Seminars in Fetal & Neonatal Medicine xxx (2017) 1e7

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier.com/locate/siny

The use of phenobarbital and other anti-seizure drugs in newborns

Mohamed El-Dib a, *, Janet S. Soul b
a
Neonatal Neurocritical Care, Department of Pediatric Newborn Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA
b
FetaleNeonatal Neurology Program, Department of Neurology, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA

a b s t r a c t
Keywords: Neonatal seizures constitute the most frequent presenting neurologic sign encountered in the neonatal
Anticonvulsants intensive care unit. Despite limited efficacy and safety data, phenobarbital continues to be used near-
Neonatal seizures
universally as the first-line anti-seizure drug (ASD) in neonates. The choice of second-line ASDs varies
EEG
Hypoxiceischemic encephalopathy
by provider and institution, and is still not supported by sufficient scientific evidence. In this review, we
discuss the available evidence supporting the efficacy, mechanism of action, potential adverse effects, key
pharmacokinetic characteristics such as interaction with therapeutic hypothermia, logistical issues, and
rationale for use of neonatal ASDs. We describe the widely used neonatal ASDs, namely phenobarbital,
phenytoin, midazolam, and levetiracetam, in addition to potential ASDs, including lidocaine, topiramate,
and bumetanide.
© 2017 Published by Elsevier Ltd.

1. Introduction discussed in more detail below, other controversies persist. The

Neonatal seizures affect 1 to 3.5 per 1000 live-born newborns,
and their most frequent etiologies are hypoxiceischemic enceph-
alopathy (HIE), ischemic stroke, and intracranial hemorrhage [1e3].
Management of neonatal seizures varies widely, due in part to the
lack of sufficient efficacy and safety data on available ASDs, in
addition to evidence suggesting that some ASDs may have delete-
rious effects on brain development. In fact, no ASD has been
approved by the US Food and Drug Administration for use in
neonates.
In the absence of strong scientific evidence, most providers use
the same ASDs they have used for decades. Some clinicians employ
new ASDs in neonates after successful use in older infants and
children. In a recent US retrospective report of 9134 neonates with
seizures, the use of phenobarbital over the last decade has slightly
declined but is still used in 96% of neonates with seizures, largely as
a first-line agent. Whereas the use of phenytoin has significantly
declined, the use of levetiracetam has increased ten-fold [4].
question of whether continuous electroencephalographic (cEEG)
monitoring is needed to detect seizures has been settled by
numerous studies showing the high incidence of subclinical sei-
zures, and the difficulty of determining whether a clinical event is a
seizure [5]. A second controversy concerns the treatment of sub-
clinical (electrographic-only) seizures. Basic science data show that
electrographic-only seizures contribute to neonatal brain injury
[6,7]. In human newborns, one small study of neonates randomized
to receive ASD for EEG-proven seizures showed lower seizure
burden and better neurodevelopmental outcome at 18e24 months
than among those treated for only clinical seizures [8]. A third
controversy is the duration of ASD treatment. For many clinicians,
the duration of ASD therapy depends largely on the etiology, but
there is no consensus on ASD treatment duration in newborns with
acute symptomatic etiologies, whose seizures often subside within
days of onset [9].
3. Choice of anti-seizure drugs

2. Controversies in the treatment of neonatal seizures
Apart from the controversy regarding ASD choice, which is
* Corresponding author. Department of Pediatric Newborn Medicine, Brigham
and Women's Hospital, Harvard Medical School, 75 Francis Street, CWN-418, Bos-
ton, MA 02115, USA.
E-mail address: MEl-Dib@bwh.harvard.edu (M. El-Dib).
Of course, clinicians prefer the most efficacious ASD with the
least adverse effects. Unfortunately, currently available data
regarding efficacy and safety are inadequate. Studies on neonatal
ASD efficacy need to be examined carefully with specific questions
in mind regarding the methodology and results. What definition of
efficacy was used: cessation of seizures or decreased seizure
burden, and, if the latter, what magnitude of seizure reduction?
How long was that effect maintained? Was there a control group in

http://dx.doi.org/10.1016/j.siny.2017.07.008
1744-165X/© 2017 Published by Elsevier Ltd.

