Epilepsy & Behavior 39 (2014) 97–104

Contents lists available at ScienceDirect

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

Developmental stage affects cognition in children with recently-diagnosed symptomatic focal epilepsy Linda M. Gonzalez a,b,c,⁎, Upeka S. Embuldeniya a,c, A. Simon Harvey d, Jacquie A. Wrennall a,b, Renee Testa c,e, Vicki A. Anderson a,b,f, Amanda G. Wood a,g a

Australian Centre for Child Neuropsychology Studies, Murdoch Childrens Research Institute, Melbourne, Australia RCH Mental Health, The Royal Children's Hospital, Melbourne, Australia School of Psychology and Psychiatry, Monash University, Victoria, Australia d Department of Neurology, The Royal Children's Hospital, Melbourne, Australia e Psychology Department, Sunshine Hospital, Melbourne, Australia f School of Psychological Sciences, University of Melbourne, Melbourne, Australia g School of Psychology, University of Birmingham, UK b c

a r t i c l e

i n f o

Article history: Received 1 May 2014 Revised 3 August 2014 Accepted 6 August 2014 Available online 18 September 2014 Keywords: Age at onset Pediatric epilepsy Development Critical periods Cognition Symptomatic focal epilepsy

a b s t r a c t This study explored the impact of developmental stage on cognitive function in children with recently-diagnosed epilepsy. In keeping with a neurodevelopmental framework, skills in a critical developmental period were expected to be more vulnerable than those stable at the time of seizure onset. We studied children with earlyonset (EO) symptomatic focal epilepsy (onset: 3–5 years; n = 18) and compared their performance with that of the group with late-onset (LO) epilepsy (onset: 6–8 years performance of; n = 8) on a range of cognitive tasks. Performance of both groups was compared with normative standards. ‘Critical’ and ‘stable’ classifications were based on developmental research. Nonparametric analyses revealed that skills in a critical developmental period for the group with EO epilepsy fell below normative standards (Phonological Processing: p = .007, Design Copying: p = .01, Visuomotor Precision:, p = .02) and fell below the performance of the group with LO epilepsy (Design Copying: p = .03, Visuomotor Precision: p = .03). There were no differences between the group with EO epilepsy and the group with LO epilepsy on measures of receptive vocabulary and memory, which were proposed to be in a stable developmental period across both groups. Auditory span, as measured by Word Order, was reduced for both the group with EO epilepsy (p = .02) and the group with LO epilepsy (p = .02) relative to normative standards, but the groups did not differ from each other. These results are consistent with a prolonged period of critical development for this skill. These findings support the notion that skills in a critical phase of development are particularly vulnerable following the onset of symptomatic focal epilepsy in childhood. Crown Copyright © 2014 Published by Elsevier Inc. All rights reserved.

1. Introduction Seizure activity during critical periods of brain development has the potential to cause significant cognitive problems for the child, which may have lifelong consequences. Infants are particularly prone to seizures, with events having the propensity to be frequent and/or prolonged [1]. Furthermore, seizures produce distinct maladaptive anatomical and physiological changes in the developing brain [2,3]. The resultant disruption to neural networks has been correlated with adverse behavioral outcomes [3]. It follows that there may also be a lasting impact of early-onset seizures on cognitive development. Adult and child studies across a range of epilepsy syndromes consistently suggest that intellectual and cognitive impairment is greater ⁎ Corresponding author at: RCH Mental Health, The Royal Children's Hospital, Flemington Road, Parkville, Victoria 3052, Australia. E-mail address: [email protected] (L.M. Gonzalez).

http://dx.doi.org/10.1016/j.yebeh.2014.08.006 1525-5050/Crown Copyright © 2014 Published by Elsevier Inc. All rights reserved.

when seizures begin early rather than later in life [4–12]. In support of this view, intellectual outcome is particularly poor if seizures begin prior to the age of five [5,13–17], with the poorest outcomes associated with seizure onset in the first year of life [11,18]. The effect of age at onset persists once factors such as seizure control [14,19], number of antiepileptic drugs (AEDs) [19], etiology [20], duration of epilepsy [14,19, 21], and extent of pathology are controlled [22], implying that developmental processes have an independent effect on cognitive outcome. Some recent studies have explored the relationship between age at seizure onset and specific cognitive functions across childhood. These studies have yielded mixed results, with age at seizure onset found to predict motor function [23], attention [6], language [8], nonverbal reasoning [16], memory [24], and executive functions [25,26]. However, several studies have reported the absence of such effects across these cognitive domains [27–31]. These inconsistencies not only may reflect differences in methodology and sample characteristics but also may be due to limitations of utilizing age at seizure onset as a marker of

