Original Article

The Use of High-Dose Epinephrine for Patients With

Out-of-Hospital Cardiopulmonary Arrest Refractory to
Prehospital Interventions
Mary D. Patterson, MD,* Douglas A. Boenning, MD,y Bruce L. Klein, MD,y Susan Fuchs, MD,z
Kathleen M. Smith, MD,x Mary A. Hegenbarth, MD,k Douglas W. Carlson, MD,{
Steven E. Krug, MD,# and Elliott M. Harris, MDz

1.66 [0.833.31]). Sixteen (17%) of 95 HDE patients and 5 (8%)

Objective: To determine if high-dose epinephrine (HDE) used
of 59 SDE patients survived at least 24 hours (P = 0.14, x2 = 2.16,
during out-of-hospital cardiopulmonary arrest refractory to prehos-
relative risk = 1.99 [0.775.14]). Nine survivors to discharge re-
pital interventions improves return of spontaneous circulation, 24-
ceived HDE, and 2 received SDE (P = 0.21, Fisher exact test, rela-
hour survival, discharge survival, and neurological outcomes.
tive risk = 2.75 [0.6112.28]). There were no long-term survivors
Methods: A multicenter randomized controlled trial was conducted
among the trauma patients. Eight of 11 long-term survivors had
between May 1991 and October 1996 to compare the effectiveness of
severe neurological outcomes defined by the Glasgow Outcome
HDE versus standard-dose epinephrine (SDE) in patients having out-
Scale (2/2 SDE, 6/9 HDE; P = 0.51, Fisher exact test).
of-hospital cardiopulmonary arrest refractory to prehospital resusci-
Conclusion: HDE does not improve or diminish return of
tation efforts. Cardiopulmonary arrest was classified as medical
spontaneous circulation, 24-hour survival, long-term survival, or
or traumatic. Two hundred thirty patients were enrolled in 7
neurological outcome compared with SDE in out-of-hospital cardio-
pediatric emergency departments. Ages ranged from newborn to 22
pulmonary arrest.
years. Seventeen patients met exclusion criteria. Patients were as-
signed to receive HDE (0.1 mg/kg for the initial dose and 0.2 mg/kg Key Words: pediatric cardiopulmonary arrest, epinephrine,
for subsequent doses) or SDE (0.01 mg/kg). The main end points resuscitation
evaluated were return of spontaneous circulation, 24-hour survival,
discharge survival, and neurological outcome.
Results: One hundred twenty-seven patients received HDE (32
trauma patients), and 86 patients received SDE (27 trauma patients).
Among medical patients, 24 (25%) of 95 experienced return of
T he outcome of pediatric cardiopulmonary arrest and re-
suscitation is poor. The long-term survival of pediatric
patients requiring cardiopulmonary resuscitation (CPR) is
spontaneous circulation in the HDE group as compared with 9
0% to 17%.115 Furthermore, the neurological outcome of
(15%) of 59 in the SDE group (P = 0.14, x2 = 2.17, relative risk =
these survivors is dismal. In studies by Schindler et al10 and
Hickey et al,16 there were no neurologically intact pediatric
*Division of Emergency Medicine, Childrens Hospital Medical Center survivors when return of spontaneous circulation (ROSC) did
ML 2008, Cincinnati, OH; yEmergency Medicine Trauma Center, not occur in the field. In response to this experience, it has
Childrens National Medical Center, Washington, DC; zDivision of been suggested that survival and outcome could be improved
Emergency Medicine, Childrens Memorial Hospital, Northwestern by modifying a number of factors, including the dose of
University Medical School, Chicago, IL; xDivision of Emergency
Medicine, Mary Bridge Childrens Health Center, Tacoma, WA;
epinephrine.8,1721
kEmergency Medicine Section, The Childrens Mercy Hospital, Epinephrine is believed to be effective in cardiac arrest
University of Missouri-Kansas City, School of Medicine, Kansas City, primarily because of its dose-dependent a-adrenergic effect.
MO; {Division of Pediatric Emergency Medicine, Washington Uni- In animal studies, epinephrine increases coronary perfusion
versity School of Medicine, St Louis Childrens Hospital, St Louis, MO pressure as well as myocardial and cerebral blood flow.2123
and #Department of Emergency Medicine, Cooper Hospital/University
Medical Center, Camden, NJ.
In some animal studies, high-dose epinephrine (HDE) was
This study was supported in part by a grant from the Nations Capital found to have increased effectiveness for ROSC at doses
Affiliate of the American Heart Association. 10 to 20 times the Advanced Cardiac Life Support recom-
Address correspondence and reprint requests to Mary D. Patterson, MD, Di- mended dose (0.01 mg/kg).24 These studies have shown
vision of Emergency Medicine, Childrens Hospital Medical Center ML
significant increases in regional myocardial blood flow,
2008, 3333 Burnet Ave, Cincinnati, OH 45229. E-mail: mary.patterson@
cchmc.org. aortic diastolic pressure, cerebral blood flow, and ROSC in
Copyright n 2005 by Lippincott Williams & Wilkins animals receiving 0.1 to 0.2 mg/kg as compared with those
ISSN: 0749-5161/05/2104-0227 receiving 0.01 mg/kg of epinephrine.22,2529 Furthermore,

