YEBEH-04335; No of Pages 3

Epilepsy & Behavior xxx (2015) xxxxxx

Contents lists available at ScienceDirect

Epilepsy & Behavior

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

Comparison of the effectiveness of four antiepileptic drugs in the

treatment of status epilepticus according to four different efcacy criteria
Juliane Redecker 1, Matthias Wittstock 2, Reiner Benecke 3, Johannes Rsche
Klinik und Poliklinik fr Neurologie, Universittsmedizin Rostock, Gehlsheimer Str. 20, 18147 Rostock, Germany

a r t i c l e

i n f o

Article history:
Revised 8 April 2015
Accepted 19 April 2015
Available online xxxx
Keywords:
Status epilepticus
Phenytoin
Valproate
Levetiracetam
Lacosamide
Criteria of efcacy

a b s t r a c t
The preliminary data presented here shall give an impression on how different criteria for the identication of an
antiepileptic drug (AED) with a possible or certain treatment effect can have an inuence on the results of retrospective case series. We present a data subset from a large retrospective study which, when completed, will cover
all treatment episodes of status epilepticus (SE) at the neurological department of the Universittsmedizin
Rostock from January 2010 to June 2013. We compare and contrast the results of four different efcacy criteria
for the effectiveness of phenytoin (PHT), valproate (VPA), levetiracetam (LEV), and lacosamide (LCM): criterion
1 = the last AED administered before SE termination; criterion 2 = the last drug introduced into the antiepileptic
therapy within 72 h before SE termination and without changes in the comedication; criterion 3 = the last drug
introduced into the antiepileptic therapy or increased in dose within 24 h before SE termination without changes
in the comedication; and criterion 4 = the last drug introduced into the antiepileptic therapy within 72 h before
SE termination, even allowing changes in the comedication. Thirty-seven treatment episodes in 32 patients (13
male and 19 female, mean age at rst episode: 68 years, SD: 17) could be analyzed. In 31 episodes, at least one
AED was given intravenously. Efcacy rates in the whole case series according to all four criteria were not significantly different between the four AEDs, but there was a considerable difference in the efcacy rates of each AED
when evaluating them with the different efcacy criteria. Our data show that statistically signicant results
concerning the efcacy of different AEDs in different subtypes of SE may depend on the outcome criteria. Therefore, efcacy criteria for the effectiveness of AEDs in the treatment of SE should be standardized.
This article is part of a Special Issue entitled Status Epilepticus.
2015 Elsevier Inc. All rights reserved.

1. Introduction
Randomized controlled studies of the treatment of status epilepticus
(SE) are difcult to perform because of ethical reasons and logistical
challenges. One study concerning third-line treatment of SE was
stopped prematurely because of difculties in recruiting patients [1].
Besides several studies on rst-line treatment of SE, there is only one
prospective randomized trial concerning second-line treatment of SE.
When SE was not responsive to diazepam (DZP), valproate (VPA) as a
second-line treatment was as effective as phenytoin (PHT) [2].
Another randomized double-blind study comparing fosphenytoin,
levetiracetam (LEV), and VPA in the treatment of benzodiazepineresistant SE is under way [3]. Therefore, the evidence for treatment
Corresponding author. Tel.: +49 381 4944768; fax: +49 381 4944794.
E-mail addresses: juliane.redecker@uni-rostock.de (J. Redecker),
matthias.wittstock@med.uni-rostock.de (M. Wittstock),
reiner.benecke@med.uni-rostock.de (R. Benecke), johannes.roesche@med.uni-rostock.de
(J. Rsche).
1
Tel.: +49 381 4944769; fax: +49 381 4944794.
2
Tel.: +49 381 4944791; fax: +49 381 4944792.
3
Tel.: +49 381 4949511; fax: +49 381 4949512.

