GI Consult: Drug Hepatotoxicity

Twelve Commonly Asked Questions About Drug Hepatotoxicity

For both clinical and medicolegal reasons, a familiarity with drug-induced liver disease is vital for the practitioner.

By Charles Maltz, MD

This series of discussions in question-and-answer form is prepared for EMERGENCY MEDICINE by Dr. Maltz, who is a clinical assistant professor of medicine at Cornell Medical School in New York City and is attending physician in the department of emergency medicine and division of gastroenterology and hepatic diseases in the department of medicine at New York Presbyterian Hospital.

1. How important is the problem of drug-induced liver injury?

About 10% of all cases of hepatitis in young adults and 40% of cases in patients older than 50 are caused by medications. The numbers are even higher in sicker patients who take more medications. In addition, one of every five cases of fulminant hepatic failure is caused by drugs. Because the cause of liver injury from prescribed medication is iatrogenic, the potential medicolegal consequences are obvious.

2. Is there a specific toxic effect that all drugs have on the liver?

There is no universal histologic pattern of drug-induced liver injury. However, certain medications are associated with specific patterns of injury. For example, isoniazid and acetaminophen are associated with hepatocellular necrosis; chlorpromazine, with cholestasis; glucocorticoids and valproate, with steatosis; sulfonylurea, with granuloma formation; and immunosuppressive medication for bone marrow transplantation, with venoocclusive disease. Combinations of these presentations may occur (see table below).

Common Types of Liver Injury and the Agents Associated with them Injury Drug Hepatocellular necrosis Isoniazid, phenytoin, acetaminophen, iron, troglitazone Steatosis Steroids, parenteral nutrition, valproic acid, IV tetracycline Cholestasis Chlorpromazine, rifampicin, erythromycin estolate Note: More than one type of injury may occur in the same patient and any one agent may produce different types of injury in different patients

3. Can't any drug cause liver-related adverse effects?

Not really. Liver disease caused by digoxin is almost unheard of; on the other hand, every physician is familiar with the risk of liver injury from isoniazid or methotrexate. Also, certain medications can cause liver disease in the form of a generalized toxic syndrome. Phenytoin, for example, may produce fever, rash, and adenopathy as part of a pseudomononucleosis syndrome.

4. Are certain persons or patient populations more susceptible to drug-induced liver damage?

Both old age and female sex are associated with an increased risk of drug-induced hepatotoxicity. In alcoholic persons, increased production of a toxic intermediate by a stimulated P450 enzyme system makes them more susceptible to acetaminophen-induced hepatotoxicity. Depleted stores of glutathione in the liver in malnourished alcoholic persons may also contribute to this effect.

5. Is there a particular pattern in enzyme levels that indicates drug-induced hepatotoxicity?

Liver chemistry findings reflect the histopathologic characteristics of drug-induced liver damage. Cholestasis, for example, is typically indicated by elevated bilirubin levels as well as jaundice, whereas hepatocellular damage may produce only increased alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels. A combination of cholestasis, hepatocellular damage, and steatosis (fat deposition in hepatocytes) may be found in the same patient.

6. Is drug-induced hepatotoxicity simply an immune-mediated phenomenon?

In general, no. Some hepatotoxins, such as phenytoin and quinidine, produce syndromes that appear to be immune mediated in nature, but most do not.

It is more useful to classify hepatotoxins as either intrinsic or idiosyncratic agents. Acetaminophen and inorganic iron (ferrous sulfate and ferrous gluconate) are examples of intrinsic hepatotoxins. In any person, a significant overdose of these drugs will cause hepatic necrosis; however, most hepatotoxins cause liver disease only in certain persons. The reasons for this are not completely understood, although it is believed that the underlying metabolic state of the liver plays a pivotal role. This metabolic state is a reflection of a person's age, sex, genetic makeup, nutritional status, and use of other medications.

7. If a drug is going to cause a hepatotoxic reaction, will it do so within a certain period after the patient starts taking the drug?

It depends on the medication. An overdose of acetaminophen will produce symptoms within a few hours. Typically, intrinsic hepatotoxins—as well as Amanita phalloides, a poisonous mushroom with hepatotoxic effects—will produce nausea and vomiting, followed by a period of well-being and then signs and symptoms of liver dysfunction. Because most drug hepatotoxins are idiosyncratic in nature, however, the period between the first dose of the medication and the appearance of liver injury will vary.

