INTRODUCTION

 Benign epilepsy of childhood (BFE) is defined by the International League against Epilepsy¹   classification scheme as an idiopathic age and localization-related epileptic syndrome with a combination of clinical and EEG characteristics. It is characterized by brief, simple partial and hemi-facial motor seizures with associated somatosensory symptoms that can evolve into generalized seizures. EEG shows high-voltage spikes often followed by a slow wave ². The two most commonly described benign partial epilepsies of childhood are benign childhood epilepsy with centro-temporal spikes (BFECTS) and childhood epilepsy with occipital paroxysms (BEOP). They account for   8–23% of all childhood epilepsy³.

Children with BFECTS showed significantly poorer results in total IQ, visual perception, spatial orientation, short-term memory, some fine-motor tasks, as well as impairment of the reading, spelling, auditory verbal learning and auditory discrimination. Those data indicated a language dysfunction that was closely associated with academic difficulties. ^4,5,6 The objective of this study is to use psychometric studies, P300, Contingent   negative  variation (CNV)  and Quantitative

 

EEG analysis for objective assessment of possible alterations of the cognitive functions in children BFECTS.

Aim of work: To assess possible alterations in cognitive and psychological functions in school children having BFE in comparison to age and sex matched healthy children.

 

SUBJECTS AND METHODS

 

The study was carried-out on 30 school children suffering from BFECTS (group 1) as well as 30 age and sex matched healthy controls (group 2). The diagnosis of BFECTS was based on the presence of the Nocturnal brief somatosensory seizures, involving the tongue, the inner cheeks, lips and gums, speech arrest, pooling of saliva and occasionally tonic or tonic-clonic spread. Patient age ranged from 7 – 15 years of either sex. Any patient who had history of brain damage, cerebral palsy, active local pathology, mental retardation, psychomotor seizures ,temporal lobe epilepsy, Landau-Kleffner , Rett's syndrome or fragile X syndrome were excluded from the study. All the children had normal Brain MRI examination. All patients underwent general and neurological history taking and full neurological examination.

 

Neurophysiologic methods:

Digital EEG Recordings were carried out on a Dantec, Medtronic PL-EEG, 2003 machine. Twenty one Ag/AgCl electrodes were applied according to the international 10-20 system. One hour of continuous recording under standard conditions was obtained for each patient. The EEG was then automatically divided in to 20 second epochs and then data were automatically analyzed by the computer software (Windsor – 2003) for spike frequency / epoch (sf/e) which is the mean number of spikes/ epoch, source localization and Frequency analysis.

 

P300 was carried out using the auditory oddball paradigm. Patients were instructed to raise the right hand as a reaction to the target tone but not the frequent one. The montage used was Cz – linked earlobes. Band pass was 0.5-70 Hz. Stimulus intensity was adjusted at 60 dB above the subject’s hearing threshold. Responses to 30 target and 120 non-target tones were obtained.

 

CNV was carried out using three active electrodes applied at Fz, Cz, and Pz. Linked earlobes were used as the reference electrode. Eye movements were monitored by the electrode on the infra-orbital margin. The patient was instructed to press a response switch immediately after the presentation of the imperative stimulation. An auditory Tone-Burst was used as warning stimulation and LED goggles stimulation was used as imperative stimulation. Stimulation frequency 0.1 Hz, delay time 2 seconds. Sensitivity was set at 50µv. Band pass was 0.01 - 20 Hz.

 

Cognitive Assessment:

 

The Wechsler Intelligence Scale for Children (WISC) ⁷ translated to Arabic language was used for cognitive assessment. The Scale is divided into 2 main groups (1) Verbal IQ (VIQ) that has 6 sub-tests ; Information, Comprehension, Arithmetic, Similarities, Vocabulary, and Digit Span; (2) Performance IQ (PIQ) that has 6 subtests; Picture Completion, Picture Arrangement, Block Design, Object Assembly, Coding and Mazes. Each test has raw scores according to the age of the child, after measuring the raw tests; they are converted to scaled scores according to the appropriate table to the child’s age and then the total is converted into a Total IQ (TIQ) measure.

 

Statistical Methods:

 

      IBM SPSS statistics (V. 21.0, IBM Corp., USA, 2012) was used for data analysis. Comparison between two independent mean groups for parametric data were done using Student t- test. Comparison between independent groups for non-parametric data was done using Wilcoxon Rank Sum test. Ranked Spearman correlation test was used to study the possible association between each two variables among each group for non-parametric data. The probability of error at 0.05 was considered sig., while at 0.01 and 0.001 are highly significant.

