A psychotropic drug may be needed to treat symptoms of depressive disorders in a patient with developmental disability (DD) and epilepsy, despite concerns that the seizure threshold may be lowered and seizures exacerbated. For patients with the symptoms of these disorders, the mainstay of treatment is an antidepressant.

Diagnostic criteria

The diagnosis of a major depressive disorder (MDD) can be made if the patient has five or more of these symptoms at any one time:29

  • persistent depressed mood (>2 weeks)
  • anhedonia
  • weight loss or gain
  • sleep dysfunction
  • agitation or psychomotor retardation
  • guilt
  • concentration difficulty
  • recurrent thoughts of death or suicide

Dysthymic disorder is a low-grade chronic depression with many of the same features of an MDD but without the severity. Many practitioners who deal frequently with epileptic patients have emphasized the pleomorphic quality of the patients' depressive symptoms. Blumer30 coined the term interictal dysphoric disorder to describe patients with these symptoms, emphasizing especially their irritability, mood variability, and anxiety.

Behavioral dyscontrol also can be a significant problem in patients with DDs and comorbid epilepsy. In particular, overarousal and dysfunction of the serotonin system have been implicated as frequent primary factors. Beta-blockers may be useful for modulating overstimulation. Aggression may be another area in which antidepressants may be useful.31

Types of antidepressants

Antidepressants constitute a wide variety of medications.32 The "classic" agents were discovered in the 1950s and include tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs).

The TCAs used most frequently include:

  • amitriptyline
  • nortriptyline
  • imipramine
  • desipramine
  • doxepin
  • clomipramine

This class also includes two tetracyclic antidepressants:

  • amoxapine
  • maprotiline

The MAOIs include:

  • phenelzine
  • selegiline
  • tranylcypromine
  • isocarboxazid

In 1988, the introduction of the selective serotonin reuptake inhibitors (SSRIs) provided a new class of antidepressants with a unique and advantageous side effect profile. Among the SSRIs are:

  • fluoxetine
  • paroxetine
  • sertraline
  • citalopram
  • fluvoxamine

Other new antidepressants include:32

  • venlafaxine (combined serotonin-norepinephrine reuptake inhibitor [SNRI])
  • bupropion (noradrenergic-dopaminergic compound)
  • nefazodone and trazodone (5-hydroxytryptamine-2 antagonists)
  • mirtazapine (noradrenergic antidepressant

Efficacy in patients with epilepsy

The efficacy of antidepressants to treat depression in patients with epilepsy has scant substantiation in the literature, but the data that do exist appear very encouraging. Anecdotal studies by Blumer30 corroborate the utility of antidepressants in low dosages. Blumer frequently used the TCA imipramine in dosages up to 150 mg per day and, for more refractory cases, added SSRIs.30 (Concerns about this combination are discussed later on this page.)

Ojemann et al.33,34 performed two retrospective studies on the use of psychotropics in patients with epilepsy. The first focused on epilepsy patients with depressive symptoms treated with the TCA doxepin. Depressive symptoms decreased by 89%, and seizure frequency improved in 79%, with the suggestion of a link between both effects.33 In the second review, a wide variety of psychiatric syndromes and psychotropic drugs were evaluated but with a similar finding (i.e., 86% improvement and 58% decrease in seizure frequency).34

Robertson and Trimble35 conducted a double-blind placebo-controlled study using amitriptyline and nomifensine. All patients improved somewhat. In the second part of the study, in which all nonresponders were enrolled without a control group and in which antidepressant dosages were increased, a 65% response rate resulted.35

Favale et al.36 used fluoxetine (20 mg per day) to treat 17 patients with epilepsy and symptoms of depression. Of significant interest was the complete remission of seizure activity in 6 patients, with the remaining patients having a 30% reduction in seizure frequency.36 No increase in carbamazepine levels occurred during the study.

Side effects

A wide variety of side effects are associated with the use of antidepressants, depending on the specific group of medications employed.

MAOIs have the most serious restrictions, including dietary and drug interactions, along with hypotension, weight gain, and the like.

The TCAs as a group and the tertiary amines in particular (amitriptyline-imipramine) cause tachycardia, hypotension, dry mouth, and constipation and have a rather low therapeutic index.

