Unbekannt

Psychopharmacology (2004) 173:153–159DOI 10.1007/s00213-003-1711-8 O R I G I N A L I N V E S T I G A T I O N Jessica Werth Cook · Bonnie Spring ·Dennis E. McChargue · Belinda Borrelli ·Brian Hitsman · Raymond Niaura · Nancy J. Keuthen ·Jean Kristeller Influence of fluoxetine on positive and negative affectin a clinic-based smoking cessation trial Received: 7 July 2003 / Accepted: 30 October 2003 / Published online: 15 January 2004 Springer-Verlag 2004 Abstract Rationale: Fluoxetine improves affect in clin- or negative affect. Conclusions: Results indicate that ical syndromes such as depression and premenstrual 60 mg of fluoxetine improves both positive and negative dysphoric disorder. Little is known about fluoxetine’s mood states after quitting smoking and that diminished influence on mood changes after quitting smoking, which positive affect may be an overlooked affective response to often resemble sub-clinical depression. Objectives: The present study, a re-analysis of previously published data(Niaura et al. 2002), examined fluoxetine’s effect on Keywords Fluoxetine · Smoking cessation · Positive changes in negative and positive affect following quitting smoking. Methods: Adult smokers (n=175) withoutclinically significant depression were randomized on adouble-blind basis to receive fluoxetine hydrochloride (30 or 60 mg daily) or placebo for 10 weeks in combinationwith cognitive-behavioral therapy (CBT) for smoking Abstaining from nicotine, even during short periods of cessation. We postulated that fluoxetine would benefi- time, is accompanied by a cluster of affective responses cially influence post-cessation changes in positive and resembling sub-clinical depression (Gilbert et al. 1998).
negative affect. Results: Mood change across treatment Irritability, anxiety, and depressed mood, for example, are was analyzed using mixed linear modeling controlling for commonly reported mood disturbances that are experi- initial level of nicotine dependence, plasma fluoxetine enced after quitting smoking (Hughes and Hatsukami metabolites, and change in cotinine (a nicotine metabo- 1986; American Psychiatric Association 1994; Gilbert et lite) at each visit. Relative to placebo, those on 60 mg al. 1998). Reductions in positive affect, however, have fluoxetine experienced an elevation in positive affect that largely been overlooked as a possible response to increased across time [t(526)=2.50, P=0.01], and a smoking cessation. That neglect is surprising, given the reduction in negative affect that returned to baseline integral relationship between deficient positive mood across time [t(524)=2.26, P=0.02]. There were no differ- states and depression (Clark and Watson 1991; Coyne ences between 30 mg and placebo on changes in positive 1994), and evidence that quitting smoking heightens therisk of depressive episodes (Borrelli et al. 1996; Tsoh etal. 2002). Positive affect declines after quitting smoking J. W. Cook · B. Spring ()) · D. E. McChargueDepartment of Psychology, (Gilbert et al. 1998; Lerman et al. 2002; Cook et al. 2003) 1007 West Harrison Avenue, Chicago, IL 60607-7137, USA and persists across 30 days of abstinence (Gilbert et al.
1998). Further, empirical evidence that positive and negative affect are linked to different neural underpin-nings (Davidson 1992) and have different psychological B. Borrelli · B. Hitsman · R. NiauraCenters for Behavioral and Preventive Medicine, correlates (Watson et al. 1988) supports the independence of these constructs. Thus, low positive affect may be an Suite 500, One Hoppin Street, Providence, RI 02903, USA important and neglected response to quitting smoking,which, in combination with elevated negative affect, constitutes a distressing affective syndrome that follows OCD Clinic, Massachusetts General Hospital, 149 13th Street, #9106, Charlestown, MA 02129, USA Similarities between cessation-induced mood distur- bance and other affective syndromes (e.g. depression, Department of Psychology, Indiana State University, premenstrual dysphoric disorder) suggest that antidepres- sant medications might also ameliorate unpleasant mood reinforcing value of approach behavior to rewards (Depue states after quitting smoking. A variety of antidepressants and Collins 1999), fluoxetine might buffer the losses in with dopaminergic, noradrenergic, and/or serotonergic positive affect associated with quitting smoking.
