Journal of Clinical Sleep Medicine, Vol.5, No. 5, 2009 471
Sleepwalking or somnambulism, is a parasomnia consisting
of a series of complex behaviors usually initiated during
arousals from slow wave sleep and commonly culminate
in walking with an altered state of consciousness and impaired
judgment.1
Sleep related eating disorder (SRED) consists of
recurrent episodes of involuntary eating during arousals from
sleep.1
Parasomnias such as sleepwalking, SRED, and sleepdriving
can coexist and are rare side effects of zolpidem. In a
2005 National Institutes of Health consensus statement for the
treatment of chronic insomnia in adults zolpidem was considered
a hypnotic with limited risk.2
Two post-marketing studies
of zolpidem reported sleepwalking incidences of 7 of 1972 patients
(0.3%)3
and 1 of 96 patients (1%).4
We present a patient
with zolpidem-induced sleepwalking, SRED, and sleep-driving.
A fluorine-18-flourodeoxyglucose positron emission tomography
(18F-FDG-PET) was obtained one month after discontinuation
of zolpidem. A second 18F-FDG-PET was acquired the
following day, 1 h after oral administration of zolpidem 10 mg
(Figure 1). The cerebral glucose metabolism rates of the 2 studies
were then compared, using statistical parametric mapping
analysis. We also review the literature regarding unintended effects
of zolpidem use.
Case Report
The patient is a 51-year-old African American woman with
past medical history of hypertension, mild obstructive sleep apnea,
hyperlipidemia, and depression. Previous diagnostic polysomnogram
revealed an apnea-hypopnea index of 10 events/h.
Medications included paroxetine 20 mg once a day, extended
release metoprolol 25 mg twice per day, and simvastatin 40 mg
once a day. History for alcohol, tobacco, or illicit drug use was
negative. The patient reported no personal or family history of
sleepwalking or other parasomnias. The patient did not have a
history of daytime eating disorder. At age 44 the patient was
started on non-extended release zolpidem 10 mg at bedtime for
insomnia. A few weeks after starting zolpidem. she began sleep
related walking, eating, and one episode of driving.
Episodes of sleepwalking, SRED, and sleeptalking occurred
3 nights per week, 1-2 h after sleep onset. The patient would
speak incoherently using short phrases with her eyes closed and
would then open her eyes when questioned by her husband. She
would also leave her bedroom to go to her kitchen where she
would eat a loaf of bread, cold cereal, or leftover food. The following
morning she would have abdominal fullness, find her
kitchen messy, and have complete amnesia for the event. The
patient would also leave her home and walk on her front porch
or on her front lawn. As a preventive measure, she installed
nocturnal alarms on her doors to wake her or her family from
sleep if she opened one. Other reported events included one
occasion of urination in the hallway, and one episode when
the patient drove her automobile 10 miles from her home and
Zolpidem-Induced Sleepwalking, Sleep Related Eating Disorder, and Sleep-Driving:
Fluorine-18-Flourodeoxyglucose Positron Emission TomographyAnalysis, and a
Literature Review of Other Unexpected Clinical Effects of Zolpidem
Romy Hoque, M.D.; Andrew L. Chesson, Jr., M.D.
Sleep Disorders Center, Department of Neurology, Louisiana State University School of Medicine, Shreveport, LA
Submitted for publication December, 2008
Submitted in final revised form April, 2009
Accepted for publication April, 2009
Address correspondence to: Romy Hoque, M.D., Department of Neurology,
Louisiana State University School of Medicine, 1501 Kings Highway,
Shreveport, LA; Email: romy.hoque@gmail.com
Case Report
Zolpidem is a hypnotic which acts at the GABAA
receptor and is indicated
for short-term insomnia. Sleep related disorders including
somnambulism, sleep related eating and sleep-driving have been reported
with zolpidem. A 51-year-old insomniac who used zolpidem 10
mg nightly starting at 44 years of age is described. A few weeks after
starting zolpidem she began walking, eating, and had one episode of
driving while asleep. Episodes of sleep related eating, sleepwalking,
and sleeptalking occurred 3 nights per week, 1 to 2 h after sleep onset.
After her evaluation, the patient’s zolpidem was gradually discontinued,
and all sleep related activities immediately ceased. An 18F-FDGPET
was obtained 2 months after discontinuation of zolpidem. The
following day, FDG was administered 1 h after oral administration of
10 mg zolpidem, and then a second PET was performed. We report
the results and a review of the literature regarding other unintended
effects seen with zolpidem use.
Keywords: Zolpidem, fluorine-18-flourodeoxyglucose positron emission
tomography, sleep related eating disorder, sleepwalking, sleep-
driving
Citation: Hoque R; Chesson AL. Zolpidem-induced sleepwalking,
sleep related eating disorder, and sleep-driving: fluorine-18-flourodeoxyglucose
positron emission tomography analysis, and a literature
review of other unexpected clinical effects of zolpidem. J Clin Sleep
Med 2009;5(5):471-476.
Downloadedfromjcsm.aasm.orgby109.81.215.43onJuly23,2020.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2020AmericanAcademyofSleepMedicine.Allrightsreserved.
Journal of Clinical Sleep Medicine, Vol.5, No. 5, 2009 472
R Hoque, AL Chesson
was found asleep behind the wheel by police. She had a vague
recollection of this event, but thought that she was dreaming.
After her evaluation, the patient’s zolpidem was gradually withdrawn;
all sleep related activities immediately ceased and have
not recurred during 6 months of follow-up.
18F-FDG-PET Analysis
The first PET study in our patient was performed after discontinuation
of zolpidem for 2 months. A second PET study
FDG was administered 1 h after the administration of 10 mg
zolpidem. The patient was asleep in the scanner during both
PET studies. Statistical parametric mapping comparison between
our patient’s 2 studies demonstrated no significant differences
(Figure 1).
