Вы находитесь на странице: 1из 6

Evaluation of a First Seizure

STEPHEN M. ADAMS, MD, and PAUL D. KNOWLES, MD, University of Tennessee College of Medicine,
Chattanooga Unit, Chattanooga, Tennessee
Seizure is a common presentation in the emergency care setting, and new-onset epilepsy is the most common cause
of unprovoked seizures. The patient history and physical examination should direct the type and timing of labora-
tory and imaging studies. No single sign, symptom, or test clearly differentiates a seizure from a nonseizure event
(e.g., syncope, pseudoseizure). Electroencephalography is recommended for patients presenting with a first seizure, and
neuroimaging is recommended for adults. Neuroimaging also should
be performed in children with risk factors such as head trauma, focal
neurologic deficits, or a history of malignancy. Magnetic resonance
imaging is preferred over computed tomography except when acute
intracranial bleeding is suspected. The most common laboratory
findings associated with a seizure are abnormal sodium and glucose
levels. Patients with a normal neurologic examination, normal test
results, and no structural brain disease do not require hospitalization
or antiepileptic medications. Treatment with antiepileptic medica-

ILLUSTRATION BY mark lefkowitz

tions reduces the one- to two-year risk of recurrent seizures but
does not reduce the long-term risk of recurrence and does not affect
remission rates. Regardless of etiology, a seizure diagnosis severely
limits a patient’s driving privileges, although laws vary by state. (Am
Fam Physician 2007;75:1342-1347, 1348. Copyright © 2007 American
Academy of Family Physicians.)

Patient Information:

bout 2 to 5 percent of Americans a recognizable cause (e.g., head injury, brain

A handout on seizures,
written by the authors of
experience an afebrile seizure, tumor), and idiopathic seizures are those
this article, is provided on and seizures account for approxi- for which no abnormality is found. Acute
page 1348. mately 1 to 2 percent of all emer- symptomatic seizures are caused by a recent
gency department visits.1 The self-reported or current event, whereas remote symp-
prevalence of epilepsy is 1.1 to 2.2 percent in tomatic seizures are caused by a chronic
the United States.2 Most patients (57 percent) abnormality such as an old stroke. A pro-
who present with a first seizure are younger voked seizure is caused by an identifiable
than 25 years, and 71 percent of this subset transient disturbance such as an electrolyte
is 15 years or younger; 58 percent of patients abnormality (e.g., hypocalcemia). In the
with a first seizure are men.3 year following an acute symptomatic sei-
zure diagnosis, patients have a higher risk
Seizure Types of death than those without this diagnosis.
Seizures are categorized based on presen- Idiopathic seizures are not associated with
tation and etiology. A generalized seizure increased risk of death.4
involves all areas of the brain (both hemi-
spheres), whereas a partial (focal) seizure Etiology
involves only one area of the brain. A first There are multiple causes of seizure
seizure is twice as likely to be a generalized (Table 15), but new-onset epilepsy is the most
seizure as a partial seizure. Most generalized common cause of a first seizure.3 One in six
seizures occur when the patient is awake, patients who present with a single seizure will
but one in four occurs during sleep.3 Partial have an identifiable potential cause such as
seizures can be further classified as simple pre- or perinatal brain injury (4.4 percent),
(i.e., no loss of consciousness) or complex cerebrovascular disease (3.9 percent), head
(i.e., loss of consciousness). injury (3.2 percent), brain tumor (1.7 per-
Symptomatic seizures are those that have cent), or alcohol use (0.3 percent).3

Downloaded from the American Family Physician 
Web site at www.aafp.org/afp. Copyright © 2007 American Academy of Family Physicians. For the private, noncommercial
use of one individual user of the Web site. All other rights reserved. Contact copyrights@aafp.org for copyright questions and/or permission requests.

