Dr. Levenson is professor in the Departments of Psychiatry, Medicine, and Surgery, chair of the Division of Consultation-Liaison Psychiatry, and vice chair for clinical affairs in the Department of Psychiatry at Virginia Commonwealth University School of Medicine in Richmond.

Disclosure: Dr. Levenson is on the depression advisory board for Eli Lilly.



Important psychiatric issues affecting diagnosis and management arise in patients with neurological illness more often than any other area of medicine. These include cognitive impairment either as a primary feature or a secondary complication of a known neurological disorder; other psychiatric symptoms as a manifestation or complication of neurological disease; and physical neurological symptoms that do not correspond to any recognized pattern of neurological disease, ie, conversion disorder or somatization disorder. In addition, behavioral, cognitive, or emotional symptoms may occur as a complication of drug therapy of neurological disease.1,2 In previous columns, psychiatric issues in stroke3 and Parkinson’s disease and multiple sclerosis4 were reviewed. This column reviews psychiatric issues related to epilepsy.



Epileptic seizures are the result of transient cerebral dysfunction caused by abnormal electrical activity in the brain, presenting as sudden recurring attacks of motor, sensory, or psychic manifestations with or without loss of consciousness or generalized convulsions. Consequently, psychiatrists must consider epilepsy when determining whether psychiatric symptoms are due to epilepsy, when treating psychiatric complications of epilepsy or when treating psychiatric complications of epilepsy with anticonvulsants, and when prescribing psychiatric medication that may adversely affect epilepsy or interact with anticonvulsants.

In developed countries most cases of epilepsy are idiopathic; specific etiologies include include perinatal trauma, head trauma, central nervous system (CNS) infection, CNS degenerative disorders, cerebrovascular disease, brain tumors, and substance misuse. Epilepsy is more common in developing nations, because of  increased rates of birth trauma and head injury, lack of health services,  high rates of  alcohol and substance misuse, and  poor sanitation leading to high rates of CNS infection (eg, neurocysticercosis which is the leading cause of seizures in adults in endemic areas).  In some benign childhood seizures, an anticonvulsant is unnecessary, but most patients with epilepsy require treatment with anticonvulsants which  sometimes can eventually be withdrawn and sometimes must be continued indefinitely. In approximately one-third of patients with epilepsy, anticonvulsants fail to achieve adequate control.


Clinical Features

Epilepsy is heterogeneous in etiology and in its clinical features, but an individual patient’s seizures are usually stereotypical. The key clinical distinction is between focal and generalized seizures. Tonic-clonic seizures usually begin with no warning and are characterized by sudden loss of consciousness and dramatic motor activity (tonic, ie, sustained muscle contractions lasting approximately 10–20 seconds, followed by clonic, ie, repetitive muscle contractions lasting approximately 30 seconds). Autonomic changes may include an increase in blood pressure and pulse rate, apnea, mydriaisis, incontinence, piloerection, cyanosis and perspiration. In the post ictal period the patient is drowsy and confused and abnormal neurological signs are often present.

Partial seizures may be simple (without impairment in consciousness) or complex (with impairment of consciousness). Simple partial seizures may be manifest in many different ways, including focal muscle contractions, somatoensory experiences (eg, numbness, paresthesias), vertigo, visual disturbances (eg, micropsia, macropsia, and visual hallucinations), other hallucinations (eg, auditory, olfactory, gustatoryl, and tactile), language disturbance, emotional outbursts, unpleasant, epigastric sensations, or motor automatisms such as bicycling or sexual movements and vocalization.

In complex partial seizures, the patient frequently experiences an aura preceding the seizure lasting seconds to minutes. The content of the aura may consist of  hallucinations; intense affective symptoms such as fear, depression, or depersonalization;  cognitive dysfunction;  dreamy states, flashbacks and distortions of familiarity with events (déjà vu or jamais vu).1 A prodrome of nervousness or irritability may begin hours or even days before a seizure. This aura is followed by disturbance of consciousness and a seizure usually lasting 60–90 seconds, which may or may not generalize into a tonic-clonic seizure. Automatisms may occur such as chewing, swallowing, lip smacking, grimacing, fumbling with objects, walking or trying to stand up. Post-ictal confusion typically lasts >10 minutes.

