A chronic illness like epilepsy is a stress factor for the child and his family. The child and family need to recruit coping or behavioral, emotional, and cognitive strategies to decrease the illness related “mismatch” with the environment.^98,218 Adaptive or maladaptive use of these coping strategies depends on the complex interaction of a “multivariate multiprocess system”¹⁴³ that involves the child and family, as well as the social, educational, and medical environments.

Several studies have demonstrated that maternal and family functioning are important predictor(s) of psychological functioning in children with a chronic illness, whereas illness severity merely mediates (moderates) the effects of psychosocial variables; the reader is directed to two reviews on this subject.^248,269 In pediatric epilepsy, maternal depression and family functioning primarily reflect the presence of comorbid behavioral problems in the child rather than illness severity.^164,232 Limited family mastery and lower levels of confidence in managing child problems also predicted behavior problems in children with epilepsy at baseline and 24 months later.¹¹

From the psychodynamic perspective, a child with epilepsy is faced with several stresses. The first stress factor, unpredictability, is a hallmark of epilepsy because the child never knows when a seizure is going to happen.²⁷⁶ From the child’s perspective, this unpredictability can lead to a sense of lack of control, fear, anxiety, or need to be dependent and protected by significant adults.

During development, children acquire self-esteem as they begin to master their environment and can do more things by themselves. This sense of mastery fosters a feeling of competence and control while decreasing the child’s need for dependence. In contrast, the need for dependence produces a sense of helplessness and lack of self-esteem that can lead to feelings of inadequacy, poor self-worth, and ultimately to depression.¹⁴⁴

In addition to unpredictability, the lapse in consciousness during a seizure or the child’s experience of having his limbs do things that he has no control over further increases the sense of lack of control. Similarly, the unpleasant sensation of epigastric distress during the aura of a complex partial seizure can exacerbate the sense of lack of control, fearfulness, and dependence. Furthermore, the manifestations of a seizure are often perceived by others as frightening and grotesque.¹⁶³ The child’s peers might, therefore, react to a seizure in a negative manner.¹⁶³ This could increase the child’s feelings of being different.

Several older studies have shown that children with epilepsy, particularly those with temporal lobe epilepsy (TLE), were more dependent than children with other chronic illnesses.^96,109,243 In these studies, dependence was related to the presence of psychiatric disturbances,^96,109,243 a sense of lack of control, and attribution of failures and successes (particularly those in the social realm) to external unknown sources.¹⁶³ More recent studies have shown that children with newly diagnosed epilepsy feel shame and guilt about having epilepsy¹⁹⁶ and that adolescents with epilepsy are aware of a stigma to this illness.⁹ Caplin et al.³⁶ demonstrated a positive correlation of self-efficacy with positive self-concept and positive attitude toward illness in youth with epilepsy and a negative association of child self-efficacy with child worries and symptoms of depression. Austin et al.¹¹ have shown that greater worry, negative attitude, poor self-concept, and symptoms of depression were associated with an increased sense of stigma in a large sample of children with chronic epilepsy.

Furthermore, the child has to contend with the possible behavioral and cognitive side effects of AEDs. Young children
and sometimes adolescents are often unaware of the connection between these effects and their medication. These side effects could, therefore, contribute to the child’s feeling of not being in control. Finally, despite the previously described stressors, children with epilepsy, like their peers, need to function adequately in the academic and social environment. These are far from trivial stressors.

Three main factors affect the coping skills of children with a chronic illness: the marginality of the sick status, the episodic nature of the disease, and the age of onset of the disorder.^100,202 If a child has visible physical impairments, a disease that is not episodic, and onset of the disorder from the preschool period, she self-identifies as sick and others identify her as different. If the disease has no visible impairments, is episodic, and begins after kindergarten, the child does not feel different from other children. When ill, the child tries to regain and/or achieve normal status.

The emotional burden on the child increases the less visible the handicap and the more normal the child is expected to be.¹⁰⁰ The need to meet these expectations becomes increasingly taxing for the child with epilepsy with recurrent rather than well-controlled seizures.

From the conceptual perspective, children’s understanding of the nature of their disorder is age-related and can impact how well they cope with the illness.²²⁰ Thus, 5- to 7-year-old children are aware of the external aspects of their disorder.²³ For example, they take medication, have repeated blood tests, and visit the doctor. Older children with epilepsy are more aware of the brain’s role in their disorder, and this knowledge sometimes becomes a source of concern for them.

