Psychogenic nonepileptic seizures in children: A review



One of the considerations when a child presents with paroxysmal events is psychogenic nonepileptic seizures (PNES). PNES are discernible changes in behavior or consciousness that resemble epileptic seizures but are not accompanied by electrophysiologic changes. They are usually understood as the manifestation of a conversion disorder that reflects underlying psychological distress. There is a lack of population-based data on the prevalence or incidence of PNES in pediatric populations. The prevalence of PNES in children would appear to be lower than that in the adult population, but the prevalence of PNES seems to increase with age, and nonepileptic paroxysmal events are more likely to be PNES in adolescence than earlier in childhood. In terms of manifestation, PNES in childhood have been described using various categorizations and terminology, making comparisons across studies difficult. There is some evidence that events are more likely to involve unresponsiveness in younger children and prominent motor symptoms in older children. The most common precipitating factors would appear to be school-related difficulties and interpersonal conflict within the child's family. In terms of psychopathology, children with PNES are at high risk for symptoms of depression and anxiety. Accurate diagnosis of PNES in children is likely to involve taking a comprehensive description of the episodes, garnering the child's medical/developmental history, video-electroencephalography (video-EEG) to rule out epileptic seizures, and an evaluation of family functioning. The importance of effective and sensitive communication of the diagnosis of PNES has been emphasized and management approaches will typically involve multidisciplinary efforts to safely manage the events at home and at school. Interventions to reduce the effect of precipitating psychosocial stressors and the involvement of a mental health professional to treat comorbid psychopathology will also form part of an effective management plan. Outcome at follow-up is reported to be largely positive, although studies have not been able to follow all children, and few studies have focused on predictors of a good outcome. Future controlled intervention studies using a range of outcome measures are needed to identify efficacious approaches and predictors of best outcome.

The consequences of a misdiagnosis of epilepsy are significant, and like many medical conditions epilepsy has its mimics and borderlands (Smith & Bialer, 2012). A significant minority of children referred because of suspected epileptic seizures may not actually have the condition (e.g., Kotagal et al., 2002; Udall et al., 2006; Kim et al., 2012). One of the considerations in adult and pediatric populations when an individual is suspected of having epilepsy is nonepileptic seizures or paroxysmal nonepileptic events. Nonepileptic seizures present in both psychiatric and nonpsychiatric conditions. The nonpsychiatric conditions in pediatric populations consist of physiologic and organic disorders; the most common include inattention/daydreaming, staring, sleep myoclonus, stereotyped movements, hypnic jerks, tonic posturing, parasomnias, and movement disorders (Kotagal et al., 2002; Kim et al., 2012). Of those with a psychiatric basis, psychogenic nonepileptic seizures (PNES) are some of the most common nonepileptic events, with an increasing prevalence with advancing age in childhood (Kotagal et al., 2002).

PNES involve observable abrupt paroxysmal changes in consciousness or behavior that present as similar to epileptic seizures but are not accompanied by the electrophysiologic changes associated with epilepsy (Szab́ et al., 2012), and there is strong suspicion or positive evidence for a psychogenic cause (Bodde et al., 2009). PNES are usually interpreted as a manifestation of conversion disorder (Reuber, 2008). According to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition-Text Revision (DSM-IV-TR; American Psychiatric Association, 2000), PNES are classified as a conversion disorder with seizures or convulsions. The diagnostic criteria for conversion disorder include the following: symptoms affecting voluntary motor or sensory function suggest a neurologic or other medical condition but psychological factors are deemed to be associated with the initiation or exacerbation of the symptoms (American Psychiatric Association, 2000). The symptoms are not intentionally produced or feigned and cannot be explained by a general medical condition or by the direct effects of a substance, or a culturally sanctioned behavior or experience (American Psychiatric Association, 2000). Although PNES can be conceptualized as the manifestation of a conversion disorder, a conversion disorder need not necessarily include psychogenic nonepileptic seizures. The recognition of PNES are important, as children incorrectly diagnosed as having epilepsy instead of PNES may undergo unnecessary investigations and take unnecessary antiepileptic medications. Furthermore, PNES in children are associated with significant impairment in functioning and psychopathology. Despite the potential negative impact of misdiagnosis, there is inadequate information in the literature regarding clinical characteristics, treatment, and outcome of PNES in children and adolescents (Patel et al., 2007).

