Effects of Topiramate on Urge to Drink and the
Subjective Effects of Alcohol: A Preliminary Laboratory
Robert Miranda Jr, James MacKillop, Peter M. Monti, Damaris J. Rohsenow, Jennifer
Tidey, Chad Gwaltney, Robert Swift, Lara Ray, and John McGeary
Background: Topiramate was recently reported to be efficacious in reducing drinking rates and
craving among individuals with alcohol dependence in a randomized controlled trial, but dose
effects could not be determined. This laboratory study systematically examined the dose-depen-
dent effects of topiramate on cue-elicited craving and other putative mechanisms of its pharmaco-
therapeutic effects on drinking.
Methods: Male and female heavy drinkers (n = 61) were randomized to 1 of 3 medication
conditions (200 mg⁄d; 300 mg⁄d; placebo) in a double-blind study. Participants reached the target
dose after a 32-day titration period, then were stabilized for approximately 1 week. All then par-
ticipated in a laboratory assessment of alcohol cue reactivity and of the subjective effects of a
moderate dose of alcohol.
Results: Both doses of topiramate reduced the frequency of heavy drinking during the titration
period as compared to placebo. However, topiramate did not affect self-reported craving for alco-
hol during the titration period, during the cue reactivity protocol, or in response to the alcohol
challenge procedure. Topiramate reduced the stimulating effects of alcohol ingestion compared to
placebo, but only in the 200 mg group.
Conclusions: The results of this study support previous findings that topiramate reduces drink-
ing, but the behavioral mechanism underlying this effect does not appear to be attenuation of
craving for alcohol as measured using the approaches employed in this study. Rather, the results
tentatively suggest that topiramate may exert its beneficial effects by altering the subjective experi-
ences of alcohol consumption. Limitations of the current study are discussed and complementary
methods are recommended for future studies, such as the use of behavioral economic paradigms
and ecological momentary assessment.
Key Words: Topiramate, Medication Action, Cue Reactivity, Alcohol Challenge, Craving.
ment for alcohol dependence based primarily on results of 1
double-blind randomized clinical trial (Johnson et al., 2003).
In this 12-week study, individuals with alcohol dependence
who received topiramate in doses titrated up to 300 mg⁄day
reported reductions in their quantity and frequency of alcohol
consumption, with large magnitude effect size differences
compared to placebo. In addition, results of 2 open-label tri-
als found similarly positive results on alcohol use after treat-
ment with topiramate (250 to 300 mg⁄d; Rubio et al., 2004;
Raguraman et al., 2005). Although the efficacy cannot be
OPIRAMATE IS AN anticonvulsant medication that
was recently identified as a potentially efficacious treat-
determined from open-label studies, these latter reports pro-
vide further support for the tolerability and safety of topira-
mate for treating alcoholism.
Despite considerable interest in topiramate as a promising
pharmacotherapy for alcohol dependence, the mechanisms by
which this medication reduces drinking are not well character-
ized. Pharmacotherapeutic strategies for treating alcoholism
generally aim at reducing withdrawal symptoms, reducing
craving or urge to drink, attenuating the pleasurable effects of
alcohol ingestion, and intensifying the unpleasant effects of
alcohol ingestion (Davidson et al., 1999). Identifying the
mechanisms by which different pharmacological agents
impact drinking would allow for more targeted use of medica-
tions and expand our understanding of the underlying neuro-
biology of alcoholism. Moreover, different medications may
reduce drinking through different mechanisms. Understand-
ing how particular medications affect drinking would suggest
which types and combinations of medications may be most
beneficial for treating alcohol use disorders.
It has been hypothesized that topiramate reduces alcohol
intake by attenuating alcohol craving and motivation to drink
(Johnson et al., 2003, 2004). This hypothesis is based largely
From the Center for Alcohol and Addiction Studies, Brown Univer-
sity (RM, JMK, JT, CG, LR), Providence Veterans Affairs Medical
Center and Center for Alcohol and Addiction Studies, Brown Univer-
sity (PMM, DJR, RS, JMG) Providence, Rhode Island.
Received for publication July 17, 2007; accepted November 20, 2007.
Reprint requests: Robert Miranda, Jr., Ph.D., Brown University,
Box S121-5, Providence, RI 02912; Fax: 401-863-6697; E-mail: robert_
Copyright ? 2008 by the Research Society on Alcoholism.
Alcoholism: Clinical and Experimental Research
Vol. 32, No. 3
Alcohol Clin Exp Res, Vol 32, No 3, 2008: pp 1–91
on the medication’s neuropharmacological actions, which
include facilitation of gamma aminobutyric acid (GABA)
neurotransmission and blockade of AMPA⁄kainite glutamate
receptors. Acute dopamine release from mesocorticolimbic
neurons plays a critical role in determining the motivational
significance of stimuli in general (see Hyman and Malenka,
2001) and in alcohol reward (e.g., Dodd et al., 2000; Weiss
et al., 1993; Weiss & Porrino, 2002). According to Johnson
et al. (2003), because mesocorticolimbic DA release is under
tonic inhibitory control via GABAergic neurons and excit-
atory control via glutamatergic neurons, topiramate’s concur-
rent GABAergic agonism and glutamatergic antagonism may
inhibit acute motivation for alcohol (i.e., craving). The one
published placebo-controlled clinical trial of topiramate’s
effects on drinking reported that topiramate reduced urge to
drink and this attenuation of craving was significantly corre-
lated with decreased alcohol use (Johnson et al., 2003).
Although the report from Johnson et al. (2003) provides
provisional support for the hypothesis that topiramate
reduces alcohol use by attenuating craving, no controlled lab-
oratory study has systematically tested this hypothesis. Fur-
thermore, other putative mechanisms of topiramate’s actions,
such as alterations in the subjective reinforcing effects of alco-
hol, have not been examined. In the Johnson et al. (2003)
study of topiramate’s effects, it is unclear whether topiramate
first reduced craving which in turn reduced drinking, or
whether topiramate affected another aspect of alcohol use
which reduced drinking and these changes subsequently atten-
uated desire for alcohol.