Sanger et al.²²⁰ found that 41% of epileptic children and adolescents between the ages of 5 and 16 years identify epilepsy as a disease that involves the brain. Some children¹⁰ fear that they might die during a seizure, and this fear makes them more anxious and dependent. Many children have misconceptions about seizure disorders and lack disease-related information. Only 41% of the children identified epilepsy as a disease involving the brain.²²⁰

Compared with children with asthma and diabetes, those with epilepsy have more unanswered questions about their illness and feel excluded from discussions with doctors.¹¹⁹ Supporting this finding, McNelis et al.¹⁵⁷ reported that more than half of children with new-onset seizures want more information on seizures, the cause of seizures, treatment, handling future seizures, and protection from injury 6 months after their first seizure. They want to talk to other people with seizures, and need help handling seizures at school. More than 30% have fears or concerns about having another seizure and how to tell others about their seizures.

The impact and ramifications of the unpredictability of seizures is as important for the parents as for the child. It causes a sense of anxiety and is confounded by fear that the child could get hurt¹⁰ or die²⁷⁷ during a seizure. This unpredictability results in the difficult task of having to tread the fine line between what might be perceived as an overanxious or overprotective parent versus a neglectful parent. Thus, mothers of children with epilepsy are more stressed than are parents of children with other chronic illnesses.⁶⁸

Parents vacillate between the need to protect and make sure their child is safe at all times versus the need to allow the child autonomy. The way parents cope with this stress affects the child’s sense of dependence/independence, competence, and self-esteem. The findings of studies on the relationship of maternal anxiety and maternal adjustment with child quality of life, severity of epilepsy, and child adaptation highlight the role psychosocial variables play in how families cope with having a child with epilepsy.

For example, a study conducted on a relatively small sample described mothers as more emotionally involved in the lives of their children with epilepsy compared with their children without epilepsy.¹¹¹ This study also found increased maternal criticism toward the children with epilepsy, and this finding was related to more antisocial behaviors and poorer self-esteem in the children with epilepsy compared with their siblings.

Williams et al.²⁶⁶ found that a significantly higher rate of parents of children with epilepsy sleep with their children than do parents of children with diabetes. They suggested that anxious parents might be more likely to perceive more risks for their children and misinterpret information about their child’s condition, such as the need to sleep with the child to monitor for seizures. In a later study, Williams and colleagues²⁶⁸ demonstrated that families most vulnerable to reduced quality of life were those in which a child has poorly controlled epilepsy, comorbid disabilities, and increased parental anxiety. Similarly, Austin et al.¹⁵ reported that maternal adjustment and attitude was significantly related to the severity of the child’s seizures.

In a large sample of children with epilepsy, Shore et al.²³¹ used a conceptual model based on prior findings in the literature to investigate the associations among maternal and child characteristics, maternal beliefs, and maternal adaptation. They found that most mothers adapt relatively well to their children having epilepsy. However, more than one-third of mothers were depressed and felt inadequate at managing their child’s seizure disorder and maintaining the family’s usual leisure activities. Child behavior problems, maternal satisfaction with family, and maternal learned helplessness had the strongest associations with maternal outcomes. This study did not examine the relationship between maternal and child adaptation.

In terms of child functioning and adaptation, Lotham and Pianka¹⁵⁴ found that measures of child–mother interaction and child self-reliance were significantly related to adjustment measures in children with epilepsy. Chapieski et al.⁴⁸ reported that poor adaptive functioning, based on the Vineland Adaptive Behavior Scale²³⁸ in children with epilepsy was significantly associated with maternal anxiety, as well as with an overprotective and overly directive parenting style.

The few studies that have included both mothers and fathers find similar concerns in both parents of children with new-onset epilepsy.²³³ Although these concerns decrease over a 6–month period, mothers maintain a higher level of concern than do fathers. Fathers have also reported that mothers are more concerned than they are about their child’s epilepsy.⁶⁶ These parental differences can influence what children feel about the impact of their illness on their parents and family functioning. Therefore, studies are needed to determine how differences in parental concerns, anxiety level, and attitude are reflected in child functioning and adaptation.

