Epilepsy & Behavior 29 (2013) 166–171

Contents lists available at ScienceDirect

Epilepsy & Behavior

journal homepage: www.elsevier.com/locate/yebeh

Determinants of intelligence in childhood-onset epilepsy: A

single-center study
Jungmee Park a,1, Mi-Sun Yum a,1, Hae-won Choi a, Eun Hee Kim a, Hyo Won Kim b, Tae-Sung Ko a,⁎
a
Department of Pediatrics, Asan Medical Center Children's Hospital, University of Ulsan College of Medicine, Seoul, Republic of Korea
b
Department of Psychiatry, Asan Medical Center Children's Hospital, University of Ulsan College of Medicine, Seoul, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: The purpose of this study was to quantify the intelligence of children with epilepsy and to determine the clinical
Received 29 May 2013 factors associated with intellectual impairment.
Revised 8 July 2013 The medical records of patients diagnosed with childhood-onset epilepsy at a single tertiary medical center in
Accepted 12 July 2013
Korea between 2006 and 2011 were retrospectively reviewed. The Korean Education Development Institute—
Available online 22 August 2013
Wechsler Intelligence Scale for Children or Korean Wechsler Intelligence Scale for adults was used to quantify
Keywords:
the level of intelligence. Age at seizure onset, etiology, epilepsy duration, number of seizures in the last year,
Cognition use of antiepileptic drugs, EEG/MRI findings, and epilepsy classification were recorded. The association between
Intelligence clinical factors and the intelligence was determined using logistic regression.
Epilepsy Three hundred and twenty-two patients were included in the analysis. One hundred and seventy-six (54.7%)
Children patients had low intelligence (intelligence quotient [IQ] b 80) with 18 (5.6%) defined as borderline mental
Mental retardation retardation (IQ 70–79), 47 (14.6%) as mild mental retardation (IQ 60–69), and 111 (34.5%) as moderate-to-
Seizure severe mental retardation (IQ b 60). Epilepsy duration, number of seizures in the last year, and epilepsy
classification were significantly associated with low intelligence in multivariate logistic regression (p b 0.05).
However, when analyzed according to etiology, these factors were not associated with low intelligence in
children with idiopathic epilepsy.
The most important factors associated with low intelligence in childhood-onset epilepsy are the underlying eti-
ology and, in cryptogenic and symptomatic epilepsy, seizure burden. The results of this study underscore the im-
portance of seizure control to alleviate the harmful impact of epilepsy on cognition.
© 2013 Elsevier Inc. All rights reserved.

1. Introduction
Cognition is the mental process of knowing from the environment
and includes aspects of awareness, perception, memory, reasoning,
and judgment. Cognition and intelligence are often considered synony-
mous. However, intelligence is only one aspect of cognition and is eval-
uated by a standardized intelligence test [1]. Epilepsy is a disorder of the
brain that is caused by abnormal electrical activity in neurons and is
closely associated with cognitive comorbidities, which are major con-
cerns in children with epilepsy [2,3].
The prevalence of mental retardation is more frequent in children
with epilepsy than in the general population [4,5], and a large number
Abbreviations: AED, Antiepileptic drug; IGE, Idiopathic generalized epilepsy; ILRE,
Idiopathic localization-related epilepsy; LGS, Lennox–Gastaut syndrome; EE, Epileptic
encephalopathy.
⁎ Corresponding author at: Department of Pediatrics, Asan Medical Center Children's
Hospital, University of Ulsan College of Medicine, 86 Asanbyeongwon-gil, Songpa-gu,
Seoul 138-736, Republic of Korea. Fax: +82 2 473 3725.
E-mail address: tsko@amc.seoul.kr (T.-S. Ko).
1
These two authors contributed equally in this work.
of studies have shown that poor achievement and learning disability
are associated with childhood epilepsy [6–10]. Childhood-onset epilep-
sy is associated with greater risk of cognitive impairment compared
with adult-onset epilepsy [11]. Even in patients with benign childhood
epilepsy with centrotemporal spikes (BCECTS), which has been associ-
ated with normal cognitive function and benign progression, many
studies have found a negative impact of epilepsy on cognition including
intelligence and school difficulties [12–14].
Epilepsy is not a single disease entity but a complex group of disor-
ders embracing a large number of diseases that cause neuronal excit-
ability. Multiple interrelated factors influence cognitive function in
childhood epilepsy. Brain pathology or etiological factors have a critical
impact on cognition. Patients with symptomatic epilepsy are more like-
ly to have cognitive impairment than patients without detectable brain
lesions [15], and patients with temporal lobe epilepsy often suffer mem-
ory deficits [16]. The frequency, duration, and severity of seizures can
also impact cognitive function, and a younger age at onset of seizures
is a risk factor for intellectual impairment [17,18]. There are also
treatment-related factors that can impact cognitive function, including
the use of antiepileptic drugs (AEDs) and sequelae of epilepsy surgery
[19].