The primary objective of the present study was to systemat-
ically examine the effects of topiramate on several putative
mechanisms of pharmacotherapy action using 2 well-estab-
lished human laboratory paradigms, one that focused on
alcohol craving and one that examined subjective responses
to alcohol. The cue reactivity paradigm was developed to
examine urge to drink in response to alcohol-associated stim-
uli in a controlled setting, as exposure to alcohol cues can sim-
ulate a high-risk situation for alcohol use. Indeed, the cue
reactivity paradigm has demonstrated substantial utility in
eliciting urge to drink (e.g., Monti et al., 1987; for a review,
see Carter and Tiffany, 1999) and reactivity to alcohol cues
has been shown to predict treatment outcome (Monti et al.,
2000). Similarly, an important advantage of the alcohol
administration paradigm is that it allows for the assessment
of a medication’s action on the subjective effects of alcohol
ingestion under carefully controlled alcohol administration
conditions. Furthermore, by assessing effects in close proxim-
ity to drinking, bias because of memory decay involved in ret-
rospective reporting is prevented.
In the present study, participants were randomly assigned
to receive 1 of 3 targeted medication conditions (200 mg⁄d,
300 mg⁄d, or placebo). We tested the dose-dependent effects
of topiramate because Johnson et al. (2003) found that signifi-
cant reductions in drinking emerged while participants were
at the 200 mg⁄day dose, however, doses continued to escalate.
As such, the effects of time and dose on drinking were con-
founded in that study and the potential efficacy of doses stabi-
lized at lower than 300 mg⁄day remains unclear. In the
present study, medication was titrated over a 32-day period to
1 of 2 stable target doses or placebo. After dose stabilization,
participants completed a laboratory session that involved in
vivo exposure to alcohol beverage cues followed by alcohol
administration procedures. We examined the dose-dependent
effects of topiramate on: (i) alcohol use and craving during
the titration period; (ii) craving in response to alcohol cue
exposure in the laboratory; and (iii) craving and other subjec-
tive responses to alcohol administration in the laboratory.
Based on previous findings, it was hypothesized that topira-
mate would reduce alcohol use and craving during the titra-
tion period, and attenuate urge to drink in response to
alcohol cues in the laboratory. Although no previous labora-
tory studies have been conducted on topiramate and alcohol
consumption, we explored its effects on the stimulant and sed-
ative effects of alcohol, since these should be related to the
reinforcement value of drinking (Swift et al., 1994).
MATERIALS AND METHODS
To be eligible for the study, participants needed to be between 21
and 65 years of age, consume 18 to 60 standard alcoholic drinks⁄
week during the 90 days prior to enrollment, if male (14 to 53
drinks⁄week, if female), to score <14 on the Beck Depression Inven-
tory-II (Beck et al., 1996), and, for purposes of the alcohol challenge,
to like beer. Potential participants were excluded if they had recent
treatment or expressed interest in current treatment for alcoholism, if
they had medical contraindications to study participation, if they
used medications that could affect mood or drinking, or if they were
living with someone who was enrolled in the study.
Of 78 individuals enrolled in the study and sequentially random-
ized into medication condition, 14 voluntarily withdrew from the
study for various reasons, including side effects (n = 6), misgivings
about participating in a medication trial (n = 2), or miscellaneous
personal reasons (e.g., relocation; n = 6). Three additional partici-
pants in the 300 mg group who completed the study were determined
to have zero topiramate levels by serum assay. These participants
were considered to be noncompliant and their data were excluded
from analyses, leaving the final sample at 61 participants (PLA
n = 20, 200 mg n = 20, 300 mg n = 21).
Participants were typically white (89%), with smaller proportions
of minority participants (5% African-American, 2% Native Ameri-
can, 5% biracial); ethnically, participants were 90% non-Hispanic.
The percentage with current alcohol use disorders was not signifi-
cantly different across conditions: overall, 14% met criteria for alco-
hol abuse and 46% met criteria for dependence according to the
Diagnostic and Statistical Manual of Mental Disorders, Fourth Edi-
tion, Text Revision (DSM-IV-TR; American Psychiatric Association,
2000). 1 participant met criteria for current marijuana dependence,
but no other current substance use disorders were present. Diagnostic
interviews were conducted by a trained doctoral level psychologist
using the Structured Clinical Interview for DSM-IV (First et al.,
1995). Other axis I psychopathology was not assessed.
Medication and Compliance
Participants underwent a 32-day titration period, followed by up
to 7 days at target dose (see Table 1) during which time the labora-
tory session took place. The 7-day window permitted some flexibility
in scheduling the laboratory session. Modal duration at target dose
MIRANDA ET AL.
at the time of the laboratory session was 4 days. Immediately follow-
ing the laboratory session, participants underwent a 7-day dose
reduction period (see Table 2) for safety reasons. The placebo cap-
sules were compounded by the same pharmacy and were identical in
all dimensions except the topiramate content. All medication was
compounded from bulk topiramate to our specifications. Specifically,
the amounts of topiramate as described were compounded with Avi-
cel filler, which is a nonlactose, pharmacologically inert, microcrystal-
line cellulose filler. Active and placebo capsules were identical in
appearance across all doses and conditions, and all participants were
instructed to take 2 capsules daily, 1 in the morning and 1 in the
evening. The capsules were individually labeled by dose (e.g., Day 4,
Morning) and individually sealed in plastic wrappers.
Medication compliance was assessed using 2 methods. Electronic
medication bottle caps [Medication Event Monitoring System
(MEMS); Aardex Inc., Geneva, Switzerland] prompted participants
to take capsules and recorded the date⁄time that the medication bot-
tles were opened. In addition, blood samples were taken during week
3 and at the laboratory session and analyzed for plasma levels of
Alcohol use prior to and during the study was assessed using the
Timeline Follow Back (Sobell et al., 1979), with heavy drinking days
defined as 6⁄4 standard drinks in a day for males⁄females (Flannery
et al., 2002). Craving on average over days during the titration period
was assessed using 2 measures, the Drinking Obsessions subscale of
the Obsessive Compulsive Drinking Scale (OCDS; Anton et al.,
1995; Bohn et al., 1996) and the Penn Alcohol Craving Scale (PACS;
Flannery et al., 1999), which explicitly assesses experiential craving.