In addition to the previously reviewed child and parent psychological variables, sociodemographic variables, such as social class, current and previous family stresses, and the presence of other children might also contribute to both the child’s psychosocial adjustment and the family’s coping skills.¹¹⁰ Within the family context, there is a debate in the literature on the impact of chronic illness on healthy siblings; this topic is reviewed by Sharpe.²²⁸ Hoare¹⁰⁸ demonstrated that the siblings of children with new-onset epilepsy are not more maladapted than the siblings of children with new-onset diabetes. However, the siblings of children with chronic epilepsy were more maladapted than the siblings of children with chronic diabetes. These findings led Hoare to conclude that chronic epilepsy has a marked impact on the behavior of the affected child, the siblings, and the family.¹⁰⁸

In terms of the external environment, both the child and parents must deal with the school system. From the psychodynamic perspective, the occurrence of seizures at school can increase the child’s sense of lack of control and present the child with the difficulty of dealing with stigma from peers and
teachers. The child is also often faced with repeated questions by peers after having a seizure at school and also with teachers’ lack of knowledge and misconceptions about epilepsy.²⁴ Even though there is no medical indication, teachers sometimes send children home or to an emergency room when they have a seizure. As a result of repeated absences, the child has more work to catch up, and this may increase the already difficult work load.²⁰² Teachers also might underestimate the child’s intellectual potential due to misconceptions about epilepsy.

From the parental perspective, their efforts to help the child deal with the school system might increase their stress and that of their child. Hoare and Kerley¹¹⁰ have shown that families of children with epilepsy who attend special schools are under more stress than those of children in regular schools. Alternatively, Austin et al.¹⁸ described a significant association between academic achievement and family mastery in children with epilepsy.

Regarding academic achievement, Fastenau et al.⁷² demonstrated a moderating effect of family factors, particularly organized/supportive versus disorganized/unsupportive family structure, on the impact of neuropsychological deficits on the writing and reading skills of children with epilepsy. Supporting the importance of family functioning in the academic performance of children with epilepsy, Mitchell et al.¹⁷⁵ found that parental education rather than seizure variables plays an important role in the IQ of children with epilepsy. Taken together, these psychosocial factors could increase the demands facing the child and affect his academic functioning,⁷² self-esteem,¹⁶² and related social and school functioning.⁴²

From the social perspective, the problems of adolescents with epilepsy deserve special mention. During adolescence, youth acquire more independence, yet vacillate between dependence and independence. Uncontrolled epilepsy could affect this process by increasing dependency. Alternatively, if adolescents experiment with the decision-making, they might make decisions that are potentially harmful for their illness.³³ For example, to conform to what they perceive as the expectations of peers, adolescents might be noncompliant with AEDs, go to sleep late, or use alcohol and drugs. These behaviors could lead to poor seizure control and possible denial of a driver’s license. These circumstances can lead to a significant increase in family stress.¹⁴⁶

In terms of coping skills, a study of a large sample of adolescents with epilepsy has shown increased negative attitude with increasing age and poor seizure control⁹⁹ using the Child Attitude Towards Illness Scale.¹³ The authors of this study suggest that as youth get older, a better understanding of the restrictions and limitations of the illness might underlie the increased negative attitude. Despite seizure control, the mothers of adolescents with epilepsy report significantly more psychological, social, and school adjustment problems than do the mothers of adolescents with asthma.¹⁴ Also, above and beyond the effects of seizure control, Sbarra et al.²²⁴ found that adolescent perception of controlling maternal behavior was significantly associated with youth externalizing behaviors.

Dealing with the stigma of epilepsy during adolescence might also pose additional coping challenges, as reviewed by MacLeod.¹⁵⁷ Westbrook et al.²⁶³ found that only one-third of 64 adolescents with epilepsy confirmed that they felt stigmatized, and this finding was associated with poor self-esteem. Similarly, using the Quality of Life in Epilepsy for Adolescents (QOLIE-AD-48), which includes a stigma scale, Devinsky et al.⁵⁷ found relatively low stigma and good quality of life in youth with epilepsy. In contrast, a 37-item questionnaire designed to measure familiarity with epilepsy, knowledge about epilepsy, and perceptions reflecting stigma filled out by 19,441 adolescents revealed lack of knowledge and misperceptions about epilepsy that suggest a high rate of stigma.¹⁹ In addition, the finding that 70% of the epilepsy patients in the Westbrook et al.²⁶³ study said that they do not talk with anyone about their epilepsy implies that they do not talk about their illness because of stigma.