Please cite this article in press as: El-Dib M, Soul JS, The use of phenobarbital and other anti-seizure drugs in newborns, Seminars in Fetal &
Neonatal Medicine (2017), http://dx.doi.org/10.1016/j.siny.2017.07.008
2 M. El-Dib, J.S. Soul / Seminars in Fetal & Neonatal Medicine xxx (2017) 1e7
the study? Was the medication used as first-line therapy or second-
and third-line therapies? Seizures from acute symptomatic etiol-
ogies typically have a crescendoedecrescendo pattern [10].
Therefore, a medication used late in the course of acute seizures
(e.g. as second or third line drug) may be misinterpreted as having
good efficacy, when in fact the seizures are subsiding naturally.
Another caveat in some of the studies reported in literature is the
use of cEEG monitoring. Without cEEG for detection of seizures at
baseline, as well as determination of drug response, efficacy cannot
be determined with certainty as clinical events might not have
been seizures. Moreover, medications may cause electroclinical
uncoupling with persistence of the electrographic component
without clinical manifestations. To measure efficacy accurately,
drugs need to be tested early in the course of seizures, compared
with a control group, and their effect monitored by cEEG. Finally,
drug efficacy ideally should be associated with improved long-term
neurodevelopmental outcome and decreased risk of later epilepsy.
Published data on neonatal ASDs should be interpreted cautiously
after considering these factors.
4. Effect of hypothermia
Therapeutic hypothermia is standard of care in neonatal HIE, the
most frequent cause of neonatal seizures, and needs to be consid-
ered with regard to its interaction with ASDs. Data indicate that
hypothermia itself is likely to reduce seizure burden [11,12]. In
addition, hypothermia may alter ASD pharmacodynamics and
pharmacokinetics. Hypothermia mainly affects biotransformation
of drugs via reducing the activity of the hepatic cytochrome P450
(CYP450) enzymes, and it can affect enzymatic conjugation. The
effect of hypothermia is exaggerated in cases with hypo-
xiceischemic (HI)-induced hepatic and renal dysfunction hindering
drug metabolism and elimination [13,14]. Similar concerns are
evident during rewarming, but in the opposite direction.
Rewarming could be associated with increased metabolism and
accelerated elimination of drugs leading to significant decrease in
drug concentrations [14].
5. Widely used anti-seizure drugs
In this review, we discuss widely used ASDs, namely pheno-
barbital, phenytoin, midazolam, and levetiracetam, and other po-
tential ASDs for newborns, including lidocaine, topiramate and
bumetanide. For each of these medications, we discuss mechanism
of action, rationale of use, potential adverse effects, logistical issues,
and any interaction with hypothermia. A summary of efficacy,
adverse effects, effect of hypothermia and logistical issues is pre-
sented in Table 1 [15e46].
5.1. Phenobarbital
Phenobarbital has been used as the standard first-line of therapy
for neonatal seizures for decades, and is the drug most frequently
used by providers worldwide [4,47e49]. This is likely attributed to
available efficacy data, predictable pharmacokinetics with long
half-life making it feasible to monitor, and likely most of all, long
experience in its use.
Phenobarbital is metabolized by CYP2C19, which may be
depressed by therapeutic hypothermia. Although factors affecting
renal or hepatic elimination, that may occur in neonatal encepha-
lopathy, could also affect phenobarbital metabolism, the pharma-
cokinetics of phenobarbital are not generally altered by
hypothermia [23,24], so its dosing may not need adjustment for
hypothermia.
The most frequently adverse effect of phenobarbital is central
Please cite this article in press as: El-Dib M, Soul JS, The use of phenobarbital and other anti-seizure drugs in newborns, Seminars in Fetal &
Neonatal Medicine (2017), http://dx.doi.org/10.1016/j.siny.2017.07.008
nervous system (CNS) depression and potential for respiratory
depression requiring additional support. Moreover, there is evi-
dence from multiple animal studies that phenobarbital, as well as
other ASDs, induces apoptosis in rodent neurons. This apoptosis
involves the cortex, hypothalamus, thalamus, and basal ganglia as
well as developing white matter [18e20]. Of note, the average dose
of phenobarbital used in these animal studies was 75 mg/kg, which
significantly exceeds doses typically used in neonates. In addition,
both phenobarbital and phenytoin have been shown to disrupt the
synaptic maturation of neonatal rat brain and impair behavior [21].
Long-term effects in rats exposed to similarly very high loading
doses of phenobarbital include schizophrenia-like behavioral ab-
normalities, and impaired learning, memory, and social interaction
[22]. It is difficult to determine whether phenobarbital has similar
deleterious effects on human neonates at typical doses.
Phenobarbital has been shown to be efficacious, at least
partially, in controlling seizures. In a randomized, crossover trial
conducted by Painter et al., when 59 neonates with EEG-confirmed
neonatal seizures were randomized to receive either phenobarbital
or phenytoin, phenobarbital controlled seizures (80% reduction in
severity) in 43% of patients, which increased to 57% when both
phenobarbital and phenytoin were used in combination [15]. In
another small trial using EEG diagnosis of seizures, 50% of patients
responded (complete cessation or 80% decrease in burden of sei-
zures) to loading doses of phenobarbital up to 40 mg/kg [16]. In a
recent retrospective report, phenobarbital was completely effective
in 62.6% and partially effective in 16.5% of neonates to eliminate
both clinical and electrographic seizures; this report did not include
newborns with status epilepticus [17]. Animal studies suggested
that use of phenobarbital during therapeutic hypothermia after HIE
insult could be neuroprotective [50]. However, phenobarbital has
not been shown to improve neurologic outcome in human studies
[51,52]. Additionally, a recent Cochrane review reported insufficient
evidence to support use of prophylactic phenobarbital following
neonatal encephalopathy [53].
Neonates receiving phenobarbital (or any other ASD) are already
at high risk of developmental delay and behavioral impairment
because of both the primary hypoxiceischemic injury and/or the
effect of repeated seizures on the developing brain. It remains
unknown whether phenobarbital worsens or potentially improves
long-term developmental outcome, when effective in controlling
neonatal seizures.
5.2. Phenytoin
Traditionally, phenytoin/fosphenytoin has been the most widely
used second-line ASD for neonatal seizures, with fosphenytoin
preferred over phenytoin due to reduced adverse effects such as
local irritation. However, in a recent survey, the use of phenytoin
has declined [4]. Adverse effects of phenytoin include arrhythmia,
hypotension, and CNS depression [25]. Large doses of phenytoin of
50 mg/kg (far exceeding the typical 20 mg/kg loading dose in
humans) was shown to cause similar apoptosis and synaptic
disruption in the developing brain of different animal models as did
phenobarbital [20,21].
Hypothermia decreases the activities of both CYP2C9 and
CYP2C19 that metabolize phenytoin. In children with traumatic
brain injury, increased phenytoin concentrations extended beyond
the period of therapeutic hypothermia, warranting careful and
prolonged monitoring of phenytoin concentrations [26]. Although
the pharmacokinetics of phenytoin during neonatal therapeutic
hypothermia have not been studied, it is possible that this combi-
nation could be associated with prolonged clearance and
bradycardia.
Available data show that phenytoin is about as effective as
 M. El-Dib, J.S. Soul / Seminars in Fetal & Neonatal Medicine xxx (2017) 1e7 3

Table 1
Comparison of efficacy, adverse effects and logistical issues among widely used anti-seizure drugs in neonates.

