Anxiety and Epilepsy: What Neurologists and Epileptologists Should Know
- First Online:
- Cite this article as:
- Munger Clary, H.M. Curr Neurol Neurosci Rep (2014) 14: 445. doi:10.1007/s11910-014-0445-9
- 445 Views
Although there has been increasing recognition of psychiatric comorbidity in epilepsy, most research and attention in this area has focused on depression. However, comorbid anxiety in epilepsy is highly prevalent, affecting more than 40 % of patients in some reports. Many important outcomes are significantly impacted by anxiety in epilepsy, including quality of life, mortality, and seizure status. Recent evidence from epidemiologic studies suggests a bidirectional association of anxiety and epilepsy, and there is mounting evidence for possible common pathophysiology underlying anxiety and epilepsy. Despite this importance, anxiety is under-recognized and undertreated in clinical practice. A variety of anxiety symptoms are seen in epilepsy, including symptoms exclusively before, during or after seizures (peri-ictal anxiety), symptoms resembling primary anxiety disorders, and anxiety directly related to epilepsy or its treatment. Key therapeutic approaches include pharmacotherapy or cognitive behavioral therapy for most forms of interictal anxiety and better seizure control for peri-ictal anxiety.
KeywordsAnxietyEpilepsyPsychiatric comorbidityEpidemiologyBidirectional relationTreatmentQuality of lifePrognosisPanicPhobiaObsessive compulsive disorderSuicideAntiepileptic drugsPost-traumatic stress disorderPeri-ictal anxietyIctal anxietyPostictal anxietyPre-ictal anxietySelective serotonin reuptake inhibitorSerotonin-norepinephrine reuptake inhibitorCognitive behavior therapyInterictal anxiety
Increasing attention on psychiatric comorbidity in epilepsy during recent years has mostly focused on depression, which has been demonstrated to be highly prevalent and to affect patient quality of life, seizure outcome, and suicide risk . Although data is somewhat more limited for anxiety in epilepsy, substantial evidence suggests anxiety has a similar prevalence and impact on patients’ lives, yet it is less frequently recognized or treated . This review will summarize current data on the complicated relationship between anxiety and epilepsy, types of anxiety symptoms seen in patients with epilepsy, and approach to treatment.
How Common is Anxiety in Epilepsy?
Multiple studies have demonstrated that a substantial portion of adults and children with epilepsy experience anxiety, significantly more than in the general population and some other chronic diseases. In adults with epilepsy, prevalence estimates for anxiety range from 11 % to nearly 50 %, depending upon the population assessed and type of anxiety measure used . A recent population-based study in England demonstrated prevalence of generalized anxiety disorder to be 12.5 % in epilepsy, based on a structured psychiatric interview, with adjusted odds of generalized anxiety disorder 2.6 for patients with epilepsy compared with the general population. This was much higher than the odds of generalized anxiety among those with asthma or diabetes [4•]. Two population-based studies in the United States found that people with self-reported epilepsy were about twice as likely to report diagnosis of anxiety than those without epilepsy [odds ratio (OR) 2.3, prevalence ratio 1.61] [5, 6]. A study in a community sample of epilepsy patients in Brazil found similar increased odds of anxiety compared with community controls without epilepsy (OR 2.1), and the prevalence of anxiety symptoms in epilepsy was 39.4 % on a self-report scale . Other studies in primary care populations and tertiary epilepsy populations have found prevalence of anxiety to range from 11 %–25 % for Diagnostic and Statistical Manual of Mental Disorders (DSM) or International Classification of Diseases-10 based psychiatric diagnoses, and 20.5 %–46 % for self-report instruments [8–13].
Similar to adults, children with epilepsy also experience high rates of anxiety. A United States population-based survey of parent-reported diagnoses showed that 17 % of children with epilepsy had been diagnosed with anxiety, compared with 3 % of those without epilepsy . Children with new onset epilepsy at 2 Midwestern pediatric neurology centers had 35.8 % prevalence of DSM-IV anxiety disorders compared with 22 % among first cousin controls . Anxiety was more common than depression in these studies and additional studies of childhood psychiatric comorbidity in the United States, Jordan, and Nigeria, with anxiety prevalence as high as 48.5 % [16–18].
Impact of Anxiety in Epilepsy: Quality of Life, Side Effects, Cognition, and Cost
Anxiety has an enormous impact on quality of life in epilepsy. In fact, studies have consistently demonstrated psychiatric comorbidity to be a greater predictor of quality of life than seizure control or severity [19–22]. Even though numerous studies have shown anxiety and depression to be independent predictors of poor quality of life in epilepsy, multivariable analysis by groups in South Korea, Hong Kong, and mainland China found anxiety to be the most important predictor of quality of life in models including depression and seizure factors [23–25]. A recent study in children ages 8–18 years demonstrated generalized anxiety, but not depression, to be an independent predictor of child-reported quality of life .