Pediatric Emergency Care  Volume 21, Number 4, April 2005 227

Copyr ight Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Patterson et al
2 series of pediatric inpatients experienced an improved rate
of ROSC after receiving 0.2 mg/kg of epinephrine as the
third dose during resuscitation.4,30
However, there have been no large prospective clinical
trials that established the effectiveness of HDE in cardio-
pulmonary arrest (CPA) of the pediatric patient. A survey of
pediatric emergency medicine specialists around the time
of this study demonstrated that almost 75% of respondents
were using HDE in pediatric CPA despite the lack of
randomized controlled trials in the pediatric population.31
Therefore, it was important to systematically examine the
use of HDE in pediatric CPR. Patients presenting in CPA
having failed to respond to prehospital resuscitation are
a group that has a particularly poor prognosis.10,16 We
hypothesized that HDE given to these patients would in-
crease ROSC and long-term survival. Neurologic outcome
might also be improved by decreasing the length of arrest.
Our hypothesis was that the use of HDE (0.1 mg/kg) in out-
of-hospital pediatric CPA refractory to prehospital resusci-
tation efforts would result in (1) a higher rate of ROSC, that
is, the return of a cardiac rhythm with a measurable blood
pressure and 24-hour survival, (2) improved long-term
survival, that is, survival to discharge or meeting discharge
criteria after CPA, and (3) improved neurological outcome
in survivors of CPA.
METHODS
Participants and Randomization
As pediatric CPA is a relatively rare event, this study
was designed as a multicenter collaborative trial involving
7 tertiary pediatric centers. The study was conducted by
members of the Pediatric Emergency Medicine Collaborative
Research Committee of the Section on Emergency Medicine
of the American Academy of Pediatrics.
Patients eligible for the study included infants and
children presenting to the collaborating pediatric emergency
departments in CPA, that is, patients who were apneic and
pulseless. (This included older children and adolescents with
chronic underlying conditions who received their usual care
at 1 of these facilities.) The exclusion criteria were:
a. Patients having a respiratory arrest alone or responding to
prehospital interventions before emergency department
arrival. (As we were specifically interested in the effect of
HDE as compared with standard-dose epinephrine [SDE]
in patients refractory to prehospital interventions, patients
were excluded if they had a perfusing rhythm upon
presentation to the emergency department.)
b. Patients with do not resuscitate orders.
c. Patients with evidence of prolonged cessation of cir-
culation, that is, rigor mortis, dependent lividity, and core
temperature less than 338C (not suffering from exposure
or near drowning).
228
Copyr ight Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Pediatric Emergency Care  Volume 21, Number 4, April 2005
Randomization method varied by institution and by
whether enrollment occurred at the time of prehospital con-
tact or emergency department arrival.
Six centers randomized patients by using individual
envelopes designating high or standard dosage that had been
generated using a table of random digits. Institution 3 used
an acute medical record system. In this center, those
patients whose medical record number ended in an odd digit
were assigned to the study group, whereas patients with even
medical record numbers were assigned to the standard-dose
group. Those patients assigned to the standard-dose group
received 0.01 mg/kg as outlined in the applicable Advanced
Cardiac Life Support recommendations at the time. Patients
receiving SDE were eligible to have their doses doubled for
the second and subsequent doses at the physicians discretion
as per Pediatric Advanced Life Support (PALS) guidelines
in effect at the time of the study32 (Fig. 1).
Epinephrine, 0.1 mg/kg (10 times the then-current
Advanced Cardiac Life Support dose), was administered as
the initial dose to patients in the high-dose group. A con-
centration of 1:1000 epinephrine was used preferentially, as
this required a smaller volume. If the patient had no re-
sponse to the first dose, a second dose of epinephrine was
administered after 5 minutes. The second dose of epineph-
rine, as well as all subsequent doses, was 0.2 mg/kg. Thus,
even in those standard-dose patients who had their second
dose doubled, there was still a 10-fold difference with re-
spect to high-dose patients (Fig. 1).
Institutions 2 and 6 attempted to randomize in the
prehospital setting; however, they were not always success-
ful. The remaining centers randomized when the patient
FIGURE 1. Idealized epinephrine dosage schema. EMS
indicates emergency medical services.
n 2005 Lippincott Williams & Wilkins
Pediatric Emergency Care  Volume 21, Number 4, April 2005
arrived in the emergency department. All the hospitals
participating in this study were served by emergency medical
systems with advanced life support capabilities.
The study was not blinded in all centers secondary to
the technical difficulty in doing so. Blinding would have
required transferring 1:10,000 epinephrine to alternative sy-
ringes. Because this concentration is packaged as a single-
dose unpreserved solution, questions of contamination were
raised. Whereas pharmacists in 2 of the centers (1 and 5) felt
comfortable in guaranteeing the potency of repackaged
epinephrine, other pharmacists were reluctant to do so.
Weights of patients presenting in cardiac arrest were
estimated using the Broselow tape, a tool widely used by
prehospital personnel and previously validated.33 Patients
were weighed subsequent to the resuscitation effort. In cases
where actual weights were not obtained, estimated weights
were used to calculate the dose of epinephrine received.
For both groups, epinephrine was administered via an
intravenous (IV) line, intraosseous (IO) line, or endotracheal
(ET) tube. Doses administered via the ET tube were doubled
(to 0.02 mg/kg in the control group and 0.2 and 0.4 mg/kg
in the study group). At the time this study was initiated, the
ET dose was not clearly defined by the then-current PALS
guidelines except that it should be at a minimum equal to the
IV dose and optimally a larger amount.29,33,34 All other
aspects of the resuscitation were conducted according to
then-current Advanced Cardiac Life Support guidelines.24
Resuscitation continued until spontaneous circulation was
restored or until the physician in charge terminated the effort.
Data Collection
Information analyzed from the resuscitation was ob-
tained from the resuscitation record, other information
available from the medical record, prehospital records, and
autopsy results.
Main Outcome Measures
Specific end points evaluated include the return of a
cardiac rhythm with a palpable pulse for greater than 20
minutes and admission to the intensive care unit (sustained
ROSC) as per the pediatric Utstein guidelines.35 All sur-
vivors of resuscitation were then evaluated for 24-hour
survival, survival to discharge, and neurological outcome. To
determine neurological function, long-term survivors were
evaluated using the Glasgow Outcome Scale at time of
discharge. The Glasgow Outcome scale defines neurological
outcome by placing patients into 1 of 5 categories: death,
persistent vegetative state, severe disability, moderate dis-
ability, and good recovery.36
Although this scale has been primarily used in adults,
the defining criteria for each category can be modified to
include age-appropriate activities as described by the Denver
Developmental Scale. This scale has the advantage of being
n 2005 Lippincott Williams & Wilkins
Copyr ight Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
The Use of High-Dose Epinephrine for Patients
easily applied by a wide number of observers with remark-
able consistency.37 In 1 study of those patients categorized as
being in a persistent vegetative state at 1 month, only 28%
progressed to the severe disability classification by 1 year.38
The predictive ability of this scale for the most severely
affected categories is helpful when attempting to determine
whether the use of HDE results in a better or worse
neurological outcome.
Ethical Considerations
Prospective informed consent could not be obtained in
this study because of the severe and life-threatening nature of
CPA. At the time this study was conducted, Food and Drug
Administration regulations permitted the waiver of informed
consent if the following 4 conditions applied:
1. The human subject is confronted by a life-threatening
situation.
2. Informed consent cannot be obtained from the subject
because of an inability to communicate with or obtain
legally effective consent from the subject.
3. Time is not sufficient to obtain consent from the subjects
legal representative.
4. There is no alternative method of approved or generally
recognized therapy that provided an equal or greater
likelihood of saving the life of the patient.39,40
As the research described complied with these 4 con-
ditions, the requirement for informed consent was waived
for this investigation by the institutional review boards of
all the participating institutions. The study was approved
by the institutional review boards of all the participating
institutions.
The study was terminated in October of 1996 second-
ary to implementation of new regulations on waiver of in-
formed consent by the Department of Health and Human
Services and the Food and Drug Administration. These reg-
ulations imposed severe new restrictions on this type of re-
suscitation research.41
Information sheets were provided to the parents of
patients who received HDE as part of this study.
Sample Size and Analysis
At the time this study was begun, it was generally
accepted that 10% or less of pediatric patients having out-of-
hospital CPA refractory to prehospital interventions would
survive to discharge. It was predicted that given the poor
prognosis of this group of patients, the use of HDE would
increase survival by only 10% to 15% under optimal con-
ditions. At an a of 0.05 and a power of 0.80 with an
estimated delta of 14% for long-term survival, it was cal-
culated that 120 patients would be required in each group to
achieve statistical significance. Dichotomous variables were
analyzed by means of x2 or Fisher exact test. Continuous
variables were analyzed using the Student t test. Relative
229
Patterson et al
risks (RRs) and 95% confidence intervals (CIs) were cal-
culated when applicable.
RESULTS
The study was conducted from May 1991 to October
1996. The study was halted at that time because of the
implementation of new regulations on waiver of informed
consent by the Department of Health and Human Services
and the Food and Drug Administration. Two hundred thirty
patients, ranging in age from newborn to 22 years (including
some chronically ill older pediatric patients who continued to
receive care at a tertiary center), were enrolled. Seventeen
patients met exclusion criteria and were excluded from
subsequent analysis (8 patients were found to have rigor/
dependent lividity, 2 patients presented to the emergency
department with a perfusing rhythm, and 7 patients arrested
in the emergency department). Two hundred thirteen patients
were included in the analysis. One hundred fifty-four pa-
tients had CPA as a result of a medical etiology and 59
patients as the result of a traumatic etiology (Fig. 2). Among
the medical patients, 95 were enrolled in the HDE group and
59 in the SDE group. Of the trauma patients, 32 were
assigned to the HDE group and 27 to the SDE group. There
were no significant differences between the HDE and SDE
groups in terms of age, sex, proportion of patients with
underlying conditions, proportion of witnessed arrests, rate
of prehospital intubation, administration of prehospital epi-
nephrine, proportion of patients receiving bystander CPR,
FIGURE 2. Study profile. ROSC determined if patient had a perfusing rhythm for more than 20 minutes and admitted to an
inpatient service. Long-term survival determined when patient is ready for hospital discharge.
230
Copyr ight Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Pediatric Emergency Care  Volume 21, Number 4, April 2005
or proportion of patients receiving initial doses of epineph-
rine via the IV, ET, or IO routes (Table 1). The rhythms seen
at hospital presentation were similar to what has been re-
ported in other studies of pediatric CPA, with most (93.5%)
presenting in asystole. The pattern of enrollment by insti-
tution is demonstrated in Table 2. The study was analyzed in
an intention-to-treat manner. Fourteen patients enrolled in
the HDE group received initial doses of epinephrine <0.1
mg/kg. The mean initial dose received in those 14 patients
was 0.04 mg/kg. Eleven patients enrolled in the SDE group
received initial doses >0.02 mg/kg. The mean initial dose of
these 14 patients was 0.05 mg/kg. All of these patients were
included in the analysis of the group to which they were
originally assigned.
The mean initial dose of epinephrine administered to
medical patients on hospital presentation in the HDE group
was 0.13 mg/kg, and the mean initial dose in the SDE group
was 0.02 mg/kg (t test, P < 0.001). The mean of all doses
of epinephrine administered to patients in the HDE group
was 0.16 (range 0.010.50) mg/kg and to patients in the
SDE group was 0.02 mg/kg (range 0.010.10 mg/kg; t test,
P < 0.001).
Those medical patients experiencing ROSC included
25% (24/95) in the HDE group compared with 15% (9/59)
of patients in the SDE group. This difference was not
statistically significant (P = 0.14, x2 = 2.17, RR = 1.66
[0.833.31]). There was no statistical difference in the
proportion of patients that survived 24 hours between the
n 2005 Lippincott Williams & Wilkins
Pediatric Emergency Care  Volume 21, Number 4, April 2005 The Use of High-Dose Epinephrine for Patients

