Psychogenic Non-Epileptic Seizures

Andres M. Kanner

W. Curt LaFrance Jr.

Tim Betts

Introduction

It has been known since the earliest recorded medical writings⁵¹^,¹⁰⁰^,¹⁴¹^,¹⁴⁸ that not all seizures in humans are of epileptic origin. From ancient times onward, instruction was given in medical texts about how to determine the clinical difference between epileptic and nonepileptic events. Many modern authorities make an assumption that it is only in recent years, with the advent of the electroencephalogram (EEG), that we have been able to detect this difference, but this is not so. Our forbearers had their own conception of what was epilepsy and what was not. Gowers’ meticulous description of the hysterical convulsion bears a close relationship to a type of nonepileptic seizure that we recognize today.

The advent of the EEG and, from the late 1960s, continuous EEG monitoring (with or without video recording) led to an increased interest in distinguishing between seizures of epileptic and nonepileptic origin. It is undoubtedly true that the ability to monitor an EEG over a long period of time and register an EEG during a seizure (and to be able, in tranquility, using video, to retrospectively look at the behavior that the patient displayed during a seizure) has enormously helped our ability to distinguish between events that are epileptic and those that are not.

Nonepileptic seizures (NESs) are paroxysmal events that mimic (or are confused with) epileptic seizures, but which do not result from epileptic activity. NESs can be the expression of organic or psychogenic processes. The type of organic disorders often confused with epileptic seizures include convulsive syncope, various forms of sleep and movement disorders, and migrainous processes and are reviewed in Section XI of this book. The decision as to whether a patient’s seizures belong in the domain of epilepsy or nonepileptic events may have to be based on various sets of criteria, in addition to the EEG data. The distinction between epilepsy and nonepilepsy cannot always be made with complete confidence, and the physician working in this field must be able to tolerate some degree of uncertainty. Epileptic and nonepileptic seizures also may coexist. This chapter will focus on nonepileptic seizures of psychogenic origin.

Nomenclature

The most commonly used term for nonepileptic seizures is pseudoseizure; this term first came into use in the 1960s and was widely used in the 1970s and 1980s. Yet, psychogenic NESs have been labeled with various eponyms over the last two centuries. For example, Charcot referred to these attacks as “hysteroepilepsy”⁷⁷ in the 19th century, and he had recommended the use of ovarian compressors for their treatment, based largely on the theory of the ancient Greeks that such attacks were due to the “wandering of the uterus throughout the body, away from its normal location in the pelvis.” Other terms used more recently have included nonepileptic seizures, psychogenic seizures, and hysterical seizures.

There is an increasing tendency to abandon the term pseudoseizure as it is considered to have a pejorative connotation and to imply falseness. Deliberate deceit plays only a small part in the genesis of most nonepileptic seizures (i.e., cases of malingering),²² although, sadly, many clinicians think otherwise. The term pseudoseizure also conveys the implication that anything that is not epilepsy is not interesting and that no further attention need be paid to the attacks.

There is an ongoing debate on the use of the term seizure as many patients are unable to discriminate between the concepts of psychogenic versus epileptic processes and get fixated on the word seizure. To avert such potential confusion the term psychogenic nonepileptic event has been suggested. Such terms help differentiate the nonepileptic nature of these attacks from epileptic seizures and convey their probable psychogenic cause.

Psychogenic seizures is a term much used in the United States and on the continent of Europe to describe psychologically based nonepileptic seizures. It is correct providing that there is agreement about what it means and the range of disorders it covers. Unfortunately, the term also has been used to refer to those epileptic seizures that are triggered by an emotional experience.⁴⁴ Potential confusion could be avoided by using the terms nonepileptic psychogenic seizure and epileptic psychogenic seizure.

The terms hysterical seizure, hysterical epilepsy, and hysterical convulsion should not be used. The word hysteria is no longer used in psychiatric parlance because it is meaningless. It has been so much abused and misused that it should be abandoned,¹⁰⁰ and the term conversion disorder should be used instead. Not all nonepileptic attacks are related to conversion disorder.