The compulsive subscale of the OCDS was not used because its con-
tent is largely not craving-related. In addition, to allow direct com-
parisons with Johnson et al. (2003), the Automaticity of Drinking
and Interference due to Drinking subscales of the OCDS, as derived
by Bohn et al. (1996), were also used. During the titration period,
side effects were rated for severity (i.e., none, mild, moderate, and
severe) using an interview that assessed 38 specific side effects (John-
son et al., 2005; Levine and Schooler, 1986) and used open-ended
questions to assess any additional unexpected side effects.
The immediate state of craving was assessed using the Alcohol
Urge Questionnaire (AUQ; Bohn et al., 1995), an 8-item measure val-
idated for use in laboratory research (MacKillop, 2006). During the
cue reactivity procedure, the AUQ was complemented by the modi-
fied Alcohol Attention Scale (AAS), two 10-point Likert-type scales
assessing attention to the sight and smell of water and alcohol cues
(Rohsenow et al., 1992). These scales were modified by omitting an
item pertaining to thoughts about drinking. In addition, psychophys-
iological arousal was assessed via measures of mean arterial pressure
(MAP) and heart rate in beats per minute (BPM) using a Criticare?
Scholar II 507EP blood pressure monitor.
During the alcohol administration, breath alcohol (BrAC) was
assessed using an Intoximeters? Alco-Sensor IV breathalyzer. Stimu-
lation and sedation were assessed using the Biphasic Alcohol Effects
Scale (BAES; Martin et al., 1993). Craving was assessed using the
AUQ and affect was assessed using the Positive and Negative Affect
Schedule (PANAS; Watson et al., 1988).
The measures administered during the laboratory session are pre-
sented in Table 3.
All procedures were approved by the Brown University Institu-
tional Review Board. Participants were recruited via advertisements
placed in the community. Potential participants underwent a tele-
phone screening and those who met initial criteria were invited to
attend an in-person screening interview that included a physical
examination and urine toxicology screen. Individuals who met crite-
ria provided written informed consent and received their first week of
medication. The instructional set was that the study was examining
the effects of 2 doses of topiramate on craving in response to alcohol
cues and while drinking modest doses of alcohol; no instructions
about potential effects on participants’ drinking in their daily life
were given. Subsequently, participants attended weekly sessions dur-
ing which they were assessed for alcohol use, craving, and side effects,
and were given their next 7-day supply of medication.
The individual laboratory session took place in the afternoon, typi-
cally starting at noon, and lasted approximately 5 hours. Participants
Table 1. 200 and 300 mg Topiramate Titration Schedules
200 mg⁄d300 mg⁄d
Table 2. Baseline Descriptive Statistics for the 3 Experimental Conditions
(total n = 61)
(n = 20)
(n = 20)
(n = 21)
Alcohol use disorder
Years of education
% Drinking days
64.50% (19.39) 77.28% (25.32) 67.99% (21.83)
40.11 (27.39)48.83 (29.34)
37. 76 (13.18)
Note: Percentage or mean (±SD) is provided, as appropriate.
Table 3. Laboratory Session Experiential and Psychophysiological
Cue reactivity paradigm Alcohol Urge Questionnaire (AUQ)
Alcohol Attention Scale (AAS)
Alcohol administration AUQ
Biphasic Alcohol Effects Scale (BAES)
Positive and Negative Affect Scales (PANAS)
Note: MAP⁄BPM = mean arterial pressure⁄beats per minute.
MECHANISMS OF TOPIRAMATE ACTION
were transported by taxi to the laboratory, where they completed
their weekly assessments, had blood drawn to determine topiramate
levels and alcohol breath testing to ensure recent abstinence. They
then underwent the cue reactivity and alcohol challenge protocols.
No participant had a positive breath alcohol level upon arrival. All
experimental manipulations took place in a sound-attenuated testing
room equipped with a one-way mirror to monitor participants’
behavior. The cue reactivity procedure was similar to that of previ-
ously published protocols (e.g., McGeary et al., 2006; Monti et al.,
1993). Participants were fitted with the blood pressure cuff on their
nondominant arm and habituated to the inflation cycle (approxi-
mately every 40 seconds) during an audio taped instruction period,
which included a demonstration of the procedures without a bever-
age. Trials were presented in the same order for all participants
because of known carryover effects (Monti et al., 1987).
Participants first underwent a 3-minute relaxation period (‘‘please
sit quietly and do nothing’’) to collect baseline levels of urge and
physiological arousal. Following relaxation, a research assistant
(RA) entered the room with a tray containing a glass half full of
water and a commercially labeled bottle of water, both covered by an
inverted pitcher. The pitcher was removed, the RA left the room,
and the audio tape instructed the participant to sniff the glass of
water when s⁄he heard high tones and stop sniffing when s⁄he heard
low tones. This protocol included thirteen 5-second olfactory expo-
sures during each 3-minute trial, with variable intervals between each
exposure. In this way, participants were exposed to visual, tactile,
olfactory, and proprioceptive stimuli associated with the beverage.
At the end of the trial, the participant completed an assessment
packet (described above in Laboratory Assessments), the RA
removed the tray, bottle, and glass, and another 3-minute relaxation
period began. Participants next underwent two 3-minute alcohol cue
exposure trials that were identical to the water trial except the glass
of water was replaced with their preferred alcohol beverage (Staiger
and White, 1991), and the bottle of water was replaced with the
appropriate commercially-labeled alcohol bottle. Two alcohol trials
were conducted to gain a stable estimate of participants’ reactions to
alcohol cues and because two exposures have proven sensitive to dif-
ferential effects in previous studies (e.g., Rohsenow et al., 2000). Con-
sistent with recommendations by Wertz and Sayette (2001),
participants were clearly informed they would receive alcohol to con-
sume following the cue exposure procedure.
