Chapter 48. Drug-Induced Liver Disease

• A remarkable array of therapeutic drugs has been shown to cause hepatic injuries of many types. There are no specific diagnostic tests that establish the presence of a drug-induced liver injury; therefore, the diagnosis is often one of exclusion.
• Evidence of hepatic injury appearing days to months after adding a therapeutic drug with known or suspected potential to injure the liver suggests possible drug-induced injury.
• Improvement following discontinuation of a drug suspected to have caused liver injury (deceleration) is often helpful in diagnosis.
• In patients with few if any signs or symptoms of liver injury, hepatotoxicity may be detected only by biochemical tests showing elevated aminotransferase, bilirubin, or alkaline phosphatase levels.
• A drug-induced etiology should be considered in any patient who develops acute hepatitis with jaundice (hepatocellular injury) or jaundice associated with pruritus and elevated alkaline phosphatase levels (cholestatic or mixed injury).

Almost all drugs in current use have been suggested as a cause of some liver alterations, most often asymptomatic elevations of aminotransferases. Awareness of drug-induced reactions affecting the liver is receiving increased attention. A number of drugs with proven efficacy have been withdrawn from the market because of the recognition of a few cases of severe liver injury. The economic consequences after removing an approved drug from the market or markedly restricting its use are tremendous. Drug-induced injuries that lead to clinically severe liver disease garner much attention even if only a few cases are recognized (or even suggested). A challenge is to improve the methods by which the risk of drug-related hepatotoxicity with a specific drug is assessed (and hopefully predicted) so that effective plans to minimize risk can be developed.

It is well recognized that with many drugs, transient biochemical changes (usually increases in aminotransferases) occur in patients in whom there are no symptoms suggesting liver injury and that with continue use, these usually minor elevations subside as the liver adapts to the newly introduced agent. Determining the threshold of tolerance (risk-benefit factor) for mild to moderate injury caused by drugs that are effective in the treatment of serious illnesses is a difficult but important undertaking.

Hepatic injury that occurs rarely (<1 in 10,000) is designated as an idiosyncratic reaction that is defined as a response or reaction that is host dependent and dose independent. The frequency of hepatotoxicity is influenced by age, patterns of use (short courses or long-term administration), age, sex, ethnicity, and increasingly determination of genetic factors that govern metabolism of the drug and the immune responses to the drug or its metabolic products. Furthermore, transport of specific drugs into the hepatocyte from the blood and from the hepatocyte into bile is dependent on genetically influenced transporters. Evidence of clinically apparent severe events from an approved drug is rare. Significant events may not be detected in the preapproval development during which a limited (several thousand) number of patients are exposed and evidence of liver injury appears after release when large numbers of patients have been treated. The causes of these rare reactions have remained largely unknown. The challenge for further research is to identify why an agent that is well tolerated by thousands of patients may cause a devastating injury in a few. The present focus on genetic variations in drug metabolism, immune responses, and transporters is receiving much attention.

Drug-induced liver disease is usually indistinguishable (clinically and histologically) from liver injury of other causes and may only be detected through awareness of the possibility, suspicion, careful history, exclusion of other likely causes, and inquisitive persistence by the clinician.

Careful attention must be given to the time of onset, duration of therapy, and nature of the hepatic injury, as well as consideration of the roles of other drugs that may enhance production or accumulation of reactive intermediates or interfere with drug transport. Most clinicians prescribe a relatively limited panel of drugs. It is essential that the clinician become familiar with the hepatic risk profile and patterns of manifestation of the agents he or she uses. Difficulties in establishing a diagnosis are compounded by the status of the underlying disease, concomitant illnesses, and polypharmacy.

Considerable supportive evidence implicating a drug as the cause of a liver injury is found if there is resolution of manifestations of liver injury (deceleration) following withdrawal. There are problems in a patient who is receiving many drugs in deciding which agent to remove when evidence of liver injury is found.

The profile of frequency, type of reaction, and possible extent of injury for individual drugs must be considered. Minimal biochemical abnormalities, especially minor elevations of serum aminotransferases (less than three times the upper limit of normal), do not necessarily indicate that clinically important liver injury is occurring. Patients with elevated aminotransferase levels and clinical jaundice are in a group at considerable risk of an adverse outcome.

