Psychiatric comorbidity in children and youth with epilepsy: An association with executive dysfunction?

Epilepsy & Behavior 56 (2016) 88–94

Contents lists available at ScienceDirect

Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh

Psychiatric comorbidity in children and youth with epilepsy: An association with executive dysfunction? Kristin Å. Alfstad a,⁎, Halvor Torgersen a, Betty Van Roy b, Erik Hessen a,c,i, Berit Hjelde Hansen d, Oliver Henning a, Jocelyne Clench-Aas e, Petter Mowinckel f, Leif Gjerstad g,h, Morten I. Lossius a a

National Centre for Epilepsy, Division for Surgery and Clinical Neuroscience, Oslo University Hospital, Norway Department of Pediatrics and Adolescent Medicine, Akershus University Hospital, Lørenskog, Norway Department of Neurology, Akershus University Hospital, Lørenskog, Norway d Division of Mental Health, Akershus University Hospital, Lørenskog, Norway e The National Institute of Health, Department of Mental Health, Norway f Department of Paediatrics, Oslo University Hospital, Norway g Department of Neurology, Division for Surgery and Clinical Neuroscience, Oslo University Hospital, Norway h Faculty of Medicine, University of Oslo, Norway i Department of Psychology, University of Oslo, Norway b c

a r t i c l e

i n f o

Article history: Received 27 September 2015 Revised 2 January 2016 Accepted 4 January 2016 Available online xxxx Keywords: Children Epilepsy Executive function Psychiatry

a b s t r a c t Objectives: Psychopathology in children and youth with epilepsy has previously been related to executive dysfunction, but the nature of the association is uncertain. We sought to explore risk factors for psychiatric disorders in children and youth with epilepsy, with emphasis on executive dysfunction, along with seizure-related and psychosocial factors. Methods: The cohort consisted of one hundred and one consecutive patients aged 10–19 years with focal (n = 52) or genetic generalized (n = 49) epilepsy. All were screened for psychiatric symptoms, using part of an extensive questionnaire, the Strengths and Difficulties Questionnaire (SDQ) for both patients and their parents. Participants scoring in the borderline or abnormal range on the SDQ received a psychiatric interview (Kiddie-SADS-PL). All participants underwent a neuropsychological examination, and those with general cognitive abilities (IQ) b 70 were excluded. Results: Forty-seven of 101 participants (46.5%) had a SDQ score in the borderline or abnormal range and underwent a psychiatric evaluation. Of these, 44 (93.6%) met the criteria for a psychiatric diagnosis, the most common being ADHD and anxiety. An executive deficit was identified in 26.8% of the participants with a psychiatric diagnosis, but in only 5.4% of those without such a diagnosis (p = 0.003). Multivariate logistic regression analysis showed that executive dysfunction was an independent risk factor for having a psychiatric disorder (OR 8.2, CI 1.8–37.2, p = 0.006), along with male gender (OR 2.9, CI 1.2–7.3, p = 0.02), and early seizure onset (0.86—that is one year older equals risk of psychiatric disorder reduced by 14%—CI 0.77–0.96, p = 0.01). Other epilepsyrelated or psychosocial factors were not significantly associated with psychiatric disorders. Conclusions: Multiple factors are associated with psychiatric problems in children and youth with epilepsy. In this study, executive dysfunction, male gender, and early epilepsy onset were independent risk factors for having a psychiatric disorder. An evaluation of psychiatric and cognitive problems is important to enable a positive longterm outcome in childhood epilepsy. © 2016 Elsevier Inc. All rights reserved.

1. Introduction Psychiatric comorbidity affects quality of life in children and youth with epilepsy (CWE) and is important to address in clinical care [1]. Psychopathology occurs frequently, but prevalence rates show a wide variation (21–60%) depending on the population studied and differences in methodology [2]. Identification of risk factors may enable ⁎ Corresponding author. Tel.: +47 67501000. E-mail address: [email protected] (K.Å. Alfstad).

http://dx.doi.org/10.1016/j.yebeh.2016.01.007 1525-5050/© 2016 Elsevier Inc. All rights reserved.

implementation of preventive measures at an early stage, and this might improve the prognosis. Different risk factors for behavioral problems in CWE have been investigated, but many questions remain unresolved. Psychiatric problems have been examined in relation to various epilepsy syndromes. Whereas some report more psychopathology in focal epilepsy than in generalized (absence) epilepsy [3], others find no difference between syndromes [4]. Underlying structural brain abnormalities that cause epilepsy have been found to be associated with psychopathology in several studies [5]. Poorly controlled seizures or antiepileptic drug

