Acute atomoxetine effects on the EEG of children with Attention-Deficit/Hyperactivity Disorder

Abstract

Although stimulant medications are the most commonly-used treatments for Attention-Deficit/Hyperactivity Disorder (AD/HD), as many as 20% of treated children do not respond clinically to stimulants. This study investigated the effects of an acute dose of atomoxetine, a selective noradrenaline reuptake inhibitor (SNRI), on the electroencephalogram (EEG) and performance of children with AD/HD. An initial pre-medication EEG was recorded during an eyes-closed resting condition. Within two weeks, a second EEG was recorded 1 h after ingestion of 20 mg of atomoxetine. Data were Fourier transformed to provide absolute and relative power estimates for the delta, theta, alpha, beta and gamma bands. Compared to controls, the unmedicated AD/HD children had significantly elevated global absolute and relative delta, with reduced global relative alpha, and absolute and relative gamma, and many topographic differences. Atomoxetine produced significant global increases in absolute and relative beta, with several topographic changes in other bands, and a significant reduction in omission errors on a Continuous Performance Task. These results indicate that SNRIs can produce substantial normalisation of the AD/HD EEG profile, together with behavioural performance improvements. Although EEG changes induced by acute administration of psychostimulants (methylphenidate/dexamphetamine) and atomoxetine are not identical, both classes of AD/HD drugs produce similar EEG band changes. Further analysis of EEG responses to SNRIs and psychostimulants could reveal common neurophysiological processes closely linked to clinical improvement of AD/HD symptoms in response to pharmacotherapy, providing translational markers for clinical efficacy studies and potential translational biomarkers for AD/HD drug discovery.

Introduction

Attention-Deficit/Hyperactivity Disorder (AD/HD) is a persistent problem currently affecting 4%–6% of school-age children (American Psychiatric Association (APA), 1994, Lindgren et al., 1990, Pelham et al., 1992), which interferes with many aspects of normal development and functioning in a child's life. In Australia and North America, stimulant medications are widely used in the treatment of AD/HD, with numerous controlled trials indicating that approximately 80% of patients have clinically-significant benefits (Swanson et al., 1993, Wilens and Biederman, 1992). However, there has been a considerable effort to develop alternative pharmacotherapies to psychostimulants in AD/HD treatment, particularly for a sub-set of patients who do not tolerate psychostimulants well. Treatments with norepinephrine (NE) reuptake inhibitor tricyclic antidepressant drugs, such as desipramine, have been successful in adolescent AD/HD patients, but their potentially-serious adverse cardiac effects (arrhythmias) limit their clinical use (Biederman et al., 1986, Smith et al., 2000). Currently the only non-stimulant drug FDA-approved for treatment of AD/HD is atomoxetine, a prototype of the selective NE reuptake inhibitors (SNRIs). In randomized, placebo-controlled trials in children and adolescents, atomoxetine has demonstrated a statistically-significant reduction in core AD/HD symptoms, and improvement in social and family functioning, compared with the placebo group (Michelson et al., 2001, Pliszka, 2007). In addition, it has been reported that reboxetine, a highly selective NRI, is a clinically-effective AD/HD treatment (Toren et al., 2007).

Electroencephalogram (EEG) studies of children with and without AD/HD typically report that AD/HD children have increased theta activity, often maximal in the frontal regions (Barry et al., 2003, Chabot and Serfontein, 1996, Clarke et al., 1998, Clarke et al., 2001b, Clarke et al., 2001c, Clarke et al., 2001d, Clarke et al., 2002c, Janzen et al., 1995, Lazzaro et al., 1998, Satterfield et al., 1972), increased posterior delta (Clarke et al., 1998, Clarke et al., 2001b, Clarke et al., 2001d, Matousek et al., 1984), and decreased alpha and beta activity (Barry et al., 2003, Callaway et al., 1983, Dykman et al., 1982), also most apparent in the posterior regions (Barry et al., 2003, Clarke et al., 1998, Clarke et al., 2001b, Clarke et al., 2001d, Lazzaro et al., 1998). There is emerging evidence that gamma activity (35–45 Hz) is also reduced in AD/HD, particularly in posterior and hemispheric regions (Barry et al., unpublished data).

A number of studies have investigated changes in the EEG due to stimulant medications, with the majority of studies finding that the stimulants result in some normalisation of the EEG. Swartwood et al. (1998) and Lubar et al. (1999) failed to find changes in EEG power due to stimulant medication, but Chabot et al. (1999) found that 56.9% of a group of children with AD/HD showed normalisation of the EEG after the administration of a stimulant, while 33.8% remained unchanged, and 9.3% showed an increase in EEG abnormality. Loo et al. (1999) found that, after administration of methylphenidate, good responders had decreased theta and alpha, and increased beta activity in the frontal regions, while poor responders showed the opposite EEG changes. Clarke et al., 2002a, Clarke et al., 2003, Clarke et al., 2007) found that stimulant medications resulted in normalisation of the EEG, with a reduction in theta activity and an increase in beta activity. These results were interpreted as indicating that stimulants acted at a cortical level by increasing arousal, although complete normalisation was not found in the entire sample. However, the effects of non-stimulants on the EEG of children with AD/HD has not been widely investigated. Clarke et al. (2008) examined the effects of imipramine hydrochloride (Tofranil, a non-stimulant) on the EEG of children with AD/HD who were non-responsive to stimulants. Although the patients showed clinical improvement over 6 months while on this medication, there were no significant changes in the EEG pattern. There are no EEG studies of SNRIs in AD/HD, and hence the aim of this study was to investigate the effect of an acute dose of atomoxetine on the EEG profile of children with AD/HD.

One tool that has been used to evaluate acute medication effects in children with AD/HD is the continuous performance task (CPT; Riccio et al., 2001). Used with a range of variants, the CPT has been reported as being the most popular clinic-based measure of sustained attention and vigilance in children with AD/HD (DuPaul et al., 1992). Since its first use in 1956 (Rosvold et al., 1956), the CPT has undergone considerable development, with the A–X variant being the most widely used today (Riccio et al., 2001). This requires the subject to watch a continuous stream of random targets and non-targets, and to respond only when a target (X) follows a specified non-target (A). Counts of omission errors (failures to respond to an X following an A) and commission errors (responses to an X not preceded by an A, and other inappropriate responses) are commonly derived. Research has shown that CPTs provide a measure of attention (or inattention in the case of children with AD/HD), and to some degree, a measure of impulsivity (Swanson, 1985), two of the main cognitive deficits in AD/HD. CPTs have excellent face validity, measure concentration well, and provide an objective measure of responsivity to a medication (Benedict et al., 1994). To date, the effects of atomoxetine on CPT performance have not been reported, so the second aim of this study was to evaluate changes in CPT scores as a result of an acute dose of atomoxetine. Indications of substantive EEG and CPT effects with atomoxetine would encourage further investigations of the effectiveness of SNRIs in the treatment of AD/HD.