Psychiatric Disorders Associated With Epilepsy

Updated: Jul 27, 2022
  • Author: Janet K Lee, MD; Chief Editor: Jose E Cavazos, MD, PhD, FAAN, FANA, FACNS, FAES  more...
  • Print


The International League Against Epilepsy (ILAE) and the International Bureau for Epilepsy (IBE) define epilepsy as a disorder of the brain characterized by an enduring predisposition to generate epileptic seizures and by the biologic, cognitive, psychological, and social consequences of this condition. This association may reflect the anatomical and neurobiological source of both epileptic seizures and the behavioral manifestations.

Antiseizure medications (ASMs) can play a role in the genesis of psychiatric symptoms; alternatively, some psychotropic medications can lower the seizure threshold and provoke epileptic seizures.

Indeed, there is a general agreement that the incidence of neurobehavioral disorders is higher in patients with epilepsy than in the general population. At any given time, the higher prevalence of neurobehavioral disorders may largely be accounted for by patients with medically intractable epilepsy. Recent population-based studies suggest there is likely a bidirectional relationship between epilepsy and certain neurobehavioral disorders as patients with primary psychiatric disorders have a greater risk of developing epilepsy. [107] Many, but not all, authors also accept the proposition that the link between neurobehavioral disorders and temporal lobe or complex partial epilepsy is particularly strong.

Go to Epilepsy and Seizures for an overview of this topic. Additionally, go to Psychogenic Nonepileptic Seizures for complete information on this topic.

Factors in the relationship between epilepsy and behavioral disorders

Mechanisms for a relationship between epilepsy and behavioral disorders include the following:

  • Common neuropathology
  • Genetic predisposition
  • Developmental disturbance
  • Ictal neurophysiologic effects
  • Inhibition or hypometabolism surrounding the epileptic focus
  • Secondary epileptogenesis
  • Alteration of receptor sensitivity
  • Secondary endocrinologic alterations
  • Primary, independent psychiatric illness
  • Consequence of medical or surgical treatment
  • Consequence of psychosocial burden of epilepsy

Multiple interacting biologic and psychosocial factors determine the risk for the development of either schizophreniform psychoses or major depression in patients with epilepsy, and behavioral disorders in epilepsy have multiple risk factors and multifactorial etiologies. [1]

Role of the neurologist in the psychiatric management of patients with epilepsy

As neurologists, we tend to focus on seizure control, and psychiatric comorbidities are often underestimated. Recognizing psychiatric manifestations is an area that needs improvement. Once symptoms are identified, the following questions arise: [2]

  • Are the symptoms related to the occurrence of seizures (preictal, ictal, postictal)?
  • Are the symptoms related to ASMs?
  • Is the onset of symptoms associated with the remission of seizures in patients who had previously failed to respond to ASMs?

The close association between epilepsy and psychiatry has a long history. The traditional approach to epilepsy care has been to focus on the seizures and their treatment. Concentrating only on the treatment of the seizures, which occupy only a small proportion of the patient's life, does not address many of the issues that have an adverse impact on the quality of life of the patient with epilepsy.

Sackellares and Berent stated that comprehensive care of the epileptic patient requires "attention to the psychological and social consequences of epilepsy as well as to the control of seizures." [3]

Although undoubtedly important in the care of the patient with epilepsy, advances in neurologic diagnosis and treatment tended to obscure the behavioral manifestations of epilepsy until Gibbs drew attention to the high incidence of behavioral disorders in patients with temporal lobe epilepsy. [4]

Frequency of psychiatric disorders in patients with epilepsy

It is estimated that 20–30% of patients with epilepsy have psychiatric disturbances. [5]

Of patients with intractable complex partial seizures, 70% may have 1 or more diagnoses consistent with the Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition (DSM-III-R); 58% of these patients have a history of depressive episodes, 32% have agoraphobia without panic or other anxiety disorder, and 13% have psychoses. [6]

The risk of psychosis in patients with epilepsy may be 6–12 times that of the general population, with a prevalence of about 7–8%; in patients with treatment-refractory temporal lobe epilepsy, the prevalence has been reported to range from 0% to 16%. [7, 108, 109]

Differences in the rates may result from differences in the populations studied, time periods investigated, and diagnostic criteria.

The most common psychiatric conditions in epilepsy are depression, anxiety, and psychoses. [8, 9, 10, 11, 12, 13]  (See the Table below.)

Table. (Open Table in a new window)

Psychiatric Disorder


Patients With Epilepsy

Major depressive disorder



Anxiety disorder



Mood/anxiety disorder



Suicidal Ideation






Table. Prevalence Rates of Psychiatric Disorders in Patients With Epilepsy and the General Population (2007 data) [8]

The psychiatric symptoms characteristic of the neurobehavioral syndrome of epilepsy (ie, Morel syndrome) tend to be distinguished in the following ways:

  • Atypical for the psychiatric disorder
  • Episodic
  • Pleomorphic

Psychotic Disorders

Psychotic disorders are severe mental disorders that cause abnormal thinking and perception. Individuals who experience psychosis lose relation with reality. Symptoms are described as either positive, such as hallucinations, delusions, and disorganized behaviors, or as negative, such as diminished range of emotion, reduced speech, and inability to initiate and sustain goal-directed activities.

The lifetime prevalence of psychotic disorder in the general population is 3% where studies have shown this number is higher in patients with epilepsy.