Anti-seizure drug Efficacy Adverse effects Effect of hypothermia Logistical issues

Phenobarbital Partial efficacy data.
As first-line drug, may control
seizures in up to 43e63% of
newborns [15e17]
Phenytoin Partial efficacy data.
As first-line drug may control
seizures in up to 45% of
newborns [15]
Levetiracetam Limited retrospective data.
As second- or third-line drug,
may be effective in up to 32%
[28]
Midazolam Very limited clinical data.
Retrospective reports describe
0e100% efficacy as second- or
third-line drug [16,32,33]
Lidocaine Limited retrospective data.
Retrospective reports describe
efficacy of 20e91% as second-
or third-line drug [16,33,37,38]
Topiramate Very limited human data but
promising animal data.
Case series describe efficacy but
total n ¼ 15 [41e43]
Bumetanide Promising experimental data.
Limited human data from one
small trial [46], awaiting data
from a larger RCT
(NCT00830531)
May cause CNS depression and
respiratory depression. Animal
studies report neuronal
apoptosis, altered synaptic
maturation, long-term
behavior, learning, memory and
social interaction [18e22]
Hypotension and CNS
depression [25].
Apoptosis and synaptic
disruption in animal models
[20,21]
Few known adverse effects,
including somnolence [29].
Conflicting animal data but
some showed increased brain
injury and apoptosis [30,31]
Possible increased length of
stay, mortality, and poor short-
term neurological outcome
[34]. Possible apoptosis, similar
to other benzodiazepines [18
e20]
Risk of arrhythmias,
particularly with cardiac
dysfunction, hyperkalemia, and
phenytoin [39,40]
Anorexia, metabolic acidosis,
lethargy, irritability, and
cognitive dulling reported in
older children [44]
Potential adverse effects
include fluid and electrolyte
imbalance, hemodynamic
instability, nephrocalcinosis,
azotemia, and ototoxicity
Minimal effect; dose does not
need to be adjusted [23,24]
Hypothermia might decrease
clearance and augment
bradycardic effect [26]
Likely not affected by
hypothermia
No effect with limited data in
neonates [35,36]
Hypothermia decreases
clearance so modified dosing is
needed [37]
Increased half-life and
concentration [45]
Unknown
Long half-life may be increased
with hepatic dysfunction
common with systemic HI
insult, therefore dose may need
to be decreased
Poor enteral bioavailability (IV
use preferred) and non-linear
pharmacokinetics with variable
elimination [27]
Available in both IV and PO
formulation
Administered as continuous IV
infusion
Administered as continuous IV
infusion. Should be avoided
with cardiac dysfunction or
phenytoin administration
IV formulation not
commercially available
IV formulation available, and
pharmacokinetic data available
but effective dose is not known

CNS, central nervous system; IV, intravenous; PO, per oral; RCT, randomized controlled trial.

phenobarbital in treating neonatal seizures, as evidenced by the assumption has not been studied. Adverse effects such as mild
only randomized trial comparing these two ASDs, in which sedation, feeding difficulty, mild apnea and bradycardia, and
phenytoin was effective in 45% of cases (compared to 43% with decreased urine output have been reported in neonates receiving
phenobarbital) [15]. The challenges of using phenytoin are related LEV [29,54]. However, no clinically significant adverse effects have
to poor enteral bioavailability and non-linear pharmacokinetics, been documented [28,57e59].
leading to significant variability in elimination and hence, serum In rodent models, LEV showed mixed results regarding brain
concentrations [27]. injury versus a possible neuroprotective role. A loading dose of
80 mg/kg and maintenance daily dose of 40 mg/kg for a week in rat
models of HI was associated with decreased apoptosis suggestive of
5.3. Levetiracetam
a neuroprotective role [60]. This could be due its effect on
increasing expression of superoxide dismutase and glutathione
Levetiracetam (LEV) use in neonatal seizures has increased ten-
peroxidase [30]. In another HI model, LEV 50 m/kg every12 h for 3
fold over the last decade [4]. The reasons for its increased use are
days increased brain injury significantly when used with normo-
probably related to its the wide safety margin and efficacy reported
thermia, but not under hypothermic conditions [31]. Yet another
in older infants and children, availability of pharmacokinetic data,
murine model showed different dose-related effects where a low
and the availability of both parenteral and enteral formulations
dose of 7 mg/kg/day for one week did not have a neuroprotective
[54]. However, RCT data are lacking to demonstrate its efficacy or
effect, whereas a high dose of 70 mg/kg/day for one week was
safety as an ASD in neonates.
associated with apoptosis, whether used with or without hypo-
The mechanism of action of LEV is not fully known. It likely
thermia [61].
works via its effect on a synaptic vesicle protein expressed
Several small retrospective studies have reported efficacy
throughout the brain (synaptic vesicle glycoprotein 2A, SV2A),
(32e82%) of LEV mainly as a second-line treatment both in human
thereby impeding release of neurotransmitters [55]. In newborns
term and preterm neonates; however, the interpretation of these
with HIE, levetiracetam clearance increases significantly from day
studies is limited by methodological issues [28,57e59]. In small
1e7, as half-life decreases from 18.5 to 9.1 h [54]. LEV differs from
retrospective cohorts of 22 term and 12 preterm neonates, a
other ASDs in that 66% of the drug is excreted unchanged in urine
loading dose of levetiracetam was reported to result in complete
and the rest undergoes enzymatic hydrolysis to inactive metabo-
cessation of seizures in 32% and 36% of patients, respectively, with
lites without involving the hepatic cytochrome P450 [56]. This
increasing seizure reduction over time [58,59]. A larger retrospec-
might make the LEV pharmacokinetics less affected by therapeutic
tive cohort of 32 neonates with HIE who received levetiracetam,
hypothermia or hepatic dysfunction than other ASDs, but this