Finally, the child and family also must interact with the medical system. Two studies, one conducted in the mid 1980s⁵⁰ and one in 2004 (Smith et al., in press) found a marked discrepancy between the concerns of parents of children with epilepsy regarding their children’s needs compared to how the treating pediatric neurologists and pediatricians viewed the parents’ concerns. Whereas the parents were mainly worried about their children’s behavior and performance at school, physicians thought the parents were concerned primarily about the stigma of epilepsy and seizure control. This continued discrepancy, despite increasing evidence for behavioral, cognitive, linguistic, and quality-of-life difficulties experienced by children and adolescents with epilepsy, underscores the stresses faced by parents of children with epilepsy.

In summary, from the psychological perspective, the child’s ability to cope with epilepsy is a function of his intrapsychic experiences, such as sense of mastery and understanding of the illness, how the child’s parents and family deal with the illness, as well as how the school, peer group, and medical system respond to the illness.

The landmark Isle of Wight epidemiologic study on psychopathology in children found an increased rate in children with TLE.²¹⁸ However, very few children in that study met criteria for TLE. Subsequent studies, albeit nonepidemiologic, of psychopathology by seizure type in childhood inconsistently found a relationship between type of epilepsy and psychopathology.

These discrepancies reflect limitations inherent in including relatively small sample sizes of children with epilepsy, and, therefore, small numbers of children with different seizure types. This, in turn, reduces the power to examine possible confounding effects of other seizure variables (e.g., seizure control, age of onset, duration of illness, type and number of AEDs), as well as of comorbid cognitive and linguistic deficits, demographic variables, and premorbid behavior problems.

Moreover, studies differ in whether they classify children by epilepsy syndrome, type of epilepsy, or localization-related epilepsy, or include type of epilepsy as one of several seizure severity measures. In addition, they also vary in the technique used to collect behavioral information on the children (e.g., structured psychiatric interviews, questionnaires), the source of information (e.g., child, parent(s), teacher), and the measures (psychiatric diagnoses, scores).

With these limitations in mind, the findings of earlier studies suggested that a relationship exists between type of seizure disorder and behavioral problems. For example, Nuffield¹⁹⁰ described neurotic traits in childhood absence epilepsy. Several studies suggested that children with TLE have characteristic behavioral disturbances,^{94,110,151,217,241,246} particularly if they have left-sided foci.^151,241 Their behavior problems, hyperactivity, aggression, and antisocial behavior were
associated with male gender, poor seizure control, low IQ, and childhood rages.^{94,110,151,217,241} However, other researchers were unable to identify differences in measures of psychopathology between children with temporal lobe and generalized epilepsy.^103,265

Using structured psychiatric interviews, a recent study on a large sample of children with different types of seizure disorder has demonstrated similar rates of DSM-IV psychiatric diagnoses in 60% of the children with complex partial seizures and 58% of those with childhood absence epilepsy, compared with normal children.¹⁹⁸ Other than schizophrenia-like psychosis in 10% of children with complex partial seizures rather than childhood absence epilepsy,^41,150 children with complex partial seizures and childhood absence epilepsy have similar rates of disruptive, affective/anxiety, and both disruptive and affective/anxiety disorders. Controlling for seizure and demographic variables, IQ was the single predictor of a psychiatric diagnosis in children with complex partial and childhood absence groups.¹⁹⁸

A long-term follow-up study of young adults with a history of childhood absence demonstrated that, compared with a control group of young adults with a juvenile rheumatoid arthritis history, they had a significantly higher rate of behavioral and school/academic problems.²⁷⁰ This finding was unrelated to seizure variables. Thus, childhood absence epilepsy, considered by most neurologists to be completely benign, has both a high rate of psychopathology in childhood and poor outcome regarding behavior and academics by young adulthood.