In addition to quality of life, comorbid anxiety in epilepsy affects other subjective experiences of the patient, including antiepileptic drug (AED) side effects and memory complaints. Among a group of Australian first seizure patients, some of whom were on AEDs and others who were not, symptoms of anxiety and depression measured with the Hospital Anxiety and Depression Scale (HADS) predicted report of adverse AED effects better than AED status . A group from Mexico demonstrated significant positive correlation of scores on the HADS with scores on the Liverpool Adverse Events Profile (LAEP), the established measure of adverse effects from AEDs . Among patients at United States epilepsy centers, DSM anxiety diagnosis was a significant predictor of higher LAEP score, while AED polytherapy was not, even when mood related questions of the LAEP were excluded . Other studies have demonstrated a strong association of anxiety symptoms with self-reported cognitive complaints. Among first seizure patients in Australia, anxiety was strongly correlated with cognitive complaints, while there was no association between cognitive complaints and objective measures of attentional processing . In a multivariable regression analysis of factors associated with memory complaints among Hong Kong adults with epilepsy, anxiety was the most important independent predictor of memory complaints, explaining more than twice as much variance as objective memory performance, and depression was not an independent predictor .
In addition to the impact on quality of life, adverse effects, and cognitive complaints, new evidence suggests anxiety in epilepsy is associated with increased economic costs. Assessment of factors associated with emergency department (ED) visits among epilepsy patients in London revealed that those with higher levels of anxiety symptoms on the HADS self-report instrument visited the ED significantly more often than the wider epilepsy population . Refractory focal epilepsy patients at the Penn Epilepsy Center with symptoms of anxiety missed more outpatient visits per year than those without anxiety or depression, and those with both anxiety and depression were more likely to be admitted to the hospital .
Impact of Anxiety in Epilepsy: Mortality and Seizure Outcome
Beyond the important implications of anxiety in epilepsy for patient quality of life, adverse effects, memory, and cost, anxiety is also linked to increased risk of mortality from suicide and poor seizure outcome despite medical or surgical therapy. Multiple groups have demonstrated higher odds of suicidal ideation and/or suicide attempt among epilepsy patients with anxiety compared with those without anxiety. Specifically, in a community sample of epilepsy patients in Brazil, the odds of lifetime suicidal thoughts was 3.6 times higher in anxious patients than those without anxiety, and similar increased odds of suicide risk was found in Australian epilepsy patients with anxiety (OR 3.58) [34, 35]. In the Brazilian cohort, odds of actual suicide attempts were even higher: 6.9 for those with anxiety compared with those without. A prospective study of patients with anxiety disorders showed that history of epilepsy was significantly associated with shorter time to suicide attempt, while other physical health problems including cardiac disease, autoimmune disease, cancer, diabetes, headache, and head injury were not .
Given the 2008 U.S. Food and Drug Administration alert on suicidality and antiepileptic drugs, it is reasonable to question whether AED therapy may confound the association of anxiety with increased suicide risk. Analysis of suicide risk among US veterans newly prescribed AED monotherapy demonstrated that the strongest predictor of suicide-related behavior was presence of depression, anxiety, or post-traumatic stress disorder (PTSD) prior to initiation of AED therapy, and no specific AEDs were associated with suicide behavior . Furthermore, a separate analysis among US veterans found that the likelihood of suicide-related behavior was highest in the month prior to initiation of AED therapy, with a gradual reduction in risk after initiation of AED treatment .
Multiple studies have demonstrated that epilepsy patients with anxiety are less likely to attain seizure control during medical therapy than those without anxiety. Significantly higher prevalence of anxiety was demonstrated among drug-resistant epilepsy patients in Hong Kong compared with drug responsive patients (44 % vs 16 %), and among patients with more than 1 seizure per month in the United Kingdom compared with seizure free patients (44 % vs 13 %) [39, 40]. Similarly, separate investigations have shown the presence of anxiety symptoms to be associated with poor seizure control among patients with generalized epilepsy [41, 42]. Not only is there a cross-sectional association between anxiety symptoms and poor seizure control during medical management, but some evidence suggests history of anxiety (or psychiatric disorders including anxiety) predicts subsequent poor seizure control. A prospective study of patients with newly diagnosed single seizure demonstrated that higher self-reported anxiety one month after the seizure predicted subsequent seizure recurrence . Higher pretreatment neuropsychiatric symptoms (measured with a self-report scale shown to correlate with validated scales for anxiety, depression, and memory) were independently associated with seizure recurrence at 1 year among newly treated Australian epilepsy patients . In a large Scottish cohort, prior or concurrent psychiatric history at the time of epilepsy diagnosis was associated with adjusted odds of 2.17 for later development of pharmacoresistant epilepsy .
