Anxiety represents a particular challenge for persons with epilepsy. Anxiety has been referred to as the “forgotten comorbidity” in epilepsy, because historically, clinicians have minimized its impact, reasoning that just having epilepsy, with its inherent threat of spontaneously occurring seizures, is anxiety provoking in itself (
1). However, that view is insufficient to account for the large extent of symptom overlap that is seen in clinical practice. Anxiety may be present as an ictal phenomenon or as an interictal comorbid condition, yet the high prevalence of the co-occurring conditions suggests that the overlap is meaningful beyond the baseline stress of having chronic epilepsy. Anxiety seems to be equally likely whether or not the epilepsy is well controlled (
2).
Anxiety itself is heterogeneous, and the distinctions between the presence of symptoms and being diagnosed as having a formal or psychiatric disorder are important considerations. Treatments are usually directed toward symptoms; however, the evidence base regarding the treatment of anxiety, with or without epilepsy, is focused on outcomes related to specific diagnoses. Although this approach is common for most psychiatric illnesses, given the variability in presentation and the wide range of symptom expression especially in the context of epilepsy, treatment decisions are often difficult.
Unfortunately, the complexity of symptom identification combined with uncertainty within the evidence base typically leads to undertreatment of anxiety in people with epilepsy; this situation is similar to the undertreatment of other psychiatric illnesses in the context of epilepsy (
3). Few clinicians in either psychiatry or neurology have experience and comfort with the multifaceted approach required to address this comorbid condition (
4,
5). Anxiety disorders themselves may necessitate a complex array of treatment strategies that combine medical and nonmedical interventions. Lifestyle changes and involvement of allied health clinicians may be necessary to fully manage anxiety symptoms. Additionally, the medications used in the treatment of anxiety may have notable side effects, including cognitive dysfunction and addiction.
Despite the complexity in the diagnosis and treatment of anxiety comorbid with epilepsy, the potential therapeutic options are numerous, even if there is conflicting evidence in the literature. The present review is an effort to address this conundrum and to provide appropriate context and direction for managing this comorbid condition. Our aim was to search the literature with the goal of discovering current phenomenological conceptualizations and treatment strategies to create a highly informative narrative literature review. A broad PubMed search of the literature through the year 2021—using the subject terms “anxiety,” “epilepsy,” “seizures,” “psychiatry,” and “treatment”—was implemented. We further refined the search to highlight the use of “medication” to provide an increased focus on specific classes of pharmacological treatment. We also added “anticonvulsant” and “antidepressant” to the search terms without searching for specific medications. We selected studies that offered theoretical insight and actionable evidence, while maintaining an overarching approach to select papers based on their value in offering insight into present-day assessment and treatment strategies.
Historical Links Between Anxiety and Epilepsy
Anxiety and epilepsy are among the oldest referenced conditions in the medical literature. Epilepsy was described in ancient Greece by Hippocrates, as well as in texts that predate that society (
6). The frequency with which the two conditions overlap is noteworthy from both a phenomenological and an etiological standpoint. The observed sudden crescendo and burst of anxiety or panic symptoms have similarities to observed manifestations of epilepsy that also involve paroxysmal expressions of disease.
Connections between panic symptoms or “ictal fear” and seizure foci in the brain were noted by Dr. John Hughlings Jackson, an English neurologist during the late 19th century (
1), who inferred that epilepsy was tied to specific lesions in the brain that led to a “discharge of neurons” beyond the norm (
7). The idea of attributing behavioral disorders to higher-order brain centers was particularly insightful, and subsequent efforts clarified the role of the temporal lobe in terms of anxiety symptomatology overlapping with epilepsy. In the early 20th century, the Spanish psychiatrist Juan José López-Ibor noted that patients with panic disorders experienced an “autonomous anxiety” that could not be voluntarily induced or reduced (
8); concepts of a seizure aura reinforced the notion of autonomous anxiety, and some clinicians have posited panic attacks to be “partial seizures with psychic content” (
9). In subsequent research efforts, Dr. Pierre Gloor, a Swiss neurologist, studied 50 patients with temporal lobe epilepsy and found that direct stimulation of temporal lobe seizure foci showed a strong correlation with experiential aspects of fear (
10).
