مدارک و شواهد برای واکنش پذیری نشانه بیشتر در میان افراد سیگاری با وابستگی پایین در مقابل افراد سیگاری با وابستگی بالا

Abstract Introduction Cue reactivity paradigms are well-established laboratory procedures used to examine subjective craving in response to substance-related cues. For smokers, the relationship between nicotine dependence and cue reactivity has not been clearly established. The main aim of the present study was to further examine this relationship. Methods Participants (N = 90) were between the ages 18–40 and smoked ≥10 cigarettes per day. Average nicotine dependence (Fagerström Test for Nicotine Dependence; FTND) at baseline was 4.9 (SD = 2.1). Participants completed four cue reactivity sessions consisting of two in vivo cues (smoking and neutral) and two affective imagery cues (stressful and relaxed), all counterbalanced. Craving in response to cues was assessed following each cue exposure using the Questionnaire of Smoking Urges-Brief (QSU-B). Differential cue reactivity was operationally defined as the difference in QSU scores between the smoking and neutral cues, and between the stressful and relaxed cues. Results Nicotine dependence was significantly and negatively associated with differential cue reactivity scores in regard to hedonic craving (QSU factor 1) for both in vivo and imagery cues, such that those who had low FTND scores demonstrated greater differential cue reactivity than those with higher FTND scores (β = −.082; p = .037; β = −.101; p = .023, respectively). Similar trends were found for the Total QSU and for negative reinforcement craving (QSU factor 2), but did not reach statistical significance. Discussion Under partially sated conditions, less dependent smokers may be more differentially cue reactive to smoking cues as compared to heavily dependent smokers. These findings offer methodological and interpretative implications for cue reactivity studies.

1. Introduction Cue reactivity paradigms are well-established and specific laboratory procedures to examine craving in response to drug-paired cues. Cue-specific craving is most commonly measured with various self-report indices of craving (Carter and Tiffany, 1999 and Ferguson and Shiffman, 2009) and is commonly viewed as a form of stimulus control (i.e., the ability of environmental cues to elicit craving) that develops following repeated pairings between drug administration and specific environmental and/or affective stimuli. Cue reactivity methods are valid methods to experimentally test the likelihood of relapse and treatment outcome (Donny et al., 2008, Ferguson and Shiffman, 2009, Payne et al., 2006, Swan et al., 1996 and Waters et al., 2004). Also, to the extent that treatments are designed with the purpose of diminishing craving, cue reactivity methodology could serve as an early method to test the potential efficacy of treatments prior to large clinical trials (Davies et al., 2000 and Waters et al., 2004). Some investigators have proposed that cue reactivity could be used as treatment itself, (i.e., through cue exposure/extinction), though the therapeutic significance of this approach as a stand-alone intervention has been questioned elsewhere (Brandon et al., 1995, Conklin and Tiffany, 2002a and Conklin and Tiffany, 2002b). Numerous studies have examined factors that may influence cue reactivity among smokers, including perceived drug availability (Wertz and Sayette, 2001), affect (Taylor, Harris, Singleton, Moolchan and Heishman, 2000), level of nicotine deprivation (Geier et al., 2000 and Payne et al., 1996), gender (Niaura et al., 1998, Saladin et al, under review and Waters et al., 2004), and manipulation of cues (Conklin and Tiffany, 2001). Another important factor that has been examined is the relationship between nicotine dependence and cue reactivity (Davies et al., 2000, Donny et al., 2008, Knott et al., 2008, McClernon et al., 2008, Payne et al., 1996, Shadel et al., 2000, Shiffman and Paty, 2006 and Smolka et al., 2006). However, results from these studies are equivocal and no clear relationship exists. On one hand, heavily dependent smokers could be more cue reactive than minimally dependent smokers, since greater nicotine/cigarette exposure in the former group should lead to greater neuroadaptations in brain reward systems that would, in turn, augment sensitivity to smoking-related cues (Robinson and Berridge, 1993). Indeed, evidence for this relationship exists. Two studies of treatment (Payne et al., 1996) and non-treatment seeking (Donny et al., 2008) smokers have shown a positive correlation between dependence and craving in response to smoking-related cues, suggesting that heavier smokers are more cue reactive. Corroborating evidence also comes from two imaging studies that demonstrated increased responding to cues among smokers with greater levels of dependence (McClernon et al., 2008 and Smolka et al., 2006), though one of these studies also found a negative correlation between dependence and fMRI reactivity in other brain areas (McClernon et al., 2008). Alternatively, other models of addiction (Stewart, de Wit and Eikelboom, 1984) allow, at least under some conditions, that nicotine dependence would be inversely associated with cue reactivity. For example, low-dependent smokers smoke less frequently and often within a relatively narrow range of stimuli, whereas heavily dependent smokers smoke more frequently and irrespective of specific environmental cues. Thus, for heavily dependent smokers, few stimuli become unique predictors of nicotine administration. Support for this notion comes from literature on “chippers,” i.e., people who smoke no more than five cigarettes per day on at least four days per week (Shiffman, Paty, Kassel, Gnys and Zettler-Segal, 1994). Recent research by Shiffman and Paty (2006) suggest that “chippers”, are under significantly greater stimulus control than are heavy smokers. These researchers were able to correctly predict smoking (yes or no) on the basis of distinct situational stimuli more so among chippers (83% of the time) than heavy smokers (65%). Though chippers represent a distinct group of smokers towards an extreme end on the continuum of regular smoking, an inverse relationship between stimulus control and level of dependence may still hold among more frequent smokers. For example, Hogarth, Mogg, Bradley, Duka and Dickinson (2003) demonstrated that light daily smokers (people who smoke fewer than 20 cigarettes per day) have a higher attentional bias to cigarette cues than do heavy smokers, again suggesting that it is possible that lower dependent individuals are under greater stimulus control than their high-dependent counterparts ( Hogarth et al., 2003). Finally, indirect data from our own lab suggest that cue reactivity procedures could be most sensitive among smokers low in dependence and thus under greater stimulus control ( Carpenter et al., 2009). The purpose of the present study was to further examine the relationship between nicotine dependence and cue-elicited craving. With few exceptions (Davies et al., 2000), previous literature in this area has largely ignored the possibility that craving is multidimensional (Shadel, Niaura, Brown, Hutchison and Abrams, 2001), and is frequently thought to include both hedonic craving (i.e., anticipation of positive outcomes) and craving as a function of negative reinforcement (i.e., anticipation of withdrawal relief) (Davies et al., 2000, King and Epstein, 2005 and Tiffany and Drobes, 1991). Given the possibility that low-dependent smokers often do not experience withdrawal (Shiffman et al., 1994 and Shiffman et al., 1994) but rather smoke under tightly bound and usually positively-valenced stimuli, it follows that the conditioned response for low-dependent smokers would likely be limited to hedonic craving only. We specifically examined whether nicotine dependence and cue reactivity are inversely related, and whether this relationship is specific to hedonic craving, withdrawal craving, or both. Data from this report derive from a larger study examining gender and menstrual cycle phase effects on craving and cue reactivity, tested among non-treatment seeking smokers.

