Brain gray matter deficits at 33-year follow-up in adults with attention-deficit/hyperactivity disorder established in childhood

Abstract

Context: Volumetric studies have reported relatively decreased cortical thickness and gray matter volumes in adults with attention-deficit/hyperactivity disorder (ADHD) whose childhood status was retrospectively recalled. We present, to our knowledge, the first prospective study combining cortical thickness and voxel-based morphometry in adults diagnosed as having ADHD in childhood.

Objectives: To test whether adults with combined-type childhood ADHD exhibit cortical thinning and decreased gray matter in regions hypothesized to be related to ADHD and to test whether anatomic differences are associated with a current ADHD diagnosis, including persistent vs remitting ADHD.

Design: Cross-sectional analysis embedded in a 33-year prospective follow-up at a mean age of 41.2 years.

Setting: Research outpatient center.

Participants: We recruited probands with ADHD from a cohort of 207 white boys aged 6 to 12 years. Male comparison participants (n = 178) were free of ADHD in childhood. We obtained magnetic resonance images in 59 probands and 80 comparison participants (28.5% and 44.9% of the original samples, respectively).

Main outcome measures: Whole-brain voxel-based morphometry and vertexwise cortical thickness analyses.

Results: The cortex was significantly thinner in ADHD probands than in comparison participants in the dorsal attentional network and limbic areas (false discovery rate < 0.05, corrected). In addition, gray matter was significantly decreased in probands in the right caudate, right thalamus, and bilateral cerebellar hemispheres. Probands with persistent ADHD (n = 17) did not differ significantly from those with remitting ADHD (n = 26) (false discovery rate < 0.05). At uncorrected P < .05, individuals with remitting ADHD had thicker cortex relative to those with persistent ADHD in the medial occipital cortex, insula, parahippocampus, and prefrontal regions.

Conclusions: Anatomic gray matter reductions are observable in adults with childhood ADHD, regardless of the current diagnosis. The most affected regions underpin top-down control of attention and regulation of emotion and motivation. Exploratory analyses suggest that diagnostic remission may result from compensatory maturation of prefrontal, cerebellar, and thalamic circuitry.

Figure 1

(A) t-map of the significant cortical thinning in probands with ADHD (n=59) compared to comparisons (n=80). (B) t-map of the significant cortical thinning in probands with persistent ADHD (n=17) compared to non-ADHD comparisons (n=57). False Discovery Rate (FDR) threshold depends on the data and is different for the right and left hemispheres. Here the t-statistics at the lowest FDR threshold are projected across each hemisphere for each comparison.

Figure 2

Exploratory uncorrected analyses (p<0.05) reveal regions in which remitted probands (n=27) exhibit thicker cortex than probands with persistent ADHD (n=17). See Author e-Table 7 for peaks and coordinates of clusters.

Figure 3

Comparisons (N=80) exhibit greater gray matter density (left) and cortical thickness (right) in the bilateral dorsal attentional network than probands (n=59) with childhood combined type ADHD. Images are in radiological convention, right is left and left is right.

Figure 4

Voxel-based morphometry reveals that comparisons (N=80) exhibit significantly greater gray matter density (FDR <0.05) in right ventral caudate, right thalamus, bilateral cerebellum than probands (n=59) with childhood combined type ADHD. Images are in radiological convention, right is left and left is right.