There is a need for systematic treatment studies that investigate the efficacy of antidepressants in epilepsy patients. In particular, the efficacy of the new antidepressive agents needs confirmation. The prophylaxis of depression in the treatment of epilepsy involves the identification of vulnerable patients (Table 7). Patients at risk for the development of depression are those who have a familial predisposition for affective disorders, or have had previous episodes of depression. Temporal lobe epilepsy and drug-resistant seizures are risk factors, as is a recent history of temporal lobe resection. These patients should ideally be treated with AEDs that have a favorable psychotropic profile, preferably as monotherapy. If potentially depressogenic AEDs are used, these should be titrated slowly, and patients should be monitored closely. However, it is not possible to predict the psychotropic effects of AEDs in individual patients, because drug effects on mood are modulated by the patient's biologic and biographic predisposition. If a patient becomes depressed, it is always necessary to analyze recent changes to the AED regimen, because this may lead to an obvious treatment. In some cases, an effective AED may lead to an "alternative" depression via the mechanism of "forced normalization". Such a depressive reaction does not necessarily require the discontinuation of the responsible AED. Depending on therapeutic alternatives, in patients with difficult-to-treat seizures, the add-on treatment with an antidepressant should be considered. Antidepressants should be used when a depressive episode is severe and long lasting, in particular when there are no promising options of modifying the AED treatment regimen. It is important that antidepressants, when indicated, be used properly, meaning that adequate dosages are prescribed over a sufficiently long period. Antidepressants with low proconvulsive risks are recommended. However, the risk of antidepressants provoking seizures is generally overestimated. Further, antidepressants that are effective in the treatment of depression also may enhance the control of seizures, because patients sleep better and are likely to be more compliant with their AEDs. SSRIs have a relatively low epileptogenic potency. Experimental research using animal models showed that there may be a dose-dependent antiepileptic effect of SSRIs. Other drugs with minor seizure risks are the monoamine oxidase inhibitors, as well as atypical antidepressants such as viloxazine and trazodone. Because of their strong proconvulsive effects, among the classic antidepressants, maprotiline and clomipramine should be avoided. In addition to pharmacodynamic interactions, pharmacokinetic interactions need consideration. For example, with fluoxetine, but not with paroxetine, there is a significant risk of an increase of serum levels of some AEDs. To avoid intoxication with antidepressants (proconvulsive effects are dose dependent) and pseudoresistance due to underdosage (when enzyme-inducing AEDs are prescribed at the same time), or noncompliance (information brochures for patients often warn of seizures), serum drug levels should be checked (of AEDs and antidepressants). Individually, depending on the epileptic syndrome, EEG monitoring may be helpful as an indicator of proconvulsive effects. Tricyclic antidepressants should always be carefully titrated in patients with epilepsy. With the SSRIs, this is not obligatory. However, in patients who have a polytherapy and who are not free from side effects, antidepressants of the new generation also should be introduced slowly. Because of their superior tolerability, SSRIs are generally preferred to the tricyclic drugs. An exception may be patients with a good response to tricyclics in the past. The most important adverse effects of SSRIs are sexual dysfunction and agitation. Alternatively, nefazodone or mirtazapine may be used, which are sedating and have no sexual side effects. In cases of mild depression, St. John's wort extracts may be used. These herbal drugs are effective and have few side effects. However, in a recent study, it was shown that serum levels of some drugs were significantly influenced by St. John's wort. This is important to know, because patients often take these drugs without consulting their epileptologist (Table 8). Of utmost importance for the success of an antidepressive treatment is the information and education of the patient and his relatives. The prerequisite of good compliance with treatment recommendations is the understanding of the nature of depression, and why an antidepressant drug is necessary. Antidepressants are effective only when taken regularly over a long period. Epilepsy patients who do not respond to an antidepressant drug treatment should ideally be seen by a psychiatrist. Pharmacotherapy of a depressive disorder is not regarded as an alternative to professional social support and psychotherapy. Of course, medical and nonmedical treatment strategies should be used complementarily in the management of depression in a person with epilepsy.