guidelines for benzodiazepine-resistant SE is mainly based on observational studies, case series, case reports, and retrospective database
analyses. There is an extensive diversity of approaches used to determine the termination drug in a treatment episode of refractory status
epilepticus. A review of topiramate (TPM) in SE [4] describes eight
different criteria for a possible or certain treatment effect of an antiepileptic drug (AED), which were different from the criteria commonly used in prospective randomized controlled trials in SE
(e.g., [5]). In another review on LEV as second-line treatment of SE
[6], seven different criteria for a treatment effect of an AED were described. The time frame for the attribution of a treatment effect to the
administration of a new AED ranged from 3 min to 72 h.
In a meta-analysis [7] of published studies concerning the relative effectiveness of lacosamide (LCM), LEV, phenobarbital (PB), PHT, and VPA
in the treatment of benzodiazepine-resistant convulsive SE, only about
half of the papers cited a specied time frame in which they considered
the seizure termination to be successful. The most commonly stated
specication was the termination of seizures within 30 min of infusion
(6 papers, 22.2%). However, the time frame in other studies ranged
from 3 min to 48 h. One study even linked the time of the cessation of
the SE with the end of the infusion. In this meta-analysis, the authors

http://dx.doi.org/10.1016/j.yebeh.2015.04.038
1525-5050/ 2015 Elsevier Inc. All rights reserved.

Please cite this article as: Redecker J, et al, Comparison of the effectiveness of four antiepileptic drugs in the treatment of status epilepticus
according to four different efcacy criteria, Epilepsy Behav (2015), http://dx.doi.org/10.1016/j.yebeh.2015.04.038

J. Redecker et al. / Epilepsy & Behavior xxx (2015) xxxxxx

tried to control the effect of different criteria with statistical methods.

Unfortunately, this cannot be done without knowing the severity of
the effect. The scope of the problem may be estimated from a recent discussion concerning the efcacy of LCM in absence status epilepticus
(ASE). d'Orsi et al. [8] administered 400 mg LCM to a patient with ASE
after two boluses of a benzodiazepine. Twenty-four hours later, the
ASE resolved. d'Orsi et al. [8] used the outcome criterion of the study
mentioned above [5] (i.e., status epilepticus is considered aborted if
ictal patterns cease within 20 min following treatment and remain suppressed for 40 min). Consequently, they considered LCM not to be effective in their patient with ASE. Sodemann et al. [9] published a case of a
patient with ASE refractory to diazepam and LEV. After a loading dose
of LCM 400 mg, the patient recovered within 2 h. The authors considered LCM as effective in this case. Using the same criterion as d'Orsi
et al. [8], LCM would be considered as ineffective in this case as well.
Hottinger et al. [10] considered an AED as effective in SE if clinical improvement and electroencephalographic resolution of refractory status
epilepticus occurred within 24 h after starting with the new AED with
no requirement for further AEDs. Using this criterion, LCM would have
been effective in both treatment episodes.
The preliminary data presented here shall give an impression on
how different criteria for the identication of an AED with a possible
or certain treatment effect can have an inuence on the results of retrospective case series.
2. Methods
We present a data subset from a large retrospective study which,
when completed, will cover all treatment episodes of SE at the neurological department of the Universittsmedizin Rostock from January
2010 to June 2013. This study was approved by the local ethics board
at Rostock University under the identier A-2013-0099.
We identify the patients treated by searching the electronic medical
records of our clinic for the term status epilepticus. We then manually
review the medical les of these patients to determine at which time a
certain AED was administered, which AED was effective in terminating
the epileptic condition, and the time of termination. When treatment
had been started in another department or hospital, the treatment episode was only included if a thorough documentation of the treatment
process was available. When treatment had been started by the emergency service in the prehospital setting, the data were usually sufcient
to be included in the analysis. When reviewing the medical le, we
checked the correctness of the diagnosis of SE. We classify an antiepileptic drug as effective for the termination of SE using four different criteria
described below.
After collecting all the data, we compare and contrast the results of
four different efcacy criteria for the effectiveness of PHT, VPA, LEV,
and LCM. The criteria by which we evaluate the episodes are as follows:
Criterion 1 = the last AED administered before SE termination;
Criterion 2 = the last drug introduced into the antiepileptic therapy
within 72 h before SE termination and without changes in the
comedication;
Criterion 3 = the last drug introduced into the antiepileptic therapy
or increased in dose within 24 h before SE termination without
changes in the comedication; and
Criterion 4 = the last drug introduced into the antiepileptic therapy
within 72 h before SE termination, even allowing changes in the
comedication.
We intend to perform subgroup analyses of efcacy in different subtypes of SE by classifying subtypes of nonconvulsive SE according to the
system of Shorvon [11]. In this preliminary analysis, we perform subgroup analyses for the three most frequent subtypes in our series so
far, i.e., epilepsia partialis continua, nonlimbic complex partial SE, and
nonconvulsive SE in the postictal phase of generalized tonicclonic

