• Liver disease
• Gastrointestinal disease
• Effects on specific AEDs
• Carbamazepine
• Ethosuximide
• Felbamate
• Gabapentin
• Lamotrigine
• Levetiracetam
• Oxcarbazepine
• Phenobarbital
• Phenytoin
• Tiagabine
• Topiramate
• Valproic acid
• Zonisamide
• Choosing and monitoring AEDs
• References

The capacity for the liver to metabolize drugs depends on hepatic blood flow and liver enzyme activity.1 Hepatic metabolic enzymes are diverse. They are classified as:

• Phase I, involved in nonsynthetic or functionalization metabolism, which includes oxidation, reduction, and hydrolysis reactions
• Phase II, involved in synthetic or conjugation metabolism, which includes glucuronidation, methylation, and acetylation reactions

Multiple forms of cytochrome P-450 are involved in Phase I metabolism. There may be genetic polymorphic distribution of these enzymes. Therefore, some individuals may be inherently fast or slow metabolizers. It has become critical to know which of the isoenzymes of cytochrome is involved in a specific drug’s metabolism and whether there is genetic polymorphism for metabolism.

Drug-metabolizing enzymes may also be induced or inhibited. When enzymes are induced, drugs are metabolized more quickly. This increases the clearance, shortens the half-life, and decreases the bioavailability. Conversely, when drug-metabolizing enzymes are inhibited, drugs are metabolized more slowly, which decreases the clearance, increases the half-life, and increases the bioavailability.

The liver produces albumin and alpha glycoprotein, to which some drugs are extensively bound. An alteration in the ability of the liver to synthesize proteins decreases the amount of bound drug and increases the drug’s free fraction, which binds at receptor sites, exerting its pharmacologic activity. Liver disease also may alter the binding characteristics of plasma proteins in a manner similar to what happens with renal disease, so that even if plasma proteins are normal, plasma protein binding may be decreased. By altering the binding characteristics, the percent bound will be changed. An increase in free fraction also makes more drug available for metabolism, which increases clearance if the hepatocytes are functioning. If the hepatocytes are not functioning, there is an increase in the free drug concentration.1,2

Several patient factors alter the liver’s ability to metabolize drugs:

• food consumption (For example, the presystemic clearance of high-extraction drugs such as propranolol is significantly reduced and bioavailability significantly increased when taken within 3 hours after eating.)
• age
• sex
• race
• pregnancy
• hormones
• circadian variability

Another factor is concurrently administered medications. Drugs have the ability to induce and inhibit the enzymes responsible for drug metabolism. Drug interactions may occur when new drugs are added or if a concurrent drug is discontinued. Concurrent drug administration also may change protein binding.1,2

Liver disease causes multiple pathophysiologic changes that influence drug disposition. Decreased hepatic blood flow, extrahepatic and intrahepatic blood shunting, and loss of hepatocytes alter the ability to metabolize drugs. The bioavailability of administered medications increases, effectively increasing the dose. Decreased protein synthesis decreases the percentage of drug bound to plasma proteins and increases the amount of “free” or unbound drug. The increase in free fraction makes more drug available to the receptor site and more drug available for metabolism, thereby increasing its clearance. Increased clearance does not occur if hepatocytes are not capable of metabolizing the drug, however. An increase in free fraction and a decrease in hepatocyte function result in an increase in free drug concentration.

Liver diseases that alter drug disposition include:

• chronic liver disease (e.g., cirrhosis)
• acute hepatitis
• drug-induced hepatotoxicity
• cholestasis
• infiltrative or neoplastic disease1-3

In liver disease, there is no measure of residual liver function comparable to creatinine clearance in renal disease. Endogenous biochemical markers such as AST, ALT, bilirubin, and INR are qualitative but not quantitative for liver function Clearance tests of exogenous markers such as aminopyrine, indocyanine green, and lorazepam have not proved to be clinically useful. The Child-Pugh classification of liver disease can be used clinically to indicate mild (type A), moderate (type B), or severe (type C) disease, but this does not predict drug metabolism.1,2