1525-5050/$ – see front matter © 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.yebeh.2013.07.010
 J. Park et al. / Epilepsy & Behavior 29 (2013) 166–171 167

Children undergo a process of brain maturation and intellectual 3. Results

growth, and, as such, the intellectual impact of epilepsy is one of
the most critical aspects of treatment. The aim of this study was to 3.1. Baseline characteristics
evaluate intelligence in patients with childhood-onset epilepsy, to
determine the prevalence of low intelligence in patients with Clinical characteristics are summarized in Table 1. Age at the time
childhood-onset epilepsy, and to determine the factors that are associ- of cognitive evaluation ranged from 4 to 29 years (mean ± SD,
ated with intellectual impairment. 12.4 ± 4.8 years), and age at onset of seizures ranged from 0 to
16 years (mean ± SD, 5.8 ± 4.2 years). Epilepsy duration at the time
of cognitive evaluation ranged from 0 to 29 years (mean ± SD, 6.5 ±
2. Methods 5.1 years). Intelligence was normal in 146 (45.3%) patients and low in
176 (54.7%) patients. Three neurologically devastated patients could
2.1. Subjects not perform the Wechsler Intelligence Scale and were excluded for
the risk factor analysis.
Between January 2006 and March 2011, a total of 1834 patients with
childhood-onset epilepsy over 5 years old have been actively managed 3.2. Risk factors for low intelligence
in the pediatric neurology clinic at Asan Medical Center in Seoul, Korea.
Among them, the medical records of 322 patients with childhood-onset Mean IQ (Table 2) and low intelligence (IQ b 80; Table 3) were
epilepsy who were evaluated by an intelligence test were retrospective- strongly associated with multiple risk factors including age at onset of
ly reviewed. Demographic data and clinical history were obtained from seizures, epilepsy duration, AED number, number of seizures in the pre-
the medical records. ceding year, MRI abnormalities, epilepsy classification, and etiology. In
multivariate logistic regression model with backward elimination, epi-
lepsy duration, number of seizures in the preceding year, and epilepsy
2.2. Cognitive assessment classification were significant risk variables for low intelligence. Longer
epilepsy duration was associated with higher risk of low intelligence
The Korean Education Development Institute—Wechsler Intelli- compared with epilepsy duration of b 1 year, and ten or more seizures
gence Scale for Children was used to evaluate intelligence in patients in the preceding year were associated with higher risk of low intelli-
younger than 18 years of age, and the Korean Wechsler Intelligence gence than zero or one seizure in the preceding year. Epilepsy classifica-
Scale for adults was used to evaluate intelligence in patients aged tion was also a strong risk variable (Table 3). The etiology of epilepsy
18 years or above. A licensed psychologist or psychometrician ad-
ministered all evaluations. The level of intelligence was categorized
Table 1
as normal (intelligence quotient [IQ] ≥ 80), borderline mental retar-
Patient characteristics.
dation (IQ 70–79), mild mental retardation (IQ 60–69), moderate-
to-severe mental retardation (IQ b 60), or neurologically devastated Age at onset of seizuresa
b5 years 147 (47.7%)
(not testable). When repeated tests were undertaken, data from the
5–9 years 103 (33.4%)
first evaluation were used. ≥10 years 58 (18.8%)
Epilepsy durationa
b1 year 37 (12.0%)
2.3. Epilepsy variables 1–5 years 101 (32.8%)
6–9 years 100 (32.5%)
≥10 years 70 (22.7%)
The following information was obtained from the medical records: Number of seizures in preceding year
1) age at first unprovoked seizure (b5 years, 5–9 years, ≥10 years), 0–1 196 (60.9%)
2) epilepsy duration (b1 year, 1–5 years, 6–9 years, ≥ 10 years), 2–9 66 (20.5%)
≥10 60 (18.6%)
3) number of AEDs (0, 1, 2, ≥ 3), 4) number of seizures in the year
Number of antiepileptic drugs
preceding the cognitive evaluation (0–1, 2–9, ≥ 10), 5) EEG findings 0 52 (16.1%)
(focal, generalized, focal and generalized, normal), 6) MRI findings 1 151 (46.9%)
(abnormal, normal), 7) epilepsy classification (idiopathic localization- 2 48 (14.9%)
related epilepsy [ILRE], symptomatic focal epilepsy, cryptogenic focal ≥3 71 (22.0%)
Epilepsy classification
epilepsy, idiopathic generalized epilepsy [IGE], Lennox–Gastaut syn- ILRE 25 (7.8%)
drome/epileptic encephalopathy [LGS/EE]), and 8) etiology of epilepsy CFE 119 (37.0%)
(idiopathic, cryptogenic, symptomatic). Epilepsy duration was defined SFE 96 (29.8%)
as time between seizure onset and the day of cognitive evaluation. IGE 55 (17.1%)
LGS/EE 27 (8.4%)
Antiepileptic drug status, EEG findings, and MRI findings were deter-
MRI findingsb
mined at the time of cognitive evaluation. Normal 172 (56.2%)
Abnormal 134 (43.8%)
Etiology of epilepsy
2.4. Statistics Idiopathic 80 (24.8%)
Cryptogenic 135 (41.9%)
Symptomatic 107 (33.2%)
Kruskal–Wallis or Mann–Whitney tests were used to determine if Overall global cognitive function
the IQ differed across the different categories of clinical characteristics. Normal (IQ ≥ 80) 146 (45.3%)
Low intelligence was defined as an IQ score b 80, and logistic regression Borderline mental retardation (IQ 70–79) 18 (5.6%)
Mild mental retardation (IQ 60–69) 47 (14.6%)
analysis was used to identify the risk factors for low intelligence. Univar-
Moderate-to-severe mental retardation (IQ b 60) 111 (34.5%)
iate logistic regression analysis was used to identify the variables that
were significantly associated with low intelligence, and these were en- ILRE, idiopathic localization-related epilepsy; CFE, cryptogenic focal epilepsy; SFE,
symptomatic focal epilepsy; IGE, idiopathic generalized epilepsy; LGS/EE, Lennox–
tered into a multiple logistic regression. Backward elimination was used Gastaut syndrome/epileptic encephalopathy.
to identify significant risk factors for low intelligence. Statistical signifi- a
14 patients who did not know the age at seizure onset were excluded.
cance was set as p b 0.05. b
13 patients who did not have brain magnetic resonance images were excluded.
168 J. Park et al. / Epilepsy & Behavior 29 (2013) 166–171

