You are on page 1of 9

ISSN 1541–6933 (Print) ISSN 1556–0961 (Online)

Volume 5 • Number 1 • 2006 A Journal of Acute and Emergency Care

Eelco F. M. Wijdicks, MD

The Official Journal of the
wn m
Do .co

Indexed and Abstracted in Index Medicus and MEDLINE

an nal
ea ou

, R naJ
ar ma
Se Hu

NCC_5_1_cvr 1 7/17/06, 12:07 PM

Neurocritical Care
Copyright © 2006 Humana Press Inc.
All rights of any nature whatsoever are reserved.
ISSN 1541-6933/06/5:71–78 ISSN 1556-0961 (Online)
DOI: 10.1385/Neurocrit. Care 2006;05:71–78

Translational Research

Levetiracetam is Neuroprotective in Murine Models of Closed

Head Injury and Subarachnoid Hemorrhage
Haichen Wang,1,2 Junling Gao,1–5 Timothy F. Lassiter,3 David L . McDonagh,1,2,4 Huaxin Sheng,1,4 David S.Warner,1,4
John R . Lynch,1,2 and Daniel T. Laskowitz* ,1,2,4
Multidisciplinary Neuroprotection Laboratories, and the 2Departments of Medicine (Neurology), 3Pharmacy,
and 4Anesthesiology, Duke University Medical Center, Durham, NC; 5Department of Histology and Embryology,
North China Coal Medical College, Tangshan, Hebei, China

Objectives: Prophylactic treatment with antiepileptic drugs is common practice following
subarachnoid hemorrhage (SAH) and traumatic brain injury. However, commonly used
antiepileptic drugs have multiple drug interactions, require frequent monitoring of serum
levels, and are associated with adverse effects that may prompt discontinuation. In the
current study, we test the hypothesis that levetiracetam, an anticonvulsant with favorable
interaction and adverse event profiles, is neuroprotective in clinically relevant models of
SAH and closed head injury (CHI).
Methods: A single intravenous dose of vehicle, low-dose (18 mg/kg), or high-dose
(54 mg/kg) levetiracetam was administered intravenously followed CHI. Functional
assessments were performed on a daily basis, and histological assessments performed at
24 hours. In a separate series of experiments, mice were randomized to receive intravenous
administration of vehicle, low-dose, or high-dose levetiracetam every 12 hours for 3 days
following SAH. Functional endpoints were assessed daily, followed by measurement of
MCA luminal diameter on day 3.
Results: A single dose of levetiracetam improved functional and histological outcomes
after CHI. This effect appeared specific for levetiracetam and was not associated with
fosphenytoin treatment. Treatment with levetiracetam also improved functional outcomes
and reduced vasospasm following SAH.
Conclusion: Levetiracetam is neuroprotective in clinically relevant animal models of SAH
and CHI. Levetiracetam may be a therapeutic alternative to phenytoin following acute
brain injury in the clinical setting when seizure prophylaxis is indicated.
Key Words: Subarachnoid hemorrhage; traumatic brain injury; posttraumatic epilepsy;
*Correspondence and reprint levetiracetam; vasospasm; neuroprotection.
requests to: (Neurocrit. Care 2006;05:71–78)
Daniel T. Laskowitz
Box 2900
Duke University Medical Center
Durham, NC 27710
E-mail: Introduction matic brain injury accompanied by loss
Patients with brain injury have an in- of consciousness, depressed skull frac-
creased risk of seizures, which may be ture, parenchymal contusion, or pro-
especially detrimental in the acute care longed amnesia may be at high risk of
setting. In particular, patients with sub- seizures in the acute care setting. For
arachnoid hemorrhage (SAH) and trau- these reasons, antiepileptic drug (AED)