Patients with both liver disease and kidney disease may have further changes in the protein binding of drugs because patients with renal failure retain a high-molecular-weight protein that displaces drugs such as phenytoin. They also may have a low albumin. There is thus a significant potential for liver disease to alter the pharmacokinetics of AEDs, as well as their pharmacodynamics. For example, some patients with chronic liver disease (e.g., alcoholic cirrhosis) may have chronic mild to moderate encephalopathy, which could make these patients more sensitive to the CNS side effects of AEDs.1-3

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74. With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

Orally administered medications must dissolve and go into solution in the GI tract before absorption across the gut wall into circulating blood. These processes depend upon many factors, including:4

• lipid solubility
• concentration of nonionized molecules
• gastric blood flow
• gastric emptying time
• gut pH
• timing of meals
• meal content
• bowel length

GI diseases that alter these factors may therefore affect drug absorption and disposition.

In addition to playing a role in absorption, the GI tract also contributes to drug metabolism. For example, two enzymes that are found in the gut, alcohol dehydrogenase and CYP 3A4, contribute to xenobiotic metabolism.

It is difficult to perform prospective trials in patients with acute GI disease. For example, it would be hard to study the pharmacokinetics and pharmacodynamics of a drug in a patient with diarrhea secondary to influenza. Even those with chronic GI diseases may have confounding factors that make pharmacokinetic and pharmacodynamic evaluation difficult.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

Carbamazepine is extensively metabolized (primarily by CYP3A4) to an active metabolite, which is further metabolized. Carbamazepine induces its own metabolism.22-23

Effect of liver disease

Carbamazepine is a low-extraction drug but is extensively metabolized by the liver. Therefore, it is expected that the metabolism of carbamazepine is sensitive to decreased hepatic function but not to changes in hepatic blood flow. A reduction in protein binding can occur in patients with liver disease.24

Effect of GI disease

The effect of GI disease has not been studied in patients receiving carbamazepine. However, because carbamazepine is poorly water-soluble and the immediate-release tablet (Tegretol) results in erratic and prolonged absorption, it is likely that GI disease will alter carbamazepine absorption. Diseases that alter GI transit can affect the absorption of the osmotic-pump controlled-release preparation of carbamazepine (Tegretol-XR). In normal volunteers, Wilding demonstrated significant variability in GI transit time of the osmotic-pump controlled-release formulation, which resulted in variability in carbamazepine absorption.25 Absorption variability from the immediate-release tablets is also likely because the absorption is slow, erratic, and prolonged. Absorption from the suspension and the sustained-dosage form (Carbatrol) is unknown.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

There is no first-pass metabolism but the drug is extensively metabolized in the liver to inactive metabolites.19 CYP3A4 is the main isoenzyme involved in the metabolism of ethosuximide, although CYP2E also has an important role and CYP2B may be involved.20-21

Effect of liver disease

Because of the extensive liver metabolism, significant impairment of liver function may decrease the clearance of ethosuximide.

Effect of GI disease

The effects of GI disease on ethosuximide are unknown.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

Felbamate is a low-extraction drug and is metabolized mainly by hydroxylation and conjugation.30-31

Effect of liver disease

Because felbamate is not extensively protein-bound, the free fraction is not significantly changed in liver disease. The effect of liver disease on the metabolism of felbamate is unknown. The mixture of renal and hepatic elimination may minimize the effect of liver disease. Patients with impairment of both kidney and liver function would be expected to have a reduced clearance of felbamate.