It is difficult to establish accurate estimates of the incidence or prevalence of PNES in either adult or pediatric populations. Most studies originate from neurology/epilepsy centers, and only cases for which video-electroencephalography (video-EEG) has confirmed a diagnosis of PNES are included. Therefore, reported figures are likely to be an underestimate (Reuber, 2008). Sigurdardottir and Olafsson (1998) reported an incidence of 1.4 in 100,000 in Iceland for individuals ages 15–54 years, with the highest incidence of 3.4 per 100,000 in the 15–24 year age group. It is thought that the prevalence of PNES in children is lower than that in adults (Kim et al., 2012). PNES represent 5–20% of an outpatient adult epilepsy population (Scott, 1982; Gates et al., 1991) and 15–20% of adult referrals to epilepsy centers (Lesser, 1996; Gumnit, 2001). Estimates of the prevalence of PNES in children with suspected epilepsy have varied from 1% to 9% (Smith et al., 2002; Pakalnis & Paolicchi, 2003; Udall et al., 2006; Kutluay et al., 2010; Asadi-Pooya & Emani, 2012; Szabó et al., 2012). As well as estimates of the prevalence of PNES in populations with suspected epilepsy, estimates of the proportion of those with paroxysmal nonepileptic events who have PNES have also been reported and range from 11% to 46% (Kotagal et al., 2002; Kutluay et al., 2010; Kim et al., 2012).

The terms used to describe PNES have changed over the years (Dickinson & Looper, 2012), and terminology continues to vary in published studies. Terms such as pseudoseizures and hysteria are used less frequently and are perceived as pejorative by many patients (Stone et al., 2003). The term psychogenic nonepileptic seizures (PNES) is commonly used between clinicians, but the term nonepileptic seizures is most frequently used when communicating with patients (LaFrance et al., 2006). In a study of health professionals in the United Kingdom, the terms most commonly used at diagnosis were nonepileptic attacks (62%), nonepileptic attack disorder (52%), and nonepileptic seizures (47%) (Mayor et al., 2011). The term PNES was reportedly used only by a small minority (5%) of clinicians (Mayor et al., 2011). However, the terms psychogenic seizures and psychogenic nonepileptic seizures (PNES) dominate the recent medical literature and are used throughout this review to denote the fact that the seizures are not epileptic but are thought to have a psychogenic basis. However, other terms that have referred to the same phenomena were used in the literature search carried out as part of this review.


The electronic database PubMed was searched on January 25, 2013 using the following search terms in the English language in “Title/Abstract”; “pseudoseizures,” “non-epileptic attack disorder,” “NEAD,” “psychogenic non-epileptic seizures,” “non-epileptic seizures,” “nonepileptic seizures,” and “psychogenic seizures.” To meet the criteria for inclusion in this review, studies had to refer to children (majority of participants were aged 0–18 years), contain more than 10 participants, and include some children who were experiencing PNES. The results of the search process are shown in Figure 1. In the context of this review, PNES were understood as events mimicking epileptic seizures that were thought to result from psychogenic processes, and studies were included only if it was clear that the phenomenon being referred to was PNES. In total, 24 studies met criteria for inclusion (see Table 1).

Table 1. Studies of PNES in children
AuthorTotal sample sizeAge range of children with PNES (mean)Total with PNES (gender)PNES onlyEpilepsy and PNES
  1. NR, not reported.

Szabó et al. (2012)5688–18 (14.8)27 (6 M 21 F)189
Asadi-Pooya and Emani (2012)198NR3 (NR)30
Kim et al. (2012)1436–1915 (NR)114
Gupta et al. (2011)45NR4 (NR)40
Salpekar et al. (2010)4810–17 (14)24 (11 M 13 F)222
Verrotti et al. (2009)366–1736 (10 M 26 F)036
Chinta et al. (2008)177–13 (10.7)17 (4 M 13 F)170
Kaciński et al. (2007)4511–1945 (4 M 41 F)450
Patel et al. (2007)1,9675–1968 (31 M 37 F)5926
Udall et al. (2006)223NR12 (NR)93
Vincentiis et al. (2006)69NR21 (13 M 9 F)021
Bhatia and Sapra (2005)1106–1150 (22 M 28 F)5012
Kotagal et al. (2002)8835–1862 (28 M 34 F)4121
Smith et al. (2002)207NR6 (NR)6NR
Gudmundsson et al. (2001)178–15 (12.75)17 (2 M 15 F)170
Irwin et al. (2000)356–18 (14.1)35 (11 M 24 F)2411
Wyllie et al. (1999)349–18 (14)34 (9 M 25 F)304
Selbst and Clancy (1996)106–1710 (3 M 7 F)100
Carmant et al. (1996)143(12)8 (NR)85
Lancman et al. (1994)435–16 (12.4)43 (11 M 32 F)430
Wyllie et al. (1991)188–18(6 M 12 F)180
Wyllie et al. (1990)218–18 (14.5)21 (6 M 15 F)210
Golden et al. (1985)85NR16 (NR)160
Holmes et al. (1980)53(14.33)11 (3 M 8 F)38
Figure 1.