Table 1
Types of status epilepticus.
Type of status epilepticus

Number of episodes

Generalized tonicclonic status epilepticus

Epilepsia partialis continua
Limbic complex partial status epilepticus
Nonlimbic complex partial status epilepticus
Nonconvulsive status epilepticus in the postictal
phase of tonicclonic seizures
Aura continua

2
9
2
9
6
3

seizures. We used two-tailed 2-tests with the Yates adjustment for

small samples to evaluate statistical differences between efcacy rates
of different AEDs. We analyzed the position of the rst administration
in the course of treatment. For this purpose, we counted every administration of an AED as a new treatment attempt. This means that a second
administration, e.g., of a benzodiazepine, was counted as a second treatment attempt. To look for differences between rst positions of the four
AEDs, we performed MannWhitneyWilcoxon tests. Since these data
are very preliminary, we did not perform a BonferroniHolmes procedure to correct for multiple statistical tests.

3. Results
Thirty-seven treatment episodes in 32 patients (13 male and 19
female, mean age at rst episode: 68 years; SD: 17) could be analyzed.
In 31 episodes, at least one AED was given intravenously (i.e., PHT,
n = 12; VPA, n = 21; LEV, n = 26; and LCM, n = 6). In 24 of these treatment episodes, at least one benzodiazepine had been given previously
without success. Lacosamide was never given before a benzodiazepine
had been tried. The types of SE are listed in Table 1, with the classication of nonconvulsive SE according to the system of Shorvon [11]. In the
group as a whole, efcacy rates according to all four criteria were not
signicantly different between the four AEDs, but the efcacy rates of
each AED differed considerably when using different efcacy criteria
(see Table 2). The positions of the rst treatment attempt and the dosages are given in Table 3. Probably because of the small sample size,
no difference between the positions of the rst administration of the
AEDs reached signicance (all p N 0.1). The treatment delay between
the rst symptoms of SE and the rst treatment attempt is given in
Table 4. In the episodes with nonlimbic complex partial SE, PHT was
used three times; VPA, ve times; and LEV, seven times. According to
our efcacy criterion 1, PHT was never effective; VPA, in only one
case; and LEV, in all seven episodes where it was administered. This difference reached signicance (p b 0.01). According to criterion 1, LEV
seemed to be more effective in nonlimbic complex partial SE than in
epilepsia partialis continua and nonconvulsive SE in the postictal
phase of generalized tonicclonic seizures (p b 0.05). No other signicant differences of efcacy rates were revealed by the subgroup
analyses.

Table 2
The number of treatment episodes and number of effective administrations according to
four different criteria.
AED

Criterion 1

Criterion 2

Criterion 3

Criterion 4

PHT
VPA
LEV
LCM

12
21
26
6

16.7%
28.6%
50%
50%

0%
14.3%
19.2%
16.7%

0%
19%
34.6%
33.3%

16.7%
19%
23%
16.7%

Criterion 1 = the last AED administered before SE termination; criterion 2 = the last drug
introduced into the antiepileptic therapy within 72 h before SE termination and without
changes in the comedication; criterion 3 = the last drug introduced into the antiepileptic
therapy or increased in dose within 24 h before SE termination without changes in the
comedication; and criterion 4 = the last drug introduced into the antiepileptic therapy
within 72 h before SE termination, even allowing changes in the comedication.

Please cite this article as: Redecker J, et al, Comparison of the effectiveness of four antiepileptic drugs in the treatment of status epilepticus
according to four different efcacy criteria, Epilepsy Behav (2015), http://dx.doi.org/10.1016/j.yebeh.2015.04.038

J. Redecker et al. / Epilepsy & Behavior xxx (2015) xxxxxx

Table 3
Positions of the rst treatment attempt and dosages of intravenously administered AEDs.
AED (i.v.) with the number
of treatment episodes

Position of the rst treatment attempt

with the AED, median (range)

First dosage in mg,

mean (SD)

Total dosage in 24 h in episodes where SE was not

terminated by the rst dosage in mg, mean (SD)