Functional seizure is also sometimes used, but the term is imprecise and has more than one meaning. An epileptic seizure, for instance, that occurs or is precipitated at a propitious time (e.g., before a stressful event) so that the patient avoids a threatening situation could legitimately be called functional. Blumer³⁶ has suggested the term paroxysmal somatoform disorder. In some ways this is an attractive term. However, it does imply that all nonepileptic events of psychological origin relate to the somatoform disorders when they clearly do not. The term could be used as part of a classification of psychologically based nonepileptic seizures, and this would be more acceptable.

Psychogenic nonepileptic seizure (PNES) is the term that has been chosen for this chapter and is coming into common use. It may not be ideal, but it is less pejorative and more meaningful than pseudoseizure; however, clinicians must always ensure that the patients and family understand that the
inclusion of the term seizure does not imply a diagnosis of epilepsy.

Epidemiology of Psychogenic Nonepileptic Seizures

There are relatively little data on the incidence and prevalence rates of PNESs. Various estimates of prevalence have been given by various investigators in different settings. It is important to try to determine the influence that the setting has both on the way that seizures present and on their frequency. Prevalence will be different if patients with new-onset seizures are being assessed in a primary care setting compared with reassessment of patients who have an intractable seizure disorder in a tertiary referral center.

An indirect estimate of the possible incidence of nonepileptic seizures in a primary care setting can be derived from the work of Sander et al.,¹²¹ which was a careful phenomenologic study of patients presenting to their general practitioner with a seizure disorder and who were carefully assessed and followed. It was shown that some 28% of patients suspected of having epilepsy after presentation to their general practitioner either did not have it (7%) or the diagnosis could not be substantiated even a year after the initial presentation (21%). In the only nationwide population-based study done to date, Sigurdardottir and Olafsson found the population-based incidence of PNESs to be 1.4 per 100,000 in Iceland, compared to an incidence of 35 per 100,000 for true epilepsy among persons older than 15 years.¹²⁹ In a retrospective study carried out in Hamilton County, OH, investigators established the incidence of PNESs between 1995 and 1998. They found a mean incidence of 3.03 per 100,000, with the highest incidence in 1998 (4.6 per 100,000). Most patients with the diagnosis of PNES were aged 25 to 45 years (4.38 per 100,000).¹³⁸

The prevalence rates of PNESs have been estimated to range from 10% to 40% among patients referred to epilepsy centers for the evaluation of poorly controlled seizures.²⁶^,³⁰^,⁷⁴^,⁷⁷ The prevalence of comorbid PNES and epilepsy varies according to the type of population studied. For example, in a study of 1,590 patients who underwent a video-EEG (V-EEG) monitoring study at the Epilepsy Center of the University of Alabama in Birmingham, 514 (32.3%) were diagnosed with PNES and 29 (5.3%) of these patients were found to have both PNES and epilepsy.⁹⁶ Henry and Drury conducted a V-EEG in 145 patients who had temporal interictal EEG spikes and reported ictal semiology characteristic of temporal lobe seizures for presurgical evaluation of medically refractory seizures. PNESs were unexpectedly identified in 12 (8%) of these patients.⁵⁷ On the other hand, a study from Germany found that among 329 consecutive patients in whom the diagnosis of PNES was established, 206 (62%) had only PNES, and 123 (37%) had PNES and epilepsy.¹¹⁸ In a critical review of the literature, Ramsey et al. came to the conclusion that the comorbid occurrence of epilepsy and PNES was rare with prevalence rates ranging between 4% and 10%.¹¹⁶ Relatively high prevalence rates can be expected when studies are carried out in populations of cognitively impaired or in pediatric patients, however. Neill and Alvarez, for example, reported prevalence rates of up to 40%.¹⁰⁴ Kotagal et al. found that 25% of pediatric patients with PNES also suffered from epilepsy.⁷⁵ Higher prevalence rates of comorbid PNES and epilepsy in these two patient populations is not surprising, as they learn that fewer demands are placed on them when they have a seizure. Hence, they are likely to experience a PNES when facing situations that cause distress or that they consciously or unconsciously are trying to avoid.