TABLE 1. Characteristics of HDE and SDE Patients*

Medical Trauma
HDE (n = 95) SDE (n = 59) HDE (n = 32) SDE (n = 27)
y
Mean age (mo) 35.9 (5.9) 31.5 (7.7) 53.3 (10.1) 74.3 (12.1)
Male children (%) 58 (61) 32 (54) 20 (63) 19 (70)
Witnessed arrest (%) 22 (23) 14 (24) 25 (78) 21 (78)
Prehospital intubation (%) 61 (64) 35 (59) 23 (72) 19 (70)
Prehospital epinephrine (%) 50 (53) 28 (47) 10 (32) 11 (41)
Bystander CPR (%) 31 (33) 21 (37) 4 (13) 4 (15)
Route of first dose of epinephrine at hospital:
IV (%) 24 (25) 10 (17) 11 (34) 11 (41)
IO (%) 48 (50) 28 (47) 10 (32) 9 (33)
ET (%) 23 (24) 21 (35) 10 (32) 7 (26)
Underlying conditions (%) 26 (27) 11 (19) 0 (0) 2 (7)
Demographic and exposure characteristics of HDE and SDE patients.
*P > 0.05 for comparisons of all characteristics.
y
SEM; routes of administration are calculated based on the first dose of epinephrine received at the emergency department.