DISCUSSION
Zolpidem is an imidazopyridine drug indicated for shortterm
insomnia at a dosage usually ranging from 5 to 10 mg
per day.5
Though considered a non-benzodiazepine since its
imidazopyridine structure differs from benzodiazepine fusion
of benzene and diazepine, zolpidem is a benzodiazepine receptor
agonist with high binding affinity for the GABAA
(gammaamino
butyric acid type A) receptor expressing the α1
subunit.
Benzodiazepines and benzodiazepine receptor agonists like
zolpidem bind to the GABAA
receptor at sites that are distinct
from the GABAbinding site, thereby allosterically affecting the
activity of the ligand-operated chloride channel.
GABA is the main inhibitory neurotransmitters in the mammalian
central nervous system (CNS). GABAA
receptors exist
as pentameric protein complexes, assembled from a combination
of at least 19 subunits from 7 distinct gene families (α, β,
γ, δ, ε, θ, and π). Synaptic GABAA
receptors are responsible
for modulating benzodiazepine sensitivity and typically contain
α1, 2, 3, or 5
, β2 or 3
, and the γ2
subunits.6
GABAA
receptor sensitivity
to benzodiazepines is mediated through α subunits. Benzodiazepines
bind to synaptic GABAA
receptors containing α1
, α2
,
α3
, or α5
subunits with comparable affinity. The GABAA
receptor
expressing the α1
subunit corresponds to the benzodiazepine
ω1
receptor.7
GABAA
receptors containing the α1
, α2
, α3
, or α5
subunits correspond to ω2
benzodiazepine receptors. The ω3
benzodiazepine receptor is not related to the GABAA
receptor.
Extrasynaptic GABAA
receptors are primarily composed of α4-6
subunits in combination with δ subunits, and are insensitive to
benzodiazepines. The current benzodiazepine receptor nomenclature
(ω1
, ω2
, and ω3
) replaced the previous anatomical localization
classification (central benzodiazepine receptor type 1,
central BZ-1; central benzodiazepine receptor type 2, central
BZ-2; and peripheral benzodiazepine receptor type 3, BZ3)
because of the existence of “central” benzodiazepine receptors
with peripheral localization, and “peripheral” benzodiazepine
receptors with central localization.
Zolpidem was developed as a drug with a structure different
from benzodiazepines, allowing affinity for only a given
subset of central benzodiazepine receptors resulting in hypnotic
properties without additional anticonvulsant and myorelaxant
properties of benzodiazepines. In contrast to benzodiazepines
Figure 1—18-fluorine-flourodeoxyglucose-positron emission tomography (18F-FDG-PET) of a patient with zolpidem induced sleepwalking,
sleep related eating disorder, and sleep-driving. A: 18F-FDG-PET off zolpidem. B: 18F-FDG-PET on zolpidem. FDG was administered to patient
1 h after ingestion of 10 mg zolpidem. Statistical parametric mapping comparison of the 2 sequences shows no significant differences.
Downloadedfromjcsm.aasm.orgby109.81.215.43onJuly23,2020.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2020AmericanAcademyofSleepMedicine.Allrightsreserved.
Journal of Clinical Sleep Medicine, Vol.5, No. 5, 2009 473
Zolpidem’s Unexpected Clinical Effects
like clonazepam, diazepam and flunitrazepam, which lack selectivity
for the ω1
, ω2
, or ω3
benzodiazepine receptor subtypes;
zolpidem has a high affinity for ω1
.8
A possible explanation for zolpidem-induced nocturnal
events is that after an arousal from sleep into wakefulness, nocturnal
activity (i.e., walking, eating, or driving) occurred and
was subsequently not recalled after returning to sleep because
of the sedation-mediated amnestic properties of zolpidem. Another
possibility is that an arousal occurred out of slow wave
sleep with the parasomnia occurring in electroencephalographically
verifiable sleep. We felt our patient experienced the later,
given her incoherent interactions with her husband during her
nocturnal events. Patients who do not recall waking events on
zolpidem are typically cognitively functional, and retain the
ability to speak in coherent short phrases.9
Sleepwalking is a relatively common condition affecting 10%
of adults.1
Recently hotels across the United Kingdom reported
an increase in the number of hotel guests found to be sleepwalk-
ing.10
Though the incidence of zolpidem induced sleepwalking
has been reported to be low, it is possible that many cases of
unexplained sleepwalking may be secondary to zolpidem given
its widespread use.3,4
Along with sleepwalking, SRED, and sleep-driving parasomnias,
zolpidem has been anecdotally reported to produce a
range of unexpected beneficial effects. These include improvement
in the following conditions: post-stroke Broca’s aphasia;
blepharospasm; quadriparesis of central pontine myelinolysis;
catatonia of schizoaffective disorder; dementia with apraxia;
post-anoxic minimally conscious states; bradykinesia, akinesia,
and dystonia in Parkinson disease; post-levodopa dyskinesias
in Parkinson disease; vertical saccadic eye movements and
parkinsonism in progressive supranuclear palsy; restless legs
syndrome; post-anoxic spasticity; and spinocerebellar ataxia
(Table S1 summarizes the available reports of improvement in
varied neurological conditions with zolpidem use). Effects were
usually noted within 30 min of ingestion of the non-extended
release formulation and lasted for 2 to 4 h, corresponding with
a time to peak plasma contraction of approximately 1.2 h and a
half-life of approximately 2.5 h.
Zolpidem effects might be mediated through its anti-anxiety
effects, its benzodiazepine receptor agonist properties, its
GABAergic activity, or some combination of all three. For example,
symptoms of Parkinson disease worsen with anxiety.