K.Å. Alfstad et al. / Epilepsy & Behavior 56 (2016) 88–94

(AED) polytherapy has also been shown to increase the risk of adverse outcome, but not consistently [2]. Age, gender, and socioeconomic factors are likely to be influential, but even here, findings are inconclusive [2]. The distribution of various psychiatric disorders varies between genders, and some disorders have a later age of onset, contributing to the complex picture. Behavioral deficits may also be present before seizure onset, supporting the importance of neurobiological factors [6]. A cognitive deficit in the domain of executive functions (EFs) has also been found to be a possible factor in behavioral problems in CWE [7]. Neuropsychological function in focal epilepsies has been considered to be closely linked to the region of the brain affected and to mirror the function of that area. However, more widespread deficits are often found. Executive functions, located in the frontal lobe, involve the ability to control and regulate cognition and behavior and to adapt to changing situations. Problems in EF could thus be expected to occur especially in frontal lobe epilepsy (FLE), but such deficits can be found even in temporal lobe epilepsy (TLE) syndromes. For instance, the neuropsychological profile of idiopathic generalized epilepsy, especially juvenile myoclonic epilepsy, shares some common characteristics with FLE [8]. Whether the level of executive function in CWE with psychiatric disorders is lower than that in CWE with no psychiatric disorders is unclear. Our study sought to study this question in a clinically well-defined CWE cohort, with psychiatric interviews providing valid psychiatric diagnoses, complemented by a standardized, well-validated neuropsychological assessment. The aim was to investigate associations between psychiatric disorders and executive dysfunction, along with epilepsyrelated and psychosocial factors in children and youth with epilepsy. 2. Methods 2.1. Patient inclusion Patients aged between 10 years and 19 years were included consecutively in the study during hospital stays at the National Centre for Epilepsy, the only tertiary epilepsy center in Norway, from January 2012 to June 2014. Informed written consent was obtained from parents or participants of legal age; children gave their assent. The patients/parents who declined participation gave written consent that clinical data from their journals could be used by the research group. This enabled comparisons of demographic and epilepsy-specific information between eligible nonparticipants and the study group. The study was approved by the Regional Ethics Committee (2011/1636/ REK sør-øst B). 2.2. Clinical data Participants and parents were interviewed, and all clinical records were extensively reviewed. All but one participant had a 24-hour EEG recording, and video surveillance for seizure detection during sleep is routinely used at the hospital. Electroencephalogram findings were classified as epileptiform activity present or absent during recording. All EEGs were interpreted by a neurologist experienced in neurophysiology. All participants had MRI, except nine participants with either benign childhood epilepsy with centrotemporal spikes (BECTS) or genetic generalized epilepsies (GGE). The International League Against Epilepsy (ILAE) classification of seizures and epilepsies was used [9]. Two experienced neurologists classified each case independently; interclassification reliability was 93%. For nonconsensus or difficult cases, a child neurologist reviewed the cases. Genetic generalized epilepsies included were: generalized epilepsy with febrile seizures plus (GEFS+), childhood absence epilepsy (CAE), juvenile absence epilepsy (JAE), juvenile myoclonic epilepsy (JME), generalized tonic–clonic seizures (GTCs) only, and GGE not otherwise specified (nos).

89

Focal epilepsies were classified as: temporal lobe (TLE), frontal lobe (FLE), BECTS, or focal epilepsy from anterior brain regions, but difficult to subclassify (focal epilepsy nos). Patients with focal epilepsies from the posterior regions (parietal or occipital) were excluded, as were patients with intellectual disability (See Fig. 1 for more details on inclusion and exclusion criteria and examination protocol). Seizure frequency was dichotomized, and comparisons made between the group of participants that had not experienced generalized tonic–clonic seizures (GTCs) during the previous 6 months and the group of participants that had experienced one or more GTCs during the same period. The numbers of participants on AEDs with possibly the highest potential for cognitive (i.e., phenobarbital, benzodiazepines, topiramate, zonisamide) or behavioral (levetiracetam) side effects were also analyzed [10,11]. Participants and parents completed a 110-item health questionnaire covering different topics, including sociodemographic and psychosocial conditions, school participation and well-being, and physical and mental health. The questionnaire also contained questions on family income and if the children were living with both parents or not. A low-income group was defined as below 60% of the median income in Norway (European Union definition), and the variable “single parent” was used if the child's parents were not living together. 2.3. Psychiatric evaluation In order to assess psychiatric symptoms, the Strengths and Difficulties Questionnaire (SDQ), both self (patient) and parent reports, were administered as part of the health questionnaire. The SDQ is a behavioral screening tool, widely used in both epidemiological and clinical studies, and has good psychometric properties [12]. The questionnaire consists of 25 items and can be divided into 5 subscales, covering emotional symptoms, conduct problems, hyperactivity, and peer problems, as well as prosocial behavior. The items can be rated “not true” (0), “somewhat true” (1), or “certainly true” (2), and the scores on the first 4 subscales listed are combined to obtain a total difficulty score (0–40). The total scores can be classified as either “normal”, “borderline”, and “abnormal”. Using British norms (Norwegian norms are not established) on the self report, a score of 0–15 is rated as being in the normal range, 16–19 is borderline, and N19 as abnormal; in the parent report, the corresponding scores are: normal, 0–13;, borderline, 14–16; and abnormal, N16. Patients scoring as borderline or abnormal in either the self or parent reports underwent a psychiatric examination for a clinically validated diagnosis. This methodological approach has been previously used also by other research groups [13]. The Schedule for Affective Disorders and Schizophrenia for School-Age Children—Present and Lifetime version (Kiddie-SADS-PL), a semistructured diagnostic interview, was used to assess psychiatric disorders according to DSM-IV criteria [14]. The examinations were performed by one of two experienced child psychiatrists, blinded to SDQ findings. Ten participants were examined jointly by both examiners in order to validate the evaluation. Both patients and parents were interviewed, and the study reports on the prevalence of present diagnoses. The psychiatric diagnoses were grouped as: externalizing (disruptive behavior disorder and substance abuse), internalizing (depressive disorders, anxiety disorders, adjustment and stress disorders), and neuropsychiatric disorders (ADHD, Tourette's syndrome, tics, obsessive compulsive disorder, psychosis, and autism spectrum disorder). Patients with a psychiatric disorder requiring further treatment after the hospital stay were referred to the local psychiatric health-care system. 2.4. Neuropsychological assessment The patients underwent a neuropsychological assessment, with the main focus on three cognitive domains: verbal intelligence, delayed