Clinical recommendation rating References

Neuroimaging is recommended for adults presenting with a first C 1, 7, 17, 20

unprovoked seizure and for children with risk factors.
Immediate neuroimaging is recommended for patients who have an C 1, 7
increased risk of acute intracranial pathology.
Routine laboratory tests in adults should include glucose and C 1
sodium measurements and a pregnancy test if applicable;
immunocompromised patients should receive a lumbar puncture.
Electroencephalography is recommended for patients presenting with C 1, 7, 21
a first seizure.
Magnetic resonance imaging is preferred over computed tomography C 7, 16
unless acute intracranial bleeding is suspected.
Patients with a normal neurologic examination and no known C 1, 7
structural brain disease or comorbidities do not require
hospitalization or antiepileptic medication.

A = consistent, good-quality patient-oriented evidence; B = inconsistent or limited-quality patient-oriented evi-

dence; C = consensus, disease-oriented evidence, usual practice, expert opinion, or case series. For information
about the SORT evidence rating system, see page 1289 or http://www.aafp.org/afpsort.xml.

Table 1. Selected Causes of Seizure

Alcohol use and withdrawal Medications

Brain injury or abnormality Analgesics (e.g., local anesthetics,
Brain tumor meperidine [Demerol], tramadol [Ultram])
Cerebral dysgenesis Antibiotics (e.g., beta-lactam antibiotics,
Cerebrovascular, degenerative, isoniazid [INH; Nydrazid], quinolones,
inflammatory/autoimmune diseases some human immunodeficiency virus
Genetic disorders
Immunomodulators (e.g., cyclosporine
Head trauma
[Sandimmune], interferons, tacrolimus
Peri- or prenatal brain injury [Prograf])
Central nervous system infections Psychotropics (e.g., antidepressants,
Cysticercosis antipsychotics, lithium, stimulants)
Encephalitis Theophylline
Meningitis Metabolic disorders
Contrast agents Electrolyte abnormalities
Epilepsy Inborn errors of metabolism
Fever (children) Liver or kidney failure
Illicit drug use and withdrawal Pyridoxine deficiency

Information from reference 5.

Evaluation The history should initially focus on

When evaluating a patient who has just experi- determining whether a seizure actually
enced a seizure, the physician should first ver- occurred and evaluating the circumstances
ify that the patient has normal vital signs and and characteristics of the event. The behav-
adequate oxygenation and that there is no fur- iors of the patient during the event and
ther seizure activity. There is no standardized evidence of partial onset may be important
algorithm for the evaluation of every patient in identifying a specific form of epilepsy. A
with a first seizure. Instead, a careful history history of trauma or symptoms of infection
and physical examination should determine (e.g., stiff neck, fever, headache) also helps
imaging and laboratory testing decisions.6 direct the evaluation. The patient should

May 1, 2007 ◆ Volume 75, Number 9 www.aafp.org/afp American Family Physician  1343
First Seizure