Table 2 was a highly significant risk factor for low intelligence; therefore, we
Comparison of IQ according to clinical characteristics (N = 319)a. reanalyzed the risk factors for low intelligence for each etiologic
N Mean SD p value subgroup (Table 4). In the subgroup with idiopathic epilepsy (n = 80),
no variables were significantly associated with low intelligence. By con-
Age at onset of seizuresb b5 years 146 63.93 28.29 b0.001
5–9 years 102 79.35 32.26 trast, in the subgroup with cryptogenic epilepsy (n = 135), epilepsy
N10 years 57 87.84 27.72 duration and frequency of seizures were significant risk factors for low
Epilepsy durationb b1 year 37 98.89 21.70 b0.001 intelligence, and in the subgroup with symptomatic epilepsy (n =
1–5 years 101 85.34 30.09
107), epilepsy duration and AED number were significant risk factors
6–9 years 99 65.67 26.91
≥10 years 68 53.76 25.57 for low intelligence (Table 4). Intelligence quotient in subgroups with
Number of antiepileptic 0 51 79.41 28.80 b0.001 idiopathic, cryptogenic, and symptomatic epilepsy is shown in Fig. 1
drugs 1 151 83.44 29.96 according to epilepsy duration.
2 48 67.13 28.39
≥3 69 47.68 21.26
Number of seizures in 0–1 195 77.42 29.73 b0.001
4. Discussion
preceding year 2–9 65 79.11 31.21
≥10 59 49.56 25.16
MRI findingsc Normal 172 79.27 31.04 b0.001 Cognitive deficits are common in childhood epilepsy and have a
Abnormal 134 64.07 29.28 long-term detrimental impact on the quality of life. There have been a
EEG findings Focal 144 70.40 30.74 0.148 few population-based studies on the prevalence of intellectual impair-
Generalized 39 79.36 33.00
ments in childhood-onset epilepsy. Murphy et al. [20] reported that
Focal + generalized 42 66.36 34.04
Normal 94 75.99 29.33 30% of 10-year-old children with epilepsy had mental retardation, and
Epilepsy classification ILRE 25 105.28 13.83 b0.001 Camfield and Camfield [4] reported that approximately 20% of children
CFE 119 69.85 30.82 in a population-based epilepsy cohort had mental retardation (IQ b 70).
SFE 94 66.49 27.38 In addition, Berg et al. [15] reported that 26.4% of patients with child-
IGE 54 88.35 28.59
LGS/EE 27 44.33 22.65
hood epilepsy in a community-based cohort had subnormal cognitive
Etiology of epilepsy Idiopathic 79 93.71 26.02 b0.001 function (IQ b 80). The prevalence of low IQ in patients with childhood-
Cryptogenic 135 66.93 31.07 onset epilepsy might differ across populations because of different
Symptomatic 105 63.96 27.69 exposure to factors that influence intelligence such as education and re-
p values were obtained from Kruskal–Wallis tests for comparison of three or more groups habilitation systems. In this study, we reported the prevalence of and
and Mann–Whitney tests for comparison of two groups. risk factors for low intelligence in patients with childhood-onset epilep-
ILRE, idiopathic localization-related epilepsy; CFE, cryptogenic focal epilepsy; SFE,
sy from a single Korean tertiary center setting and found that over half
symptomatic focal epilepsy; IGE, idiopathic generalized epilepsy; LGS/EE, Lennox–
Gastaut syndrome/epileptic encephalopathy.
the cohort had low intelligence (IQ b 80). The higher prevalence as
a
Patients who were neurologically devastated (not testable) were excluded. compared with previous studies is likely due to the fact that our data
b
14 patients who did not know the age at seizure onset were excluded. are from a single tertiary center. Because some of the parents of children
c
13 patients who did not have brain magnetic resonance image were excluded. with epilepsy were reluctant to give informed consent, we could not