In children with new-onset seizures, those with prior unrecognized seizures and subsequent development of partial or absence seizures have higher rates of behavior problems (based on Child Behavior Checklist [CBCL],¹ parent,¹² and teacher⁶⁵ total) internalizing, attention, thought, and somatic complaints scores than those with other types of seizures. Other studies on children with new-onset seizures, however, have not shown a relationship between severity of behavioral problems and type of epilepsy.¹⁹⁵ In fact, prior cognitive and behavioral problems, rather than seizure variables, predicted whether children with new-onset seizures continued to have behavior problems over time.¹⁹⁷

A study on localization-related epilepsy describes parental reports of behavioral problems in 39% of 63 children with TLE.⁹⁷ However, this study did not include control groups of children with other types of epilepsy. A study comparing small samples of children with temporal, frontal, and absence epilepsy found increased attentional problems based on parent CBCL scores in those with frontal lobe epilepsy.¹⁰⁴

Among the localization-related epilepsies, even the so-called “benign” epilepsies, benign epilepsy with centrotemporal spikes (BECTS) or benign occipital epilepsy (BOE) may be associated with psychopathology. In the last decade, more reports focus on the subtle behavioral alterations commonly seen with these syndromes. Nonspecific behavioral problems have been reported in 30% of children with centrotemporal spikes,²⁷⁵ and these may be more prevalent in those with atypical features of BECTS.²⁶⁰ Symptoms of attention-deficit hyperactivity disorder (ADHD) have also been reported and may be predicted by certain EEG characteristics.¹⁶¹ Parents report more problems with attention, temper, and impulsiveness in children with BECTS compared with controls.⁵¹ Of interest, centrotemporal spikes were found in 5% to 6% of a large sample of ADHD patients without a history of epilepsy,¹¹⁷ and spike frequency appeared to be related to severity of symptoms.

In summary, irrespective of type of epilepsy, children with epilepsy, both new-onset and chronic epilepsy, have high rates of a wide range of behavioral disturbances. The small study sample sizes, wide age range, and different methods of assessing behavioral abnormalities in these studies highlights the importance of conducting prospective studies of large representative samples of children to demonstrate the consistency of the behavioral problems and their course, and whether type of epilepsy plays a role in the behavioral profile of children and adolescents with epilepsy.

Historically, Lindsay et al.’s¹⁵¹ long-term outcome study of 100 pediatric TLE cases revealed that poor seizure control was associated with increased antisocial behavior in these children. Remission of seizures by age 12 years was associated with a better adult psychosocial outcome than remission after age 12.¹⁵⁰

Hermann et al.¹⁰³ demonstrated that seizure control was the most robust predictor of total and externalizing CBCL scores in a large sample of children with epilepsy, controlling for the effects of demographic, biologic, and psychosocial variables. Numerous subsequent studies have examined this relationship using different behavioral instruments, control groups, informants, measures of seizure control, sample sizes, and youth cognitive ability. Some of these studies were cross-sectional and few were longitudinal. Only a few studies have controlled for possible confounding seizure variables to varying degrees. Retrospective and prospective treatment studies have also investigated the relationship between seizure control and behavioral outcome following intervention with AEDs and epilepsy surgery.

Due to space limitations, this section briefly summarizes the findings of the larger cross-sectional and longitudinal studies (including postsurgical intervention), conducted in children with epilepsy, that have examined the association with seizure control. It also indicates how these findings vary by type of behavioral instrument and informant.

In terms of cross-sectional studies there was no statistically significant relationship of seizure frequency with the presence and type of psychiatric disorders, as well as with CBCL scores in a large sample of children with complex partial seizures who have average IQ scores.⁴³ The DSM-IV psychiatric diagnoses of these children were based on information obtained on the child by a structured psychiatric interview administered separately to the child and to the parent.

In contrast to these findings, Schoenfeld et al.²²⁵ reported significantly higher total and internalizing CBCL scores in children with complex partial seizures with the highest seizure frequency. However, they did not examine the possible effect of IQ on these findings.

Using CBCL scores, Austin et al.¹⁶ also found significantly higher internalizing scores in children with epilepsy, aged 8 to 13 years, who had more severe seizures, a measure that includes seizure control. Follow-up of these children 4 years later revealed significantly higher internalizing CBCL scores in the girls with epilepsy who had more severe seizures both at baseline and follow-up.⁹ These authors concluded that girls with more severe seizures might be more at risk for behavior problems as they moved through the teen years.