Methodology diagram.


Some of the studies in Table 1 focus specifically on children with PNES and describe semiology of events, interventions, and outcomes in some detail. Other studies, usually on a larger scale, focus on children where epilepsy is suspected or focus on all possible paroxysmal events and include less detailed descriptions or analyses of the subgroup with PNES. One study focuses on pediatric conversion disorder (i.e., Kozlowska et al., 2011) and describes a subgroup of those with PNES, whereas another is drawn from a larger sample with conversion disorder (i.e., Gupta et al., 2011). The majority of studies are retrospective, and none of the studies focus on controlled interventions. Of a total of 539 children in the studies where gender is reported, 67% were female. However, it has been suggested that the gender distinction in favor of females is apparent only in adolescence (Kotagal et al., 2002; Patel et al. 2007; Kutluay et al., 2010). The youngest child with PNES in the published studies was 5 years of age.

Manifestation/semiology of PNES in pediatric populations

PNES can include a wide spectrum of clinical presentations and their semiology has been reported in published studies using various terminology and categorizations, and in different levels of detail. The most comprehensive description of PNES in children was provided by Szabó et al. (2012) and involved analysis of 75 events in 27 children. Based on a classification of PNES used in a study of adults (Seneviratne et al., 2010), Szabó et al. (2012) reported that 18 events were classified as rhythmic motor PNES, only half the frequency previously described in adults (Seneviratne et al., 2010). The frequency of complex motor PNES (13%) and mixed PNES (4%) is similar in children and adults. Dialeptic PNES were found more frequently in children (29%) than in adults. Patients were unresponsive during 34% of events. The most frequent motor sign was tremor (25%) involving the upper limbs more frequently than the lower. Pelvic thrusting was seen in only two events. Emotional manifestations were observed during 32 events (43%), and the emotions were negative in almost all children.

Patel et al. (2007) reported that based on video-EEG, children with PNES could be divided into three groups: patients with prominent motor activity, patients with subtle motor activity, and patients with both types of events. Subtle motor activity only was more likely in those younger than 13 years compared to a group 13 years or older. Prominent motor activity only was significantly more likely in the older age group. Kotagal et al. (2002) reported that children with PNES in their study could be divided into two groups with respect to manifestation, consisting of those with unresponsive events (child became unresponsive with reduction or the absence of spontaneous movements) or motor events (child exhibited motor phenomena, often consisting of bizarre, irregular, jerking, or thrashing movements of the extremities, not typical of any of the known types of epileptic seizures). In addition, Verrotti et al. (2009) documented that events were mainly of two types. In prepubertal patients, episodes were characterized primarily by unresponsive events, whereas the pubertal group primarily exhibited prominent motor events.

A number of authors have used terminology similar or the same as descriptors for epileptic seizures when describing PNES in children. Bhatia and Sapra (2005) reported that PNES that mimicked generalized tonic–clonic seizures were most common (60%), followed by the focal–clonic type (10%) and akinetic type (10%). Irwin et al. (2000) reported that of 35 children with PNES, 11 had tonic–clonic PNES, 10 had prolonged blank spells during which they stared straight ahead unresponsively, 10 had episodes during which they slid to the floor and remained still and unresponsive (“swoons”), and 4 had a mixture of events. Holmes et al. (1980) reported that partial “seizures” occurred in 62% of 11 children with PNES, whereas generalized “seizures” occurred in 38% of the group. Chinta et al. (2008) reported that upper limb movements were observed in 71% of 17 children, asymmetrical clonic movements in both hands in 5 children, and unilateral clonic movement in 3 children.

