Brain structural magnetic resonance imaging predictors of brief intervention response in individuals with alcohol use disorder

Abstract

Aims: Magnetic resonance imaging (MRI) studies have identified brain structural predictors of treatment response in individuals with alcohol use disorder (AUD) but with varying findings and primarily in male veterans. The present study investigated cortical surface area and thickness (CT) as predictors of brief intervention response in community-based adults with AUD.

Methods: Sixty-five non-treatment-seeking adults with AUD (44.6% male, aged 33.2 ± 1.3 years) underwent an MRI and received a brief intervention comprising personalized feedback and motivational interviewing, with follow-up ~6-8 weeks later to quantify changes in drinks/week (DPW), the primary outcome. Eighteen bilateral a priori regions of interest (ROIs) were used to predict DPW at follow-up, adjusting for baseline drinking. Significant predictors were examined with secondary outcomes, percent drinking and heavy drinking days, and in relation to out-of-scanner measures of impulsivity and comorbidities.

Results: Participants exhibited significant decreases in alcohol consumption in response to the brief intervention. Eight bilateral CT ROIs in the frontal, temporal, and occipital lobes, most notably medial orbitofrontal, middle temporal, and lateral occipital gyri, predicted DPW; however, only three predicted the secondary outcomes. Significant associations were observed between CT in frontal and occipital regions and impulsivity (delay discounting, lack of premeditation), executive functioning, anxiety, and stress.

Conclusions: Thinner frontal, temporal, and occipital ROIs predicted poorer brief intervention response, with notable overlap with brain regions previously implicated in AUD. Clarifying whether these regions reflect premorbid or acquired differences and, if the latter, the potential for recovery of cortical gray matter following drinking reductions are future priorities.

Keywords: MRI; alcohol; alcohol use disorder; brief intervention; neural biomarkers; neural correlates.

Figure 1

Line graphs demonstrating the changes in (a) drinks per week, (b) percent of drinking days, and (c), percent of heavy drinking days. Each diamond denotes the mean and standard error. * P < .05, ** P < .01, *** P < .005, **** P < .001.

Figure 2

Significant bilateral a priori cortical thickness regions of interest in relation to changes in drinks per week at follow-up following the intervention. Regions that were implicated in secondary outcome analyses are indicated by ^* for percent of drinking days and ^** for percent of drinking and heavy drinking days.