Effects of topiramate on cue-induced cigarette craving and the response to a smoked cigarette in briefly abstinent smokers
- First Online:
- Cite this article as:
- Reid, M.S., Palamar, J., Raghavan, S. et al. Psychopharmacology (2007) 192: 147. doi:10.1007/s00213-007-0755-6
- 154 Views
Clinical studies have shown that topiramate, an anticonvulsant medication, may be effective as a treatment for smoking cessation. However, less is known about topiramate effects on nicotine withdrawal and craving and its interactions with a smoked cigarette.
The objective of this study was to investigate the effects of topiramate treatment on abstinence-related nicotine withdrawal, cue-induced cigarette craving, and the acute effects of a smoked cigarette.
Materials and methods
Fifteen female and 25 male cigarette smokers were randomly assigned to 9-day treatment with topiramate (final titration dose, 75 mg/day) or placebo. On the last day of treatment, after a 3-h smoke-free abstinence period, participants were evaluated for symptoms of nicotine withdrawal and then underwent cigarette and neutral cue reactivity testing. Thirty minutes after completing cue exposure testing, participants were then evaluated for the acute effects of a smoked cigarette. Cue reactivity and acute smoking measures included subjective ratings of cigarette craving and withdrawal and physiological measures of skin conductance and temperature, heart rate, and blood pressure. In addition, smoking topography was measured using a puff volume apparatus.
Topiramate treatment enhanced subjective ratings of withdrawal after the 3-h abstinence period and reduced pre-cue skin conductance levels. Cigarette cue exposure resulted in a moderate increase in craving, which was unaffected by treatment. Topiramate treatment enhanced the rewarding effects of a smoked cigarette, even while participants smoked less per puff and achieved lower plasma nicotine levels.
Results suggest that topiramate enhances both nicotine withdrawal and reward. These findings question the utility of topiramate treatment for smoking cessation.
Cigarette smoking is the single greatest contributor to preventable death in the United States, with over 430,000 deaths attributed to cigarettes and other forms of tobacco annually (CDC 2005). Cigarette smoking cessation treatment, similar to most other forms of drug dependence, includes a comprehensive approach with both medication and counseling. First line pharmacotherapies for smoking cessation include nicotine replacement therapy (NRT), sustained release bupropion (Zyban®), and the recently approved nicotine receptor partial agonist varenicline (Chantix®). These medications achieve abstinence rates of approximately 30–50% upon the completion of treatment, and approximately 15–25% at 6–12 months after the initial quit date (Fiore et al. 1994; Jorenby et al. 1999, 2006; Hughes et al. 1999; Nides et al. 2006). These results, in particular the long-term abstinence rates, indicate the need for more effective medications for smoking cessation. A pair of recent clinical trials has provided preliminary evidence that topiramate may be an effective smoking cessation treatment medication (Johnson et al. 2005; Khazaal et al. 2006). In a placebo-controlled trial of topiramate treatment of alcohol dependence, subjects receiving topiramate were more likely to quit cigarette smoking despite not being explicitly treated for smoking cessation (Johnson et al. 2005). In a small open-label trial of topiramate as a smoking cessation treatment medication, 9 of 13 patients reduced their daily smoking, and six of them quit smoking within the first 4 weeks of treatment (Khazaal et al. 2006).
