DOI: 10.1542/peds.

2007-0930
2008;121;e58 Pediatrics
Byarugaba
Arthur Mpimbaza, Grace Ndeezi, Sarah Staedke, Philip J. Rosenthal and Justus
Prolonged Seizures in Ugandan Children: A Randomized Clinical Trial
Comparison of Buccal Midazolam With Rectal Diazepam in the Treatment of

http://pediatrics.aappublications.org/content/121/1/e58.full.html
located on the World Wide Web at:
The online version of this article, along with updated information and services, is

of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.
Boulevard, Elk Grove Village, Illinois, 60007. Copyright 2008 by the American Academy
published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point
publication, it has been published continuously since 1948. PEDIATRICS is owned,
PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly
at Indonesia:AAP Sponsored on May 28, 2014 pediatrics.aappublications.org Downloaded from at Indonesia:AAP Sponsored on May 28, 2014 pediatrics.aappublications.org Downloaded from
ARTICLE
Comparison of Buccal Midazolam With Rectal
Diazepam in the Treatment of Prolonged Seizures in
Ugandan Children: A Randomized Clinical Trial
Arthur Mpimbaza, MMed
a
, Grace Ndeezi, MMed
a
, Sarah Staedke, MD
b
, Philip J. Rosenthal, MD
c
, Justus Byarugaba, MMed
a
a
Department of Pediatrics and Child Health, Faculty of Medicine, Makerere University, Kampala, Uganda;
b
London School of Hygiene and Tropical Medicine,
London, United Kingdom;
c
Department of Medicine, San Francisco General Hospital, University of California, San Francisco, California
Financial Disclosure: Both study drugs were donated by Roche Products Limited, Nairobi, Kenya.
ABSTRACT
OBJECTIVE. Our goal was to compare the efcacy and safety of buccal midazolam with
rectal diazepam in the treatment of prolonged seizures in Ugandan children.
METHODS. This was a single-blind, randomized clinical trial in which 330 patients were
randomly assigned to receive buccal midazolam or rectal diazepam. The trial was
conducted in the pediatric emergency unit of the national referral hospital of
Uganda. Consecutive patients who were aged 3 months to 12 years and presented
while convulsing or who experienced a seizure that lasted 5 minutes were ran-
domly assigned to receive buccal midazolam plus rectal placebo or rectal diazepam
plus buccal placebo. The primary outcome of this study was cessation of visible
seizure activity within 10 minutes without recurrence in the subsequent hour.
RESULTS. Treatment failures occurred in 71 (43.0%) of 165 patients who received rectal
diazepam compared with 50 (30.3%) of 165 patients who received buccal midazo-
lam. Malaria was the most common underlying diagnosis (67.3%), although the risk
for failure of treatment for malaria-related seizures was similar: 35.8% for rectal
diazepam compared with 31.8% for buccal midazolam. For children without malaria,
buccal midazolam was superior (55.9% vs 26.5%). Respiratory depression occurred
uncommonly in both of the treatment arms.
CONCLUSION. Buccal midazolam was as safe as and more effective than rectal diazepam
for the treatment of seizures in Ugandan children, although benets were limited to
children without malaria.
P
ROLONGED SEIZURES ARE responsible for 15% of visits to pediatric emergency
departments in sub-Saharan Africa.
13
The cause of seizures in Africa differs from
that in developed countries, because infectious diseases that are unique to the tropics
are common underlying factors, in addition to simple febrile convulsions and epi-
lepsy.
4,5
In malaria-endemic areas, falciparum malaria remains the leading cause of
seizures in children who present to emergency departments.
6
In children with
cerebral malaria, seizures have been shown to increase the risks for death and
neurologic sequelae.
7,8
Seizures that last for 5 minutes are termed prolonged.
9
Unlike brief seizures, those that are prolonged are associated with an increased risk
for poor outcome
10,11
; therefore, prolonged seizures warrant urgent treatment that is
focused on early and safe seizure termination, prevention of recurrence, and iden-
tication and treatment of precipitating conditions and secondary complications.
12
In most of sub-Saharan Africa, diazepam remains the rst-line treatment for seizures in children.
13
Although this
drug is fairly efcacious, inexpensive, readily available, and easy to administer via the rectal route when intravenous
access is not available, rectally administered diazepam results in variable plasma concentrations and fails to terminate
30% of seizures.
14
An advantage of rectal administration of diazepam is that the risk for respiratory depression, the
www.pediatrics.org/cgi/doi/10.1542/
peds.2007-0930
doi:10.1542/peds.2007-0930
This trial has been registered at
www.controlled-trials.com (identier
ISRCTN13964268).
