A Placebo-controlled Trial of Brain Activity Changes Following Psychostimulant Medication in Adolescent Combined-subtype ADHD

By conservative estimate, Attention-Deficit Hyperactivity Disorder (ADHD) is a common disorder that occurs in at least 3 to 6% of school-aged children (Barkley, 1990). It is marked by symptoms of inattention and hyperactive-impulsive behavior that are developmentally inappropriate. These symptoms must occur in multiple settings and impair functioning to a clinically significant degree (DSM-IV; American Psychiatric Association, 1994). ADHD is believed to be a developmental disorder that begins in early childhood, which distinguishes it from disorders of attention and behavioral control that arise later as consequence of brain injury. Although ADHD was once thought to affect only children, considerable evidence has accumulated to show that ADHD frequently continues into adulthood. Approximately 60% of ADHD youth (about 3-4% of all adults) have clinically significant residual problems as adults (Faraone, 1999; Hechtman, 2000; Mannuzza & Klein, 1999). Although inattention problems appear to persevere into adulthood unchanged, studies suggest that some persons' ADHD hyperactive/impulsive symptoms diminish following teenage years. However, the mechanism for this decrease is not currently known (Wender, 1987; Weiss & Hechtman, 1986, 1993; Brown, 1995).

Expert consensus holds that a deficit in response inhibition is a core feature of ADHD (Barkley, 1998). Supporting this, neuropsychological research suggests that ADHD persons' commonly perform poorly on tasks that require inhibition of response. These tasks, called Go/No-Go or Stop Signal tasks, require continual manual responding to successively presented stimuli and an occasionally withheld response following a predefined signal. Numerous neuropsychological studies have shown that ADHD children, teens, and adults respond more variably on Go trials on such tasks. They also have longer reaction times and more failures to inhibit responses on No-Go trials (Koshack et al., 2003; Oosterlaan, et al., 1998; Wodushek & Neumann, 1999). Psychostimulant medications like methylphenidate improve performance deficits (Bedard et al., 2003).

The link between abnormalities of brain structure/function and cognitive test performance in ADHD persons is well-established, if poorly understood. There is evidence for smaller right frontal lobe volume, abnormalities in basal ganglia size, and reductions in white matter tracts in persons with ADHD (Castellanos et al., 1994; Castellanos et al., 1996; Filipek et al., 1997; Hynd et al.,1993; Mataro et al., 1997). Prefrontal cortex and basal ganglia are involved in planning, initiation, execution, and supervision of ongoing cognitive and motor functions (Fuster, 1997). Structural brain abnormalities correlate with deficits in these cognitive functions and with impulsive behaviors that are symptoms of ADHD (Casey et al., 1997; Mataro et al., 1997; Semrud-Clikeman et al., 2000). Functional imaging techniques, which quantify neural systems operation with respect to cognitive processes, are particularly valuable tools used to characterize dysfunctional neural networks. The most frequently employed method is electrophysiology, in which small voltage changes on the scalp occurring during task performance are recorded and averaged into waveforms that depict brain function to events of interest. These average EEG waveforms are called event-related potentials (ERPs). ERPs to Go/No-Go tasks are well characterized in healthy, non-clinical samples (Logan, 1994). The most frequently observed component to No-Go trials is a large negative potential occurring approximately 275 msec following stimulus presentation (N2) that has a fronto-central distribution. The large N2 has been interpreted as a reflection of early response inhibition processes. Electrophysiological studies find that the large N2 often seen with successful inhibition is diminished in ADHD children and adolescents (Overtoom et al., 2002; Pilszka et al., 2000; Yong-Liang et al., 2000). A literature review reveals no comparable studies of ADHD adults, so the degree of N2 abnormality in older ADHD persons is not known.

ERP source localization suggests that the No-Go N2 arises from activity in inferior prefrontal cortex (Bokura, 2001; Pliszka, 2000). The few FMRI studies of ADHD are consistent with this finding (Garavan et al., 1999; Liddle et al., 2001; Watanabe et al., 2002). Rubia et al. (1999) find less brain activity in right superior-lateral prefrontal cortex, right inferior prefrontal cortex, and left caudate nucleus in unmedicated ADHD adolescent boys relative to controls. Vaidya et al. (1998) report similar findings for ADHD children, and also showed that methylphenidate increased activity in the striatum for ADHD participants. These results support a fronto-temporal N2 decrease in ADHD in brain areas implicated by structural studies as abnormal.

Durston and colleagues also found evidence that compared to non-ADHD, ADHD children also activate a more diffuse network of posterior and dorsolateral prefrontal cortex regions to subserve successful response inhibition (Durston et al., 2003). Increasing age also significantly alters the pattern of activity seen on Go/No-Go tasks for healthy, non-ADHD persons in these brain regions. Children activate more volume of prefrontal cortex compared to adults (Casey et al., 1997). There also is evidence that healthy adults and teenagers use different neural networks to achieve comparable response inhibition task performance (Rubia et al., 2000). Healthy adults show modulation of ventral prefrontal activity depending on task difficulty, whereas children appear to maximally activate these areas generally (Durston et al., 2002). In a sample of ADHD participants ages 8 to 20, age was found to correlate positively with left inferior frontal cortex, and correlate negatively with hemodynamic activity in left dorsolateral prefrontal cortex (Tamm et al., 2002). In other words, there was increasing use of cortex known to be a primary substrate of response inhibition, while reliance on cortex that is a primary substrate of working memory and other "on-line" cognitive processes decreased with age. These differences suggest that age is an important variable in understanding response inhibition neural activity in both ADHD and non-ADHD persons.