The improvement noted in Parkinson disease with zolpidem use
may be secondary to its anxiolytic effect through a GABAergic
effect on the limbic system or elsewhere. The improvement
seen in blepharospasm, catatonia, and restless legs syndrome
may be caused by the benzodiazepine ω1
receptor agonist activity
of zolpidem. However, opposing this theory of purely
benzodiazepine agonist mediated effects, is that parasomnias
like sleepwalking are often treated with benzodiazepines like
clonazepam; yet zolpidem seems to induce parasomnias in a
susceptible subpopulation.
The action of zolpidem via synaptic GABAA
receptors with
α1 subunits may produce different clinical responses depending
upon regional distribution of receptor subtypes. Benzodiazepines
bind to all the synaptic GABAA
receptors, which are
expressed throughout the nervous system. Even though zolpidem
is a preferred α1
agonist, α1
subunits are expressed widely
throughout the CNS.11
Benzodiazepine-insensitive extrasynaptic
GABAA
receptors containing α4-6
subunits show much more
regional specificity than benzodiazepine-sensitive synaptic
GABAA
receptors containing α1, 2, 3, or 5
.
Zolpidem has a less recognized but limited binding affinity
to ω2
benzodiazepine receptors. ω1
and ω2
receptors are also
widely expressed throughout the human brain.12
At higher doses
these lower binding affinities may be expressed resulting in
unexpected clinical outcomes. For example, anecdotally there
appears to be differential efficacy of high dose zolpidem (70
mg/d) for blepharospasm, and low dose zolpidem (5-10 mg/d)
for parkinsonian features. (Surprisingly at the high doses used
by Garretto et. al for blepharospasm and Evidente for early onset
Parkinson disease, 5 of 6 patients reported no somnolence,
and only one patient had to discontinue the medication secondary
to drug-induced diarrhea.13,14
Somnolence was overcome
with slow dose titration.)
The potential clinical significance of preferred GABAA
α1
subunit/ω1
receptor activation is unclear. For example, it was
previously thought that zolpidem did not possess significant
myorelaxant properties similar to benzodiazepines. However,
anecdotal reports of efficacy for zolpidem in post-anoxic
spasticity, parkinsonian dyskinesias/tremors, blepharospasm,
and restless legs syndrome provides anecdotal evidence to the
contrary. Zolpidem may affect many neurological diseases
through binding at a variety of locations simultaneously (Figure
2). Zolpidem binding at one anatomical location is unlikely
to explain all of its myriad effects. Also, electrophysiological
studies suggest that different GABA subunit combinations may
mediate different physiological and pharmacological properties
of the ligand-operated ion channel.11
Therefore, even though
zolpidem has a high affinity for GABAA
receptors with the α1
subunit, different pharmacological responses may results from
different subunit combinations with the α1
subunit. As a result,
clinical efficacy in a given disease is difficult to correlate with
binding and receptor activation at a single GABAA
/benzodiazepine
receptor type, at a single anatomic site, or at a single
dose.
An intriguing theory on the etiology of sleepwalking and
SRED concerns the presence of theoretical cerebral pattern
generators (CPGs).15,16
CPG are thought to be neuronal collections
in the brain, brainstem or spinal cord that can potentially
control innate motor behaviors essential for survival like
feeding and locomotion. Diffuse zolpidem cortical binding
may cause release of CPGs associated with evolutionarily conserved
motor patterns such as walking and eating, leading to
subsequent disorders of arousal like somnambulism and SRED.
Since some CPGs may reside in the cortex, zolpidem use also
release cortical patterns associated with overlearned behaviors,
such as driving.
In an attempt to identify zolpidem-induced changes in cerebral
glucose metabolic rates, an 18F-FDG-PET was performed in our
patient on and off zolpidem. Gillin et al. compared the effects
of 10 mg zolpidem and placebo on cerebral glucose metabolic
rates in 12 young normal volunteers (mean age: 22.5 y) using
18F-FDG-PET.17
In that study FDG was administered about 1
h after oral administration of zolpidem while the patient was in
electroencephalographically (EEG) verifiable stage 2 sleep and
at a time of expected zolpidem peak concentrations (1.2 ± 0.2 h)
Downloadedfromjcsm.aasm.orgby109.81.215.43onJuly23,2020.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2020AmericanAcademyofSleepMedicine.Allrightsreserved.
Journal of Clinical Sleep Medicine, Vol.5, No. 5, 2009 474
somnias are thought to emerge. Future 18F-FDG-PET studies
in patients with zolpidem induced parasomnias could be attempted
during EEG verified SWS on and off zolpidem to identify
differences in glucose metabolic rates not seen in our analysis.
However, this may prove to be difficult given the variable
presence of SWS and the need to wait 1 h after FDG injection
prior to the PET scan. By the time the patient receives the FDG
injection and the scan is performed, the patient may no longer
be in SWS. And though parasomnias tend to emerge in SWS,
they can potentially arise from any NREM stage.
Temporal resolution is not a major limitation of PET studies
in that in order to scan a brain the cortical area is divided into
thirds and scanned in 3 successive 5-min sessions that are then
compiled together to form an entire cortical scan. As a result,
the PET findings in any particular cortical area are an estimation
of glucose metabolic rate over a five minute time window.
Despite this small time window, 18F-FDG-PET findings would
be difficult to correlate with a particular arousal in a period of
SWS. Similar studies may be performed on and off medicain
plasma (and presumably brain). Gillin et al. found that across
all cortical areas glucose metabolic rates were not significantly
different on placebo versus zolpidem. Our patient’s results were
similar to those seen in Gillin’s normal volunteers.
Compared to wake, whole brain cortical glucose metabolic
rates decrease in NREM and REM stage sleep.18
One would expect
a decline in cortical glucose metabolic rate with the use of
sleep-inducing hypnotics like zolpidem. However, our results,
along with those of Gillin show otherwise. The reasons for this
are unknown.
