The Use of Psychotropic Drugs in Epilepsy: A Review of Pharmacokinetic, Pharmacodynamic, and Safety Issues
Andres M. Kanner
Barry E. Gidal
Patients with epilepsy constitute a population that is at high risk for comorbid psychopathology (1). As shown in Table 4.1, prevalence rates of the four major psychiatric disorders, depression, anxiety disorders, schizophrenia, and attention-deficit hyperactivity disorder (ADHD), are significantly higher among patients with epilepsy than in the general population (2,3,4,5).
The comorbid psychiatric disorders are more frequent among patients with poorly controlled seizures (6,7,8). The comorbid psychiatric disorders can also be encountered in seizure-free patients and in patients with milder forms of epilepsy, such as patients with idiopathic generalized epilepsy (9). In a population-based study, Ettinger et al. found a prevalence of symptoms of depression in approximately 36.5% of 775 people with epilepsy, which is a significantly higher prevalence than that found in people with diabetes, asthma, and those who describe themselves as being completely healthy (10). Likewise, in children with epilepsy, for example, the prevalence rate of ADHD is relatively high, ranging between 15% and 30% among different studies. For example, McDermott et al. carried out a population-based analysis of psychiatric symptomatology among children with epilepsy, children who have cardiac disorders, and healthy controls using the Behavioral Problem Index (11). Behavioral problems were significantly more frequent among children with epilepsy (28.1%) than among children with cardiac disorders (12.6%) and healthy controls (4.9%). Headstrong and oppositional behavior was again identified more frequently in children with epilepsy (28.1%) than in children with cardiac disorders (18.6%) and healthy controls (8.6%). Antisocial behavior was reported in 18.2% of children with epilepsy, 11.6% in children with cardiac disorders, and 8.8% in healthy controls. It is clear from the examples cited in the preceding text that patients with epilepsy may require the use of psychotropic medication for the management of their comorbid psychiatric disorders.
Antidepressant drugs are the most frequently prescribed psychotropic drugs in patients with epilepsy; they are used for the management of mood disorders, anxiety disorders including panic disorders, generalized anxiety disorders, obsessive-compulsive disorder, and social phobia. Central nervous system (CNS) stimulants are the other class
of drugs that is quite frequently prescribed for the treatment of ADHD; finally, antipsychotic medications are used less frequently and are usually prescribed for the management of psychotic disorders, as well as bipolar illness and severe anxiety or behavioral disorders.
of drugs that is quite frequently prescribed for the treatment of ADHD; finally, antipsychotic medications are used less frequently and are usually prescribed for the management of psychotic disorders, as well as bipolar illness and severe anxiety or behavioral disorders.
TABLE 4.1 Prevalence Rates of Psychiatric Disorders in Patients with Epilepsy and the General Population | |||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
Despite the relatively frequent need to prescribe psychotropic drugs for patients with epilepsy, there are very few controlled studies on the use of these drugs in this patient population. To date, for example, there has been only one controlled study of antidepressant drugs for the management of major depression. This study was published in 1988 (12). Furthermore, several of the psychiatric comorbid conditions have an atypical presentation in patients with epilepsy, such as some of the depressive and psychotic disorders. Therefore, the assumption that these types of comorbid psychiatric disorders respond in the same manner as primary psychiatric disorders remains an unproven fact with sparse data to support it. In fact, it is safe to say that the practice of psychopharmacological therapy in patients with epilepsy continues to be empirical to this date. The aim of this chapter is to review the safety of the use of psychotropic drugs in patients with epilepsy and in the general population, particularly with respect to their proconvulsant property. The second part of this chapter reviews the pharmacokinetic and pharmacodynamic interactions of the four classes of drugs with antiepileptic drugs (AEDs), with special attention on the adverse events mediated by AEDs that can be enhanced by the psychotropic medications.