SSRIs and SNRIs have a more benign side effect profile. They occasionally cause headache, gastrointestinal effects, akathisia, agitation, hypomania and, possibly, sexual dysfunction, but they are considerably safer in the event of overdosage. 32

Do they induce seizures?

The single most disconcerting factor in the use of antidepressants in patients with concomitant epilepsy is their potential to worsen seizure control. All these medications decrease the seizure threshold to varying degrees. The seizure incidence with various antidepressants is listed in Table: Antidepressants and Seizure Incidence

Methods of measurement

Evaluating the extent of this problem is difficult and extremely complex, given the many methods used to determine the frequency of seizure induction.

One method often used includes recording the incidence of seizures seen in patients after overdose. Extrapolation to the risk in patients whose levels are in the therapeutic range is suspect at best.

Other methods involve the use of animal models and may not allow generalization to humans. For example, maprotiline in animal studies was thought to be relatively safe but, when marketed, was found to be one of the more epileptogenic antidepressant compounds.17

Finally, the base rate of the natural occurrence of a seizure in the general population must be factored into the discussion. From a population-based study in Rochester, Minnesota, the overall incidence of a first seizure was calculated to be approximately 0.086%.37 Any drug considered to be a severe risk for seizure induction would, therefore, have to demonstrate a rate higher than the rate of spontaneous occurrence.

Risk factors

Inclusion of patients exhibiting risk factors for seizure induction can confound the data in assessing the seizure induction potential of antidepressants. These risk factors include:17,37

  • previous history of a seizure
  • central nervous system trauma
  • neoplasia
  • cerebrovascular disease
  • mental retardation
  • dementia
  • drug abuse
  • withdrawal states
  • psychiatric illness (e.g., bulimia, obsessive-compulsive disorder)

The TCAs have an overall seizure induction potential of 0.1–4.0%37 The primary factor involved appears to be serum levels. In the experience of Preskorn and Fast60 and in their review of the literature, they noted that only those TCA levels exceeding the therapeutic range were associated with seizure induction.

Also important is that TCAs are metabolized via the 2D6 cytochrome and that some patients may have congenital deficiencies of this enzyme.32 In this case, serum levels may be elevated. In addition, drugs that inhibit the 2D6 cytochrome (e.g., paroxetine and fluoxetine) will cause TCA serum level increases.32 Thus, serum levels must be followed carefully. The rapidity of dosage escalation is another factor; gradual increase in antidepressant dosages in epileptic patients is recommended.37

Amitriptyline and imipramine have a seizure induction potential (approximately 0.1–0.6%) that is documented to be higher than that of their respective metabolites nortriptyline and desipramine (approximately 0.1%). However, in overdose, the incidence of seizures with the metabolites increases to 22.0% for desipramine and 17.9% for nortriptyline.17,52 As already mentioned, Ojemann's study33 found that seizures were reduced with doxepin.

Choosing an antidepressant

The process of choosing an antidepressant for a patient with concomitant epilepsy must take into consideration three major factors.

The first factor is the age of the patient. The phenomenology of depression in children and adolescents is thought to be similar to that found in adults. However, children appear to respond differently to antidepressants.38,39 TCAs in general have not been shown to ameliorate depressive symptoms in double-blind studies of children, and sudden unexplained death has been reported with the use of desipramine in this patient population.40 Thus, the use of TCAs in children and adolescents is of concern. The SSRIs are showing more promise with efficacy and safety. To date, however, only two of the SSRIs—fluvoxamine and sertraline—have been approved by the U.S. Food and Drug Administration for use in children, and both for symptoms of obsessive-compulsive disorder. In double-blind studies, the data on safety and utility of fluoxetine and paroxetine were positive for symptoms of MDD in children, data that may result in eventual U.S. Food and Drug Administration approval.38–40

The second factor concerns the seizure induction potential of antidepressants. Table: Antidepressants and Seizure Incidence summarizes the clinical reports of relative rates of seizures in patients administered antidepressants.12,41–59

The third factor is possible drug-drug interactions. Table: Interactions: Antidepressants and Antiepileptic Drugs displays the most frequently reported interactions between antidepressants and AEDs.