One possible explanation may be that PET is too insensitive
a tool to detect subtle localized or generalized glucose metabolic
rate differences on and off zolpidem in the normal brain. If
such differences could be identified, then patients who are susceptible
to developing parasomnias could be identified prior to
use. More importantly, this may also provide insight into other
extraordinary anecdotal effects of zolpidem.
A limitation of our PET analysis was that the scan was not
performed in EEG verified slow wave sleep (SWS), when paraR
Hoque, AL Chesson
Figure 2—Regional distribution of zolpidem binding, and the potential clinical consequences. Zolpidem is a benzodiazepine receptor
agonist with high binding affinity for the GABAA
(gamma-amino butyric acid type A) receptor expressing the α1
subunit. Benzodiazepines
and benzodiazepine receptor agonists like zolpidem bind to the GABAA
receptor at sites that are distinct from the GABA binding site,
thereby allosterically affecting the channel. GABAA
receptor sensitivity to benzodiazepines is mediated through α subunits. Zolpidem’s
action via synaptic GABAA
receptors with α1 subunits may produce different clinical responses depending upon regional distribution of
receptor subtypes. Benzodiazepines bind to all the synaptic GABAA
receptors, which are expressed throughout the nervous system. Even
though zolpidem is a preferred α1
agonist, α1
subunits are expressed widely throughout the CNS. Given zolpidem’s many binding sites,
the improvement noted across a range of neurological disorders are difficult to localize to binding at a single anatomic location. GABAA
α1
subunits/ω1
benzodiazepine receptors are widely distributed throughout the central nervous system, in many more areas than indicated
in this simple schematic figure.
Downloadedfromjcsm.aasm.orgby109.81.215.43onJuly23,2020.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2020AmericanAcademyofSleepMedicine.Allrightsreserved.
Journal of Clinical Sleep Medicine, Vol.5, No. 5, 2009 475
al patients may also provide insight into potentially identifiable
pharmacogenetic vulnerabilities/susceptibilities. These exciting
and unexplored avenues of research may be used in the treatment
of disease previously thought untreatable.
Acknowledgments
We thank David Lilien, M.D., for performing the 18F-FDGPET
studies; James Patterson, M.D., for performing a statistical
parametric mapping comparison of the two 18F-FDG-PET
studies; Eduardo Gonzalez-Toledo, M.D., Ph.D., for help in
creating Figure 1; and Richard Zweig, M.D., for comments on
the manuscript.
Disclosure Statement
This was not an industry supported study. The authors have
indicated no financial conflicts of interest.
References
1.	 American Academy of Sleep Medicine. The international classification
of sleep disorders: diagnostic & coding manual. 2nd ed.
Westchester, IL: American Academy of Sleep Medicine, 2005.
2.	 National Institutes of Health State of the Science Conference
statement on Manifestations and Management of Chronic Insomnia
in Adults, June 13-15, 2005. Sleep 2005;28:1049-57.
3.	 Ganzoni E, Santoni JP, Chevillard V, Sebille M, Mathy B. Zolpidem
in insomnia: a 3-year post-marketing surveillance study in
Switzerland. J Int Med Res 1995;23:61-73.
4.	 Sauvanet JP, Maarek L, Roger M, Renaudin J, Louvel E, Orofiamma
B. Open long-term trials with zolpidem in insomnia. In:
Sauvanet JP, Langer SZ, Morselli PL, eds. Imidazopyridines in
sleep disorders. New York: Raven Press, 1988:339-49.
5.	 Roehrs T, Vogel G, Vogel F, et al. Eligibility requirements in hypnotic
trials. Sleep 1985;8:34-9.
6.	 Harrison NL. Mechanisms of sleep induction by GABA(A) receptor
agonists. The J Clin Psychiatry 2007;68(Suppl 5):6-12.
7.	 Holm KJ, Goa KL. Zolpidem: an update of its pharmacology,
therapeutic efficacy and tolerability in the treatment of insomnia.
Drugs 2000;59:865-89.
8.	 Langer SZ, Arbilla S, Scatton B, Niddam R, Dubois A. Receptors
involved in the mechanism of action of zolpidem. In: Sauvanet JP,
Langer SZ, Morselli PL, eds. Imidazopyridines in sleep disorders.
New York: Raven Press, 1988:55-69.
9.	 Canaday BR. Amnesia possibly associated with zolpidem administration.
Pharmacotherapy 1996;16:687-9.
10.	 Collis R. Hotels wake up to sleepwalking clients. 2007 [cited
2007 June 13, 2008]; Available from: http://www.iht.com/ar-
ticles/2007/11/15/travel/trfreq16.php
11.	 Pirker S, Schwarzer C, Wieselthaler A, Sieghart W, Sperk G.
GABA(A) receptors: immunocytochemical distribution of 13
subunits in the adult rat brain. Neuroscience 2000;101:815-50.
12.	 Dennis T, Dubois A, Benavides J, Scatton B. Distribution of central
omega 1 (benzodiazepine1) and omega 2 (benzodiazepine2)
receptor subtypes in the monkey and human brain. An autoradiographic
study with [3H]flunitrazepam and the omega 1 selective
ligand [3H]zolpidem. J Pharmacol Exp Ther 1988;247:309-22.
13.	 Garretto NS, Bueri JA, Rey RD, Arakaki T, Nano GV, Mancuso
M. Improvement of blepharospasm with Zolpidem. Mov Disord
2004;19:967-8.
14.	 EvidenteVG.Zolpidemimprovesdystoniain“Lubag”orX-linked
dystonia-parkinsonism syndrome. Neurology 2002;58:662-3.
tion in NREM and REM to assess differences in brain cortical
glucose metabolic rates, though the procedural limitations described
above would still apply.
Single photon emission computed tomography (SPECT)
studies have been used successfully to show increased cerebral
blood flow in a range of cortical areas after zolpidem administration
despite a more limited spatial resolution than PET.