Impact of Psychotropic Drugs on the Risk of Seizure Occurrence
Before reviewing the available data, it is important to consider new evidence that suggests the existence of a bidirectional relationship between epilepsy and psychiatric disorders, in particular depression and ADHD (13,14,15,16). This bidirectional relation implies that not only are patients with epilepsy at greater risk of developing depression or ADHD, but patients with ADHD and depression are also at greater risk of developing epilepsy. Clearly, this bidirectional relationship does not imply that the psychiatric disorder causes epilepsy; rather, it implies the existence of common pathogenic mechanisms in epilepsy and psychiatric disorders.
For example, in a study carried out in Iceland in which all children and adults with epilepsy were identified, children with ADHD (inattentive type) were 2.5 times more likely to develop epilepsy than a control group (13). Three population case–control studies have established that people with epilepsy have three to seven times higher risk of developing epilepsy compared with controls (14,15,16). People with a history of suicidality were found to have a 5.5% higher risk of developing epilepsy relative to controls (16). These data are important because they examine the causal role of seizure occurrence played by psychotropic drugs used to treat depression or ADHD in a new light. In other words, it is important to reevaluate the inherent risk of seizure occurrence associated with psychiatric disorders, before assuming that seizure occurrence had resulted from psychotropic drugs used for their treatment.
For example, in a study carried out in Iceland in which all children and adults with epilepsy were identified, children with ADHD (inattentive type) were 2.5 times more likely to develop epilepsy than a control group (13). Three population case–control studies have established that people with epilepsy have three to seven times higher risk of developing epilepsy compared with controls (14,15,16). People with a history of suicidality were found to have a 5.5% higher risk of developing epilepsy relative to controls (16). These data are important because they examine the causal role of seizure occurrence played by psychotropic drugs used to treat depression or ADHD in a new light. In other words, it is important to reevaluate the inherent risk of seizure occurrence associated with psychiatric disorders, before assuming that seizure occurrence had resulted from psychotropic drugs used for their treatment.
This is a question of significant clinical importance, as some classes of psychotropic drugs are in fact known to lower the seizure threshold and could potentially worsen the seizure disorder in patients with established epilepsy, whereas it is possible that seizures attributed to antidepressants and CNS stimulants may have in fact been the expression of the inherent risk of depression and ADHD alluded to in the preceding text. Furthermore, the concern of using antidepressants and CNS stimulants for fear that they may cause seizures or worsen seizures in people with established epilepsy has resulted on many occasions in patients’ comorbid depressive and ADHD going untreated.
Antidepressant Medication
As stated in the preceding text, antidepressant medications are commonly used currently for the management of major depressive disorders, panic disorders, generalized anxiety disorders and obsessive-compulsive disorders, and, finally, ADHD in children and adults who cannot tolerate or do not respond to CNS stimulants. An increased serotonergic (5-hydroxytryptamine [5-HT]) with or without noradrenergic (norepinephrine [NE]) and dopaminergic (dopamine [DA]) CNS activity have been the basic mechanisms of action of most classes of antidepressant drugs. Some of the antidepressants have a mixed serotonergic/noradrenergic action (i.e., imipramine, venlafaxine, duloxetine), whereas others have a pure serotonergic (fluoxetine, sertraline, paroxetine, citalopram) or noradrenergic (desipramine) effect while acting on dopaminergic and noradrenergic receptors (i.e., bupropion). It is of note that there is a growing body of literature that suggests an antiepileptic effect mediated by noradrenergic and serotonergic mechanisms (Chapter 3). For example, increments of either NE or 5-HT transmission with the selective serotonin reuptake inhibitor (SSRI) sertraline resulted in a dose-dependent seizure-frequency reduction in the genetically epilepsy-prone rats (GEPRs) that correlated with the extracellular thalamic serotonergic thalamic concentration (17). In addition, the 5-HT precursor 5-HTP has been shown to have anticonvulsant effects in GEPRs when combined with a monoamine oxidase inhibitor (MAOI) (18), whereas SSRIs and MAOIs have been found to exert anticonvulsant effects in genetically prone epilepsy mice (19) and baboons (20,21) and in non–genetically prone cats (22), rabbits (23), and rhesus monkeys (24). Conversely, drugs that interfere with the release or synthesis of NE or 5-HT exacerbate seizures in the GEPRs. These include NE storage vesicle inactivators, reserpine or tetrabenazine, the NE false transmitter α-methyl-m-tryosine, the NE synthesis inhibitor α-methyl-Δ-tyrosine, and the 5-HT synthesis inhibitor Δ-chlorophenylalanine.