98

L.M. Gonzalez et al. / Epilepsy & Behavior 39 (2014) 97–104

development, a variable enmeshed with illness duration, medication, and cumulative burden of seizures [32]. One approach, to separate developmental effects from illness variables, has been to study cognition in children with new-onset epilepsy. Studies utilizing such methodologies have identified difficulties across a range of cognitive skills [30,33], with attention and information processing being most consistently impaired [34,35]. These studies highlight the cognitive burden experienced by children with new-onset epilepsy; however, they do not account for developmental factors. That is, previous research has largely overlooked the impact of the timing of seizure-onset occurrence and the possibility of different outcomes for specific skills at different developmental stages. A handful of studies have considered developmental stage, but this approach is yet to be applied to a cohort with new-onset epilepsy. Upton and Thompson [36] stratified their sample of adults with frontal lobe epilepsy (FLE) according to three groups with different ages at seizure onset that correspond to stages of executive development. The two measures of executive function employed yielded inconsistent results but did not suggest a lasting effect of age at seizure onset. Of note, their design may have overlooked a delay in development. Consistent with that possibility, Hernandez et al. [37] studied a sample of children with FLE and found that these children had specific deficits in aspects of executive function that were more apparent in younger as opposed to older children, suggesting a delay in these abilities rather than a discrete impairment. Developmental effects may be particularly apparent in executive functions because of the prolonged developmental trajectory of the neural architecture supporting this aspect of cognition. This concept can be applied to other cognitive domains that emerge earlier in childhood. Dennis [38] provided a useful heuristic to examine developmental delays and deficits by recognizing potentially different outcomes for skills in an emerging, developing, or established phase at the time of insult. Empirical studies testing this framework suggest that skills in an emerging phase of development are associated with poorer outcome than more established skills in generalized brain insults [39,40]. While Dennis' original model does not easily accommodate nonlinear patterns of development which have been, she and colleagues [41] have recently suggested an update to accommodate periods of relative stability and critical bursts of change resulting in a wider range of potential developmental outcomes following early childhood brain insult. The normal developmental trajectory of specific cognitive abilities is an important consideration in predicting cognitive impairment in clinical populations. Given that the peak incidence for the onset of focal epilepsy occurs in early childhood [42,43], skills undergoing rapid development at this time may provide an important insight into cognitive outcome for this group. Such skills include, but are not limited to, receptive vocabulary, associative memory, visuospatial function, phonological processing, deductive reasoning, and auditory span. Specifically, there is strong evidence to suggest that the first year of life represents a key period for language function, particularly receptive vocabulary [44]. Similarly, the hippocampus, which supports associative memory, undergoes a critical period of development in infancy, with maturation continuing into middle childhood [45,46]. Despite ongoing refinement, the neural foundations for receptive language and associative memory are laid down in infancy, indicating an early critical period for these skills. Although the foundations of visuospatial and phonological skills are also laid early in life, there is strong evidence to suggest that these skills undergo significant structural and functional development between the ages of 3 and 5 years, indicating a critical period for these abilities [47–49]. In terms of deductive reasoning, although rudimentary skills are apparent in preschoolers [50], the critical period for these abilities does not occur until middle–late childhood when children are better able to process complexity [51]. Auditory span has a more gradual and prolonged developmental course from the age of three through to nine years [52]. Distinct developmental spurts have not been described, with maturation characterized by a more linear progression, suggesting that the critical period for auditory span is more prolonged [52–54].

The present study aimed to examine the impact of seizure onset during early childhood and to determine whether there are differential effects for specific cognitive skills depending on developmental stage. Consistent with Dennis et al.'s [41] model, cognitive abilities were classified as ‘stable’ or ‘critical’, and performance was compared for two age groups. Three- to five-year-old children with new-onset epilepsy (group with early-onset (EO) epilepsy) were compared with a group of six- to eight-year-old children with new-onset epilepsy (group with late-onset (LO) epilepsy). It was expected that skills in a ‘critical’ but not ‘stable’ developmental phase would fall below normative standards irrespective of seizure onset. Differences between the group with EO epilepsy and the group with LO epilepsy were expected for skills, where one group was in a ‘critical’ developmental phase and the other was in a ‘stable’ developmental phase. The group in the ‘critical’ phase was expected to perform more poorly than that in the ‘stable’ phase. No between-group differences were expected for skills classified as ‘critical’ or ‘stable’ in both the group with EO epilepsy and the group with LO epilepsy. 2. Method 2.1. Participants Twenty-eight children with symptomatic focal epilepsy participated in the study. Symptomatic focal epilepsies were defined as recurrent, unprovoked seizures with a known or presumed focal basis (including those with or without impairment of consciousness or awareness, as well as those evolving into bilateral convulsive seizures). Diagnosis, including localization, was made by the child's treating neurologist, supplemented by EEG results, MRI findings, and other clinical investigations as necessary. Any child with a current developmental or psychiatric diagnosis or a known or suspected diagnosis of intellectual disability (ID) was excluded from the study. Parent ratings of adaptive function were used to exclude children with ID, rather than IQ, given that IQ is enmeshed with the specific cognitive domains central to this study. Children with composite adaptive behavior ratings b70 on the Vineland Adaptive Behavior Scales — Second Edition (VABS-II; [55]) were excluded. Two children were excluded on this basis, resulting in a sample size of 26. Several other studies have utilized adaptive behavior to infer IQ scores in individuals with ID [56–58]. All children were assessed as close as possible to time of diagnosis but up to a maximum of two years since diagnosis. Participants were divided into two groups based on age at seizure onset: (EO: 3–5 years; n = 18) and (LO: 6–8 years; n = 8). Age at onset and age at assessment (i.e., chronological age) both fell in the same age bracket (3 years, 0 months–5 years, 11 months or 6 years, 0 months–8 years, 11 months). This was to preclude the confounding effect of children having an early age at onset but late age at assessment. Although the division between the group with EO epilepsy and the group with LO epilepsy reflects a critical point in development for many cognitive domains, change is gradual rather than associated with a fixed age or time point. Thus, the cutoff between the group with EO epilepsy and the group with LO epilepsy is somewhat arbitrary, and the distribution of the sample within the group with EO epilepsy and that with LO epilepsy becomes important, particularly for five- and six-year-old patients. The age distribution of the sample is depicted in Fig. 1. Of the seven five-year-old participants, five were between 5 years, 1 month and 5 years, 6 months of age. The six-year-old child was 6 years, 10 months of age. Thus, there was minimal clustering around the cutoff. Current seizure frequency was classified as follows: daily, weekly, monthly–quarterly, or yearly or longer (yearly plus). Classifications were made on the basis of parent report as well as information obtained from the child's medical file. Seizure focus included the following: frontal lobe (EO: n = 7, LO: n = 4); temporal lobe (EO: n = 8, LO: n = 1); parietal lobe (EO: n = 1, LO: n = 1); occipital lobe (EO: n = 1, LO: n = 1); and hypothalamus (EO: n = 1, LO: n = 1). Chi-square revealed

L.M. Gonzalez et al. / Epilepsy & Behavior 39 (2014) 97–104

Fig. 1. Age distribution for the group with early-onset epilepsy and for that with late-onset epilepsy.