Today, anxiety is known to be pervasive in the context of numerous chronic medical illnesses including epilepsy. Descriptions of a fear network that involves mesial temporal lobe pathology and seizure semiology cements the anxiety-epilepsy relationship (
11). The high incidence of panic disorder among patients with ictal fear, in addition to the fact that nearly one-half of patients with chronic epilepsy also have an associated mood disorder, suggests that the magnitude of this relationship is strong (
12,
13). Ultimately, anxiety and epilepsy appear to be inexorably tied, and studies clarifying this relationship continue to appear (
14,
15).
Characterization of Anxiety in Epilepsy
Anxiety symptoms are most often described as generalized worry or fear, physiological responses (elevated heart rate, hyperventilation, sweating, and tremor), panic, avoidance, fatigue, irritability, or insomnia. However, a report of anxiety symptoms does not necessarily mean that an individual has an anxiety disorder. Anxiety disorders represent categorical sets of psychiatric symptoms that are ultimately associated with administratively applied diagnostic codes. A specific anxiety disorder diagnosis indicates that an individual has an aggregation of symptoms that causes functional impairment and is not better attributed to other medical or psychiatric illnesses. Even if a sufficient number of symptoms are present, they may not cause a functional impairment that results in a categorical anxiety disorder diagnosis. In research studies, anxiety symptoms are usually validated either by using standardized questionnaires or by using structured diagnostic interviews. According to DSM-5, established anxiety disorders include agoraphobia, panic disorder, specific phobia, separation anxiety, social anxiety (including selective mutism), and generalized anxiety disorder (
16).
Anxiety disorders are among the most common psychiatric disorders throughout the lifespan, yet they tend to follow a developmental pattern with an initial anxiety disorder diagnosis occurring during childhood or adolescence, which is much earlier than for diagnoses of other psychiatric disorders, such as mood disorders (
17,
18). In a population-based sample, 5%–18% of children, 0.3%–12.9% of preadolescents, and 0.6%–7% of adolescents met diagnostic criteria for an anxiety disorder (
19). The estimated mean age at onset for anxiety disorders is about 11 years of age, but the prevalence rates of specific anxiety disorders vary. In early childhood, specific phobia and separation anxiety are the most likely anxiety disorders to appear, with a median age of 7 years. Social anxiety has a later median age at onset, around 13 years. Generalized anxiety disorder, obsessive-compulsive disorder, and panic disorder tend to have even later ages at onset, typically late adolescence or early adulthood (
20,
21).
There is some evidence that anxiety disorders are more common in patients with focal epilepsy than in those with generalized epilepsy (
22). A recent cross-sectional study conducted in Norway found relatively more obsessive symptoms among children and adolescents with focal epilepsy, whereas more phobias were present with generalized epilepsy (
23). Epilepsy localized to the temporal lobe is commonly associated with anxiety disorders and symptoms, likely resulting from a neuroanatomical relationship between the hippocampus and amygdala, which are critical in the pathophysiology of fear and anxiety as well as seizure activity (
24).
Unfortunately, in the context of epilepsy, the distinction of symptoms versus disorders may become problematic, because categorizing anxiety symptoms as peri-ictal versus interictal may necessitate some arbitrary distinctions. The timing of symptoms necessitates detailed characterization of seizure events and the epilepsy subtype itself, which may not always be possible. Chemical and physiological precursors to a seizure event may occur over hours or days, and therefore clear delineations of interictal periods may be impossible. As a result, rigorous clinical psychiatric or psychological diagnoses may be provisional given the categorical requirement that anxiety disorders be considered independent from other etiologies. This uncertainty in the characterization of anxiety is superimposed upon uncertainty regarding epilepsy subtypes, where localization or etiologies may be elusive. Thus, evidence for treatment, even if specifically directed toward symptoms in a broad context of established epilepsy, must also be considered provisional in nature.
Treatment Strategies for Anxiety Co-Occurring With Epilepsy
Anxiety symptoms that appear to be purely ictal in nature would clearly improve by directing treatment toward seizures, presumably with anticonvulsant drugs. However, given the large temporal and phenomenological convergence between epilepsy and anxiety, an approach that accounts for overlapping pathophysiological processes may be intuitive (
25). Directing treatment to either anxiety or epilepsy would then be expected to improve both conditions. However, current practice has not incorporated approaches that treat anxiety as a way to improve seizures (
26).
