Seizure (2006) 15, 221—226

www.elsevier.com/locate/yseiz

Cognitive effects of seizures

Guy Vingerhoets *

Laboratory for Neuropsychology, Department of Internal Medicine-Section Neurology, Ghent University,De Pintelaan 185-4K3, B-9000 Ghent, Belgium

Received 24 January 2006; accepted 15 February 2006

KEYWORDSCognition;Seizures;Epilepsy;Neuropsychology;Intelligence

Summary We aimed to review recent prospective and cross-sectional studiesregarding the gradual and chronic effects of (cumulative) seizures on cognition. Incontrast with the increasing evidence of structural changes in the brain associatedwith repeated seizures, its functional repercussions remain unclear. Methodologicaldifficulties of cross-sectional and prospective studies are addressed. It appears thatall but one of the prospective studies available on children are limited to measures ofintelligence. Most studies revealed no significant adverse effects, although thereappears to be a subgroup of about 10—25% of children that shows a clinicallysignificant intellectual decline. Children with generalized symptomatic epilepsies,frequent seizures, high antiepileptic drug use, and early onset of epilepsy appear atrisk, although psychosocial factors may also play an important role. Five of the sixprospective studies on adults report evidence of amild decline in cognition in patientswith a (longstanding) history of pharmacoresistant epilepsy. The adverse effect oncognitive abilities, memory in particular, seems somewhat more robust than that onmeasures of intelligence. A significant association between cognitive decline andseizure related variables is rarely substantiated in prospective research and cross-sectional studies show contradicting results. Taken together, the data suggest a mildbut measurable decline of intellectual performance in children and adults. Decline ofspecific cognitive abilities in children is impossible to evaluate given the very littledata available. In adults, memory appears to be the most vulnerable cognitivefunction. Due to many confounding variables, the effect of seizures per se is difficultto estimate, but appears limited.# 2006 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.

Cognitive effects of seizures

Patients with epilepsy are at significant risk forcognitive impairment and behavioral abnormal-

* Tel.: +32 9 240 45 87; fax: +32 9 240 45 55.E-mail address: guy.vingerhoets@ugent.be.

1059-1311/$ — see front matter # 2006 British Epilepsy Associationdoi:10.1016/j.seizure.2006.02.012

ities.1 Although it appears safe to say that the likelyreason for the cognitive impairment is the neuro-pathology underlying the epilepsy, it remainsunclear whether one of its clinical manifestations,i.e. seizures, can cause cognitive decline per se.By cognitive decline we mean the gradual andchronic loss of mental abilities over time, not the

. Published by Elsevier Ltd. All rights reserved.

222 G. Vingerhoets

reversible postictal cognitive deterioration that canbe observed shortly after focal and generalizedseizures.2

This apparently simple question has provenextraordinarily complex, and the problem hasbeen addressed from very different angles withoutproviding a definite answer.3 The aim of this reviewis to offer a short state of affairs on this clinicallyand therapeutically important question by sum-marizing the results of recent prospective andcross-sectional studies in children and adults.The main reason for studying the data on childrenseparately is that there are reasons to believethat the immature brain may react differently tothe long-term effects of seizures than does themature brain.1,4

Do seizures cause brain damage?

One of the first questions that come to mind iswhether there is evidence that seizures can cause(permanent) brain damage that could be responsi-ble for the cognitive decline. The substantial experi-mental research on chronic rodent models ofepilepsy suggests that the brain’s vulnerability toseizure-induced injury is age-specific: whereas theimmature brain is highly prone to develop seizures,it is more resistant to seizure induced damage thanthe adult brain.4 Although experimental research, inwhich the animals can be sacrificed to investigatethe effects of seizures on the developing brain, hasshown that seizure induced damage appears lessprominent in the immature nervous system, thereis an abundance of evidence that there are signifi-cant long-term adverse consequences of earlylife seizures. Kindled seizures induce progressivecellular and metabolic alterations correlated with(hippocampal) neuronal loss, neoneurogenesis andsynaptic reorganization, increased susceptibility toevoked and spontaneous seizures, and behavioraland cognitive deficits that worsen as a function ofthe cumulative number of seizures.3—8

