Download Risk Factors for Apnea in Pediatric Patients Transported by

PEDIATRICS/ORIGINAL RESEARCH
Risk Factors for Apnea in Pediatric Patients Transported
by Paramedics for Out-of-Hospital Seizure
Nichole Bosson, MD, MPH; Genevieve Santillanes, MD; Amy H. Kaji, MD, PhD; Andrea Fang, MD;
Tasha Fernando, BS; Margaret Huang, MD; Jumie Lee, CPNP, MSN; Marianne Gausche-Hill, MD
Study objective: Apnea is a known complication of pediatric seizures, but patient factors that predispose children are
unclear. We seek to quantify the risk of apnea attributable to midazolam and identify additional risk factors for apnea in
children transported by paramedics for out-of-hospital seizure.
Methods: This is a 2-year retrospective study of pediatric patients transported by paramedics to 2 tertiary care centers.
Patients were younger than 15 years and transported by paramedics to the pediatric emergency department (ED) for
seizure. Patients with trauma and those with another pediatric ED diagnosis were excluded. Investigators abstracted
charts for patient characteristics and predefined risk factors: developmental delay, treatment with antiepileptic
medications, and seizure on pediatric ED arrival. Primary outcome was apnea defined as bag-mask ventilation or
intubation for apnea by paramedics or by pediatric ED staff within 30 minutes of arrival.
Results: There were 1,584 patients who met inclusion criteria, with a median age of 2.3 years (Interquartile range 1.4
to 5.2 years). Paramedics treated 214 patients (13%) with midazolam. Seventy-one patients had apnea (4.5%): 44
patients were treated with midazolam and 27 patients were not treated with midazolam. After simultaneous evaluation
of midazolam administration, age, fever, developmental delay, antiepileptic medication use, and seizure on pediatric ED
arrival, 2 independent risk factors for apnea were identified: persistent seizure on arrival (odds ratio [OR]¼15; 95%
confidence interval [CI] 8 to 27) and administration of field midazolam (OR¼4; 95% CI 2 to 7).
Conclusion: We identified 2 risk factors for apnea in children transported for seizure: seizure on arrival to the pediatric
ED and out-of-hospital administration of midazolam. [Ann Emerg Med. 2014;63:302-308.]
Please see page 303 for the Editor’s Capsule Summary of this article.
A feedback survey is available with each research article published on the Web at www.annemergmed.com.
A podcast for this article is available at www.annemergmed.com.
0196-0644/$-see front matter
Copyright © 2013 by the American College of Emergency Physicians.
http://dx.doi.org/10.1016/j.annemergmed.2013.09.015
INTRODUCTION
Background
Seizure is the most common chief complaint for pediatric
patients in the out-of-hospital setting, accounting for
approximately 15% of all emergency medical services (EMS)
transports of pediatric patients in the United States.1 Given the
significant morbidity and mortality from prolonged seizures,
including risk of permanent neurologic sequelae, treatment is
recommended for any seizure lasting more than 10 minutes.1-3
The sooner the seizure is treated, the more likely it is to be
controlled.1 For this reason, initiating treatment for persistent
seizures in the out-of-hospital setting is common practice in EMS
systems in the United States. Benzodiazepines are first-line
therapy, and their out-of-hospital administration shortens overall
seizure duration.4,5 Traditionally, diazepam was most often used
to treat pediatric seizures. However, in the past decade,
midazolam has emerged as a favored therapy because of its ease of
administration by multiple routes and its rapid onset and
302 Annals of Emergency Medicine
clearance, making it less likely to result in adverse effects,
including cardiovascular depression and apnea.2,6-8 Los Angeles
County EMS Agency protocol was changed in July 2009,
designating midazolam (through the intravenous, intramuscular,
or intranasal route) as the treatment for pediatric seizure.
Importance
Respiratory depression is a potential complication of
treatment with benzodiazepines, especially in children.5,6,9-13
The high risk of respiratory depression in several studies has
caused some to question the use of benzodiazepines as first-line
therapy for seizure.12 Others have argued that the risk of
complications with benzodiazepines may be outweighed by the
benefits of reduced seizure duration, which itself could lead to
respiratory compromise.4 There is little knowledge of what other
factors may contribute to respiratory depression in children with
seizure. It has been suggested that children with developmental
delay, those receiving previous seizure medications, and those
with high fever may be more susceptible, but this is not well
Volume 63, no. 3 : March 2014
Bosson et al
Risk Factors for Apnea in Pediatric Patients
Editor’s Capsule Summary
What is already known on this topic
Seizing children can develop apnea.