actions have demonstrated efficacy for promoting smok- The present study examined the effect of 30 and 60 mg ing cessation (Spring et al. 1995; Hurt et al. 1997; Hall et fluoxetine, relative to placebo, on acute change in positive al. 1998; Niaura et al. 2002). Studies on the effects of the and negative moods during nicotine abstinence. We agents have, for the most part, focused on the alleviation postulated that fluoxetine would buffer the increase, or of negative mood or depressive symptoms. Specifically, even diminish negative affect during smoking cessation.
euthymic smokers treated with nortryptiline (catechol- Moreover, we postulated that fluoxetine would buffer aminergic and serotonergic actions) or bupropion (dopa- post-cessation losses in positive affect, or, potentially, minergic and noradrenergic actions), report lower levels increase positive affect during smoking cessation.
of post-cessation negative affect than smokers who didnot receive these treatments (Hall et al. 1998; Shiffman etal. 2000; Lerman et al. 2002). Effects of antidepressants on positive affect are less understood (Zald and Depue2001). Although Shiffman et al. (2000) found that bupropion attenuated a decrease in positive affect in Subjects were 175 euthymic male and female smokers recruited to euthymic smokers during 72-h nicotine deprivation, three of 16 sites in a double-blind, placebo-controlled, multicenter Lerman and colleagues (2002) did not detect an effect trial examining the influence of fluoxetine on smoking cessation.
of bupriopion on post-cessation positive moods.
The present study is a re-analysis of previously published data The influence of fluoxetine, a selective serotonin (Niaura et al. 2002), and used the three-site subset that measured reuptake inhibiting antidepressant (SSRI), on mood positive and negative affect across visits. Subjects’ mean age was42.6 years (SD=9.4). Most were female (57.1%), Caucasian following smoking cessation has yet to be examined.
(94.6%), and married (52.6%). Subjects averaged 30.1 cigarettes Although fluoxetine, given for 3 weeks prior to a quit per day (SD=12.9) for a mean of 25.1 years (SD=9.1) prior to attempt reduced depression among euthymic smokers treatment. Enrollees were moderately dependent on cigarettes, as (Dalack et al. 1995), fluoxetine’s effect on affective indicated by a mean score of 6.8 (SD=1.8) on the FagerstromTolerance Questionnaire (Fagerstrom 1978). At entry, cotinine, a distress during nicotine abstinence remains unknown. In nicotine metabolite, averaged 288.7 ng/ml (SD=152.9). Subject other clinical syndromes, including major depression eligibility criteria, as well as the cessation outcomes of the multi- (Gram 1994), subsyndromal depression (Gram 1994), and center trial, are reported elsewhere (Niaura et al. 2002).
premenstrual dysphoric disorder (Cohen et al. 2002),fluoxetine has been shown to alleviate negative affect. In depression, the drug’s mood benefit is not confined topatients who present solely with depressed mood, but occurs also among patients with comorbid anxietydisorder (Sonawalla et al. 2002). In addition to engen- Positive affect was assessed at each visit via the Positive andNegative Affect Scale (PANAS; Watson et al. 1988). The PANAS dering improvements in depression, fluoxetine reduces a is a self-report state mood questionnaire consisting of 20 adjectives full range of negative moods, including anxiety and that are rated on 5-point scales ranging from 1 (very slightly or not anger-hostility (Sonawalla et al. 2002).
at all) to 5 (extremely). The positive affect subscale, manifested by To our knowledge, there have been no prior studies feelings of activation, elation, enthusiasm, and enjoyment, consistsof 10 items, with potential scores ranging from 10 to 50. The examining whether fluoxetine can prevent the dysphoric Positive Affect subscale has been shown to possess high internal mood and the decreased positive affect that typically follow nicotine abstinence. There are, however, threereasons to expect such effects. First, similar preventive benefits have been demonstrated for fluoxetine as atreatment of premenstrual dysphoric disorder, such that Negative affect was assessed at each visit via the negative affect the drug prevents intermittent worsening of negative subscale of the PANAS (Watson et al. 1988). The negative affect mood (Cohen et al. 2002). Second, although fluoxetine scale, encompassing feelings of distress, hostility, nervousness, has yet to be established as a treatment for smoking scorn, and gloominess, is comprised of ten words with scoresranging from 10 to 50. The negative affect scale possesses high cessation, it has been shown to yield a modest cessation internal consistency (Watson et al. 1988).
advantage to smokers who are trying to quit (Niaura et al.