Although a similar relation might be expected inhumans, it has proven much more difficult to sub-stantiate this, and different reviews of the availabledata have come to somewhat different conclusions.Some underline that human cross-sectional morpho-logical and biochemical studies on brain damagefollowing status epilepticus or brief seizures showinconsistent results, that prospective data are lack-ing, and that the functional significance of thestructural and biochemical changes is unclear.4

Others argue that given the emerging data fromhuman magnetic resonance imaging and neuropsy-chological studies, patients can no longer be reas-

sured with confidence that only prolonged seizures,as in status epilepticus, can cause brain damage,whereas repeated brief seizures do not.8 In additionto the neuropathologically documented hippocam-pal neuronal loss in the majority of patients withtemporal lobe epilepsy, magnetic resonance ima-ging revealed more widespread associated neuronaldamage in extrahippocampal regions. Repeatedmagnetic resonance imaging showed progressivehippocampal reduction following repeated seizuresin several case-studies,9,10 and an associationbetween seizure frequency and hippocampal volumeloss in prospective group studies,11,12 although this isnot always confirmed.13—15 A recent study compared37 patients with childhood-onset versus 16 patientswith late-onset temporal epilepsy and 62 healthycontrols using magnetic resonance imaging volu-metrics and neurocognitive assessment.16 A substan-tial reduction in brain volume, especially evident inwhite matter tissue, and extending to extra-tem-poral regions was found among the childhood-onsetpatients compared to the late-onset patients andhealthy controls. This reduction was also associatedwith significantly poorer cognitive status. It appearsfrom these findings that the presence of recurrentseizures in the developing human brain is associatedwith an adverse and widespread neurodevelopmen-tal impact on both brain structure and function ifassessed after a longstanding disease history in thematured brain.

Methodological issues

Before going into the recent data on the relationbetween seizures and cognition, it is relevant topoint out the methodological difficulties that areassociated with this question. In patients with loca-lization-related symptomatic epilepsy it is difficultto disentangle the effects of epilepsy as a symptomfrom those of the underlying lesion causing theepilepsy. This covariance provides a major disad-vantage for cross-sectional studies to address theeffect of a single variable (i.e. seizures) on cogni-tion. Measuring a ‘state’ rather than a ‘change’,cross-sectional research cannot differentiatebetween the effect of the initial underlying lesionand the long-term effects of the disease. As a result,they tend to overestimate the effects of seizures.Patients with idiopathic epilepsy might provide abetter model to study the effects of seizures in theabsence of a confounding identifiable cerebral dis-ease. However, these patients usually respond wellto anticonvulsive treatment and many are onlymildly cognitively impaired. The few cross-sectionalneuropsychological studies on idiopathic epilepsy

Cognitive effects of seizures 223

failed to find a relation between cognitive impair-ment and seizure frequency, although a significantrelation with interictal epileptiform activity wasobserved.17 Some cross-sectional studies, however,were able to document the contribution of seizurefrequency either statistically or by strategic com-parison of patient subgroups.

Prospective studies may be better able to objec-tivate intra-individual changes over time and tocorrelate these changeswith specific epileptic vari-ables, but they too are subject to confounders thatinterfere with the interpretation. Duration ofepilepsy is a composite factor that reflects theinfluence of several factors in combination, therelative explanatory power of each remaining tobe clarified.18,19 The longer one suffers fromintractable epilepsy, the higher the (cumulative)effect of the seizures, the more prolonged theexposure to interictal discharges or anticonvulsantdrugs, and the greater the risk of seizure relatedclosed-head injuries. In addition, seizure frequencyitself is often related to increased or multiple anti-epileptic drug dosage and uncontrolled seizuresburden the psychological and psychosocial wellbeing of the patient which also negatively affectscognitive performance.20 These interrelations mayonly be disentangled in multiple longitudinal stu-dies with repeated assessments in carefullyselected homogeneous groups of patients. And eventhen, homogeneous epilepsy syndromes may pre-sent with different cognitive profiles dependent onthe age at onset of epilepsy.1 A specific issue inlongitudinal research with repeated measurementsis the problem of test—retest variation. In normals,repeated testing gives rise to improved cognitiveperformance that is attributed to learning. Themagnitude of this test—retest effect is dependenton the number of repetitions, the time intervalbetween testing, the use of alternative test ver-sions, and the cognitive function under investiga-tion. The absence of such a test—retest effect inpatient groups is often interpreted as a reducedlearning potential and an argument for cognitiveimpairment. The use of a matched control groupthat undergoes exactly the same procedure is theonly valid way to assess the magnitude and rele-vance of a (absent) test—retest effect.

