Epilepsy & Behavior 89 (2018) 99–104

Contents lists available at ScienceDirect

Epilepsy & Behavior

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

The cognitive phenotype of idiopathic generalized epilepsy

Belén Abarrategui ⁎,1, Beatriz Parejo-Carbonell, Maria Eugenia García García 2,
Daniela Di Capua 3, Irene García-Morales
Unidad de Epilepsia, Hospital Clínico San Carlos, Calle Profesor Martín Lagos s/n, 28040 Madrid, Spain

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

Article history: Objective: Dysexecutive traits have been described in idiopathic generalized epilepsy (IGE), but studies mainly fo-
Received 31 July 2018 cused on juvenile myoclonic epilepsy (JME). To better understand the neuropsychology of IGE, more research is
Revised 7 October 2018 needed on syndromes other than JME, controlling potential confounding factors as the cognitive effects of
Accepted 7 October 2018 valproate and epileptic discharges (ED). We describe the neuropsychological profile of a group of patients with
Available online 6 November 2018
different syndromes of IGE including simultaneous video electroencephalography (EEG).
Methods: We performed a comprehensive cognitive and neuropsychiatric evaluation with video-EEG on 61
Keywords:
Generalized epilepsy
adults with IGE (JME 19; IGE with generalized tonic–clonic seizures [GTCS] alone [IGE-GTCS] 22; childhood ab-
Absences sence epilepsy [CAE] or juvenile absences epilepsy [JAE] persisting in adulthood 20). We compared results be-
Juvenile myoclonic epilepsy tween patients (globally and by syndrome) and a control group of 21 individuals (similar age, educational
Executive dysfunction epilepsy level); p-values were adjusted for multiple testing according to a 0.05 false discovery rate.
Cognition epilepsy Results: Patients obtained significantly lower results than controls on visuospatial working memory, processing
Neuropsychology epilepsy speed, cognitive flexibility and strategy, abstract visuospatial reasoning, arithmetic, and acquired knowledge.
While CAE/JAE showed the lowest scores on cognitive assessment and highest anxiety index, IGE-GTCS showed
the most favorable scores. Most tests were not influenced by valproate intake, and the dose did not correlate with
cognitive performance in the test that yielded differences between patients and controls. Epileptic discharges
during assessment were not frequent (10 patients, 1–4 tests).
Significance: Our findings suggest that patients with IGE have significantly lower abilities in various executive
functions and acquired knowledge, compared to population of same age and education. The low frequency of
ED on simultaneous video-EEG and absence of correlation of scores with valproate dose reinforce that the ob-
tained results are due to a cognitive phenotype in IGE. This phenotype may be influenced by syndrome, and pa-
tients with CAE/JAE persisting in the adult may have a wider neuropsychiatric impairment.
© 2018 Elsevier Inc. All rights reserved.

1. Introduction combination of absences, generalized tonic–clonic (GTC), and myo-

Idiopathic generalized epilepsies (IGE) account for 15–20% of cases
of epilepsy [1]. They comprise a group of different syndromes with
onset during childhood or adolescence and characterized by a variable
Abbreviations: IGE, idiopathic generalized epilepsy; JME, juvenile myoclonic epilepsy;
ED, epileptic discharges; IGE-GTCS, IGE with generalized tonic–clonic seizures alone; CAE,
childhood absence epilepsy; JAE, Juvenile Absences epilepsy; VPA, valproic acid; GTC,
generalized tonic–clonic; SW, spike-and-wave; PSW, polyspike-and-wave; ILAE,
International League Against Epilepsy; WAIS, Wechsler Intelligence Scale for Adults; IQ,
intelligence quotient; ISRA, Inventario de Situaciones y Respuestas de Ansiedad; NDDI-E,
Neurological Disorders Depression Inventory for Epilepsy; AEDs, antiepileptic drugs; TCI,
transient cognitive impairment; ICD, International Classification of Diseases.
⁎ Corresponding author.
E-mail address: belen.abarrategui@salud.madrid.org (B. Abarrategui).
1
Present address: Neurología, Servicio de Medicina Interna, Hospital Universitario de
Fuenlabrada. Calle Camino del Molino 2, 28942 Fuenlabrada, Madrid, Spain.
2
Present address: UCL Institute of Neurology, Queen Square, London, WC1N 3BG, UK.
3
Present address: Neurología, Hospital de Especialidades Eugenio Espejo, Av. Gran
Colombia, Quito 170136, Ecuador.
clonic seizures. They have a good response to antiepileptic drugs, partic-
ularly to valproic acid (VPA), and present typical reflex traits such as
photosensitivity and praxis induction [2,3]. Their hallmark on electroen-
cephalography (EEG) are generalized epileptiform discharges (ED) of
spike-and-wave (SW) or polyspike-and-wave (PSW) that are typically
induced by non-REM sleep, hyperventilation, intermittent photic stim-
ulation, and certain praxis [2].
Main IGE syndromes are childhood absences epilepsy (CAE), juve-
nile absences epilepsy (JAE), juvenile myoclonic epilepsy (JME), and
IGE with generalized tonic–clonic seizures (GTCS) alone (IGE-GTCS) [4].
From a cognitive point of view, IGE is characterized by the absence of
intellectual disability and focal neurological deficits. Beyond these gen-
eral traits, cognition in IGE has been mainly studied in patients with JME
[5]. Most publications assessing cognition in JME support the presence
of an alteration on executive functions, such as working memory, ab-
stract reasoning, and cognitive flexibility [5–7]. The analysis of this liter-
ature outlines the importance of performing complete and
comprehensive assessments of cognitive functions in order to

https://doi.org/10.1016/j.yebeh.2018.10.007
1525-5050/© 2018 Elsevier Inc. All rights reserved.
100 B. Abarrategui et al. / Epilepsy & Behavior 89 (2018) 99–104

