RETIRED

Practice Parameter Evaluating an apparent unprovokedfirst seizure in adults (an evidence-based review)Report of the Quality Standards Subcommittee of the AmericanAcademy of Neurology and the American Epilepsy Society

A Krumholz MDS Wiebe MDG Gronseth MDS Shinnar MD PhDP Levisohn MDT Ting MDJ Hopp MDP Shafer RNH Morris MDL Seiden MDG Barkley MDJ French MD

ABSTRACT

Objective The Quality Standards Subcommittee of the American Academy of Neurology developspractice parameters as strategies for patient care based on analysis of evidence For this practiceparameter the authors reviewed available evidence relevant to evaluating adults presenting with anapparent unprovoked first seizure

Methods Relevant questions were defined and addressed by multiple searches of medical literatureEach article was then reviewed abstracted and classified using an established evidence scoringsystem Conclusions and recommendations were based on a standard three-tiered scheme of evi-dence classification

Results For adults presenting with a first seizure a routine EEG revealed epileptiform abnormalitiesin approximately 23 of patients and these were predictive of seizure recurrence A brain imagingstudy (CT or MRI) was significantly abnormal in 10 of patients indicating a possible seizure etiologyLaboratory tests such as blood counts blood glucose and electrolyte panels were abnormal in up to15 of individuals but abnormalities were minor and did not cause the seizure Overt clinical signs ofinfection such as fever typically predicted significant CSF abnormalities on lumbar puncture Toxicol-ogy screening studies were limited but report some positive tests

Recommendations EEG should be considered as part of the routine neurodiagnostic evaluation ofadults presenting with an apparent unprovoked first seizure (Level B) Brain imaging with CT orMRI should be considered as part of the routine neurodiagnostic evaluation of adults presentingwith an apparent unprovoked first seizure (Level B) Laboratory tests such as blood counts bloodglucose and electrolyte panels (particularly sodium) lumbar puncture and toxicology screeningmay be helpful as determined by the specific clinical circumstances based on the history physicaland neurologic examination but there are insufficient data to support or refute recommendingany of these tests for the routine evaluation of adults presenting with an apparent first unpro-voked seizure (Level U) Neurologyreg 2007691996ndash2007

INTRODUCTION

Seizures are among the most common seri-ous neurologic disorders cared for by neurol-ogists Annually approximately 150000adults will present with a first seizure in theUnited States1 It is estimated that 40 to 50of these incident seizures recur to be classi-fied as epilepsy a condition of recurrent un-

provoked seizures12 The lifetime cumulativerisk of developing recurrent unprovoked sei-zures or epilepsy by the age of 80 years rangesfrom 14 to 331 Since even one seizure isa frightening traumatic event with seriouspotential consequences such as loss of driv-ing privileges limitations for employmentand bodily injury information about opti-

From the University of Maryland School of Medicine (AK TT JH) Baltimore University of Calgary Foothills Medical Centre (SW)Calgary Alberta Canada University of Kansas Medical Center (GG) Kansas City KS Departments of Neurology and Pediatrics (SS)Comprehensive Epilepsy Management Center Montefiore Medical Center Albert Einstein College of Medicine Bronx NY ChildrenrsquosHospital (PL) Denver CO Beth Israel Deaconess Medical Center (PS) Boston MA Fletcher Allen Health CareUniversity of VermontCollege of Medicine (HM) Burlington Peachtree Neurological Clinic (LS) Atlanta GA Henry Ford Medical Hospital (GB) DetroitMI and Department of Neurology (JF) New York University School of Medicine New York

Approved by the Quality Standards Subcommittee on October 28 2006 by the Practice Committee on July 16 2007 and by the AAN Boardof Directors on July 19 2007

Disclosure The authors report no conflicts of interest

Address correspondence andreprint requests to theAmerican Academy ofNeurology 1080 MontrealAvenue St Paul MN 55116guidelinesaancom

Supplemental data atwwwneurologyorg

SPECIAL ARTICLE

1996 Copyright copy 2007 by AAN Enterprises Inc

RETIRED

RETIRED

mal evidence-based approaches for evaluat-ing and treating adults presenting with aseizure is important This practice parameteraddresses evidence regarding methods forevaluating a first seizure in adults

One major study estimates the annualcost of epilepsy in the United States at$125 billion in 1995 with the majority ofdirect cost attributed to diagnostic testsmedical care and drugs prescribed at thetime of the initial evaluation for a seizuredisorder or epilepsy3 Misdiagnosis maylead to ineffective management choices45

and excessive and unnecessary costs3 Er-rors are not only expensive but may alsoresult in harm to the patient

Errors in diagnosis seizure classifica-tion and prognosis are known to lead toinappropriate decisions on the use orchoice of antiepileptic drugs and to otherserious patient management errors45 Asingle seizure can be the first manifestationof epilepsy which is characterized by re-current unprovoked seizures (two or more)or may be a symptom of a brain tumor asystemic disorder an infection or a syn-drome that deserves special attention andtreatment6-8

In this analysis we focus on the methodsand procedures that complement the stan-dard initial history physical and neuro-logic examination In particular we assessthe yield and value of various diagnosticprocedures such as EEG CT or MRI andspecific laboratory or diagnostic tests in-cluding blood counts blood glucose elec-trolytes lumbar puncture and toxicologyscreening79

For the purposes of this practice param-eter we considered studies of adults (ap-pendix 4) defined here as individuals over18 years of age Another recent practice pa-rameter has already addressed evaluationof a first nonfebrile seizure in children10

We attempted to include only patientswith an apparent unprovoked first seizureof any type and we specifically excludedpatients with epilepsy per se The defini-tion of epilepsy we used requires recurrent(two or more) unprovoked seizures6 We

defined the first seizure using the Interna-tional League Against Epilepsy criteria asdid the pediatric guideline10 to include asingle or multiple seizures within 24 hourswith recovery of consciousness betweenthe seizures6 We have therefore excludedpatients who would be diagnosed with epi-lepsy at the time of initial presentationHowever we recognize that seizures and epi-lepsy syndromes overlap may be difficult toseparate and pose similar diagnostic issuesparticularly at initial presentation Studies ofindividuals presenting with presumed newonset seizures report that about 50 of theseindividuals after careful history and ques-tioning actually have had previous seizuresand merit a diagnosis of epilepsy11 Althoughthe recurrence rates differ in individuals aftera first seizure as opposed to a new diagnosisof epilepsy both patient groups are at signif-icant risk for seizure recurrence2 We also ex-cluded adults presenting with a seizure as aknown consequence of an acute conditionsuch as immediate cerebral trauma orstroke6 In addition we considered only pa-tientswho had returned to their normal base-line level of function in order to avoidincluding patients with an acute symptom-atic or provoked seizure We limited ouranalysis in this manner because these disor-ders are diagnostically and therapeuticallydifferent but this substantially reduced thenumber of relevant studies

DESCRIPTION OF THE ANALYTICAL PROCESSThis is an evidence-based appraisal of currently availableclinical measures pertinent to the initial evaluation of a pre-sumed new unprovoked seizure in adults who have returnedto their previous baseline functional levels The appraisalconsists of a systematic review of the literature published inEnglish based on established standards12

Our literature search was conducted by the University ofMinnesota using methodology and filters that increase theyield of evidence-based articles Our search used MedLine1966 to November 2004 and also included CINAHL andThe Cochrane Trials Register

All citations and abstracts were printed and screened bytwo reviewers for any mention of patients with a first sei-zure a first presentation or a new diagnosis of seizure orepilepsy using established criteria (appendix 5) To be in-cluded in the review studies had to report results of anydiagnostic or monitoring intervention pertinent to a first ornew seizure in adults or adolescents (18 years of age) withat least 10 patients as total sample size Studies with mixed

Neurology 69 November 20 2007 1997

RETIRED

RETIRED

age populations were reviewed for data pertaining to pa-tients 18 years of age when possible

Literature search results In our search we found 793articles and obtained all in abstract form Each abstract wasreviewed by two committee members We identified 157 ar-ticles for review of the full text article (an article was in-cluded for review if selected by at least one committeemember based on criteria in appendix 4)

These 157 full text articles were obtained and reviewedby two committee members using established criteria Arti-cles were accepted or rejected when agreed on by both re-viewers using inclusion and exclusion criteria in appendix 4When there was disagreement between the reviewers a thirdreviewer cast the determining vote There were 10 instancesof disagreement resolved by a third party Of the 157 articlesreviewed 39 were selected as acceptable An additional 33studies from the same time period were identified from re-view articles and other sources these were subjected to thesame process and 14 were selected for inclusion

Each accepted study was abstracted by one investigatorand agreed to by a second Key data elements sought forextraction from each study included study patient and in-tervention characteristics In addition for all diagnostictests sensitivity specificity and positive and negative pre-dictive value with its gold standard were sought All eligiblearticles were scored on features pertinent to study designexecution and reporting with a range of possible scores asstandardized by the American Academy of Neurology Qual-ity Standards Subcommittee12

In order to score the evidence we used the Method ofScreening Intervention and Prognosis approved by the Qual-ity Standards Subcommittee (QSS) of the AAN (Appendices2 and 3)12 Of the 53 articles one was ranked as Class I 11 asClass II and the remaining 41 as Class III or IV (see appendixE-1 on theNeurologyreg Web site at wwwneurologyorg)

Goals of immediate evaluation of a first seizure Af-ter an adult who presents with a first seizure isstabilized and returns to baseline function a phy-sician must determine if the event was a seizureand if so whether this was the first such event Ahistory of prior seizures supports a diagnosis ofepilepsy and indicates a higher risk for seizure re-currence210 A careful and complete history phys-ical and neurologic examination are criticallyimportant at the time of initial presentation Thedetermination as to whether a seizure occurred istypically based on the history obtained from a re-liable observer The approach in adults is similarto what was presented in the practice parameterfor evaluating a first nonfebrile seizure in chil-dren2 as noted in appendix 5 A good historyphysical and neurologic examination may allowa physician to make the diagnosis of a seizurewithout additional diagnostic or laboratory test-ing However adults as well as children canpresent with episodic disorders that can be con-fused with seizures or epilepsy Such episodic dis-orders include syncope migraine drug reactionor intoxication and mental disorders such as psy-

chogenic seizures5 No single test clinical findingor symptom is reliable in discriminating betweenan initial seizure and such nonepileptic events Inaddition the reliability of witnesses to the event isvariable and witnesses are not always available

The next goal of assessment is to determine acause for the seizure For some patients the historyphysical and neurologic examination prove ade-quate to discern a probable cause or provide infor-mation to guide the physician to consider otherdiagnostic testing Some disorders causing seizuresrequire prompt diagnosis and acute treatment andothers strongly influence prognosis and impact deci-sions regarding initiation and maintenance of anti-epileptic drug therapy In particular provokedseizures are the result of acute precipitating disor-ders such as meningitis intoxications trauma ormetabolic derangements including hypoglycemiaand may require prompt intervention to reverse po-tentially damaging dangerous causes In contrastunprovoked seizuresmay also have causes but theseare not acute precipitating conditions requiring im-mediate action Their basis may be cryptogenic (noknown cause) remote symptomatic (due to a pre-existing brain injury or lesion such as a tumor orstroke) or idiopathic (genetic)6

ANALYSIS OF EVIDENCE Electroencephalogra-phy Should an EEG be routinely ordered in an adult pre-sentingwithanapparentunprovoked first seizureEvidence

Of 1 Class I and 10 Class II articles reviewed (witha total of 1766 patients) (table 1)13-24 assessing theyield of EEG EEGs were reported as abnormalin 12 to 73 (average yield 51) and reportedas significantly abnormal in 8 to 50 (average29) The abnormality considered as significant byauthors was the presence of epileptiform activity inthe form of spikes or sharp waves as interpreted bythe local or reading electroencephalographer in pa-tients clinically judged to present with a new onsetseizure evidence similar to that reported for chil-dren10 This yield is substantial However it is alsoclear from the evidence that a normal EEG does notexclude the presence of a seizure disorder Indeedon average about 50 of individuals clinically diag-nosed with a seizure have a normal EEG (table 1)

With respect to the value of the EEG as a predic-tor of risk of seizure recurrence and prognosis stud-ies support the EEG as useful (table 1) This issimilar to findings reported in children10 Of the oneClass I13-14 and seven Class II studies15-21 that con-sider the EEG as a predictor of risk of seizure recur-rence five studies a majority including the onlyClass I study13-15 demonstrate that epileptiformEEG activity in particular generalized spike andwave type discharges or focal spikes are associated

1998 Neurology 69 November 20 2007

RETIRED

RETIREDwith a greater risk for seizure recurrence13-161921

Therefore the weight of the evidence supports theEEG as of value in predicting seizure recurrence

What is this degree of increased risk for seizurerecurrence associated with epileptiform EEG ab-normalities In the only Class I study patientswith idiopathic seizure disorders and generalizedspike wave EEG abnormalities had an actual550 rate of seizure recurrence at 60 monthsfollow-up while the expected recurrence rate forthese patients was calculated in the same study tobe less 482 a small but statistically significantdifference1314 A meta-analysis in adults and chil-dren by Berg and Shinnar found that individualswith epileptiform EEG abnormalities were morelikely to have a seizure recurrence with a pooledrelative risk of 20 (95CI 16 26)2 and this issimilar to a more recent study16 Our own meta-analysis of the Class I and II studies (table 1)shows similar results with the estimated post-testprobability of a seizure recurrence in patientswith epileptiform EEG abnormalities of 495compared to only 274 in individuals whoseEEGs are completely normal The data show nosignificance for other more nonspecific EEG ab-normalities such as focal or diffuse slowing forpredicting seizure recurrence Several Class IIIand IV articles are more strongly in support ofepileptiform EEG abnormalities as a predictor ofseizure recurrence risk

Some of the study variability in risks for sei-zure recurrence may relate to the timing of theEEG time of entry into the clinical study after theinitial event duration of the recording or treat-

ment with antiepileptic medications In regard tothe timeframe of the EEGs this was specified inonly three studies (table 1)172024 and varied fromwithin 48 hours20 to a mean of 150 (range 1 to 36)days24 following the presenting seizure The ef-fects of antiepileptic drug treatment were specifi-cally considered in one study that carefullyrandomized treatment it still found the EEG pre-dictive of seizure recurrence after a first seizure16

Furthermore if there were a treatment bias wewould expect it to be a bias toward demonstrat-ing the EEG as less predictive of seizure recur-rence since patients with abnormal EEGs wouldmore likely get treated with antiepileptic drugsthat could in turn reduce seizure recurrence2

Studies of patients who presented with a his-tory of previous seizures were excluded in ouranalysis It is well described that the EEG is valu-able for establishing the nature of a seizure disor-der and guiding optimal therapy in such patientsor in patients with recurrent seizures or estab-lished epilepsy24825 For example specific gener-alized spike wave abnormalities on EEG arenoted to be of particular importance in properlydiagnosing determining prognosis and guidingtherapy for patients with juvenile myoclonic epi-lepsy and absence epilepsy These types of pri-mary generalized epilepsies are associated with apoor response to some antiepileptic drugs but ex-cellent response to others241025

Conclusion For adults presenting with an appar-ent unprovoked first seizure analysis of the evi-dence from 1 Class I and 10 Class II studiesindicates that the EEG is probably helpful It has a

Table 1 EEG in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) ClassNo studied()

No abnormal()

No signabnormal ()

Duration offollow-up mo

17 147 (16 and over) II 147 (100) 82 (56) 51 (35) 30 to 54

19 76 (mean 20) II 76 (100) 16 (21) 12 (16) 12 to 24 (mean 19)

23 56 (16 to 89) II 43 (77) 14 (33) 8 (19) NA

16 397 (72 over 16) II 397 (100) 221 (56) 199 (50) 24

22 107 (17 and over) II 103 (96) 75 (73) 29 (28) NA

13 14 173 (70 over 19) I 168 (97) 74 (44) 13 (8)dagger 64 mean (1990)

18 306 (16 and over) II 295 (96) 158 (54) 79 (27) 48

20 132 (13 to 86 mean 33) II 132 (100) 59 (45) 39 (30) 48

24 91 (15 or older mean 50) II 91 (100) 63 (69) 19 (21) NA

21 157 (17 to 84) II 157 (100) 111 (71) 42 (27) 24

15 157 (15 to 85 mean 38) II 157 (100) 19 (12) 19 (12) 24

Total 1799 1766 (98) 892 (51) 510 (29)

Epileptiform activity as interpreted by the local or reading electroencephalographerdaggerSignificant abnormal is only generalized spike waveNA not applicable or not available

Neurology 69 November 20 2007 1999

RETIRED

RETIRED

substantial yield with about 29 of EEGs dem-onstrating significant abnormalities2526 and theseabnormalities predict the risk for seizure recur-rence In addition EEG is regarded as a standardfor the initial classification of seizures since itforms a basis for the ldquoclinical and electroencepha-lographic classification of epileptic seizuresrdquo2526

Recommendations 1 The EEG (routine) shouldbe considered as part of the neurodiagnostic eval-uation of the adult with an apparent unprovokedfirst seizure because it has a substantial yield(Level B)82526

2 The EEG (routine) should be considered aspart of the neurodiagnostic evaluation of theadult with an apparent unprovoked first seizurebecause it has value in determining the risk forseizure recurrence (Level B)

Neuroimaging studies Should a brain imaging study(CT or MRI) be routinely ordered in an adult presenting

with an apparent unprovoked first seizure Evidence Ofthe seven Class II studies1518-20222327 (table 2) thatconsidered the yield and value (with or withoutcontrast agents) of the CT or MRI in adults ini-tially presenting with a seizure (1092 patients) theCT was reported as abnormal in 1 to 57 (av-erage yield 15) and was significantly abnormalin 1 to 47 (average 10) These significantabnormalities affected patient management andincluded previously unrecognized brain tumorsvascular lesions and cerebral cysticercosisTherefore six studies using CT1518-202327 and oneusing CT and MRI22 were of value for assessingadult patients with a first seizure Two of theseClass II studies also indicated that the finding ofan abnormal CT was associated with a greaterrisk of seizure recurrence1820 but the other studiesdid not address this issue The 13 Class III and 8Class IV studies of brain imaging we consideredfurther support the value of the CT or MRI indetermining a seizure cause

Interestingly almost all of the Class II studies

report on CT rather than MRI (table 2) One rea-son for this is that some of the studies are olderbut another is that many of these studies are ofpatients seen acutely in emergency departmentswhere the CT is the procedure of choice becauseof the relative speed and ease of obtaining thestudy and its effectiveness in excluding cata-strophic problems that may require immediate at-tention7 MRIs seem to be done in a moreselective way and most of the studies consideringMRI are not prospective or well controlled limit-ing them as Class I or II evidence However oneMRI study of new onset seizures indicates thatthe yield of MRI is at least as high and likelyhigher than CT but that study was not classifiedas Class I or II evidence because it included pa-tients with provoked seizures and those present-ing after multiple seizures or with pre-existingepilepsy28 In general although these includedbrain imaging studies varying in their methodsand conclusions the great majority support thatneuroimaging can determine potentially impor-tant treatable causes of seizures in a significantnumber of patients particularly in older patientsA previous QSS study of this issue in children didnot recommend neuroimaging as a standard10 butour review of studies in adults indicates that neuro-imaging deserves greater consideration in adultsand particularly in older adults Not only is the po-tential to discover significant abnormalities such astumors greater in adults728 but the risk associatedwithMRIs is less since adults are much less likely torequire conscious sedation than are children10

A practice parameter from the AmericanAcademy of Neurology in 1996 also consideredthe issue of neuroimaging in the patient present-ing with seizures29 That study similarly con-cluded that the decision for neuroimaging andtiming of the study should be driven by individualclinical circumstances In general when an imag-ing study is necessary in a non-emergent situa-

Table 2 Neuroimaging with CT or MRI in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) Class Modality No studied () No abnormal ()No signabnormal ()

19 100 (mean 20) II CT 100 (100) 17 (17) 17 (17)

23 56 (16 to 89) II CT 50 (89) 14 (28) 2 (4)

22 107 (17 and over) II CT and MRI 45 (42) 25 (56) 21 (47)

18 408 (16 and over) II CT 375 (92) 3 (1) 3 (1)

20 132 (13 to 86 mean 33) II CT 85 (64) 9 (11) 9 (11)

27 132 (adults mean 46) II CT 119 (90) 68 (57) 34 (29)

15 157 (15 to 85 mean 38) II CT 154 (98) 4 (3) 4 (3)

Total 1092 928 (85) 140 (15) 90 (10)

2000 Neurology 69 November 20 2007

RETIRED

RETIRED

tion MRI is more sensitive and is more likely toshow significant abnormalities than is CT91029

Conclusion For adults presenting initially with anapparent unprovoked first seizure the evidencefrom seven Class II studies indicates that a brain im-aging study either a CT orMRI is probably usefulIt has a significant yield of about 10 which maylead to the diagnosis of disorders such as a braintumor stroke cysticercosis or other structural le-sions and may have some value in determining therisk for seizure recurrence

Recommendation Brain imaging using CT orMRI should be considered as part of the neurodi-agnostic evaluation of adults presenting with anapparent unprovoked first seizure (Level B)

Laboratory studies Based on review of clinicalstudies expert opinion and position papers wechose to assess the value of the following labora-tory tests blood glucose blood counts electro-lytes lumbar puncture and toxicology screeningin adults with a first seizure79182330-32

Should blood counts blood glucose and electrolytepanels be routinely ordered in an adult with an apparent

unprovoked first seizure EvidenceOf the two Class IIstudies1823 (464 patients) (table 3) that assessedthe yield and value of blood counts blood glu-cose and electrolyte panels abnormalities werereported in from 0 to 15 for each of these testsbut no clinically significant abnormalities werenoted by the authors (table 3) In one Class IIstudy blood count abnormalities were reportedin 15 of patients but all these were judged bythe author to be incidental and clinically insignif-icant such as minor increased white blood cellcounts that resolved23 That same study reporteda 10 incidence of hypoglycemia but again allthose abnormalities were judged as mild and clin-ically insignificant with regard to the seizure Inboth the Class II studies routine electrolytes wereobtained in a high percentage of patients with

particular attention to serum calcium and sodium(table 3) Serum calcium was normal in all pa-tients in which it was measured In one study so-dium abnormalities of hyponatremia were notedin 7 of patients but again all were judged bythe author not to be of clinical significance23 Theother Class II study also reported no significantsodium abnormalities18

There were four Class III studies that showed ahigher incidence of significant laboratory abnormal-ities particularly for serum sodium and glucose30-33

All these studies were based in emergency depart-ments and the patients were obviously ill withmorecomorbidities than those in our Class II studies1823

A large proportion were patients with acute symp-tomatic or provoked rather than apparent unpro-voked seizures as required by our inclusion criteria(appendix 4) Among the reported acute symptom-atic causes for the seizureswere alcoholwithdrawalacute stroke tumor sepsis or trauma Also somestudies had many patients with other exclusion cri-teria (appendix 4) such as focal neurologic deficitsand itwas not specifiedwhether patients returned totheir normal level of function after the seizure30-33

Despite these limitations and acknowledgment thatyield is very low7 these four Class III studies wereconsistent in proposing some value to routinescreening of blood glucose for hypoglycemia and se-rum electrolytes for hyponatremia because unantic-ipated and clinically relevant hypoglycemia andhyponatremia were found in about 1 of thesepatients30-33

The two Class II studies as well the four ClassIII we identified do not provide convincing evi-dence to support or refute significant value forblood glucose blood count and electrolytes asroutine for adults at initial presentation with epi-leptic seizures This finding is similar to the QSSreport on initial seizure evaluation in children10

and a previous evidence-based analysis10

Table 3 Laboratory diagnostic tests in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) Class Modality No studied ()No abnormal() 95 CI

No sign abnormal() 95 CI

23 56 (16 to 89) II Blood count 55 (98) 8 (15) 76ndash26 0 (0) 0ndash65

Electrolytes 55 (98) 4 (7) 29ndash173 0 (0) 0ndash65

Glucose 50 (89) 5 (10) 43ndash21 0 (0) 0ndash71

Calcium 35 (63) 0 (0) 0ndash99 0 (0) 0ndash99

18 408 (16 and older) II Blood count 371 (91) NA 0 (0) 0ndash1

Electrolytes calcium 371 (91) NA 0 (0) 0ndash1

Total 464

NA not available

Neurology 69 November 20 2007 2001

RETIRED

RETIRED

In general most studies emphasize that the his-tory and physical examination can guide decisionmaking because they often predict patients whowill demonstrate significantly abnormal labora-tory tests23 For example patients with persistingaltered mental status fever or focal neurologicdeficits warrant more extensive evaluation andare more likely to have abnormal laboratory find-ings7 A practice parameter from the AmericanCollege of Emergency Physicians which generallysupports our findings still recommends obtaininga blood glucose determination and serum sodiumin patients with a first seizure7 However thatrecommendation was based on evidence fromstudies that we excluded or downgraded as ourinclusion criteria require because they involvedlarge numbers of patients with obvious acutesymptomatic seizures

Conclusion Data from two Class II and fourClass III studies showed that in adults presentingwith an apparent unprovoked first seizure al-though some abnormal laboratory results are re-ported there is not sufficient evidence to supportor refute recommending routine testing of bloodglucose blood counts or electrolyte panels Thenecessity for such studies should be guided by spe-cific clinical circumstances based on the historyphysical and neurologic examination

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure bloodglucose blood counts and electrolyte panels(particularly sodium) may be helpful in specificclinical circumstances but there are insufficientdata to support or refute routine recommenda-tion of any of these laboratory tests (Level U)

Should a lumbar puncture be routinely performed inan adult presenting with an apparent unprovoked first sei-

zure Evidence There were no Class I or II studiesbut of the two Class III studies considering thevalue of a lumbar puncture in patients with a firstseizure both were from emergency departmentsand reported significant lumbar puncture abnor-malities in up to 8 of patients3033 However notall patients in these two studies had unprovokedseizures or returned to baseline function as ourinclusion criteria require A substantial numberhad acute symptomatic seizures and level of con-sciousness was not well specified Another majorlimitation of these studies is that lumbar punc-tures were performed selectively They were per-formed in only 68 of the patients in one study30

and 24 of the other33 and this depended on suchpresenting clinical features as fever or at the dis-cretion of the staff Consequently lumbar punc-tures were not routinely performed and the

decision to perform a lumbar puncture was di-rected by the clinical history and physical find-ings Moreover in one study the determination ofwhether the lumbar puncture showed a ldquosignifi-cantrdquo abnormality was based largely on whetherthe patient was admitted to the hospital ratherthan whether the test showed a cause for the sei-zure30 In other studies of first seizures that ex-cluded patients with acute symptomatic seizureslumbar puncture was not considered as an impor-tant routine test1823

In particular there are no Class I II or otherconvincing studies to support a lumbar puncturein patients who are alert oriented afebrile andnot immunocompromised7 There is some evi-dence to support a lumbar puncture after a firstseizure in patients who are immunocompromisedeven if they are afebrile734

Conclusion Data from two Class III studies re-vealed significant abnormalities in up to 8 of amixed group of patients presenting to an emer-gency department with a first seizure Howeverthe studies selectively performed lumbar punc-tures based on clinical findings and included pa-tients who did not meet our inclusion criteriasuch as those with acute symptomatic causes fortheir seizures or who had not returned to theirnormal baseline function

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure lumbarpuncture may be helpful in specific clinical circum-stances such as patients who are febrile but thereare insufficient data to support or refute recom-mending routine lumbar puncture (Level U)

Should toxicologic screening be routinely ordered inan adult presenting with an apparent unprovoked first sei-

zure Evidence Seizures are reported as a consequenceof drug intoxication particularly with tricyclic anti-depressants cocaine and other stimulants735 In aseries of patients with acute medical complicationsof cocaine intoxication seizures often first seizuresaccounted for 10 of the presenting symptoms36

Several studies of emergency department admis-sions for first seizures including both acute symp-tomatic and unprovoked seizures indicated about3 may relate to drug toxicity or abuse73233 andone study advocated toxicology screening for allpatients with unexplained first seizures30 Still arecent evidence-based review by the AmericanAcademy of Emergency Physicians Clinical Pol-icy Committee considering management of adultpatients presenting to an emergency departmentwith seizures did not find sufficient evidence in thelimited existing prospective or retrospective studiesto recommend routine toxicology screening7

2002 Neurology 69 November 20 2007

RETIRED

RETIRED

In considering this issue we identified twoClass III studies considering the value of toxicol-ogy screening in adult patients presenting with afirst seizure3536 Although both studies reportedsome first seizures in patients with abnormal tox-icology screening neither study investigated theuse of routine screening in first seizure patients Inaddition the studies did not consider mainly un-provoked seizures in which the patient returnedto baseline function as our inclusion criteria spec-ify (appendix 4)

Conclusion In two Class III studies consideringthe value of toxicology screening in adult patientspresenting with a seizure some patients with ap-parent unprovoked first seizure were includedbut neither study investigated the use of routinetoxicology screening for such patients

Recommendation In the adult presenting with anapparent unprovoked seizure toxicology screen-ing may be helpful in specific clinical circum-stances but there are insufficient data to supportor refute a routine recommendation for toxicol-ogy screening (Level U)

CONCLUSIONS AND RECOMMENDATIONSDiagnosis of seizures determination of possiblecauses classification of seizure type and possibleseizure syndrome and assessment of the risk forrecurrence of seizures require timely selection andinterpretation of appropriate diagnostic tools in-cluding EEGs CTs or MRIs blood studies andlumbar puncture This report assesses the evi-dence available to answer the question ldquoWhat arethe most appropriate tools for evaluating a firstunprovoked seizure in adults at the time of initialpresentationrdquo

For the adult initially presenting with an ap-parent unprovoked seizure diagnostic evalua-tions influence aspects of patient care andmanagement including drug treatment patientand family counseling and the need for immedi-ate hospitalization and subsequent follow-upThis practice parameter reviews the published lit-erature concerning the yield and value of studiesfollowing an initial seizure in adults and classifiesthe strength of available evidence There is insuf-ficient Class I evidence to provide any recommen-dation with the highest degree of clinical certainty(see appendix 3) However there is sufficientClass I and Class II evidence that a routine EEGshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with an ap-parent unprovoked first seizure and predicts therisk of seizure recurrence (Level B) There is alsosufficient Class II evidence that a brain imaging

study either a CT or MRI helps determine thepresence of a significant abnormality such as abrain tumor Brain imaging using CT or MRIshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with a firstseizure (Level B) The CT has a major advantagebecause of the speed with which it can be ob-tained so its value is mainly in the emergency sit-uation but theMRI is regarded by expert opinionas having a higher yield and is the preferable pro-cedure in non-emergency or elective situa-tions91029 However both MRI and CT are ofvalue The decision to perform other studies in-cluding blood glucose blood counts electrolytepanels lumbar puncture or toxicology screeningfor the purpose of determining the cause of theseizure or identifying potentially treatable condi-tions will depend on the specific or individualizedclinical circumstances based on the history phys-ical and neurologic examination There are in-sufficient data to support or refute the use of anyone of these tests or procedures as a routine rec-ommendation (Level U)

The history physical and neurologic exami-nation are acknowledged by expert consensus tobe required for the initial diagnosis and classifica-tion of seizures26 These standards for the initialdiagnosis and classification of seizures form thebasis for the ldquoclinical and electroencephalo-graphic classification of epileptic seizuresrdquo21

Indeed information from the history includingthe type of the seizure21 together with findings fromthe physical and neurologic examination also aid inthe determination of a probable cause or risk of re-currence If the history reveals that there have beenother seizures in the past then this indicates epilepsyrather than a single isolated seizure Although thesubsequent diagnostic evaluation may be similarto that for a single unprovoked seizure there aredifferent prognostic and therapeutic implicationsfor a patient with epilepsy27 Evidence of pre-existing seizures provides information for the po-tential classification of an epilepsy syndrome25

Such information is of importance in establishingpossible causes associated comorbidities prog-nosis and optimal treatment of a seizure disorderor epilepsy syndrome2526 Moreover several fea-tures of the clinical history and physical and neuro-logic examination such as the presence of specificgeneralized or focal neurologic deficits are predic-tive of higher risks of seizure recurrence210

RECOMMENDATIONS FOR FUTURE RE-SEARCH The history physical and neurologicexaminations form the basis for the diagnosis and

Neurology 69 November 20 2007 2003

RETIRED

RETIRED

classification of epileptic seizures26 and this is ac-knowledged by expert consensus26 Still future re-search is warranted to determine the specificaspects of the history physical and neurologicexamination that are most useful in both initialdiagnosis of a seizure and subsequent manage-ment Also it would be helpful to understandhow this type of information should best guideperformance and timing of other neurodiagnosticor laboratory studies An international group re-cently recommended a broader definition of epi-lepsy as a brain disorder with a history of at leastone seizure and ldquocharacterized by an enduringpredisposition to generate epileptic seizuresrdquo8

This type of approach will require validation andinvestigation including determination of the truevalue of diagnostic studies such as EEG in estab-lishing the presence of an epileptic syndrome aftera first seizure In regard to the EEG one limita-tion of currently available studies is the variabilityin the timing of the EEG after incident seizureThere is some evidence in children that an EEGdone within 24 hours of a presenting seizure givesa higher yield of significant abnormalities10

Although all types of seizures are considered inthis analysis most patients had a convulsive sei-zure This may be because convulsive seizures aremore likely to prompt acute medical attentionthan a simple or complex partial seizure Morestudies specifically focusing attention on all typesof initial seizures not just convulsive seizures areneeded

Other factors deserve consideration in futureinvestigations For instance the costs of these var-ious tests do not receive much attention in studiesof patients with first seizure and should Also theelderly are a growing proportion of the generalpopulation and warrant special consideration infuture investigations Another unresolved issuerelates to the need for immediate hospitalizationof patients with a first seizure Some guidelinesexist but well designed studies are needed7

Prediction of seizure recurrence after a first un-provoked seizure is important to guide decisionsregarding antiepileptic drug therapy patientmanagement and counseling Many factors arereported to predict seizure recurrence includingthe etiology of the seizures and association withneurologic abnormalities2 Although the EEG andbrain imaging studies are reported to predict sei-zure recurrence studies vary in support of thatobservation and in respect to the specific EEG ab-normalities or brain imaging findings that aremost significant Some of this variability regard-ing EEG findings may relate to the timing of the

study after the initial event and to the decisionwhether to treat with antiepileptic medicationsThe value of specific findings on the EEG brainimaging and other variables deserve furtherstudy in regard to seizure recurrence risks Alsofuture studies of the clinical utility for such neuro-diagnostic procedures should employ standard-ized tests and prediction algorithms37

Future studies are needed to analyze the valueof expert input into the history physical neuro-logic examination or diagnostic testing such as theEEG or brain imaging The expertise and qualifica-tions of individuals interpreting such information ortests particularly the EEG or brain imaging varyconsiderably and may influence interpretationsand outcomes37 These are not simple ldquopositiverdquo orldquonegativerdquo tests like a pregnancy test they requireexpertise for proper interpretation Some other di-agnostic studies deserve further analysis in patientspresenting with new onset seizures In particularroutine electrolytes including serum sodium andglucosewere reported in some studies to be of valuebut this was not consistent or confirmed by otherreports Also toxicologic screening was reported asimportant in some limited series but was not stud-ied in a manner to allow adequate evidence-basedverification or refutation of its routine utility

Studies of patient management and counselingat the time of first diagnosis of a single seizure orepilepsy are recommended For example whichpatients should receive antiepileptic drug treat-ment and if so exactly what type and whenshould they get it Also when is the patient bestadvised regarding driving laws and other socialissues and how and by whom should this bedone This is not a trivial issue For instance aBritish study noted that only 21 of all adult firstseizure patients received the correct advice aboutdriving limitations23