that there were no differences in seizure type as a function of age-atonset group. In order to reduce the number of cells in this analysis, parietal, occipital, and hypothalamic foci were grouped together in an ‘other’ category. Participants were also divided into a group with lesional epilepsy and a group with nonlesional epilepsy on the basis of MRI findings. Chi-square did not reveal any differences between the group with EO epilepsy and that with LO epilepsy for this variable or for seizure frequency, bilateral spread, and gender, indicating that the distribution of these variables did not differ significantly between both groups. Mann–Whitney U tests indicate that groups were comparable in regard to socioeconomic status (SES), VABS-II composite score, epilepsy duration, and number of AEDs. As expected, age at seizure onset (z = .000, p b .001) and age at assessment (z = .000, p b .001) differed significantly between the group with EO epilepsy and that with LO epilepsy. Descriptive and seizure characteristics are given in Table 1. 2.2. Materials Socioeconomic status was determined using the first revision of the Australian and New Zealand Standard Classification of Occupations [59] and the Australian Socioeconomic Index 2006, which allows computation of occupational status scores [60] ranging from 0 (low) to

99

100 (high). To measure specific cognitive domains, we administered participants a comprehensive neuropsychological assessment incorporating standardized measures. For each measure, normative data were available for the full age range of this study, allowing the same measures to be administered to all participants. Subtests were chosen from the Developmental Neuropsychological Assessment — Second Edition [61] and the Kaufman Assessment Battery for Children — Second Edition [62]. The Peabody Picture Vocabulary Test — Fourth Edition [63] was administered in its entirety. These measures are well validated and widely used psychometric tools and were selected primarily because they also cover the age range of the current study through middle childhood, which allows for the sample to be followed over time. The selected subtests were assigned to a specific cognitive domain based on the psychometric and theoretical underpinnings of each subtest. Each measure was then classified into a ‘stable’ or a ‘critical’ developmental phase for the group with EO epilepsy and that with LO epilepsy in keeping with Dennis et al. [41]. These classifications primarily reflected qualitative changes in development (i.e., the acquisition of new skills) and do not imply that the groups are necessarily equivalent in terms of absolute ability. These determinations were made on the basis of review of the developmental literature [e.g., 64,65]. The measures, associated cognitive processes, and prescribed developmental stage are illustrated in Fig. 2. 2.3. Procedure Participants were recruited both prospectively and retrospectively through The Royal Children's Hospital (RCH), Melbourne, Australia between June 2008 and October 2009. Ethics approval was obtained from the Human Research Ethics Committee, as well as through Monash University. Research was conducted in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. Participants' parents provided informed written consent prior to participation, and children over the age of 5 years gave verbal assent. The assessment protocol was administered in a set order. 2.4. Statistical analyses All scores were converted to standardized age-scaled scores. All measures had a normative mean of 10 (SD = 3), except for the PPVT-4

Table 1 Demographic and seizure characteristics of the group with EO epilepsy and the group with LO epilepsy.

Demographic characteristics Adaptive function Seizure characteristics

Demographic characteristics Current seizure frequency

Seizure localization

Lesion status Seizure type ⁎⁎ p b .001.

Age at assessment ⁎⁎ SES VABS composite Age at seizure onset ⁎⁎ Duration of epilepsy Number of AEDs

Males Daily Weekly Monthly–quarterly Yearly plus Frontal Temporal Other Lesional Nonlesional Unilateral Bilateral spread

Group with EO epilepsy n = 18

Group with LO epilepsy n=8

Mann–Whitney U test

p

n (SD)

n (SD)

z

4.49 (0.88) 55.51 (27.04) 90.56 (12.86) 3.46 (0.55) 1.13 (0.76) 1.50 (0.79)

7.95 (0.78) 62.13 (19.09) 96.63 (17.74) 6.80 (0.67) 1.18 (0.72) 1.13 (0.64)

−4.01 −0.67 −1.00 0.00 −1.50 −1.11

b.001 .53 .39 b.001 .13 .34

n (%)

n (%)

Chi-square

p

9 2 (11.0) 4 (22.3) 9 (50.0) 3 (16.7) 7 (38.9) 8 (44.4) 3 (16.7) 11 (61.0) 7 (39.0) 8 (44.0) 10 (56.0)

5 1 (12.5) 1 (12.5) 5 (62.5) 1 (12.5) 4 (50.0) 1 (12.5) 3 (37.5) 3 (38.0) 5 (62.0) 2 (25.0) 6 (75.0)

.35 .51

.56 .92

2.84

.24

1.24

.27

.89

.38

100

L.M. Gonzalez et al. / Epilepsy & Behavior 39 (2014) 97–104

Fig. 2. Critical classification versus stable classification for cognitive measures.

which had a mean of 100 (SD = 15). Although data were normally distributed, nonparametric techniques were employed, given the small sample size, particularly of the group with LO epilepsy. Differences between the whole sample and normative means were examined utilizing the Wilcoxon signed-rank test, where the performance of each case was compared with a dummy variable set to the population mean. This process was repeated for the group with EO epilepsy and that with LO epilepsy. All analyses were two-tailed. Effect size for all comparisons was measured by the r statistic which, as suggested by Field [66], is an appropriate measure of magnitude for nonparametric data, where r = z/√N. Field [66] defines a small effect as r = 0.1, a medium effect as r = 0.3, and a large effect as r = 0.5. The Mann–Whitney U test was utilized to examine differences between the group with EO epilepsy and that with LO epilepsy. Each measure was compared individually, and then as a nonparametric alternative to MANOVA, a composite variable was created for PPVT-4 and Atlantis, which were both classified as ‘stable’ for the group with EO epilepsy and that with LO epilepsy. These measures were converted to z-scores prior to this analysis to ensure that both were on the same scale. Similarly, a composite variable was created for all measures classified as ‘critical’ for the group with EO epilepsy and ‘stable’ for the group with LO epilepsy, those being Visuomotor Precision, Design Copying, and Phonological Processing. The Mann–Whitney U test was then performed to explore differences between the group with EO epilepsy and that with LO epilepsy on these composite variables.