In the adult population Mostacci et al. (2011) have comprehensively compared ictal characteristics in PNES and epileptic seizures, but there have been fewer studies that have compared both events in the pediatric population. PNES typically last longer than epileptic seizures (Reuber, 2008), and Carmant et al. (1996) reported that psychogenic episodes were longer in duration than epileptic seizures in a group of 13 children with both conditions. Szabó et al. (2012) found that PNES lasted on average 269 seconds (range 1–3,417 seconds) compared with 83 seconds in a group of children with epileptic seizures. Gupta et al. (2011) reported that typical features of epileptic seizures such as frothing at the mouth, incontinence of urine, and tongue biting were absent in four cases of PNES in children who also had epilepsy. Verrotti et al. (2009) reported that none of 36 children with idiopathic epilepsy had incontinence, injury, or tongue biting (all possible in epileptic seizures) in episodes of PNES. However, in some children with epilepsy and PNES, the episodes do mimic the children's epileptic seizures (Holmes et al., 1980). Vincentiis et al. (2006) reported that 10 children (47.6%) had events that mimicked their own epileptic seizures and Irwin et al. (2000) reported on PNES events in 11 children with PNES and epilepsy and 7 had events similar to their epileptic seizures. Prolonged PNES have been reported in the pediatric population. These events have been described as resembling epileptogenic status (e.g., Papavasiliou et al., 2004), and the term “nonepileptic psychogenic status” (NEPS) has been used to describe this phenomenon, although there is a lack of consensus regarding definitions and classification parameters for these events (Dworetzky et al., 2010).

Precipitating factors of PNES in children

Establishing the cause of PNES may be helpful with respect to understanding why they have developed and what might be useful in terms of effective treatment. A number of studies have attempted to identify precipitating psychosocial stressors in PNES in children, and the categories used have varied significantly across the studies. The main areas in which difficulties have been reported include school-related difficulties (academic difficulties and bullying) family/interpersonal conflict, and physical/sexual abuse. The main categories of psychosocial stressor that have been reported in published studies of PNES in children are summarized in Table 2:

Table 2. Main precipitating factors of PNES in published studies
AuthorSchool-related difficulties, %Stressful family environment including interpersonal conflict, %Sexual abuse/physical abuse, %Fear of rejection/need for attention, %No cause identified,%
  1. N/A, not applicable.

Verrotti et al. (2009)318819N/A
Patel et al. (2007)466717N/A22
Vincentiis et al. (2006)N/A6619N/AN/A
Bhatia and Sapra (2005)303082612
Pakalnis and Paolicchi (2003)N/AN/A36N/AN/A
Gudmundsson et al. (2001)41N/A24N/AN/A
Irwin et al. (2000)202533N/A21
Wyllie et al. (1999)94438N/AN/A
Lancman et al. (1994)N/A3312N/A19
Wyllie et al. (1990)N/A435N/A29

In studies of adults with PNES, a history of trauma—particularly that arising from sexual or physical abuse—appears to be common (Reuber et al., 2007). However, the frequency of physical/sexual abuse in children would appear to be somewhat less common, with sexual/physical abuse occurring in less than half the cases in all of the published studies. Verrotti et al. (2009) reported that school phobia and fear of examinations were the most common precipitating or perpetuating factors (31%), followed by fear of rejection and need for attention (19%). Patel et al. (2007) also reported that the most commonly identified stressors were school difficulties (46%), with family discord (42%) and interpersonal conflicts (25%) also prominent. Vincentiis et al. (2006) reported that the most common precipitating factor was “inadequate family setting,” characterized by a stressful environment at home not directly related to the child but affecting the child's subjective perception of wellbeing. Bhatia and Sapra (2005) reported that school phobia (30%) was the most common precipitating or perpetuating factor, followed by fear of rejection or need for attention (26%). Gudmundsson et al. (2001) reported that 7 of 17 pediatric patients were found to have “academic difficulties,” and in 4 of 17 cases a past history of confirmed or strongly suspected prior traumatizing sexual experiences. Wyllie et al. (1999) reported that psychosocial stressors included severe family stress (44%), sexual abuse (32%), physical abuse (6%), and school failure (9%). Irwin et al. (2000) reported that in five cases no cause could be found for the PNES, although attention gained from their PNES seemed to perpetuate the problem. Verrotti et al. (2009) suggest that a previous history of epilepsy, mostly as a personal history rather than a family history, may represent an important risk factor for PNES in children.

Psychopathology in children with PNES

Children with PNES are at high risk for psychopathology, particularly symptoms of depression and anxiety. Some of the reviewed studies have reported on the presence/absence of psychiatric disorders, whereas others have reported the presence of psychopathology in more detail using international classification systems (i.e., DSM-IV/International Classification of Diseases, Tenth Revision [ICD-10]). Table 3 summarizes the main findings from studies that have reported on rates of psychopathology.

Table 3. Main forms of psychopathology in children with PNES
AuthorCriteria/measuresDepression, %Separation anxiety/school phobia, %Panic, %GAD, %PTSD, %None, %
  1. N/R, not reported; GAD, generalized anxiety disorder; PTSD, posttraumatic stress disorder.

  2. a


  3. b

    Depression and anxiety disorders combined.

  4. c

    Any anxiety disorder.