Dr Mpimbaza contributed to study design
and coordination, collected data,
supervised patient enrollment and follow-
up, and analyzed and interpreted data; Dr
Byarugaba contributed to study design,
supervised coordination of the study,
enrollment, and clinical care of the
patients, and interpreted data; Dr Ndeezi
contributed to study design and
coordination, supervised enrollment and
follow-up of the patients, and interpreted
data; Dr Staedke contributed to study
design, analysis, and data interpretation;
and Dr Rosenthal contributed to study
design and coordination, analysis, and data
interpretation. All authors contributed to
the writing of the manuscript.
Key Words
midazolam, diazepam, seizures, Ugandan,
children
Abbreviations
ACUacute care unit
RRrelative risk
CIcondence interval
IQRinterquartile range
Accepted for publication May 29, 2007
Address correspondence to Arthur Mpimbaza,
MMed, Makerere University, Department of
Pediatrics and Child Health, Faculty of
Medicine, PO Box 7072, Kampala, Uganda.
E-mail: arthurwakg@yahoo.com
PEDIATRICS (ISSNNumbers: Print, 0031-4005;
Online, 1098-4275). Copyright 2008 by the
American Academy of Pediatrics
e58 MPIMBAZA et al
at Indonesia:AAP Sponsored on May 28, 2014 pediatrics.aappublications.org Downloaded from
principal concern with intravenous administration of the
drug, is low
15
; however, because of the tendency of
diazepam to accumulate in adipose tissue, repeated
doses of rectal diazepam can cause marked respiratory
depression.
16,17
This is of particular concern in units with-
out infrastructure to provide ventilatory support. An
additional concern is that a combination of diazepam, in
multiple doses, and phenobarbital has been associated
with an increased risk for mortality from respiratory
depression in children with cerebral malaria.
17
Midazolam, an inexpensive benzodiazepine with
anticonvulsive activity, can be administered via mul-
tiple routes, including topical application in the buccal
cavity.
18
Buccal midazolam has been found to be as
effective as rectal diazepam in control of seizures in
developed countries.
19,20
Recently, in a randomized,
controlled trial, buccal midazolam was as safe as and
more efcacious than rectal diazepam for the treat-
ment of seizures in children who presented to hospi-
tals in Great Britain.
21
Potential advantages of buccal
midazolam over rectal diazepam include improved ef-
cacy, at least in developed countries, ease of admin-
istration, and safety. To evaluate further the use of
buccal midazolam in Africa, we compared the efcacy
and safety of buccal midazolam with rectal diazepam
in Kampala, Uganda.
METHODS
Study Design
We conducted a single-blind, randomized clinical trial to
compare the efcacy and safety of buccal midazolam
versus rectal diazepam.
Study Site and Population
The study was conducted between November 2005
and June 2006 in the acute care unit (ACU), the
pediatric emergency unit of Mulago Hospital, the na-
tional referral hospital in Kampala, Uganda. In Kam-
pala, malaria is mesoendemic (25% palpable spleen
rate, 25% parasitemia rate), occurring perennially
with peaks during 2 rainy seasons (A. Talisuna,
MBchB, MSc, PhD, personal communication, 1994).
The ACU is the emergency unit for review and admis-
sion of acutely ill children up to the age of 12 years.
The unit admits between 30 and 70 patients daily.
Consecutive patients who were aged 3 months to 12
years and presented to the ACU while convulsing or
who experienced a seizure that lasted 5 minutes while
in the unit were screened for enrollment. Patients were
enrolled when they fullled the following criteria: (1) 3
months to 12 years of age; (2) no documented evidence
of having received intravenous diazepam or intravenous
phenobarbital within 24 hours before presentation; (3)
documented seizure persisting at the time of administra-
tion of study drug; and (4) provision of informed consent
to continue participation in the study.
In view of ethical and practical limitations, informed
consent was waived on emergency presentation. Writ-
ten consent to continue participation in the study was
subsequently sought from parents or legal guardians as
soon as was practically possible after initial evaluation
and treatment. The study was approved by the Uganda
National Council for Science and Technology and the
institutional review boards of Makerere University, Kam-
pala, and the University of California, San Francisco; it
was overseen by a data and safety monitoring board that
consisted of 2 pediatricians and 1 pharmacist, all in
Kampala. The study was reviewed on 3 occasions.
Study Procedures
Screening and subsequent enrollment of patients was
done consecutively. A child with a suspected seizure was
transferred to a resuscitation room, where the study
doctor rapidly assessed the patient to conrm genuine
convulsive activity, examined the patients airway for
gastric contents, and screened the patient for enroll-
ment. If the patient qualied for enrollment, then a
parent or legal guardian was briey informed of the
study procedures. If they agreed to proceed, then the
patient was randomly assigned to 1 of 2 treatment arms:
rectal diazepam and buccal placebo or rectal placebo and
buccal midazolam.
