ANTIEPILEPTIC MEDICATION IMPROVES SCHOOL SUCCESS IN CHILDREN WITH FOCAL EPILEPTIFORM DISCHARGE WITHOUT EPILEPTIC SEIZURES
Objectives: The aim of this paper was to show that suppressing focal epileptiform discharges improves school success in children without epileptic seizures, but with focal epileptiform discharges and poor school success.
Study design: In the prospective study lasting from 1993 to 2009 year. 30 children with poor school success, and focal epileptiform discharges, without epileptic seizures, were followed. Electroencephalographic recordings, psychology testing and behavioral scales were performed at baseline and each 12 months during the follow-up period. Children were followed up without antiepileptic medication for one year, and then they were followed with active drug treatment (valproate 20 mg/kg/day, karbamazepine 20 mg/kg/day, or lamotrigine 5-8 mg/kg/day) from 12 months up to five years. The primary hypothesis to be tested was that the school success would improve in children by the suppression of focal epileptiform discharges using antiepileptic medication.
Results: The global rating of school success improved only in patients who showed significant electroencephalography normalization during antiepileptic drug medication.
Conclusions: Our data show that suppressing focal epileptiform discharges in children without epileptic seizures improves school success.
Key words: children, school success, focal epileptiform discharges, antiepileptic medication
Previous studies have suggested that cognitive disfunctionss and behavior problems (BP) are common in children with epilepsy, and are related to the type of epilepsy, type of medications, the underlying neurological disorder, family environment, parenting behaviors, and teachers and school friends behavior (1,2,3,4). Focal (Rolandic) epileptiform spikes are described in up to 5.7% of children without overt epileptic seizures (ES). Prospectively, the incidence of epileptic discharges (ED) in the child psychiatric client population is found to be increased three to five times over what would be expected from epidemiologic studies (5). In children with a first recognized ES, higher than expected (compared with their healthy siblings) rates of cognitive and BP are found in the six months before the first recognized ES (32.1%) (4).
It seem they exist before the onset of ES in children with a normal intellectual quotient (IQ), and there is no comorbidity in the form of Atenttion Deficiency Hyperactivity Disorder (ADHD) (6). Interictal ED, or sub-clinical ED, occur in up to 80% of patients with ongoing epilepsy, although they may not be seen in every EEG (7). This leads us to wonder, do we undertreat patients with epilepsy by aiming only to suppress clinically obvious ES, and what is the nature of the mechanism that might be related to both the ES and behavioral disturbances? Aicardi proposes that transient cognitive impairment (TCI) caused by ED might be this mechanism. Support for this hypothesis includes empirical evidence; ED not accompanied by clinically observable ES may result in brief episodes of impaired cognitive and reduced psychosocial functioning (7). It is uncertain whether these ED are truly subclinical, or cause brief consciousness disruptions.
The only way to determine whether ED causes cognitive and BP in children without ES is by finding whether cognitive and BP reduce and school success increase when ED are suppressed.
We aimed to test whether ED in children without ES cause poor school success by performing a follow up, crossover study to examine whether suppressing ED can improve school success, improving cognitive functions and the behavior in children with ED without ES.
For the prospective study lasting from 01.01.1993 to 31.12.2009, the study group included 32 children with ED, without ES and antiepileptic, or any other, medication. The participants were 16 boys and 14 girls with a mean age of nine years and three months (range 8-15years) recruited after their first EEG at the Institute for Child and Youth Health Care in Vojvodina.
All children sent to the EEG department from their pediatrician and/or psychologist due to BP and poor school success. Only children with poor school success, according to inclusion criteria have been included consecutively.
Benign epileptiform discharges (BEDs) in the Rolandic region coexist in pediatric patients with intractable localization-related epilepsy, secondary to hippocampal sclerosis (8). To avoid its effects in our study, only children with normal CNS MRI were included in our study.