We recommend that future studies address theissues that we raise here To best determine howto optimally evaluate the adult presenting with afirst seizure those studies should be structured toemphasize the use of large well-characterizedsamples clearly defined subjects and outcomesand standardized data collection methods

MISSION STATEMENT OF QSS The missionof the QSS is to prioritize develop and publishevidence-based practice parameters related to thediagnosis treatment and prognosis of neurologicdisorders

The QSS is committed to using the most rigor-ous methods available within our budget in col-

2004 Neurology 69 November 20 2007

RETIRED

RETIRED

laboration with other available AAN resourcesto most efficiently accomplish this mission

DISCLAIMER This statement is provided as aneducational service of the American Academy ofNeurology It is based on an assessment of currentscientific and clinical information It is not intendedto include all possible proper methods of care for aparticular neurologic problem or all legitimate crite-ria for choosing to use a specific procedure Neitheris it intended to exclude any reasonable alternativemethodologies The AAN recognizes that specificpatient care decisions are the prerogative of the pa-tient and physician caring for the patient based onall of the circumstances involved

CONFLICT OF INTEREST STATEMENT TheAmerican Academy of Neurology is committed toproducing independent critical and truthful clin-ical practice guidelines (CPGs) Significant effortsare made to minimize the potential for conflicts ofinterest to influence the recommendations of thisCPG To the extent possible the AAN keeps sep-arate those who have a financial stake in the suc-cess or failure of the products appraised in theCPGs and the developers of the guidelines Con-flict of interest forms were obtained from all au-thors and reviewed by an oversight committeeprior to project initiation AAN limits the participa-tion of authors with substantial conflicts of interestThe AAN forbids commercial participation in orfunding of guideline projects Drafts of the guide-lines have been reviewed by at least threeAANcom-mittees a network of neurologists Neurology peerreviewers and representatives from related fieldsTheAANGuidelineAuthorConflict of Interest Pol-icy can be viewed at wwwaancom

APPENDIX 1

Quality Standards Subcommittee members Jacque-line French MD FAAN (Co-Chair) Gary S Gronseth MD(Co-Chair) Charles E Argoff MD Stephen Ashwal MDFAAN (ex-officio) Christopher Bever Jr MD MBA FAANJohn D England MD FAAN Gary M Franklin MD MPH(ex-officio) GaryH FridayMDMPH FAAN Larry BGold-stein MD FAAN Deborah Hirtz MD (ex-officio) Robert GHollowayMDMPH FAANDonald J IversonMD FAANLeslie A Morrison MD Clifford J Schostal MD David JThurmanMDMPHWilliam JWeinerMD FAAN SamuelWiebe MD

APPENDIX 2AAN classification of evidence for rating of screen-ing article

Class I A statistical population-based sample of patientsstudied at a uniform point in time (usually early) during the

course of the condition All patients undergo the intervention ofinterest The outcome if not objective is determined in an eval-uation that is masked to the patientsrsquo clinical presentations

Class II A statistical non-referral-clinic-based sample ofpatients studied at a uniform point in time (usually early)during the course of the condition Most patients undergothe intervention of interest The outcome if not objective isdetermined in an evaluation that is masked to the patientsrsquoclinical presentations

Class III A sample of patients studied during the course ofthe condition Some patients undergo the intervention of in-terest The outcome if not objective is determined in anevaluation by someone other than the treating physician

Class IV Expert opinion case reports or any study notmeeting criteria for Class I to III

APPENDIX 3

Classification of recommendationsA Established as effective ineffective or harmful (or es-

tablished as usefulpredictive or not usefulpredictive)for the given condition in the specified population(Level A rating requires at least two consistent Class Istudies)

B Probably effective ineffective or harmful (or probablyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level B ratingrequires at least one Class I study or at least two consis-tent Class II studies)

C Possibly effective ineffective or harmful (or possiblyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level C ratingrequires at least one Class II study or two consistentClass III studies)

U Data inadequate or conflicting given current knowl-edge treatment (test predictor) is unproven (Studiesnot meeting criteria for Class IndashClass III)

In exceptional cases one convincing Class I study may suf-fice for an ldquoArdquo recommendation if 1) all criteria are met themagnitude of effect is large (relative rate improved outcome5 and the lower limit of the confidence interval is 2)

APPENDIX 4

Exclusion criteria

bull Letters and case reportsbull Non-English language studiesbull Animal studiesbull Pharmacodynamicpharmacokinetic studiesbull Studies dealing primarily with established or chronicepilepsy patients

bull Studies dealing primarily with acute provoked seizures

Inclusion criteria

bull Study designs observational (prospective retrospectiveand cross-sectional) or interventional (randomized con-trolled trials [RCTs] nonrandomized controlled trials[nRCTs] and uncontrolled case series [UCS])

bull At least 10 patients adults with a first seizure a first pre-sentation with epilepsy or seizures or a first diagnosis ofepilepsy

Neurology 69 November 20 2007 2005

RETIRED

RETIRED

bull Studies addressing any of the following diagnostic in-terventions neuropsychological assessment imagingwith CT MRI or PET scans EEG (standard videoinvasive ambulatory) lumbar puncture toxic screensand laboratory tests (hematologybiochemistry)

bull Patients should be over the age of 18 or the studyshould include a substantial proportion of subjectsover the age of 18

bull Studies reported in English onlybull Studies dealing primarily with apparent unprovokedfirst seizures

APPENDIX 5

Outline for seizure assessment Features of a seizureAssociated factors

AgeMedical history ndash previous history of similar episodesprior stroke brain tumor systemic illness mental ill-ness drug or alcohol abuse

Family historyDevelopmental statusBehaviorHealth at seizure onset ndash febrile ill exposed to illnesscomplaints of not feeling well sleep deprived

Precipitating events other than illness ndash trauma alcoholmedications illicit drugs toxins

Symptoms during seizure (ictal)Aura Subjective sensationsBehavior Mood or behavioral changes before the seizurePreictal symptoms Described by patient or witnessedVocal Cry or gasp slurring of words garbled speechMotor Head or eye turning eye deviation posturingjerking (rhythmic) stiffening automatisms (purpose-less repetitive movements such as picking at clothinglip smacking) generalized or focal movements

Respiration Change in breathing pattern cessation ofbreathing cyanosis

Autonomic Pupillary dilation drooling change in respira-tory or heart rate incontinence pallor vomiting

Loss of consciousness or inability to understand or speak

Symptoms following a seizure (postictal)Amnesia for eventsConfusionLethargySleepinessHeadaches and muscle achesTransient focal weakness (Toddrsquos paresis)Nausea or vomitingBiting of tongue

Adapted from Hirtz et al10

Received February 20 2007 Accepted in final form June 72007

REFERENCES1 Hauser WA Hesdorffer DC Epilepsy frequency

causes and consequences New York Demos Publica-tions 1990

2 Berg AT Shinnar S The risk of seizure recurrence fol-lowing a first unprovoked seizure a quantitative re-view Neurology 199141965ndash972

3 Begley CE Famulari M Annegers JF et al The cost ofepilepsy in the United States an estimate frompopulation- based and survey data Epilepsia 200041342ndash352

4 Grunewald RA Chroni E Panayiotopoulos CP De-layed diagnosis of juvenile myoclonic epilepsy J Neu-rol Neurosurg Psychiatry 199255497ndash499

5 Krumholz A Nonepileptic seizures diagnosis andmanagement Neurology 1999S76ndash83

6 Commission on Epidemiology and Prognosis Interna-tional League Against Epilepsy Guidelines for epidemio-logical studies on epilepsy Epilepsia 199334592ndash596

7 American College of Emergency Physicians (ACEP)Clinical Policies Committee and the Clinical PoliciesSubcommittee on Seizures Clinical Policy Critical is-sues in the evaluation of adult patients presenting to anemergency department with seizures Ann Emerg Med200443605ndash625

8 Fisher RS BoasWvE BlumeW et al Epileptic seizuresand epilepsy Definitions proposed by the InternationalLeague Against Epilepsy (ILAE) and the InternationalBureau for Epilepsy (IBE) Epilepsia 200546470ndash472

9 Agency for Healthcare Research and Quality (AHRQ)Management of newly diagnosed patients with epi-lepsy a systematic review of the literature (AHRQPublication No 01-E038) Rockville MD Agency forHealth Research Quality US Department of Healthand Human Services 2001

10 Hirtz D Ashwal S Berg A et al Practice parameterevaluating a first nonfebrile seizure in children Neu-rology 200055616ndash623

11 Jallon P Loiseau P Loiseau J Newly diagnosed unpro-voked epileptic seizures Presentation at diagnosis inCORALE study Epilepsia 200142464ndash475

12 Quality Standards Subcommittee and the Therapeuticsand Technology Assessment Subcommittee of theAmerican Academy of Neurology Clinical PracticeGuideline Process Manuel St Paul MN AmericanAcademy of Neurology 2004

13 Hauser WA Anderson VE Loewenson RB McRob-erts SM Seizure recurrence after a first unprovoked sei-zure N Engl J Med 1982307522ndash528

14 Hauser WA Rich SS Annegers JF Anderson VE Sei-zure recurrence after a 1st unprovoked seizure an ex-tended follow-up Neurology 1990401163ndash1170

15 van Donselaar CA Schimsheimer R Geerts AT et alValue of the electroencephalogram in adult patientswith untreated idiopathic first seizures Arch Neurol199249231ndash237

16 First Seizure Trial Group Randomized clinical trial onthe efficacy of antiepileptic drugs in reducing the risk ofrelapse after a first unprovoked tonic-clonic seizureNeurology 199343478ndash483

17 Bora I Seckin B Zarifoglu S Turin F Sadikogllu S OgulE Risk of recurrence after first unprovoked tonic-clonicseizure in adults J Neurol 1995242157ndash163

18 Hopkins A Garman A Clarke C The first seizure inadult life Value of clinical features electroencephalog-raphy and computerized tomographic scanning in pre-diction of seizure recurrence Lancet 1988721ndash726

19 Das CP Sawhney IMS Lal V et al Risk of recurrenceof seizures following a single unprovoked idiopathicseizure Neurol India 200046357ndash360

2006 Neurology 69 November 20 2007

RETIRED

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

RETIRED

mal evidence-based approaches for evaluat-ing and treating adults presenting with aseizure is important This practice parameteraddresses evidence regarding methods forevaluating a first seizure in adults

One major study estimates the annualcost of epilepsy in the United States at$125 billion in 1995 with the majority ofdirect cost attributed to diagnostic testsmedical care and drugs prescribed at thetime of the initial evaluation for a seizuredisorder or epilepsy3 Misdiagnosis maylead to ineffective management choices45

and excessive and unnecessary costs3 Er-rors are not only expensive but may alsoresult in harm to the patient

Errors in diagnosis seizure classifica-tion and prognosis are known to lead toinappropriate decisions on the use orchoice of antiepileptic drugs and to otherserious patient management errors45 Asingle seizure can be the first manifestationof epilepsy which is characterized by re-current unprovoked seizures (two or more)or may be a symptom of a brain tumor asystemic disorder an infection or a syn-drome that deserves special attention andtreatment6-8

In this analysis we focus on the methodsand procedures that complement the stan-dard initial history physical and neuro-logic examination In particular we assessthe yield and value of various diagnosticprocedures such as EEG CT or MRI andspecific laboratory or diagnostic tests in-cluding blood counts blood glucose elec-trolytes lumbar puncture and toxicologyscreening79

For the purposes of this practice param-eter we considered studies of adults (ap-pendix 4) defined here as individuals over18 years of age Another recent practice pa-rameter has already addressed evaluationof a first nonfebrile seizure in children10

We attempted to include only patientswith an apparent unprovoked first seizureof any type and we specifically excludedpatients with epilepsy per se The defini-tion of epilepsy we used requires recurrent(two or more) unprovoked seizures6 We

defined the first seizure using the Interna-tional League Against Epilepsy criteria asdid the pediatric guideline10 to include asingle or multiple seizures within 24 hourswith recovery of consciousness betweenthe seizures6 We have therefore excludedpatients who would be diagnosed with epi-lepsy at the time of initial presentationHowever we recognize that seizures and epi-lepsy syndromes overlap may be difficult toseparate and pose similar diagnostic issuesparticularly at initial presentation Studies ofindividuals presenting with presumed newonset seizures report that about 50 of theseindividuals after careful history and ques-tioning actually have had previous seizuresand merit a diagnosis of epilepsy11 Althoughthe recurrence rates differ in individuals aftera first seizure as opposed to a new diagnosisof epilepsy both patient groups are at signif-icant risk for seizure recurrence2 We also ex-cluded adults presenting with a seizure as aknown consequence of an acute conditionsuch as immediate cerebral trauma orstroke6 In addition we considered only pa-tientswho had returned to their normal base-line level of function in order to avoidincluding patients with an acute symptom-atic or provoked seizure We limited ouranalysis in this manner because these disor-ders are diagnostically and therapeuticallydifferent but this substantially reduced thenumber of relevant studies

DESCRIPTION OF THE ANALYTICAL PROCESSThis is an evidence-based appraisal of currently availableclinical measures pertinent to the initial evaluation of a pre-sumed new unprovoked seizure in adults who have returnedto their previous baseline functional levels The appraisalconsists of a systematic review of the literature published inEnglish based on established standards12

Our literature search was conducted by the University ofMinnesota using methodology and filters that increase theyield of evidence-based articles Our search used MedLine1966 to November 2004 and also included CINAHL andThe Cochrane Trials Register

All citations and abstracts were printed and screened bytwo reviewers for any mention of patients with a first sei-zure a first presentation or a new diagnosis of seizure orepilepsy using established criteria (appendix 5) To be in-cluded in the review studies had to report results of anydiagnostic or monitoring intervention pertinent to a first ornew seizure in adults or adolescents (18 years of age) withat least 10 patients as total sample size Studies with mixed

Neurology 69 November 20 2007 1997

RETIRED

RETIRED

age populations were reviewed for data pertaining to pa-tients 18 years of age when possible

Literature search results In our search we found 793articles and obtained all in abstract form Each abstract wasreviewed by two committee members We identified 157 ar-ticles for review of the full text article (an article was in-cluded for review if selected by at least one committeemember based on criteria in appendix 4)

These 157 full text articles were obtained and reviewedby two committee members using established criteria Arti-cles were accepted or rejected when agreed on by both re-viewers using inclusion and exclusion criteria in appendix 4When there was disagreement between the reviewers a thirdreviewer cast the determining vote There were 10 instancesof disagreement resolved by a third party Of the 157 articlesreviewed 39 were selected as acceptable An additional 33studies from the same time period were identified from re-view articles and other sources these were subjected to thesame process and 14 were selected for inclusion

Each accepted study was abstracted by one investigatorand agreed to by a second Key data elements sought forextraction from each study included study patient and in-tervention characteristics In addition for all diagnostictests sensitivity specificity and positive and negative pre-dictive value with its gold standard were sought All eligiblearticles were scored on features pertinent to study designexecution and reporting with a range of possible scores asstandardized by the American Academy of Neurology Qual-ity Standards Subcommittee12

In order to score the evidence we used the Method ofScreening Intervention and Prognosis approved by the Qual-ity Standards Subcommittee (QSS) of the AAN (Appendices2 and 3)12 Of the 53 articles one was ranked as Class I 11 asClass II and the remaining 41 as Class III or IV (see appendixE-1 on theNeurologyreg Web site at wwwneurologyorg)

Goals of immediate evaluation of a first seizure Af-ter an adult who presents with a first seizure isstabilized and returns to baseline function a phy-sician must determine if the event was a seizureand if so whether this was the first such event Ahistory of prior seizures supports a diagnosis ofepilepsy and indicates a higher risk for seizure re-currence210 A careful and complete history phys-ical and neurologic examination are criticallyimportant at the time of initial presentation Thedetermination as to whether a seizure occurred istypically based on the history obtained from a re-liable observer The approach in adults is similarto what was presented in the practice parameterfor evaluating a first nonfebrile seizure in chil-dren2 as noted in appendix 5 A good historyphysical and neurologic examination may allowa physician to make the diagnosis of a seizurewithout additional diagnostic or laboratory test-ing However adults as well as children canpresent with episodic disorders that can be con-fused with seizures or epilepsy Such episodic dis-orders include syncope migraine drug reactionor intoxication and mental disorders such as psy-

chogenic seizures5 No single test clinical findingor symptom is reliable in discriminating betweenan initial seizure and such nonepileptic events Inaddition the reliability of witnesses to the event isvariable and witnesses are not always available

The next goal of assessment is to determine acause for the seizure For some patients the historyphysical and neurologic examination prove ade-quate to discern a probable cause or provide infor-mation to guide the physician to consider otherdiagnostic testing Some disorders causing seizuresrequire prompt diagnosis and acute treatment andothers strongly influence prognosis and impact deci-sions regarding initiation and maintenance of anti-epileptic drug therapy In particular provokedseizures are the result of acute precipitating disor-ders such as meningitis intoxications trauma ormetabolic derangements including hypoglycemiaand may require prompt intervention to reverse po-tentially damaging dangerous causes In contrastunprovoked seizuresmay also have causes but theseare not acute precipitating conditions requiring im-mediate action Their basis may be cryptogenic (noknown cause) remote symptomatic (due to a pre-existing brain injury or lesion such as a tumor orstroke) or idiopathic (genetic)6

ANALYSIS OF EVIDENCE Electroencephalogra-phy Should an EEG be routinely ordered in an adult pre-sentingwithanapparentunprovoked first seizureEvidence

Of 1 Class I and 10 Class II articles reviewed (witha total of 1766 patients) (table 1)13-24 assessing theyield of EEG EEGs were reported as abnormalin 12 to 73 (average yield 51) and reportedas significantly abnormal in 8 to 50 (average29) The abnormality considered as significant byauthors was the presence of epileptiform activity inthe form of spikes or sharp waves as interpreted bythe local or reading electroencephalographer in pa-tients clinically judged to present with a new onsetseizure evidence similar to that reported for chil-dren10 This yield is substantial However it is alsoclear from the evidence that a normal EEG does notexclude the presence of a seizure disorder Indeedon average about 50 of individuals clinically diag-nosed with a seizure have a normal EEG (table 1)

With respect to the value of the EEG as a predic-tor of risk of seizure recurrence and prognosis stud-ies support the EEG as useful (table 1) This issimilar to findings reported in children10 Of the oneClass I13-14 and seven Class II studies15-21 that con-sider the EEG as a predictor of risk of seizure recur-rence five studies a majority including the onlyClass I study13-15 demonstrate that epileptiformEEG activity in particular generalized spike andwave type discharges or focal spikes are associated

1998 Neurology 69 November 20 2007

RETIRED

RETIREDwith a greater risk for seizure recurrence13-161921

Therefore the weight of the evidence supports theEEG as of value in predicting seizure recurrence

What is this degree of increased risk for seizurerecurrence associated with epileptiform EEG ab-normalities In the only Class I study patientswith idiopathic seizure disorders and generalizedspike wave EEG abnormalities had an actual550 rate of seizure recurrence at 60 monthsfollow-up while the expected recurrence rate forthese patients was calculated in the same study tobe less 482 a small but statistically significantdifference1314 A meta-analysis in adults and chil-dren by Berg and Shinnar found that individualswith epileptiform EEG abnormalities were morelikely to have a seizure recurrence with a pooledrelative risk of 20 (95CI 16 26)2 and this issimilar to a more recent study16 Our own meta-analysis of the Class I and II studies (table 1)shows similar results with the estimated post-testprobability of a seizure recurrence in patientswith epileptiform EEG abnormalities of 495compared to only 274 in individuals whoseEEGs are completely normal The data show nosignificance for other more nonspecific EEG ab-normalities such as focal or diffuse slowing forpredicting seizure recurrence Several Class IIIand IV articles are more strongly in support ofepileptiform EEG abnormalities as a predictor ofseizure recurrence risk

Some of the study variability in risks for sei-zure recurrence may relate to the timing of theEEG time of entry into the clinical study after theinitial event duration of the recording or treat-

ment with antiepileptic medications In regard tothe timeframe of the EEGs this was specified inonly three studies (table 1)172024 and varied fromwithin 48 hours20 to a mean of 150 (range 1 to 36)days24 following the presenting seizure The ef-fects of antiepileptic drug treatment were specifi-cally considered in one study that carefullyrandomized treatment it still found the EEG pre-dictive of seizure recurrence after a first seizure16

Furthermore if there were a treatment bias wewould expect it to be a bias toward demonstrat-ing the EEG as less predictive of seizure recur-rence since patients with abnormal EEGs wouldmore likely get treated with antiepileptic drugsthat could in turn reduce seizure recurrence2

Studies of patients who presented with a his-tory of previous seizures were excluded in ouranalysis It is well described that the EEG is valu-able for establishing the nature of a seizure disor-der and guiding optimal therapy in such patientsor in patients with recurrent seizures or estab-lished epilepsy24825 For example specific gener-alized spike wave abnormalities on EEG arenoted to be of particular importance in properlydiagnosing determining prognosis and guidingtherapy for patients with juvenile myoclonic epi-lepsy and absence epilepsy These types of pri-mary generalized epilepsies are associated with apoor response to some antiepileptic drugs but ex-cellent response to others241025

Conclusion For adults presenting with an appar-ent unprovoked first seizure analysis of the evi-dence from 1 Class I and 10 Class II studiesindicates that the EEG is probably helpful It has a

Table 1 EEG in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) ClassNo studied()

No abnormal()

No signabnormal ()

Duration offollow-up mo

17 147 (16 and over) II 147 (100) 82 (56) 51 (35) 30 to 54

19 76 (mean 20) II 76 (100) 16 (21) 12 (16) 12 to 24 (mean 19)

23 56 (16 to 89) II 43 (77) 14 (33) 8 (19) NA

16 397 (72 over 16) II 397 (100) 221 (56) 199 (50) 24

22 107 (17 and over) II 103 (96) 75 (73) 29 (28) NA

13 14 173 (70 over 19) I 168 (97) 74 (44) 13 (8)dagger 64 mean (1990)

18 306 (16 and over) II 295 (96) 158 (54) 79 (27) 48

20 132 (13 to 86 mean 33) II 132 (100) 59 (45) 39 (30) 48

24 91 (15 or older mean 50) II 91 (100) 63 (69) 19 (21) NA

21 157 (17 to 84) II 157 (100) 111 (71) 42 (27) 24

15 157 (15 to 85 mean 38) II 157 (100) 19 (12) 19 (12) 24

Total 1799 1766 (98) 892 (51) 510 (29)

Epileptiform activity as interpreted by the local or reading electroencephalographerdaggerSignificant abnormal is only generalized spike waveNA not applicable or not available

Neurology 69 November 20 2007 1999

RETIRED

RETIRED

substantial yield with about 29 of EEGs dem-onstrating significant abnormalities2526 and theseabnormalities predict the risk for seizure recur-rence In addition EEG is regarded as a standardfor the initial classification of seizures since itforms a basis for the ldquoclinical and electroencepha-lographic classification of epileptic seizuresrdquo2526

Recommendations 1 The EEG (routine) shouldbe considered as part of the neurodiagnostic eval-uation of the adult with an apparent unprovokedfirst seizure because it has a substantial yield(Level B)82526

2 The EEG (routine) should be considered aspart of the neurodiagnostic evaluation of theadult with an apparent unprovoked first seizurebecause it has value in determining the risk forseizure recurrence (Level B)

Neuroimaging studies Should a brain imaging study(CT or MRI) be routinely ordered in an adult presenting

with an apparent unprovoked first seizure Evidence Ofthe seven Class II studies1518-20222327 (table 2) thatconsidered the yield and value (with or withoutcontrast agents) of the CT or MRI in adults ini-tially presenting with a seizure (1092 patients) theCT was reported as abnormal in 1 to 57 (av-erage yield 15) and was significantly abnormalin 1 to 47 (average 10) These significantabnormalities affected patient management andincluded previously unrecognized brain tumorsvascular lesions and cerebral cysticercosisTherefore six studies using CT1518-202327 and oneusing CT and MRI22 were of value for assessingadult patients with a first seizure Two of theseClass II studies also indicated that the finding ofan abnormal CT was associated with a greaterrisk of seizure recurrence1820 but the other studiesdid not address this issue The 13 Class III and 8Class IV studies of brain imaging we consideredfurther support the value of the CT or MRI indetermining a seizure cause

Interestingly almost all of the Class II studies

report on CT rather than MRI (table 2) One rea-son for this is that some of the studies are olderbut another is that many of these studies are ofpatients seen acutely in emergency departmentswhere the CT is the procedure of choice becauseof the relative speed and ease of obtaining thestudy and its effectiveness in excluding cata-strophic problems that may require immediate at-tention7 MRIs seem to be done in a moreselective way and most of the studies consideringMRI are not prospective or well controlled limit-ing them as Class I or II evidence However oneMRI study of new onset seizures indicates thatthe yield of MRI is at least as high and likelyhigher than CT but that study was not classifiedas Class I or II evidence because it included pa-tients with provoked seizures and those present-ing after multiple seizures or with pre-existingepilepsy28 In general although these includedbrain imaging studies varying in their methodsand conclusions the great majority support thatneuroimaging can determine potentially impor-tant treatable causes of seizures in a significantnumber of patients particularly in older patientsA previous QSS study of this issue in children didnot recommend neuroimaging as a standard10 butour review of studies in adults indicates that neuro-imaging deserves greater consideration in adultsand particularly in older adults Not only is the po-tential to discover significant abnormalities such astumors greater in adults728 but the risk associatedwithMRIs is less since adults are much less likely torequire conscious sedation than are children10

A practice parameter from the AmericanAcademy of Neurology in 1996 also consideredthe issue of neuroimaging in the patient present-ing with seizures29 That study similarly con-cluded that the decision for neuroimaging andtiming of the study should be driven by individualclinical circumstances In general when an imag-ing study is necessary in a non-emergent situa-

Table 2 Neuroimaging with CT or MRI in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) Class Modality No studied () No abnormal ()No signabnormal ()

19 100 (mean 20) II CT 100 (100) 17 (17) 17 (17)

23 56 (16 to 89) II CT 50 (89) 14 (28) 2 (4)

22 107 (17 and over) II CT and MRI 45 (42) 25 (56) 21 (47)

18 408 (16 and over) II CT 375 (92) 3 (1) 3 (1)

20 132 (13 to 86 mean 33) II CT 85 (64) 9 (11) 9 (11)

27 132 (adults mean 46) II CT 119 (90) 68 (57) 34 (29)

15 157 (15 to 85 mean 38) II CT 154 (98) 4 (3) 4 (3)

Total 1092 928 (85) 140 (15) 90 (10)

2000 Neurology 69 November 20 2007

RETIRED

RETIRED

tion MRI is more sensitive and is more likely toshow significant abnormalities than is CT91029

Conclusion For adults presenting initially with anapparent unprovoked first seizure the evidencefrom seven Class II studies indicates that a brain im-aging study either a CT orMRI is probably usefulIt has a significant yield of about 10 which maylead to the diagnosis of disorders such as a braintumor stroke cysticercosis or other structural le-sions and may have some value in determining therisk for seizure recurrence

Recommendation Brain imaging using CT orMRI should be considered as part of the neurodi-agnostic evaluation of adults presenting with anapparent unprovoked first seizure (Level B)

Laboratory studies Based on review of clinicalstudies expert opinion and position papers wechose to assess the value of the following labora-tory tests blood glucose blood counts electro-lytes lumbar puncture and toxicology screeningin adults with a first seizure79182330-32

Should blood counts blood glucose and electrolytepanels be routinely ordered in an adult with an apparent

unprovoked first seizure EvidenceOf the two Class IIstudies1823 (464 patients) (table 3) that assessedthe yield and value of blood counts blood glu-cose and electrolyte panels abnormalities werereported in from 0 to 15 for each of these testsbut no clinically significant abnormalities werenoted by the authors (table 3) In one Class IIstudy blood count abnormalities were reportedin 15 of patients but all these were judged bythe author to be incidental and clinically insignif-icant such as minor increased white blood cellcounts that resolved23 That same study reporteda 10 incidence of hypoglycemia but again allthose abnormalities were judged as mild and clin-ically insignificant with regard to the seizure Inboth the Class II studies routine electrolytes wereobtained in a high percentage of patients with

particular attention to serum calcium and sodium(table 3) Serum calcium was normal in all pa-tients in which it was measured In one study so-dium abnormalities of hyponatremia were notedin 7 of patients but again all were judged bythe author not to be of clinical significance23 Theother Class II study also reported no significantsodium abnormalities18

There were four Class III studies that showed ahigher incidence of significant laboratory abnormal-ities particularly for serum sodium and glucose30-33

All these studies were based in emergency depart-ments and the patients were obviously ill withmorecomorbidities than those in our Class II studies1823

A large proportion were patients with acute symp-tomatic or provoked rather than apparent unpro-voked seizures as required by our inclusion criteria(appendix 4) Among the reported acute symptom-atic causes for the seizureswere alcoholwithdrawalacute stroke tumor sepsis or trauma Also somestudies had many patients with other exclusion cri-teria (appendix 4) such as focal neurologic deficitsand itwas not specifiedwhether patients returned totheir normal level of function after the seizure30-33

Despite these limitations and acknowledgment thatyield is very low7 these four Class III studies wereconsistent in proposing some value to routinescreening of blood glucose for hypoglycemia and se-rum electrolytes for hyponatremia because unantic-ipated and clinically relevant hypoglycemia andhyponatremia were found in about 1 of thesepatients30-33

The two Class II studies as well the four ClassIII we identified do not provide convincing evi-dence to support or refute significant value forblood glucose blood count and electrolytes asroutine for adults at initial presentation with epi-leptic seizures This finding is similar to the QSSreport on initial seizure evaluation in children10

and a previous evidence-based analysis10

Table 3 Laboratory diagnostic tests in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) Class Modality No studied ()No abnormal() 95 CI

No sign abnormal() 95 CI

23 56 (16 to 89) II Blood count 55 (98) 8 (15) 76ndash26 0 (0) 0ndash65

Electrolytes 55 (98) 4 (7) 29ndash173 0 (0) 0ndash65

Glucose 50 (89) 5 (10) 43ndash21 0 (0) 0ndash71

Calcium 35 (63) 0 (0) 0ndash99 0 (0) 0ndash99

18 408 (16 and older) II Blood count 371 (91) NA 0 (0) 0ndash1

Electrolytes calcium 371 (91) NA 0 (0) 0ndash1

Total 464

NA not available

Neurology 69 November 20 2007 2001

RETIRED

RETIRED

In general most studies emphasize that the his-tory and physical examination can guide decisionmaking because they often predict patients whowill demonstrate significantly abnormal labora-tory tests23 For example patients with persistingaltered mental status fever or focal neurologicdeficits warrant more extensive evaluation andare more likely to have abnormal laboratory find-ings7 A practice parameter from the AmericanCollege of Emergency Physicians which generallysupports our findings still recommends obtaininga blood glucose determination and serum sodiumin patients with a first seizure7 However thatrecommendation was based on evidence fromstudies that we excluded or downgraded as ourinclusion criteria require because they involvedlarge numbers of patients with obvious acutesymptomatic seizures

Conclusion Data from two Class II and fourClass III studies showed that in adults presentingwith an apparent unprovoked first seizure al-though some abnormal laboratory results are re-ported there is not sufficient evidence to supportor refute recommending routine testing of bloodglucose blood counts or electrolyte panels Thenecessity for such studies should be guided by spe-cific clinical circumstances based on the historyphysical and neurologic examination

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure bloodglucose blood counts and electrolyte panels(particularly sodium) may be helpful in specificclinical circumstances but there are insufficientdata to support or refute routine recommenda-tion of any of these laboratory tests (Level U)

Should a lumbar puncture be routinely performed inan adult presenting with an apparent unprovoked first sei-

zure Evidence There were no Class I or II studiesbut of the two Class III studies considering thevalue of a lumbar puncture in patients with a firstseizure both were from emergency departmentsand reported significant lumbar puncture abnor-malities in up to 8 of patients3033 However notall patients in these two studies had unprovokedseizures or returned to baseline function as ourinclusion criteria require A substantial numberhad acute symptomatic seizures and level of con-sciousness was not well specified Another majorlimitation of these studies is that lumbar punc-tures were performed selectively They were per-formed in only 68 of the patients in one study30

and 24 of the other33 and this depended on suchpresenting clinical features as fever or at the dis-cretion of the staff Consequently lumbar punc-tures were not routinely performed and the

decision to perform a lumbar puncture was di-rected by the clinical history and physical find-ings Moreover in one study the determination ofwhether the lumbar puncture showed a ldquosignifi-cantrdquo abnormality was based largely on whetherthe patient was admitted to the hospital ratherthan whether the test showed a cause for the sei-zure30 In other studies of first seizures that ex-cluded patients with acute symptomatic seizureslumbar puncture was not considered as an impor-tant routine test1823

In particular there are no Class I II or otherconvincing studies to support a lumbar puncturein patients who are alert oriented afebrile andnot immunocompromised7 There is some evi-dence to support a lumbar puncture after a firstseizure in patients who are immunocompromisedeven if they are afebrile734

Conclusion Data from two Class III studies re-vealed significant abnormalities in up to 8 of amixed group of patients presenting to an emer-gency department with a first seizure Howeverthe studies selectively performed lumbar punc-tures based on clinical findings and included pa-tients who did not meet our inclusion criteriasuch as those with acute symptomatic causes fortheir seizures or who had not returned to theirnormal baseline function

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure lumbarpuncture may be helpful in specific clinical circum-stances such as patients who are febrile but thereare insufficient data to support or refute recom-mending routine lumbar puncture (Level U)

Should toxicologic screening be routinely ordered inan adult presenting with an apparent unprovoked first sei-

zure Evidence Seizures are reported as a consequenceof drug intoxication particularly with tricyclic anti-depressants cocaine and other stimulants735 In aseries of patients with acute medical complicationsof cocaine intoxication seizures often first seizuresaccounted for 10 of the presenting symptoms36

Several studies of emergency department admis-sions for first seizures including both acute symp-tomatic and unprovoked seizures indicated about3 may relate to drug toxicity or abuse73233 andone study advocated toxicology screening for allpatients with unexplained first seizures30 Still arecent evidence-based review by the AmericanAcademy of Emergency Physicians Clinical Pol-icy Committee considering management of adultpatients presenting to an emergency departmentwith seizures did not find sufficient evidence in thelimited existing prospective or retrospective studiesto recommend routine toxicology screening7

2002 Neurology 69 November 20 2007

RETIRED

RETIRED

In considering this issue we identified twoClass III studies considering the value of toxicol-ogy screening in adult patients presenting with afirst seizure3536 Although both studies reportedsome first seizures in patients with abnormal tox-icology screening neither study investigated theuse of routine screening in first seizure patients Inaddition the studies did not consider mainly un-provoked seizures in which the patient returnedto baseline function as our inclusion criteria spec-ify (appendix 4)

Conclusion In two Class III studies consideringthe value of toxicology screening in adult patientspresenting with a seizure some patients with ap-parent unprovoked first seizure were includedbut neither study investigated the use of routinetoxicology screening for such patients

Recommendation In the adult presenting with anapparent unprovoked seizure toxicology screen-ing may be helpful in specific clinical circum-stances but there are insufficient data to supportor refute a routine recommendation for toxicol-ogy screening (Level U)

CONCLUSIONS AND RECOMMENDATIONSDiagnosis of seizures determination of possiblecauses classification of seizure type and possibleseizure syndrome and assessment of the risk forrecurrence of seizures require timely selection andinterpretation of appropriate diagnostic tools in-cluding EEGs CTs or MRIs blood studies andlumbar puncture This report assesses the evi-dence available to answer the question ldquoWhat arethe most appropriate tools for evaluating a firstunprovoked seizure in adults at the time of initialpresentationrdquo