In humans, the evidence is less robust; however, in a double-blind crossover study, imipramine therapy resulted in a significant reduction in seizure frequency in patients with absence and myoclonic-astatic epilepsy despite the discontinuation of other antiepileptic medications (25). Open trials with the tricyclic drugs doxepin and nortriptyline and the SSRIs fluoxetine and citalopram have also suggested a significant reduction in seizure frequency in patients with refractory epilepsy (26,27,28), but
these results need to be confirmed in controlled studies.
these results need to be confirmed in controlled studies.
The first two classes of antidepressant medications to be developed were the tricyclic antidepressants (TCAs) and tetracyclic antidepressants (TTAs). The former class includes eight drugs: amitriptyline, imipramine, desipramine, doxepin, clomipramine, nortriptyline, protriptyline, and trimipramine maleate, whereas the TTAs include amoxapine and maprotiline. These two classes of antidepressants were reported to cause seizures in nonepileptic patients, primarily in the setting of overdoses or in patients who have high serum concentrations attributed to a genetically determined slow metabolic rate. Among the family of TCAs, however, clomipramine was identified as an agent with a significantly higher incidence of seizures (1%), and therefore it is not recommended for use in people with epilepsy (29). Jobe estimated the risk ratio for epilepsy in patients taking TCAs to be 1.5, compared with a risk ratio of 3.7 for people with depression (30). On the other hand, the two TTAs, amoxapine and maprotiline, were also found to have higher incidence of seizures and are also not recommended in this patient population (31). A low initial dose and slow titration is recommended with the use of TCAs in patients with epilepsy. Furthermore, serum concentrations of most antidepressants of this class are available, and it is recommended that they be measured once the desired dose is reached to identify those patients who may be “slow metabolizers.”
The next class of antidepressant drugs to come onto the market was the MAOIs, which include four drugs: phenelzine, isocarboxazid, tranylcypromine, and selegiline. At present, MAOIs have been restricted to the management of certain patients with atypical depressions or borderline personality disorders, and since the advent of the SSRIs, they have been used as a third-line treatment drug. These medications have been found to be safe in people with epilepsy and their limited use pertains rather to the potential adverse events associated with the ingestion of certain tyramine-containing foods that can result in hypertensive crisis (32).
The third class of antidepressants, which is currently preferred as a first line treatment for mood and anxiety disorders is the SSRIs, which includes six drugs: fluoxetine, sertraline, paroxetine, fluvoxamine, citalopram, and escitalopram. In line with the evidence cited in the preceding text linking 5-HT anticonvulsant effects, there are some animal data that have suggested a positive pharmacodynamic synergistic interaction (anticonvulsant) between SSRIs such as fluoxetine and AEDs such as phenytoin or carbamazepine (33,34,35). However, other studies have failed to confirm these observations. Rapeport et al. did not find a meaningful pharmacodynamic interaction between the serotonin reuptake inhibitor sertraline and either phenytoin or carbamazepine (36,37). Whereas evidence for a positive, synergistic dynamic interaction is lacking, there does not appear to be evidence to support a substantial risk of adverse, proconvulsant interaction between SSRIs and AEDs.
In line with the observations that patients with depression may have a 3% to 7% higher risk of developing unprovoked seizures or epilepsy, Kahn et al. completed a study in which they compared the incidence of seizures among patients randomized to three SSRIs (escitalopram, fluoxetine, and fluvoxamine) with placebo as part of the regulatory studies for the treatment of depression (38). They found that patients randomized to the antidepressants were significantly less likely to have had an epileptic seizure compared with controls (24).
The experience with the newer antidepressant drugs in the family of serotonin norepinephrine reuptake inhibitor (SNRI), venlafaxine and duloxetine, is very limited, but the incidence of seizures in the regulatory studies compared with placebo-exposed patients does not appear to be increased.