For randomization, a person independent of the study
used a computer to generate a list of sequential random
treatment codes, representing each treatment arm, using
variable blocks of 4 or 6. Each assigned treatment code
and the corresponding sequential study treatment num-
ber were written on a piece of paper in the order of the
randomization list and placed in an opaque envelope
that was labeled with the treatment number, sealed, and
placed in a box for the nurse who was responsible for
treatment allocation. Investigators were not aware of a
patients treatment allocation. Two placebos were de-
signed (Department of Pharmacy, Faculty of Medicine,
Makerere University) to have a color that matched the
study drug; normal saline for buccal midazolam and
distilled water colored with riboavin, giving it a yellow
tinge, for rectal diazepam. The volumes of placebos were
equivalent to those of study drugs. The concentration of
riboavin administered was below its recommended
daily dietary allowance. Study drugs and placebos were
prepackaged by a pharmacist who was not involved with
patient care and kept in 2 boxes that corresponded to the
2 treatment arms: each included study drug and placebo,
separated into aliquots for 4 different age-based dosing
categories. Although the study team was not aware
which treatment a patient received, they were aware of
the treatment code to which a patient was assigned,
potentially introducing bias, so we considered this to be
a single-blind trial. Boxes were stored at 5 to 10C and
emptied and relled every 7 days. Diazepam and rectal
placebo were packaged in 2-mL glass syringes, whereas
midazolam and buccal placebo were packaged in 2-mL
plastic syringes. Both midazolam and diazepam are sta-
ble under these conditions for up to 1 month.
2224
When
a patient was due to receive treatment, the study nurse
opened the top randomization envelope, noted the
treatment code, and selected age-based dosages from the
box with the patients assigned treatment code. Both
drugs were administered at 0.5 mg/kg (2.5 mg for 311
PEDIATRICS Volume 121, Number 1, January 2008 e59
at Indonesia:AAP Sponsored on May 28, 2014 pediatrics.aappublications.org Downloaded from
months of age; 5 mg for ages 14 years; 7.5 mg for ages
59 years; and 10 mg for ages 1012 years).
Treatment was administered simultaneously by 2
study nurses. For buccal treatments, a syringe was
placed between the teeth and cheek, the drug or placebo
was administered, and the cheek was gently massaged.
For rectal treatments, the drug or placebo was adminis-
tered via a tube inserted 3 to 4 cm into the rectum and
the tube was ushed with air to ensure complete deliv-
ery of the drug. The buttocks were then held together for
5 minutes to prevent expulsion. During a seizure, oxy-
gen was administered by nasal prongs. Peripheral oxy-
gen saturation and blood pressure were recorded on
study drug administration and at 5, 10, 20, 40, and 60
minutes thereafter. All children in the study had a ran-
dom blood sugar level determined with a glucometer
(Sensorex Apex Biotechnology Corp, Hsinchu, Taiwan)
during the course of study drug administration. Two
thick blood lms were prepared, 1 stained with Field
stain and read in the ACU laboratory by the laboratory
technologist on duty and the other stained with Giemsa
and read subsequently by an experienced malaria mi-
croscopist for conrmation. Patients were followed up
for 24 hours after study drug administration or until
cessation of the study as a result of loss to follow-up or
death.
Outcome Measurements
The primary study outcome was cessation of visible sei-
zure activity within 10 minutes, without recurrence in
the subsequent hour. When the convulsion persisted
beyond 10 minutes or recurred within 1 hour, the child
was categorized as having treatment failure and treated
with intravenous diazepam. Secondary outcome mea-
sures included proportion with cessation of convulsions
within 10 minutes, time to cessation of convulsion
within 10 minutes, proportion with seizure recurrence
in subsequent hour and within 24 hours after initial
control, and time to recurrence within the respective
time periods. We also assessed the risk for respiratory
depression, dened as a persistent decrease in oxygen
saturation to 92% or a decrease in respiratory effort
sufcient to warrant assisted breathing. This approach
was consistent with the hospitals policies for assisted
ventilation.
Sample Size
On the basis of a similar study conducted in Great Brit-
ain,
19
we estimated that the proportion of patients whose
seizure would not successfully terminate within 10 min-
utes would be 41% after rectal diazepam compared with
25% after buccal midazolam. With these estimates, a
study with a sample size of 176 patients in each arm was
required to show a 16% difference between the 2 treat-
ment groups with 90% power and a 2-tailed signicance
of 5%.