When felbamate was initially approved by the U.S. Food and Drug Administration (FDA), it was felt to be extremely well tolerated. However, after 100,000 patient exposures, the use of felbamate was associated with the occurrence of aplastic anemia and hepatotoxicity. These side effects are potentially fatal. Hepatotoxicity occurs less often than aplastic anemia. The incidence of hepatotoxicity with felbamate is felt to be comparable to the incidence with valproic acid. The patients developing hepatotoxicity were predominantly female, were equally divided between adult and pediatric patients, and had a broad range of time to presentation after starting felbamate therapy. The role of concomitant therapy with other drugs that may cause liver disease is unclear. A reactive metabolite, atropaldehyde, may be responsible for the hepatotoxicity. HLA studies may identify high risk patients.

Felbamate is still available, but the approval of other AEDs has relegated it to deep reserve status. It is unlikely that it would be used for patients with liver disease without careful consideration. Felbamate may be considered for patients who have failed other therapies. It is also useful in treating the Lennox-Gastaut syndrome.

Patients who start taking felbamate should be closely monitored for any signs of bone marrow suppression such as easy bruising, sore throat, or infections, and for the prodromal symptoms of liver disease, such as lethargy, dark urine, and jaundice.

Effect of GI disease

The effect of GI disease on the pharmacokinetics of felbamate is unknown.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

Gabapentin is rapidly absorbed, reaching a peak in 2 to 4 hours. It binds to a leucine-amino acid protein in the gut and is actively absorbed. The bioavailability of gabapentin decreases with increase in dose. It is not metabolized, is not bound to plasma proteins, and is excreted unchanged in the urine.32

Effect of liver disease

Liver disease should not affect gabapentin because it is not hepatically metabolized and is not protein-bound.

Effect of GI disease

It is unknown whether GI disease affects the leucine-amino protein that is responsible for the active absorption of gabapentin.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

Lamotrigine is rapidly absorbed, reaching peak concentrations in 1 to 3 hours. It is well absorbed and has an oral bioavailability of 98%. The protein binding of lamotrigine is only 55%. It is metabolized by glucuronide conjugation and less than 1% is excreted renally.33-35

Effect of liver disease

One report describes the metabolism of lamotrigine in seven subjects with Gilbert’s syndrome, a benign deficiency in the enzyme bilirubin uridine diphosphate glucuronyl transferase. Although clearance was reduced and half-life was prolonged, these effects were considered clinically insignificant.36

Because lamotrigine is extensively metabolized, loss of hepatocyte function may decrease clearance. Phase I metabolism is generally affected more by loss of hepatocyte function than is phase II, however. Protein binding of lamotrigine would not be expected to be significantly altered by liver disease.

Effect of GI disease

The effect of GI disease on the pharmacokinetics of lamotrigine is unknown.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

Levetiracetam is rapidly and almost completely absorbed. Protein binding is <10 % and 66% of levetiracetam is excreted renally as unchanged drug. Metabolism of the remaining drug occurs by hydrolysis of the acetamide group and is not cytochrome P450 dependent.41

Effect of liver disease

The clearance of levetiracetam is reported to be unaffected by liver disease.

Effect of GI disease

The effect of GI disease on levetiracetam is unknown.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

More than 96% of an oral dose of oxcarbazepine is absorbed. Oxcarbazepine is rapidly converted to a monohydroxy derivative (MHD), the active component, by cytosol arylketone reductase. The MHD is further metabolized by glucuronide conjugation. There is no evidence of autoinduction, but the metabolism of oxcarbazepine to an inactive metabolite may be induced. About 60% of oxcarbazepine and 40% of MHD are bound to plasma proteins.42-43

Effect of liver disease

Oxcarbazepine is only about 60% protein-bound and is rapidly converted to its active metabolite MHD by non-liver enzymes. Therefore, it is unlikely that liver disease will alter the pharmacokinetics of the parent drug. MHD does undergo metabolism to an inactive metabolite by glucuronide conjugation. Phase II metabolism is less likely to be affected by liver disease than phase I pathways (e.g., CYP). Therefore, it is unlikely that liver disease will affect the pharmacokinetics of oxcarbazepine or its active metabolite MHD. Patients with liver impairment are unlikely to require dosage adjustments.