72 Wang et al.

prophylaxis is a common practice and has been incorpo- in previously validated murine models of closed head
rated into the standard of care in many neurocritical care injury (CHI) and SAH (13–15).
units (1,2). However, the prophylactic use of anticonvul-
sant drugs is often associated with the need for frequent Methods
monitoring of drug levels, as well as the possibility of These studies were approved by the Duke University
drug interactions and adverse events. Animal Care and Use Committee. The care and handling
Phenytoin is often considered the agent of choice for of the animals comply with National Institutes of Health
seizure prophylaxis because of its ease of administra- guidelines.
tion. Disadvantages of prophylactic phenytoin use in-
clude its association with allergic reactions (predomi-
nantly rash), drug fever, thrombocytopenia, hepatitis,
Closed Head Injury Model
cardiovascular toxicities, and fluctuating serum levels This murine CHI model (13,14) was adapted from a
that require frequent monitoring and dose adjustments previously described model of closed cranial trauma for
(3–5). Fosphenytoin is a phenytoin prodrug that is asso- the rat (16,17). Twelve- to fourteen-week-old C57Bl/6J
ciated with a much higher cost, but lower incidence of male mice (Jackson Laboratories, Bar Harbor, ME) were
symptomatic hypotension and bradyarrhythmias dur- used. The trachea was intubated after anesthesia induc-
ing intravenous administration (6). However, both of tion with 4.6% isoflurane and the lungs were mechani-
these agents are associated with numerous drug interac- cally ventilated with 1.6% isoflurane in 30% O2/70% N2.
tions, and the association between phenytoin and ad- Rectal temperature was maintained at 37ºC. The animal
verse reactions such as thrombocytopenia, fever, and was positioned in a stereotactic device, the scalp was in-
rash may lead to early discontinuation. Moreover, recent cised and the skull exposed. A concave 3-mm metallic
observations suggest that exposure to phenytoin may be disc was glued to the skull immediately caudal to
associated with impaired functional outcomes when bregma. A 2.0-mm diameter pneumatic impactor (Air-
used in the management of aneurysmal SAH (7). Power, Inc. High Point, NC) was used to deliver a single
Given these concerns, there remains a clinical need midline impact to the disc surface. The impactor was
for an agent with reduced drug interactions, need for discharged at 6.8 ± 0.2 m/second with a head displace-
monitoring, and adverse events. Levetiracetam belongs ment of 3 mm. After impact, the animals were allowed
to the pyrrolidine class of drugs, and was approved in to recover spontaneous ventilation and the tracheas
1999 as an adjunct in the treatment of partial complex were extubated. Following recovery, mice were allowed
seizures. Interestingly, levetiracetam has a different anti- free access to food and water.
convulsant profile from most other AEDs, and has no ef-
fect on the maximal electroshock or pentylenetrazol Immunohistochemistry: Fluoro-Jade B Staining
models which are traditionally used in the screening of Sagittal sections (40 µm) were cut on a vibratome, col-
anticonvulsant candidates (8). Levetiracetam has mini- lected in cryoprotectant solution containing ethylene
mal drug interactions, and there is no interaction with glycol, sucrose, and sodium phosphate. Every eighth
hepatic CYP450 enzymes (9). Although adverse effects section was mounted onto a charged slide and stained
are infrequent, administration of levetiracetam may be with Fluoro-Jade B (18) to mark degenerating neurons.
associated with agitation and emotional lability, and Slides were immersed in 100% ethanol for 3 minutes
should be monitored closely in patients with psychiatric followed by 1 minute in 70% alcohol and 1 minute in
conditions (10). distilled water. The slides were then transferred to a so-
Preliminary data also suggest that levetiracetam may lution of 0.06% potassium permanganate for 15 minutes
be neuroprotective in the setting of acute brain injury. In followed by a 1-minute rinse in distilled water. Slides
an in vitro hippocampal slice paradigm, levetiracetam were then stained in a 0.001% Fluoro-Jade B (Histochem,
reduced high voltage activated calcium currents, and re- Jefferson, AR) solution prepared in 0.1% acetic acid for
versed the inhibition of negative allosteric modulators of 30 minutes, the slides were rinsed for 1 minute in dis-
-aminobutyric acid and glycine-gated currents (11). tilled water three times. Slides were then dried over-
These results were recently extended to a rodent model night in the dark. The following day, the slides were
of focal ischemia, where intraperitoneal administration cleared by immersion in xylene and coverslipped with
of levetiracetam was associated with a dose-dependent DPX (Fluka, Milwaukee, WI). Slides containing hippo-
reduction in infarct volume (12). However, this was not campus were examined for degenerating neurons using
a study that could be easily translated into the clinical an epifluorescent microscope (Nikon, Japan) with a me-
setting, as drug was administered prior to ischemic in- dium band blue excitation (Nikon B-2A, 450–490 nm)
jury. In the current study, we assessed whether post- filter set. Degenerating neurons were quantified at ×20
insult intravenous administration of levetiracetam is magnification by counting the total number of Fluoro-
associated with histological or functional improvement Jade B-positive neurons in every eighth section of brain
Neurocritical Care ♦ Volume 5, 2006
Levetiracetam is Neuroprotective 73

hippocampus by an observer blinded to group ately after SAH and at 12-hour intervals for the first
assignment. 3 days as previously described.