An early study performed by Perez and Trimble reported that about half of epileptic patients with psychosis could be diagnosed with schizophrenia. [15]  More recent studies bring up the challenges of comparing psychosis in epilepsy and schizophrenia. Regardless, likely a bidirectional relationship exists between epilepsy and psychosis, as patients with schizophrenia have a higher incidence of epilepsy and the incidence of schizophrenia is higher in patients with epilepsy. [110]

Neuroanatomy and neurobiology of psychosis of epilepsy

The etiology and pathogenesis of psychosis in epilepsy remains poorly understood. However, the neurotransmitter dopamine has been proven to have a clear role in schizophrenia and psychosis. Additionally, neuroanatomical changes have also been observed in patients with psychosis and include the following:

  • Asymmetry of amygdala and anterior segment of the hippocampus [16]
  • Rule of hippocampal-amygdala complex in pathogenesis of schizophrenia [17]
  • Smaller gray matter volume in the left and middle temporal gyri and left posterior superior temporal gyrus [18]
  • Rule of bilateral middle frontal gyrus (prefrontal cortex) in overt psychosis occurring with schizophrenia [19]
  • Superior temporal cortex and dysfunction of corollary discharges in auditory hallucination [20]

In studies looking at rat models, dopamine has been implicated as a factor in the pathogenesis of epilepsy. Al-Tajir et al showed dopamine activity likely has both antiepileptic and proconvulsant properties. While dopamine activity appears to be antiepileptic in the forebrain, reduced D1 receptor activation can lower the seizure threshold and promote seizure activity. [111]

This suggests a likely role of dopamine in the pathogenesis of seizures and psychosis. Additional studies have looked at the structural underpinnings of psychosis of epilepsy. Studies have shown that:

  • Patients with psychosis of epilepsy had larger hippocampal fissures bilaterally and no significant differences in the volume of hippocampal regions. [112]
  • Patients with temporal lobe epilepsy and psychosis of epilepsy have significantly smaller brain volume than people with temporal lobe epilepsy alone [21]

A systematic review by Allebone et al reviewed 12 studies investigating structural and functional neuroimaging literature in psychosis of epilepsy. They found that there are inconsistent findings particularly in research that examines hippocampal volumes; that there is evidence to suggest there are structural changes in extramesial temporal regions and whole-brain structural networks; and that frontotemporal hyperperfusion can be seen in functional imaging studies during post-ictal psychosis. From this, they concluded that there is growing evidence to support a more complex pathophysiology that goes beyond the initial hypothesis that a single focal structural lesion will be identified and more likely that psychosis of epilepsy represents a network disorder. [113]

Nakahara et al also suggest there is likely a role in abnormal neural network connectivity, specifically misguidance of hippocampal mossy fiber projections, underpinning both schizophrenia and epilepsy. [114]

Interestingly, a meta-analysis published in 2018 (The Brain Consortium) that analyzed shared heritability between common psychiatric and neurologic disorders using genome-wide association studies of patients found no genetic correlation or common genetic risk between schizophrenia and focal or generalized epilepsy. Although they suggest this may be in part due to modest power of the data available. [115]

Classification of psychosis of epilepsy

Kanner states that various classifications have been proposed for psychoses associated with epilepsy. He asserts that for the neurologist, the most useful might be that which distinguishes among psychoses closely linked to seizures (ictal or postictal psychosis), those linked to seizure remission (alternative psychosis), psychoses with a more stable and chronic course (eg, interictal psychosis), and iatrogenic psychotic processes related to antiseizure medications. [22]

Ictal psychosis

Ictal psychosis describes psychotic symptoms that present during a seizure. Symptoms are sudden in onset and last for the duration of the seizure. This may also occur in non-convulsive status epilepticus or status epilepticus. [116]  

Postictal psychosis

Postictal psychosis is most common and describes onset of psychotic symptoms following a seizure. In these cases, psychotic symptoms are typically self limited in duration, consist of positive symptoms, and can be dominated by persecutory delusions, religious delusions, systematic hallucinations, and have strong affective components. [116, 117]  When making this diagnosis it is important to make the distinction between postictal psychosis, which is characterized by well-systematized delusions and hallucinations in a setting of preserved orientation and alertness, and postictal confusion. [23]

Criteria proposed by Stagno for postictal psychosis include the following: [24]

  • Psychotic or other psychiatric symptoms occur after a seizure or, more frequently, a series of seizures, after a lucid interval, or within 7 days of the seizure(s)
  • The event may be psychosis, depression, or elation or may be an anxiety-related symptom
  • The event is time-limited, lasting days or, rarely, weeks; no significant clouding of consciousness occurs

Logsdail and Toone believe that clouding of consciousness, disorientation, or delirium may be noted, and, if consciousness is unimpaired, delusions and hallucinations are present; a mixture of both also may be noted. [25]

Clouding should not be attributed to other medical or psychiatric causes (eg, drug intoxication, complex partial status epilepticus, metabolic disturbance).

Interictal psychosis

So et al distinguished between self-limited postictal psychosis and the unremitting chronic interictal psychosis. Interictal psychotic phenomena, particularly hallucinations and delusions, are common in patients with long-standing epileps. [26, 27, 28]

Although many etiologies of psychosis in epilepsy have been proposed, the significance of such factors as the type of seizure, epilepsy classification, lateralization of foci, and age at onset of epilepsy remains uncertain. [29, 30, 31, 32]

Tarulli et al documented cases of patients who had multiple episodes of postictal psychosis before developing interictal psychosis. [33]  They concluded that a progression from postictal to interictal psychosis may be at play and that increased awareness and prompt treatment of postictal psychosis may inhibit or prevent the development of some instances of interictal psychosis.