demonstrate the selective affectation of executive skills [5]. Although it
is essential to avoid confounding factors like the presence of ED during
testing, only a minority of studies are performed with simultaneous
video-EEG [8].
On the other hand, published research regarding cognitive evalua-
tion in syndromes different than JME is much less extensive [9,10]. Spe-
cifically, absences epilepsy that persists in adulthood has received much
lesser attention in literature.
The aim of this study was to describe the complete neuropsycholog-
ical profile of a group of patients with different syndromes of IGE.
2. Material and methods
We recruited prospectively adult patients with IGE from the Epi-
lepsy Unit of Hospital Clínico San Carlos (Madrid) between 2014 and
2016. Diagnosis and syndromic classification were based on clinical his-
tory and EEG findings according to the International League Against Ep-
ilepsy (ILAE) [11]. Age of inclusion was 18–55 years. All participants had
at least one previous EEG showing typical generalized ED of SW or PSW,
and a normal neuroimaging. Their native language was Spanish. Exclu-
sion criteria were as follows: a diagnosis of schizophrenia, bipolar or ob-
sessive–compulsive disorder; regular treatment with topiramate in a
daily dose of ≥100 mg/day [12]; and/or medical past history with poten-
tial effect over cognition.
The study included anamnesis, neurological examination, video-EEG
registration, comprehensive cognitive assessment, and neuropsychiat-
ric evaluation.
Cognitive assessment comprised test of attention and executive
function, episodic memory, acquired knowledge, naming, visuospatial
and visuoperceptive skills (Table 1). General cognitive abilities were ex-
amined by a short form (seven tests) of the WAIS-III battery (Wechsler
Intelligence Scale for Adults) that allowed calculating the intelligence
quotient (IQ) [13,14]. The total duration of this cognitive assessment
was approximately 2 h.
Neuropsychiatric evaluation addressed different anxiety traits with
the ISRA (Inventario de Situaciones y Respuestas de Ansiedad) test
[15], and depressive symptoms with the NDDI-E (Neurological Disor-
ders Depression Inventory for Epilepsy) [16] and Beck [17].
The whole study was performed during video-EEG monitoring. The
video-EEG was recorded with a 32-channel XLTEK system during 5–
24 h, according to the international 10–20 system. Two experienced
epileptologist (IGM and BA) reviewed page-by-page (10-second
pages) the whole recordings on referential average and longitudinal bi-
polar montages with 0.5- to 70-Hz bandwidth, and manually marked
every ED registered during the cognitive evaluation.
Healthy volunteers were recruited to constitute a control group of
similar age and formal education level than the patient group. Exclusion
criteria were family history of epilepsy, a psychiatric diagnosis (similar
than those for patients) and medical history or current treatment with
potential effect over cognition. They were submitted to the same cogni-
tive and neuropsychiatric evaluation protocol than patients.
Statistical analysis was performed with SPSS 22.0. Firstly, we com-
pared cognitive and neuropsychiatric raw scores between patients and
controls. Secondly, patients were subdivided by syndrome and results
of each group were compared with those of controls, in order to analyze
the specific neuropsychological profiles. In this second part, we used
test scores adjusted for age and education according to normative data
for the Spanish population NEURONORMA [18], instead of raw scores,
in order to adjust for differences in education and especially in age
among subgroups.
For this analysis, Student t-test/Mann–Whitney U test were used ac-
cording to the normal or abnormal distribution of each test results. The
resulting p-values were adjusted for multiple testing according to a false
discovery rate of 0.05 by the Benjamini and Hochberg procedure [19].
This study was approved by the Clínico San Carlos Hospital Ethics
Committee. All participants signed the written informed consent.
3. Results
3.1. Sample description
Sixty-one patients with IGE were included, of whom 34 were
women (55.7%). Mean age was 32.3 ± 9.7 years and median number
of years of formal education was 14 [12–16]. Syndromic classification
of patients was JME in 19 and IGE-GTCS in 22. Twenty other patients
presented a CAE (10) or JAE (10) persisting in adult life (see Table 2).
Thirty-seven (60.7%) patients were treated with monotherapy and 23
(37.7%) with polytherapy. The most common antiepileptic drugs
(AEDs) at the time of the study were VPA (n = 30, 49.2%, median
daily dose 1100 mg [875–1525]), levetiracetam (42.6%), and
lamotrigine (31.1%). Fourteen patients (22.9%) were pharmacoresistant
according to the ILAE criteria [20].

Table 2
Table 1 Demographic and clinical data by syndrome.
Neuropsychological test battery for the cognitive assessment.
Global IGE group (n = 61)
Attention and executive functions Age at epilepsy onset (y) 13 [11–17]
Digit Span (forward and backward) (WAIS) Duration of epilepsy (y) 18 [8–25]
Visual Span (Corsi Cubes; forward and backward) JME (n = 19)
Coding (WAIS) Age (y) 33 ±8.1
Stroop (words, colors, interference) Education (y) 14 [11–15]
Tower of London Age at epilepsy onset (y) 14 [12–16]
Five Points Test Duration of epilepsy (y) 18 [14–25]
Similarities (WAIS) VPA daily dose (mg) 950 [600–1150]
Matrix (WAIS) Pharmacoresistance 36.8%
Arithmetic (WAIS) CAE/JAE (n = 20)
Boston Naming test Age (y) 34.4 ±10.7
Verbal Fluency (phonemic, excluded letter) Education (y) 12 [12–16]
Episodic memory Age at epilepsy onset (y) 9.5 [6.2–13]
Rey–Osterrieth Figure (copy, recall at 3′, recall at 30′, recognition) Duration of epilepsy (y) 24.5 [11.2–37]
Wechsler Memory Scale: List of words (learning, interference, recall at 3′, recall VPA daily dose (mg) 1550 [1287–2000]
at 30′) Pharmacoresistance 25%
Acquired knowledge IGE-GTCS (n = 22)
Information (WAIS) Age (y) 29.9 ±10.0
Visuospatial and visuoperceptive function Education (y) 15 [13–17]
Benton Judgment of Line Orientation Age at epilepsy onset (y) 15.5 [12.7–20]
Visual Object and Space Perception Battery (Object decision and Progressive Duration of epilepsy (y) 8 [4–21]
Silhouettes) VPA daily dose (mg) 1000 [850–1300]
Picture completion (WAIS) Pharmacoresistance 9.1%

WAIS: Wechsler Intelligence Scale for Adults. y: years.