The EEG, psychological testing and behavioral scales were performed during the baseline and at each follow-up period of every 12 months. The children were followed up without AEM for one year, with AEM (karbamazepine-KBZ, valproic acid-VPA or lamotrigine-LTG) for 12 months.
During the first year, one boy was excluded from further observation, due to the appearance of ES.
During the third year other boy was excluded by his parent’s decision. KBZ was introduced in children with clear unilateral ED, and VPA was given to children with bilateral ED showing some tendency to generalization during sleep. Both (KBZ and VPA) were given in two daily doses, 20 mg/kg/day. In children without an improvement in school success with persistant ED after one year, AEM was replaced with LTG (5-8 mg/kg/day, in two daily doses). Only monotherapy was used. At entry, history, a physical and neurological examination, EEG, standard biochemical tests, and IQ tests (Wechsler Intelligence Scale for Children, WISC-III) were performed. Behavioral scales were completed by parents and teachers. The children were assessed at the end of each treatment phase (12 months, 24 months, and 36 months), at which point the following were recorded: a physical examination, AED blood levels, EEG, psychological tests, behavioral scales for parents and teachers, and documentation of adverse events.
EEG monitoring was performed for 30 minutes using an Oxford Medical 12-channel digital recording system. All children had sleep – wake records within the 30 minutes of recording after partial sleep deprivation. The EEG recordings were analyzed visually. EEG was analyzed from author of this paper and described according to standard rools.
EEG was analyzed after first visit, and then every six months according to follow up.The EEG features of ED were frequent spike and wave discharges in the centrotemporal region with the electrical field of the epileptiform discharges not distributed widely; they commonly occur in runs, mostly unilateral, are not blocked by activating movements or eye opening, and are activated during the first sleep stages. A continuous run of epileptiform waveforms would be considered as one ED if not interrupted by normal activity of more than one second. The discharges were quantified by the “frequency of discharges” (number per minute) and the “discharge time” (duration in seconds per hour). The minimum duration allocated to any single ED was one second.
Behavior was assessed with the Conners Rating Scales (CRS) for parents and teachers. The CRS are analytically-derived scales for assessing behavior problems in children.
Table 1. Inclusion/exclusion criteria
The parent’s rating scale consists of 93 questions and the teacher’s rating scale of 39 questions. The raw scores are translated into t scores. The t scores have a mean of 50 and a standard deviation of 10. Higher scores denote more serious BP (9). The parent’s rating scale has eight subscales designated as I-antisocial, II-anxious, III-conduct disorder, IV-immature, V-learning problem, VI-obsessive/compulsive, VII-psychosomatic, and VIIIrestless/ disorganized. The teacher’s rating scale has six subscales labeled as I-anxious/passive, II-asocial, III-conduct problem, IV-daydream, V-emotional/indulgence, and VI-hyperactivity. The rating forms were completed by the same person on all occasions. Only children with learning problem and poor sucess where included.
School success was described as “excellent” with median mark 4,5-5, “very good” with median mark 4-4.5, “good” with median mark 3-4, “poor” with median mark 2-3, and “insufficient” with median mark lower than 2.
The primary hypothesis tested was that the behavioral scales in generall, and school success would improve in children with ED during the active AEM.
In our crossover study, the patients acted as their own controls. The changes in the global rating of school success were analyzed by a repeated-measurement multivariate analysis of variance (MANOVA) between the same children in different periods of the study. To identify the most relevant subscale, a univariate test was used. A p value of <0.05 was considered significant. All statistical tests were two-tailed.
The analysis was by intention to treat.
Results: All of the children had ED at baseline, 29 (80.5%) unilateral and 7 (19.5%) bilateral with some intent to secondary generalization. In all 30 of the participants that were followed, there was no improvement in the school success and behavior (CRS) during the first year without introducing an AEM according to non-pharmacological treatment and psychomotor education. All of the 30 children had persisting ED after the follow-up period of one year without AEM.
