Dalziel et al.

BMC Pediatrics (2017) 17:152

DOI 10.1186/s12887-017-0887-8

STUDY PROTOCOL Open Access

A multicentre randomised controlled trial

of levetiracetam versus phenytoin for
convulsive status epilepticus in children
(protocol): Convulsive Status Epilepticus
Paediatric Trial (ConSEPT) - a PREDICT study
Stuart R. Dalziel1,2*, Jeremy Furyk3,4, Megan Bonisch1, Ed Oakley5,6,7, Meredith Borland8, Jocelyn Neutze9,
Susan Donath6, Cynthia Sharpe1, Simon Harvey5, Andrew Davidson5, Simon Craig10, Natalie Phillips11,
Shane George12,13,14, Arjun Rao15, Nicholas Cheng16, Michael Zhang17, Kam Sinn18, Amit Kochar19,
Christine Brabyn20, Franz E. Babl5,6,7 and On Behalf of the PREDICT research network21

Abstract
Background: Convulsive status epilepticus (CSE) is the most common life-threatening childhood neurological
emergency. Despite this, there is a lack of high quality evidence supporting medication use after first line
benzodiazepines, with current treatment protocols based solely on non-experimental evidence and expert opinion.
The current standard of care, phenytoin, is only 60% effective, and associated with considerable adverse effects.
A newer anti-convulsant, levetiracetam, can be given faster, is potentially more efficacious, with a more tolerable
side effect profile. The primary aim of the study presented in this protocol is to determine whether intravenous (IV)
levetiracetam or IV phenytoin is the better second line treatment for the emergency management of CSE in
children.
Methods/Design: 200 children aged between 3 months and 16 years presenting to 13 emergency departments in
Australia and New Zealand with CSE, that has failed to stop with first line benzodiazepines, will be enrolled into this
multicentre open randomised controlled trial. Participants will be randomised to 40 mg/kg IV levetiracetam infusion
over 5 min or 20 mg/kg IV phenytoin infusion over 20 min. The primary outcome for the study is clinical cessation
of seizure activity five minutes following the completion of the infusion of the study medication. Blinded
confirmation of the primary outcome will occur with the primary outcome assessment being video recorded and
assessed by a primary outcome assessment team blinded to treatment allocation. Secondary outcomes include:
Clinical cessation of seizure activity at two hours; Time to clinical seizure cessation; Need for rapid sequence
induction; Intensive care unit (ICU) admission; Serious adverse events; Length of Hospital/ICU stay; Health care costs;
Seizure status/death at one-month post discharge.
Discussion: This paper presents the background, rationale, and design for a randomised controlled trial comparing
levetiracetam to phenytoin in children presenting with CSE in whom benzodiazepines have failed. This study will
provide the first high quality evidence for management of paediatric CSE post first-line benzodiazepines.
(Continued on next page)

* Correspondence: sdalziel@[Link]
1
Starship Childrens Hospital, Private Bag 92024, Auckland 1142, New Zealand
2
Liggins Institute, University of Auckland, Auckland, New Zealand
Full list of author information is available at the end of the article

The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License ([Link] which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
([Link] applies to the data made available in this article, unless otherwise stated.
Dalziel et al. BMC Pediatrics (2017) 17:152 Page 2 of 9

(Continued from previous page)

Trial registration: Prospectively registered with the Australian and New Zealand Clinical Trial Registry (ANZCTR):
ACTRN12615000129583 (11/2/2015). UTN U11111144-5272. ConSEPT protocol version 4 (12/12/2014).
Keywords: Convulsive status epilepticus, Paediatrics, Emergency medicine, Levetiracetam, Phenytoin, Intervention
study, Randomised controlled trial