Subarachnoid Hemorrhage Model Testing of Functional Deficits

SAH was induced as previously described (15,19,20). An automated rotarod (Ugo Basile, Comerio, Italy)
Twelve- to fourteen-week-old C57Bl/6J male mice was used to assess vestibulomotor function (14). On the
(Jackson Laboratories, Bar Harbor, ME) were induced in day prior to CHI, mice underwent two consecutive con-
a chamber with 4.6% isoflurane. The trachea was intu- ditioning trials at a set rotational speed (16 rpm) for 60
bated and anesthesia maintained by ventilation with seconds and then three additional trials with an acceler-
1.6% isoflurane in 30% O2/70% N2. The right common ating rotational speed. The average time to fall from the
carotid was exposed by a midline neck incision. rotating cylinder in the latter three trials was recorded
Following isolation and ligation of the external carotid as baseline latency. On days 1–5 post-CHI, the mice had
artery (ECA), a blunted 5-0 monofilament (Ethicon, three consecutive daily trials with accelerating rotational
Somerville, NJ) nylon suture was inserted into the ECA speed (intertrial interval = 15 minutes). The average la-
and advanced distal to the right anterior cerebral artery tency to fall from the rod was recorded. Mice unable to
(ACA)–middle cerebral artery (MCA) bifurcation to per- grasp the rotating rod were given a latency value of
forate the right ACA. The suture was then withdrawn, 0 seconds.
allowing reperfusion and SAH. In sham-operated mice Functional deficit (3–21) was also assessed daily by
the monofilament was advanced without perforation. an examiner blinded to group assignment, as we have
After removal of the filament, the skin was closed with described previously (20). The motor component score
4-0 suture (Surgical Specialties, Reading, PA) and isoflu- was derived from spontaneous activity, symmetry of
rane was discontinued. After recovery of spontaneous limb movements, climbing, and balance and coordina-
ventilation, the trachea was extubated, and mice were tion with each activity scored on a 0–3 scale. The sensory
allowed free access to food and water. component score was derived from examination of body
proprioception, vibrissae, and tactile responses, with
Cerebral Vascular Casting and MCA Luminal each component scored on a 1–3 scale. A score of 21 rep-
Measurement After SAH resented normal function.
Cerebral vascular casting was performed 72 hours
Statistical Analysis
after SAH as previously described (15,19,20). Three days
following SAH, mice were intubated after induction Serial tests of functional performance, including ro-
with 4.6% isoflurane. A thoracotomy was performed to tarod performance and clinical severity scores, were
allow cannulation of the proximal aorta. Flexible plastic compared with repeated measures analysis of variance
tubing (0.76-mm internal diameter, Helix Medical, (ANOVA) with time as the repeated variable and using
Netherlands) was used to deliver infusion solutions by Dunnett’s post hoc method for correcting multiple com-
manual pulsatile syringe pressure (60–80 mmHg). parisons. The numbers of Fluoro-Jade B positive cells
Physiological saline (20 mL) was infused followed by and neurological scores were compared among groups
30 mL of 10% formalin and 5 mL of gelatin-India ink with the Kruskal–Wallis H statistic. Between groups, dif-
solution passed through a 0.2-µm filter. The proximal ferences were compared by the Mann–Whitney U statis-
aorta and vena cava were ligated following gelatin ink tic. MCA luminal diameters were compared with
infusion to prevent leaking. The body was then refriger- one-way ANOVA. Parametric values are expressed as
ated for 24 hours to allow gelatin solidification, after mean ± standard deviation (SD). Significance was
which the brain was harvested and stored in 4% for- assumed if p < 0.05.
maldehyde. The MCA was imaged using a video camera
linked to a dissecting microscope controlled by an
image analyzer (MCID-M5, Imaging Research Inc., St.
Catherine’s, Ontario, Canada). The narrowest diameter of Levetiracetam is Associated With Improved
the MCA observed within 1-mm distal to the ACA–MCA Functional and Histological Outcomes
bifurcation was determined by digital measurement. Following CHI
To assess whether administration of levetiracetam re-
Drug Administration duced functional deficit in a clinically relevant paradigm
In the CHI paradigm, a single intravenous adminis- of CHI, mice were randomized after injury to receive a
tration of either vehicle or drug was administered by tail single intravenous injection of high-dose levetiracetam
vein 30 minutes following pneumatic impact. In the (54 mg/kg; n = 14); low-dose levetiracetam (18 mg/kg;
SAH model, vehicle or drug was administered immedi- n = 14), or saline control (n = 9). Although levetiracetam
Neurocritical Care ♦ Volume 5, 2006
74 Wang et al.