PNESs are much more common in women than men; in most series 60% to 75% of patients are women.¹²^,⁵⁸^,⁷⁷^,⁷⁸^,⁹¹^,⁹³^,¹¹⁰^,¹¹⁷ The onset of PNESs most often occurs in the third and fourth decades of life,¹²^,⁵³^,⁷⁷^,⁸⁹ though they have been described in both children and the elderly.⁷^,⁴²^,⁶⁸^,¹⁵²^,¹⁵³

Clinical Manifestations

PNESs can mimic convulsive and nonconvulsive epileptic seizures; they may present as isolated events or occur in clusters and, often, may mimic status epilepticus. The advent of V-EEG has facilitated the recognition of PNESs among neurologists and epileptologists, and it appears that nonneurologists are increasingly becoming more aware of this type of paroxysmal event. There have been multiple review articles on the diagnosis of PNES⁷^,²¹^,⁵⁴^,⁶⁶^,⁶⁸^,⁷¹^,⁹²^,¹²⁶ and various investigators have tried to emphasize the clinical differences between PNESs and epileptic seizures with lists and tables of distinguishing features.⁵³^,⁷⁸^,⁸⁹ While certain features of convulsive PNESs may be highly suggestive of this diagnosis, there are no clinical phenomena that are 100% specific to one or the other.¹⁵¹ Because our ability to recognize PNESs is not infallible and in some patients, with different kinds of events, it is impossible to decide if they have epilepsy or not, a diagnosis cannot be established on the bases of clinical phenomena alone, but has to be corroborated with electrographic recordings, and occasionally, functional neuroimaging studies with ictal single photon emission computed tomography (SPECT) may be necessary to reach a correct diagnosis (see below). In short, the diagnosis of a PNES is based on a considered judgment of all relevant history, seizure phenomena, electrophysiologic data, and results of other investigations and remains a clinical one. Using one or two phenomena in a differential diagnosis table to make a rapid diagnosis would be injudicious and could lead to inadvertent misdiagnosis. Furthermore, several studies published in the last 15 years have demonstrated that clinical phenomena associated with PNESs can occur in certain types of epileptic seizures, a fact that still remains underrecognized by many clinicians, including neurologists, and has often lead to the inverse diagnostic problem (i.e., patients with epilepsy being falsely diagnosed with PNES). For example, in a study of 100 consecutive patients that were undergoing a diagnostic V-EEG, Parra et al.¹¹⁰ found that referring physicians correctly suspected a diagnosis of epileptic seizures in only 9 (43%) of 21 patients, while 12 (57%) patients were incorrectly thought to have PNESs. This misdiagnosis was especially likely in patients with clinical seizures of mesial frontal or parietal lobe origin. To confuse matters further, there are a number of patients who suffer from both epileptic seizures and PNESs, with coexistence of both types of disorders ranging from 7% to 37%. Most studies have found incidence rates ranging between 10% and 20%.¹⁴^,³⁸^,⁵⁸^,⁷²^,⁹⁰ This point is examined in greater detail in Chapter 208.

Needless to say, reaching a correct diagnosis is of the essence as a misdiagnosis of PNESs as epileptic seizures, and vice versa, exposes patients to ineffective, costly, and potentially dangerous treatments. Indeed, several investigators have found that 69% to 78% of patients with PNESs have been treated with antiepileptic drugs (AEDs)¹⁴^,¹²⁰ and up to 30% of patients have been misdiagnosed at some point with status epilepticus and admitted to an intensive care unit where they were treated with parenteral medications that are associated with potential serious adverse events, and which in some cases caused respiratory arrest.⁵⁹^,¹¹⁷^,¹²⁰

In general, a diagnosis of PNES is easier to suspect in convulsivelike events, while in nonconvulsive events a diagnosis of PNES may be impossible without concurrent electrographic recordings. In the next section we review the value and
limitations of clinical phenomena in suspecting the diagnosis of PNES.

Convulsive Psychogenic Nonepileptic Seizures

Motor Phenomena

In convulsivelike PNES, motor signs include cloniclike, myo-cloniclike, and toniclike movements of the extremities and trunk. However, violent thrashing of the extremities and/or of the entire body, opisthotonic arching of the back, pelvic thrusting motions, and side-to-side head movements are the typical motor phenomena associated with PNESs. Frontal lobe seizures have been shown to present with the same type of motor phenomena; for example, Geyer et al. found pelvic thrusting to be equally common in frontal lobe seizures and PNESs⁴⁸ (also see below).