Please cite this article in press as: El-Dib M, Soul JS, The use of phenobarbital and other anti-seizure drugs in newborns, Seminars in Fetal &
Neonatal Medicine (2017), http://dx.doi.org/10.1016/j.siny.2017.07.008
4 M. El-Dib, J.S. Soul / Seminars in Fetal & Neonatal Medicine xxx (2017) 1e7
mostly as second line of therapy for clinical or electrographic sei-
zures, reported that seizures abated within 72 h in 84% of subjects
[57]. These data should be interpreted cautiously as most seizures
were diagnosed clinically, drug response was not monitored by
cEEG, and the vast majority of these infants were first treated with
other ASD medications. Thus it is unclear to what degree the
seizure cessation in these studies is related to LEV efficacy or nat-
ural resolution of neonatal seizures, particularly with time
advancement. In a retrospective study that used cEEG as a standard
to determine seizure burden in 23 neonates, LEV was associated
with complete cessation of seizures in only 32% of the patients
within 24 h, while being used mainly as a second- and third-line
agent [28]. Thus current evidence suggests that LEV works at best
in around 30% of patients with a slow response. An ongoing phase II
randomized trial testing levetiracetam in neonates (NCT01720667)
will hopefully provide more information on LEV efficacy and safety.
While awaiting these results, LEV may be considered as a second-
or a third-line ASD to treat neonatal seizures, and there is insuffi-
cient evidence to use it as a neuroprotective agent.
5.4. Midazolam
Midazolam is one of the benzodiazepines most widely used in
treating neonatal seizures. Like phenobarbital, its anti-seizure ef-
fect is due to its action at GABA receptors [62]. Midazolam is
metabolized by CYP3A enzyme; hence hepatic dysfunction in ne-
onates with HIE may lead to greater concentrations of midazolam
[35]. In addition, renal impairment in these patients may affect
elimination of its active metabolite. Although neonatal studies have
not documented a significant effect of hypothermia on midazolam
pharmacokinetics, these studies were limited by small sample size
[35,36].
Potential adverse effects of midazolam include sedation and
need for mechanical ventilation, particularly for midazolam infu-
sion or repeated doses. The routine use of midazolam for sedation
in the neonatal intensive care unit has been discouraged due to
concerns for increased length of hospital stay, mortality, and poor
short-term neurological outcome [34]. Although other benzodiaz-
epines are associated with apoptotic degeneration in animal
models, midazolam has not been specifically studied [18e20].
The efficacy of midazolam as a second- or a third-line therapy
has been reported to vary widely, from 0% to 100%, with data
limited by small sample size and other methodological limitations.
One retrospective study reported that all 13 neonates with EEG-
confirmed seizures, refractory to phenobarbital/phenytoin,
responded to midazolam bolus and infusion within 2 h [32].
However, this effect was not reproducible in other studies, and
might have been related to its late administration at a time of
natural resolution of seizures. In a larger retrospective cohort,
midazolam had a good response (seizure cessation for >4 h) in only
12.7% of the 161 neonates who received it as a second line of
treatment [33]. It should be noted that aEEG used in this latter
study is not reliable for detection of all (especially brief) seizures,
which further limits data on midazolam efficacy [63]. In the only
small prospective randomized controlled trial of midazolam as a
second-line treatment of phenobarbital-non-responsive seizures
detected by video-EEG monitoring, none of the six babies treated
with midazolam had significant decrease in seizure burden
(defined as <80%) in 48 h [16]. Based on available data, the use of
midazolam should be reserved for refractory neonatal seizures,
anticipating need for increased respiratory support.
5.5. Lidocaine
Lidocaine works as an ASD by inhibiting voltage-gated sodium
Please cite this article in press as: El-Dib M, Soul JS, The use of phenobarbital and other anti-seizure drugs in newborns, Seminars in Fetal &
Neonatal Medicine (2017), http://dx.doi.org/10.1016/j.siny.2017.07.008
channels, which inhibits neuronal depolarization, and has been
used largely as an intravenous infusion in newborns [64]. Although
widely used in Europe, there is very limited use of lidocaine in the
USA [65]. The main concern for its use is the risk of cardiac ar-
rhythmias, whose incidence was originally reported to be up to
4.8% [39]. A more recent report showed an incidence <2% and as
low as 0.4% when a reduced-dose regimen was adopted. However,
there is consensus that lidocaine should be avoided in the presence
of cardiac dysfunction or treatment with phenytoin/fosphenytoin
to avoid the increased risk of arrhythmias [66].
Lidocaine is largely metabolized by CYP1A2 and its clearance is
related to hepatic blood flow [40]. Since reduced hepatic flow
during hypothermia is associated with decreased clearance by 24%,
a specific dosing regimen has been suggested for this clinical sit-
uation [37].
Lidocaine has been used as a second- and a third-line ASD in
multiple, small, single-center studies with reported efficacy
ranging from 20% to 91% [16,33,37,38]. Data have been largely
derived from retrospective studies, with just two small prospective
cohorts [16,37]. When used to treat persistent seizures in spite of
phenobarbital and midazolam treatment in 22 neonates with HIE
receiving therapeutic hypothermia, it was reported to have 91%
efficacy. In that study, efficacy was defined as 80% decrease in the
burden of aEEG electrographic seizures within 4 h of administra-
tion without administration of additional rescue ASD [37]. The
limitations of this study are the use of aEEG instead of cEEG to
detect seizures and the late initiation of lidocaine following
phenobarbital and midazolam. Other reports have shown signifi-
cantly less efficacy than this study. In a large retrospective cohort,
only 37 out of the 186 (19.9%) who received lidocaine as a second
line after phenobarbital had a good response (cessation of aEEG
seizures for >4 h) [33] Lidocaine will continue to have limited
applicability in the USA because of its modest efficacy data and the
concerns for cardiac arrhythmias.
5.6. Topiramate
Although used widely as an ASD in children and adults, top-
iramate has not been used much in neonates, mainly due to the lack
of a parenteral formulation and the paucity of human neonatal data
[67]. An intravenous formulation of topiramate has been tested in
Phase 1 studies in adults [68], so there is a possibility of testing this