Twenty participants got KBZ, and teen received VPA.
At the 24-month follow-up, twelve (40%) children had a reduced frequency, 12 (40%) had a reduced duration, ten (33.3%) had a reduced frequency and reduced duration of the FRED, whereas 10 (33.3%) had no change of ED and 6 (20%) had no ED. Six children (20%), all without ED, had a clear bettering in school success increasing mark score for 0,67 (range 0.2-1.2). The difference between the frequency and or the duration of persisting ED separately for KBM and VPA was not significant (X2 =2.5, df =1, not significant (NS)). There was no difference in the global rating of behavior (combining parent’s and teacher’s CRS) when comparing KBZ and VPA for the total group of patients (MANOVA: F = 0.78; df = 14; NS). A significant improvement in the global rating of behavior in the children without ED (P<0.05) was found subscalw V/ learning problem.
Children without whose bettering of school success parents have not been pleased have been switched to LTG (18 with unilateral ED, and 6 with bilateral ED). The children with an improvement in school success, without ED (6), continued with KBZ. All of them were followed for the next 12 months. After a 36-month follow-up period, 23 (76.7%) children with unilateral ED (6 on CBZ and 17 on LTG) had no ED. Six (20%) children with bilateral ED, as well as one (3.3%) with unilateral ED still had them on LTG. The children with unilateral ED and LTG medication were more likely to show improvements in school success (for 0.57 increased median mark (range 0,3-0,9)) after ED were suppressed, than patients with bilateral ED (for 0.43 increased median mark (range 0.2-0,87)). Patients on LTG were more likely to have an improvement in school succes if ED were suppressed than patients on KBZ and VPA (MANOVA: F = 2.08; df = 14; P<0.001). There was no other effect caused by the random assignment (MANOVA: F=1.11; df = 14; NS).
A search for studies on poor school success coexisting with epileptiform discharges in single observations and uncontrolled reports claim an improvement of cognitive functioning by suppressing discharges with AEM in patients with epilepsy (10,11,12).
In our study, ED disappearing in children without ES was associated with a global rating of school success improve. The controlled study design, with standardized behavioral questionnaires, general school success scoring throug median mark and an appropriate number of patients, avoided methodological pitfalls.
The benefit of pharmacotherapy for treating the neuropsychiatric symptoms in children with ED but without overt ES remains to be clarified (13). Our results are concordant with other studies. In a double- blind crossover study of a possible relationship between the clinical responses and changes of the amount of epileptiform activity in EEG, it was shown that school success improved during AEM.
LTG depressed seizures in about half of the patients studied, but gave improvements of behavior in all of the patients (14). The relationship between the suppression of spike-and-wave episodes and the improvements in school success in some patients who take KBZ and/or LTG is not yet completely clear. It could be a result of a reduction of epileptiform activity, a direct mood-elevating effect of AEM, or a combination of these effects.
There are also reports that VPA had a cognitive enhancing effect, probably by reducing epileptiform discharges, but we did not prove this effect on school success (15). Ronen aimed to determine whether VPA improves the cognitive performance and behavior in children with learning and behavioral problems associated with electrographic ED but without clinical seizures in a randomized, double-blind, single-crossover trial on eight participants with different learning and BP (16). The participants underwent neuropsychological testing under video EEG and the parent and teacher behavior checklist during each treatment phase. Clinically, none of the children improved on VPA. On formal testing, the children were more distractible, had an increased delay in the response time, and showed lower memory scores while on VPA. In addition, the parents reported higher internalizing scores on the parent and teacher behavior check list CBCL while the children were on VPA.(16) Our data do not support the use of VPA in children with ED and poor school success.