Out-of-phase and asynchronous cloniclike movements in the upper extremities (as opposed to in-phase and synchronous during epileptic seizures) is another “typical” motor sign suggestive of PNESs.⁴⁶ However, Leis et al. were not able to make this same distinction in their analysis of the motor phenomena of PNESs.⁸⁹

The absence of any facial cloniclike activity in the presence of generalized cloniclike activity can be another “helpful” sign in suspecting a diagnosis of PNES. Family members need to be instructed ahead of time to look for such sign, however.

Incontinence and Self-injury

Contrary to old beliefs, patients with PNES do get hurt. Indeed, in a study by Reuber et al., almost 60% of patients had experienced PNES-related injuries, 32.3% had reported urinary incontinence, and 31.5% tongue biting (patients with both PNESs and epileptic seizures were not included in these percentages).¹²⁰ Incontinence has been reported to range between 10% and 44% in PNESs, most of which consist of urinary incontinence, but fecal incontinence has been also described.⁹⁵^,¹¹²^,¹²⁰ Bruising and minor lacerations from falls followed by tongue biting (that may be severe enough to cause bleeding) are among the more common injuries reported in patients with PNESs. Tongue biting, which often does occur in a tonic–clonic seizure, also can occur in complex partial seizures, in faints (if the patient falls and bites the tongue as a result of the fall), and in PNESs (either as a result of falling or as a result of deliberate injurious behavior, not necessarily to simulate a seizure but because the patient is responding to inwardly driven motivation to try to hurt himself or herself). The site of the bite may yield clues but can be misleading. In a tonic–clonic seizure the bite is usually on the side of the tongue and is deep, although it may occur at the tip of it. In tongue injury as a result of falling, the bite is usually at the tip. In complex partial seizures the bite may be at the tip or occasionally at the side of the tongue or may involve the cheek. However, patients having PNESs also can bite the side of the tongue, tip of the tongue, and cheek. Finally, serious injuries, including facial bone fractures, are less common but have also been reported and tend to occur in patients with a psychiatric history of suicide attempts.¹¹²

Vocalizations

Vocalizations can be often seen in convulsivelike PNESs, but occasionally occur in non–convulsivelike events. Vocalizations associated with shedding of tears tend to suggest a strong possibility of PNESs.¹⁷ In convulsivelike events, vocalizations consist of shouting, screaming, and sobbing often associated with understandable speech.⁴⁶^,⁵³^,⁸⁹^,⁹⁵ In contrast, in epileptic seizures, vocalizations are typically more primitive, such as grunting or a simple shout.⁴⁶ Vocalizations are more likely to occur in the middle of PNESs, unlike epileptic seizures, where vocalizations usually occur at the onset of the event.

Nonconvulsive Psychogenic Nonepileptic Seizures

Nonconvulsive PNESs can mimic either absence and/or partial complex seizures; more often than not, a correct diagnosis is unlikely to be reached until patients are referred for V-EEG with a presumed diagnosis of “intractable epilepsy.” Patients may be unresponsive while exhibiting a motionless stare and they may exhibit semipurposeful movements simulating motor automatisms of complex partial seizures.⁵³ Maintenance of eyes closed and responsiveness to “selective” external stimulation, including verbally, in the course of an event during which they “appear” to be unresponsive are two clinical “clues” that should raise the suspicion of a PNES. More commonly, patients remain nonverbal but display active avoidance to both noxious and nonnoxious stimuli and/or resist attempts at eye opening, but this type of avoidant behavior may also be identified in complex partial seizures.

Other Clinical Characteristics of Psychogenic Nonepileptic Seizures

Sudden collapse to the ground with apparent unconsciousness but no movement can be part of epilepsy and can occur without warning. It is usually a primary generalized seizure (either a tonic seizure in which the patient falls stiffly or an atonic one in which the patient collapses with sudden loss of muscle tone). In both seizures recovery of consciousness (unless the patient has injured the head in falling) is swift and there is little postictal confusion (unless, again, the patient’s head has been struck in the fall). Occasionally sudden loss of consciousness with falling can be seen in seizures originating in the temporal or frontal lobe.²⁸ The fall may be flaccid (but can occur with a tonic spasm), and consciousness usually is regained fairly quickly. The patient who falls in a flaccid way and remains apparently unconscious for some time is unlikely to have epilepsy. Epileptic seizures with falls, if monitored via EEG, may not be detected either because the patient is in bed and the characteristic falling is therefore not seen, or because the fall itself produces such artifact in the recording device that the ictal nature of the event is obscured. It also is possible that the preliminary collapse was due to epilepsy but the subsequent withdrawn behavior is a psychological elaboration of the brief epileptic experience. This may not be an uncommon occurrence.

