Pharmacotherapy, Epilepsy, and Comorbidities: Causes for Concern

Many persons with epilepsy have comorbid conditions that are treated with concomitant medications. However, patients for whom first-generation antiepileptic drugs (AEDs) are prescribed have been shown to be at high risk for drug interactions with medications that involve the cytochrome P-450 pathway, specifically antipsychotics, contraceptives, tricyclic antidepressants (TCAs), calcium channel blockers, and warfarin.

Recently published research by Patsalos and Perucca¹ showed that medication interactions are common in elderly patients with epilepsy. Now, compelling new data from 2 retrospective cohort studies have shown that all adults and children with epilepsy are at risk for drug interactions.

The 2 new studies, led by Jacqueline French, MD,² director of the Penn Epilepsy Center and assistant dean for clinical trials at the University of Pennsylvania in Philadelphia, and Barry Gidal, PharmD, RPh,³ professor of pharmacy and neurology at the University of Wisconsin in Madison, evaluated the potential risk for drug interactions to create greater awareness of the problem within the medical community.

Both studies used retrospective claims from the PharMetrics Anonymous Patient-Centric Database. The first study² evaluated drug use in adults; the second study³ evaluated children aged 17 years and younger. The results are cause for concern.

POTENTIAL DRUG INTERACTIONS IN ADULTS

In the first study,² adults with epilepsy made up a cohort of 11,206 patients who were beginning monotherapy or adjunctive therapy with either phenytoin (Dilantin), valproic acid (Depakote), levetiracetam (Keppra), lamotrigine (Lamictal), gabapentin (Neurontin), carbamazepine, or topiramate (Topamax). Patients had to have received treatment for a minimum of 30 days. Overlap times for treatment with concomitant medications were defined as 5 consecutive days or more for antibiotics and antifungals, and 30 consecutive days or more for all other medications.

Patients were classified into the following groups: young adults (aged 18 to 34 years); adults (aged 35 to 54 years); older adults (aged 55 to 64 years); and the elderly (aged 65 years or older). Data from 2001 through 2004 were evaluated including each patient's course of treatment, history of clinical patterns, comorbidities, treatment approaches, and the cost of medical and pharmaceutical claims.

Not surprisingly, the results showed that the mean concomitant medication use more or less doubled with age. Concomitant drug use increased from 2.4 in young men to 6.2 in older men and from 4.0 in young women to 7.9 in older women. The mean number of concomitant medications was significantly higher among women than it was among men, specifically among patients aged 18 to 64 years.

Of note, the mean time to addition of a first concomitant medication was 67 days. The potential that patients would be taking 1 or more medications in addition to AEDs at 10 months was more than 80% irrespective of age.

Statin use was highest in the elderly group (34.7%); in older adults, selective serotonin reuptake inhibitors (SSRIs) were the most common concomitant medication. Other medication classes used in more than 10% of the study population included antipsychotics, contraceptives, calcium channel blockers, sulfonylureas, and TCAs. Warfarin use also was common in elderly and older adult patients.

Phenytoin and carbamazepine were the most frequently used first-generation AEDs.

"This is relevant because there are quite a number of first-generation AEDs that affect the way other drugs are handled by the body. They either speed up or induce the metabolism of certain drugs, or they slow down their metabolism," said French. "I'm referring to very common drugs-drugs that are still widely used in the United States and familiar to everyone, such as phenytoin, carbamazepine, valproic acid, and phenobarbital," she said.

The problem arises when the older drugs are used-particularly those that increase metabolism of other prescription drugs-but the treating physician is not aware of the impact. "For example, if an antiepileptic medication is prescibed for a healthy 25-year-old-and many people with epilepsy are in that age range-it's not uncommon to think that there's no need to worry about the effects on other prescription drugs. Clinicians think, what are the chances that a healthy 25-year-old will be taking drugs for chronic illnesses other than epilepsy?" French commented.

This thinking suggests a false sense of confidence, which can have far reaching therapeutic consequences. "To give you an idea of the impact of these drug interactions, if a patient is taking phenytoin and is then given a cholesterol-lowering drug, the concentration of that cholesterol-lowering drug will be reduced by at least half if not two thirds," French explained. "So the standard dosage that the primary care doctor knows to prescribe will be very diluted and may not have the impact that it's supposed to have," she said. "In this case it is easy to rectify because cholesterol monitoring will show that the cholesterol levels are not being reduced.

"However, to take another example, specifically antidepressants; some of these drugs are diluted by a half or more," French continued. "So primary care doctors, who prescribe antidepressants most frequently these days, may not be expert enough to know whether the antidepressant is having the anticipated effect. Consequently, patients may be undertreated."

Probably the most obvious and common example of the problem is the concomitant use of oral contraceptives and AEDs, French said. The effectiveness of the oral contraceptive is compromised, and the result may be unwanted pregnancy.

POTENTIAL DRUG INTERACTIONS IN CHILDREN

In the second study,³ which evaluated the potential for drug interactions in 4663 children aged 17 years and younger, it was learned that a high percentage of patients used 5 or more medications. The mean number of concomitant medications with an associated high risk of interactions was 1.43.

The patients began treatment with either an enzyme-inducing AED (EIAED) (N = 2145)-specifically, phenytoin, carbamazepine, phenobarbital, or primidone-or a non-enzyme-inducing AED (NEIAED) (N = 2509)-specifically divalproex sodium, valpro-ate, tiagabine (Gabitril), levetiracetam, lamotrigine, gabapentin, topiramate, oxcarbazepine (Trileptal), and zonisamide (Zonegran).