is not dosed by weight clinically, these doses were cho- functional outcome and increased evidence of hippo-
sen as they are in the range commonly used in clinical campal neurodegeneration.
practice (1000–3000 mg. daily, delivered in divided
doses). Treatment with high-dose levetiracetam was as- Levetiracetam is Associated With Improved
sociated with a significant improvement in vestibulo- Functional Outcomes and Reduction
motor function as assessed by rotarod as compared to in Vasospasm Following SAH
low-dose levetiracetam or control animals. This effect Routine seizure prophylaxis is also commonly em-
was sustained throughout the 5-day testing period. The ployed in the setting of aneurysmal SAH. To assess
performance of animals treated with lowdose levetirace- whether levetiracetam was neuroprotective in this model,
tam was not significantly different than control mice we randomized three groups of animals to receive
treated with vehicle (Figure 1A). high-dose (54 mg/kg; n = 14) or low-dose levetiracetam
To address whether this beneficial effect of levetirace-
tam was nonspecific and caused by early suppression of
seizures, we performed a parallel experiment with fos-
phenytoin. As before, mice were randomized into three
groups following CHI. Thirty minutes following impact,
mice received a single intravenous administration of
either vehicle (n = 14), a standard therapeutic loading
dose of fosphenytoin (18 mg/kg phenytoin equivalents;
n = 14), or high-dose fosphenytoin (54 mg/kg phenytoin
equivalents; n = 15). Interestingly, although admini-
stration of 18 mg/kg fosphenytoin did not have any
beneficial effect in this model, high-dose fosphenytoin
was associated with impaired motor performance that
was sustained throughout the 5-day testing period
(Figure 1B).
To assess whether the beneficial effects of levetirace-
tam on functional performance were associated with
histological evidence of neuroprotection, we performed
CHI on a separate cohort of animals (n = 8) that received
a single intravenous dose of vehicle (n = 8), low-dose
levetiracetam (18 mg/kg; n = 7), or high-dose levetirace-
tam (54 mg/kg; n = 7). At 24 hours after injury, animals
were euthanized. Hippocampal sections were stained
with Fluoro-Jade B to quantify the magnitude of neuro-
nal injury. We have previously demonstrated that histo-
logical evidence of neuronal injury is correlated with
vestibulomotor (rotarod) and neurocognitive (Morris
water maze) performance (14). Fewer Fluoro-Jade
B positive degenerating neurons were present in the
high-dose (11 ± 1) and low-dose (15 ± 1) levetiracetam
groups versus vehicle (26 ± 2, p < 0.01) (see Figure 2). In
contrast, treatment with low-dose fosphenytoin (18 mg/
kg; n = 5) did not affect histological evidence of hippo-
campal injury (28 ± 2). Furthermore, high-dose fosphe-
nytoin (54 mg/kg; n = 5) increased Fluoro-Jade B posi-
tive cells (39 ± 3). Taken together, these results suggest Fig. 1. (A) A single intravenous administration of high-dose leveti-
that a single administration of levetiracetam following racetam improved post traumatic vestibulomotor function measured
CHI improves functional outcomes and reduces by rotarod testing (*p < 0.05 high-dose levetiracetam versus con-
evidence of neuronal degeneration after head injury. trol). Low-dose levetiracetam was not significantly different than
This beneficial effect was not present with standard control. (B) Administration of fosphenytoin at a standard intrave-
nous loading dose (18 mg/kg phenytoin equivalents) did not confer
doses of fosphenytoin, suggesting that the beneficial any functional benefit following closed head injury. High-dose fosphe-
effects of levetiracetam are not caused by a nonspecific nytoin (54 mg/kg phenytoin equivalents) was associated with impaired
anticonvulsant effect. To the contrary, high-dose fosphe- functional performance as compare to vehicle-treated animals (*p <
nytoin was associated with deleterious effects on 0.05 versus control). All values are expressed as mean ± SD.
Neurocritical Care ♦ Volume 5, 2006
Levetiracetam is Neuroprotective 75