Topiramate (Topamax®), a sulfamate-substituted fructopyranose derivative, is from a new generation anti-epileptic medications with both GABAergic and glutamatergic properties. Its anti-epileptic activity has been attributed to several mechanisms which inhibit neuronal activity, including suppression of voltage-sensitive Na+ channels (Zona et al. 1997), enhancement of GABA mediated Cl− conductance into the neuron (White et al. 1997), elevation of cerebral GABA levels (Kuzniecky et al. 2002), and antagonism of the AMPA glutamate receptor (Gibbs et al. 2000; Ängehagen et al. 2001). The therapeutic effects of topiramate in the treatment of addiction have been hypothesized to be due to the inhibition of dopamine release in the mesolimbic dopamine pathways, via glutamate receptor antagonism and/or the enhancement of GABA neurotransmission (Johnson 2004). The involvement of glutamate in the rewarding properties of nicotine has been demonstrated in numerous studies. Nicotine stimulates glutamate levels in the nucleus accumbens (Reid et al. 2000) and VTA (Schilstrom et al. 2000), nicotine stimulation of nucleus accumbens dopamine release is blocked by glutamate receptor antagonists (Miyasato 2001; Sziraki et al. 1998), and nicotine self-administration is inhibited by the metabatropic glutamate receptor antagonist 2-methyl-6-(phenylethynyl)-pyridine (Paterson et al. 2003). GABAergic modulation of nicotine-induced dopamine activity and reward is also well documented. Direct and indirect GABAergic agonists baclofen and gamma-vinyl GABA are potent inhibitors of nicotine-induced dopamine release (Fadda et al. 2003), nicotine-conditioned locomotor sensitization (Bevins et al. 2001) and nicotine self-administration (Fattore et al. 2002; Patterson and Markou 2002). These findings indicate that glutamatergic antagonists and GABAergic agonists reduce the acute rewarding and reinforcing effects of nicotine.
Results from these pre-clinical and clinical studies indicate the need to further investigate topiramate effects on cigarette smoking behavior. Furthermore, little is known about topiramate effects on nicotine withdrawal and cigarette craving during smoking cessation. Therefore, the current study characterized the effects of topiramate on the subjective and physiological responses to nicotine withdrawal, cue-elicited cigarette craving, and the acute effects of a smoked cigarette in cigarette smokers.
Materials and methods
Healthy individuals who smoked at least 15 cigarettes/day were recruited from the New York City area by newspaper advertisement. To obtain study participants similar to individuals that enroll in smoking cessation treatment, more appropriate for medication-screening tests (see Perkins et al. 2006), participants interested in quitting smoking were recruited. All participants gave informed consent before their entry into the study. Participants were informed that the investigational medication being tested was not a proven treatment and that this was not a smoking cessation study, but that they would receive referral for smoking cessation treatment after completing the study. All enrolled participants had normal results for physical, laboratory, and psychiatric examinations. Exclusion criteria included evidence of current depression (BDI >17), current drug or alcohol dependence, current axis-I psychiatric disorders which require medication treatment, and current, or within the last 6 months, enrollment in smoking cessation treatment. Screening, treatment, and testing were performed at the VA New York Harbor Healthcare System (VANYHHS), Manhattan Campus, and the participants were compensated for time and travel. This study was approved by the VANYHHS Human Subjects Subcommittee and the New York University School of Medicine Institutional Review Board and was performed in accordance with the ethical standards of the 1964 Declaration of Helsinki.
Topiramate n = 19
Placebo n = 21
43.0 ± 13.7
43.3 ± 11.2
13.7 ± 1.8
13.4 ± 2.0
198 ± 48
190 ± 53
Heavy smoker (>25/day)
Light smoker (15–25 per day)
Cigarettes per day
21.9 ± 10.5
21.7 ± 9.0
22.9 ± 14.1
26.4 ± 11.9
6.6 ± 2.2
7.3 ± 1.9
Number of prior quit attempts
4.4 ± 6.2
5.4 ± 6.7
5.2 ± 5.0
5.5 ± 6.3
Study eligible patients were randomly selected to receive topiramate (Topamax®) or placebo in a randomized, double-blind, dose-escalating design. Medication assignment was based on a randomization schedule using a 1:1 ratio and permutation blocks of four prepared by a study statistician. Topiramate dose titration used the following schedule: 25 mg in the AM and placebo in the PM on days 1–3 (BID dosing), 25 mg BID on days 4–7, and 50 mg in the AM and 25 mg in the PM(BID dosing) on days 8–9 to achieve a final test dose of 75 mg/day. Placebo treatment followed the same dispensing schedule. The test dose of topiramate was selected based on prior studies of topiramate (25–50 mg, single dose) interactions with intravenous nicotine (Sofuoglu et al. 2006) and on prior double-blind (average 165 mg/day) and open-label (average 90 mg/day) treatment studies with topiramate in healthy, non-psychiatric co-morbid smokers (Khazaal et al. 2006; Anthenelli et al. 2006). In addition, we wished to limit topiramate-related nervous system side effects (e.g., fatigue, difficulty concentrating) previously reported with doses greater than 100 mg/day (Anthenelli et al. 2006; Mecarelli et al. 2001). Commercially available topiramate and placebo medications were packaged and blinded by the research pharmacist employing an over-encapsulation process with 00 size, opaque capsules. Each treatment arm was subject to stratification based on heavy (>25 cigarettes/day) and light (15–25 cigarettes/day) smoking habits to assure an equal balance of smoking severity across groups. All study patients ingested their first dose of medication at the research clinic and were observed by staff for 1 h before leaving. Patients were instructed to continue their usual daily smoking habits and not to attempt to quit smoking while in the study. A clinical safety visit was scheduled 4 days after the start of medication treatment. At all visits ad hoc, smoking behavior and compliance with the daily medication regimen was assessed by self-report.