Hilger et al found that patients with interictal psychosis were younger at onset of first psychosis, had a shorter interval between epilepsy onset and first psychotic episode, and more frequently exhibited schizophreniform traits. [118]

Factors in the development of psychosis

The following variables are believed to have particularly strong links to the development of psychotic phenomena in patients with epilepsy:

  • Family history of psychosis - Patients who had a positive family history of psychosis were extremely susceptible to psychosis, suggestive of a genetic predisposition
  • Age at onset of epilepsy - Patients with interictal psychosis showed a significantly earlier onset of epilepsy [34, 35, 36, 37, 38]
  • Type of seizure - The existence of complex partial seizure (mostly temporal lobe epilepsy and left sided epileptic focus) may be strongly associated with interictal psychoses [39, 40, 41]
  • Intelligence - Patients with borderline intellectual functioning tend to develop psychotic symptoms relatively frequently [34, 35]
  • Partial complex seizures, especially in patients with temporal lobe epilepsy and left sided epileptic focus; other studies have also seen an association with frontal lobe epilepsy.
  • Structural brain lesions (eg, small tumors, hamartomas) [41]
  • Mesial temporal lobe gangliogliomas [41]
  • Left-handedness, especially in women [41]

Schmitz et al studied risk factors and classified them by the following system [1] :

  • biologic factors
  • earlier onset of epilepsy
  • more severe epilepsy
  • psychosocial factors
  • disturbed family background
  • lack of interpersonal relationships
  • social dependency
  • professional failure
  • more frequent temporal lobe and unclassifiable epilepsies and less frequent generalized epilepsies

With regard to the last item above, no significant differences in types of epilepsies between patients with epilepsy and psychosis and patients with epilepsy without psychiatric disease have been found.

Trimble and Schmitz believe that the conclusions presented in the literature on risk factors are highly controversial. [41]


When diagnosing psychosis of epilepsy it is important to make the distinction with schizophrenia. Schizophrenia typically presents with an insidious onset in late teens and early 20s. Diagnosis requires the presence of positive and negative symptoms for at least 6 months.

Stagno reported that persistent interictal psychoses of epilepsy and the schizophrenia-like psychoses of epilepsy are distinguishable from schizophrenia in the traditional psychiatric sense by the following. [24]

  • onset typically occurs 10–20 years postepilepsy diagnosis
  • lack of negative symptoms of schizophrenia, particularly flattening of affect and personality deterioration
  • better premorbid personality
  • paranoid delusions
  • delusions of reference
  • more benign and variable course
  • command hallucinations are uncommon

In a review study of patients with epilepsy who developed psychosis, Tandon and DeQuardo found that the patients’ psychoses were usually a form of schizophrenia, most commonly paranoid schizophrenia. [42]


Status epilepticus and ictal abnormalities are treated in the same way as nonpsychiatric epileptic events. Postictal events are treated by improving seizure control. So et al believe that postictal psychosis remits spontaneously even without treatment but that the use of effective neuroleptics may shorten the duration. [23] Interictal psychosis is treated with antipsychotic drugs. However, studies are limited. Medications that lower the seizure threshold should be avoided. Some studies indicate that risperidone, molindone, and fluphenazine may have better profiles than older antipsychotic medications; clozapine has been reported to confer a particularly high risk of seizures. Additional factors should be considered when starting an antipsychotic medication including side effects of each drug (metabolic, extrapyramidal, cardiovascular, and hormonal) and possible interactions with concurrent antiseizure treatments. Patients with interictal psychosis are more likely to have recurrent episodes and the overall duration of treatment should be determined on a case-by-case basis with consideration of treatment guidelines for psychosis and schizophrenia. [119]  

While not a treatment option for psychosis per se, there has been mixed results on epilepsy-related psychosis following surgical resection. In patients with treatment refractory epilepsy and psychosis who underwent resective surgery for treatment of epilepsy, Buranee et al found significantly reduced prevalence in psychosis. [120] However, their study did identify three patients who developed de novo psychoses following surgery. Other studies have similar results with some patients having improvement in epilepsy-related psychoses, some with no change, and some with de novo or worsening psychoses. [121]  Devinsky et al found the prevalence of de novo psychosis following surgery occurred at 1.1%, which has been speculated to be related to forced normalization, unsuccessful epilepsy surgeries, or surgical complications. [122]

Forced normalization

Treatment of any of the psychoses of epilepsy should take into consideration the phenomenon termed forced normalization, which is a concept described by Landolt in the 1950s. When the electroencephalogram (EEG) in psychotic patients is normalized, often with anticonvulsant medicines, the psychiatric problem worsens.


Bipolar Affective Disorders

Bipolar affective disorder is chronic psychiatric disease with severe changes in mood with a wide spectrum of clinical manifestations. A number of studies have demonstrated that affective disorders in epilepsy represent a common psychiatric comorbidity; however, most of the neuropsychiatric literature focuses on depression, which is more prominent. [45]

The incidence of developing bipolar affective disorder in epilepsy is 1.69 cases per 1000 persons-year, compared with 0.07 in the general population. [46]

Bipolar symptoms were 1.6–2.2 times more common in subjects with epilepsy than with migraine, asthma, or diabetes mellitus and are 6.6 times more likely to occur than in healthy subjects. A total of 49.7% of patients with epilepsy who screened positive for bipolar symptoms were diagnosed with bipolar disorder by a physician, nearly twice the rate seen in other disorders. [47]


Depression is a mental state or chronic mental disorder characterized by feelings of sadness, loneliness, despair, low self-esteem, and self-reproach. Accompanying signs include psychomotor retardation (or, less frequently, agitation), withdrawal from social contact, and vegetative states, such as loss of appetite and insomnia.

Depression is the most frequent psychiatric comorbidity seen in patients with epilepsy. It is more likely to occur in patients with partial seizure disorders of temporal and frontal lobe origin. It is also more frequent in patients with poorly controlled seizures. [48]

Two possibilities exist: (1) depression is a reaction to the epilepsy or (2) depression is a part of the epilepsy.

