Epileptologists frequently encounter patients who present with paroxysmal events that, despite resembling epileptic episodes, are actually nonepileptic. Indeed, as many as 50% of patients referred to specialist epilepsy centres may turn out not to have epilepsy (Francis and Baker, 1999). While some nonepileptic seizures may be attributable to physical causes other than epilepsy (see Gates and Erdahl, 1993), a demonstrable organic basis is absent in many such cases. Of these, some are attributable to an identifiable psychiatric illness, such as psychosis, that can produce seizure-like symptoms. In other cases, however, nonepileptic seizures4 occur as an isolated psychiatric problem in their own right. Identifying such cases represents a considerable challenge to neurologists working within this domain. At present, there are few, if any, reliable criteria for an inclusive diagnosis of nonepileptic attack disorder. As a result, current diagnostic practice is essentially based on the exclusion of epilepsy and other physical disorders (see Brown and Trimble, 2000). The concept of dissociation is particularly important in relation to nonepileptic attacks as it sheds light on both the mechanisms and, potentially, the differential diagnosis of these phenomena.
Although nonepileptic seizures have a presentation as diverse as that associated with epilepsy itself, it is possible to impose some order on their general semiology. In a sample of 110 patients with well-documented nonepileptic attacks, Meierkord et al. (1991) report that approximately one-third of all cases involved a collapse with limpness, while two-thirds involved prominent motor activity such as limb thrashing. In this study, decreased responsivity to verbal stimulation was evident in three-quarters of all cases, while purposeful or semi-purposeful motor behaviour
4 Much has been written about the relative merits of the various terms used to describe these phenomena. Many have argued that terms such as 'pseudoseizures' are pejorative because they imply that the events in question are not subjectively compelling (Betts, 1990). In contrast, the neutral alternative term 'nonepileptic seizures' has been criticized for being imprecise, as it fails to distinguish between the many different types of paroxysmal events that are nonepileptic (Kuyk et al., 1999). My own view is that the term 'somatoform seizures' would be more appropriate than either of the above, as it is nonpejorative, has descriptive precision and emphasizes the importance of excluding physical illness in the differential diagnosis of these events. For the sake of descriptive continuity, however, the terms 'nonepileptic seizures' and 'nonepileptic attacks' will be adopted here; the term 'nonepileptic attack disorder' will be used to describe the condition characterized by these events. In the present context, it should be assumed that the term refers specifically to those events that cannot be attributed to either an organic cause or an identifiable psychiatric illness.
was observed in 44%. Just over two-thirds of all cases presented with stereotyped seizures. These observations are broadly consistent with those reported by Betts and Boden (1992). In this study, three types of 'emotional' nonepileptic attacks were identified in addition to those deliberately simulated for primary or secondary gain and those attributable to a recognizable psychiatric attack disorder (e.g. panic disorder). Approximately 21% of these individuals experienced so-called swoon attacks, in which the individual characteristically sinks to the floor and lies inert and unresponsive for the duration of the attack. Roughly 33% displayed so-called tantrum attacks, involving a sudden drop to the floor followed by screaming and thrashing of the limbs. Finally, 46% displayed so-called abreactive attacks, involving gasping, limb-thrashing, pelvic thrusting and stiffening of the body with back arching.
In all cases, nonepileptic attacks involve a temporary loss of behavioural, sensory or cognitive control that occurs in the context of intact neuropsychological functioning, as evidenced by a normal EEG during the nonepileptic ictus. The absence of paroxysmal brain discharges serves as the principal feature that distinguishes nonepileptic from 'genuine' epileptic events. By itself, however, the EEG cannot provide a completely reliable basis for the identification of epileptic and nonepileptic seizures (Brown and Trimble, 2000), underlining the potential value of dissociation as a criterion for an inclusive diagnosis of nonepileptic attack disorder.
Several converging lines of evidence indicate that these events involve a dissociative psychological mechanism (see Kuyk et al., 1997). In the first instance, nonepi-leptic seizures are commonly found in the context of other forms of dissociative psychopathology. Bowman (1993) and Bowman and Markand (1996) found that the vast majority of individuals with nonepileptic attacks meet criteria for DSM-IV dissociative disorders such as dissociative amnesia, identity disturbance and depersonalization. Post-traumatic stress disorder, commonly assumed to involve a dissociative mechanism, was also particularly common in this group of patients (Bowman, 1993; Bowman and Markand, 1996). Other studies have found that nonepileptic seizures frequently occur alongside other unexplained physical symptoms (Krishnamoorthy et al., 2001; Meierkord et al., 1991), suggesting that they may be one aspect of a broader tendency to express psychological distress somatically, so-called 'somatization' (Lipowski, 1968). A number of authorities have suggested that dissociation is an important aspect of this phenomenon also (Nemiah, 1991). Eating disorder symptoms, which have been linked to a dissociative process (Pettinati et al., 1985), also appear to be particularly common in patients with nonepileptic seizures (Krishnamoorthy et al., 2001).