On the baseline testing day, each patient smoked one cigarette (of their preferred brand) before entering the hospital, and then 30 min later a number of subjective rating scales were completed: questionnaire of smoking urges-brief (QSU), the withdrawal scale for tobacco (WST), and the cigarette evaluation questionnaire (CEQ). In addition, patients reported on the use of cigarettes/day and any other nicotine-containing products over the preceding week, and exhaled carbon monoxide (CO) was measured. After completing the baseline assessments, subjects began day 1 of study medication treatment.
Test day evaluation
On the test day, study patients were tested after 3 h of smoking abstinence. This duration of abstinence was based on previous studies demonstrating moderate levels of craving and withdrawal, which remain susceptible to cue stimulation test procedures, after a similar period of abstinence (Hatsukami et al. 1984; Hughes 1992; Killen and Fortmann 1997; Tiffany et al. 2000; Rose et al. 2001). The abstinence period was completed while at the hospital, and compliance was verified by: direct observation by the research staff, self-report of no smoking, and exhaled CO measures (CO <15 ppm) taken at the end of the 3-h period. Subjects took their last dose of study medication upon arrival at the clinic, and during the 3-h abstinence period, they reported on their use of cigarettes and any other nicotine-containing products over the preceding 9 days of treatment. A light lunch was offered. Testing comprised of cue exposure tests followed by a cigarette smoking test.
All testing was done in an isolated room with negative pressure ventilation. In addition, smoke from exposed (open air) burning cigarettes was managed by high volume carbon and hepa filter air purifiers placed adjacent to the testing chair as well as a smokeless ashtray with carbon filter. Cue testing began between 11:00 a.m. and 12 noon and took approximately 90 min to complete. During the cue tests, the patients were seated in a comfortable, high back chair with arm rests and a small, mobile writing table and television set in front of them. Initially, a blood pressure monitoring cuff and electrodes for skin conductance and skin temperature were placed on each patient’s non-writing arm and fingertips, respectively. Thereafter, the patients were given a 3-min resting period, and then, baseline blood pressure and heart rate measures were obtained followed by completion of the subjective rating scales (QSU, WST, choice procedure). Once the cardiovascular measures and subjective rating scales were completed, 10 min of baseline skin conductance and skin temperature were recorded before commencing with cue exposure testing. The cue exposure sessions lasted 10 min and involved a combination of tactile, olfactory, visual, and audio cues modeled after previous studies (Drobes and Tiffany 1997; Mucha et al. 1999; Hutchinson et al. 1999; Shiffman et al. 2003). Continuous measurements of skin conductance and skin temperature were collected during cue exposure. Immediately after the completion of cue exposure, the subjective rating scales (QSU, WST, choice procedure) were completed, and then, blood pressure and heart rate were measured. There was a 30-min resting period between the neutral and cigarette cue tests. The neutral and cigarette cues were presented in a random order, which was counterbalanced across treatment and employed a block size of four (e.g., 50% of the patients each treatment group had an active cue then neutral sequence, and vice versa).