2002), especially among those who have higher baselinelevels of depressive symptomatology (Hitsman et al.
1999). Such effects might partly arise from the drug’s The eight-item Fagerstrom Tolerance Questionnaire (FTQ; Fager- mood control properties. Third, administration of fluox- strom 1978) was administered to measure degree of behavioral responses suggestive of nicotine dependence (e.g. smoking many prevents a rise in brain reward threshold during nicotine cigarettes, smokes early in the morning). Scores ranged from 0 to11, with higher values suggesting greater nicotine dependence.
deprivation (i.e. reverses diminished responding to re- Correlations between the FTQ and measures of nicotine intake warding stimuli; Harrison et al. 2001). To the extent that support the construct validity of the scale (Fagerstrom and diminished positive affect is associated with diminished Saliva cotinine concentration was measured at each visit. Saliva model in a backwards manner and the model was refit. Two a-priori cotinine samples were analyzed by SciCor Laboratory (Indianapo- group contrasts were specified: 30 mg versus placebo; 60 mg versus Carbon monoxide, assessed via an ecolyzer (Model EC-50, It is important to note that mixed-effects regression modeling Vitalograph Corporation), was also assessed at each visit.
does not place restrictions on the number of observations perindividual, so that participants with missing data at a particularassessment time were not excluded from analyses. Instead, model parameters were estimated using all available data. Essentially, themodel assumes that the data that are present for a given subject Compliance was verified by assays of plasma concentrations of reasonably reflect that subject’s deviation from the usual fixed- fluoxetine metabolites (fluoxetine and norfluoxetine) at visits 5 (3 effects regression part of the model (i.e. the regressors multiplied weeks after starting drug) and 9 (end of medication). Assays were by their coefficients). In the present study, the following assess- performed after study completion using gas chromatography (Nash ments of positive and negative affect were missing: 18 at visit 3, 24 at visit 4, 32 at visit 5, 39 at visit 6, 51 at visit 7, 62 at visit 8, 72 atvisit 9. Broken down by group, there were 116 total data pointsmissing from the placebo group, 83 from the 30 mg group, and 99 Axis I disorders were assessed via Structured Clinical Interview forDSM-IV, patient version (SCID; Spitzer et al. 1992). The SCID hasmoderate construct validity, as shown by its favorable comparison with other diagnostic assessment methods (Williams et al. 1992).
Individuals with current axis I disorders other than nicotine dependence were excluded from the sample.
Baseline smoking history and sociodemographic variables All study sites were approved by appropriate ethics committees.
were compared across conditions using one way analyses Participants first attended a screening session where they provided of variance for continuously scaled variables and chi- informed consent and were assessed for axis I disorders via theSCID (Spitzer et al. 1992). Those who were eligible for the study square tests for differences between proportions when began the first of nine sessions of individual cognitive behavioral variables were dichotomous. As shown in Table 1, treatment aimed at achieving smoking cessation by the develop- smokers randomized to the different treatment conditions ment of coping skills and relapse prevention. Subjects were showed no significant differences on age, gender, baseline randomized on a double-blind basis to receive 10 weeks of placebo(n=60), 30 mg (n=57) or 60 mg (n=58) of fluoxetine, which began positive and negative affect, nicotine dependence, number during the second visit. Participants quit smoking 14 days after the of years smoked, cotinine level, and history of depression.
beginning of the medication phase (just prior to visit 4), so that a Baseline PANAS scores are comparable to those reported therapeutic drug level would be achieved before quitting smoking.
in other euthymic samples (Watson et al. 1988).
Assessments of positive and negative affect were collected at eachvisit.
Abstinence was determined via self-report, cotinine, and Mood change across seven time points (visits 3–9) was analyzedusing mixed linear modeling, implemented via SAS PROC carbon monoxide assessments collected at each visit. At MIXED. Random effects regression models used random inter- visit 5, abstinent smokers comprised 36.7% of those of cepts, linear and quadratic trend model with autoregressive errors.