 B. Abarrategui et al. / Epilepsy & Behavior 89 (2018) 99–104 101

The control group was formed by 21 individuals, of whom 12 were IGE-GTCS all through the assessment (Table 4). Within this group,
women (57.1%). Mean age was 33.2 ± 9.0 years and median education those patients whose epilepsy was classified as CAE had a lower IQ
14 [12–17.5] years. Mann–Whitney comparisons between patients and (89 ± 9.8), which was the lowest among all patients with IGE (p =
controls regarding age, sex distribution, and education yielded no signif- 0.022).
icant differences (p N 0.6 in all comparisons).
3.3.2. JME
3.2. Neuropsychological evaluation: patients versus controls The group with JME scored significantly lower than controls in pro-
cessing speed (Coding and Stroop). Means were lower than controls in
Patients obtained significantly lower results than controls on visuo- the same executive skills that were described for the global group of IGE
spatial working memory (Visuospatial Span backward), processing and also on acquired knowledge, but differences did not reach statistical
speed (Stroop, Coding, Rey Figure-time to copy), cognitive flexibility significance (Table 4). Their total IQ was within the normal range (98.2
and strategy (Five Points), abstract visuospatial reasoning (Matrix), ar- ± 12.9), not significantly inferior to controls (109.5 ± 10.2, p = 0.051).
ithmetic, and acquired knowledge (Information). Table 3 shows mean
scores for each test by group. 3.3.3. IGE-GTCS
The patient group showed lower IQ as measured by the WAIS The group with IGE-GTCS showed the most favorable scores among
(99.2 ± 13.1) when compared with that of the control group all patients. In executive functions their scores were lower than controls
(109.5 ± 10.2, p = 0.025), although both were within the normal without statistically significant differences. Their IQ (103.7 ± 14.5) was
range. neither significantly lower than controls (109.5 ± 10.2, p = 0.660)
(Table 4).
3.3. Neuropsychological profiles by syndrome Direct comparisons group-to-group among syndromes did not yield
significant differences.
3.3.1. CAE/JAE persisting in adulthood
Patients with CAE/JAE scored significantly worse than controls in vi- 3.4. Neuropsychiatric evaluation: results and relation with cognition
suospatial working memory (Visuospatial Span backward), processing
speed (Coding), cognitive flexibility and strategy (Five Points), abstract Assessment of anxiety did not show statistically significant differ-
visuospatial as well as verbal reasoning (Matrix, Similarities), visuospa- ences between patients and controls, although scores were higher in pa-
tial orientation (Benton), arithmetic, and acquired knowledge (Infor- tients (Table 5). Patients with CAE/JAE persisting in adulthood
mation). These patients showed lower scores than those with JME and presented the highest anxiety scores on ISRA test.
Seventeen percent of patients had major depression as revealed by
Table 3 NDDI-E (they obtained a score N13) [16]. None of the controls scored
Results of the cognitive assessment (patients and controls). within the depression range. Severity of depressive symptoms assessed
by Beck showed that only one patient had severe depression.
Tests Patient group Control group p
No association was found among anxiety, depression, and cognitive
Total IQ 99.2 ±13.1 109.5 ±10.2 0.025
performance in the global group of patients as well as in the group with
Digit Span (forward) 5.6 ±1.0 5.9 ±1.3 0.421
Visuospatial Span (forward) 5.5 ±1.0 5.8 ±0.8 0.421 CAE/JAE.
Digit Span (backward) 4.6 ±1.1 4.6 ±1.0 0.989
Visuospatial Span (backward) 5.1 ±1.2 5.6 ±0.7 0.046 3.5. Video-EEG
Coding 66.5 ±15.6 78 ±16.6 0.045
Stroop (words) 106.5 ±14.6 119.2 ±12.5 0.019
Stroop (colors) 73.5 ±13.1 78.9 ±10.9 0.230
Background EEG was normal in all patients. In most cases (51 of the
Stroop (colors-words) 48 ±12.8 53.3 ±9.9 0.230 61 patients, 83.6%), we did not register any ED during cognitive assess-
Stroop Interference Index 3.8 [−3.0–10.2] 7.1 [6.0–8.7] 0.522 ment. Ten patients presented generalized ED at least on one test, and no
TOL latency time (s) 38 [27.5–57] 54 [30–66] 0.421 patient presented them on more than 4 (1–4 tests). The Tower of Lon-
TOL total correct exercises 4 [2-6] 6 [2.5–6.5] 0.448
don was the test with more frequent ED: six patients, three of which
TOL excess moves 34.4 ±17.2 30.6 ±18.7 0.559
TOL time for execution (s) 228 [191–291] 233 [199–272] 0.927 had a JME, two a CAE/JAE persisting in adult life, and one an IGE-GTCS.
Rey-O Figure: Copy time 145 [120–176] 110 [85–140] 0.046 Those patients who had ED during the Tower of London needed a longer
Rey-O Figure: Copy Quality 34 [32–35] 34 [34–35] 0.395 time to execute the test. They also tended to present a lower mean on
Five points: 1st minute 16.2 ±5.0 19.3 ±4.7 0.053 the test (“excess moves”, Table 6), but without significant differences
Five points: 1st + 2nd minutes 26.9 ±8.7 31.4 ±6.1 0.121
Five points: repetitions 1 [0–2] 1 (0–2) 0.989
(p = 0.088).
Five points: strategy 18 [11–32] 28 [22–33.7] 0.046
Similarities 20.2 ±4.2 22.3 ±3.9 0.152 3.6. Effect of VPA
Matrix 16.7 ±4.3 19.8 ±2.8 0.019
Arithmetic 13.2 ±4.2 15.9 ±3.7 0.046
Patients treated with VPA (n = 30, 49.2%) showed lower scores than
Phonemic Fluency (P) 15.7 ±4.4 17.7 ±4.6 0.205
Excluded letter Fluency (A) 10.3 ±4.3 11 ±3.9 0.722 those with a different treatment on the test Visuospatial Span (working
Boston 52.9 ±5.3 54.3 ±3.1 0.421 memory, 9 [7–11] and 7 [5–9], p = 0.038) and on the test Codification
Wechsler list- 1st trial 6 [5-7] 6 [5-6] 0.989 (visuomotor coordination, 10 [9–11] and 8 [6–9.3], p = 0.000).
Wechsler list- Total learning 36.3 ±3.4 36.3 ±3.9 0.989 The remaining nine tests that yielded some differences between pa-
Wechsler list- Contrast* 1 [0–2] 1 [0.5–2] 0.996
tients and controls (Table 3) were not affected by the VPA intake. The
Wechsler list: % recall at 30’ 88.8 [81.8–100] 91.6 [83.3–100] 0.623
Rey-O Figure (recall at 3′) 22.8 ±4.7 23.5 ±5.5 0.749 VPA dose did not show any significant correlation with cognitive perfor-
Rey-O Figure (recall at 30′) 22.8 ±4.6 22.8 ±5.2 0.989 mance (Table 7) within the global group of patients.
Rey-O Figure (Recognition) 20.2 ±2.3 20.8 ±1.3 0.521 We also analyzed the correlations between the test scores and VPA
Information 16.9 ±5.0 20.4 ±3.3 0.046
dose within the group with CAE/JAE, whose mean daily VPA dose was
Benton 25 ±3.3 26.5 ±2.7 0.176
Object decision (VOSP) 17.9 ±1.4 17.8 ±1.4 0.836 higher and whose cognitive scores were poorer. We did not find any
Progressive Silhouettes (VOSP) 7.5 ±2.4 6.8 ±2.1 0.421 correlation in the nine tests in where the results were actually poorer
Picture completion 20.4 ±2.3 21.4 ±1.6 0.175 in patients with CAE/JAE than in controls (Table 7). Concerning the re-
IQ: Intelligence Quotient; TOL: Tower of London; Rey-O: Rey–Osterrieth; VOSP: Visual Ob- maining test, a significant correlation was found among higher VPA
ject and Space Perception. Test with statistically significant differences are marked in bold. dose and a worse performance in attention (forward digit span, R =
102 B. Abarrategui et al. / Epilepsy & Behavior 89 (2018) 99–104