What question this study addressed
When is such apnea more likely?
What this study adds to our knowledge
In this multicenter retrospective study of 1,584
children with out-of-hospital seizures, predictors of
the 71 who developed apnea were continuing seizure
on emergency department arrival and receiving
midazolam.
How this is relevant to clinical practice
Although apnea can result from both ongoing
seizures and midazolam, these data suggest that the
benefit of using midazolam to stop seizures in
children outweighs the risk of persistent seizures.
studied.14 Furthermore, prolonged seizure is likely a risk factor
for apnea.15 Children with persistent seizure activity are more
likely to receive medication, so the true contribution of
benzodiazepines in the onset of apnea remains unclear. It is
important to account for these potential risk factors when
evaluating the risk of apnea during paramedic treatment of outof-hospital seizure. By identifying patients at higher risk of apnea,
protocols may be tailored with consideration of these special
populations for additional monitoring.
Goals of This Investigation
We sought to quantify the risk of apnea in children presenting
with out-of-hospital seizure treated with midazolam by
paramedics and to identify other risk factors associated with
apnea in this population.
MATERIALS AND METHODS
Study Design and Setting
We conducted a multicenter retrospective chart review of
pediatric patients with seizure who were transported by
paramedics to the pediatric emergency department (ED) at
Harbor-UCLA Medical Center and Los Angeles County/
University of Southern California Medical Center during a 2-year
period, from January 2010 to December 2011. The study was
approved with waiver of informed consent by the institutional
review board at both institutions. Harbor-UCLA Medical Center
and Los Angeles County/University of Southern California
Medical Center are public teaching facilities serving Los Angeles
County, with an approximate population of 10 million. Both
hospitals are designated pediatric referral centers, as well as
pediatric trauma centers, with an annual pediatric ED patient
Volume 63, no. 3 : March 2014
volume of 21,000 and 25,000 at Harbor-UCLA Medical Center
and Los Angeles County/University of Southern California
Medical Center, respectively. In Los Angeles County, paramedic
crews consist of 2 paramedic-level emergency medical technicians
(paramedics) stationed at local fire departments. Protocols in Los
Angeles County support administration of midazolam by
paramedics for seizure termination in any patient actively seizing.
The dose may be repeated as needed with online medical control
guidance until the seizure terminates. Pediatric patients are
defined as aged 14 years or younger, and Los Angeles County
pediatric treatment protocols are specific to this age group.
Pediatric airway management in the field consists of bag-mask
ventilation for children younger than 12 years or less than 40 kg.
Intubation is available by protocol for children aged 12 years and
older or greater than 40 kg. The average time from dispatch to
hospital arrival for pediatric patients transported by paramedics in
Los Angeles County is approximately 20 minutes.16
Selection of Participants
Inclusion criteria were patients younger than 15 years and
transported by paramedics to the pediatric ED with the diagnosis
of seizure. Patients with an initial chief complaint of seizure, but
who were found to have another diagnosis on pediatric ED
evaluation or whose primary complaint was traumatic injury, were
excluded. At Harbor-UCLA Medical Center, patients were
identified by chief complaint from the base-hospital log of all
paramedic transports to the ED. Out-of-hospital care coordinators
record all patient transports in a county-sponsored database. At
Los Angeles County/University of Southern California Medical
Center, an electronic medical record of pediatric ED patients is
available. Patients with a chief complaint of seizure or a related
discharge diagnosis were identified from an electronic record of all
patients treated in the pediatric ED during the study period and
screened for those patients transported by paramedics for out-ofhospital seizure. The following diagnoses were included in the
search criteria: nonconvulsive epilepsy International Classification
of Diseases, Ninth Revision (ICD-9) 345.00; epilepsy, convulsion,
generalized ICD-9 345.1; epilepsy, petit mal status ICD-9 345.2;
epilepsy, grand mal status ICD-9 345.3; seizure, grand mal ICD-9
345.3; epilepsy, without impairment of consciousness ICD-9
345.5; epilepsy (unspecified) ICD-9 345.9; epileptic convulsion
(unspecified) ICD-9 345.9; epileptic seizure (unspecified) ICD-9
345.9; febrile convulsion ICD-9 780.31; febrile seizure ICD-9
780.31; seizure, febrile ICD-9 780.31; complex febrile convulsion
ICD-9 780.32; and other convulsions (seizure not otherwise
specified) ICD-9 780.39.17 An initial chief complaint of seizure
identified by out-of-hospital personnel and a pediatric ED
diagnosis of seizure were required to meet inclusion in the study;
therefore, the final patient selection at each institution was similar.