Prognosis after epilepsy surgery is also worse among those with pre-existing psychiatric symptoms including anxiety. Three recent studies have shown an association of preoperative psychiatric history with poor seizure outcome after temporal lobectomy [46–48]. None of the 3 studies reported specific statistical analysis on the effect of anxiety separate from other psychiatric diagnoses, but anxiety was the second most common type of psychiatric diagnosis in all 3 studies, with depression being the most common diagnosis. Among patients followed at the São Paulo, Brazil epilepsy center, those with anxiety were approximately twice as likely to have an unfavorable, non-Engel 1A seizure outcome than a favorable Engel 1A outcome . One study did specifically assess the effect of preoperative anxiety diagnosis on seizure outcome after temporal lobectomy using a Cox proportional hazards model, demonstrating an independent association of preoperative anxiety with persistent auras or more severe seizures postoperatively . Interestingly, this analysis found a significant negative impact on seizure outcome of preoperative anxiety diagnosis or personality disorder, but not other psychiatric comorbidities. These associations of anxiety and other psychiatric comorbidities with seizure outcomes, particularly the findings demonstrating pre-existing anxiety symptoms to be associated with poor subsequent treatment outcomes, are intriguing and raise the question of whether there may be common pathophysiology underlying epilepsy and psychiatric comorbidity including anxiety.
Why is Anxiety Common in Epilepsy?
Anxiety and Epilepsy: A Bidirectional Relation
Recent epidemiologic studies have demonstrated that not only are patients with epilepsy more likely to be anxious than those without epilepsy, but risk of anxiety is elevated in the years prior to onset of epilepsy, as well as the years following epilepsy onset. Among US veterans above the age of 65 who later developed epilepsy, premorbid anxiety diagnoses were significantly more common than among controls (11.7 % vs 8.5 %, respectively) . Children assessed at the time of first unprovoked seizure reported significantly higher anxiety compared with published norms, and a separate study of new onset childhood epilepsy found that 45 % of comorbid psychiatric disorders were diagnosed prior to first recognized seizure, with one-third of these pre-existing diagnoses being anxiety disorders [51, 15]. A Swedish population-based case-control study demonstrated that those with a history of hospitalization for anxiety disorders were more than twice as likely to develop unprovoked seizures compared with controls (OR 2.6) [52•]. The most definitive study demonstrating a truly bidirectional association between anxiety and epilepsy in a single population was a matched, longitudinal cohort study of the UK General Practice Research Database [53••]. Individuals with new onset epilepsy were carefully matched to 4 controls without epilepsy, and the incidence rate of anxiety disorders was measured for cases and controls in each of the 3 years prior to, and after date of epilepsy onset. The incidence rate for anxiety was significantly higher in cases than controls for all years measured, with incidence rate ratios near 2 for each of the 3 years preceding epilepsy onset and following epilepsy onset. Similar findings were seen for depression and psychosis in this cohort. Overall, the epidemiologic data demonstrating a bidirectional relation between epilepsy and anxiety (or other psychiatric disorders) further raises questions regarding possible common pathophysiology underlying the disorders.
Possible Underlying Causes: Evidence for Shared Genetic Basis and Neurobiological Hypotheses
Recent studies suggest the possibility of a common underlying genetic cause for anxiety and epilepsy. A longitudinal study of new onset childhood epilepsy demonstrated that first degree family history of unprovoked seizure was significantly associated with presence of anxiety disorder at 9-year follow-up, using a DSM-oriented parent-report scale [54••]. The familial clustering of anxiety and epilepsy supports the possibility of underlying common pathophysiology and suggests that the mechanism may in part be genetic. A study of gene polymorphisms in Brazilian temporal lobe epilepsy patients may suggest a link between genetics and established neurobiological hypotheses of epilepsy and anxiety. Genotyping of serotonin transporter gene polymorphisms was carried out in 155 patients with temporal lobe epilepsy, and the presence of a specific allele (C allele of 5-HT1A C-1019G polymorphism) was found to be an independent risk factor for anxiety diagnosed by structured clinical interview .