Background seizures is to resolve spontaneously by 35 min, with

Convulsive status epilepticus (CSE) is the most common those not doing so requiring medication for termination
life-threatening childhood neurological emergency [1]. It [11, 12]. This revised definition has been adopted by re-
has an annual incidence of 1723 cases per 100,000 chil- cent trials on paediatric CSE [13]. Early medication use
dren per year, with 22% of patients requiring Rapid Se- and cessation of seizures in CSE is important. There is a
quence Induction (RSI) and Intensive Care Unit (ICU) wealth of animal evidence suggesting that longer sei-
admission [2]. Mortality following paediatric CSE is re- zures are harmful and result in irreversible brain damage
ported at 35% and neurological sequelae occur in up to and poorer outcomes [1].
34% of children [3]. A survey of attending Paediatric Emergency Physi-
Management guidelines for paediatric CSE recom- cians in Australia and New Zealand confirmed that
mend early and prompt use of anticonvulsant medica- benzodiazepines are universally recommended for first
tion [4, 5]. Recommendations from the Advanced Life line treatment in CSE and that 88% would use pheny-
Support Group [4], the Scottish Intercollegiate Guide- toin as a second line agent, in keeping with guideline
lines Network [5], the Status Epilepsy Working Party in recommendations. However there was a large vari-
the United Kingdom [6], and major textbooks [7, 8], are ation in third line agents, reflecting that the majority
broadly similar and universally adopt a stepwise ap- of consultants (68%) would try another agent prior to
proach to treatment; [Link] doses of benzodiazepine; RSI [14]. A retrospective review of CSE management
[Link] line anticonvulsant, with all recommending at eight large paediatric emergency departments (EDs)
phenytoin or fosphenytoin; and [Link] termination of in New Zealand and Australia over five years identi-
CSE with RSI intubation with thiopentone and ICU fied 542 patients with CSE and found phenytoin re-
admission. While there is reasonable evidence to support sulted in cessation of seizures in only 60% of the 315
the use of benzodiazepines in CSE there is a paucity of patients who received it as a second line anticonvulsant
evidence concerning the type and efficacy of second line for CSE [15]. This success rate is comparable with other
anticonvulsant medication used with management reported series [1517].
guidelines based only on expert opinion [4, 5, 9]. In addition to its less than optimal effectiveness,
The aetiology and outcomes of CSE in children is phenytoin has a number of features that make it less
different to that of adults, thus adult evidence cannot be than ideal to be used in CSE. Phenytoin is a potent
expected to be directly applicable to paediatric practice. inducer of hepatic enzymes resulting in reduced levels
A large population based study of CSE reported that less of a number of other anticonvulsants and non-
than a quarter of the children had a previous history of anticonvulsant drugs. Its adverse events include hep-
CSE. Of those with a first presentation of CSE over half atotoxicity, pancytopenia and Stevens-Johnson-Syndrome.
were previously neurologically normal, a third of epi- Phenytoin can cause cardiac arrhythmias, hypotension,
sodes were due to prolonged febrile convulsions, 17% of phlebitis, and severe soft tissue injury from extravasation
episodes were due to central nervous system (CNS) and purple glove syndrome. Because of its cardiotoxicity it
infection or acute metabolic derangement, with the has to be given slowly (1 mg/kg/min) [16, 18]. Further-
remainder of episodes idiopathic or associated with a more, phenytoin cannot be mixed with dextrose, a com-
pre-existing CNS abnormality [2]. mon component of paediatric intravenous (IV) fluids [16].
The term CSE was traditionally defined as 30 min of a In North America fosphenytoin, the prodrug of pheny-
continuous generalised tonic-clonic convulsion or recur- toin, is increasingly used instead of phenytoin [8, 16].
rent tonic-clonic convulsions without recovery of con- Although fosphenytoin can be administered more rap-
sciousness between each convulsion [10]. Recently a idly, the additional requirement to be metabolised into
revised operational definition based on the indication to the active phenytoin means that it does not offer any
commence treatment has defined CSE as seizures of a true time advantages over phenytoin. However, fosphe-
duration of five minutes or more [11]. This shortening nytoin has a number of other advantages such as the
of seizure duration for CSE definition is due to evidence ability to be administered intramuscularly and decreased
that the natural history for typical generalised convulsive infusion related adverse events, although deaths due to
Dalziel et al. BMC Pediatrics (2017) 17:152 Page 3 of 9

fosphenytoin infusions have been reported. Furthermore, Design

there are no data to show that fosphenytoin is more ef- This is a RCT comparing IV levetiracetam with IV
fective than phenytoin in stopping seizures. Importantly, phenytoin in children presenting to EDs with CSE who
for the purposes of this study, fosphenytoin is neither are still seizing after two doses of benzodiazepines. The
available nor approved for use in New Zealand and study will follow the Consolidated Standards of Report-
Australia. ing Trials (CONSORT) guidelines.
Newer antiepileptic drugs such as levetiracetam, val-
proate and lacosamide have been proposed [9, 16, Participants
17], and have been reported as effective in case re- 200 children aged between 3 months and 16 years pre-
ports and small case series in adults and in children senting with CSE to EDs. The study is ongoing with 147
[1922]. The most promising, Levetiracetam, a broad participants enrolled as of April 2017.
spectrum, antiepileptic drug has been approved for
use for over a decade and is widely used internation- Setting
ally for maintenance seizure prophylaxis for both The study is taking place in 13 EDs in New Zealand and
focal and generalised seizure disorders in both chil- Australia that are members of the Paediatric Research in
dren and adults. An IV formulation of levetiracetam Emergency Departments International Collaborative
is available for those unable to take oral preparations (PREDICT), in New Zealand; Kids First Childrens Hos-
and appears to have an excellent safety profile includ- pital, Auckland, Waikato Hospital, Hamilton, and Star-
ing rapid IV use in children [16, 17, 20, 21, 2326]. ship Childrens Hospital, Auckland; in Australia; Princess
In adults, IV infusions of levetiracetam have been well Margaret Hospital for Children, Perth, WA, Womens
tolerated [27], including at dosages and rates of infu- and Childrens Hospital, Adelaide, SA, Royal Childrens
sion greater than recommended [28]. Hospital, Melbourne, VIC, Monash Medical Centre,
Levetiracetam has the following potential advantages Clayton, VIC, Childrens Hospital at Westmead, Sydney,
when compared to phenytoin for use in CSE. Levetirace- NSW, Sydney Childrens Hospital, Sydney, NSW, John
tam is easy to administer and can be given as a five- Hunter Hospital, Newcastle, NSW, Gold Coast Univer-
minute infusion into a peripheral IV cannula without the sity Hospital, Southport, QLD, Lady Cilento Childrens
increased risk of serious adverse events (including Hospital, Brisbane, QLD, and Townsville Hospital,
hypotension, cardiac arrhythmias, extravasation or Townsville, QLD. The annual paediatric census of the
death). Furthermore, levetiracetam is compatible with participating 13 EDs is approximately 500,000. The cen-
both dextrose and normal saline infusion and has limited tral site for the study is Starship Childrens Hospital,
drug interactions. Auckland, New Zealand.
On the basis of efficacy from the limited cohort data
with levetiracetam, and concerns around low pheny- Time frame
toin efficacy and serious adverse events, IV levetirace- Three years.
tam is being increasingly used as a second line
anticonvulsant in CSE in children. However, good Interventions
quality evidence for IV levetiracetam use in CSE is Participants will be administered 40 mg/kg IV levetirace-
lacking, and now is an ideal opportunity to compare it tam infusion over 5 min (100 mg/ml levetiracetam (Kep-
to phenytoin, the current recommended standard of pra, UCB Pharma), maximum 3 g, diluted 1:1 with 0.9%
care, in the robust environment of a randomised sodium chloride to a minimum volume of 10 ml) or
controlled trial (RCT). 20 mg/kg IV phenytoin infusion over 20 min (50 mg/ml
phenytoin (DBL Phenytoin, Hameln Pharmaceuticals)
Methods/Design maximum 1 g, diluted 1:4 with 0.9% sodium chloride to
Aim a minimum volume of 20 ml). The primary outcome is
The primary aim of the study is to determine whether assessed 5 min following the end of the study interven-
IV levetiracetam or IV phenytoin is the better second tion infusion. However, if seizures persist the alternative
line treatment for the emergency management of CSE in medication will be administered; IV phenytoin infusion
children. Specifically, we hypothesise that children if levetiracetam given first (LP regimen), IV levetirace-
treated with IV levetiracetam for CSE will do better than tam if phenytoin given first (PL regimen). See Fig. 1 for
children treated with IV phenytoin in terms of time to study protocol.
clinical cessation of seizure activity, need for RSI for on-
going seizure management, need for ICU admission, ser- Allocation concealment
ious adverse events, length of hospital stay, health care A computer generated randomisation code using block
costs, and long-term outcome. randomisation was created by a statistician independent
Dalziel et al. BMC Pediatrics (2017) 17:152 Page 4 of 9

