Generalidades en el diagnóstico de epilepsia en pediatríakindermed.info/ · ! 10,5 millones de...

Generalidades en el diagnóstico de

epilepsia en pediatría

Dr. Alvaro Retamales M.

Neurólogo Infantil

- Miércoles, 13 de agosto de 2014 -

Temario

¤  1. Definiciones y clasificación

¤  2. Semiología

¤  3. Exámenes complementarios

¤  4. Diagnósticos diferenciales

Epilepsia y crisis epilépticas

¤ Epilepsia:

¤  Trastorno cerebral caracterizado por:

¤  Predisposición a generar crisis epilépticas

¤  Las consecuencias neurobiológicas,

cognitivas, psicológicas y sociales derivadas

de ella.

¤  Requiere al menos 1 crisis epiléptica

Epileptic seizures and epilepsy Definitions proposed by the International League Against Epilepsy (ILAE) and the International Bureau for Epilepsy. Epilepsia (2005)

¤  10,5 millones de niños <15

años tienen epilepsia.

¤  80%de pacientes con

epilepsia son menores de 15

años.

¤ Aproximadamente un 60% de

los niños con epilepsia en la

niñez dejarán de padecerla

en la adultez.

¤ 6 a 7% de los niños

menores de 15 años

presentarán una crisis a lo

largo de su vida

(provocada o no)

¤  Aumenta a 7 – 12% en

países con bajo ingreso.

¤ Crisis epiléptica:

¤  Episodio causado por una

descarga anormal, excesiva y

sincrónica de un grupo de

neuronas que produce una

alteración en la función del

paciente.

Epileptic seizures and epilepsy Definitions proposed by the International League Against Epilepsy (ILAE) and the International Bureau for Epilepsy. Epilepsia (2005)

¤ Otros conceptos:

¤  Epilepsia es una ENFERMEDAD.

¤  Evidencia clara de un síndrome epiléptico, se requiere 1 sola

crisis, aún cuando el riesgo de recurrencia sea bajo.

¤  2 crisis se consideran independientes si tienen >24 hrs

¤  No hay límite de tiempo para una 2da crisis

¤  Epilepsia “resuelta”

¤  Información “imperfecta”: concepto de epilepsia “probable”

Otras definiciones

¤  Síndrome:

¤  Conjunto de síntomas y signos relacionados entre sí, que permiten

realizar una unidad diagnóstica.

¤  Síndrome epiléptico:

¤  Asociaciones particulares de manifestaciones clínicas y

electroencefalográficas frecuentemente edad dependiente, que

permiten:

¤  Orientar estudio etiológico

¤  Definir conducta terapéutica

¤  Definir pronóstico.

Clasificación internacional de crisis epilépticas (ILAE 1981) I. Crisis focales

A. Crisis parciales simples 1. Con signos motores 2. Con síntomas somatosensoriales 3. Con signos y síntomas autonómicos 4. Con síntomas psíquicos

B. Crisis parciales complejas 1. Inicio parcial simple 2. Parcial compleja desde el inicio

C. Crisis parciales secundariamente generalizadas 1. Parciales simples con GS. 2. Parciales complejas con GS. 3. Parcial simple, luego parcial compleja y GS

Clasificación internacional de crisis epilépticas (ILAE 1981) II. Crisis generalizadas

A. Ausencias

B. Mioclónicas

C. Clónicas

D. Tónicas

E. Tónico-clónicas

F. Atónicas

III. Crisis no clasificables.

¤  Etiología

¤  Idiopáticas

¤  Sintomáticas

¤  Criptogénicas

¤  Localización:

¤  Síndromes epilépticos

relacionados con la localización

¤  Síndromes epilépticos

generalizados

¤  Síndromes epilépticos de

localización indeterminada

¤  Síndromes especiales

Clasificación de 1989

Evolución de las clasificaciones

1969  Clasificación de crisis epilépticas y epilepsia

1981  Revisión de la clasificación de crisis

1985  Revisión de la clasificación de epilepsia

1989  Revisión de la clasificación de epilepsia

2001  Propuesta del esquema diagnóstico de las crisis y epilepsias

2001 Glosario de semiología ictal de las crisis

2006  Reporte del core group de ILAE

2010  Revisión de terminología y conceptos de las crisis y epilepsias

2014 Definición práctica de epilepsia

Crisis epilépticas Semiología de las crisis

¤ Cuándo sospecharlas:

¤  Inicio abrupto

¤  Patrón estereotipado

¤  Eventos involuntarios e incontrolables, breves

¤  Fin abrupto y generalmente reconocible

¤  Conductas ictales no propositivas

¤  Compromiso post ictal: amnesia, sueño, alteraciones

neurológicas transitorias.

Crisis epilépticas

¤  Crisis Tónico clónico generalizadas:

¤  Involucran el tronco y probablemente el córtex prefrontal y

ganglios basales.

¤  Inicialmente existen mecanismos desinhibitorios (fase tónica)

con posterior instalación periódica de mecanismos

inhibitorios (fase clónica).

¤  En algunos casos, pueden precederse por mioclonías (crisis

clónico-tónico-clónicas)

Engel J. Report of the ILAE Classification Core Group. Epilepsia, 47(9):1558–1568, 2006

¤  Crisis mioclónicas:

¤  Contracción muscular súbita, breve (<350 ms).

¤  Pueden ser generalizadas, en la cara o en 1 o más

extremidades.

¤  Más breves que las clonías y no son rítmicas.

¤  Ocasionalmente ocasionan caídas (crisis astáticas)

¤  Compromiso de conciencia?

¤  Correlato EEG con actividad espiga o poliespiga onda

generalizada.

¤  Espasmos epilépticos:

¤  Habitualmente en lactantes menores.

¤  Contracción muscular de predominio axial.

¤  2 fases: rápida de contracción y lenta de relajación

¤  Pueden ser en flexión, extensión o mixtos.

¤  Característicamente en salvas con diferencias de <1 minuto.

¤  Predominio al despertar.

¤  Característicos del síndrome de West.

¤  Atónicas:

¤  Pérdida súbita del tono muscular

¤  Pueden provocar caídas

¤  Compromiso de conciencia breve

¤  Pueden asociarse a mioclonías (crisis mioclónico astáticas)

¤  Correlato EEG con actividad espiga onda lenta a 1 o más

Hz o bien con una onda lenta generalizada seguida de

depresión de voltaje.

¤  Ausencias:

¤  Cese brusco y breve de las actividades.

¤  Mirada fija (“perdida”).

¤  Frecuentes automatismos motores (parpadeo, “chupeteo”)

¤  Difícil distinción con crisis parciales complejas (Frontales)

¤  A veces en contexto de síndromes específicos.

¤  Correlato EEG: actividad espiga onda lenta a 3 cps.

Exámenes complementarios

¤  Descrito en humanos por Hans Berger a

partir de 1929.

¤  Se basa en las diferencias de potencial

entre 2 electrodos (canal) ubicados de

manera estandarizada en el cuero

cabelludo.

¤  Actividad fisiológica vs actividad

anormal:

¤  Epileptiforme

¤  No epileptiforme

Electroencefalograma

Por lo tanto, la ubicación del foco en el montaje bipolar está definida por la inversión de fase en 2 canales adyacentes.

¤  No “hace” el diagnóstico de epilepsia.

¤  Primer EEG standard en crisis focales: sensibilidad 50%.

¤  La sensibilidad es de 84% con el 3er EEG y 92% con el 4to.

¤  Otros factores: edad temprana, EEG de sueño, epilepsia generalizada y

registro post crisis.

¤  Espigas focales en regiones inferiores o mesiales pueden no verse en EEG

de superficie.

¤  EEG continuamente normales levantan la sospecha de un TPNE.

¤ Procedimientos de activación aumentan la

sensibilidad:

¤  EFI y HV: aumentan probabilidad de encontrar descargas

generalizadas.

¤  Privación de sueño: aumenta la probabilidad de encontrar

anomalías focales.

La privación de sueño por sí misma es activadora.