The conceptual framework of differentiating types of hepatic adverse reactions into those that are predominantly hepatocellular (elevated aminotransferase levels) and those that are predominantly cholestatic injury (elevated alkaline phosphatase levels) is useful. With some drugs, there are elements of both patterns of injury, leading to a category of a mixed reaction.

Cholestatic disorders range from mild elevations of alkaline phosphatase levels found on routine biochemical testing to symptomatic cholestatic syndromes with jaundice and pruritus that resemble primary biliary cirrhosis and primary sclerosing cholangitis. Hepatic steatosis (fatty liver) of both the microvesicular and macrovesicular types is frequently noted in liver biopsy and is a near-constant finding in obese and diabetic patients. Drugs that cause mitochondrial injury may lead to microvesicular fat accumulation. A few drugs (eg, amiodarone) lead to an acquired phospholipidosis. Several drugs are established causes of hepatic granulomatous inflammation indistinguishable from those found in a variety of infections and sarcoidosis.

Tumors induced or promoted by therapeutic drugs range from benign hepatic adenomas, associated with long-term use of oral contraceptives, to angiosarcomas, cholangiocarcinomas, and hepatocellular carcinomas.

Several helpful observations regarding drug-induced liver disease include:

1. Clinical manifestations of drug-induced hepatotoxicity are often clinically indistinguishable from those of liver disease caused by other etiologies.

2. The risk of developing serious liver injury from a drug-induced injury that is predominantly hepatocellular is considerably greater than that in patients who have cholestatic injury.

3. The appearance of clinically apparent liver disease in patients who have predominantly hepatocellular injury with clinical jaundice requires immediate attention.

Hepatic injury from a drug may have a somewhat specific "signature" as regards time of onset, type of injury, and propensity to develop severe disease. However, exceptions occur and many drugs, even those that predominantly cause hepatocellular injury, will on occasion present with a mixed hepatocellular and cholestatic injury or even predominant cholestatic injury.

It may be difficult or impossible to assess the contribution of a drug to hepatic injury in a patient who has an underlying disease or a condition that itself is known to cause liver injury (eg, disseminated carcinoma or fungal disease, chronic hepatitis or cirrhosis, HIV infection and AIDS, and multiorgan involvement from an autoimmune process). Furthermore, in settings in which many drugs are used, especially in the elderly, assigning attribution to any one agent is difficult. Additional consideration must be given to a role for herbal compounds, which may be the initiator of the hepatic findings. Obtaining an accurate history from the patient of the use of herbal products may be especially difficult. The variable timing of stopping, restarting, and substitution of drugs adds to the difficulties. Knowing the relative hepatic risks of individual agents the patient is receiving and the expected time of onset is helpful. Dilemmas are heightened when one or more of the drugs suspected to have caused hepatotoxicity are required in the treatment of a serious underlying illness.

Symptoms and Signs

Early symptoms sometimes associated with drug-induced injuries are nonspecific and include loss of appetite, fatigue, lassitude, and occasionally a dull discomfort more prominent in the right upper quadrant of the abdomen. All these symptoms may be attributed to an underlying condition. There may be few clinical signs or symptoms suggesting liver injury, even in a patient who has biochemical and histologic evidence of considerable damage. These are the same symptoms that are often found in patients who have chronic viral hepatitis B or C or alcohol-induced liver disease. With a few drugs, there is the concomitant presence of fever, rash, or eosinophilia—the hallmarks of hypersensitivity reactions.

Genetic Testing

For several drugs that have been implicated in the pathogenesis of liver injury, there is evidence that genetically controlled metabolic or immune response pathways play important roles in determining which individuals are likely to have an adverse reaction affecting the liver (for example, flucloxacillin, isoniazid, diclofenac, and ximelagatran). The expression of genes controlling the metabolism of a drug (heightened or dampened) influences individual susceptibility to injury, and the patterns of genes affected may allow prediction of a group of individuals at increased risk. Much attention has been paid to variations in the levels of individual cytochronic p450s involved in metabolism.

Extensive evaluations of genetic polymorphisms focusing on enzymes that metabolize or transport metabolic products and those that influence the responses to injury are being undertaken as part of efforts to unravel the mystery of what causes a rare reaction designated as idiosyncratic.