placebo, 35.1% of those on 30 mg and 46.1% of those As recommended, this variance covariance structure for the taking 60 mg. At visit 6, abstinent smokers comprised longitudinal data was selected after comparisons with several otherpotential structures (Verbecke and Molenberghs 2000). Nicotine 38.3% of those on placebo, 35.1% of the 30 mg group, dependence, change in cotinine, and fluoxetine blood levels were and 39.7% of the 60 mg group. At visit 7, abstinent utilized as covariates. Change in cotinine at each visit was analyzed smokers comprised 25% of the placebo group, 26.3% of as a time varying covariate, which statistically removed the the 30 mg group, and 25.9% of the 60 mg group. At visit influence of nicotine exposure on positive and negative affect 8, abstinent smokers comprised 26.7% of those on across time. Both time and time squared terms were included in allanalyses. Non-significant interaction terms were removed from the placebo 19.3% of those on 30 mg, and 29.3% of those and smoking variable meansand standard deviations by on 60 mg. At visit 9, abstinent smokers comprised 21.7%of the placebo group, 14% of the 30 mg group, and 24.9%of the 60 mg group.
Simple change score analysis (visitŸbaseline) measuredchange in positive and negative affect at each visit,consistent with our interest in examining fluoxetine’sinfluence on acute changes in affect after quittingsmoking and evidence that mood change represents aclinically relevant affective disturbance during nicotineabstinence (Piasecki et al. 2000). Visits 1 and 2 wereaveraged together to create baseline positive and negativeaffect because treatment did not require smoking reduc- Fig. 1 Regression function showing estimated positive affect tion at these measurement points and fluoxetine was not change by condition across time, covariate adjusted for change in administered until after visit 2. Prior to creating baseline cotinine, fluoxetine blood levels and nicotine dependence. Dis- scores, it was established that positive and negative affect tances represent real time. Participants quit smoking between visit 3 did not change across visits 1 and 2. Mean positive affect and 4, 14 days after beginning of medication phase at visits 1 and 2 was 32.27 and 32.79, respectively(P=0.48) and mean negative affect at visits 1 and 2 was 15.86 and 16.52, respectively (P=0.11).
A significant group by time interaction was present in thelongitudinal analysis of change in positive affect (see Pearson correlations were computed to identify appropri- Table 2). Specifically, the 60 mg versus placebo com- ate covariates. The association between change in positive and negative affect and the following potential covariates [t(526)=2.50, P=0.01]. Relative to placebo, those on were examined: history of depression, nicotine depen- 60 mg fluoxetine experienced an increase in positive dence (FTQ), number of cigarettes smoked daily, gender, affect that grew across time (see Fig. 1). There were no plasma concentrations of fluoxetine metabolites (com- significant differences between 30 mg and placebo.
posite average of visits 5 and 9), and change in cotinine Parallel analyses using residualized positive affect as an from baseline to each visit. Only nicotine dependence, outcome variable yielded the same results.
fluoxetine blood levels, and change in cotinine weresignificantly correlated with change in positive andnegative affect after cessation (all r<0.05), and were therefore retained as covariates. Associations betweenchange in positive and negative affect at each visit were The negative affect model showed a significant linear also examined. Correlation analysis showed that the association between change in positive and negative P=0.02] interaction with time (see Table 3). Relative to affect was moderate and significant at all time points placebo, those on 60 mg fluoxetine experienced a [r=Ÿ0.21 to Ÿ0.41, P<0.05], although not to the extent that decrease in negative affect. Across time, however, this they appeared to reflect the same construct.
advantage appeared to diminish (see Fig. 2). There were positive affect from visit 3through visit 9, determined by negative affect from visit 3through visit 9, determined by negative affect (linear, P=0.04; quadratic, P=0.04). Re-sults suggest that positive and negative affect changes areinfluenced by fluoxetine rather than feelings of successafter quitting smoking.
The current results are the first to show that fluoxetineimproves post-quit positive and negative mood states. Thefindings accord with growing evidence that antidepres-sants may be used to reduce affective distress followingquitting smoking (Hall et al. 1998; Shiffman et al. 2000;Lerman et al. 2002). Notably, fluoxetine produced areduction in negative affect and a rise in positive affect.
That observation is particularly noteworthy when consid- Fig. 2 Regression function showing estimated negative affect ering that other antidepressants have simply attenuated change by condition across time, covariate adjusted for change in increases in negative affect (nortriptyline, bupropion) and cotinine, fluoxetine blood levels and nicotine dependence. Dis- decreases in positive affect (bupropion) while abstaining tances represent real time. Participants quit smoking between visit 3 from nicotine. The current results showing that fluoxetine and 4, 14 days after beginning of medication phase increased post-cessation positive affect suggest that thedrug may be unique in not only buffering against an no significant differences between 30 mg and placebo.
aversive mood response to nicotine abstinence, but Parallel analyses using residualized negative affect as an actually producing improvement in positive affect fol- outcome variable yielded the same results.