¤  Incidencia de 2 a 4% de alteraciones epileptiformes en

pacientes sin epilepsia.

¤  Esto puede aumentar hasta un 25% en pacientes estudiados

por TDAH.

¤  La mayoría presentan actividad epileptiforme de tipo

benigna (alto % centrotemporales).

Frequency of Epileptiform Discharges in the Sleep-Deprived Electroencephalogram in Children Evaluated for Attention-Deficit Disorders. Journal of Child Neurology 26(1) 6-11. 2011.

Seminars in Neurology, Volume 23, Number 1, 2003

Special Article

CME Practice parameter: Evaluating afirst nonfebrile seizure in children

Report of the Quality Standards Subcommittee of theAmerican Academy of Neurology, the Child Neurology

Society, and the American Epilepsy SocietyD. Hirtz, MD; S. Ashwal, MD; A. Berg, PhD; D. Bettis, MD; C. Camfield, MD; P. Camfield, MD;

P. Crumrine, MD; R. Elterman, MD; S. Schneider, MD; and S. Shinnar, MD, PhD

Article abstract—Objective: The Quality Standards Subcommittee of the American Academy of Neurology developspractice parameters as strategies for patient management based on analysis of evidence. For this practice parameter, theauthors reviewed available evidence on evaluation of the first nonfebrile seizure in children in order to make practicerecommendations based on this available evidence. Methods: Multiple searches revealed relevant literature and eacharticle was reviewed, abstracted, and classified. Recommendations were based on a three-tiered scheme of classification ofthe evidence. Results: Routine EEG as part of the diagnostic evaluation was recommended; other studies such aslaboratory evaluations and neuroimaging studies were recommended as based on specific clinical circumstances. Conclu-sions: Further studies are needed using large, well-characterized samples and standardized data collection instruments.Collection of data regarding appropriate timing of evaluations would be important.NEUROLOGY 2000;55:616–623

The Quality Standards Subcommittee (QSS) of theAmerican Academy of Neurology (AAN) seeks to de-velop scientifically sound, clinically relevant practiceparameters for physicians for diagnostic procedures,treatment modalities, and clinical disorders. Practiceparameters are strategies for patient managementthat might include diagnosis, symptom, treatment,or procedure evaluation. They consist of one or morespecific recommendations based on the analysis ofevidence.

Every year, an estimated 25,000 to 40,000 USchildren experience their first nonfebrile seizure, adramatic and frightening event.1-4 This practice pa-rameter reviews available evidence concerning thevalue of diagnostic testing after a first nonfebrileseizure in a child, and provides recommendationsbased on this evidence. It addresses the evaluation ofchildren age 1 month to 21 years who have experi-enced a first nonfebrile seizure that cannot be ex-

plained by an immediate, obvious provoking causesuch as head trauma or intracranial infection. Re-ports concerning serum laboratory studies, CSF ex-amination, EEG, CT, and MRI are reviewed. Thisparameter concerns diagnostic evaluation; a subse-quent parameter will focus on treatment of the firstnonfebrile seizure.

The seizure types covered by this parameter in-clude partial (simple or complex partial, or partialwith secondary generalization), generalized tonic-clonic, or tonic seizures. We are specifically not in-cluding children diagnosed with epilepsy, defined astwo or more seizures without acute provocation. Forthis reason, myoclonic and atonic seizures are ex-cluded because they typically are not recognized un-til there have been multiple occurrences. We definedthe first seizure using the International LeagueAgainst Epilepsy (ILAE) criteria to include multipleseizures within 24 hours with recovery of conscious-