and for Visuomotor Precision, z = − 1.85, p = .06, r = .36 although there was a medium effect for this latter measure. This analysis was repeated for the group with EO epilepsy and that with LO epilepsy and yielded a different pattern of findings, suggesting that it is meaningful to divide the whole sample into such groups. Performance of the group with EO epilepsy fell below normative standards for Phonological Processing: z = −2.67, p = .007, r = .62; Visuomotor Precision: z = −3.05, p = .002, r = .72; Design Copying: z = −2.57, p = .01, r = .61; and Word Order: z = − 2.31, p = .02, r = .54 but not PPVT-4: z = − 1.49, p = .14, r = .35; Atlantis: z = − 1.50, p = .13, r = .35; or Riddles: z = − 1.69, p = .09, r = .39. Medium effect sizes were observed for these nonsignificant measures indicating a trend for the group with EO epilepsy to perform below normative standards. Performance of the group with LO epilepsy fell significantly below normative means for Riddles: z = −2.26, p = .02, r = .80 and Word Order: z = −2.32, p = .02, r = .82 but not PPVT-4: z = 0.00, p = 1.00, r = 0.00; Atlantis: z = − 1.38, p = .17; Phonological Processing: z = −0.41, p = .68, r = .14, r = .49; Visuomotor Precision: z = − 0.85, p = .40, r = .30; or Design Copying: z = −0.95, p = .0.34, r = .33. A large effect was observed for Atlantis and medium effects for the visuospatial measures. This series of analyses is depicted in Fig. 3. All results have been converted to z-scores to allow for direct comparison. 3.2. Group differences as a function of developmental stage

3. Results 3.1. Differences between performances of groups with epilepsy and normative standards The Wilcoxon signed-rank test comparing the performance of the whole sample (n = 26) with normative standards revealed that there were significant differences for Atlantis: z = − 2.01, p = .04, r = .39; Phonological Processing: z = − 2.62, p = .009, r = .51; Riddles: z = − 2.57, p = .01, r = .50; and Word Order: z = − 3.16, p = .002, r = .62. Results were more marginal for Design Copy: z = −1.97, p = .05, r = .38 and nonsignificant for PPVT-4: z = − 1.25, p = .21, r = .24

Mann–Whitney U tests examined the difference between the group with EO epilepsy and the group with LO epilepsy for each of the dependent variables (see Table 2). There were no significant between-groups differences for PPVT-4: z = − 0.67, p = .50, r = . 13 and Atlantis: z = − 0.42 p = .67, r = .08, which were both classified as ‘stable’ across groups. Consistent with these negligible effects, the composite variable combining these ‘stable’ measures did not differ significantly between groups, z = − .03, p = .98, r = .01. The group with EO epilepsy performed significantly more poorly than the group with LO epilepsy for Visuomotor Precision: z = −2.24, p = .03, r = .44 and Design Copying: z = − 2.19, p = .03, r = .43,

L.M. Gonzalez et al. / Epilepsy & Behavior 39 (2014) 97–104

Fig. 3. Performance of the whole sample the group with EO epilepsy, and the group with LO epilepsy relative to normative standards.

both of which were classified as ‘critical’ for the younger sample and ‘stable’ for the older sample. There were no significant differences between both groups for Phonological Processing: z = − 1.48, p = .14, r = .30, although there was a medium effect size. The composite measure for tasks with a critical period between 3 and 5 years was significantly different between groups, z = −2.59, p = .01, r = .51. There was no significant group difference on Riddles: z = − 0.79, p = .43, r = .15, which was the only measure classified as ‘critical’ in the group with LO epilepsy but ‘stable’ in the group with EO epilepsy. Both groups performed at comparable levels on Word Order: z = −.93, p = .93, r = .18, which was the only measure classified as ‘critical’ for both groups. As this latter result may also reflect a nonspecific reduction in information processing ability, the Spearman rho correlation between Word Order and duration of epilepsy, number of AEDs, seizure frequency, seizure localization, lesion status, and seizure type was examined. There were no significant associations. 4. Discussion This study revealed that developmental stage influences cognitive function in children with recently-diagnosed epilepsy. The distinction between critical and stable developmental periods was informative, with most analyses upholding the expected pattern of results, vulnerability for those in a critical phase of development and resilience for Table 2 Descriptive and nonparametric statistics for cognitive measures in the group with EO epilepsy and the group with LO epilepsy. Group with EO epilepsy mean (SD)

Group with LO epilepsy mean (SD)

PPVT 91.11 (18.81) 99.63 (20.15) Atlantis 8.33 (4.07) 7.88 (3.80) 7.50 (2.97) 11.25 (3.96) Visuomotor Precision combined+ 6.39 (4.60) 10.63 (4.24) Design copying total+ Phonological processing 7.39 (3.52) 9.50 (2.27) Riddles 8.39 (3.57) 7.63 (2.62) Word Order 7.83 (3.50) 7.50 (2.56) +

p b .05.