  5. d

    Overanxious disorder.

Verrotti et al. (2009)ICD-10/DSM-IV19N/R118358
Patel et al. (2007)N/R257a59
Vincentiis et al. (2006)DSM-IV/ICD-1062bN/R
Bhatia and Sapra (2005)ICD-10N/R321212216
Pakalnis and Paolicchi (2003)N/R3241cNR
Wyllie et al. (1999)DSM-IV322486d8N/R

The studies in Table 3 suggest that a significant proportion of children with PNES will have a psychiatric disorder as well as conversion disorder. Verrotti et al. (2009) reported that 15 (42%) of 36 children with PNES had at least one psychiatric disorder based on DSM-IV (American Psychiatric Association, 2000) criteria. Vincentiis et al. (2006) used DSM-IV and ICD-10 (World Health Organization, 1992) classification systems and reported a similar rate of psychopathology. Bhatia and Sapra (2005) reported that only 8 (16%) of 50 children did not have any psychiatric disorder. Wyllie et al. (1999) used DSM-IV criteria and reported that 32% of 34 children had “mood” disorders including major depression, bipolar disorder, and dysthymic disorder. Eight of the 34 had separation anxiety and/or school refusal and two had brief reactive psychosis or schizophreniform disorder.

Salpekar et al. (2010) compared 24 children with epilepsy with 24 with PNES using a battery of behavioral questionnaires. They found that children with PNES had significantly higher scores on the Childhood Somatization Inventory (CSI; Walker et al., 2009) and the Functional Disability Inventory (Walker & Greene, 1991) and their parents reported more somatic problems on the Child Behavior Checklist (Achenbach, 1991). The groups did not differ on measures of depression, anxiety, or alexithymia, and Salpekar et al. (2010) suggest that measures of somatization and functional disability may be promising tools for differentiating the behavioral profile of PNES from that of epilepsy. They further suggest that child-specific self-efficacy might be more impaired among children with PNES. Chinta compared 17 children with PNES, 17 with epilepsy, and healthy controls using the Childhood Psychopathology Measurement Scale (CPMS; Malhotra et al., 1988). A significantly higher number of children in the PNES group and children in the epilepsy group were found to have CPMS scores above the at-risk cut-off score. Pakalnis and Paolicchi (2003) and Kozlowska et al. (2011) emphasize that children with PNES may also experience other symptoms of conversion disorder including pain, sensory, and motor symptoms.

Assessment and diagnosis in children with PNES

Accurately identifying and diagnosing PNES is a challenge for both psychiatrists and neurologists, and the diagnosis of PNES currently relies on the exclusion of epilepsy (Dickinson & Looper, 2012). Up to one fourth of patients with suspected epilepsy seen at epilepsy centers who had “failed” to achieve seizure control after two or more antiepileptic drugs (AEDs) had been tried were found to not have epileptic seizures after video-EEG (Alsaadi et al., 2004). Epilepsy is usually ruled out using video-EEG telemetry to analyze typical events (Alsaadi et al., 2004), and video-EEG has been considered the gold standard for the diagnosis of PNES (Iriarte et al., 2003; Reuber & Mayor, 2012), although the subjective component of video-EEG analysis has been highlighted (Benbadis et al., 2009). Biologic markers have been sought to aid discrimination between PNES and epileptic seizures. Wyllie et al. (1985) reported prolonged postictal creatine kinase (CK) elevations in 6 of 12 patients with generalized tonic–clonic seizures but no elevations in 6 patients with PNES. However, elevations were also absent in patients with complex partial, focal motor, absence, or tonic epileptic seizures. Therefore, normal postictal CK levels do not rule out the possibility that an event is epileptic in nature, and serum prolactin level estimation may not have the necessary sensitivity and specificity for the differentiation of psychogenic nonepileptic from true epileptic events.

Viewing recording of a typical PNES can be a significant advantage with respect to clarifying the nature of the event and may alleviate the necessity of using video-EEG analysis. The increased availability of modern technology such as smart phones and digital cameras can facilitate recording of events. In the absence of such evidence, “seizure induction” via a range of methods may be employed in an attempt to induce PNES. Provocative techniques are sometimes used at epilepsy clinics to aid in the diagnosis of PNES (Benbadis et al., 2000), and include the use of intravenous saline, a tuning fork, and a patch applied to the skin. Dericioğlu et al. (1999) reported that use of seizure induction techniques (verbal suggestion or intravenous saline infusion) can avoid the use of video-EEG analysis in approximately half of all adult patients with PNES. Wyllie et al. (1990) reported that six children who did not have a PNES during EEG recording had episodes induced in response to suggestion and intravenous saline injection. Ethical concerns about the use of placebo induction have been raised (Benbadis et al., 2000), and given that these techniques involve deception, suggestion would appear to be the most appropriate seizure induction technique in both adult and pediatric populations. Benbadis et al. (2000) describe the use of an effective procedure involving the use of suggestion, hyperventilation, and photic stimulation in 19 adults with suspected PNES, and because these techniques are also used to induce epileptic seizures, there is an absence of deception and thus could be seen as more ethical than placebo induction.