During the cigarette cue session, each patient was presented with a lighter, ashtray, and three to four new packs of cigarettes, including one of their preferred brand. All items were placed on the table in front of them. Initially, the patients handled each of the packs and then selected their preferred brand as if they were about to smoke. Then, the research assistant opened the selected pack, and the patient, using his/her non-writing hand, removed a cigarette and held it in his/her hand, smelled the tobacco in the cigarette, and lit the cigarette (with the aid of a research assistant so as not to inhale smoke). The lit cigarette was then placed in the ashtray in front of the patient, who viewed the lit cigarette and smelled the smoke for approximately 30 s before extinguishing it in the ashtray (paraphernalia phase lapsed time, 5 min). The patient then watched a video depicting scenes of people smoking: restaurant/bar with two people smoking cigarettes, a person having a cigarette after completing a meal, co-workers having a cigarette break outside an office building, and people speaking about the pleasures of smoking (video lapsed time, 5 min). After the video, the patient re-lit the cigarette in the ashtray and smelled the smoke of the burning cigarette for another 20–30 s.
This session was similar to the cigarette cue session in procedure but not in content. During the neutral cue session, the items on the table included seashells, string, computer disks, a plastic spoon, pinecones, and rocks; the patient handled each of the items, and instead of a cigarette, the patient smelled a cinnamon stick or container of nutmeg to assure a comparable level of olfactory stimulation. The video depicted similar neutral items, as well as people in office and outdoor surroundings speaking about basic work and home routines.
Cigarette smoking test
The smoking test was limited to one cigarette to assess the pharmacokinetic, physiological, and subjective effects of a single-dose unit (one cigarette) and to retain the viability of the post-smoking choice procedure. The cigarette smoking tests were performed 30 min after completion of the neutral and cigarette cue exposure sessions (starting between 1 and 2 p.m.) in the same testing room. Before testing, patients were allowed to stretch their legs and have a glass of water if they wished. The smoking tests took approximately 20 min.
Initially, a blood pressure monitoring cuff and electrodes for skin conductance and skin temperature were placed on each patient’s non-writing arm and finger tips, respectively. After a resting period of 3 min in the testing chair, baseline blood pressure and heart rate were measured, subjects completed the subjective rating scales (QSU, WST, choice procedure), and then, 10 min of baseline skin conductance and skin temperature were recorded. Patients then smoked a cigarette of their preferred brand using the controlled puff volume apparatus. The patients were instructed to smoke as much as they wanted to and were not required to finish the entire cigarette. Number of puffs and volume per puff were measured during smoking, as well as continuous measurements of skin conductance and skin temperature. Immediately after cigarette smoking was completed (indicated by the subject expressing that they were finished), patients completed the cigarette evaluation questionnaire (CEQ), the subjective rating scales (QSU, WST, choice procedure), and blood pressure and heart rate levels were measured. Fifteen minutes after completing the cigarette smoking procedure, a blood sample was obtained.
After testing was completed, all study patients were scheduled for a follow-up visit 5 days later, and referral for smoking cessation treatment at local hospitals and clinics in New York City was provided.
Skin conductance (μS) and skin temperature (°F) were recorded using fingertip electrodes connected to a PC computer via a RS232 beltpack/interface unit (Biograph, Thought Technology, Montreal, Canada). Skin conductance was recorded using Ag–Ag/Cl paste electrodes with a constant excitation set at 0.5 V d.c.. Skin temperature was recorded using an electric thermister that monitors absolute temperature. Heart rate and blood pressure were measured using a standard blood pressure cuff and stethoscope, taken in an upright sitting position. Comparison of skin conductance and temperature values employed means of the baseline period (5 min) and the last minute of the cue presentation or cigarette smoking paradigms.
Questionnaire of smoking urges-brief (QSU)
The QSU (Cox et al. 2001), based on an earlier version of this questionnaire (Tiffany and Drobes 1991), includes subject-rated questions related to desire to smoke a cigarette. Factor 1, items related to intention and expected positive outcome of smoking (appetitive craving) and Factor 2, items related to anticipation of relief from negative affect (withdrawal craving), were evaluated.
Withdrawal scale for tobacco (WST)
The WST is a modified version of the Minnesota withdrawal scale (Hughes et al. 1991; Hatsukami et al. 1997; Hughes and Hatsukami 1986; Eissenberg et al. 1996) in which the insomnia/disrupted sleep item was removed, as it is not relevant in cue reactivity testing. In this subject-rated scale, the first item: “craving for a cigarette”, and the total of items two to seven: “irritability/frustration/anger”, “anxiety”, “difficulty concentrating”, “restlessness/impatience”, “increased appetite”, and “depression” were evaluated as measures of craving and withdrawal, respectively.