RESULTS

 

The mean TIQ in group (1) was 90.1±13.9% ; in group (2) the mean was 90.1 ± 13.8% .The mean VIQ in group (1) was 93.3±14.4% ; in group (2) the mean was 93.6 ± 13.8% .  Mean PIQ in group (1) was 88.6±13.4%, in group (2) 88.3 ± 14.0 %. There was no statistically significant differences between the 2 groups regarding the sores of the WISC or any of the subsets (p= 0.933, 0.920, 0.925 respectively) (Table 1).

All patients had the characteristic large amplitude spikes with a big positive component in the centro-temporal region. Eight patients had bilateral spikes, 10 had right sided spikes and 12 had left sided ones. The sf/e ranged from 1 to 5 with a mean of 2.66±1.29 sf/e.

P300: The mean P300 latency (P3 L) in group (1) 339.8± 28.0, in group (2) the mean was 336.5±17.1 ms.  The difference between the P3 L in group (1) compared to group (2) was statistically non- significant. (p = 0.577). The mean P300 amplitude (P3A) in group (1) 15.7±9.2 µv, in group (2) the mean was 22.6 ± 14.7 µv. There was a significantly lower P3A in group (1) compared to group (2) (p = 0.024).

CNV: CNV in group (1) the mean was 61.2± 44.7µv, in group (2) the mean was 50.3 ± 25.3 µv. The difference between the CNV in group (1) compared to group (2) was statistically non- significant. (p = 0.965).

Frequency of Fits: There was highly significant inverse correlation between frequency of the fits and Picture Arrangement, Block Design, Object Assembly & Coding of PIQ  (p= 0.004, 0.005, 0.002, 0.002 respectively); on the other hand there was no significant correlation between frequency of fits and the VIQ (p=0.555) and any of its subsets or the Total IQ (p=0.469). The correlation between the P3L, P3A, and CNV and the frequency of fits was statistically non-significant. (p=0.211, 0.301, 0.532 respectively).

Patients who had right sided spikes had significantly lower scores of the Total IQ (right =87.3±13.8, left = 98.75±4.86, p=0.0217), Total VIQ (right = 83.8±11.35, left = 101.41 ±7.99,p=0.0005) namely the Arithmetic (Right = 7.6± 2.27, left = 13.08± 2.93, p=0.0001), Similarities ( right =7.5± 1.53, left = 10.83±2.2, p=0.0006) and Vocabulary (right = 4.8± 1.3, left = 8.5± 2.27, p=0.0002) sub-tests. The localization of the spikes in the EEG did not have a significant effect on the Total PIQ (right = 92.00±2.34, left = 96.41±6.35, p=0.178).

Treatment: Compared to untreated patients, the treated ones had highly significant higher scores of the Total IQ  (untreated = 84.3± 15.0, treated = 98.5± 6.5, p=0.001), the Total VIQ (treated= 102.3+ 10.2,  untreated= 87.1+ 13.7, p=0.002) namely the Comprehension, Arithmetic, Similarities & Vocabulary subtests of VIQ Tests.(table 2), the total PIQ ( treated = 94.9± 6.8, untreated =84.4± 15.0 , p=0.016) namely the Object Assembly( treated patients=11.1+ 3.8,untreated=8.1+ 3.8 p=0.041) .

Treated patients had a significantly shorter mean P3 L (327.4± 21.3 ms) and higher mean P3A (28.0± 15.2 µv) compared to untreated ones (P3 L = 348.1± 29.4 ms, P3A =14.4± 9.3 µv p= 0.033, 0.009 respectively).

There was no significant difference of CNV and sf/e between the treated (54.9+ 38.9µv, 2.94±0.8 respectively) and untreated (65.4+ 48.9 µv, 2.25+1.71 respectively) patients (p= 0.658, 0.214).

Duration of treatment: There was a significant inverse correlation between duration of treatment and the scores of the Total IQ ( p=0.018) , the Information (p=0.019) and Comprehension  ( p= 0.035) sub-tests of the VIQ test, the Total PIQ (r= p=0.005) mainly the Picture arrangement ( p=0.004), Block Design (p= 0.005), Object Assembly (p= 0.001) and Coding ( p= 0.002).There was a statistically  non- significant  correlation between duration of treatment and CNV amplitude (p= 0.100), P3 L and P3A (p= 0.697, 0.17 respectively)

Mono-therapy versus poly-therapy: There was no significant difference between the patients on mono-therapy and those on poly-therapy in the Total IQ (84.5± 14.8; 84.2± 15.4, p=0.973) or VIQ (85.8±10.6; 88.8± 17.5, p=0.672) or PIQ Tests (85.9± 17.6; 82.6± 11.9, p=0.674) or any of their sub-tests.