Plioplys et al. (2007) emphasize that PNES are not a default diagnosis based on negative video-telemetry, as the mental health clinician also needs to show evidence of an underlying conversion disorder. The importance of taking a detailed history (i.e., history of paroxysmal events, child medical/developmental history, and family history) from multiple sources (e.g., child, parents and school staff) has been emphasized (Plioplys et al., 2007), and may allow health care professionals to differentiate between PNES and epilepsy (Reuber & Mayor, 2012). Plioplys et al. (2007) argue that interviewing the child and parents separately may be helpful and also suggest that it is important to address the degree to which the parents are invested in the child having medical problems (or not having a psychological problem).

The delivery of a diagnosis of PNES is a skilled exercise (Stone et al., 2003), and it is becoming increasingly clear how important effective communication of the diagnosis of PNES is (Reuber & Mayor, 2012). It has been reported that making and communicating the diagnosis of PNES in the context of the adult population leads to immediate and marked reduction in PNES-related health care demand, which emphasizes the importance of communication of the diagnosis as a therapeutic measure (Razvi et al., 2012). Most authors emphasize that it is crucial to make it explicit that attacks are real and the certainty of a psychogenic basis for the events should be clearly conveyed to the child and family. Plioplys et al. (2007) advocate that initial feedback should be given to the parents alone to allow them to discuss their reactions before information is given to the child, although the efficacy of this approach has not been evaluated and it is possible that approaches may differ depending on the age of the young person. Plioplys et al. (2007) further add that it is important to help parents understand PNES in a biopsychosocial context and to emphasize to parents that the child is not faking the events. Irwin et al. (2000) emphasize that the diagnosis should be unambiguously communicated to families, including an explanation of the role of stress or emotional factors in the cause of events. Understanding and acceptance can be facilitated by the use of videotaped events recorded during an evaluation (Wyllie et al., 1999). The rationale for a treatment program that focuses on psychological rather than medical causes should be clear, and it is important for the diagnosis to be conveyed positively, both with regard to the expectations of a good prognosis and recovery and in terms of there being no shame or blame attached to a nonorganic cause (Irwin et al., 2000). A proportion of children with PNES will also have epilepsy (e.g., Kotagal et al., 2002), and it is important that the manifestation of both events are made clear to the family and videotape review can aid the recognition of epileptic seizures and PNES in children with both conditions (Wyllie et al., 1990).

Interventions for PNES in children

PNES are a heterogeneous group of disorders, and choice of treatment modality is likely to reflect this heterogeneity (Kanner, 2003). There are now a number of studies suggesting that cognitive behavioral therapy (CBT) may be an effective approach in treating PNES in the adult population (Kuyk et al., 2008; La France et al., 2006), including one randomized controlled trial (RCT) (Goldstein et al., 2010). In the pediatric population, most authors recommend prompt intervention from a child mental health professional and Irwin et al. (2000) argue that their involvement is important to prevent families from becoming entrenched in a medical approach and to prevent the behavioral pattern of PNES from becoming fully incorporated into the child's personality and way of life. It has been argued that the first treatment goal in PNES is to ensure acceptance of the diagnosis, and that PNES are not epileptic seizures that require emergency treatment (Kanner, 2003). Many parents may not be prepared for their child to be discharged immediately from the care of a neurologist. Therefore, there should be a transition process from neurology to psychology and/or psychiatry, as premature discharge often contributes to resistance to accept a diagnosis of a psychogenic process (Kanner, 2003; Plioplys et al., 2007). The continued involvement of a neurologist for a period after the diagnosis of PNES can also ensure that epileptic seizures have not been missed and ensure that both child and family are seen by appropriate mental health professionals.