After the conclusion of the cue reactivity procedure, participants
underwent an alcohol challenge (e.g., Drobes et al., 2004; King et al.,
1997; Swift et al., 1994). Individuals were given high-alcohol content
(7.2%) beer in a volume that was designed for them to reach a
blood-alcohol level of 0.06% based on a nomogram incorporating
their gender and weight (Rohsenow et al., 2006). Total alcohol
administered was .647 g⁄kg for males and .548 g⁄kg for females; this
was typically 2 to 3 16 oz. bottles of 7.2% alcohol beer. Participants
were asked to consume the alcohol over a 20-minute period, which
was followed by a 20-minute absorption period. Participants were
then assessed for BrAC after rinsing their mouths with water to elimi-
nate residual oral alcohol. At this time, the assessment packet was
administered to participants who were at a BrAC of 0.06 g % or
higher. For those who were still below 0.06 g %, BrAC was assessed
10 minutes later to allow further absorption. At that point, the
assessment packet was administered. Following the assessment, par-
ticipants were provided with a meal and were permitted to relax in
the laboratory until their BrAC had dropped to below 0.04 g %, at
which time they were provided with the dose reduction medication
and transported home in a taxi. All individuals were compensated
for their participation.
All data were initially examined for distribution normality and out-
liers, and the psychophysiological data were examined for movement
artifacts. Potential differences on dependent variables between condi-
tions were examined using univariate analyses of variance (ANO-
VAs). Differences in side effects among the 3 groups were examined
using Kruskal–Wallis H-tests and significant effects were clarified
using follow-up chi-square tests. Consistent with Johnson et al.
(2003), effects of topiramate during the titration period were exam-
ined on a weekly basis and titration effects were examined using 3
(PLA, 200 mg, 300 mg) · 4 (weeks 1 to 4) analyses of covariance
(ANCOVAs), while covarying initial baseline values. Effects of topi-
ramate were examined for weeks 1 to 4 only, because after week 4
participants’ duration of enrollment varied, generating potentially
nonrepresentative data. Topiramate’s effects on alcohol cue reactivity
were examined using 3 (PLA, 200 mg, 300 mg) · 3 (Water cues,
Alcohol Cues 1, Alcohol Cues 2) mixed ANOVAs. Medication
effects on reactions to alcohol were examined using one-way
between-subjects analyses of covariance (ANCOVAs), while covary-
ing pre-administration values and BrAC. For all analyses, planned
follow-up pairwise comparisons and simple effects tests were con-
ducted to characterize significant main effects and interactions. All
analyses were conducted using SPSS 14.0 (SPSS, Inc., Chicago, IL,
USA), using partial eta-squared (gp2) as an index of effect size, and a
significance criterion of p < 0.05. Multiple-test error-correction was
not employed because of the exploratory nature of this study.
Less than 1% of data were missing and no data were
imputed. All dependent variables were normally distributed,
without outliers. Movement artifacts were removed from the
MAP and BPM data. Univariate ANOVAs revealed no dif-
ferences between the conditions in terms of baseline values of
drinking quantity and frequency measures or of other depen-
dent variables. Of those participants who withdrew because
of side effects, 4 were in the 300 mg condition, 1 was in the
200 mg condition, and 1 was in the placebo condition. No
medication-related serious adverse events occurred.
According to MEMS data, all participants met the medica-
tion compliance criterion of taking 80% of their assigned
medication, with a mean compliance rate of 96.5%
(range = 82 to 100%) for the full duration of the study,
Table 4. Effects of Topiramate on Drinking and Craving During the Titration
mate · Time
Alcohol use variables
% Drinking days
% Heavy drinking days
Craving and related variables
Penn Alcohol Craving
Automaticity of drinking
Interference of drinking
aThe degrees of freedom for the Penn Alcohol Craving Scale are:
(2, 56), (3, 168), and (6, 168), respectively.
?p < 0.10, * p < 0.05, ** p < 0.01.
MIRANDA ET AL.
including the decreasing titration period. Of 41 participants
randomized to active drug, 32 were verified as compliant by
serum assay; samples from the remaining 9 participants
hemolyzed in transit and were nonviable for analysis. The
300 mg group had significantly higher topiramate serum lev-
elscomparedto 200 mg
(SD = 1.41) vs. 300 mg M = 6.98 mcg⁄ml (SD = 2.16); F
(1, 23) = 12.55, p < 0.005, gp2= 0.35).
With respect to side effects once participants reached their
target dose, significant main effects of medication condition
were evident for dizziness (v2[2, N = 60] = 7.00, p < 0.05),
difficulty concentrating (v2[2, N = 60] = 10.71, p < 0.01),
and tingling in fingers and toes (v2[2, N = 60] = 14.48,
p < 0.001), which is similar to findings of Johnson et al.
(2003). Dizziness was more frequent in the 300 mg group than
the 200 mg group (p < 0.05). Likewise, difficulty concentrat-
ing was more frequent in the 300 mg group compared to
PLA (p < 0.05). Tingling in fingers and toes was more fre-
quent in the both topiramate groups compared to PLA
(p < 0.05), but there was no difference between topiramate
doses on this effect. No other differences between groups in
side effects were present. At the conclusion of the study, no
differences were evident across conditions in terms of dizzi-
ness and difficulty concentrating (p > 0.20), although the 2
topiramate groups reported residual mild tingling in fingers
and toes significantly more than PLA (p < 0.05).