For the adult initially presenting with an ap-parent unprovoked seizure diagnostic evalua-tions influence aspects of patient care andmanagement including drug treatment patientand family counseling and the need for immedi-ate hospitalization and subsequent follow-upThis practice parameter reviews the published lit-erature concerning the yield and value of studiesfollowing an initial seizure in adults and classifiesthe strength of available evidence There is insuf-ficient Class I evidence to provide any recommen-dation with the highest degree of clinical certainty(see appendix 3) However there is sufficientClass I and Class II evidence that a routine EEGshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with an ap-parent unprovoked first seizure and predicts therisk of seizure recurrence (Level B) There is alsosufficient Class II evidence that a brain imaging

study either a CT or MRI helps determine thepresence of a significant abnormality such as abrain tumor Brain imaging using CT or MRIshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with a firstseizure (Level B) The CT has a major advantagebecause of the speed with which it can be ob-tained so its value is mainly in the emergency sit-uation but theMRI is regarded by expert opinionas having a higher yield and is the preferable pro-cedure in non-emergency or elective situa-tions91029 However both MRI and CT are ofvalue The decision to perform other studies in-cluding blood glucose blood counts electrolytepanels lumbar puncture or toxicology screeningfor the purpose of determining the cause of theseizure or identifying potentially treatable condi-tions will depend on the specific or individualizedclinical circumstances based on the history phys-ical and neurologic examination There are in-sufficient data to support or refute the use of anyone of these tests or procedures as a routine rec-ommendation (Level U)

The history physical and neurologic exami-nation are acknowledged by expert consensus tobe required for the initial diagnosis and classifica-tion of seizures26 These standards for the initialdiagnosis and classification of seizures form thebasis for the ldquoclinical and electroencephalo-graphic classification of epileptic seizuresrdquo21

Indeed information from the history includingthe type of the seizure21 together with findings fromthe physical and neurologic examination also aid inthe determination of a probable cause or risk of re-currence If the history reveals that there have beenother seizures in the past then this indicates epilepsyrather than a single isolated seizure Although thesubsequent diagnostic evaluation may be similarto that for a single unprovoked seizure there aredifferent prognostic and therapeutic implicationsfor a patient with epilepsy27 Evidence of pre-existing seizures provides information for the po-tential classification of an epilepsy syndrome25

Such information is of importance in establishingpossible causes associated comorbidities prog-nosis and optimal treatment of a seizure disorderor epilepsy syndrome2526 Moreover several fea-tures of the clinical history and physical and neuro-logic examination such as the presence of specificgeneralized or focal neurologic deficits are predic-tive of higher risks of seizure recurrence210

RECOMMENDATIONS FOR FUTURE RE-SEARCH The history physical and neurologicexaminations form the basis for the diagnosis and

Neurology 69 November 20 2007 2003

RETIRED

RETIRED

classification of epileptic seizures26 and this is ac-knowledged by expert consensus26 Still future re-search is warranted to determine the specificaspects of the history physical and neurologicexamination that are most useful in both initialdiagnosis of a seizure and subsequent manage-ment Also it would be helpful to understandhow this type of information should best guideperformance and timing of other neurodiagnosticor laboratory studies An international group re-cently recommended a broader definition of epi-lepsy as a brain disorder with a history of at leastone seizure and ldquocharacterized by an enduringpredisposition to generate epileptic seizuresrdquo8

This type of approach will require validation andinvestigation including determination of the truevalue of diagnostic studies such as EEG in estab-lishing the presence of an epileptic syndrome aftera first seizure In regard to the EEG one limita-tion of currently available studies is the variabilityin the timing of the EEG after incident seizureThere is some evidence in children that an EEGdone within 24 hours of a presenting seizure givesa higher yield of significant abnormalities10

Although all types of seizures are considered inthis analysis most patients had a convulsive sei-zure This may be because convulsive seizures aremore likely to prompt acute medical attentionthan a simple or complex partial seizure Morestudies specifically focusing attention on all typesof initial seizures not just convulsive seizures areneeded

Other factors deserve consideration in futureinvestigations For instance the costs of these var-ious tests do not receive much attention in studiesof patients with first seizure and should Also theelderly are a growing proportion of the generalpopulation and warrant special consideration infuture investigations Another unresolved issuerelates to the need for immediate hospitalizationof patients with a first seizure Some guidelinesexist but well designed studies are needed7

Prediction of seizure recurrence after a first un-provoked seizure is important to guide decisionsregarding antiepileptic drug therapy patientmanagement and counseling Many factors arereported to predict seizure recurrence includingthe etiology of the seizures and association withneurologic abnormalities2 Although the EEG andbrain imaging studies are reported to predict sei-zure recurrence studies vary in support of thatobservation and in respect to the specific EEG ab-normalities or brain imaging findings that aremost significant Some of this variability regard-ing EEG findings may relate to the timing of the

study after the initial event and to the decisionwhether to treat with antiepileptic medicationsThe value of specific findings on the EEG brainimaging and other variables deserve furtherstudy in regard to seizure recurrence risks Alsofuture studies of the clinical utility for such neuro-diagnostic procedures should employ standard-ized tests and prediction algorithms37

Future studies are needed to analyze the valueof expert input into the history physical neuro-logic examination or diagnostic testing such as theEEG or brain imaging The expertise and qualifica-tions of individuals interpreting such information ortests particularly the EEG or brain imaging varyconsiderably and may influence interpretationsand outcomes37 These are not simple ldquopositiverdquo orldquonegativerdquo tests like a pregnancy test they requireexpertise for proper interpretation Some other di-agnostic studies deserve further analysis in patientspresenting with new onset seizures In particularroutine electrolytes including serum sodium andglucosewere reported in some studies to be of valuebut this was not consistent or confirmed by otherreports Also toxicologic screening was reported asimportant in some limited series but was not stud-ied in a manner to allow adequate evidence-basedverification or refutation of its routine utility

Studies of patient management and counselingat the time of first diagnosis of a single seizure orepilepsy are recommended For example whichpatients should receive antiepileptic drug treat-ment and if so exactly what type and whenshould they get it Also when is the patient bestadvised regarding driving laws and other socialissues and how and by whom should this bedone This is not a trivial issue For instance aBritish study noted that only 21 of all adult firstseizure patients received the correct advice aboutdriving limitations23

We recommend that future studies address theissues that we raise here To best determine howto optimally evaluate the adult presenting with afirst seizure those studies should be structured toemphasize the use of large well-characterizedsamples clearly defined subjects and outcomesand standardized data collection methods

MISSION STATEMENT OF QSS The missionof the QSS is to prioritize develop and publishevidence-based practice parameters related to thediagnosis treatment and prognosis of neurologicdisorders

The QSS is committed to using the most rigor-ous methods available within our budget in col-

2004 Neurology 69 November 20 2007

RETIRED

RETIRED

laboration with other available AAN resourcesto most efficiently accomplish this mission

DISCLAIMER This statement is provided as aneducational service of the American Academy ofNeurology It is based on an assessment of currentscientific and clinical information It is not intendedto include all possible proper methods of care for aparticular neurologic problem or all legitimate crite-ria for choosing to use a specific procedure Neitheris it intended to exclude any reasonable alternativemethodologies The AAN recognizes that specificpatient care decisions are the prerogative of the pa-tient and physician caring for the patient based onall of the circumstances involved

CONFLICT OF INTEREST STATEMENT TheAmerican Academy of Neurology is committed toproducing independent critical and truthful clin-ical practice guidelines (CPGs) Significant effortsare made to minimize the potential for conflicts ofinterest to influence the recommendations of thisCPG To the extent possible the AAN keeps sep-arate those who have a financial stake in the suc-cess or failure of the products appraised in theCPGs and the developers of the guidelines Con-flict of interest forms were obtained from all au-thors and reviewed by an oversight committeeprior to project initiation AAN limits the participa-tion of authors with substantial conflicts of interestThe AAN forbids commercial participation in orfunding of guideline projects Drafts of the guide-lines have been reviewed by at least threeAANcom-mittees a network of neurologists Neurology peerreviewers and representatives from related fieldsTheAANGuidelineAuthorConflict of Interest Pol-icy can be viewed at wwwaancom

APPENDIX 1

Quality Standards Subcommittee members Jacque-line French MD FAAN (Co-Chair) Gary S Gronseth MD(Co-Chair) Charles E Argoff MD Stephen Ashwal MDFAAN (ex-officio) Christopher Bever Jr MD MBA FAANJohn D England MD FAAN Gary M Franklin MD MPH(ex-officio) GaryH FridayMDMPH FAAN Larry BGold-stein MD FAAN Deborah Hirtz MD (ex-officio) Robert GHollowayMDMPH FAANDonald J IversonMD FAANLeslie A Morrison MD Clifford J Schostal MD David JThurmanMDMPHWilliam JWeinerMD FAAN SamuelWiebe MD

APPENDIX 2AAN classification of evidence for rating of screen-ing article

Class I A statistical population-based sample of patientsstudied at a uniform point in time (usually early) during the

course of the condition All patients undergo the intervention ofinterest The outcome if not objective is determined in an eval-uation that is masked to the patientsrsquo clinical presentations

Class II A statistical non-referral-clinic-based sample ofpatients studied at a uniform point in time (usually early)during the course of the condition Most patients undergothe intervention of interest The outcome if not objective isdetermined in an evaluation that is masked to the patientsrsquoclinical presentations

Class III A sample of patients studied during the course ofthe condition Some patients undergo the intervention of in-terest The outcome if not objective is determined in anevaluation by someone other than the treating physician

Class IV Expert opinion case reports or any study notmeeting criteria for Class I to III

APPENDIX 3

Classification of recommendationsA Established as effective ineffective or harmful (or es-

tablished as usefulpredictive or not usefulpredictive)for the given condition in the specified population(Level A rating requires at least two consistent Class Istudies)

B Probably effective ineffective or harmful (or probablyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level B ratingrequires at least one Class I study or at least two consis-tent Class II studies)

C Possibly effective ineffective or harmful (or possiblyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level C ratingrequires at least one Class II study or two consistentClass III studies)

U Data inadequate or conflicting given current knowl-edge treatment (test predictor) is unproven (Studiesnot meeting criteria for Class IndashClass III)

In exceptional cases one convincing Class I study may suf-fice for an ldquoArdquo recommendation if 1) all criteria are met themagnitude of effect is large (relative rate improved outcome5 and the lower limit of the confidence interval is 2)

APPENDIX 4

Exclusion criteria

bull Letters and case reportsbull Non-English language studiesbull Animal studiesbull Pharmacodynamicpharmacokinetic studiesbull Studies dealing primarily with established or chronicepilepsy patients

bull Studies dealing primarily with acute provoked seizures

Inclusion criteria

bull Study designs observational (prospective retrospectiveand cross-sectional) or interventional (randomized con-trolled trials [RCTs] nonrandomized controlled trials[nRCTs] and uncontrolled case series [UCS])

bull At least 10 patients adults with a first seizure a first pre-sentation with epilepsy or seizures or a first diagnosis ofepilepsy

Neurology 69 November 20 2007 2005

RETIRED

RETIRED

bull Studies addressing any of the following diagnostic in-terventions neuropsychological assessment imagingwith CT MRI or PET scans EEG (standard videoinvasive ambulatory) lumbar puncture toxic screensand laboratory tests (hematologybiochemistry)

bull Patients should be over the age of 18 or the studyshould include a substantial proportion of subjectsover the age of 18

bull Studies reported in English onlybull Studies dealing primarily with apparent unprovokedfirst seizures

APPENDIX 5

Outline for seizure assessment Features of a seizureAssociated factors

AgeMedical history ndash previous history of similar episodesprior stroke brain tumor systemic illness mental ill-ness drug or alcohol abuse

Family historyDevelopmental statusBehaviorHealth at seizure onset ndash febrile ill exposed to illnesscomplaints of not feeling well sleep deprived

Precipitating events other than illness ndash trauma alcoholmedications illicit drugs toxins

Symptoms during seizure (ictal)Aura Subjective sensationsBehavior Mood or behavioral changes before the seizurePreictal symptoms Described by patient or witnessedVocal Cry or gasp slurring of words garbled speechMotor Head or eye turning eye deviation posturingjerking (rhythmic) stiffening automatisms (purpose-less repetitive movements such as picking at clothinglip smacking) generalized or focal movements

Respiration Change in breathing pattern cessation ofbreathing cyanosis

Autonomic Pupillary dilation drooling change in respira-tory or heart rate incontinence pallor vomiting

Loss of consciousness or inability to understand or speak

Symptoms following a seizure (postictal)Amnesia for eventsConfusionLethargySleepinessHeadaches and muscle achesTransient focal weakness (Toddrsquos paresis)Nausea or vomitingBiting of tongue

Adapted from Hirtz et al10

Received February 20 2007 Accepted in final form June 72007

REFERENCES1 Hauser WA Hesdorffer DC Epilepsy frequency

causes and consequences New York Demos Publica-tions 1990

2 Berg AT Shinnar S The risk of seizure recurrence fol-lowing a first unprovoked seizure a quantitative re-view Neurology 199141965ndash972

3 Begley CE Famulari M Annegers JF et al The cost ofepilepsy in the United States an estimate frompopulation- based and survey data Epilepsia 200041342ndash352

4 Grunewald RA Chroni E Panayiotopoulos CP De-layed diagnosis of juvenile myoclonic epilepsy J Neu-rol Neurosurg Psychiatry 199255497ndash499

5 Krumholz A Nonepileptic seizures diagnosis andmanagement Neurology 1999S76ndash83

6 Commission on Epidemiology and Prognosis Interna-tional League Against Epilepsy Guidelines for epidemio-logical studies on epilepsy Epilepsia 199334592ndash596

7 American College of Emergency Physicians (ACEP)Clinical Policies Committee and the Clinical PoliciesSubcommittee on Seizures Clinical Policy Critical is-sues in the evaluation of adult patients presenting to anemergency department with seizures Ann Emerg Med200443605ndash625

8 Fisher RS BoasWvE BlumeW et al Epileptic seizuresand epilepsy Definitions proposed by the InternationalLeague Against Epilepsy (ILAE) and the InternationalBureau for Epilepsy (IBE) Epilepsia 200546470ndash472

9 Agency for Healthcare Research and Quality (AHRQ)Management of newly diagnosed patients with epi-lepsy a systematic review of the literature (AHRQPublication No 01-E038) Rockville MD Agency forHealth Research Quality US Department of Healthand Human Services 2001

10 Hirtz D Ashwal S Berg A et al Practice parameterevaluating a first nonfebrile seizure in children Neu-rology 200055616ndash623

11 Jallon P Loiseau P Loiseau J Newly diagnosed unpro-voked epileptic seizures Presentation at diagnosis inCORALE study Epilepsia 200142464ndash475

12 Quality Standards Subcommittee and the Therapeuticsand Technology Assessment Subcommittee of theAmerican Academy of Neurology Clinical PracticeGuideline Process Manuel St Paul MN AmericanAcademy of Neurology 2004

13 Hauser WA Anderson VE Loewenson RB McRob-erts SM Seizure recurrence after a first unprovoked sei-zure N Engl J Med 1982307522ndash528

14 Hauser WA Rich SS Annegers JF Anderson VE Sei-zure recurrence after a 1st unprovoked seizure an ex-tended follow-up Neurology 1990401163ndash1170

15 van Donselaar CA Schimsheimer R Geerts AT et alValue of the electroencephalogram in adult patientswith untreated idiopathic first seizures Arch Neurol199249231ndash237

16 First Seizure Trial Group Randomized clinical trial onthe efficacy of antiepileptic drugs in reducing the risk ofrelapse after a first unprovoked tonic-clonic seizureNeurology 199343478ndash483

17 Bora I Seckin B Zarifoglu S Turin F Sadikogllu S OgulE Risk of recurrence after first unprovoked tonic-clonicseizure in adults J Neurol 1995242157ndash163

18 Hopkins A Garman A Clarke C The first seizure inadult life Value of clinical features electroencephalog-raphy and computerized tomographic scanning in pre-diction of seizure recurrence Lancet 1988721ndash726

19 Das CP Sawhney IMS Lal V et al Risk of recurrenceof seizures following a single unprovoked idiopathicseizure Neurol India 200046357ndash360

2006 Neurology 69 November 20 2007

RETIRED

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

RETIRED

age populations were reviewed for data pertaining to pa-tients 18 years of age when possible

Literature search results In our search we found 793articles and obtained all in abstract form Each abstract wasreviewed by two committee members We identified 157 ar-ticles for review of the full text article (an article was in-cluded for review if selected by at least one committeemember based on criteria in appendix 4)

These 157 full text articles were obtained and reviewedby two committee members using established criteria Arti-cles were accepted or rejected when agreed on by both re-viewers using inclusion and exclusion criteria in appendix 4When there was disagreement between the reviewers a thirdreviewer cast the determining vote There were 10 instancesof disagreement resolved by a third party Of the 157 articlesreviewed 39 were selected as acceptable An additional 33studies from the same time period were identified from re-view articles and other sources these were subjected to thesame process and 14 were selected for inclusion

Each accepted study was abstracted by one investigatorand agreed to by a second Key data elements sought forextraction from each study included study patient and in-tervention characteristics In addition for all diagnostictests sensitivity specificity and positive and negative pre-dictive value with its gold standard were sought All eligiblearticles were scored on features pertinent to study designexecution and reporting with a range of possible scores asstandardized by the American Academy of Neurology Qual-ity Standards Subcommittee12

In order to score the evidence we used the Method ofScreening Intervention and Prognosis approved by the Qual-ity Standards Subcommittee (QSS) of the AAN (Appendices2 and 3)12 Of the 53 articles one was ranked as Class I 11 asClass II and the remaining 41 as Class III or IV (see appendixE-1 on theNeurologyreg Web site at wwwneurologyorg)

Goals of immediate evaluation of a first seizure Af-ter an adult who presents with a first seizure isstabilized and returns to baseline function a phy-sician must determine if the event was a seizureand if so whether this was the first such event Ahistory of prior seizures supports a diagnosis ofepilepsy and indicates a higher risk for seizure re-currence210 A careful and complete history phys-ical and neurologic examination are criticallyimportant at the time of initial presentation Thedetermination as to whether a seizure occurred istypically based on the history obtained from a re-liable observer The approach in adults is similarto what was presented in the practice parameterfor evaluating a first nonfebrile seizure in chil-dren2 as noted in appendix 5 A good historyphysical and neurologic examination may allowa physician to make the diagnosis of a seizurewithout additional diagnostic or laboratory test-ing However adults as well as children canpresent with episodic disorders that can be con-fused with seizures or epilepsy Such episodic dis-orders include syncope migraine drug reactionor intoxication and mental disorders such as psy-

chogenic seizures5 No single test clinical findingor symptom is reliable in discriminating betweenan initial seizure and such nonepileptic events Inaddition the reliability of witnesses to the event isvariable and witnesses are not always available

The next goal of assessment is to determine acause for the seizure For some patients the historyphysical and neurologic examination prove ade-quate to discern a probable cause or provide infor-mation to guide the physician to consider otherdiagnostic testing Some disorders causing seizuresrequire prompt diagnosis and acute treatment andothers strongly influence prognosis and impact deci-sions regarding initiation and maintenance of anti-epileptic drug therapy In particular provokedseizures are the result of acute precipitating disor-ders such as meningitis intoxications trauma ormetabolic derangements including hypoglycemiaand may require prompt intervention to reverse po-tentially damaging dangerous causes In contrastunprovoked seizuresmay also have causes but theseare not acute precipitating conditions requiring im-mediate action Their basis may be cryptogenic (noknown cause) remote symptomatic (due to a pre-existing brain injury or lesion such as a tumor orstroke) or idiopathic (genetic)6

ANALYSIS OF EVIDENCE Electroencephalogra-phy Should an EEG be routinely ordered in an adult pre-sentingwithanapparentunprovoked first seizureEvidence

Of 1 Class I and 10 Class II articles reviewed (witha total of 1766 patients) (table 1)13-24 assessing theyield of EEG EEGs were reported as abnormalin 12 to 73 (average yield 51) and reportedas significantly abnormal in 8 to 50 (average29) The abnormality considered as significant byauthors was the presence of epileptiform activity inthe form of spikes or sharp waves as interpreted bythe local or reading electroencephalographer in pa-tients clinically judged to present with a new onsetseizure evidence similar to that reported for chil-dren10 This yield is substantial However it is alsoclear from the evidence that a normal EEG does notexclude the presence of a seizure disorder Indeedon average about 50 of individuals clinically diag-nosed with a seizure have a normal EEG (table 1)

With respect to the value of the EEG as a predic-tor of risk of seizure recurrence and prognosis stud-ies support the EEG as useful (table 1) This issimilar to findings reported in children10 Of the oneClass I13-14 and seven Class II studies15-21 that con-sider the EEG as a predictor of risk of seizure recur-rence five studies a majority including the onlyClass I study13-15 demonstrate that epileptiformEEG activity in particular generalized spike andwave type discharges or focal spikes are associated

1998 Neurology 69 November 20 2007

RETIRED

RETIREDwith a greater risk for seizure recurrence13-161921

Therefore the weight of the evidence supports theEEG as of value in predicting seizure recurrence

What is this degree of increased risk for seizurerecurrence associated with epileptiform EEG ab-normalities In the only Class I study patientswith idiopathic seizure disorders and generalizedspike wave EEG abnormalities had an actual550 rate of seizure recurrence at 60 monthsfollow-up while the expected recurrence rate forthese patients was calculated in the same study tobe less 482 a small but statistically significantdifference1314 A meta-analysis in adults and chil-dren by Berg and Shinnar found that individualswith epileptiform EEG abnormalities were morelikely to have a seizure recurrence with a pooledrelative risk of 20 (95CI 16 26)2 and this issimilar to a more recent study16 Our own meta-analysis of the Class I and II studies (table 1)shows similar results with the estimated post-testprobability of a seizure recurrence in patientswith epileptiform EEG abnormalities of 495compared to only 274 in individuals whoseEEGs are completely normal The data show nosignificance for other more nonspecific EEG ab-normalities such as focal or diffuse slowing forpredicting seizure recurrence Several Class IIIand IV articles are more strongly in support ofepileptiform EEG abnormalities as a predictor ofseizure recurrence risk

Some of the study variability in risks for sei-zure recurrence may relate to the timing of theEEG time of entry into the clinical study after theinitial event duration of the recording or treat-

ment with antiepileptic medications In regard tothe timeframe of the EEGs this was specified inonly three studies (table 1)172024 and varied fromwithin 48 hours20 to a mean of 150 (range 1 to 36)days24 following the presenting seizure The ef-fects of antiepileptic drug treatment were specifi-cally considered in one study that carefullyrandomized treatment it still found the EEG pre-dictive of seizure recurrence after a first seizure16

Furthermore if there were a treatment bias wewould expect it to be a bias toward demonstrat-ing the EEG as less predictive of seizure recur-rence since patients with abnormal EEGs wouldmore likely get treated with antiepileptic drugsthat could in turn reduce seizure recurrence2

Studies of patients who presented with a his-tory of previous seizures were excluded in ouranalysis It is well described that the EEG is valu-able for establishing the nature of a seizure disor-der and guiding optimal therapy in such patientsor in patients with recurrent seizures or estab-lished epilepsy24825 For example specific gener-alized spike wave abnormalities on EEG arenoted to be of particular importance in properlydiagnosing determining prognosis and guidingtherapy for patients with juvenile myoclonic epi-lepsy and absence epilepsy These types of pri-mary generalized epilepsies are associated with apoor response to some antiepileptic drugs but ex-cellent response to others241025

Conclusion For adults presenting with an appar-ent unprovoked first seizure analysis of the evi-dence from 1 Class I and 10 Class II studiesindicates that the EEG is probably helpful It has a

Table 1 EEG in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) ClassNo studied()

No abnormal()

No signabnormal ()

Duration offollow-up mo

17 147 (16 and over) II 147 (100) 82 (56) 51 (35) 30 to 54

19 76 (mean 20) II 76 (100) 16 (21) 12 (16) 12 to 24 (mean 19)

23 56 (16 to 89) II 43 (77) 14 (33) 8 (19) NA

16 397 (72 over 16) II 397 (100) 221 (56) 199 (50) 24

22 107 (17 and over) II 103 (96) 75 (73) 29 (28) NA

13 14 173 (70 over 19) I 168 (97) 74 (44) 13 (8)dagger 64 mean (1990)

18 306 (16 and over) II 295 (96) 158 (54) 79 (27) 48

20 132 (13 to 86 mean 33) II 132 (100) 59 (45) 39 (30) 48

24 91 (15 or older mean 50) II 91 (100) 63 (69) 19 (21) NA

21 157 (17 to 84) II 157 (100) 111 (71) 42 (27) 24

15 157 (15 to 85 mean 38) II 157 (100) 19 (12) 19 (12) 24

Total 1799 1766 (98) 892 (51) 510 (29)

Epileptiform activity as interpreted by the local or reading electroencephalographerdaggerSignificant abnormal is only generalized spike waveNA not applicable or not available

Neurology 69 November 20 2007 1999

RETIRED

RETIRED

substantial yield with about 29 of EEGs dem-onstrating significant abnormalities2526 and theseabnormalities predict the risk for seizure recur-rence In addition EEG is regarded as a standardfor the initial classification of seizures since itforms a basis for the ldquoclinical and electroencepha-lographic classification of epileptic seizuresrdquo2526

Recommendations 1 The EEG (routine) shouldbe considered as part of the neurodiagnostic eval-uation of the adult with an apparent unprovokedfirst seizure because it has a substantial yield(Level B)82526

2 The EEG (routine) should be considered aspart of the neurodiagnostic evaluation of theadult with an apparent unprovoked first seizurebecause it has value in determining the risk forseizure recurrence (Level B)

Neuroimaging studies Should a brain imaging study(CT or MRI) be routinely ordered in an adult presenting

with an apparent unprovoked first seizure Evidence Ofthe seven Class II studies1518-20222327 (table 2) thatconsidered the yield and value (with or withoutcontrast agents) of the CT or MRI in adults ini-tially presenting with a seizure (1092 patients) theCT was reported as abnormal in 1 to 57 (av-erage yield 15) and was significantly abnormalin 1 to 47 (average 10) These significantabnormalities affected patient management andincluded previously unrecognized brain tumorsvascular lesions and cerebral cysticercosisTherefore six studies using CT1518-202327 and oneusing CT and MRI22 were of value for assessingadult patients with a first seizure Two of theseClass II studies also indicated that the finding ofan abnormal CT was associated with a greaterrisk of seizure recurrence1820 but the other studiesdid not address this issue The 13 Class III and 8Class IV studies of brain imaging we consideredfurther support the value of the CT or MRI indetermining a seizure cause

Interestingly almost all of the Class II studies

report on CT rather than MRI (table 2) One rea-son for this is that some of the studies are olderbut another is that many of these studies are ofpatients seen acutely in emergency departmentswhere the CT is the procedure of choice becauseof the relative speed and ease of obtaining thestudy and its effectiveness in excluding cata-strophic problems that may require immediate at-tention7 MRIs seem to be done in a moreselective way and most of the studies consideringMRI are not prospective or well controlled limit-ing them as Class I or II evidence However oneMRI study of new onset seizures indicates thatthe yield of MRI is at least as high and likelyhigher than CT but that study was not classifiedas Class I or II evidence because it included pa-tients with provoked seizures and those present-ing after multiple seizures or with pre-existingepilepsy28 In general although these includedbrain imaging studies varying in their methodsand conclusions the great majority support thatneuroimaging can determine potentially impor-tant treatable causes of seizures in a significantnumber of patients particularly in older patientsA previous QSS study of this issue in children didnot recommend neuroimaging as a standard10 butour review of studies in adults indicates that neuro-imaging deserves greater consideration in adultsand particularly in older adults Not only is the po-tential to discover significant abnormalities such astumors greater in adults728 but the risk associatedwithMRIs is less since adults are much less likely torequire conscious sedation than are children10

A practice parameter from the AmericanAcademy of Neurology in 1996 also consideredthe issue of neuroimaging in the patient present-ing with seizures29 That study similarly con-cluded that the decision for neuroimaging andtiming of the study should be driven by individualclinical circumstances In general when an imag-ing study is necessary in a non-emergent situa-

Table 2 Neuroimaging with CT or MRI in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) Class Modality No studied () No abnormal ()No signabnormal ()

19 100 (mean 20) II CT 100 (100) 17 (17) 17 (17)

23 56 (16 to 89) II CT 50 (89) 14 (28) 2 (4)

22 107 (17 and over) II CT and MRI 45 (42) 25 (56) 21 (47)

18 408 (16 and over) II CT 375 (92) 3 (1) 3 (1)

20 132 (13 to 86 mean 33) II CT 85 (64) 9 (11) 9 (11)

27 132 (adults mean 46) II CT 119 (90) 68 (57) 34 (29)

15 157 (15 to 85 mean 38) II CT 154 (98) 4 (3) 4 (3)

Total 1092 928 (85) 140 (15) 90 (10)

2000 Neurology 69 November 20 2007

RETIRED

RETIRED

tion MRI is more sensitive and is more likely toshow significant abnormalities than is CT91029

Conclusion For adults presenting initially with anapparent unprovoked first seizure the evidencefrom seven Class II studies indicates that a brain im-aging study either a CT orMRI is probably usefulIt has a significant yield of about 10 which maylead to the diagnosis of disorders such as a braintumor stroke cysticercosis or other structural le-sions and may have some value in determining therisk for seizure recurrence

Recommendation Brain imaging using CT orMRI should be considered as part of the neurodi-agnostic evaluation of adults presenting with anapparent unprovoked first seizure (Level B)

Laboratory studies Based on review of clinicalstudies expert opinion and position papers wechose to assess the value of the following labora-tory tests blood glucose blood counts electro-lytes lumbar puncture and toxicology screeningin adults with a first seizure79182330-32

Should blood counts blood glucose and electrolytepanels be routinely ordered in an adult with an apparent

unprovoked first seizure EvidenceOf the two Class IIstudies1823 (464 patients) (table 3) that assessedthe yield and value of blood counts blood glu-cose and electrolyte panels abnormalities werereported in from 0 to 15 for each of these testsbut no clinically significant abnormalities werenoted by the authors (table 3) In one Class IIstudy blood count abnormalities were reportedin 15 of patients but all these were judged bythe author to be incidental and clinically insignif-icant such as minor increased white blood cellcounts that resolved23 That same study reporteda 10 incidence of hypoglycemia but again allthose abnormalities were judged as mild and clin-ically insignificant with regard to the seizure Inboth the Class II studies routine electrolytes wereobtained in a high percentage of patients with

particular attention to serum calcium and sodium(table 3) Serum calcium was normal in all pa-tients in which it was measured In one study so-dium abnormalities of hyponatremia were notedin 7 of patients but again all were judged bythe author not to be of clinical significance23 Theother Class II study also reported no significantsodium abnormalities18

There were four Class III studies that showed ahigher incidence of significant laboratory abnormal-ities particularly for serum sodium and glucose30-33

All these studies were based in emergency depart-ments and the patients were obviously ill withmorecomorbidities than those in our Class II studies1823

A large proportion were patients with acute symp-tomatic or provoked rather than apparent unpro-voked seizures as required by our inclusion criteria(appendix 4) Among the reported acute symptom-atic causes for the seizureswere alcoholwithdrawalacute stroke tumor sepsis or trauma Also somestudies had many patients with other exclusion cri-teria (appendix 4) such as focal neurologic deficitsand itwas not specifiedwhether patients returned totheir normal level of function after the seizure30-33

Despite these limitations and acknowledgment thatyield is very low7 these four Class III studies wereconsistent in proposing some value to routinescreening of blood glucose for hypoglycemia and se-rum electrolytes for hyponatremia because unantic-ipated and clinically relevant hypoglycemia andhyponatremia were found in about 1 of thesepatients30-33

The two Class II studies as well the four ClassIII we identified do not provide convincing evi-dence to support or refute significant value forblood glucose blood count and electrolytes asroutine for adults at initial presentation with epi-leptic seizures This finding is similar to the QSSreport on initial seizure evaluation in children10

and a previous evidence-based analysis10

Table 3 Laboratory diagnostic tests in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) Class Modality No studied ()No abnormal() 95 CI

No sign abnormal() 95 CI

23 56 (16 to 89) II Blood count 55 (98) 8 (15) 76ndash26 0 (0) 0ndash65

Electrolytes 55 (98) 4 (7) 29ndash173 0 (0) 0ndash65

Glucose 50 (89) 5 (10) 43ndash21 0 (0) 0ndash71

Calcium 35 (63) 0 (0) 0ndash99 0 (0) 0ndash99

18 408 (16 and older) II Blood count 371 (91) NA 0 (0) 0ndash1

Electrolytes calcium 371 (91) NA 0 (0) 0ndash1

Total 464

NA not available

Neurology 69 November 20 2007 2001

RETIRED

RETIRED

In general most studies emphasize that the his-tory and physical examination can guide decisionmaking because they often predict patients whowill demonstrate significantly abnormal labora-tory tests23 For example patients with persistingaltered mental status fever or focal neurologicdeficits warrant more extensive evaluation andare more likely to have abnormal laboratory find-ings7 A practice parameter from the AmericanCollege of Emergency Physicians which generallysupports our findings still recommends obtaininga blood glucose determination and serum sodiumin patients with a first seizure7 However thatrecommendation was based on evidence fromstudies that we excluded or downgraded as ourinclusion criteria require because they involvedlarge numbers of patients with obvious acutesymptomatic seizures

Conclusion Data from two Class II and fourClass III studies showed that in adults presentingwith an apparent unprovoked first seizure al-though some abnormal laboratory results are re-ported there is not sufficient evidence to supportor refute recommending routine testing of bloodglucose blood counts or electrolyte panels Thenecessity for such studies should be guided by spe-cific clinical circumstances based on the historyphysical and neurologic examination

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure bloodglucose blood counts and electrolyte panels(particularly sodium) may be helpful in specificclinical circumstances but there are insufficientdata to support or refute routine recommenda-tion of any of these laboratory tests (Level U)

Should a lumbar puncture be routinely performed inan adult presenting with an apparent unprovoked first sei-

zure Evidence There were no Class I or II studiesbut of the two Class III studies considering thevalue of a lumbar puncture in patients with a firstseizure both were from emergency departmentsand reported significant lumbar puncture abnor-malities in up to 8 of patients3033 However notall patients in these two studies had unprovokedseizures or returned to baseline function as ourinclusion criteria require A substantial numberhad acute symptomatic seizures and level of con-sciousness was not well specified Another majorlimitation of these studies is that lumbar punc-tures were performed selectively They were per-formed in only 68 of the patients in one study30

and 24 of the other33 and this depended on suchpresenting clinical features as fever or at the dis-cretion of the staff Consequently lumbar punc-tures were not routinely performed and the

decision to perform a lumbar puncture was di-rected by the clinical history and physical find-ings Moreover in one study the determination ofwhether the lumbar puncture showed a ldquosignifi-cantrdquo abnormality was based largely on whetherthe patient was admitted to the hospital ratherthan whether the test showed a cause for the sei-zure30 In other studies of first seizures that ex-cluded patients with acute symptomatic seizureslumbar puncture was not considered as an impor-tant routine test1823

In particular there are no Class I II or otherconvincing studies to support a lumbar puncturein patients who are alert oriented afebrile andnot immunocompromised7 There is some evi-dence to support a lumbar puncture after a firstseizure in patients who are immunocompromisedeven if they are afebrile734