Fluoxetine’s enhancement of positive affect is inter- esting to consider given that human pharmacology studies have focused primarily on SSRI modulation of negativeaffect (Zald and Depue 2001). The dearth of research We ran parallel analyses among continuous abstainers examining SSRI influences on positive affect may versus non-continuous abstainers to examine whether partially result from the view that positive mood states fluoxetine’s beneficial mood effects were secondary to reflect little more than the opposite of negative moods feelings of success associated with quitting smoking.
along a single affective dimension (Russell and Carroll Those identified as continuous abstainers were abstinent 1999). Our data, however, showed that changes in post- at visit 6 and maintained continuous abstinence through quit positive and negative affect were only moderately visit 9. Fluoxetine produced similar trends in positive and intercorrelated. That fluoxetine exerted a more sustained negative affect in both groups. That is, 60 mg fluoxetine influence on positive affect than on negative affect also increased positive affect and decreased negative affect suggests that these two parameters of affective responses after quitting, regardless of smoking status. To examine are at least partially independent. Although those on the impact of smoking cessation on changes in affect 60 mg fluoxetine experienced reductions in negative further, continuous abstinence was statistically controlled affect during the first few weeks after quitting smoking, in both regression models. Even after controlling for improvements in negative affect dissipated by end of abstinence, those taking 60 mg fluoxetine experienced a treatment (see Fig. 2). In contrast, fluoxetine’s influence linear increase in positive affect (P=0.01) and decrease in on positive affect grew stronger across time (see Fig. 1), resulting in assessments of positive mood that exceeded (Clark and Watson 1991; Coyne 1994). In the context of pre-quit levels. The positive affect pattern that emerged smoking cessation, our data show that 60 mg fluoxetine, for the 60 mg group is strikingly different than reported relative to placebo, improved positive and negative affect losses in positive affect in the placebo group and in others during nicotine abstinence. In view of prior findings that abstaining from nicotine (Gilbert et al 1998; Lerman et al.
link mood problems during smoking cessation with 2002; Cook et al. 2003). Fluoxetine appears to have not relapse, it is plausible that beneficial effects of antide- only reversed typical reported losses in positive affect pressants on smoking cessation (Spring et al. 1995; Hurt following smoking, but stimulated a positive mood et al. 1997; Hall et al. 1998; Niaura et al. 2002) may be improvement that continued to grow 8 weeks after modulated via effects on positive and negative affect.
Clinically, the current findings indicate that fluoxetine is Animal models of brain reward function are relevant to helpful in alleviating affective distress triggered by understanding positive mood effects in humans to the quitting smoking, and might therefore prove helpful in extent that positive emotions can be viewed as subjective preventing post-cessation decline into depression (Bor- cues that motivate approach behavior toward rewards relli et al. 1996; Tsoh et al. 2002). An important question (Depue and Collins 1999). Findings from animal studies that remains to be examined in future research is whether indicate that fluoxetine heightens responsivity to rewards, fluoxetine’s effects on post-cessation affect mediate and suggest dopaminergic (DA) and serotonergic (5-HT) mechanisms of action. In animal studies, nicotine absti-nence leads to a decrement in reward functioning that is Acknowledgements This study was supported in part by DA14144 reversed or prevented by fluoxetine (Harrison et al. 2001), to Jessica Werth Cook, VA Merit Review, NIH HL63307and HL59348 to Bonnie Spring, 1 K08 DA00467 to Dennis a finding analogous to fluoxetine’s enhancement of McChargue, NHLBI R01 62165 to Belinda Borrelli, National positive affect in the present study. Fluoxetine’s positive Cancer Institute, Transdisciplinary Tobacco Use Research Center mood enhancing effects might plausibly come about via Grant, P50 CA84719 to Brian Hitsman and Raymond Niaura, 5-HT/DA interaction. Fluoxetine administration directly Forest Laboratories, Pfizer Inc. to Nancy Keuthen, NIH-NCCAM-R21AT00416-01, Metanexus Institute, NIDA, and Fetzer Institute enhances 5-HT activity, which has been shown to Grant to Jean Kristeller. Additional funding provided by the facilitate dopamine release in the nucleus accumbens National Institute on Drug Abuse, and the Robert Wood Johnson (Benloucif and Galloway 1991; De Deuwaerdere et al.