Fig. 2. (A) Representative images of Fluoro-Jade B-positive neurons in the hippocampus. Fluro-Jade B stains degenerating neurons. At 24
hours after CHI, mice treated with both high (c) and low (b) dose levetiracetam demonstrated fewer Fluoro-Jade B positive degenerating
neurons than the vehicle control (a). (B) Neuronal injury was then quantified by counting the total number of Fluoro-Jade B-positive neu-
rons present throughout the entire hippocampus, which was sampled at every eighth section (approximately 7 sections/brain). The total
number of degenerating neurons was less in animals treated with high- or low-dose levetiracetam, but greater in high-dose fosphenytoin
treated group as compared to vehicle control or low-dose fosphenytoin-treated animals (**p < 0.01 for treatment group versus vehicle).
Values = mean ± SD.

(18 mg/kg; n = 13), or vehicle (n = 10) following experi- Discussion

mental SAH. To recreate the clinical setting of delayed In the current study, we demonstrate that administra-
ischemic deficit following aneurysmal SAH, drug or ve- tion of levetiracetam improves functional performance
hicle was administered intravenously by tail vein imme-
and reduces histological evidence of neuronal injury
diately following SAH, and at 12-hour intervals for the
and vasospasm in murine models of CHI and SAH, re-
first 3 days. Both high- and low-dose levetiracetam were
spectively. Although seizure prophylaxis has not been
associated with functional improvements as assessed by
rotarod and neuroseverity score. This effect was sus- shown to prevent epileptogenesis or the later develop-
tained throughout the 3-day testing period (Figure 3). ment of seizure disorder following head trauma, a num-
Interestingly, this functional improvement was also ber of studies have demonstrated the efficacy of this
associated with a reduction in luminal narrowing approach in the early suppression of seizures (2). Current
(Figure 4), suggesting that, in addition to any direct recommendations suggest the use for 1 week following
neuroprotective effects, levetiracetam administration re- injury (1). In the setting of aneurysmal SAH, seizures
duced vasopasm in this model. occur with a prevalence of 4–10% and are more likely to
Neurocritical Care ♦ Volume 5, 2006
76 Wang et al.

Fig. 4. Vasospasm, defined as the reduction of MCA diameter, was

attenuated in the groups treated with both high- and low-dose leve-
tiracetam (**p < 0.01 versus vehicle).