Fig. 1 Participant flow for ConSEPT. Management indicated by dashed lines and boxes without fill will only occur if CSE is on-going at that time
point. Min = minutes, LP = Levetiracetam phenytoin, PL = Phenytoin levetiracetam, LEVE = levetiracetam, PHY = Phenytoin

to the study for each site and placed in sequentially num- second line anticonvulsants (phenytoin, levetiracetam,
bered opaque, sealed and signed, envelopes for each site phenobarbitone or paraldehyde) in the last 24 h, a man-
by central study pharmacists. Randomisation is stratified agement plan stating refractory to phenytoin, known
by site and age (5 years of age and >5 years of age). contraindication or allergy to levetiracetam or pheny-
Stratification by age is utilised to account for the different toin, CSE due to an obvious major head injury or CSE
aetiology of CSE within the paediatric age range. due to eclampsia in late pregnancy.

Inclusion criteria Primary outcome

Children aged between 3 months and 16 years of age The primary outcome for the study is clinical cessation
who are currently in CSE, following two doses of benzo- of seizure activity five minutes following the completion
diazepines (given by parents, paramedics, or hospital of the infusion of the study medication (primary efficacy
staff ), who present to a study ED. CSE is defined as a outcome). As study medications have different optimal
child who is unresponsive with continuing abnormality infusion rates this will be 10 min after starting study in-
of movement (increased tone or jerking) of greater than fusions in the case of levetiracetam and 25 min after
five minutes duration, or two or more recurrent convul- starting study infusions in the case of phenytoin. Blinded
sions without recovery of consciousness between convul- confirmation of the primary outcome will occur with the
sions, or three or more convulsions within the preceding primary outcome assessment being video recorded and
hour, and currently experiencing a convulsion. This def- assessed by a primary outcome assessment team blinded
inition encompasses the International League Against to treatment allocation.
Epilepsy (ILAE) seizure types of generalised tonic-clonic
convulsions, secondarily generalised tonic-clonic convul- Secondary outcomes
sions, and complex partial status epilepticus, but not ab- Secondary outcomes include: [Link] cessation of seiz-
sence, myoclonic, tonic and simple partial status ure activity at two hours following the commencement
epilepticus. of the study infusions without the need for further seiz-
ure management after the initial agent (levetiracetam or
Exclusion criteria phenytoin); [Link] cessation of seizure activity at two
Exclusion criteria include previous randomisation, regu- hours following the commencement of the study treat-
lar phenytoin or levetiracetam use, administration of ment regimen without the need for RSI or further
Dalziel et al. BMC Pediatrics (2017) 17:152 Page 5 of 9