Conclusion Data from two Class III studies re-vealed significant abnormalities in up to 8 of amixed group of patients presenting to an emer-gency department with a first seizure Howeverthe studies selectively performed lumbar punc-tures based on clinical findings and included pa-tients who did not meet our inclusion criteriasuch as those with acute symptomatic causes fortheir seizures or who had not returned to theirnormal baseline function

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure lumbarpuncture may be helpful in specific clinical circum-stances such as patients who are febrile but thereare insufficient data to support or refute recom-mending routine lumbar puncture (Level U)

Should toxicologic screening be routinely ordered inan adult presenting with an apparent unprovoked first sei-

zure Evidence Seizures are reported as a consequenceof drug intoxication particularly with tricyclic anti-depressants cocaine and other stimulants735 In aseries of patients with acute medical complicationsof cocaine intoxication seizures often first seizuresaccounted for 10 of the presenting symptoms36

Several studies of emergency department admis-sions for first seizures including both acute symp-tomatic and unprovoked seizures indicated about3 may relate to drug toxicity or abuse73233 andone study advocated toxicology screening for allpatients with unexplained first seizures30 Still arecent evidence-based review by the AmericanAcademy of Emergency Physicians Clinical Pol-icy Committee considering management of adultpatients presenting to an emergency departmentwith seizures did not find sufficient evidence in thelimited existing prospective or retrospective studiesto recommend routine toxicology screening7

2002 Neurology 69 November 20 2007

RETIRED

RETIRED

In considering this issue we identified twoClass III studies considering the value of toxicol-ogy screening in adult patients presenting with afirst seizure3536 Although both studies reportedsome first seizures in patients with abnormal tox-icology screening neither study investigated theuse of routine screening in first seizure patients Inaddition the studies did not consider mainly un-provoked seizures in which the patient returnedto baseline function as our inclusion criteria spec-ify (appendix 4)

Conclusion In two Class III studies consideringthe value of toxicology screening in adult patientspresenting with a seizure some patients with ap-parent unprovoked first seizure were includedbut neither study investigated the use of routinetoxicology screening for such patients

Recommendation In the adult presenting with anapparent unprovoked seizure toxicology screen-ing may be helpful in specific clinical circum-stances but there are insufficient data to supportor refute a routine recommendation for toxicol-ogy screening (Level U)

CONCLUSIONS AND RECOMMENDATIONSDiagnosis of seizures determination of possiblecauses classification of seizure type and possibleseizure syndrome and assessment of the risk forrecurrence of seizures require timely selection andinterpretation of appropriate diagnostic tools in-cluding EEGs CTs or MRIs blood studies andlumbar puncture This report assesses the evi-dence available to answer the question ldquoWhat arethe most appropriate tools for evaluating a firstunprovoked seizure in adults at the time of initialpresentationrdquo

For the adult initially presenting with an ap-parent unprovoked seizure diagnostic evalua-tions influence aspects of patient care andmanagement including drug treatment patientand family counseling and the need for immedi-ate hospitalization and subsequent follow-upThis practice parameter reviews the published lit-erature concerning the yield and value of studiesfollowing an initial seizure in adults and classifiesthe strength of available evidence There is insuf-ficient Class I evidence to provide any recommen-dation with the highest degree of clinical certainty(see appendix 3) However there is sufficientClass I and Class II evidence that a routine EEGshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with an ap-parent unprovoked first seizure and predicts therisk of seizure recurrence (Level B) There is alsosufficient Class II evidence that a brain imaging

study either a CT or MRI helps determine thepresence of a significant abnormality such as abrain tumor Brain imaging using CT or MRIshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with a firstseizure (Level B) The CT has a major advantagebecause of the speed with which it can be ob-tained so its value is mainly in the emergency sit-uation but theMRI is regarded by expert opinionas having a higher yield and is the preferable pro-cedure in non-emergency or elective situa-tions91029 However both MRI and CT are ofvalue The decision to perform other studies in-cluding blood glucose blood counts electrolytepanels lumbar puncture or toxicology screeningfor the purpose of determining the cause of theseizure or identifying potentially treatable condi-tions will depend on the specific or individualizedclinical circumstances based on the history phys-ical and neurologic examination There are in-sufficient data to support or refute the use of anyone of these tests or procedures as a routine rec-ommendation (Level U)

The history physical and neurologic exami-nation are acknowledged by expert consensus tobe required for the initial diagnosis and classifica-tion of seizures26 These standards for the initialdiagnosis and classification of seizures form thebasis for the ldquoclinical and electroencephalo-graphic classification of epileptic seizuresrdquo21

Indeed information from the history includingthe type of the seizure21 together with findings fromthe physical and neurologic examination also aid inthe determination of a probable cause or risk of re-currence If the history reveals that there have beenother seizures in the past then this indicates epilepsyrather than a single isolated seizure Although thesubsequent diagnostic evaluation may be similarto that for a single unprovoked seizure there aredifferent prognostic and therapeutic implicationsfor a patient with epilepsy27 Evidence of pre-existing seizures provides information for the po-tential classification of an epilepsy syndrome25

Such information is of importance in establishingpossible causes associated comorbidities prog-nosis and optimal treatment of a seizure disorderor epilepsy syndrome2526 Moreover several fea-tures of the clinical history and physical and neuro-logic examination such as the presence of specificgeneralized or focal neurologic deficits are predic-tive of higher risks of seizure recurrence210

RECOMMENDATIONS FOR FUTURE RE-SEARCH The history physical and neurologicexaminations form the basis for the diagnosis and

Neurology 69 November 20 2007 2003

RETIRED

RETIRED

classification of epileptic seizures26 and this is ac-knowledged by expert consensus26 Still future re-search is warranted to determine the specificaspects of the history physical and neurologicexamination that are most useful in both initialdiagnosis of a seizure and subsequent manage-ment Also it would be helpful to understandhow this type of information should best guideperformance and timing of other neurodiagnosticor laboratory studies An international group re-cently recommended a broader definition of epi-lepsy as a brain disorder with a history of at leastone seizure and ldquocharacterized by an enduringpredisposition to generate epileptic seizuresrdquo8

This type of approach will require validation andinvestigation including determination of the truevalue of diagnostic studies such as EEG in estab-lishing the presence of an epileptic syndrome aftera first seizure In regard to the EEG one limita-tion of currently available studies is the variabilityin the timing of the EEG after incident seizureThere is some evidence in children that an EEGdone within 24 hours of a presenting seizure givesa higher yield of significant abnormalities10

Although all types of seizures are considered inthis analysis most patients had a convulsive sei-zure This may be because convulsive seizures aremore likely to prompt acute medical attentionthan a simple or complex partial seizure Morestudies specifically focusing attention on all typesof initial seizures not just convulsive seizures areneeded

Other factors deserve consideration in futureinvestigations For instance the costs of these var-ious tests do not receive much attention in studiesof patients with first seizure and should Also theelderly are a growing proportion of the generalpopulation and warrant special consideration infuture investigations Another unresolved issuerelates to the need for immediate hospitalizationof patients with a first seizure Some guidelinesexist but well designed studies are needed7

Prediction of seizure recurrence after a first un-provoked seizure is important to guide decisionsregarding antiepileptic drug therapy patientmanagement and counseling Many factors arereported to predict seizure recurrence includingthe etiology of the seizures and association withneurologic abnormalities2 Although the EEG andbrain imaging studies are reported to predict sei-zure recurrence studies vary in support of thatobservation and in respect to the specific EEG ab-normalities or brain imaging findings that aremost significant Some of this variability regard-ing EEG findings may relate to the timing of the

study after the initial event and to the decisionwhether to treat with antiepileptic medicationsThe value of specific findings on the EEG brainimaging and other variables deserve furtherstudy in regard to seizure recurrence risks Alsofuture studies of the clinical utility for such neuro-diagnostic procedures should employ standard-ized tests and prediction algorithms37

Future studies are needed to analyze the valueof expert input into the history physical neuro-logic examination or diagnostic testing such as theEEG or brain imaging The expertise and qualifica-tions of individuals interpreting such information ortests particularly the EEG or brain imaging varyconsiderably and may influence interpretationsand outcomes37 These are not simple ldquopositiverdquo orldquonegativerdquo tests like a pregnancy test they requireexpertise for proper interpretation Some other di-agnostic studies deserve further analysis in patientspresenting with new onset seizures In particularroutine electrolytes including serum sodium andglucosewere reported in some studies to be of valuebut this was not consistent or confirmed by otherreports Also toxicologic screening was reported asimportant in some limited series but was not stud-ied in a manner to allow adequate evidence-basedverification or refutation of its routine utility

Studies of patient management and counselingat the time of first diagnosis of a single seizure orepilepsy are recommended For example whichpatients should receive antiepileptic drug treat-ment and if so exactly what type and whenshould they get it Also when is the patient bestadvised regarding driving laws and other socialissues and how and by whom should this bedone This is not a trivial issue For instance aBritish study noted that only 21 of all adult firstseizure patients received the correct advice aboutdriving limitations23

We recommend that future studies address theissues that we raise here To best determine howto optimally evaluate the adult presenting with afirst seizure those studies should be structured toemphasize the use of large well-characterizedsamples clearly defined subjects and outcomesand standardized data collection methods

MISSION STATEMENT OF QSS The missionof the QSS is to prioritize develop and publishevidence-based practice parameters related to thediagnosis treatment and prognosis of neurologicdisorders

The QSS is committed to using the most rigor-ous methods available within our budget in col-

2004 Neurology 69 November 20 2007

RETIRED

RETIRED

laboration with other available AAN resourcesto most efficiently accomplish this mission

DISCLAIMER This statement is provided as aneducational service of the American Academy ofNeurology It is based on an assessment of currentscientific and clinical information It is not intendedto include all possible proper methods of care for aparticular neurologic problem or all legitimate crite-ria for choosing to use a specific procedure Neitheris it intended to exclude any reasonable alternativemethodologies The AAN recognizes that specificpatient care decisions are the prerogative of the pa-tient and physician caring for the patient based onall of the circumstances involved

CONFLICT OF INTEREST STATEMENT TheAmerican Academy of Neurology is committed toproducing independent critical and truthful clin-ical practice guidelines (CPGs) Significant effortsare made to minimize the potential for conflicts ofinterest to influence the recommendations of thisCPG To the extent possible the AAN keeps sep-arate those who have a financial stake in the suc-cess or failure of the products appraised in theCPGs and the developers of the guidelines Con-flict of interest forms were obtained from all au-thors and reviewed by an oversight committeeprior to project initiation AAN limits the participa-tion of authors with substantial conflicts of interestThe AAN forbids commercial participation in orfunding of guideline projects Drafts of the guide-lines have been reviewed by at least threeAANcom-mittees a network of neurologists Neurology peerreviewers and representatives from related fieldsTheAANGuidelineAuthorConflict of Interest Pol-icy can be viewed at wwwaancom

APPENDIX 1

Quality Standards Subcommittee members Jacque-line French MD FAAN (Co-Chair) Gary S Gronseth MD(Co-Chair) Charles E Argoff MD Stephen Ashwal MDFAAN (ex-officio) Christopher Bever Jr MD MBA FAANJohn D England MD FAAN Gary M Franklin MD MPH(ex-officio) GaryH FridayMDMPH FAAN Larry BGold-stein MD FAAN Deborah Hirtz MD (ex-officio) Robert GHollowayMDMPH FAANDonald J IversonMD FAANLeslie A Morrison MD Clifford J Schostal MD David JThurmanMDMPHWilliam JWeinerMD FAAN SamuelWiebe MD

APPENDIX 2AAN classification of evidence for rating of screen-ing article

Class I A statistical population-based sample of patientsstudied at a uniform point in time (usually early) during the

course of the condition All patients undergo the intervention ofinterest The outcome if not objective is determined in an eval-uation that is masked to the patientsrsquo clinical presentations

Class II A statistical non-referral-clinic-based sample ofpatients studied at a uniform point in time (usually early)during the course of the condition Most patients undergothe intervention of interest The outcome if not objective isdetermined in an evaluation that is masked to the patientsrsquoclinical presentations

Class III A sample of patients studied during the course ofthe condition Some patients undergo the intervention of in-terest The outcome if not objective is determined in anevaluation by someone other than the treating physician

Class IV Expert opinion case reports or any study notmeeting criteria for Class I to III

APPENDIX 3

Classification of recommendationsA Established as effective ineffective or harmful (or es-

tablished as usefulpredictive or not usefulpredictive)for the given condition in the specified population(Level A rating requires at least two consistent Class Istudies)

B Probably effective ineffective or harmful (or probablyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level B ratingrequires at least one Class I study or at least two consis-tent Class II studies)

C Possibly effective ineffective or harmful (or possiblyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level C ratingrequires at least one Class II study or two consistentClass III studies)

U Data inadequate or conflicting given current knowl-edge treatment (test predictor) is unproven (Studiesnot meeting criteria for Class IndashClass III)

In exceptional cases one convincing Class I study may suf-fice for an ldquoArdquo recommendation if 1) all criteria are met themagnitude of effect is large (relative rate improved outcome5 and the lower limit of the confidence interval is 2)

APPENDIX 4

Exclusion criteria

bull Letters and case reportsbull Non-English language studiesbull Animal studiesbull Pharmacodynamicpharmacokinetic studiesbull Studies dealing primarily with established or chronicepilepsy patients

bull Studies dealing primarily with acute provoked seizures

Inclusion criteria

bull Study designs observational (prospective retrospectiveand cross-sectional) or interventional (randomized con-trolled trials [RCTs] nonrandomized controlled trials[nRCTs] and uncontrolled case series [UCS])

bull At least 10 patients adults with a first seizure a first pre-sentation with epilepsy or seizures or a first diagnosis ofepilepsy

Neurology 69 November 20 2007 2005

RETIRED

RETIRED

bull Studies addressing any of the following diagnostic in-terventions neuropsychological assessment imagingwith CT MRI or PET scans EEG (standard videoinvasive ambulatory) lumbar puncture toxic screensand laboratory tests (hematologybiochemistry)

bull Patients should be over the age of 18 or the studyshould include a substantial proportion of subjectsover the age of 18

bull Studies reported in English onlybull Studies dealing primarily with apparent unprovokedfirst seizures

APPENDIX 5

Outline for seizure assessment Features of a seizureAssociated factors

AgeMedical history ndash previous history of similar episodesprior stroke brain tumor systemic illness mental ill-ness drug or alcohol abuse

Family historyDevelopmental statusBehaviorHealth at seizure onset ndash febrile ill exposed to illnesscomplaints of not feeling well sleep deprived

Precipitating events other than illness ndash trauma alcoholmedications illicit drugs toxins

Symptoms during seizure (ictal)Aura Subjective sensationsBehavior Mood or behavioral changes before the seizurePreictal symptoms Described by patient or witnessedVocal Cry or gasp slurring of words garbled speechMotor Head or eye turning eye deviation posturingjerking (rhythmic) stiffening automatisms (purpose-less repetitive movements such as picking at clothinglip smacking) generalized or focal movements

Respiration Change in breathing pattern cessation ofbreathing cyanosis

Autonomic Pupillary dilation drooling change in respira-tory or heart rate incontinence pallor vomiting

Loss of consciousness or inability to understand or speak

Symptoms following a seizure (postictal)Amnesia for eventsConfusionLethargySleepinessHeadaches and muscle achesTransient focal weakness (Toddrsquos paresis)Nausea or vomitingBiting of tongue

Adapted from Hirtz et al10

Received February 20 2007 Accepted in final form June 72007

REFERENCES1 Hauser WA Hesdorffer DC Epilepsy frequency

causes and consequences New York Demos Publica-tions 1990

2 Berg AT Shinnar S The risk of seizure recurrence fol-lowing a first unprovoked seizure a quantitative re-view Neurology 199141965ndash972

3 Begley CE Famulari M Annegers JF et al The cost ofepilepsy in the United States an estimate frompopulation- based and survey data Epilepsia 200041342ndash352

4 Grunewald RA Chroni E Panayiotopoulos CP De-layed diagnosis of juvenile myoclonic epilepsy J Neu-rol Neurosurg Psychiatry 199255497ndash499

5 Krumholz A Nonepileptic seizures diagnosis andmanagement Neurology 1999S76ndash83

6 Commission on Epidemiology and Prognosis Interna-tional League Against Epilepsy Guidelines for epidemio-logical studies on epilepsy Epilepsia 199334592ndash596

7 American College of Emergency Physicians (ACEP)Clinical Policies Committee and the Clinical PoliciesSubcommittee on Seizures Clinical Policy Critical is-sues in the evaluation of adult patients presenting to anemergency department with seizures Ann Emerg Med200443605ndash625

8 Fisher RS BoasWvE BlumeW et al Epileptic seizuresand epilepsy Definitions proposed by the InternationalLeague Against Epilepsy (ILAE) and the InternationalBureau for Epilepsy (IBE) Epilepsia 200546470ndash472

9 Agency for Healthcare Research and Quality (AHRQ)Management of newly diagnosed patients with epi-lepsy a systematic review of the literature (AHRQPublication No 01-E038) Rockville MD Agency forHealth Research Quality US Department of Healthand Human Services 2001

10 Hirtz D Ashwal S Berg A et al Practice parameterevaluating a first nonfebrile seizure in children Neu-rology 200055616ndash623

11 Jallon P Loiseau P Loiseau J Newly diagnosed unpro-voked epileptic seizures Presentation at diagnosis inCORALE study Epilepsia 200142464ndash475

12 Quality Standards Subcommittee and the Therapeuticsand Technology Assessment Subcommittee of theAmerican Academy of Neurology Clinical PracticeGuideline Process Manuel St Paul MN AmericanAcademy of Neurology 2004

13 Hauser WA Anderson VE Loewenson RB McRob-erts SM Seizure recurrence after a first unprovoked sei-zure N Engl J Med 1982307522ndash528

14 Hauser WA Rich SS Annegers JF Anderson VE Sei-zure recurrence after a 1st unprovoked seizure an ex-tended follow-up Neurology 1990401163ndash1170

15 van Donselaar CA Schimsheimer R Geerts AT et alValue of the electroencephalogram in adult patientswith untreated idiopathic first seizures Arch Neurol199249231ndash237

16 First Seizure Trial Group Randomized clinical trial onthe efficacy of antiepileptic drugs in reducing the risk ofrelapse after a first unprovoked tonic-clonic seizureNeurology 199343478ndash483

17 Bora I Seckin B Zarifoglu S Turin F Sadikogllu S OgulE Risk of recurrence after first unprovoked tonic-clonicseizure in adults J Neurol 1995242157ndash163

18 Hopkins A Garman A Clarke C The first seizure inadult life Value of clinical features electroencephalog-raphy and computerized tomographic scanning in pre-diction of seizure recurrence Lancet 1988721ndash726

19 Das CP Sawhney IMS Lal V et al Risk of recurrenceof seizures following a single unprovoked idiopathicseizure Neurol India 200046357ndash360

2006 Neurology 69 November 20 2007

RETIRED

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

RETIREDwith a greater risk for seizure recurrence13-161921

Therefore the weight of the evidence supports theEEG as of value in predicting seizure recurrence

What is this degree of increased risk for seizurerecurrence associated with epileptiform EEG ab-normalities In the only Class I study patientswith idiopathic seizure disorders and generalizedspike wave EEG abnormalities had an actual550 rate of seizure recurrence at 60 monthsfollow-up while the expected recurrence rate forthese patients was calculated in the same study tobe less 482 a small but statistically significantdifference1314 A meta-analysis in adults and chil-dren by Berg and Shinnar found that individualswith epileptiform EEG abnormalities were morelikely to have a seizure recurrence with a pooledrelative risk of 20 (95CI 16 26)2 and this issimilar to a more recent study16 Our own meta-analysis of the Class I and II studies (table 1)shows similar results with the estimated post-testprobability of a seizure recurrence in patientswith epileptiform EEG abnormalities of 495compared to only 274 in individuals whoseEEGs are completely normal The data show nosignificance for other more nonspecific EEG ab-normalities such as focal or diffuse slowing forpredicting seizure recurrence Several Class IIIand IV articles are more strongly in support ofepileptiform EEG abnormalities as a predictor ofseizure recurrence risk

Some of the study variability in risks for sei-zure recurrence may relate to the timing of theEEG time of entry into the clinical study after theinitial event duration of the recording or treat-

ment with antiepileptic medications In regard tothe timeframe of the EEGs this was specified inonly three studies (table 1)172024 and varied fromwithin 48 hours20 to a mean of 150 (range 1 to 36)days24 following the presenting seizure The ef-fects of antiepileptic drug treatment were specifi-cally considered in one study that carefullyrandomized treatment it still found the EEG pre-dictive of seizure recurrence after a first seizure16

Furthermore if there were a treatment bias wewould expect it to be a bias toward demonstrat-ing the EEG as less predictive of seizure recur-rence since patients with abnormal EEGs wouldmore likely get treated with antiepileptic drugsthat could in turn reduce seizure recurrence2

Studies of patients who presented with a his-tory of previous seizures were excluded in ouranalysis It is well described that the EEG is valu-able for establishing the nature of a seizure disor-der and guiding optimal therapy in such patientsor in patients with recurrent seizures or estab-lished epilepsy24825 For example specific gener-alized spike wave abnormalities on EEG arenoted to be of particular importance in properlydiagnosing determining prognosis and guidingtherapy for patients with juvenile myoclonic epi-lepsy and absence epilepsy These types of pri-mary generalized epilepsies are associated with apoor response to some antiepileptic drugs but ex-cellent response to others241025

Conclusion For adults presenting with an appar-ent unprovoked first seizure analysis of the evi-dence from 1 Class I and 10 Class II studiesindicates that the EEG is probably helpful It has a

Table 1 EEG in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) ClassNo studied()

No abnormal()

No signabnormal ()

Duration offollow-up mo

17 147 (16 and over) II 147 (100) 82 (56) 51 (35) 30 to 54

19 76 (mean 20) II 76 (100) 16 (21) 12 (16) 12 to 24 (mean 19)

23 56 (16 to 89) II 43 (77) 14 (33) 8 (19) NA

16 397 (72 over 16) II 397 (100) 221 (56) 199 (50) 24

22 107 (17 and over) II 103 (96) 75 (73) 29 (28) NA

13 14 173 (70 over 19) I 168 (97) 74 (44) 13 (8)dagger 64 mean (1990)

18 306 (16 and over) II 295 (96) 158 (54) 79 (27) 48

20 132 (13 to 86 mean 33) II 132 (100) 59 (45) 39 (30) 48

24 91 (15 or older mean 50) II 91 (100) 63 (69) 19 (21) NA

21 157 (17 to 84) II 157 (100) 111 (71) 42 (27) 24

15 157 (15 to 85 mean 38) II 157 (100) 19 (12) 19 (12) 24

Total 1799 1766 (98) 892 (51) 510 (29)

Epileptiform activity as interpreted by the local or reading electroencephalographerdaggerSignificant abnormal is only generalized spike waveNA not applicable or not available

Neurology 69 November 20 2007 1999

RETIRED

RETIRED

substantial yield with about 29 of EEGs dem-onstrating significant abnormalities2526 and theseabnormalities predict the risk for seizure recur-rence In addition EEG is regarded as a standardfor the initial classification of seizures since itforms a basis for the ldquoclinical and electroencepha-lographic classification of epileptic seizuresrdquo2526

Recommendations 1 The EEG (routine) shouldbe considered as part of the neurodiagnostic eval-uation of the adult with an apparent unprovokedfirst seizure because it has a substantial yield(Level B)82526

2 The EEG (routine) should be considered aspart of the neurodiagnostic evaluation of theadult with an apparent unprovoked first seizurebecause it has value in determining the risk forseizure recurrence (Level B)

Neuroimaging studies Should a brain imaging study(CT or MRI) be routinely ordered in an adult presenting

with an apparent unprovoked first seizure Evidence Ofthe seven Class II studies1518-20222327 (table 2) thatconsidered the yield and value (with or withoutcontrast agents) of the CT or MRI in adults ini-tially presenting with a seizure (1092 patients) theCT was reported as abnormal in 1 to 57 (av-erage yield 15) and was significantly abnormalin 1 to 47 (average 10) These significantabnormalities affected patient management andincluded previously unrecognized brain tumorsvascular lesions and cerebral cysticercosisTherefore six studies using CT1518-202327 and oneusing CT and MRI22 were of value for assessingadult patients with a first seizure Two of theseClass II studies also indicated that the finding ofan abnormal CT was associated with a greaterrisk of seizure recurrence1820 but the other studiesdid not address this issue The 13 Class III and 8Class IV studies of brain imaging we consideredfurther support the value of the CT or MRI indetermining a seizure cause

Interestingly almost all of the Class II studies

report on CT rather than MRI (table 2) One rea-son for this is that some of the studies are olderbut another is that many of these studies are ofpatients seen acutely in emergency departmentswhere the CT is the procedure of choice becauseof the relative speed and ease of obtaining thestudy and its effectiveness in excluding cata-strophic problems that may require immediate at-tention7 MRIs seem to be done in a moreselective way and most of the studies consideringMRI are not prospective or well controlled limit-ing them as Class I or II evidence However oneMRI study of new onset seizures indicates thatthe yield of MRI is at least as high and likelyhigher than CT but that study was not classifiedas Class I or II evidence because it included pa-tients with provoked seizures and those present-ing after multiple seizures or with pre-existingepilepsy28 In general although these includedbrain imaging studies varying in their methodsand conclusions the great majority support thatneuroimaging can determine potentially impor-tant treatable causes of seizures in a significantnumber of patients particularly in older patientsA previous QSS study of this issue in children didnot recommend neuroimaging as a standard10 butour review of studies in adults indicates that neuro-imaging deserves greater consideration in adultsand particularly in older adults Not only is the po-tential to discover significant abnormalities such astumors greater in adults728 but the risk associatedwithMRIs is less since adults are much less likely torequire conscious sedation than are children10

A practice parameter from the AmericanAcademy of Neurology in 1996 also consideredthe issue of neuroimaging in the patient present-ing with seizures29 That study similarly con-cluded that the decision for neuroimaging andtiming of the study should be driven by individualclinical circumstances In general when an imag-ing study is necessary in a non-emergent situa-

Table 2 Neuroimaging with CT or MRI in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) Class Modality No studied () No abnormal ()No signabnormal ()

19 100 (mean 20) II CT 100 (100) 17 (17) 17 (17)

23 56 (16 to 89) II CT 50 (89) 14 (28) 2 (4)

22 107 (17 and over) II CT and MRI 45 (42) 25 (56) 21 (47)

18 408 (16 and over) II CT 375 (92) 3 (1) 3 (1)

20 132 (13 to 86 mean 33) II CT 85 (64) 9 (11) 9 (11)

27 132 (adults mean 46) II CT 119 (90) 68 (57) 34 (29)

15 157 (15 to 85 mean 38) II CT 154 (98) 4 (3) 4 (3)

Total 1092 928 (85) 140 (15) 90 (10)

2000 Neurology 69 November 20 2007

RETIRED

RETIRED

tion MRI is more sensitive and is more likely toshow significant abnormalities than is CT91029

Conclusion For adults presenting initially with anapparent unprovoked first seizure the evidencefrom seven Class II studies indicates that a brain im-aging study either a CT orMRI is probably usefulIt has a significant yield of about 10 which maylead to the diagnosis of disorders such as a braintumor stroke cysticercosis or other structural le-sions and may have some value in determining therisk for seizure recurrence

Recommendation Brain imaging using CT orMRI should be considered as part of the neurodi-agnostic evaluation of adults presenting with anapparent unprovoked first seizure (Level B)

Laboratory studies Based on review of clinicalstudies expert opinion and position papers wechose to assess the value of the following labora-tory tests blood glucose blood counts electro-lytes lumbar puncture and toxicology screeningin adults with a first seizure79182330-32

Should blood counts blood glucose and electrolytepanels be routinely ordered in an adult with an apparent

unprovoked first seizure EvidenceOf the two Class IIstudies1823 (464 patients) (table 3) that assessedthe yield and value of blood counts blood glu-cose and electrolyte panels abnormalities werereported in from 0 to 15 for each of these testsbut no clinically significant abnormalities werenoted by the authors (table 3) In one Class IIstudy blood count abnormalities were reportedin 15 of patients but all these were judged bythe author to be incidental and clinically insignif-icant such as minor increased white blood cellcounts that resolved23 That same study reporteda 10 incidence of hypoglycemia but again allthose abnormalities were judged as mild and clin-ically insignificant with regard to the seizure Inboth the Class II studies routine electrolytes wereobtained in a high percentage of patients with

particular attention to serum calcium and sodium(table 3) Serum calcium was normal in all pa-tients in which it was measured In one study so-dium abnormalities of hyponatremia were notedin 7 of patients but again all were judged bythe author not to be of clinical significance23 Theother Class II study also reported no significantsodium abnormalities18

There were four Class III studies that showed ahigher incidence of significant laboratory abnormal-ities particularly for serum sodium and glucose30-33

All these studies were based in emergency depart-ments and the patients were obviously ill withmorecomorbidities than those in our Class II studies1823

A large proportion were patients with acute symp-tomatic or provoked rather than apparent unpro-voked seizures as required by our inclusion criteria(appendix 4) Among the reported acute symptom-atic causes for the seizureswere alcoholwithdrawalacute stroke tumor sepsis or trauma Also somestudies had many patients with other exclusion cri-teria (appendix 4) such as focal neurologic deficitsand itwas not specifiedwhether patients returned totheir normal level of function after the seizure30-33

Despite these limitations and acknowledgment thatyield is very low7 these four Class III studies wereconsistent in proposing some value to routinescreening of blood glucose for hypoglycemia and se-rum electrolytes for hyponatremia because unantic-ipated and clinically relevant hypoglycemia andhyponatremia were found in about 1 of thesepatients30-33

The two Class II studies as well the four ClassIII we identified do not provide convincing evi-dence to support or refute significant value forblood glucose blood count and electrolytes asroutine for adults at initial presentation with epi-leptic seizures This finding is similar to the QSSreport on initial seizure evaluation in children10

and a previous evidence-based analysis10

Table 3 Laboratory diagnostic tests in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) Class Modality No studied ()No abnormal() 95 CI

No sign abnormal() 95 CI

23 56 (16 to 89) II Blood count 55 (98) 8 (15) 76ndash26 0 (0) 0ndash65

Electrolytes 55 (98) 4 (7) 29ndash173 0 (0) 0ndash65

Glucose 50 (89) 5 (10) 43ndash21 0 (0) 0ndash71

Calcium 35 (63) 0 (0) 0ndash99 0 (0) 0ndash99

18 408 (16 and older) II Blood count 371 (91) NA 0 (0) 0ndash1

Electrolytes calcium 371 (91) NA 0 (0) 0ndash1

Total 464

NA not available

Neurology 69 November 20 2007 2001

RETIRED

RETIRED

In general most studies emphasize that the his-tory and physical examination can guide decisionmaking because they often predict patients whowill demonstrate significantly abnormal labora-tory tests23 For example patients with persistingaltered mental status fever or focal neurologicdeficits warrant more extensive evaluation andare more likely to have abnormal laboratory find-ings7 A practice parameter from the AmericanCollege of Emergency Physicians which generallysupports our findings still recommends obtaininga blood glucose determination and serum sodiumin patients with a first seizure7 However thatrecommendation was based on evidence fromstudies that we excluded or downgraded as ourinclusion criteria require because they involvedlarge numbers of patients with obvious acutesymptomatic seizures

Conclusion Data from two Class II and fourClass III studies showed that in adults presentingwith an apparent unprovoked first seizure al-though some abnormal laboratory results are re-ported there is not sufficient evidence to supportor refute recommending routine testing of bloodglucose blood counts or electrolyte panels Thenecessity for such studies should be guided by spe-cific clinical circumstances based on the historyphysical and neurologic examination

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure bloodglucose blood counts and electrolyte panels(particularly sodium) may be helpful in specificclinical circumstances but there are insufficientdata to support or refute routine recommenda-tion of any of these laboratory tests (Level U)

Should a lumbar puncture be routinely performed inan adult presenting with an apparent unprovoked first sei-

zure Evidence There were no Class I or II studiesbut of the two Class III studies considering thevalue of a lumbar puncture in patients with a firstseizure both were from emergency departmentsand reported significant lumbar puncture abnor-malities in up to 8 of patients3033 However notall patients in these two studies had unprovokedseizures or returned to baseline function as ourinclusion criteria require A substantial numberhad acute symptomatic seizures and level of con-sciousness was not well specified Another majorlimitation of these studies is that lumbar punc-tures were performed selectively They were per-formed in only 68 of the patients in one study30

and 24 of the other33 and this depended on suchpresenting clinical features as fever or at the dis-cretion of the staff Consequently lumbar punc-tures were not routinely performed and the

decision to perform a lumbar puncture was di-rected by the clinical history and physical find-ings Moreover in one study the determination ofwhether the lumbar puncture showed a ldquosignifi-cantrdquo abnormality was based largely on whetherthe patient was admitted to the hospital ratherthan whether the test showed a cause for the sei-zure30 In other studies of first seizures that ex-cluded patients with acute symptomatic seizureslumbar puncture was not considered as an impor-tant routine test1823

In particular there are no Class I II or otherconvincing studies to support a lumbar puncturein patients who are alert oriented afebrile andnot immunocompromised7 There is some evi-dence to support a lumbar puncture after a firstseizure in patients who are immunocompromisedeven if they are afebrile734

Conclusion Data from two Class III studies re-vealed significant abnormalities in up to 8 of amixed group of patients presenting to an emer-gency department with a first seizure Howeverthe studies selectively performed lumbar punc-tures based on clinical findings and included pa-tients who did not meet our inclusion criteriasuch as those with acute symptomatic causes fortheir seizures or who had not returned to theirnormal baseline function

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure lumbarpuncture may be helpful in specific clinical circum-stances such as patients who are febrile but thereare insufficient data to support or refute recom-mending routine lumbar puncture (Level U)

Should toxicologic screening be routinely ordered inan adult presenting with an apparent unprovoked first sei-

zure Evidence Seizures are reported as a consequenceof drug intoxication particularly with tricyclic anti-depressants cocaine and other stimulants735 In aseries of patients with acute medical complicationsof cocaine intoxication seizures often first seizuresaccounted for 10 of the presenting symptoms36

Several studies of emergency department admis-sions for first seizures including both acute symp-tomatic and unprovoked seizures indicated about3 may relate to drug toxicity or abuse73233 andone study advocated toxicology screening for allpatients with unexplained first seizures30 Still arecent evidence-based review by the AmericanAcademy of Emergency Physicians Clinical Pol-icy Committee considering management of adultpatients presenting to an emergency departmentwith seizures did not find sufficient evidence in thelimited existing prospective or retrospective studiesto recommend routine toxicology screening7

2002 Neurology 69 November 20 2007

RETIRED

RETIRED

In considering this issue we identified twoClass III studies considering the value of toxicol-ogy screening in adult patients presenting with afirst seizure3536 Although both studies reportedsome first seizures in patients with abnormal tox-icology screening neither study investigated theuse of routine screening in first seizure patients Inaddition the studies did not consider mainly un-provoked seizures in which the patient returnedto baseline function as our inclusion criteria spec-ify (appendix 4)

Conclusion In two Class III studies consideringthe value of toxicology screening in adult patientspresenting with a seizure some patients with ap-parent unprovoked first seizure were includedbut neither study investigated the use of routinetoxicology screening for such patients

Recommendation In the adult presenting with anapparent unprovoked seizure toxicology screen-ing may be helpful in specific clinical circum-stances but there are insufficient data to supportor refute a routine recommendation for toxicol-ogy screening (Level U)