rier solution that allows for safer intravenous adminis-

tration without significant cardiovascular side effects
(6,26). The remaining toxicity profile is similar to phe-
nytoin. There is emerging evidence that prophylactic
treatment with phenytoin may be deleterious in patients
with acute brain injury. For example, a recent study has
demonstrated that exposure to phenytoin is associated
with impaired neurocognitive performance following
SAH (7). These observations would be consistent with
preclinical data suggesting adverse effects of phenytoin
Fig. 3. (A) Animals treated with either high- or low-dose levetirace- on recovery from acute brain injury (27). In the current
tam had improved rotarod performance as compared to vehicle- study, we found that therapeutic doses of phenytoin ex-
treated animals after subarachnoid hemorrhage (*p < 0.05 for either
erted no beneficial effect in the setting of acute brain in-
dose compared to vehicle). This effect was sustained throughout the
3-day testing period. (B) Functional improvement was also demon- jury, and in fact high-dose phenytoin was associated
strated by serial neurological clinical assessments (*p < 0.05 in both with impaired recovery.
treated groups versus vehicle). Levetiracetam has several advantages over phenytoin
for use in seizure prophylaxis, as it is well tolerated,
does not require monitoring of levels, has minimal drug
occur soon after the ictus (21,22). The rapid increase in interactions or adverse effects (28,29), and has been re-
blood pressure associated with seizures may be espe- cently approved for intravenous administration in the
cially detrimental in the setting of a patient with an un- acute setting. Levetiracetam has a distinct anticonvul-
secured aneurysm. Thus, although there is insufficient sant profile from most other classical AEDs. For exam-
evidence to support or dismiss the use of routine seizure ple, levetiracetam lacks activity in traditional seizure
prophylaxis in SAH (23,24), this has been incorporated models used to screen anticonvulsant candidates (8).
into routine practice at many institutions. Levetiracetam binds throughout the central nervous
Traditionally, phenytoin is the most common AED system to a presynaptic vesicle protein called SV2A (30).
administered for seizure prophylaxis following brain The function of this protein is unknown, but it may
injury. However, phenytoin has a number of drug modulate the vesicle fusion process (31). The binding of
interactions, requires close monitoring of drug levels, levetiracetam to SV2A correlates with its anticonvulsant
and suspicion of adverse events such as drug fever, rash, effects (30), and there is evidence that levetiracetam may
or thrombocytopenia often necessitates discontinuation selectively interact with this integral protein under
(3–5,25). Fosphenytoin is a newer drug that is converted pathophysiological conditions such as may occur with
to phenytoin in vivo, but is delivered in a nontoxic car- acute brain injury or seizure activity. Levetiracetam may
Neurocritical Care ♦ Volume 5, 2006
Levetiracetam is Neuroprotective 77