Table 3
Univariate and multivariate logistic regression of the clinical characteristics associated with low intelligence (IQ b 80).

Univariate Multivariable

Odds ratio 95% CI p value Odds ratio 95% CI p value

a
Age at onset of seizures b5 years 5.202 2.692 10.053 b0.001
5–9 years 1.657 0.840 3.270 0.145
≥10 years 1
Epilepsy durationa b1 year 1 1
1–5 years 3.545 1.270 9.896 0.016 3.583 1.207 10.637 0.022
6–9 years 12.424 4.438 34.776 b0.001 9.560 3.206 28.508 b0.001
≥10 years 30.933 10.002 95.665 b0.001 21.241 6.607 68.285 b0.001
Number of antiepileptic drugs 0 1
1 0.693 0.367 1.307 0.257
2 1.969 0.882 4.399 0.098
≥3 6.588 2.782 15.600 b0.001
Number of seizures in preceding year 0–1 1 1
2–9 0.886 0.506 1.551 0.672 1.067 0.542 2.104 0.851
≥10 6.025 2.812 12.908 b0.001 7.912 2.868 21.823 b0.001
EEG findings Normal 1 0.156
Focal 1.513 0.898 2.548 0.119
Generalized 0.806 0.380 1.709 0.574
Focal + generalized 1.696 0.807 3.563 0.163
Epilepsy classification ILRE 1 1
CFE 33.120 4.336 252.994 0.001 23.177 2.776 193.517 0.004
SFE 52.800 6.822 408.673 b0.001 31.887 3.773 269.473 0.001
IGE 9.000 1.117 72.513 0.039 6.440 0.730 56.780 0.094
LGS/EE 300 25.504 3528.817 b0.001 128.105 9.992 1642.374 b0.001