ASD in newborns in the future. Topiramate has multiple mecha-
nisms of action, including modulation of glutamate activity at the
a-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) and
kainate receptors in addition to the modulation of GABA activity at
GABA receptors [69,70]. Its neuroprotective role has been demon-
strated in numerous animal models for HI brain injury, stroke, re-
fractory seizures, and periventricular leukomalacia, making it a
potentially promising ASD for newborns with symptomatic sei-
zures caused by acute brain injury [44,45,71,72].
The pharmacokinetics of topiramate have been described in
infants and in neonatal animal models [73] and one human
neonatal study [74]. Topiramate is metabolized chiefly by CYP3A4,
which makes it susceptible to interactions with drugs or therapies
affecting cytochrome P450, including therapeutic hypothermia.
When pharmacokinetics were studied in 13 term neonates
receiving therapeutic hypothermia, there was an increased half-life
and concentration of topiramate compared with normothermic
neonates. However, the majority of neonates had concentrations
within the therapeutic range without apparent adverse effects [74].
Adverse effects reported in older infants and children include
anorexia, metabolic acidosis, lethargy, irritability, and cognitive
dulling [75]. Although there are no clinical data, topiramate has
potential for less apoptosis compared with other ASDs [76].
 M. El-Dib, J.S. Soul / Seminars in Fetal & Neonatal Medicine xxx (2017) 1e7
The only clinical data available are from several case series
which describe promising efficacy and safety [41e43]. These data
showed cessation of refractory seizures in three out of six patients
in two reports [42,43] and all three patients in another small report
[41]. Although promising, the extremely limited data available in
neonates means that its efficacy and safety remain unknown.
5.7. Bumetanide
Bumetanide, a loop diuretic used for decades in newborns,
blocks a sodium/potassium/chloride (NKCC1) co-transporter that is
highly expressed in immature neurons, thereby altering the chlo-
ride gradient and resulting in GABA-mediated inhibition instead of
depolarization [67].
The potential adverse effects of bumetanide include fluid and
electrolyte imbalance (hyponatremia, hypochloremia, hypo-
phosphatemia, and hypocalcemia), nephrocalcinosis, azotemia, and
ototoxicity [67]. These adverse effects have been well documented
by studies of bumetanide as a diuretic. Notably, ototoxicity was
reported in a single clinical trial of bumetanide, although no other
significant adverse events were reported [56].
The ability of bumetanide to augment the anti-seizure effect of
phenobarbital was demonstrated in in-vitro preparations [77] and
in a rat model of hypoxic neonatal seizures [78]. In addition,
bumetanide was found to augment the neuroprotective effect of
phenobarbital in HIE models with hypothermia [79]. In humans,
although a beneficial effect was demonstrated in one case report
[80], there is not enough evidence from current published data to
demonstrate safety or efficacy. An open-label Phase I/II trial was
conducted in Europe for neonates with HIE whose seizures did not
respond to a loading dose of phenobarbital. This trial aimed to
evaluate the efficacy and safety of four doses of bumetanide using
bivariate Bayesian sequential dose-escalation design. The study
was stopped prematurely after 14 babies were enrolled due to
hearing loss in three out of 11 survivors and reported lack of effi-
cacy [46], although other authors argued that this small trial did
achieve the primary efficacy endpoint with an expected rate of
ototoxicity [81]. A larger double-blind RCT with dose-escalation
design to obtain pharmacokinetic and safety data in newborns
with refractory seizures (NCT00830531) has recently completed
enrollment and results are anticipated soon. This trial aimed to test
bumetanide early in the course of seizures and used continuous
EEG monitoring and a standard therapy control group to determine
both drug response and safety. This design and the larger number of
subjects may provide more robust data regarding the safety and
utility of this medication for neonatal seizures.
6. Conclusion
Management of neonatal seizures continues to be highly
controversial and variable, largely because of insufficient data to
guide drug choice in neonates. Data regarding the efficacy and
safety of ASDs are largely driven from retrospective reports using
highly variable methodologies and definitions of efficacy. Apoptosis
reported in animal models was induced by very high doses of ASD
and has not been demonstrated in humans. ASD safety in humans
needs to be evaluated both in the neonatal period as well as in long-
term neurodevelopmental studies. In the absence of data from
well-designed clinical trials and prospective studies, ASD choice
and dosing will continue to be guided by clinical experience. Given
these limitations, ASDs should be used carefully in conjunction
with continuous EEG monitoring to guide management, and dis-
continued when the risk of recurrence of seizures declines. This
review highlights the critical need for better evidence from ongoing
and future trials to guide management of neonatal seizures.
Please cite this article in press as: El-Dib M, Soul JS, The use of phenobarbital and other anti-seizure drugs in newborns, Seminars in Fetal &
Neonatal Medicine (2017), http://dx.doi.org/10.1016/j.siny.2017.07.008
5
6.1. Practice points
 Phenobarbital continues to be the most widely used first-line
ASD in neonates, despite only partial efficacy and long-term
safety data.
 Therapeutic hypothermia may alter effects and metabolism of
ASDs, requiring adjustment of ASD dosing.
 Newer ASDs are often used to treat neonatal seizures, despite
insufficient data regarding efficacy, safety, and even
pharmacokinetics.
6.2. Research directions
 Efficacy and safety of new ASDs should be tested in rigorously
designed clinical trials.
 Short- and long-term safety of all ASDs needs further study in
newborn infants.
Funding sources
None.
Conflict of interest statement
None declared.
References
[1] Lanska MJ, Lanska DJ. Neonatal seizures in the United States: results of the
national hospital discharge survey, 1980e1991. Neuroepidemiology 1996;15:
117e25.
[2] Glass HC, Pham TN, Danielsen B, Towner D, Glidden D, Wu YW. Antenatal and
intrapartum risk factors for seizures in term newborns: a population-based
study, California 1998e2002. J Pediatr 2009;154. 24e28.e1.
[3] Glass HC, Shellhaas RA, Wusthoff CJ, Chang T, Abend NS, Chu CJ, et al.
Contemporary profile of seizures in neonates: a prospective cohort study.
J Pediatr 2016;174. 98e103.e1.