Absence seizures are abrupt, brief episodes of decreased awareness which occur without any warning, aura or post-ictal symptoms. A simple absence seizure is characterized by only an alteration in consciousness of around 15 seconds, ending abruptly with the patient resuming previous activity, often unaware that a seizure has occurred. A complex absence seizure includes additional signs such as change in postural tone, minor clonic movements, minor automatisms, or autonomic symptoms.


Differential Diagnosis

The differential diagnosis of epilepsy includes syncope, psychogenic spells associated with several different psychiatric diagnoses, transient ischemic attacks, arrhythmias, recurrent pulmonary emboli, hypoglycemia, cataplexy, acute dystonia, and other paroxysmal disorders. For attacks occurring only during sleep, night terrors, rapid eye movement sleep behavior disorder, periodic limb movements, sleepwalking, and other parasomnias should also be considered.

The most common problem in the differential diagnosis of epilepsy is its distinction from psychogenic spells. The term “pseudoseizures” is unfortunate and misleading. It is typically assigned to patients whose seizures’ phenomenology is not consistent with epilepsy and/or have been demonstrated to occur without epileptiform activity on the simultaneous EEG recording during video-EEG monitoring. Thus, “pseudoseizures” is not a diagnosis, but merely designates “not-epilepsy,” and not all “pseudoseizures” are psychogenic. The “pseudo-“ prefix carries a pejorative connotation that the patient’s symptoms are illegitimate. Prigatano and colleagues5 found that reports by health care providers that patient’s seizures were not “real” (ie, true epilepsy) restimulated feelings associated with their not being believed when they reported being sexually abused as children.

Seizure-like spells may occur as part of many psychiatric diagnoses including conversion disorder, panic disorder, hyperventilation syndrome, somatization disorder, posttraumatic stress disorder, dissociative disorders, and mental retardation, as well as in factitious disorder6 and malingering.7 The presence of confirmed epilepsy does not rule out the presence of psychogenic spells as well; it is not unusual for a patient to have both. That a spell appears to be precipitated by hyperventilation does not establish the etiology as psychogenic, because hyperventilation can induce seizures in person with epilepsy. In fact, hyperventilation has been long uitlized as a method to provoke epileptiform activity during diagnostic electroencephalography (EEG).

Distinguishing psychogenic spells from epilepsy  can often be made on the basis of a careful history and examination. Clinical clues include other symptoms and signs of  psychiatric disorders, atypical seizure phenomenology, especially the occurrence of frequent and prolonged seizures in the face of normal interictal intellectual function and EEG; seizures that mainly occur and are witnessed in medical settings  and never alone; and preservation of awareness  during an apparent generalized seizure (eg, resistance to attempted eye opening). It is widely believed that tongue-biting, urinary incontinence, and injury from seizures are diagnostic of epilepsy, but this is fallacious. A survey8 of 102 consecutive patients diagnosed with psychogenic seizures by video-EEG monitoring revealed that during typical attacks of psychogenic seizures, 40% reported injuries, 44% reporting tongue biting, and 44% reported urinary incontinence. A history of previous head injury is frequent in both epilepsy and psychogenic spells. Previous childhood sexual abuse is very common in patients with psychogenic seizures but not always present9 and not distinctive since childhood abuse is common in the general population, Rosenberg and colleagues10 found that histories of sexual and physical abuse, other traumas, and PTSD were common in patients with intractable seizures, both in those with epilepsy and those with psychogenic seizures, albeit more frequent in the latter.

The gold standard for diagnosis remains observation of attacks during video-EEG recording. A normal EEG during or immediately after an apparent generalized seizure provides strong evidence that the patient’s seizure is not epileptic, but making a specific psychiatric diagnosis requires identifying positive psychiatric evidence as well.


