POSSIBLE BEHAVIOURAL DISORDERS

The phenomenology of psychiatric disorders associated with epilepsy is wide and variable (Choudhary et al. 2014). Adding to the diagnostic difficulty of neurobehavioural disorders in epilepsy is the fact that several psychiatric symptoms, recognised as diagnostic criteria, can occur secondary to seizure activity, EEG abnormalities or antiepileptic drugs (AEDs). Moreover, some authors have considered cognition and language as intrinsic components of behaviour since they play an important role in social functioning, as well as in the sense of competence and self-esteem (Austin and Caplan 2007). Consequently, in some articles it is difficult to discern between pure behavioural and cognitive problems.

The neurobehavioural comorbidities being various, the attempts of classification have been controversial. Some researchers contend that psychiatric syndromes in epilepsy can be characterised by atypical features that are poorly represented by conventional classification systems such as the fourth edition (text revised) of the Diagnostic and Statistical Manual of Mental Disorders (DSM IV-R) and the International Classification of disease (ICD) Version 10 (Krishnamoorthy et al. 2007). In 2007, the International League Against Epilepsy (ILAE) Commission on Psychobiology of Epilepsy developed a classification proposal of neuropsychiatric disorders in epilepsy with the aim to separate disorders comorbid with epilepsy and those that reflect ongoing epileptiform activity from epilepsy-specific disorders, and to attempt to subclassify the epilepsy-specific disorders alone (Krishnamoorthy et al. 2007). The classification was followed by international clinical practice consensus statements for the treatment of neuropsychiatric conditions associated with epilepsy, to provide guidance on the management of these conditions (Kerr et al. 2011a). These documents put together adult and childhood epilepsies.

Generally, there is a wide consensus regarding a greater association between epilepsy and the following behavioural disorders: attention-deficit–hyperactivity disorder (ADHD), autism spectrum disorders (ASD), anxiety, depression and psychosis.

ADHD is one of the most common comorbidities, being reported in about 30–40% of children with epilepsy compared to 3–6% of controls, especially in preschool and school-age children, without sex differences (Dunn and Kronenberger 2005; Hermann et al. 2007a; Parisi et al. 2010; Cohen et al. 2013; Socanski et al. 2013). Differently from children without epilepsy, children with epilepsy are most likely to present with an inattentive form (Reilly 2011). The symptoms of ADHD may not be the same as those of ADHD in other conditions, because of the comorbidity with psychiatric disorders, underlying brain abnormality, frequent epileptiform discharges, seizures and adverse effects of medication. A common pathogenic mechanism has been hypothesised: the animal models of ADHD suggest that synaptic abnormality in excitatory glutamergic transmission may contribute to vulnerability for epilepsy and ADHD (Jensen et al. 2009).

The increased prevalence of anxiety and depression has been found in epidemiological and clinical studies, and is frequently comorbid (Berg et al. 2011a; Russ et al. 2012). The reported prevalence of these internalising disorders from samples obtained in epilepsy clinics is higher than the prevalence reported in epidemiological surveys, namely 12–20% for depression and 10–20% for anxiety, compared to 1–7% of controls, going beyond 30% in patients with lower IQ, language delays, or neuropsychological deficits such as academic difficulties (Caplan et al. 2005b; Cushner-Weinstein et al. 2008; Ekinci et al. 2009; Austin et al. 2010b; Reilly 2011). The higher risk for anxiety includes specific phobias, panic disorder, and posttraumatic stress symptoms (Dunn et al. 2009). Age has been found to be a significant risk factor for both depression and anxiety, adolescents being more at risk than younger children, although findings are discordant (Ekinci et al. 2009). It also seems that adolescents report more symptoms than parents observe. For children and adolescents with primary depression, early age at onset and the severity of depressive symptoms are associated with non-adherence to treatment or lack of treatment and are predictive of both pharmaco-resistant epilepsy and recurrence of depressive symptoms during their adult age (Pereira and Valente 2013). The sex difference in the increased prevalence of anxiety and depression is inconsistent though the prevalence of depression is high in female adolescents (Pereira and Valente 2013). The symptoms are similar to those observed in peers without seizures (Vega et al. 2011). However, some symptoms may relate specifically to epilepsy: brief episodes of anxiety or fear can occur as an aura in children with focal epileptiform discharges frequently originating from the temporal lobe (Salpekar and Dunn 2007). For symptoms of depression and anxiety, some authors suggested distinguishing between epilepsy-related preictal, ictal, postictal and interictal symptoms and comorbid disorders unrelated to epilepsy. Specific epilepsy-related factors do not seem to be a major factor in the causation of anxiety or depression (Pereira and Valente 2013), and findings have shown that depression is more important than seizure frequency in determining perceived QOL (Hecimovic et al. 2012).