Mendez et al compared patients with epilepsy to matched controls without epilepsy but with a similar degree of disability from other chronic medical diseases and found that while 55% of the patients with epilepsy reported depression, only 30% of the matched controls reported depression. [49]

Mendez et al concluded that depression is related to a specific epileptic psychosyndrome.

On the other hand, Robertson concluded that with few exceptions, the phenomenology of the depression to a large degree is not attributed to neuroepilepsy variables; however, not all studies have found this difference. [50]

In patients with refractory epilepsy, the presence of depression is one of the most important variables to have an impact on their quality of life, even more than the frequency and severity of the seizures. [123]

Several studies have documented that the quality of life improves significantly in patients with epilepsy who are made seizure free. If those patients are excluded, Boylan et al have found that the quality of life is related to depression but not to degree of seizure control. [51]

Despite its high prevalence in patients with epilepsy, depression very often remains unrecognized and untreated. The reasons for clinicians’ failure to recognize depressive disorders in patients with epilepsy include the following: [52]

  • Patients tend to minimize their psychiatric symptoms for fear of being further stigmatized
  • The clinical manifestations of certain types of depressive disorders in epilepsy differ from depressive disorders in patients without epilepsy and therefore are not recognized by physicians
  • Clinicians usually fail to inquire about psychiatric symptoms
  • Patients and clinicians tend to minimize the significance of symptoms of depression because they consider them to be a reflection of a normal adaptation process to this chronic disease [53]
  • The concern that antidepressant drugs may lower the seizure threshold has generated among clinicians a certain reluctance to use psychotropic drugs in patients with epilepsy

Another concern identified by Gill et al is the unclear diagnostic accuracy of depression screening tools used for people with epilepsy. Only one validated epilepsy-specific tool exists, the neurological disorders depression inventory for epilepsy (NDDI-E), which could affect the diagnosis and surveillance of depression. [124]

Risk factors for the development of depression in patients with epilepsy include the following:

  • temporal lobe (but not frontal lobe) partial complex seizures
  • vegetative auras
  • family history of psychiatric illness, particularly depression
  • laterality effects, which are controversial

Physiologic factors associated with epilepsy and depression

Decreased serotonergic, noradrenergic, and GABAergic functions have been identified as pivotal etiologic mechanisms in depression and have been the basis for antidepressant pharmacologic treatments. [54]  Decreased activity of these same neurotransmitters has been shown to facilitate the kindling process of seizure foci, to exacerbate seizure severity, and to intensify seizure predisposition in some animal models of epilepsy.

Therefore, parallel changes of serotonin, norepinephrine, dopamine, and GABA may be operant in the pathophysiology of depressive disorders and epilepsy. Jobe et al have presented evidence that some types of depression and some types of epilepsy may be associated with decreased noradrenergic and serotonergic transmission in the brain. [55]

Flor-Henry speculated that depression might be related to right (nondominant) foci, a finding confirmed by a few other investigators. [56]

Some authors have suggested that elation is associated with right-sided lesions and depression or sadness with left-sided lesions. Most studies that find a relationship between laterality and depression have found depression to be more common with left-sided foci.

Lopez-Rodriguez et al found that major depressive episodes were statistically more frequent in patients with left temporal lobe seizures than in patients with right temporal lobe seizures. [57]

Other authors report no laterality differences in depression rates.

More recent studies have linked other neurophysiological signaling pathways to epilepsy and depression. Prior studies have proposed a role of glucocorticoid receptors in the pathogenesis of depression, D’Alessio et al looked directly at glucocorticoid receptors with quantitative analysis in patients who underwent temporal lobe resection for refractory temporal lobe epilepsy with or without depression. Patients with comorbid depression had lower expression of glucocorticoid receptors compared to those without. [125]  Expression of glucocorticoid receptors also negatively correlated to severity of depression. Sadeghi et al, found in a rat model after induced status epilepticus, NMDA receptors pathologically signal leading to increased nitric oxide production and clinically leads to fluoxetine treatment resistant depression. [126]  Other studies have also looked into the association between NMDR in mediating depression in epilepsy models.

Studies have also investigated neuroinflammation from a range of causes including limbic encephalitis and chronic stress as a possible mechanism in the pathogenesis of epilepsy, behavioral changes, and depression in epilepsy. These studies also discuss possible targets for treatment. [127, 128, 129, 130, 131, 132, 133]

Other factors associated with depression in epilepsy

One of the variables linking depression and epilepsy is a family history of depression.

A greater frequency of depression has been found in patients with seizures originating in limbic structures; also, a frontal lobe dysfunction has been associated with depression.

The quality of life is often suboptimal for patients with epilepsy, and this may adversely affect mood. [58, 59, 60, 61, 62]

Increased financial stress, life stressors, and poor adjustment to seizures are predictive of increased depression. [63]  Additional risk factors include older age, female gender, low education level, not being employed, poor antiseizure medication adherence, polytherapy, stigma, and anxiety. Yang et al also found that a shorter epilepsy duration was associated with a lower risk of depression. [134]

The lack of control over the illness may be an additional risk factor for depression [59, 65]

Depression in epilepsy may also result from iatrogenic causes (pharmacologic and surgical).

The ASMs most frequently associated with iatrogenic depressive symptoms include the following: [66]

Depressive disorder can also occur following the discontinuation of ASMs with positive psychotropic properties, such as carbamazepineoxcarbazepinevalproic acid, and lamotrigine.

Frequency of depression in epilepsy

In patients with epilepsy, the reported rates of depression range from 8% to 48% (mean 29%, median 32%); the prevalence of depression in the general population in different epidemiologic studies ranges from 6% to 17%. [67]

In a study of patients with epilepsy who were admitted to a psychiatric hospital, Betts found that depression was the most common psychiatric diagnosis.