The SSRIs (fluoxetine, paroxetine, sertraline, fluvoxamine, citalopram) and the SNRI venlafaxine are the drugs of choice for patients with comorbid depression and epilepsy. They have an overall seizure induction potential of between 0.10% and 0.26%.17,37,49 (The rate can be higher in overdosage).61,62 SSRIs appear to be relatively safe, especially considering the aforementioned rate of spontaneous seizure occurrence (0.086%). Fluoxetine, however, may cause more seizure induction than do other SSRIs.

The TCA clomipramine and the tetracyclic antidepressants amoxapine and maprotiline have an unacceptable rate of seizure induction. They are, therefore, considered contraindicated in patients with epilepsy.

Of the other antidepressants available, bupropion remains the most controversial for use in patients with epilepsy. Bupropion provides an excellent example of the critical importance of drug dosage. At dosages exceeding 450 mg per day, the rate was 0.60–2.19%.50 This was reduced to 0.4% with the sustained-release preparation at 400 mg per day and can be decreased even further to 0.1% when 300 mg per day of the sustained-release preparation is used. Given these figures and the variety of antidepressants from which to choose, other alternatives are recommended for patients with epilepsy.

The remaining antidepressants, including trazodone and nefazodone, probably are safe for use in patients with comorbid epilepsy. The MAOIs appear relatively safe.17 Few data are available regarding seizure induction by mirtazapine and reboxetine. Phenytoin and carbamazepine may hypothetically lower mirtazapine levels, and cases of agranulocytosis or neutropenia have been associated with this agent.17 Combining it with another bone marrow–suppressing drug, such as carbamazepine, should be avoided until more data are available.17

Interactions with AEDs

Other concerns regarding the use of antidepressants in patients with epilepsy focus on potential drug interactions with AEDs. Both classes of drugs undergo hepatic metabolism. Biotransformation by the hepatic microsomal cytochrome P450 oxidases appears to be the process most responsible for drug-drug interactions. Antidepressants and AEDs share several common cytochrome P450 enzymes (3A4, 1A2, 2C19, 2C9, and, possibly, 2B6)12,63–68 and also influence the 2D6 cytochrome. Drugs can influence these enzymes by inducing or inhibiting them. The serum levels of medications that act as a substrate for the cytochrome may be decreased if an inducing agent is added, whereas an inhibitory compound has the opposite effect.59

The AEDs most commonly responsible for interactions with antidepressants are the cytochrome-inducing drugs69–79, and the inhibitory agent valproic acid. Clinically significant elevations of these AEDs have been observed when used in combination with fluoxetine, although available data are conflicting.36 Other SSRIs also have been implicated, but less frequently and with less clinical relevance. For this reason, fluoxetine should not be used in combination with carbamazepine, phenytoin, or phenobarbital, or, at the least, serum monitoring is required.

The SSRIs—in particular fluoxetine, paroxetine and, to a lesser extent, sertraline—also interact with TCAs, increasing their levels by as much as threefold.80 Caution and careful monitoring of drug levels are recommended with these combinations.

The serotonin syndrome is a rare disorder that may result from drug interactions.81,82 It is caused by excessive serotoninergic stimulation. Symptoms include:

  • restlessness
  • myoclonus
  • hyperthermia
  • convulsions
  • death (possible)

Most reports have documented the interaction between SSRIs and an MAOI, but similar results have been documented with TCAs, in particular imipramine and clomipramine. Treatment is symptomatic and syndrome resolution usually takes place in 24 hours if the offending agent has been removed.83 Some AEDs have serotonin effects and theoretically can cause this syndrome, as has been reported with carbamazepine.84

Adapted from: Barry JJ and Huynh N. Psychotropic drug use in patients with epilepsy and developmental disabilities. In: Devinsky O and Westbrook LE, eds. Epilepsy and Developmental Disabilities. Boston: Butterworth-Heinemann; 2001;205–217. With permission from Elsevier (www.elsevier.com).

Authored By: 
John J. Barry MD
Nga Huynh PharmD
Reviewed By: 
Steven C. Schachter MD
Monday, May 31, 2004