[200 mgM = 4.45 mcg⁄ml
Effects of Topiramate During the Titration Period
The results of the 3 · 4 mixed ANCOVAs on the alcohol
use (TLFB) and craving (OCDS subscales) variables are pre-
sented in Table 4. There was a significant topiramate · time
interaction on drinks⁄week such that topiramate reduced
drinking as dose increased (see Fig. 1). Follow-up simple
effects tests indicated no significant differences between medi-
cation conditions at weeks 1 and 2, but differences were sig-
nificant at weeks 3 and 4 (p < 0.05). At week 3, the 300 mg
group reported significantly fewer drinks⁄week compared to
the other groups (p < 0.05), which did not significantly differ
from eachother. Atweek4,the 300 mggroupreportedsignifi-
cantly fewer drinks⁄week compared to PLA (p < 0.01), but
not to the 200 mg group, and PLA and 200 mg were not sig-
nificantly different from each other.
There was a significant main effect of topiramate and a sig-
nificant topiramate · time interaction on percent heavy
drinking days. Pairwise comparisons revealed that there was
significantly less heavy drinking days in the 300 mg
(M = 29.83%, SE = 3.76%) and 200 mg (M = 28.31%,
SE = 3.87%) topiramategroups
(M = 41.80%, SE = 3.83%; p < 0.05), but there was no
significant difference between active doses on this measure.
The interaction effect and simple effects tests revealed no dif-
ferences between groups at week 1, but significant differences
emerged in weeks 2 to 4 (Fig. 1). At week 2, percent heavy
drinking days for the 200 mg group were significantly lower
than for both the 300 mg and PLA groups (p < 0.05), which
were not significantly different from each other. At week 3,
there were fewer percent heavy drinking days in the 300 mg
group than the PLA group (p < 0.01) and a trend toward
fewer heavy drinking days in the 200 mg group compared to
the PLA group (p = 0.06), but the 2 medication groups were
not significantly different from each other. At week 4, both
medication groups reported fewer heavy drinking days than
the PLA group (ps < .05), but were nonsignificantly different
from each other.
With regard to craving during the titration period, no sig-
nificant main effects of medication or time were evident for
any of the craving variables (i.e., OCDS- Obsessions, OCDS-
Automaticity of Drinking, OCDS-Interference from Drink-
ing, and PACS), nor were there any interaction effects.
Effects of Topiramate on Reactions to Alcohol Cues
The 3 · 3 (medication condition · cue type) mixed ANO-
VA revealed a significant main effect of cue type on craving
(F [2, 116] = 25.26, p < 0.001, gp2= 0.30), but no effect of
medication condition (F [2, 58] = 0.04), and no medication
condition · cue type interaction (F [4, 116] = 0.89). The sig-
nificant main effect of cue type reflected the characteristic
Fig. 1. Effects of topiramate on alcohol use over the course of the titration
period for measures with significant interaction effects. Asterisks reflect fol-
low-up simple effects tests; *p < 0.05, **p < 0.01. Specific comparisons
between conditions at each time point are provided in the text.
MECHANISMS OF TOPIRAMATE ACTION
increase in craving during the alcohol cue exposure trials com-
pared to the water cue exposure trial (ps < 0.001), but no dif-
ferences were evident between the 2 alcohol exposure trials.
Similarly, a main effect for alcohol cues was found on the
AAS (F [2, 116] = 11.03, p < 0.001, gp2= 0.16), with sig-
nificantly greater attention paid to the alcohol cue exposure
trials compared to the water cues (p < 0.001), but no differ-
ence between the 2 alcohol cue exposure trials on the atten-
tion scale.No effectof
condition · cue type interaction effect was evident for the
AAS (ps >0 .67). In terms of effects on psychophysiological
arousal, a significant main effect of alcohol cues on MAP was
present (F [2, 114] = 7.52, p < 0.001, gp2= 0.12), but no
main effect of medication (F [2, 57] = 2.22, p = 0.12) or
interaction effect (F [4, 114] = 1.83, p = 0.13) was observed.
The significant effect reflected greater MAP during the alco-
hol cue exposure trials compared to the water cue exposure
trial (ps < 0.005), but no difference between the 2 alcohol tri-
als. For BPM, neither main effects nor a medication interac-
tion were evident (ps > 0.12).
Effects of Topiramate on Reactions to Alcohol Ingestion
Participants’ BrAC following the absorption period at the
estimated peak was M = 0.08 g % (SD = 0.01), with no sig-
nificant differences by gender or medication. Fifteen partici-
pants required 10 to 20 minutes of additional absorption
time. The ANCOVA revealed a significant medication effect
on postconsumption stimulation, F (2, 55) = 3.60, p < 0.05,
gp2= 0.12. Pairwise comparisons indicated that the 200 mg
group, but not the 300 mg group, reported significantly lower
stimulation compared to PLA (p = 0.01). No significant
effects of medication condition were evident on postconsump-
tion sedation (F (2, 55) = 2.21), urge to drink (F [2,
54] = 0.10), positive affect (F [2, 55] = 0.35), or negative
affect (F [2, 54] = 0.86).
The goals of this preliminary study were to clarify the
mechanisms of topiramate’s effects in reference to drinking
using several approaches. The results reveal a number of
findings that converge with previous clinical and preclinical
reports and several novel findings. Topiramate was gener-
ally well tolerated and significantly reduced drinking and
heavy drinking over the course of the medication titration
period, with results generally showing the strongest effect
as the dose increased. These results were consistent with
previous findings that topiramate treatment decreased
(Johnson et al., 2003; Raguraman et al., 2005; Rubio
et al., 2004). Similarly, in preclinical research, chronic
administration of topiramate to mice and rats has been
demonstrated to result in decreased ethanol consumption
and ethanol preference (Gabriel and Cunningham, 2005;
Knapp et al., 2004; Nguyen et al., 2004).
with alcohol dependence
There are a number of specific aspects of topiramate’s
effects on drinking that were interesting in this study. Topira-
mate did not affect participants’ frequency of drinking, but
rather how much they drank during an episode, as reflected
by the largest effect being on percentage of heavy drinking
days, similar to the findings by Johnson et al. (2003). In addi-
tion, significant decreases were clearly evident in both medica-
tion groups by the third week of the study, suggesting that
doses of topiramate as low as 100 to 175 mg may be effective
in reducing heavy drinking. Interestingly, effects were evident
in the 200 mg group as early as the second week (75 mg),
however, this was not the case for the 300 mg group. It is
not clear why this discrepancy was observed, although it is
possible that either an unmeasured pharmacogenetic variable
associated with greater sensitivity to topiramate was dispro-
portionately present in the 200 mg group, or that the result is
a function of random variation. Both of these interpretations
must be speculative at this point.