1996), and increase sensitivity to reward (Sasaki-Adamsand Kelley 2001). Conversely, decreased 5-HT releaseinhibits DA activity (Ichikawa et al. 1995), a mechanism that has been linked with acquisition of anhedonia(Harrison et al. 2001; Zagen et al. 2001). Plausibly, American Psychiatric Association (1994) Diagnostic and statistical therefore, reduction in positive affect during nicotine manual of mental disorders (4th edn). APA, Washington D.C.
Benloucif S, Galloway MP (1991) Facilitation of dopamine release deprivation may be engendered by decreased 5-HT in vivo be serotonin agonists: studies with microdialysis. Eur J activity with consequent inhibition of DA release. We posit that a relatively high dose of fluoxetine (60 mg) may Borrelli B, Niaura R, Keuthen NJ, Goldstein MG, DePue JD, be needed to induce sufficient 5-HT activation to Murphy C, Abrams DB (1996) Development of major depres-sive disorder during smoking-cessation treatment. J Clin normalize DA release in the nucleus accumbens, thereby Clark LA, Watson D (1991) Tripartite model of anxiety and Study limitations include the selective nature of the depression: psychometric evidence and taxonomic implica- sample in the respect that participants were motivated to quit smoking and generally in good physical and mental Cohen LS, Miner C, Brown EW, Freeman E., Halbreich U, Sundell K, McCray S (2002) Premenstrual daily fluoxetine for health. Generalizability to populations with greater phys- premenstrual dysphoric disorder: a placebo-controlled, clinical ical and psychological comorbidities cannot be assumed.
trial using computerized diaries. Obstet Gynecol 100:435–444 Nor is it known to what degree study results would Cook JW, Spring B, McChargue D, Hedeker D (2003) Hedonic generalize to populations of smokers who were attempt- capacity, cigarette craving, and positive moods. NicotineTobacco Res (in press) ing to quit without behavioral assistance. Another limi- Coyne JC (1994) Self-reported distress: analog or ersatz depres- tation is that other mood measures were not examined, potentially limiting the generalizability with regard to Dalack GW, Glassman AH, Rivelli S, Covey L, Stetner F (1995) other assessments of affect. A final limitation is that Mood, major depression, and fluoxetine response in cigarettesmokers. Am J Psychiatry 152:398–403 missing data increased over time and differed between Davidson RJ (1992) Anterior asymmetry and the nature of emotion.
groups. We therefore utilized mixed linear modeling, an analytic strategy designed for handling time and group De Deurwaerdere P, Bonhomme N, Lucas G, Le Moal M, differences in missing data. Random effects regression Spampinato U (1996) Serotonin enhances striatal dopamine creates model parameters using all available data, there- outflow in vivo through dopamine uptakes sites. Neurochem-istry 66:210–215 fore estimating rate of change for missing participants.
Depue R, Collins P (1999) Neurobiology and the structure of More generally, this study suggests the utility of SSRIs personality: dopamine, facilitation of incentive motivation, and for treating not only elevated negative affect, but also extraversion. Behav Brain Sci 22:491–569 deficient positive affect, a common feature of depression Fagerstrom KO (1978) Measuring degree of physical dependence to assumptions yield differing outcomes. J Consult Clin Psychol tobacco smoking with reference to individualization of treat- Piasecki TM, Niaura R, Shadel WG, Abrams D, Goldstein M, Fiore Fagerstrom KO, Schneider NG (1989) Measuring nicotine depen- MC, Baker TB (2000) Smoking withdrawal dynamics in dence: a review of the Fagerstrom tolerance questionnaire. J unaided quitters. J Abnorm Psychol 109:74–86 Russell JA, Carroll JM (1999) On the bipolarity of positive and Gilbert DG, McClernon FJ, Rabinovich NE, Plath LC, Jensen RA, Meliska CJ (1998) Effects of smoking abstinence on mood and Sasaki-Adams DM, Kelley AE (2001) Serotonin-dopamine inter- craving in men: influences of negative-affect-related personal- actions in the control of conditioned reinforcement and motor ity traits, habitual nicotine intake and repeated measurements.