2 groups: 17% (16/95) in the HDE group and 8% (5/59) in Of interest, none of the patients with good neurological
the SDE group: P = 0.14, x2 = 2.16, RR = 1.98 (0.7685.14). outcome were intubated in the prehospital phase, and only 1
Overall, 11 (7.1%) of 154 of medical patients survived to of these received epinephrine in the prehospital phase of
discharge. There were 9 of 95 or 9.5% long-term survivors in care. Four of the survivors in the HDE group received
the HDE medical group as compared with 2 of 59 or 3.4% of bystander CPR (1 with good neurological outcome) as com-
patients in the SDE medical group. This was also not a pared with 1 of the survivors in the SDE group (P = 0.32,
statistically significant difference (P = 0.21, Fisher exact test, Fisher exact test) (Table 3). There was no statistical dif-
RR = 2.75 [0.6112.28]). The number needed to treat with ference in the number of neurologically intact survivors
HDE would be approximately 16 patients for 1 survivor to between the 2 groups, although obviously, the numbers are
discharge. very small (P = 0.51, Fisher exact test).
Of the 9 survivors in the HDE group, 6 suffered severe Among medical patients, a total of 36 patients ex-
neurological outcome as defined by the Glasgow Outcome perienced a witnessed arrest (22 in the HDE group and 14
Scale. The other 3 survivors in the HDE group had moderate in the SDE group). A comparison of outcomes in witnessed
deficits or good outcome. Both of the survivors in the SDE versus unwitnessed arrests is demonstrated in Table 4. The
group had severe deficits. Therefore, the number needed only significant difference identified was for ROSC and
to treat for 1 survivor with good neurological outcome is 24-hour survival in the HDE group.
33 patients. Among HDE patients, 28 (29%) of 95 as compared
with 14 (24%) of 59 SDE patients received at least 1 pre-
hospital dose of epinephrine via the IV or IO route. The
TABLE 2. Enrollment by Institution (Medical Patients Only) mean dose administered was 0.05 mg/kg in the HDE group
Institution HDE SDE ROSC Survivors and 0.01 mg/kg in the SDE group (P < 0.05). Twenty-seven
(28%) of 95 HDE patients and 20 (34%) of 59 of SDE
1* 22 14 3 0 patients received at least 1 prehospital dose of epinephrine
2y 9 7 4 0 via the ET route. The mean dose received was 0.07 mg/kg
3z 34 23 19 9 in the HDE patients and 0.03 mg/kg in the SDE patients.
4 2 2 1 0 (Some patients in both groups received prehospital epineph-
5* 4 5 0 0 rine by more than 1 route.)
6y 18 8 5 2 On analysis of etiologies of arrest, it was determined
7 6 0 1 0 that 39% (37/95) of the HDE patients and 37% (22/59) of
Totals 95 59 33 11 the SDE patients represented probable cases of sudden in-
*Blinded. fant death syndrome (SIDS) (P = 0.84, x2 = 0.04, RR = 1.04
y
Randomized in prehospital setting. [0.691.58]). An analysis was conducted excluding the SIDS
z
Randomized by acute medical record number.
cases. Obviously, the proportion of patients surviving in each

n 2005 Lippincott Williams & Wilkins 231

Copyr ight Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 232
Patterson et al

TABLE 3. Survivor Profile

Neurological
Age Status Before Bystander Prehospital Prehospital Hospital Neurological
Patient Treatment (mo) Sex PMH Arrest CPR Intubation Epinephrine Rhythm** Outcome Final Diagnosis
1 HDE 69 M Sickle cell disease, Normal A Severe Anoxic
smoke inhalation encephalopathy
2 HDE 30 M Near drowning Normal + A Severe Near-drowning
3 HDE 14 M Former 27-wk Unknown + + B Severe Tracheitis
premature infant,
seizures, tracheostomy,
G-tube
4 HDE 141 M Former premature Unknown A Moderate Laryngospasm,
infant, RAD RAD
5 HDE 8 M URI, found Normal + V Severe Anoxic brain
unresponsive damage, seizures
6 HDE 171 M Severe HA Normal + + A Severe Arteriovenous
malformation
7 HDE 38 M Former premature Normal A Severe Persistent
infant URI vegetative state
8 HDE 144 M Mylomeningocele Normal + + A Good Aspiration
seizures pneumonia
9 HDE 151 M Cold water Normal A Good Near drowning
drowning
10 SDE 0.03 M 28-wk Premature Unknown A Severe Respiratory
newborn distress,
prematurity
11 SDE 3 F URI Normal A Severe ALTE
A indicates asystole; B, severe bradycardia; V, ventricular fibrillation; RAD, reactive airway disease; URI, upper respiratory infection; HA, headache; ALTE, acute life threatening event; PMH,
past medical history.
*Rhythm at hospital presentation.

Copyr ight Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Pediatric Emergency Care  Volume 21, Number 4, April 2005

n 2005 Lippincott Williams & Wilkins

Pediatric Emergency Care  Volume 21, Number 4, April 2005 The Use of High-Dose Epinephrine for Patients

TABLE 4. Witnessed Versus Unwitnessed Arrest

ROSC (%) 24-h Survival (%) Discharged (%)
Witnessed arrest, HDE (n = 22) 12 (54) 9 (41) 5 (23)
Unwitnessed arrest, HDE (n = 73) 12 (16) 7 (9) 4 (5)
RR (95% CI) 1.84* (1.152.94) 1.53*y (1.072.18) 1.22y (0.971.54)
Witnessed arrest, SDE (n = 14) 5 (36) 4 (28) 2 (14)
Unwitnessed arrest, SDE (n = 45) 4 (9) 1 (2) 0 (0)
RR (95% CI) 1.42y (0.952.12) 1.37y (0.981.91) 1.17y (0.941.44)
Witnessed arrest, HDE (n = 22) 12 (54) 9 (41) 5 (23)
SDE (n = 14) 5 (36) 4 (28) 2 (14)
RR (95% CI) 1.53 (0.693.40) 1.43 (0.543.77) 1.59y (0.367.11)
Unwitnessed arrest, HDE (n = 73) 12 (16) 7 (9) 4 (5)
SDE (n = 45) 4 (9) 1 (2) 0 (0)
RR (95% CI) 1.85 (0.635.38) 4.31y (0.5533.93) 0.94y (0.891.00)
*P < 0.05.
y
Fisher exact test.