z

p

r

−0.67 −0.42 −2.24 −2.19 −1.48 −0.79 −0.93

.50 .67 .03 .03 .14 .43 .93

.13 .08 .44 .43 .30 .15 .18

101

stable skills. These results were demonstrated both in relation to normative standards and for comparisons between the group with EO epilepsy and the group with LO epilepsy and despite a small sample size and use of nonparametric statistical techniques. As groups with different ages at seizure onset were comparable in regard to SES, adaptive function, and seizure variables, results suggest that developmental stage is an independent determinant of cognitive function in children with recently-diagnosed epilepsy. This study provides evidence that skills in a critical phase of development are more vulnerable than those in a stable phase of development at the time of seizure onset. Analyses comparing the performances of the group with EO epilepsy and the group with LO epilepsy with normative standards were entirely consistent with predictions and showed that each group performed below expectations if skills were in a critical but not stable pattern of development, whereas some of these findings were not as clear utilizing the whole sample. Comparisons between the group with EO epilepsy and the group with LO epilepsy also indicated poorer outcome for the group with EO epilepsy for skills in a critical stage of development. These findings are consistent with the predictions of Dennis et al. [41] and fit the pattern described for generalized acquired brain injuries in childhood [39]. It is notable that such developmental vulnerabilities were demonstrated in a sample of children with symptomatic partial epilepsy, which could be argued to be a focal brain disorder. However, results suggest that the cognitive profile associated with symptomatic focal epilepsy may mirror that seen in generalized neurodevelopmental insults, owing to the effect of electrical and functional disturbance on developing cognitive networks [2,67]. The finding of significant cognitive deficits in a recently-diagnosed cohort is consistent with a number of other studies [19,68] and further underscores that cognitive deficits are primary features of focal epilepsy syndromes and not just the consequence of chronic seizures. This study suggests that the presence of a critical developmental period at the time of seizure onset is associated with cognitive impairment early in the course of childhood focal epilepsy. The distinction between critical and stable developmental stages appeared to better support results than did the more simple between group comparisons, which lends further weight to a developmental interpretation of these data and argues against a spurious effect related to the cutoff between the group with EO epilepsy and the group with LO epilepsy. Specifically, differences were not observed between the age groups on measures proposed to be in the same developmental phase, be that stable (memory and receptive language) or critical (auditory span). Again, skills in a stable phase were characterized by resilience, and those in a critical period were more vulnerable. The ageappropriate performances for each of the age groups on measures of memory and receptive language are consistent with the notion that the fundamental neural circuitry of these abilities is in place early in life [69,70], and these foundations afford some protection against the onset of seizures during childhood. Conversely, the more generalized deficits seen for auditory span are consistent with a more protracted developmental trajectory, possibly reflecting that a number of other abilities must also be acquired to advance this skill. A developmental interpretation of these data is further supported by the lack of association between auditory span and seizure variables. For measures where the groups were in different developmental phases, results were consistent with predictions when the group with early-onset epilepsy was in a critical phase, but not so when the developmental spurt was expected to occur in the group with late-onset epilepsy. In terms of vulnerability for the group with early-onset epilepsy, findings indicate that, when epilepsy begins between the ages of 3 and 5 years, visuospatial skills are particularly impaired. The selective vulnerability of visuospatial skills in early childhood has also been shown in other populations including children who have sustained strokes [71] or have undergone cranial radiation therapy for leukemia [72,73]. These findings are consistent with studies suggesting a qualitative shift in visuospatial processing and its neural substrates during late

102

L.M. Gonzalez et al. / Epilepsy & Behavior 39 (2014) 97–104

preschool resulting in an increased capacity for abstract spatial representation [74]. Functional gains likely correspond to changes in neural circuitry, with visuospatial tasks shifting from recruitment of frontal and subcortical areas to bilateral parietal areas between the ages of 3 and 5 years [48]. Findings were slightly attenuated for phonological processing, which was also hypothesized to be in a critical phase for the group with EO epilepsy and in a stable phase for older children. While the expected pattern was observed for comparisons against normative standards, the difference between the group with EO epilepsy and the group with LO epilepsy, while not statistically significant, was associated with a medium effect size. These more marginal results may be related to statistical power and increased influence of specific cases. This may be particularly relevant for phonological processing with some evidence to suggest that the critical period may not occur until the age of six for some individuals [75] and may be linked to commencement of formal reading instruction [76]. Composite variables were utilized to increase reliability, with findings highlighting the vulnerability of skills in a critical phase of development. Although there is a prevailing view that disruption to development during a critical period has the potential to delay or derail skill acquisition, the mechanisms underpinning such vulnerability are less well understood [77]. Anderson et al. [40] raised the possibility that the complexity of the skill and underlying neural network may influence outcome following early brain injury, with more circumscribed abilities, such as visuospatial skills, associated with better outcome compared with widely distributed abilities such as attention and executive functions. The range of potential outcomes proposed by Dennis et al. [41] is consistent with the view that developmental trajectories may differ for specific skills. At a structural level, a recent DTI study in children with fetal alcohol syndrome correlated abnormal white matter trajectories with specific cognitive functions, including receptive vocabulary, in children between 5 and 15 years of age [78]. While comparable research has not been undertaken in children with epilepsy, disruption to white matter trajectories is one possible mechanism underlying the vulnerability of rapidly developing skills. Further research is clearly required to explore the range of outcomes and underlying neural processes. Elucidating developmental processes in childhood epilepsy is a complex endeavor. While results are largely consistent with the central contention that cognitive skills are vulnerable to seizure onset in critical developmental periods, there are some inconsistent findings particularly for deductive reasoning. As expected, there was a difference between normative standards for the group with late-onset epilepsy but not for the group with early-onset epilepsy, but the difference between the groups was not significant, and the effect size was small. It is possible that the critical period for this specific skill may occur later in childhood or be just beginning between the ages of 6 and 8 years, which would further diminish the power of this study to detect an effect. Results support the influence of developmental variables but should not be taken to imply that these are the only factors that impact cognition in this population. Rather, developmental variables should be considered alongside other factors known to influence cognition in this group including seizure frequency, duration of epilepsy, AEDs, and EEG findings. With the exception of auditory span, exploratory analysis between these variables and cognitive performance was not undertaken in the current study as groups did not differ on a range of demographic and seizure variables. However, this lack of difference may reflect limited power as visual inspection suggests relatively more temporal lobe cases for the group with early-onset epilepsy and a greater proportion of frontal lobe cases for the group with late-onset epilepsy. While this variable would have been explored in a larger sample, other studies found groups with TLE or FLE to be comparable on all but very specific cognitive measures [79,80]. There is no compelling literature to suggest that the number of cases with bilateral spread of seizures significantly impacts the cognitive domains under study, but these issues would benefit from specific attention in future research.

