• Frontal Lobe;
  • Focal epilepsy;
  • Cognition;
  • Behavior


  1. Top of page
  2. Abstract

Summary: Rationale: Few studies have looked at long-term epileptic and cognitive outcome of frontal lobe epilepsy (FLE) in children. Most are limited by inclusion of lesional and nonlesional patients.

Goal: To define the epileptic and functional outcome of children with nonlesional FLE.

Methods: We reviewed medical records and neuropsychological evaluations of patients with nonlesional FLE diagnosed between 1994 and 2004. We included children with either focal or regional frontal EEG and/or functional imaging abnormalities. We reviewed their charts for seizure and neuropsychological outcome.

Results: We included 21 children. Twelve (57.1%) presented with daily seizures. Seizures were nocturnal in 8 of 21, secondarily generalized in 6 of 21, adversive in 5 of 21, and focal motor in 6 of 21. Although, initial seizure control was poor in 14 of 21, long-term control was achieved in 10 of 21 after 14.6 ± 22.3 months. Early development was normal in 12 of 21 but at later formal neuropsychological evaluation only 3 of 12 still had a normal profile. The majority of children had learning difficulties requiring special education prior to seizure onset (6 of 10). A clearly defined regression after seizure onset was observed in three children. The majority exhibited attention deficit and hyperactivity or impulsivity (14 of 21), behavioral problems (8 of 21), and cognitive impairments (10 of 21). Early seizure control was associated with a better cognitive outcome.

Conclusion: Nonlesional FLE is associated with poor seizure and behavioral outcomes. Whether this is secondary to MRI-silent developmental lesions or to the progressive repercussion of seizures on frontal lobe functions remains uncertain. A prospective study with early neuropsychological assessment could help confirm the latter.

Frontal lobe epilepsy (FLE) has a significant impact on the children's behavior, learning, and development (1–5). Executive function, attention, and fine motor deficits are some of the disabilities described with frontal lobe dysfunction. These have been previously reported in adult patients with lesional FLE as well as in children with frontal lobe injury (2,3). Our group previously compared 16 children with FLE, eight with temporal lobe epilepsy, and eight with generalized absence on different neuropsychological tests (2). We demonstrated that children with FLE are more impaired in attention, impulse control, and working memory than the two other groups. But little data is available on the epileptic and neuropsychological outcome of children with nonlesional FLE (4).

The aim of our study was to define these outcomes in children with nonlesional FLE. We wanted to evaluate the impact of FLE on school performance and neuropsychological profile and correlate these changes with the long-term epileptic outcome in the absence of a structural lesion.


  1. Top of page
  2. Abstract


We included 21 children seen at the Ste-Justine Hospital Comprehensive Epilepsy Centre's outpatient clinic between 1994 and 2004. We included all children, with either focal EEG findings involving only frontal lobe electrodes or with regional EEG changes (involving frontal and temporal or central electrodes) and focal functional imaging abnormalities limited to the frontal lobes. Exclusion criteria were the presence of a frontal lobe lesion or of nonlocalizing electroencephalographic abnormalities, including spike and wave complexes. We reviewed the charts for demographic data, seizure history, and neuropsychological evaluations and included in the analysis: medical problems, family history of epilepsy, early development, seizure frequency, functional imaging changes, EEG findings, drug treatment, and response to therapy. We reviewed the neuropsychological profile of all children. Cognition, behavior, and school performance were recorded. These characteristics were correlated with seizure control.


  1. Top of page
  2. Abstract

Clinical characteristics

The mean age of the group at seizure onset was 6.7 years (±3.9). Three children were less then 1 year old and two older than 13 years at diagnosis (median 6.0 years). There were followed for a mean period of 9.4 ± 3.5 years. Most of these children had no history of febrile convulsion (19 of 20, 95%) and no family history of epilepsy (14 of 20, 70%). One boy was adopted. Two children were born of a twin pregnancy and both were discordant with their sibling for epilepsy (Table 1).

