Emergency Department Admission of Children
With Unprovoked Seizure: Recurrence
Within 24 Hours
Yoshimi Sogawa, MD and Joseph Maytal, MD
Little is known about acute recurrence of seizures,
and children with unprovoked seizure tend to be
admitted for observation on the basis of the risk
factors known for epilepsy. The purpose of this study
is to define the clinical profile of pediatric seizure
patients who are likely to be admitted and to analyze
the incidence and the risk factors of acute recur-
rence of seizures in the admitted patients within 24
hours after admission. A retrospective chart review
was performed on pediatric patients who arrived to
the Schneider Children’s Hospital Emergency De-
partment because of an unprovoked seizure who
were not on antiepileptic medication. Fifty-nine per-
cent of children who arrived to the emergency
department were admitted. Emergency department
physicians tend to admit children with younger age
(P < 0.01), multiple seizures (P < 0.001), and
children who were treated in the emergency depart-
ment (P < 0.001). Twenty percent of admitted
patients had one or more seizures within 24 hours.
Multiple seizures before the emergency department
arrival was a significant risk factor for acute recur-
rence (33%, P < 0.05). Based on our findings that
20% of admitted children had one or more seizures
within 24 hours, we think it is justified to admit and
observe the children with seizures who are not on
antiepileptic medications if the follow-up cannot be en-
sured. © 2006 by Elsevier Inc. All rights reserved.
Sogawa Y, Maytal J. Emergency department admission of
children with unprovoked seizure: recurrence within 24
hours. Pediatr Neurol 2006;35:98-101.
From Schneider Children’s Hospital, Division of Pediatric Neurology,
New Hyde Park, New York.
98 PEDIATRIC NEUROLOGY Vol. 35 No. 2
doi:10.1016/j.pediatrneurol.2006.01.007 ● 0887-8994/06/$—see front matter
Introduction
Seizure is one of the most anxiety-provoking conditions
for parents, and a common reason for emergency depart-
ment visits [1], especially if the seizure is of new onset or
the child is not on antiepileptic medication. Approximately
4% to 6% of children will have a seizure by 16 years of
age [2]. Although all physicians agree that children who
had a seizure should be evaluated, the criteria for admis-
sion in these children, who have often returned to their
baseline level of consciousness by the time they are
evaluated, are not clear. Many parents express the fear of
their children having another seizure at home after leaving
the emergency department. Studies [3,4] have described
the incidence of recurrent seizures over long periods of
time and clinical profiles of children who later develop
epilepsy. The common practice decision regarding admis-
sion of seizure patients is often based on the known risk
factors of epilepsy, such as symptomatic etiology and
history of previous seizure. Although it seems reasonable,
this approach is based on the assumption that the risk
factors of acute recurrence are similar to those of remote
recurrence. The purpose of this study is twofold: first, to
define the clinical profile of pediatric seizure patients who
are likely to be admitted for observation or treatment, and
second, to analyze the incidence, and the risk factors of
acute recurrence of seizures in the admitted children
within 24 hours after admission.
Materials and Methods
A retrospective, emergency department chart review of all pediatric
patients (age under 21 years) who arrived at the Schneider Children’s
Hospital Emergency Department because of an unprovoked seizure
during the year 2001 was conducted. The study excluded children who
were on antiepileptic medications. Inpatient charts were also reviewed
Communications should be addressed to:
Dr. Sogawa; Montefiore Medical Center; Epilepsy Center; 111 East
210th Street; Bronx, NY 10467.
E-mail: Ysogawa@aol.com
Received September 7, 2005; accepted January 30, 2006.
© 2006 by Elsevier Inc. All rights reserved.
 1014
ED patients with diagnosis of
seizure

117 897
Patients with unprovoked seizure who were Patients were on anti-epileptic medication
not on anti-epileptic medication or
Seizure was not the primary reason for ED visit
or
Patients had an acute provoked seizure

69 48
Admitted for observation or further Discharged from
treatment ED

Recurrent seizure within 24 Recurrent seizure within 24

hours hours
YES NO
14 (20%) 55 (80%)

Figure 1. Study population. ED ⫽ Emergency Department.