Table 4
Cognitive evaluation scores by syndrome.

JME CAE/JAE IGE-GTCS Control group p1 p2 p3

Total IQ 98.2 ±12.9 94.5 ±9.8 103.7 ±14.5 109.5 ±10.2 0.051 0.000 0.660
Digit Span (forward) 8 (8.0–10.0) 8 (8.0–10.0) 10 (8.0–10.8) 10 (8.0–10.0) 0.820 0.599 0.860
Visuospatial Span (forward) 11 (8.0–11.0) 9 (5.3–11.0) 10 (8.8–10.0) 10 (9.0–11.5) 0.931 0.368 0.836
Digit Span (backward) 11 (9.0–11.0) 8 (8.0–11.0) 10 (7.8–11.3) 10 (7.5–11.0) 0.690 0.941 0.860
Visuospatial Span (backward) 12 (6.0–12.0) 7 (6.0–12.0) 9 (6.0–12.0) 12 (8.0–12.0) 0.430 0.046 0.690
Coding 9 (6.0–10.0) 9 (6.3–10.0) 10 (8.8–12.0) 11 (9.0–12.5) 0.038 0.009 0.806
Stroop (words) 7 (5.0–11.0) 8 (6.0–11.0) 10 (6.8–11.0) 11 (9.5–13.5) 0.038 0.084 0.589
Stroop (colors) 9 (7.0–12.0) 8.5 (5.0–10.8) 10.5 (7.8–13.0) 10 (8.3–13.0) 0.412 0.091 0.860
Stroop (colors–words) 10 (8.0–12.0) 8.5 (6.0–11.0) 8.5 (6.0–13.3) 12 (8.5–14.0) 0.530 0.089 0.690
Stroop Interference Index 5.7 (–1.1–8.2) 6 (–2.5–10.5) 3.6 (–5.0–13.1) 7.1 (0.1–11.8) 0.783 0.387 0.860
TOL latency time (s) 10 (9.0–12.3) 10.5 (10.0–12.0) 12 (10.0–13.0) 10 (9.0–12.0) 0.931 0.599 0.589
TOL total correct 10.5 (8.0–13.0) 9 (8.0–10.0) 10.5 (8.0–13.0) 13 (8.5–13.0) 0.821 0.387 0.860
TOL excess moves 9 (7.8–12.0) 8.5 (6.0–10.0) 10 (7.8–11.0) 10 (7.0–12.0) 0.943 0.368 0.860
TOL time for execution (s) 9.5 (7.8–13.0) 9 (7.3–10.8) 11 (10.0–12.0) 10 (8.5–11.0) 0.821 0.434 0.836
Rey-O Figure: Copy time 10 (7.5–10.0) 10 (8.8–12.0) 10 (9.0–10.8) 12 (10.0–13.5) 0.065 0.147 0.589
Rey-O Figure: Copy Quality 10 (9.0–12.0) 9 (7.3–11.0) 11 (9.0–12.0) 11 (10.0–12.0) 0.530 0.089 0.860
Five points: 1st minute 14 (12.0–17.0) 16 (13.5–19.8) 17 (14.0–20.0) 18 (16.0–20.8) 0.065 0.368 0.589
Five points: 1st + 2nd minutes 23 (19.0–31.0) 26.5 (22.3–30.0) 28.5 (21.0–33.0) 29.5 (27.0–35.5) 0.091 0.236 0.660
Five points: Repetitions 1 (0–2.0) 1 (0.0–2.0) 1 (0.0–2.3) 1 (0.0–2.0) 0.932 0.966 0.860
Five points: Strategy 15 (11.0–34.0) 18 (10.8–28.0) 21.5 (12.3–36.3) 28 (22.3–33.8) 0.065 0.043 0.690
Similarities 11 (9.0–12.0) 10 (9.0–12.0) 11.5 (9.0–13.3) 12 (10.5–13.0) 0.453 0.043 0.860
Matrix 9 (7.0–10.0) 8 (7.3–10.0) 10 (7.8–12.3) 11 (10.0–12.0) 0.057 0.000 0.589
Arithmetic 10 (7.0–13.0) 9.5 (6.3–11.8) 12 (10.0–13.3) 13 (10.5–15.0) 0.091 0.030 0.690
Phonemic Fluency (P) 9 (6.0–10.0) 9.5 (8.3–10.0) 10 (7.8–11.0) 10 (9.0–12.5) 0.186 0.387 0.860
Excluded letter Fluency (A) 10 (8.0–12.2) 9 (8.0–10.0) 9.5 (8.8–11.0) 10 (8.5–12.0) 0.936 0.441 0.860
Boston 10.5 (9.0–11.0) 10 (7.0–11.5) 10 (9.0–14.0) 11 (9.0–14.0) 0.765 0.266 0.860
Wechsler list—1st trial 9 (5.0–11.0) 10 (8.0–11.0) 10.5 (8.0–13.0) 9 (8.0–11.0) 0.784 0.961 0.836