SPECT has been used to show increased regional blood flow
in the frontal cortex in Broca aphasia, the cerebellum in spinocerebellar
ataxia, and the contralateral hemisphere in hemiparetic
patients.19-21
In normal baboon models, SPECT has been
used to demonstrate that zolpidem does not cause changes in
regional cerebral blood flow in normal baboons. However, in
baboons with cortical injuries, zolpidem increased blood flow
to the injured areas.22
Zolpidem mediated increase in regional
cerebral blood flow to injured cortical areas on SPECT was attenuated
by the use of flumazenil, a benzodiazepine receptor
antagonist.23
The baboon studies correlate to the case report of
Brefel-Courbon et al. of a patient in a post-anoxic minimally
conscious state showing arousal on clinical exam and increased
cerebral glucose metabolism on 18F-FDG-PET in the bilateral
post-rolandic territories and frontal lobes after zolpidem ad-
ministration.24
The normal baboon SPECT study findings also
correlate with the 18F-FDG-PET findings in our neurologically
intact patient and Gillin’s normal volunteer cohort.17
To date no large scale randomized controlled trials exist assessing
the efficacy of zolpidem for aphasia, blepharospasm,
catatonia, central pontine myelinolysis, dementia with apraxia,
Parkinson disease, progressive supranuclear palsy, restless legs
syndrome, post-anoxic spasticity, or spinocerebellar ataxia. The
clinical benefit of zolpidem for patients in minimally conscious
states is currently being explored in clinical trials.25
These results
may also help to further understand sleep-wake mechanisms
and the function of hypnotics.
The anecdotal benefits of zolpidem have provided hope that
damage to brain tissue after strokes anoxic insults previously
thought to be permanent may actually be reversible. Zolpidem
may reactivate cortical areas that have undergone injury-induced
dormancy, or there may be more redundancy built into
our brains than previously believed, e.g. CPGs. GABAergic
hypnotics like zolpidem through diffuse cortical binding may
somehow unmask this redundancy.
Future studies may also shed light on whether different susceptibilities
to zolpidem induced parasomnias and its other effects
may depend upon the formulation used. For example, Chiang et
al. reported 2 patients who experienced zolpidem induced sleepwalking
and SRED on only the extended release formulation and
not the non-extended release formulation.26
Validation of these
anecdotal findings and investigations into the new sublingual
formulation of zolpidem may provide insight into how formulation
dependent pharmacokinetics may influence an individual’s
susceptibility to zolpidem-induced parasomnias.27
Investigations into the mechanisms of action of GABAergic
induced parasomnias may overturn therapeutic nihilism for a
variety of neurological disease. Capitalizing upon zolpidem’s
myriad anecdotal serendipitous effects, basic science research
using animal models of non–sleep-wake related neurological
disorders may provide us with a of understanding how the brain
reorganizes itself after injury. Also genetic analysis of individuZolpidem’s
Unexpected Clinical Effects
Downloadedfromjcsm.aasm.orgby109.81.215.43onJuly23,2020.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2020AmericanAcademyofSleepMedicine.Allrightsreserved.
Journal of Clinical Sleep Medicine, Vol.5, No. 5, 2009 476
31.	 JarryC,FontenasJP,Jonville-BeraAP,Autret-LecaE.Beneficialeffect
of zolpidem for dementia. Ann Pharmacother 2002;36:1808.
32.	 Clauss RP, Guldenpfennig WM, Nel HW, Sathekge MM, Venkannagari
RR. Extraordinary arousal from semi-comatose state on
zolpidem. A case report. S Afr Med J 2000;90:68-72.
33.	 Clauss R, Nel W. Drug induced arousal from the permanent vegetative
state. NeuroRehabilitation 2006;21:23-8.
34.	 Cohen SI, Duong TT. Increased arousal in a patient with anoxic
brain injury after administration of zolpidem. Am J Phys Med
Rehabil 2008;87:229-31.
35.	 Shames JL, Ring H. Transient reversal of anoxic brain injuryrelated
minimally conscious state after zolpidem administration:
a case report. Arch Phys Med Rehabil 2008;89:386-8.
36.	 Daniele A, Albanese A, Gainotti G, Gregori B, Bartolomeo P.
Zolpidem in Parkinson’s disease. Lancet 1997;349:1222-3.
37.	 Ruzicka E, Roth J, Jech R, Busek P. Subhypnotic doses of zolpidem
oppose dopaminergic-induced dyskinesia in Parkinson’s disease.
Mov Disord 2000;15:734-5.
38.	 Farver DK, Khan MH. Zolpidem for antipsychotic-induced parkinsonism.
Ann Pharmacother 2001;35:435-7.
39.	 Daniele A, Moro E, Bentivoglio AR. Zolpidem in progressive supranuclear
palsy. New Engl J Med 1999;341:543-4.
40.	 Mayr BJ, Bonelli RM, Niederwieser G, Koltringer P, Reisecker
F. Zolpidem in progressive supranuclear palsy. Eur J Neurol
2002;9:184-5.
41.	 Gericke CA, Ludolph AC. Chronic abuse of zolpidem. JAMA
1994;272:1721-2.
42.	 Bezerra ML, Martinez JV. Zolpidem in restless legs syndrome.
Eur Neurol 2002;48:180-1.
43.	 Mendelson WB. Sleepwalking associated with zolpidem. J Clin
Psychopharmacol 1994;14:150.
44.	 Harazin J, Berigan TR. Zolpidem tartrate and somnambulism.
Mil Med 1999;164:669-70.
45.	 Morgenthaler TI, Silber MH. Amnestic sleep-related eating disorder
associated with zolpidem. Sleep Med 2002;3:323-7.
46.	 Sattar SP, Ramaswamy S, Bhatia SC, Petty F. Somnambulism due
to probable interaction of valproic acid and zolpidem. Ann Pharmacother
2003;37:1429-33.