for those children who were admitted (Fig 1). Schneider Children’s
Hospital is a 154-bed children’s hospital with an annual emergency
department census of 25,000 patient visits. Patients were identified using
the International Classification of Diseases code, ninth edition (ICD-9) as
part of their emergency department or final hospital discharge diagnosis.
The study was exempted from the institutional human subjects review
board.
The clinical profile of these admitted children included diagnosis
(epilepsy, new-onset seizure), etiology (symptomatic, idiopathic), age,
duration (less than 5 minutes, longer than 5 minutes), multiple seizures
before emergency department arrival (more than one seizure in 24 hours),
and emergency department treatment. History of pervasive developmen-
tal disorder and global developmental delay were considered symptom-
atic etiology. The history of attention-deficit hyperactivity syndrome and
isolated speech delay were considered idiopathic etiology. In those
children who were admitted, the rate of acute seizure recurrence was
evaluated and correlated with potential risk factors such as diagnosis,
etiology, age, duration of seizure, multiple seizures before emergency
department, emergency department treatment, and electroencephalo-
graphic results.
A comparison of two groups (admitted children vs discharged chil-
dren, inpatients with acute recurrence vs inpatients without acute recur-
rence) was made using chi-square or an unpaired t test analysis.
Statistical significance was established at P ⬍ 0.05.
Results
Clinical Profile of Hospital Admission
A total (Table 1) of 117 children were analyzed (male:
female ⫽ 69:48, mean age 6 years and 8 months old).
Fifty-nine percent of children were admitted for observa-
tion. There is no significant difference regarding admis-
sion rate between the children with epilepsy and new-
onset seizure. There is no significant difference regarding
the admission rate between the children with symptomatic
etiology and idiopathic etiology. The mean age of admit-

Table 1. Clinical profile of hospital admission

Total Admission 69/117 Admission Rate (%) in

(59.0%) Each Group P Value

Diagnosis Epilepsy 40/62 (64.5%) ⱕ0.2

New onset seizure 29/55 (52.7%)
Etiology Symptomatic 28/42 (66.7%) ⱕ1
Idiopathic 41/75 (54.7%)
Age (mean) 3 years 8 months ⬍0.01
Duration Less than 5 minutes 54/91 (59.9%) ⱕ1
Longer than 5 minutes 14/22 (63.6%)
Not documented 1/4 (25.0%)
Multiple seizures Single 42/85 (49.4%) ⬍0.001
Multiple 27/32 (84.4%)
Emergency department treatment No treatment 48/95% (50.5%) ⬍0.001
Treatment 21/22 (95.5%)