From the National Institute of Neurological Disorders and Stroke (Dr. Hirtz), National Institutes of Health, Bethesda, MD; Department of Pediatrics (Dr.Ashwal), Loma Linda University School of Medicine, Loma Linda, CA; Department of Biological Sciences (Dr. Berg), Northern Illinois University, Dekalb; Dr.David Bettis, Boise, ID; Department of Pediatric Neurology (Drs. C. Camfield and P. Camfield), IW Killam Hospital for Children, Halifax, Nova Scotia;Department of Neurology (Dr. Crumrine), Children’s Hospital of Pittsburgh, PA; Dr. Roy Elterman, Dallas, TX; Dr. Sanford Schneider, Riverside, CA; andthe Departments of Neurology and Pediatrics (Dr. Shinnar), Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, NY.Approved by the Quality Standards Subcommittee April 1, 2000. Approved by the Practice Committee May 3, 2000. Approved by the AAN Board of DirectorsJune 9, 2000.Received September 27, 1999. Accepted in final form June 13, 2000.Address correspondence and reprint requests to QSS, American Academy of Neurology, 1080 Montreal Avenue, St. Paul, MN 55116; phone: 800-879-1960

zure patients.40,41 In these studies, interpretationwas difficult in the presence of diffuse postictal slow-ing,42 and an EEG done at that time was not helpfulin determining which patients should be admitted tothe hospital.40

A recent analysis of selected findings from severalof the Class I studies referred to above25-27,30,31 con-cluded that an EEG should not be routinely per-formed after a first seizure because it does not yieldsufficient information to alter treatment decisions.43

To reach this conclusion, the authors did not con-sider evidence that the EEG result does in fact altertreatment decisions. They assumed a treatmentthreshold to be at an 80% risk of recurrence, andused a univariate analysis. However, where the EEGis used as one of several variables, it can identifychildren with very high and very low recurrencerisks.25,26,32,35 The EEG is not used solely to determinerecurrence, but also helps differentiate a seizurefrom other events, is essential to the diagnosis of asyndrome, and provides information on long-termprognosis; it influences the decision to perform sub-sequent neuroimaging studies44 and may influencecounseling about management of the child.

Conclusions. The majority of evidence from ClassI and Class II studies confirms that an EEG helps indetermination of seizure type, epilepsy syndrome,and risk for recurrence, and therefore may affectfurther management decisions. Experts commonlyrecommend that an EEG be performed after all firstnonfebrile seizures.39,45-47 It is not clear what the op-timal timing should be for obtaining an EEG. Al-though an EEG done within 24 hours of the seizureis most likely to show abnormalities,34 physiciansshould be aware that some abnormalities such aspostictal slowing that can be seen on EEG donewithin 24 to 48 hours of a seizure may be transientand must be interpreted with caution.

There is no evidence that the EEG must be donebefore discharge from the emergency department;the study may be arranged on an outpatient basis.Epileptiform EEG abnormalities may be useful inconfirming that the event was a seizure; however, anEEG abnormality by itself is not sufficient to make adiagnosis that an epileptic seizure occurred, nor canits absence rule out a seizure.46,47 The EEG is neces-sary to determine the epilepsy syndrome and thediagnosis of an epilepsy syndrome may be helpful indetermining the need for imaging studies.34 The EEGis also useful in predicting the prognosis forrecurrences.20,39,45-47

It is not clear what the optimal timing should befor obtaining an EEG. Although an EEG done within24 hours of the seizure is most likely to show abnor-malities, physicians should be aware that some ab-normalities such as postictal slowing that can beseen on EEG done within 24 to 48 hours of a seizuremay be transient and must be interpreted withcaution.

Recommendations.

• The EEG is recommended as part of the neurodi-agnostic evaluation of the child with an apparentfirst unprovoked seizure. (Standard)

Neuroimaging studies. Evidence—CT scans.There were five Class I studies regarding imaging byCT scan after a first seizure; the data pertained tochildren32 and adults24,42,48 with first seizures, and toadults and children over age 6 with both new onsetand established seizures.14 In the single Class Istudy of first seizures in children, the abnormalities(mostly atrophy) found in 12 children were “withouttherapeutic consequences” (95% CI 0, 3%).32 In one ofthe adult studies, 1.3% of the patients who had CTscans were diagnosed with tumors,24 and in another,of 62 patients there were three tumors seen on CT,all in patients with abnormal neurologic examina-tions.42 Of 119 adults who had CT scans after a firstgeneralized seizure, 20 had abnormalities that war-ranted therapeutic intervention.48 In the Class Istudy in which 19 CT scans were done in selectivecases (first seizures if greater than age 6 years, headtrauma, or focal seizure), there was one significantabnormality (age of the patient was not given), asubdural hematoma, not predicted by history andphysical examination.14