One treatment paradigm that may be relevant involves understanding the role of the hypothalamic-pituitary-adrenal (HPA) axis. Any medication that reduces sympathetic overactivity may help to reduce stress and, at least intuitively, reduce the seizure threshold (
15). Although potentially effective, the evidence base does not yet support this approach. Another approach involves using treatments that are presumed to target mesial temporal lobe structures involved in anxiety, such as the amygdala (
27). The amygdala is an important component of the neural circuitry related to mood disorders, which phenomenologically overlaps with anxiety to such a degree that similar treatments are often used (
28).
Given the lack of an evidence base, the most common approach is to treat anxiety with a similar strategy used in treating anxiety in those without epilepsy. However, data related to the treatment of anxiety and epilepsy together are still very limited. Fortunately, two classes of medications are commonly used for both epilepsy and psychiatric illnesses. Anticonvulsants and benzodiazepines are commonly used in both neurology and psychiatry. This fact suggests that many neurologists are already well versed in medications that may be very effective in treating this comorbid condition.
Ultimately, the most efficient approach may be to take advantage of medications already used in this patient population, namely anticonvulsants, for the treatment of anxiety. Although sound strategies may blend the use of anticonvulsants with other classes of psychiatric medications, modern regimens of anticonvulsants already use multiple agents to synergize differing mechanisms of action; these combinations not only improve seizure outcomes, but they may potentially and efficiently address comorbid conditions. However, it may be that judicious combinations of different classes of medications coupled with nonpharmacological strategies may yield the best outcomes.
Nonpharmacological Strategies
Most nonpharmacological strategies involve psychotherapeutic approaches, but additional lifestyle changes may be important. Appropriate diet and activity level, including exercise, will improve wellness and mood. Appropriate sleep will also improve anxiety, as well as increase the threshold for seizures. Reduced sleep makes one less capable of managing anxiety-provoking events that happen during the daytime, potentially causing a vicious cycle, which can further interfere with sleep. Anxiety symptoms commonly cause initial insomnia, often resulting from rumination or anticipating anxious moments, whether or not formal diagnoses of anxiety or mood disorders are present.
Psychotherapy may involve a variety of strategies, but primarily includes talk therapy that is problem focused or insight oriented. More practical strategies include supportive psychotherapy or problem-focused approaches. Patients with epilepsy are good candidates for talk therapy given that they tend to require lifestyle adjustments to address the possibility of seizure activity occurring, and inherent anxiety may be present in attempting to make such accommodations. Additional strategies include the use of relaxation techniques, such as abdominal breathing, progressive muscle relaxation, or imagery, to mitigate anxiety symptoms.
The most evidence-based psychotherapeutic intervention for the treatment of anxiety disorders among patients without epilepsy is cognitive-behavioral therapy, which involves a structural approach and has been extensively studied (
29,
30). A computer-assisted approach to cognitive-behavioral therapy in those with anxiety has had some encouraging initial results (
31). Behavioral therapy has shown effectiveness in the treatment of obsessive-compulsive disorder (
32). Other psychotherapies, including psychodynamic therapy, are still used, although evidence of efficacy may not be as robust for these therapies (
33).
Pharmacological Strategies
Although psychotherapy has documented efficacy, it may be that pharmacotherapy is essential, particularly for severe symptoms. A meta-analysis of pharmacological and psychotherapeutic treatments for anxiety disorders found greater efficacy with pharmacological treatments than psychotherapeutic approaches (
34). There appears to be some benefit of adding psychotherapy to pharmacotherapy for certain anxiety disorders; however, studies of such combination therapies are still lacking (
35). Psychotherapeutic approaches, even with less established efficacy, may still play a role in improving quality of life among patients with anxiety disorders and can be used alone when considering factors such as patient preference and cost effectiveness.
Antidepressants.