seizure management (comparison of LP versus PL regi- Within each study sites resuscitation area opaque
mens); [Link] to clinical seizure cessation from com- study boxes for children 5 years of age and >5 years of
mencement of study treatment regimen; [Link] for RSI age are stored. When potential patients are moved to
with thiopentone for on-going seizure management after the resuscitation areas clinical staff complete a study
administration of study treatment regimen; [Link] ad- Clinical Research Form (CRF) addressing inclusion and
mission; [Link] adverse events (primary safety out- exclusion criteria for the study. If all inclusion criteria,
come) including death, airway complications, and and no exclusion criteria are present, then clinical staff
cardiovascular instability (cardiac arrest, arrhythmia and should open opaque study boxes. Boxes will be opened
hypotension requiring intervention); [Link] of Hos- at the time the second dose of benzodiazepine is given,
pital/ICU stay; [Link] care costs (total costs associated or on arrival if two doses have already been given, in
with CSE admission); [Link] status at one month post order to allow nursing staff appropriate time to draw up
discharge, or two months post randomisation (whichever the infusions of second line anticonvulsant agents.
is the earliest); [Link] at one month one post dis- Opaque study boxes contain: An opaque, sealed and
charge, or two months post randomisation (whichever is signed, envelope containing randomisation allocation
the earliest). and infusions instructions; A Timer; A video device and
instructions; Seizure charts; Study information sheet and
consent form.
Study process According to the randomisation allocation clinical staff
At all sites patients arriving in clinically diagnosed CSE will draw up and administer a levetiracetam (5-min infu-
are assigned an Australasian Triage Category score of 1, sion) or phenytoin infusion (20-min infusion) (see Fig. 1,
as per current procedure, and are immediately taken to time 0 = start of study infusion). While the first study
the resuscitation area for management. Standard seizure medication is being administered clinical staff will draw
care is initiated in accordance with each sites clinical up the alternative study medication.
practice guidelines, including establishment of IV or Five minutes following the completion of study medi-
intraosseous (IO) access. All clinical practice guidelines, cation infusion a formal assessment of seizure activity
Advanced Life Support Group guideline [4], and the will be performed by the most senior treating physician.
Scottish Intercollegiate Guidelines Network guideline This assessment is video recorded to allow blinded con-
[5], recommend two doses of benzodiazepine prior to firmation of the primary outcome. The participant will
initiation of second line seizure medications. For the be examined for the following: i)Increased tone; ii)Jerk-
purposes of the study sites can give their usual benzodi- ing movements (including nystagmoid jerking eye move-
azepine type (diazepam, lorazepam, or midazolam), ments); iii)Level of consciousness according to the Alert,
route (rectal, buccal, oral, intranasal, IV, IO, or intra- Voice, Pain, Unresponsive (AVPU) scale. Continued seiz-
muscular) and dose. Doses given by parents and/or para- ure activity is defined as presence of either increased
medic staff are regarded as an effective benzodiazepine tone or jerking movements. If seizure activity is present
dose for the purposes of the study. The minimum dose then the alternative study medication is to be infused
and route of each benzodiazepine is detailed in Table 1. (phenytoin if levetiracetam given or levetiracetam if

Table 1 Benzodiazepine dosing prior to enrolment in ConSEPT

Benzodiazepine Route Minimum dose prior to trial enrolment Recommended dose*
Diazepam IV/IO 0.1 mg/kg 0.25 mg/kg
Total dose 5 mg Max 10 mg
PR 0.1 mg/kg 0.5 mg/kg
Total dose 5 mg Max 10 mg
Midazolam IV/IO 0.1 mg/kg 0.15 mg/kg
Total dose 2 mg Max 10 mg
IM 0.1 mg/kg 0.2 mg/kg
Total dose 2 mg Max 10 mg
Buccal 0.1 mg/kg 0.5 mg/kg
Total dose 2 mg Max 10 mg
Intranasal 0.1 mg/kg 0.5 mg/kg
Total dose 2 mg Max 10 mg
Lorazepam IV/IO 0.05 mg/kg 0.1 mg/kg
Total dose 2 mg Max 4 mg
*As per Advanced Paediatric Life Support guidelines (Australia/New Zealand) [4]
IV Intravenous, IO Intraosseous, IM Intramuscular, PR Per Rectum
Dalziel et al. BMC Pediatrics (2017) 17:152 Page 6 of 9