Irwin et al. (2000) have published one of the most comprehensive accounts of management approaches of PNES in children. They suggest that the initial focus should be on containment and safe management of the PNES while exploring the likely precipitating stressors and dealing with these. They argue that it is crucial for any management plan to be explicitly shared with and understood by families and all other professionals involved with the child, including school and community pediatric services, to facilitate consistency in management. Initially a focus on “living safely” with the seizures can reduce the heightened emotional arousal and attention that is focussed on the events (Irwin et al., 2000). Stress management techniques including relaxation therapy and cognitive approaches can be helpful, and a focus on acquiring social skills, raising self-esteem, and helping the child to confront or resolve issues with others either in the family or outside may also be needed (Irwin et al., 2000). Direct interventions may involve offering additional support to improve the child's academic performance, or putting programs in place to deal with difficult peer relationships and bullying (Irwin et al., 2000). Plioplys et al. (2007) indicate that short-term treatment goals are to stop or significantly decrease frequency and duration of events, provide treatment for underlying psychiatric disorders, stop/modify antiepileptic drug (AED) use (depending on epilepsy status), and return the child to typical daily activities. Collaboration between the medical team and school is essential to ensure success (Plioplys et al., 2007).

There are no controlled intervention studies aimed at treating PNES in the pediatric population, although a number of descriptive studies have commented on the interventions or management approaches employed. Verrotti et al. (2009) reported that after diagnosis, all of the children in their study commenced psychotherapy (explanation, reassurance, suggestion, confrontation, and discussion of problems associated with the origin of the PNES) and 6 (17%) of these children were put on appropriate drug therapy (anxiolytic and/or antidepressants). Vincentiis et al. (2006) reported that in 10 of 21 patients with epilepsy and PNES, psychoactive drug therapy was initiated. In Irwin et al. (2000), management of PNES in most cases involved referral to a child psychologist, whereas those with epilepsy and PNES also continued to be reviewed by a pediatric neurologist. Wyllie et al. (1999) reported that a subgroup of students with PNES with separation anxiety and school refusal benefitted from family counselling, classroom setting modifications based on results of neuropsychological assessments, and in some cases treatment with methylphenidate or tricyclic antidepressants for attention deficit/hyperactivity disorder (ADHD) or separation anxiety. Chinta et al. (2008) indicated that all children with PNES were managed by communicating the diagnosis to the families as well as an unambiguous explanation of the role of stress and emotional factors in the cause of PNES. Direct management techniques were also used, such as providing additional academic support to a child with academic problems, recommending change of school for a child with significant adjustment problems at school, and asking parents to reduce evidence of open marital conflict for a child who is experiencing stress due to an increase in parental conflict (Chinta et al., 2008). In addition, parents were asked to reduce secondary gain, if any, and to normalize the child's activities if the child showed significant functional impairment (Chinta et al., 2008).

Outcome of PNES in children

Despite the fact that there have been no systematic studies of treatments/interventions for children with PNES, a number of studies have reported on follow-ups of affected children and these studies suggest that outcomes may be more favorable than in the adult population (e.g., Wyllie et al., 1991). The follow-up times in the studies have varied or have not been reported, and some studies have reported only on participants with complete freedom from events, whereas others have also included reference to those who had experienced a reduction in events but not total remission. Table 4 summarizes studies where children have been followed after the diagnosis of PNES:

Table 4. Outcome of PNES in children
 Total in original samplePNES free, %Not PNES freeUnknown/lost at follow-up, %
Improvement, %No improvement, %
  1. N/R, not reported.

Chinta et al. (2008)173547017
Bhatia and Sapra (2005)50722080
Pakalnis and Paolicchi (2003)223623041
Kotagal et al. (2002)62181468
Gudmundsson et al. (2001)1759–6337–41N/R
Irwin et al. (2000)356620106
Wyllie et al. (1999)3444N/R1838
Lancman et al. (1994)43232849
Wyllie et al. (1991)18721711
Wyllie et al. (1990)21671914

Comparisons between studies with respect to rates of remission and improvement in symptoms are difficult owing to the significant number of children lost to follow-up in some studies. Nevertheless, of those who have been followed, it would appear that the majority of children have experienced significant reductions or are free from events. Wyllie et al. (1999) were able to follow 21 of 34 patients and of these 15 (72%) were free of PNES during the preceding 9–55 months and 6 (28%) had persistent PNES. Two patients with epilepsy were available for follow-up and both were free of PNES. Two additional patients were taking AEDs despite the lack of epilepsy diagnosis. Kotagal et al. (2002) reported that when 20 patients followed up a median time of 8.35 months, 65% of them were free from events. Holmes et al. (1980) reported that after 1 year, all eight children with PNES without epilepsy had a decreased frequency of events, whereas one of three children with epilepsy and PNES continued to have episodes of PNES and the other two were free of PNES. Chinta et al. (2008) reported that 6 of 17 children were seizure free after 3 to 6 months, and 8 of 17 experienced a >50% reduction in the frequency of symptoms. Pakalnis and Paolicchi (2003) found that all patients available for follow-up without epilepsy had a resolution of PNES at 12–18 months, but that five children with both epilepsy and PNES continued to have PNES. Gudmundsson et al. (2001) reported that 59% of 17 children were seizure free at 6 months and 63% at 12 months. Gudmundsson et al. (2001) also reported good school attendance in 14 of 16 children at follow-up, and none of the 17 children remained on AEDs. The reasons for better outcome in children may include lack of prolonged chronicity, less severe psychopathology, and greater effectiveness with earlier intervention (Wyllie et al., 1990).