Cigarette evaluation questionnaire (CEQ)
The CEQ (Westman et al. 1992; Rose et al. 1998, 2000, 2001) is a subject-rated scale that assesses the patient’s response to a smoked cigarette. There are ten items which can be scored in five domains: satisfaction, psychological reward, alleviation from withdrawal, sensory feelings, and reduction in craving.
The choice procedure (Kidorf et al. 1995) asks subjects to hypothetically choose between smoking a cigarette now or receiving a small amount of money (from 10 cents up $6 in increments of 10 cents) available at the end of the day. A crossover ($) value, at and above which patients prefer money, is obtained.
Controlled puff volume apparatus
The controlled puff volume apparatus (University Research Glass, Chapel Hill, NC) is a tabletop device which holds a lit cigarette inside a volumetric glass chamber with a constant air supply to keep the cigarette from extinguishing. When not in use by the subject, smoke is exhausted through an outlet tube that feeds into the ventilation system. When used by the subject, puff volume is quantified using the intake cylinder. Number of puffs and volume per puff per cigarette were measured.
For each parameter, a multivariate approach to repeated measures analysis of variance (ANOVA) was used (mixed repeated and between groups analyses). Significant interactions were followed up with post hoc analyses of simple effects (e.g., three-way interaction between medication, cue, and time would be followed by comparisons between treatment groups at individual time points) with Bonferroni corrections for repeated measures. Order effects were initially tested for as a factor in the cue reactivity repeated measures ANOVA (no effect was found).
The effects of a 3-h smoking abstinence period on subjective ratings of cigarette craving and withdrawal were analyzed using the day 1 (30-min post-smoke, non-abstinent) the day 9 pre-cue (3-h post-smoke, abstinent) measures collected at the beginning of each study visit. Topiramate versus placebo conditions were compared in a two (day 1, day 9) by two (placebo, topiramate) repeated measures ANOVA, followed by post hoc comparisons of the day 9 values in a one-way ANOVA. The effects of treatment medication on cue-induced cigarette craving, withdrawal, and physiological measures were tested using the day 9 cue-induced cigarette craving tests. Topiramate versus placebo conditions were compared in a two (neutral cue, active cue) by two (pre-cue, post-cue) by two (placebo, topiramate) repeated measures ANOVA. The effects of the smoked cigarette on self-reported cigarette craving, withdrawal, alternate choice ($), heart rate, blood pressure, and physiological measures were compared using data collected on day 9 in a two (pre-smoke, post-smoke) by two (placebo, topiramate) repeated measures ANOVA. The effects of the smoked cigarette on reward and satiation (CEQ) were compared using the day 1 (30 -min post-smoke) and the day 9 (post-test cigarette) measures in a two (day 1, day 9) by two (placebo, topiramate) repeated measures ANOVA, followed by post hoc comparisons of the day 9 values in a one-way ANOVA. In addition, pairwise comparisons of the number of puffs, volume per puff, total puff volume during smoking, and plasma nicotine levels were analyzed in a one-way ANOVA.
Subjects smoked approximately a pack a day at the beginning of treatment (day 1: cigarettes/day: topiramate = 22.7 + 11.1, placebo = 19.5 + 12.3; exhaled CO: topiramate = 15.7 + 9.9 ppm, placebo = 16.7 + 7.2 ppm) and showed a minor, but non-significant, decrease by the end of treatment (day 9: cigarettes/day: topiramate = 15.1 + 8.3, placebo = 15.7 + 10.2; exhaled CO: topiramate = 14.3 + 10.6 ppm, placebo = 15.4 + 9.6 ppm). There was no difference between topiramate and placebo in ad lib smoking rates or exhaled CO levels. Adherence to the medication treatment regimen was similar in both groups (full adherence: topiramate = 98%, placebo = 99%). Adverse events, primarily rated as mild with some mild to moderate, were reported by 14 patients (74%) in the topiramate group and 15 patients (68%) in the placebo group. The most common event was drowsiness, which was reported by 22% patients in the topiramate group and 18% patients in the placebo group. All other adverse event rates were 10% or less. There were no differences in adverse event rates between the topiramate and placebo groups.