also have neuroprotective properties, and has been tiracetam for seizure prophylaxis in the neurocritical
shown to reduce infarct volume in a rodent model of care setting. The apparent neuroprotective and anti-
focal ischemia (12). In the current study, high-dose vasospastic effect of levetiracetam in this murine model
levetiracetam also improved functional outcomes and warrants future confirmatory trials in other animal and/
reduced histological evidence of neuronal injury or human trials.
following CHI. This is unlikely to be the result of its an-
ticonvulsant properties, as only a single dose given 30 Acknowledgments
minutes following brain trauma was efficacious, whereas This work was supported by the Institute for the
treatment with fosphenytoin had no comparable effect. Study of Aging. Junling Gao was partly supported by
Our observation that supratherapeutic doses of phenyt- Natural Science Foundation of Hebei Province, China
oin were associated with impaired outcome is consistent (grant 301404). Dr. Laskowitz is a consultant for UCB
with a recent clinical study demonstrating adverse ef- Pharma, which provided levetiracetam. We would like
fects of phenytoin in patients with SAH (7). to thank Dr. Heather Vita for her thoughtful review of
In the current study, we also observed that levetirace- this manuscript.
tam improves functional recovery in a murine SAH
model. Although this might have been partly because of References
a direct neuroprotective effect against cerebral ischemia, 1. Chang BS, Lowenstein DH. Practice parameter: Antiepileptic
there was also a direct reduction in vasospasm, defined drug prophylaxis in severe traumatic brain injury. Report of the
Quality Standards Subcommittee of the American Academy of
by measurement of MCA luminal diameter. Although Neurology. Neurology 2003;60:10–16.
this has never been described before, it is possible that 2. Temkin NR, Dikmen SS, Wilensky AJ, Keihm J, Chabal S, Winn
these affects are mediated by nitric oxide (NO). Vascular HR. A randomized, double-blind study of phenytoin for the pre-
endothelium smooth muscle tone is mediated by a bal- vention of post-traumatic seizures. N Engl J Med 1990;323:
ance of NO (32) and endothelial-derived constriction 3. Korman LB, Olson MJ. Phenytoin-induced hepatitis, rhabdomy-
factors (33). NO depletion is associated with SAH, and olysis, and renal dysfunction. Clin Pharm 1989;8:514–515.
NO replacement reverses cerebral vasospasm (34–36). 4. Conger LA, Grabski WJ. Dilantin hypersensitivity reaction. Cutis
Interestingly, administration of levetiracetam has been 1996;57(4):223–226.
5. Patsalos PN. Clinically important drug interactions in epilepsy:
demonstrated to upregulate astrocytic production of in- interactions between antiepileptic drugs and other drugs. Lancet
ducible NO synthase in a concentration-dependent fash- Neurol 2003;2(8):473–481.
ion (37). In a recent study utilizing in vivo microdialysis, 6. Browne TR, Kugler AR, Eldon MA. Pharmacology and pharma-
levetiracetam, in a dose-dependent fashion, directly cokinetics of fosphenytoin. Neurology 1996;46(Suppl) 1:S7–S10.
7. Naidech AM, Kreiter KT, Janua N, et al. Phenytoin exposure is
upregulated NO production in the cerebellar nuclei of associated with functional and cognitive disability after sub-
rats (38). Thus, it is plausible the palliative effects of arachnoid hemorrhage. Stroke 2005;36:583–587.
levetiracetam on reducing vasospasm and improving 8. Klitgaard H, Matagne A, Gobert J, Wülfert E. Evidence for a
unique profile of levetiracetam in rodent models of seizures and
functional recovery following SAH may be mediated by
epilepsy. Eur J Pharmacol 1998;353:191–206.
its upregulation of NO synthesis. In summary, leveti- 9. Dooley M, Plosker GL, Levetiracetam. A review of its adjunctive
racetam was well tolerated in two murine models of use in the management of partial onset seizures. Drugs 2000;60:
acute brain injury. A single intravenous administration 871–893.
10. Mula M, Trimble MR, Yuen A, Liu RS, Sander JW. Psychiatric ad-
improved functional and histological endpoints follow- verse events during levetiracetam therapy. Neurology 2003;5:
ing CHI, an effect that was not replicated with fosphe- 704–706.
nytoin. Following experimentally induced SAH, admin- 11. Rigo JM, Hans G, Nguyen L, et al. The anti-epileptic drug leveti-
istration of levetiracetam was associated with both racetam reverses the inhibition by negative allosteric modulators
of neuronal GABA- and glycine-gated currents. Br J Pharmacol
functional improvement and reduction in vasospasm as 2002;136:659–672.
compared to vehicle-treated controls. These results may 12. Hannon E, Klitgaard H. Neuroprotective properties of the novel
have implications for the use of levetiracetam for sei- antiepileptic drug levetiracetam in the rat middle cerebral arteryoc-
zure prophylaxis in the neurocritical care setting. clusion model of focal cerebral ischemia. Seizure 2001;10:287–293.
13. Lynch JR, Pineda JA, Morgan D, et al. Apolipoprotein E affects
In summary, we demonstrate that levetiracetam is the central nervous system response to injury and the develop-
well tolerated in two murine models of acute brain in- ment of cerebral edema, Ann Neurol 2002;51:113–117.
jury. The beneficial effects of a single intravenous 14. Lynch JR, Wang H, Mace B, et al. A novel therapeutic derived
administration of levetiracetam on functional and from apolipoprotein E reduces brain inflammation and improves
outcome after closed head injury. Exp Neurol 2005;192:109–116.
histological outcomes post head injury was not repli- 15. McGirt MJ, Lynch J, Sheng H, Laskowitz DT, Pearlstein RD,
cated with fosphenytoin. Following experimentally in- Warner DS. Simvastatin increases endothelial nitric oxide syn-
duced SAH, administration of levetiracetam was associ- thase and ameliorates cerebral vasospasm following subarach-
noid hemorrhage. Stroke 2002;33:2950–2956.
ated with both functional improvement and reduction
16. Foda MA, Marmarou A. A new model of diffuse brain injury in
in vasospasm as compared to vehicle-treated animals. rats. Part II: morphological characterization. J Neurosurg 1994;80:
These results may have implications for the use of leve- 301–313.