Children with new-onset seizures and subsequent recurrent seizures have significantly higher parent-¹⁷ and teacher-based total and internalizing CBCL scores,⁶² suggesting a relationship between seizure control and severity of behavioral problems. In contrast, Oostrom et al.¹⁹⁵ reported lack of persistence in behavior problems based on parent CBCL and teacher reports in their 3-, 12-, and 48-month follow-up study of children with new-onset epilepsy. Despite similar mean parent and teacher ratings, Oostrom et al.¹⁹⁵ found no agreement between parent and teacher ratings for individual children.

In terms of surgical studies, several older studies noted that some children with TLE become less aggressive, and their behavior improves once they attain seizure control after temporal lobectomy.^{54,60,89,118,149} None of these studies, however, used standardized psychiatric instruments and control groups to follow the postoperative behavioral changes of a representative
sample of children undergoing temporal lobectomy. In addition, these studies did not include a control group of medically treated children with epilepsy.

Using the CBCL,¹ Lendt et al.¹⁴⁵ found a significant decrease in the CBCL externalizing, internalizing, attention, and thought scores that had been within the clinical range 3 months after epilepsy surgery on 13 temporal and 11 extratemporal lobectomies, two hemispherectomies, and two callosotomies, compared with a control group of medically treated children with intractable epilepsy. Postsurgical seizure control rather than age, sex, onset, and duration of epilepsy; the site of the focus; and changes in AED regimen accounted for this finding.

In a 1-year follow-up study of a large sample of 30 children who underwent epilepsy surgery, Smith et al.²³⁷ found no significant change in CBCL scores 1 year after surgery, compared with medically treated children with intractable epilepsy. Seizure control was unrelated to these findings.

In summary, the findings of cross-sectional and follow-up studies of children with new-onset and chronic epilepsy demonstrate an association of seizure frequency with the presence and severity of behavioral disturbances. However, recent evidence that verbal IQ predicts psychopathology, children with new-onset seizures have behavior problems prior to the onset of seizures, and children who undergo epilepsy surgery continue to have behavior problems irrespective of postsurgical seizure control suggest that psychopathology is a comorbid component of pediatric epilepsy irrespective of illness variables.

Although there are numerous reports on AED effects on behavior and cognition in children with epilepsy, no double-blind, randomized, controlled studies have examined these relationships in these children. For detailed information on extant studies and reports, the reader is referred to two reviews, one on the mechanisms of action of AEDs and the relevance to neurobehavioral adverse effects,²²² and the other on the cognitive side effects of AEDs in children with epilepsy (see also Chapter 201).¹⁵³

This brief summary of the adverse effects of AEDs on cognition and behavior emphasizes the importance of controlling for these variables in neurobehavioral studies of children with epilepsy. In addition, different children appear to be vulnerable to these AED adverse effects. Thus, as described in adults with epilepsy,^138,181 increased risk for AED-related behavioral and cognitive side effects in children with epilepsy include AED polytherapy,^{32,95,134,151,155} mental retardation or learning disabilities,^32,95 and a family history of psychiatric disorders.^28,30 Similarly, as shown in adults with epilepsy,²⁵⁰ children with a past history of behavioral problems, as well as of prior adverse behavioral and cognitive effects to AEDs, might also be more vulnerable to these adverse effects.

It is important to note that children on high doses of combinations of AEDs might appear to be depressed because they are tired, apathetic, listless, and poorly motivated. Other children on polytherapy or high-dose monotherapy might respond with irritability, low frustration tolerance, poor impulse control, and hyperactivity.

However, AED polytherapy might also reflect the type or types of epilepsy, as well as comorbid cognitive and linguistic difficulties, all of which might increase the likelihood of behavioral problems. Therefore, clinicians should be aware that the functional impact of AED behavioral and cognitive side effects on the individual child’s daily functioning and quality of life might be as severe as that of uncontrolled seizures.

In terms of specific behavioral side effects, AEDs that can trigger depression in children include phenobarbital,^28,30 primidone,¹⁰⁵ and valproic acid.¹⁰⁵ Adult studies also find depression associated with topiramate¹⁸² and levetiracetam.¹⁸⁸ Manic behavior has been described in children with epilepsy treated with felbamate⁴⁵ and ethosuximide.