One approach utilizing genetic testing is to identify a group of individuals who have experienced an adverse reaction and compare the gene profile of those patients with those who have received the drug and have not had a problem. Thus far, there is insufficient experience to fully assess the value of this approach, which is presently the focus of considerable research.

The development of reliable tests to detect hepatitis A–E has made the task of excluding viral hepatitis more straightforward and certain. The finding of antimitochondrial antibodies in a patient who has biochemical evidence of cholestasis and jaundice may go far to resolve whether the patient has primary biliary cirrhosis or a drug-induced syndrome that resembles the disorder. However, it must be recalled that patients with liver disorders of all types may develop superimposed drug-induced injury, which, in patients with cirrhosis and decreased hepatic reserve, is more likely to manifest as clinically significant disease.

Drug-Induced Chronic Hepatitis

Several drugs cause chronic hepatitis syndromes that may be largely indistinguishable from autoimmune hepatitis. Drugs that cause a pattern of chronic hepatitis include minocycline, nitrofurantoin, and methyldopa. Misdiagnosing those patients as having autoimmune hepatitis may lead to the institution of corticosteroid therapy while there is continuation of the drug. Corticosteroids serve to blunt the manifestations of the underlying ongoing injury while continued drug use leads to further damage. Generally drugs that cause chronic hepatitis are taken for extended intervals.

The majority of patients who develop drug-induced chronic hepatitis are women, and many have increased serum globulin levels as well as the presence of autoantibodies. The clinical onset of the hepatic injury may be as an acute hepatitis in a patient in whom liver biopsy changes suggest long-standing disease (acute on chronic pattern). The histologic changes may be indistinguishable from autoimmune hepatitis. In some patients chronic hepatitis may present as insidious progressive hepatic failure in a patient who develops hepatosplenomegaly and evidence of portal hypertension or ascites.

The vast majority of drug-induced elevations of serum aminotransferases occur in minimally symptomatic or asymptomatic patients who require no treatment beyond identification and withdrawal of the causation. Removal of the causative agent is generally followed by resolution of the biochemical abnormalities over several days to weeks (deceleration). Some drugs cause a hepatic reaction that closely mimics autoimmune injury. Even in these situations, withdrawal of the likely culprit is generally followed by gradual resolution. With some drugs that cause symptomatic hepatitis, especially if the increase in aminotransferases is associated with the concomitant presence of clinical jaundice, careful evaluation and often hospitalization to allow careful monitoring is required.

If drug-induced acute liver failure occurs, the patient should be viewed as a candidate for liver transplantation and referred to a transplant-capable center. Corticosteroid therapy has not definitely been established to be useful in the treatment of drug-induced liver injury. These agents have often been used in those patients with drug-induced liver injuries who also manifest hypersensitivity reactions, for example in patients with drug-induced injury caused by halothane or diphenylhydantoin.

In many patients who have drug-induced hepatocellular injury, a gradual decrease in the aminotransferase levels occurs within 1–2 weeks of removing the drug. The decrease may not occur immediately, especially with drugs whose parent compound or metabolic products accumulate in the hepatocytes. For the first several days after drug withdrawal, processes already in place may proceed, causing early concern as to whether the drug was the cause. Continued evidence of liver injury may persist for weeks to months and cause a pattern of chronic injury. The longer abnormalities persist, the greater the possibility that permanent hepatic injury is present or will develop or that the diagnosis of a drug-induced injury was not correct.

Rechallenge with a suspected drug to establish a diagnosis of drug-induced injury is seldom necessary and if the initial event has been severe (such as a presentation with clinical jaundice and marked elevations of aminotransferases), may be dangerous.

Acetaminophen

Acetaminophen hepatotoxicity is the most frequent cause of acute liver failure in the United States and is likely the cause of (or a major contributor to) elevations of aminotransferases often found on routine evaluations. Ingestion of excessive amounts of acetaminophen (>10–15 g), used in suicidal attempts, predictably leads to liver injury ranging from elevated aminotransferases to acute liver failure and in some patients death or the need for liver transplantation. In therapeutic doses (≤4 g/day), acetaminophen is usually safe and well tolerated. The safety margin of patients who are regular users of alcohol appears to be diminished, and these individuals are at increased risk of developing acetaminophen-induced liver injury.