[4] Ahmad KA, Desai SJ, Bennett MM, Ahmad SF, Ng YT, Clark RH, et al. Changing
antiepileptic drug use for seizures in US neonatal intensive care units from
2005 to 2014. J Perinatol 2016.
[5] Murray DM, Boylan GB, Ali I, Ryan CA, Murphy BP, Connolly S. Defining the
gap between electrographic seizure burden, clinical expression and staff
recognition of neonatal seizures. Archs Dis Childh Fetal Neonatal Ed 2008;93:
F187e91.
[6] Wasterlain CG. Recurrent seizures in the developing brain are harmful. Epi-
lepsia 1997;38:728e34.
[7] Stafstrom CE, Chronopoulos A, Thurber S, Thompson JL, Holmes GL. Age-
dependent cognitive and behavioral deficits after kainic acid seizures. Epi-
lepsia 1993;34:420e32.
[8] Srinivasakumar P, Zempel J, Trivedi S, Wallendorf M, Rao R, Smith B, et al.
Treating EEG seizures in hypoxic ischemic encephalopathy: a randomized
controlled trial. Pediatrics 2015;136:e1302e9.
[9] Fitzgerald MP, Kessler SK, Abend NS. Early discontinuation of antiseizure
medications in neonates with hypoxiceischemic encephalopathy. Epilepsia
2017;58:1047e53.
[10] Painter MJ, Scher MS, Stein AD, Armatti S, Wang Z, Gardiner JC, et al.
Phenobarbital compared with phenytoin for the treatment of neonatal sei-
zures. N Engl J Med 1999;341:485e9.
[11] Srinivasakumar P, Zempel J, Wallendorf M, Lawrence R, Inder T, Mathur A.
Therapeutic hypothermia in neonatal hypoxic ischemic encephalopathy:
electrographic seizures and magnetic resonance imaging evidence of injury.
J Pediatr 2013;163:465e70.
[12] Low E, Boylan GB, Mathieson SR, Murray DM, Korotchikova I, Stevenson NJ,
et al. Cooling and seizure burden in term neonates: an observational study.
Archs Dis Childh Fetal Neonatal Ed 2012;97:F267e72.
[13] Tortorici MA, Kochanek PM, Poloyac SM. Effects of hypothermia on drug
disposition, metabolism, and response: a focus of hypothermia-mediated al-
terations on the cytochrome P450 enzyme system. Crit Care Med 2007;35:
2196e204.
[14] Zhou J, Poloyac SM. The effect of therapeutic hypothermia on drug meta-
bolism and response: cellular mechanisms to organ function. Expert Opin
Drug Metab Toxicol 2011;7:803e16.
[15] Bartha AI, Shen J, Katz KH, Mischel RE, Yap KR, Ivacko JA, et al. Neonatal
seizures: multicenter variability in current treatment practices. Pediatr Neurol
2007;37:85e90.
6 M. El-Dib, J.S. Soul / Seminars in Fetal & Neonatal Medicine xxx (2017) 1e7
[16] Glass HC, Kan J, Bonifacio SL, Ferriero DM. Neonatal seizures: treatment
practices among term and preterm infants. Pediatr Neurol 2012;46:111e5.
[17] Hellstrom-Westas L, Boylan G, Agren J. Systematic review of neonatal seizure
management strategies provides guidance on anti-epileptic treatment. Acta
Paediatr 2015;104:123e9.
[18] Van Den Broek MPH, Groenendaal F, Toet MC, Van Straaten HLM, Van
Hasselt JGC, Huitema ADR, et al. Pharmacokinetics and clinical efficacy of
phenobarbital in asphyxiated newborns treated with hypothermia: a ther-
mopharmacological approach. Clin Pharmacokinet 2012;51:671e9.
[19] Shellhaas RA, Ng CM, Dillon CH, Barks JDE, Bhatt-Mehta V. Population phar-
macokinetics of phenobarbital in infants with neonatal encephalopathy
treated with therapeutic hypothermia. Pediatr Crit Care Med 2013;14:
194e202.
[20] Bittigau P, Sifringer M, Genz K, Reith E, Pospischil D, Govindarajalu S, et al.
Antiepileptic drugs and apoptotic neurodegeneration in the developing brain.
Proc Natl Acad Sci U. S. A 2002;99:15089e94.
[21] Forcelli PA, Kim J, Kondratyev A, Gale K. Pattern of antiepileptic drug-induced
cell death in limbic regions of the neonatal rat brain. Epilepsia 2011;52:
e207e11.
[22] Kaushal S, Tamer Z, Opoku F, Forcelli PA. Anticonvulsant drug-induced cell
death in the developing white matter of the rodent brain. Epilepsia 2016;57:
727e34.
[23] Forcelli PA, Janssen MJ, Vicini S, Gale K. Neonatal exposure to antiepileptic
drugs disrupts striatal synaptic development. Ann Neurol 2012;72:363e72.
[24] Frankel S, Medvedeva N, Gutherz S. Comparison of the long-term behavioral
effects of neonatal exposure to retigabine or phenobarbital in rats. Epilepsy
Behav 2016;57(Pt A):34e40.
[25] Painter MJ, Scher MS, Stein AD, Armatti S, Wang Z, Gardiner JC, et al.
Phenobarbital compared with phenytoin for the treatment of neonatal sei-
zures. N Engl J Med 1999;341:485e9.
[26] Boylan GB, Rennie JM, Chorley G, Pressler RM, Fox GF, Farrer K, et al. Second-
line anticonvulsant treatment of neonatal seizures: a video-EEG monitoring
study. Neurology 2004;62:486e8.
[27] Spagnoli C, Seri S, Pavlidis E, Mazzotta S, Pelosi A, Pisani F. Phenobarbital for
neonatal seizures: response rate and predictors of refractoriness. Neuro-
pediatrics 2016;47:318e26.
[28] Barks JD, Liu YQ, Shangguan Y, Silverstein FS. Phenobarbital augments hy-
pothermic neuroprotection. Pediatr Res 2010;67:532e7.
[29] Meyn Jr DF, Ness J, Ambalavanan N, Carlo WA. Prophylactic phenobarbital and
whole-body cooling for neonatal hypoxiceischemic encephalopathy. J Pediatr
2010;157:334e6.
[30] Sarkar S, Barks JD, Bapuraj JR, Bhagat I, Dechert RE, Schumacher RE, et al. Does
phenobarbital improve the effectiveness of therapeutic hypothermia in in-
fants with hypoxiceischemic encephalopathy? J Perinatol 2012;32:15e20.
[31] Young L, Berg M, Soll R. Prophylactic barbiturate use for the prevention of
morbidity and mortality following perinatal asphyxia. Cochrane Database Syst
Rev 2016;5. CD001240.
[32] Zeller B, Giebe J. Pharmacologic management of neonatal seizures. Neonatal
Netw 2015;34:239e44.
[33] Empey PE, De Mendizabal NV, Bell MJ, Bies RR, Anderson KB, Kochanek PM,
et al. Therapeutic hypothermia decreases phenytoin elimination in children
with traumatic brain injury. Crit Care Med 2013;41:2379e87.
[34] Bourgeois BF, Dodson WE. Phenytoin elimination in newborns. Neurology
1983;33:173e8.
[35] Sharpe CM, Capparelli EV, Mower A, Farrell MJ, Soldin SJ, Haas RH. A seven-
day study of the pharmacokinetics of intravenous levetiracetam in neonates:
marked changes in pharmacokinetics occur during the first week of life.
Pediatr Res 2012;72:43e9.
[36] Mruk AL, Garlitz KL, Leung NR. Levetiracetam in neonatal seizures: a review.
J Pediatr Pharmacol Ther 2015;20:76e89.
[37] Patsalos PN. Clinical pharmacokinetics of levetiracetam. Clin Pharmacokinet
2004;43:707e24.
[38] Merhar SL, Schibler KR, Sherwin CM, Meinzen-Derr J, Shi J, Balmakund T, et al.
Pharmacokinetics of levetiracetam in neonates with seizures. J Pediatr
2011;159. 152e154.e3.
[39] Venkatesan C, Young S, Schapiro M, Thomas C. Levetiracetam for the treat-
ment of seizures in neonatal hypoxic ischemic encephalopathy. J Child Neurol
2017;32:210e4.
[40] Khan O, Chang E, Cipriani C, Wright C, Crisp E, Kirmani B. Use of intravenous