seizure-like movements while uncon-

Table 2. Risk Factors for Acute scious.10 Historic features suggestive of sei-
Intracranial Pathology in Adults zure include tongue biting, presence of an
with a First Unprovoked Seizure aura, sensation of epigastric fullness, post-
ictal confusion, and focal neurologic signs.
Acquired immunodeficiency syndrome Tongue biting, especially lateral, is highly
Acute head trauma specific but not sensitive for generalized
Age older than 40 years seizures.11 Events precipitated by an emo-
Fever tionally stressful event or preceded by light-
History of anticoagulation headedness, sweating, prolonged standing,
History of malignancy chest pain, palpitations, or slow heart rate
New focal neurologic deficit are more likely to be syncopal.12
Partial (focal) seizure
Persistent altered mental status Diagnostic Testing
Persistent headache Adults who present to the emergency depart-
ment after an unprovoked first seizure should
note: Immediate neuroimaging is important if there is
an increased risk of acute intracranial pathology. receive immediate neuroimaging of the brain
Information from reference 13.
if feasible, although testing at a later date may
be acceptable in certain patients if follow-up
is reliable.1 Patients at an increased risk of
be asked about medication, illicit drug, acute intracranial pathology (Table 213) need
and alcohol use. A history of neurologic or immediate neuroimaging.1,7 Current prac-
developmental disorders or a family history tice guidelines allow a well-appearing child
of epilepsy may help narrow the differential without risk factors to be discharged from
diagnosis. The physical examination should the emergency department without emergent
include a thorough neurologic and mental neuroimaging.7
status evaluation. A lumbar puncture is indicated for patients
with a history or examination results sug-
Differential Diagnosis gestive of central nervous system infection
No sign, symptom, or test clearly differ- and in patients who are immunocompro-
entiates a seizure from a nonseizure event mised. New-onset seizures may be the only
(e.g., syncope, pseudoseizure).7 Up to symptom of central nervous system infec-
20 per­­cent of patients diagnosed with epi- tion in patients with human immunode-
lepsy actually have pseudoseizures. Eye clo- ficiency virus.14 Lumbar puncture is not
sure throughout the event is rare in true routinely recommended in afebrile chil-
seizures but common in pseudoseizures, and dren but should be considered in children
a history of fibromyalgia or chronic pain syn- younger than six months and in those who
drome is predictive of pseudoseizures.8 have persistently altered mental status or
In older children and adults, a serum meningeal signs.7
prolactin measurement, if obtained within Glucose abnormalities and hyponatremia
10 to 20 minutes of the event, is useful are the most common laboratory findings
in differentiating a generalized tonic-clonic associated with seizures.1 Practice guide-
seizure or complex partial seizure from a lines recommend testing children based on
pseudoseizure. The sensitivity of an elevated individual clinical circumstances and rou-
prolactin level is 60 percent for generalized tinely measuring serum glucose and sodium
tonic-clonic seizures and 46 percent for levels in adults. Pregnancy testing should
complex partial seizures.9 be performed in premenopausal women,
Syncope may be difficult to differentiate and toxicology testing should be performed
from seizures, particularly if the event was when substance abuse is suspected.1,7 Infants
unwitnessed. Up to 90 percent of patients with new-onset seizures should be tested for
with syncope have myoclonic or other electrolyte abnormalities.

1344  American Family Physician www.aafp.org/afp Volume 75, Number 9 ◆ May 1, 2007
First Seizure

Electroencephalography (EEG) is recom- of seizure with or without secondary gener-