Factors that predispose patients to hepatic injury from acetaminophen in nonsuicidal situations include the dose ingested and are more likely to occur in a patient who has been a regular user of alcohol. First and most important is the dose of acetaminophen. Patients may have underestimated or understated the amount ingested, especially since acetaminophen is present in many widely used combination products. Ingestion of 4 g of acetaminophen given as a single agent or in combination with opioids in healthy, asymptomatic individuals was shown to elevate aminotransferase levels to greater than three times the upper limit of normal in 31–44%. Second is the role of concomitant use of alcohol. Chronic use of alcohol may induce changes in levels of CYP2E1 (the major cytochrome P450 subspecies involved in the metabolism of ethanol and in the metabolism of acetaminophen). Therefore, in this setting, the alcohol-induced increased CYP2E1 potentially leads to more acetaminophen being metabolized to yield the reactive metabolite N-acetyl-p-benzoquinone imine (NAPQI). In addition, the intracellular concentration of glutathione (which has the major role in binding to NAPQI, leading to its excretion as mercapturic acid) is possibly lowered in patients who regularly use alcohol, further promoting the potential for injury.

N-acetylcysteine has been shown to be effective in reducing the extent of hepatic injury from acetaminophen.

Isoniazid

Isoniazid (INH) has been recognized as an established cause of liver injury for the past five decades. Elevations of aminotransferase levels are found in 10–20% of patients who receive the drug, with the onset of these elevations within several weeks of beginning treatment. In most patients, the elevated aminotransferase levels are not associated with symptoms suggestive of liver injury, and often the levels return to normal during continued treatment (adaptation). In patients with elevated ALT levels, discontinuation of INH is usually followed by resolution over several weeks. A few patients develop clinically evident liver disease. Many years of observation of patients receiving INH allow the following conclusions to be made:

1. Patients older than 50 years at the time of therapy are at increased risk of developing clinically important INH-induced liver injury (∼1%).

2. Children younger than age 16 years seldom show evidence of hepatic injury from INH.

3. Prodromal signs and symptoms are vague and include loss of appetite. If clinically apparent jaundice develops, there is an approximately 10% mortality.

4. Continued use of INH after the appearance of non-specific symptoms and signs of liver injury and, especially after the presence of jaundice, is associated with an increased likelihood of development of severe injury.

5. Patients who have hepatitis B, hepatitis C, and HIV are at increased risk of developing liver injury from INH.

6. Combining INH with several other drugs also used to treat tuberculosis (eg, rifampin and pyrazinamide) increases the likelihood of injury.

The principal mechanism underlying INH-induced hepatic injury stems from the effects of an intermediary metabolite. The specific toxic intermediate has not been definitely established but is likely a hydrazine derivative. Genetic polymorphisms of the genes for N-acetyltransferase and CYP2E1, both of which have important roles in the metabolism of INH, have been shown to influence susceptibility. Slow acetylators of INH by the enzyme NAT2 are at higher risk of developing hepatotoxicity than are rapid acetylators and are more likely to develop severe injury. Individuals who are CYP2E1 c1/c1 (wild type) have higher levels of CYP2E1 activity than do those who are CYP2E1 c1/c2 or c2/c2 and are at increased risk.

Awareness of the risk of INH-induced liver injury and regular monitoring of aminotransferase levels in patients receiving the drug have proven effective in identifying evidence of hepatic injury early, at a time when withdrawal of the drug leads to resolution.

Minocycline

Minocycline, a tetracycline derivative used in many situations including the long-term treatment of acne, has been established to cause several types of liver damage including acute hepatitis, occasionally with features of a hypersensitivity reaction including Stevens-Johnson syndrome with fever and lymphadenopathy, chronic hepatitis closely simulating autoimmune hepatitis, and, rarely, acute liver failure. In some patients, cholestatic features may be prominent or even predominate.

Of particular interest is a minocycline-induced syndrome that simulates autoimmune hepatitis. The majority of the affected patients are women who often have taken the drug for months. Hyperglobulinemia and the presence of antinuclear antibodies and anti-DNA antibodies are found in some. The role of minocycline in the causation or unmasking of an autoimmune hepatitis is important in that misdiagnosis and treatment of presumed autoimmune hepatitis type 1 with corticosteroids may mask the minocycline-induced liver injury, which then progresses relatively unnoticed.