Of the 14 Class II studies, nine involved childrenonly (n ! 2559),17,19,49-56 four were of adults only (n !666),24,42,57,58 and one involved children and adults(n ! 109).59 Only a small percentage of children inthese studies (0 to 7%) had lesions on CT that al-tered or influenced management. These were mostcommonly brain tumors, communicating or obstruc-tive hydrocephalus, one subarachnoid and one poren-cephalic cyst, and three children with cysticercosis.The yield of abnormality on CT when the neurologicexamination and EEG were normal was 5 to 10%.50,54

In a Class II study in which seven children (14% ofchildren with nonfebrile seizures) had CT scans thatinfluenced management, five had focal or complexpartial seizures. Abnormalities on neuroimagingwere associated with a higher recurrence risk.54 Inone study of febrile and nonfebrile children, CTscans were always normal in the absence of de-fined risk factors such as known neurologic diagno-sis, age "5 months, or focal deficit.57 Focal lesions onCT scans tended to be more commonly found inadults (18 to 34%)40,48,57,58 than in children (0 to12%),17,32,49,52,54-56 particularly when ordered for spe-cific clinical indications. At least three studies pro-vided evidence that MRI scanning was preferable toCT51,54,60 in children following nonfebrile seizures.

Evidence—MRI. There was one Class I report re-garding MRI in children presenting with a first sei-zure54 and another Class I report of newly diagnosedepilepsy in children.60 Of 411 children who presentedwith a first seizure, 218 had neuroimaging studies.Four had lesions seen on MRI or CT (two brain tu-mors, two neurocysticercosis) that potentially altered

September (1 of 2) 2000 NEUROLOGY 55 619

Special Article

zure patients.40,41 In these studies, interpretationwas difficult in the presence of diffuse postictal slow-ing,42 and an EEG done at that time was not helpfulin determining which patients should be admitted tothe hospital.40

A recent analysis of selected findings from severalof the Class I studies referred to above25-27,30,31 con-cluded that an EEG should not be routinely per-formed after a first seizure because it does not yieldsufficient information to alter treatment decisions.43

To reach this conclusion, the authors did not con-sider evidence that the EEG result does in fact altertreatment decisions. They assumed a treatmentthreshold to be at an 80% risk of recurrence, andused a univariate analysis. However, where the EEGis used as one of several variables, it can identifychildren with very high and very low recurrencerisks.25,26,32,35 The EEG is not used solely to determinerecurrence, but also helps differentiate a seizurefrom other events, is essential to the diagnosis of asyndrome, and provides information on long-termprognosis; it influences the decision to perform sub-sequent neuroimaging studies44 and may influencecounseling about management of the child.

Conclusions. The majority of evidence from ClassI and Class II studies confirms that an EEG helps indetermination of seizure type, epilepsy syndrome,and risk for recurrence, and therefore may affectfurther management decisions. Experts commonlyrecommend that an EEG be performed after all firstnonfebrile seizures.39,45-47 It is not clear what the op-timal timing should be for obtaining an EEG. Al-though an EEG done within 24 hours of the seizureis most likely to show abnormalities,34 physiciansshould be aware that some abnormalities such aspostictal slowing that can be seen on EEG donewithin 24 to 48 hours of a seizure may be transientand must be interpreted with caution.

There is no evidence that the EEG must be donebefore discharge from the emergency department;the study may be arranged on an outpatient basis.Epileptiform EEG abnormalities may be useful inconfirming that the event was a seizure; however, anEEG abnormality by itself is not sufficient to make adiagnosis that an epileptic seizure occurred, nor canits absence rule out a seizure.46,47 The EEG is neces-sary to determine the epilepsy syndrome and thediagnosis of an epilepsy syndrome may be helpful indetermining the need for imaging studies.34 The EEGis also useful in predicting the prognosis forrecurrences.20,39,45-47

It is not clear what the optimal timing should befor obtaining an EEG. Although an EEG done within24 hours of the seizure is most likely to show abnor-malities, physicians should be aware that some ab-normalities such as postictal slowing that can beseen on EEG done within 24 to 48 hours of a seizuremay be transient and must be interpreted withcaution.