8. How do I diagnose drug-induced liver injury?

Because there are no characteristic liver function abnormalities, the diagnosis requires a thorough medication history, appropriate blood studies to rule out other causes (such as viral hepatitis), and knowledge of which drugs are most likely to cause liver injury.

Often, because the diagnosis is presumptive, the suspected medication may have to be discontinued temporarily. One case may be fairly straightforward, yet another may be extremely difficult to evaluate. In a patient who starts taking phenytoin, for example, the appearance of liver function abnormalities along with a rash, fever, and eosinophilia presents no diagnostic challenge. In contrast, a patient with liver dysfunction who has undergone a bone marrow transplant is usually taking several potentially hepatotoxic medications; in such a patient, the liver disease may also have an infectious cause.

9. A best-selling drug for type 2 diabetes and a promising antibiotic were recently either recalled or given severely restricted indications by the Food and Drug Administration because they were hepatotoxic. Can you elaborate on the reasons?

Troglitazone, a type 2 diabetes agent that appeared to have a more desirable metabolic effect than the sulfonylureas, was approved for use in the United States, Europe, and Japan over the last 5 years. Results of premarketing trials showed that in as many as 2% of patients receiving the medication, ALT and AST levels were elevated to more than three times normal levels. Postmarketing surveillance revealed hepatic failure in several patients who required liver transplantation; it also found one death. The drug was recalled within the last year.

A similar fate befell trovafloxacin, a new quinolone with gram-positive, gram-negative, and anaerobic activity that was introduced several years ago. Because of its broad spectrum of action, the drug was prescribed for many patients. However, with widespread use came more than 14 cases of hepatic failure. Some patients required liver transplantation; others died. Trovafloxacin therapy is now limited to hospitalized patients who have no better antibiotic alternatives.

The mechanism of troglitazone—and trovafloxacin—associated hepatotoxicity is idiosyncratic; the period from initiation of therapy to toxic reaction is variable.

10. What about the use of potentially hepatotoxic drugs in patients with preexisting liver disease? Are these patients more likely to have a hepatotoxic reaction?

There is no evidence that patients with preexisting liver disease are more predisposed to hepatotoxic reactions. However, alcohol consumption does increase the toxicity of acetaminophen and isoniazid. In addition, a person who has decreased liver reserve resulting from liver damage caused by cirrhosis may not be able to tolerate additional damage from a toxic drug reaction.

11. How can I avoid causing a hepatotoxic reaction in my patient?

Short of prescribing no medications at all and advising your patient to take no over-the-counter or herbal preparations, there is no sure way to avoid such a reaction. The best defense is being aware of the relative risk of a hepatotoxic reaction from the various medications you prescribe. Patients who are alcoholic should be cautioned about excessive acetaminophen intake. Isoniazid prophylaxis should be administered according to appropriate guidelines. Always obtain a thorough medication history, including use of over-the-counter drugs and herbal preparations (see table below) in addition to prescribed drugs, when evaluating patients who have abnormal liver function test results. Also keep in mind that toxic effects are becoming increasingly common as new drugs are more widely prescribed.

Herbal Preparations Associated with Hepatotoxic Reactions Chaparral (Larrea tridentata) Germander (Teucrium chamaedrys) Chinese herbs (jin bu huan, ma huang) Valerian root (Valeriana officinalis) Skullcap (Scutellaria galericulata) Mistletoe (Viscum species) Herbal teas (pyrrolizidine alkaloids) Senna (Cassia angustifolia) Comfrey (Symphytum officinale)

12. How should hepatotoxic drug reactions be treated?

The possibility of biliary tract disease and viral or alcoholic causes of liver function abnormalities first must be ruled out. The suspected medication should be discontinued, if possible, and supportive care provided as needed. Although it is tempting to administer steroid therapy to a patient whose condition is deteriorating, there is no evidence to support the contention that such an approach yields beneficial results. Perhaps steroids should be given only to patients who appear to be having a generalized allergic reaction along with liver dysfunction. Finally, a liver transplant team should be notified if the liver failure appears to be worsening and is accompanied by coagulopathy or encephalopathy.

Printable version of this article

E-Mail this Article