In the multivariable analysis, missing values were accounted for using a single imputation with the Markov chain Monte Carlo (MCMC) method.
ILRE, idiopathic localization-related epilepsy; CFE, cryptogenic focal epilepsy; SFE, symptomatic focal epilepsy; IGE, idiopathic generalized epilepsy; LGS/EE, Lennox–Gastaut syndrome/
epileptic encephalopathy.
a
14 patients who did not know the age at seizure onset were excluded.
 J. Park et al. / Epilepsy & Behavior 29 (2013) 166–171 169

Table 4
Univariate and multivariate logistic regression of the clinical characteristics associated with low intelligence in patients with idiopathic, cryptogenic, and symptomatic epilepsy.

Idiopathic Cryptogenic Symptomatic

(total n = 80; (total n = 135; low intelligence n = 84) (total n = 107; low intelligence n = 76)
low intelligence
n = 16)

Univariate Univariate Multivariate Univariate Multivariate

Odds p value Odds p value Odds 95% CI p value Odds p value Odds 95% CI p value
ratio ratio ratio ratio ratio

Age at onset of seizures b5 years 4.312 2.824 0.050 5.197 0.011

5–9 years 2.875 0.875 0.804 1.803 0.384
≥10 years 1 0.251 1 0.011 1 0.017
Epilepsy duration b1 year 1 0.058 1 b0.001 1 b0.001 1 0.009 1 0.023
1–5 years 1.529 5.895 0.032 7.910 1.320 47.416 0.024 3.333 0.190 1.736 0.258 11.674 0.570
6–9 years 5.909 15.500 0.001 16.538 2.772 98.681 0.002 8.333 0.021 5.355 0.866 33.125 0.071
≥10 years 10.400 52.500 b0.001 55.022 7.662 395.109 b0.001 20.000 0.004 8.449 1.111 64.264 0.039
Number of 0 1 0.068 1 0.001 1 0.024 1 0.013
antiepileptic 1 0.171 1.161 0.792 1.846 0.251 1.733 0.558 5.382 0.342
drugs 2 0.720 3.122 0.092 2.031 0.292 1.853 0.427 8.031 0.410
≥3 0.300 10.286 0.001 12.923 0.002 11.546 2.041 65.314 0.006
Number of seizures 0–1 1 0.154 1 0.009 1 b0.001 1 0.115
in preceding year 2–9 0.587 0.657 0.351 0.692 0.255 1.874 0.469 1.263
≥10 3.520 19.149 0.005 18.616 2.072 167.226 0.009 4.053
EEG findings Normal 1 0.824 1 0.904 1
Focal 0.524 1.265 1.680
Generalized 0.733 1.091 1.458
Focal + generalized 1.455 3.125