Statins & Ezetimibe

The era of the widespread, often long-term use of statin drugs to lower low-density lipoprotein and cholesterol levels through inhibition of HMG-CoA began in the 1980s. Increases of aminotransferase level to more than three times the upper limit of normal occurs in 1–3% of patients receiving statins. The extent of the elevation appears directly related to the dose.

Statin-induced aminotransferase elevations are noted within weeks to months of initiating therapy, and the increases appear to be linked in some as yet undetermined way to the changes in lipids. There have been remarkably few instances of clinically significant liver injury reliably attributable to statins. The risk of statin-induced acute liver failure is less than 1 in 1 million, and cases of statin-induced significant liver injury are infrequent in large series of patients with acute liver failure.

In asymptomatic patients receiving statins who have elevations of aminotransferases, withdrawing the statin is followed by a return of ALT to normal. In some patients, adaptation occurs during continued treatment or upon treatment with a reduced dose. Increases of bilirubin associated with the elevations of aminotransferases are extremely uncommon.

There is no convincing evidence that patients with abnormal baseline aminotransferase levels, underlying steatohepatitis, or chronic hepatitis C are at increased risk of hepatotoxicity from statins. A few reports suggesting autoimmune hepatitis may be triggered by statins have been reported.

Ezetimibe, an inhibitor of intestinal uptake of intestinal and biliary cholesterol, is another agent widely used to lower cholesterol and is often given along with a statin. There have been isolated reports of ezetimibe in individual hepatic injury. Overall, the agent has an excellent safety record.

Amoxicillin–Clavulanate

Hepatotoxicity from the extensively used combination of amoxicillin and clavulanic acid is widely recognized and is attributed to the clavulanic acid component. Amoxicillin–clavulanate is among the most frequently identified drugs causing liver injury.

The frequency of hepatic injury from amoxicillin–clavulanate is low (∼1–3 in 100,000), and the range of clinical and laboratory manifestations is broad. The drug is generally used as short duration therapy (7–10 days). The appearance of evidence of liver injury is often delayed for days to weeks after initiation of therapy; therefore, many cases of hepatic injury appear well after the course of the therapy has been completed, and the pivotal role of amoxicillin–clavulanate may be overlooked. The types of liver injury from amoxicillin–clavulanate include hepatocellular necrosis, cholestatic reactions, and, in some patients, a mixed hepatocellular-cholestatic presentation. Some patients (one to two-thirds) have signs of hypersensitivity, with rash and fever at the time hepatic injury is recognized, whereas others present with jaundice and evidence of a bland cholestasis. In general, the liver injury from amoxicillin–clavulanate is mild and self-limiting, with a gradual and complete resolution of the process over several days to weeks.

Advancing age and prolonged therapy are factors associated with the development of the cholestatic injury. Predominantly hepatocellular injury is more frequent in younger patients. A few instances of severe hepatocellular injury leading to death or the need for transplantation have been reported.

Herbal Hazards

Hepatic injury from a variety of herbal products from around the world has been associated with hepatotoxicity. In most instances, these products have not been well characterized chemically and have not been subjected to clinical trials. For example, Pyrrolizidine alkaloids used in herbal teas in Africa and the Caribbean have been associated with the development of sinusoidal obstruction syndrome. Extracts of kava, a plant found in the South Pacific, is widely used to alleviate stress, anxiety, and sleeplessness. There are multiple reports of hepatotoxicity including instances of acute liver failure, which have been attributed to the use of kava. There is a dose relationship between kavalactones and the risk of hepatic injury.

Patients ingesting greater than 250 mg a day of kavalactones are at increased risk. The variable processes of extraction and variation in kava products exemplify the difficulties of assessing risk with herbal problems. Several herbal products for weight loss including Hydroxycut have been banned following recognition of hepatic injury.

The important lesson for the clinician is to inquire (often repeatedly) about the use of herbals and vitamins. For example, excessive use of vitamin A can lead to sinusoidal compression, fibrosis, cirrhosis, and ascites.