Psychogenic Nonepileptic Seizures as “Drop Attack”

Physical and emotional causes of collapse are considered in other chapters (e.g., syncope in Chapter 271 and hyperventilation in Chapter 285). A not-uncommon cause of collapse and flaccid unconsciousness of a psychological type of nonepileptic seizure is the so-called swoon.¹⁹ Here the patient, when threatened by external events or unpleasant memories or by a situation reminiscent of an unpleasant experience, will close the eyes, sink to the floor, and lie inert and not moving for various lengths of time, sometimes for over an hour. Passive movement and eye opening may be resisted, but usually the patient just lies inert and unresponsive and comes to quite quickly with little in the way of postictal symptoms. Presumably this is a form of learned behavior; it is a common childhood behavior that may have persisted into adult life. There is some evidence²⁰^,⁶¹ that it also may be a common mechanism for avoiding flashbacks and reawakened memories in adults with a chronic posttraumatic stress disorder related to previous childhood sexual abuse,
particularly if it was the way that the individual dealt with the abuse at the time it was occurring. Some abused people cope with their abuse by lying inert and retreating into themselves or by splitting their consciousness so that they almost seem to be watching events from a corner of the room and become passive observers of themselves; this may become a persistent unconscious learned pattern of behavior to deal with stressful events and later may be mistaken for epilepsy.

Deliberate Simulation of Unconsciousness

Deliberate simulation of unconsciousness, which may involve falling, is part of the repertoire of people with somatoform disorders or with factitious disorder or malingering (deliberately simulated disease). In the case of factitious disorder, individuals simulate illness to maintain the sick role; malingerers do so for social or financial gain. It is easy to sustain for long periods of time and difficult to detect, particularly if the patient does not resist passive movement or eye opening and does not respond to noxious stimuli. Such behavior (often seen as attention-seeking behavior in nurses and other caring professionals who are particularly prone to factitious disorder) is sometimes mistaken for epilepsy. Some professional simulators may inject themselves with insulin beforehand to add further authenticity to their performance.

Nonstereotypic Events?

PNESs have typically been thought of as nonstereotypic events (within patients), while the reverse has been considered to be a clinical sign supportive of a diagnosis of epileptic seizures. While the absence of a stereotypic semiology within patients should raise suspicions of a diagnosis of PNES, the reverse does not rule out this diagnosis. In fact, in four studies, 60% to 90% of patients with PNESs had spells that varied little from event to event.³⁸^,⁴⁸^,⁵³^,⁹⁹ However, even when clinical features and their sequence is relatively stereotypic, the duration of seizures in patients with PENSs tends to be more variable than in those with epileptic seizures.

Events of Long Duration

PNESs are more often than not significantly longer in duration than epileptic seizures and often lead to a misdiagnosis of status epilepticus (see below). A rapid recovery of cognitive functions following a prolonged convulsivelike or non–convulsivelike event should raise suspicions of possible PNESs (see below).

Psychogenic Nonepileptic Seizures Only Occur in Awake State

PNESs can only occur when the patient is awake, and events occurring out of sleep, documented with electrographic recordings, cannot be considered as PNESs. Having said that, clinicians cannot accept the patients’ assertion of events occurring out of sleep without electrographic corroboration, as PNESs have been described in patients that “appear to be asleep” but in reality are awake.¹⁶

Psychogenic Nonepileptic Seizures Mimicking Status Epilepticus

One of the most serious consequences of PNES is their mimicking “status epilepticus” that is not properly diagnosed, which in turn results in unnecessary admissions to intensive care units, aggressive use of parenteral AEDs, and placement of endotracheal tubes. Iatrogenic damage (respiratory arrests, toxic effects from anticonvulsants) are common, and because PNESs do not respond to AEDs, prolonged hospital stays with expensive and dangerous treatments (including paralysis, anesthesia, and assisted ventilation) often occur.⁵⁹^,¹¹⁷^,¹²⁰ For example, in a study by Reuber et al., 51% of patients with PNESs presented as “pseudostatus” (lasting more than 30 minutes) and 27.8% were admitted to intensive care units.¹²⁰ Howell et al.⁵⁹ suggested that perhaps 50% of patients admitted to emergency care in the United Kingdom in status epilepticus do not actually have epilepsy.