Effect of GI disease

The effect of GI disease on oxcarbazepine is unknown.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed and revised March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

In the liver, phenobarbital is parahydroxylated and subsequently conjugated to glucuronic acid. The extent of glucuronide formation of phenobarbital varies widely. About 60–80% of the drug is metabolized by the liver, but alkalinization of the urine increases the amount of phenobarbital excreted unchanged by the kidney.5,6

Effects of liver disease

Although a significant amount of phenobarbital is excreted unchanged in the urine, the clearance of phenobarbital is altered in patients with liver disease. In cirrhotic patients, the half-life of phenobarbital is 130 ± 15 hours, compared to 86 ± 3 hours in healthy controls. The half-life of only one dose of phenobarbital is not altered in patients with acute hepatitis, however. Patients with cirrhosis have a decreased ability to form the parahydroxy metabolite of phenobarbital.7 It is unlikely that there are any clinically relevant changes in protein binding of phenobarbital in patients with liver disease.

Effects of GI disease

The presence of food and neutralizing agents or the occurrence of rapid gastric emptying slows phenobarbital absorption.6 Therefore, GI disease may alter the absorption of phenobarbital.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

Fosphenytoin is a diphosphate ester of phenytoin that is very water-soluble, making it a convenient intravenous dosage form. Its molecular weight is 1.5 times that of phenytoin. The commercial product is labeled in terms of phenytoin equivalents and should be dosed on that basis. It is rapidly converted systemically to phenytoin.8-10

Effects of liver disease

Changes in hepatic blood flow do not alter phenytoin clearance because it is a low-extraction drug.10 However, loss of functional hepatocytes decreases phenytoin metabolism. Phenytoin accumulates as hepatic dysfunction increases.11 Because of decreased albumin production, liver disease is associated with decreased protein binding capacity for phenytoin. Further, bilirubin may compete for binding sites of the albumin molecule, further increasing the unbound concentration.12-17

The effect of liver diseases on fosphenytoin was evaluated in four patients with liver dysfunction, four patients with renal dysfunction, and four control subjects. There was no difference in the time to peak fosphenytoin concentrations, but the time to achieve peak plasma concentrations of phenytoin was faster in the patients with liver and renal impairment because of decreased protein binding of fosphenytoin and phenytoin.18

Effects of GI disease

The absorption of phenytoin is pH dependent and is maximal in the duodenum.9 Diseases that alter gut pH and decrease duodenal dwell time are likely to alter the absorption of phenytoin.

Nasogastric feedings, which increase GI transit, decrease the absorption of phenytoin. The mechanism of this interaction remains unclear. It may result from binding of phenytoin to proteins in the feeding or from decreased time in the duodenum, where absorption is optimal.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

Tiagabine is rapidly and almost completely absorbed, with peak concentrations achieved in 30 to 90 minutes. Tiagabine is 96% bound to albumin and alpha glycoprotein and is extensively metabolized by the hepatic CYP3A4 isoenzyme.39

Effect of liver disease

The pharmacokinetics of tiagabine were compared in four subjects with mild liver impairment, three subjects with moderate liver impairment, and matched controls. The Cmax, Cmin, AUC, and elimination half-lives were all higher in the patients with liver impairment. The free fractions were also increased in the hepatically impaired subjects. The recommendation is that the dose, dosing interval, or both be reduced when tiagabine is given to patients with liver impairment.40

Effect of GI disease

The effect of GI disease on the pharmacokinetics of tiagabine is unknown.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

The absorption of topiramate is rapid and nearly complete. Topiramate is poorly bound (9–17%) to plasma proteins. Both renal and hepatic elimination occurs. In the absence of enzyme induction, more than 80% of a single radiolabeled dose of topiramate is excreted unchanged in the urine.37