47.	 Sharma A, Dewan VK. A case report of zolpidem-induced
somnambulism. Primary care companion to J Clin Psychiatry
2005;7:74.
48.	 Najjar M. Zolpidem and amnestic sleep related eating disorder. J
Clin Sleep Med 2007;3:637-8.
49.	 Sansone RA, Sansone LA. Zolpidem, somnambulism, and nocturnal
eating. Gen Hosp Psychiatry 2008;30:90-1.
50.	 Shadan FF, Poceta JS, Kline LE. Zolpidem for postanoxic spasticity.
South Med J 2004;97:791-2.
15.	 Tassinari CA, Rubboli G, Gardella E, et al. Central pattern generators
for a common semiology in fronto-limbic seizures and
in parasomnias. A neuroethologic approach. Neurol Sci 2005;26
Suppl 3:s225-32.
16.	 Yuste R, MacLean JN, Smith J, Lansner A. The cortex as a central
pattern generator. Nature Rev 2005;6:477-83.
17.	 Gillin JC, Buchsbaum MS, Valladares-Neto DC, et al. Effects
of zolpidem on local cerebral glucose metabolism during nonREM
sleep in normal volunteers: a positron emission tomography
study. Neuropsychopharmacology 1996;15:302-13.
18.	 Buchsbaum MS, Hazlett EA, Wu J, Bunney WE Jr. Positron
emission tomography with deoxyglucose-F18 imaging of sleep.
Neuropsychopharmacology 2001;25(5 Suppl):S50-6.
19.	 Cohen L, Chaaban B, Habert MO. Transient improvement of
aphasia with zolpidem. New Engl J Med 2004;350:949-50.
20.	 Clauss RP, Nel WH. Effect of zolpidem on brain injury and diaschisis
as detected by 99mTc HMPAO brain SPECT in humans.
Arzneimittelforschung 2004;54:641-6.
21.	 Clauss R, Sathekge M, Nel W. Transient improvement of spinocerebellar
ataxia with zolpidem. New Engl J Med 2004;351:511-2.
22.	 Clauss RP, Dormehl IC, Oliver DW, Nel WH, Kilian E, Louw WK.
Measurement of cerebral perfusion after zolpidem administration
in the baboon model. Arzneimittelforschung 2001;51:619-22.
23.	 Clauss RP, Dormehl IC, Kilian E, Louw WK, Nel WH, Oliver
DW. Cerebral blood perfusion after treatment with zolpidem and
flumazenil in the baboon. Arzneimittelforschung 2002;52:740-4.
24.	 Brefel-Courbon C, Payoux P, Ory F, et al. Clinical and imaging
evidence of zolpidem effect in hypoxic encephalopathy. Ann
Neurol 2007;62:102-5.
25.	 May 2008 [cited; Available from: www.regentherapeutics.com
26.	 Chiang A, Krystal A. Report of two cases where sleep related
eating behavior occurred with the extended-release formulation
but not the immediate-release formulation of a sedative-hypnotic
agent. J Clin Sleep Med 2008;4:155-6.
27.	 Roth T, Krystal AD, Maguire Y, Singh N, Maytom M. Pharmacokinetics
of the sublingual zolpidem tartrate 3.5 mg lozenge
compared to the oral zolpidem tartrate 10 mg tablet. Sleep
2008;31(Abstract Supplement):A235.
28.	 Thomas P, Rascle C, Mastain B, Maron M, Vaiva G. Test for catatonia
with zolpidem. Lancet 1997;349:702.
29.	 Wang WT, Chen YY, Wu SL, Wei TS, Liu SY. Zolpidem dramatically
improved motor and speech function in a patient with
central pontine myelinolysis. Eur J Neurol 2007;14:e9-10.
30.	 Tsai MJ, Tsai YH, Huang YB. Compulsive activity and anterograde
amnesia after zolpidem use. Clin Toxicol (Philadelphia,
PA) 2007;45:179-81.
R Hoque, AL Chesson
Downloadedfromjcsm.aasm.orgby109.81.215.43onJuly23,2020.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2020AmericanAcademyofSleepMedicine.Allrightsreserved.
Journal of Clinical Sleep Medicine, Vol.5, No. 5, 2009 S1
Zolpidem’s Unexpected Clinical Effects
Table S1—Zolpidem: A Spectrum of Unintended Effects and Potential Uses. SRED: sleep-related eating disorder.
Reference Age, sex, and clinical state Dose Description of response
Aphasia
Cohen et al. 200419
52, female, stroke of left
insula, putamen, and superior
temporal gyrus with resultant
Broca aphasia
Zolpidem 10 mg once/d
Time to response: 20 minutes
Duration of response: unknown
Patient regained effective speech,
naming, and repetition.
Blepharospasm,
Meige syndrome
(oromandibular
dystonia and
blepharospasm)
Garretto et al. 200413
Three patients
Patient 1: 57, woman, Meige
syndrome
Patient 2: 63, male,
blepharospasm
Patient 3: 66, male, Meige
syndrome
Patient 1: Zolpidem 10 mg every 2-3 h
(70 mg/d)
Time to response: 30 min
Duration of response: 3 h
Patient 2: Zolpidem 10 mg 3 times/d
Time to response: 20-30 min
Duration of response: 3 h
Patient 3: Zolpidem 10 mg twice a day
Time to response: 30 min
Duration of response: 3 h
Blepharospasm improved in all 3.
Effect on oromandibular dystonia
in the 2 patients with Meige
syndrome was not mentioned.
Catatonia
Thomas et al. 199728
21, woman, schizoaffective
disorder. One month
after discontinuation of
haloperidol, patient developed
severe oppositional behavior,
mutism, staring, posturing,
and refusal of food.
Zolpidem 10 mg once/d
Time to response: 15 min
Duration of response: 4 h
All behavior and motor symptoms
resolved on zolpidem.