category is higher when the SIDS cases are excluded; how-
ever, there is still no statistical difference between the HDE
and SDE patients (Table 5).
Fifty-nine patients were enrolled as a result of a
traumatic etiology, 32 in the HDE group and 27 in the SDE
group. The patients experienced ROSC in 25% (8/32) of
HDE cases and 22% (6/27) of SDE cases (P = 0.80,
x2 = 0.06, RR = 1.13 [0.452.84]). There were no long-term
survivors in either group of trauma patients.
The dose of epinephrine calculated for use in analysis
was based on the actual weight of the patient obtained after
the arrest when that value was available. In the remainder of
the patients, the dose was calculated based on the estimated
weight. In the medical group, actual weights were obtained
on 34 of 95 HDE patients and 30 of 59 SDE patients. The
difference in weight between the actual weight and the es-
timate was calculated as a percentage of the actual weight in
these patients ([actual weight  estimated weight]/actual
weight). The mean differences between the actual and es-
timated weights were 0.02 and 0.08 for the HDE and
SDE groups, respectively (P = 0.38; 95% CI = 0.08 to
0.21). The 25 to 75 quartile range was 0.17 to 0.11 and
TABLE 5. AnalysisSIDS Cases Excluded
ROSC 24-h Survival Discharged
(%) (%) (%)
SIDS cases
excluded
HDE (n = 58) 21 (36) 13 (22) 9 (15)
SDE (n = 37) 8 (21) 5 (13) 2 (5)
RR (95% CI) 1.59 (0.833.05) 1.50 (0.683.30) 2.29* (0.648.23)
*Fisher exact test.
0.18 to 0.05 in the HDE and SDE groups, respectively. In
the HDE group, 18 weights were underestimated, and 14
weights were overestimated. In the SDE group, 15 weights
were overestimated, and 10 weights were underestimated
based on actual weights obtained. Estimated weights were
consistent with actual weights in 2 HDE and 5 SDE patients.
As this study was not blinded, the number of doses of
epinephrine administered to patients in each group after
hospital arrival was examined to try and ascertain whether
any bias was introduced by the physicians knowledge of the
patients assignment (ie, did the physician try harder in
resuscitating a particular group of patients). The mean
number of doses administered in patients not experiencing
ROSC was 3.38 (SD 1.41) in the HDE group and 3.32 (SD
1.35) in the SDE group. Among those patients experiencing
ROSC, the mean number of doses in the HDE group was
3.33 (SD 1.46) and 3.00 (SD 1.58) in the SDE group. Neither
of these differences was significant. The only significant
difference in number of doses of epinephrine administered
occurred among HDE survivors and HDE patients experi-
encing ROSC but not discharged. The survivors received a
mean of 2.33 doses of epinephrine, whereas the nonsurvivors
received 3.93 doses (2-tailed t test, P = 0.006).
It would have been desirable to evaluate outcomes with
respect to no flow time, prehospital interval, time until
first dose of epinephrine, etc. Whereas we attempted to col-
lect these intervals as per the pediatric Utstein guidelines,
the information was often unavailable or of poor quality.
Therefore, it is impossible to compare data concerning no
flow intervals, prehospital times, etc, between the 2 groups.
More patients were enrolled in the HDE medical group
than would have been anticipated by pure randomization.
In 6 of 7 centers, patients were enrolled by randomization
envelopes in which patients were assigned to groups based

n 2005 Lippincott Williams & Wilkins 233

Copyr ight Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Patterson et al Pediatric Emergency Care  Volume 21, Number 4, April 2005

on random number tables. In 1 center (3), patients were It is possible that a difference in treatment outcome
assigned based on the last digit of the acute medical record would have become significant if we had been able to enroll
number. The number of patients enrolled in this center before a larger number of patients. However, previous pediatric
and following the modification of PALS guidelines (1994) studies by Dieckmann and Vardis44 and Carpenter and
was examined to attempt to ascertain enrollment bias. The
study was conducted over a period of 65 months, 31 (48%)
months of the period elapsing before PALS guideline modi-
fication. Overall, 50 (53%) of 95 HDE medical patients were
enrolled after January 1, 1994. At institution 3, a total of
34 patients were enrolled in the HDE group with 14 (41%)
enrolled after January 1994. In the SDE group, 20 (34%) of
59 patients were enrolled after January 1994. At the same
center, 7 (30%) of the 23 patients enrolled in the SDE group
were enrolled after January 1994. It appears, therefore, that
the number of patients enrolled at the center using a less
desirable method of randomization followed the overall
trend of patient enrollment at the other centers.
In an attempt to determine if a biased randomization
process had affected the results of the study, a statistical
method of balancing the data was attempted. In examining the
data, it appeared that patients entered at institution 1 were not
balanced after 1995; therefore, all patients entered from
institution 1 after 1995 were eliminated. Because no patients
from institution 7 were entered in the standard-dose group,
those patients were also eliminated. The data from institutions
3 and 6 were balanced by randomly selecting 24 patients from
the HDE group of institution 3 and 8 patients in the HDE group
from institution 6. This brings the total number of patients
available for analysis to 115 patients: 59 high-dose and 56
standard-dose patients. The results of the analysis after this
attempt to balance the data are shown in Figure 3. As shown,
there is not a marked difference in the x2 values obtained for
ROSC, survival, or with respect to the relative risk.