Effect of liver disease

The effect of liver impairment on the metabolism of topiramate was evaluated in patients with moderate to severe hepatic impairment as defined by Child-Pugh.1 When compared to healthy subjects, the plasma clearance was reduced by 26% and the half-life was prolonged by 36%.38 Both these results were clinically insignificant, but these changes may be clinically significant in selected patients. Liver disease would not be expected to cause significant changes in the protein binding of topiramate. Because the protein binding of topiramate is only 9–17% and the alteration in metabolism by liver disease is not clinically significant, dosage adjustments are not required for patients with liver disease.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

Some authors have argued that some of the metabolites, though active, may not contribute much to the efficacy of valproic acid because they are present only in low concentrations in plasma and brain. The metabolites of valproic acid also may contribute to the rare, but potentially fatal, hepatotoxicity.26

Valproic acid is a low-extraction drug and its clearance is independent of hepatic blood flow.26-27

Effect of liver disease

The protein binding of valproic acid is decreased in alcoholic cirrhosis and viral hepatitis. The elimination half-life is increased but there is no change in total clearance because clearance of the unbound drug is reduced. The result is an increase in unbound drug with no change in the total drug concentration.28

Liver disease alters the profile of valproic acid metabolites but it is not possible to distinguish between benign and life-threatening hepatic adverse reactions based on the profile of the valproic acid metabolites.29 Valproic acid may cause hepatotoxicity, especially in young children with polytherapy or inborn errors of metabolism. Two retrospective reviews demonstrated that the risks of fatal hepatotoxicity associated with valproic acid are highest in children less than 2 years of age who are taking multiple medications:

In these reports, no patient over age 10 developed fatal hepatotoxicity with valproic acid, but case reports have associated valproic acid with fatal hepatotoxicity in adults. The hepatotoxicity may be the result of a metabolite that is normally not present or present in low concentrations. This is suggested by the fact that the children receiving polytherapy had severe epilepsy associated with mental retardation, neurologic deficits, congenital anomalies, and other developmental delays. These factors, as well as the comedication, may alter the metabolism of valproic acid.26,29 Animal studies indicate that the mechanism by which valproic acid induces hepatotoxicity is different from its mechanism for antiepileptic activity.

The use of L-carnitine supplementation in the treatment and prophylaxis of valproic-acid–induced liver disease remains controversial. A recent consensus panel recommended L-carnitine supplementation for various groups:

• patients with certain secondary carnitine-deficiency syndromes
• patients with symptomatic valproate-associated hyperammonemia
• patients with multiple risk factors for valproic-acid toxicity or renal associated syndrome
• infants and young children taking valproic acid
  
• patients with epilepsy using the ketogenic diet who have hypocarnitinemia
• patients on dialysis
• premature infants receiving total parenteral nutrition

Effect of GI disease

Changes in gastric pH could change the dissolution of the enteric coating of valproic acid. Since the absorption of valproic acid is not site-specific, it is difficult to predict the effect of GI disease.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

Zonisamide has a long half-life, around 63 hours in an uninduced patient and around 27 to 38 hours in an induced patient. The half-life would be expected to be greater than 63 hours in a patient taking enzyme-inhibiting drugs. About 30% of a dose of zonisamide is excreted unchanged in the urine. Zonisamide is only about 40% protein-bound.44-46

Effect of liver disease

The pharmacokinetics of zonisamide in patients with liver dysfunction have not been formally evaluated. Because the drug is about 70% metabolized by reduction and acetylation, however, it is anticipated that the metabolism of zonisamide would be reduced in these patients.

Drugs like zonisamide with long half-lives require longer intervals between dosage adjustments to achieve steady state. Rapid titration of zonisamide has resulted in an increase in CNS side effects. Because the half-life of zonisamide is likely to be even more prolonged in patients with liver dysfunction, there should be a longer interval between dosage adjustments. Patients should be carefully evaluated clinically.