phenytoin given). Five minutes following the completion verbally confirming the presence or absence of jerking
of the second infusion (if required) a formal assessment movements; [Link] of jerking movements by re-
of seizure activity will again be performed by the most cording the eyes of the patient approximately 10 s re-
senior treating physician. cording verbally confirming the presence or absence of
If at any stage seizure activity has ceased (as per above nystagmoid jerking eye movements; [Link] study ID.
definition) the time is recorded and participants will fin- Prior to the videos being reviewed by the blinded pri-
ish the infusion they are currently receiving. No further mary outcome assessment committee they will be
infusions will be commenced if participants remain seiz- reviewed by the study management team and edited so
ure free. If seizure activity recommences and partici- that any part of the video that confirms study medica-
pants have only received one study infusion they can be tion is removed (i.e. syringe driver with study medication
treated with the other medication if this is felt to be ap- labelled accidentally included in video recording).
propriate by the treating team.
Clinical or research staff will collect the following data: Adverse events
Demographics; Date of presentation; Date and time of An independent three member Data Monitoring Com-
onset of seizure; Benzodiazepine type, dose, route and mittee (DMC), comprised of two clinicians each with
time given; Highest recorded temperature during resus- both emergency medicine and ethics experience, and a
citation, at home or with ambulance service; Adverse biostatistician, has been established. The DMC will re-
events occurring prior to starting study medications re- ceive interim reports every 6 months of adverse events:
quiring an intervention (airway repositioning, oral or Episodes of airway repositioning, oral or nasal airway
nasal airway placement, application of positive pressure placement, application of positive pressure or ventilation
or ventilation with bag mask, tracheal intubation, fluid with bag mask, fluid boluses, and extravasation of IV/IO
bolus, chest compressions, cardiac defibrillation); Ad- fluids in the first 2 h after starting study medication in-
verse events occurring anytime in the first two hours fusions; Episodes of tracheal intubation in the first 48 h;
after starting study infusions (in addition to above aller- All episodes of chest compressions, cardiac defibrillation,
gic reaction, IV/IO access tissued, extravasation of IV in- allergic reactions, or purple glove syndrome.
fusions, purple glove syndrome, any other clinical events The following are considered Serious Adverse Events
deemed significant). (SAEs): Death; Serious airway complications in the first
Trained research nurses will visit participants daily 24 h, defined as the unexpected use of an endotracheal
while they remain in-patients and contact families one tube, LMA; and cricothyrotomy. Unexpected is defined
month following discharge collecting the following data: as the use of these interventions when it was not part of
Past medical history; Epilepsy/seizure history; Medica- a planned RSI following failure of medical management,
tion history; Background of presenting event; Family his- nor airway support required by a patient who develops a
tory; Length of stay in hospital/ICU; IV and nasogastric compromised airway secondary to seizure activity or first
fluids use; Ventilator support; Medications; Seizures; Ad- line CSE medications e.g. benzodiazepines; Cardiovascu-
verse events; Seizure classification during admission; lar instability (cardiac arrest or arrhythmia requiring
Neurological investigations. electrical cardioversion); Any other event that is a life-
threatening event. SAEs will be reported to the principal
Blinded confirmation of primary outcome investigator within 24 h, and will be reported to the
In order to increase the robustness of the primary out- chair of the DMC within 48 h. The DMC will receive an
come assessment seizure continuation or cessation is interim analysis of trial data following the recruitment
video recorded and will be independently assessed by a and follow-up of the first 100 participants. The study
blinded primary outcome assessment committee (com- will be terminated early if: [Link] DMC, with regards to
prising three study physicians, including at least one currently available evidence, following the death or car-
study ED physician and one study neurologist). At the diac arrest of a participant due to a study medication,
time of primary outcome assessment the treating team thought that the risks for individual participants out-
will record the senior treating physician assessing the weighed the benefits of continuing the study; [Link] in-
following: [Link] of tone in lower limbs (i.e. dependent DMC, with regards to currently available
flexion of bilateral ankles for clonus, or flexion of bilat- evidence, following the analysis from the first 100 partic-
eral elbows) - approximately 10 s recording verbally con- ipants, thought that the risks for individual participants
firming the presence or absence of increased tone; outweighed the benefits of continuing the study.
[Link] of jerking movements by recording the
hands of the patient (in cases of unilateral seizures the Consent and ethical considerations
affected side, in cases of predominantly lower limb sei- Due to the life threatening nature of CSE, and the need
zures the lower limbs) approximately 20 s recording for urgent timely treatment, it is not possible to gain
Dalziel et al. BMC Pediatrics (2017) 17:152 Page 7 of 9

informed consent prior to randomisation and treatment participants to be lost due to exclusions, failure to enrol,
in this study. Delayed retrospective consent can be or refusal of consent.
sought in New Zealand if consent prior to the interven- Analysis will be by intention to treat. Results from un-
tion is impracticable and/or undesirable [29] and in adjusted comparisons between groups will be reported,
Australia if prospective consent is not practicable, there together with analyses adjusted for possible imbalances
is potential benefit to the patient, risk is low, the re- between groups for results with appropriate data distri-
search has merit and there is no reason to suspect the butions. Categorical outcome variables (including the
parents would not give consent [30]. Ethics approval for primary outcome) will be compared with chi-squared
the study and the accompanying consent process has tests (unadjusted) and logistic regression (adjusted).
been granted by the four ethics committees with govern- Continuous outcome variables will be analysed using
ance for the 13 study sites. Thus written informed con- survival analysis and Cox regression. Continuous out-
sent to remain in the study is sought from parents and come variables with skewed distributions will be log-
guardians at the earliest possible time after emergency transformed. Continuous variables will be compared
stabilisation of the CSE, i.e. after seizure cessation or with unpaired t tests (unadjusted) and linear regression
seizure termination by RSI and intubation, by either (adjusted). Continuous outcomes variables with skewed
trained research or clinical staff. Data for children whose distributions after log-transformation will be compared
parents and guardians do not wish for their child to re- with Mann-Whitney tests. Differences for categorical
main in the study is destroyed, apart from demographic and unskewed continuous data will be reported as odds
data, and will not be available for data analysis. ratios (95% confidence interval (CI)) or difference be-
The use of videos during resuscitation has been stand- tween means (95% CI) respectively. Differences between
ard of care in some of the PREDICT EDs, where they log-transformed data will be reported as a ratio of geo-
have been used for resuscitation research and found to metric means (95% CI). Differences between skewed
be acceptable to families [31]. Consent to use the video continuous data will be reported as difference between
recordings is a separate item on the consent form i.e. medians (95% CI). Planned subgroup analysis will be
families can take part in the study but not have their undertaken by focal or generalised onset of CSE, febrile
childs video recordings used. If families do not consent or afebrile CSE, and type of benzodiazepine used. Sensi-
to the video recordings these are deleted immediately at tivity analyses will be undertaken using a modified
the time of consent. intention-to-treat dataset (excluding those participants
Due to the study being undertaken exclusively in randomised but in whom seizure activity stopped prior
paediatric participants informed consent is not being to the start of the first study infusion) and a per-
sought from participants, but only from their parents protocol dataset.
and guardians. De-identified data will be collected by trained research
Two members of the DMC, one in each country, are nurses, managed using REDCap electronic data capture
available to talk with the parent/guardian(s) on request tools, including data range checks, securely hosted at
if the parent/guardian(s) have concerns about the con- The University of Auckland, Auckland, New Zealand
sent process. [32], and analysed using Stata 12 (Statagroup, College
Station, Texas, USA). Starship Childrens Hospital, Auck-
land, New Zealand, is the co-ordinating centre for the
Sample size, power and statistical methods study. Study sites will be audited for data collection and
Using pilot data indicating a phenytoin seizure cessation management by the co-ordinating centre. No independ-
rate of 60% [15] a total of 91 participants will be re- ent audit of data is planned.
quired to be randomised into each arm for the study to A per-protocol analysis of efficacy will be undertaken
have at least 80% power to detect a total difference in as a sensitivity analysis. A further sensitivity analysis will
seizure cessation rates between levetiracetam and pheny- be undertaken using the blinded confirmation of the pri-
toin of 20% (alpha = 0.05). The 80% seizure cessation mary outcome data.
rate for levetiracetam that this study is powered for is at
the conservative range of seizure cessation rates re- Discussion
ported in retrospective series (75100%) [19, 21, 22]. To Limitations
allow for loss to follow-up a total of 100 participants will The primary outcome does not include electroencephal-
be randomised into each arm of the study. ography (EEG) confirmation of seizure termination.
Given that the five-year pilot study showed an average While it is possible that a number of participants may
of eight possible participants per site per year the study have the termination of seizure status misclassified fol-
will require three years to complete (8 participants 13 lowing the study infusion the primary end-point of the
sites 3 years = 312). This allows for a third of possible study is a pragmatic end-point and reflective of the real
Dalziel et al. BMC Pediatrics (2017) 17:152 Page 8 of 9