Changes in drinking were not accompanied by changes
in weekly reports of craving for alcohol using validated
self-report measures. This finding is consistent with obser-
vations from the laboratory portion of the study. Expo-
suretoalcohol cues increased
psychophysiological reactions, indicating the validity of
the cue reactivity procedure. However, topiramate did not
affect the subjective or physiological responses to alcohol
cues. Similarly, during the alcohol administration proce-
dure, topiramate did not affect urge to drink alcohol
during a moderately intoxicated state (mean of 0.08 g%
BrAC). Thus, in contrast with Johnson et al. (2003), the
present findings do not suggest that topiramate exerts its
beneficial effects on drinking outcomes by attenuating
craving, at least not according to the measures and experi-
mental paradigms employed.
There are a number of possible reasons why these data
diverge from the previous study showing that topiramate
reduced craving (Johnson et al., 2003). One possibility, as
noted above, is that Johnson et al. (2003) combined data from
both abstinent and non-abstinent people at each time point,
confounding the craving measure. Also, it is unclear whether
topiramate may have initially reduced alcohol use via a vari-
able other than craving, and decreased drinking could have
subsequently resulted in reduced desire for alcohol. Further-
more, the previous clinical trials studied treatment-seeking
alcohol dependent individuals, who are clearly different from
our nontreatment-seeking heavy drinkers (of whom 60% met
criteria for an alcohol use disorder). Treatment-seeking indi-
viduals may be particularly motivated to experiencing reduc-
tions in their desire for alcohol and thus may be more
sensitive to the effects of medication on craving than individu-
topiramate reduced drinking via an alternative mechanism,
participants may have attributed those reductions to
decreased desire. Although it is unclear which of these
possible factors played a role, if any, what is clear is that in
this study topiramate reduced drinking and did so in the
MIRANDA ET AL.
absence of effects on craving in each of the approaches used,
suggesting alternative mechanisms of action.
Interestingly, suggestive findings were evident with regard
to the effect of topiramate on alcohol’s subjective effects.
Topiramate significantly reduced the stimulating effects of
alcohol ingestion, although this was restricted to the 200 mg
group compared to placebo. These findings, albeit prelimin-
ary, suggest that in the current study topiramate may have
affected subjective effects of alcohol consumption by attenu-
ating its positively reinforcing properties, but only at the
lower 200 mg dose. Such relationships would be consistent
with our finding that topiramate reduced drinking quantity
but not frequency, and with our interpretation that this effect
reflects a reduction in the reinforcing effects of alcohol once
drinking has begun.
There are, however, several reasons to be cautious about
these results. As noted, the effect on stimulation only
reflected a significant effect for the 200 mg group. In
addition, the observed effects were of relatively small mag-
nitude, suggesting that these findings per se are insufficient
to fully explain the effects on drinking. Furthermore, topi-
ramate’s effect was only studied at estimated peak BrAC
after 1 dose of alcohol. It is possible that topiramate has
more pronounced effects at lower or higher blood-alcohol
levels and it is also possible that focusing on topiramate’s
effects during the ascending or descending limb of the
medication’s effects on experiential intoxication (Ray &
Hutchison, in press; Ray et al., 2007). As such, the current
findings are suggestive that topiramate may exert its
impact via effects on the subjective effects of alcohol, but
should be considered preliminary and supportive of the
need for more fine-grained and extensive examinations of
its effects on subjective intoxication.
This study had a number of limitations that should be
considered. As a preliminary study, the sample size was rela-
tively small, which resulted in statistical power to only
detect moderate-to-large effect size effects. In addition,
although the study sought to study topiramate’s mecha-
nisms more clearly than in previous clinical trials, its scope
was nonetheless limited to 2 laboratory paradigms. As such,
future studies would be well served to examine topi-
ramate using alternative methodological approaches. Recent
alcoholism pharmacotherapy studies have found that behav-
ioral economic paradigms for assessing alcohol reinforce-
ment (e.g., alcohol self-administration under conditions of
cost) have proven to be very useful for understanding medi-
cation effects (e.g., Drobes et al., 2003; O’Malley et al.,
2002). Importantly, behavioral economic measures of moti-
vation objectively assess drug motivation but have only
modest-to-moderate correlations with subjective craving
(e.g., MacKillop et al., 2007; O’Malley et al., 2002; Sayette
et al., 2001). Thus, such paradigms may be more sensitive
to medication effects that are independent of subjective
craving. In addition, because topiramate does not appear
to have an acute effect on alcohol use at low doses but
more gradually reduces drinking (Johnson et al., 2003;
Raguraman et al., 2005; Rubio et al., 2004), as was evident
in this study, it would be informative to have fine-grained
assessment of the processes that take place in initiating and
sustaining those changes in future studies. Ecological
momentary assessment, or close to real-time assessment in
individuals daily life using structured or electronic daily
diaries, has been profitably used in understanding the mech-
anisms of naltrexone (e.g., Armeli et al., 2006; Kranzler
et al., 2004; Tidey et al., in press) and could be a promising
approach for studying topiramate. Taken together, future
studies that use larger sample sizes, additional experimental
paradigms, and higher resolution assessment appear to be
A final consideration that is worthy of note pertains to the
actual levels of drinking observed in the 2 topiramate groups.