Some investigators have suggested that PNES mimicking status epilepticus occurs mainly in women whose seizures relate to a posttraumatic stress disorder from previous sexual abuse,²⁰^,²⁹ in individuals with hyperventilation attacks that are easily prolonged by inadvertent reinforcement, or in patients with conversion or factitious disorder (e.g., Münchhausen syndrome or malingering). Accordingly, emergency room staff should be encouraged to observe seizures critically before starting treatment, to use a pulse oximeter to distinguish patients who need immediate intubation from those who can be safely left to breathe without assistance, and to obtain arterial blood gases, which may also detect those who are hyperventilating.

Epileptic Seizures Mimicking Psychogenic Nonepileptic Seizures

As stated above, a study by Parra et al.¹¹⁰ suggested that as clinicians have become increasingly aware of PNESs, epileptic seizures with atypical features are being misdiagnosed as PNESs. The error rate in diagnosing nonepileptic seizures in patients who clearly have epilepsy probably lies between 5% and 10%.¹¹⁶^,¹⁵¹ Complex partial seizures of mesial frontal origin can present with complex automatisms, very often with “bizarre” features that mimic the typical phenomena of PNESs. These include kicking or pedaling motions of the lower extremities; thrashing and flailing motions of the extremities or entire body, at times with very violent thrusts; and affect-laden vocalizations, such as screams and loud cries. Other phenomena that may contribute to the diagnostic confusion include a preserved consciousness in seizures originating in supplementary sensory motor area, despite a bilateral tonic posturing, and the absence of postictal lethargy or confusion. Kanner et al.⁷⁰ compared the clinical phenomena of supplementary sensory motor seizures with those of PNESs and found clinical phenomena suggestive of PNESs in 82% of all seizures recorded in 91% of the frontal lobe epilepsy patients studied.

The atypical clinical phenomena of these seizures are frequently coupled with undetected interictal and ictal epileptiform activity on scalp recordings. The difficulty in identifying any epileptiform discharges results from the great amount of muscle and movement artifact masking the underlying electrical activity. The inability of the angle subtended by scalp electrodes to identify the source of the epileptic activity in mesial or orbitofrontal regions is another contributing cause to this problem. Thus, it is not surprising that these seizures are frequently confused with PNESs.⁷⁰^,¹⁰³^,¹²³^,¹⁴⁷^,¹⁵¹ Morris et al.¹⁰³ have demonstrated the presence of a subtle rhythmic theta pattern in parasagittal leads, buried within muscle and movement artifact. The use of high-frequency filters may be necessary to recognize such pattern. In addition, the presence of a structural lesion on magnetic resonance imaging (MRI) in mesial or orbitofrontal regions should raise suspicions of epileptic seizures in patients diagnosed with PNESs.

There are certain clinical features that may help the clinician distinguish this type of frontal lobe seizure from PNES: (a) frontal lobe seizures are very short in duration (<30 seconds); (b) they are stereotypic, including the complex automatisms; (c) they often occur out of sleep, whereas PNESa always occur in the awake state; and (d) they often display a tonic posturing in abduction of the upper extremities, a sign that was found to be specific to epileptic seizures involving the supplementary sensory-motor area and never seen in PNES.⁷⁰^,¹⁰³^,¹⁴⁷ A
cautionary note is in order: PNESs may be reported by patients to arise out of sleep, but V-EEG monitoring demonstrates an awake EEG pattern preceding the onset of the PNESs.¹⁶^,¹³⁹ Benbadis et al.¹⁶ coined the term “pre-ictal pseudosleep” to describe the state of wakefulness while appearing asleep in these patients with PNES.

Comorbidity of Epilepsy and Psychogenic Nonepileptic Seizures

This topic is reviewed in great detail in Chapter 208. Briefly, quoted rates for the coexistence of epilepsy and PNES vary from <10% to >90% of the population studied.⁶⁶^,⁶⁸^,⁷¹^,⁹²^,¹¹⁶ The literature is confusing because different populations of patients have been studied by different investigators and because of the uncertainty about the diagnosis of PNESs. It is impossible to be dogmatic about how commonly patients with established epilepsy develop PNESs, particularly because the association at times must be purely coincidental, but coexistence of epilepsy with PNESs is not a common problem. Probably no more than 5% of people with established epilepsy develop nonepileptic seizures.