Effect of GI disease

The effect of GI disease on zonisamide is unknown.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed and revised March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

Adverse effects of AEDs on the liver range from elevation of liver enzymes to hepatic disease. Elevations of liver enzymes are much more common than liver disease. For example, up to 50% of patients taking AEDs will have an elevation of gamma-glutamyl transferase.47 In a study of 786 patients taking carbamazepine, 14% had an AST that was 2 to 3 times normal and 9% had an increase in bilirubin, but all were asymptomatic for liver disease.48

Valproic acid may be associated with a dose-dependent increase in SGOT,49 and up to half of patients taking valproic acid will development hyperammonemia.50,51 The increase in ammonia is seen more often in patients also taking other AEDs. It usually is not associated with liver enzyme abnormalities or liver disease.

Serious hepatotoxicity caused by aromatic AEDs, such as carbamazepine, phenytoin, and phenobarbital, is very rare, with an incidence of less than 1 case in 3,000 exposures. This type of hepatotoxicity usually begins within 2 to 8 weeks after the initiation of therapy and presents with a rash, fever, and internal organ involvement. It may be part of an AED hypersensitivity syndrome.52,53 The hepatotoxicity associated with valproic acid and felbamate appears to be different histopathologically from the type associated with the aromatic AEDs.

Mild to moderate elevations of liver enzymes are common with AEDs but true hepatotoxicity is rare. Elevation of liver enzymes is a poor predictor of impending liver toxicity. Patients should be carefully monitored clinically.

Adverse GI effects of AEDs

GI symptoms that may result from AED use include nausea, vomiting, and indigestion. These are usually seen in the early stages of dosage titration and may be ameliorated by slowing the titration. GI symptoms associated with valproic acid may be reduced by using the enteric coated form (Depakote). The use of valproic acid has been associated with pancreatitis.54,55

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).

Reviewed March 2004 by Steven C. Schachter, MD, epilepsy.com Editorial Board.

1. Brouwer KLR, Dukes GE, Powell JR. Influence of liver function on drug disposition. In. Evans WE, Schentag JJ, Jusko WJ (eds.). Applied Pharmacokinetics: Principles of Therapeutic Drug Monitoring 3rd ed. Vancouver, WA. Applied Therapeutics, Inc., 1992, 6-1 to 6-59.

2. Rodighiero V. Effects of liver disease on pharmacokinetics. An update. Clin Pharmacokinet. 1999,37(5):399-431.

3. Boggs J, Waterhouse E, DeLorenzo RJ. The use of antiepileptic medications in renal and liver disease. In. Wyllie E (ed). The Treatment of Epilepsy: principles and Practice, 2nd ed. Baltimore, Williams & Wilkins, 1996, 753-762.

4. Rowland M, Tozer. Absorption. Clinical Pharmacokinetics: Concepts and Applications 2nd ed. Philadelphia, Lea & Febiger, 1989, 113-130.

5. Anderson GD, Levy RH. Phenobarbital: Chemistry and Biotransformation. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:371- 377.

6. Dodson WE, Rust Jr RS. Phenobarbital: Absorption, distribution, and excretion. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:379- 387.

7. Alvin J, McHorse T, Hoyumpa A, Bush MT, Schenker S. The effect of liver disease in man on the disposition of phenobarbital. J Pharmacol Exp Ther. 1975;192:224-235.

8. Browne TR, LeDuc B. Phenytoin: Chemistry and Biotransformation. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:283-300.

9. Treiman DM, Woodbury DM. Phenytoin: Absorption, distribution, and excretion. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:301-314.

10. Tozer TN, Winter ME. In, Evans WE, Schentag JJ, Jusko WJ (eds.). Applied Pharmacokinetics: Principles of Therapeutic Drug Monitoring 3rd ed. Vancouver, WA. Applied Therapeutics, Inc., 1992, 25-1 to 25-44.

11. Kutt H, Winters W, Scherman R, McDowell R. Diphenylhydantoin and phenobarbital toxicity. The role of liver disease. Arch Neurol. 1964;11:649-656.

12. Hooper WD, Bochner F, Endre MJ, Tyrer JH. Plasma protein binding of diphenylhydantoin. Effects of sex, hormones, renal and hepatic disease. Clin Pharmacol Ther. 1974;15:276-282.