CONCLUSIONS AND RECOMMENDATIONSDiagnosis of seizures determination of possiblecauses classification of seizure type and possibleseizure syndrome and assessment of the risk forrecurrence of seizures require timely selection andinterpretation of appropriate diagnostic tools in-cluding EEGs CTs or MRIs blood studies andlumbar puncture This report assesses the evi-dence available to answer the question ldquoWhat arethe most appropriate tools for evaluating a firstunprovoked seizure in adults at the time of initialpresentationrdquo

For the adult initially presenting with an ap-parent unprovoked seizure diagnostic evalua-tions influence aspects of patient care andmanagement including drug treatment patientand family counseling and the need for immedi-ate hospitalization and subsequent follow-upThis practice parameter reviews the published lit-erature concerning the yield and value of studiesfollowing an initial seizure in adults and classifiesthe strength of available evidence There is insuf-ficient Class I evidence to provide any recommen-dation with the highest degree of clinical certainty(see appendix 3) However there is sufficientClass I and Class II evidence that a routine EEGshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with an ap-parent unprovoked first seizure and predicts therisk of seizure recurrence (Level B) There is alsosufficient Class II evidence that a brain imaging

study either a CT or MRI helps determine thepresence of a significant abnormality such as abrain tumor Brain imaging using CT or MRIshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with a firstseizure (Level B) The CT has a major advantagebecause of the speed with which it can be ob-tained so its value is mainly in the emergency sit-uation but theMRI is regarded by expert opinionas having a higher yield and is the preferable pro-cedure in non-emergency or elective situa-tions91029 However both MRI and CT are ofvalue The decision to perform other studies in-cluding blood glucose blood counts electrolytepanels lumbar puncture or toxicology screeningfor the purpose of determining the cause of theseizure or identifying potentially treatable condi-tions will depend on the specific or individualizedclinical circumstances based on the history phys-ical and neurologic examination There are in-sufficient data to support or refute the use of anyone of these tests or procedures as a routine rec-ommendation (Level U)

The history physical and neurologic exami-nation are acknowledged by expert consensus tobe required for the initial diagnosis and classifica-tion of seizures26 These standards for the initialdiagnosis and classification of seizures form thebasis for the ldquoclinical and electroencephalo-graphic classification of epileptic seizuresrdquo21

Indeed information from the history includingthe type of the seizure21 together with findings fromthe physical and neurologic examination also aid inthe determination of a probable cause or risk of re-currence If the history reveals that there have beenother seizures in the past then this indicates epilepsyrather than a single isolated seizure Although thesubsequent diagnostic evaluation may be similarto that for a single unprovoked seizure there aredifferent prognostic and therapeutic implicationsfor a patient with epilepsy27 Evidence of pre-existing seizures provides information for the po-tential classification of an epilepsy syndrome25

Such information is of importance in establishingpossible causes associated comorbidities prog-nosis and optimal treatment of a seizure disorderor epilepsy syndrome2526 Moreover several fea-tures of the clinical history and physical and neuro-logic examination such as the presence of specificgeneralized or focal neurologic deficits are predic-tive of higher risks of seizure recurrence210

RECOMMENDATIONS FOR FUTURE RE-SEARCH The history physical and neurologicexaminations form the basis for the diagnosis and

Neurology 69 November 20 2007 2003

RETIRED

RETIRED

classification of epileptic seizures26 and this is ac-knowledged by expert consensus26 Still future re-search is warranted to determine the specificaspects of the history physical and neurologicexamination that are most useful in both initialdiagnosis of a seizure and subsequent manage-ment Also it would be helpful to understandhow this type of information should best guideperformance and timing of other neurodiagnosticor laboratory studies An international group re-cently recommended a broader definition of epi-lepsy as a brain disorder with a history of at leastone seizure and ldquocharacterized by an enduringpredisposition to generate epileptic seizuresrdquo8

This type of approach will require validation andinvestigation including determination of the truevalue of diagnostic studies such as EEG in estab-lishing the presence of an epileptic syndrome aftera first seizure In regard to the EEG one limita-tion of currently available studies is the variabilityin the timing of the EEG after incident seizureThere is some evidence in children that an EEGdone within 24 hours of a presenting seizure givesa higher yield of significant abnormalities10

Although all types of seizures are considered inthis analysis most patients had a convulsive sei-zure This may be because convulsive seizures aremore likely to prompt acute medical attentionthan a simple or complex partial seizure Morestudies specifically focusing attention on all typesof initial seizures not just convulsive seizures areneeded

Other factors deserve consideration in futureinvestigations For instance the costs of these var-ious tests do not receive much attention in studiesof patients with first seizure and should Also theelderly are a growing proportion of the generalpopulation and warrant special consideration infuture investigations Another unresolved issuerelates to the need for immediate hospitalizationof patients with a first seizure Some guidelinesexist but well designed studies are needed7

Prediction of seizure recurrence after a first un-provoked seizure is important to guide decisionsregarding antiepileptic drug therapy patientmanagement and counseling Many factors arereported to predict seizure recurrence includingthe etiology of the seizures and association withneurologic abnormalities2 Although the EEG andbrain imaging studies are reported to predict sei-zure recurrence studies vary in support of thatobservation and in respect to the specific EEG ab-normalities or brain imaging findings that aremost significant Some of this variability regard-ing EEG findings may relate to the timing of the

study after the initial event and to the decisionwhether to treat with antiepileptic medicationsThe value of specific findings on the EEG brainimaging and other variables deserve furtherstudy in regard to seizure recurrence risks Alsofuture studies of the clinical utility for such neuro-diagnostic procedures should employ standard-ized tests and prediction algorithms37

Future studies are needed to analyze the valueof expert input into the history physical neuro-logic examination or diagnostic testing such as theEEG or brain imaging The expertise and qualifica-tions of individuals interpreting such information ortests particularly the EEG or brain imaging varyconsiderably and may influence interpretationsand outcomes37 These are not simple ldquopositiverdquo orldquonegativerdquo tests like a pregnancy test they requireexpertise for proper interpretation Some other di-agnostic studies deserve further analysis in patientspresenting with new onset seizures In particularroutine electrolytes including serum sodium andglucosewere reported in some studies to be of valuebut this was not consistent or confirmed by otherreports Also toxicologic screening was reported asimportant in some limited series but was not stud-ied in a manner to allow adequate evidence-basedverification or refutation of its routine utility

Studies of patient management and counselingat the time of first diagnosis of a single seizure orepilepsy are recommended For example whichpatients should receive antiepileptic drug treat-ment and if so exactly what type and whenshould they get it Also when is the patient bestadvised regarding driving laws and other socialissues and how and by whom should this bedone This is not a trivial issue For instance aBritish study noted that only 21 of all adult firstseizure patients received the correct advice aboutdriving limitations23

We recommend that future studies address theissues that we raise here To best determine howto optimally evaluate the adult presenting with afirst seizure those studies should be structured toemphasize the use of large well-characterizedsamples clearly defined subjects and outcomesand standardized data collection methods

MISSION STATEMENT OF QSS The missionof the QSS is to prioritize develop and publishevidence-based practice parameters related to thediagnosis treatment and prognosis of neurologicdisorders

The QSS is committed to using the most rigor-ous methods available within our budget in col-

2004 Neurology 69 November 20 2007

RETIRED

RETIRED

laboration with other available AAN resourcesto most efficiently accomplish this mission

DISCLAIMER This statement is provided as aneducational service of the American Academy ofNeurology It is based on an assessment of currentscientific and clinical information It is not intendedto include all possible proper methods of care for aparticular neurologic problem or all legitimate crite-ria for choosing to use a specific procedure Neitheris it intended to exclude any reasonable alternativemethodologies The AAN recognizes that specificpatient care decisions are the prerogative of the pa-tient and physician caring for the patient based onall of the circumstances involved

CONFLICT OF INTEREST STATEMENT TheAmerican Academy of Neurology is committed toproducing independent critical and truthful clin-ical practice guidelines (CPGs) Significant effortsare made to minimize the potential for conflicts ofinterest to influence the recommendations of thisCPG To the extent possible the AAN keeps sep-arate those who have a financial stake in the suc-cess or failure of the products appraised in theCPGs and the developers of the guidelines Con-flict of interest forms were obtained from all au-thors and reviewed by an oversight committeeprior to project initiation AAN limits the participa-tion of authors with substantial conflicts of interestThe AAN forbids commercial participation in orfunding of guideline projects Drafts of the guide-lines have been reviewed by at least threeAANcom-mittees a network of neurologists Neurology peerreviewers and representatives from related fieldsTheAANGuidelineAuthorConflict of Interest Pol-icy can be viewed at wwwaancom

APPENDIX 1

Quality Standards Subcommittee members Jacque-line French MD FAAN (Co-Chair) Gary S Gronseth MD(Co-Chair) Charles E Argoff MD Stephen Ashwal MDFAAN (ex-officio) Christopher Bever Jr MD MBA FAANJohn D England MD FAAN Gary M Franklin MD MPH(ex-officio) GaryH FridayMDMPH FAAN Larry BGold-stein MD FAAN Deborah Hirtz MD (ex-officio) Robert GHollowayMDMPH FAANDonald J IversonMD FAANLeslie A Morrison MD Clifford J Schostal MD David JThurmanMDMPHWilliam JWeinerMD FAAN SamuelWiebe MD

APPENDIX 2AAN classification of evidence for rating of screen-ing article

Class I A statistical population-based sample of patientsstudied at a uniform point in time (usually early) during the

course of the condition All patients undergo the intervention ofinterest The outcome if not objective is determined in an eval-uation that is masked to the patientsrsquo clinical presentations

Class II A statistical non-referral-clinic-based sample ofpatients studied at a uniform point in time (usually early)during the course of the condition Most patients undergothe intervention of interest The outcome if not objective isdetermined in an evaluation that is masked to the patientsrsquoclinical presentations

Class III A sample of patients studied during the course ofthe condition Some patients undergo the intervention of in-terest The outcome if not objective is determined in anevaluation by someone other than the treating physician

Class IV Expert opinion case reports or any study notmeeting criteria for Class I to III

APPENDIX 3

Classification of recommendationsA Established as effective ineffective or harmful (or es-

tablished as usefulpredictive or not usefulpredictive)for the given condition in the specified population(Level A rating requires at least two consistent Class Istudies)

B Probably effective ineffective or harmful (or probablyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level B ratingrequires at least one Class I study or at least two consis-tent Class II studies)

C Possibly effective ineffective or harmful (or possiblyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level C ratingrequires at least one Class II study or two consistentClass III studies)

U Data inadequate or conflicting given current knowl-edge treatment (test predictor) is unproven (Studiesnot meeting criteria for Class IndashClass III)

In exceptional cases one convincing Class I study may suf-fice for an ldquoArdquo recommendation if 1) all criteria are met themagnitude of effect is large (relative rate improved outcome5 and the lower limit of the confidence interval is 2)

APPENDIX 4

Exclusion criteria

bull Letters and case reportsbull Non-English language studiesbull Animal studiesbull Pharmacodynamicpharmacokinetic studiesbull Studies dealing primarily with established or chronicepilepsy patients

bull Studies dealing primarily with acute provoked seizures

Inclusion criteria

bull Study designs observational (prospective retrospectiveand cross-sectional) or interventional (randomized con-trolled trials [RCTs] nonrandomized controlled trials[nRCTs] and uncontrolled case series [UCS])

bull At least 10 patients adults with a first seizure a first pre-sentation with epilepsy or seizures or a first diagnosis ofepilepsy

Neurology 69 November 20 2007 2005

RETIRED

RETIRED

bull Studies addressing any of the following diagnostic in-terventions neuropsychological assessment imagingwith CT MRI or PET scans EEG (standard videoinvasive ambulatory) lumbar puncture toxic screensand laboratory tests (hematologybiochemistry)

bull Patients should be over the age of 18 or the studyshould include a substantial proportion of subjectsover the age of 18

bull Studies reported in English onlybull Studies dealing primarily with apparent unprovokedfirst seizures

APPENDIX 5

Outline for seizure assessment Features of a seizureAssociated factors

AgeMedical history ndash previous history of similar episodesprior stroke brain tumor systemic illness mental ill-ness drug or alcohol abuse

Family historyDevelopmental statusBehaviorHealth at seizure onset ndash febrile ill exposed to illnesscomplaints of not feeling well sleep deprived

Precipitating events other than illness ndash trauma alcoholmedications illicit drugs toxins

Symptoms during seizure (ictal)Aura Subjective sensationsBehavior Mood or behavioral changes before the seizurePreictal symptoms Described by patient or witnessedVocal Cry or gasp slurring of words garbled speechMotor Head or eye turning eye deviation posturingjerking (rhythmic) stiffening automatisms (purpose-less repetitive movements such as picking at clothinglip smacking) generalized or focal movements

Respiration Change in breathing pattern cessation ofbreathing cyanosis

Autonomic Pupillary dilation drooling change in respira-tory or heart rate incontinence pallor vomiting

Loss of consciousness or inability to understand or speak

Symptoms following a seizure (postictal)Amnesia for eventsConfusionLethargySleepinessHeadaches and muscle achesTransient focal weakness (Toddrsquos paresis)Nausea or vomitingBiting of tongue

Adapted from Hirtz et al10

Received February 20 2007 Accepted in final form June 72007

REFERENCES1 Hauser WA Hesdorffer DC Epilepsy frequency

causes and consequences New York Demos Publica-tions 1990

2 Berg AT Shinnar S The risk of seizure recurrence fol-lowing a first unprovoked seizure a quantitative re-view Neurology 199141965ndash972

3 Begley CE Famulari M Annegers JF et al The cost ofepilepsy in the United States an estimate frompopulation- based and survey data Epilepsia 200041342ndash352

4 Grunewald RA Chroni E Panayiotopoulos CP De-layed diagnosis of juvenile myoclonic epilepsy J Neu-rol Neurosurg Psychiatry 199255497ndash499

5 Krumholz A Nonepileptic seizures diagnosis andmanagement Neurology 1999S76ndash83

6 Commission on Epidemiology and Prognosis Interna-tional League Against Epilepsy Guidelines for epidemio-logical studies on epilepsy Epilepsia 199334592ndash596

7 American College of Emergency Physicians (ACEP)Clinical Policies Committee and the Clinical PoliciesSubcommittee on Seizures Clinical Policy Critical is-sues in the evaluation of adult patients presenting to anemergency department with seizures Ann Emerg Med200443605ndash625

8 Fisher RS BoasWvE BlumeW et al Epileptic seizuresand epilepsy Definitions proposed by the InternationalLeague Against Epilepsy (ILAE) and the InternationalBureau for Epilepsy (IBE) Epilepsia 200546470ndash472

9 Agency for Healthcare Research and Quality (AHRQ)Management of newly diagnosed patients with epi-lepsy a systematic review of the literature (AHRQPublication No 01-E038) Rockville MD Agency forHealth Research Quality US Department of Healthand Human Services 2001

10 Hirtz D Ashwal S Berg A et al Practice parameterevaluating a first nonfebrile seizure in children Neu-rology 200055616ndash623

11 Jallon P Loiseau P Loiseau J Newly diagnosed unpro-voked epileptic seizures Presentation at diagnosis inCORALE study Epilepsia 200142464ndash475

12 Quality Standards Subcommittee and the Therapeuticsand Technology Assessment Subcommittee of theAmerican Academy of Neurology Clinical PracticeGuideline Process Manuel St Paul MN AmericanAcademy of Neurology 2004

13 Hauser WA Anderson VE Loewenson RB McRob-erts SM Seizure recurrence after a first unprovoked sei-zure N Engl J Med 1982307522ndash528

14 Hauser WA Rich SS Annegers JF Anderson VE Sei-zure recurrence after a 1st unprovoked seizure an ex-tended follow-up Neurology 1990401163ndash1170

15 van Donselaar CA Schimsheimer R Geerts AT et alValue of the electroencephalogram in adult patientswith untreated idiopathic first seizures Arch Neurol199249231ndash237

16 First Seizure Trial Group Randomized clinical trial onthe efficacy of antiepileptic drugs in reducing the risk ofrelapse after a first unprovoked tonic-clonic seizureNeurology 199343478ndash483

17 Bora I Seckin B Zarifoglu S Turin F Sadikogllu S OgulE Risk of recurrence after first unprovoked tonic-clonicseizure in adults J Neurol 1995242157ndash163

18 Hopkins A Garman A Clarke C The first seizure inadult life Value of clinical features electroencephalog-raphy and computerized tomographic scanning in pre-diction of seizure recurrence Lancet 1988721ndash726

19 Das CP Sawhney IMS Lal V et al Risk of recurrenceof seizures following a single unprovoked idiopathicseizure Neurol India 200046357ndash360

2006 Neurology 69 November 20 2007

RETIRED

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

RETIRED

substantial yield with about 29 of EEGs dem-onstrating significant abnormalities2526 and theseabnormalities predict the risk for seizure recur-rence In addition EEG is regarded as a standardfor the initial classification of seizures since itforms a basis for the ldquoclinical and electroencepha-lographic classification of epileptic seizuresrdquo2526

Recommendations 1 The EEG (routine) shouldbe considered as part of the neurodiagnostic eval-uation of the adult with an apparent unprovokedfirst seizure because it has a substantial yield(Level B)82526

2 The EEG (routine) should be considered aspart of the neurodiagnostic evaluation of theadult with an apparent unprovoked first seizurebecause it has value in determining the risk forseizure recurrence (Level B)

Neuroimaging studies Should a brain imaging study(CT or MRI) be routinely ordered in an adult presenting

with an apparent unprovoked first seizure Evidence Ofthe seven Class II studies1518-20222327 (table 2) thatconsidered the yield and value (with or withoutcontrast agents) of the CT or MRI in adults ini-tially presenting with a seizure (1092 patients) theCT was reported as abnormal in 1 to 57 (av-erage yield 15) and was significantly abnormalin 1 to 47 (average 10) These significantabnormalities affected patient management andincluded previously unrecognized brain tumorsvascular lesions and cerebral cysticercosisTherefore six studies using CT1518-202327 and oneusing CT and MRI22 were of value for assessingadult patients with a first seizure Two of theseClass II studies also indicated that the finding ofan abnormal CT was associated with a greaterrisk of seizure recurrence1820 but the other studiesdid not address this issue The 13 Class III and 8Class IV studies of brain imaging we consideredfurther support the value of the CT or MRI indetermining a seizure cause

Interestingly almost all of the Class II studies

report on CT rather than MRI (table 2) One rea-son for this is that some of the studies are olderbut another is that many of these studies are ofpatients seen acutely in emergency departmentswhere the CT is the procedure of choice becauseof the relative speed and ease of obtaining thestudy and its effectiveness in excluding cata-strophic problems that may require immediate at-tention7 MRIs seem to be done in a moreselective way and most of the studies consideringMRI are not prospective or well controlled limit-ing them as Class I or II evidence However oneMRI study of new onset seizures indicates thatthe yield of MRI is at least as high and likelyhigher than CT but that study was not classifiedas Class I or II evidence because it included pa-tients with provoked seizures and those present-ing after multiple seizures or with pre-existingepilepsy28 In general although these includedbrain imaging studies varying in their methodsand conclusions the great majority support thatneuroimaging can determine potentially impor-tant treatable causes of seizures in a significantnumber of patients particularly in older patientsA previous QSS study of this issue in children didnot recommend neuroimaging as a standard10 butour review of studies in adults indicates that neuro-imaging deserves greater consideration in adultsand particularly in older adults Not only is the po-tential to discover significant abnormalities such astumors greater in adults728 but the risk associatedwithMRIs is less since adults are much less likely torequire conscious sedation than are children10

A practice parameter from the AmericanAcademy of Neurology in 1996 also consideredthe issue of neuroimaging in the patient present-ing with seizures29 That study similarly con-cluded that the decision for neuroimaging andtiming of the study should be driven by individualclinical circumstances In general when an imag-ing study is necessary in a non-emergent situa-

Table 2 Neuroimaging with CT or MRI in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) Class Modality No studied () No abnormal ()No signabnormal ()

19 100 (mean 20) II CT 100 (100) 17 (17) 17 (17)

23 56 (16 to 89) II CT 50 (89) 14 (28) 2 (4)

22 107 (17 and over) II CT and MRI 45 (42) 25 (56) 21 (47)

18 408 (16 and over) II CT 375 (92) 3 (1) 3 (1)

20 132 (13 to 86 mean 33) II CT 85 (64) 9 (11) 9 (11)

27 132 (adults mean 46) II CT 119 (90) 68 (57) 34 (29)

15 157 (15 to 85 mean 38) II CT 154 (98) 4 (3) 4 (3)

Total 1092 928 (85) 140 (15) 90 (10)

2000 Neurology 69 November 20 2007

RETIRED

RETIRED

tion MRI is more sensitive and is more likely toshow significant abnormalities than is CT91029

Conclusion For adults presenting initially with anapparent unprovoked first seizure the evidencefrom seven Class II studies indicates that a brain im-aging study either a CT orMRI is probably usefulIt has a significant yield of about 10 which maylead to the diagnosis of disorders such as a braintumor stroke cysticercosis or other structural le-sions and may have some value in determining therisk for seizure recurrence

Recommendation Brain imaging using CT orMRI should be considered as part of the neurodi-agnostic evaluation of adults presenting with anapparent unprovoked first seizure (Level B)

Laboratory studies Based on review of clinicalstudies expert opinion and position papers wechose to assess the value of the following labora-tory tests blood glucose blood counts electro-lytes lumbar puncture and toxicology screeningin adults with a first seizure79182330-32

Should blood counts blood glucose and electrolytepanels be routinely ordered in an adult with an apparent

unprovoked first seizure EvidenceOf the two Class IIstudies1823 (464 patients) (table 3) that assessedthe yield and value of blood counts blood glu-cose and electrolyte panels abnormalities werereported in from 0 to 15 for each of these testsbut no clinically significant abnormalities werenoted by the authors (table 3) In one Class IIstudy blood count abnormalities were reportedin 15 of patients but all these were judged bythe author to be incidental and clinically insignif-icant such as minor increased white blood cellcounts that resolved23 That same study reporteda 10 incidence of hypoglycemia but again allthose abnormalities were judged as mild and clin-ically insignificant with regard to the seizure Inboth the Class II studies routine electrolytes wereobtained in a high percentage of patients with

particular attention to serum calcium and sodium(table 3) Serum calcium was normal in all pa-tients in which it was measured In one study so-dium abnormalities of hyponatremia were notedin 7 of patients but again all were judged bythe author not to be of clinical significance23 Theother Class II study also reported no significantsodium abnormalities18

There were four Class III studies that showed ahigher incidence of significant laboratory abnormal-ities particularly for serum sodium and glucose30-33

All these studies were based in emergency depart-ments and the patients were obviously ill withmorecomorbidities than those in our Class II studies1823

A large proportion were patients with acute symp-tomatic or provoked rather than apparent unpro-voked seizures as required by our inclusion criteria(appendix 4) Among the reported acute symptom-atic causes for the seizureswere alcoholwithdrawalacute stroke tumor sepsis or trauma Also somestudies had many patients with other exclusion cri-teria (appendix 4) such as focal neurologic deficitsand itwas not specifiedwhether patients returned totheir normal level of function after the seizure30-33

Despite these limitations and acknowledgment thatyield is very low7 these four Class III studies wereconsistent in proposing some value to routinescreening of blood glucose for hypoglycemia and se-rum electrolytes for hyponatremia because unantic-ipated and clinically relevant hypoglycemia andhyponatremia were found in about 1 of thesepatients30-33

The two Class II studies as well the four ClassIII we identified do not provide convincing evi-dence to support or refute significant value forblood glucose blood count and electrolytes asroutine for adults at initial presentation with epi-leptic seizures This finding is similar to the QSSreport on initial seizure evaluation in children10

and a previous evidence-based analysis10

Table 3 Laboratory diagnostic tests in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) Class Modality No studied ()No abnormal() 95 CI

No sign abnormal() 95 CI

23 56 (16 to 89) II Blood count 55 (98) 8 (15) 76ndash26 0 (0) 0ndash65

Electrolytes 55 (98) 4 (7) 29ndash173 0 (0) 0ndash65

Glucose 50 (89) 5 (10) 43ndash21 0 (0) 0ndash71

Calcium 35 (63) 0 (0) 0ndash99 0 (0) 0ndash99

18 408 (16 and older) II Blood count 371 (91) NA 0 (0) 0ndash1

Electrolytes calcium 371 (91) NA 0 (0) 0ndash1

Total 464

NA not available

Neurology 69 November 20 2007 2001

RETIRED

RETIRED

In general most studies emphasize that the his-tory and physical examination can guide decisionmaking because they often predict patients whowill demonstrate significantly abnormal labora-tory tests23 For example patients with persistingaltered mental status fever or focal neurologicdeficits warrant more extensive evaluation andare more likely to have abnormal laboratory find-ings7 A practice parameter from the AmericanCollege of Emergency Physicians which generallysupports our findings still recommends obtaininga blood glucose determination and serum sodiumin patients with a first seizure7 However thatrecommendation was based on evidence fromstudies that we excluded or downgraded as ourinclusion criteria require because they involvedlarge numbers of patients with obvious acutesymptomatic seizures

Conclusion Data from two Class II and fourClass III studies showed that in adults presentingwith an apparent unprovoked first seizure al-though some abnormal laboratory results are re-ported there is not sufficient evidence to supportor refute recommending routine testing of bloodglucose blood counts or electrolyte panels Thenecessity for such studies should be guided by spe-cific clinical circumstances based on the historyphysical and neurologic examination

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure bloodglucose blood counts and electrolyte panels(particularly sodium) may be helpful in specificclinical circumstances but there are insufficientdata to support or refute routine recommenda-tion of any of these laboratory tests (Level U)

Should a lumbar puncture be routinely performed inan adult presenting with an apparent unprovoked first sei-

zure Evidence There were no Class I or II studiesbut of the two Class III studies considering thevalue of a lumbar puncture in patients with a firstseizure both were from emergency departmentsand reported significant lumbar puncture abnor-malities in up to 8 of patients3033 However notall patients in these two studies had unprovokedseizures or returned to baseline function as ourinclusion criteria require A substantial numberhad acute symptomatic seizures and level of con-sciousness was not well specified Another majorlimitation of these studies is that lumbar punc-tures were performed selectively They were per-formed in only 68 of the patients in one study30

and 24 of the other33 and this depended on suchpresenting clinical features as fever or at the dis-cretion of the staff Consequently lumbar punc-tures were not routinely performed and the

decision to perform a lumbar puncture was di-rected by the clinical history and physical find-ings Moreover in one study the determination ofwhether the lumbar puncture showed a ldquosignifi-cantrdquo abnormality was based largely on whetherthe patient was admitted to the hospital ratherthan whether the test showed a cause for the sei-zure30 In other studies of first seizures that ex-cluded patients with acute symptomatic seizureslumbar puncture was not considered as an impor-tant routine test1823

In particular there are no Class I II or otherconvincing studies to support a lumbar puncturein patients who are alert oriented afebrile andnot immunocompromised7 There is some evi-dence to support a lumbar puncture after a firstseizure in patients who are immunocompromisedeven if they are afebrile734

Conclusion Data from two Class III studies re-vealed significant abnormalities in up to 8 of amixed group of patients presenting to an emer-gency department with a first seizure Howeverthe studies selectively performed lumbar punc-tures based on clinical findings and included pa-tients who did not meet our inclusion criteriasuch as those with acute symptomatic causes fortheir seizures or who had not returned to theirnormal baseline function

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure lumbarpuncture may be helpful in specific clinical circum-stances such as patients who are febrile but thereare insufficient data to support or refute recom-mending routine lumbar puncture (Level U)

Should toxicologic screening be routinely ordered inan adult presenting with an apparent unprovoked first sei-

zure Evidence Seizures are reported as a consequenceof drug intoxication particularly with tricyclic anti-depressants cocaine and other stimulants735 In aseries of patients with acute medical complicationsof cocaine intoxication seizures often first seizuresaccounted for 10 of the presenting symptoms36

Several studies of emergency department admis-sions for first seizures including both acute symp-tomatic and unprovoked seizures indicated about3 may relate to drug toxicity or abuse73233 andone study advocated toxicology screening for allpatients with unexplained first seizures30 Still arecent evidence-based review by the AmericanAcademy of Emergency Physicians Clinical Pol-icy Committee considering management of adultpatients presenting to an emergency departmentwith seizures did not find sufficient evidence in thelimited existing prospective or retrospective studiesto recommend routine toxicology screening7

2002 Neurology 69 November 20 2007

RETIRED

RETIRED

In considering this issue we identified twoClass III studies considering the value of toxicol-ogy screening in adult patients presenting with afirst seizure3536 Although both studies reportedsome first seizures in patients with abnormal tox-icology screening neither study investigated theuse of routine screening in first seizure patients Inaddition the studies did not consider mainly un-provoked seizures in which the patient returnedto baseline function as our inclusion criteria spec-ify (appendix 4)

Conclusion In two Class III studies consideringthe value of toxicology screening in adult patientspresenting with a seizure some patients with ap-parent unprovoked first seizure were includedbut neither study investigated the use of routinetoxicology screening for such patients

Recommendation In the adult presenting with anapparent unprovoked seizure toxicology screen-ing may be helpful in specific clinical circum-stances but there are insufficient data to supportor refute a routine recommendation for toxicol-ogy screening (Level U)

CONCLUSIONS AND RECOMMENDATIONSDiagnosis of seizures determination of possiblecauses classification of seizure type and possibleseizure syndrome and assessment of the risk forrecurrence of seizures require timely selection andinterpretation of appropriate diagnostic tools in-cluding EEGs CTs or MRIs blood studies andlumbar puncture This report assesses the evi-dence available to answer the question ldquoWhat arethe most appropriate tools for evaluating a firstunprovoked seizure in adults at the time of initialpresentationrdquo

For the adult initially presenting with an ap-parent unprovoked seizure diagnostic evalua-tions influence aspects of patient care andmanagement including drug treatment patientand family counseling and the need for immedi-ate hospitalization and subsequent follow-upThis practice parameter reviews the published lit-erature concerning the yield and value of studiesfollowing an initial seizure in adults and classifiesthe strength of available evidence There is insuf-ficient Class I evidence to provide any recommen-dation with the highest degree of clinical certainty(see appendix 3) However there is sufficientClass I and Class II evidence that a routine EEGshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with an ap-parent unprovoked first seizure and predicts therisk of seizure recurrence (Level B) There is alsosufficient Class II evidence that a brain imaging

study either a CT or MRI helps determine thepresence of a significant abnormality such as abrain tumor Brain imaging using CT or MRIshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with a firstseizure (Level B) The CT has a major advantagebecause of the speed with which it can be ob-tained so its value is mainly in the emergency sit-uation but theMRI is regarded by expert opinionas having a higher yield and is the preferable pro-cedure in non-emergency or elective situa-tions91029 However both MRI and CT are ofvalue The decision to perform other studies in-cluding blood glucose blood counts electrolytepanels lumbar puncture or toxicology screeningfor the purpose of determining the cause of theseizure or identifying potentially treatable condi-tions will depend on the specific or individualizedclinical circumstances based on the history phys-ical and neurologic examination There are in-sufficient data to support or refute the use of anyone of these tests or procedures as a routine rec-ommendation (Level U)

The history physical and neurologic exami-nation are acknowledged by expert consensus tobe required for the initial diagnosis and classifica-tion of seizures26 These standards for the initialdiagnosis and classification of seizures form thebasis for the ldquoclinical and electroencephalo-graphic classification of epileptic seizuresrdquo21

Indeed information from the history includingthe type of the seizure21 together with findings fromthe physical and neurologic examination also aid inthe determination of a probable cause or risk of re-currence If the history reveals that there have beenother seizures in the past then this indicates epilepsyrather than a single isolated seizure Although thesubsequent diagnostic evaluation may be similarto that for a single unprovoked seizure there aredifferent prognostic and therapeutic implicationsfor a patient with epilepsy27 Evidence of pre-existing seizures provides information for the po-tential classification of an epilepsy syndrome25

Such information is of importance in establishingpossible causes associated comorbidities prog-nosis and optimal treatment of a seizure disorderor epilepsy syndrome2526 Moreover several fea-tures of the clinical history and physical and neuro-logic examination such as the presence of specificgeneralized or focal neurologic deficits are predic-tive of higher risks of seizure recurrence210

RECOMMENDATIONS FOR FUTURE RE-SEARCH The history physical and neurologicexaminations form the basis for the diagnosis and

Neurology 69 November 20 2007 2003

RETIRED

RETIRED

classification of epileptic seizures26 and this is ac-knowledged by expert consensus26 Still future re-search is warranted to determine the specificaspects of the history physical and neurologicexamination that are most useful in both initialdiagnosis of a seizure and subsequent manage-ment Also it would be helpful to understandhow this type of information should best guideperformance and timing of other neurodiagnosticor laboratory studies An international group re-cently recommended a broader definition of epi-lepsy as a brain disorder with a history of at leastone seizure and ldquocharacterized by an enduringpredisposition to generate epileptic seizuresrdquo8

This type of approach will require validation andinvestigation including determination of the truevalue of diagnostic studies such as EEG in estab-lishing the presence of an epileptic syndrome aftera first seizure In regard to the EEG one limita-tion of currently available studies is the variabilityin the timing of the EEG after incident seizureThere is some evidence in children that an EEGdone within 24 hours of a presenting seizure givesa higher yield of significant abnormalities10

Although all types of seizures are considered inthis analysis most patients had a convulsive sei-zure This may be because convulsive seizures aremore likely to prompt acute medical attentionthan a simple or complex partial seizure Morestudies specifically focusing attention on all typesof initial seizures not just convulsive seizures areneeded

Other factors deserve consideration in futureinvestigations For instance the costs of these var-ious tests do not receive much attention in studiesof patients with first seizure and should Also theelderly are a growing proportion of the generalpopulation and warrant special consideration infuture investigations Another unresolved issuerelates to the need for immediate hospitalizationof patients with a first seizure Some guidelinesexist but well designed studies are needed7

Prediction of seizure recurrence after a first un-provoked seizure is important to guide decisionsregarding antiepileptic drug therapy patientmanagement and counseling Many factors arereported to predict seizure recurrence includingthe etiology of the seizures and association withneurologic abnormalities2 Although the EEG andbrain imaging studies are reported to predict sei-zure recurrence studies vary in support of thatobservation and in respect to the specific EEG ab-normalities or brain imaging findings that aremost significant Some of this variability regard-ing EEG findings may relate to the timing of the

study after the initial event and to the decisionwhether to treat with antiepileptic medicationsThe value of specific findings on the EEG brainimaging and other variables deserve furtherstudy in regard to seizure recurrence risks Alsofuture studies of the clinical utility for such neuro-diagnostic procedures should employ standard-ized tests and prediction algorithms37

Future studies are needed to analyze the valueof expert input into the history physical neuro-logic examination or diagnostic testing such as theEEG or brain imaging The expertise and qualifica-tions of individuals interpreting such information ortests particularly the EEG or brain imaging varyconsiderably and may influence interpretationsand outcomes37 These are not simple ldquopositiverdquo orldquonegativerdquo tests like a pregnancy test they requireexpertise for proper interpretation Some other di-agnostic studies deserve further analysis in patientspresenting with new onset seizures In particularroutine electrolytes including serum sodium andglucosewere reported in some studies to be of valuebut this was not consistent or confirmed by otherreports Also toxicologic screening was reported asimportant in some limited series but was not stud-ied in a manner to allow adequate evidence-basedverification or refutation of its routine utility

Studies of patient management and counselingat the time of first diagnosis of a single seizure orepilepsy are recommended For example whichpatients should receive antiepileptic drug treat-ment and if so exactly what type and whenshould they get it Also when is the patient bestadvised regarding driving laws and other socialissues and how and by whom should this bedone This is not a trivial issue For instance aBritish study noted that only 21 of all adult firstseizure patients received the correct advice aboutdriving limitations23

