result on a measurement of insulin resistance (eg, HOMA [homeostasis model-assessment estimate]) coupled with a positive imaging study makes NASH highly likely, and some physicians would forgo biopsy in this setting. At present, there is no consensus about when to order biopsy in patients with suspected NASH.
Viral hepatitis. Always ask the standard questions about sexual activity, injection drug use, and history of transfusions before 1992. In addition, ask women whether they had any type of gynecological surgery before that year; blood was sometimes given during cesarean sections, hysterectomies, and dilations and curettage without the patients' being aware of it.
Keep in mind that patients in their 50s or 60s who experimented briefly with drugs 30 or more years ago often will respond negatively to questions about injection drug use. Yet needle sharing was common in that era, particularly during the Vietnam War, and a new aminotransferase elevation could be the first evidence of a hepatitis C virus infection that a patient acquired from a needle he used 30 years earlier. If you suspect hepatitis C, order a test for antibody to hepatitis C virus and consider a polymerase chain reaction test for HCV RNA, which is specific for the presence of the virus.
To determine the risk of hepatitis B, ask about patients' immigration history. Those who emigrated from parts of Asia where the disease is endemic (eg, China, Korea, Taiwan, Singapore, Thailand, Vietnam, and Malaysia) have a greatly increased risk of hepatitis B acquired through perinatal transmission-as do patients from sub-Saharan Africa. Patients who emigrated from Eastern European countries also have an increased risk, although in their case the infection is more likely to have been acquired through tainted blood (testing of blood donors was initiated only recently in some of these countries). If clues in the history point to hepatitis B, order tests for hepatitis B surface antigen, hepatitis B surface antibody, and antibody to hepatitis B core antigen. The presence of hepatitis B surface antigen in an asymptomatic person is indicative of the presence of chronic hepatitis B.
Other viral infections. Ask about recent viral or flu-like illnesses. Enteroviruses can cause elevations of the aminotransferases that persist for a week or so. Mononucleosis can cause acute liver injury and produce aminotransferase elevations.
Autoimmune disease. Inquire about a family history of autoimmune disease of any kind. A family history of autoimmunity is disproportionately present in patients with autoimmune hepatitis, and such a history can be a clue to that diagnosis. Patients with autoimmune hepatitis may also have a history of a recent injury to the liver (such as a newly prescribed hepatotoxic drug). Whereas liver enzyme levels return to normal in most persons when the source of the injury is withdrawn (eg, the new drug is stopped), in persons with a genetic predisposition to autoimmunity, the injury inaugurates a cycle of autoimmune destruction that continues even after the initial injury has ceased.
When the family history suggests autoimmune hepatitis, order tests for anti-smooth muscle antibodies and antinuclear antibodies. Classic autoimmune hepatitis is associated with positive results on both tests. A definitive diagnosis, however, must be based on biopsy results. Keep in mind that presentations of autoimmune hepatitis tend to peak at age 15 to 25 years and then again after age 45 or 50 years. Refer patients with established chronic liver disease to a hepatologist.
RARER CAUSES OF AMINOTRANSFERASE ELEVATIONS
When the workup does not support any of the more common causes of a persistent AST/ALT elevation, or when there are suggestive clues in the history, you need to consider other, more rare possibilities. These include exercise-induced rhabdomyolysis, celiac disease, Wilson disease, α1-antitrypsin deficiency, hemochromatosis, and macroenzymes. Refer to a specialist any patient in whom one of these disorders (other than exercise-induced rhabdomyolysis) is diagnosed or strongly suspected.
Exercise-induced rhabdomyolysis. This is still a fairly rare cause of aminotransferase elevations, although it is becoming more common-especially in highly active persons, such as military personnel-and the diagnosis is easily overlooked.
Ask specific questions about patients' exercise habits. Typically, patients in whom exercise is causing aminotransferase elevations work out daily, engage in vigorous aerobic activity, and exercise for more than an hour at a time. They may not have typical symptoms of rhabdomyolysis-eg, they may not complain of muscle pain.
Be sure to ask patients whether they exercised on the day of their blood test. Generally, exercise-associated elevations do not show up if blood is drawn 2 or 3 days after the patient last exercised. Conversely, such elevations are much more likely when blood is drawn shortly after a patient exercised.
Often, the sole clue to this diagnosis is the patient's history of frequent vigorous exercise (together with the lack of another obvious cause of an aminotransferase elevation). A simple initial test for exercise-associated rhabdomyolysis is to ask the patient not to exercise for 2 weeks and then repeat the liver enzyme measurement. If the levels are significantly lower after 2 weeks without exercising, the diagnosis is highly likely. Keep in mind that affected patients may be compulsive about exercise; they are likely to resist if you ask them to go a week or two without exercising.