90

K.Å. Alfstad et al. / Epilepsy & Behavior 56 (2016) 88–94

Fig. 1. Flow chart, inclusion and exclusion criteria.

memory, and executive functions (EFs). For assessing verbal IQ, the Wechsler Abbreviated Scale of Intelligence was used [15]; delayed memory assessment consisted of Rey auditory verbal learning test (delayed recall) [16,17] and Rey complex figure Test (delayed reproduction) [16,18]; executive functions were assessed using Delis–Kaplan Executive Function System (number–letter switching, letter fluency, and color–word interference test) [19] (Table 1). The raw scores from the tests of memory and EF were converted to T-scores, a normally distributed scale with a mean score of 50 and a standard deviation (SD) of 10. Composite memory and executive scores were computed based on average T-scores for the memory and executive tests. For the cutoff between normal and abnormal neuropsychological functions, we used 1.5 SD below normative mean (IQ ≤ 78 and T-score ≤ 35). An experienced neuropsychologist was responsible for the study test protocol and testing. The results from the neuropsychological assessment were summarized in a written report provided to all participants.

and significant factors were then included in the multivariate analysis. We applied Hosmer's step-down procedure to all variables significant at the 0.25 level in the bivariate analyses, removing the least significant one at a time until significant variables remained in the final model [20]. Results of bivariate and multivariate analysis are presented with odds ratios (OR) with 95% confidence intervals (CI) and p-values. All tests were two-sided and performed at a 5% significance level. The Statistical Package for Social Sciences (SPSS version 21) was used.

2.5. Statistical analysis

3.1. Psychiatric findings

Group comparisons were tested using Pearson's Chi-square for categorical variables and independent sample t-tests for continuous variables. Logistic regression analysis was performed to search for risk factors for psychopathology. Bivariate analyses were first performed,

Borderline or abnormal scores on the SDQ self or parent report were recorded in 47 of 101 participants (46.5%), all of whom underwent clinical diagnostic interview. A psychiatric diagnosis was confirmed in 44 of these, i.e., 93.6% with a borderline/abnormal SDQ score fulfilled the criteria of a validated psychiatric diagnosis. In the total cohort, a prevalence of psychiatric disorders was thus found to be 43.6%. The most common diagnoses were ADHD (diagnosed in 31.7% of all participants) and anxiety (diagnosed in 23.8%). Other diagnoses are described in Table 4. Psychiatric diagnoses were more frequent in boys (57.1%) than girls (30.8%, p = 0.008). More than one diagnosis was present in 26.7% of participants, and more boys (36.7%) had ≥ 2 diagnoses than girls (17.3%, p = 0.03).

Table 1 Neuropsychological tests used in the study. Cognitive domain

Tests

References

General IQ and verbal intelligence

Wechsler abbreviated scale of intelligence (WASI) (word comprehension and similarities) Rey auditory verbal learning test (RAVLT), delayed recall Rey complex figure test (RCFT), delayed reproduction Delis–Kaplan executive function system (D–KEFS) (number–letter switching, letter fluency, and color word interference test) (inhibition/Stroop 3)

[15]

Verbal and visual memory

Executive functions

[16,17] [16,18] [19]

3. Results Initially 123 participants were included, but 22 had to be excluded because of various reasons, and 101 were included in the final analysis (Fig. 1, Flow Chart). After classification, 52 participants were diagnosed with focal epilepsy and 49 with GGE. Further background characteristics are provided in Table 2.

3.2. Neuropsychological findings The mean full-scale IQ of all participants was 93.9 (SD 12.3). On the executive score, 13.9% of participants scored below cutoff, 23.8% were below cutoff on verbal IQ, and 20.8% were below cutoff on the memory score. Poor EF was significantly more prevalent in the group with a

K.Å. Alfstad et al. / Epilepsy & Behavior 56 (2016) 88–94 Table 2 Background characteristics of participants. Variables

Frequency n (%) (n = 101)

Gender male Age at epilepsy onset mean (range, SD) Age at examination mean (range, SD) Epilepsy duration mean (range, SD) Epilepsy syndrome –Temporal lobe epilepsy –Frontal lobe epilepsy –Benign childhood epilepsy with centrotemporal spikes –Focal epilepsy not otherwise specified –Absence epilepsy –Juvenile myoclonic epilepsy –Generalized tonic–clonic seizures only –Generalized epilepsy with febrile seizures plus –Genetic generalized epilepsy not otherwise specified Etiology –Structural pathology –Genetic –Unknown Seizure frequency –No generalized tonic–clonic seizures during last 6