Williams studied 2000 patients with epilepsy and found that depressed mood was part of the attack in 21. According to Williams, depressed mood was the second most common emotion constituting part of the attack, with fear being the most common. [68]  Others have found similar results.

Characteristics of depression in patients with epilepsy

Characteristics of patients with epilepsy who also have depression include the following:

  • fewer neurotic traits
  • more psychotic traits
  • higher trait and state anxiety scores
  • more abnormal affect and chronic dysthymic disorder
  • high hostility scores, especially for self-criticism and guilt
  • sudden onset and brief duration of symptoms

Perhaps 10%–20% of persons with epilepsy have a peri-ictal prodrome consisting of depressed-irritable mood, sometimes with anxiety or tension and headaches. Although Williams noted in his patients that the mood disturbance would persist for 1 hour to 3 days after the ictus, postictal affective syndromes have received little attention in the literature. [68]

Blumer has defined an interictal dysphoric disorder in patients with epilepsy in which symptoms tend to be intermittent. [69]

On average, the patients tend to have 5 of the following symptoms (range 3–8):

  • depressed mood
  • anergia
  • pain
  • insomnia
  • fear
  • anxiety
  • paroxysmal irritability
  • euphoric moods

Kanner has noted that the symptoms of depression in patients with epilepsy are different from those in patients without epilepsy. He believes that patients with epilepsy who are felt to warrant antidepressant therapy often do not meet formal DSM criteria for a mood disorder and concludes that the problem of depression in epilepsy may be underestimated by using screening instruments designed for use in psychiatric patients. [70]

Kanner continued with this research using the DSM-IV criteria. Most symptoms presented with a waxing and waning course, with symptom-free periods. He referred to this form of depression as “dysthymic-like disorder of epilepsy.”

Caplan et al believe that depression in children and adolescents with epilepsy tends to have a different presentation from that seen in adults with epilepsy, although some adolescents with depression may present with a syndrome similar to that seen in adults. They reported that children with depression often do not appear sad and that the depression may be manifested by the following: [71]

  • irritability
  • oppositionality
  • aggression
  • anger

For this reason, special instruments are used to assess depression in children.

Thome-Souza et al reported that depression in children with epilepsy may be underdiagnosed and untreated for longer periods than in adults. They found that 70.5% of children and adolescents in the study had psychiatric disorders and that the most frequent psychiatric disorder in children was attention-deficit/hyperactivity disorder (ADHD) and the most frequent psychiatric disorder in adolescents was depression. They found that family history was also an important determinant in mood disorders in children and adolescents, [72] Kwong et al found that in adolescents risk factors for depression included medical comorbidities, female gender, frequent seizures, and younger age of seizure onset/duration of epilepsy. Anxiety and depression were highly associated with each other. [135]

Preictal symptoms of depression

Categorizing depression in patients with epilepsy as depression occurring peri-ictally (preictally, ictally, or postictally) and interictally may be useful.

Preictal symptoms of depression are believed to present as symptoms of irritability, poor frustration tolerance, motor hyperactivity, and aggressive behavior in children with epilepsy. However, very few studies have been performed in the literature. [73]

Ictal symptoms of depression

Ictal symptoms are the clinical expression of a simple partial seizure. Psychiatric symptoms occur in approximately 25% of auras. The most frequent symptoms include feelings of anhedonia, guilt, and suicidal ideation. [74]

Postictal symptoms of depression

Postictal symptoms of depression have been recognized for a long time, but they have been poorly studied in a systematic manner. [75]

Interictal symptoms of depression

For patients with epilepsy to experience depressive episodes that fail to meet any of the DSM-IV-TR criteria is not unusual. Kraepelin and Bleuler were the first to describe affective symptoms of prominent irritability, intermixed with euphoric mood, fear, and symptoms of anxiety, as well as anergia, pain, and insomnia. [76, 77, 78]

In 1986, Mendez et al used the term atypical depression in epilepsy patients using the DSM-III-R criteria.


Depressive symptoms are often the most common comorbidity among people with epilepsy, However, multiple studies indicate that symptoms continue to remain underdiagnosed and untreated, [136]   The treatment of mood disorders in patients with epilepsy includes reevaluation of the anticonvulsant regimen, cautious but aggressive use of antidepressants, and psychotherapy. The role of neurostimulation and neuromodulation is also being investigated.

First and foremost, treatment involves seizure control with appropriate anticonvulsant therapies. A phenomenon analogous to alternative psychosis, worsening of behavior with better seizure control, has been reported in epilepsy-associated mood disorders.

There is evidence that some anticonvulsant therapies, including vagus nerve stimulation, valproate, carbamazepine, and lamotrigine, also have antidepressant effects and may prove effective in treating depression in patients with epilepsy. Phenobarbital is known to produce depression. While topiramate, vigabatrin, tiagabine, levetiracetam, and zonisamide may have negative effects on mood.

According to Schmitz, vigabatrin has been linked to psychoses and to major depression, and phenytoin has been associated with toxic encephalopathies. [79]

McConnell and Duncan cite some patients in whom phenytoin had been linked to depression and mania. A case has been made that the GABAergic drugs may be associated with an increased incidence of psychiatric problems. [80]

While there has been concern that antidepressants can have proconvulsant effects, evidence of this is somewhat controversial and has been primarily associated with overdoses. Antidepressants may be necessary to effectively treat depression in these patients. When an antidepressant is prescribed, adverse effects and drug interactions must be evaluated. Newer serotonin and/or noradrenaline reuptake inhibitors should be regarded as safe and may be used in patients with epilepsy. Citalopram (owing to its lack of drug interactions) and multireceptor-active compounds such as nefazodone or venlafaxine are suggested as first-line treatments. However, there is consensus that amoxapine, bupropionmaprotiline, and clomipramine should be avoided given their proconvulsant effects. [137, 138]  Additionally tricyclic antidepressants are not recommended as first line treatment given side effects and the increased seizure risk specifically in children.