Data Analysis
Data were analyzed as intention to treat. Data were
entered on standardized case record forms and doubly
entered into EpiInfo 6.04 (Centers for Disease Control
and Prevention, Atlanta, GA). SPSS 12 (SPSS Inc, Chi-
cago, IL) and Stata 8.0 (Stata Corp, College Station, TX)
were used for data analysis. The Mann-Whitney test was
FIGURE 1
Study prole.
a
Some patients had more than 1 rea-
son for exclusion.
TABLE 1 Baseline Characteristics of Enrolled Children
Characteristic Rectal Diazepam
(N 165)
Buccal Midazolam
(N 165)
Demographics
Male, n (%) 82 (49.7) 84 (50.9)
Age, median (IQR), mo 18.0 (11.536.0) 17.0 (10.530.0)
Clinical features
Axillary temperature, mean SD, C 38.2 1.15 38.2 1.16
No. of convulsions before treatment,
median (IQR)
a
2 (14) 3 (14)
Classication of convulsion, n (%)
Febrile (37.5C ) 115 (69.7) 121 (73.3)
Generalized 134 (81.2) 135 (81.8)
Focal 31 (18.8) 30 (18.2)
Laboratory results
Positive blood-smear result for
malaria parasites, n (%)
106 (64.2) 116 (70.3)
Random blood sugar level, N 162 160
Median (IQR), mg/dL 117 (83.0152.7) 106 (80.1165.5)
a
During 24 hours before enrollment.
e60 MPIMBAZA et al
at Indonesia:AAP Sponsored on May 28, 2014 pediatrics.aappublications.org Downloaded from
used to compare medians, and the
2
test was used to
assess the relative risk (RR) for treatment failure.
RESULTS
Patient Recruitment and Baseline Characteristics
Of 525 patients screened, 195 (37.1%) were excluded
(Fig 1). The most common reason for exclusion was
termination of seizure before enrollment. Baseline char-
acteristics of study patients were similar in the 2 treat-
ment arms (Table 1). The majority (94.8%) of children
enrolled in the study were between the ages of 3 months
and 5 years. Of the 330 patients enrolled, 304 (92.1%)
presented with a history of fever as reported by the
parent or guardian and 236 (71.5%) had a documented
temperature 37.5C. Considering the nature of con-
vulsions, 269 (81.5%) were generalized and 61 (18.5%)
were focal. Of the generalized convulsions, 249 (92.5%)
were tonic-clonic, 18 (6.9%) were tonic, and 3 patients
experienced myoclonic jerks. Severe malaria (excluding
cerebral malaria) was the most frequent clinical diagno-
sis, followed by cerebral malaria, bacterial meningitis,
pneumonia, and epilepsy (Fig 2).
Primary Outcome
Comparing the 2 treatment arms, 71 (43.0%) patients
who received rectal diazepam experienced treatment
failure compared with 50 (30.3%) who received buccal
midazolam (RR: 1.42; 95% condence interval [CI]:
1.061.90; P .016; Table 2). When we considered only
those with malaria, the risk for treatment failure did not
differ between the 2 treatment arms (35.8% vs 31.8%;
RR: 1.12; 95% CI: 0.781.62; P .534). For those
without malaria, the risk for treatment failure was sig-
nicantly higher when the patient had received rectal
diazepam (55.9%) compared with buccal midazolam
(26.5%; RR: 2.11; 95% CI: 1.263.54; P .002). On the
last interim review, when 300 patients had been en-
rolled, the data and safety monitoring board identied a
signicant difference between treatment arms and rec-
ommended that the study be halted when 330 patients
had been enrolled.
Secondary Outcome
For initial cessation of seizures, 114 (69.1%) seizures
terminated within 10 minutes in the diazepam arm com-
pared with 125 (75.8%) in the midazolam arm (RR:
0.91; 95% CI: 0.801.04; P .175). The median time to
cessation of the seizure was 4.4 minutes (interquartile
range [IQR]: 2.726.58) for rectal diazepam and 4.8
minutes (IQR: 3.026.52; P .518) for buccal midazo-
lam (Table 3). Of the 114 children whose seizure was
initially controlled within 10 minutes by rectal diaze-
pam, 20 (17.5%) experienced a seizure recurrence in the
subsequent hour compared with 10 (8%) of 125 chil-
dren in the buccal midazolam arm (RR: 2.19; 95% CI:
1.074.50; P .026). Of children who experienced a
seizure recurrence within 1 hour after initial control, the
median time to recurrence was 20 minutes (IQR: 11.0
47.2) for rectal diazepam and 25 minutes (IQR: 2.75
36.7) for buccal midazolam (P .492). Considering the
risk for seizure recurrence after initial control during the
24 hours after treatment, 51 (46.3%) who were treated
with rectal diazepam recurred versus 47 (39.1%) who
were treated with buccal midazolam. The median time
to recurrence within 24 hours differed signicantly be-
tween the 2 treatment arms: 1.8 hours (IQR: 0.933.48)
for rectal diazepam and 5.11 hours (IQR: 1.0810.0; P
80
25
10 9
7
34
91
23
8 7 6
30
0
20
40
60
80
100
120
S
e
v
e
r
e
m
a
l
a
r
i
a
C
e
r
e
b
r
a
l
m
a
l
a
r
i
a
M
e
n
i
n
g
i
t
i
s
P
n
e
u
m
o
n
i
a
E
p
i
l
e
p
s
y
O
t
h
e
r
s
N
o
.