Central pontine
myelinolysis
Wang et al. 200729
51, female, post-surgical
complication resulting in
central pontine myelinolysis
with spastic quadriparesis and
pseudobulbar palsy
Zolpidem 10 mg once/d
Time to response: 20 min
Duration of response: 4 h
Regained facial expression,
effective speech, and smooth limb
movements
Compulsive behavior
while awake
Tsia et al. 200730
Three previously healthy
patients
Patient 1: 34, female
Patient 2: 40, female
Patient 3: 50, female
Zolpidem 10 mg once/d
All 3 patients exhibited compulsive
behaviors (cleaning, shopping, and
eating) with anterograde amnesia for
the events.
Events were not considered to
be sleepwalking because patients
were able to communicate
fluently and perform normally
beyond the compulsive activity.
Events stopped with cessation of
zolpidem in all 3 patients.
Dementia with
apraxia
Jarry et al. 200231
60, woman, alcoholic,
reduced cognition, memory
loss, apraxia, inability to join
in conversation
Zolpidem 10 mg once/d
Time to response: 45-60 min
Duration of response: 3 h
Improvement in cognition and
praxis. Regained the ability to
perform simple chores around the
home.
Minimally conscious
states
Claus et al. 200032
28, male, motor vehicle
accident resulting in postanoxic
minimally conscious
state
Zolpidem 10 mg once/d,
Time to response: 15 min
Duration of response not explicitly
stated
Interacted spontaneously and
responded to simple questions
with short appropriate answers
Downloadedfromjcsm.aasm.orgby109.81.215.43onJuly23,2020.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2020AmericanAcademyofSleepMedicine.Allrightsreserved.
Journal of Clinical Sleep Medicine, Vol.5, No. 5, 2009 S2
R Hoque, AL Chesson
Reference Age, sex, and clinical state Dose Description of response
Claus et al. 200633
Three patients:
Patient L: 31, male,
Patient N: 31, male,
Patient G: 29, male.
All 3 had motor vehicle
accidents resulting in postanoxic
minimally conscious
state
Doses unknown.
Time to response: L: 20-30 min, N: 1
h; G: 20-30 min. Duration of response
was 4 h for all 3 patients.
L: meaningful interactions/
conversations with caregivers,
purposeful response to stimuli,
voluntary behavior.
N: Cessation of random
screaming episodes. Watched
television and laughed at funny
scenes. Stated his name and age.
G: Meaningful interactions with
caregivers. Lifted hands, counted
to 5, and smiled on command.
Brefel-Courbon et al.
200724
48, male, suicide attempt by
hanging resulting in postanoxic
encephalopathy with
minimally conscious state
Zolpidem 10 mg once/d
Time to response: 20 min
Duration of response: 2-3 h
Increased arousal; communicated
with family; swallowed food;
moved spontaneously in bed
Cohen et al. 200834
35, male, cardiac arrest
resulting in post-anoxic
encephalopathy with
minimally conscious state
Zolpidem 5 mg once a day → 10 mg in
the morning and 5 mg in the evening →
10 mg twice/d,
Time to response: 1 h
Duration of response: 3-6 h depending
on dosing
Zolpidem 5 mg once a day:
Increased arousal and more
interactive with caregivers. Could
toss a football
Zolpidem 10 mg/ 5 mg:
Walked with assistance
Zolpidem 10 mg/ 10 mg:
Spoke more frequently and
fluently. Walked unassisted.
Shames et al. 200835
50, woman, cardiac arrest
resulting in post-anoxic
encephalopathy with
minimally conscious state
Zolpidem 10 mg once/d
Time to response: 30 min
Duration of response: 3 h
Increased arousal. Alert and
oriented to self and place.
Cessation of athetoid movements.
Regained full voluntary
movements of all 4 extremities.
Parkinson disease,
Parkinsonian
features
Daniele et al. 199736
10 patients, mean age 69.9 Zolpidem 10 mg once/d
In the responders (6/10 patients):
Time to response: 45-60 min
Duration of response: 2-4 h
6/10 patients showed motor
improvement in facial expression,
rigidity, akinesia, bradykinesia,
posture and gait
Ruzicka et al. 200037
45, female, Parkinson disease
with generalized choreic
dyskinesia post-levodopa
administration
Zolpidem 5 mg per day
Time to response: 30 min
Duration of response: 2 h
Limb and trunk dyskinesia ceased
on zolpidem
Farver et al. 200138
31, male, schizoaffective
disorder, illicit substance
abuse (cocaine,
amphetamines, marijuana)
Patient developed
parkinsonian rest tremor on
antipsychotic medications.
Tremor failed to respond to
lowering the antipsychotic
dose, adding anticholinergic
or dopaminergic medications,
or switching to novel
antipsychotics.
Zolpidem 5 mg 4 times/d The patient’s tremors resolved
within 1 month of zolpidem.
With withdrawal of zolpidem the
tremors returned. They ceased
once again when zolpidem was
restarted.
Downloadedfromjcsm.aasm.orgby109.81.215.43onJuly23,2020.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2020AmericanAcademyofSleepMedicine.Allrightsreserved.
Journal of Clinical Sleep Medicine, Vol.5, No. 5, 2009 S3
Zolpidem’s Unexpected Clinical Effects
Reference Age, sex, and clinical state Dose Description of response
Evidente 200214
Three patients with X-linked
early onset parkinsonism with
dystonia (Lubag syndrome)
Patient 1: 41, male
Patient 2: 38, male
Patient 3: 36, male
Patient 1: Zolpidem 10 mg every 2 h
Time to response: 15 min
Duration of response: 2 h
Patient 2: Zolpidem 10 mg twice/d
Time to response: 40 min
Duration of response: 3 h
Patient 3: Zolpidem 10 mg twice/d
Time to response: 45 min
Duration of response: 2 h
All three patients had
improvement in dystonia and
parkinsonism on zolpidem
Progressive
supranuclear palsy
(PSP)
Daniele et al. 199939
10 patients with PSP, mean
age: unknown
Crossover trial: All patients received
zolpidem 5 mg once a day, zolpidem 10
mg once a day, levodopa-carbidopa 250
mg-25 mg, and placebo.