DISCUSSION
This study is the first prospective, randomized, con-
trolled examination of the use of HDE in out-of-hospital
CPA in the pediatric population. Unfortunately, although
there may be a trend toward an increased rate of ROSC in the
HDE medical patients, there was no significant improvement
in survival to discharge or neurological outcome. The overall
survival rate of 7.1% is consistent with other relatively
recent studies of pediatric CPA.1215 The small number of
survivors makes it extremely difficult to draw any con-
clusions regarding positive prognostic factors for out-of-
hospital arrest. The hypothesis that the use of HDE would
result in improved rates of ROSC, long-term survival, and
neurological function was not supported by the data. How-
ever, there was also no evidence that the use of HDE resulted
in more survivors with severe neurological deficits. A recent FIGURE 3. Step 1 of balancing data: elimination of patients
enrolled after 1995 at institution 1 and all patients from
study of inpatient pediatric CPA has suggested that HDE institution 7. Step 2: random selection and elimination of
may actually worsen survival. However, this study included excess HDE patients from institutions 3 and 6. *P value of
only 68 patients and 4 patients that survived to discharge.12 Fisher exact test.

234 n 2005 Lippincott Williams & Wilkins

Copyr ight Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Pediatric Emergency Care  Volume 21, Number 4, April 2005
Stenmark45 examining the use of HDE in pediatric CPA also
did not demonstrate a difference in outcome. These inves-
tigations were limited by the retrospective nature of the
studies as well as the smaller number of patients (65 and
51 patients, respectively).42,43 Furthermore, our results are
similar to what has been found with larger numbers in adult
out-of-hospital arrest studies of HDE.
Youngs large (599 patients) study of out-of-hospital
CPA demonstrated an 8.6% long-term survivor rate. Half of
the survivors required no epinephrine.13 A recent study by
Engdahl et al14 of out-of hospital CPA similarly confirmed
only a 5% survival rate. Some studies of inpatient CPA
have showed somewhat better results with survival rates of
15% to 54%.15,46,47 However, the study of Perondi et al12 of
inpatient CPA had only a 6% survival rate. The outcome of
pediatric CPA, particularly out-of hospital CPA, remains
grim.
There are certain limitations to this study which by the
nature of the subject are unavoidable. Patients presenting
to the emergency department differ markedly in the cause
of the arrest as well as the amount of information available
regarding the time of arrest, time until CPR, etc. It is
recognized that because of the wide variability of these
patients presentations, generalizations concerning treatment
and outcome are difficult. In our analysis, we have con-
centrated on patients experiencing CPA as a result of a
medical cause, given the acknowledged and uniformly poor
outcome of patients having traumatic arrest. Unfortunately,
most children with out-of-hospital CPA have prolonged
hypoxia and are unlikely to be successfully resuscitated,
particularly if prehospital efforts have not resulted in
ROSC.16 As mentioned previously, although we attempted
to collect information on no flow time, interval to CPR,
interval to emergency medical services arrival, and scene
transport time, this information was often unavailable or of
poor quality. Although no flow time is believed to be
a significant indication of prognosis in resuscitation, this
information is unfortunately not consistently available. The
only consistently available prehospital intervals recorded
are typically the time of the 911 call, dispatch time, and
time of arrival at the hospital. Obviously, these intervals are
confounded by distance to the scene, time at the scene, and
transport time.
The study was blinded in some but not all centers. It
would have been preferable to study this proposal in a
blinded as well as randomized fashion, as there is a possi-
bility that bias will be introduced by the physicians knowl-
edge of the dose of epinephrine being used. As mentioned
previously, blinding was not feasible secondary to limita-
tions of the drug formulation. An additional concern is the
documentation of accurate weights upon which to base cal-
culation of epinephrine doses. As weights are often esti-
mated during a cardiac arrest, we planned to weigh patients
n 2005 Lippincott Williams & Wilkins
Copyr ight Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
The Use of High-Dose Epinephrine for Patients
subsequent to the arrest and recalculate the dose of epi-
nephrine that was actually administered. Compliance was
not universal as demonstrated in our results. However, the
comparison of actual versus estimated weights is somewhat
reassuring in that the estimates generally fell within 15% to
20% of the actual weight. Although not optimal, this rep-
resents the reality for most pediatric resuscitations for which
an actual weight is rarely available.
The protocol called for repeated epinephrine doses to
be administered at 5-minute intervals as opposed to 3- to 5-
minute intervals. In theory, this longer interval may favor
HDE dosage schema.
It is also noted that institution 3 had a higher per-
centage of survivors than any other institutions. In exam-
ining the possible reasons for this, it is noted that this
institution is located in an inner-city environment with rel-
atively short transport times. Of the 57 patients presenting to
this institution, 21 (37%) were intubated in the prehospital
setting as compared with an overall prehospital intubation
rate of 62% for the entire study. In addition, 10 (17%) of 57 of
patients presenting at this institution had received prehospital
epinephrine as compared with an overall rate of 51% of all
enrolled medical patients. This likely implies a scoop and
run strategy with shorter scene times and transport times
as compared with other institutions. It is also consistent with
the study by Gausche et al11 demonstrating no improvement
in survival for prehospital intubation of pediatric patients.
Another possible confounder is that we do not have data on
the number of patients that required reintubation on hospital
presentation. Data from the study by Gausche et al suggest
that a significant number of prehospital intubations are in-
correctly placed or dislodged. This raises questions with
respect to the adequacy of ventilatory support as well as the
administration of epinephrine via the ET route.