13. Olsen GD, Bennett WM, Porter GA. Morphine and phenytoin binding to plasma proteins in renal and hepatic failure. Clin Pharmacol Ther. 1975;17:677-684.

14. Reidenberg MN, Affrime M. Influence of disease on binding of drugs to plasma proteins. Ann NY Acad Sci. 1973;226:115-126.

15. Blaschke TF, Meffin PJ, Melmon KL, Rowland M. Influence of acute viral hepatitis on phenytoin kinetics and protein binding. Clin Pharmacol Ther. 1975;17:685-691.

16. Lunde PKM, Rane A, Yaffe SJ, Lund L. Sjoqvist F. Plasma protein binding of diphenylhydantoin in man. Interaction with other drugs and the effect of temperature and plasma dilution. Clin Pharmacol Ther. 1970;11:846-855.

17. Rane A, Lunde PKM, Jalling B, Yaffe S, Sjoqvist F. Plasma protein binding of diphenylhydantoin in normal and hyperbilirubinemic infants. J Pediatr. 1971;78:877-882.

18. Aweeka FT, Gottwald MD, Gambertoglio JG, Wright TL, Boyer TD, Pollock AS, Eldon MA, Kugler AR, Alldredge BK. Pharmacokinetics of fosphenytoin in patients with hepatic and renal disease. Epilepsia. 1999;40(6):777-82.

19. Garnett WR. Ethosuximide. In, Taylor WJ, Caviness MHD (eds). A Textbook for the Clinical Application of Therapeutic Drug Monitoring. Abbott Laboratories, Diagnostics Division, Irving, 1986, 225-235.

20. Pisani F, Bialer M. Ethosuximide: Chemistry and Biotransformation. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:655-658.

21. Mialer M, Xiaodong S, Perucca E. Ethosuximide: Absorption, Distribution, and Excretion. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:659-665.

22. Faigle JW, Feldmann KF. Carbamazepine: Chemistry and Biotransformation. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:499-513.

23. Morselli PL. Carbamazepine: Absorption, Distribution, and Excretion. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:515-528.

24. Hooper WD, Dubetz DK, Bochner F, Cotter LM, et al. Plasma protein binding of carbamazepine. Clin Pharmacol Ther. 1975;17:433-440.

25. Wilding IR, Davis SS, Hardy JG, et al. Relationship between systemic drug absorption and gastrointestinal transit after the simultaneous oral administration of carbamazepine as a controlled-release system and as a suspension of 15N-labelled drug to healthy volunteers. Br J Clin Pharmacol. 1991;32:573-9.

26. Baillie TA, Sheffels PR. Valproic Acid: Chemistry and Biotransformation. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:589 - 604.

27. Levy RH, Shen DD. Valproic Acid: Absorption, Distribution, and Excretion. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:605-619.

28. Klotz U, Rapp T, Mueller WA. Disposition of valproic acid in patients with liver disease. Eur J Clin Pharmacol. 1978;13:55-60.

29. Siemes H, Nau H, Schulze K, et al. Valproate (VPA) metabolites in various clinical conditions of probable VPA-associated hepatotoxicity. Epilepsia 1993;34:332-346.

30. Kucharczk N. Felbamate: Chemistry and Biotransformation. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:799- 806.

31. Perhach JL, Shumaker RC. Felbamate: Absorption, Distribution, and Excretion. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:807-812.

32. McLean MJ. Gabapentin: Chemistry, Absorption, Distribution, and Excretion. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:843 - 849.

33. Garnett WR. Lamotrigine: Pharmacokinetics. J Child Neurol. 1997;12(Suppl.1):S10-S15.

34. Dickins M, Sawyer DA, Morley TJ, Parsons DN. Lamotrigine: Chemistry and Biotransformation. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:871-75.

35. Parsons DN, Dickins M, Morley TJ. Lamotrigine: Absorption, Distribution, and Excretion. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:877-881.