conduct cognitive function tests in all children with epilepsy, and chil-
dren with clinically suspected mental retardation were more likely to
be included. Hence LGS/EE (8.4%) was relatively frequent, and ILRE
(7.8%) was relatively underrepresented compared with other popula-
tion-based studies.
A proper understanding of the factors that are negatively related to
intelligence in patients with epilepsy will allow these variables to be
properly evaluated and, in the case of correctable risk factors, treated.
The risk factors that lead to intellectual impairment are multiple and
closely related to each other; therefore, we used multivariate logistic
regression to evaluate significant risk factors. After multivariate analysis,
Fig. 1. IQ according to epilepsy duration in patients with idiopathic, cryptogenic, and
symptomatic epilepsy.
a longer duration of epilepsy, a higher number of seizures in the
preceding year, and epilepsy classification were significant risk factors
for low intelligence. Since the longer duration of epilepsy and more
frequent seizures constitute greater accumulation of epileptic seizures,
epilepsy duration and/or frequency of seizures can be interpreted as
the seizure burden. Seizures are thought to damage the brain through
anoxia, lactic acidosis, and excessive excitatory neurotransmitters [19],
and this might underlie the associations between seizure burden and
low intelligence. In a previous study of adults with epilepsy, history of
status epilepticus and history of repeated generalized tonic–clonic sei-
zures (N100 in a lifetime) were associated with intellectual dysfunction
[21].
Antiepileptic drug usage was not a significant risk factor for low
intelligence in the multivariate analysis. Antiepileptic drugs decrease
neuronal excitability, interfere with normal neuronal networks, and
induce cognitive deficits [22]. Polypharmacy, higher target doses, higher
serum level of AEDs, and rapid titration are associated with adverse
cognitive effects [23,24]. In our study cohort, multiple AED usage may
represent pharmacoresistance, which overlaps with the frequency of
seizures. Therefore, the influence of seizure burden, number of seizures
in the preceding year, and duration of epilepsy on intelligence may can-
cel out the effect of multiple AEDs, leaving the independent effect of the
number of AEDs on cognition as insignificant.
In children, the age at seizure onset is an important factor for cogni-
tive function. Children with age of seizure onset b5 years showed sig-
nificantly lower IQ regardless of epilepsy classification [15,25], and,
among surgically treated patients with temporal lobe epilepsy, intellec-
tual impairment was more frequent in patients with seizure onset be-
fore 1 year of age [17]. In this study, age at seizure onset was not a
risk factor for low intelligence in multivariate analysis. This could be
explained by an offset between the age at seizure onset and the duration
of epilepsy. Nearly half of the patients were aged 5 years or under at the
time of seizure onset, and these patients are likely to have a longer time
between seizure onset and cognitive evaluation and, therefore, a greater
exposure to seizures prior to the cognitive evaluation. Moreover, epilep-
sy duration, another biomarker of pharmacoresistance, may be more
important than the age at seizure onset. Consistent with this result,
Berg at al. reported that lower IQ level was significantly associated
with age at onset only in patients with pharmacoresistant epilepsy [26].
170 J. Park et al. / Epilepsy & Behavior 29 (2013) 166–171
Intellectual impairment may also differ according to seizure type
and epileptic syndrome. Symptomatic or cryptogenic focal epilepsy
and LGS/EE were associated with higher risk of cognitive dysfunction
than idiopathic epilepsy, especially ILRE, supporting a previous report
[27].
The etiology of epilepsy was one of the most significant risk factors
for low intelligence; therefore, we repeated the risk analysis in each eti-
ological subgroup. In patients with idiopathic epilepsy, there were no
significant risk factors for low intelligence. According to Henkin et al.
[28], poor attention, verbal memory deficits, word fluency difficulties,
and poor control of fine motor responses are not caused by seizure bur-
den or treatments but are innate characteristics of IGE itself. In addition,
the distinct cognitive difficulties in epilepsy could be due to the fre-
quent interictal epileptiform discharges from the epileptic focus as pre-
vious studies indicated [12,13,29,30]. Even though several studies have
reported subtle cognitive and behavioral deficits, such as verbal dys-
function [31–34], spatial perception difficulties [35] and learning dis-
abilities [14,34,36] in BCECTS, and attention, verbal learning and
memory deficits in IGE [28], the intelligence of such patients was rela-
tively normal in these studies. This finding suggests that the cognitive
impact of idiopathic epilepsy may be different from symptomatic/
cryptogenic epilepsies and limited to other domains of cognition apart
from intelligence.
Nevertheless, in this study, the insignificant difference of IQ
according to clinical factors could be explained by the mild clinical fea-
tures of IGE or ILRE, relatively late onset, less frequent seizures, and
good seizure control with minimal AEDs. By contrast, seizure frequency
and epilepsy duration were significant risk factors for low intelligence
in patients with cryptogenic epilepsy, and AED number and epilepsy
duration were significant risk factors for low intelligence in patients
with symptomatic epilepsy. In the symptomatic group, taking more
than three AEDs might reflect the refractoriness of seizures more so
than seizure frequency, and this could explain the difference between
the two groups. Furthermore, the LGS/EE in 27 patients with profound
mental deterioration, frequent seizures, and refractoriness to conven-
tional AEDs were categorized as symptomatic or cryptogenic. Potential
mechanisms of cognitive regression in LGS/EE are repetitive severe
seizures and persistent spike waves [37], and the inclusion of LGS/
EE patients could contribute to the significance of seizure burden
or pharmacoresistance.
Electroencephalogram findings did not show any significant associ-
ation with low intelligence. Epileptiform discharges of EEG at one time
point may not reflect the ongoing status of brain function, and epilepti-