This study would have been enhanced by inclusion of a large control group, which would have allowed for critical and stable periods to be determined statistically rather than inferred from the literature. The sample size of the current study is small, particularly for the group with late-onset epilepsy, highlighting the need for further research to confirm this pattern of results. Notwithstanding this limitation, the distribution of our sample is consistent with epidemiological studies of focal epilepsy, which indicate that the peak incidence is in early childhood [42,43]. In the current study, efforts were made to target analyses to reduce the chance of Type 1 error, and effect sizes were explored to guard against Type 2 error. Furthermore, nonparametric methods were employed to ensure that all analyses were appropriate for the small sample size. It would also have been optimal to assess children closer to the time of onset. The delays were largely attributable to issues around recruitment and/or diagnosis and were not easily controlled. Finally, any developmental study is enhanced by longitudinal follow-up, which allows for exploration of the course of deficits and the possibility of different outcomes for different skills. It is intended that this cohort will be reviewed to determine whether the deficits highlighted in this study persist or resolve over time. In summary, this study demonstrates that developmental stage influences the nature of cognitive impairment in children with recentlydiagnosed symptomatic focal epilepsy. Specifically, skills in a critical phase of development were vulnerable, whereas those in a stable developmental period were more resilient. Results were independent of illness variables and clearly demonstrate that cognitive deficits are a core characteristic of symptomatic focal epilepsy, with individual deficits likely to differ depending on what skills are in a critical developmental phase at the time of onset. These findings highlight the need for early neuropsychological assessment to identify skills in a critical period of development at the time of diagnosis and to guide targeted intervention to support vulnerable skills. This study suggests that a central issue in understanding the cognitive effects of focal epilepsy is not so much about age at onset but rather which skills are in a critical period of development at seizure onset, a question best answered by early neuropsychological assessment. Acknowledgments The authors wish to thank all of the children and families who participated in this study. Dr. Embuldeniya received financial support from Monash University during the completion of her DPsych program. This research was supported by the Victorian Government Operational Infrastructure Scheme. Dr. Amanda Wood received support from Australian Rotary Health (Geoffrey Betts Postdoctoral Award). Conflict of interest statement None of the authors have any conflict of interest to disclose. We, the authors, confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. References [1] Sankar R, Rho JM. Do seizures affect the developing brain? Lessons from the laboratory. J Child Neurol 2007;22(5 Suppl.):21S–9S. [2] Holmes GL. EEG abnormalities as a biomarker for cognitive comorbidities in pharmacoresistant epilepsy. Epilepsia 2013;54(Suppl. 2):60–2. [3] Ben-Ari Y, Holmes GL. Effects of seizures on developmental processes in the immature brain. Lancet Neurol 2006;5(12):1055–63. [4] Strauss E, Loring D, Chelune G, Hunter M, Hermann B, Perrine K, et al. Predicting cognitive impairment in epilepsy: findings from the Bozeman epilepsy consortium. J Clin Exp Neuropsychol 1995;17(6):909–17. [5] O'Leary DS, Lovell MR, Sackellares JC, Berent S, Giordani B, Seidenberg M, et al. Effects of age of onset of partial and generalized seizures on neuropsychological performance in children. J Nerv Ment Dis 1983;171(10):624–9. [6] Montamedi G, Meador KJ. Epilepsy and cognition. Epilepsy Behav 2003;2003(4): S25–38.