Case studies describe psychosis in children treated with topi-ramate,²⁰⁶ levetiracetam,^134,274 vigabatrin,³⁵ and zonis-amide.¹⁷⁷ Several case reports describe irritability, hyper-activity, and aggression in children with epilepsy on gabapentin,^129,172 phenobarbital,^34,261 vigabatrin,^3,230 and benzodiazepines.^81,249

Slowing of cognitive processes and motor reaction time are attributed to the older AEDs, such as phenobarbital and benzodiazepines.^{26,34,105,261} In adults with epilepsy, these effects are dose-related for carbamazepine, phenytoin, and valproic acid.⁵⁹ However, there have been no well designed studies on the new AEDs and cognition in children. The impaired word retrieval and cognitive slowing found in children treated with topiramate,^91,229 and the possible association with the speed of titration²²¹ must be studied in children.

To date, most clinicians have been impressed by the lack of adverse cognitive effects in children treated with lamot-rigine.²²¹ The slow titration used in children with epilepsy to avoid the rash associated with lamotrigine treatment might contribute to this observation. However, well-designed studies are needed to examine the effects of the new AEDs on cognition.

In summary, double-blind, randomized studies using well-established behavioral and cognitive instruments sensitive to change are needed to determine which AEDs have behavioral and cognitive adverse effects and for which children with epilepsy. Studies of behavior and cognition in children with epilepsy, however, should include large samples of children to examine the possible role played by AEDs in the studies’ findings. AED polytherapy and high therapeutic blood levels might be associated with more general behavioral problems, such as irritability, aggression, and hyperactivity, as well as slowing of cognitive processes in children. Children with epilepsy at risk for the adverse neurobehavioral effects of AEDs are those with mental retardation, learning disabilities, poor prior behavioral and cognitive responses to AEDs, psychiatric history of past behavioral and cognitive problems, and family history of psychiatric disorders.

Increasing evidence suggests connectivity and structural changes caused by seizures in the brain of young animals despite the lack of cell loss (see review by Holmes¹¹⁴), an age-related effect on synaptic function,¹⁴¹ and long-term cognitive and behavioral changes.²²³ Maternal deprivation in the young animal has synergistic effects, with recurrent seizures in inducing long-term damage in the developing brain.¹²¹ Although the impact of early seizures on brain development might reflect the underlying cause of the seizures rather than the effects of seizures,¹¹³ the findings of animal studies support the previously reviewed clinical data suggesting that the child’s immediate environment (i.e., home, school), rearing practices, recurrent ongoing seizures, and AEDs might impact the development of behavior and cognition.

The age of onset of epilepsy might be an indicator of the severity of the underlying pathology. Thus, early-onset epilepsy is often associated with mental retardation, and children with mental retardation have an earlier onset of epilepsy.²¹⁸ These age relationships imply that epilepsy and mental retardation in children with early onset are epiphenomena of underlying pathologic processes in the brain.

In terms of psychopathology, two epidemiologic studies have demonstrated significantly higher rates of psychopathology
in children with epilepsy who have mental retardation²¹⁸ or complicated epilepsy, including learning disabilities⁵⁵ than those with normal intelligence or uncomplicated epilepsy. A population study of children with both mental retardation and epilepsy also revealed high rates of severe behavior problems.²³⁹

Nevertheless, it is still unclear how and to what degree recurrent seizures, both frequent and infrequent, and the need for high doses of AED polytherapy impair ongoing development of the brain and of emotional, social, cognitive, and linguistic skills in early childhood. It also remains to be determined through prospective studies if the behavioral, cognitive, and linguistic outcome in developmentally disabled children with earlier onset of epilepsy is worse than in children with later onset of their epilepsy. Whereas some studies suggest that epilepsy surgery and subsequent seizure control for early-onset intractable epilepsy changes the developmental course,⁷⁸ others do not.^{44,133,135,237}

Regarding later onset of epilepsy, two of the most common types of pediatric epilepsy, complex partial seizure disorder and primary generalized epilepsy, typically begin in middle childhood. During middle childhood, acceleration occurs in the development of children’s emotional, social, cognitive, and linguistic skills and continued maturation through adolescence. Ongoing seizures, high doses of multiple AEDs, and prolonged seizures might interrupt the functioning of the neural circuits involved in these maturational processes.

In children with onset of epilepsy in middle childhood, although the presence of psychopathology does not appear to be associated with age of onset,⁴³ earlier studies identify an association between early-onset and impaired cognitive functioning.^2,26,70,191 Yet, these findings might also reflect the confounding effects of poor seizure control, as well as AED polytherapy and high blood levels of treating drugs.²⁶