In children occur also psychogenic non-epileptic seizures (PNES), even if no formal guidelines for diagnosing in children exist, and little is known about the clinical practice of diagnosing PNES in the paediatric setting (Wichaidit et al. 2014). PNES are paroxysmal events that resemble clinical signs and symptoms of epileptic seizures but are not associated with ictal epileptiform abnormalities in the EEG pattern. Some studies focused on the development or on the remission of PNES after cranial surgery for epilepsy, especially for right temporal dysplasia. The clinical semiology of PNES is very complex and belongs to the complicated phenomenon defined as ‘dissociation’, for example, ‘a disruption of and/or discontinuity in the normal integration of consciousness, memory, identity, emotion, perception, body representation, motor control, and behavior’ (Roberts and Reuber 2014).

Bipolar disorders are rarely diagnosed in the paediatric epilepsy population, rating around 1% (Berg 2011). Bipolar disorder symptoms have been more often reported as postictal phenomena and occasionally as ictal phenomena. The phenomenological understanding of childhood mania has changed over recent years to the extent that a bipolar disorder spectrum allows for better description of the patient presentations most commonly encountered in clinical practice. Symptoms such as explosive rage, extreme irritability, mood lability and agitation may not meet full criteria for traditional mania but may be consistent with a bipolar disorder spectrum especially in terms of treatment selection (Nishida et al. 2005; Schmitz 2005).

There does not appear to be a significant increase of oppositional defiant disorder (ODD) or conduct disorder in children with uncomplicated epilepsy, although results are not univocal (Jones et al. 2007).

Suicidal ideation has been reported in around 20% of children with epilepsy, compared to about 5% in general population (Caplan et al. 2005b; Wagner et al. 2009; Hesdorffer et al. 2012b). Adolescents with epilepsy are at higher risk, both as compared with general population and as compared with adults with older-onset epilepsy. The risk seems to be associated with the severity of epilepsy and with intellectual disability, but other factors are likely to be important and need to be discovered through further investigation and clinical practice advances (Rodenburg et al. 2011).

METHODOLOGICAL ISSUES

Consensus neurobehavioural diagnoses have been generally made using the DSM-IV criteria together with standardised scales. The ADHD Rating Scale-IV (ADHD-RS-IV; DuPaul et al. 1998) and Conners’ Rating Scales (CRS-R, Conners 1997) have been used for ADHD diagnosis. The Childhood Depression Inventory (CDI, Kovacs 1985), the Short Mood and Feelings Questionnaire (SMFQ) (Angold et al. 1987), the Child or Adolescent Symptom Inventory (Sprafkin et al. 2002), the Beck Depression Inventory (BDI) (Beck et al. 1988) and the Children Depression Rating-Scale Revised (CDRS-R) (Poznanski and Mokros 1996) have been used to rate mood disorders. So far, the instruments most commonly used to assess anxiety disorders have been the Revised Children’s Manifest Anxiety Scale (RCMAS) (Reynolds and Richmond 1985), the Multidimensional Anxiety Scale for Children (MASC, March et al. 1997), the Screen for Child Anxiety Related Emotional Disorders (SCARED, Birmaher et al. 1997).

Among self-rating scales, the Child Behaviour Check List (CBCL) (Achenbach 1991) for children, parent and teacher has been highly used, confirming behavioural disorders associated with epilepsy, although with different results (Rantanen et al. 2012).

To rate parenting stress in relation to children with special needs and/or epilepsy, the Parenting Stress Index (PSI) has been developed (Abidin 1990).