Methods of Measurement
Charts were reviewed by 2 investigators at each institution; at
Harbor-UCLA, a nurse practitioner in the division of pediatric
emergency medicine and an EMS fellow board-certified in
emergency medicine, and at Los Angeles County/University of
Annals of Emergency Medicine 303
Risk Factors for Apnea in Pediatric Patients
Southern California Medical Center an emergency medicine
resident and an attending physician board-certified in pediatric
emergency medicine. In addition, at Harbor-UCLA, a medical
student initiated the data collection instrument with basic
demographic information, which was reviewed and then
completed by the investigators above. Abstractors were trained by
the principal investigator to use a single data dictionary identical
at both sites that contained definitions for each variable and
coded response to be entered on the abstraction form.
The investigators were not blinded to the purpose of the
study. The charts were abstracted for study variables: paramedicadministered midazolam, route of administration, number of
doses of medication administered, and total dose of medication
administered. During the entire study period, paramedics could
administer midazolam intravenously, intranasally, or
intramuscularly at a dose of 0.1 mg/kg.
Paramedics used the Broselow tape to estimate patient weight
for medication dosing. In addition, investigators extracted
information on sex, age, weight, fever, history of developmental
delay, seizure on arrival to the pediatric ED, treatment with rectal
diazepam by a caregiver, and current treatment with an
antiepileptic drug. Weight was recorded as the measured value in
the ED. If the measured value was not available but an estimated
value was recorded, then this was used; otherwise, weight was
recorded as missing. Fever was considered present if a temperature
measured at any point during the patient’s stay in the pediatric ED
was documented at or above 38 C (100.4 F) or if a parent or
guardian reported a measured or tactile fever. It was considered
absent if no temperature exceeded 38 C (100.4 F) and there was
no documentation of fever reported at home. A patient was
considered to be receiving an antiepileptic drug if there was
documentation of currently prescribed medication. If
noncompliance was reported, the variable was coded as positive if
the last dose was within 1 week of presentation, considering
negligible amounts of drug after 5 half-lives. Developmental delay
was defined as neurologic delay; it was considered present if
documented on any pediatric ED visit or inpatient stay and absent
if not documented. Seizure on arrival to the pediatric ED was
based on initial triage and physician documentation and was used
as a proxy for prolonged seizure.
Age was coded as a continuous variable, whereas other
variables were coded as binary predictors. With a designated,
standardized collection instrument, data were tabulated and
entered into a Microsoft Excel spreadsheet (Microsoft, Redmond,
WA) and back-checked by the principal investigator for accuracy.
This was accomplished by evaluating for any inconsistency, using
the inherent redundancy of the data collection instrument, and
revisiting any chart reviews in which inconsistencies were
identified. The data were then imported into SAS for analysis
(version 9.3; SAS Institute, Inc., Cary, NC).
Outcome Measures
Outcome variables for airway management for apnea were
defined as bag-mask ventilation or intubation by paramedics or
by pediatric ED staff within 30 minutes of arrival. For patients in
304 Annals of Emergency Medicine
Bosson et al
whom airway intervention was required, the chart review
included the respiratory status of the patient before airway
management and the documented indications for intubation if
performed to determine whether apnea was present. Outcomes
were abstracted and then confirmed by agreement of the 2
investigators at that site, and any discrepancy was discussed and
resolved with a third (for Los Angeles County/University of
Southern California Medical Center patients, this was the
principal investigator at Harbor-UCLA Medical Center; for
Harbor-UCLA Medical Center patients, this was the senior
investigator at Harbor-UCLA Medical Center). There were 2
patients at Harbor-UCLA Medical Center and 5 at Los Angeles
County/University of Southern California Medical Center who
required such review. After review, they were not included in the
group with apnea. The discrepancies were due to the indication for
intubation in 6 cases and the timing of airway management 1 case.
