Neural activity in response to smoking-related and neutral pictures in three different states in smokers compared to non-smoking controls

Background: Tobacco smoking is the leading preventable cause of morbidity and premature death worldwide and despite many smokers wanting to quit, relapse rates are high. Cue-induced craving plays a prominent role in relapse of tobacco use disorder (TUD). Therefore, elucidating the underlying neural correlates may shed light on relapse mechanisms. Here, we investigated the effect of smoking state on cue-induced neural activity measured by BOLD fMRI, implementing a block design with smoking-related and neutral images, and made comparisons with non-smoking controls.

Methods: Fifty-one male subjects with TUD underwent BOLD fMRI scans in three different smoking states: baseline (smoking as usual), withdrawal (24 hours smoking abstinence), and satiation (directly after smoking). As a control, fifty matched non-smokers underwent BOLD fMRI scans once. Smoking-related and neutral pictures were presented in a block design. Craving, valence and arousal were rated on visual analog scales before and after the task and withdrawal symptoms were assessed during every state. TUD severity was assessed with the Fagerström Test for?Nicotine Dependence. Preprocessing was performed with SPM. In the first-level analysis, a random-effect general linear model with?task-related regressors was implemented to compare the BOLD signal between smoking-related and neutral?images. Second-level random effect analysis was used to assess?activation across subjects and states with t-tests using family-wise error correction for multiple comparisons.?Repeated measures ANOVAs were performed to evaluate correlations between cue-induced neural activity and self-reported withdrawal?symptoms, TUD severity and craving before and after?the task.

Results: In smokers, the BOLD response to smoking compared to neutral images?was greater in the right occipital gyrus and left prefrontal cortex during baseline. During withdrawal, greater activation was found bilaterally in the occipital gyrus, left fusiform gyrus and right cuneus, prefrontal cortex and temporal gyrus. Compared to controls, smokers showed higher activity in the right frontal gyrus during satiation. Furthermore, TUD severity, withdrawal symptoms and craving at both time points at baseline were positively correlated with activity in the right occipital gyrus and left prefrontal cortex. Concerning visual analog scales, there was a significant main effect of time on craving and arousal in smokers, with higher ratings after the task. A significant main effect of smoking state on craving, valence and arousal was found, with higher values during?withdrawal compared to baseline and satiation. Controls rated lower for arousal and higher for valence than smokers, while craving was not present.

Conclusion: Compared to neutral?pictures, smoking pictures elicited higher BOLD responses in?visual processing areas during baseline and withdrawal, with prominent recruitment of the extended visual system during the latter. Additionally, frontal brain areas involved in inhibitory control were activated. Interestingly, this response was not seen in the satiated state, however, premotor areas were found to elicit higher responses in this state compared to controls. TUD severity, stronger craving and withdrawal symptoms were associated with higher activity in the same visual processing and inhibitory control areas. These results imply that the stronger a smoker’s addiction and craving is, the stronger the responses induced by smoking cues are.