Cue reactivity tests
There was a significant effect of cigarette cues on subjective ratings of WST craving [F(1,152) =4.075, p < 0.05], whereas other measures showed weaker, non-significant increases in response to smoking cues. Post hoc analyses of the individual items in the QSU (e.g., urge to smoke) and WST (e.g., anxiety) did not reveal any significant cue effects. In addition, there was a main effect of topiramate treatment on WST withdrawal [F(1,152) =8.770, p < 0.01] which was higher in the topiramate group irrespective of cue type. However, there were no significant interactions of medication treatment with cue exposure for any of the subjective measures.
Physiological measures: cigarette cue and smoking tests
Cigarette cue test
Heart Rate (bpm)
72 ± 8
72 ± 7
69 ± 19
69 ± 20
Diastolic blood pressure (mmHg)
107 ± 11
107 ± 12
99 ± 25
101 ± 26
Systolic blood pressure (mmHg)
71 ± 11
72 ± 9
63 ± 19
66 ± 17
Skin temperature (°F)
88 ± 3
87 ± 4
86 ± 8
85 ± 8
Skin conductance (μS)
0.49 ± 0.50a
delta: 0.01 ± 0.14
0.89 ± 0.64
delta: −0.06 ± 0.38
Heart rate (bpm)
70 ± 10
74 ± 11
66 ± 19
74 ± 21
Diastolic blood pressure (mmHg)
106 ± 12
111 ± 12
100 ± 25
103 ± 26
Systolic blood pressure (mmHg)
71 ± 8
76 ± 8
66 ± 17
69 ± 17
Skin temperature (°F)
82 ± 8
81 ± 8
83 ± 10
82 ± 9
Skin conductance (μS)
0.45 ± 0.80
Delta: −0.05 ± 0.16
0.58 ± 0.59
delta: −0.17 ± 0.37
Cigarette smoking tests
Number of puffs
8.9 ± 5.0
10.4 ± 4.5
Average puff volume (cc)
31.4 ± 10.8a
41.9 ± 17.4
Total puff volume (cc)
291 ± 197a
434 ± 243
Plasma nicotine (ng/ml)
18.7 ± 8.8a
26.5 ± 7.7
Day 9 smoking-induced heart rate, blood pressure, skin temperature, and skin conductance levels are also shown in Table 2. Pairwise comparison of pre-smoke values revealed no differences in baseline heart rate, blood pressure, skin temperature, or skin conductance. Due to variability in baseline skin conductance levels, analyses of smoking effects on skin conductance employed change from baseline values. There was no effect of smoking on skin conductance and no interaction with medication treatment. Analyses of smoking effects on heart rate, blood pressure, and skin temperature revealed a significant effect of smoking on heart rate [F(1,66) = 4.078, p < 0.05] and, at a trend level, systolic blood pressure [F(1,74) = 3.327, p = 0.07], but no interaction with medication treatment was found.
In the present study, topiramate had both positive and negative effects on patients’ subjective ratings, depending on their cigarette smoking status. When patients were in a state of brief smoking abstinence, topiramate enhanced ratings of nicotine withdrawal. In contrast, when patients smoked a cigarette, topiramate enhanced the rewarding effects of nicotine. In both cases, the effects treatment medication augmented procedurally induced response measures, demonstrating that topiramate has robust effects on various components of smoking behavior.