L.M. Gonzalez et al. / Epilepsy & Behavior 39 (2014) 97–104 [7] Lespinet V, Bresson C, N'Kaoua B, Rougier A, Claverie B. Effect of age of onset of temporal lobe epilepsy on the severity and the nature of preoperative memory deficits. Neuropsychologia 2002;40(9):1591–600. [8] Jambaque I. Neuropsychology of temporal lobe epilepsy in children. In: Jambaque I, Lassonde M, editors. Neuropsychology of childhood epilepsy. New York: Plenum; 2001. [9] Hessen E, Lossius MI, Reinvang I, Gjerstad L. Predictors of neuropsychological impairment in seizure-free epilepsy patients. Epilepsia 2006;47(11):1870–8. [10] Hermann B, Seidenberg M, Bell B, Rutecki P, Sheth R, Ruggles K, et al. The neurodevelopmental impact of childhood-onset temporal lobe epilepsy on brain structure and function. Epilepsia 2002;43(9):1062–71. [11] Cormack F, Cross JH, Isaacs E, Harkness W, Wright I, Vargha-Khadem F, et al. The development of intellectual abilities in pediatric temporal lobe epilepsy. Epilepsia 2007;48(1):201–4. [12] Black LC, Schefft BK, Howe SR, Szaflarski JP, Yeh HS, Privitera MD. The effect of seizures on working memory and executive functioning performance. Epilepsy Behav 2010;17(3):412–9. [13] Saykin AJ, Gur RC, Sussman NM, O'Connor MJ, Gur RE. Memory deficits before and after temporal lobectomy: effect of laterality and age of onset. Brain Cogn 1989; 9(2):191–200. [14] Rantanen K, Eriksson K, Nieminen P. Cognitive impairment in preschool children with epilepsy. Epilepsia 2011;52(8):1499–505. [15] Pavone P, Bianchini R, Trifiletti RR, Incorpora G, Pavone A, Parano E. Neuropsychological assessment in children with absence epilepsy. Neurology 2001;56(8):1047–51. [16] Hoie B, Mykletun A, Sommerfelt K, Bjornaes H, Skeidsvoll H, Waaler PE. Seizurerelated factors and non-verbal intelligence in children with epilepsy. A population-based study from Western Norway. Seizure 2005;14(4):223–31. [17] Berg AT, Langfitt JT, Testa FM, Levy SR, DiMario F, Westerveld M, et al. Global cognitive function in children with epilepsy: a community-based study. Epilepsia 2008; 49(4):608–14. [18] Vanderlinden L, Lagae LG. Clinical predictors for outcome in infants with epilepsy. Pediatr Neurol 2004;31(1):52–5. [19] Kaaden S, Helmstaedter C. Age at onset of epilepsy as a determinant of intellectual impairment in temporal lobe epilepsy. Epilepsy Behav 2009;15(2):213–7. [20] Vasconcellos E, Wyllie E, Sullivan S, Stanford L, Bulacio J, Kotagal P, et al. Mental retardation in pediatric candidates for epilepsy surgery: the role of early seizure onset. Epilepsia 2001;42(2):268–74. [21] Dikmen S, Matthews CG. Effect of major motor seizure frequency upon cognitive–intellectual functions in adults. Epilepsia 1977;18(1):21–9. [22] Korman B, Krsek P, Duchowny M, Maton B, Pacheco-Jacome E, Rey G. Early seizure onset and dysplastic lesion extent independently disrupt cognitive networks. Neurology 2013;81(8):745–51. [23] van Mil SG, Reijs RP, van Hall MH, Aldenkamp AP. Neuropsychological profile of children with cryptogenic localization related epilepsy. Child Neuropsychol 2008;14(4): 291–302. [24] Mabbott DJ, Smith ML. Memory in children with temporal or extra-temporal excisions. Neuropsychologia 2003;41(8):995–1007. [25] Riva D, Avanzini G, Franceschetti S, Nichelli F, Saletti V, Vago C, et al. Unilateral frontal lobe epilepsy affects executive functions in children. Neurol Sci 2005;26(4): 263–70. [26] Hoie 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(6):519–25. [27] Sturniolo MG, Galletti F. Idiopathic epilepsy and school achievement. Arch Dis Child 1994;70(5):424–8. [28] Sonmez F, Atakli D, Sari H, Atay T, Arpaci B. Cognitive function in juvenile myoclonic epilepsy. Epilepsy Behav 2004;5(3):329–36. [29] Gonzalez LM, Mahdavi N, Anderson VA, Harvey AS. Changes in memory function in children and young adults with temporal lobe epilepsy: a follow-up study. Epilepsy Behav 2012;23(3):213–9. [30] Fastenau PS, Shen J, Dunn DW, Perkins SM, Hermann BP, Austin JK. Neuropsychological predictors of academic underachievement in pediatric epilepsy: moderating roles of demographic, seizure, and psychosocial variables. Epilepsia 2004;45(10):1261–72. [31] Bailet LL, Turk WR. The impact of childhood epilepsy on neurocognitive and behavioral performance: a prospective longitudinal study. Epilepsia 2000;41(4):426–31. [32] Hermann BP, Jones JE, Jackson DC, Seidenberg M. Starting at the beginning: the neuropsychological status of children with new-onset epilepsies. Epileptic Disord 2012; 14(1):12–21. [33] Hermann B, Jones J, Sheth R, Dow C, Koehn M, Seidenberg M. Children with new-onset epilepsy: neuropsychological status and brain structure. Brain 2006;129(Pt 10): 2609–19. [34] Oostrom KJ, Smeets-Schouten A, Kruitwagen CL, Peters AC, Jennekens-Schinkel A. Not only a matter of epilepsy: early problems of cognition and behavior in children with “epilepsy only”—a prospective, longitudinal, controlled study starting at diagnosis. Pediatrics 2003;112(6 Pt 1):1338–44. [35] Bhise VV, Burack GD, Mandelbaum DE. Baseline cognition, behavior, and motor skills in children with new-onset, idiopathic epilepsy. Dev Med Child Neurol 2010;52(1): 22–6. [36] Upton D, Thompson PJ. Age at onset and neuropsychological function in frontal lobe epilepsy. Epilepsia 1997;38(10):1103–13. [37] Hernandez MT, Sauerwein HC, Jambaque I, De Guise E, Lussier F, Lortie A, et al. Deficits in executive functions and motor coordination in children with frontal lobe epilepsy. Neuropsychologia 2002;40(4):384–400. [38] Dennis M. Language and the young damaged brain, in clinical neuropsychology and brain function: research, measurement and practice. In: Boll T, B.K.B., editors. Washington, D.C.: American Psychological Association; 1989. p. 89–123.