Wechsler list—Total learning 8 (6.0–11.0) 8 (7.0–9.8) 9.5 (7.8–11.0) 9 (7.0–11.0) 0.765 0.599 0.860
Wechsler list—Contrast* 10 (6.0–11.0) 8 (6.0–10.0) 8 (6.0–10.0) 8 (6.0–10.5) 0.821 0.794 0.860
Wechsler list: % Recall at 30′ 9 (7.8–12.2) 10.5 (8.0–13.0) 11 (8.0–13.0) 12 (9.0–13.0) 0.530 0.599 1.000
Rey-O Figure (Recall at 3′) 10 (9.0–11.0) 11 (9.0–12.0) 10 (9.0–12.0) 10 (9.5–13.0) 0.821 0.941 0.860
Rey-O Figure (Recall at 30′) 10 (9.0–11.5) 11 (9.0–12.0) 10.5 (9.0–12.0) 10 (9.0–12.5) 0.943 0.794 0.860
Rey-O Figure (Recognition) 9 (8.0–10.0) 9 (9.0–10.0) 10 (8.0–11.0) 9.5 (9.0–10.0) 0.680 0.751 0.860
Information 9 (8.0–11.0) 9 (7.0–11.0) 10 (8.0–12.0) 12 (10.5–12.5) 0.091 0.043 0.589
Benton 11 (8.0–12.0) 8 (7.0–9.3) 12 (9.0–13.0) 12 (9.5–13.0) 0.680 0.009 0.926
Object decision 10 (9.0–13.0) 10 (9.0–13.0) 10 (9.0–13.0) 10 (9.0–13.0) 0.784 0.961 0.860
Progressive Silhouettes 8.5 (6.8–12.0) 8.5 (6.8–12.0) 8 (6.3–11.8) 9 (8.0–12.0) 0.722 0.465 0.690
Picture completion 11 (9.0–12.0) 11 (10.0–12.8) 11 (8.0–13.0) 11.5 (9.0–13.3) 0.821 0.599 0.860

Mann Whitney comparisons between each group and controls are shown. p1 JME compared to controls; p2 CAE/JAE to controls; p3 IGE-GTCS to controls. IQ: Intelligence Quotient; TOL:
Tower of London; Rey-O: Rey–Osterrieth;

− 0.794; p = 0.006) and verbal fluency (phonemic fluency R =
−0.790, p = 0.007; excluded letter fluency R = −0.643; p = 0.043).
4. Discussion
The cognitive profile of patients with IGE has been focus of increas-
ing research in the last decade, with most studies centered on JME [7,
8]. As far as we know, the present study is one of the first in performing
a comprehensive cognitive and neuropsychiatric assessment on a com-
plete group of patients with IGE with mixed syndromes and simulta-
neous video-EEG.
In the cognitive assessment, patients with IGE showed a significantly
worse performance selectively in executive skills, acquired knowledge,
and overall cognitive ability (IQ) compared with controls. This finding
is consistent with previous research on the field: Chowdhury et al.
[10] evaluated IQ and executive functions in 36 patients with IGE with
mixed syndromes. They found that patients scored significantly worse
than controls in nonverbal reasoning, working memory, and IQ that is
in line with our work, as well as in verbal fluency and attention.
Loughman et al. [9] performed a comprehensive cognitive evaluation
of a group of 76 patients with IGE with mixed syndromes. They reported
poorer results compared with controls in overall cognitive ability, ac-
quired knowledge, and speed of information processing, which is

Table 5
Anxiety and depression assessment.