clinical world practice and clinical decision points. EEG Australia. SRDs time was part funded by the Health Research Council of New
confirmation of seizure activity is not routinely available Zealand (HRC13/556). The study sponsor is Starship Childrens Health, Private
Bag 92,024, Auckland 1142, New Zealand. Neither the funder, nor the
in any of the study sites EDs, or indeed internationally sponsor, have any role in study design, collection, data management,
in EDs. analysis, interpretation of data, proposed writing of reports or decision to
In addition, the lack of EEG confirmation of seizure submit for publication. These tasks are the responsibility of the study
investigators.
activity may possibly result in some pseudo seizures or
seizure mimics enrolled in the study. In reality this is Availability of data and material
very unlikely and if such conditions were to be enrolled Not applicable. This is a protocol manuscript, not a research findings report.
All investigators will have access to the final dataset. Following the
the presence of randomisation will make the effect min- completion of the study results will be disseminated via medical conferences
imal on the overall study results. and publication in a peer reviewed medical journal with authorship
Those assessing the primary end-point are not blinded determined according to International Committee of Medical Journal Editors
(ICMJE) criteria. There are no current plans to use professional writers or to
to the assigned intervention group and it is possible that make the final participant-level dataset publically available. Participants, who
this lack of blinding could introduce bias. As the two wished to be informed of results, will be provided with a summary of the re-
study interventions have different optimal infusion search findings at the time of publication.

times, manufacturing presentations (vials and ampoules), Authors contributions

and due to manufacturing technical difficulties related to The PREDICT network was responsible for identifying the research question.
phenytoins high pH we could not instigate a blinded SRD designed the study. SRD, FEB, EO, MB, JN, CS, SH, SD, AD, MB refined the
study design and developed the research protocol. All authors contributed
study. However, due to the life threatening nature of to the development of the protocol, the implementation of the study at
CSE it is unlikely that a physician would report that seiz- participating sites and the enrolment of patients, SRD and JF were
ure activity had terminated when in fact it had not. Fur- responsible for the drafting of this paper. All authors provided comments on
the drafts and have read and accepted the final version. SRD, FEB, EO, MB, JN
thermore, the independent video confirmation of the and SD comprise the study steering committee with responsibility for all
primary outcome assessment will also reduce this pos- aspects of the study. SRD takes responsibility for the manuscript as a whole.
sible bias.
Competing interests
The authors declare that they have no competing interests.
Time line
The study commenced recruitment in March 2015, with Consent for publication
Not applicable.
the first patient enrolled on the 19th March 2015. Re-
cruitment is expected to finish in 2018. This study is ex- Ethics approval and consent to participate
pected to provide robust evidence for second line Ethical approval for this study has been obtained from the Northern B
Health and Disability Ethics Committee, Auckland, New Zealand (13/NTB/83/
management of CSE in children. AM01), Childrens Health Services Queensland Human Research Ethics
Committee (HREC), Brisbane, Queensland, Australia (HREC/13/QRCH/167),
Abbreviations Princess Margaret Hospital for Children HREC, Perth, Western Australia,
CI: Confidence interval; CNS: Central nervous system; ConSEPT: Convulsive Australia (2013081EP), and Womens and Childrens Hospital Network HREC,
Status Epilepticus Paediatric Trial; CONSORT: Consolidated Standards of Adelaide, South Australia, Australia (HREC/13/WCHN/134), following peer
Reporting Trials; CRF: Clinical research form; CSE: Convulsive status review. All important protocol modifications will be reported to the four
epilepticus; DMC: Data Monitoring Committee; ED: Emergency department; ethics committees with governance for the 13 study sites.
EEG: Electroencephalography; ICU: Intensive Care Unit; ILAE: International
League Against Epilepsy; IO: Intraosseous; IV: Intravenous; LP: Levetiracetam
phenytoin; PL: Phenytoin levetiracetam; PREDICT: Paediatric Research in Publishers Note
Emergency Departments International Collaborative; RCT: Randomised Springer Nature remains neutral with regard to jurisdictional claims in
control trial; RSI: Rapid sequence induction; SAE: Serious adverse event published maps and institutional affiliations.