Primary Data Analysis
The study outcomes (intubation or bag-mask ventilation) for
each group were calculated as proportions with exact binomial
confidence intervals (CIs). Adjusted odds ratios (ORs) were
determined with multiple logistic regression. All variables were
selected a priori according to previous knowledge of potential risk
factors for apnea and potential effect modifiers. They were
introduced simultaneously into the model to determine the
adjusted OR for apnea for each variable. Age was entered as a
continuous variable and all other variables were binary. Model fit
was determined by assessing the Hosmer-Lemeshow fit statistic,
and model predictors were retained according to previous
knowledge, as well as statistical significance. The interaction
between seizing on arrival and receiving a field medication was
considered, and effect measure modification of these 2 variables
was assessed by comparing ORs for apnea, given one variable across
categories of another. Finally, a propensity score–adjusted model
was performed to confirm the results of the multivariate regression
because of the large differences between the treatment groups.17
The propensity score–adjusted analysis also allowed adjustment by
additional potential confounding factors (weight and treatment
with rectal diazepam by a caregiver), given the limitations of the
multivariate model because of the rarity of the outcome.
Figure. Patient identification and inclusion.
Volume 63, no. 3 : March 2014
Bosson et al
Risk Factors for Apnea in Pediatric Patients
Table 1. Characteristics of the study population (n¼1,584).
Total Population
Characteristics
Age, y
Weight, kg
Sex
Male
Female
Fever
Treatment with an antiepileptic drug
Developmental delay
Seizure on pediatric ED arrival
Rectal diazepam administered
No. (Median)
2.3
14
875
709
1,074
374
300
153
60
Midazolam (n[214)
%, IQR
No. (Median)
%, IQR
No. (Median)
1.4–5.2
11–20
3.2
15
1.8–6.2
12–20
2.2
13
55
45
68
24
19
10
4
RESULTS
Characteristics of Study Subjects
As seen in the Figure, 2,403 patients were reviewed for study
inclusion, of whom 815 were excluded. The remaining 1,584
pediatric patients transported by paramedics during the study
period make up the study cohort, 770 patients transported to
Harbor-UCLA Medical Center and 813 patients transported to
Los Angeles County/University of Southern California Medical
Center. Table 1 shows the characteristics of the study population
as a whole and by treatment group. Of the 214 patients (14%)
treated with midazolam, paramedics administered the drug
intravenously to 87 (41%), intramuscularly to 53 (25%),
intranasally to 57 (27%), and by multiple routes to 17 (8%).
Paramedics dosed midazolam accurately; the median intravenous
dose was 0.09 mg/kg (95% CI 0.08 to 0.1 mg/kg) and this did
not differ across age groups. There was a 22% admission rate.
Discharge diagnoses were simple febrile seizure (50%), complex
febrile seizure (7%), epilepsy (27%), first-time seizure (8%), and
seizure type undefined (8%).
Main Results
There were 71 patients with apnea (4.5%): 44 patients (62%)
treated with midazolam and 27 (38%) not treated with midazolam.
Table 2 shows patient outcomes. For the 53 patients (3.6%)
who required intubation, the median time to intubation was 13
minutes (Interquartile range [IQR] 8 to 20) from pediatric ED
arrival. The majority of patients (70%) who required airway
intervention in the pediatric ED did not receive medications there
No airway intervention
Any apneic event
Apnea in the field or at pediatric ED arrival
Field BMV
Pediatric ED BMV
Pediatric ED intubation
BMV, Bag-mask ventilation.
Volume 63, no. 3 : March 2014
98
116
125
92
74
86
15
46
54
58
43
35
40
7
777
593
949
282
226
67
45
%, IQR
1.3–4.9
11–20
57
43
69
21
17
5
3
before airway management for apnea. Paramedics did not intubate
any patients in the field.
We evaluated the effect of the administration of midazolam in
the field, age, fever, developmental delay, treatment with an
antiepileptic drug, seizure on pediatric ED arrival, and the
interaction between midazolam administration and seizure on
pediatric ED arrival on the risk of apnea. The final model,
simultaneously evaluating the above covariates, demonstrated 2
independent risk factors for apnea: persistent seizure on arrival
(OR¼15; 95% CI 8 to 27) and administration of field
midazolam (OR¼4; 95% CI 2 to 7) (Table 3). The interaction
term was not a statistically significant predictor for apnea
(OR¼1.5; 95% CI 0.6 to 4.1), so it was not included in the final
model. The results of the model including the interaction term
have been provided in Table E1 (available online at http://www.
annemergmed.com). A propensity score–adjusted model
confirmed the association of field midazolam administration with
apnea (OR¼4.6; 95% CI 2.6 to 8.3). The Hosmer-Lemeshow
goodness-of-fit statistics for the propensity score–adjusted and
unadjusted models demonstrated a P value of .2 and .1,
respectively, indicating adequate fit for both models.