In our study, the patients taking KBZ were also more likely to show an improvement when the ED were suppressed. This could be due to the choice of first-line AEDs for focal epileptic discharges rather than due to an independent effect (17). We found a significant behavioral improvement in the group using LTG in the patients with suppressed ED. LTG does not appear to adversely affect cognition in epileptic patients. In a recent study, low-dose lamotrigine had a positive effect on reaction time measurements and on one of six mood scales in healthy volunteers; however, the number of volunteers tested was small (18,19). Furthermore, several uncontrolled studies reported improved cognition and behavior, without clinically significant cognitive effects in adjunctive therapy for children with epilepsy (20).
Epileptiform discharges may cause psychosocial disturbances by directly interacting with cognitive and behavioral function. With the use of electroencephalography- linked cognitive tests, TCI has been found in up to 50% of patients investigated. Our results provide evidence that ED particularly may play a role in the underlying mechanisms of learning process. Generalized bursts lasting at least three seconds are most likely to produce demonstrable TCI, but they can also be found during briefer, focal ED.
TCI may impair day-to-day psychosocial function. It is well established that children with focal EEG abnormalities or complex partial seizures are particularly vulnerable to psychiatric and behavioral disturbances (8). If discharges are causing cognitive changes such as TCI and behavioral changes, they are, strictly speaking, neither subclinical nor interictal.
Marston et al. administered AEM to reduce subclinical discharges with the goal of improving psychosocial functioning in 10 children with TCI (21).
In eight of them, the psychosocial functioning was improved in conjunction with the reduced interictal ED. Binnie hypothesizes that episodes of TCI can adversely affect interpersonal interactions and social functioning if it should cause the child to miss important social or emotional cues during interactions with peers, or if the child experiences interruptions in the flow of conversations, leading to failure to respond appropriately.
Table 2. Patient characteristics
The increased rates of attention problems found in this study also indirectly lend support for the hypothesis that transient cognitive impairment could account for the increased BP found in children with epilepsy (22). Limitations of the Marston study, however, were the small sample size and the global measurement of psychosocial functioning (21). In contrast, epileptiform discharges were not found to be related to BP in children from the general population (23).
In clinical research, behavioral scales, such as the CRS or the Achenbach Child Behavior Checklist, are often used to compare t scores quantitatively before and after intervention rather than qualitatively defining what is a normal or abnormal score (24,25). This may obtain a significant result that may or may not be clinically relevant, but this problem is inherent to all research using behavioral scales in children with mild problems. Conclusion: The benefit of AEM for children without epilepsy with poor school success and epileptic discharges must be clarified. Our study showed that poor school success can be related to the presence of the ED, and if so, AEM is justified, with the best effect on ED suppression and school success in our group during using LTG.
- Besag FM. Behavioral aspects of pediatric epilepsy syndromes. Epilepsy Behav 2004;5(Suppl. 1):S3–S13.
- Keene DL, Manion I, Whiting S, et al. A survey of behavior problems in children with epilepsy. Epilepsy Behav 2005;6:581–6.
- Austin J, Harezlak J, Dunn D, et al. Behavior problems in children before first recognized seizures. Pediatrics 2001;107:115–22.
- Deonna T. Cognitive and Behavioural Correlates of Epileptic Activity in Children. Journal od Child Psychology and psychiatry 2006. Vol 34.Issue 5:611-20.
- Holtmann M Schmidt M H. Behavior Problems in Nonepileptic Children with Rolandic Epileptiform Discharges. Epilepsia 2003. Vol 44, Issue 6:875.
- Rasmussen NH, Hansen LK, Sahlholdt L. Comorbidity in children with epilepsy. I: Behaviour problems, ADHD and intelligence. Ugeskr Laeger. 2007 May 7;169(19):1767-70.
- Pressler R, Robinson R, Wilson G, and Binnie C. Treatment od interictal epileptiform discharges can improve behavior in children with behaviotal problems and epilepsy. J Pediatr 2005;146:112-7.