 

 

Mono-therapy and poly-therapy had statistically non-significant effects on P3 L, P3A or CNV (P3 L =348.3+ 28.9, 348± 32.0, p= 0.984; P3A = 24.6± 2.9; 32.2± 13.4, p= 0.528; CNV = 56.1± 53.8, 77.0± 42.6, p=0.244). There was no significant difference of sf/e between patients on mono (2.9±1.19) or poly-therapy (3.0±0.09) p=0.79.

 

Psychometric scores neurophysiology results: The Total IQ and the total VIQ were not significantly related to the sf/e (p= 0.390, 0.836 respectively), yet, there was a highly significant inverse correlation between sf/e and the PIQ (p=0.007) namely the Picture Completion, Picture Arrangement, Block Design & Coding tests (p=0.001, 0.046, 0.011, 0.032 respectively) . There was as well a highly significant inverse correlation between sf/e and comprehension subset of VIQ Tests (p=0.006).

Neurophysiology inter-relations: The relation between the P3 L, P3A, CNV and sf/e was statistically non-significant (p= 0.739, 0.668 and 0.72 respectively)

Patients who had left sided spikes had significantly lower P3A (left =16.75±9.91, right = 28.84±17.31, p= 0.0431). Localization of the spikes did not have a statistically significant effect on P3 L (left = 342.83±15.3, right = 347±36.37, p= 0.72) or the CNV (left = 56.43, right = 64.35, p= 0.721).

 D

Table 1. Comparison between group (1) and group (2) in WISC

Verbal IQ		group (1)	group (2)	p value
	Information	11.2± 2.6	11.03± 2.8	0.816
	Comprehension	9.5± 2.5	9.4± 2.5	0.959
	Arithmetic	11.5± 3.9	10.7± 4.0	0.982
	Similarities	9.7± 2.6	9.6± 2.6	0.911
	Vocabulary	5.5± 2.8	6.8± 2.8	1.00
	Digit Span	5.6± 2.4	5.6± 2.4	1.00
Performance IQ	Picture Completion	10.8± 4.2	11.0± 4.0	0.858
	Picture Arrangement	9.2± 2.2	9.2± 2.2	0.954
	Block Design	10.4± 2.6	10.4± 2.5	0.96
	Object Assembly	9.4±3.9	9.3± 4.0	0.982
	Coding	4.9± 4.3	4.8± 4.3	0.963
	Mazes	5.1± 1.5	5.1± 1.5	0.933

 

 

 

 

 

 

 

 

 

Table 2. Comparison between Verbal IQ scores in treated and untreated patients

	Information	Comprehension	Arithmetic	Similarities	Vocabulary	Digit Span	Total
Untreated	10.3± 3.3	8.6± 2.8	9.0± 3.1	8.7± 2.5	5.7± 2.5	5.0± 2.1	87.1± 13.7
Treated	12.0± 1.6	10.7± 1.1	13.2± 3.9	10.9± 2.4	8.4± 2.5	6.4± 2.7	102.3± 10.2
P-Value	0.066	0.009**	0.006**	0.03*	0.01**	0.195	0.002**

*Significant at p<0.05  **Significant at p<0.01 

DISCUSSION

 

Several investigators reported a decline in the short and long term memories, attention, concentration, speed of mental processing and even motor control in epileptic patients. These abnormalities may be caused by the process of epileptogenesis, the seizures and/or antiepileptic drugs. BECTS is a form of epilepsy with no demonstrable anatomical lesion showing spontaneous seizure remission. Despite the normal global intellectual abilities, compared to the control groups, children with BECTS, showed some significant impairments of specific regions of their cognitive functions, academic performances and speech abilities. ^7,8,9 Miziara et al , 2012 ⁹ reported lower scores of the digits and similarities sub-tests of the WISC, significantly poorer results of the total IQ, visual perception and spatial orientation, short-term memory, and  some fine-motor tasks. Similarly  Staden et al., 1998 ⁵ showed that BFECTS children showed poor standard scores in two or more of the 12 language tests, whereas the median scores for the IQ were within the average range for the group as a whole. In this study the mean WISC in group (1) were not significantly different from those of group (2), yet within group (1) , the higher the frequency of seizures; the lower the scores of the PIQ. The relation was statistically significant. The scores of the PIQ were also significantly inversely related to frequency of spikes in EEG, with a specifically significant deterioration in PIQ in patients with higher spike frequency. Patients with Right sided spikes showed lower TIQ and VIQ scores. Previous investigators  reported that in BFECTS children with unilateral spikes (regardless of lateralization) scored higher total IQ and PIQ than those with bilateral spikes, hence concluded that the cognitive symptoms associated with BFE are thought to be related to the spike localization in the brain cortex and functional hemisphere asymmetry.^11,12 More recent studies, using combined EEG/MEG that allows for more accurate spike localization, showed a correlation between the location of spikes and selective cognitive deficits in BFE ^13,14 In contrast to the WISC scores those with right side spikes had higher mean P3A probably because that clinical psychometric scales and event related potentials measure   different aspects of cognitive functions .