As in the adult study, a difference between the sexes became apparent. Although 74.7% of the girls and 75.4% of boys were taking at least 1 concomitant medication, the girls who took EIAEDs were significantly more likely to be taking more concomitant medications than the boys taking EIAEDs: 21.3% versus 16.8%. However, no significant difference in mean concomitant medication prescribing patterns was found between the EIAED and NEIAED groups. The most common and relevant concomitant medications were SSRIs, antipsychotics, contraceptives, TCAs, antibiotics, and antifungals.

The data also showed that as patients aged, the number of concomitant medications decreased significantly, from a mean of 3.76 for neonates to 2.33 for adolescents. Furthermore, this decrease was maintained regardless of the type of AED prescribed, and it occurred largely in patients aged 0 to 10 years. After age 10, medication use stabilized.

Overall, patients with epilepsy aged 0 to 17 years had a higher rate of medication use than patients of the same age in the general population^4,5: 54% were taking more than 2 medications versus 6% in the general population.

CLINICAL IMPLICATIONS

The number of concomitant medications prescribed in these populations was unexpected by French and her team. "We were somewhat surprised. The number was very, very high, even in patients who were reasonably young," French said. Despite their ages, a large number of the pediatric patients with epilepsy are receiving pharmacological therapy for other medical conditions. This could have far reaching effects, particularly in patients being treated for epilepsy and depression.

"A recently published article in Lancet Neurology noted a 2.5% incidence of suicide in the epileptic population. This rate is very much higher than it is in the general population," French said. The study by Chistensen and colleagues,⁶ a large, Danish, population-based, case-controlled study, showed that the rate of suicide among persons with epilepsy was 3 times higher than that in the general population. Even after adjusting for psychiatric disease and socioeconomic factors, the relative risk was 1.99.

Patients at highest risk were those in whom epilepsy had been diagnosed within the past 6 months. The authors also noted that the risk of suicide was particularly high among patients with a history of comorbid psychiatric disease.

Christensen and colleagues also found that the greatest number of suicides occurred among women, whereas in the general population suicide is higher among men. Furthermore, the risk of suicide among those studied decreased with age, a trend that is contrary to that seen in the general population. The authors theorize that this may be due to age-related effects of quality of life and age-related differences in the underlying pathology of epilepsy.

"I think the incidence of suicide is even high for people with neurological disorders. Depression is extremely common, and obviously if there is such a high suicide rate, getting adequate treatment in a timely fashion is very important," French said.

But there are other factors involved in the decision to prescribe AEDs, such as cost. "Even if the physician does understand what the interactions between the various drugs are and decides to give twice as much of, say, a cholesterol-lowering agent [to compensate for drug interactions], that's not going to be inexpensive for the patient," French said. "In previous years there was a trade-off with respect to cost because the new AEDs were that much more expensive than the older ones. However, the newer generation of AEDs will soon be generic, so there really is no excuse to prescribe older AEDs based solely on cost," French said. "While some patients can benefit from the older drugs, given all of the new data, prescribing AEDs that don't negatively impact the other drugs [they take] seems like a reasonably good idea," she said.

GOING FORWARD

When asked how these data might be incorporated into clinical decision making and clinical guidelines, French said, "I can only speak for the Academy of Neurology, because I was involved in creating its evidence-based guidelines for using the newer AEDs. Essentially, the guidelines are based on the evidence of the effect of the drug-whether the data are class 1, 2, or 3," she said. "It's very hard to classify the adverse effects associated with the various drugs in the same way. We don't have a classification scheme for the data that we're talking about here. Therefore it is very difficult to incorporate," she said.

French and her team are currently assessing how best to proceed with their findings so that the data may better inform clinical practice. "Possibilities include gathering a cohort of patients and looking at the doses of concomitant medications that they're taking in conjunction with which AED they're receiving and using measurable clinical outcomes to determine the effects," French said.

"We believe that the risk of drug interactions in the epileptic population is a much bigger issue than people realize," French said. "Therefore, it is important to think not only about the drugs that the patient is taking today but also what she or he may be taking next year or the year after next, and that physicians prescribe with that thought in mind."

References:

References

1.

Patsalos PN, Perucca E. Clinically important drug interactions in epilepsy: general features and interactions between antiepileptic drugs.

Lancet Neurol.

2003;2:347-356.

2.

French J, Grossman P, Gidal B, et al. Potential drug interactions in adult patients with epilepsy: risk of concomitant medication use.

Neurology.

2007; 68(suppl 1):A123.

3.

Gidal B, French J, Grossman P, et al. Potential drug interactions in children with epilepsy: an overview of concomitant medication use by age and gender using a retrospective claims database.

Neurology.

2007;68(suppl 1):A123.

4.

American Society of Health-System Pharmacists (ASHP). Snapshot of medication use in the U.S./ ASHP Research Report. December 2000. Available at:

www.ashp.org/s_ashp/docs/files/PR_snapshot.pdf

. Accessed October 23, 2007.

5.

Kaufman DW, Kelly JP, Rosenberg L, et al. Recent patterns of medication use in the ambulatory adult population of the United States: the Slone survey.

JAMA.

2002;287:337-344

6.

Chistensen J, Vestergaard M, Mortensen PB, et al. Epilepsy and the risk of suicide: a population-based case control study.

Lancet Neurol.

2007;6:693-698.