ADHD comorbidity can matter when assessing cortical thickness abnormalities in patients with bipolar disorder

Abstract

Objectives: Attention-deficit hyperactivity disorder (ADHD) is prevalent in patients with bipolar disorder (BP), but very few studies consider this when interpreting magnetic resonance imaging findings. No studies, to our knowledge, have screened for or controlled for the presence of ADHD when examining cortical thickness in patients with BP. We used a 2 × 2 design to evaluate the joint effects of BP and ADHD on cortical thickness and uncover the importance of ADHD comorbidity in BP subjects.

Methods: The study included 85 subjects: 31 healthy controls, 17 BP-only, 19 ADHD-only, and 18 BP/ADHD. All patients with BP were subtype I, euthymic, and not taking lithium. Groups did not differ significantly in age or gender distribution. We used cortical thickness measuring tools combined with cortical pattern matching methods to align sulcal/gyral anatomy across participants. Significance maps were used to check for both main effects of BP and ADHD and their interaction. Post-hoc comparisons assessed how the effects of BP on cortical thickness varied as a function of the presence or absence of ADHD.

Results: Interactions of BP and ADHD diagnoses were found in the left subgenual cingulate and right orbitofrontal cortex, demonstrating that the effect of BP on cortical thickness depends on ADHD status.

Conclusions: Some brain abnormalities attributed to BP may result from the presence of ADHD. Diagnostic interactions were found in regions previously implicated in the pathophysiology of BP, making it vital to control for an ADHD comorbid diagnosis when attempting to isolate neural or genetic abnormalities specific to BP.

Fig. 1

Regions of interaction between bipolar disorder (BP) and attention-deficit hyperactivity disorder (ADHD) diagnoses. BA = Brodmann's area; CON = controls. *Indicates significant differences (p < 0.05) in cortical thickness between groups. **Indicates trend level differences (p < 0.10) in cortical thickness between groups.

Fig. 2

Significance maps of cortical thickness associated with main effects and subgroup comparisons. ^aThe color bar reflects the p-value associated with comparisons using the general linear model performed at each cortical surface point. Main effects of (A) groups with bipolar disorder (BP) [BP-only and BP/attention-deficit hyperactivity disorder (ADHD)] versus those without BP (ADHD-only and controls), where p < 0 (blue/purple) signifies regions in which the presence of BP is associated with cortical thinning, and (B) groups with ADHD (ADHD-only and BP/ADHD) versus those without ADHD (BP-only and controls), where p < 0 signifies regions in which the presence of ADHD is associated with cortical thinning, and p > 0 (red/yellow) is associated with increased cortical thickness. Pairwise comparisons of (C) BP-only versus healthy controls, where p < 0 indicates regions in which BP-only was associated with cortical thinning relative to controls, and (D) ADHD-only versus BP/ADHD, where p < 0 indicate regions in which BP/ADHD is associated with cortical thinning relative to ADHD-only.