MDMA (ecstasy) use is associated with reduced BOLD signal change during semantic recognition in abstinent human polydrug users: a preliminary fMRI study

Abstract

3,4-methylenedioxymethamphetamine (MDMA) users have impaired verbal memory, and voxel-based morphometry has shown decreased grey matter in Brodmann area (BA) 18, 21 and 45. Because these regions play a role in verbal memory, we hypothesized that MDMA users would show altered brain activation in these areas during performance of a functional magnetic resonance imaging (fMRI) task that probed semantic verbal memory. Polysubstance users enriched for MDMA exposure participated in a semantic memory encoding and recognition fMRI task that activated left BA 9, 18, 21/22 and 45. Primary outcomes were percent blood oxygen level-dependent signal change in left BA 9, 18, 21/22 and 45, accuracy and response time. During semantic recognition, lifetime MDMA use was associated with decreased activation in left BA 9, 18 and 21/22 but not 45. This was partly influenced by contributions from cannabis and cocaine use. MDMA exposure was not associated with accuracy or response time during the semantic recognition task. During semantic recognition, MDMA exposure was associated with reduced regional brain activation in regions mediating verbal memory. These findings partially overlap with previous structural evidence for reduced grey matter in MDMA users and may, in part, explain the consistent verbal memory impairments observed in other studies of MDMA users.

Figure 1. Semantic Task

Subjects encoded a block of 5 words or pseudo-words followed by a recall task using novel homophones as distracter images. Subjects were instructed to press a button corresponding to right or left screen location of previously encoded items. Semantic encode: Activation during semantic encoding was defined as (word encode minus pseudo-word encode). Semantic recognition: Activation during semantic recognition was defined as (word recognition minus pseudo-word recognition).

Figure 2. Activation to Word Encode and Word Recognition

Task-induced regional brain activation maps for word encode and word recognition. Maps were thresholded at a false discovery rate (FDR) of < 0.05 (see methods for details). Rows show (top to bottom) regional brain activation to Brodmann Area (BA) 9, 18, 21/22 and 45. White arrowheads point to BA in each brain view. For each condition (word encode or word recognition) activation maps are overlayed on a group average structural brain in (left to right) coronal, sagittal, and axial sections. Right side of figure is left side of brain for coronal and axial sections. Right side is posterior for sagittal sections. Scale bar indicates t values ranging from −8.00 to plus 8.00. Warm colors are positive t, indicating increased activation with task activity. Cool colors are negative t, indicating decreased activation with task activity. All regions of interest used for analysis showed increased task-induced activation.

Figure 3. Regional Semantic percent BOLD signal change

Bar chart indicating regional semantic percent BOLD signal change as the percent signal change in a region of interest for word encode or recognition minus pseudo-word encode or recognition.