Epilepsy and Psychosis

Psychotic symptoms may be coincident with seizures when both are the result of brain disease or injury, especially with subcortical or temporal lobe lesions. Examples of acute causes of psychosis and seizures include encephalitis, CNS vasculitis, alcohol withdrawal, hyponatremia, and drug toxicity (eg, lidocaine, cocaine). But psychotic symptoms may also be a consequence of some forms of epilepsy, particularly complex partial seizures. Psychotic symptoms can occur ictally, postictally, or interictally. Brief psychotic symptoms can occur in  nonconvulsive status epilepticus, most commonly with partial complex status.11 In such cases, other features of complex partial seizures may be present as well, such as automatisms (eg, lip smacking, picking at clothes), mutism, altered consciousness, or amnesia. Postictal psychosis follows an increase in the frequency of seizures, usually with a nonpsychotic period of 1–7 days between the last seizure and the psychosis. Manic grandiosity and religious and mystical features are often present in postictal psychosis and it often resolves within a few days. Chronic interictal psychosis can occur even in the absence of frequent seizures, but usually occurs in patients with poorly controlled seizures. It is often schizophreniform including auditory hallucinations, and is usually self-limiting but can last for a few weeks. Unlike postictal psychosis, interictal psychosis is sometimes ameliorated by the occurrence of one or more seizures. Chronic interictal psychosis differs from schizophrenia in having better preservation of affect, and by mood swings, mystical experiences, and visual hallucinations.

Psychosis can also be an iatrogenic consequence of the treatment of epilepsy. Psychosis is a potential side effect of anticonvulsants, most frequently with levetiracetam and topiramate, but also with phenytoin, valproate, lamotrigine, zonisamide, pregabalin, and vigabatrin. One suggested mechanism is that control of seizures causes psychotic symptoms through “forced normalization.”12 Abrupt discontinuation of anticonvulsants can also cause acute psychosis. Finally, temporal lobectomy for medically intractable epilepsy may precipitate a schizophrenia-like psychosis. A retrospective study13 found this occurring in 11 of 320 patients, with those who had bilateral functional and structural abnormalities, particularly of the amygdala, at particular risk for the development of such psychoses.


Epilepsy and Depression

Depression is very common in patients with epilepsy, with its lifetime prevalence estimated between 6% and 30% in population-based studies and up to 50% among epilepsy patients in tertiary centers.14 The etiology of  depression in epilepsy is multifactorial, with neurobiological, psychological, social, and iatrogenic factors all relevant.1,15,16  The risk of depression is greater in patients with high seizure frequency and symptomatic focal epilepsy,15 especially complex partial seizures. The stress of living with a stigmatized chronic illness can  have profound negative effects on health-related quality of life. Learned helplessness giving rise to depression may occur in patients with epilepsy as a result of the unpredictability and unavoidability of seizures, further exacerbated by occupational disruption and losing driving privileges until seizure-free for an extended period. Anticonvulsants can be a cause of depression as well. The relationship between depression and epilepsy is bi-directional (ie, each is a risk factor for the other). In patients with epilepsy, depression may be a stronger predictor of health-related quality of life, than seizure frequency and severity, employment, or driving status.17 Finally, for patients with intractable epilepsy, comorbid depression may improve with vagal nerve stimulation.


Epilepsy and Anxiety

Similarly anxiety in epilepsy is very common and best considered as multifactorial in origin. Preexisting vulnerability, neurobiological factors including seizure focus location, iatrogenic influences (anticonvulsants, epilepsy surgery), and psychosocial factors are all likely to play a role, but with considerable individual differences The prevalence of anxiety in a community sample of adults with epilepsy was 20.5% and was associated with a current history of depression, perceived side effects of antiepileptic medication, lower educational attainment, chronic ill health, female gender, and unemployment, but was not associated with the duration of epilepsy.18 While the relationship between epilepsy and depression has received much attention, less is known about anxiety disorders in epilepsy. Anxiety can have a profound influence on the quality of life of patients with epilepsy. Anxiety in epilepsy may be ictal, postictal, or interictal.19 In particular, anticipatory anxiety about having a seizure, in the absence of a warning, can lead to agoraphobic-like symptoms and behavior, to avoid embarrassment, shame, inconvenience, and stigma.


Epilepsy and Violent Behavior

As noted, complex partial seizures may cause emotional symptoms and automatic motor behavior and this can very occasionally result in undirected violent behavior. However, in the overwhelming majority of cases this is in response to being restrained during a seizure. One should be very cautious before attributing other violent assaults to a seizure. True ictal violence is rare, and most cases are characterized by spontaneous, non-directed, stereotyped aggressive behaviors.  One should be very cautious before attributing violence to a seizure. Characteristics of ictal violence include: the seizure episode is sudden, without provocation, and lasts at most a few minutes; automatisms and other stereotypic phenomena of the patient’s typical seizures accompany the aggressive act, and the act is associated with these phenomena from one seizure to the next; the patient’s consciousness is impaired;  the behavior is poorly directed and is unskilled; purpose and interpersonal interaction are absent.20 To confirm that violent behavior is attributable to a seizure disorder requires documenting aggression during epileptic automatisms during video-EEG monitoring.  Non-seizure EEG abnormalities (such as sharp waves) are non-specific findings and should not be used as evidence that violence is ictal. 