      The slower information processing, defective memory and reduced attention levels in group (2), is further indicated by the significant reduction in P300 amplitudes and the trend towards a higher  P300 latencies compared to group (1) .Although the frequency of the seizures was significantly correlated to the PIQ yet did not  significantly affect the P300 latencies or P300 amplitude. This suggests that probably the consequences of the seizure expression and the process of epileptogenesis are responsible, indifferent ways  for the clinical psychometric profiles and stimulus conductivity and  processing within the CNS not only in BFECTS but also in other forms of epilepsy as reported by previous investigators. ^15,16,17  The improvement in the cognitive functioning in the treated children under CBZ mono-therapy, as indicated by the lower P300 and CNV amplitudes and shorter P300 latencies (compared to the untreated), as well as the previously reported remission of the P300 abnormalities after complete recovery of  BFECTS further point out to role played by an integrated pathophysiological epileptogenic mechanism that initiates seizures and alters the cognitive aspects of those patients.

Not only the cognitive functions are impaired in BFECTS but also a possible disorder of the regulatory mechanisms related to decision making, motor readiness and execution were reported ¹⁸.  Maalouf et al ¹⁹ 2006 found that performance and adaptation to perturbing mechanical constraints imposed by a robotic device were significantly impaired in children with BFECTS reflecting impaired motor control. In our study the epileptic patients had a trend towards higher amplitude of nearly all the components of the CNV which are slow cortical potentials related to motor preparation, decision making, somatosensory, feedback and motor execution.  The abnormalities of the CNV amplitudes were independent of the seizure frequency, mode and duration of treatment. In temporal lobe epilepsy lower mean CNV amplitudes, not related to treatment, were recorded ²⁰ probably due to different pathophysiological mechanisms underlying the epileptic syndromes.

      Although treated patients, compared to untreated ones, had significantly higher scores in the total, performance and verbal IQs, as well as, shorter P3 l and higher P300a , yet among the treated patients, our data agree with previous investigators CBZ and probably other anti-epileptic drugs had both a desirable effect on cognitive functioning, through improvement of the underlying dysfunction caused by epileptogenesis, and an undesirable effect reflecting chronic impairment associated with longer duration of treatment .²¹  Patients on mono-therapy did not significantly score better in WISC or P300 and CNV compared to those on poly-therapy . Previous investigators reported results similar to ours’, in contrast others reported that patients on mono-therapy shorter P300 latencies^22,23. The results differed according to the type of drug used for mono-therapy, the drug combination in poly-therapy, where phenobarbital was found to have the most adverse effect on P 300 followed by CBZ, valproate and phenytoin. The effects are dose dependent.

 

In conclusion abnormalities in psychometric tests and event related potentials recorded in this study provide objective evidence about cognitive deficits and educational problems in BFECTS among school children, hence challenging the term “Benign”. Those impairments probably result from the pathophysiological process underlying the epileptogenesis as well as the impact of repeated seizures activity on the stimulus processing and conduction within the brain. Electrophysiological and psychometric studies measure different aspects of the cognitive processing and should be used as complementary, not alternative, tests for full investigation of probable alteration in cognitive functions.

 

 

[Disclosure: Authors report no conflict of interest]

 

REFERENCES

 

1.      Berg AT, Berkovic SF,  Brodie MJ, Buchhalter J, Cross HJ, Van Emde Boas W et al . Revised terminology and concepts for organization of seizures and epilepsies: Report of the ILAE Commission on Classification and Terminology, 2005–2009. Epilepsia. 2010; 51: 676- 85.

2.      Kavros PM, Clarke T, Strug LJ, Halperin JM, Dorta NJ, Pal DK. Attention impairment in rolandic epilepsy: Systematic review.  Epilepsia. 2008 Sep;49(9):1570-80.

3.      Holmes GL .  Benign focal epilepsies of childhood. Epilepsia. 1993; 34: S49–61.

4.      Weglage J, Demsky A, Pietsch M. Neuropsychological, intellectual, and behavioral findings in patients with centro-temporal spikes with and without seizures. Dev Med Child Neurol. 1997; 39:646–51.