3. Results The participants (N = 90; 53 males, 74 white) had a mean age of 30 (SD = 6), had been smoking for an average of 11.3 (5.9) years, and had an average of 3.4 (5.7) prior quit attempts. Participants reported smoking an average of 19.3 (7.9) cigarettes per day, with a mean FTND score of 4.9 (2.1). Absolute and differential QSU data can be found in Table 1. In an effort to describe cue-elicited craving across the range of dependence, a median split was conducted to differentiate between low- (FTND <5; n = 37) and high-dependent (FTND ≥5; n = 53) smokers. Low-dependent smokers smoked an average of 14.9 (4.1) cigarettes per day, vs. 22.4 (8.4) among high-dependent smokers. All smokers, including those both high and low in dependence, reported significantly greater craving (total and within each factor) in response to both the active in vivo and imagery cues (p < .001) as compared to their respective control cues, thus demonstrating that our procedures effectively elicited craving. Table 1. Absolute and differential cue reactivity responding by level of dependence (mean (SD)). In vivo cues Imagery cues Neutral Smoking Differencea Relaxed Stressful Differencea Overall (N = 90) Total 3.44 (1.48) 3.98 (1.51) 0.54 3.23 (1.91) 3.85 (1.49) 0.62 QSU F1 4.56 (1.81) 5.19 (1.72) 0.63 4.26 (1.91) 4.99 (1.69) 0.73 QSU F2 2.32 (1.45) 2.76 (1.62) 0.44 2.19 (1.46) 2.71 (1.60) 0.52 Low FTND (n = 37) Total 2.93 (1.17) 3.61 (1.34) 0.68 2.68 (1.22) 3.43 (1.29) 0.75 QSU F1 4.03 (1.61) 4.91 (1.67) 0.88 3.65 (1.72) 4.58 (1.59) 0.93 QSU F2 1.83 (0.99) 2.31 (1.31) 0.48 1.71 (1.00) 2.27 (1.33) 0.56 High FTND (n = 53) Total 3.81 (1.56) 4.25 (1.58) 0.44 3.62 (1.59) 4.16 (1.54) 0.54 QSU F1 4.94 (1.85) 5.41 (1.72) 0.47 4.69 (1.93) 5.28 (1.17) 0.59 QSU F2 2.67 (1.61) 3.09 (1.75) 0.42 2.55 (1.63) 3.03 (1.71) 0.48 a Note: p < .001 for all comparisons between in vivo cues (smoking vs. neutral) and imagery cues (stressful vs. relaxed). Table options For all cues, absolute craving (craving in response to each cue alone) was higher among smokers with greater nicotine dependence (see Table 1). However, differential, cue-elicited craving (i.e., the difference between active and control cues) was higher for all cues among smokers with lower nicotine dependence. The relationship between overall differential cue-elicited craving (Total QSU) and FTND was non-significant, for both the in vivo cue (β = −.041; p = .253) and the stressful imagery cue (β = −.059; p = .153). In support of our hypotheses, FTND was significantly and negatively associated with differential hedonic craving (QSU factor 1) for both in vivo (β = −.082; p = .037; see Fig. 1) and imagery cues (β = −.101; p = .023; see Fig. 2). There was no relationship between dependence and negative reinforcement craving (QSU factor 2) for any cue. We considered the possibility of a ceiling effect among high-dependent smokers; i.e., that high-dependent smokers could not significantly respond to cues because their ambient craving (i.e., in response to neutral/relaxed cues) was high. This does not appear to be the case. As Table 1 shows, all smokers, including those both high and low in nicotine dependence, demonstrated heightened cue-elicited craving in response to each active cue, as compared to each inactive cue. FTND by differential cue reactivity of QSU F1 elicited by in vivo cues. Fig. 1. FTND by differential cue reactivity of QSU F1 elicited by in vivo cues. Figure options FTND by differential cue reactivity of QSU F1 elicited by imagery cues. Fig. 2. FTND by differential cue reactivity of QSU F1 elicited by imagery cues.