Control Patients JME CAE/JAE IGE-GTCS p1 p2 p3 p4

ISRA T 17.5 [13–34] 19 [8.5–25.5] 16 [11.3–30.8] 29 [14.5–41] 17 [12.5–29.5] 0.531 0.764 0.130 0.989
ISRA C 7 [4–12.5] 8 [3–14.5] 6 [2.3–10.8] 13.5 [4.5–21] 8 [3–13] 0.69 0.764 0.130 0.989
ISRA F 3 [1–8.5] 4 [2–9.5] 6 [3.3–9.8] 5.5 [2.7–13] 3 [1–6] 0.531 0.764 0.161 0.989
ISRA M 5 [3–8.5] 6 [4–9.5] 7.5 [2.5–8.8] 7 [4.7–11.5] 6 [3–9] 0.53 0.764 0.140 0.989
F1 9 [7–13.7] 11 [7.5–14] 11 [6–17] 14 [10–18] 11 [7–13] 0.53 0.764 0.106 0.989
F2 2 [1–3] 3 [2–4] 2 [2–4] 3.5 [2–4.2] 3 [1–3] 0.53 0.764 0.106 0.989
F3 2.5 [1–5.7] 2.5 [1–6.7] 1 [1–6] 5 [1–7.2] 2 [0–5] 0.877 0.764 0.344 0.989
F4 1 [0.25–1.7] 2 [0.25–3] 1 [0–3] 2 [1–4] 2 [0–2] 0.53 0.861 0.106 0.989
Beck 3.5 [2–6.7] 4.5 [1–9] 2 [0–7] 7 [0–10.5] 5.5 [1.2–9.7] 0.975 0.764 0.835 0.989

ISRA test evaluates different types of anxious responses—cognitive (ISRA C), physiological (ISRA F), and behavioral-motor (M); and anxiety provoked by situations of evaluation (F1), in-
terpersonal relations (F2), phobic stimuli (F3), and daily life (F4). It has also a global score (ISRA T). p1 patients (globally) compared to controls; p2 JME to controls; p3 CAE/JAE to controls;
p4 IGE-GTCS to controls.
 B. Abarrategui et al. / Epilepsy & Behavior 89 (2018) 99–104 103

Table 6
Performance in the Tower of London test (TOL).
With epileptic Without epileptic
discharges discharges
(n = 6) (n = 55)
TOL correct exercises 8.5 [6–10.8] 10 [8–13]
TOL excess moves 6.5 [5.8–8.3] 9.5 [7.8–11]
TOL latency time (s) 10 [8.3–12] 11 [10–12]
TOL execution time (s) 8 [4.3–8.3] 10 [9–12]
TOL solving time (s) 7.5 [3.5–11.8] 10 [9–12]
TOL slow exercises 1 [0.7–2.5] 0 [0–1]
Comparison between patients with and without epileptic discharges during the test.
consistent with our results; however, they found wider differences on
long-term retrieval, differently to this study and previous publications.
In contrast to the cited works, the present study included simulta-
neous video-EEG during the neuropsychological assessment. We con-
sidered it essential to exclude the apparition of seizures and to
identify the presence of ED during testing that could lead to transient
cognitive impairment (TCI) [21]. We observed that ED were not very
frequent (10/61–16.4% — in 1–4 tests). Indeed, the test Tower of London
that provoked the highest burden of ED was not among the tests in
where the results were inferior in patients compared with that in con-
trols. This suggests that the described cognitive profile of patients with
IGE is not justified by TCI, even if some slowness effect of ED is possible
(patients with ED during the test Tower of London obtained a lower
score on speed in that test).
The contribution of the cognitive adverse effects of VPA was another
essential factor to control; we obtained that VPA intake was associated
with a worse performance on one of the processing speed test and on
the visual working memory test, but not in the remaining nine tests
that marked differences between patients and controls. Additionally,
the dose did not correlate with the cognitive results. Our results suggest
that even with a possible effect on processing speed and working mem-
ory, VPA intake does not justify globally the cognitive outcome of the
patients. This observation, along with the one about TCI, points to the
presence of a fixed cognitive phenotype in IGE. This concept of pheno-
type is reinforced by a previous finding of Chowdhury et al. [10] that rel-
atives of patients with IGE had a similar profile of dysexecutive deficits,
but with a milder affectation than patients.
Phenotype could be influenced by the specific syndrome in view of
our results that were obtained by comparing patients with different
syndromes to controls. In the same way than prior studies, the direct
comparison among syndromes failed to define significant differences
[5,9]. In the present study, patients with CAE/JAE that persists in adult-
hood obtained the lowest results versus controls, showing a broader
and more noticeable cognitive impairment on acquired knowledge
and executive functioning, that also affected other skills (as revealed
by the visuospatial Benton test). Another finding within this group
was that those with CAE had the lowest IQ. Patients with CAE/JAE that
persists in adulthood form a group not extensively studied in literature.
Two decades ago, in a clinical series of adults with absences, the authors
observed that some of these patients exhibited “cognitive disturbances
p consisting of forgetfulness and mental slowing” [22]. However, this ob-
servation had not been translated into formal neuropsychological work.
From a psychosocial perspective, Wirrell et al. [23] described signifi-
0.204 cantly poorer academic-personal and behavioral outcomes in patients
0.088 with CAE compared with those of another group with another chronic
0.335
disease (juvenile rheumatoid arthritis). The cognitive difficulty, and
0.030
0.204 also the anxiety traits evidenced in the present work, could be in the
0.045 base of these psychosocial outcomes.
The group of patients with JME, which has been widely reported to
have a characteristic dysexecutive profile [6,7], showed few significant
differences compared with the controls in our study. This could be par-
tially explained by the loss of statistical power derived of subdividing
the sample into groups by syndrome. Means of the group with JME
were generally lower than that of the controls in the test of executive
function, but differences were not statistically significant. In previous
literature on JME, other studies with less than 20–25 patients (Iqbal et
al. [8] with 22 patients; Roebling et al. [24] with19 patients) found
few significant differences among all the executive functions that
were assessed. Undoubtedly, in clinical practice the possibility of a po-
tential executive difficulty merits attention in the follow-up of these pa-
tients. However, our results remind that dysexecutive problems are not
constantly and markedly present in JME. Also psychosocial outcomes in
JME have been demonstrated to be heterogeneous and not constantly
negative [25].
Lastly, in the case of patients with IGE-GTCS, their neuropsychologi-
cal profile was closer to the control group. However, even mild difficulty
in patients with this syndrome may impact their educational years: in a
semi-structured interview of up to 40 adult patients with IGE-GTCS by
Camfield and Camfield [26], 77% reported significant learning problems
during their school years and 70% considered that their performance on
school was “below average”.
With regard to psychiatric comorbidity, in this work, 16.9% of pa-
tients had major depression as revealed by NDDI-E. This proportion is
not far from the 13.4% (with International Classification of Diseases
[ICD]-10 criteria) reported by Akanuma et al. [27] in the biggest study
on the subject so far (157 patients with IGE), taking into account that
NDDI-E is a screening tool with a specificity of 78% [16]; NDDI-E was se-
lected for the study because of its meticulous design to detect depres-
sion in people with epilepsy, whose special characteristics can
sometimes be difficult to detect with other conventional tests: intermit-
tent dysthymic course, diagnostic confusion with adverse effects of
AEDs [28]. The low score obtained on Beck test by patients in our
study shows that depressive symptoms were relatively mild, which is
congruent with previous authors who reported milder depressive
symptoms in IGE compared to temporal lobe epilepsy [29]. However,
even mild depression associated with epilepsy has demonstrated a sig-
nificant impact on quality of life [30].
A novel finding of our study was that anxiety was more relevant in
patients with CAE/JAE. It has been demonstrated that anxiety affects