form discharges in BCECTS cannot be equivalent to those of epileptic en-
cephalopathy, such as LGS.
There are several limitations to this study, including the retrospec-
tive design, potential selection bias arising from the use of data from a
single tertiary center, and omission of potential confounders such as
psychiatric comorbidities, social economic status, family resources,
and history of education. However, the inclusion of a relatively large
population of children with epilepsy in a single tertiary center can be
the strength of this study.
In conclusion, this study reinforces the previous findings that the
most important risk factors for cognitive function in childhood-onset
epilepsy were the underlying neuropathology and seizure burden.
However, the seizure burden does not affect the intelligence of patients
with idiopathic epilepsy, and the role of AEDs was relatively small. Phy-
sicians should pay attention to the cognitive function of children with
epilepsy and try to control seizures with appropriate AEDs to alleviate
the harmful impact of epilepsy on cognition.
Acknowledgment
The authors would like to thank Sun-Ok Kim for statistical analysis.
No funds were received in support of this work. No benefits in any
form related to this manuscript have been received from any commer-
cial party.
References
[1] Sparrow SS, Davis SM. Recent advances in the assessment of intelligence and cogni-
tion. J Child Psychol Psychiatry 2000;41:117–31.
[2] Berg AT. Epilepsy, cognition, and behavior: the clinical picture. Epilepsia 2011;52(Suppl. 1):
7–12.
[3] Lhatoo SD, Sander JW. The epidemiology of epilepsy and learning disability. Epilepsia
2001;42(Suppl. 1):6–9 [discussion 19–20].
[4] Camfield C, Camfield P. Preventable and unpreventable causes of childhood-onset
epilepsy plus mental retardation. Pediatrics 2007;120:e52–5.
[5] Sillanpaa M. Epilepsy in children: prevalence, disability, and handicap. Epilepsia
1992;33:444–9.
[6] Fastenau PS, Shen Jianzhao, Dunn DW, Austin JK. Academic underachievement
among children with epilepsy: proportion exceeding psychometric criteria for
learning disability and associated risk factors. J Learn Disabil 2008;41:195–207.
[7] Mitchell WG, Chavez JM, Lee H, Guzman BL. Academic underachievement in chil-
dren with epilepsy. J Child Neurol 1991;6:65–72.
[8] Seidenberg M, Beck N, Geisser M, Giordani B, Sackellares JC, Berent S, et al. Academic
achievement of children with epilepsy. Epilepsia 1986;27:753–9.
[9] Fowler MG, Johnson MP, Atkinson SS. School achievement and absence in children
with chronic health conditions. J Pediatr 1985;106:683–7.
[10] Fastenau PS, Shen J, Dunn DW, Perkins SM, Hermann BP, Austin JK. Neuropsycholog-
ical predictors of academic underachievement in pediatric epilepsy: moderating roles
of demographic, seizure, and psychosocial variables. Epilepsia 2004;45:1261–72.
[11] Hermann BP, Seidenberg M, Bell B. The neurodevelopmental impact of childhood
onset temporal lobe epilepsy on brain structure and function and the risk of progres-
sive cognitive effects. Prog Brain Res 2002;135:429–38.
[12] Goldberg-Stern H, Gonen OM, Sadeh M, Kivity S, Shuper A, Inbar D. Neuropsychological
aspects of benign childhood epilepsy with centrotemporal spikes. Seizure 2010;19:
12–6.
[13] Nicolai J, Aldenkamp AP, Arends J, Weber JW, Vles JS. Cognitive and behavioral
effects of nocturnal epileptiform discharges in children with benign childhood epi-
lepsy with centrotemporal spikes. Epilepsy Behav 2006;8:56–70.
[14] Baglietto MG, Battaglia FM, Nobili L, Tortorelli S, De Negri E, Calevo MG, et al. Neuro-
psychological disorders related to interictal epileptic discharges during sleep in
benign epilepsy of childhood with centrotemporal or rolandic spikes. Dev Med
Child Neurol 2001;43:407–12.
[15] Berg AT, Langfitt JT, Testa FM, Levy SR, DiMario F, Westerveld M, et al. Global cogni-
tive function in children with epilepsy: a community-based study. Epilepsia 2008;49:
608–14.
[16] Bell B, Lin JJ, Seidenberg M, Hermann B. The neurobiology of cognitive disorders in
temporal lobe epilepsy. Nat Rev Neurol 2011;7:154–64.
[17] 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;48:201–4.
[18] 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;42:268–74.
[19] Meador KJ, Gilliam FG, Kanner AM, Pellock JM. Cognitive and behavioral effects of
antiepileptic drugs. Epilepsy Behav 2001;2:SS1–SS17.
[20] Murphy CC, Trevathan E, Yeargin-Allsopp M. Prevalence of epilepsy and epileptic
seizures in 10-year-old children: results from the Metropolitan Atlanta Develop-
mental Disabilities Study. Epilepsia 1995;36:866–72.
[21] Dodrill CB. Correlates of generalized tonic–clonic seizures with intellectual, neuro-
psychological, emotional, and social function in patients with epilepsy. Epilepsia
1986;27:399–411.
[22] Sankar R, Holmes GL. Mechanisms of action for the commonly used antiepileptic
drugs: relevance to antiepileptic drug-associated neurobehavioral adverse effects.
J Child Neurol 2004;19(Suppl. 1):S6–S14.
[23] Smith DB. Cognitive effects of antiepileptic drugs. Adv Neurol 1991;55:197–212.
[24] Shorvon SD, Reynolds EH. Reduction in polypharmacy for epilepsy. Br Med J 1979;2:
1023–5.
[25] 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;171:624–9.
[26] Berg AT, Zelko FA, Levy SR, Testa FM. Age at onset of epilepsy, pharmacoresistance,
and cognitive outcomes: a prospective cohort study. Neurology 2012;79:1384–91.
[27] Motamedi G, Meador K. Epilepsy and cognition. Epilepsy Behav 2003;4(Suppl. 2):
S25–38.
[28] Henkin Y, Sadeh M, Kivity S, Shabtai E, Kishon-Rabin L, Gadoth N. Cognitive function
in idiopathic generalized epilepsy of childhood. Dev Med Child Neurol 2005;47:
126–32.
[29] Bedoin N, Ferragne E, Lopez C, Herbillon V, De Bellescize J, des Portes V. Atypical
hemispheric asymmetries for the processing of phonological features in children
with rolandic epilepsy. Epilepsy Behav 2011;21:42–51.
[30] Bedoin N, Ciumas C, Lopez C, Redsand G, Herbillon V, Laurent A, et al. Disengage-
ment and inhibition of visual–spatial attention are differently impaired in children
with rolandic epilepsy and Panayiotopoulos syndrome. Epilepsy Behav 2012;25:
81–91.
[31] Ay Y, Gokben S, Serdaroglu G, Polat M, Tosun A, Tekgul H, et al. Neuropsychologic
impairment in children with rolandic epilepsy. Pediatr Neurol 2009;41:359–63.
 J. Park et al. / Epilepsy & Behavior 29 (2013) 166–171 171

