Alcohol sensitizes cerebral responses to the odors of alcoholic drinks: an fMRI study

Abstract

Background: Small, priming doses of alcohol enhance desire to drink, and thus play a role in the loss of control of alcohol consumption. Using functional magnetic resonance imaging (fMRI), we previously showed that alcoholic drink odors (AO; subjects' drinks of choice) induce greater nucleus accumbens (NAc) activity than non-appetitive odors (NApO; grass, leather) in subjects at risk for alcoholism. Here we hypothesized that priming exposure to alcohol would enhance responses to AO in the NAc and orbitofrontal cortex in comparison to NApO (grass, leather) and to the appetitive control odors (ApCO) of chocolate and grape.

Methods: Ten hazardous drinkers (mean age = 22.7; SD = 2.9, average drinks per drinking day = 5.9, SD = 2.3; drinking days/90 days = 50.4, SD = 13.7) were scanned on a 1.5 T GE Signa MR scanner during intravenous infusion of lactated Ringer's or 6% ethanol in lactated Ringer's that was pharmacokinetically modeled to achieve a constant breath alcohol concentration (BrAC) of 50 mg% throughout imaging. During scanning, subjects sniffed AO, NApO, and ApCO.

Results: Alcohol infusion enhanced the contrast between AO and NApO in the NAc, and in orbitofrontal, medial frontal, and precuneus/posterior cingulate regions. The contrast between AO and appetitive control odors (ApCO; chocolate and grape) was similarly larger in the orbital, medial frontal, precuneus, and posterior cingulate/retrosplenial areas, with the most robust finding being a potentiated response in the posterior cingulate/retrosplenial area. The orbital region is similar to an area previously shown to manifest satiety-related decreases in activity induced by food cues.

Conclusions: The results suggest that priming exposure to alcohol renders a limbic network more responsive to alcohol cues, potentially enhancing desire to drink.

Fig. 1

Odor stimulation paradigm. Subjects sniffed during 2-second periods of odor stimulation (light gray tracing) across a 36-second epoch (high phase of black dashed tracing) during which the alcoholic odors (representing the subject’s two most-preferred drinks), non-appetitive odors (grass/leather) or appetitive control odors (grape/chocolate) were presented. These periods were contrasted against sniffing during odorless baseline periods (low phase of the black dashed tracing). Each 36-second epoch (odor or control) consisted of four valve (odor or sham) openings (light gray tracing) following auditory commands (gray insert).

Fig. 2

Ratings of perceived “High” and “Intoxication” under alcohol and placebo infusions (left scale). Dashed reference line represents the modeled BrAC clamp (right scale). Time 1 = Baseline (immediately before the infusion pump start). Arrow = start of scanning.

Fig. 3

(a) Trend-level activation (display threshold, p < 0.05) in which activity from AO was greater than activity from NApO in ventromedial frontal cortex and posterior cingulate. (b) Greater AO than NApO activation during alcohol infusion in right nucleus accumbens/olfactory tubercle area, ventromedial frontal cortex, posterior cingulate/retrosplenial cortex, and precuneus area (display threshold, p < 0.005). (c) Greater AO activation compared to ApCO (grape and chocolate) during alcohol infusion in ventromedial frontal cortex and posterior cingulate/retrosplenial areas (display threshold, p < 0.005).

Fig. 4

Orbitofrontal cortex effects as a result of AO, relative to NApO and ApCO, under placebo and alcohol infusion (display threshold, p < 0.05).

Fig. 5

(a) Region of interest (ROI) analysis showing ROI volume for nucleus accumbens in stark white (y = 8 mm) and the average effect in this ROI for the contrasts, (b) AO compared to NApO; (c) AO compared to ApCO.

Fig. 6

Greater activation during alcohol infusion, as compared to placebo infusion, in the n = 7 paired analysis (display threshold, p < 0.005). Significant effects for the comparisons: (a) [Alcohol odors > Non-Appetitive odors] (AO > NApO) in posterior cingulate and precuneus; (b) [Alcohol odors > Appetitive Control odors] (AO > ApCO) in posterior cingulate.

Fig. 7

Plot of the [AO > ApCO] effect under alcohol as a function of number of heavy drinking days within a 90-day period (mean-centered across the sample) in the retrosplenial cortex peak maximum [−8, −34, 28]. Blue arrow indicates the specific retrosplenial cluster of voxels from which the local maximum was extracted and plotted (display threshold, p < 0.005).