Time to response to zolpidem: 40-60 min
Duration of response to zolpidem: 2 h
Zolpidem, unlike levodopacarbidopa
or placebo, improved
saccadic eye movements in 4/10
patients on Zolpidem 10 mg/d
Mayr et al. 2002 40
61, male Zolpidem 5 mg/d
Time to response and duration of
response: unknown
Patient showed improvement
in vertical gaze palsy and
parkinsonism. Patient showed
response for 4 weeks. Second
trial of zolpidem 2 months later
showed no benefit.
Psychostimulant
effect
Gericke et al. 199441
33, male, depression Zolpidem 80 mg/d. Patient selfincreased
the dose from 10 mg per
day due to insomnia and accidently
discovered a stimulant effect at higher
doses with improvement in depression.
Time to stimulant response: 20 min
Duration of response: 5-6 h
Generalized tonic-clonic seizure
occurred at 80 mg per day.
Restless legs
syndrome
Bezerra et al. 200242
8 patients, 3 male,
Mean age: 50.8
Zolpidem 10 mg once/d
Time to response: mean of 4 days
Continued response on medication with
no relapses
Positive response was noted in
all patients, regardless of age and
severity of restless legs syndrome
Sleepwalking with
or without sleep
related eating
disorder (SRED)
Sauvanet et al. 19884
Post-marketing study showed
1 of 96 patients developed
sleepwalking on zolpidem
Dose unknown Data not available
Mendelson 199443
20, male, participant in
clinical study of nocturnal
effects of sedative
medications
History of sleepwalking as a
child.
Zolpidem 10 mg at bedtime before
polysomnogram.
Investigator initiated auditory
tone aroused patient out of stage 4
sleep. Patient stood up and walked
on the bed. He was confused when
sleep technician entered room and
asked him to sit.
He had vague memory of the
events
Ganzoni et al. 19993
Post-marketing study showed
7 of 1972 patients developed
sleepwalking on zolpidem
Dose unknown Data not available
Downloadedfromjcsm.aasm.orgby109.81.215.43onJuly23,2020.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2020AmericanAcademyofSleepMedicine.Allrightsreserved.
Journal of Clinical Sleep Medicine, Vol.5, No. 5, 2009 S4
R Hoque, AL Chesson
Reference Age, sex, and clinical state Dose Description of response
Harazin et al. 199944
46, male, shift work related
insomnia
Negative history for
sleepwalking
Zolpidem 10 mg at bedtime Within 4 days of initiation of
zolpidem he would arise at night
to prepare a meal and eat it. He
had no memory of the events.
Medication was discontinued.
Unknown whether events
recurred off zolpidem.
Morgenthaler, et. al.
200245
Five patients
Mean age: 61.4, 3 male
All 5 patients had restless legs
syndrome, 3 had obstructive
sleep apnea
All 5 patients had a negative
history for sleepwalking
Zolpidem 10-20 mg once/d All exhibited sleepwalking and
SRED
4/5 had vague memory of the
events.
Events ceased in all patients with
zolpidem discontinuation
Sattar et al. 200346
47, male, bipolar disorder.
Negative history for
sleepwalking.
Citalopram 40 mg once/d, valproic
acid 250 mg twice/d, Zolpidem 5 mg at
bedtime
Sleepwalking started when
valproic acid was added to
citalopram and zolpidem. He had
no memory of the events.
Events ceased when valproic acid
was withdrawn and returned when
valproic acid was restarted.
Sharma et al. 200547
19, male, schizoaffective
disorder
Negative history for
sleepwalking
Zolpidem 10 mg at bedtime Within a few days of initiation of
zolpidem he started to sleepwalk
and talk incoherently. He had no
memory of the events.
Events ceased with zolpidem
discontinuation.
Najjar 200748
46, female, depression,
obstructive sleep apnea on
continuous positive airway
pressure therapy
Negative history for
sleepwalking or parasomnias
Zolpidem extended-release form 6.25
mg once/d
Sleepwalking and SRED began
immediately after initiation
of zolpidem use and ceased
immediately after it was
discontinued.
Chiang et al. 200826
Two patients:
Patient 1: 75, female
Patient 2: 70, female
Both patients had restless legs
syndrome, obstructive sleep
apnea , and used continuous
positive airway pressure
therapy
Both patients had a negative
history for sleepwalking or
SRED
Both patients had sleepwalking and
SRED on zolpidem extended release
12.5 mg once/d. Neither patient
experienced sleepwalking or SRED
on the immediate release form of
zolpidem.
Both patients experienced
sleepwalking and SRED
immediately after initiation of
extended release zolpidem use
and ceased immediately after it
was discontinued
Sansone et al. 200849
51, female, Zolpidem 10 mg at bedtime Arose at night to prepare a meal
and eat it. She had no memory of
the events.
Events stopped with cessation of
zolpidem.
Post-anoxic
spasticity
Shadan et al. 200450
28, male, cardiac arrest
resulting in post-anoxic
spasticity
Zolpidem 10 mg once a day
Time to response: 20 min
Duration of response: 2-4 h
Decrease in muscle rigidity,
spasticity and dystonic posturing
Spinocerebellar
ataxia
Claus et al. 200421
5 members of a family, 3
male, Mean age: 33
Zolpidem 10 mg once/d
Time to response: 1 h
Duration of response: unknown
4/5 patients showed improvement
of ataxia with zolpidem
Downloadedfromjcsm.aasm.orgby109.81.215.43onJuly23,2020.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2020AmericanAcademyofSleepMedicine.Allrightsreserved.