PNES can be debilitating in childhood and are associated with school-related difficulties and significant psychopathology. For children who also have epilepsy, PNES can be even more impairing than the epileptic seizures. Differing views regarding etiology, terminology, and lack of standardized criteria can hamper treatment and diagnosis (Plioplys et al., 2007). Correct recognition, referral, and treatment can save on health resources and alleviate significant distress in affected children and their families. Almost all of the published studies on PNES in pediatric populations have involved small samples, retrospective analysis of data, and lack of comprehensive description of interventions. It is difficult to establish accurate estimates of the prevalence or incidence of PNES in childhood, and there is lack of population-based data. Furthermore, it is not clear if PNES is likely to be primarily identified in relation to a suspicion of epilepsy or whether the phenomenon is widely diagnosed outside of epilepsy centers. It would appear that PNES represent an increasing proportion of paroxysmal nonepileptic events as children transition to adolescence. In terms of manifestation of PNES, there is some suggestion that younger children may be more likely to display more subtle motor activity or unresponsiveness and older children more prominent motor activity, but there is a lack of agreed criteria with respect to classification of PNES in children.

Reaching a correct diagnosis of PNES as early as possible would appear to be key to successful treatment. Appropriate diagnosis will involve taking a comprehensive history of events, child development, and family functioning, whereas video-EEG may be needed to rule out epilepsy. Detailed neuropsychiatric assessment is likely to be helpful to identify comorbid psychopathology. The most likely precipitating psychosocial stressors are school-related problems, interpersonal/family problems, and a history of physical or sexual abuse. Symptoms of depression and anxiety are common among children with PNES and should be a consideration in an assessment. For some children, clear communication of the diagnosis to child and family may be all that is needed to achieve optimum outcome (Gudmundsson et al., 2001). However, for many children there will be a need for follow-up intervention, which is likely to be multidisciplinary in nature. This intervention is likely to involve support around the management of the PNES, psychoeducation for child, family and school about PNES, management strategies, or psychotherapeutic intervention to address precipitating factors (depending on whether they are ongoing or historical), and psychotherapy or psychopharmacology to treat comorbid mental health and behavioral difficulties. In children with epilepsy and PNES, treatment planning will need to address both disorders. Parents may be unwilling to accept the diagnosis of PNES, and children may continue to take antiepileptic medication; this risk is particularly great if the diagnosis is not handled sensitively and appropriate supports made available. The lack of a clear-cut etiologic model that supports the psychogenic origin of PNES can make a one-size-fits-all intervention a difficult task to achieve, but collaboration between neurology and psychiatry/psychology professionals is paramount (Baslet, 2012).

The lack of current guidelines or evidence for treatment of PNES in children should not deter attempts to develop effective interventions and protocols notwithstanding that treatments may need to be individualized (Baslet, 2012). It may be that children should be divided into different treatment groups depending on likely etiology or manifestations of PNES (Reuber, 2008). Recent results of CBT interventions in adults offer optimism with respect to treatment in children, but studies in pediatric populations are needed, as the pattern of psychosocial stressors and comorbidities differ between adult and pediatric groups. Successful outcomes in clinical practice will be dependent on adaptation and flexible delivery of interventions (Baslet, 2012). As well as examining reduction in psychogenic events, it will be important to measure quality of life, school attendance and school achievement, cognitive functioning, behavioral and emotional well-being, and family functioning in future outcome studies. Descriptive and intervention studies also need to include measures of family functioning including parental stress and parental emotional well-being. Epilepsy is associated with high rates of maternal depression (Ferro & Speechley, 2009) and it is likely that parents of children with PNES are similarly afflicted. Because it will be difficult to gather large numbers of affected children, multicenter studies are needed with respect to prospective studies of incidence, prevalence, and treatment. PNES in children have a good prognosis provided that diagnosis is made early and sensitively, a cause is identified, and there is appropriate psychological intervention.


None of the authors have any conflict of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.