Neurocritical Care ♦ Volume 5, 2006

78 Wang et al.

17. Marmarou A, Foda MA, van den Brink W, Campbell J, Kita H, analysis of data from randomized clinical trials. Epilepsy Res
Demetriadou K. A new model of diffuse brain injury in rats. Part I: 2005;64(1–2):1–11.
pathophysiology and biomechanics. J Neurosurg 1994;80:291–300. 29. Briggs DE, French JA. Levetiracetam safety profiles and tolera-
18. Schmued LC, Hopkins KJ. Fluoro-Jade B: a high affinity fluores- bility in epilepsy patients. Expert Opin Drug Saf 2004;3(5):415–424.
cent marker for the localization of neuronal degeneration. Brain 30. Lynch BA, Lambeng N, Nocka K, et al. The synaptic vesicle
Res 2000;874(2):123–130. protein SV2A is the binding site for the antiepileptic drug leveti-
19. Parra A, McGirt MJ, Sheng H, Laskowitz DT, Pearlstein RD, racetam. Proc Natl Acad Sci USA 2004;101(26):9861–9866.
Warner DS. Mouse model of subarachnoid hemorrhage associ- 31. Crowder KM, Gunther JM, Jones TA, et al. Abnormal neuro-
ated cerebral vasospasm: Methodological analysis. Neurol Res transmission in mice lacking synaptic vesicle protein 2A (SV2A).
2002;24:510–516. Proc Natl Acad Sci USA 1999;96(26):15,268–15,273.
20. Gao J, Wang H, Sheng H, et al. A novel apoE-derived therapeutic 32. Ignarro LJ, Byrns RE, Buga GM, Wood KS. Mechanisms of
reduces vasospasm and improves outcome in a murine model of endothelium-dependent vascular smooth muscle relaxation elic-
subarachnoid hemorrhage. Neurocritl Care 2006;4:25–31. ited by bradykinin and VIP. Am J Physiol 1987;253:H1074–H1082.
21. Hasan D, Schonck RSM, Avezaat CJJ, Tanghe HLJ, van Gijn J. van 33. Yanagisawa M, Kurihara H, Kimura S, Goto K, Masaki T. A novel
der Lugt PJM. Epileptic seizures after subarachnoid hemorrhage. peptide vasoconstrictor, endothelin, is produced by vascular
Ann Neurol 1993;33:286–291. endothelium and modulates smooth muscle Ca2+ channels. J
22. Rose FC, Sarner M. Epilepsy after ruptured intracranial aneu- Hypertens Suppl 1988;6:S188–S191.
rysm. Brit Med J 1965;5426:18–21. 34. Ito Y, Isotani E, Mizuno Y, Azuma H, Hirakawa K. Effective im-
23. Meek PD, Davis SN, Collins DM, et al. Guidelines for nonemer- provement of the cerebral vasospasm after subarachnoid hemor-
gency use of parenteral phenytoin products: proceedings of an rhage with low-dose nitroglycerin. J Cardiovasc Pharmacol 2000;
expert panel consensus process. Arch Intern Med 1999;159(22): 35:45–50.
2639–2644. 35. Pluta RM, Oldfield EH, Boock RJ. Reversal and prevention of ce-
24. Rhoney DH, Tipps LB, Murry KR, Basham MC, Michael DB, rebral vasospasm by intracarotid infusions of nitric oxide donors
Coplin WM. Anticonvulsant prophylaxis and timing of seizures in a primate model of subarachnoid hemorrhage. J Neurosurg
after aneurysmal subarachnoid hemorrhage. Neurology 2000; 1997;87:746–751.
55:258–265. 36. Wolf EW, Banerjee A, Soble-Smith J, Dohan FC, Jr., White RP,
25. Holtzer CD, Reisner-Keller LA. Phenytoin-induced thrombocy- Robertson JT. Reversal of cerebral vasospasm using an intrathecally
topenia. Ann Pharmacother 1997;31:435–437. administered nitric oxide donor. J Neurosurg 1998;89:279–288.
26. Ramsay RE, DeToledo J. Intravenous administration of fosphe- 37. Cardile V, Pavone A, Gulino R, Renis M, Scifo C, Perciavalle V.
nytoin: options for the management of seizures. Neurology 1996; Expression of brain-derived neurotrophic factor (BDNF) and in-
46:17–19. ducible nitric oxide synthase (iNOS) in rat astrocyte cultures
27. Goldstein LB. Potential effects of common drugs on stroke recov- treated with Levetiracetam. Brain Res 2003;976(2):227–233.
ery. Arch Neurol 1998;55:454–456. 38. Dagonnier M, Laute MA, Pandolfo M, Manto M. Effects of leveti-
28. Gidal BE, Baltes E, Otoul C, Perucca E. Effect of levetiracetam on racetam on the production of nitric oxide--an in vivo study.
the pharmacokinetics of adjunctive antiepileptic drugs: a pooled J Neurol 2005;252(6):727–730.

Neurocritical Care ♦ Volume 5, 2006