Sogawa and Maytal: Unprovoked Seizure Recurrence 99

Table 2. Risk factors for acute seizure recurrence within 24 hours after admission
Overall Recurrence Rate
14/69 (20%)
Diagnosis Epilepsy
New onset seizure
Etiology Symptomatic
Idiopathic
Duration Less than 5 minutes
Longer than 5 minutes
Multiple seizures Single
Multiple
Emergency department treatment No treatment
Treatment
EEG Abnormal
Normal
Not done
ted children was 3 years and 8 months (2 months old to 18
years old), and it was 8 years old (9 months old to 20 years
old) for the discharged children. The mean age of admitted
children was significantly younger than discharged chil-
dren (P ⬍ 0.01). Most (77.8%) seizures lasted less than 5
minutes. There is no significant difference regarding
admission rate between the children with longer than 5
minutes seizure and children with less than 5 minutes
seizure. Nine of 10 children with seizures longer than 10
minutes were admitted. Children with multiple seizures
before arrival to the emergency department were more
likely to be admitted than those with single seizure (P ⬍
0.001). Children who received antiepileptic medication in
the emergency department were more likely to be admitted
(P ⬍ 0.001). Sixteen children received lorazepam, four
children received lorazepam and phenytoin, one child
received lorazepam and phenobarbital, and one child
received diazepam. Eighty-six of 117 children (73.5%)
had computed tomographic scan in the hospital, not
necessarily in the emergency department. Computed to-
mographic scan was obtained in 81.2% (56 of 67) of
admitted children, and the results were normal in 41 and
abnormal in 15. Computed tomographic scan was obtained
in 62.5% (30 of 48) of discharged children; the results
were normal in 27 and abnormal in 3. The results of
computed tomographic scans were not analyzed, as we
could not determine if the decision of admission was made
before or after the computed tomographic result.
Risk Factors for Acute Seizure Recurrence Within 24
Hours After Admission
Fourteen (Table 2) (20%) children had a seizure recur-
rence within 24 hours. Multiple seizures before the emer-
gency department arrival was the only significant risk
factor that correlated with recurrence of seizure within 24
hours. Electroencephalographic studies were obtained in
59 of 69 (85.5%) children, and the results were normal in
33 children and abnormal in 26 children. Epileptiform
activity was observed in 18 of 26 abnormal electroen-
100 PEDIATRIC NEUROLOGY Vol. 35 No. 2
Recurrence Rate (%)
With Each Risk Factor P Value
8/40 (20.0%) ⱕ1
6/29 (20.7%)
5/28 (17.9%) ⱕ1
9/41 (22.0%)
14/54 (25.9%) Not analyzed
0/15 (0%)
5/42 (11.9%) ⬍0.05
9/27 (33.3%)
9/48 (18.8%) ⱕ1
5/21 (23.8%)
7/26 (27.0%) ⱕ1
5/33 (15.2%)
1/9 (11.1%)
cephalograms. Neither abnormal electroencephalogram
nor epileptiform electroencephalogram were correlated
with recurrence of acute seizures. Treatment in the emer-
gency department did not lower the acute recurrence rate.
Discussion
Seizure is a common reason for emergency department
visit, and 4% of all emergency department visits to our
hospital were for children who had had a seizure, which is
similar to the previous report from the Boston Children’s
Hospital [1]. In the same study, Sharma et al. suggested
that emergent neuroimaging should be considered for the
children with new-onset seizure, if they have conditions
predisposing them to intracranial abnormalities, or if they
were ⬍33 months old with focal seizures. In other words,
well-appearing children with new-onset afebrile seizure
for whom these criteria do not apply can be safely
discharged from the emergency department without neu-
roimaging if follow-up can be assured. The Quality
Standards Subcommittee [5] recommended that an elec-
troencephalogram be performed as part of the evaluation
of a child with a first unprovoked seizure, although it does
not influence the decision regarding treatment after a first
seizure. It is becoming clear that admission in order to
obtain the neuroimaging or electroencephalogram is not
appropriate. The main reasons in our hospital to admit
children through the emergency department were (1) those
children who have acute medical problems or have not
returned to their baseline condition; (2) those children who
have higher risk of acute recurrence of seizures; and (3)
those children whose follow-up can not be ensured.
Little is known about risk factors for acute recurrence of
seizures, and children with seizures tend to be admitted for
observation on the basis of the risk factors known for
epilepsy. Shinnar et al. [3] have reported that the risk
factors for seizure recurrence included a remote symptom-
atic etiology, an abnormal electroencephalogram, a seizure
occurring while asleep, a history of prior febrile seizures,
and Todd’s paresis. Lizana et al. [4] also reported that a
symptomatic etiology increased the risk of recurrence,
whereas a patient age of 3 to 10 years decreased this risk.
Our data confirmed that the emergency department phy-
sicians tend to admit children with multiple seizures,
younger age, and children who received antiepileptic
medication in the emergency department. The data of the
present study also confirmed that remote symptomatic
etiology did not affect the admission rate. Based on our
findings that 20% of admitted children had one or more
seizures within 24 hours, we think it is necessary to follow
closely the children with unprovoked seizures who are not
on antiepileptic medication. If the follow-up cannot be
ensured, it may be justified to admit and observe those
children.
It is not clear why the emergency department treatment
did not lower the acute recurrence rate in this study. It may
be that in most cases antiepileptic medication was used as
one-time dose to stop an ongoing seizure, and not to
prevent acute recurrence.
The limitations of this study include its retrospective
design as well as lack of follow-up of 48 children who
were discharged from the emergency department. Clinical
data were limited to what was available in the emergency
department records, and some of the risk factors that
Shinnar et al. [3] mentioned, such as Todd’s paresis and
sleep status, were frequently not mentioned in the emer-
gency department records. This information as well as the
electroencephalographic results may only be available
after extensive evaluation, typically performed by the
child neurology service. It is possible that some of the
discharged children had acute recurrent seizures at home.
The external validity of our results may be limited by
virtue of the study setting, a tertiary care children’s
hospital emergency department. Compared with a commu-
nity hospital emergency department, the setting of this
study will result in an overrepresentation of complex or
severely ill patients. Annegers and Begley [6] have esti-
mated the direct cost for pediatric single seizures as over
$43 million, or approximately $2700 per person. These
include emergency services, inpatient hospitalizations,
outpatient visits, electroencephalography, computed to-
mography, and magnetic resonance imaging. Inpatient
cost accounted for approximately 18%, which is far larger
than the cost of emergency services of 4%. Prospective
studies to better identify the risk factors of acute recur-
rence are warranted to develop management strategies for
children who presented to the emergency department
because of seizure. Future study should include the fol-
low-up of discharged children and a focus to identify
low-risk children who may be safely managed as outpa-
tients.
References
[1] Sharma S, Riviello JJ, Harper MB, Baskin MN. The role of
emergency neuroimaging in children with new-onset afebrile seizures.
Pediatrics 2003;111:1-5.
[2] Chiang VW. Seizures. In: Fleisher GR, Ludwig S, eds. Textbook
of pediatric emergency medicine, 4th ed. Philadelphia: Lippincott,
Williams & Wilkins, 2000:573-9.
[3] Shinnar S, Berg AT, Moshe SL, et al. The risk of seizure
recurrence after a first unprovoked afebrile seizure in childhood: An
extended follow up. Pediatrics 1996;98:216-25.
[4] Lizana RJ, Garcia CE, Marina CLL, Lopez VM, Gonzales MM,
Hoyos MA. Seizure recurrence after a first unprovoked seizure in
childhood: A prospective study. Epilepsia 2000;41:1005-13.
[5] Practice parameter: evaluating a first nonfebrile seizure in
children. Report of the Quality Standards Subcommittee of the American
Academy of Neurology, the Child Neurology Society, and the American
Epilepsy Society. Neurology 2000;55:616-23.
[6] Annegers JF, Begley CE. Costs of pediatric epilepsy. In: Pellock
JM, Dodson WE, Bourgeois BF, eds. Pediatric epilepsy, 2nd ed. New
York: Demos, 2001:569-74.
Sogawa and Maytal: Unprovoked Seizure Recurrence 101