levetiracetam for management of acute seizures in neonates. Pediatr Neurol
2011;44:265e9.
[41] Abend NS, Gutierrez-Colina AM, Monk HM, Dlugos DJ, Clancy RR. Levetir-
acetam for treatment of neonatal seizures. J Child Neurol 2011;26:465e70.
[42] Khan O, Cipriani C, Wright C, Crisp E, Kirmani B. Role of intravenous levetir-
acetam for acute seizure management in preterm neonates. Pediatr Neurol
2013;49:340e3.
[43] Kilicdag H, Daglioglu K, Erdogan S, Guzel A, Sencar L, Polat S, et al. The effect of
levetiracetam on neuronal apoptosis in neonatal rat model of hypoxic
ischemic brain injury. Early Hum Dev 2013;89:355e60.
[44] Komur M, Okuyaz C, Celik Y, Resitoglu B, Polat A, Balci S, et al. Neuroprotective
effect of levetiracetam on hypoxic ischemic brain injury in neonatal rats.
Childs Nerv Syst 2014;30:1001e9.
[45] Griesmaier E, Stock K, Medek K, Stanika RI, Obermair GJ, Posod A, et al.
Levetiracetam increases neonatal hypoxiceischemic brain injury under
normothermic, but not hypothermic conditions. Brain Res 2014;1556:10e8.
Please cite this article in press as: El-Dib M, Soul JS, The use of phenobarbital and other anti-seizure drugs in newborns, Seminars in Fetal &
Neonatal Medicine (2017), http://dx.doi.org/10.1016/j.siny.2017.07.008
[46] Strasser K, Lueckemann L, Kluever V, Thavaneetharajah S, Hoeber D, Bendix I,
et al. Dose-dependent effects of levetiracetam after hypoxia and hypothermia
in the neonatal mouse brain. Brain Res 2016;1646:116e24.
[47] Holmes GL, Riviello Jr JJ. Midazolam and pentobarbital for refractory status
epilepticus. Pediatr Neurol 1999;20:259e64.
[48] Welzing L, Junghaenel S, Weiss V, Roth B, Mueller C, Wiesen MHJ. Disposition
of midazolam in asphyxiated neonates receiving therapeutic hypothermia e a
pilot study. Klin Padiatr 2013;225:398e404.
[49] Van Den Broek MPH, Van Straaten HLM, Huitema ADR, Egberts T, Toet MC, De
Vries LS, et al. Anticonvulsant effectiveness and hemodynamic safety of
midazolam in full-term infants treated with hypothermia. Neonatology
2015;107:150e6.
[50] Ng E, Taddio A, Ohlsson A. Intravenous midazolam infusion for sedation of
infants in the neonatal intensive care unit. Cochrane Database Syst Rev
2017;1. Cd002052.
[51] Castro Conde JR, Hernandez Borges AA, Domenech Martinez E, Gonzalez
Campo C, Perera Soler R. Midazolam in neonatal seizures with no response to
phenobarbital. Neurology 2005;64:876e9.
[52] Weeke LC, Toet MC, Van Rooij LGM, Groenendaal F, Boylan GB, Pressler RM,
et al. Lidocaine response rate in aEEG-confirmed neonatal seizures: retro-
spective study of 413 full-term and preterm infants. Epilepsia 2016;57:
233e42.
[53] El-Dib M, Chang T, Tsuchida TN, Clancy RR. Amplitude-integrated electroen-
cephalography in neonates. Pediatr Neurol 2009;41:315e26.
[54] Donovan MD, Griffin BT, Kharoshankaya L, Cryan JF, Boylan GB. Pharmaco-
therapy for neonatal seizures: current knowledge and future perspectives.
Drugs 2016;76:647e61.
[55] Vento M, De Vries L, Alberola A, Blennow M, Steggerda S, Greisen G, et al.
Approach to seizures in the neonatal period: a European perspective. Acta
Paediatr Int J Paediatr 2010;99:497e501.
[56] Van Rooij LGM, Toet MC, Rademaker KMA, Groenendaal F, De Vries LS. Cardiac
arrhythmias in neonates receiving lidocaine as anticonvulsive treatment. Eur J
Pediatr 2004;163:637e41.
[57] Weeke LC, Schalkwijk S, Toet MC, Van Rooij LGM, De Vries LS, Van Den
Broek MPH. Lidocaine-associated cardiac events in newborns with seizures:
incidence, symptoms and contributing factors. Neonatology 2015;108:130e6.
[58] Van Den Broek MPH, Huitema ADR, Van Hasselt JGC, Groenendaal F, Toet MC,
Egberts TCG, et al. Lidocaine (lignocaine) dosing regimen based upon a pop-
ulation pharmacokinetic model for preterm and term neonates with seizures.
Clin Pharmacokinet 2011;50:461e9.
[59] Van Den Broek MPH, Rademaker CMA, Van Straaten HLM, Huitema ADR,
Toet MC, De Vries LS, et al. Anticonvulsant treatment of asphyxiated new-
borns under hypothermia with lidocaine: efficacy, safety and dosing. Archs
Dis Childh Fetal Neonat Ed 2013;98:F341e5.
[60] Hellstrom-Westas L, Westgren U, Rosen I, Svenningsen NW. Lidocaine for
treatment of severe seizures in newborn infants. I. Clinical effects and cerebral
electrical activity monitoring. Acta Paediatr Scand 1988;77:79e84.
[61] Soul J. Novel medications for neonatal seizures: bumetanide and topiramate.
J Pediatr Neurol 2009;7:85e93.
[62] Clark AM, Kriel RL, Leppik IE, White JR, Henry TR, Brundage RC, et al. Intra-
venous topiramate: safety and pharmacokinetics following a single dose in
patients with epilepsy or migraines taking oral topiramate. Epilepsia 2013;54:
1106e11.
[63] Kudin AP, Debska-Vielhaber G, Vielhaber S, Elger CE, Kunz WS. The mecha-
nism of neuroprotection by topiramate in an animal model of epilepsy. Epi-
lepsia 2004;45:1478e87.
[64] White HS, Brown SD, Woodhead JH, Skeen GA, Wolf HH. Topiramate enhances
GABA-mediated chloride flux and GABA-evoked chloride currents in murine
brain neurons and increases seizure threshold. Epilepsy Res 1997;28:167e79.
[65] Glauser TA, Miles MV, Tang P, Clark P, McGee K, Doose DR. Topiramate
pharmacokinetics in infants. Epilepsia 1999;40:788e91.
[66] Filippi L, la Marca G, Fiorini P, Poggi C, Cavallaro G, Malvagia S, et al. Top-
iramate concentrations in neonates treated with prolonged whole body hy-
pothermia for hypoxic ischemic encephalopathyEpilepsia 2009;50:2355e61.
[67] Korinthenberg R, Schreiner A. Topiramate in children with West syndrome: A
retrospective multicenter evaluation of 100 patients. J Child Neurol 2007;22:
302e6.
[68] Glier C, Dzietko M, Bittigau P, Jarosz B, Korobowicz E, Ikonomidou C. Thera-
peutic doses of topiramate are not toxic to the developing rat brain. Exp
Neurol 2004;187:403e9.
[69] Liu Y, Barks JD, Xu G, Silverstein FS. Topiramate extends the therapeutic
window for hypothermia-mediated neuroprotection after stroke in neonatal
rats. Stroke 2004;35:1460e5.
[70] Koh S, Jensen FE. Topiramate blocks perinatal hypoxia-induced seizures in rat
pupsAnn Neurol 2001;50:366e72.
[71] Cha BH, Silveira DC, Liu X, Hu Y, Holmes GL. Effect of topiramate following
recurrent and prolonged seizures during early development. Epilepsy Res
2002;51:217e32.
[72] Follett PL, Deng W, Dai W, Talos DM, Massillon LJ, Rosenberg PA, et al.
Glutamate receptor-mediated oligodendrocyte toxicity in periventricular
leukomalacia: a protective role for topiramate. J Neurosci 2004;24:4412e20.
[73] Riesgo R, Winckler MI, Ohlweiler L, Ranzan J, Becker M, Salvador S, et al.
Treatment of refractory neonatal seizures with topiramate. Neuropediatrics
2012;43:353e6.
[74] Kundak AA, Okumus N, Dilli D, Erol S, Zenciroglu A. Topiramate use in the
 M. El-Dib, J.S. Soul / Seminars in Fetal & Neonatal Medicine xxx (2017) 1e7 7