Despite our desire to obtain an autopsy report on every
patient, this was not always possible. In cases of sudden death,
the autopsy is at the discretion and responsibility of the
medical examiner. In some cases, an autopsy was not per-
formed. In other cases, a partial report or no autopsy report
was ever available to medical personnel. In the autopsies
that were available, no mention was made of the type of
myocardial necrosis which has been reported in animal
models receiving HDE.
Our largest concern was our inability to enroll as many
patients as originally planned. With the change of the re-
quirements governing waiver of informed consent, it was no
longer possible to continue the study as initially designed.
We also perceived a reluctance to enroll patients in the SDE
group subsequent to the revision of the PALS guidelines
in 1994. We did modify the SDE protocol after these rec-
ommendations were published to allow patients randomized
to the SDE group to receive higher dose of epinephrine if
they required multiple doses. Although the study was still
235
Patterson et al
randomized, we were concerned about a possible enrollment
bias. It appears that approximately one half of the total
number of HDE patients were enrolled after the PALS
guideline modification. However, approximately one third
of the total SDE patients were enrolled after the PALS
guideline modification. Thus, it is possible that patients
who would have been enrolled in the SDE group by the
randomization scheme were not enrolled at all.
The outcome of the survivors is problematic from an
ethical perspective. Although there is no statistical evidence
that there was a disproportionate number of neurologically
devastated survivors in either group, the fact remains that
of 11 survivors, only 2 (both in the high-dose group) ex-
perienced a good neurological recovery. Similar results
were seen in the study by Engdahl et al14 of out-of-hospital
CPA with only 3 survivors with good neurological outcome.
In the study by Young et al,13 the majority of survivors
with good neurological outcome presented to the emer-
gency department with a perfusing rhythm (these patients
responded to prehospital interventions). The survival of
neurologically devastated patients is tragic for their fami-
lies, the individual, and society. The burden is considerable
and begs the question as to how we can better identify
patients during a resuscitation who are unlikely to recover
neurologically.
Given the current state of the requirements for waiver
of informed consent, it seems unlikely that a similar trial
will be conducted in the future. The Final Rule guidelines
on waiver of informed consent implemented in 1996 im-
posed significant restrictions on this type of resuscitation
research.41 The requirements for community approval, inde-
pendent monitoring boards, etc, may preclude this type of
multicenter trial and make it unlikely that a similar multi-
center trial will be feasible.
Moreover, it is apparent that to improve the outcome of
these patients, we will need to explore other modifications
of therapy. It is possible that the addition of other therapies
or mechanical adjuncts to CPR may improve outcome.
Modifications of CPR such as active compression decom-
pression CPR, interposed abdominal counterpulsation, and
circumferential vest CPR may result in modest improve-
ments in survival, although no pediatric studies have been
conducted using these devices. Some studies have shown
improvement in survival with bystander CPR.16,48 This sug-
gests that an increased emphasis should be placed on pub-
lic education and bystander CPR. Finally, it is clear that
prevention of cardiac arrest is preferable to treatment of
cardiac arrest. Although the etiology of SIDS is still unclear,
certainly, some progress in prevention has been made
(the Back to Sleep campaign).49,50 Similar strides are
necessary to decrease the incidence of CPA caused by other
common etiologies of pediatric CPA such as trauma and
drowning.
236
Copyr ight Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Pediatric Emergency Care  Volume 21, Number 4, April 2005
ACKNOWLEDGMENTS
The authors thank Murray Pollock, MD, for assistance
with study design. Joye Carter, MD, chief medical examiner,
Washington, DC, for assistance in obtaining autopsy re-
sults, Regina ODonnell, PhD, and Judy Bean, PhD, for
statistical assistance, and Maryam Gholizadeh, Kristine Alix
Bailie, and Sonya Worosz for assistance in data collection
and entry.
REFERENCES
1. Friesen RM, Duncan P, Tweed WA, et al. Appraisal of pediatric car-
diopulmonary resuscitation. Can Med Assoc J. 1982;126(9):10551058.
2. Eisenberg M, Bergner L, Hallstrom A. Epidemiology of cardiac arrest
and resuscitation in children. Ann Emerg Med. 1983;12(11):672674.
3. Gillis J, Dickson D, Rieder M, et al. Results of inpatient pediatric
resuscitation. Crit Care Med. 1986;14(5):469471.
4. Goetting MG, Paradis NA. High dose epinephrine in refractory pediatric
cardiac arrest [see comments]. Crit Care Med. 1989;17(12):12581262.
5. Lewis JK, Minter MG, Eshelman SJ, et al. Outcome of pediatric
resuscitation. Ann Emerg Med. 1983;12(5):297299.
6. Nichols DG, Kettrick RG, Swedlow DB, et al. Factors influencing
outcome of cardiopulmonary resuscitation in children. Pediatr Emerg
Care. 1986;2(1):15.
7. Torphy DE, Minter MG, Thompson BM. Cardiorespiratory arrest and
resuscitation of children. Am J Dis Child. 1984;138(12):10991102.
8. Zaritsky A, Nadkarni V, Getson P, et al. CPR in children. Ann Emerg
Med. 1987;16(10):11071111.
9. Sirbaugh PE, Pepe PE, Shook JE, et al. A prospective, population-based
study of the demographics, epidemiology, management, and outcome
of out-of-hospital pediatric cardiopulmonary arrest. Ann Emerg Med.
February 1999;33(2):174184.
10. Schindler MB, Bohn D, Cox PN, et al. Outcome of out-of-hospital
cardiac or respiratory arrest in children. N Engl J Med. November 14,
1996;335(20):14731479.
11. 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. February 9, 2000;283(6):783790.
12. Perondi MB, Reis AG, Paiva EF, et al. A comparison of high-dose and
standard-dose epinephrine in children with cardiac arrest. N Engl J Med.
April 22, 2004;350(17):17221730.
13. Young KD, Gausche-Hill M, McClung CD, et al. A prospective
population based study of the epidemiology and outcome of out-of
hospital pediatric cardiopulmonary arrest. Pediatrics. July 2004;114(1):
157164.
14. Engdahl J, Axelsson A, Bang A, et al. The epidemiology of cardiac