Evidence from animal and human studies supports multiple neurobiological hypotheses for common pathophysiology of anxiety and epilepsy, including abnormalities in neurotransmitter systems (serotonin, norepinephrine, and GABA), as well as involvement of limbic circuits (particularly the amygdala) [1, 3, 56, 57]. Serotonin imbalance has been repeatedly implicated in human epilepsy, anxiety disorders, and animal models of epilepsy and behavioral abnormalities. A PET study in patients with severe mesial temporal lobe epilepsy showed reduced serotonin type 1A receptor binding potential in the seizure focus and its limbic connections . Similarly, a PET study in patients with panic disorder found reduced serotonin type 1A receptor binding in the anterior cingulate, posterior cingulate, and raphe regions compared with controls . Strains of the genetic epilepsy-prone rat (GEPR), a model of epilepsy and behavioral disorders including anxiety, were found to have innate pre-and postsynaptic serotonin and norepinephrine transmission deficits [60, 61]. GABA may also play a role in the pathophysiology of anxiety and epilepsy, considering the known antiepileptic and anxiolytic effects of benzodiazepines, which act as GABA agonists, and that panic symptoms were induced by the benzodiazepine antagonist flumazenil in patients with panic disorder . A flumazenil PET study in patients with panic disorder demonstrated global reduction in GABA-A receptor binding at the benzodiazepine site . In addition to the potential causative role of neurotransmitter system dysfunction in epilepsy and anxiety, other evidence suggests the amygdala, a key structure involved in mesial temporal epilepsy, is also instrumental in generating anxiety symptoms. Imaging studies have demonstrated reduced amygdala volume to be associated with ictal fear, and subclinical anxiety symptoms in a sample of healthy adults were predicted by left amygdala volumes [63–65].
Types of Anxiety Symptoms in Epilepsy
Types of anxiety symptoms in epilepsy
Pre-ictal anxiety (seizure prodrome)
Interictal: Symptoms of primary DSM-IV TR disorders
Social phobia (social anxiety disorder)
Panic with or without agoraphobia
Obsessive compulsive symptoms
Interictal: Anxiety specifically pertaining to epilepsy
Anxiety about epilepsy diagnosis
Anxiety/fear of seizure occurrence
AED side effect
Peri-ictal anxiety has been demonstrated to occur in a substantial number of patients with epilepsy, some of whom may experience exclusively pre-ictal anxiety, ictal anxiety, or postictal anxiety, while others may experience more than one of these anxiety types [3, 66]. A study of seizure prediction in focal epilepsy showed an association of higher self-reported anxiety with increased risk of seizure the following day, in a multiple regression model . Similarly, an older investigation found that self-rated anxiety worsened on the day preceding seizure among patients with a variety of epilepsy types .
Ictal anxiety has been reported in as many as 20 % of epilepsy patients, with anxiety being the most common ictal emotion reported by some studies . Ictal anxiety often presents as sudden onset intense fear, which may be clinically similar to panic attacks. There are multiple reported cases in the literature of patients with refractory temporal lobe epilepsy having been previously misdiagnosed with panic attacks . Clinically, panic attacks tend to be of longer duration (5–10 minutes) than seizures, and they only come from wakefulness, while seizures with ictal fear may start directly from peaceful sleep . However, the distinction often is not straightforward, and a subset of patients who experience ictal fear also have a comorbid diagnosis of panic disorder .
Anxiety symptoms may occur exclusively in the postictal period, appearing 12–72 hours after a seizure, or worsening of interictal anxiety symptoms may be seen after seizures . In the largest study of postictal psychiatric symptoms to date, anxiety was the most prevalent postictal symptom, as 45 % of patients reported significant anxiety after more than 50 % of seizures . In this study, the median duration of individual anxiety symptoms was 6–24 hours. There was also a clear relation to the presence of interictal psychiatric symptoms: >90 % of patients with interictal psychiatric symptoms reported postictal exacerbation. Interictal anxiety was associated with postictal exacerbation of symptoms and the appearance of additional distinct anxiety symptoms exclusively in the postictal period.