Cognitive consequences of seizures inchildren

In a recent review, Dodrill21 listed nine longitudinalstudies done with children. Five of these studieswere published prior to 1940. They used relativelyshort test—retest intervals of 1 or 2 years, and were

all using older measures of intelligence. In addition,information on seizure frequency was either com-pletely lacking or unclear. We will not include theseolder studies here, but will focus on the four morerecent studies in Dodrill’s review and on threeadditional prospective studies that are of interest.

Bourgeois et al.22 investigated 72 children (meanage: 7.5 years) who suffered from active seizuredisorders over a 4-year interval with age-appropri-ate IQ tests. No group changes in IQ were detected,but a subgroup (11%) with frequent seizures, toxicdrug levels, and early onset of epilepsy showed adecline of more than 10 IQ points. Ellenberg et al.23

tested 83 children between 4 and 7 years old withrelatively few seizures with age-appropriate IQ testsover a test—retest period of 3 years. No losses inintelligence were detected. Aldenkamp et al.24 per-formed a study on 45 outpatients with an average of9.3 years of age, most of them had active seizures,with a Dutch form of the WISC-R. The average test—retest interval measured 4.2 years and no significantgroup changes were reported although a slightdecrease in IQ scores was observed. In 24% of thechildren however, full-scale IQ losses of more than 9IQ points were registered. Neyens et al.25 comparedthe intellectual development of 11 children withepilepsy aged 7—15 years with 39 normal controls onthree occasions over a 1.5-year period. Compared tocontrols, children with epilepsy already showedlower intelligence levels at the first assessmentand showed significantly less gain in full-scale IQover trials. The reduced gain was more frequent inchildren with a recent onset of seizures than in thosewho had a longer duration of epilepsy. Seizure fre-quency was not associated with differences in full-scale IQ gain. Bjornaes et al.26 tested 17 candidatesfor epilepsy surgery (mean age 10.2 years) on a 3.5years (average) interval prior to surgery with aNorwegian version of the WISC-R. All children hadactive seizures of moderate severity and showed asignificant decrease in performance IQ and full-scaleIQ. The same group published a later study thatreported on the post surgical outcome of 13 of thesechildren.27 Although, on average, the childrenshowed a general decline in full-scale IQ after boththe drug treatment period and the subsequent sur-gery, children who became seizure free after sur-gery showed no further decline in IQ. A recentprospective study repeatedly investigated 42 chil-dren with newly diagnosed idiopathic or cryptogenicepilepsy and 30 healthy gender-matched classmatecontrols over a 3.5-year period using a comprehen-sive neuropsychological battery and behavior check-lists.28 Most of these children became seizure-freewithin the first 2 years after diagnosis, either spon-taneously or with anti-epileptic drug treatment.

224 G. Vingerhoets

After controlling for the possible influence ofrepeating a grade at school, increased susceptibilityto proactive interference, the phenomenon thatprevious learning inhibits new learning of similarmaterial, was the only cognitive variable on whichthe epileptic group performed worse. None of theepilepsy-related variables, including seizure remis-sion or not, had clinically meaningful effects oncognition or behavior. Instead, in the 19% of childrenwith persistent cognitive deficits, poor parenting,unhappy family situations, and prior existence ofbehavioral problems were overrepresented. Theeffect of psychosocial variables on cognition hasbeen corroborated recently.29

