Poster #104 - Shared and distinct brain structure and behavioral correlates of ADHD and overweight/obesity

• In Event: 3-206 - Poster Session 13
  In Poster Session: PS 13 Section - Health, Growth, Injury

Integrative Statement

Attention-deficit/hyperactivity disorder (ADHD) and overweight/obesity are prevalent in pediatric populations. Cross-sectional and longitudinal studies support inter-relationships between these conditions. However, the biobehavioral basis of this observed overlap is not well understood. We aimed to investigate both neural and behavioral correlates of ADHD and overweight/obesity in a large sample of school-aged children.

Participants included 207 children between the ages of 8-12 years with ADHD (n=107) and typically developing (TD) controls (n=100). Within each diagnostic group, participants were classified as lean weight (Lean; BMI <50th percentile; 77 ADHD, 53 TD) or overweight (OW; >=85th percentile; 30 ADHD, 47 TD). For 181 of these participants (98 ADHD, 93 TD), structural MRI was obtained and volume of prefrontal and subcortical regions of interests involved in reward and self-regulation (ROIs: caudate, putamen, dorsolateral prefrontal cortex [dlPFC], orbitofrontal cortex [OFC], and anterior cingulate cortex [ACC]) were compared. Participants also completed two delay discounting tasks assessing preference for smaller immediate rewards over larger delayed rewards (money and game time), and a go/no-go task assessing response control (response variability and response inhibition).

Results of a 2 Diagnosis x 2 Weight Group x 2 Sex analysis of covariance (total cerebral volume as covariate) revealed a main effect of diagnosis for the ACC (p<.001; Lean TD=OW TD > Lean/OW ADHD) and a main effect of weight group for the OFC (p=.027; Lean TD > OW TD = Lean/OW ADHD). Further, a Diagnosis x Weight Group interaction was observed for the dlPFC (p=.043; Lean TD > OW TD = Lean/OW ADHD; see Figure 1). There was no evidence of diagnosis or weight effects on subcortical volumes and these results hold with age and ADHD medication as additional covariates.
Behavioral analyses indicated greater monetary delay discounting among overweight children, regardless of diagnosis (main effect of weight group, p=.005; see Figure 2), whereas children with ADHD showed greater delay discounting of game time (main effect of diagnosis, p=.020). Higher BMI z score was associated with greater monetary delay discounting across the whole sample (r=.194, p=.001). In the go/no-go task, greater response variability was observed in children with ADHD (main effect of diagnosis: p<.001), with no weight group effects.

Our results support both shared and distinct neurobehavioral correlates of ADHD and overweight/obesity in children. Whereas obesity/overweight was associated with reduced volumes in the OFC and dlPFC, this impact was only observed in TD children and not in ADHD children. ADHD itself was associated with reduced ACC volume, regardless of weight. This suggests that while obesity/OW is associated with structural brain differences in the general population, that it has little impact in the context of ADHD. Further, the differential sensitivity of the delay discounting tasks to weight group and ADHD suggests that obesity/overweight may be associated with heightened delay discounting of hypothetical and more abstract rewards whereas ADHD is associated with heightened discounting of more immediate and tangible rewards. Together these results support alterations in reward and response control function and underlying circuitry in both heavier children and those with ADHD.

Authors

• Keri Shiels Rosch, Kennedy Krieger Institute, Johns Hopkins University
  Presenting Author

• Afroditi Papantoni, Johns Hopkins School of Medicine
  Non-Presenting Author

• Stewart H Mostofsky, Kennedy Krieger Institute
  Non-Presenting Author

• Karen E. Seymour, Kennedy Krieger Institute
  Non-Presenting Author

• Susan Carnell, Johns Hopkins School of Medicine
  Non-Presenting Author