o
f

p
a
t
i
e
n
t
s
Diazepam
Midazolam
FIGURE 2
Distribution of clinical diagnosis in the 2 treatment groups.
TABLE 2 Risk for Treatment Failure
Outcome Rectal Diazepam
(N 165)
Buccal Midazolam
(N 165)
P
All patients
Treatment failure, n (%) 71 (43.0) 50 (30.3) .016
a
Patients with malaria, N 106 116
Treatment failure, n (%) 38 (35.8) 37 (31.8) .534
Patients without malaria, N 59 49
Treatment failure, n (%) 33 (55.9) 13 (26.5) .002
b
a
RR: 1.42 (95% CI: 1.061.90); P .016.
b
RR: 2.11 (95% CI: 1.263.54); P .002.
TABLE 3 Secondary Outcomes After Treatment
Outcome Rectal Diazepam
(N 165)
Buccal Midazolam
(N 165)
P
Initial 10 min
Stopped convulsing, n (%) 114 (69.1) 125 (75.8) .175
Time to stop convulsing,
median (IQR), min
4.35 (2.726.58) 4.75 (3.026.52) .518
1 h after control in rst 10 min, N 114 125
Recurred in next 1 h, n (%) 20 (17.5) 10 (8) .026
a
Time to recur within 1 h, N 20 10
Median (IQR), min 20.0 (11.047.2) 25.0 (2.7536.7) .492
24 h after control in rst 10 min
b
Recurred in next 24 h, N 110 120
n (%) 51 (46.3) 47 (39.1) .270
Time to recur within 24 h, N 51 47
Median (IQR), h 1.81 (0.563.48) 5.11 (1.0810.3) .001
c
a
RR: 2.19 (95% CI: 1.074.50); P .026.
b
9 patients who received phenobarbitone 1 hour after seizure control were excluded
c
Mann-Whitney test.
PEDIATRICS Volume 121, Number 1, January 2008 e61
at Indonesia:AAP Sponsored on May 28, 2014 pediatrics.aappublications.org Downloaded from
.001) for buccal midazolam. This difference was signi-
cant even for children with malaria: 2.10 hours (IQR:
0.763.40) for rectal diazepam versus 5.18 hours (IQR:
0.6311.2; P .028) for buccal midazolam. After treat-
ment, parents were asked which route of treatment
administration they preferred; 184 (56%) favored the
buccal route compared with 111 (34%) who favored the
rectal route. The remaining parents did not mind which
route was used.
Safety
Only 4 (1.2%) children experienced respiratory depres-
sion. These patients included 2 in the diazepam group (1
who had cerebral malaria and died after seizure cessa-
tion and 1 who had meningitis and recovered after re-
suscitation and treatment with antibiotics) and 2 pa-
tients in the buccal midazolam arm (1 who had status
epilepticus with raised intracranial pressure and had a
full recovery after treatment with mannitol and antibi-
otics and 1 who had severe malaria and also fully recov-
ered).
Adverse Events
One patient, a 2-year-old girl who was admitted with
severe malaria and multiple convulsions, developed
aphasia 12 hours after receiving buccal midazolam. This
event was considered to be unlikely to be related to the
midazolam treatment, given that aphasia is a known
complication of multiple convulsions. Another patient,
who received buccal midazolam in addition to oral phe-
nobarbitone, experienced intense pruritus that resolved
on oral antihistamines. The pruritus was deemed possi-
bly related to midazolam treatment.
Deaths
Twenty children died during the course of the study, 12
of 165 in the rectal diazepam arm and 8 of 165 in the
buccal midazolam arm (RR: 1.5; 95% CI: 0.633.57; P
.356). Severe malaria accounted for 10 (50%) of the
deaths; 8 of the children who died had cerebral malaria,
and 2 had severe malaria complicated by severe anemia
and multiple convulsions. Severe malnutrition and sep-
ticemia each accounted for 3 (15%) of the deaths. Two
(10%) children died of pneumonia; both of these chil-
dren had HIV and had clinical evidence of severe im-
mune suppression. Meningitis accounted for the re-
maining 2 (10%) deaths.