For many years, antidepressants have been the mainstay of treatment for anxiety among those without epilepsy. Particularly with the advent of serotonin selective reuptake inhibitors (SSRIs), antidepressants have become accessible for a wider population, because previous generations of antidepressants were hampered by intolerable side effects or risk of toxicity. For persons with epilepsy, older antidepressants, such as tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors, are regarded as too risky either because they lower the seizure threshold or have too many drug-drug interactions, respectively. Today, there are many antidepressant-based treatment options for anxiety and associated depression, although most of the evidence base does not include studies in persons with epilepsy.
At present, first-line pharmacological treatment of anxiety disorders among those without epilepsy consists of the administration of serotonergic antidepressants in both adults and children. In numerous clinical trials, SSRIs and serotonin-norepinephrine reuptake inhibitors (SNRIs) have proven to be effective in the treatment of generalized anxiety disorder in those without epilepsy (
36,
37). In particular, duloxetine, venlafaxine, paroxetine, and escitalopram have shown high evidence of both efficacy and tolerability in widespread trials (
37–
39). Other antidepressants, including mirtazapine, sertraline, and fluoxetine, have also shown efficacy and tolerability in a range of studies (
40). Regardless of the specific agent, SSRIs and SNRIs typically exert anxiolytic effects 2–4 weeks after initiation of medication administration, although variability in this time course may occur (
41). In cases where comorbid depression exists, serotonergic antidepressants can simultaneously provide effective treatment for both anxiety and depression.
Treatment strategies directed toward panic disorder may offer particular insight for persons with epilepsy, because panic also involves a sudden escalation or paroxysm of symptoms. However, as is the case for anxiety and depression, a variety of medications have been shown to have similar efficacy. SSRIs, SNRIs, and TCAs all have good response rates as measured by the Clinical Global Impression scale or the Panic Disorder Severity Scale, and they decrease the frequency of panic attacks (
42). Compared with placebo, SSRIs decreased the risk of nonresponse to treatment by 25%, and SNRIs decreased the risk by 39% (
42). SSRIs are typically considered first-line therapy, with fluoxetine, sertraline, paroxetine, fluvoxamine, citalopram, and escitalopram found to be effective in randomized clinical trials for the treatment of panic disorder (
43).
Nevertheless, some variation may exist between and within classes of SSRIs and SNRIs. In a meta-analysis, fluoxetine, fluvoxamine, and paroxetine demonstrated better clinical response and tolerability in the pediatric population than sertraline and venlafaxine (
44). Although antidepressants can provide relief and improve function among children and adolescents with anxiety disorders, the potential for adverse effects requires additional precautions for this age group. In the United States, antidepressant medications, as well as anticonvulsant medications, carry class-wide warnings of increased risk of suicidality among youths and young adults (
45).
In most cases, this apparent paradox may best be described as antidepressant-induced activation, that is, hyperarousal that may present as disinhibition, impulsivity, insomnia, restlessness, hyperactivity, and irritability (
46). This activation can result in medication discontinuation and be difficult to differentiate from anxiety symptoms. Rapid titration of antidepressant dose or the use of high doses of SSRIs appears to increase the risk of activation syndrome; therefore, such symptoms could be mitigated by using low doses and slowly titrating doses (
46). Frequent follow-up and reassessment should be done any time that doses of antidepressant medications are increased.
Alpha-adrenergic agonists.
One strategy that is gaining popularity is to use medications that influence the autonomic nervous system. Typical symptoms of anxiety and panic include tachycardia, tachypnea, diaphoresis, and tremor, all of which are consistent with hypersympathetic autonomic states. The autonomic nervous system is driven by the HPA axis, which is a vital part of the body’s stress-response network, with effects on numerous brain regions, including the prefrontal cortex, the hippocampus, and the amygdala (
47). Dysfunction of the HPA axis plays a significant role in the state of the adrenergic system and, not surprisingly, the onset and persistence of anxiety disorders (
48). The relationship between the HPA axis and stress response is physiologically relevant for establishing a seizure threshold but also may alter brain function among people without a diagnosis of epilepsy. Some speculation suggests that overactivity of the HPA axis and subsequent adrenergic state may play an etiological role in some forms of epilepsy (
49). Given such a situation, it may be intuitive to consider that medications that reduce sympathetic overactivity could also be used to treat anxiety symptoms comorbid with epilepsy.
