Behavioural cue reactivity to alcohol and non-alcohol-related stimuli among individuals with severe alcohol use disorder: an fMRI study with a visual task

Patients with alcohol use disorder (AUD) have difficulties controlling alcohol cravings and thus exhibit increased use and earlier relapse. Although patients tend to respond more strongly to alcohol-related images compared with non-alcohol-related images, few researchers examined the factors that modulate cravings. Here, we examined whole-brain blood oxygen level-dependent (BOLD) responses to behavioural cues in individuals with AUD and healthy controls (HCs). The participants included 24 patients with severe AUD and 15 HCs. We presented four beverage images (juice, drinking juice, sake, and drinking sake) and compared participant BOLD responses between the two groups. Multiple comparisons revealed that the AUD group had lower BOLD responses compared with the HC group to images of drinking juice in the left precuneus (p = 0.036) and the left posterior cingulate cortex (PCC) (p = 0.044) and higher BOLD responses to images of drinking sake in the left PCC (p = 0.044). Furthermore, compared to the HCs, the AUD patients had decreased BOLD responses associated with cue reactivity to drinking juice in the left precuneus during the period from 15 to 18 s (p = 0.004, df = 37) and 18 to 21 s (p = 0.002, df = 37). Using the Spearman correlation, we found a significant negative correlation between BOLD responses in the left PCC of the AUD patients and Mini–Mental State Examination (MMSE) scores (r = −0.619, p = 0.001). Our findings suggest that HCs and severe AUD patients differ in their responses not to images of alcoholic beverages but those related to alcohol drinking behavior. Thus, these patients appear to have different patterns of brain activity. This information may aid clinicians in developing treatments for patients with AUD.


42
According to the World Health Organization (WHO), 3.3 million deaths every year result 43 from the harmful use of alcohol, which represents 5.9% of all deaths in the world [1]. The harmful 44 use of alcohol is a causal factor in more than 200 diseases and injuries [2]. Overall, 5.1% of the 45 global burden of disease and injury is attributable to alcohol, as measured in disability-adjusted 46 life years [1]. Beyond health problems, the harmful use of alcohol is associated with significant 47 social and economic losses for individuals and society at large. 48 As the American Psychiatric Association (APA) guidelines suggest, not only abstinence from 49 alcohol use but also reduction or moderation of alcohol use may be appropriate initial goals of 50 treatment for alcohol use disorder (AUD) from a harm reduction perspective. In substance use 51 research, however, exposure to substance-related cues (such as the sight or smell of alcoholic 52 beverages) has been found to evoke elevations in subjective craving and physiological arousal 53 and increase the likelihood of substance use [3]. Thus, craving appears to play an important role 54 in predicting negative outcomes such as increased use and earlier relapse in AUD patients [4]. 55 However, some patients are not sensitive to the craving before their relapse. Furthermore, the 56 mechanisms underlying craving remain unclear. 57 Previous studies investigating craving have suggested that the posterior cingulate cortex 58 (PCC) plays a role in craving and relapse to alcohol use [5,6]. The PCC is a primary node of the 59 default mode network (DMN) [7,8], which might be related to deficits in brain function in AUD 60 patients. The PCC codes relevant information from visual sensory systems to evaluate emotional 4 61 content [9] and is involved in internally directed cognition, such as memory retrieval and planning 62 [10]. This region plays a crucial role in integrating incoming episodic memory with existing 63 knowledge to create a coherent representation of the event [11]. While the PCC is a well-studied 64 region with AUD patients in task-based studies on craving, and AUD patients have been found to 65 have greater responsiveness to alcohol-related images than non-alcohol beverage images [12], few 66 studies have examined the factors that contribute to craving in AUD patients.

67
Given the findings of several reports using human brain imaging, we hypothesized that the 68 presentation of stimuli involving alcohol drinking behavior may be accompanied by stronger brain 69 activation in AUD patients compared with stimuli that simply contain alcoholic beverages. To 70 address the question of whether and how the PCC and other brain regions related to the DMN are 71 implicated in the deficits, we examined possible alterations in brain function in patients with AUD 72 using functional MRI (fMRI) with beverage image cues. We anticipated that such alterations might 73 be correlated with changes in cognitive function in the patient group.    The methods were based on a previous study [16]. In the informed consent process, we 113 explained to all participants that the tasks included the presentation of images, but we did not 7 114 describe the content of the images. The participants completed eye tests, and the participants with 115 weak sight were given glasses to wear during the MRI.

116
The participants were asked to lie in a supine position on a bed inside the MRI scanner  severe AUD in Japan often drink this kind of alcohol because it has a high alcohol content and is 128 cheap, portable, and convenient to drink; 19 of the 24 AUD participants in this study stated that 129 sake was their preferred alcoholic beverage (Fig 1),. Means, SDs, and frequencies for clinical and demographic variables are presented in Table   214 1. We found no significant group differences in the demographic data with respect to age, sex, and

Main effects of group and image and stimulus-by-group interactions 221
As shown in Table 2   diagram. There was a group difference between the distribution trends.

259
In the left PCC, the median of the BOLD responses to Image B in the AUD participants 260 was lower than that in HCs, but the median of the BOLD responses to Image D in the AUD 261 participants was higher than that in the HCs. In the left precuneus, however, this inverse 262 relationship was absent.

264
The time course of BOLD activation of behavioural cue reactivity 265 for drinking juice vs. drinking sake 266 We found significant group differences in the time course of BOLD responses with a linear precuneus, Image B in the left PCC, or Image D in the left PCC (Fig 3a-d).  (Fig 4). In the AUD participants, we found no statistically significant between-group differences 297 in the correlations between brain activation elicited by beverage images and demographic 298 measurements (age, sex, handedness, and education) or clinical measurements (duration of illness, 299 number of hospital admissions, past history of neurologic symptoms, preferred alcoholic beverage, 300 and dose of antipsychotics, acamprosate, and antidepressant drugs). In the HC participants, we 301 also found no statistically significant between-group differences in the correlations between brain 302 activation and demographic measurements.

305
We collected fMRI data for admitted AUD patients and HCs while they viewed substantial 306 and behavioural visual cues of alcohol and non-alcohol beverages. To the best of our knowledge, 307 this is the first study to demonstrate biological differences in brain function associated with strong 308 visual behavioural cues regarding alcohol and non-alcohol beverages in AUD patients. patients.

343
In interpreting the current study, it is important to consider several possible limitations. 344 First, as most of the patients we recruited were severe AUD patients (23 severe and one moderate 345 according to the DSM-5 criteria), selection bias should be considered. Mild and moderate AUD 346 patients should be examined using the same approach to determine whether the BOLD activity 347 that we observed was a state-dependent endophenotype. Second, given that the sample size of 348 participants was small, the statistical power for assessing brain activity was limited. Future studies 349 with larger numbers of participants are needed to determine whether brain BOLD activity can Our study showed that, compared with controls, patients with severe AUD exhibited signs 361 of functional brain impairment. Taking a harm reduction approach, the APA and domestic 362 guidelines in Japan suggest that not only abstinence from alcohol use but also reduction or 363 moderation of alcohol use may be appropriate initial goals for AUD treatment. Clinically, 364 individuals with severe AUD generally have great difficulty moderating alcohol use. This may be 365 related to the neural differences indicated by our study. Even clinicians who are able to form strong 366 relationships with patients may face challenges in helping those who want to drink alcohol in 367 moderation. Our results may help to explain the challenges encountered in such moderation efforts.