You may also want to check the patient's creatine kinase (CK) level. A finding of CK elevation also makes exercise-induced rhabdomyolysis quite likely; however, normal CK levels do not rule out the diagnosis.
Celiac disease. Although the pathogenesis is unknown, celiac disease can be associated with chronic hepatitis. The liver disease resolves when patients follow a gluten-free diet. Because celiac disease can be asymptomatic or cause only mild symptoms, it is a possibility to consider in a patient in whom tests for all other potential causes of an aminotransferase elevation have had negative results. Before ordering the standard tests for endomysial antibodies, you may want to simply prescribe a gluten-free diet and see if the enzyme levels normalize.
Wilson disease. This rare inherited disease is characterized by a defect in the transport of copper through the liver. It often first presents with mild liver abnormalities. Because Wilson disease is progressive and invariably fatal unless timely treatment is initiated, making the diagnosis as early as possible is crucial. Suspect Wilson disease particularly in younger patients, although the illness may not be evident until later in life. When the diagnosis is suspected, order a measurement of serum ceruloplasmin. A low level of ceruloplasmin heightens suspicion; however, it is not confirmatory.
If a low ceruloplasmin level is accompanied by a high urinary copper level and neuropsychiatric symptoms (eg, parkinsonism) or the presence of orangish brown or orangish green rings in the iris (Kayser-Fleischer rings), the diagnosis of Wilson disease is confirmed. However, Kayser-Fleischer rings are difficult to detect; moreover, neither neuropsychiatric symptoms nor Kayser-Fleischer rings are present until later in the disease course. Thus, any patient in whom low ceruloplasmin levels are detected should be referred to a specialist.
α1-Antitrypsin deficiency. This is another inherited disease-one that, unlike Wilson disease, typically first presents when patients are older. (It can also cause neonatal hepatitis.) A defect in the way the protein α1-antitrypsin unfolds prevents its being transported out of the liver. Thus, the defective protein accumulates in the liver and a deficiency state develops elsewhere in the body. In affected patients, the accumulated protein in the liver produces hepatitis and eventually cirrhosis, and the system-wide protein deficiency may cause emphysema. Clues from the history include a family history of liver disease, sometimes accompanied by a family history of emphysema.
When α1-antitrypsin deficiency is suspected, order a measurement of the patient's α1-antitrypsin level and a determination of his α1-antitrypsin phenotype. Any α1-antitrypsin level below the normal range-or a homozygous phenotype-suggests the diagnosis.
Hemochromatosis. This inherited disorder is suggested by a family history or by a high ferritin level. If a patient has a family history of hemochromatosis and there is no other obvious explanation for his aminotransferase elevation, order iron studies. Results typical in hemochromatosis include high ferritin and serum iron levels and increased transferrin saturation. If the results of iron studies are consistent with hemochromatosis, the next step in the workup is to order genetic testing for the C282Y and H63D mutations of the HFE gene. If the results of the genetic test show that the patient is homozygous for the C282Y mutation, most experts now consider this to confirm the diagnosis. Formerly, biopsy was regarded as necessary; although this is no longer the case, I still recommend biopsy in the index case in a family. Referral to a specialist is recommended when results of iron studies suggest hemochromatosis.
Macroenzymes. This is a rare biochemical abnormality of one of the liver enzymes (usually AST) in which the enzyme in question combines with a protein and as a consequence "reads" at a very high level. If an otherwise healthy patient has an isolated, very significant elevation (usually more than 5 times the upper limit of normal) and all other liver enzyme levels are normal, suspect macroenzymes. To confirm the diagnosis, a blood sample must be sent to a laboratory specially equipped to test for the condition (eg, the Mayo Clinic laboratory). Fortunately, macroenzymes are relatively benign.
Lewis JH, Mortensen ME, Zweig S, et al; Pravastatin in Chronic Liver Disease Study Investigators. Efficacy and safety of high-dose pravastatin in hypercholesterolemic patients with well-compensated chronic liver disease: results of a prospective, randomized, double-blind, placebo-controlled, multicenter trial.
2007 Aug 1; [Epub ahead of print].
Watkins PB, Kaplowitz N, Slattery JT, et al. Aminotransferase elevations in healthy adults receiving 4 grams of acetaminophen daily: a randomized controlled trial.
Amar PJ, Schiff ER. Acetaminophen safety and hepatotoxicity-where do we go from here?
Exp Opin Drug Saf.