Although both groups exhibited significant decreases on mul-
tiple drinking variables, absolute drinking remained high
(e.g., >20 drinks⁄week). Indeed, clinically speaking, these
individuals were still drinking at problematic levels. This
underscores the essential role of motivation in changing exces-
sive alcohol use. In its absence, although topiramate reduces
alcohol use, its effects were not so robust as to reach healthy
drinking levels. Recognizing this, it is likely that topiramate
will best fit into a multifaceted clinical approach, where moti-
vated individuals receive both psychosocial and pharmacolog-
Despite the various considerations of the methods and
findings, the present findings are significant as the first
human laboratory study to examine the biobehavioral
mechanisms of topiramate’s effects on drinking. The pres-
ent findings suggest that while topiramate may not reduce
the likelihood that an individual will drink, it may hold
promise for reducing the likelihood that he or she will
drink heavily. In this respect, the clinical praxis of topira-
mate pharmacotherapy may be similar to naltrexone (e.g.,
O’Malley et al., 1996; Rohsenow, 2004). Furthermore, the
present findings do not support the notion that the effects
of topiramate on drinking were a function of its effects
on craving in the paradigms used. Rather, topiramate
may exert its beneficial effects by altering the subjective
experiences of alcohol ingestion, although this hypothesis
requires further examination. Considerable further research
is clearly necessary to understand the mechanisms under-
lying topiramate’s effects on drinking.
This research was supported in part by a Career Develop-
ment Award (1K23 AA014966) and a grant (5R01 AA7850-
17) from the National Institute of Alcohol Abuse and
Alcoholism, and a Research Career Scientist Award and a
Senior Research Career Scientist Award from the Department
of Veterans Affairs. The authors are very grateful for the
research assistance of Amy Christian, John-Paul Massaro,
Suzanne Sales, Timothy Souza, and Lynn Taylor, MD.
MECHANISMS OF TOPIRAMATE ACTION
American Psychiatric Association (2000) Diagnostic and Statistical Manual of
Mental Disorders, 4th ed. Text Revision. American Psychiatric Association,
Anton RF, Moak DH, Latham P (1995) The Obsessive Compulsive Drinking
Scale: a self-rated instrument for the quantification of thoughts about alco-
hol and drinking behavior. Alcohol Clin Exp Res 19:92–99.
Armeli S, Feinn R, Tennen H, Kranzler HR (2006) The effects of naltrexone
on alcohol consumption and affect reactivity to daily interpersonal events
among heavy drinkers. Exp Clin Psychopharmacol 14:199–208.
Beck AT, Steer RA, Brown GK (1996) Manual for the Beck Depression
Inventory-II. Psychological Corporation, San Antonio, TX.
Bohn MJ, Barton BA, Barron KE (1996) Psychometric properties and validity
of the obsessive-compulsive drinking scale. Alcohol Clin Exp Res 20:817–
Bohn MJ, Krahn DD, Staehler BA (1995) Development and initial validation
of a measure of drinking urges in abstinent alcoholics. Alcohol Clin Exp
Carter BL, Tiffany ST (1999) Meta-analysis of cue-reactivity in addiction
research. Addiction 94:327–340.
Davidson D, Palfai T, Bird C, Swift R (1999) Effects of naltrexone on alcohol
self-administration in heavy drinkers. Alcohol Clin Exp Res 23:195–203.
Dodd PR, Beckmann AM, Davidson MS, Wilce PA (2000) Glutamate-
mediated transmission, alcohol, and alcoholism. Neurochem Int 37:509–
Drobes DJ, Anton RF, Thomas SE, Voronin K (2003) A clinical laboratory
paradigm for evaluating medication effects on alcohol consumption: nal-
trexone and nalmefene. Neuropsychopharmacology 28:755–764.
Drobes D, Anton R, Thomas SE, Voronin K (2004) Effects of naltrexone and
nalmefene on subjective response to alcohol among non-treatment-seeking
alcoholics and social drinkers. Alcohol Clin Exp Res 28:1362–1370.
First MB, Gibbon M, Spitzer RL, Williams JBW (1995) User’s guide for the
Structured Clinical Interview for. DSM-IV Axis I Disorders (SCID-I, Ver-
sion 2.0.). New York State Psychiatric Institute, Biometrics Research
Department, New York.
Flannery BA, Allen JP, Pettinati HM, Rohsenow DJ, Cisler RA, Litten RZ
(2002) Using acquired knowledge and new technologies in alcoholism treat-
ment trials. Alcohol Clin Exp Res 26:423–429.
Flannery BA, Volpicelli JR, Pettinati HM (1999) Psychometric properties of
the Penn Alcohol Craving Scale. Alcohol Clin Exp Res 23:1289–1295.
Gabriel KI, Cunningham CL (2005) Effect of topiramate on ethanol and sac-
charin consumption and preferences in C57BL⁄6J mice. Alcohol Clin Exp
Hyman SE, Malenka RC (2001) Addiction and the brain: the neurobiology of
compulsion and its persistence. Nat Rev Neurosci 2:695–703.
Johnson BA, Ait-Daoud N, Bowden CL, DiClemente CC, Roche JD, Lawson
K, Javors MA, Ma JZ (2003) Oral topiramate for treatment of alcohol
dependence: a randomized controlled trial. Lancet 361:1677–1685.
Johnson BA, Ait-Daoud N, Roache JD (2005) The COMBINE SAFTEE: a
structured instrument for collecting adverse events adapted for clinical stud-
ies in the alcoholism field. J Stud Alcohol 15:157–167.
Johnson BA, Swift RM, Ait-Daoud N, DiClemente CC, Javors MA, Malcolm
RJ (2004) Development of novel pharmacotherapies for the treatment of
alcohol dependence: focus on antiepileptics. Alcohol Clin Exp Res 28:295–
King AC, Volpicelli JR, Frazer A, O’Brien CP (1997) Effect of naltrexone on
subjective alcohol response in subjects at high and low risk for future alco-
hol dependence. Psychopharmacology 129:15–22.
Knapp CM, Markley L, Ciraulo DA, Kometsky C (2004) Topiramate and
zonisamide reduce the consumption of an ethanol-sucrose solution in the
rat. Alcohol Clin Exp Res 28:59.
Kranzler HR, Armeli S, Feinn R, Tennen H (2004) Targeted naltrexone treat-
ment moderates the relations between mood and drinking behavior among
problem drinkers. J Consult Clin Psychol 72:317–327.