A stratified analysis demonstrated that among those who were
not seizing on arrival (n¼1,431), the OR for apnea if patients
were given midazolam was 16 (95% CI 7 to 39). Among patients
who were seizing on arrival (n¼153), the OR for apnea if
patients were given midazolam was 1.5 (95% CI 0.8 to 3.1).
Thus, there appears to be effect measure modification of
Table 3. Adjusted ORs for apnea.
Table 2. Frequencies for patient outcomes and airway
management (n¼1,584).
Outcome
No Midazolam (n[1,370)
Variable
No.
%
1,513
71
48
27
49
53
95.5
4.5
3.0
1.7
3.1
3.6
Seizure on pediatric ED arrival
Treatment with midazolam
Developmental delay
Age (per year younger)
Treatment with antiepileptic drug
Fever
Adjusted OR for Apnea (95% CI)*
15
4
1.9
1.1
1.0
0.8
8–27
2–7
0.8–4.3
1.0–1.3
0.4–2.2
0.4–1.5
*Adjusted OR generated by the simultaneous entry of covariates in the logistic
regression model. Hosmer-Lemeshow goodness-of-fit statistics P value 0.1, Akaike
information criterion¼406.8.
Annals of Emergency Medicine 305
Bosson et al
Risk Factors for Apnea in Pediatric Patients
persistent seizure on the association of midazolam with apnea.
Midazolam administration was no longer associated with apnea
among patients who were seizing on arrival, possibly because of
small sample size.
Of the 60 patients (4%) treated with rectal diazepam by
caregivers before EMS arrival, 8 had apnea, of whom half
received additional treatment with midazolam by paramedics.
We included treatment with rectal diazepam in the propensity
score–adjusted analysis reported above. In addition, neither
inclusion of all patients who received rectal diazepam by
caregivers in the treated group nor their exclusion from the
analysis affected the study results.
patients were noted to be apneic in the field or on pediatric ED
arrival. Thus, most patients who became apneic did not receive
medication in the pediatric ED before airway management, and
the decision to treat with benzodiazepines in the pediatric ED
was affected by the outcome in the field. We lacked the statistical
power to include parental administration of rectal diazepam in
our regression model. However, we included this treatment in
the propensity score–adjusted model, and we performed a
sensitivity analysis in which inclusion of these patients in the
treatment group yielded results similar to those we reported
above. There may also be other unknown confounders that were
not included in our model.
LIMITATIONS
This study has several limitations. Although it represents a
large number of children treated for seizure by multiple EMS
provider agencies and transported to 2 different hospitals, it is
limited to a single urban area of Los Angeles County and may not
be generalizable to all populations.
The retrospective nature of the study could lead to bias, given
that the investigators were not blinded to the outcome when
extracting the data and we did not calculate the interrater
reliability between reviewers to confirm reliability of study
variables. However, we attempted to mitigate this bias by
choosing concrete outcomes (rather than hypoventilation). In
addition, the definite outcome of ventilation by bag-mask or
intubation for apnea in the field or within 30 minutes of pediatric
ED arrival was selected to avoid misclassification of primary
respiratory illnesses such as asthma and pneumonia, which would
otherwise be a concerning source of bias. However, this could
have introduced a different source of bias and affected our
specificity if data for patients intubated for other reasons were
included as positive outcomes. We believe that these
misclassifications were unlikely to occur because of our rigorous
chart review performed by trained investigators at each institution.
Misclassification of some exposure measurements may also
have occurred. We considered seizure on arrival as a proxy for
prolonged seizure because timing of seizure as estimated by
witnesses is fraught with error. In addition, fever was considered
present if measured in the pediatric ED or if reported by parents;
although fever reported by parents has been shown to be
accurate, this may not hold true in our population of pediatric
seizure patients.18,19 Misclassification, if present, could lead to
residual confounding by these factors in the model. However, the
primary exposure, midazolam administration, as well as the other
covariates, was determined directly from record review and
should not be subject to measurement error.
