When Dietrich Blumer, MD, attended medical school in Zurich in the 1950s, the curriculum included extensive teaching of psychiatric disorders associated with epilepsy. So when he came to the United States, in 1957, he was "absolutely shocked" that psychiatrists didn't play a routine role in the care of epileptic patients. Today, the situation is better—but not much. Over the past 13 years, Dr. Blumer, who is Professor of Psychiatry at the University of Tennessee College of Medicine, has evaluated and treated at the Memphis epilepsy center every patient who had neuropsychiatric problems. "It's been wonderful," he says, but he wishes there were more than a handful of other epilepsy programs that have made a similar commitment to psychiatric care of this population. In addition to his clinical work, Dr. Blumer has authored dozens of papers and was editor of the 1984 text, Psychiatric Aspects of Epilepsy(American Psychiatric Press).

NPR: As you and others have noted, psychiatric problems are very common in epilepsy patients. Are these problems being adequately recognized and treated?

DB: No, there is terrible neglect. I asked a very fine neurologist who directs a university epilepsy center, "What do you do with the psychiatric problems in the epileptic population?" He said, "I don't know." He has, of course, a social worker and he certainly has a neuropsychologist. Neuropsychologists are very important for the cognitive assessments of patients with epilepsy, especially for the surgical candidates. But they are not familiar with the intermittent psychiatric problems of the patients and they don't treat with medication. If the patients have severe psychiatric problems, they may be referred to psychiatrists who are not trained to carry out modern treatment for these patients.

We have sort of rediscovered what the old psychiatrists knew: Patients with epilepsy have intermittent and pleomorphic affective changes. In chronic epilepsy, more than 50% of the patients have some significant dysphoria. But these are missed with neuropsychologic assessments that are cross-sectional analyses. Yes, you may document some depressive traits, but you need to get a history of events to identify the specific psychiatric disorders of epilepsy that need to be treated.

NPR: Why are psychiatrists so uninterested in epilepsy?

DB: It's not taught during their training, and they feel it's the province of neurology. They feel incompetent; moreover, they are afraid they might worsen the seizure condition with psychotropic drugs. Here in Memphis, our psychiatry residents are all trained to recognize and treat the psychiatric problems of patients with epilepsy and they become quite competent in dealing with these problems. But that's probably rather unique.

For years, when everything was psychoanalytically and psychosocially explained, the belief was that epileptics only have troubles because people are not accepting of them—it's stigmatizing to have seizures in public, they can't drive, and all of that. These are important considerations, of course. But the major problem of patients with chronic epilepsy is of a neuropsychiatric nature, most importantly in the form of an interictal dysphoric disorder. The patients not only have depressive moods, but also are intermittently, for hours or days, irritable, anxious, listless, fearful of certain situations; they have headaches or other pains, insomnia, and may experience brief euphoric moods. If at least three of the eight symptoms are present, each to a bothersome degree, the diagnosis of interictal dysphoric disorder can be made. The disturbance cannot be explained as psychosocial impairment.

NPR: Do you treat dysphoria in this population the same way you would in a nonepileptic population?

DB: It is treated by the addition of modest amounts of antidepressant. The antidepressants are still often shunned in epilepsy patients because of fears that they lower the seizure threshold and could cause more seizures. We have seen patients who come in after repeated suicide attempts, and nobody has ever put them on an antidepressant.

However, we have never seen a seizure breakthrough because of the use of an antidepressant. We use modest amounts of a tricyclic first, such as 100 to 150 mg of imipramine or nortriptyline. This helps the entire range of the symptoms of the dysphoric disorder, including the insomnia. It's very important that the patients can sleep well. And if that doesn't work, we add an SSRI [selective serotonin reuptake inhibitor]. My colleague at Stanford, John Barry [MD], treats with the SSRI first and then he adds the tricyclic if necessary. So you can do it either way. But the antidepressants are most effective for the interictal dysphoric disorder and even for the psychosis.

If that doesn't work, we add a small amount of neuroleptic, like risperidone. And we have been very pleased about how this works.

NPR: You alluded to interictal psychosis. How do the manifestations of this psychosis differ from other psychotic disorders?

DB: People have said that they could not be differentiated from schizophrenic psychosis, but that's really not true. For one thing, the baseline is very different—the persons affected are not schizoid. They have a lot of affect and still like to shake hands with you. They can have delusions and hallucinations, as we see in schizophrenics, but the basic personality differs vastly. And almost invariably, there are heightened affective changes. The psychotic state tends to develop out of a dysphoric disorder. There are a few patients who don't have much of a dysphoric disorder in the first place and just present with the psychosis; but that's by and large the exception.

The concept of forced normalization posits that psychiatric changes tend to occur when epilepsy is treated successfully. We presume that inhibitory mechanisms whose precise nature is still not clarified develop in the course of chronic epilepsy and may result in dysphoric and finally even psychotic symptoms.

NPR: Do you consider forced normalization inevitable? Are there any treatment approaches that can minimize the risk of this happening?

DB: We presented data at the 1999 American Epilepsy Society meeting showing that 9% of patients treated with vagus nerve stimulation had severe psychiatric complications that invariably occurred when seizures that had been very frequent stopped, or were at least reduced by 75%. Six patients became severely dysphoric—five with episodes of catastrophic rage and four with psychotic symptoms; one developed a psychotic episode. All five suicides from our epilepsy center over the past 13 years occurred when a very long-standing seizure disorder was finally controlled, by surgery, medication, or vagus nerve stimulation.