In the literature, completely different methods have been used for diagnosing nonepileptic seizures, based sometimes on EEG criteria (but not necessarily EEGs recorded during the ictus itself), on clinical criteria, or on a mixture of both. It is also sometimes not clear, when the prevalence of the two conditions is compared, whether the investigator is referring to patients with nonepileptic seizures and present epilepsy or merely a history of it in the past. It is likely, as Ramsey et al. have concluded,¹¹⁶ that if the history is one of past as opposed to present epilepsy, the proportion of patients with both epilepsy and nonepilepsy is likely to be higher, particularly in patients with learning difficulty.¹⁰⁴

Sometimes the question can only be answered by withdrawing the antiepileptic medication and seeing if epilepsy re-emerges. If it does, one must pose the question, “Was this all epilepsy that I was trying to treat, or am I dealing with somebody who has controllable epilepsy but who has a psychological need for it?”

A review of the literature suggests that 10% to 30% of patients who appear to have established nonepileptic seizures also have a past history of epileptic seizures, which means that even when it is obvious that a patient’s seizures are nonepileptic, slow and cautious reduction of anticonvulsant medication is indicated.⁶⁶^,⁷¹ Rapid reduction of anticonvulsant medication, even in patients who do not have epilepsy, may precipitate pharmacologic withdrawal seizures, which will make the picture even more difficult to assess. A final consideration: The literature suggests that if one is assessing a patient with an unknown attack disorder that has two different presentations, 5% to 30% of the time one may be dealing both with epilepsy and a nonepileptic seizure.¹¹⁶

Diagnostic Evaluation of Psychogenic Nonepileptic Seizures

Neurologic Workup

The diagnostic workup of PNES consists first and foremost of neurophysiologic studies aimed at capturing the patient’s typical events and documenting the absence of a concurrent epileptic ictal pattern. However, the fact that a patient has PNES does not rule out the presence of comorbid neurologic disorders, including a coexisting or past epileptic seizure disorder. In fact, people with a correct diagnosis of NES have been found to have a clearcut history or evidence on examination of cerebral injury,⁷²^,⁷⁸^,¹¹⁸ whereas a significant percentage of people with epilepsy have no such history. Once the diagnosis of NES is established, a psychogenic cause must be investigated with psychiatric and neuropsychological evaluations. Yet, although many people with proven NES have a recognizable psychiatric illness or history or evidence of emotional precipitation of seizures, so do many people with epilepsy. The diagnosis of an NES cannot depend solely on these features but is based on the judicious and considered review of all the evidence available to the clinician. This is particularly important when certain psychological or psychiatric conditions are present in a patient who has an undiagnosed attack disorder. For example, a history of sexual abuse is not uncommon in people with proven epilepsy and may be pathoplastic in the seizure content.¹⁹^,⁵²

The neurologic evaluation of patients with suspected PNES must include V-EEG or ambulatory EEG studies to capture the typical event, and neuroimaging studies including brain MRI to rule out the presence of structural pathology that may be associated with a past or present comorbid neurologic disorder. In addition, the presence of a structural lesion in mesial-frontal regions of the brain may alert the clinician to be cautious on reaching a false-positive diagnosis of PNES and rule out the presence of epileptic seizures mimicking PNESs. In some patients functional neuroimaging studies such as ictal SPECT may be necessary when the distinction between PNES and epileptic seizures with atypical features may be difficult.

Video-electroencephalographic Monitoring

The electrographic diagnosis of PNESs is based on the recording of one of the patient’s typical events in the absence of any electrographic change from their baseline “awake” activity. To ensure that a typical event has been captured, the recordings have to be shown to family members who have witnessed the patient’s events. Furthermore, it is important to enquire from these family members about the occurrence (in the present or in the past) of events different than those captured during the V-EEG. This will minimize the risk of failing to identify a comorbid epileptic seizure disorder that may be either occurring simultaneously with PNES or a prior seizure disorder that may have remitted with AEDs. Clearly, such information is pivotal on the decision to discontinue AEDs.

Only gold members can continue reading. Log In or Register to continue