[32] Danielsson J, Petermann F. Cognitive deficits in children with benign rolandic epilepsy of
childhood or rolandic discharges: a study of children between 4 and 7 years of age with
and without seizures compared with healthy controls. Epilepsy Behav 2009;16:646–51.
[33] Croona C, Kihlgren M, Lundberg S, Eeg-Olofsson O, Eeg-Olofsson KE. Neuropsycho-
logical findings in children with benign childhood epilepsy with centrotemporal
spikes. Dev Med Child Neurol 1999;41:813–8.
[34] Staden U, Isaacs E, Boyd SG, Brandl U, Neville BG. Language dysfunction in children
with rolandic epilepsy. Neuropediatrics 1998;29:242–8.
[35] Volkl-Kernstock S, Willinger U, Feucht M. Spacial perception and spatial memory
in children with benign childhood epilepsy with centro-temporal spikes (BCECTS).
Epilepsy Res 2006;72:39–48.
[36] Miziara CS, de Manreza ML, Mansur L, Reed UC, Guilhoto LM, Serrano VA, et al.
Impact of benign childhood epilepsy with centrotemporal spikes (BECTS) on school
performance. Seizure 2012;21:87–91.
[37] Nabbout R, Dulac O. Epileptic encephalopathies: a brief overview. J Clin Neurophysiol
2004;20:393–7.