Recommendations.

• The EEG is recommended as part of the neurodi-agnostic evaluation of the child with an apparentfirst unprovoked seizure. (Standard)

Neuroimaging studies. Evidence—CT scans.There were five Class I studies regarding imaging byCT scan after a first seizure; the data pertained tochildren32 and adults24,42,48 with first seizures, and toadults and children over age 6 with both new onsetand established seizures.14 In the single Class Istudy of first seizures in children, the abnormalities(mostly atrophy) found in 12 children were “withouttherapeutic consequences” (95% CI 0, 3%).32 In one ofthe adult studies, 1.3% of the patients who had CTscans were diagnosed with tumors,24 and in another,of 62 patients there were three tumors seen on CT,all in patients with abnormal neurologic examina-tions.42 Of 119 adults who had CT scans after a firstgeneralized seizure, 20 had abnormalities that war-ranted therapeutic intervention.48 In the Class Istudy in which 19 CT scans were done in selectivecases (first seizures if greater than age 6 years, headtrauma, or focal seizure), there was one significantabnormality (age of the patient was not given), asubdural hematoma, not predicted by history andphysical examination.14

Of the 14 Class II studies, nine involved childrenonly (n ! 2559),17,19,49-56 four were of adults only (n !666),24,42,57,58 and one involved children and adults(n ! 109).59 Only a small percentage of children inthese studies (0 to 7%) had lesions on CT that al-tered or influenced management. These were mostcommonly brain tumors, communicating or obstruc-tive hydrocephalus, one subarachnoid and one poren-cephalic cyst, and three children with cysticercosis.The yield of abnormality on CT when the neurologicexamination and EEG were normal was 5 to 10%.50,54

In a Class II study in which seven children (14% ofchildren with nonfebrile seizures) had CT scans thatinfluenced management, five had focal or complexpartial seizures. Abnormalities on neuroimagingwere associated with a higher recurrence risk.54 Inone study of febrile and nonfebrile children, CTscans were always normal in the absence of de-fined risk factors such as known neurologic diagno-sis, age "5 months, or focal deficit.57 Focal lesions onCT scans tended to be more commonly found inadults (18 to 34%)40,48,57,58 than in children (0 to12%),17,32,49,52,54-56 particularly when ordered for spe-cific clinical indications. At least three studies pro-vided evidence that MRI scanning was preferable toCT51,54,60 in children following nonfebrile seizures.

Evidence—MRI. There was one Class I report re-garding MRI in children presenting with a first sei-zure54 and another Class I report of newly diagnosedepilepsy in children.60 Of 411 children who presentedwith a first seizure, 218 had neuroimaging studies.Four had lesions seen on MRI or CT (two brain tu-mors, two neurocysticercosis) that potentially altered

¤  Indicadas ante sospecha de epilepsias focales.

¤  Examen neurológico alterado

¤  Epilepsias de muy dificil manejo

¤  Cambios en el patrón habitual de crisis.

¤  Existe:

¤  Resonancia magnética

¤  Tomografía axial computada (TAC)

¤  Otras técnicas

Neuroimágenes

Special Article

on long-term treatment and management.20,22 Fac-tors to be considered include the age of the child, theneed for sedation to perform the study, the EEGresults, a history of head trauma, and other clinicalcircumstances such as a family history of epilepsy.

Recommendations.