Table 7
Influence of VPA intake and VPA daily dose on scores. Results are shown for the test that yielded differences between patients and controls. IQ: Intelligence Quotient; TOL: Tower of London.

Test Without VPA With VPA p Correlation with dose p Correlation in CAE/JAE p

Total IQ 103 (88.5–110.0) 97 (88.0–105.0) 0.462 −0.233 0.532 −0.479 0.456

Visuospatial span (backwards) 12 (6.0–12.0) 6 (6.0–12.0) 0.142 −0.258 0.456 −0.547 0.384
Coding 10 (9.0–11.0) 8 (6.0–9.3) 0.000 −0.049 0.946 −0.627 0.374
Rey Figure: copy time 10 (9.0–11.0) 9.5 (6.8–10.3) 0.282 0.012 0.953 −0.624 0.384
Stroop (words) 9 (7.0–11.0) 7 (5.5–11.0) 0.714 −0.016 0.953 −0.27 0.374
Five points: Strategy 18 (13.0–34.0) 17.5 (11.0–28.0) 0.731 −0.063 0.939 −0.459 0.456
Similarities 11 (9.0–12.0) 11 (9.0–12.5) 0.971 −0.067 0.939 −0.28 0.555
Arithmetic 10 (8.0–13.0) 10.5 (7.0–13.3) 0.971 −0.104 0.939 −0.252 0.555
Matrix 10 (8.0–11.0) 8 (7.0–10.0) 0.127 −0.068 0.939 −0.255 0.555
Information 9 (7.0–11.0) 10 (8.5–12.0) 0.296 −0.098 0.939 −0.551 0.393
Benton 10 (8.0–12.0) 10 (8.0–12.0) 0.962 −0.432 0.355 0.000 1000
104 B. Abarrategui et al. / Epilepsy & Behavior 89 (2018) 99–104
quality of life independently from depression in people with epilepsy
[31]. Further, the work by Kampf et al. [32] revealed that the only factor
that affected significantly their social functioning was anxiety. Our re-
sults in the group with CAE/JAE reinforce the importance of addressing
this comorbidity especially in this group of patients with IGE.
5. Conclusions
Our findings suggest that patients with IGE have significantly lower
abilities in various executive functions and acquired knowledge com-
pared with those of a population of their same age and level of educa-
tion, and that these lower abilities are due to a cognitive phenotype;
the low frequency of ED during assessment and the absence of correla-
tion with VPA dose do not justify the results. This phenotype may be in-
fluenced by the specific syndrome, and patients with CAE/JAE persisting
in adulthood may have a wider cognitive impairment and greater levels
of anxiety. For future research into neuropsychiatric aspects in IGE,
studies with bigger sample size could permit direct comparison
among groups to confirm our results concerning the influence of the
syndrome, as well as to describe the specific features of the affective dis-
orders and anxiety in IGE as it has been extensively made in temporal
lobe epilepsy.
Conflict of interest
The authors declare no conflicts of interest.
Acknowledgments
The authors gratefully acknowledge the statistical support by Irene
Serrano and Cristina Fernández from the Clinical Epidemiology Unit of
Hospital Clínico. We would also like to acknowledge Isabel Graciani
and the video-EEG nursing team for their support with the caring and
technical aspects of video-EEG.
First author's work was partially funded by the Fundación para la
Investigación Biomédica Hospital Clínico San Carlos (Grant Numbers, on
the research project “Epilepsy, cognition and psychiatric comorbidity”).
References
[1] Jallon P, Latour P. Epidemiology of idiopathic generalized epilepsies. Epilepsia 2005;
46(s9):10–4.
[2] Panayiotopoulos CP. Idiopathic generalised epilepsies. In: Panayiotopoulos CP, edi-
tor. A clinical guide to epileptic syndromes and their treatment. London: Springer
Healthcare; 2010. p. 394–6.
[3] Yacubian EM, Wolf P. Praxis induction. Definition, relation to epilepsy syndromes,
nosological and prognostic significance. A focused review. Seizure 2014;23:247–51.
[4] Scheffer IE, Berkovic S, Capovilla G, Connolly MB, French J, Guilhoto L, et al. ILAE clas-
sification of the epilepsies: position paper of the ILAE Commission for Classification
and Terminology. Epilepsia 2017;58:512–21.
[5] Loughman A, Bowden SC, D'Souza W. Cognitive functioning in idiopathic generalised
epilepsies: a systematic review and meta-analysis. Neurosci Biobehav Rev 2014;43:
20–34.
[6] Pascalicchio TF, de Araujo Filho GM, da Silva Noffs MH, Lin K, Caboclo LO, Vidal-
Dourado M, et al. Neuropsychological profile of patients with juvenile myoclonic ep-
ilepsy: a controlled study of 50 patients. Epilepsy Behav 2007;10:263–7.
[7] Thomas RH, Walsh J, Church C, Sills GJ, Marson AG, Baker GA, et al. A comprehensive
neuropsychological description of cognition in drug-refractory juvenile myoclonic
epilepsy. Epilepsy Behav 2014;36:124–9.
[8] Iqbal N, Caswell H, Muir R, Cadden A, Ferguson S, Mackenzie H, et al. Neuropsycho-
logical profiles of patients with juvenile myoclonic epilepsy and their siblings: an ex-
tended study. Epilepsia 2015;56:1301–8.