49 (48.0) 8.1 (b1–17 years, 4.1) 14.1 (10–19 years, 2.4) 6.0 (1–16 years, 4.0) 13 (12.9) 19 (18.8) 12 (11.9) 8 (7.9) 17 (16.8) 7 (6.9) 5 (5.0) 3 (3.0) 17 (16.8)

41 (40.6) 53 (52.5) 7 (6.9) 32 (31.7) 94.0 (71–125, 12.3) 91.1 (68–122, 12.0) 44.2 (24–65, 8.4) 44.3 (19–72, 11.4) 30 (29.7) 16 (15.8) 44 (43.6)

Table 3 Comparison of participants with or without a psychiatric disorder. Psychiatric disorder n (%)/mean (SD) (n = 44)

No psychiatric disorder n (%)/mean (SD) (n = 57)

28 (63.6) 16 (36.4) 6.8 (3.8) 13.5 (2.1) 6.7 (4.4)

21 (36.8) 36 (63.2) 9.2 (4.1) 14.5 (2.6) 5.4 (3.6)

p-Value

22 (50.0) 22 (50.0) 24 (54.5)

30 (52.6) 27 (47.4) 31 (54.4)

ns

5 (11.4)

9 (15.8)

ns

15 (34.1) 12 (27.3)

26 (45.6) 20 (35.1)

ns ns

92.7 (11.6) 89.0 (11.5) 42.0 (9.6)

95.1 (12.8) 92.7 (12.2) 45.8 (7.1)

ns ns 0.03

43.1 (12.4)

45.3 (10.6)

ns

16 (36.4) 9 (20.5)

14 (24.6) 7 (12.3)

ns ns

0.008

0.003 0.03 ns ns

Abbreviations used: ns = not significant, GGE = genetic generalized epilepsy, GCTs = generalized tonic–clonic seizures, AED = antiepileptic drug.

Psychiatric disorder

Total n (%) (n = 101)

Boys (n = 49)

Girls (n = 52)

p-Value

ADHD Anxiety disordersa Disruptive behavior disordersb Autism spectrum disordersc Depressive disordersd ≥2 psychiatric diagnoses

32 (31.7) 24 (23.8) 10 (9.9) 9 (8.9) 7 (6.9) 27 (26.7)

22 (44.9) 14 (28.6) 8 (16.3) 6 (12.2) 6 (12.2) 18 (36.7)

10 (19.2) 10 (19.2) 2 (3.8) 3 (5.8) 1 (1.9) 9 (17.3)

0.06 ns ns ns 0.04 0.03

55 (54.5)

psychiatric disorder (26.8%) than in those without (5.4%, p = 0.003). However, similar associations were not found for low verbal ability or poor memory (Table 3). There was a higher prevalence of executive

Gender –Boys –Girls Age at epilepsy onset (years) Age at examination (years) Epilepsy duration Epilepsy syndrome –Focal epilepsy –GGE Seizure frequency (no GTC in last 6 months) MRI structural lesion (missing = 4) Antiepileptic drugs –≥2 AED (polytherapy) EEG no epileptiform activity (missing = 1) Full scale IQ Verbal IQ Composite executive score (missing = 4) Composite memory score (missing = 3) Single parent (missing = 2) Low income (missing = 15)

Table 4 Most frequent psychiatric disorders, as a proportion (%) of whole cohort and related to gender.

Abbreviations used: ns = not significant, ADHD = attention-deficit hyperactivity disorder. a Anxiety disorders included: panic disorder, agoraphobia, separation anxiety, social anxiety, specific phobia, generalized anxiety disorder, anxiety disorder not otherwise specified. b Disruptive behavior disorders included: conduct disorder, oppositional defiant disorder, disruptive behavior disorder not otherwise specified. c Autism spectrum disorders included: Asperger's disorder, pervasive developmental disorder not otherwise specified. d Depressive disorders included: major depressive disorder, dysthymia, depression not otherwise specified.

14 (13.9) 49 (48.5) 38 (37.6)

months Antiepileptic drugs –2 or 3 antiepileptic drugs –1 antiepileptic drug –No antiepileptic drugs EEG—no epileptiform activity (missing = 1) Full scale IQ mean (range, SD) Verbal IQ mean (range, SD) Composite executive score mean (range, SD) (missing = 4) Composite memory score mean (range, SD) (missing = 3) Single parent (missing = 2) Low income (missing = 15) Psychiatric disorder