• If a neuroimaging study is obtained, MRI is thepreferred modality.50,51,54,60,62,63,65 (Guideline)

Emergent neuroimaging should be performed in achild of any age who exhibits a postictal focal deficit(Todd’s paresis) not quickly resolving, or who has notreturned to baseline within several hours after theseizure.46,66 (Option)

• Nonurgent imaging studies with MRI should beseriously considered in any child with a significantcognitive or motor impairment of unknown etiol-ogy, unexplained abnormalities on neurologic ex-amination, a seizure of partial (focal) onset with orwithout secondary generalization, an EEG thatdoes not represent a benign partial epilepsy ofchildhood or primary generalized epilepsy, or inchildren under 1 year of age.20,34 (Option)

Summary. In the child with a first nonfebrile sei-zure, diagnostic evaluations influence therapeuticdecisions, how families are counseled, and the needfor hospital admission and/or specific follow-upplans. This practice parameter has reviewed thepublished literature concerning the usefulness ofstudies following a first nonfebrile seizure in chil-dren, and has classified the strength of the availableevidence. There is sufficient Class I evidence, whichinvolves a well executed prospective study, to pro-vide a recommendation with the highest degree ofclinical certainty—i.e., a Standard—that an EEG beobtained in all children in whom a nonfebrile seizurehas been diagnosed, to predict the risk of recurrenceand to classify the seizure type and epilepsy syn-drome. The decision to perform other studies, includ-ing LP, laboratory tests, and neuroimaging, for thepurpose of determining the cause of the seizure anddetecting potentially treatable abnormalities, willdepend on the age of the patient and the specificclinical circumstances. Children of different agesmay require different management strategies.20,22

Future research. For most of the questions ad-dressed by this parameter, evidence was insufficientfor making a strong recommendation for a standardor guideline, particularly for laboratory studies. Inorder to generate definitive evidence regarding thevalue of routine (or selective) laboratory testing andthe use of routine neuroimaging studies, sufficientlylarge samples allowing for adequate statistical powerto provide precise estimates (i.e., with narrow confi-dence intervals) are needed. Neuroimaging studiesare needed to understand the significance of neuro-nal migration defects in the context of a first seizure,and are important because of the improved technical

ability of current MRI. In addition, prospective col-lection of data using standardized treatment proto-cols and standardized data collection instruments isessential. Results of studies will only be helpful ifthe patient sample and factors that resulted in inclu-sion into or exclusion from the sample are well de-scribed and documented. Ideally, large consecutiveseries of well-characterized patients are needed forthe results to be accurate and generalizable. Finally,future studies should present separate data fromchildren and adults, and it would be optimal for re-sults in children to be presented by age groupings.

Appropriate timing as well as the choice of evalu-ative studies have not been adequately studied. Chil-dren may present as actively having a seizure whenbrought to the emergency department, as postictal,or as alert with a history of a possible seizure epi-sode having occurred hours, days, or weeks previ-ously. Data regarding the appropriate timing oflaboratory testing, neuroimaging, or EEG studies re-quire adequate prospective studies of these specificquestions, with clearly defined entry criteria and acommon protocol for type and timing of evaluations.

Research studies with adequate sample sizes andappropriate protocols that provide answers to thesequestions may serve to reduce the expense and dis-comfort of unnecessary testing in children with firstseizures, and, more importantly, by identifying ap-propriate candidates, may improve the care andmanagement that these children receive.

Disclaimer. This statement is provided as an edu-cational service of the American Academy of Neurol-ogy. It is based on an assessment of current scientificand clinical information. It is not intended to includeall possible proper methods of care for a particularneurologic problem or all legitimate criteria forchoosing to use a specific procedure. Neither is itintended to exclude any reasonable alternativemethodologies. The AAN recognizes that specific pa-tient care decisions are the prerogative of the patientand the physician caring for the patient, based on allof the circumstances involved.

AcknowledgmentThe authors thank the members of the Quality Standards Sub-committee of the AAN, and Wendy Edlund, the AAN ClinicalPolicy Administrator, for their time, expertise, and efforts in de-veloping this document.

Appendix 1Classification of evidence

Class I. Must have all of a–d:a. Prospective study of a well defined cohort which includes a

description of the nature of the population, the inclusion/exclusioncriteria, demographic characteristics such as age and sex, andseizure type.

b. The sample size must be adequate with enough statisticalpower to justify a conclusion or for identification of subgroups forwhom testing does or does not yield significant information.

c. The interpretation of evaluations performed must be doneblinded to outcome.

d. There must be a satisfactory description of the technologyused for evaluations (e.g., EEG, MRI).