Table 1. Clinical characteristics of the study population
Total number21100%
Gender, male/female14/767/33
Age at seizure onset, years6.71 ± 3.9 
Mean follow-up, years9.36 ± 3.5 
Perinatal history:
 Twin 2 9.5%
 Preterm 1 4.8%
 Anoxia grade 1 1 4.8%
 Maternal diabetes and HTA 1 4.8%
Medical problems:
 Migraine 314.3
 Parasomnia 1 4.8
 Dystonia 1 4.8
Early development:
 Language delay 628.6
 Motor delay 314.3
 Global delay 1 4.8
 Febrile seizure1/20 5.0
 Family history of seizure6/2030.0

Early development, as assessed by the medical questionnaire, was normal in 57.1% (12 of 21). The other children had delays: one with language and motor delay, five with language delay alone, two with motor delay alone, and one with global developmental delay. There was no focal or lateralized abnormality on neurological examination.

Seizure history

The seizures types observed were variable. We recorded secondarily generalized tonic and/or clonic seizures in six children, absence-like or myoclonic seizures in three. Partial seizures were also observed: focal motor in six, adversive in five, complex partial in four, and gelastic in three children. These seizures were only nocturnal in 8 of 21 patients (38.1%) (Table 2).

Table 2. EEG findings and epileptic outcome in our nonlesional FLE population
Seizure's characteristics
EEG lateralization
 Bilateral 628.6%
SPECT and PET scan
Functional imaging concordant with EEG lateralization9/9100%
Seizure frequency at onset
 ≤1/month 628.6%
 >1/month ≤ 1/week 1 4.8%
 >1/week ≤ 1/day 2 9.5%
Seizure control at last follow-up visit
 No seizure1047.6%
 ≤1/month 2 9.5%
 >1/month ≤ 1/week 314.3%
 >1/week 628.6%

At seizure onset, 57.1% of children had daily seizures, five with more than 10 per day. These five children remained poorly controlled. In contrast, the six children with monthly seizures at presentation were all seizure-free at the last follow-up.

The most frequent choice of initial therapy was carbamazepine (CBZ) in 11 children (52.4%). The other choices were valproic acid (VPA) in five, phenytoin (PHT) in four, clobazam (CLB) in two, and phenobarbital (PB) in one patient (two patients were started with two drugs). The initial response was more than 50% reduction in seizure frequency in 7 of 11 on CBZ, 2 of 5 VPA, 2 of 4 PHT, 1 of 2 CLB, and 0 of 1 PB. Of these 12 initial responders, seven became and remained seizure-free. At last follow-up, five children were on one anticonvulsant, while 10 children required more then three drugs and 8 of 10 were still not seizure-free. During the follow-up period, children had tried a combination of anticonvulsants including CBZ in 15, CLB in 16, VPA in 12, lamotrigine (LTG) in nine, topiramate (TPM) in five, PHT and levetiracetam (LEV) in four. These drugs were discontinued in 3 of 4 children on PHT (75%), 7 of 12 on VPA (58.3%), 7 of 16 on CLB (43.8%), 6 of 15 on CBZ (40%), 1 of 4 on LVT (25%), 1 of 5 on TPM (20.0%), and 1 of 9 on LTG (11.1%) because of lack of efficacy or side effects in a number of children. Four children remained on lamotrigine despite poor efficacy due to positive behavioral effects.

Complete seizure control was achieved in 10 children (47.6%) in a mean period of 14.6 months (±22.32 SD). The EEG became normal in 57.1% of the children after a mean time of 1.67 year of treatment (±0.78). Normalization of the EEG was associated with seizure freedom in seven of the 12 children (58.3%). At the time of EEG normalization nine patients were on CBZ, five on CLB, four on LTG, three on VPA, two on TMP, and one on PHT and LEV.

Neuropsychological profile

Sixteen out of 21 patients developed a learning disability. This was not necessarily due to an intellectual deficit, as 52.4% had a normal IQ (WPPSI-R, WISC-III). Seizure control did not predict academic performance, as six of the children with learning disability were seizure-free. Four out of six seizure-free children had a normal IQ and two a borderline normal IQ. And four had an attention deficit disorder with or without hyperactivity (Table 3).