Variables

91

dysfunction within all different subgroups of psychiatric diagnoses (externalizing, internalizing, and neuropsychiatric disorders) than in the group with no psychiatric disorder. Further analyses were not performed at the subgroup level of psychiatric diagnoses because of small numbers. 3.3. Seizure findings Mean age at epilepsy onset was 8.1 years (SD 4.1). Mean age at epilepsy onset was significantly lower in those with a psychiatric diagnosis (6.8 years, SD 3.8) than in those with no psychiatric diagnosis (9.2 years, SD 4.1, p = 0.003) (Table 3). Mean age of onset in children with focal epilepsies was 8.8 years (SD 3.8), and for those with GGE, the mean age at onset was 7.4 years (SD 4.4) (ns). There was no significant difference between focal epilepsy and generalized epilepsy regarding the presence of a psychiatric diagnosis. The mean epilepsy duration was 6.0 years (SD 3.9), and half the cohort had an epilepsy duration ≤4 years. A more prolonged period of having had epilepsy was not associated with an increased risk of being diagnosed with a psychiatric disorder. Although age at examination was significantly lower in those with a psychiatric disorder (Table 3), this correlated with age at epilepsy onset, and regression analysis with this variable was not performed. Generalized tonic–clonic seizure frequency, epilepsy syndrome, and MRI findings had no significant impact on psychiatric comorbidity (Table 3). Comparison of study participants with those who had declined participation in the study revealed no significant differences regarding gender, age at seizure onset, or epilepsy syndrome. However, GTC frequency was significantly different between these two groups; no GTC during the last 6 months was more often seen in the nonstudy group (84.4%) than in the participants (54.5%, p = 0.002). 3.4. AED findings The mean number of AEDs used per patient was 1.4 (range 0–3, SD 0.72); AED monotherapy vs polytherapy had no significant impact on psychiatric comorbidity (Tables 2 and 3). The use of AEDs with maybe the greatest risk of cognitive side effects (phenobarbital (n = 0), benzodiazepines (n = 7), topiramate (n = 9), or zonisamide (n = 4)) did not differ between children with or without a psychiatric diagnosis, or executive deficit, or between the groups of focal and generalized epilepsies. A possible negative effect of levetiracetam (n = 27) on behavior was also analyzed, but no difference was found regarding the presence or absence of a psychiatric disorder.

92

K.Å. Alfstad et al. / Epilepsy & Behavior 56 (2016) 88–94

3.5. Psychosocial findings Low family income or living with a single parent did not differ significantly between the group with a psychiatric disorder and those with no psychiatric comorbidity (Table 3). 3.6. Independent risk factors Bivariate logistic regression analysis showed that executive dysfunction (OR = 6.5, CI 1.7–25.1, p = 0.007), male gender (OR = 3.0, CI 1.3–6.8, p = 0.008), and early seizure onset (0.86; that is one year older equals risk of psychiatric disorder reduced by 14%, CI 0.77– 0.95, p = 0.005) were significantly associated with psychiatric comorbidity (Table 5). These three factors remained significant in the multivariate analysis: executive dysfunction (OR 8.2, CI 1.8–37.2, p = 0.006), male gender (OR 2.9, CI 1.2–7.3, p = 0.02), and early seizure onset (0.86—that is one year older equals risk of psychiatric disorder reduced by 14%—CI 0.77–0.96, p = 0.01) (Table 5). 4. Discussion The main finding of our study is that executive dysfunction was an independent risk factor for psychiatric comorbidity. There are several possible explanations for the association between psychopathology and executive dysfunction. Impairment in EF occurs in various neuropsychiatric disorders in children in the general population, and the deficit could be considered an integral part of the psychiatric disorder [21]. Executive function includes a wide range of cognitive processes, such as planning, problem solving, flexibility, and regulation of behavior [22]. It has been argued that dysfunction in this domain could cause increasing difficulties with age, as children are expected to develop independence and learn to demonstrate appropriate social behavior, judgment, and complex reasoning as they grow older [23]. A lower level of functioning in this domain is likely to cause problems in everyday life and complicate interactions with peers. Executive dysfunction could thus also represent an independent risk factor for different psychiatric disorders in CWE. It is also of importance to note the possible bidirectional relationship between executive dysfunction and psychiatric disorders; having a psychiatric disorder could also be a predictor of executive difficulties. Behavioral problems have also been shown to be associated with lower EF in other studies of children and adults with epilepsy [7,24]. Baum et al. investigated 229 children with a first recognized seizure by neuropsychological testing, and evaluated behavioral problems using the child behavior check list (CBCL). The authors suggest a possible connection between cognitive and emotional dysregulations related to executive dysfunction and to depressive and anxious mood symptoms [7]. However, in their study, the psychiatric symptoms were not validated into psychiatric diagnoses. The prevalence of executive dysfunction in CWE is not well established. In our study, 13.9% of children and teenagers exhibited executive deficit, with the cutoff used (1.5 SD). In a population-based Norwegian study using the 1.3 SD (i.e., the 10th percentile) as cutoff, a higher prevalence (31%) of executive dysfunction was found as compared with our study [25]. This higher prevalence is likely to be due to using the lower cutoff and inclusion of patients who probably Table 5 Risk factors for psychiatric disorders. Variable

Bivariate analysis OR (95% CI)

Executive score (missing = 4) Male gender Age at epilepsy onset

Multivariate analysis p-Value

OR (95% CI)

p-Value

6.5 (1.7–25.1)

0.007

8.2 (1.8–37.2)

0.006

3.0 (1.3–6.8) 0.86 (0.77–0.95)

0.008 0.005

2.9 (1.2–7.3) 0.86 (0.77–0.96)