DISCUSSION
In this study of treatment for prolonged seizures in
African children, patients who received buccal mida-
zolam were more likely than those who received rectal
diazepam to have successful control of seizures, de-
ned as cessation of convulsive activity within 10
minutes, without recurrence in the subsequent hour.
In subgroup analysis, the risk for treatment failure
differed only in children who did not have a diagnosis
of malaria. The risk for respiratory depression was
minimal and similar in both treatment arms, although
this risk might be greater with multiple doses of either
drug. In summary, in this study of an urban African
population, buccal midazolam was superior to rectal
diazepam for treatment of children with prolonged
convulsions, but the benet of midazolam was con-
ned to children with convulsions that were not as-
sociated with malaria.
Our study provides a useful summary of causes of
prolonged seizures in African children. Malaria was by
far the most common problem underlying presentation
with prolonged seizures, with prevalence similar to that
seen in other hospital-based studies of seizures from
Nigeria and Kenya.
1,25
Our study population seems to be
representative of sub-Saharan Africa but very different
from that in Great Britain, in which buccal midazolam
was shown to have superior efcacy over rectal diaze-
pam for control of prolonged seizures.
21
It was of interest
to determine whether the efcacy benets of buccal
midazolam applied also to Africa, where the need for
simple therapies for seizures is particularly great, be-
cause intravenous therapy is routinely unavailable. It is
encouraging that buccal midazolam was highly efca-
cious, because this regimen offers simple administration,
relatively low cost, lack of need for refrigeration, and
low risk for respiratory depression. Indeed, buccal mida-
zolam provides a more socially acceptable route of drug
administration than rectal diazepam and avoids the need
for intravenous access,
26,27
which is often unavailable in
many district hospitals in Africa and can be challenging
to establish in a convulsing child; therefore, the buccal
route of administration offers an appealing alternative
for seizure control in the community and also in situa-
tions in which establishing intravenous access is prob-
lematic or not possible.
Additional options for anticonvulsant therapy in Af-
rica include intramuscular paraldehyde and intranasal
lorazepam. A recent study showed that intranasal loraz-
epam was as effective as intramuscular paraldehyde in
stopping seizures in Malawi
28
; however, both of these
agents have important drawbacks, including high cost
and need for refrigeration of parenteral lorazepam and
risk for sterile abscesses with intramuscular paralde-
hyde.
16,29
Considering available options, buccal midazo-
lam seems to be the superior simple therapy for pro-
longed seizures in African children.
An effective anticonvulsant should rapidly control
seizures and also prevent recurrent seizure activity
12
;
therefore, we judged treatment success on the basis of
both rapid seizure termination and persistence of anti-
convulsive effects. It is interesting that the effects of
rectal diazepam and buccal midazolam differed primarily
not in their initial effects but in prevention of seizure
recurrences in the subsequent 1 hour after initial con-
trol. The number of seizures that failed to stop in the
initial 10 minutes was higher in patients who received
diazepam, but this difference was not statistically signif-
icant. The regimens differed more greatly in the risk for
seizure recurrence in the subsequent hour, with the risk
for recurrence with rectal diazepam signicantly higher
than that with buccal midazolam, which was similar to
that in other studies.
21
Over 24 hours, the risk for seizure
recurrence was similar with midazolam or diazepam, but
e62 MPIMBAZA et al
at Indonesia:AAP Sponsored on May 28, 2014 pediatrics.aappublications.org Downloaded from
the time to recurrence in patients who received mida-
zolam was signicantly longer. The difference between
diazepam and midazolam was likely attributable to the
shorter duration of action of diazepam in the brain (ac-
tive half-life of diazepam: 1 hour),
30
which is explained
by the rapid decline in diazepam brain concentration
levels during the redistribution of diazepam from cere-
bral gray matter into white matter, brainstem, and body
fat.
14,16
The active half-life of midazolam is 2 to 3 hours in
healthy children and 5 hours in very ill children.
3133
The benets of buccal midazolam were pronounced
in children who did not have malaria diagnosed at the
time of presentation with seizures, but there was no
difference in efcacy in children who had convulsions
associated with malaria. These results suggest that the
cause of convulsions in severe malaria differs from that
in other conditions.
34
Indeed, after omission of children
with malaria, the efcacies of rectal diazepam and buccal
midazolam were remarkably similar in Uganda and
Great Britain.
21
CONCLUSIONS
Our results suggest that buccal midazolam offers a prom-
ising alternative to rectal diazepam for the treatment of
seizures in African children. Midazolam offers benets
over diazepam in ease of use, improved efcacy over 1
hour, and a more prolonged anticonvulsive effect. For
children with malaria-related seizures, midazolam re-
mains a suitable alternative to diazepam considering its
ease of use and that the underlying diagnosis may not be
known at the time of treatment. Thus, in addition to its
immediate benets, buccal midazolam may be a more
effective bridge to long-acting agents in children who
need prolonged anticonvulsant therapy.