We recommend that future studies address theissues that we raise here To best determine howto optimally evaluate the adult presenting with afirst seizure those studies should be structured toemphasize the use of large well-characterizedsamples clearly defined subjects and outcomesand standardized data collection methods

MISSION STATEMENT OF QSS The missionof the QSS is to prioritize develop and publishevidence-based practice parameters related to thediagnosis treatment and prognosis of neurologicdisorders

The QSS is committed to using the most rigor-ous methods available within our budget in col-

2004 Neurology 69 November 20 2007

RETIRED

RETIRED

laboration with other available AAN resourcesto most efficiently accomplish this mission

DISCLAIMER This statement is provided as aneducational service of the American Academy ofNeurology It is based on an assessment of currentscientific and clinical information It is not intendedto include all possible proper methods of care for aparticular neurologic problem or all legitimate crite-ria for choosing to use a specific procedure Neitheris it intended to exclude any reasonable alternativemethodologies The AAN recognizes that specificpatient care decisions are the prerogative of the pa-tient and physician caring for the patient based onall of the circumstances involved

CONFLICT OF INTEREST STATEMENT TheAmerican Academy of Neurology is committed toproducing independent critical and truthful clin-ical practice guidelines (CPGs) Significant effortsare made to minimize the potential for conflicts ofinterest to influence the recommendations of thisCPG To the extent possible the AAN keeps sep-arate those who have a financial stake in the suc-cess or failure of the products appraised in theCPGs and the developers of the guidelines Con-flict of interest forms were obtained from all au-thors and reviewed by an oversight committeeprior to project initiation AAN limits the participa-tion of authors with substantial conflicts of interestThe AAN forbids commercial participation in orfunding of guideline projects Drafts of the guide-lines have been reviewed by at least threeAANcom-mittees a network of neurologists Neurology peerreviewers and representatives from related fieldsTheAANGuidelineAuthorConflict of Interest Pol-icy can be viewed at wwwaancom

APPENDIX 1

Quality Standards Subcommittee members Jacque-line French MD FAAN (Co-Chair) Gary S Gronseth MD(Co-Chair) Charles E Argoff MD Stephen Ashwal MDFAAN (ex-officio) Christopher Bever Jr MD MBA FAANJohn D England MD FAAN Gary M Franklin MD MPH(ex-officio) GaryH FridayMDMPH FAAN Larry BGold-stein MD FAAN Deborah Hirtz MD (ex-officio) Robert GHollowayMDMPH FAANDonald J IversonMD FAANLeslie A Morrison MD Clifford J Schostal MD David JThurmanMDMPHWilliam JWeinerMD FAAN SamuelWiebe MD

APPENDIX 2AAN classification of evidence for rating of screen-ing article

Class I A statistical population-based sample of patientsstudied at a uniform point in time (usually early) during the

course of the condition All patients undergo the intervention ofinterest The outcome if not objective is determined in an eval-uation that is masked to the patientsrsquo clinical presentations

Class II A statistical non-referral-clinic-based sample ofpatients studied at a uniform point in time (usually early)during the course of the condition Most patients undergothe intervention of interest The outcome if not objective isdetermined in an evaluation that is masked to the patientsrsquoclinical presentations

Class III A sample of patients studied during the course ofthe condition Some patients undergo the intervention of in-terest The outcome if not objective is determined in anevaluation by someone other than the treating physician

Class IV Expert opinion case reports or any study notmeeting criteria for Class I to III

APPENDIX 3

Classification of recommendationsA Established as effective ineffective or harmful (or es-

tablished as usefulpredictive or not usefulpredictive)for the given condition in the specified population(Level A rating requires at least two consistent Class Istudies)

B Probably effective ineffective or harmful (or probablyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level B ratingrequires at least one Class I study or at least two consis-tent Class II studies)

C Possibly effective ineffective or harmful (or possiblyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level C ratingrequires at least one Class II study or two consistentClass III studies)

U Data inadequate or conflicting given current knowl-edge treatment (test predictor) is unproven (Studiesnot meeting criteria for Class IndashClass III)

In exceptional cases one convincing Class I study may suf-fice for an ldquoArdquo recommendation if 1) all criteria are met themagnitude of effect is large (relative rate improved outcome5 and the lower limit of the confidence interval is 2)

APPENDIX 4

Exclusion criteria

bull Letters and case reportsbull Non-English language studiesbull Animal studiesbull Pharmacodynamicpharmacokinetic studiesbull Studies dealing primarily with established or chronicepilepsy patients

bull Studies dealing primarily with acute provoked seizures

Inclusion criteria

bull Study designs observational (prospective retrospectiveand cross-sectional) or interventional (randomized con-trolled trials [RCTs] nonrandomized controlled trials[nRCTs] and uncontrolled case series [UCS])

bull At least 10 patients adults with a first seizure a first pre-sentation with epilepsy or seizures or a first diagnosis ofepilepsy

Neurology 69 November 20 2007 2005

RETIRED

RETIRED

bull Studies addressing any of the following diagnostic in-terventions neuropsychological assessment imagingwith CT MRI or PET scans EEG (standard videoinvasive ambulatory) lumbar puncture toxic screensand laboratory tests (hematologybiochemistry)

bull Patients should be over the age of 18 or the studyshould include a substantial proportion of subjectsover the age of 18

bull Studies reported in English onlybull Studies dealing primarily with apparent unprovokedfirst seizures

APPENDIX 5

Outline for seizure assessment Features of a seizureAssociated factors

AgeMedical history ndash previous history of similar episodesprior stroke brain tumor systemic illness mental ill-ness drug or alcohol abuse

Family historyDevelopmental statusBehaviorHealth at seizure onset ndash febrile ill exposed to illnesscomplaints of not feeling well sleep deprived

Precipitating events other than illness ndash trauma alcoholmedications illicit drugs toxins

Symptoms during seizure (ictal)Aura Subjective sensationsBehavior Mood or behavioral changes before the seizurePreictal symptoms Described by patient or witnessedVocal Cry or gasp slurring of words garbled speechMotor Head or eye turning eye deviation posturingjerking (rhythmic) stiffening automatisms (purpose-less repetitive movements such as picking at clothinglip smacking) generalized or focal movements

Respiration Change in breathing pattern cessation ofbreathing cyanosis

Autonomic Pupillary dilation drooling change in respira-tory or heart rate incontinence pallor vomiting

Loss of consciousness or inability to understand or speak

Symptoms following a seizure (postictal)Amnesia for eventsConfusionLethargySleepinessHeadaches and muscle achesTransient focal weakness (Toddrsquos paresis)Nausea or vomitingBiting of tongue

Adapted from Hirtz et al10

Received February 20 2007 Accepted in final form June 72007

REFERENCES1 Hauser WA Hesdorffer DC Epilepsy frequency

causes and consequences New York Demos Publica-tions 1990

2 Berg AT Shinnar S The risk of seizure recurrence fol-lowing a first unprovoked seizure a quantitative re-view Neurology 199141965ndash972

3 Begley CE Famulari M Annegers JF et al The cost ofepilepsy in the United States an estimate frompopulation- based and survey data Epilepsia 200041342ndash352

4 Grunewald RA Chroni E Panayiotopoulos CP De-layed diagnosis of juvenile myoclonic epilepsy J Neu-rol Neurosurg Psychiatry 199255497ndash499

5 Krumholz A Nonepileptic seizures diagnosis andmanagement Neurology 1999S76ndash83

6 Commission on Epidemiology and Prognosis Interna-tional League Against Epilepsy Guidelines for epidemio-logical studies on epilepsy Epilepsia 199334592ndash596

7 American College of Emergency Physicians (ACEP)Clinical Policies Committee and the Clinical PoliciesSubcommittee on Seizures Clinical Policy Critical is-sues in the evaluation of adult patients presenting to anemergency department with seizures Ann Emerg Med200443605ndash625

8 Fisher RS BoasWvE BlumeW et al Epileptic seizuresand epilepsy Definitions proposed by the InternationalLeague Against Epilepsy (ILAE) and the InternationalBureau for Epilepsy (IBE) Epilepsia 200546470ndash472

9 Agency for Healthcare Research and Quality (AHRQ)Management of newly diagnosed patients with epi-lepsy a systematic review of the literature (AHRQPublication No 01-E038) Rockville MD Agency forHealth Research Quality US Department of Healthand Human Services 2001

10 Hirtz D Ashwal S Berg A et al Practice parameterevaluating a first nonfebrile seizure in children Neu-rology 200055616ndash623

11 Jallon P Loiseau P Loiseau J Newly diagnosed unpro-voked epileptic seizures Presentation at diagnosis inCORALE study Epilepsia 200142464ndash475

12 Quality Standards Subcommittee and the Therapeuticsand Technology Assessment Subcommittee of theAmerican Academy of Neurology Clinical PracticeGuideline Process Manuel St Paul MN AmericanAcademy of Neurology 2004

13 Hauser WA Anderson VE Loewenson RB McRob-erts SM Seizure recurrence after a first unprovoked sei-zure N Engl J Med 1982307522ndash528

14 Hauser WA Rich SS Annegers JF Anderson VE Sei-zure recurrence after a 1st unprovoked seizure an ex-tended follow-up Neurology 1990401163ndash1170

15 van Donselaar CA Schimsheimer R Geerts AT et alValue of the electroencephalogram in adult patientswith untreated idiopathic first seizures Arch Neurol199249231ndash237

16 First Seizure Trial Group Randomized clinical trial onthe efficacy of antiepileptic drugs in reducing the risk ofrelapse after a first unprovoked tonic-clonic seizureNeurology 199343478ndash483

17 Bora I Seckin B Zarifoglu S Turin F Sadikogllu S OgulE Risk of recurrence after first unprovoked tonic-clonicseizure in adults J Neurol 1995242157ndash163

18 Hopkins A Garman A Clarke C The first seizure inadult life Value of clinical features electroencephalog-raphy and computerized tomographic scanning in pre-diction of seizure recurrence Lancet 1988721ndash726

19 Das CP Sawhney IMS Lal V et al Risk of recurrenceof seizures following a single unprovoked idiopathicseizure Neurol India 200046357ndash360

2006 Neurology 69 November 20 2007

RETIRED

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

RETIRED

tion MRI is more sensitive and is more likely toshow significant abnormalities than is CT91029

Conclusion For adults presenting initially with anapparent unprovoked first seizure the evidencefrom seven Class II studies indicates that a brain im-aging study either a CT orMRI is probably usefulIt has a significant yield of about 10 which maylead to the diagnosis of disorders such as a braintumor stroke cysticercosis or other structural le-sions and may have some value in determining therisk for seizure recurrence

Recommendation Brain imaging using CT orMRI should be considered as part of the neurodi-agnostic evaluation of adults presenting with anapparent unprovoked first seizure (Level B)

Laboratory studies Based on review of clinicalstudies expert opinion and position papers wechose to assess the value of the following labora-tory tests blood glucose blood counts electro-lytes lumbar puncture and toxicology screeningin adults with a first seizure79182330-32

Should blood counts blood glucose and electrolytepanels be routinely ordered in an adult with an apparent

unprovoked first seizure EvidenceOf the two Class IIstudies1823 (464 patients) (table 3) that assessedthe yield and value of blood counts blood glu-cose and electrolyte panels abnormalities werereported in from 0 to 15 for each of these testsbut no clinically significant abnormalities werenoted by the authors (table 3) In one Class IIstudy blood count abnormalities were reportedin 15 of patients but all these were judged bythe author to be incidental and clinically insignif-icant such as minor increased white blood cellcounts that resolved23 That same study reporteda 10 incidence of hypoglycemia but again allthose abnormalities were judged as mild and clin-ically insignificant with regard to the seizure Inboth the Class II studies routine electrolytes wereobtained in a high percentage of patients with

particular attention to serum calcium and sodium(table 3) Serum calcium was normal in all pa-tients in which it was measured In one study so-dium abnormalities of hyponatremia were notedin 7 of patients but again all were judged bythe author not to be of clinical significance23 Theother Class II study also reported no significantsodium abnormalities18

There were four Class III studies that showed ahigher incidence of significant laboratory abnormal-ities particularly for serum sodium and glucose30-33

All these studies were based in emergency depart-ments and the patients were obviously ill withmorecomorbidities than those in our Class II studies1823

A large proportion were patients with acute symp-tomatic or provoked rather than apparent unpro-voked seizures as required by our inclusion criteria(appendix 4) Among the reported acute symptom-atic causes for the seizureswere alcoholwithdrawalacute stroke tumor sepsis or trauma Also somestudies had many patients with other exclusion cri-teria (appendix 4) such as focal neurologic deficitsand itwas not specifiedwhether patients returned totheir normal level of function after the seizure30-33

Despite these limitations and acknowledgment thatyield is very low7 these four Class III studies wereconsistent in proposing some value to routinescreening of blood glucose for hypoglycemia and se-rum electrolytes for hyponatremia because unantic-ipated and clinically relevant hypoglycemia andhyponatremia were found in about 1 of thesepatients30-33

The two Class II studies as well the four ClassIII we identified do not provide convincing evi-dence to support or refute significant value forblood glucose blood count and electrolytes asroutine for adults at initial presentation with epi-leptic seizures This finding is similar to the QSSreport on initial seizure evaluation in children10

and a previous evidence-based analysis10

Table 3 Laboratory diagnostic tests in patients with new onset seizure (Class I and II articles)

Reference No subjects (age y) Class Modality No studied ()No abnormal() 95 CI

No sign abnormal() 95 CI

23 56 (16 to 89) II Blood count 55 (98) 8 (15) 76ndash26 0 (0) 0ndash65

Electrolytes 55 (98) 4 (7) 29ndash173 0 (0) 0ndash65

Glucose 50 (89) 5 (10) 43ndash21 0 (0) 0ndash71

Calcium 35 (63) 0 (0) 0ndash99 0 (0) 0ndash99

18 408 (16 and older) II Blood count 371 (91) NA 0 (0) 0ndash1

Electrolytes calcium 371 (91) NA 0 (0) 0ndash1

Total 464

NA not available

Neurology 69 November 20 2007 2001

RETIRED

RETIRED

In general most studies emphasize that the his-tory and physical examination can guide decisionmaking because they often predict patients whowill demonstrate significantly abnormal labora-tory tests23 For example patients with persistingaltered mental status fever or focal neurologicdeficits warrant more extensive evaluation andare more likely to have abnormal laboratory find-ings7 A practice parameter from the AmericanCollege of Emergency Physicians which generallysupports our findings still recommends obtaininga blood glucose determination and serum sodiumin patients with a first seizure7 However thatrecommendation was based on evidence fromstudies that we excluded or downgraded as ourinclusion criteria require because they involvedlarge numbers of patients with obvious acutesymptomatic seizures

Conclusion Data from two Class II and fourClass III studies showed that in adults presentingwith an apparent unprovoked first seizure al-though some abnormal laboratory results are re-ported there is not sufficient evidence to supportor refute recommending routine testing of bloodglucose blood counts or electrolyte panels Thenecessity for such studies should be guided by spe-cific clinical circumstances based on the historyphysical and neurologic examination

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure bloodglucose blood counts and electrolyte panels(particularly sodium) may be helpful in specificclinical circumstances but there are insufficientdata to support or refute routine recommenda-tion of any of these laboratory tests (Level U)

Should a lumbar puncture be routinely performed inan adult presenting with an apparent unprovoked first sei-

zure Evidence There were no Class I or II studiesbut of the two Class III studies considering thevalue of a lumbar puncture in patients with a firstseizure both were from emergency departmentsand reported significant lumbar puncture abnor-malities in up to 8 of patients3033 However notall patients in these two studies had unprovokedseizures or returned to baseline function as ourinclusion criteria require A substantial numberhad acute symptomatic seizures and level of con-sciousness was not well specified Another majorlimitation of these studies is that lumbar punc-tures were performed selectively They were per-formed in only 68 of the patients in one study30

and 24 of the other33 and this depended on suchpresenting clinical features as fever or at the dis-cretion of the staff Consequently lumbar punc-tures were not routinely performed and the

decision to perform a lumbar puncture was di-rected by the clinical history and physical find-ings Moreover in one study the determination ofwhether the lumbar puncture showed a ldquosignifi-cantrdquo abnormality was based largely on whetherthe patient was admitted to the hospital ratherthan whether the test showed a cause for the sei-zure30 In other studies of first seizures that ex-cluded patients with acute symptomatic seizureslumbar puncture was not considered as an impor-tant routine test1823

In particular there are no Class I II or otherconvincing studies to support a lumbar puncturein patients who are alert oriented afebrile andnot immunocompromised7 There is some evi-dence to support a lumbar puncture after a firstseizure in patients who are immunocompromisedeven if they are afebrile734

Conclusion Data from two Class III studies re-vealed significant abnormalities in up to 8 of amixed group of patients presenting to an emer-gency department with a first seizure Howeverthe studies selectively performed lumbar punc-tures based on clinical findings and included pa-tients who did not meet our inclusion criteriasuch as those with acute symptomatic causes fortheir seizures or who had not returned to theirnormal baseline function

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure lumbarpuncture may be helpful in specific clinical circum-stances such as patients who are febrile but thereare insufficient data to support or refute recom-mending routine lumbar puncture (Level U)

Should toxicologic screening be routinely ordered inan adult presenting with an apparent unprovoked first sei-

zure Evidence Seizures are reported as a consequenceof drug intoxication particularly with tricyclic anti-depressants cocaine and other stimulants735 In aseries of patients with acute medical complicationsof cocaine intoxication seizures often first seizuresaccounted for 10 of the presenting symptoms36

Several studies of emergency department admis-sions for first seizures including both acute symp-tomatic and unprovoked seizures indicated about3 may relate to drug toxicity or abuse73233 andone study advocated toxicology screening for allpatients with unexplained first seizures30 Still arecent evidence-based review by the AmericanAcademy of Emergency Physicians Clinical Pol-icy Committee considering management of adultpatients presenting to an emergency departmentwith seizures did not find sufficient evidence in thelimited existing prospective or retrospective studiesto recommend routine toxicology screening7

2002 Neurology 69 November 20 2007

RETIRED

RETIRED

In considering this issue we identified twoClass III studies considering the value of toxicol-ogy screening in adult patients presenting with afirst seizure3536 Although both studies reportedsome first seizures in patients with abnormal tox-icology screening neither study investigated theuse of routine screening in first seizure patients Inaddition the studies did not consider mainly un-provoked seizures in which the patient returnedto baseline function as our inclusion criteria spec-ify (appendix 4)

Conclusion In two Class III studies consideringthe value of toxicology screening in adult patientspresenting with a seizure some patients with ap-parent unprovoked first seizure were includedbut neither study investigated the use of routinetoxicology screening for such patients

Recommendation In the adult presenting with anapparent unprovoked seizure toxicology screen-ing may be helpful in specific clinical circum-stances but there are insufficient data to supportor refute a routine recommendation for toxicol-ogy screening (Level U)

CONCLUSIONS AND RECOMMENDATIONSDiagnosis of seizures determination of possiblecauses classification of seizure type and possibleseizure syndrome and assessment of the risk forrecurrence of seizures require timely selection andinterpretation of appropriate diagnostic tools in-cluding EEGs CTs or MRIs blood studies andlumbar puncture This report assesses the evi-dence available to answer the question ldquoWhat arethe most appropriate tools for evaluating a firstunprovoked seizure in adults at the time of initialpresentationrdquo

For the adult initially presenting with an ap-parent unprovoked seizure diagnostic evalua-tions influence aspects of patient care andmanagement including drug treatment patientand family counseling and the need for immedi-ate hospitalization and subsequent follow-upThis practice parameter reviews the published lit-erature concerning the yield and value of studiesfollowing an initial seizure in adults and classifiesthe strength of available evidence There is insuf-ficient Class I evidence to provide any recommen-dation with the highest degree of clinical certainty(see appendix 3) However there is sufficientClass I and Class II evidence that a routine EEGshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with an ap-parent unprovoked first seizure and predicts therisk of seizure recurrence (Level B) There is alsosufficient Class II evidence that a brain imaging

study either a CT or MRI helps determine thepresence of a significant abnormality such as abrain tumor Brain imaging using CT or MRIshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with a firstseizure (Level B) The CT has a major advantagebecause of the speed with which it can be ob-tained so its value is mainly in the emergency sit-uation but theMRI is regarded by expert opinionas having a higher yield and is the preferable pro-cedure in non-emergency or elective situa-tions91029 However both MRI and CT are ofvalue The decision to perform other studies in-cluding blood glucose blood counts electrolytepanels lumbar puncture or toxicology screeningfor the purpose of determining the cause of theseizure or identifying potentially treatable condi-tions will depend on the specific or individualizedclinical circumstances based on the history phys-ical and neurologic examination There are in-sufficient data to support or refute the use of anyone of these tests or procedures as a routine rec-ommendation (Level U)

The history physical and neurologic exami-nation are acknowledged by expert consensus tobe required for the initial diagnosis and classifica-tion of seizures26 These standards for the initialdiagnosis and classification of seizures form thebasis for the ldquoclinical and electroencephalo-graphic classification of epileptic seizuresrdquo21

Indeed information from the history includingthe type of the seizure21 together with findings fromthe physical and neurologic examination also aid inthe determination of a probable cause or risk of re-currence If the history reveals that there have beenother seizures in the past then this indicates epilepsyrather than a single isolated seizure Although thesubsequent diagnostic evaluation may be similarto that for a single unprovoked seizure there aredifferent prognostic and therapeutic implicationsfor a patient with epilepsy27 Evidence of pre-existing seizures provides information for the po-tential classification of an epilepsy syndrome25

Such information is of importance in establishingpossible causes associated comorbidities prog-nosis and optimal treatment of a seizure disorderor epilepsy syndrome2526 Moreover several fea-tures of the clinical history and physical and neuro-logic examination such as the presence of specificgeneralized or focal neurologic deficits are predic-tive of higher risks of seizure recurrence210

RECOMMENDATIONS FOR FUTURE RE-SEARCH The history physical and neurologicexaminations form the basis for the diagnosis and

Neurology 69 November 20 2007 2003

RETIRED

RETIRED

classification of epileptic seizures26 and this is ac-knowledged by expert consensus26 Still future re-search is warranted to determine the specificaspects of the history physical and neurologicexamination that are most useful in both initialdiagnosis of a seizure and subsequent manage-ment Also it would be helpful to understandhow this type of information should best guideperformance and timing of other neurodiagnosticor laboratory studies An international group re-cently recommended a broader definition of epi-lepsy as a brain disorder with a history of at leastone seizure and ldquocharacterized by an enduringpredisposition to generate epileptic seizuresrdquo8

This type of approach will require validation andinvestigation including determination of the truevalue of diagnostic studies such as EEG in estab-lishing the presence of an epileptic syndrome aftera first seizure In regard to the EEG one limita-tion of currently available studies is the variabilityin the timing of the EEG after incident seizureThere is some evidence in children that an EEGdone within 24 hours of a presenting seizure givesa higher yield of significant abnormalities10

Although all types of seizures are considered inthis analysis most patients had a convulsive sei-zure This may be because convulsive seizures aremore likely to prompt acute medical attentionthan a simple or complex partial seizure Morestudies specifically focusing attention on all typesof initial seizures not just convulsive seizures areneeded

Other factors deserve consideration in futureinvestigations For instance the costs of these var-ious tests do not receive much attention in studiesof patients with first seizure and should Also theelderly are a growing proportion of the generalpopulation and warrant special consideration infuture investigations Another unresolved issuerelates to the need for immediate hospitalizationof patients with a first seizure Some guidelinesexist but well designed studies are needed7

Prediction of seizure recurrence after a first un-provoked seizure is important to guide decisionsregarding antiepileptic drug therapy patientmanagement and counseling Many factors arereported to predict seizure recurrence includingthe etiology of the seizures and association withneurologic abnormalities2 Although the EEG andbrain imaging studies are reported to predict sei-zure recurrence studies vary in support of thatobservation and in respect to the specific EEG ab-normalities or brain imaging findings that aremost significant Some of this variability regard-ing EEG findings may relate to the timing of the

study after the initial event and to the decisionwhether to treat with antiepileptic medicationsThe value of specific findings on the EEG brainimaging and other variables deserve furtherstudy in regard to seizure recurrence risks Alsofuture studies of the clinical utility for such neuro-diagnostic procedures should employ standard-ized tests and prediction algorithms37

Future studies are needed to analyze the valueof expert input into the history physical neuro-logic examination or diagnostic testing such as theEEG or brain imaging The expertise and qualifica-tions of individuals interpreting such information ortests particularly the EEG or brain imaging varyconsiderably and may influence interpretationsand outcomes37 These are not simple ldquopositiverdquo orldquonegativerdquo tests like a pregnancy test they requireexpertise for proper interpretation Some other di-agnostic studies deserve further analysis in patientspresenting with new onset seizures In particularroutine electrolytes including serum sodium andglucosewere reported in some studies to be of valuebut this was not consistent or confirmed by otherreports Also toxicologic screening was reported asimportant in some limited series but was not stud-ied in a manner to allow adequate evidence-basedverification or refutation of its routine utility

Studies of patient management and counselingat the time of first diagnosis of a single seizure orepilepsy are recommended For example whichpatients should receive antiepileptic drug treat-ment and if so exactly what type and whenshould they get it Also when is the patient bestadvised regarding driving laws and other socialissues and how and by whom should this bedone This is not a trivial issue For instance aBritish study noted that only 21 of all adult firstseizure patients received the correct advice aboutdriving limitations23

We recommend that future studies address theissues that we raise here To best determine howto optimally evaluate the adult presenting with afirst seizure those studies should be structured toemphasize the use of large well-characterizedsamples clearly defined subjects and outcomesand standardized data collection methods

MISSION STATEMENT OF QSS The missionof the QSS is to prioritize develop and publishevidence-based practice parameters related to thediagnosis treatment and prognosis of neurologicdisorders

The QSS is committed to using the most rigor-ous methods available within our budget in col-

2004 Neurology 69 November 20 2007

RETIRED

RETIRED

laboration with other available AAN resourcesto most efficiently accomplish this mission

DISCLAIMER This statement is provided as aneducational service of the American Academy ofNeurology It is based on an assessment of currentscientific and clinical information It is not intendedto include all possible proper methods of care for aparticular neurologic problem or all legitimate crite-ria for choosing to use a specific procedure Neitheris it intended to exclude any reasonable alternativemethodologies The AAN recognizes that specificpatient care decisions are the prerogative of the pa-tient and physician caring for the patient based onall of the circumstances involved

CONFLICT OF INTEREST STATEMENT TheAmerican Academy of Neurology is committed toproducing independent critical and truthful clin-ical practice guidelines (CPGs) Significant effortsare made to minimize the potential for conflicts ofinterest to influence the recommendations of thisCPG To the extent possible the AAN keeps sep-arate those who have a financial stake in the suc-cess or failure of the products appraised in theCPGs and the developers of the guidelines Con-flict of interest forms were obtained from all au-thors and reviewed by an oversight committeeprior to project initiation AAN limits the participa-tion of authors with substantial conflicts of interestThe AAN forbids commercial participation in orfunding of guideline projects Drafts of the guide-lines have been reviewed by at least threeAANcom-mittees a network of neurologists Neurology peerreviewers and representatives from related fieldsTheAANGuidelineAuthorConflict of Interest Pol-icy can be viewed at wwwaancom

APPENDIX 1

Quality Standards Subcommittee members Jacque-line French MD FAAN (Co-Chair) Gary S Gronseth MD(Co-Chair) Charles E Argoff MD Stephen Ashwal MDFAAN (ex-officio) Christopher Bever Jr MD MBA FAANJohn D England MD FAAN Gary M Franklin MD MPH(ex-officio) GaryH FridayMDMPH FAAN Larry BGold-stein MD FAAN Deborah Hirtz MD (ex-officio) Robert GHollowayMDMPH FAANDonald J IversonMD FAANLeslie A Morrison MD Clifford J Schostal MD David JThurmanMDMPHWilliam JWeinerMD FAAN SamuelWiebe MD

APPENDIX 2AAN classification of evidence for rating of screen-ing article

Class I A statistical population-based sample of patientsstudied at a uniform point in time (usually early) during the

course of the condition All patients undergo the intervention ofinterest The outcome if not objective is determined in an eval-uation that is masked to the patientsrsquo clinical presentations

Class II A statistical non-referral-clinic-based sample ofpatients studied at a uniform point in time (usually early)during the course of the condition Most patients undergothe intervention of interest The outcome if not objective isdetermined in an evaluation that is masked to the patientsrsquoclinical presentations

Class III A sample of patients studied during the course ofthe condition Some patients undergo the intervention of in-terest The outcome if not objective is determined in anevaluation by someone other than the treating physician

Class IV Expert opinion case reports or any study notmeeting criteria for Class I to III

APPENDIX 3

Classification of recommendationsA Established as effective ineffective or harmful (or es-

tablished as usefulpredictive or not usefulpredictive)for the given condition in the specified population(Level A rating requires at least two consistent Class Istudies)

B Probably effective ineffective or harmful (or probablyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level B ratingrequires at least one Class I study or at least two consis-tent Class II studies)

C Possibly effective ineffective or harmful (or possiblyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level C ratingrequires at least one Class II study or two consistentClass III studies)

U Data inadequate or conflicting given current knowl-edge treatment (test predictor) is unproven (Studiesnot meeting criteria for Class IndashClass III)

In exceptional cases one convincing Class I study may suf-fice for an ldquoArdquo recommendation if 1) all criteria are met themagnitude of effect is large (relative rate improved outcome5 and the lower limit of the confidence interval is 2)

APPENDIX 4

Exclusion criteria

bull Letters and case reportsbull Non-English language studiesbull Animal studiesbull Pharmacodynamicpharmacokinetic studiesbull Studies dealing primarily with established or chronicepilepsy patients

bull Studies dealing primarily with acute provoked seizures

Inclusion criteria

bull Study designs observational (prospective retrospectiveand cross-sectional) or interventional (randomized con-trolled trials [RCTs] nonrandomized controlled trials[nRCTs] and uncontrolled case series [UCS])

bull At least 10 patients adults with a first seizure a first pre-sentation with epilepsy or seizures or a first diagnosis ofepilepsy

Neurology 69 November 20 2007 2005

RETIRED

RETIRED

bull Studies addressing any of the following diagnostic in-terventions neuropsychological assessment imagingwith CT MRI or PET scans EEG (standard videoinvasive ambulatory) lumbar puncture toxic screensand laboratory tests (hematologybiochemistry)

bull Patients should be over the age of 18 or the studyshould include a substantial proportion of subjectsover the age of 18

bull Studies reported in English onlybull Studies dealing primarily with apparent unprovokedfirst seizures

APPENDIX 5

Outline for seizure assessment Features of a seizureAssociated factors

AgeMedical history ndash previous history of similar episodesprior stroke brain tumor systemic illness mental ill-ness drug or alcohol abuse

Family historyDevelopmental statusBehaviorHealth at seizure onset ndash febrile ill exposed to illnesscomplaints of not feeling well sleep deprived

Precipitating events other than illness ndash trauma alcoholmedications illicit drugs toxins

Symptoms during seizure (ictal)Aura Subjective sensationsBehavior Mood or behavioral changes before the seizurePreictal symptoms Described by patient or witnessedVocal Cry or gasp slurring of words garbled speechMotor Head or eye turning eye deviation posturingjerking (rhythmic) stiffening automatisms (purpose-less repetitive movements such as picking at clothinglip smacking) generalized or focal movements

Respiration Change in breathing pattern cessation ofbreathing cyanosis

Autonomic Pupillary dilation drooling change in respira-tory or heart rate incontinence pallor vomiting

Loss of consciousness or inability to understand or speak

Symptoms following a seizure (postictal)Amnesia for eventsConfusionLethargySleepinessHeadaches and muscle achesTransient focal weakness (Toddrsquos paresis)Nausea or vomitingBiting of tongue

Adapted from Hirtz et al10

Received February 20 2007 Accepted in final form June 72007

REFERENCES1 Hauser WA Hesdorffer DC Epilepsy frequency

causes and consequences New York Demos Publica-tions 1990

2 Berg AT Shinnar S The risk of seizure recurrence fol-lowing a first unprovoked seizure a quantitative re-view Neurology 199141965ndash972

3 Begley CE Famulari M Annegers JF et al The cost ofepilepsy in the United States an estimate frompopulation- based and survey data Epilepsia 200041342ndash352

4 Grunewald RA Chroni E Panayiotopoulos CP De-layed diagnosis of juvenile myoclonic epilepsy J Neu-rol Neurosurg Psychiatry 199255497ndash499

5 Krumholz A Nonepileptic seizures diagnosis andmanagement Neurology 1999S76ndash83

6 Commission on Epidemiology and Prognosis Interna-tional League Against Epilepsy Guidelines for epidemio-logical studies on epilepsy Epilepsia 199334592ndash596

7 American College of Emergency Physicians (ACEP)Clinical Policies Committee and the Clinical PoliciesSubcommittee on Seizures Clinical Policy Critical is-sues in the evaluation of adult patients presenting to anemergency department with seizures Ann Emerg Med200443605ndash625

8 Fisher RS BoasWvE BlumeW et al Epileptic seizuresand epilepsy Definitions proposed by the InternationalLeague Against Epilepsy (ILAE) and the InternationalBureau for Epilepsy (IBE) Epilepsia 200546470ndash472

9 Agency for Healthcare Research and Quality (AHRQ)Management of newly diagnosed patients with epi-lepsy a systematic review of the literature (AHRQPublication No 01-E038) Rockville MD Agency forHealth Research Quality US Department of Healthand Human Services 2001

10 Hirtz D Ashwal S Berg A et al Practice parameterevaluating a first nonfebrile seizure in children Neu-rology 200055616ndash623

11 Jallon P Loiseau P Loiseau J Newly diagnosed unpro-voked epileptic seizures Presentation at diagnosis inCORALE study Epilepsia 200142464ndash475

12 Quality Standards Subcommittee and the Therapeuticsand Technology Assessment Subcommittee of theAmerican Academy of Neurology Clinical PracticeGuideline Process Manuel St Paul MN AmericanAcademy of Neurology 2004

13 Hauser WA Anderson VE Loewenson RB McRob-erts SM Seizure recurrence after a first unprovoked sei-zure N Engl J Med 1982307522ndash528

14 Hauser WA Rich SS Annegers JF Anderson VE Sei-zure recurrence after a 1st unprovoked seizure an ex-tended follow-up Neurology 1990401163ndash1170

15 van Donselaar CA Schimsheimer R Geerts AT et alValue of the electroencephalogram in adult patientswith untreated idiopathic first seizures Arch Neurol199249231ndash237

16 First Seizure Trial Group Randomized clinical trial onthe efficacy of antiepileptic drugs in reducing the risk ofrelapse after a first unprovoked tonic-clonic seizureNeurology 199343478ndash483

17 Bora I Seckin B Zarifoglu S Turin F Sadikogllu S OgulE Risk of recurrence after first unprovoked tonic-clonicseizure in adults J Neurol 1995242157ndash163

18 Hopkins A Garman A Clarke C The first seizure inadult life Value of clinical features electroencephalog-raphy and computerized tomographic scanning in pre-diction of seizure recurrence Lancet 1988721ndash726

19 Das CP Sawhney IMS Lal V et al Risk of recurrenceof seizures following a single unprovoked idiopathicseizure Neurol India 200046357ndash360