As illustrated before, cross-sectional studies faceimportant methodological problems in evaluatingthe cognitive effects of unique variables of theepileptic syndrome. Recently, a number of large-scale cross-sectional studies were published thatlooked at specific variables and may contribute tothe discussion. Nolan et al.30 investigated 169 chil-dren, aged 0—18 years, and vigorously categorizedeach patient according to 6 different epilepsy syn-dromes, with age appropriate intelligence tests.Younger age of onset, higher seizure frequency(especially if daily), and the intake of more thantwo antiepileptic drugs significantly were asso-ciated with a lower full-scale IQ and contributedto 26% of its variance. After accounting for thesecovariates, significant differences in full-scale IQwere found between the syndrome groups, withbest performances for the generalized idiopathicepilepsy group, the central epilepsy group, andthe temporal lobe epilepsy group. Children withgeneralized symptomatic epilepsy performed mostpoorly and children with either non-localized partialepilepsy or frontal lobe epilepsy performed inbetween. Every group’s average full-scale IQ fellbelow the age-adjusted normedmean. These resultsseem to indicate that seizures have an impact, andthat syndromal type can be used as a separatevariable when considering intellectual potential.Similar findings on type of epilepsy have beenreported in a recent study with educational under-achievement being more prominent in children withlocalized and symptomatic generalized epilepsies,suggesting a dominant impact of the underlyingbrain dysfunction.31 Hoie et al.32 compared non-verbal intelligence in 183 children between 6 and12 years old with 126 healthy controls matched forage and gender and related these data with seizurerelated factors. Severe non-verbal intelligence pro-blems were present in 43% of the patients and in 3%of the controls. The problems were especially com-mon in children with symptomatic epilepsy, unde-termined epilepsy syndromes, myoclonic seizures,

early seizure debut, high seizure frequency, and inchildren with polytherapy. Caplan et al.33 investi-gated the effect of seizure related factors on intel-lectual and linguistic abilities in 101 children withcomplex partial seizures and 102 normal childrenbetween 5 and 16 years old. Interestingly, principalcomponent analysis of the seizure related variablesrevealed four components: a duration/onset com-ponent, an EEG component, a prolonged seizures/febrile convulsions component, and a seizure fre-quency/number of antiepileptic drugs component.Patients showed significantly lower IQ scores andlinguistic abilities than controls. The componentsprolonged seizures/febrile convulsions and seizurefrequency/number of antiepileptic drugs both con-tributed significantly to the IQs’ observed variance.Previous research in a comparable group showedthat frequency of seizure activity in the past year,rather than age at seizure onset, was the bestpredictor for behavioral difficulties.34 Riva et al.35

assessed eight children (aged 6—13 years) withfrontal epilepsy and regular seizures on IQ and awide range of frontal tests. Although IQ was gen-erally spared, the children showed a great variety ofexecutive dysfunctions. Frequent seizures corre-lated with difficulties in attention and the inabilityto inhibit impulsive responses.

Cognitive consequences of seizures inadults

Dodrill’s review of prospective investigations lists 13studies on adults.21 We will focus on the only fivestudies that offered precise data on seizure typesand frequencies and included patients who actuallysuffered from uncontrolled seizures. In addition, wewill present one recently published longitudinalstudy. Dodrill and Wilensky36 compared ninepatients with a history of status epilepticus beforeand after a 5-year study period with nine matchedcontrols with no status on an intelligence test. Full-scale IQ improved significantly more in the non-status group than in the status group. Selwaet al.37 investigated 28 patients with temporal lobeepilepsy and high seizure frequency on an intelli-gence and memory scale at intervals ranging from 1to 8 years. A normal improvement in full-scale andperformance IQ was noted and no decline in memoryperformance was found. Holmes et al.38 assessed 35patients with high frequent intractable seizuresover a 10-year period with an intelligence test anda neuropsychological battery. Whereas IQ showedno change, 6 out of 17 neuropsychological variablesshowed subtle but significant losses, especially mea-sures of visual memory, attention, problem solving,