Although peripheral effects of hypersympathetic states are driven by epinephrine, in the central nervous system, norepinephrine is an important neurotransmitter for purposes of activation and directed attention. Reducing the effects of sympathetic nerve activity through the modulation of norepinephrine is a commonly used strategy. Powerful medications, such as beta blockers, are used to reduce sympathetic activity in hypertension as well as other states of overactivation. Propranolol is an agent often used for hyperexcitability and also for short-term anxiety control (e.g., in the context of social exposure or public speaking) (
50). Although there have only been case reports describing the use of propranolol in those with epilepsy, for example, to prevent startle-induced seizures (
51), it has been used with good success in populations with intellectual and developmental disability (
52). Those with intellectual and developmental disability are more likely to have epilepsy, and the overlap in terms of hyperexcitable neural networks may reflect the need for agents to reduce adrenergic activity.
A more common approach is to use an alpha-2 agonist, such as clonidine or guanfacine. Alpha-2 agonist activity provides negative feedback on the presynaptic terminal and reduces the release of norepinephrine in the synapse. Alpha-2 adrenergic receptors are widespread throughout the brain, although this receptor subtype has been observed to be preferentially distributed in certain locations (
53,
54). Clonidine and guanfacine may differentially bind to alpha-2 receptor subtypes, which could explain differences in the sedative effects of clonidine and guanfacine, although both agents will lead to sedation at higher doses (
54).
The effects of these medications on impulsivity is marked, and both clonidine and guanfacine have been widely used to reduce the hyperactivity and impulsivity associated with attention deficit hyperactivity disorder. Attention may be enhanced, even in a population with focal epilepsy (
55). Reducing the activity of the alpha-adrenergic system may also be useful for anxiety states such as posttraumatic stress disorder (PTSD), where reducing the peripheral effects of adrenaline may help to mitigate feedforward anxiety states (
56). In a placebo-controlled trial, guanfacine was found to be effective for the treatment of PTSD, with few side effects (
57). Although formal clinical trials of alpha-2 agonists in epilepsy have not been conducted, some cases have shown positive results (
58).
Benzodiazepines.
Benzodiazepines have a long history of use in the treatment of anxiety disorders and epilepsy. Benzodiazepines are commonly used as standing medications for the treatment of epilepsy, although it is unclear whether such usage moderates co-occurring anxiety symptoms (
59). While serotonergic antidepressants have largely replaced benzodiazepines as first-line therapy for anxiety disorders because of clinical issues—including dependence, rebound anxiety, and discontinuation syndrome—benzodiazepines continue to provide utility, particularly in the management of acute panic attacks (
60). Clonazepam has been used in combination with antidepressants as an adjunct for depression (
61).
Benzodiazepines are often well tolerated and have the advantage of a rapid onset of action compared with SSRIs. In studies performed under laboratory conditions, patients with panic disorder who were induced to have panic attacks through carbon dioxide inhalation were less likely to have panic symptoms when they received an acute dose of clonazepam (
62,
63). Alprazolam has been shown to reduce or eliminate spontaneous and situational panic attacks starting in the first week of treatment (
64). Subsequent trials have also shown lorazepam and clonazepam to be effective for panic disorder (
65).
While both lorazepam and diazepam have been validated for use in the treatment of panic disorder, long-acting benzodiazepines—including oxazepam, chlordiazepoxide, and clorazepate—may also be effective despite less of an evidence base. However, discontinuation of either short- or long-acting medications may present significant risks among persons with epilepsy. Abruptly stopping benzodiazepines could lead to breakthrough seizures, in addition to other withdrawal symptoms (
66). Discontinuation of long-acting benzodiazepines may be associated with a later onset of withdrawal symptoms than that following discontinuation of short-acting benzodiazepines (
67).