To assess social skills, there are the CBCL, the CRS-R and the Social Skills Rating System (SSRS, Gresham and Elliott 1990). Together with the SSRS, the Strength and Difficulties Questionnaire (SDQ, Goodman 1997) is an example of possible and practical assessment tools for a broader perspective in the assessment of social competence, since they include all aspects (Dafoulis and Kalyva 2012). Studies have also provided evidence for a high sensitivity and specificity of SDQ among children with epilepsy (Hysing et al. 2007). Furthermore, since expectations of social competence increase as a function of child’s age and developmental level, the division into subgroups according to their age has been suggested (Rantanen et al. 2012).

A novel approach using multiple instruments to assess ASD, developmental delay and behavioural/emotional difficulties has been proposed (Eom et al. 2014), including the Social Communication Questionnaire (SCQ) (Rutter et al. 2003), the Modified Checklist for Autism in Toddlers (mCHAT) (Robins et al. 2001), and the SDQ.

QOL scales specific for children with epilepsy have been developed to cover physical, psychological, social and academic functioning. Examples of tools include the Quality of Life in Epilepsy for Adolescents (QOLIE-AD-48) (Cramer et al. 1999), the Quality of Life in Childhood Epilepsy (QOLCE) (Sabaz et al. 2000), the Impact of Paediatric Epilepsy Scale (IPES) (Camfield et al. 2001) and the Impact of Childhood Illness Scale (Hoare and Russell 1995). The National Institute of Neurological Disorders and Stroke (NINDS) provided leadership to develop the Neuro-QOL, which provides psychometrically sound and clinically relevant QOL measures for children and adults with neurological disorders (Cella et al. 2012; Gershon et al. 2012).

The best way of obtaining accurate information on the prevalence of neurobehavioural comorbidities in children with epilepsy is to analyse data from well-designed, prospective epidemiological studies on large populations of children, including control groups. While interviewing patients, some strategies can be used that are both efficient and accurate (Salpekar and Dunn 2007). With younger children, a calm and reassuring stance will yield the most information, and a gradual process of general questions leading to more specific questions is often most effective. Especially for adolescents, to obtain a behavioural description every attempt should be made to interview the patient independently, since they may not reveal their symptoms in the presence of caregivers.

A lifespan perspective on behavioural disorders

Evidence has emerged suggesting bidirectional relations between epilepsy and various neurobehavioural comorbidities in paediatric epilepsy (Lin et al. 2012; Helmstaedter et al. 2014). Psychiatric disorders can either precede or follow the onset of epilepsy, which is the case for depression, anxiety, ADHD and psychosis (Jones et al. 2005; Hesdorffer et al. 2006; Wotton and Goldacre 2012). Seizure incidence has been shown to be higher in psychiatric patients, males and aged 6 to 12 years, even in those without seizure risk factors (McAfee et al. 2007).

THE ONSET

In a study, more than 40% of children with epilepsy met diagnostic criteria for a DSM-IV disorder before the diagnosis of epilepsy and the first recognised seizure (Jones et al. 2005), even without clinical evidence of brain structural abnormalities (Almane et al. 2014). The increased prevalence may be consistent with the presence of underlying neurobiological influences independent of seizures, epilepsy syndrome and AEDs (Jones et al. 2007).

ADHD is more prevalent in new-onset epilepsy than in healthy controls (31% vs 6%), even in idiopathic syndromes (Hermann et al. 2007a). More than 15% of the children with new-onset seizures fell in the clinical range of internalising and externalising problems (Dunn et al. 2009). About 25% of children with new-onset idiopathic epilepsies need special education services before clinical seizure onset (Berg et al. 2005). Behavioural complications do not necessarily respect traditional syndrome groupings in children with new- or recent-onset epilepsy, showing shared findings in localisation-related and generalised epilepsies, even if social difficulties and internalising behaviours are more frequent in the latter (Bhise et al. 2009; Almane et al. 2014). At the time of the first seizure episodes, the child must begin to adapt to an unpredictable and frightening disorder, causing him/her persisting worries about having another seizure (Choudhary et al. 2014). Anxiety, stress and emotions can trigger seizures and increase seizure frequency. This, in turn, can also trigger anticipatory anxiety and evoke stress, anger and frustration leading to a vicious circle of anxiety and seizures fuelling each other, further increasing the risk for behavioural problems (Lathers and Schraeder 2006).