ACKNOWLEDGMENTS
This investigation received support from the Fogarty
International Center of the National Institutes of Health
(grant D43 TWO1506). Financial support was also pro-
vided by the Nufeld Foundation, United Kingdom. Dr
Rosenthal is a Doris Duke Charitable Foundation Distin-
guished Clinical Scientist. The sponsors of this study had
no role in study design or collection, analysis, or inter-
pretation of data or in the writing of the report.
We thank the study clinical team, Aggrey Dhabangi,
Jolly Rubambarama, Julian Eyotaru, Florence Pido, Rose
Nakikwaku, and Maria Rutaro, and the laboratory team,
Maxwell Kilama, Regina Nakafeero, and Felix Jurua. We
also thank Grant Dorsey for providing methodologic and
statistical guidance and all of the families and children
who participated in this study. Finally, we thank Prof
Richard Odome Odoi (Head Department of Pharmacy
Medical School, Makerere University), Paul Musoke,
and Benjamin Mwesigwe, who designed and prepared
the placebos that were used in the study.
REFERENCES
1. Waruiru CM, Newton CR, Forster D, et al. Epileptic seizures
and malaria in Kenyan children. Trans R Soc Trop Med Hyg.
1996;90:152155
2. Iloeje SO. Paediatric neurologic emergencies at the University
of Nigeria Teaching Hospital, Enugu. West Afr J Med. 1997;16:
8084
3. Mulowoza M. Knowledge, Attitudes and Practices About Convul-
sions Among Carers of Children Presenting to Mulago Hospital [dis-
sertation]. Kampala, Uganda: Makerere University; 2004
4. Birbeck GL. Neurologic disease in a rural Zambian hospital.
Trop Doct. 2001;31:8285
5. Ofovwe GE, Ibadin MO, Okunola PO, Ofoegbu B. Pattern of
emergency neurologic morbidities in children. J Natl Med Assoc.
2005;97:488492
6. Ogutu BR, Newton CR. Management of seizures in children
with falciparum malaria. Trop Doct. 2004;34:7175
7. Molyneux ME, Taylor TE, Wirima JJ, Borgstein A. Clinical
features and prognostic indicators in paediatric cerebral
malaria: a study of 131 comatose Malawian children. Q J Med.
1989;71:441459
8. Brewster DR, Kwiatkowski D, White NJ. Neurological sequelae
of cerebral malaria in children. Lancet. 1990;336:10391043
9. Lowenstein DH, Bleck T, Macdonald RL. Its time to revise the
denition of status epilepticus. Epilepsia. 1999;40:120122
10. Metsaranta P, Koivikko M, Peltola J, Eriksson K. Outcome after
prolonged convulsive seizures in 186 children: low morbidity,
no mortality. Dev Med Child Neurol. 2004;46:48
11. Scott RC. The acute management of seizures in childhood. Hosp
Med. 2004;65:748752
12. Bleck TP. Management approaches to prolonged seizures and
status epilepticus. Epilepsia. 1999;40(suppl 1):S59S63; discus-
sion S64S66
13. Idro R, Aketch S, Gwer S, Newton CR, Maitland K. Research
priorities in the management of severe Plasmodium falciparum
malaria in children. Ann Trop Med Parasitol. 2006;100:95108
14. Ogutu BR, Newton CR, Crawley J, et al. Pharmacokinetics and
anticonvulsant effects of diazepam in children with severe
falciparum malaria and convulsions. Br J Clin Pharmacol. 2002;
53:4957
15. Appleton R, Sweeney A, Choonara I, Robson J, Molyneux E.
Lorazepam versus diazepam in the acute treatment of epileptic
seizures and status epilepticus. Dev Med Child Neurol. 1995;37:
682688
16. Scott RC, Neville BG. Pharmacological management of convul-
sive status epilepticus in children. Dev Med Child Neurol. 1999;
41:207210
17. Crawley J, Waruiru C, Mithwani S, et al. Effect of phenobar-
bital on seizure frequency and mortality in childhood cerebral
malaria: a randomised, controlled intervention study. Lancet.
2000;355:701706
18. Wallace SJ. Nasal benzodiazepines for management of acute
childhood seizures? Lancet. 1997;349:222
19. 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;353:623626
20. Baysun S, Aydin OF, Atmaca E, Gu rer YK. A comparison of
buccal midazolam and rectal diazepam for the acute treatment
of seizures. Clin Pediatr (Phila). 2005;44:771776
21. McIntyre J, Robertson S, Norris E, et al. Safety and efcacy of
buccal midazolam versus rectal diazepam for emergency treat-
ment of seizures in children: a randomised controlled trial.