In the treatment of epilepsy-related depression, priority should be given to optimizing seizure control, since seizure remission tends to accompany improved psychosocial functioning.

Maguire et al performed a metaanalysis of 10 studies on effectiveness of antidepressants in treating depressive symptoms associated with epilepsy and concluded that evidence remains very limited. There is no available comparative data to inform the choice of antidepressant drug or classes of drug for efficacy or safety for treating people with epilepsy and depression. [139]

A recent unblinded study performed by Gilliam et al in 2019 compared sertraline and CBT for treating depression in epilepsy and effect on seizure frequency.  This study found both to be effective compared to placebo control groups. They also found that sertraline was not associated with an increased risk of GTC seizures or suicidality. [140]

Additional studies have looked into non pharmacologic interventions. Studies suggest that psychotherapy, specifically cognitive behavioral therapy and mindfulness-based CBT, can improve quality of life and prevent major depressive episodes and suicidality in patients with epilepsy." [141, 140]

However, given that patients may have limited access to psychotherapy, other interventions have been investigated. Small trials with targeted self-management, [142]  telephone based education for self management of depression, [143] mindfulness-based interventions, [144]  and electronic CBT [145]  have been effective in improving depression scores in patients with epilepsy.

Additional studies have seen improvement in psychiatric comorbidities and quality of life with implementation of multidisciplinary management programs that include involvement of epileptologists, psychiatrists, and social media based education. [146]

Neuromodulation and neurostimulation treatments including electroconvulsive therapy, vagus nerve stimulation, and repetitive transcranial magnetic stimulation have been under investigation for treatment of depression in patients with epilepsy. Electroconvulsive therapy is not contraindicated and may prove effective for epilepsy patients with severe, treatment-resistant, or psychotic depression. [147]  Recent studies have shown an increase in seizure threshold with ECT and may even have a role in refractory status epilepticus. [148]

Vagus nerve stimulation has been approved for drug-refractory epilepsy. In small studies, Elger et al showed that treatment with vagal nerve stimulation improves depression in epileptics independent of effects on seizure frequency. [103] Since these initial studies, follow-up has shown VNS to have antidepressant effects with response rates ranging from 30–53% and with sustained efficacy. [147, 149]

Repetitive transcranial magnetic stimulation is a treatment primarily used for depression that was approved in 2007. While there has been the reported side effect of seizure with rTMS, studies have shown that there is no additional risk of rTMS induced seizure in people with epilepsy compared to those without. Additionally, there is ongoing investigation in the role of low-frequency rTMS to treat refractory seizures. Other studies have looked at the effect of targeted responsive neurostimulation and mood.

A small study performed by Azmoodeh et al in 2019, evaluated transcranial direct current stimulation in patients with temporal lobe epilepsy for treatment of depression, anxiety, and stress. Patients received 10 sessions targeting the left dorsolateral prefrontal cortex and results of the study showed significant decrease in depression, anxiety, and stress. [150]  Animal models have seen similar encouraging results. [151]

Cranially implanted responsive neurostimulators (RNS) are a new option for treating refractory epilepsy, and available data showed no effect on mood. [104]

It is imperative that depression be recognized and treated in patients with epilepsy. Further prospective studies of new treatment options for depression in this patient population are needed. [81]


In a carefully selected series of patients with epilepsy, Williams found that only 165 of 2000 patients had complex, including emotional, ictal experiences. [82]

Of those 165 patients, only 3 described elation. Mania and hypomania are rare in association with epilepsy.

Manic-depressive illness is also rare; of 66 patients with epilepsy and major depression, only 2 had bipolar disorder. This rarity is probably, to some degree, secondary to the antimanic effect of drugs such as carbamazepine and valproate. However, mania was uncommonly associated with epilepsy even before the use of modern ASMs.


Suicidal Behaviors

Suicidality (completed suicide, suicide attempt, and suicidal ideation) is significantly more frequent among people with epilepsy than in the general population. [81, 83, 84, 85, 86, 87] This is also true in studies looking at teenagers and young adults. [152]

The risk of suicide in the general population averages about 1.4%. Depression is one of the psychiatric disorders that increases the risk of suicide. The risk of suicide in depressed patients is believed to be around 15%.

On average, the risk of suicide in patients with epilepsy is about 13% (prevalence rate 5–10 times that of the general population). Although some authors question its methodological and patient selection techniques, most authors cite Barraclough's meta-analysis, which revealed that the risk of suicide in patients with temporal lobe epilepsy is increased to as much as 25-fold that of the general population. [88]

The relationship between epilepsy and suicidality is complex, multifactorial, and may share underlying neurobiology. A study published by Hesdorffer et al found that prior to the onset of epilepsy, there was an increased risk for suicide attempts when comparing patients who later developed epilepsy compared to control patients. Furthermore the strength of this association was similar for those with and without diagnosed psychiatric disorders. [153]  Similarly Fazel et al found an independent effect of epilepsy on suicide but also reported that patients with epilepsy and a comorbid psychiatric diagnosis had a higher risk of mortality (from all causes) compared to the general population and those with epilepsy and without psychiatric comorbidities. [106]  

The incidence of suicidal phenomena linked to specific ASMs has not been systematically well studied. These data may either reflect the natural course of an underlying, recurrent psychiatric illness with no real effect from ASMs or could suggest that ASMs lower the threshold for manifesting psychiatric symptoms in individuals who are vulnerable because of a genetic or historical predisposition to psychiatric disorders.