NPR: So this occurs regardless of modality?

DB: Yes. Michael Trimble [MD, FRCP, of the Institute of Neurology, University College of London] has written about the new, more effective anticonvulsants, pointing out that psychotic and depressive episodes are associated with their use. And it seems plausible that this has to do with the suppression of seizures. In other words, it seems that seizure inhibition may have a psychotoxic effect. We don't know why this happens in some patients and not in others with similar chronicity of the epilepsy.

NPR: One might assume that neuroleptics would be the appropriate treatment for these psychoses, but you've suggested that antidepressants are more effective.

DB: Yes, we use the same treatment as for the dysphoric disorder. If the dysphoric disorder is severe, we already add an atypical neuroleptic to the double antidepressant medication. If a patient is psychotic or suicidal, we promptly treat with two antidepressants, then add a small amount of the neuroleptic if necessary. In premodern times, the interictal psychoses were viewed as an expansion of dysphoric disorder, so it wasn't just our idea. We merely rediscovered it.

NPR: I've seen speculation that there is a relationship between epilepsy and panic disorder. Do you think there is a neurobiological connection?

DB: I think the mechanisms are different. Of course, both are highly intermittent, as is the dysphoric disorder. But they're quite different.

I'll tell you where we see dysphoria: in PMS [premenstrual syndrome]. They gave it the same name—premenstrual dysphoric disorder. And the profile of symptoms is exactly like that of the interictal dysphoric disorder. It makes sense, because the premenstrual period is a time that favors paroxysmal events. For instance, more than two thirds of women with epilepsy have most or all of their seizures in the premenstrual period.

I wrote an article on that with Andy Herzog [MD, MSc] of Harvard. He had pointed out that during the menstrual cycle there are changes in the ratio of progesterone and estrogen. Progesterone is protective against seizures, and estrogen favors seizure activity. And premenstrually, the ratio is changed in favor of estrogen, facilitating seizures and dysphoric symptoms in women with epilepsy, and dysphoric symptoms—ie, PMS—in women without epilepsy. And we proposed in that article that PMS should be treated like the interictal dysphoric disorder—with the combination of an antiepileptic and an antidepressant, not just with the latter.

The SSRIs are effective to a significant degree in 60% of the women with premenstrual dysphoric disorder. But they often lose their effectiveness after 10 to 12 months. If you combine the SSRI with an antiepileptic drug, this works wonderfully. We now have a series of 50 cases that we haven't published yet.

NPR: What issues in the psychiatric treatment of epilepsy would you most like to see addressed in future research?

DB: The prevalence of epilepsy grows over the years and is highest in old people, but we really don't know about their associated psychiatric problems, which surely must be there. So we need to see what happens in older people. There's also the interesting overlap between epileptic seizures and nonepileptic seizures; so many patients with epilepsy have both.

NPR: I didn't realize that there was so much overlap between the two types.

DB: My next major paper will focus on the problem of the so-called pseudoseizures. In one series, of over 60 patients, one third had both types of seizures. But that's fairly well recognized; other people have looked at this and come up with the same finding.

NPR: What do you make of the phenomenon?

DB: Orrin Devinsky [MD] reported patients in whom pseudoseizures actually ended up in an epileptic event. We had several patients who developed pseudoseizures after surgery, and they all had the dysphoric disorder. The presence of epilepsy with a dysphoric disorder seems to make patients more vulnerable to have the nonepileptic seizures. For patients with nonepileptic seizures, we use antidepressants combined with propranolol or with a benzodiazepine. A majority have suffered the pain of severe abuse in the past, and these patients benefit from good psychotherapy. It is no surprise that the majority of nonepileptic seizures tend to be preceded, often in a crescendo, by a prodrome of pain.

NPR: Have the newer anticonvulsants, such as lamotrigine or gabapentin, had any impact on psychiatric symptoms of epilepsy?

DB: Well, as I mentioned, Dr. Trimble has shown how vigabatrin and other newer anticonvulsants may bring about psychiatric disorders. He looked particularly at the emergence of psychosis when seizures are suppressed.

We've seen the same thing. Topiramate, for instance, is a very interesting drug. But it may be so potent in seizure suppression that patients may become psychotic. And the drug may have to be discontinued.

The crucial point is that once you suppress the seizures, inhibition predominates and then you see the psychiatric complications. We had a patient who became psychotic with topiramate; we took him off, and he became psychotic with gabapentin and finally with valproate. I still see him, he got married and holds a good job but he has one to two complex partial seizures per week. We can't suppress them all; otherwise he's psychotic. But he functions very well.

Suggested Reading
1. Blumer, D. Antidepressant and double antidepressant treatment for the affective disorder of epilepsy. J Clin Psychiatry.1997;58:3-11.
2. Blumer D. Dysphoric disorders and paroxysmal affects: recognition and treatment of epilepsy-related psychiatric disorders. Harv Rev Psychiatry.2000;8:8-17.
3. Blumer D, Herzog AG, Himmelhoch J, et al. To what extent do premenstrual and interictal dysphoric disorder overlap? Significance for therapy. J Affect Disord.1998;48:215-225.
4. Blumer D, Wakhlu S, Montouris G, Wyler AR. Treatment of the interictal psychoses. J Clin Psychiatry.2000;61:110-122.

Return to table of contents