[9] Loughman A, Bowden SC, D'Souza WJ. A comprehensive assessment of cognitive
function in the common genetic generalized epilepsy syndromes. Eur J Neurol
2017;24:453–60.
[10] Chowdhury FA, Elwes RDC, Koutroumanidis M, Morris RG, Nashef L, Richardson MP.
Impaired cognitive function in idiopathic generalized epilepsy and unaffected family
members: an epilepsy endophenotype. Epilepsia 2014;55:835–40.
[11] Proposal for revised classification of epilepsies and epileptic syndromes Commission
on Classification and Terminology of the International League Against Epilepsy,
Epilepsia 1989;30:389–99.
[12] Loring DW, Williamson DJ, Meador KJ, Wiegand F, Hulihan J. Topiramate dose effects
on cognition: a randomized double-blind study. Neurology 2011;76:131–7.
[13] Sattler JM, Ryan JJ. Assessment of children: revised and updated third edition. WAIS-
III supplement. San Diego, CA: Sattler; 1998.
[14] Ryan JJ, Ward LC. Validity, reliability, and standard errors of measurement for two 7-
subtest short forms of the Wechsler Adult Intelligence Scale-III. Psychol Assess 1999;
11:207–11.
[15] Miguel-Tobal JJ, Cano-Vindel A. Inventario de situaciones y respuestas de ansiedad
(ISRA): Manual. Madrid: TEA; 2002.
[16] Di Capua D, Garcia-Garcia ME, Reig-Ferrer A, Fuentes-Ferrer M, Toledano R, Gil-
Nagel A, et al. Validation of the Spanish version of the Neurological Disorders De-
pression Inventory for Epilepsy (NDDI-E). Epilepsy Behav 2012;24:493–6.
[17] Sanz J, Perdigon L, Vazquez C. Adaptacion espanola del inventario para la depresion
de Beck—II (BDI-II): propiedades psicometricas en poblacion general. Clin Salud
2003;14:249–80.
[18] Pena-Casanova J, Blesa R, Aguilar M, Gramunt-Fombuena N, Gómez-Ansón B, Oliva
R, et al. Spanish multicenter normative studies (NEURONORMA Project): methods
and sample characteristics. Arch Clin Neuropsychol 2009;24:307–19.
[19] Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and pow-
erful approach to multiple testing. J R Stat Soc B 1995;57:289–300.
[20] Kwan P, Arzimanoglou A, Berg AT, Brodie MJ, Allen Hauser W, Mathern G, et al. Def-
inition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the
ILAE Commission on Therapeutic Strategies. Epilepsia 2010;51:1069–77.
[21] Drane DL, Ojemann JG, Kim MS, Gross RE, Miller JW, Faught Jr RE, et al. Interictal ep-
ileptiform discharge effects on neuropsychological assessment and epilepsy surgical
planning. Epilepsy Behav 2016;56:131–8.
[22] Michelucci R, Rubboli G, Passarelli D, Riguzzi P, Volpi L, Parmeggiani L, et al.
Electroclinical features of idiopathic generalised epilepsy with persisting absences
in adult life. J Neurol Neurosurg Psychiatry 1996;61:471–7.
[23] Wirrell EC, Camfield CS, Camfield PR, Dooley JM, Gordon KE, Smith B. Long-term
psychosocial outcome in typical absence epilepsy. Sometimes a wolf in sheeps'
clothing. Arch Pediatr Adolesc Med 1997;151:152–8.
[24] Roebling R, Scheerer N, Uttner I, Gruber O, Kraft E, Lerche H. Evaluation of cognition,
structural, and functional MRI in juvenile myoclonic epilepsy. Epilepsia 2009;50:
2456–65.
[25] Syvertsen MR, Thuve S, Stordrange BS, Brodtkorb E. Clinical heterogeneity of juve-
nile myoclonic epilepsy: follow-up after an interval of more than 20 years. Seizure
2014;23:344–8.
[26] Camfield P, Camfield C. Idiopathic generalized epilepsy with generalized tonic-clonic
seizures (IGE-GTC): a population-based cohort with N20 year follow up for medical
and social outcome. Epilepsy Behav 2010;18:61–3.
[27] Akanuma N, Hara E, Adachi N, Hara K, Koutroumanidis M. Psychiatric comorbidity in
adult patients with idiopathic generalized epilepsy. Epilepsy Behav 2008;13:248–51.
[28] Kanner AM. Depression in epilepsy: prevalence, clinical semiology, pathogenic
mechanisms, and treatment. Biol Psychiatry 2003;54:388–98.
[29] Sarkis RA, Pietras AC, Cheung A, Baslet G, Dworetzky B. Neuropsychological and psy-
chiatric outcomes in poorly controlled idiopathic generalized epilepsy. Epilepsy
Behav 2013;28:370–3.
[30] Perrine K, Hermann BP, Meador KJ, Vickrey BG, Cramer JA, Hays RD, et al. The rela-
tionship of neuropsychological functioning to quality of life in epilepsy. Arch Neurol
1995;52:997–1003.
[31] Johnson EK, Jones JE, Seidenberg M, Hermann BP. The relative impact of anxiety, de-
pression, and clinical seizure features on health-related quality of life in epilepsy.
Epilepsia 2004;45:544–50.
[32] Kampf C, Walter U, Rösche J. The impact of anxiety, seizure severity, executive dys-
function, subjectively perceived psychological deficits, and depression on social
function in patients with epilepsy. Epilepsy Behav 2016;57(Pt A):5–8.