Frequent risks associated with suicidality include the following: [81]

  • Current or past history of mood and anxiety disorders
  • Family psychiatric history of mood disorders, particularly of suicidal behavior
  • Past suicidal attempts
  • Social stressors including unemployment and stigma [155]
  • High antiseizure load (sum of prescribed daily doses/defined daily dose ratios for each ASM) and frequent seizures have been associated with suicide risk independent of depression [156]

In January 2008, the US Food and Drug Administration (FDA) issued an alert regarding the association between suicidality and ASMs, having concluded that there was a statistically significant 1.8-fold increased risk of suicidality with exposure to ASMs. This conclusion was based on the results of a meta-analysis that included data from 199 randomized clinical trials of 11 ASMs: carbamazepine, felbamate, gabapentin, lamotrigine, levetiracetam, oxcarbazepine, pregabalin, tiagabine, topiramate, valproate, and zonisamide. The meta-analysis encompassed 43,892 patients treated for epilepsy, psychiatric disorders, and other disorders, predominantly pain. In the study, suicidality occurred in 4.3 of 1000 patients treated with ASMs in the active arm, compared with 2.2 of 1000 patients in the comparison arm. The results of this meta-analysis must be considered with great caution, and more research is necessary. [81, 93, 94]

The FDA has decided to insert suicide warnings in the package inserts of all ASMs; thus, physicians need to identify patients with increased risk of suicide. [95]


Anxiety Disorders

Anxiety is an experience of fear or apprehension in response to anticipated internal or external danger, accompanied by muscle tension, restlessness, sympathetic hyperactivity, and/or cognitive signs and symptoms (hypervigilance, confusion, decreased concentration, or fear of losing control).

Anxiety is common in patients with epilepsy; of 49 patients with epilepsy attending a tertiary epilepsy care center, 57% had high-level anxiety.

Anxiety in patients with epilepsy can be ictal, postictal, or interictal. There appears to be a bidirectional link between anxiety and epilepsy as seizures may trigger anxiety symptoms while anxiety may also trigger seizures.

GABA is the most important inhibitory transmitter in the central nervous system. Evidence suggests that the abnormal functioning of GABA receptors could be of great importance in the pathophysiology of epilepsy and anxiety disorders. [82, 81]

Differentiating between spontaneous fear and reactive fear (ie, reaction to the knowledge that a seizure may occur) can be difficult. Panic disorder can produce paroxysmal symptoms, which can be confused with epileptic events and may go unrecognized in patients with epilepsy. Anxiety also may be related to nonepileptic attack disorder.

Symptoms of anxiety in epilepsy

Symptoms of anxiety in epilepsy may result or be exacerbated by psychological reactions, including responses to the unpredictability of seizures and restrictions of normal activities. This results in low self-esteem, stigmatization, and social rejection. [1, 83, 84]  According to Goldstein and Harden, epileptic events can produce symptoms indistinguishable from those of primary anxiety disorder. [85]

Fear and anxiety are often associated with simple partial seizures. Torta and Keller estimated that fear occurs as an aura in as many as 15% of patients, [11]  and Goldstein and Harden concluded from several studies that fear is one of the most common ictal emotions. [7]

Ictal anxiety symptoms manifest as fear or panic, sometimes with other characteristics of temporal discharges, such as depersonalization and déjà vu, as well as other psychological and autonomous phenomena.

Anxiety in association with type of epilepsy and frequency of seizures

The highest rates of psychiatric comorbidities, including anxiety, are reported in patients with chronic, refractory seizure disorders [86, 87]

Interestingly, however, Goldstein et al found that patients with epilepsy with high seizure frequency had lower anxiety scores than did patients with lower seizure frequency. [88]

The risk of anxiety is higher in focal (more frequent in temporal lobe) epilepsy than in generalized epilepsy. In patients with temporal lobe epilepsy, Trimble et al reported that 19% of the patients were diagnosed with anxiety and 11% were diagnosed with depression.

Edeh and Toone found that patients with temporal lobe epilepsy scored higher for anxiety than did those with focal, nontemporal lobe epilepsy. [4] Furthermore, direct electrical stimulation of the amygdala during steroelectroencephalography recording of patients undergoing presurgical evaluation for refractory epilepsy was shown to induce anxiety and fear. [157]

Anxiety can also be seen in frontal lobe epilepsy.

Ictal and interictal anxiety

Anxiety in epileptic patients may occur as an ictal phenomenon, as normal interictal emotion or as part of an accompanying anxiety disorder, as part of an accompanying depressive disorder, or in association with nonepileptic, seizurelike events as part of an underlying primary anxiety disorder.

Interictal anxiety has a great influence on the quality of life of patients, since most of them have a permanent fear of new discharges. Torta and Keller have estimated that as many as 66% of patients with epilepsy report interictal anxiety. Goldstein and Harden proposed 2 major psychological mechanisms for this, as follows:

  • fear of seizure recurrence (seizure phobia)
  • issues surrounding locus of control

They concluded that documented cases of actual seizure phobia are rare but that a sense of dispersed locus of control can cause profound problems in patients with epilepsy.


Similar to management of depression and depressive symptoms, anxiety symptoms can improve with better seizure control. However, targeted pharmacological management including SSRI and SNRIs and psychotherapy should also be considered. [158]

Several studies have shown that pregabalin, used as an adjunct for partial seizures, has been an effective, rapidly active, and safe treatment for generalized anxiety disorder. Gabapentin has also shown to have anxiolytic properties.


Although, as shown above, studies looking into the association between anxiety and epilepsy have been performed, because of the difficulty in separating the anxiety that accompanies a chronic disease from pathologic anxiety, studies investigating anxiety in epilepsy have nonetheless been relatively few.