Cognitive effects of seizures 225

and perception. Dodrill39 compared 35 patientswith detailed estimates of seizure frequency withor without secondary generalization with 35healthy controls over a 10-year period with anintelligence test and a neuropsychological battery.He found less improvement in the epilepsy group,but the frequency of partial seizures did not corre-latewith changes in performance. Twoout of 20 testvariables correlated with the number of general-ized seizures, and the 4 patients that experienced astatus in the course of the study period sufferedvisual and verbal memory loss. Helmstaedteret al.40 investigated 102 adults with temporal lobeepilepsy on measures of memory over an averageperiod of 57 months. Twelve percent of thesepatients were seizure free at the follow-up testing.Fifty percent of the patients showed a significantdecline in verbal and figural memory function. In arecent study, Andersson-Roswall et al.41 compared36 patients with partial epilepsy and 30 healthycontrols over a 3—4-year period on measures ofintelligence and verbal and visual memory. IQscores were significantly lower in the patient groupat baseline, but showed no significant changes atfollow-up, whereas the controls showed increasedscores at retesting. Verbal memory capacity andretention of visuospatial material was also lower inthe patients at baseline. At follow-up a decline inverbal memory was observed in the patients, whichwas absent in the controls. An association withseizure related variables was not found.

Recent cross-sectional neuropsychological stu-dies that compared chronic epilepsy patients withhealthy controls have documented impaired perfor-mances of memory, language, executive function,and motor speed.18,42 In some studies, the degree ofneuropsychological morbidity was correlated withthe duration of epilepsy,18 in others this relation wasnot confirmed.42,43

Conclusions

It appears that all but one of the prospective studiesavailable on children are limited to measures ofintelligence. Since IQ-tests were not designed toinvestigate brain—behavior relationships, thesemeasures may underestimate changes in a broaderrange of cognitive functions. Only in the presurgicalgroup with refractory epilepsy26 a significantdecline of group level IQ was noted. The otherstudies revealed no significant adverse effects,although there appears to be a subgroup of about10—25% of children that shows a clinically significantintellectual or cognitive decline. This subgroup con-tains more children with frequent seizures, high

antiepileptic drug use, and early onset of epilepsy,although psychosocial factors may also play animportant role. As expected, cross-sectional studiesreveal more robust findings of intellectual and cog-nitive impairment. Children with generalized symp-tomatic epilepsies appear to be a high risk group forintellectual and educational underachievement.

Five of the six prospective studies on adults reportevidence of a mild decline in cognition in patientswith a (longstanding) history of pharmacoresistantepilepsy. The adverse effect on cognitive abilities,memory in particular, seems somewhat more robustthan that on measures of intelligence. A significantassociation between cognitive decline and seizurerelated variables, however, is rarely substantiatedin prospective research. Cross-sectional studiesreport a wider variety of neurocognitive deficits,but show contradicting results regarding the influ-ence of seizure related variables.

Taken together, the data suggest a mild but mea-surable decline of intellectual performance in chil-dren and adults. Decline of specific cognitiveabilities in children is impossible to evaluate giventhe very little data available. In adults, memoryappears to be the most vulnerable cognitive func-tion. Due to many confounding variables, the effectof seizures per se is difficult to estimate, butappears limited. Future prospective research shouldtake an effort in accurately describing seizure typeand frequency, and by comparing groups of differentepilepsy syndromes.

References

1. Elger CE, Helmstaedter C, Kurthen M. Chronic epilepsy andcognition. Lancet (Neurol) 2004;3:663—72.

2. Helmstaedter C, Elger CE, Lendt M. Postictal courses ofcognitive deficits in focal epilepsies. Epilepsia 1994;35:1073—8.

3. Sutula T, Pitkanen A. Do Seizures damage the Brain? ProgBrain Res 2002;135:1—520.

4. Haut SR, Veliskova J, Moshe SL. Susceptibility of immatureand adult brains to seizure effects. Lancet (Neurol) 2004;3:608—17.

5. Najm IM,Wang I, Shedid D, Luders HO, Ng TC, Comair YG. MRSmetabolic markers of seizures and seizure-induced neuronaldamage. Epilepsia 1998;39:244—50.

6. Lukoyanov NV, Sa MJ, Madeira MD, Paula-Barbosa MM. Selec-tive loss of hilar neurons and impairment of initial learning inrats after repeated administration of electroconvulsive shockseizures. Exp Brain Res 2004;154:192—200.

