Peri-ictal versus interictal anxiety disturbances
A 34-year-old man with a 15-year history of refractory temporal lobe epilepsy (TLE), associated with a cavernous angioma in the right mesial temporal region, was evaluated for recurrent clusters of two to four seizures over a 12-h period. Seizures presented as a panicky feeling of moderate severity that lasted between 20 and 40 s, at times associated with hyperventilation, followed by rubbing of his thumb and index fingers bilaterally, and mouthing movements, during which he became somewhat confused for a period of 30–45 s.
During the immediate postictal period, he denied any confusion or dysphasic symptoms. Twenty-four to 36 h after a cluster of seizures, the patient usually became very irritable and fearful of staying alone or leaving the house. He had difficulty falling asleep, felt helpless, could not find pleasure in any activity, and experienced passive suicidal ideation. These symptoms tended to last for 24–48 h in duration and remitted spontaneously. These seizures were treated as panic attacks for 4 years, and the diagnosis of epilepsy was only reached after a secondarily generalized tonic-clonic seizure (SGTCS).
Interictally, the patient was described as “a very tense individual,” prone to moodiness, irritability, and poor frustration tolerance, but of lesser severity than that witnessed postictally. Further, he had a history of panic attacks, described as a sudden sensation of “impending doom” associated with marked tachycardia, diffuse diaphoresis, nausea, and dizziness that in most cases lasted 10–20 min, but also could have persisted for up to 45 min. These panic attacks occurred in cycles of 4–6 months' duration every 2–3 years, and have been prevented with the use of selective serotonin reuptake inhibitors (SSRIs). On the other hand, the ictal fear and postictal psychiatric symptoms were not modified by this medication.
This case illustrates the clinical differences between certain ictal, postictal, and interictal symptoms of anxiety. Anxiety and mood disorders are the most common psychiatric comorbidity in patients with epilepsy. Population-based studies have estimated the prevalence of anxiety in epilepsy patients to range between 15% and 25% (2,3). These prevalence rates are somewhat higher in studies carried out in hospital-derived populations. In a recently published study of 174 consecutive patients with epilepsy from five epilepsy centers, a current DSM-IV diagnosis of anxiety disorder was found in 30% of patients (4).
The various forms of anxiety disorders (generalized anxiety disorder, panic disorder, phobias, obsessive compulsive disorder, and posttraumatic stress disorder) can present interictally with the same clinical manifestations of anxiety disorders seen in the general population. However, the peri-ictal presentations of anxiety symptoms often differ from their interictal manifestations.
Ictal fear or panic is the most frequent ictal psychiatric symptom (5). It is the sole or predominant clinical expression of a simple partial seizure (aura) or the initial symptom of a complex partial seizure that usually has a mesial temporal lobe origin. It has been estimated that psychiatric symptoms occur in 25% of “auras,” 60% of which present as fear or panic, and 20% as symptoms of depression (5). Frequently, patients fail to recognize and report associated symptoms indicative of an ictal event such as transient confusion or subtle automatisms, as in the case presented here, which often results in its misdiagnosis and treatment as a panic disorder. The correct diagnosis is reached only after the patient has suffered from a generalized tonic-clonic seizure.
Yet, a careful history can help distinguish a panic disorder occurring interictally from ictal panic. Ictal panic is typically brief (<30 s in duration), is stereotypical, occurs out of context to concurrent events, and is associated with other ictal phenomena, such as periods of confusion of variable duration and subtle or overt automatisms. The intensity of the sensation of fear is mild to moderate and rarely reaches the intensity of a panic attack.
By contrast, interictal panic attacks consist of episodes 15–20 min in duration that at times may persist for several hours. During an interictal panic attack, the feeling of fear or panic is very intense (feeling of “impending doom”) and is associated with a variety of autonomic symptoms, including tachycardia, diffuse diaphoresis, and shortness of breath (6). Patients may become completely absorbed by the panicky experience to the point that they may not be able to report what is going on around them. However, there is no confusion or loss of consciousness as in complex partial seizures. It is not infrequent for patients to become extremely apprehensive about experiencing another panic attack, which may then lead to the development of full-blown agoraphobia.
The misdiagnosis of ictal fear as a panic disorder may be compounded by the failure to identify an electrographic ictal pattern in simple partial seizures of mesial temporal origin, above all when the seizure focus is in the amygdala. In such cases, EEG recordings with sphenoidal electrodes placed under fluoroscopic guidance may be necessary to demonstrate the epileptiform activity (7).