0.02 0.01

had intellectual disability [25]. In another publication by the same group, EF deficit was reported in all epilepsy syndromes apart from BECTS [26]. Children with epilepsy (aged 6–12 years) showed significantly lower mean values in EF than controls [26]. In a study of children with new onset seizures, 10% of the cohort was found to have executive dysfunction, measured as 1 SD below the mean, while 5.7% of the cohort had a deficit if a cutoff of 1.5 SD was used [7]. This low prevalence is probably due to the population-based design and perhaps also because the participants were investigated after their first recognized seizure. Using a cutoff mean N 1.3 SD lower than healthy sibling controls, another study reports an executive deficit frequency of 13.9% [27]. Comparing different studies is difficult because of the various batteries of neuropsychological tests used. Another methodological issue is the choice of cutoff used to define a neuropsychological deficit. While the optimal cutoff is debatable, the criterion we chose has been employed in previous studies of cognitive impairment and is commonly recommended in the diagnosis of mild cognitive impairment [28]. Thus, the prevalence of executive dysfunction in various groups of CWE is largely unknown, but is likely to show considerable variation depending on the population studied and the methodology used. Another cognitive domain that has been examined in relation to psychopathology is language. An association between poor language skills and psychiatric disorders was demonstrated in a study of children with focal seizures [29]. A correlation was seen between low verbal IQ and having a psychiatric disorder [29]. In our study, the association between low verbal IQ and a psychiatric disorder was borderline significant, and such a deficit could be of importance, but was not visible in our study setup. Several other risk factors for psychopathology have been investigated in previous studies. A clear gender difference was observed in our cohort, with more boys having a psychiatric diagnosis. In contrast to this, a community-based UK study found that female gender predicted the presence of depression [30], and investigations of gender as a risk factor for a psychiatric diagnosis have shown inconsistent findings [2]. Many studies have used screening questionnaires for psychiatric symptoms, with no validated psychiatric diagnoses [31]. Some psychiatric disorders may be more frequent in boys, while other disorders are more common in girls, and this difference would not be apparent in studies without validation into different psychiatric disorders. The age range of participants included can also influence the results. Some studies demonstrate a higher prevalence of ADHD in young children, whereas the typical age for onset of depression is during adolescence [32]. In our study we detected more boys with ADHD (p = 0.06), no gender difference regarding anxiety, and for other psychiatric disorders, the numbers were too low for meaningful analysis. It should be noted that we used a clinical cohort from an epilepsy center (i.e., with an inherent selection bias). However, our findings are consistent with those of other studies, in both epidemiological and clinical cohorts, which also report ADHD and anxiety disorders as being the most frequent psychiatric conditions in childhood epilepsy [5]. The prevalence of psychiatric symptoms, as measured by a borderline/abnormal score on the SDQ, was obtained for 46.5% of participants in our study. Almost 94% of participants with a borderline or abnormal SDQ score fulfilled the criteria of a psychiatric diagnosis. The number of children with a validated psychiatric diagnosis in our cohort (43.6%) is in the range reported from other studies [33]. The symptom load in our study population is underlined by 27 (26.7%) of the participants having more than one psychiatric diagnosis. This is very similar to the results found in a population-based study, where 26% of the children had more than one neurobehavioral condition [34]. The frequency of CWE with several psychiatric diagnoses has otherwise been the focus of very few studies. How seizure variables influence psychopathology is uncertain. We found that children with early seizure onset had a significantly higher occurrence of psychiatric comorbidity. Most previous studies have found that age at seizure onset does not predict psychopathology [35]. As age at seizure onset is related to epilepsy syndrome, methodological

K.Å. Alfstad et al. / Epilepsy & Behavior 56 (2016) 88–94

issues may again be relevant. If age at epilepsy onset is used as a categorical variable, the question of cutoff arises as there is no accepted definition of “young age”. We therefore chose to treat age at onset as a continuous variable. Other seizure variables, such as seizure frequency, showed no association with psychiatric disorders in our study, and seizure-related variables have not consistently been shown to predict psychopathology in children with normal intelligence in other studies [5,35]. Regarding sociodemographic factors, psychosocial burden has been proposed as a risk factor for psychiatric problems in some studies of the epilepsy population [1,31]. Psychiatric disorders did not occur significantly more often in participants from low-income or single-parent families in our study. As discussed above, many different factors may contribute to the psychiatric comorbidity seen in CWE. Antiepileptic drugs can have side effects that contribute to both psychiatric symptoms [11,36] as well as cognitive problems [10,11]. In our study, use of those AEDs with perhaps the greatest risk of cognitive or behavioral side effects was not significantly different in participants with or without a psychiatric disorder or executive deficit. Furthermore, AED polytherapy was not associated with a higher prevalence of psychiatric disorders. Our study was not designed to analyze the effects of specific AEDs on psychiatric disorders, but whether the use of some AEDs might have had a noticeable effect on outcome was considered. When comparing the group of participants in the study with those who declined to participate, a significant difference in seizure frequency was found. The participants had a greater likelihood of having experienced GTCs during the last 6 months than the group who declined participation. This might cause a bias towards more severe epilepsy in the study group. One strength of our study is that all patients were thoroughly examined according to epilepsy diagnosis and syndrome. Classifying epilepsy in children can be challenging, and we conducted an extensive review of all available information. Neuropsychological testing with examination of important domains is another strength. An appropriate test battery is likely to provide more reliable results than a more narrow investigative strategy [37]. Nevertheless, it is important to be aware that EF is a complex concept and not all features were tested in our study. The same neuropsychological test battery was applied to all participants. Only children and adolescents scoring in the borderline or abnormal range on the SDQ total score were examined by psychiatric interview. A normal score obviously does not exclude a psychiatric disorder, and therefore, we could have missed some cases, although the screening properties of the SDQ are fairly robust [12]. In addition, children scoring in the borderline range were also examined, and both self and parent reports were considered, thereby probably increasing sensitivity. Many previous studies have examined only selected factors; however, when elements are multifactorial, it is important to include as many relevant factors as possible. In our cohort, all patients were thoroughly examined according to epilepsy diagnosis/syndrome, psychiatric status, and neuropsychological functioning in one model. Our study design does not investigate causality, but it seems plausible that an executive deficit could increase the risk of developing psychiatric problems. The nature of associations between cognitive deficits and psychiatric disorders are still largely unexplored. In searching for psychopathology risk factors, it is important to include both the psychiatric and neuropsychological status in CWE and analyze these together with epilepsy-related and sociodemographic factors, preferably at seizure onset. Mental disorders are common in people with epilepsy, and onset during childhood is not uncommon. With proper diagnosis and treatment, it is possible to improve the prognosis and reduce the burden of epilepsy. Recognition of both psychiatric and cognitive functions is necessary in order to ensure a positive long-term outcome in childhood epilepsy [1]. Executive function plays a critical role in managing everyday life, in developing social relationships, and in handling interactions