arrest in children and young adults. Resuscitation. August 2003;58(2):
131138.
15. Horisberger T, Fischer E, Fanconi S. One-year survival and neurological
outcome after pediatric cardiopulmonary resuscitation. Intensive Care
Med. March 2002;28(3):365368.
16. Hickey RW, Cohen DM, Strausbaugh S, et al. Pediatric patients
requiring CPR in the prehospital setting [see comments]. Ann Emerg
Med. 1995;25(4):495501.
17. Ludwig S, Fleisher G. Pediatric cardiopulmonary resuscitation: a review
and a proposal. Pediatr Emerg Care. 1985;1(1):4044.
18. Ludwig S, Kettrick RG, Parker M. Pediatric cardiopulmonary
resuscitation. A review of 130 cases. Clin Pediatr (Phila). 1984;23(2):
7175.
19. Orlowski JP. The effectiveness of pediatric cardiopulmonary resuscita-
tion [editorial]. Am J Dis Child. 1984;138(12):10971098.
20. Zaritsky A. Drug therapy of cardiopulmonary resuscitation in children.
Drugs. 1989;37(3):356374.
21. Zaritsky A. Selected concepts and controversies in pediatric cardiopul-
monary resuscitation. Crit Care Clin. 1988;4(4):735754.
22. Michael JR, Guerci AD, Koehler RC, et al. Mechanisms by which
n 2005 Lippincott Williams & Wilkins
Pediatric Emergency Care  Volume 21, Number 4, April 2005
epinephrine augments cerebral and myocardial perfusion during car-
diopulmonary resuscitation in dogs. Circulation. 1984;69(4):822835.
23. Holmes HR, Babbs CF, Voorhees WD, et al. Influence of adrenergic
drugs upon vital organ perfusion during CPR. Crit Care Med. 1980;
8(3):137140.
24. Standards and guidelines for cardiopulmonary resuscitation (CPR) and
emergency cardiac care (ECC). National Academy of Sciences
National Research Council [published erratum appears in JAMA.
October 3, 1986;256(13):1727]. JAMA. 1986;255(21):29052989.
25. Jackson RE, Joyce K, Danosi SF, et al. Blood flow in the cerebral cortex
during cardiac resuscitation in dogs. Ann Emerg Med. 1984;13(9 pt 1):
657659.
26. Brown CG, Werman HA, Davis EA, et al. Comparative effect of graded
doses of epinephrine on regional brain blood flow during CPR in a
swine model. Ann Emerg Med. 1986;15(10):11381144.
27. Brown CG, Werman HA, Davis EA, et al. The effects of graded doses of
epinephrine on regional myocardial blood flow during cardiopulmonary
resuscitation in swine. Circulation. 1987;75(2):491497.
28. Kosnik JW, Jackson RE, Keats S, et al. Dose-related response of
centrally administered epinephrine on the change in aortic diastolic
pressure during closed-chest massage in dogs. Ann Emerg Med. 1985;
14(3):204208.
29. Ralston SH, Tacker WA, Showen L, et al. Endotracheal versus
intravenous epinephrine during electromechanical dissociation with
CPR in dogs. Ann Emerg Med. 1985;14(11):10441048.
30. Goetting MG, Paradis NA. High-dose epinephrine improves outcome
from pediatric cardiac arrest [see comments]. Ann Emerg Med. 1991;
20(1):2226.
31. Schunk JE, Corneli HM. What dose of epinephrine is being used in
pediatric resuscitation? Results of a survey. Pediatr Emerg Care.
1993;9(1):47.
32. Chameides L, eds. Textbook of Pediatric Advanced Life Support. Dallas,
TX: American Heart Association; 1988:2 4.
33. Lubitz DS, Seidel JS, Chameides L, et al. A rapid method for estimating
weight and resuscitation drug dosages from length in the pediatric age
group. Ann Emerg Med. 1988;17(6):576581.
34. Prete MR, Hannan CJ Jr, Burkle FM Jr. Plasma atropine concentrations
via intravenous, endotracheal, and intraosseous administration. Am J
Emerg Med. 1987;5(2):101104.
35. Zaritsky A, Nadkarni V, Hazinski MF, et al. Recommended guidelines
for uniform reporting of pediatric advanced life support: the pediatric
Utstein style. A statement for healthcare professionals from a task
force of the American Academy of Pediatrics, the American Heart
Association, and the European Resuscitation Council. Pediatrics.
October 1995;96(4 pt 1):765779.
n 2005 Lippincott Williams & Wilkins
Copyr ight Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
The Use of High-Dose Epinephrine for Patients
36. Jennett B, Bond M. Assessment of outcome after severe brain damage.
Lancet. 1975;1(7905):480484.
37. Teasdale G, Knill-Jones R, van der Sande J. Observer variability in
assessing impaired consciousness and coma. J Neurol Neurosurg
Psychiatry. 1978;41(7):603610.
38. Heiden JS, Small R, Caton W, et al. Severe head injury. Clinical
assessment and outcome. Phys Ther. 1983;6(12):19461951.
39. Abramson NS, Meisel A, Safar P. Informed consent in resuscitation
research. JAMA. 1981;246(24):28282830.
40. Abramson NS, Meisel A, Safar P. Deferred consent. A new approach
for resuscitation research on comatose patients. JAMA. 1986;255(18):
24662471.
41. Protection of human subjects: informed consent and waiver of informed
consent requirements in certain emergency research; final rules. Fed
Regist. 1996;61:5149851533.
42. Dieckmann RA, Vardis R. High-dose epinephrine in pediatric out-of-
hospital cardiopulmonary arrest. Pediatrics. June 1995;95(6):901913.
43. Carpenter TC, Stenmark KR. High-dose epinephrine is not superior to
standard-dose epinephrine in pediatric in-hospital cardiopulmonary
arrest. Pediatrics. March 1997;99(3):403408.
44. Gueugniaud PY, Mols P, Goldstein P, et al. A comparison of repeated
high doses and repeated standard doses of epinephrine for cardiac arrest
outside the hospital. European Epinephrine Study Group. N Engl J Med.
November 26, 1998;339(22):15951601.
45. Brown CG, Martin DR, Pepe PE, et al. A comparison of standard-dose
and high-dose epinephrine in cardiac arrest outside the hospital. The
Multicenter High-Dose Epinephrine Study Group. N Engl J Med.
October 8, 1992;327(15):10511055.
46. Reis AG, Nadkarni V, Perondi MB, et al. A prospective investigation
into the epidemiology of in-hospital pediatric cardiopulmonary re-
suscitation using the international Utstein reporting style. Pediatrics.
February 2002;109(2):200209.
47. Suominen P, Olkkola KT, Voipio V, et al. Utstein style reporting of in-
hospital paediatric cardiopulmonary resuscitation. Resuscitation. June
2000;45(1):1725.
48. Kuisma M, Suominen P, Korpela R. Paediatric out-of-hospital cardiac
arrestsepidemiology and outcome. Resuscitation. October 1995;30(2):
141150.
49. Gibson E, Fleming N, Fleming D, et al. Sudden infant death syndrome
rates subsequent to the American Academy of Pediatrics supine sleep
position. Med Care. June 1998;36(6):938942.
50. Willinger M, Hoffman HJ, Hartford RB. Infant sleep position and risk
for sudden infant death syndrome: report of meeting held January 13 and
14, 1994, National Institutes of Health, Bethesda, MD. Pediatrics. May
1994;93(5):814819.
237