Clobazam is a long-acting benzodiazepine with an indication by the U.S. Food and Drug Administration (FDA) for adjunct treatment for seizures associated with Lennox-Gastaut syndrome (
68). Clobazam is structurally different from other benzodiazepines in that it has nitrogen atoms at the first and fifth positions of the diazepine ring as opposed to the first and fourth positions, which is typical for other benzodiazepines. Clobazam has demonstrated efficacy in the treatment of anxiety among adults and children, although most of this usage has been in Europe and Asia (
69). Multiple randomized studies of clobazam conducted with patients with anxiety have demonstrated rapid decreases in anxiety symptoms as measured by the Hamilton Anxiety Rating Scale and by improvements in performance on global rating scales in clobazam groups compared with placebo groups (
70,
71). Studies comparing clobazam with diazepam, lorazepam, alprazolam, and buspirone have shown comparable efficacy in reducing anxiety symptoms (
72–
74). Studies have also demonstrated efficacy in treating primary anxiety disorder among children (
75). Clobazam seems to have a reduced side-effect profile compared with other benzodiazepines, with fewer sedating and amnestic effects, which is likely related to its structural differences from other benzodiazepines. The reduced side-effect profile increases its attractiveness as an option for the treatment of anxiety (
76). However, few studies exploring the use of clobazam for concomitant treatment for anxiety and epilepsy have been conducted.
Anticonvulsants.
There is significant appeal to use anticonvulsants in the treatment of anxiety comorbid with epilepsy. It is a safe presumption that most patients will already be treated with anticonvulsants targeting seizures. Additionally, the mechanisms of many anticonvulsants include the modulation of gamma-aminobutyric acid type A (GABA
A) receptors. GABAergic neurons are ubiquitous in the nervous system and are often interneurons that are critical for providing inhibition. A lack of inhibition is related to seizure propagation, and therefore many anticonvulsants have efficacy related to their GABAergic effects. A lack of inhibition is also intrinsic to escalating anxiety symptoms, as seen with panic attacks. Benzodiazepines as a class have direct effects on GABAergic receptors as well; therefore, it is intuitive to consider that a wide range of anticonvulsants may be useful in the independent treatment of anxiety (
77).
However, despite the fact that many anticonvulsants influence GABAergic neurons, they have not been commonly used in a deliberate manner for that purpose. Even though anxiety is a frequent psychiatric comorbid condition in those with epilepsy, few studies specifically investigating treatment options for anxiety among patients with epilepsy have been conducted; the evidence from these studies has been mixed (
76,
78,
79). The literature is somewhat more robust regarding the use of anticonvulsants for patients with anxiety without epilepsy. Anticonvulsants have been studied for depression and are often first-line agents for bipolar disorder (
80). In epilepsy, some anticonvulsants appear to independently improve depressive or other mood symptoms (
81,
82). Selected anticonvulsants are further described below.
Lamotrigine
Although lamotrigine is not typically considered to affect the GABAergic system, it has widespread effects on other networks that overlap with antidepressants, such as the serotonergic system. Lamotrigine has been studied as an adjunct to antidepressants for mood disorders, as well as for generalized anxiety disorder and PTSD (
83). In a placebo-controlled trial, it was also shown to be effective in one outcome measure for depression among adults with epilepsy (
84). In a small study of adults with focal seizures, patients showed improvements in state and trait anxiety after 5 weeks of lamotrigine administration (
85).
Pregabalin and gabapentin
Both pregabalin and its precursor, gabapentin, have been considered to be GABAergic medications, although they have prominent effects on presynaptic voltage-gated calcium channels. In any case, both agents have been studied for their effect on anxiety. The literature contains conflicting information to some degree, mostly because samples were not selected for pure anxiety disorders. Some evidence shows that pregabalin use could reduce the need for benzodiazepines and may help with benzodiazepine withdrawal (
86). Some small studies have shown efficacy for anxiety in the context of epilepsy (
87,
88).
Valproate
For decades, valproate has been commonly used for epilepsy and for psychiatric illnesses. Valproate is viewed as a broad-spectrum anticonvulsant, as well as a versatile mood stabilizer. It has effects on GABAergic neurons but also has a number of other presumed mechanisms that have not been well established. Valproate is an adjunctive treatment for mood disorders, and it is often a primary treatment for bipolar disorder. Despite its widespread usage, few studies have formally addressed the direct effects of valproate on anxiety symptoms, although there are many observational reports (
89).