Psychotropic Drugs and Seizure Risk

Patients with epilepsy are frequently prescribed psychotropics due to the high psychiatric comorbidity described above. Many psychotropics, especially antidepressants and antipsychotics, can lower the seizure threshold.21 Controlled studies of seizure frequency with individual psychotropics in psychiatric populations without epilepsy are infrequent, and nonexistent in patients with comorbid psychiatric disorders and epilepsy. Consequently, case reports form a large part of the available literature, so estimates of the frequency of psychotropic drugs causing seizures and aggravating epilepsy are far from precise. In general, if a patient’s epileptic seizures are well-controlled on anticonvulsants, most psychotropic drugs can be used without significant increased risk. The risk of seizure induced by a psychotropic is increased in patients with treatment-resistant epilepsy, concurrent use of other drugs that lower the seizure threshold, electrolyte and other metabolic derangements, and  blood levels that rise too high because of rapid dose titration, slow metabolism, or and drug-drug interactions.22

How much risk for increased seizures do antidepressants pose? Antidepressants at therapeutic doses in non-epileptic patients exhibit a seizure risk close to that reported for the first spontaneous seizure in the general population (≤0.1%).  The risk of seizures with bupropion SR is low at doses ≤450 mg/day. The risk of seizures with bupropion has been overstated in many sources; a systematic review concluded the risk was lower than that associated with tricyclic antidepressants and phenothiazines.23

Both typical and atypical antipsychotics can lower the seizure threshold, increasing the chances of seizure induction. Of the typical antipsychotics, chlorpromazine appears to be associated with the greatest risk of seizure provocation, while high potency typical antipsychotics like haloperidol and fluphenazine are associated with a lower risk. Among the atypical antipsychotics, clozapine is the most frequently associated with seizures, with a risk of approximately 1% to 2%. Consequently, low-potency typical antipsychotics and clozapine ideally should be avoided in patients with epilepsy.24
Cholinomimetics may also reduce the seizure threshold. While many physicians believe that psychostimulants lower seizure threshold, evidence for this is lacking. Finally, many other psychiatric drugs are potent anticonvulsants (benzodiazepines, mood stabilizers other than lithium).


Electroconvulsive Therapy and Epilepsy

Electroconvulsive therapy (ECT) has anticonvulsant activity, as indicated by a progressive increase in seizure threshold and decrease in seizure length during the course of ECT treatment, but this effect is  too short-lived to make it an option for treating intractable epilepsy. Rarely, ECT has induced status epilepticus, but there is no evidence that spontaneous seizure frequency increases with ECT in epileptic patients.25

A key clinical question is how to treat the patient whose psychiatric disorder requires ECT but who also has epilepsy, ie, how to elicit therapeutic seizures in the face of concomitant treatment with anticonvulsants. While there is surprisingly little published literature addressing this issue, experts suggest that most epileptic patients can be successfully treated with ECT without having to alter their anticonvulsant regimen. For those who either do not obtain seizures or have extremely short ones, a number of techniques are recommended to consider, after consultation with a neurologist.25 It is not known whether various anticonvulsants differentially affect seizure threshold and duration in ECT. PP