2006 Neurology 69 November 20 2007

RETIRED

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

RETIRED

In general most studies emphasize that the his-tory and physical examination can guide decisionmaking because they often predict patients whowill demonstrate significantly abnormal labora-tory tests23 For example patients with persistingaltered mental status fever or focal neurologicdeficits warrant more extensive evaluation andare more likely to have abnormal laboratory find-ings7 A practice parameter from the AmericanCollege of Emergency Physicians which generallysupports our findings still recommends obtaininga blood glucose determination and serum sodiumin patients with a first seizure7 However thatrecommendation was based on evidence fromstudies that we excluded or downgraded as ourinclusion criteria require because they involvedlarge numbers of patients with obvious acutesymptomatic seizures

Conclusion Data from two Class II and fourClass III studies showed that in adults presentingwith an apparent unprovoked first seizure al-though some abnormal laboratory results are re-ported there is not sufficient evidence to supportor refute recommending routine testing of bloodglucose blood counts or electrolyte panels Thenecessity for such studies should be guided by spe-cific clinical circumstances based on the historyphysical and neurologic examination

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure bloodglucose blood counts and electrolyte panels(particularly sodium) may be helpful in specificclinical circumstances but there are insufficientdata to support or refute routine recommenda-tion of any of these laboratory tests (Level U)

Should a lumbar puncture be routinely performed inan adult presenting with an apparent unprovoked first sei-

zure Evidence There were no Class I or II studiesbut of the two Class III studies considering thevalue of a lumbar puncture in patients with a firstseizure both were from emergency departmentsand reported significant lumbar puncture abnor-malities in up to 8 of patients3033 However notall patients in these two studies had unprovokedseizures or returned to baseline function as ourinclusion criteria require A substantial numberhad acute symptomatic seizures and level of con-sciousness was not well specified Another majorlimitation of these studies is that lumbar punc-tures were performed selectively They were per-formed in only 68 of the patients in one study30

and 24 of the other33 and this depended on suchpresenting clinical features as fever or at the dis-cretion of the staff Consequently lumbar punc-tures were not routinely performed and the

decision to perform a lumbar puncture was di-rected by the clinical history and physical find-ings Moreover in one study the determination ofwhether the lumbar puncture showed a ldquosignifi-cantrdquo abnormality was based largely on whetherthe patient was admitted to the hospital ratherthan whether the test showed a cause for the sei-zure30 In other studies of first seizures that ex-cluded patients with acute symptomatic seizureslumbar puncture was not considered as an impor-tant routine test1823

In particular there are no Class I II or otherconvincing studies to support a lumbar puncturein patients who are alert oriented afebrile andnot immunocompromised7 There is some evi-dence to support a lumbar puncture after a firstseizure in patients who are immunocompromisedeven if they are afebrile734

Conclusion Data from two Class III studies re-vealed significant abnormalities in up to 8 of amixed group of patients presenting to an emer-gency department with a first seizure Howeverthe studies selectively performed lumbar punc-tures based on clinical findings and included pa-tients who did not meet our inclusion criteriasuch as those with acute symptomatic causes fortheir seizures or who had not returned to theirnormal baseline function

Recommendation In the adult initially presentingwith an apparent unprovoked first seizure lumbarpuncture may be helpful in specific clinical circum-stances such as patients who are febrile but thereare insufficient data to support or refute recom-mending routine lumbar puncture (Level U)

Should toxicologic screening be routinely ordered inan adult presenting with an apparent unprovoked first sei-

zure Evidence Seizures are reported as a consequenceof drug intoxication particularly with tricyclic anti-depressants cocaine and other stimulants735 In aseries of patients with acute medical complicationsof cocaine intoxication seizures often first seizuresaccounted for 10 of the presenting symptoms36

Several studies of emergency department admis-sions for first seizures including both acute symp-tomatic and unprovoked seizures indicated about3 may relate to drug toxicity or abuse73233 andone study advocated toxicology screening for allpatients with unexplained first seizures30 Still arecent evidence-based review by the AmericanAcademy of Emergency Physicians Clinical Pol-icy Committee considering management of adultpatients presenting to an emergency departmentwith seizures did not find sufficient evidence in thelimited existing prospective or retrospective studiesto recommend routine toxicology screening7

2002 Neurology 69 November 20 2007

RETIRED

RETIRED

In considering this issue we identified twoClass III studies considering the value of toxicol-ogy screening in adult patients presenting with afirst seizure3536 Although both studies reportedsome first seizures in patients with abnormal tox-icology screening neither study investigated theuse of routine screening in first seizure patients Inaddition the studies did not consider mainly un-provoked seizures in which the patient returnedto baseline function as our inclusion criteria spec-ify (appendix 4)

Conclusion In two Class III studies consideringthe value of toxicology screening in adult patientspresenting with a seizure some patients with ap-parent unprovoked first seizure were includedbut neither study investigated the use of routinetoxicology screening for such patients

Recommendation In the adult presenting with anapparent unprovoked seizure toxicology screen-ing may be helpful in specific clinical circum-stances but there are insufficient data to supportor refute a routine recommendation for toxicol-ogy screening (Level U)

CONCLUSIONS AND RECOMMENDATIONSDiagnosis of seizures determination of possiblecauses classification of seizure type and possibleseizure syndrome and assessment of the risk forrecurrence of seizures require timely selection andinterpretation of appropriate diagnostic tools in-cluding EEGs CTs or MRIs blood studies andlumbar puncture This report assesses the evi-dence available to answer the question ldquoWhat arethe most appropriate tools for evaluating a firstunprovoked seizure in adults at the time of initialpresentationrdquo

For the adult initially presenting with an ap-parent unprovoked seizure diagnostic evalua-tions influence aspects of patient care andmanagement including drug treatment patientand family counseling and the need for immedi-ate hospitalization and subsequent follow-upThis practice parameter reviews the published lit-erature concerning the yield and value of studiesfollowing an initial seizure in adults and classifiesthe strength of available evidence There is insuf-ficient Class I evidence to provide any recommen-dation with the highest degree of clinical certainty(see appendix 3) However there is sufficientClass I and Class II evidence that a routine EEGshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with an ap-parent unprovoked first seizure and predicts therisk of seizure recurrence (Level B) There is alsosufficient Class II evidence that a brain imaging

study either a CT or MRI helps determine thepresence of a significant abnormality such as abrain tumor Brain imaging using CT or MRIshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with a firstseizure (Level B) The CT has a major advantagebecause of the speed with which it can be ob-tained so its value is mainly in the emergency sit-uation but theMRI is regarded by expert opinionas having a higher yield and is the preferable pro-cedure in non-emergency or elective situa-tions91029 However both MRI and CT are ofvalue The decision to perform other studies in-cluding blood glucose blood counts electrolytepanels lumbar puncture or toxicology screeningfor the purpose of determining the cause of theseizure or identifying potentially treatable condi-tions will depend on the specific or individualizedclinical circumstances based on the history phys-ical and neurologic examination There are in-sufficient data to support or refute the use of anyone of these tests or procedures as a routine rec-ommendation (Level U)

The history physical and neurologic exami-nation are acknowledged by expert consensus tobe required for the initial diagnosis and classifica-tion of seizures26 These standards for the initialdiagnosis and classification of seizures form thebasis for the ldquoclinical and electroencephalo-graphic classification of epileptic seizuresrdquo21

Indeed information from the history includingthe type of the seizure21 together with findings fromthe physical and neurologic examination also aid inthe determination of a probable cause or risk of re-currence If the history reveals that there have beenother seizures in the past then this indicates epilepsyrather than a single isolated seizure Although thesubsequent diagnostic evaluation may be similarto that for a single unprovoked seizure there aredifferent prognostic and therapeutic implicationsfor a patient with epilepsy27 Evidence of pre-existing seizures provides information for the po-tential classification of an epilepsy syndrome25

Such information is of importance in establishingpossible causes associated comorbidities prog-nosis and optimal treatment of a seizure disorderor epilepsy syndrome2526 Moreover several fea-tures of the clinical history and physical and neuro-logic examination such as the presence of specificgeneralized or focal neurologic deficits are predic-tive of higher risks of seizure recurrence210

RECOMMENDATIONS FOR FUTURE RE-SEARCH The history physical and neurologicexaminations form the basis for the diagnosis and

Neurology 69 November 20 2007 2003

RETIRED

RETIRED

classification of epileptic seizures26 and this is ac-knowledged by expert consensus26 Still future re-search is warranted to determine the specificaspects of the history physical and neurologicexamination that are most useful in both initialdiagnosis of a seizure and subsequent manage-ment Also it would be helpful to understandhow this type of information should best guideperformance and timing of other neurodiagnosticor laboratory studies An international group re-cently recommended a broader definition of epi-lepsy as a brain disorder with a history of at leastone seizure and ldquocharacterized by an enduringpredisposition to generate epileptic seizuresrdquo8

This type of approach will require validation andinvestigation including determination of the truevalue of diagnostic studies such as EEG in estab-lishing the presence of an epileptic syndrome aftera first seizure In regard to the EEG one limita-tion of currently available studies is the variabilityin the timing of the EEG after incident seizureThere is some evidence in children that an EEGdone within 24 hours of a presenting seizure givesa higher yield of significant abnormalities10

Although all types of seizures are considered inthis analysis most patients had a convulsive sei-zure This may be because convulsive seizures aremore likely to prompt acute medical attentionthan a simple or complex partial seizure Morestudies specifically focusing attention on all typesof initial seizures not just convulsive seizures areneeded

Other factors deserve consideration in futureinvestigations For instance the costs of these var-ious tests do not receive much attention in studiesof patients with first seizure and should Also theelderly are a growing proportion of the generalpopulation and warrant special consideration infuture investigations Another unresolved issuerelates to the need for immediate hospitalizationof patients with a first seizure Some guidelinesexist but well designed studies are needed7

Prediction of seizure recurrence after a first un-provoked seizure is important to guide decisionsregarding antiepileptic drug therapy patientmanagement and counseling Many factors arereported to predict seizure recurrence includingthe etiology of the seizures and association withneurologic abnormalities2 Although the EEG andbrain imaging studies are reported to predict sei-zure recurrence studies vary in support of thatobservation and in respect to the specific EEG ab-normalities or brain imaging findings that aremost significant Some of this variability regard-ing EEG findings may relate to the timing of the

study after the initial event and to the decisionwhether to treat with antiepileptic medicationsThe value of specific findings on the EEG brainimaging and other variables deserve furtherstudy in regard to seizure recurrence risks Alsofuture studies of the clinical utility for such neuro-diagnostic procedures should employ standard-ized tests and prediction algorithms37

Future studies are needed to analyze the valueof expert input into the history physical neuro-logic examination or diagnostic testing such as theEEG or brain imaging The expertise and qualifica-tions of individuals interpreting such information ortests particularly the EEG or brain imaging varyconsiderably and may influence interpretationsand outcomes37 These are not simple ldquopositiverdquo orldquonegativerdquo tests like a pregnancy test they requireexpertise for proper interpretation Some other di-agnostic studies deserve further analysis in patientspresenting with new onset seizures In particularroutine electrolytes including serum sodium andglucosewere reported in some studies to be of valuebut this was not consistent or confirmed by otherreports Also toxicologic screening was reported asimportant in some limited series but was not stud-ied in a manner to allow adequate evidence-basedverification or refutation of its routine utility

Studies of patient management and counselingat the time of first diagnosis of a single seizure orepilepsy are recommended For example whichpatients should receive antiepileptic drug treat-ment and if so exactly what type and whenshould they get it Also when is the patient bestadvised regarding driving laws and other socialissues and how and by whom should this bedone This is not a trivial issue For instance aBritish study noted that only 21 of all adult firstseizure patients received the correct advice aboutdriving limitations23

We recommend that future studies address theissues that we raise here To best determine howto optimally evaluate the adult presenting with afirst seizure those studies should be structured toemphasize the use of large well-characterizedsamples clearly defined subjects and outcomesand standardized data collection methods

MISSION STATEMENT OF QSS The missionof the QSS is to prioritize develop and publishevidence-based practice parameters related to thediagnosis treatment and prognosis of neurologicdisorders

The QSS is committed to using the most rigor-ous methods available within our budget in col-

2004 Neurology 69 November 20 2007

RETIRED

RETIRED

laboration with other available AAN resourcesto most efficiently accomplish this mission

DISCLAIMER This statement is provided as aneducational service of the American Academy ofNeurology It is based on an assessment of currentscientific and clinical information It is not intendedto include all possible proper methods of care for aparticular neurologic problem or all legitimate crite-ria for choosing to use a specific procedure Neitheris it intended to exclude any reasonable alternativemethodologies The AAN recognizes that specificpatient care decisions are the prerogative of the pa-tient and physician caring for the patient based onall of the circumstances involved

CONFLICT OF INTEREST STATEMENT TheAmerican Academy of Neurology is committed toproducing independent critical and truthful clin-ical practice guidelines (CPGs) Significant effortsare made to minimize the potential for conflicts ofinterest to influence the recommendations of thisCPG To the extent possible the AAN keeps sep-arate those who have a financial stake in the suc-cess or failure of the products appraised in theCPGs and the developers of the guidelines Con-flict of interest forms were obtained from all au-thors and reviewed by an oversight committeeprior to project initiation AAN limits the participa-tion of authors with substantial conflicts of interestThe AAN forbids commercial participation in orfunding of guideline projects Drafts of the guide-lines have been reviewed by at least threeAANcom-mittees a network of neurologists Neurology peerreviewers and representatives from related fieldsTheAANGuidelineAuthorConflict of Interest Pol-icy can be viewed at wwwaancom

APPENDIX 1

Quality Standards Subcommittee members Jacque-line French MD FAAN (Co-Chair) Gary S Gronseth MD(Co-Chair) Charles E Argoff MD Stephen Ashwal MDFAAN (ex-officio) Christopher Bever Jr MD MBA FAANJohn D England MD FAAN Gary M Franklin MD MPH(ex-officio) GaryH FridayMDMPH FAAN Larry BGold-stein MD FAAN Deborah Hirtz MD (ex-officio) Robert GHollowayMDMPH FAANDonald J IversonMD FAANLeslie A Morrison MD Clifford J Schostal MD David JThurmanMDMPHWilliam JWeinerMD FAAN SamuelWiebe MD

APPENDIX 2AAN classification of evidence for rating of screen-ing article

Class I A statistical population-based sample of patientsstudied at a uniform point in time (usually early) during the

course of the condition All patients undergo the intervention ofinterest The outcome if not objective is determined in an eval-uation that is masked to the patientsrsquo clinical presentations

Class II A statistical non-referral-clinic-based sample ofpatients studied at a uniform point in time (usually early)during the course of the condition Most patients undergothe intervention of interest The outcome if not objective isdetermined in an evaluation that is masked to the patientsrsquoclinical presentations

Class III A sample of patients studied during the course ofthe condition Some patients undergo the intervention of in-terest The outcome if not objective is determined in anevaluation by someone other than the treating physician

Class IV Expert opinion case reports or any study notmeeting criteria for Class I to III

APPENDIX 3

Classification of recommendationsA Established as effective ineffective or harmful (or es-

tablished as usefulpredictive or not usefulpredictive)for the given condition in the specified population(Level A rating requires at least two consistent Class Istudies)

B Probably effective ineffective or harmful (or probablyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level B ratingrequires at least one Class I study or at least two consis-tent Class II studies)

C Possibly effective ineffective or harmful (or possiblyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level C ratingrequires at least one Class II study or two consistentClass III studies)

U Data inadequate or conflicting given current knowl-edge treatment (test predictor) is unproven (Studiesnot meeting criteria for Class IndashClass III)

In exceptional cases one convincing Class I study may suf-fice for an ldquoArdquo recommendation if 1) all criteria are met themagnitude of effect is large (relative rate improved outcome5 and the lower limit of the confidence interval is 2)

APPENDIX 4

Exclusion criteria

bull Letters and case reportsbull Non-English language studiesbull Animal studiesbull Pharmacodynamicpharmacokinetic studiesbull Studies dealing primarily with established or chronicepilepsy patients

bull Studies dealing primarily with acute provoked seizures

Inclusion criteria

bull Study designs observational (prospective retrospectiveand cross-sectional) or interventional (randomized con-trolled trials [RCTs] nonrandomized controlled trials[nRCTs] and uncontrolled case series [UCS])

bull At least 10 patients adults with a first seizure a first pre-sentation with epilepsy or seizures or a first diagnosis ofepilepsy

Neurology 69 November 20 2007 2005

RETIRED

RETIRED

bull Studies addressing any of the following diagnostic in-terventions neuropsychological assessment imagingwith CT MRI or PET scans EEG (standard videoinvasive ambulatory) lumbar puncture toxic screensand laboratory tests (hematologybiochemistry)

bull Patients should be over the age of 18 or the studyshould include a substantial proportion of subjectsover the age of 18

bull Studies reported in English onlybull Studies dealing primarily with apparent unprovokedfirst seizures

APPENDIX 5

Outline for seizure assessment Features of a seizureAssociated factors

AgeMedical history ndash previous history of similar episodesprior stroke brain tumor systemic illness mental ill-ness drug or alcohol abuse

Family historyDevelopmental statusBehaviorHealth at seizure onset ndash febrile ill exposed to illnesscomplaints of not feeling well sleep deprived

Precipitating events other than illness ndash trauma alcoholmedications illicit drugs toxins

Symptoms during seizure (ictal)Aura Subjective sensationsBehavior Mood or behavioral changes before the seizurePreictal symptoms Described by patient or witnessedVocal Cry or gasp slurring of words garbled speechMotor Head or eye turning eye deviation posturingjerking (rhythmic) stiffening automatisms (purpose-less repetitive movements such as picking at clothinglip smacking) generalized or focal movements

Respiration Change in breathing pattern cessation ofbreathing cyanosis

Autonomic Pupillary dilation drooling change in respira-tory or heart rate incontinence pallor vomiting

Loss of consciousness or inability to understand or speak

Symptoms following a seizure (postictal)Amnesia for eventsConfusionLethargySleepinessHeadaches and muscle achesTransient focal weakness (Toddrsquos paresis)Nausea or vomitingBiting of tongue

Adapted from Hirtz et al10

Received February 20 2007 Accepted in final form June 72007

REFERENCES1 Hauser WA Hesdorffer DC Epilepsy frequency

causes and consequences New York Demos Publica-tions 1990

2 Berg AT Shinnar S The risk of seizure recurrence fol-lowing a first unprovoked seizure a quantitative re-view Neurology 199141965ndash972

3 Begley CE Famulari M Annegers JF et al The cost ofepilepsy in the United States an estimate frompopulation- based and survey data Epilepsia 200041342ndash352

4 Grunewald RA Chroni E Panayiotopoulos CP De-layed diagnosis of juvenile myoclonic epilepsy J Neu-rol Neurosurg Psychiatry 199255497ndash499

5 Krumholz A Nonepileptic seizures diagnosis andmanagement Neurology 1999S76ndash83

6 Commission on Epidemiology and Prognosis Interna-tional League Against Epilepsy Guidelines for epidemio-logical studies on epilepsy Epilepsia 199334592ndash596

7 American College of Emergency Physicians (ACEP)Clinical Policies Committee and the Clinical PoliciesSubcommittee on Seizures Clinical Policy Critical is-sues in the evaluation of adult patients presenting to anemergency department with seizures Ann Emerg Med200443605ndash625

8 Fisher RS BoasWvE BlumeW et al Epileptic seizuresand epilepsy Definitions proposed by the InternationalLeague Against Epilepsy (ILAE) and the InternationalBureau for Epilepsy (IBE) Epilepsia 200546470ndash472

9 Agency for Healthcare Research and Quality (AHRQ)Management of newly diagnosed patients with epi-lepsy a systematic review of the literature (AHRQPublication No 01-E038) Rockville MD Agency forHealth Research Quality US Department of Healthand Human Services 2001

10 Hirtz D Ashwal S Berg A et al Practice parameterevaluating a first nonfebrile seizure in children Neu-rology 200055616ndash623

11 Jallon P Loiseau P Loiseau J Newly diagnosed unpro-voked epileptic seizures Presentation at diagnosis inCORALE study Epilepsia 200142464ndash475

12 Quality Standards Subcommittee and the Therapeuticsand Technology Assessment Subcommittee of theAmerican Academy of Neurology Clinical PracticeGuideline Process Manuel St Paul MN AmericanAcademy of Neurology 2004

13 Hauser WA Anderson VE Loewenson RB McRob-erts SM Seizure recurrence after a first unprovoked sei-zure N Engl J Med 1982307522ndash528

14 Hauser WA Rich SS Annegers JF Anderson VE Sei-zure recurrence after a 1st unprovoked seizure an ex-tended follow-up Neurology 1990401163ndash1170

15 van Donselaar CA Schimsheimer R Geerts AT et alValue of the electroencephalogram in adult patientswith untreated idiopathic first seizures Arch Neurol199249231ndash237

16 First Seizure Trial Group Randomized clinical trial onthe efficacy of antiepileptic drugs in reducing the risk ofrelapse after a first unprovoked tonic-clonic seizureNeurology 199343478ndash483

17 Bora I Seckin B Zarifoglu S Turin F Sadikogllu S OgulE Risk of recurrence after first unprovoked tonic-clonicseizure in adults J Neurol 1995242157ndash163

18 Hopkins A Garman A Clarke C The first seizure inadult life Value of clinical features electroencephalog-raphy and computerized tomographic scanning in pre-diction of seizure recurrence Lancet 1988721ndash726

19 Das CP Sawhney IMS Lal V et al Risk of recurrenceof seizures following a single unprovoked idiopathicseizure Neurol India 200046357ndash360

2006 Neurology 69 November 20 2007

RETIRED

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

RETIRED

In considering this issue we identified twoClass III studies considering the value of toxicol-ogy screening in adult patients presenting with afirst seizure3536 Although both studies reportedsome first seizures in patients with abnormal tox-icology screening neither study investigated theuse of routine screening in first seizure patients Inaddition the studies did not consider mainly un-provoked seizures in which the patient returnedto baseline function as our inclusion criteria spec-ify (appendix 4)

Conclusion In two Class III studies consideringthe value of toxicology screening in adult patientspresenting with a seizure some patients with ap-parent unprovoked first seizure were includedbut neither study investigated the use of routinetoxicology screening for such patients

Recommendation In the adult presenting with anapparent unprovoked seizure toxicology screen-ing may be helpful in specific clinical circum-stances but there are insufficient data to supportor refute a routine recommendation for toxicol-ogy screening (Level U)

CONCLUSIONS AND RECOMMENDATIONSDiagnosis of seizures determination of possiblecauses classification of seizure type and possibleseizure syndrome and assessment of the risk forrecurrence of seizures require timely selection andinterpretation of appropriate diagnostic tools in-cluding EEGs CTs or MRIs blood studies andlumbar puncture This report assesses the evi-dence available to answer the question ldquoWhat arethe most appropriate tools for evaluating a firstunprovoked seizure in adults at the time of initialpresentationrdquo

For the adult initially presenting with an ap-parent unprovoked seizure diagnostic evalua-tions influence aspects of patient care andmanagement including drug treatment patientand family counseling and the need for immedi-ate hospitalization and subsequent follow-upThis practice parameter reviews the published lit-erature concerning the yield and value of studiesfollowing an initial seizure in adults and classifiesthe strength of available evidence There is insuf-ficient Class I evidence to provide any recommen-dation with the highest degree of clinical certainty(see appendix 3) However there is sufficientClass I and Class II evidence that a routine EEGshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with an ap-parent unprovoked first seizure and predicts therisk of seizure recurrence (Level B) There is alsosufficient Class II evidence that a brain imaging

study either a CT or MRI helps determine thepresence of a significant abnormality such as abrain tumor Brain imaging using CT or MRIshould be considered as part of the neurodiagnos-tic evaluation of the adult presenting with a firstseizure (Level B) The CT has a major advantagebecause of the speed with which it can be ob-tained so its value is mainly in the emergency sit-uation but theMRI is regarded by expert opinionas having a higher yield and is the preferable pro-cedure in non-emergency or elective situa-tions91029 However both MRI and CT are ofvalue The decision to perform other studies in-cluding blood glucose blood counts electrolytepanels lumbar puncture or toxicology screeningfor the purpose of determining the cause of theseizure or identifying potentially treatable condi-tions will depend on the specific or individualizedclinical circumstances based on the history phys-ical and neurologic examination There are in-sufficient data to support or refute the use of anyone of these tests or procedures as a routine rec-ommendation (Level U)

The history physical and neurologic exami-nation are acknowledged by expert consensus tobe required for the initial diagnosis and classifica-tion of seizures26 These standards for the initialdiagnosis and classification of seizures form thebasis for the ldquoclinical and electroencephalo-graphic classification of epileptic seizuresrdquo21

Indeed information from the history includingthe type of the seizure21 together with findings fromthe physical and neurologic examination also aid inthe determination of a probable cause or risk of re-currence If the history reveals that there have beenother seizures in the past then this indicates epilepsyrather than a single isolated seizure Although thesubsequent diagnostic evaluation may be similarto that for a single unprovoked seizure there aredifferent prognostic and therapeutic implicationsfor a patient with epilepsy27 Evidence of pre-existing seizures provides information for the po-tential classification of an epilepsy syndrome25

Such information is of importance in establishingpossible causes associated comorbidities prog-nosis and optimal treatment of a seizure disorderor epilepsy syndrome2526 Moreover several fea-tures of the clinical history and physical and neuro-logic examination such as the presence of specificgeneralized or focal neurologic deficits are predic-tive of higher risks of seizure recurrence210

RECOMMENDATIONS FOR FUTURE RE-SEARCH The history physical and neurologicexaminations form the basis for the diagnosis and

Neurology 69 November 20 2007 2003

RETIRED

RETIRED

classification of epileptic seizures26 and this is ac-knowledged by expert consensus26 Still future re-search is warranted to determine the specificaspects of the history physical and neurologicexamination that are most useful in both initialdiagnosis of a seizure and subsequent manage-ment Also it would be helpful to understandhow this type of information should best guideperformance and timing of other neurodiagnosticor laboratory studies An international group re-cently recommended a broader definition of epi-lepsy as a brain disorder with a history of at leastone seizure and ldquocharacterized by an enduringpredisposition to generate epileptic seizuresrdquo8

This type of approach will require validation andinvestigation including determination of the truevalue of diagnostic studies such as EEG in estab-lishing the presence of an epileptic syndrome aftera first seizure In regard to the EEG one limita-tion of currently available studies is the variabilityin the timing of the EEG after incident seizureThere is some evidence in children that an EEGdone within 24 hours of a presenting seizure givesa higher yield of significant abnormalities10

Although all types of seizures are considered inthis analysis most patients had a convulsive sei-zure This may be because convulsive seizures aremore likely to prompt acute medical attentionthan a simple or complex partial seizure Morestudies specifically focusing attention on all typesof initial seizures not just convulsive seizures areneeded

Other factors deserve consideration in futureinvestigations For instance the costs of these var-ious tests do not receive much attention in studiesof patients with first seizure and should Also theelderly are a growing proportion of the generalpopulation and warrant special consideration infuture investigations Another unresolved issuerelates to the need for immediate hospitalizationof patients with a first seizure Some guidelinesexist but well designed studies are needed7

Prediction of seizure recurrence after a first un-provoked seizure is important to guide decisionsregarding antiepileptic drug therapy patientmanagement and counseling Many factors arereported to predict seizure recurrence includingthe etiology of the seizures and association withneurologic abnormalities2 Although the EEG andbrain imaging studies are reported to predict sei-zure recurrence studies vary in support of thatobservation and in respect to the specific EEG ab-normalities or brain imaging findings that aremost significant Some of this variability regard-ing EEG findings may relate to the timing of the

study after the initial event and to the decisionwhether to treat with antiepileptic medicationsThe value of specific findings on the EEG brainimaging and other variables deserve furtherstudy in regard to seizure recurrence risks Alsofuture studies of the clinical utility for such neuro-diagnostic procedures should employ standard-ized tests and prediction algorithms37

Future studies are needed to analyze the valueof expert input into the history physical neuro-logic examination or diagnostic testing such as theEEG or brain imaging The expertise and qualifica-tions of individuals interpreting such information ortests particularly the EEG or brain imaging varyconsiderably and may influence interpretationsand outcomes37 These are not simple ldquopositiverdquo orldquonegativerdquo tests like a pregnancy test they requireexpertise for proper interpretation Some other di-agnostic studies deserve further analysis in patientspresenting with new onset seizures In particularroutine electrolytes including serum sodium andglucosewere reported in some studies to be of valuebut this was not consistent or confirmed by otherreports Also toxicologic screening was reported asimportant in some limited series but was not stud-ied in a manner to allow adequate evidence-basedverification or refutation of its routine utility

Studies of patient management and counselingat the time of first diagnosis of a single seizure orepilepsy are recommended For example whichpatients should receive antiepileptic drug treat-ment and if so exactly what type and whenshould they get it Also when is the patient bestadvised regarding driving laws and other socialissues and how and by whom should this bedone This is not a trivial issue For instance aBritish study noted that only 21 of all adult firstseizure patients received the correct advice aboutdriving limitations23

We recommend that future studies address theissues that we raise here To best determine howto optimally evaluate the adult presenting with afirst seizure those studies should be structured toemphasize the use of large well-characterizedsamples clearly defined subjects and outcomesand standardized data collection methods

MISSION STATEMENT OF QSS The missionof the QSS is to prioritize develop and publishevidence-based practice parameters related to thediagnosis treatment and prognosis of neurologicdisorders

The QSS is committed to using the most rigor-ous methods available within our budget in col-

2004 Neurology 69 November 20 2007

RETIRED

RETIRED

laboration with other available AAN resourcesto most efficiently accomplish this mission

DISCLAIMER This statement is provided as aneducational service of the American Academy ofNeurology It is based on an assessment of currentscientific and clinical information It is not intendedto include all possible proper methods of care for aparticular neurologic problem or all legitimate crite-ria for choosing to use a specific procedure Neitheris it intended to exclude any reasonable alternativemethodologies The AAN recognizes that specificpatient care decisions are the prerogative of the pa-tient and physician caring for the patient based onall of the circumstances involved

CONFLICT OF INTEREST STATEMENT TheAmerican Academy of Neurology is committed toproducing independent critical and truthful clin-ical practice guidelines (CPGs) Significant effortsare made to minimize the potential for conflicts ofinterest to influence the recommendations of thisCPG To the extent possible the AAN keeps sep-arate those who have a financial stake in the suc-cess or failure of the products appraised in theCPGs and the developers of the guidelines Con-flict of interest forms were obtained from all au-thors and reviewed by an oversight committeeprior to project initiation AAN limits the participa-tion of authors with substantial conflicts of interestThe AAN forbids commercial participation in orfunding of guideline projects Drafts of the guide-lines have been reviewed by at least threeAANcom-mittees a network of neurologists Neurology peerreviewers and representatives from related fieldsTheAANGuidelineAuthorConflict of Interest Pol-icy can be viewed at wwwaancom

APPENDIX 1

Quality Standards Subcommittee members Jacque-line French MD FAAN (Co-Chair) Gary S Gronseth MD(Co-Chair) Charles E Argoff MD Stephen Ashwal MDFAAN (ex-officio) Christopher Bever Jr MD MBA FAANJohn D England MD FAAN Gary M Franklin MD MPH(ex-officio) GaryH FridayMDMPH FAAN Larry BGold-stein MD FAAN Deborah Hirtz MD (ex-officio) Robert GHollowayMDMPH FAANDonald J IversonMD FAANLeslie A Morrison MD Clifford J Schostal MD David JThurmanMDMPHWilliam JWeinerMD FAAN SamuelWiebe MD

APPENDIX 2AAN classification of evidence for rating of screen-ing article

Class I A statistical population-based sample of patientsstudied at a uniform point in time (usually early) during the

course of the condition All patients undergo the intervention ofinterest The outcome if not objective is determined in an eval-uation that is masked to the patientsrsquo clinical presentations

Class II A statistical non-referral-clinic-based sample ofpatients studied at a uniform point in time (usually early)during the course of the condition Most patients undergothe intervention of interest The outcome if not objective isdetermined in an evaluation that is masked to the patientsrsquoclinical presentations

Class III A sample of patients studied during the course ofthe condition Some patients undergo the intervention of in-terest The outcome if not objective is determined in anevaluation by someone other than the treating physician

Class IV Expert opinion case reports or any study notmeeting criteria for Class I to III

APPENDIX 3

Classification of recommendationsA Established as effective ineffective or harmful (or es-

tablished as usefulpredictive or not usefulpredictive)for the given condition in the specified population(Level A rating requires at least two consistent Class Istudies)

B Probably effective ineffective or harmful (or probablyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level B ratingrequires at least one Class I study or at least two consis-tent Class II studies)

C Possibly effective ineffective or harmful (or possiblyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level C ratingrequires at least one Class II study or two consistentClass III studies)

U Data inadequate or conflicting given current knowl-edge treatment (test predictor) is unproven (Studiesnot meeting criteria for Class IndashClass III)

In exceptional cases one convincing Class I study may suf-fice for an ldquoArdquo recommendation if 1) all criteria are met themagnitude of effect is large (relative rate improved outcome5 and the lower limit of the confidence interval is 2)

APPENDIX 4

Exclusion criteria

bull Letters and case reportsbull Non-English language studiesbull Animal studiesbull Pharmacodynamicpharmacokinetic studiesbull Studies dealing primarily with established or chronicepilepsy patients

bull Studies dealing primarily with acute provoked seizures

Inclusion criteria

bull Study designs observational (prospective retrospectiveand cross-sectional) or interventional (randomized con-trolled trials [RCTs] nonrandomized controlled trials[nRCTs] and uncontrolled case series [UCS])

bull At least 10 patients adults with a first seizure a first pre-sentation with epilepsy or seizures or a first diagnosis ofepilepsy

Neurology 69 November 20 2007 2005

RETIRED

RETIRED

bull Studies addressing any of the following diagnostic in-terventions neuropsychological assessment imagingwith CT MRI or PET scans EEG (standard videoinvasive ambulatory) lumbar puncture toxic screensand laboratory tests (hematologybiochemistry)

bull Patients should be over the age of 18 or the studyshould include a substantial proportion of subjectsover the age of 18

bull Studies reported in English onlybull Studies dealing primarily with apparent unprovokedfirst seizures

APPENDIX 5

Outline for seizure assessment Features of a seizureAssociated factors

AgeMedical history ndash previous history of similar episodesprior stroke brain tumor systemic illness mental ill-ness drug or alcohol abuse

Family historyDevelopmental statusBehaviorHealth at seizure onset ndash febrile ill exposed to illnesscomplaints of not feeling well sleep deprived

Precipitating events other than illness ndash trauma alcoholmedications illicit drugs toxins

Symptoms during seizure (ictal)Aura Subjective sensationsBehavior Mood or behavioral changes before the seizurePreictal symptoms Described by patient or witnessedVocal Cry or gasp slurring of words garbled speechMotor Head or eye turning eye deviation posturingjerking (rhythmic) stiffening automatisms (purpose-less repetitive movements such as picking at clothinglip smacking) generalized or focal movements

Respiration Change in breathing pattern cessation ofbreathing cyanosis

Autonomic Pupillary dilation drooling change in respira-tory or heart rate incontinence pallor vomiting

Loss of consciousness or inability to understand or speak

Symptoms following a seizure (postictal)Amnesia for eventsConfusionLethargySleepinessHeadaches and muscle achesTransient focal weakness (Toddrsquos paresis)Nausea or vomitingBiting of tongue

Adapted from Hirtz et al10

Received February 20 2007 Accepted in final form June 72007

REFERENCES1 Hauser WA Hesdorffer DC Epilepsy frequency

causes and consequences New York Demos Publica-tions 1990

2 Berg AT Shinnar S The risk of seizure recurrence fol-lowing a first unprovoked seizure a quantitative re-view Neurology 199141965ndash972