Another class of drugs is the NE and DA reuptake blocker bupropion, which has relatively weak reuptake properties of DA and NE, but its active hydroxylated metabolite has strong reuptake properties of these neurotransmitters and seems to be responsible
for the efficacy of this agent. This a drug used for a certain type of depressive order (i.e., retarded depression) and for the management of ADHD. In an early study, bupropion appeared to be relatively safe, even at dosages above 450 mg per day (39). However, in a study of 69 patients with bulimia, 4 patients developed generalized tonic-clonic seizures (40). A rate of seizure frequency ranging from 0.36% to 0.48% has been reported in patients exposed to bupropion at a dosage of 450 mg per day or less (41). Lower doses, however, were identified with extended release formulations at doses of 300 mg per day or less in a study of 3,100 patients, yielding a seizure frequency of 0.15% (42). This lower frequency of seizures suggests that avoidance of a peak of dose effect may have an important role in minimizing the proconvulsant properties of this antidepressant medication. If this medication is to be used presently in patients with epilepsy, it should be started at a low dose and not exceed doses of 300 mg per day. In general, it is recommended that the drug be avoided.
for the efficacy of this agent. This a drug used for a certain type of depressive order (i.e., retarded depression) and for the management of ADHD. In an early study, bupropion appeared to be relatively safe, even at dosages above 450 mg per day (39). However, in a study of 69 patients with bulimia, 4 patients developed generalized tonic-clonic seizures (40). A rate of seizure frequency ranging from 0.36% to 0.48% has been reported in patients exposed to bupropion at a dosage of 450 mg per day or less (41). Lower doses, however, were identified with extended release formulations at doses of 300 mg per day or less in a study of 3,100 patients, yielding a seizure frequency of 0.15% (42). This lower frequency of seizures suggests that avoidance of a peak of dose effect may have an important role in minimizing the proconvulsant properties of this antidepressant medication. If this medication is to be used presently in patients with epilepsy, it should be started at a low dose and not exceed doses of 300 mg per day. In general, it is recommended that the drug be avoided.
In summary, to date, antidepressant medications can be safely used in patients with epilepsy, with drugs of the SSRIs being among the safest ones. TCAs can be safely used but can be started at low doses and increased slowly, and serum concentrations should be monitored to identify patients who may be slow metabolizers. Clomipramine, maprotiline, and amoxapine should, however, be avoided in patients with epilepsy. MAOIs can be safely used in patients with epilepsy, whereas bupropion should be avoided, unless there is no other option. If it were to be used, the dose should not surpass 300 mg per day.
Central Nervous System Stimulants
The CNS stimulants used for ADHD include methylphenidate in its different formulations (immediate release methylphenidate, extended release methylphenidate, Concerta) dextroamphetamine in its immediate and extended release formulations, and pemoline, which is used as a last choice currently because of its identified hepatotoxicity. There has been a concern among clinicians that the use of CNS stimulants may be associated with an increased risk of epileptic seizures, and hence these drugs are often avoided by pediatricians, pediatric neurologists, and neurologists, in patients with epilepsy. It should be remembered, however, that ADHD is the other psychiatric condition with a bidirectional relationship with epilepsy and children with ADHD were found to have a 2.5% higher risk of developing incident seizures or epilepsy (14). As stated in the preceding text, this bidirectional relationship obviously raises the question as to whether the witnessed seizures reported in the literature in patients started on CNS stimulants were an expression of the inherent risk of the ADHD and not the consequence of the treatment with a CNS stimulant. In fact, Wernicke reviewed the seizure risk in patients with attention deficit disorder who were treated with an NE agent atomoxetine, methylphenidate, and placebo (43). In the review of 31 studies that totaled 5,831 patients on atomoxetine, 1,015 patients randomized to placebo, and 523 patients randomized to methylphenidate, there were no statistical differences in the incidence of seizures among the three groups. The incidence of seizures was of 0.06% during the control phases. Currently, the consensus among pediatric epileptologists and neurologists is that the use of CNS stimulants in children and adults with epilepsy is safe.