Interictal anxiety symptoms may resemble typical symptoms associated with primary DSM-IV anxiety disorders, or they may be directly related to epilepsy factors such as AED therapy or risk of seizure occurrence. Although most of the epidemiologic and treatment outcome research has focused on measures of generalized anxiety disorder, composite measures of all anxiety disorders combined, or generic self-report anxiety symptom scales, multiple studies have demonstrated increased risk of various primary anxiety disorders in epilepsy. A population-based study of anxiety disorder prevalence in England found significantly increased odds of social phobia and agoraphobia in epilepsy compared with the general population, in addition to increased odds of generalized anxiety disorder [4•]. A study of psychiatric symptoms in children found increased risk of specific phobia, obsessions, panic attacks, and post-traumatic stress disorder(PTSD) on a caregiver report measure in 13- to 14-year-olds . Prevalence of obsessive compulsive symptoms or diagnosis of obsessive compulsive disorder was higher in epilepsy than estimated for the general population in a few studies, with prevalence ranging from 1 %–22 % depending upon the study population and specific measure of obsessive compulsive symptoms [9, 74, 75]. One of these studies was a clinic-based survey in Germany, which also found increased prevalence of social phobia, specific phobia, panic disorder, generalized anxiety disorder, and anxiety disorder not further specified compared with general population estimates . A study of epilepsy patients recruited from epilepsy support groups in England found 51 % met diagnostic criteria for post-traumatic stress disorder (PTSD), with most of those affected having post epileptic-seizure PTSD [76, 77].
In addition to PTSD symptoms as a consequence of seizures, other anxiety symptoms may be directly related to epilepsy or its treatment. Disabling fear of seizure occurrence has been reported, and anxiety regarding the diagnosis of epilepsy and diagnostic uncertainty in epilepsy may occur [3, 78]. Anxiety may be a consequence of poor social adaptation to having epilepsy, with epilepsy stigma, poor social support, and poor coping skills potentially contributing [69, 79, 80]. Anxiety is also a well-known side effect of numerous AEDs, affecting a small subset of patients who are prescribed levetiracetam, zonisamide, topiramate, phenytoin, felbamate, ethosuximide, lamotrigine, or vigabatrin [3, 66]. Anxiety may also rarely occur as an end-of-dose withdrawal phenomenon in patients taking lamotrigine .
Management of Anxiety Symptoms in Epilepsy
Although anxiety in epilepsy is common and associated with a variety of poor outcomes, it is frequently under-recognized and undertreated in the clinical setting. This may be in part due to the relative paucity of evidence on treatment of anxiety in epilepsy populations, as well as the current lack of brief self-rating instruments specifically validated in epilepsy patients. The complexity of anxiety types seen in epilepsy may also be a barrier to diagnosis and treatment. However, to better identify patients with anxiety symptoms, neurologists and epileptologists may consider asking questions related to anxiety in the review of systems at each visit, or using a simple screening instrument. The Generalized Anxiety Disorder-7 (GAD-7) is a brief scale designed to screen for the most frequent type of anxiety disorder, and it does not contain somatic symptom items that could be confused with epilepsy related cognitive symptoms or AED side effects [2, 82].
Once the presence of anxiety symptoms is recognized for an individual patient, the next step is to identify whether the symptoms are exclusively peri-ictal, or whether they may be due to AED therapy or gaps in understanding of epilepsy. Although there are no published data on the treatment of peri-ictal anxiety or other peri-ictal psychiatric symptoms, anecdotal reports suggest that postictal depressive symptoms do not respond to pharmacotherapy for depression . Current treatment strategy for peri-ictal anxiety is to try to attain seizure remission and thus, reduce the risk of symptoms [3, 83]. Iatrogenic anxiety symptoms associated with a particular AED should be treated by transitioning to an AED less likely to provoke anxiety symptoms. Other possible iatrogenic causes of anxiety may include discontinuation of an AED with anxiolytic properties (gabapentin, pregabalin, valproate, benzodiazepines, barbiturates, tiagabine, possibly lamotrigine, or vigabatrin), or addition of an enzyme inducing AED to an antidepressant with anxiolytic properties and potential to interact with enzyme inducers [3, 83]. Anxiety related to incomplete education regarding epilepsy or epilepsy workup may improve after educational interventions, such as referral to epilepsy nurse specialists for education [3, 84].
SSRIs and SNRIs: CYP interactions and FDA approved indications
CYP Inhibition relevant to AEDs?
Generalized anxiety disorder
Obsessive compulsive disorder
Likely low potential
Likely low potential
Likely low potential
Likely low potential
Given the high prevalence of anxiety in epilepsy and its critical impact on numerous aspects of patients’ lives, it is important for neurologists and epileptologists to take first steps in identifying and managing these symptoms in clinical practice.
Compliance with Ethics Guidelines
Conflict of Interest
Heidi M. Munger Clary has received a small honorarium from The Neurology Report for preparing a CME review of an epilepsy course at the 2012 American Academy of Neurology Annual Meeting. She has also received travel accommodations from the American Epilepsy Society (Young Investigator Award, 2012, travel to 2012 AES Annual Meeting) and The Neurology Report (reimbursement of travel expenses for American Academy of Neurology Annual Meeting, 2012).
Human and Animal Rights and Informed Consent
This article does not contain any studies with human or animal subjects performed by any of the authors.