5.      Staden U, Isaacs E, Boyd SG, Brandl U, Neville BG. Language dysfunction in children with Rolandic epilepsy. Neuropediatr. 1998; 29:242-8.

6.      Tedrus GM, Fonseca LC, Melo EM, Ximenes VL. Educational problems related to quantitative EEG changes in benign childhood epilepsy with centrotemporal spikes. Epilepsy Behav. 2009; 15: 486-90.

7.      Fonseca LC, Tedrus GM, Pacheco EM, Berretta MF, Campregher AA, Costa DM. Benign childhood epilepsy with centro-temporal spikes: correlation between clinical, cognitive and EEG aspects. Arq Neuropsiquiatr. 2007; 65: 569-75.

8.      Valkl-Kernstock S, Bauch-Prater S, Ponocny-Seliger E, Feucht M. Speech and school performance in with benign partial epilepsy with centro-temporal spikes (BFECTS), Seizure. 2009; 18: 320-6.

9.      Miziara CS, de Manreza ML, Mansur L, Reed UC, Guilhoto LM, Serrano VA et al. Impact of benign partial epilepsy with centro-temporal spikes (BCECTS) on school performance. Seizure. 2012; 21: 87-91.

10.    Staden U, Issacs E, Boyd SG. Language dysfunction in children with rolandic epilepsy. Neuropediatrics. 1998; 29: 242–8.

11.    Zenkov LR, Konstantinov PA, Shiriaeva IIu, Miasnikov VN, Sirazitdinova EB, Shevel'chisnkia¬ SI.  Mental and behavioral disorders in idiopathic focal epileptiform spikes. Zh Nevrol Psikhiatr. 2007; 107: 39-49.

12.    Jeong MH, Yum MS, Ko TS, You SJ, Lee EH, Yoo HK.    Neuropsychological status of children with newly diagnosed idiopathic childhood Epilepsy. Brain Dev. 2011;33:  666-71.

13.    Wolff M, Weiskopf N, Serra E, Preissl H, Birbaumer N, Kraegeloh-Mann I. Benign partial epilepsy in childhood: selective cognitive deficits are related to the location of focal spikes determined by combined EEG/MEG Epilepsia.2005; 46:1661-7.

14.    Kakisaka Y, Gupta A, Enatsu R, Wang ZI, Alexopoulos AV, Mosher JC, et al.: Magnetoencephalography revealed a unique neurophysiological profile of focal-onset epileptic spasms. Tohoku J Exp Med. 2013 ; 229 :147-51.

15.    Konishi T, Naganuma Y, Hongou K, Murakami M, Yamatani M, Yagi S. Changes of P300 latency with age in childhood epilepsy. Pediatr Neurol. 1995 ;12:132-5.

16.    Bocqillon P, Dujardin K, Betrouni N, Phalempin V, Houdayer E, Bourriez J. Attention impairment in temporal lobe epilepsy: a neurophysiological approach via analysis of the P300 wave. Hum Brain Mapp. 2009 ; 30: 2267-77.

17.    Duncan CC, Mirsky AF, Lovelace CT, Theodore WH. Assessment of the attention impairment in absence epilepsy: comparison of visual and auditory P300. Int J Psychophysiol. 2009; 73: 118-22.

18.    Hamano T, ikeda A, collura T, Comir Y, Shibasaki H. The cortical generators f contingent negative variation in humans: a study with subdural electrodes. Electroencephalogr clin Neurophysiol. 1997; 104: 257-68.

 

19.    Maalouf M, Takahashi C, Reinkensmeyer D, Cooper D, Rho J. Impaired Motor Control in Patients With Benign Focal Epilepsy of Childhood. J Child Neurol. 2006; 21:157-60.

20.    Drake M, Weate  S, Newell S. Contingent negative variation in epilepsy.Seizure. 1997;6: 297-301.

21.    Naganuma y, Konishi T, Hongou K, Murakami M, Yamatani M, Yagi S, et al. Auditory event related potential in benign childhood epilepsy with centro-temporal spikes : the effects of carbamazepine. Cin electroencephaloogr. 1994; 25: 8-12. 

22.    Ozmenek O, Nazleil B, Levento A, Bilir E.  The role of event related potentials in evaluation of subclinical cognitive dysfunctions in epileptic patients. Acta Neurol Belg. 2008, 108: 58-63.

23.    Triantfyllou N, Zalonis I, Kokoytis P, Anthracopoulos M, Siafacas A. Cognition in epilepsy: A multichannel event related potential (P300) study. Acta  Neurol  Scand. 1992; 86: 462-5.