Levine J, Schooler N (1986) SAFTEE: a technique for the systematic assess-
ment of side effects in clinical trials. Psychopharmacol Bull 22:343–381.
MacKillop J (2006) Factor structure of the alcohol urge questionnaire under
neutral conditions and during a cue-elicited urge state. Alcohol Clin Exp
MacKillop J, Menges DP, McGeary JE, Lisman SA (2007) Influences of crav-
ing and DRD4 VNTR genotype on the relative value of alcohol. Behav
Brain Funct, 3: doi:10.1186/1744-9081-3-11.
Martin CS, Earleywine M, Musty RE, Perrine MW, Swift RM (1993) Devel-
opment and validation of the Biphasic Alcohol Effects Scale. Alcohol Clin
Exp Res 17:140–146.
McGeary JE, Monti PM, Rohsenow DJ, Tidey J, Swift R, Miranda R Jr
(2006) Genetic moderators of naltrexone’s effects on alcohol cue reactivity.
Alcohol Clin Exp Res 30:1288–1296.
Monti PM, Binkoff JA, Zwick WR, Abrams DB, Nirenberg TD, Liepman
MR (1987) Reactivity of alcoholics and nonalcoholics to drinking cues. J
Abnorm Psychol 96:122–126.
Monti PM, Rohsenow DJ, Hutchison KE (2000) Toward bridging the gap
between biological, psychobiological, and psychosocial models of alcohol
craving. Addiction 95:S229–S236.
Monti PM, Rohsenow DJ, Rubonis AV, Niaura RS, Sirota AD, Colby SM,
Abrams DV (1993) Alcohol cue reactivity: effects of detoxification and
extended exposure. J Stud Alcohol 54:235–249.
Nguyen SA, DeLeon CP, Middaugh LD (2004) Topiramate effects on ethanol
consumption by C57BL⁄6 mice. Alcohol Clin Exp Res 28:16A.
O’Malley SS, Jaffe AJ, Chang G, Rode S, Schottenfeld R, Meyer RE, Roun-
saville B (1996) Six month follow-up of naltrexone and psychotherapy for
alcohol. Arch Gen Psychiatry 53:217–224.
O’Malley SS, Krishnan-Sarin S, Farren C, Sinha R, Kreek MJ (2002) Naltrex-
one decreases craving and alcohol self-administration in alcohol-dependent
subjects and activates the hypothalamo-pituitary-adrenocortial axis. Psycho-
Raguraman J, Priyadharshini KR, Chandrasekaran R (2005) Effects of topira-
mate in alcohol dependence. Aust N Z J Psychiatry 39:736–737.
Ray LA, Hutchison KE (2007) Effects of naltrexone on alcohol sensitivity and
genetic moderators of medication response: A double-blind placebo-
controlled study. Arch Gen Psychiatry 64:1069–1077.
Ray LA, Hutchison KE, MacKillop J, Miranda R Jr, Audette A, Swift R,
Monti PM (2008) The effects of naltrexone during the descending limb of
the blood alcohol curve. Amer J Addict, in press.
Rohsenow DJ (2004) What place does naltrexone have in the treatment of
alcoholism?. CNS Drugs 18:547–560.
Rohsenow DJ, Howland J, Minsky SJ, Arnedt JT (2006) Effects of heavy
drinking by maritime academy cadets on hangover, perceived sleep, and
next-day ship power plant operation. J Stud Alcohol 67:406–415.
Rohsenow DJ, Monti PM, Abrams DB, Rubonis AV, Niaura RS, Sirota
AD, Colby SM (1992) Cue elicited urge to drink and salivation in alco-
holics: Relationship to individual differences. Adv Behav Re Ther 14:195–
Rohsenow DJ, Monti PM, Hutchison KE (2000) Naltrexone’s effects on
reactivity to alcohol cues among alcoholic men. J Abnorm Psychol 109:738–
Rubio G, Ponce G, Jimenez-Arriero MA, Palomo T, Manzanares J, Ferre F
(2004) Effects of topiramate in the treatment of alcohol dependence. Phar-
Sayette MA, Martin CS, Wertz JM, Shiffman S, Perrott MA (2001) A multi-
dimensional analysis of cue-elicited craving in heavy smokers and tobacco
chippers. Addiction 96:1419–1432.
Sobell LC, Maisto SA, Sobell MB, Cooper AM (1979) Reliability of alcohol
abusers’ self-reports of drinking behavior. Behav Res Ther 17:157–160.
Staiger PK, White JM (1991) Cue reactivity in alcohol abusers: stimulus speci-
ficity and extinction of the responses. Addict Behav 16:211–221.
Swift RM, Whelihan W, Kuznetsov O, Buongiorno G, Hsuing H (1994) Nal-
trexone- induced alternations in human ethanol intoxication. Am J Psychia-
Tidey JW, Monti PM, Rohsenow DJ, Gwaltney CJ, Miranda R Jr, McGeary
JE, MacKillop J, Swift RM, Abrams DB, Shiffman S, Paty JA (2007) Mod-
erators of naltrexone’s effects on drinking, urge and alcohol effects in the
natural environment. Alcohol Clin Exp Res 32:1–9.
MIRANDA ET AL.
Watson D, Clark LA, Tellegen A (1988) Development and validation of brief Download full-text
measures of positive and negative affect: The PANAS scales. J Pers Soc Psy-
Weiss F, Lorang MT, Bloom FE, Koob GF (1993) Oral alcohol self-adminis-
tration stimulates dopamine release in the rat nucleus accumbens: Genetic
and motivational determinants. J Pharmacol Exp Ther 267:250–258.
Weiss F, Porrino LJ (2002) Behavioral neurobiology of alcohol addiction:
Recent advances and challenges. J Neurosci 22:3332–3337.
Wertz JM, Sayette MA (2001) A review of the effects of perceived drug use
opportunity on self-reported urge. Exp Clin Psychopharmacol 9:3–13.
MECHANISMS OF TOPIRAMATE ACTION