neonatal period. Pediatr Neurol 2012;46:410. author reply.
[75] Glass HC, Poulin C, Shevell MI. Topiramate for the treatment of neonatal
seizures. Pediatr Neurol 2011;44:439e42.
[76] Pressler RM, Boylan GB, Marlow, Blennow M., Chiron C., Cross J.H. N, et al.
Bumetanide for the treatment of seizures in newborn babies with hypoxic
ischaemic encephalopathy (NEMO): an open-label, dose finding, and feasi-
bility phase 1/2 trial. Lancet Neurol 2015;14:469e77.
[77] Dzhala VI, Brumback AC, Staley KJ. Bumetanide enhances phenobarbital effi-
cacy in a neonatal seizure model. Ann Neurol 2008;63:222e35.
[78] Cleary RT, Sun H, Huynh T, Manning SM, Li Y, Rotenberg A, et al. Bumetanide
enhances phenobarbital efficacy in a rat model of hypoxic neonatal seizures.
PLoS One 2013;8:e57148.
[79] Liu Y, Shangguan Y, Barks JDE, Silverstein FS. Bumetanide augments the
neuroprotective efficacy of phenobarbital plus hypothermia in a neonatal
hypoxiaeischemia model. Pediatr Res 2012;71:559e65.
[80] Kahle KT, Barnett SM, Sassower KC, Staley KJ. Decreased seizure activity in a
human neonate treated with bumetanide, an inhibitor of the Na(þ)-K(þ)-
2Cl(e) cotransporter NKCC1. J Child Neurol 2009;24:572e6.
[81] Thoresen M, Sabir H. Epilepsy: neonatal seizures still lack safe and effective
treatment. Nat Rev Neurol 2015;11:311e2.

Please cite this article in press as: El-Dib M, Soul JS, The use of phenobarbital and other anti-seizure drugs in newborns, Seminars in Fetal &
Neonatal Medicine (2017), http://dx.doi.org/10.1016/j.siny.2017.07.008