Acknowledgements Author details

1
We would like to thank participating families, ED staff and the site research Starship Childrens Hospital, Private Bag 92024, Auckland 1142, New
assistants. Zealand. 2Liggins Institute, University of Auckland, Auckland, New Zealand.
3
The Townsville Hospital, Townsville, Queensland, Australia. 4James Cook
Funding and sponsor University, Townsville, Queensland, Australia. 5Royal Childrens Hospital,
The study is funded by grants from the Health Research Council of New Melbourne, Victoria, Australia. 6Murdoch Childrens Research Institute, Victoria,
Zealand (HRC 12/525), Auckland, New Zealand; A+ Trust (Auckland District Australia. 7Department of Paediatrics, University of Melbourne, Victoria,
Health Board), Auckland, New Zealand; Queensland Emergency Medicine Australia. 8Princess Margaret Hospital, Perth, Western Australia, Australia. 9Kidz
Research Foundation, Milton, Queensland, Australia (EMPJ-105R212014- First Hospital, Auckland, New Zealand. 10Monash Medical Centre, Victoria,
FURYK); Private Practice Research and Education Trust Fund, The Townsville Australia. 11Lady Cilento Childrens Hospital, Brisbane, Queensland, Australia.
12
Hospital and Health Service, Douglas, Queensland, Australia; Eric Ormond Gold Coast University Hospital, Southport, Queensland, Australia.
13
Baker Charitable Fund, Equity Trustees, Clayton, Victoria, Australia; and University of Queensland, Brisbane, Queensland, Australia. 14Bond
Princess Margaret Hospital Foundation, Perth, Western Australia, Australia. University, Gold Coast, Queensland, Australia. 15Sydney Childrens Hospital,
The PREDICT network is supported as a Centre of Research Excellence for Randwick, New South Wales, Australia. 16Childrens Hospital at Westmead,
Paediatric Emergency Medicine by the National Health and Medical Research Sydney, New South Wales, Australia. 17John Hunter Hospital, Newcastle, New
Council, Canberra, Australian Capital Territory, Australia (NHMRC South Wales, Australia. 18Canberra Hospital, Canberra, Australian Capital
GNT1058560). The Victorian sites were supported by the Victorian Territory, Australia. 19Womens and Childrens Hospital, Adelaide, South
Governments Infrastructure Support Program, Melbourne, Victoria, Australia. Australia, Australia. 20Waikato Hospital, Hamilton, New Zealand. 21Paediatric
FEBs time was part funded by a grant from the Murdoch Childrens Research Research in Emergency Departments International Collaborative, Melbourne,
Institute and the Royal Childrens Hospital Foundation, Melbourne, Victoria, Victoria, Australia.
Dalziel et al. BMC Pediatrics (2017) 17:152 Page 9 of 9