mended for all patients with new-onset sei- alization, or no evidence of benign partial
zures.1,7 Emergent EEG testing is indicated epilepsy of childhood or primary generalized
if there is concern about status epilepticus. epilepsy on EEG.7 Seizures only provoked by
Nonconvulsive or subtle convulsive status the presence of fever (febrile seizures) do not
epilepticus may be difficult to diagnose require neuroimaging.16
clinically and may be mistaken for a pro-
longed postictal state. One fourth of patients
with treated status epilepticus who appear Neuroimaging scans reveal abnormalities in
to be seizure-free continue to have seizure 3 to 38 percent of patients with a first seizure,
activity that is only detectable with EEG. depending on patient demographics. A joint
Patients who have had a seizure and are in consensus statement from the American Col-
a drug-induced coma or who have received lege of Emergency Physicians (ACEP), the
a long-acting paralytic agent also should American Academy of Neurology (AAN), and
receive immediate EEG testing.1 others states that immediate neuroimaging is
Most other patients with a first seizure can indicated when a serious structural brain lesion
receive EEG testing at a scheduled follow-up is suspected and also should be considered for
visit. Although EEG within 24 to 48 hours patients with partial-onset seizures and for
of a seizure is more likely to show an abnor- those who are older than 40 years. Neuroimag-
mality, some early abnormalities, such as ing at a later date is acceptable for patients who
postictal slowing, may not be significant.7 have completely recovered from their seizures
and when there is no clear etiology, although
Neuroimaging immediate imaging should be performed if
The recommendations for imaging after a follow-up cannot be guaranteed.17
first seizure depend on patient age, seizure MRI is the preferred imaging method
type, and associated risk factors. because it has greater sensitivity for detect-
ing abnormalities than CT.7,16 However,
children patients with acute seizures initially should
There is insufficient evidence to recom- undergo CT because it more accurately
mend for or against routine neuroimaging detects acute bleeding and is reasonably sen-
in children who present only with a single sitive in detecting other abnormalities.7,16
unprovoked seizure, although unnecessary Intracranial disease is commonly detected
radiation exposure should be avoided. The on neuroimaging scans even if a nonbrain
estimated lifetime risk of death from radia- etiology for the seizure is apparent. A study
tion-induced malignancy caused by a single of unselected patients presenting to an urban
computed tomography (CT) scan of the head emergency department after a first seizure
at one year of age is 0.07 percent.15 related to alcohol withdrawal showed brain
Neuroimaging should be performed in abnormalities that required intervention or
children with a postictal focal neurologic change in therapy in 10 out of 259 patients
deficit that does not resolve or in children (3.9 percent). History of trauma and neuro-
who do not return to baseline neurologic logic examination findings did not predict
function within several hours. Neuroimag- these abnormalities.13
ing also should be performed in children
with head trauma or a history of malignancy. Treatment
Nonurgent magnetic resonance imaging When predicting future risk, all seizures
(MRI) should be seriously considered in chil- occurring within 24 hours are considered
dren younger than one year and in children a single seizure. Approximately one half of
with any of the following characteristics: patients who have a first unprovoked seizure
significant cognitive or motor impairment of and three out of four who have multiple
unknown etiology, unexplained abnormali- seizures will have another seizure within the
ties on neurologic examination, partial onset next eight years.18 Risk factors for seizure

May 1, 2007 ◆ Volume 75, Number 9 www.aafp.org/afp American Family Physician  1345
First Seizure

recurrence in children presenting with a Usually, patients with a history or cur-