1.     Carson AJ, Zeman A, Myles L Sharpe MC. Neurology and neurosurgery. In: Levenson, JL, ed. American Psychiatric Publishing Textbook of Psychosomatic Medicine. Washington, DC: American Psychiatric Publishing; 2005:701-732.
2.     Carson AJ, Zeman A, Myles L Sharpe MC. Neurology and Neurosurgery. In: Levenson JL, ed. Essentials of Psychosomatic Medicine. Washington, DC: American Psychiatric Publishing; 2007;313-342.
3.     Levenson JL: Psychiatric issues in Neurology, Part I: Stroke. Primary Psychiatry. 2007;14(9):37-40.
4.     Levenson JL: Psychiatric Issues in Neurology, Part 2: Parkinson’s disease and multiple sclerosis. Primary Psychiatry. 2007;14(11):35-39.
5.     Prigatano GP, Stonnington CM, Fisher RS. Psychological factors in the genesis and management of nonepileptic seizures: clinical observations. Epilepsy Behav. 2002;3(4):343-349.
6.     Savard G, Andermann F, Teitelbaum J, Lehmann H. Epileptic Munchausen’s syndrome: a form of pseudoseizures distinct from hysteria and malingering. Neurology. 1988;38(10):1628-1629.
7.     Beaumont G. Is it epilepsy? J Forensic Leg Med. 2007;14(2):99-102.
8.     Peguero E, Abou-Khalil B, Fakhoury T, Mathews G. Self-injury and incontinence in psychogenic seizures.Epilepsia. 1995;36(6):586-591.
9.     Binzer M, Stone J, Sharpe M. Recent onset pseudoseizures–clues to aetiology. Seizure. 2004;13(3):146-155.
10. Rosenberg HJ, Rosenberg SD, Williamson PD, Wolford GL 2nd. A comparative study of trauma and posttraumatic stress disorder prevalence in epilepsy patients and psychogenic nonepileptic seizure patients.Epilepsia. 2000;41(4):447-452.
11. Sachdev PS. Alternating and postictal psychoses: review and a unifying hypothesis. Schizophr Bull. 2007;33(4):1029-1037.
12.     Akanuma N, Kanemoto K, Adachi N, Kawasaki J, Ito M, Onuma T. Prolonged postictal psychosis with forced normalization (Landolt) in temporal lobe epilepsy. Epilepsy Behav. 2005;6(3):456-459.
13.    Shaw P, Mellers J, Henderson M, Polkey C, David AS, Toone BK. Schizophrenia-like psychosis arising de novo following a temporal lobectomy: timing and risk factors. J Neurol Neurosurg Psychiatry. 2004;75(7):1003-1008.
14.     Kanner AM. Depression in epilepsy: prevalence, clinical semiology, pathogenic mechanisms, and treatment. Biol Psychiatry. 2003;54(3):388-398.
15.     Kimiskidis VK, Triantafyllou NI, Kararizou E, et al. Depression and anxiety in epilepsy: the association with demographic and seizure-related variables. Ann Gen Psychiatry. 2007;6(1):28.
16.     Seethalakshmi R, Krishnamoorthy ES. Depression in epilepsy: phenomenology, diagnosis and management. Epileptic Disord. 2007;9(1):1-10.
17.     Gilliam F, Hecimovic H, Sheline Y. Psychiatric comorbidity, health, and function in epilepsy. Epilepsy Behav. 2003;4(suppl 4):S26-S30.
18.     Mensah SA, Beavis JM, Thapar AK, Kerr MP. A community study of the presence of anxiety disorder in people with epilepsy. Epilepsy Behav. 2007;11(1):118-124.
19.    Beyenburg S, Mitchell AJ, Schmidt D, Elger CE, Reuber M. Anxiety in patients with epilepsy: systematic review and suggestions for clinical management. Epilepsy Behav. 2005;7(2):161-171.
20.     Marsh L, Krauss GL. Aggression and violence in patients with epilepsy. Epilepsy Behav. 2000;1(3):160-168.
21.     Mula M, Monaco F, Trimble MR. Use of psychotropic drugs in patients with epilepsy: interactions and seizure risk. Expert Rev Neurother. 2004;4(6):953-964.
22.     Hedges D, Jeppson K, Whitehead P. Antipsychotic medication and seizures: a review. Drugs Today (Barc). 2003;39(7):551-557.
23.     Ruffmann C, Bogliun G, Beghi E. Epileptogenic drugs: a systematic review. Expert Rev Neurother. 2006;6(4):575-589.
24.     Alldredge BK. Seizure risk associated with psychotropic drugs: clinical and pharmacokinetic considerations. Neurology. 1999;53:S68-S75.
25.     Rasmussen KG, Rummans TA, Tsang TSM, Barnes RD. Electroconvulsive therapy. In: Levenson JL, ed. American Psychiatric Publishing Textbook of Psychosomatic Medicine. Washington, DC: American Psychiatric Publishing; 2005:957-978.