Table 3. Patients' characteristics
PatientAge at seizure onset, yearsSeizure-free at follow-upNormal EEG at follow-upIQNumber of AED* triedLearning disabilitiesDeterioration post seizure onset
  1. *Antiepileptic drug.

 116  YesYesNormal1No 
 214  YesYesNormal2No 
 3 4.8YesYesNormal5Yes 
 4 0.2NoNoModerate delay6Yes 
 5 2.6NoYesMild delay3Yes 
 6 7.1YesYesNormal1Yes 
 7 5  NoNoMild delay3YesX
 8 6  YesNoMild delay2Yes 
 9 6  NoNoNormal8Yes 
10 9.7NoYesNormal5Yes 
11 0.4NoNoModerate delay1Yes 
12 3.3YesNoNormal2Yes 
13 9  YesNoNormal1No 
14 5  YesYesNormal1Yes 
15 6.6NoYesNormal5Yes 
16 7  YesYesNormal3No 
17 0.3NoNoModerate delay6Yes 
18 4.7NoYesNormal6No 
19 6  NoYesModerate delay5YesX
20 4  YesYesMild delay6Yes 
21 5  NoNoNormal4YesX

For most the learning difficulties preceded seizure onset, as 60% (6 of 10) of the children aged 6 years and older at presentation, had learning difficulties requiring special education prior to seizure onset. In addition, a clearly defined regression after seizure onset was observed in only three patients; and only one of these patients was developmentally normal before. Almost all children with seizure onset before the age of 6 years went on to develop a learning disability (10 of 11, 90.9%). The majority exhibited an attention deficit disorder with hyperactivity or impulsivity (14 of 21) and behavioral problems (8 of 21). Three children had a language deficit and two had memory difficulties. Of the three children with a language disability, only one was noted to have a language delay prior to school entry.

We looked at school performance and its relationship with behavioral problems. All the children with attention deficit with or without hyperactivity required special academic support. The same thing applied for the five children with language or memory deficits. Three of the five children with an isolated impulsive behavior and two of the eight with behavioral problems (oppositional behavior, impulsivity, anxiety) did well in a regular school setting. The others required special education supports.


  1. Top of page
  2. Abstract

The principal finding of this study is that in children with nonlesional FLE, seizure control does not guarantee the absence of severe comorbidity in this population. Our data suggest that the comorbidity often precedes seizure onset. We hypothesize that subtle-MRI silent lesions involving the frontal lobes might be the underlying etiology in a number of these children. Development of new MRI techniques might help us with the diagnosis of an underlying lesional etiology (6).

However, early age of onset and persistent seizures appear to lead to additional difficulties as previously reported (1). In children, epilepsy can interfere with normal brain development (2). We know that epilepsy can disrupt learning and even brain development (7). The impact of seizures during development might be even more important in children with FLE as frontal lobes continue to develop after birth and until late adolescence (2,3). For example, 10 of our 11 patients with seizure-onset before age six had an unfavorable epileptic and neuropsychological outcome. The majority of children with nonlesional FLE require special educational support and sometime specialized schools. The reason of their educational failure seems multiple but involving behavioral more than cognitive problems.


  1. Top of page
  2. Abstract

Nonlesional FLE is not necessarily associated with a poor seizure outcome but is often associated with poor behavioral outcomes and learning disabilities. We believe that in children, FLE, even without a known etiology, can not be considered benign. Compared to temporal lobe epilepsy, these patients have less memory and language disabilities, but more pronounced attention and behavior problems associated with severe learning disabilities. Whether this is secondary to an MRI-silent developmental lesion or the repercussion of epileptic discharges on frontal lobe functions remains uncertain. We think that aggressive management may improve the long-term outcome of these patients. This not only requires early seizure control but also the recognition of the poor global prognosis of this nonlesional form of epilepsy, which should help us obtain special help from educational professionals for these children at the time of diagnosis to improve their global outcome. A prospective study with neuropsychological assessment at the time of diagnosis and an educational intervention could help demonstrate the progressive nature of nonlesional FLE.


  1. Top of page
  2. Abstract