Levetiracetam
In an open-label study, adjunctive levetiracetam was shown to improve anxiety symptoms in one subgroup (
90). A double-blind placebo-controlled trial targeting social anxiety has also been conducted, but the effect of levetiracetam was not different from that of placebo (
91). However, levetiracetam was shown to increase anxiety in at least one study (
92). In addition, it is commonly associated with a high incidence of irritability that may progress to frank depression, although this may be mitigated by supplementation with B-complex vitamins (
93).
Dibenzazepines
The dibenzazepine class of medications includes carbamazepine, oxcarbazepine, and eslicarbazepine. These compounds have similar structures, but each has been refined over the years to reduce associated side effects and potential toxicities. Oxcarbazepine has been used for depression in epilepsy, with a robust response on one outcome measure but not on two other measures (
94).
Lacosamide
Lacosamide is similar to other dibenzazepines in that it has been described as having effects due to inactivation of sodium channels. A prospective multicenter study of adults with focal refractory epilepsy found improvement with add-on therapy of lacosamide for treatment of anxiety, independent from outcomes for seizure control (
95). This study was intriguing in that anxiety improvement did not depend on improving epilepsy.
Perampanel
As a selective noncompetitive antagonist of alpha-amino-3-hydroxy-5-methyl-4-isooxazolepropionic acid receptors, perampanel has a novel mechanism for an anticonvulsant. Despite limited evidence, there is some concern about aggression as a potential adverse event, to the extent that a strong precaution for homicidal ideation is included in FDA labeling. However, perampanel’s mechanism of action appears to interact with serotonergic pathways, suggesting that at least theoretically, anxiety could be improved with this agent (
96). A retrospective review found mixed results in terms of psychiatric symptoms in epilepsy. In some studies, improvements in psychiatric symptoms were found, often when patients were taking concomitant carbamazepine (
97).
Cannabidiol.
The endocannabinoid system has received renewed interest with the advent of pharmaceutical-grade cannabidiol. Anandamide is an endogenous cannabinoid that modulates amygdala function and has effects on serotonergic transmission; therefore, it may be intuitive to suggest that anandamide has effects in neural pathways underlying anxiety (
98). The use of cannabidiol is common among the public, and a recent naturalistic study showed encouraging results regarding cannabinoid derivatives and effects on anxiety in those with epilepsy (
99). Unfortunately, high-quality prospective studies have not yet been conducted.
Other medications, such as tiagabine, zonisamide, and vigabatrin, have also been speculated to have beneficial effects on anxiety among those with epilepsy but in studies with limited samples or in less sophisticated studies (
100). Details of selected treatment studies of anxiety in those with epilepsy are summarized in
Table 1 (
59,
87,
88,
90,
95,
101–
104).
Conclusions
Anxiety and epilepsy have been linked throughout history. The significance of this overlap is so compelling that common roots in the underlying neurophysiology seem intuitive. Despite this overlap, studies clarifying the meaning of these areas of convergence are few. Several brain regions seem critical. The role of the mesial temporal lobe for depression and the role of the HPA axis for panic symptoms are well accepted. The fact that seizure disorders also commonly involve these regions compels speculation about sources of neural instability that generates paroxysms, whether electrical events or not.
The role of the HPA axis may be significant. Spikes in heart rate occur during generalized seizures, which are mechanistically driven by noradrenaline. The same activated autonomic states are also clearly involved in the context of panic symptoms and are likely pronounced to varying extents in other anxiety symptoms or disorders. Treatment of such symptoms may improve quality of life, whether or not the symptoms are sufficient to justify a formal anxiety diagnosis (
105).
We consider that the information included in this narrative review is broad based and was selected with the aim of providing meaningful insight. However, we did not have the capacity to search the non-English literature or reports that were not indexed in PubMed. It is possible that post hoc analyses of studies with primary outcomes unrelated to anxiety could have been informative as well, but such analyses were not included in this narrative review.
It may be that certain classes of medications are particularly effective for treating anxiety in the context of epilepsy. Studies with various classes of medications have been conducted, but future efforts could be directed toward clinical trials for epilepsy using specific medications with anxiety outcomes in mind. Ultimately, it may be the case that treating a comorbid psychiatric condition, for example, with an antidepressant or a non-antiepileptic drug, may actually reduce seizure frequency. If this were to occur, then addressing a comorbid condition such as anxiety may be a critical strategy to improve seizure control.