7. Pitkanen A, Sutula TS. Is epilepsy a progressive disorder?Prospects for new therapeutic approaches in temporal lobeepilepsy. Lancet 2002;1:173—81.

8. Sutula TP, Hagen J, Pitkanen A. Do epileptic seizures damagethe brain? Curr Opin Neurol 2003;16:189—95.

9. Briellmann R, Newton M, Wellard R, et al. Hippocampalsclerosis following brief generalized seizures in aduldhood.Neurology 2001;57:315—7.

226 G. Vingerhoets

10. Worrell G, Seccakova M, Jack C, et al. Rapidly progressivehippocampal atrophy: evidence for a seizure-inducedmechanism. Neurology 2002;58:1553—6.

11. Briellmann R, Berkovic S, Syngeniotis A, et al. Seizure-asso-ciated hippocampal volume loss: a longitudinal magneticresonance study of temporal lobe epilepsy. Ann Neurol2002;51:641—4.

12. Fuerst D, Shah J, Shah A, Watson C. Hippocampal sclerosis is aprogressive disorder: a longitudinal volumetric MRI study.Ann Neurol 2003;53:413—6.

13. Liu R, Lemieux L, Bell G, et al. The structural consequencesof newly diagnosed seizures. Ann Neurol 2002;52:573—80.

14. Holtkamp M, Schuchmann S, Gottschalk S, Meierkord H.Recurrent seizures do not cause hippocampal damage. JNeurol 2004;251(4):458—63.

15. Briellmann RS, Wellard RM, Jackson GD. Seizure-associatedabnormalities in epilepsy: evidence from MR imaging. Epi-lepsia 2005;46(5):760—6.

16. Hermann B, Seidenberg M, Bell B, Rutecki P, Sheth R, RugglesK, et al. The neurodevelopmental impact of childhood-onsettemporal lobe epilepsy on brain structure and function.Epilepsia 2002;43(9):1062—71.

17. Weglage J, Demsky A, Pietsch M, Kurlemann G. Neuropsycho-logical, intellectual, and behavioral findings in patients withcentrotemporal spikes with and without seizures. Dev MedChild Neurol 1997;39:646—51.

18. Oyegbile TO, Dow C, Jones J, Bell B, Rutecki P, Sheth R, et al.The nature and course of neuropsychological morbidity inchronic temporal lobe epilepsy. Neurology 2004;62:1736—42.

19. Jokeit H, Ebner A. Long term effects of refractory temporallobe epilepsy on cognitive abilities: a cross sectional study. JNeurol Neurosurg Psychiatry 1999;67:44—50.

20. Paradiso S, Hermann BP, Blumer D, Davies K, Robinson RG.Impact of depressed mood on neuropsychological status intemporal lobe epilepsy. J Neurol Neurosurg Psychiatry2001;70(2):180—5.

21. Dodrill CB. Neuropsychological effects of seizures. EpilepsyBehav 2004;5:S21—4.

22. Bourgeois BFD, Prensky AL, Palkes HS, Talent BK, Buxch SG.Intelligence in epilepsy: a prospective study in children. AnnNeurol 1983;14:438—44.

23. Ellenberg JH, Hirts DG, Nelson KB. Do seizures in childrencause intellectual deterioration? N Engl J Med 1986;314:1085—8.

24. Aldenkamp AP, Alpherts WCJ, Bruine-Seeder DD, Dekker MJA.Test—retest variability in children with epilepsy: a compar-ison of WISC-R profiles. Epilepsy Res 1990;7:165—72.

25. Neyens LG, Aldenkamp AP, Meinardi HM. Prospective follow-up of intellectual development in children with a recentonset of epilepsy. Epilepsy Res 1999;34:85—90.

26. Bjornaes H, Stabell K, Henriksen O, Loyning Y. The effects ofrefractory epilepsy on intellectual functioning in childrenand adults: a longitudinal study. Seizure 2001;10:250—9.

