Reduced short interval cortical inhibition correlates with atomoxetine response in children with attention-deficit hyperactivity disorder (ADHD)

Abstract

Clinical trials in children with attention-deficit hyperactivity disorder (ADHD) show variability in behavioral responses to the selective norepinephrine reuptake inhibitor atomoxetine. The objective of this study was to determine whether transcranial magnetic stimulation-evoked short interval cortical inhibition might be a biomarker predicting, or correlating with, clinical atomoxetine response. At baseline and after 4 weeks of atomoxetine treatment in 7- to 12-year-old children with ADHD, transcranial magnetic stimulation short interval cortical inhibition was measured, blinded to clinical improvement. Primary analysis was by multivariate analysis of covariance. Baseline short interval cortical inhibition did not predict clinical responses. However, paradoxically, after 4 weeks of atomoxetine, mean short interval cortical inhibition was reduced 31.9% in responders and increased 6.1% in nonresponders (analysis of covariance t 41 = 2.88; P = .0063). Percentage reductions in short interval cortical inhibition correlated with reductions in the ADHD Rating Scale (r = 0.50; P = .0005). In children ages 7 to 12 years with ADHD treated with atomoxetine, improvements in clinical symptoms are correlated with reductions in motor cortex short interval cortical inhibition.

Keywords: atomoxetine; attention-deficit hyperactivity disorder (ADHD); short interval cortical inhibition; transcranial magnetic stimulation.

Conflict of interest statement

Declaration of Conflicting Interests

There are no other conflicts of interest to disclose. All authors approve submission of this work.

© The Author(s) 2014.

Figures

Figure 1

ADHDRS in responders and non-responders at baseline and after 4 weeks of atomoxetine treatment. Means from ANCOVA, error bars are standard error.

Figure 2

SICI in responders and non-responders at baseline and after 4 weeks of atomoxetine treatment. Means from ANCOVA, error bars are standard error.

Figure 3

ICF in responders and non-responders at baseline and after 4 weeks of atomoxetine treatment. Means from ANCOVA, error bars are standard error.