36. Posner J, Cohen AF, Land G, Peck AW. The pharmacokinetics of lamotrigine (BW 430C) in health subjects with unconjugated hyperbilirubinemia (Gilbert’s syndrome). Br J Clin Pharmacol. 1989:28:117-120.

37. Garnett WR. Clinical pharmacology of topiramate: A review. Epilepsia. 2000;41(Suppl. 1):S61-S65.

38. Doose DR, Walker SA, Venkataramanan R, Rabinovitz M, Lever J. Topiramate pharmacokinetics in subjects with liver impairment. J Pharm Res. 1994;11(Suppl):S446.

39. Gustavson LE, Mengesl HB. Pharmacokinetics of tiagabine, a gamma-aminobutyric acid-uptake inhibitor, in healthy subjects after single and multiple doses. Epilepsia 1995;36:605-11.

40. Lau AH, Gustavson LE, Sperelakis R, Lam NP, El-Shourbagy T, Qian JX, Layden T. Pharmacokinetics and safety of tiagabine in subjects with various degrees of hepatic function. Epilepsia. 1997;38(4):445-451.

41. Patsalow PN. Pharmacokinetic profile of levetiracetam: Toward ideal characteristics. Pharmacol Ther. 2000;85(2):77-85.

42. Tecoma ES. Oxcarbazepine. Epilepsia. 1999;40(Suppl 5):S37-S46.

43. Dam M, Ostergaard LH. Oxcarbazepine. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:987-995.

44. Leppik IE. Zonisamide. Epilepsia. 1999;40(Suppl 5):S23-S29.

45. Kochak GM, Page JG, Buchanan RA, Peters R, Padgett CS. Steady-state pharmacokinetics of zonisamide, an antiepileptic agent for treatment of refractory complex partial seizures. J Clin Pharmacol. 1998;38(2):166-71.

46. Seino M, Naruto S, Ito T, Miyazaki H. Zonisamide. In Levy RH, Mattson RH, Meldrum BS (eds), Antiepileptic Drugs (4th ed). New York: Raven Press, 1995:1011-1023.

47. Krauss G, Crone N. Non-CNS side effects of antiepileptic drugs. Medscape (http://www.medscape.com/viewprogram/308), February 2000.

48. Camfield C, Camfield P, Smith E, Tibbles JA. Asymptomatic children with epilepsy: little benefit from screening for anticonvulsant-induced liver, blood, or renal damage. Neurology 1986;36:838-841.

49. Haidukewych D, John G. Chronic valproic acid and coantiepileptic drug therapy and incidence of increases in serum liver enzymes. Ther Drug Monit. 1986;8:407–410.

50. Kugoh T, Yamamoto M, Hosokawa K. Blood ammonia level during valproic acid therapy. Jpn J Psychiatry Neurol. 1986;40:663–668.

51. Zaccara G, Paganini M, Campostrini R, et al. Effect of associated antiepileptic treatment on valproate-induced hyperammonemia. Ther Drug Monit. 1985;7:185–190.

52. Schlienger RG, Shear NH, Antiepileptic drug hypersensitivity syndrome. Epilepsia. 1998;39:S3–S7.

53. Knowles SR, Shapiro LE, Shear NH. Anticonvulsant hypersensitivity syndrome: incidence, prevention and management Drug Safety. 1999;21:489–501.

54. Asconape JJ, Penry JK, Dreifuss FE, et al. Valproate-associated pancreatitis. Epilepsia. 1993;34:177–183.

55. Binek J, Hany A, Heer M. Valproic-acid–induced pancreatitis. Case report and review of the literature. J Clin Gastroenterol. 1991;13:690–693.

Adapted from: Garnett WR. Gastrointestinal and hepatic disease. In: Ettinger AB and Devinsky O, eds. Managing epilepsy and co-existing disorders. Boston: Butterworth-Heinemann; 2002;63-74.
With permission from Elsevier (www.elsevier.com).