behavior. Neuropsychopharmacology 25:440–452 Shiffman S, Johnston JA, Khayrallah M, Elash CA, Gwaltney CJ, Gram L (1994) Fluoxetine. N Engl J Med 331:1354–1361 Paty JA, Gnys M, Evoniuk G, DeVeaugh-Geiss J (2000) The Hall SM, Reus VI, Munoz RF, Sees KL, Humfleet G, Hartz DT, effect of bupropion on nicotine craving and withdrawal.
Frederick S, Triffleman E (1998) Nortriptyline and cognitive- behavioral therapy in the treatment of cigarette smoking. Arch Sonawalla SB, Farabaugh A, Johnson MW, Morray M, Delgado ML, Pingol MG, Rosenbaum JF, Fava M (2002) Fluoxetine Harrison AA, Liem YT, Markou A (2001) Fluoxetine combined treatment of depressed patients with comorbid anxiety disor- with a serotonin-1A receptor agonist reversed reward deficits observed during nicotine and amphetamine withdrawal in rats.
Spitzer RL, Williams JB, Gibbon M, First MB (1992) The structured clinical interview for DSM-III-R (SCID). History, Hitsman B, Pingitore R, Spring B, Mahableshwarkar A, Mizes JS, rationale, and description. Arch Gen Psychiatry 49:624–629 Segraves KA, Kristeller J L, Xu W (1999) Antidepressant Spring B, Wurtman J, Wurtman R, El-Khoury A, Goldberg H, pharmacotherapy helps some smokers more than others. J McDermott J, Pingitore R (1995) Efficacies of dexfenfluramine and fluoxetine on preventing weight gain after smoking Hughes JR, Hatsukami D (1986) Signs and symptoms of nicotine withdrawal. Arch Gen Psychiatry 43:289–294 Tsoh J, Humfleet G, Muæoz R, Reus V, Hartz D, Hall S (2002) Hurt RD, Sachs DP, Glover ED, Offord KP, Johnston JA, Dale LC, Development of major depression after treatment for smoking Khayrallah MA, Schroeder DR, Glover PN, Sullivan C, Croghan IT, Sullivan PM (1997) A comparison of sustained- Verbeke G, Molenberghs G (2000) Linear mixed models for release bupropion and placebo for smoking cessation. N Engl J Watson D, Clark LA, Tellegen A (1988) Development and Ichikawa J, Kuroki T, Kitchen MT, Meltzer HY (1995) R(+)-8-OH- validation of brief measures of positive and negative affect: DPAT, a 5-HT1A receptor agonist, inhibits amphetamine- the PANAS scales. J Person Soc Psychol 54:1063–1070 induced dopamine release in rat striatum and nucleus accum- Williams JBW, Gibbon, M, First MB, Spitzer RL, Davies M, Borus J, Howes MJ, Kane J, Pope HG, Rounsaville B, Wittchen HU Lerman C, Roth D, Kaufmann V, Audrain J, Hawk L, Liu A, (1992) The structured clinical interview for DSM-III-R (SCID).
Niaura R, Epstein L (2002) Mediating mechanisms for the Multisite test-retest reliability. Arch Gen Psychiatry 49:630– impact of bupropion in smoking cessation treatment. Drug Zagen A, Nakash R, Overstreet DH, Yadid G (2001) Association Nash JF, Bopp RJ, Carmichael RH, Farid KZ, and Lemberger L between depressive behavior and absence of serotonin-dopa- (1982) Determination of fluoxetine and norfluoxetine in plasma mine interaction in the nucleus accumbens. Psychopharmacol- by gas chromatography with electron-capture detection. Clin Zald DH, Depue RA (2000) Serotonergic functioning correlates Niaura R, Spring B, Borrelli B, Hedeker D, Goldstein MG, Keuthen with positive and negative affect in psychiatrically healthy N, DePue J, Kristeller J, Ockene J, Prochazka A, Chiles JA, Abrams DB (2002) Multicenter trial of fluoxetine as adjunct tobehavioral smoking cessation treatment: different missing data

Source: http://tigger.uic.edu/labs/hprl/pubs/cook04.pdf