Received: 14 November 2016 Accepted: 17 May 2017 25. Grosso S, Cordelli DM, Franzoni E, Coppola G, Capovilla G, Zamponi N, et al.
Efficacy and safety of levetiracetam in infants and young children with
refractory epilepsy. Seizure. 2007;16:34550.
26. Wheless JW, Clarke D, Hovinga CA, Ellis M, Durmeier M, McGregor A, et al.
References Rapid infusion of a loading dose of intravenous Levetiracetam with minimal
1. Novorol CL, Chin RFM, Scott RC. Outcome of convulsive status epilepticus: a dilution: a safety study. J Child Neurol. 2009;24(8):94651.
review. Arch Dis Child. 2007;92:94851. 27. Baulac M, Brodie MJ, Elger CE, Krakow K, Stockis A, Meyvisch P, et al.
2. Chin RF, Neville BG, Peckham C, Bedford H, Wade A, Scott RC. Incidence, Levetiracetam intravenous infusion as an alternative to oral dosing in
cause, and short-term outcome of convulsive status epilepticus in patients with partial-onset seizures. Epilepsia. 2007;48(3):58992.
childhood: prospective population-based study. Lancet. 2006;368(9531):222 28. Ramael S, Daoust A, Otoul C, Toublanc N, Troenaru M, Lu Z, et al.
9. Levetiracetam intravenous infusion: a randomized, placebo-controlled safety
3. Raspall-Chaure M, Chin RFM, Neville BG, Scott RC. Outcome of paediatric and pharmacokinetic study. Epilepsia. 2006;47(7):112835.
convulsive status epilepticus: a systematic review. Lancet Neurol. 2006;5: 29. National Ethics Advisory Committee. Ethical guidelines for intervention
76979. studies. Wellington: Ministry of Health; 2009.
4. Advanced Life Support Group. Advanced Paediatric Life Support: The 30. National Health and Medical Research Council, Australian Research Council,
Practical Approach. 4th ed. London: BMJ Publishing Group; 2005. Australian vice-chancellors committee. National Statement on ethical conduct
5. Scottish Intercollegiate Guidelines Network. Diagnosis and management of in human research: National Health and Medical Research Council, 2007.
epilepsies in children and young people: a national guideline. 2005. 31. Oakley E, Stocker S, Staubli G, Young S. Using video recording to identify
6. Appleton R, Choonara I, Martland T, Phillips B, Scott R, Whitehouse W. The management errors in pediatric trauma resuscitation. Pediatrics. 2006;117(3):65864.
treatment of convulsive status epilepticus in children. The status Epilepticus 32. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research
Working Party, members of the status Epilepticus Working Party. Arch Dis electronic data capture (REDCap) - a metadata-driven methodology and
Child. 2000;83(5):4159. workflow process for providing translational research informatics support. J
7. Cameron P, Jelinek G, Everitt I, Browne G, Raftos J. Tesxtbook of paediatric Biomed Inform. 2009;42(2):37781.
emergency medicine. 1st ed. Philadelphia: Churchill Livingstone; 2006.
8. Kliegman RM, Behrman RE, Jenson HB, Stanton BF. Nelson textbook of
pediatrics. 18th ed. Philadelphia: Saunders; 2007.
9. Appleton R, Macleod S, Martland T. Drug management for acute tonic-
clonic convulsions including convulsive status epilepticus in children.
Cochrane Database Syst Rev. 2008;(3):CD001905. doi:10.1002/14651858.
CD001905.pub2.
10. Epilepsy Foundation of America's Working Group on Status Epilepticus.
Recommendations of the Epilepsy Foundation of America's Working group
on status Epilepticus. JAMA. 1993;270:8549.
11. Lowenstein DH, Bleck T, Macdonald RL. It's time to revise the definition of
status epilepticus. Epilepsia. 1999;40(1):1202.
12. Lowenstein DH, Cloyd J. Out-of-hospital treatment of status epilepticus and
prolonged seizures. Epilepsia. 2007;48(Suppl 8):968.
13. Chamberlain JM, Okada P, Holsti M, Mahajan P, Brown KM, Vance C, et al.
PECARN. Lorazepam vs diazepam for pediatric status epilepticus: a
randomized controlled trial. JAMA. 2014;311(16):165260.
14. Babl F, Sheriff N, Borland M, Acworth J, Neutze J, Krieser D, et al. Emergency
management of paediatric status epilepticus in Australia and New Zealand:
practice patterns in the context of clinical practice guidelines. J Paediatr
Child Health. 2009;45(9):5416.
15. Lewena S, Pennington V, Acworth J, Thornton S, Ngo P, Mcintyre S, et al.
Emergency management of pediatric convulsive status epilepticus. Pediatr
Emerg Care. 2009;25(2):837.
16. Abend NS, Huh JW, Helfaer MA, Dlugos DJ. Anticonvulsant medications in
the pediatric emergency room and intensive care unit. Pediatr Emerg Care.
2008;24(10):70518.
17. Wheless JW, Treiman DM. The role of the newer antiepileptic drugs in the
treatment of generalized convulsive status epilepticus. Epilepsia. 2008;49(s9):
748.
18. National Patient Safety Agency. Rapid response report NPSA/2010/RRR018
Preventing fatalities from medication loading doses. London: National
Patient Safety Agency; 2010.
19. Knake S, Gruener J, Hattemer K, Klein KM, Bauer S, Oertel WH, et al.
Intravenous levetiracetam in the treatment of benzodiazepine refractory
status epilepticus. J Neurol Neurosurg Psychiatry. 2008;79:5889. Submit your next manuscript to BioMed Central
20. Michaelides C, Thibert RL, Shapiro MJ, Kinirons P, John T, Manchharam D, et and we will help you at every step:
al. Tolerability and dosing experience of intravenous levetiracetam in
children and infants. Epilepsy Res. 2008;81:1437. We accept pre-submission inquiries
21. Goraya JS, Khurana DS, Valencia I, Melvin JJ, Cruz M, Legido A, et al. Our selector tool helps you to find the most relevant journal
Intravenous Levetiracetam in children with epilepsy. Pediatr Neurol. 2008;38:
We provide round the clock customer support
17780.
22. Kirmani BF, Crisp ED, Kayani S, Rajab H. Role of intravenous levetiracetam in Convenient online submission
acute seizure management of children. Pediatr Neurol. 2009;41(1):379. Thorough peer review
23. Wheless JW. Levetiracetam in the treatment of childhood epilepsy.
Inclusion in PubMed and all major indexing services
Neuropsychiatr Dis Treat. 2007;3(4):40921.
24. Gustafson M, Ritter FJ, Frost MD, Doescher J. Review of over 400 Maximum visibility for your research
intravenous levetiracetam administrations in pediatric patients ages
newborn through 11 years of age. Epilepsia. 2007;48(s6):248380. Submit your manuscript at
[Link]/submit