103

[39] Ewing-Cobbs L, Barnes M. Linguistic outcomes following traumatic brain injury in children. Semin Pediatr Neurol 2002;9(3):209–17. [40] Anderson V, Spencer-Smith M, Wood A. Do children really recover better? Neurobehavioural plasticity after early brain insult. Brain 2011;134(Pt 8):2197–221. [41] Dennis M, Spiegler BJ, Simic N, Sinopoli KJ, Wilkinson A, Yeates KO, et al. Functional plasticity in childhood brain disorders: when, what, how, and whom to assess. Neuropsychol Rev 2014. http://dx.doi.org/10.1007/s11065-014-9261-x (in press). [42] Zarrelli MM, Beghi E, Rocca WA, Hauser WA. Incidence of epileptic syndromes in Rochester, Minnesota: 1980–1984. Epilepsia 1999;40(12):1708–14. [43] Chan D, Phuah HK, Ng YL, Choong CT, Lim KW, Goh WH. Pediatric epilepsy and first afebrile seizure in Singapore: epidemiology and investigation yield at presentation. J Child Neurol 2010;25(10):1216–22. [44] Powell DR, Diamond KE. Promoting early literacy and language development. In: Pianta RC, et al, editors. Handbook of early childhood education. New York: The Guilford Pewaa; 2012. p. 194–216. [45] Uematsu A, Matsui M, Tanaka C, Takahashi T, Noguchi K, Suzuki M, et al. Developmental trajectories of amygdala and hippocampus from infancy to early adulthood in healthy individuals. PLoS One 2012;7(10):e46970. [46] Thompson DK, Omizzolo C, Adamson C, Lee KJ, Stargatt R, Egan GF, et al. Longitudinal growth and morphology of the hippocampus through childhood: impact of prematurity and implications for memory and learning. Hum Brain Mapp 2014;35(8): 4129–39. [47] Lonigan CJ, Burgess SR, Anthony JL. Development of emergent literacy and early reading skills in preschool children: evidence from a latent-variable longitudinal study. Dev Psychol 2000;36(5):596–613. [48] Eslinger PJ, Blair C, Wang J, Lipovsky B, Realmuto J, Baker D, et al. Developmental shifts in fMRI activations during visuospatial relational reasoning. Brain Cogn 2009;69(1):1–10. [49] Dodd B, Gillon G. Exploring the relationship between phonological awareness, speech impairment and literacy. Adv Speech Lang Pathol 2001;3(2):139–47. [50] Smidts DP, Jacobs R, Anderson V. The Object Classification Task for Children (OCTC): a measure of concept generation and mental flexibility in early childhood. Dev Neuropsychol 2004;26(1):385–401. [51] Hayes BK, Lim M. Development, awareness and inductive selectivity. J Exp Psychol Learn Mem Cogn 2013;39(3):821–31. [52] Gathercole SE. Cognitive approaches to the development of short-term memory. Trends Cogn Sci 1999;3(11):410–9. [53] Jones G. Why chunking should be considered as an explanation for developmental change before short-term memory capacity and processing speed. Front Psychol 2012;3:1–8. [54] Henry LA. The development of auditory memory span; the role of rehearsal. Br J Dev Psychol 1991;9:493–511. [55] Sparrow S, Balla D, Cichetti D. Vineland Adaptive Behvior Scales. 2nd ed. Circles Pines, Minnesota: American Guidance Service; 2006. [56] Schatz J, Hamdan-Allen G. Effects of age and IQ on adaptive behavior domains for children with autism. J Autism Dev Disord 1995;25(1):51–60. [57] Moss S, Hogg J. Estimating IQ from adaptive behaviour information in people with moderate-severe intellectual disability. J Appl Res Intellect Disabil 1997; 10(1):61–6. [58] Bolte S, Poustka F. The relation between general cognitive level and adaptive behavior domains in individuals with autism with and without co-morbid mental retardation. Child Psychiatry Hum Dev 2002;33(2):165–72. [59] Statistics. ABo, Zealand. SN, ANZSCO: Australian and New Zealand standard classification of occupations. Australian Bureau of Statistics; 2006. [60] McMillan J, Beavis A, Jones FL. The AUS106: a new socioeconomic index for Australia. J Sociol 2009;45(2):123–49. [61] Korkman M, Kirk U, Kemp S. NEPSY-II: clinical and interpretative manual. 2nd ed. San Antonio, TX: Psychological Corporation; 2007. [62] Kaufman AS, Kaufman NL. Kaufman assessment battery for children. 2nd ed. Circle Pines, MN: AGS Publishing; 2004. [63] Dunn LM, Dunn DM. Peabody Picture Vocabulary Test. 4th ed. Minneapolis, MN: Pearson; 2007. [64] Stiles J, Jernigan TL. The basics of brain development. Neuropsychol Rev 2010;20(4): 327–48. [65] Hollister-Sandberg E, Spritz BL. A clinician's guide to normal cognitive development in childhood. New York: Routledge; 2010. [66] Field A. Discovering statistics using SPSS. 2nd ed. London: Sage; 2005. [67] Helmstaedter C, Kockelmann E. Cognitive outcomes in patients with chronic temporal lobe epilepsy. Epilepsia 2006;47(Suppl. 2):96–8. [68] Neyens LG, Aldenkamp AP, Meinardi HM. Prospective follow-up of intellectual development in children with a recent onset of epilepsy. Epilepsy Res 1999;34(2–3): 85–90. [69] Utsunomiya H, Takano K, Okazaki M, Mitsudome A. Development of the temporal lobe in infants and children: analysis by MR-based volumetry. Am J Neuroradiol 1999;20(4):717–23. [70] Bates E, Thal D, Finlay B, Clancy B. Early language development and its neural correlates. In: Rapin I, Segalowitz S, editors. Handbook of neuropsychology. Amsterdam: Elservier; 1992. [71] Stiles J, Reilly J, Paul B, Moses P. Cognitive development following early brain injury: evidence for neural adaptation. Trends Cogn Sci 2005;9(3):136–43. [72] Moleski M. Neuropsychological, neuroanatomical, and neurophysiological consequences of CNS chemotherapy for acute lymphoblastic leukemia. Arch Clin Neuropsychol 2000;15(7):603–30. [73] Hockenberry M, Krull K, Moore K, Gregurich MA, Casey ME, Kaemingk K. Longitudinal evaluation of fine motor skills in children with leukemia. J Pediatr Hematol Oncol 2007;29(8):535–9.

104

L.M. Gonzalez et al. / Epilepsy & Behavior 39 (2014) 97–104

[74] Kulp MT. Relationship between visual motor integration skill and academic performance in kindergarten through third grade. Optom Vis Sci 1999;76(3):159–63. [75] Wesseling R, Reitsma P. Preschool phonological representations and development of reading skills. Ann Dyslexia 2001;51:203–20. [76] Korkman M, Barron-Linnankoski S, Lahti-Nuuttila P. Effects of age and duration of reading instruction on the development of phonological awareness, rapid naming, and verbal memory span. Dev Neuropsychol 1999;16:415–31. [77] Fox SE, Levitt P, Nelson III CA. How the timing and quality of early experiences influence the development of brain architecture. Child Dev 2010;81(1):28–40.

[78] Treit S, Lebel C, Baugh L, Rasmussen C, Andrew G, Beaulieu C. Longitudinal MRI reveals altered trajectory of brain development during childhood and adolescence in fetal alcohol spectrum disorders. J Neurosci 2013;33(24):10098–109. [79] Jocic-Jakubi B, Jovic NJ. Verbal memory impairment in children with focal epilepsy. Epilepsy Behav 2006;9(3):432–9. [80] Longo CA, Kerr EN, Smith ML. Executive functioning in children with intractable frontal lobe or temporal lobe epilepsy. Epilepsy Behav 2013;26(1):102–8.