The inclusion of airway interventions within 30 minutes of
pediatric ED arrival was intended to focus the outcome on events
related to out-of-hospital care, but other treatments in the
pediatric ED may have contributed to these outcomes. We
evaluated treatment with benzodiazepines in the pediatric ED in
our regression analysis and this did not alter our findings, likely
because the median time to definitive airway management in the
pediatric ED was short, at 13 minutes, and the majority of these
306 Annals of Emergency Medicine
DISCUSSION
We determined that the occurrence of apnea in children with
out-of-hospital seizure is multifactorial, and risk factors likely
include treatment with midazolam and prolonged seizure. Other
patient factors hypothesized to influence apnea risk, including
age, fever, treatment with antiepileptic medication, and
developmental delay, did not significantly contribute to outcomes
in our patient population. To our knowledge, our study is the
first to quantify the risk of apnea in children with out-of-hospital
seizure that may be attributed to midazolam administration after
accounting for other potential confounding risk factors.
The overall risk of apnea in our cohort was 4.5%. A high risk
of apnea has previously been reported with benzodiazepine
administration in the field for treatment of pediatric seizure, but
the contribution of patient factors was not evaluated.6,13,15,20
The risk of apnea appears to be lower with midazolam than with
other benzodiazepines traditionally used,6,7,21 but this is not
consistent across studies.15,22-25 After adjusting for patient
factors, Holsti et al22 found the proportion of respiratory
complications to be similar between patients treated with
midazolam and diazepam.
Although previous studies highlight the potential for apnea in
pediatric patients with out-of-hospital seizure, simply describing
the risk of apnea in patients treated with out-of-hospital
benzodiazepines does not allow the assessment of the effect of
other contributing factors. The risk of apnea caused by
benzodiazepines in relation to the benefit of early seizure
termination requires a large cohort of patients. Orr et al9
evaluated the association of diazepam with apnea in children
treated in the ED, adjusting for patient age, weight, duration of
seizure before medication, final diagnosis, and additional
therapies and found a significant association between diazepam
and apnea, with a reported OR of 49.4. Our study differs in its
out-of-hospital setting, the benzodiazepine evaluated, and the
inclusion of other potential confounders in our model.
The Prehospital Treatment of Status Epilepticus trial, a
randomized study of adult patients treated by paramedics for outof-hospital seizure, found equal proportions of respiratory
complications in patients treated with benzodiazepines and
patients who received placebo, suggesting that other factors
contributed to overall apnea risk.5 They demonstrated that
Volume 63, no. 3 : March 2014
Bosson et al
benzodiazepines were safe and effective for treatment of out-ofhospital seizure in adults. In the Rapid Anticonvulsant
Medication Prior to Arrival Trial study with adults and children
(down to 13 kg body weight), the risk of intubation within 30
minutes in the ED in the lorazepam and midazolam groups
treated out-of-hospital were similar, although for both
medications the rate of apnea was much higher than reported
here (14.3% and 14.6%, respectively).15 Other investigators
report high proportions of apneic events in children with out-ofhospital seizure, and the possibility of increased risk in younger
children has been suggested to explain this observation.6
Therefore, we thought it was important to evaluate the risk
specific to pediatric patients.
Although treatment with benzodiazepines is a known risk factor
for apnea, we found that the association of midazolam
administration with the occurrence of apnea is greatly attenuated
after accounting for other important patient factors. As has been
suggested by previous studies, we determined prolonged seizure to
be a significant risk factor for apnea, resulting in greater risk to the
patient than administration of benzodiazepines.5,8,15 As one might
anticipate, there was an association between prolonged seizure and
field treatment with benzodiazepines. Sicker patients with complex
seizures are less likely to spontaneously terminate seizure activity
and more likely to require benzodiazepines by paramedics. We
found this to be a predominant cause of the observed association
between benzodiazepine administration and apnea. In addition,
there appears to be interaction between benzodiazepines and
prolonged seizure activity in regard to the outcome of apnea.
Although seizing on pediatric ED arrival appeared to be the greatest
risk factor for apnea, midazolam administration was an
independent risk factor. This association was not statistically
significant among patients who were seizing on pediatric ED
arrival, likely because of the small number of patients in this group.