Personality Disorders

Personality disorders in epileptic patients can cause abnormal behavior that can have a direct impact on seizure control and quality of life. The question of the relationship has a long history and remains controversial. In 1975, Woxman and Geschwind described a syndrome consisting of circumstantiality (excessive verbal output, stickiness, and hypergraphia), altered sexuality, and intensified mental life in a patient with temporal lobe epilepsy. It was called Geschwind syndrome. [89]

Bensan and Herman reported that data are insufficient to state with certainty that a consistent pattern of behavioral changes occur in patient with temporal lobe epilepsy. The complex partial epilepsy should not be diagnosed on the basis of the presence of Geschwind syndrome without any paroxysmal episodes that can be proven to be epileptic. [90]

The link of personality disorders to epilepsy was not only seen in temporal lobe epilepsy. Trinka et al found that personality disorders were present in 23% of patients with juvenile myoclonic epilepsy. [91]

Trimble has summarized the data and concluded that the personality profile of a patient with epilepsy can be explained by a complex combination of the effect of (1) dealing with a chronic illnesses, (2) ASM effects, (3) and temporal lobe pathology. He supported that certain personality disturbances in epilepsies should be viewed as associated with cerebral abnormalities that also lead to seizures. [92]


Attention-Deficit/Hyperactivity Disorder

Attention-deficit/hyperactivity disorder (ADHD) is another psychiatric comorbidity in patients with epilepsy and is more common in children. The co-occurrence may result from altered neurobiological mechanisms involved in early brain development.

The incidence of ADHD is about 7.76 cases per 1000 person-years in patients with epilepsy and 3.22 in patients without epilepsy. The incidence of epilepsy is 3.24 cases per 1000 person-year in patient with ADHD and 0.78 in those without ADHD. [93]

A neuropsychiatrist may find difficulty in differentiating impaired attention secondary to absence seizures and attention deficit as a phenotypical representation of ADHD.

Many ASMs can cause symptoms that mimic ADHD, and the most commonly implicated are the GABAergic drugs such as barbiturates, benzodiazepines, and vigabatrin. Behavioral changes should be monitored when levetiracetam is administered to patients with concomitant autism or ADHD. [96]

Methylphenidate can cause aggravate seizures in patients with ADHD, although generally it is considered safe in those who are seizure free. [94]


Psychotropic Effects of Antiseizure Medications

Knowledge about the psychotropic effects of antiseizuer medications (ASMs) is crucial and yet very limited in the epilepsy population. Psychiatric adverse events, including symptoms of depression and anxiety, can occur with any ASM. This is more common when drugs are used at high doses and when used in patients with prior or family history of psychiatric disorders. There has been an association with several ASMs, particularly barbiturates (phenobarbital and primidone), topiramate, vigabatrin, tiagabine, topiramate, zonisamide,  levetiracetam, ezogabine, and perampanel. [89, 90, 91, 92, 95]  Psychopathology can also result in patients who discontinue antiseizure medications with anxiolytic or mood stabilizing properties or when additional medications are started that have enzyme-inducing properties that accelerate metabolic clearance of psychotropic medications. [107]

Evidence suggests that lamotrigine and the vagal nerve stimulator may have antidepressant properties that could be of use in light of common comorbid depression. Carbamazepine, valproate, lamotrigine, and possibly oxcarbazepine may have mood stabilizing properties. Gabapentin, pregabalin, and tiagabine may have anxiolytic benefits. Lacosamide has no effect on depression or anxiety symptoms in epileptic patients. [97]


Psychiatric Disorders and Epilepsy Surgery

Psychiatric comorbidities can improve, worsen, have no change, or occur de novo following epilepsy surgery. A history of psychiatric disorders before epilepsy surgery is associated with poorer chance of postsurgical seizure remission. After resective surgery, only patients with good or excellent seizure control had sustained long-term improvement in mood.

Postsurgical patients had higher suicidal mortality rate compared with the general population, and people who continue to have seizures after surgery had a higher suicidal mortality rate, in contrast to those who were seizure free after surgery (4–5 times). [98]  In a series of 26 patients, gamma knife radiosurgery for mesial temporal lobe epilepsy showed no significant psychiatric changes from preoperative baseline for up to 24 months. [99]

About one third of patients present with depression after surgery (25% of whom have no history of preexisting mood disorders). Depression de novo is the highest risk factor and is found in 75% of patients. [100]  Studies have shown that the majority of patients who develop mood and anxiety disorders following surgery will have remission within the first year, while approximately 10–15% of patients may not. [107] Psychosocial factors (maladaptive personality and adjustment, financial difficulties, and poor postoperative family dynamics) may play a role in postoperative depression even in those with excellent seizure control. [100]

Unfortunately, in many instances unless a concomitant psychiatric disorder affects the process of preoperative planning and evaluation (e.g. inability to provide informed consent), presurgical psychiatric disorders go undiagnosed. This ultimately affects the ability to accurately determine incidence of postsurgical of psychiatric disorder incidence.  


Patient and Family Education

For patient education information, see Epilepsy, Depression, Schizophrenia, Bipolar Disorder, and Anxiety.

The following Web sites are useful patient and family education tools:



Psychiatric comorbidities in patients with epilepsy are relatively frequent. Despite the high prevalence rates, few data are available. Because of this, the data used are from primary psychiatric disorders, assuming it can be applicable to patients with epilepsy.

Early recognition and management of psychiatric disorders in patients with epilepsy is extremely important, because it improves the quality of life, decreases suicidality, and aids in better seizure control. Considering ASM side effects and their impact on preexisting psychiatric symptoms can help to avoid worsening of patient symptoms and help in treatmant compliance.