first unprovoked seizure include abnormal rent diagnosis of epilepsy may not hold a
EEG test results; a history of febrile seizures; commercial driver’s license. If a seizure is
remote, symptomatic etiology; seizure that caused by a medical condition (e.g., drug
occurs during sleep; and Todd’s paralysis.18 reaction, high fever, acute infectious disease,
The decision to initiate treat- dehydration, acute metabolic disturbance),
ment after a single unprovoked the patient may be allowed to hold a com-
Most states require a seizure is controversial. A ran- mercial driver’s license after full recovery
three- to 18-month seizure-  domized controlled trial showed from the condition.24 Drivers with a his-
free period before a patient that antiepileptic medications tory of epilepsy or seizures who have not
may resume driving a pri- initiated at the occurrence of a needed antiseizure medication and have
vate vehicle. first seizure reduced the inci- been seizure-free for 10 years may reapply
dence of additional seizures to operate a commercial vehicle in interstate
over the next one to two years commerce.24 Drivers with a history of a
but did not reduce the long-term recurrence single unprovoked seizure may be qualified
risk or affect remission rates.19 to drive a commercial vehicle if they have
An AAN practice guideline states that not needed antiseizure medication and have
treatment with antiepileptic medications is been seizure-free for at least five years.24
not indicated in children for the purpose
of preventing epilepsy, but treatment may The Authors
be considered if the benefits of reducing the
STEPHEN M. ADAMS, MD, is an associate professor in
risk of a second seizure outweigh the risks the Department of Family Medicine at the University of
of pharmacologic and psychosocial adverse Tennessee College of Medicine, Chattanooga Unit. Dr.
effects.20 There is no guideline for adults. Adams received his medical degree from the University of
ACEP policy states that patients evaluated in Tennessee College of Medicine, Memphis, and completed
a family medicine residency at the University of Alabama,
the emergency department for a seizure who Huntsville.
have a normal neurologic examination, no
PAUL D. KNOWLES, MD, is an assistant professor in the
known comorbidities, and no known struc- Department of Pediatrics at the University of Tennessee
tural brain abnormalities may be referred for College of Medicine, Chattanooga Unit, and is director of
outpatient follow-up without initiating an pediatric neurology at T.C. Thompson Children’s Hospital,
antiepileptic medication.1 Consultation with Chattanooga. Dr. Knowles received his medical degree
from Eastern Virginia Medical School, Norfolk, and com-
a subspecialist is indicated if uncertainty pleted a residency at Children’s Hospital, Akron, Ohio.
remains after evaluation.21 If an antiepileptic
Address correspondence to Stephen M. Adams, MD,
medication is initiated, acceptable choices University of Tennessee Health Science Center, 1100
include carbamazepine (Tegretol), phenytoin E. Third St., Chattanooga, TN 37403 (e-mail: stephen.
(Dilantin), valproic acid/divalproex (Depak- adams@erlanger.org). Reprints are not available from
ene), phenobarbital, gabapentin (Neurontin), the authors.
lamotrigine (Lamictal), oxcarbazepine (Tri- Author disclosure: Dr. Knowles has served on the speakers’
leptal), and topiramate (Topamax).22 bureaus for Novartis (Trileptal), UCB Pharma (Keppra), and
Abbott (Depakote). He has been approached by Valeant
(Diastat) but has not yet signed an agreement with them.
Driving Limitations
Most states require a three- to 18-month
seizure-free period before a patient may REFERENCES
resume driving a private vehicle, although 1. American College of Emergency Physicians Clinical
most states have an appeals process for driv- Policies Committee, Clinical Policies Subcommittee on
Seizures. Clinical policy: critical issues in the evaluation
ing restrictions.23 Some states require physi- and management of adult patients presenting to the
cians to report patients who are diagnosed emergency department with seizures. Ann Emerg Med
with epilepsy.23 State driving laws for patients 2004;43:605-25.
2. Centers for Disease Control and Prevention. Prevalence
with epilepsy are available at http://www.
of epilepsy and health-related quality of life and disability
epilepsyfoundation.org; click the driving link among adults with epilepsy—South Carolina, 2003 and
under Quick Links. 2004. MMWR Morb Mortal Wkly Rep 2005;54:1080-2.