Age, fever, developmental delay, and current treatment with
an antiepileptic drug do not appear to be associated with apnea in
these children. Unlike in previous studies, intubation for other
reasons, such as for airway protection, was not included in our
predefined composite outcome for apnea. To specifically quantify
the risk of apnea associated with benzodiazepines, we intended to
focus our analysis on risk factors for apnea rather than risk factors
for airway intervention.
In light of our findings, it would seem that the most
important factor in reducing the risk of apnea in children with
out-of-hospital seizure is termination of seizure activity. Because
benzodiazepines help terminate seizure activity, their
administration by paramedics in the field is justified and may
reduce the overall incidence of apnea in children with seizure.
Although there remains a risk of apnea associated with
benzodiazepines, it is outweighed by the risk of apnea caused by
persistent seizure activity.
In summary, we identified 2 risk factors for apnea in children
transported for seizure: prolonged seizure as manifested by
persistent seizure on pediatric ED arrival and out-of-hospital
administration of midazolam. Although treatment with
benzodiazepines is associated with occurrence of apnea,
Volume 63, no. 3 : March 2014
Risk Factors for Apnea in Pediatric Patients
persistent seizure appears to be the strongest risk factor, and
early termination of out-of-hospital seizure in children is
warranted. Paramedics should be prepared to manage the airway of
any child presenting with out-of-hospital seizure who is seizing at
paramedic arrival or requires medication for seizure termination.
The authors acknowledge Paul Hsu, PhD for his contribution to
the design of this study.
Supervising editor: Steven M. Green, MD
Author affiliations: From the Department of Emergency Medicine,
Harbor-UCLA Medical Center, and the Los Angeles Biomedical
Research Institute at Harbor-UCLA, Torrance, CA (Bosson, Kaji,
Fang, Lee, Gausche-Hill); the David Geffen School of Medicine at
UCLA, Los Angeles, CA (Bosson, Kaji, Fernando, Gausche-Hill); the
Keck School of Medicine of the University of Southern California,
Department of Emergency Medicine, Los Angeles, CA (Santillanes,
Huang); and the School of Nursing, University of California, Los
Angeles, CA (Lee).
Author contributions: MG-H conceived the study and supervised
the conduct of the study and data collection. NB, GS, AHK, and
MG-H collaborated on the design. NB, GS, TF, MH, and JL collected
the data. NB, GS, and AF managed the data, including quality
control. AHK provided statistical advice on study design and NB
analyzed the data. NB drafted the article, and all authors
contributed substantially to its revision. NB takes responsibility for
the paper as a whole.
Funding and support: By Annals policy, all authors are required to
disclose any and all commercial, financial, and other relationships
in any way related to the subject of this article as per ICMJE conflict
of interest guidelines (see www.icmje.org). The authors have stated
that no such relationships exist.
Publication dates: Received for publication April 16, 2013.
Revisions received July 15, 2013, and August 22, 2013. Accepted
for publication September 12, 2013. Available online October 10,
2013.
Presented at the Society for Academic Emergency Medicine
Western regional meeting, March 2013, Long Beach, CA.
Address for correspondence: Nichole Bosson, MD, MPH,
E-mail [email protected].
REFERENCES
1. Treatment of convulsive status epilepticus. Recommendations of the
Epilepsy Foundation of America’s Working Group on Status
Epilepticus. JAMA. 1993;270:854-859.
2. Pellock JM. Use of midazolam for refractory status epilepticus in
pediatric patients. J Child Neurol. 1998;13:581-587.
3. Verity CM. Do seizures damage the brain? the epidemiological
evidence. Arch Dis Child. 1998;78:78-84.
4. Alldredge BK, Wall DB, Ferriero DM. Effect of prehospital treatment on
the outcome of status epilepticus in children. Pediatr Neurol.
1995;12:213-216.
5. Alldredge BK, Gelb AM, Isaacs SM, et al. A comparison of lorazepam,
diazepam, and placebo for the treatment of out-of-hospital status
epilepticus. N Engl J Med. 2001;345:631-637.
Annals of Emergency Medicine 307
Risk Factors for Apnea in Pediatric Patients
6. Rainbow J, Browne GJ, Lam LT. Controlling seizures in the prehospital
setting: diazepam or midazolam? J Paediatr Child Health.
2002;38:582-586.
7. Gathwala G, Goel M, Singh J, et al. Intravenous diazepam, midazolam
and lorazepam in acute seizure control. Indian J Pediatr.
2011;79:327-332.