93

with peers [38,39]. A cognitive deficit, perhaps particularly in the executive domain, may be a part of the explanation for the long-term negative psychosocial outcomes seen in the epilepsy population, even during remission [40]. In conclusion, our study from a tertiary epilepsy center population demonstrates a psychiatric diagnosis in almost 44% of CWE. Risk factors for psychiatric comorbidity were executive deficit, male gender, and young age at seizure onset. Psychiatric and cognitive problems should be screened for routinely in children with epilepsy as they can have a substantial negative effect on long-term outcome. Acknowledgments Financial support was obtained from the National Centre for Epilepsy and the Department of Research, Division of Surgery and Clinical Neuroscience, Oslo University Hospital, and from the Norwegian Chapter of the ILAE. We thank all participants and parents for the time and effort and for their trust and openness. We thank Lucy Robertson for linguistic assistance and critical reading of the manuscript. Ethical approval We confirm that we have read the Journal's position on issues involved in the ethical publication and affirm that this report is consistent with those guidelines. Conflict of interest None of the authors has any conflict of interest related to this article to disclose. References [1] Lin JJ, Mula M, Hermann BP. Uncovering the neurobehavioral comorbidities of epilepsy over the lifespan. Lancet 2012;380:1180–92. [2] Ekinici O, Titus JB, Rodopman AA, Berkem M, Trevathan E. Depression and anxiety in children and adolescents with epilepsy: prevalence, risk factors, and treatment. Epilepsy Behav 2009;14:8–18. [3] Caplan R, Siddarth P, Gurbani S, Hanson R, Sankar R, Shields WD. Depression and anxiety disorders in pediatric epilepsy. Epilepsia 2005;46:720–30. [4] Ott D, Caplan R, Guthrie D, Siddarth P, Komo S, Shields WD, et al. Measures of psychopathology in children with complex partial seizures and primary generalized epilepsy with absence. J Am Acad Child Psychiatry 2001;40(8):907–14. [5] Hamiwka L, Jones JE, Salpekar J, Caplan R. Child psychiatry. Epilepsy Behav 2011;22: 38–46. [6] Austin JK, Harezlak J, Dunn DW, Huster GA, Rose DF, Ambrosius WT. Behavior problems in children before first recognized seizures. Pediatrics 2001;107:115–22. [7] Baum KT, Byars AW, deGrauw TJ, Dunn DW, Bates JE, Howe SR, et al. The effect of temperament and neuropsychological functioning on behavior problems in children with new-onset seizures. Epilepsy Behav 2010;17:467–73. [8] Hommet C, Sauerwein HC, De Toffol B, Lassonde M. Idiopathic epileptic syndromes and cognition. Neurosci Biobehav Rev 2006;30:85–96. [9] Berg AT, Berkovic SF, Brodie MJ, Buchhalter J, Cross JH, van Emde Boas W, et al. Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005–2009. Epilepsia 2010;51(4):676–85. [10] Mula M, Trimble MR. Antiepileptic drug-induced cognitive adverse effects: potential mechanisms and contributing factors. CNS Drugs 2009;23(2):121–37. [11] Loring DW, Marino S, Meador KJ. Neuropsychological and behavioral effects of antiepileptic drugs. Neuropsychol Rev 2007;17:413–25. [12] Goodman R. Psychometric properties of the strengths and difficulties questionnaire. J Am Acad Child Adolesc Psychiatry 2001;40:1337–45. [13] Heiervang E, Stormark KM, Lundervoll AJ, Heimann M, Goodman R, Posserud M, et al. Psychiatric disorders in Norwegian 8 to 10 year olds: an epidemiological survey of prevalence, risk factors, and service use. J Am Acad Child Adolesc Psychiatry 2007;46(4):438–47. [14] Kaufman J, Birmaher B, Brent D, Rao U, Flynn C, Moreci P, et al. Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version (K-SADS-PL): initial reliability and validity data. J Am Acad Child Adolesc Psychiatry 1997;36:980–8. [15] Wechsler D. Wechsler abbreviated scale of intelligence (WASI). San Antonio, TX: The Psychological Corporation; 1999. [16] Rey A. L'examen clinique en psychologie. Paris: Presses Universitaires de France; 1964. [17] Schmidt M. Rey Auditory and Verbal Learning Test. A handbook. Los Angeles: Western Psychological Services; 1996.