Patients with ictal panic may also suffer from interictal panic attacks, which have been identified in up to 25% of patients with epilepsy (8). Furthermore, the presence of ictal fear has been associated with a higher likelihood of suffering from interictal anxiety disorders (9).
It is not rare for interictal anxiety disorders to occur together with mood disorders. Our patient reported chronic irritability, poor frustration tolerance, and mood lability. These are the expression of an interictal “dysthymic-like disorder,” also referred to by some authors as “interictal dysphoric disorder” (10). In the study of 174 consecutive patients with epilepsy, cited previously, we identified the rate of current Axis I DSM IV disorders using a brief standardized psychiatric interview procedure, the Mini International Neuropsychiatric Interview (v 5.0) (MINI). Current anxiety disorders were identified in 30% of patients, while 22% were suffering from a depressive disorder. A majority of patients with depressive disorders (73%) also had symptoms that met the criteria of an anxiety disorder (4).
Ictal and interictal symptoms differ as well in their duration and course from postictal symptoms of anxiety, which can be relatively frequent among patients with refractory partial epilepsy. In a recently published study of 100 consecutive patients with pharmacoresistant partial epilepsy, we identified a mean of 2 ± 1 postictal symptoms of anxiety (range: 1–5; median = 2) in 45 patients (11). These symptoms occurred after >50% of patient seizures, and had a median duration of 24 h (range: 0.5–148 h). Thirty-two patients reported symptoms of generalized anxiety and/or panic, an additional 10 patients also reported symptoms of compulsions, and 29 patients experienced postictal symptoms of agoraphobia. In 44 of these 45 patients, postictal symptoms of depression also were reported. This was clearly illustrated in our patient.
A prior history of “interictal” anxiety disorder was identified in 15 of the 45 patients (33%) with postictal symptoms of anxiety, and there was a significant association between a history of anxiety and the occurrence postictally of the following two symptoms: constant worrying and panic. In addition, a significantly greater number of postictal symptoms of anxiety were identified among patients with a history of anxiety and mood disorders.
Our patient reported symptoms of postictal agoraphobia that lasted an average of 24–48 h. Postictal and interictal agoraphobia differ in that the former is of short duration (hours to days), while interictal agoraphobia is usually chronic and disabling, for which patients need to undergo behavioral therapy in addition to pharmacologic treatment.
SSRI antidepressants can prevent the occurrence of interictal panic attacks. On the other hand, there is no evidence yet that these drugs have any impact on ictal or postictal psychiatric symptoms, as shown in our patient.
Postictal and interictal psychosis
A 43-year-old woman with a 23-year history of intractable TLE secondary to bilateral mesial temporal sclerosis (MTS) had recurrent psychotic episodes every time she experienced a cluster of at least three to four secondarily generalized tonic-clonic seizures over a period of 48 h. The psychotic episodes began with insomnia identified ∼72–96 h after the last seizure. Eight to 12 h later, she experienced religious delusions (“I have been sent to earth by our Lord to finish His work”), auditory hallucinations (“I hear the voice of Christ telling me how to improve the suffering of people on earth”), and displayed a thought disorder consisting of circumstantial and tangential speech. Symptoms could last between 3 days and 2 weeks, and usually remitted after the start of neuroleptic medication.
To abort the postictal psychotic episodes, family members were instructed to look for the first signs of insomnia after a cluster of seizures, at which point they gave the patient 4 mg of risperidone, which was repeated every 12 h for the first 48 h. If symptoms persisted, the dose was increased until symptoms remitted. Otherwise, the dose was lowered to 2 mg/day on day 3 for 2 days and discontinued. The timely introduction of risperidone averted the appearance of most psychotic episodes.
However, in the last 2 years the patient began to present the same delusions and hallucinations in the absence of seizure occurrence (i.e., interictally). She required hospitalization because of agitation when the family tried to convince her that “her mind was playing tricks on her.” At the start of chronic therapy with risperidone at a dose of 4 mg/day, symptoms remitted. Of note, the patient never lost her ability to relate to others, and was able to maintain her job as an office clerk.
This case illustrates the characteristic presentation of postictal psychosis (PIP), and the evolution of PIP to an interictal psychotic disorder. Postictal psychoses correspond to ∼25% of psychotic disorders seen in patients with epilepsy (12). While the actual prevalence of PIP in the general population of patients with epilepsy is yet to be established, it has been estimated to range between 6% and 10% (13,14). Most PIP reported in the literature have consisted of those identified in the course of a video-EEG monitoring study (V-EEG), in which we estimated a 6.4% yearly incidence among patients with partial epilepsy undergoing V-EEG (12).
