A large proportion of patients undergoing video-electroencephalography monitoring (VEM) do not have epileptiform electroencephalographic (EEG) changes during their spells, and approximately one third of patients admitted to tertiary care epilepsy monitoring units are ultimately deemed to have nonepileptic spells.^1,2 In a relatively small proportion of these patients, physiological causes for their events (e.g., cardiac arrhythmia and hypoglycemia) are found, and a majority of these patients are discharged with a definitive diagnosis of psychogenic nonepileptic seizures (PNES). Many other terms, as listed in Table 20-1, are used to describe the exact same phenomenon.³ The diversity of the terms used to describe this disorder makes it difficult to gain a comprehensive survey of the available literature. Additionally, there is considerable overlap of features present in PNES and those of other functional neurologic disorders, such as psychogenic movement disorder and psychogenic pain disorder. From the observation of this overlap, a spectrum of disorders is likely represented under the umbrella of conversion and somatization disorders.

What is unique to patients diagnosed with PNES compared with other functional neurologic disorders is that dissociative symptoms often predominate in the clinical presentation. This further complicates establishing the diagnosis because behaviors similar to dissociation occur in some complex partial seizures. Indeed, PNES that include other behaviors are more reliably distinguished from epileptic seizures (Table 20-2). Currently, PNES are described in the literature as a form of conversion disorder. This leads to the issue of whether patients with PNES manifested by dissociative symptoms should be diagnosed as having a conversion disorder or a dissociative disorder that has been misdiagnosed as epilepsy. That is, the patient who presents to a psychiatrist with dissociative symptoms diagnosed as dissociative disorder may be identical to the patient who presents to a neurologist and is diagnosed with epilepsy. However, in the patient presenting to a neurologist who is diagnosed with epilepsy, the diagnosis is later corrected to PNES. This begs the question: are some forms of PNES better classified as a dissociative condition rather than a conversion disorder?

In the late 19th and early 20th centuries, dissociative symptoms were witnessed almost exclusively by neurologists.⁴ Although it is not true for all events in patients with PNES, a vast majority of such patients have dissociative symptoms. Bowman and Markand surveyed the dissociative symptoms in a series of patients with PNES and demonstrated that over their lifetime, 93% met clinical criteria for a dissociative disorder, with 91% meeting criteria the month prior to evaluation.⁵ Additionally, 67% of these patients had dissociative disorder—not otherwise specified (DDNOS), and 19% had psychogenic amnesia. Findings from this study have led clinicians and researchers to suggest that the descriptor PNES often is captured more accurately under the category of dissociative disorders rather than conversion disorders, given the high lifetime occurrence of dissociative symptoms among patients with PNES.^5–7 Unfortunately, classification systems appear to be inadequate to appropriately capture how patients manifest somatic symptoms of conversion and dissociation. Research on the etiology of PNES in patients with and without dissociative symptoms is needed to better understand and differentiate between the overlapping symptoms of these disorders.

The current version of the Diagnostic and Statistical Manual of Mental Disorders (fourth edition, text revision, DSM-IV-TR) is the most widely accepted classification scheme for the diagnosis of psychologically manifested illnesses. The DSM-IV-TR defines dissociative disorders as “a disruption of the usually integrated functions of consciousness, memory, identity, or perceptions.”⁸ In other words, dissociation can be thought of as “an alteration of consciousness that affects memory and identity.”⁹ Simply put, to dissociate is to separate the association of one thing from another. Dissociative phenomena represent a spectrum of symptoms.¹⁰ Not all forms of dissociation are pathologic. In fact, most of us have had the experience of some form of dissociation, such as daydreaming or long drives without recalling the specifics of the trip. However, dissociation can become more disruptive when there is a failure to integrate one’s thoughts, feelings, memories, and actions into a unified consciousness.¹¹ Dissociative identity disorder, for example, represents a more severe form of dissociation in which the dissociation creates multiple discrete personalities.

Dissociation is felt to be a type of consciousness either where events that occur simultaneously are separated psychologically or where events are unable to be incorporated into consciousness.¹⁰ It is generally felt that dissociative symptoms are a reaction to previous psychological stress, a defensive response to block the awareness of traumatic events.⁹ To this end, dissociation can be an adaptive means of coping with unimaginable stress. Previous theories have suggested that acute traumatic events are usually remembered quite vividly; however, chronic prolonged trauma can result in a “psychic numbing” and subsequently dissociation.⁹

There are five subclassifications of dissociative disorders, as listed in Table 20-3. A large proportion of dissociative disorders are within the DDNOS category. DSM-IV-TR specifies that dissociative symptoms are also found in other psychiatric diagnoses, including acute stress, posttraumatic stress, somatization, and conversion disorders. However, the DSM-IV-TR specifies that “when both conversion and dissociative symptoms occur in the same individual, both diagnoses should be made.”⁸ Many patients present with PNES as their underlying dissociative disorder.⁷ The DDNOS classification includes loss of consciousness not due to a discernible medical illness, perhaps making this an appropriate diagnosis when dissociation presents as PNES. However, not all PNES events present with true dissociation and would not meet clinical criteria for a dissociative disorder. The frequency of dissociative disorders classified as DDNOS reflects the difficulty of accurately describing and differentiating dissociative disorders. Aside from DDNOS, conversion disorder is the closest psychiatric classification to that of the neurologic diagnosis of PNES for most patients. In cases where PNES do not meet clinical criteria for a dissociative disorder, they may be more appropriately classified as a conversion disorder with dissociative features. The DSM-IV-TR diagnostic criteria for conversion disorder are detailed in Table 20-4. Much of the controversy in classifying PNES comes from debates about the DSM-IV-TR‘s classification of PNES as a conversion disorder subsumed under the category of somatoform disorders versus the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10), which subsumes conversion disorder under dissociative disorders.¹²

There has been little research on the underlying neural substrates in psychogenic seizures. Because it is felt that consciousness “splits,” leading to dissociative symptoms, the correlate from a neurophysiological standpoint would be a blocking of communication from the dominant hemisphere (i.e., verbal consciousness) with the other areas (unconscious) of the brain.¹³ Other physiological obscurations have been noted in somatization disorders. Specifically, abnormalities have been demonstrated in the hypothalamic-pituitary-adrenal axis.^10,14 In chronic stress states, adrenal activity is eventually suppressed, resulting in low cortisol levels.¹⁴ The direct effect of this has not specifically been established, but symptoms could be the result of loss of protective functions of cortisol.

Structural and functional imaging studies have shed some light on psychiatric disorders and may be relevant to pathologic dissociation. Dissociative disorder is associated with small hippocampi and amygdalae, which may be due to early life stress.¹⁵ In mood disorders and schizophrenia, hypometabolism is seen in the frontal lobes. Additionally, impaired amygdalar function, as seen with positron emission tomography (PET), has been postulated to interrupt formation of emotional memory by disrupting the connections with the prefrontal cortex. Functional magnetic resonance imaging (fMRI) of obsessive-compulsive disorder has shown posterior cingulate gyrus signal change activated in response to stress, which may result in a disruption of the Papez circuit and difficulty in laying down new memories.¹⁴ Obviously, because of the lack of definitive data, these observations are extrapolated from the psychiatric literature. However, the data that are known work well in the framework of what we know about neuroanatomical functioning of the brain, aiming for a more detailed understanding of the psychopathophysiological mechanisms involved in PNES.