The frequent co-occurrence of dissociative psychopathology in patients with nonepileptic attacks appears to indicate a general propensity for dissociative exper iences in these individuals. Consistent with this notion is a recent study by Kuyk et al. (1999), showing that individuals with nonepileptic attacks display elevated levels of hypnotic susceptibility. In a related vein, nonepileptic attacks can, in many such individuals, be provoked using suggestion, placebo or hypnosis (Dericioglu et al., 1999). High hypnotic susceptibility is commonly found in patients with dissociative psychopathology (Frischholz et al., 1992; Pettinati et al., 1985; Spiegel et al., 1988), and a dissociative interpretation of hypnosis has been offered by a number of authorities (Hilgard, 1977; Woody and Bowers, 1994). Bowman (1993) also found that individuals with nonepileptic seizures yield elevated scores on the Dissociative Experiences Scale (DES; Bernstein and Putnam, 1986), a self-report measure assessing everyday occurrences of dissociation, compared with nonclini-cal controls. However, in a more recent study, Alper et al. (1997) found that DES scores are also elevated in patients with complex partial seizures (see also Devinsky et al., 1989); indeed, there was no significant difference in overall DES scores between these patients and a group with nonepileptic seizures. Nevertheless, both epileptic and nonepileptic groups scored higher on the DES than typically observed in nonclinical populations. In my view, this particular finding demonstrates the danger of conflating the various definitions of dissociation within a single measure such as the DES. As the DES treats dissociation as a unitary concept, it cannot differentiate between conditions that are characterized by different forms of dissociative phenomena, such as epilepsy and nonepileptic attack disorder.
It is widely thought that traumatic experiences precipitate the development of dissociative symptoms, which serve a defensive function that protects the individual from extreme anxiety and psychological disintegration. Indeed, both DSM-IV and ICD-10 make an explicit link between traumatic events and the onset of dissociative symptoms. Moreover, a number of studies have found disproportionately high rates of physical, sexual and emotional abuse in patients with dissociative disorders (Chu and Dill, 1990; Irwin, 1994; Pribor et al., 1993). As such, evidence indicating an increased prevalence of traumatic experiences in individuals with nonepileptic seizures could be viewed as additional support for a dissociative interpretation of this phenomenon. To this end, Bowman (1993) found that 70% and 77% of her sample of 27 nonepileptic seizure patients had experienced physical or sexual abuse respectively. Similarly, Betts and Boden (1992) obtained positive sexual abuse histories from 54% of 96 patients with nonepileptic seizures.
Evidence implicating high dissociative comorbidity, hypnotic susceptibility and exposure to trauma in individuals with nonepileptic seizures provides only indirect evidence for a dissociative interpretation of this phenomenon. Although such evidence suggests that a tendency to dissociate may be a common feature of these individuals, it does not constitute conclusive proof that nonepileptic attacks are themselves dissociative. A recent study by Kuyk et al. (1999) places such an interpretation on a firmer footing. Like 'genuine' epileptic seizures, nonepileptic attacks are often associated with a dense amnesia for events occurring during the ictus. As I have hopefully demonstrated, genuine epileptic amnesia should not be considered a dissociative phenomenon because it arises from a seizure-related disruption in memory encoding rather than an inability to retrieve intact memory traces. However, Kuyk et al. (1999) have shown that the amnesia associated with nonepileptic attacks may actually be the product of such a retrieval deficit. Kuyk et al. (1999) compared a group of individuals with amnesia for events occurring during well-documented nonepileptic attacks with a group displaying amnesia following complex partial and generalized epileptic seizures. All subjects were hypnotized and given suggestions designed to facilitate the recovery of ictal events; the experimenter remained blind to group status at all times. Using a free-recall paradigm, 17 out of 20 patients with nonepileptic seizures recovered significant information concerning the designated attack; this information was verified by video recordings or third-party reports. In contrast, not one of the 17 patients with epilepsy retrieved information concerning their attack during hypnosis. Such a finding appears to demonstrate that, unlike that found in epilepsy, nonepileptic amnesia results from a process that prevents the individual from accessing memories successfully encoded during the attack. This apparent separation of intact memorial information from conscious awareness following a nonepileptic attack, coupled with the phenomenological character of these events, clearly identifies these phenomena as dissociative in sense (i) of the term.
Was this article helpful?
Hypnosis has been defined as a state of heightened suggestibility in which the subject is able to uncritically accept ideas for self-improvement and act on them appropriately. When a hypnotist hypnotizes his subject, it is known as hetero-hypnosis. When an individual puts himself into a state of hypnosis, it is known as self-hypnosis.