Common clinical characteristics of PIP among the different case series include: (a) a delay between the onset of psychiatric symptoms and the time of the last seizure, ranging between 12 and 120 h; (b) a relatively short duration, ranging from a few hours to 3–4 weeks; (c) an affect-laden symptomatology; (d) the clustering of symptoms into delusional and affective-like psychosis; (e) an increase in the frequency of secondarily generalized tonic-clonic seizures preceding the onset of PIP; (f) The onset of PIP after having seizures for a mean period of >10 years; and (g) a prompt response to low-dose neuroleptic medication or benzodiazepines (13–17).
The occurrence of PIP may have important implications for patients who are being evaluated for epilepsy surgery. Various studies have found that patients with PIP are likely to have bilateral independent interictal and ictal foci (15–17). In a study completed at our institution, the presence of PIP in the course of a V-EEG predicted the presence of bilateral independent ictal foci with an 89% probability (A.M. Kanner et al, unpublished observations). Thus, these patients require longer V-EEG monitoring studies to rule out the possibility of bilateral ictal foci. Some of these patients may require the use of intracranial electrodes, in which case prophylactic treatment with low-dose risperidone or haloperidol can avert the occurrence of PIP during the invasive V-EEG monitoring studies (12).
Distinguishing PIP from interictal psychosis has major therapeutic implications. Patients with PIP should not be started on daily doses of antipsychotic drugs, and often the psychotic episode could be aborted with a timely administration of low-dose neuroleptic drug such as risperidone or haloperidol, as shown in our case (12). In most cases, insomnia is the first sign of the psychotic episode, and psychotic symptoms may appear 8–24 h later. If the initial doses of antipsychotic medication fail to prevent the development of a full-blown PIP, symptoms can be expected to remit after a few days or at the most weeks, and antipsychotic drugs can then be discontinued.
Patients may experience discrete postictal psychotic symptoms that may go unnoticed by others or are unreported by the patient. In the study on the prevalence and clinical characteristics of postictal psychiatric symptoms cited previously (9), we found that seven of the 100 patients experienced 2.6 ± 1.1 psychotic symptoms (range: 1–5; median = 2) after >50% of their seizures. Of note, all of these patients also experienced postictal symptoms of anxiety and depression. None of the patients had been treated for these symptoms, even though, in four of these patients, at least one postictal psychotic symptom lasted a minimum of 24 h.
The patient presented in this article went on to develop an interictal psychosis. In contrast with PIP, interictal psychosis requires the use of chronic therapy with antipsychotic drugs to obtain remission of psychotic symptoms. Typically, interictal psychosis may develop in some of the patients with a history of PIP; less frequently, however, PIP has occurred in patients with a prior history of interictal psychosis (18,19).
In general, interictal psychosis of epilepsy (POE) has a more benign course and better response to pharmacologic therapy than psychosis in nonepilepsy patients. Furthermore, interictal POE is remarkable for the absence of negative symptoms, with better premorbid and rare subsequent deterioration of the patient's personality (20). This point is underscored by Slater's observation that “the delusions and hallucinations of patients with POE were empathizable (the patient remains in our world)” (21).
Aggressive behavior in epilepsy as a preictal, interictal, and paraictal process
A 13-year-old adolescent boy has a history of refractory epilepsy that initially presented as gelastic seizures at the age of 3, but these were only diagnosed at the age of 6, after he experienced his first SGTCS. Since the age of 6, he has continued to present weekly clusters of gelastic and complex partial seizures, as well as SGTC, every 2–3 weeks. A high-resolution MRI study revealed a hypothalamic hamartoma. Ictal recordings obtained in the course of a V-EEG revealed bitemporal independent regional onsets, while an ictal SPECT demonstrated an area of hyperfusion in the hypothalamus.
Since an early age, this patient was reported to be very restless and impulsive, and to display poor frustration tolerance. A neuropsychological evaluation revealed mild mental retardation (full scale IQ = 68, verbal IQ = 65, performance IQ = 76). Since the time he entered kindergarten, he was placed in special education classes. Since the age of 8, his teachers described him as being oppositional, defiant, and prone to outbursts of rage that led to recurrent violent incidents with teachers and other students.
Pharmacologic treatment consisted of trials with CNS stimulants (methylphenidate and dextroamphetamine), SSRIs (paroxetine and sertraline), and neuroleptic drugs (haloperidol, risperidone, olanzapine), with very limited improvement in his behavioral problems.