3 Begley CE Famulari M Annegers JF et al The cost ofepilepsy in the United States an estimate frompopulation- based and survey data Epilepsia 200041342ndash352

4 Grunewald RA Chroni E Panayiotopoulos CP De-layed diagnosis of juvenile myoclonic epilepsy J Neu-rol Neurosurg Psychiatry 199255497ndash499

5 Krumholz A Nonepileptic seizures diagnosis andmanagement Neurology 1999S76ndash83

6 Commission on Epidemiology and Prognosis Interna-tional League Against Epilepsy Guidelines for epidemio-logical studies on epilepsy Epilepsia 199334592ndash596

7 American College of Emergency Physicians (ACEP)Clinical Policies Committee and the Clinical PoliciesSubcommittee on Seizures Clinical Policy Critical is-sues in the evaluation of adult patients presenting to anemergency department with seizures Ann Emerg Med200443605ndash625

8 Fisher RS BoasWvE BlumeW et al Epileptic seizuresand epilepsy Definitions proposed by the InternationalLeague Against Epilepsy (ILAE) and the InternationalBureau for Epilepsy (IBE) Epilepsia 200546470ndash472

9 Agency for Healthcare Research and Quality (AHRQ)Management of newly diagnosed patients with epi-lepsy a systematic review of the literature (AHRQPublication No 01-E038) Rockville MD Agency forHealth Research Quality US Department of Healthand Human Services 2001

10 Hirtz D Ashwal S Berg A et al Practice parameterevaluating a first nonfebrile seizure in children Neu-rology 200055616ndash623

11 Jallon P Loiseau P Loiseau J Newly diagnosed unpro-voked epileptic seizures Presentation at diagnosis inCORALE study Epilepsia 200142464ndash475

12 Quality Standards Subcommittee and the Therapeuticsand Technology Assessment Subcommittee of theAmerican Academy of Neurology Clinical PracticeGuideline Process Manuel St Paul MN AmericanAcademy of Neurology 2004

13 Hauser WA Anderson VE Loewenson RB McRob-erts SM Seizure recurrence after a first unprovoked sei-zure N Engl J Med 1982307522ndash528

14 Hauser WA Rich SS Annegers JF Anderson VE Sei-zure recurrence after a 1st unprovoked seizure an ex-tended follow-up Neurology 1990401163ndash1170

15 van Donselaar CA Schimsheimer R Geerts AT et alValue of the electroencephalogram in adult patientswith untreated idiopathic first seizures Arch Neurol199249231ndash237

16 First Seizure Trial Group Randomized clinical trial onthe efficacy of antiepileptic drugs in reducing the risk ofrelapse after a first unprovoked tonic-clonic seizureNeurology 199343478ndash483

17 Bora I Seckin B Zarifoglu S Turin F Sadikogllu S OgulE Risk of recurrence after first unprovoked tonic-clonicseizure in adults J Neurol 1995242157ndash163

18 Hopkins A Garman A Clarke C The first seizure inadult life Value of clinical features electroencephalog-raphy and computerized tomographic scanning in pre-diction of seizure recurrence Lancet 1988721ndash726

19 Das CP Sawhney IMS Lal V et al Risk of recurrenceof seizures following a single unprovoked idiopathicseizure Neurol India 200046357ndash360

2006 Neurology 69 November 20 2007

RETIRED

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

RETIRED

classification of epileptic seizures26 and this is ac-knowledged by expert consensus26 Still future re-search is warranted to determine the specificaspects of the history physical and neurologicexamination that are most useful in both initialdiagnosis of a seizure and subsequent manage-ment Also it would be helpful to understandhow this type of information should best guideperformance and timing of other neurodiagnosticor laboratory studies An international group re-cently recommended a broader definition of epi-lepsy as a brain disorder with a history of at leastone seizure and ldquocharacterized by an enduringpredisposition to generate epileptic seizuresrdquo8

This type of approach will require validation andinvestigation including determination of the truevalue of diagnostic studies such as EEG in estab-lishing the presence of an epileptic syndrome aftera first seizure In regard to the EEG one limita-tion of currently available studies is the variabilityin the timing of the EEG after incident seizureThere is some evidence in children that an EEGdone within 24 hours of a presenting seizure givesa higher yield of significant abnormalities10

Although all types of seizures are considered inthis analysis most patients had a convulsive sei-zure This may be because convulsive seizures aremore likely to prompt acute medical attentionthan a simple or complex partial seizure Morestudies specifically focusing attention on all typesof initial seizures not just convulsive seizures areneeded

Other factors deserve consideration in futureinvestigations For instance the costs of these var-ious tests do not receive much attention in studiesof patients with first seizure and should Also theelderly are a growing proportion of the generalpopulation and warrant special consideration infuture investigations Another unresolved issuerelates to the need for immediate hospitalizationof patients with a first seizure Some guidelinesexist but well designed studies are needed7

Prediction of seizure recurrence after a first un-provoked seizure is important to guide decisionsregarding antiepileptic drug therapy patientmanagement and counseling Many factors arereported to predict seizure recurrence includingthe etiology of the seizures and association withneurologic abnormalities2 Although the EEG andbrain imaging studies are reported to predict sei-zure recurrence studies vary in support of thatobservation and in respect to the specific EEG ab-normalities or brain imaging findings that aremost significant Some of this variability regard-ing EEG findings may relate to the timing of the

study after the initial event and to the decisionwhether to treat with antiepileptic medicationsThe value of specific findings on the EEG brainimaging and other variables deserve furtherstudy in regard to seizure recurrence risks Alsofuture studies of the clinical utility for such neuro-diagnostic procedures should employ standard-ized tests and prediction algorithms37

Future studies are needed to analyze the valueof expert input into the history physical neuro-logic examination or diagnostic testing such as theEEG or brain imaging The expertise and qualifica-tions of individuals interpreting such information ortests particularly the EEG or brain imaging varyconsiderably and may influence interpretationsand outcomes37 These are not simple ldquopositiverdquo orldquonegativerdquo tests like a pregnancy test they requireexpertise for proper interpretation Some other di-agnostic studies deserve further analysis in patientspresenting with new onset seizures In particularroutine electrolytes including serum sodium andglucosewere reported in some studies to be of valuebut this was not consistent or confirmed by otherreports Also toxicologic screening was reported asimportant in some limited series but was not stud-ied in a manner to allow adequate evidence-basedverification or refutation of its routine utility

Studies of patient management and counselingat the time of first diagnosis of a single seizure orepilepsy are recommended For example whichpatients should receive antiepileptic drug treat-ment and if so exactly what type and whenshould they get it Also when is the patient bestadvised regarding driving laws and other socialissues and how and by whom should this bedone This is not a trivial issue For instance aBritish study noted that only 21 of all adult firstseizure patients received the correct advice aboutdriving limitations23

We recommend that future studies address theissues that we raise here To best determine howto optimally evaluate the adult presenting with afirst seizure those studies should be structured toemphasize the use of large well-characterizedsamples clearly defined subjects and outcomesand standardized data collection methods

MISSION STATEMENT OF QSS The missionof the QSS is to prioritize develop and publishevidence-based practice parameters related to thediagnosis treatment and prognosis of neurologicdisorders

The QSS is committed to using the most rigor-ous methods available within our budget in col-

2004 Neurology 69 November 20 2007

RETIRED

RETIRED

laboration with other available AAN resourcesto most efficiently accomplish this mission

DISCLAIMER This statement is provided as aneducational service of the American Academy ofNeurology It is based on an assessment of currentscientific and clinical information It is not intendedto include all possible proper methods of care for aparticular neurologic problem or all legitimate crite-ria for choosing to use a specific procedure Neitheris it intended to exclude any reasonable alternativemethodologies The AAN recognizes that specificpatient care decisions are the prerogative of the pa-tient and physician caring for the patient based onall of the circumstances involved

CONFLICT OF INTEREST STATEMENT TheAmerican Academy of Neurology is committed toproducing independent critical and truthful clin-ical practice guidelines (CPGs) Significant effortsare made to minimize the potential for conflicts ofinterest to influence the recommendations of thisCPG To the extent possible the AAN keeps sep-arate those who have a financial stake in the suc-cess or failure of the products appraised in theCPGs and the developers of the guidelines Con-flict of interest forms were obtained from all au-thors and reviewed by an oversight committeeprior to project initiation AAN limits the participa-tion of authors with substantial conflicts of interestThe AAN forbids commercial participation in orfunding of guideline projects Drafts of the guide-lines have been reviewed by at least threeAANcom-mittees a network of neurologists Neurology peerreviewers and representatives from related fieldsTheAANGuidelineAuthorConflict of Interest Pol-icy can be viewed at wwwaancom

APPENDIX 1

Quality Standards Subcommittee members Jacque-line French MD FAAN (Co-Chair) Gary S Gronseth MD(Co-Chair) Charles E Argoff MD Stephen Ashwal MDFAAN (ex-officio) Christopher Bever Jr MD MBA FAANJohn D England MD FAAN Gary M Franklin MD MPH(ex-officio) GaryH FridayMDMPH FAAN Larry BGold-stein MD FAAN Deborah Hirtz MD (ex-officio) Robert GHollowayMDMPH FAANDonald J IversonMD FAANLeslie A Morrison MD Clifford J Schostal MD David JThurmanMDMPHWilliam JWeinerMD FAAN SamuelWiebe MD

APPENDIX 2AAN classification of evidence for rating of screen-ing article

Class I A statistical population-based sample of patientsstudied at a uniform point in time (usually early) during the

course of the condition All patients undergo the intervention ofinterest The outcome if not objective is determined in an eval-uation that is masked to the patientsrsquo clinical presentations

Class II A statistical non-referral-clinic-based sample ofpatients studied at a uniform point in time (usually early)during the course of the condition Most patients undergothe intervention of interest The outcome if not objective isdetermined in an evaluation that is masked to the patientsrsquoclinical presentations

Class III A sample of patients studied during the course ofthe condition Some patients undergo the intervention of in-terest The outcome if not objective is determined in anevaluation by someone other than the treating physician

Class IV Expert opinion case reports or any study notmeeting criteria for Class I to III

APPENDIX 3

Classification of recommendationsA Established as effective ineffective or harmful (or es-

tablished as usefulpredictive or not usefulpredictive)for the given condition in the specified population(Level A rating requires at least two consistent Class Istudies)

B Probably effective ineffective or harmful (or probablyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level B ratingrequires at least one Class I study or at least two consis-tent Class II studies)

C Possibly effective ineffective or harmful (or possiblyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level C ratingrequires at least one Class II study or two consistentClass III studies)

U Data inadequate or conflicting given current knowl-edge treatment (test predictor) is unproven (Studiesnot meeting criteria for Class IndashClass III)

In exceptional cases one convincing Class I study may suf-fice for an ldquoArdquo recommendation if 1) all criteria are met themagnitude of effect is large (relative rate improved outcome5 and the lower limit of the confidence interval is 2)

APPENDIX 4

Exclusion criteria

bull Letters and case reportsbull Non-English language studiesbull Animal studiesbull Pharmacodynamicpharmacokinetic studiesbull Studies dealing primarily with established or chronicepilepsy patients

bull Studies dealing primarily with acute provoked seizures

Inclusion criteria

bull Study designs observational (prospective retrospectiveand cross-sectional) or interventional (randomized con-trolled trials [RCTs] nonrandomized controlled trials[nRCTs] and uncontrolled case series [UCS])

bull At least 10 patients adults with a first seizure a first pre-sentation with epilepsy or seizures or a first diagnosis ofepilepsy

Neurology 69 November 20 2007 2005

RETIRED

RETIRED

bull Studies addressing any of the following diagnostic in-terventions neuropsychological assessment imagingwith CT MRI or PET scans EEG (standard videoinvasive ambulatory) lumbar puncture toxic screensand laboratory tests (hematologybiochemistry)

bull Patients should be over the age of 18 or the studyshould include a substantial proportion of subjectsover the age of 18

bull Studies reported in English onlybull Studies dealing primarily with apparent unprovokedfirst seizures

APPENDIX 5

Outline for seizure assessment Features of a seizureAssociated factors

AgeMedical history ndash previous history of similar episodesprior stroke brain tumor systemic illness mental ill-ness drug or alcohol abuse

Family historyDevelopmental statusBehaviorHealth at seizure onset ndash febrile ill exposed to illnesscomplaints of not feeling well sleep deprived

Precipitating events other than illness ndash trauma alcoholmedications illicit drugs toxins

Symptoms during seizure (ictal)Aura Subjective sensationsBehavior Mood or behavioral changes before the seizurePreictal symptoms Described by patient or witnessedVocal Cry or gasp slurring of words garbled speechMotor Head or eye turning eye deviation posturingjerking (rhythmic) stiffening automatisms (purpose-less repetitive movements such as picking at clothinglip smacking) generalized or focal movements

Respiration Change in breathing pattern cessation ofbreathing cyanosis

Autonomic Pupillary dilation drooling change in respira-tory or heart rate incontinence pallor vomiting

Loss of consciousness or inability to understand or speak

Symptoms following a seizure (postictal)Amnesia for eventsConfusionLethargySleepinessHeadaches and muscle achesTransient focal weakness (Toddrsquos paresis)Nausea or vomitingBiting of tongue

Adapted from Hirtz et al10

Received February 20 2007 Accepted in final form June 72007

REFERENCES1 Hauser WA Hesdorffer DC Epilepsy frequency

causes and consequences New York Demos Publica-tions 1990

2 Berg AT Shinnar S The risk of seizure recurrence fol-lowing a first unprovoked seizure a quantitative re-view Neurology 199141965ndash972

3 Begley CE Famulari M Annegers JF et al The cost ofepilepsy in the United States an estimate frompopulation- based and survey data Epilepsia 200041342ndash352

4 Grunewald RA Chroni E Panayiotopoulos CP De-layed diagnosis of juvenile myoclonic epilepsy J Neu-rol Neurosurg Psychiatry 199255497ndash499

5 Krumholz A Nonepileptic seizures diagnosis andmanagement Neurology 1999S76ndash83

6 Commission on Epidemiology and Prognosis Interna-tional League Against Epilepsy Guidelines for epidemio-logical studies on epilepsy Epilepsia 199334592ndash596

7 American College of Emergency Physicians (ACEP)Clinical Policies Committee and the Clinical PoliciesSubcommittee on Seizures Clinical Policy Critical is-sues in the evaluation of adult patients presenting to anemergency department with seizures Ann Emerg Med200443605ndash625

8 Fisher RS BoasWvE BlumeW et al Epileptic seizuresand epilepsy Definitions proposed by the InternationalLeague Against Epilepsy (ILAE) and the InternationalBureau for Epilepsy (IBE) Epilepsia 200546470ndash472

9 Agency for Healthcare Research and Quality (AHRQ)Management of newly diagnosed patients with epi-lepsy a systematic review of the literature (AHRQPublication No 01-E038) Rockville MD Agency forHealth Research Quality US Department of Healthand Human Services 2001

10 Hirtz D Ashwal S Berg A et al Practice parameterevaluating a first nonfebrile seizure in children Neu-rology 200055616ndash623

11 Jallon P Loiseau P Loiseau J Newly diagnosed unpro-voked epileptic seizures Presentation at diagnosis inCORALE study Epilepsia 200142464ndash475

12 Quality Standards Subcommittee and the Therapeuticsand Technology Assessment Subcommittee of theAmerican Academy of Neurology Clinical PracticeGuideline Process Manuel St Paul MN AmericanAcademy of Neurology 2004

13 Hauser WA Anderson VE Loewenson RB McRob-erts SM Seizure recurrence after a first unprovoked sei-zure N Engl J Med 1982307522ndash528

14 Hauser WA Rich SS Annegers JF Anderson VE Sei-zure recurrence after a 1st unprovoked seizure an ex-tended follow-up Neurology 1990401163ndash1170

15 van Donselaar CA Schimsheimer R Geerts AT et alValue of the electroencephalogram in adult patientswith untreated idiopathic first seizures Arch Neurol199249231ndash237

16 First Seizure Trial Group Randomized clinical trial onthe efficacy of antiepileptic drugs in reducing the risk ofrelapse after a first unprovoked tonic-clonic seizureNeurology 199343478ndash483

17 Bora I Seckin B Zarifoglu S Turin F Sadikogllu S OgulE Risk of recurrence after first unprovoked tonic-clonicseizure in adults J Neurol 1995242157ndash163

18 Hopkins A Garman A Clarke C The first seizure inadult life Value of clinical features electroencephalog-raphy and computerized tomographic scanning in pre-diction of seizure recurrence Lancet 1988721ndash726

19 Das CP Sawhney IMS Lal V et al Risk of recurrenceof seizures following a single unprovoked idiopathicseizure Neurol India 200046357ndash360

2006 Neurology 69 November 20 2007

RETIRED

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

RETIRED

laboration with other available AAN resourcesto most efficiently accomplish this mission

DISCLAIMER This statement is provided as aneducational service of the American Academy ofNeurology It is based on an assessment of currentscientific and clinical information It is not intendedto include all possible proper methods of care for aparticular neurologic problem or all legitimate crite-ria for choosing to use a specific procedure Neitheris it intended to exclude any reasonable alternativemethodologies The AAN recognizes that specificpatient care decisions are the prerogative of the pa-tient and physician caring for the patient based onall of the circumstances involved

CONFLICT OF INTEREST STATEMENT TheAmerican Academy of Neurology is committed toproducing independent critical and truthful clin-ical practice guidelines (CPGs) Significant effortsare made to minimize the potential for conflicts ofinterest to influence the recommendations of thisCPG To the extent possible the AAN keeps sep-arate those who have a financial stake in the suc-cess or failure of the products appraised in theCPGs and the developers of the guidelines Con-flict of interest forms were obtained from all au-thors and reviewed by an oversight committeeprior to project initiation AAN limits the participa-tion of authors with substantial conflicts of interestThe AAN forbids commercial participation in orfunding of guideline projects Drafts of the guide-lines have been reviewed by at least threeAANcom-mittees a network of neurologists Neurology peerreviewers and representatives from related fieldsTheAANGuidelineAuthorConflict of Interest Pol-icy can be viewed at wwwaancom

APPENDIX 1

Quality Standards Subcommittee members Jacque-line French MD FAAN (Co-Chair) Gary S Gronseth MD(Co-Chair) Charles E Argoff MD Stephen Ashwal MDFAAN (ex-officio) Christopher Bever Jr MD MBA FAANJohn D England MD FAAN Gary M Franklin MD MPH(ex-officio) GaryH FridayMDMPH FAAN Larry BGold-stein MD FAAN Deborah Hirtz MD (ex-officio) Robert GHollowayMDMPH FAANDonald J IversonMD FAANLeslie A Morrison MD Clifford J Schostal MD David JThurmanMDMPHWilliam JWeinerMD FAAN SamuelWiebe MD

APPENDIX 2AAN classification of evidence for rating of screen-ing article

Class I A statistical population-based sample of patientsstudied at a uniform point in time (usually early) during the

course of the condition All patients undergo the intervention ofinterest The outcome if not objective is determined in an eval-uation that is masked to the patientsrsquo clinical presentations

Class II A statistical non-referral-clinic-based sample ofpatients studied at a uniform point in time (usually early)during the course of the condition Most patients undergothe intervention of interest The outcome if not objective isdetermined in an evaluation that is masked to the patientsrsquoclinical presentations

Class III A sample of patients studied during the course ofthe condition Some patients undergo the intervention of in-terest The outcome if not objective is determined in anevaluation by someone other than the treating physician

Class IV Expert opinion case reports or any study notmeeting criteria for Class I to III

APPENDIX 3

Classification of recommendationsA Established as effective ineffective or harmful (or es-

tablished as usefulpredictive or not usefulpredictive)for the given condition in the specified population(Level A rating requires at least two consistent Class Istudies)

B Probably effective ineffective or harmful (or probablyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level B ratingrequires at least one Class I study or at least two consis-tent Class II studies)

C Possibly effective ineffective or harmful (or possiblyusefulpredictive or not usefulpredictive) for the givencondition in the specified population (Level C ratingrequires at least one Class II study or two consistentClass III studies)

U Data inadequate or conflicting given current knowl-edge treatment (test predictor) is unproven (Studiesnot meeting criteria for Class IndashClass III)

In exceptional cases one convincing Class I study may suf-fice for an ldquoArdquo recommendation if 1) all criteria are met themagnitude of effect is large (relative rate improved outcome5 and the lower limit of the confidence interval is 2)

APPENDIX 4

Exclusion criteria

bull Letters and case reportsbull Non-English language studiesbull Animal studiesbull Pharmacodynamicpharmacokinetic studiesbull Studies dealing primarily with established or chronicepilepsy patients

bull Studies dealing primarily with acute provoked seizures

Inclusion criteria

bull Study designs observational (prospective retrospectiveand cross-sectional) or interventional (randomized con-trolled trials [RCTs] nonrandomized controlled trials[nRCTs] and uncontrolled case series [UCS])

bull At least 10 patients adults with a first seizure a first pre-sentation with epilepsy or seizures or a first diagnosis ofepilepsy

Neurology 69 November 20 2007 2005

RETIRED

RETIRED

bull Studies addressing any of the following diagnostic in-terventions neuropsychological assessment imagingwith CT MRI or PET scans EEG (standard videoinvasive ambulatory) lumbar puncture toxic screensand laboratory tests (hematologybiochemistry)

bull Patients should be over the age of 18 or the studyshould include a substantial proportion of subjectsover the age of 18

bull Studies reported in English onlybull Studies dealing primarily with apparent unprovokedfirst seizures

APPENDIX 5

Outline for seizure assessment Features of a seizureAssociated factors

AgeMedical history ndash previous history of similar episodesprior stroke brain tumor systemic illness mental ill-ness drug or alcohol abuse

Family historyDevelopmental statusBehaviorHealth at seizure onset ndash febrile ill exposed to illnesscomplaints of not feeling well sleep deprived

Precipitating events other than illness ndash trauma alcoholmedications illicit drugs toxins

Symptoms during seizure (ictal)Aura Subjective sensationsBehavior Mood or behavioral changes before the seizurePreictal symptoms Described by patient or witnessedVocal Cry or gasp slurring of words garbled speechMotor Head or eye turning eye deviation posturingjerking (rhythmic) stiffening automatisms (purpose-less repetitive movements such as picking at clothinglip smacking) generalized or focal movements

Respiration Change in breathing pattern cessation ofbreathing cyanosis

Autonomic Pupillary dilation drooling change in respira-tory or heart rate incontinence pallor vomiting

Loss of consciousness or inability to understand or speak

Symptoms following a seizure (postictal)Amnesia for eventsConfusionLethargySleepinessHeadaches and muscle achesTransient focal weakness (Toddrsquos paresis)Nausea or vomitingBiting of tongue

Adapted from Hirtz et al10

Received February 20 2007 Accepted in final form June 72007

REFERENCES1 Hauser WA Hesdorffer DC Epilepsy frequency

causes and consequences New York Demos Publica-tions 1990

2 Berg AT Shinnar S The risk of seizure recurrence fol-lowing a first unprovoked seizure a quantitative re-view Neurology 199141965ndash972

3 Begley CE Famulari M Annegers JF et al The cost ofepilepsy in the United States an estimate frompopulation- based and survey data Epilepsia 200041342ndash352

4 Grunewald RA Chroni E Panayiotopoulos CP De-layed diagnosis of juvenile myoclonic epilepsy J Neu-rol Neurosurg Psychiatry 199255497ndash499

5 Krumholz A Nonepileptic seizures diagnosis andmanagement Neurology 1999S76ndash83

6 Commission on Epidemiology and Prognosis Interna-tional League Against Epilepsy Guidelines for epidemio-logical studies on epilepsy Epilepsia 199334592ndash596

7 American College of Emergency Physicians (ACEP)Clinical Policies Committee and the Clinical PoliciesSubcommittee on Seizures Clinical Policy Critical is-sues in the evaluation of adult patients presenting to anemergency department with seizures Ann Emerg Med200443605ndash625

8 Fisher RS BoasWvE BlumeW et al Epileptic seizuresand epilepsy Definitions proposed by the InternationalLeague Against Epilepsy (ILAE) and the InternationalBureau for Epilepsy (IBE) Epilepsia 200546470ndash472

9 Agency for Healthcare Research and Quality (AHRQ)Management of newly diagnosed patients with epi-lepsy a systematic review of the literature (AHRQPublication No 01-E038) Rockville MD Agency forHealth Research Quality US Department of Healthand Human Services 2001

10 Hirtz D Ashwal S Berg A et al Practice parameterevaluating a first nonfebrile seizure in children Neu-rology 200055616ndash623

11 Jallon P Loiseau P Loiseau J Newly diagnosed unpro-voked epileptic seizures Presentation at diagnosis inCORALE study Epilepsia 200142464ndash475

12 Quality Standards Subcommittee and the Therapeuticsand Technology Assessment Subcommittee of theAmerican Academy of Neurology Clinical PracticeGuideline Process Manuel St Paul MN AmericanAcademy of Neurology 2004

13 Hauser WA Anderson VE Loewenson RB McRob-erts SM Seizure recurrence after a first unprovoked sei-zure N Engl J Med 1982307522ndash528

14 Hauser WA Rich SS Annegers JF Anderson VE Sei-zure recurrence after a 1st unprovoked seizure an ex-tended follow-up Neurology 1990401163ndash1170

15 van Donselaar CA Schimsheimer R Geerts AT et alValue of the electroencephalogram in adult patientswith untreated idiopathic first seizures Arch Neurol199249231ndash237

16 First Seizure Trial Group Randomized clinical trial onthe efficacy of antiepileptic drugs in reducing the risk ofrelapse after a first unprovoked tonic-clonic seizureNeurology 199343478ndash483

17 Bora I Seckin B Zarifoglu S Turin F Sadikogllu S OgulE Risk of recurrence after first unprovoked tonic-clonicseizure in adults J Neurol 1995242157ndash163

18 Hopkins A Garman A Clarke C The first seizure inadult life Value of clinical features electroencephalog-raphy and computerized tomographic scanning in pre-diction of seizure recurrence Lancet 1988721ndash726

19 Das CP Sawhney IMS Lal V et al Risk of recurrenceof seizures following a single unprovoked idiopathicseizure Neurol India 200046357ndash360

2006 Neurology 69 November 20 2007

RETIRED

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

RETIRED

bull Studies addressing any of the following diagnostic in-terventions neuropsychological assessment imagingwith CT MRI or PET scans EEG (standard videoinvasive ambulatory) lumbar puncture toxic screensand laboratory tests (hematologybiochemistry)

bull Patients should be over the age of 18 or the studyshould include a substantial proportion of subjectsover the age of 18

bull Studies reported in English onlybull Studies dealing primarily with apparent unprovokedfirst seizures

APPENDIX 5

Outline for seizure assessment Features of a seizureAssociated factors

AgeMedical history ndash previous history of similar episodesprior stroke brain tumor systemic illness mental ill-ness drug or alcohol abuse

Family historyDevelopmental statusBehaviorHealth at seizure onset ndash febrile ill exposed to illnesscomplaints of not feeling well sleep deprived

Precipitating events other than illness ndash trauma alcoholmedications illicit drugs toxins

Symptoms during seizure (ictal)Aura Subjective sensationsBehavior Mood or behavioral changes before the seizurePreictal symptoms Described by patient or witnessedVocal Cry or gasp slurring of words garbled speechMotor Head or eye turning eye deviation posturingjerking (rhythmic) stiffening automatisms (purpose-less repetitive movements such as picking at clothinglip smacking) generalized or focal movements

Respiration Change in breathing pattern cessation ofbreathing cyanosis

Autonomic Pupillary dilation drooling change in respira-tory or heart rate incontinence pallor vomiting

Loss of consciousness or inability to understand or speak

Symptoms following a seizure (postictal)Amnesia for eventsConfusionLethargySleepinessHeadaches and muscle achesTransient focal weakness (Toddrsquos paresis)Nausea or vomitingBiting of tongue

Adapted from Hirtz et al10

Received February 20 2007 Accepted in final form June 72007

REFERENCES1 Hauser WA Hesdorffer DC Epilepsy frequency

causes and consequences New York Demos Publica-tions 1990

2 Berg AT Shinnar S The risk of seizure recurrence fol-lowing a first unprovoked seizure a quantitative re-view Neurology 199141965ndash972

3 Begley CE Famulari M Annegers JF et al The cost ofepilepsy in the United States an estimate frompopulation- based and survey data Epilepsia 200041342ndash352

4 Grunewald RA Chroni E Panayiotopoulos CP De-layed diagnosis of juvenile myoclonic epilepsy J Neu-rol Neurosurg Psychiatry 199255497ndash499

5 Krumholz A Nonepileptic seizures diagnosis andmanagement Neurology 1999S76ndash83

6 Commission on Epidemiology and Prognosis Interna-tional League Against Epilepsy Guidelines for epidemio-logical studies on epilepsy Epilepsia 199334592ndash596

7 American College of Emergency Physicians (ACEP)Clinical Policies Committee and the Clinical PoliciesSubcommittee on Seizures Clinical Policy Critical is-sues in the evaluation of adult patients presenting to anemergency department with seizures Ann Emerg Med200443605ndash625

8 Fisher RS BoasWvE BlumeW et al Epileptic seizuresand epilepsy Definitions proposed by the InternationalLeague Against Epilepsy (ILAE) and the InternationalBureau for Epilepsy (IBE) Epilepsia 200546470ndash472

9 Agency for Healthcare Research and Quality (AHRQ)Management of newly diagnosed patients with epi-lepsy a systematic review of the literature (AHRQPublication No 01-E038) Rockville MD Agency forHealth Research Quality US Department of Healthand Human Services 2001

10 Hirtz D Ashwal S Berg A et al Practice parameterevaluating a first nonfebrile seizure in children Neu-rology 200055616ndash623

11 Jallon P Loiseau P Loiseau J Newly diagnosed unpro-voked epileptic seizures Presentation at diagnosis inCORALE study Epilepsia 200142464ndash475

12 Quality Standards Subcommittee and the Therapeuticsand Technology Assessment Subcommittee of theAmerican Academy of Neurology Clinical PracticeGuideline Process Manuel St Paul MN AmericanAcademy of Neurology 2004

13 Hauser WA Anderson VE Loewenson RB McRob-erts SM Seizure recurrence after a first unprovoked sei-zure N Engl J Med 1982307522ndash528

14 Hauser WA Rich SS Annegers JF Anderson VE Sei-zure recurrence after a 1st unprovoked seizure an ex-tended follow-up Neurology 1990401163ndash1170

15 van Donselaar CA Schimsheimer R Geerts AT et alValue of the electroencephalogram in adult patientswith untreated idiopathic first seizures Arch Neurol199249231ndash237

16 First Seizure Trial Group Randomized clinical trial onthe efficacy of antiepileptic drugs in reducing the risk ofrelapse after a first unprovoked tonic-clonic seizureNeurology 199343478ndash483

17 Bora I Seckin B Zarifoglu S Turin F Sadikogllu S OgulE Risk of recurrence after first unprovoked tonic-clonicseizure in adults J Neurol 1995242157ndash163

18 Hopkins A Garman A Clarke C The first seizure inadult life Value of clinical features electroencephalog-raphy and computerized tomographic scanning in pre-diction of seizure recurrence Lancet 1988721ndash726

19 Das CP Sawhney IMS Lal V et al Risk of recurrenceof seizures following a single unprovoked idiopathicseizure Neurol India 200046357ndash360

2006 Neurology 69 November 20 2007

RETIRED

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

RETIRED

20 Hui ACF Tang A Wong KS Recurrence after a firstuntreated seizure in the Hong Kong Chinese popula-tion Epilepsia 20014294ndash97

21 Schreiner A Pohlmann-Eden B Value of the early elec-troencephalogram after a first unprovoked seizureClin EEG 200334140ndash144

22 Forsgren L Fagerlund M Zetterlund B Electroen-cephalographic and neuroradiological findings inadults with newly diagnosed unprovoked seizures EurNeurol 19913161ndash67

23 Edmondstone WM How do we manage the first sei-zure in adults J R Coll Phys Lond 199529289ndash294

24 Neufeld M Chistik V Vishne TH Korczyn AD Thediagnostic aid of the routine EEG findings in patientspresenting with a presumed first-ever unprovoked sei-zure Epilepsy Res 200042197ndash202

25 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised classification of epilepsies and epileptic syn-dromes Epilepsia 198930389ndash399

26 Commission on Classification and Terminology of theInternational League Against Epilepsy Proposal for re-vised clinical and electroencephalographic classifica-tion of epileptic seizures Epilepsia 198122489ndash501

27 Schoenenberger RA Sabine MH Indication for computedtomography of the brain in patients with the first uncompli-cated generalized seizure BMJ 1994309986ndash989

28 Liu RSN Lemieux L Bell GS Sisodiya SM BartlettPA Shorvon SD Sander JWA Duncan JS The struc-tural consequences of newly diagnosed seizures AnnNeurol 200252573

29 Greenberg MK Barson WG Starkman S Practice pa-rameter Neuroimaging in the emergency patient pre-senting with a seizure Neurology 19964726ndash32

30 Henneman PL DeRoos F Lewis RJ Determining theneed for admission in patients with new-onset seizuresAnn Emerg Med 1994241108ndash1114

31 Tardy B Lafond P Convers P et al Adult first gener-alized seizure etiology biological tests EEG CT scanin an ED Am J Emerg Med 1995131ndash5

32 Turnbull TL Vanden Hoek TL Howes DS Eisner RFUtility of laboratory studies in the emergency depart-ment patient with new-onset seizure Ann Emerg Med199019373ndash377

33 Sempere AP Villaverde FJ Martinez-Menendez BCabeza C Pena P Tejerina First seizure in adults aprospective study from the emergency departmentActa Neurol Scand 199286134ndash138

34 Holtzman DM Kaku DA Yuen TS New-onset sei-zures associated with human immunodeficiency virusinfection causation and clinical features in 100 casesAm J Med 198987173ndash177

35 Olson KR Kearney TE Dyer JE Benowitz NK BlancPD Seizures associated with poisoning and drug over-dose Am J Emerg Med 199311565ndash568

36 Dhuna A Pascual-Leone A Langendorf F AndersonDC Epileptogenic properties of cocaine in humansNeurotoxicology 199112621ndash626

37 Gilbert DL Sethuraman G Kotagal U Buncher RMeta-analysis of EEG test performance shows widevariation among studies Neurology 200260564ndash570

Neurology 69 November 20 2007 2007

RETIRED

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology

DOI 10121201wnl000028508493652432007691996-2007 Neurology

A Krumholz S Wiebe G Gronseth et al the American Academy of Neurology and the American Epilepsy Society

ofevidence-based review) [RETIRED] Report of the Quality Standards Subcommittee Practice Parameter Evaluating an apparent unprovoked first seizure in adults (an

This information is current as of November 19 2007

ServicesUpdated Information amp

httpnneurologyorgcontent69211996fullincluding high resolution figures can be found at

Supplementary Material

httpnneurologyorgcontentsuppl2016101269211996DC3 httpnneurologyorgcontentsuppl2007111569211996DC1 httpnneurologyorgcontentsuppl2008111669211996DC2

Supplementary material can be found at

Citations httpnneurologyorgcontent69211996fullotherarticles

This article has been cited by 13 HighWire-hosted articles

Permissions amp Licensing

httpwwwneurologyorgaboutabout_the_journalpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnneurologyorgsubscribersadvertiseInformation about ordering reprints can be found online

Online ISSN 1526-632X1951 it is now a weekly with 48 issues per year Copyright All rights reserved Print ISSN 0028-3878

reg is the official journal of the American Academy of Neurology Published continuously sinceNeurology