94

K.Å. Alfstad et al. / Epilepsy & Behavior 56 (2016) 88–94

[18] Meyers JE, Meyers KR. Rey Complex Figure Test and Recognition Trial. Odessa, Florida: Psychological Assessment Resources; 1995. [19] Delis DC, Kaplan E, Kramer JH. Delis and Kaplan Executive Function System: examiner's manual. San Antonio, TX: The Psychological Corporation; 2001. [20] Hosmer DW, Lemeshow S. Applied logistic regression. 2nd ed. New York, USA: John Wiley & Sons; 2000. [21] Hosenbocus S, Chahal R. A review of executive function deficits and pharmacological management of children and adolescents. J Can Acad Child Adolesc Psychiatry 2012; 21(3):223–9. [22] Chan RCK, Shum D, Toulopoulou T, Chen EY. Assessment of executive functions: review of instruments and identification of critical issues. Arch Clin Neuropsychol 2008;23:201–16. [23] Powell KB, Voeller KKS. Prefrontal executive function syndromes in children. J Child Neurol 2004;19:785–97. [24] Walsh J, Thomas RH, Church C, Rees MI, Marson AG, Baker GA. Executive functions and psychiatric symptoms in drug-refractory juvenile myoclonic epilepsy. Epilepsy Behav 2014;35:72–7. [25] Høie B, Sommerfelt K, Waaler PE, Alsaker FD, Skeidsvoll H, Mykletun A. The combined burden of cognitive, executive function, and psychosocial problems in children with epilepsy: a population-based study. Dev Med Child Neurol 2008;50: 530–6. [26] Høie B, Mykletun A, Waaler PE, Skeidsvoll H, Sommerfelt K. Executive functions and seizure-related factors in children with epilepsy in western Norway. Dev Med Child Neurol 2006;48:519–25. [27] Fastenau PS, Johnson CS, Perkins SM, Byars AW, deGrauw TJ, Austin JK, et al. Neuropsychological status at seizure onset in children. Risk factors for early cognitive deficits. Neurology 2009;73:526–34. [28] Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, et al. The diagnosis of mild cognitive impairment due to Alzheimer's disease: recommendations from the National Institute on Aging–Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 2011;7(3):270–9.

[29] Caplan R, Siddarth P, Gurbani S, Ott D, Sankar R, Shields WD. Psychopathology and pediatric complex seizures: seizure-related, cognitive, and linguistic variables. Epilepsia 2004;45(10):1273–81. [30] Turky A, Beavis JM, Thapar AK. Psychopathology in children and adolescents with epilepsy: an investigation of predictive variables. Epilepsy Behav 2008;12:136–44. [31] Alfstad KÅ, Clench-Aas J, Van Roy B, Mowinckel P, Gjerstad L, Lossius MI. Psychiatric symptoms in children with epilepsy aged 8–13 years: effects of age and gender? Epilepsia 2011;52(7):1231–8. [32] Thome-Souza S, Kuczynski E, Assumpção Jr F, Rzezak P, Fuentes D, Fiore L, et al. Which factor may play a pivotal role on determining the type of psychiatric disorder in children and adolescents with epilepsy? Epilepsy Behav 2004;5:988–94. [33] Davies S, Heyman I, Goodman R. A population survey of mental health problems in children with epilepsy. Dev Med Child Neurol 2003;45:292–5. [34] Reilly C, Atkinson P, Das KB, Chin RF, Aylett SE, Burch V, et al. Neurobehavioral comorbidities in children with active epilepsy: a population-based study. Pediatrics 2014;133:e1586–93. [35] Austin J, Caplan R. Behavioral and psychiatric comorbidities in pediatric epilepsy: toward an integrative model. Epilepsia 2007;48(9):1639–51. [36] Halma E, de Louw AJA, Klinkenberg S, Aldenkamp AP, IJff DM, Majoie M. Behavioral side-effects of levetiracetam in children with epilepsy: a systematic review. Seizure 2014;23(9):685–91. [37] Rzezak P, Valente KD, Duchowny MS. Temporal lobe epilepsy in children: executive and mnestic impairments. Epilepsy Behav 2011;31:117–22. [38] Jurado MB, Rosselli M. The elusive nature of executive functions: a review of our current understanding. Neuropsychol Rev 2007;17:213–33. [39] Hermann BP, Jones JE, Sheth R, Koehn M, Becker T, Fine J, et al. Growing up with epilepsy: a two-year investigation of cognitive development in children with new onset epilepsy. Epilepsia 2008;49(11):1847–58. [40] Sillanpää M, Jalava M, Kaleva O, Shinnar S. Long-term prognosis of seizures with onset in childhood. N Engl J Med 1998;338:1715–22.