Lancet. 2005;366:205210
22. Pramar YV, Loucas VA, el-Rachidi A. Stability of midazolam
hydrochloride in syringes and i.v. uids. Am J Health Syst
Pharm. 1997;54:913915
23. Smith FM, Nuessle NO. Stability of diazepam injection repack-
aged in glass unit-dose syringes. Am J Hosp Pharm. 1982;39:
16871690
24. Gottwald MD, Akers LC, Liu PK, et al. Prehospital stability of
diazepam and lorazepam. Am J Emerg Med. 1999;17:333337
25. Asindi AA, Ekanem EE, Ibia EO, Nwangwa MA. Upsurge of
PEDIATRICS Volume 121, Number 1, January 2008 e63
at Indonesia:AAP Sponsored on May 28, 2014 pediatrics.aappublications.org Downloaded from
malaria-related convulsions in a paediatric emergency room
in Nigeria: consequence of emergence of chloroquine-
resistant Plasmodium falciparum. Trop Geogr Med. 1993;45:
110113
26. Reading R. Managing medication in schools. Arch Dis Child.
2005;90:12531255
27. Wilson MT, Macleod S, ORegan ME. Nasal/buccal midazolam
use in the community. Arch Dis Child. 2004;89:5051
28. Ahmad S, Ellis JC, Kamwendo H, Molyneux E. Efcacy and
safety of intranasal lorazepam versus intramuscular paralde-
hyde for protracted convulsions in children: an open random-
ised trial. Lancet. 2006;367:15911597
29. Goyal M, Wiznitzer M. Emergency management of seizures in
children. Lancet. 2006;367:15551556
30. Greenblatt DJ, Divoll M. Diazepam versus lorazepam: relation-
ship of drug distribution to duration of clinical action. Adv
Neurol. 1983;34:487491
31. Blumer JL. Clinical pharmacology of midazolam in infants and
children. Clin Pharmacokinet. 1998;35:3747
32. Byatt CM, Lewis LD, Dawling S, Cochrane GM. Accumulation
of midazolam after repeated dosage in patients receiving me-
chanical ventilation in an intensive care unit. Br Med J (Clin Res
Ed). 1984;289:799800
33. de Wildt SN, de Hoog M, Vinks AA, van der Giesen E, van den
Anker JN. Population pharmacokinetics and metabolism of
midazolam in pediatric intensive care patients. Crit Care Med.
2003;31:19521958
34. Wattanagoon Y, Srivilairit S, Looareesuwan S, White NJ. Con-
vulsions in childhood malaria. Trans R Soc Trop Med Hyg. 1994;
88:426428
e64 MPIMBAZA et al
at Indonesia:AAP Sponsored on May 28, 2014 pediatrics.aappublications.org Downloaded from
DOI: 10.1542/peds.2007-0930
2008;121;e58 Pediatrics
Byarugaba
Arthur Mpimbaza, Grace Ndeezi, Sarah Staedke, Philip J. Rosenthal and Justus
Prolonged Seizures in Ugandan Children: A Randomized Clinical Trial
Comparison of Buccal Midazolam With Rectal Diazepam in the Treatment of

Services
Updated Information &
ml
http://pediatrics.aappublications.org/content/121/1/e58.full.ht
including high resolution figures, can be found at:
References
ml#ref-list-1
http://pediatrics.aappublications.org/content/121/1/e58.full.ht
at:
This article cites 33 articles, 10 of which can be accessed free
Citations
ml#related-urls
http://pediatrics.aappublications.org/content/121/1/e58.full.ht
This article has been cited by 4 HighWire-hosted articles:
Subspecialty Collections
_disorders_sub
http://pediatrics.aappublications.org/cgi/collection/neurologic
Neurologic Disorders
_sub
http://pediatrics.aappublications.org/cgi/collection/neurology
Neurology
the following collection(s):
This article, along with others on similar topics, appears in
Permissions & Licensing
ml
http://pediatrics.aappublications.org/site/misc/Permissions.xht
tables) or in its entirety can be found online at:
Information about reproducing this article in parts (figures,
Reprints
http://pediatrics.aappublications.org/site/misc/reprints.xhtml
Information about ordering reprints can be found online:
rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.
Grove Village, Illinois, 60007. Copyright 2008 by the American Academy of Pediatrics. All
and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk
publication, it has been published continuously since 1948. PEDIATRICS is owned, published,
PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly
at Indonesia:AAP Sponsored on May 28, 2014 pediatrics.aappublications.org Downloaded from