PHT, n = 12
VPA, n = 21
LEV, n = 27
LCM, n = 6

8 (147)
5 (130)
2 (173)
14.5 (293)

504 (310)
870 (619)
1058 (567)
125 (42)

1006 (666)
2193 (1274)
2804 (1250)
367 (121)

Table 4
Time between the onset of symptoms and start of treatment.
Delay of treatment

Number of episodes

b5 min
530 min
3160 min
124 h

2
8
4
18

4. Discussion
Because of the very small sample, the data presented here are only
preliminary. Nevertheless, it shows that even statistically signicant
results concerning the efcacy of different AEDs in different subtypes
of SE may depend on the outcome criteria used for analysis. Notably in
refractory nonconvulsive SE, it is difcult to decide which outcome criterion is appropriate. When asked which criterion is the most likely to
be appropriate, there are several aspects to discuss. Criterion 1 seems
to be questionable because an AED, which has been administered several times for days or even weeks, may be identied as the termination
drug after many ineffective administrations. Criterion 4 is questionable
because an increase of the dosage of other AEDs may be more effective
than the mere presence of an additional AED in the last three days before SE termination. Therefore, we think that the choice has to be
made between criteria 2 and 3, but only criterion 1 leads to any signicant results in our preliminary analysis.
5. Conclusion
Efcacy criteria for the effectiveness of AEDs in the treatment of SE
should be standardized. Whereas for rst-line treatment, a treatment
effect within 10 min seems to be favorable and for second-line

treatment, a treatment effect within 30 min is probably reasonable,

the situation in refractory nonconvulsive SE is more complicated.
Conicts of interest
Dr. Rsche received speaker's honoraria from Eisai and UCB, served
as a medical advisor for Eisai, received travel grants from Eisai and
UCB, and received nancial support for an investigator-initiated trial
from Pzer. The other authors declare that they have nothing to
disclose.
References
[1] Sabers A, Wolf P, Moller A, Rysgaard K, Ben-Menachem E. A prospective, randomized, multicentre trial for the treatment of refractory status epilepticus; experiences
from evaluating their effect of the novel drug candidate, NS1209. Epilepsy Res 2013;
106:2925.
[2] Agarval P, Kumar N, Chandra R, Gupta G, Antony AR, Garg N. Randomized study of
intravenous valproate and phenytoin in status epilepticus. Seizure 2007;16:52732.
[3] Bleck T, Cock H, Chamberlain J, et al. The established status epilepticus trial 2013.
Epilepsia 2013;54(Suppl. 6):8992.
[4] Rsche J, Pohley I, Kampf C, Benecke R. Topiramat in der Behandlung des Status
epilepticus. Nervenheilkunde 2013;32:2259.
[5] Treiman DM, Meyers PD, Walton NY. A comparison of four treatments for
generalized convulsive status epilepticus. N Engl J Med 1998;339:7928.
[6] Zelano J, Kumlien E. Levetiracetam as alternative stage two antiepileptic drug in
status epilepticus: a systematic review. Seizure 2012;21:2336.
[7] Yasiry Z, Shorvon SD. The relative effectiveness of ve antiepileptic drugs in treatment of benzodiazepine-resistant convulsive status epilepticus: a meta-analysis of
published studies. Seizure 2014;23:16774.
[8] d'Orsi G, Pacillo F, Trivisano M, Pascarella MG, Ferrara MA, Specchio LM. Lacosamide
in absence status epilepticus. Seizure 2014;23:3978.
[9] Sodemann U, Moller HS, Blaabjerg M, Beier ChP. Successful treatment of refractory
absence status epilepticus with lacosamide. J Neurol 2014;261:20257.
[10] Hottinger A, Sutter R, Marsch St, Regg St. Topiramate as an adjunctive treatment in
patients with refractory status epilepticus. CNS Drugs 2012;26:76172.
[11] Shorvon S. What is nonconvulsive status epilepticus, and what are its subtypes?
Epilepsia 2007;48(Suppl. 8):358.

Please cite this article as: Redecker J, et al, Comparison of the effectiveness of four antiepileptic drugs in the treatment of status epilepticus
according to four different efcacy criteria, Epilepsy Behav (2015), http://dx.doi.org/10.1016/j.yebeh.2015.04.038