Cigarette craving is a prominent feature of the tobacco withdrawal syndrome and has been suggested to be a major contributor to the high rate of relapse observed in smoking cessation treatment programs. Studies have reported an association of craving with smoking cessation treatment outcome (Cummings et al. 1985; Durcan et al. 2002) and have shown that the levels of cigarette craving during the first few days of abstinence, controlling for differences in nicotine dependence, are predictive of both short- and long-term treatment outcome (Doherty et al. 1995; Westman et al. 1997; Killen et al. 1991; Killen and Fortmann 1997). In retrospective studies, nicotine withdrawal symptoms such as depression, anxiety, and frustration have also been found to be associated with relapse (Cummings et al. 1985), although in controlled laboratory studies, this association with nicotine withdrawal symptom severity was less consistent (Hughes and Hatsukami 1986; Hughes et al. 1990; Killen and Fortmann 1997). These findings suggest that the ability of a medication to reduce abstinence-induced cigarette craving, and perhaps withdrawal, may be indicative of its therapeutic potential in smoking cessation treatment. The results from studies on NRT and bupropion, two established smoking cessation treatments, address this suggestion. NRT has been shown to reduce symptoms of nicotine withdrawal after a 3-h to 3-day period of smoking abstinence (Rose et al. 1985; Jorenby et al. 1996; Tiffany et al. 2000; West and Shiffman 2001), although its effects on cigarette craving have been mixed with some reporting a drop in craving (Jorenby et al. 1996; Teneggi et al. 2002) and others finding no change (Pickworth et al. 1996; Hutchinson et al. 1999). Bupropion has also been shown to reduce abstinence-related nicotine withdrawal (Shiffman et al. 2000; Teneggi et al. 2005), although effects on craving were less consistent (Teneggi et al. 2005; Durcan et al. 2002; but see Shiffman et al. 2000; Cousins et al. 2001a). Both treatments can also affect cue-induced cigarette craving (Hutchinson et al. 1999; Shiffman et al. 2003; Brody et al. 2004), although the effect of NRT may be treatment-related but not cue-specific (Tiffany et al. 2000). In none of these studies did medication treatment enhance abstinence-induced nicotine withdrawal, as did topiramate. Moreover, topiramate treatment had no effect on cue-induced cigarette craving and was associated with an overall increase in withdrawal symptoms irrespective of cue type. These properties suggest that topiramate is not an effective treatment for managing cigarette craving and withdrawal during brief smoking cessation.
These findings are in contrast to two previous clinical trials in cigarette smokers which found that topiramate treatment was associated with high rates of smoking abstinence (Johnson et al. 2005; Khazaal et al. 2006). However, the study by Johnson et al. (2005) was not designed as a smoking cessation study, and the decrease in smoking was associated with concomitant abstinence from alcohol drinking. The study by Khazaal et al. (2006) was a small, open-label clinical trial lacking a control group, all of which limit the conclusiveness of their data. More recently, a double-blind placebo-controlled study of topiramate treatment in smoking cessation was reported (Anthenelli et al. 2006). This study found no difference in smoking abstinence rates between topiramate and placebo treatment. The findings from this clinical trial are consistent with our findings.
The acute reinforcing properties of a cigarette may be predictive of subsequent withdrawal and relapse in subjects enrolled in a smoking cessation program (Perkins et al. 2002). Indeed, studies on the subjective, physiological, and reinforcing effects of a smoked cigarette, after a designated period of abstinence, may represent a model of the “first slip” cigarette in a relapse event. Chronic pretreatment with nicotine patches has been reported to decrease ratings of reward and satisfaction derived from a smoked cigarette (Levin et al. 1994; Rose et al. 1994) and the ability of the cigarette to attenuate withdrawal (Brauer et al. 1999; Rose et al. 1985). An acute dose of bupropion has also been shown to reduce the “buzzed” and stimulatory effects of a smoked cigarette (Cousins et al. 2001a), although this was associated with an increase in the quantity of cigarettes smoked ad lib. These findings demonstrate the ability of effective smoking cessation treatments to attenuate the rewarding properties of a smoked cigarette. In the present study, however, topiramate treatment enhanced the rewarding effects of a smoked cigarette. In addition, subjects needed to smoke less, based on puff volume and plasma nicotine levels, to achieve their desired level of satisfaction. This finding is consistent with a recent report that acute topiramate treatment (25 and 50 mg) enhanced the subjective strength and rewarding effects of intravenous nicotine (0.5 and 1.0 mg/kg) (Sofuoglu et al. 2006). Studies on methamphetamine found similar results in which intravenous methamphetamine-induced (15 and 30 mg) ratings of euphoria and stimulation were enhanced by topiramate (100 and 200 mg) pretreatment (Johnson et al. 2006). This latter study also found evidence that topiramate reduced positive subjective mood (Johnson et al. 2006), similar with the present finding that topiramate produced an overall increase in symptoms of nicotine withdrawal. Taken together, these findings demonstrate that topiramate treatment enhances the rewarding effects of nicotine.