Since the age of 6, his parents noticed that 2–3 days before having a SGTCS his restlessness and impulsivity worsened in severity, and he was more likely to have outbursts of rage. However, his behavior improved dramatically during the 5– to 7 days that followed a SGTCS. A vagus nerve stimulator resulted in suppression of SGTCS for a 3-month period, after which this type of seizure recurred in clusters on a monthly basis. However, during these 3 months, his behavior worsened to the point that he was suspended from school. After a resection of the hypothalamic hamartoma, the SGTCS, complex partial, and >95% of gelastic seizures remitted for 8 months. Thereafter, they recurred with one SGTCS every 6–12 months, while gelastic seizures remitted completely. During the 4 weeks that followed the surgical procedure, the parents noticed a significant improvement in his behavior at home and school, where no further outbursts of rage were reported. His academic performance also improved to a significant degree.
It is well known that behavior disturbances are more common in children with epilepsy than in the general population. In most instances, they are the expression of attention-deficit/hyperactivity disorder (ADHD), oppositional defiant disorder (ODD), or conduct disorder (22). In a population-based study carried out on the Isle of Wight in Great Britain, Rutter and collaborators found behavioral disorders in 28.6% of children with uncomplicated seizures, and 58.3% of children with both seizures and additional CNS pathology (23). In a separate population-based study of children with seizures, cardiac disorders, and controls, McDermott and collaborators found that children with epilepsy had more behavioral problems than either the children with cardiac troubles or the controls (24). The authors identified three outcome variables: hyperactive, headstrong, and antisocial behaviors. Hyperactive behavior was found in 28.1% of the children with epilepsy, 12.6% of cardiac children, and 4.9% of controls. Headstrong or oppositional behavior was seen in 28.1% of children with epilepsy, 18.3% of cardiac children, and 8.6% of controls, while antisocial behavior was found in 18.2% of epilepsy children, 11.6% of cardiac children, and 8.8% of controls.
Our patient experienced behavioral problems since early childhood that could have been diagnosed as ADHD and ODD. However, his psychiatric symptoms differed in severity, according to their temporal relationship with the SGTCS occurrence. During the 2–3 days that preceded a seizure, symptoms exacerbated in severity, while they improved significantly during the 5 days that followed the SGTCS.
Preictal dysphoria and irritability have been reported previously. Blanchet and Fromer (25), for example, studied mood changes during 56 days in 27 patients who were asked to rate their mood on a daily basis. Mood ratings pointed to symptoms of dysphoria, anxiety, and irritability 3 days before a seizure in 22 patients. These changes were more accentuated during the 24 h preceding the seizure. It is not rare for parents to comment: “I know when my son is going to have a seizure, as he gets more restless and irritable 1 to 2 days before his seizure.”
While this patient experienced a significant improvement after his SGTCS, postictal irritability, aggressive behavior, and poor frustration tolerance are common occurrences in refractory partial epilepsy. In the study on clinical characteristics and prevalence of postictal psychiatric symptoms cited previously (9), we found postictal irritability in 30 patients and poor frustration tolerance in 36, with a median duration of 24 h for each symptom (range: 0.5–108 h).
The significant improvement of this patient's behavior disturbances after the quasi-total remission of epileptic seizures with the hypothalamic resection is evidence that his psychiatric disorder was the expression of a paraictal process. This phenomenon has been described in the literature by several authors. Palmini and collaborators reported a series of 13 patients, age 2–33, who underwent surgical resection of a hypothalamic hamartoma. All of these patients had behavioral and cognitive disturbances before surgery. After surgery, three patients became seizure free, while eight had >90% reduction in seizure frequency. All of these 11 patients experienced a dramatic improvement in behavior and cognition (26). Similar results were reported by Fohlen and collaborators (27).
Psychiatric disturbances as an expression of a paraictal process can be seen in other “epileptic encephalopathies,” such as the acquired epileptic aphasia of childhood, also known as Landau-Kleffner Syndrome (28). In these patients, an epileptic seizure focus in the intrasylvian cortex results in an expressive and receptive aphasia and auditory agnosia that is associated with brief absencelike seizures or SGTCS. At the onset of the epileptic disorder, psychiatric disturbances become very obvious and often difficult to treat, consisting of motor hyperactivity, impulsivity, aggressive behavior, insomnia, and outbursts of anger. These psychiatric disturbances tend to improve significantly after remission of the epileptic activity with multiple subpial transaction of the epileptogenic zone (28). Furthermore, these changes are seen in the first 2 weeks after surgery, before any improvement in language function is evident.