- Nair R, Ochi A, Benifla M, Rutka J. T, Snead O. C, Otsubo H. Benign epileptiform discharges in Rolandic region with mesial temporal lobe epilepsy: MEG, scalp and intracranial EEG features. Acta Neurologica Scandinavica 2006. Vol 1. Issue 1:59-64.
- Conners CK. Conners’ Rating Scales Manual. Instruments for use with children and adolescents. New York, NY: Multi-Health Systems, Inc; 1989.
- Binnie CD. Cognitive impairment during epileptiform discharges: is it ever justifiable to treat the EEG?. Lancet Neurol 2003;2:725-30.
- Tremmel L, Holtmann M, Schmidt M, Brandl U. Do subclinical epileptiform discharges really affect short-term memory in children. Zeitschrift für Kinder- und Jugendpsychiatrie und Psychotherapie 2006;34(2):139-48.
- Sabbagh S, Soria C, Escolano S, Bulteau C, Dellatolas G. Impact of epilepsy characteristics and behavioral problems on school placement in children Epilepsy & Behavior 9 (2006) 573–578.
- Holtman M, Becker K, el/Faddagh M, Shmidt MN. Typical benign epilepsy potentials in childhood (Rolandic spikes), neurobiological and neuropsychological symptoms and their clinical significance in child and adolescent psychiatry. Z Kinder Jugendpsychiatr Psychother. 2004 May;32(2):117-29
- Ramachandran Nair, Ochi A, Benifla M, et al. Benign epileptiform discharges in Rolandic region with mesial temporal lobe epilepsy: MEG, scalp and intracranial EEG features. Acta Neurologica Scandinavica 2006. Vol 1. Issue 1:59-64.
- Gordon, K.; Bawden, H.; Camfield, P.; Mann, S., and Orlik, P. Valproic acid treatment of learning disorder and severely epileptiform EEG without clinical seizures. J Child Neurol. 1996 Jan; 11(1):41-3.
- Ronen, G. M.; Richards, J. E.; Cunningham, C.; Secord, M., and Rosenbloom, D. Can sodium valproate improve learning in children with epileptiform bursts but without clinical seizures? Dev Med Child Neurol. 2000 Nov;42(11)71-5.
- Bourgeois BF. Antiepileptic drugs, learning, and behavior in childhood epilepsy. Epilepsia.1998;39:913-21.
- Eriksson AS, Knutsson E, Nergardh A. The effect of la motrigine on epileptiform discharges in young patients with drug-resistant epilepsy. Epilepsia 2001;42:230-6.
- Aldenkamp AP, Arends J, Bootsma HP, et al. Randomized double-blind parallel-group study comparing cognitive effects of a low-dose lamotrigine with valproate and placebo in healthy volunteers. Epilepsia 2002;43:19-26.
- Pressler M, Binnie C, Coleshill S, Chorley G, and Robinson R. Effect of lamotrigine on cognition in children with epilepsy.Neurology 2006;66:1495-1499
- Marston D, Besag F, Binnie CD, Fowler M Effects of transitory cognitive impairment on psychosocial functions of children with epilepsy: a therapeutic trial. Dev Med Child Neurol. 1993;35:574-81.
- Binnie CD Significance and management of transitory cognitive impairment due to subclinical EEG discharges in children. Brain Dev 1992;15:23-30.
- Okubo Y, Matsuura M, Asai T, et al. Epileptiform EEG discharges in healthy children: prevalence, emotional and behavioral corelates, and genetic influences. Epilepsia 1994;35;832-4.
- Weglage J, Demsky A, Pietsch M, Kurlemann G. Neuropsychological, intellectual, and behavioral findings in patients with centrotemporal spikes with and without seizures. Dev Med Child Neurol. 1997;39:646-51.
- Nicolai J, Albert P. Aldenkamp A, et all. Cognitive and behavioral effects of nocturnal epileptiform discharges in children with benign childhood epilepsy with centrotemporal spikes. Epilepsy and Behavior, Vol 8, Issue 1, 2006:56-70.