8. Chamberlain JM, Altieri MA, Futterman C, et al. A prospective,
randomized study comparing intramuscular midazolam with
intravenous diazepam for the treatment of seizures in children. Pediatr
Emerg Care. 1997;13:92-94.
9. Orr RA, Dimand RJ, Venkataraman ST, et al. Diazepam and intubation
in emergency treatment of seizures in children. Ann Emerg Med.
1991;20:1009-1013.
10. Chiulli DA, Terndrup TE, Kanter RK. The influence of diazepam or
lorazepam on the frequency of endotracheal intubation in childhood
status epilepticus. J Emerg Med. 1991;9:13-17.
11. Dieckmann RA. Rectal diazepam for prehospital pediatric status
epilepticus. Ann Emerg Med. 1994;23:216-224.
12. Norris E, Marzouk O, Nunn A, et al. Respiratory depression in children
receiving diazepam for acute seizures: a prospective study. Dev Med
Child Neurol. 1999;41:340-343.
13. Galustyan SG, Walsh-Kelly CM, Szewczuga D, et al. The short-term
outcome of seizure management by prehospital personnel: a
comparison of two protocols. Pediatr Emerg Care. 2003;19:221-225.
14. Seigler RS. The administration of rectal diazepam for acute
management of seizures. J Emerg Med. 1990;8:155-159.
15. Silbergleit R, Durkalski V, Lowenstein D, et al. Intramuscular versus
intravenous therapy for prehospital status epilepticus. N Engl J Med.
2012;366:591-600.
308 Annals of Emergency Medicine
Bosson et al
16. Gausche M, Lewis RJ, Stratton SJ, et al. Effect of out-of-hospital
pediatric endotracheal intubation on survival and neurological
outcome: a controlled clinical trial. JAMA. 2000;283:783-790.
17. Newgard CD, Hedges JR, Arthur M, et al. Advanced statistics: the
propensity score—a method for estimating treatment effect in
observational research. Acad Emerg Med. 2004;11:953-961.
18. Klein M, DeWitt TG. Reliability of parent-measured axillary
temperatures. Clin Pediatr (Phila). 2010;49:271-273.
19. Singhi S, Sood V. Reliability of subjective assessment of fever by
mothers. Indian Pediatr. 1990;27:811-815.
20. Warden CR, Frederick C. Midazolam and diazepam for pediatric
seizures in the prehospital setting. Prehosp Emerg Care.
2006;10:463-467.
21. Vilke GM, Sharieff GQ, Marino A, et al. Midazolam for the treatment of
out-of-hospital pediatric seizures. Prehosp Emerg Care. 2002;6:
215-217.
22. Holsti M, Sill BL, Firth SD, et al. Prehospital intranasal midazolam for
the treatment of pediatric seizures. Pediatr Emerg Care.
2007;23:148-153.
23. McIntyre J, Robertson S, Norris E, et al. Safety and efficacy of buccal
midazolam versus rectal diazepam for emergency treatment of
seizures in children: a randomised controlled trial. Lancet.
2005;366:205-210.
24. Scott RC, Besag FM, Neville BG. Buccal midazolam and rectal
diazepam for treatment of prolonged seizures in childhood and
adolescence: a randomised trial. Lancet. 1999;353:623-626.
25. Lahat E, Goldman M, Barr J, et al. Comparison of intranasal
midazolam with intravenous diazepam for treating febrile seizures in
children: prospective randomised study. BMJ. 2000;321:83-86.
Volume 63, no. 3 : March 2014
Bosson et al
Risk Factors for Apnea in Pediatric Patients
Table E1. Adjusted ORs for apnea with interaction term.
Variable
Treatment with midazolam
Seizure on pediatric ED arrival
Developmental delay
Age (per year younger)
Treatment with antiepileptic drug
Fever
Interaction term
Midazolam and seizure on
pediatric ED arrival
Adjusted OR for Apnea (95% CI)*
17
50
1.6
1.1
1.0
0.8
7–40
21–118
0.7–3.5
1.0–1.3
0.4–2.2
0.4–1.5
1.5
0.6–4.1
*Adjusted OR generated by the simultaneous entry of covariates in the logistic
regression model. Hosmer-Lemeshow goodness-of-fit statistics P value 0.2, Akaike
information criterion¼392.6.
Volume 63, no. 3 : March 2014
Annals of Emergency Medicine 308.e1