1346  American Family Physician www.aafp.org/afp Volume 75, Number 9 ◆ May 1, 2007
First Seizure

3. Jallon P, Loiseau P, Loiseau J. Newly diagnosed unpro- of radiation-induced fatal cancer from pediatric CT. AJR
voked epileptic seizures: presentation at diagnosis Am J Roentgenol 2001;176:289-96.
in CAROLE study. Coordination Active du Reseau 16. Bernal B, Altman NR. Evidence-based medicine:
Observatoire Longitudinal de l’Epilepsie. Epilepsia neuroimaging of seizures. Neuroimaging Clin N Am
2001;42:464-75. 2003;13:211-24.
4. Beghi E, Leone M, Solari A. Mortality in patients with 17. American College of Emergency Physicians, Ameri-
a first unprovoked seizure. Epilepsia 2005;46(suppl can Academy of Neurology, American Association of
11):40-2. Neurological Surgeons, American Society of Neuro-
5. Lowenstein DH. Diseases of the central nervous system. radiology. Practice parameter: neuroimaging in the
In: Braunwald E, Fauci AS, Kasper DL, Hauser SL, Longo emergency patient presenting with seizure (summary
DL, Jameson JL, eds. Harrison’s Principles of Internal statement). Ann Emerg Med 1996;28:114-8.
Medicine. 15th ed. New York, N.Y.: McGraw-Hill, 18. Shinnar S, Berg AT, Moshe SL, O’Dell C, Alemany M,
2001:2354-69. Newstein D, et al. The risk of seizure recurrence after
6. Freeman JM. Less testing is needed in the emer- a first unprovoked afebrile seizure in childhood: an
gency room after a first afebrile seizure. Pediatrics extended follow-up. Pediatrics 1996;98(2 pt 1):216-25.
2003;111:194-6. 19. Marson A, Jacoby A, Johnson A, Kim L, Gamble C,
7. Hirtz D, Ashwal S, Berg A, Bettis D, Camfield C, Cam- Chadwick D, for the Medical Research Council MESS
field P, et al. Practice parameter: evaluating a first Study Group. Immediate versus deferred antiepileptic
nonfebrile seizure in children: report of the quality drug treatment for early epilepsy and single seizures: a
standards subcommittee of the American Academy of randomised controlled trial. Lancet 2005;365:2007-13.
Neurology, the Child Neurology Society, and the Ameri- 20. Hirtz D, Berg A, Bettis D, Camfield C, Camfield P, Crum-
can Epilepsy Society. Neurology 2000;55:616-23. rine P, et al. Practice parameter: treatment of the child
8. LaFrance WC Jr, Benbadis SR. Avoiding the costs of with a first unprovoked seizure: report of the Quality
unrecognized psychological nonepileptic seizures. Neu- Standards Subcommittee of the American Academy
rology 2006;66:1620-1. of Neurology and the Practice Committee of the Child
9. Chen DK, So YT, Fisher RS. Use of serum prolactin Neurology Society. Neurology 2003;60:166-75.
in diagnosing epileptic seizures: report of the Thera- 21. Ross SD. Management of newly diagnosed patients
peutics and Technology Assessment Subcommittee with epilepsy: a systematic review of the literature.
of the American Academy of Neurology. Neurology Rockville, Md.: Agency for Healthcare Research and
2005;65:668-75. Quality, 2001. AHRQ Publication No. 01-E038.
10. Lempert T, Bauer M, Schmidt D. Syncope: a videometric 22. French JA, Kanner AM, Bautista J, Abou-Khalil B,
analysis of 56 episodes of transient cerebral hypoxia. Browne T, Harden CL, et al. Efficacy and tolerability
Ann Neurol 1994;36:233-7. of the new antiepileptic drugs I: treatment of new
onset epilepsy: report of the Therapeutics and Technol-
11. Benbadis SR, Wolgamuth BR, Goren H, Brener S,
ogy Assessment Subcommittee and Quality Standards
Fouad-Tarazi F. Value of tongue biting in the diagnosis
Subcommittee of the American Academy of Neurol-
of seizures. Arch Intern Med 1995;155:2346-9.
ogy and the American Epilepsy Society. Neurology
12. McKeon A, Vaughan C, Delanty N. Seizure versus syn- 2004;62:1252-60.
cope [Published correction appears in Lancet Neurol
23. Epilepsy Foundation. Driver information by state. Accessed
2006;5:293]. Lancet Neurol 2006;5:171-80.
January 19, 2007, at: http://www.epilepsyfoundation.
13. Earnest MP, Feldman H, Marx JA, Harris JA, Biletch M, org/advocacy/transportation/driverlicensing.cfm.
Sullivan LP. Intracranial lesions shown by CT scans in
24. Federal Motor Carrier Safety Administration. Part 391:
259 cases of first alcohol-related seizures. Neurology
qualifications of drivers and longer combination vehicle
(LCV) driver instructors. Subpart E—physical qualifi-
14. Pesola GR, Westfal RE. New-onset generalized seizures cations and examinations. Physical qualifications for
in patients with AIDS presenting to an emergency drivers. Accessed January 19, 2007, at: http://www.
department. Acad Emerg Med 1998;5:905-11. fmcsa.dot.gov/rules-regulations/administration/fmcsr/
15. Brenner D, Elliston C, Hall E, Berdon W. Estimated risks fmcsrruletext.asp?section=391.41.

May 1, 2007 ◆ Volume 75, Number 9 www.aafp.org/afp American Family Physician  1347