THE OUTCOME

Authors have proposed that the concept of cerebral reserve might have important implications in understanding the life course of people with epilepsy who have an early brain insult, raising the question whether it confers an increased risk for disrupted cognitive development and accelerated ageing effects owing to reduced cerebral reserve (Deary et al. 2004). In the context of childhood-onset chronic epilepsy, this concept might help to conceptualise the lifespan course of cognition and brain structure in epilepsy (Hermann et al. 2008).

There are relatively few studies of the behavioural outcome of epilepsy surgery in children who have used standardised behavioural measures before and after the procedure. Moreover, homogeneous data are lacking because of groups with mixed pathology and retrospective studies. Characterising the long-term behavioural outcome for people with epilepsy previously admitted to surgery is complex, given the highly individual nature of patient response to illness and its treatment (Wilson et al. 2005). Increasing evidence supports the idea that epilepsy surgery early in life may be associated with a complex psychological process of postoperative adjustment, which persists for many years (McLellan et al. 2005).

Taken together, results show that the outcome in temporal and extra-temporal surgery is very variable, with few children improving and others deteriorating, and with a better outcome in younger children (Mclellan et al. 2005; Colonnelli et al. 2012). De novo depression is a common psychiatric sequela of temporal lobe epilepsy (TLE) surgery (McLellan et al. 2005; Wrench et al. 2009). The study of the psychological profile in childhood-onset TLE approximately 13 years after seizure onset showed that surgical non-seizure-free patients had the poorest outcome in terms of lowered mood and depression (Micallef et al. 2010), and are at higher risk for postictal psychosis (Cleary et al. 2013). Lifetime psychiatric history predicts a worse seizure outcome following temporal lobectomy (Kanner et al. 2009). The outlook for hemispherectomy does not seem better, since psychopathology has been a common result either pre- or postoperatively, without post-surgical worsening or improvement, even in seizure-free patients (Danielsson et al. 2009; Colonnelli et al. 2012). Vagus nerve stimulation seems to increase QOL, even without a change of adaptive behaviour (Mikati et al. 2009; Zamponi et al. 2011).

Explicative models

Theoretical models propose that psychopathology in epilepsy can stem from a complex interplay of multiple aetiological variables (Choudhary et al. 2014). The prevalence of neurobehavioural comorbidities in epilepsy across the lifespan has triggered a concerted effort to uncover their potential mediators, including various epilepsy-related variables. Moreover, psychosocial models have been developed to depict the complexity of behavioural comorbidities in childhood epilepsy. Recent neuroimaging evidence contributes to the tracing of the neurobiological underpinnings of associated disorders, providing biological markers of increased risk.

THE ROLE OF EPILEPSY-RELATED VARIABLES

Given that neurobehavioural comorbidities are especially recurrent in complicated epilepsies, it is important to distinguish the burden of the core clinical features of epilepsy. Among many possible factors, the most relevant to clinical practice include aetiology, syndrome, epileptiform discharges, seizure, age at epilepsy onset, years of epilepsy duration and prescribed drugs (Lin et al. 2012).

The electro-clinical syndrome is a factor which has received particular interest. For decades, the syndromic model has served as the basis for investigating the cognitive and psychiatric domains most at risk. The importance of the behavioural implications of syndrome classification could be considered in two broad categories: first, the behavioural effects of epilepsy syndromes and, second, the behavioural associations of childhood syndromes (‘behavioural phenotypes’) in which epilepsy occurs as a prominent feature. Even if different syndromes seem to have diverse comorbidities, the same disorders have been documented across a wide range of epilepsy syndromes: for example, psychiatric complications of TLE have been documented also in benign epilepsy with centrotemporal spikes (BECTS). Consequently, the decade-long controversy regarding the unique psychiatric vulnerability of some syndromes has been replaced by the opinion that psychiatric disorders occur across diverse epilepsy syndromes. This variability within and across syndromes suggests that phenotypic presentations might be driven by other factors.

In the following, we provide a description of behavioural disorders of some syndromes which will not be discussed in detail in the next chapters of the book.