27. Bjornaes H, Stabell KE, Henriksen O, Roste G, My Diep L.Surgical versus medical treatment for severe epilepsy: con-sequences for intellectual functioning in children and adults.A follow-up study. Seizure 2002;11:473—82.

28. Oostrom KJ, Van Teeseling H, Smeets-Schouten A, PetersACB, Jennekens-Schinkel A. Three to four years after diag-nosis: cognition and behaviour in children with ‘epilepsy

only’. A prospective, controlled study. Brain 2005;128:1546—55.

29. Fastenau PS, Shen JZ, Dunn DW, Perkins SM, Hermann BP,Austin JK. Neuropsychological predictors of academic under-achievement in pediatric epilepsy: Moderating roles of demo-graphic, seizure, and psychosocial variables. Epilepsia2004;45(10):1261—72.

30. Nolan MA, Redoblado MA, Lah S, Sabaz M, Lawson JA, Cun-ningham AM, et al. Intelligence in childhood epilepsy syn-dromes. Epilepsy Res 2003;53:139—50.

31. Aldenkamp AP, Weber B, Overweg-Plandsoen WCG, Reijs R,van Mil S. Educational underachievement in children withepilepsy: a model to predict the effects of epilepsy oneducational achievement. J Child Neurol 2005;20(3):175—80.

32. Hoie B, Mykletun A, Sommerfelt K, Bjornaes H, Skeidsvol H,Waaler PE. Seizure-related factors and non-verbal intelli-gence in children with epilepsy. A population-based studyfrom Western Norway. Seizure 2005;14:223—31.

33. Caplan R, Siddarth P, Gurbani S, Ott D, Sankar R, Shields WD.Psychopathology and pediatric complex partial seizures:seizure-related, cognitive, and linguistic variables. Epilepsia2004;45(10):1273—81.

34. Schoenfeld J, Seidenberg M, Woodard A, Hecox K, Inglese C,Mack K, et al. Neuropsychologcial and behavioral status ofchildren with complex partial seizures. Dev Med Child Neurol1999;37:159—67.

35. Riva D, Saletti V, Nichelli F, Bulgheroni S. Neuropsychologiceffects of frontal lobe epilepsy in children. J Child Neurol2002;17(9):661—7.

36. Dodrill CB, Wilensky AJ. Intellectual impairment as an out-come of status epilepticus. Neurology 1990;40(Suppl. 2):23—7.

37. Selwa LM, Berent S, Giordani B, Henry TR, Buchtel HA, RossDA. Serial cognitive testing in temporal lobe epilepsy: long-itudinal changes with medical and surgical therapies. Epi-lepsia 1994;35:743—9.

38. Holmes MD, Dodrill CB, Wilkus RJ, Ojemann LM, Ojemann GA.Is partial epilepsy progressive? Ten-year follow-up of EEG andneuropsychological changes in adults with partial seizures.Epilepsia 1998;39:1189—93.

39. Dodrill CB. Progressive cognitive decline in adolescents andadults with epilepsy. In: Sutula T, Pitkanen A, editors. DoSeizures Damage the Brain? Prog Brain Res 2002;135:399—407.

40. Helmstaedter C, Kurthen M, Lux S, Reuber M, Elgers CE.Chronic epilepsy and cognition: a longitudinal study in tem-poral lobe epilepsy. Ann Neurol 2003;54:425—32.

41. Andersson-Roswall L, Engman E, Samuelsson H, Sjoberg-Lars-son C, Malmgren K. Verbal memory decline and adverseeffects on cognition in adult patients with pharmacoresistantpartial epilepsy: a longitudinal controlled study of 36patients. Epilepsy Behav 2004;5:677—86.

42. Martin RC, Griffith HR, Faught E, Gilliam F, Mackey M, VogtleL. Cognitive functioning in community dwelling older adultswith chronic partial epilepsy. Epilepsia 2005;46(2):298—303.

43. Strauss E, Loring D, Chelune G, Hunter M, Hermann B, PerrineK, et al. Predicting cognitive impairment in epilepsy: findingsfrom the Bozeman epilepsy consortium. J Clin Exp Neurop-sychol 1997:S52—5.