The mechanisms by which topiramate enhances the rewarding effects of nicotine are unclear. One possible explanation is that topiramate stimulates dopamine release on its own (Okada et al. 2005), thereby adding to the dopaminergic effects of nicotine. However, animal studies have shown that topiramate pretreatment reduced nicotine-induced dopamine and norepinephrine release in the nucleus accumbens (Schiffer et al. 2001). In terms of potential GABAergic or glutamatergic effects, previous clinical studies have reported that baclofen treatment inhibited the “liking” of a cigarette (Cousins et al. 2001b) and the “high” produced by i.v. cocaine (Rotherman-Fuller et al. 2004), whereas the glutamate antagonist memantine had no effect on the perception of intranasal nicotine (Thuerauf et al. 2007) and actually enhanced the “high” produced by i.v. cocaine and methamphetamine (Collins et al. 1998; Hart et al. 2002). These data would suggest that the effects of topiramate on nicotine reward are not likely mediated via its GABAergic agonist properties, and leave open the possibility that its glutamatergic antagonist properties are involved. It is also possible that negative mood and the increased level of nicotine withdrawal symptoms in the topiramate group before smoking (see Fig. 3) created a condition whereby the appreciation of a subsequent nicotine dose was greater. Indeed, drowsiness experienced by the topiramate group may have contributed to this perception. This proposed mechanism for enhanced rewarding effects is consistent with the previous study on topiramate interactions with methamphetamine (Johnson et al. 2006).
There are a number of weaknesses to this study that should be acknowledged. Topiramate was tested at only one dose (75 mg/day), which is lower than the dose used in prior smoking cessation studies (Johnson et al. 2005; Khazaal et al. 2006; Anthenelli et al. 2006) as well as the recommended dose for treating epilepsy (PDR 2005). We had considered continuing dose titration to a higher level and repeating cue tests; however, initial tests found that subjects were significantly less cue-reactive during the second test. This drop in cue reactivity is consistent with previous cigarette cue studies (Payne et al. 1996; Tiffany et al. 2000), and the cue effects in this study were generally weak. While cue-induced craving was readily detected by the WST craving item, the QSU craving items were variable and unable to detect an effect of cigarette cues. In addition, cue exposure had no effects the physiological measures. In contrast, previous cue exposure studies have demonstrated robust cue-induced craving using the QSU craving items (Tiffany et al. 2000; Cox et al. 2001; Drobes and Tiffany 1997; Elash et al. 1995), and physiological responding has been reported by many (Niaura et al. 1989; Drobes and Tiffany 1997; Tiffany et al. 2000). This difference may be due to the cue paradigm employed in our studies, which had a longer cue session duration and lacked in vivo exposure of another person smoking. The abstinence period employed in the current study was fairly brief, to retain the utility of cue reactivity testing, and therefore, did not capture the period at which post-cessation nicotine withdrawal is maximal. Finally, topiramate effects on nicotine withdrawal may have been influenced by medication side effects such as fatigue and difficulty concentrating, although this was limited by the low testing dose. Indeed, there were no group differences in adverse event rates, including drowsiness.
In summary, the present study demonstrates that topiramate treatment causes smokers to experience more symptoms of nicotine withdrawal during periods of brief cigarette abstinence. In addition, topiramate treatment enhances the rewarding effects of a smoked cigarette despite lower nicotine intake. These properties do not support the suggestion that topiramate treatment is an effective therapy for smoking cessation treatment. Indeed, such psychopharmacological properties could increase both the likelihood of a first slip cigarette and the possibility that such a slip then leads to a full relapse in abstinent smokers.
This study was supported by NIDA grant R21 DA017556 (MS Reid). All experiments were approved by the Institutional Review Board of New York University School of Medicine and the VA New York Harbor Healthcare System. None of the authors have any financial interest with the study sponsor or the study medication. We wish to thank Tom Cooper, Jed Rose, Deborah Leon, and Pooja Paunikar for their support in this study.