Positron Emission Tomography
Positron Emission Tomography (PET) studies of ADHD have mostly been restricted to older subjects due to concerns about exposing children to the radioactive isotopes necessary for this technique (Lou, 1992; Zametkin et al., 1990, 1993). The few studies that have been conducted have yielded inconsistent results. Zametkin et al. (1990) found reduced global glucose metabolism and reduced regional metabolism in 30 of the 60 areas examined in adults with a history of childhood ADHD, when compared with normal controls. The greatest decreases in metabolism occurred in premotor and superior prefrontal cortex, areas associated with control of motor activity and attention, suggesting a relationship between reduced frontal metabolic activity and ADHD symptoms (Zametkin et al., 1990). Reduced metabolism in ADHD subjects was also found in the striatum and the thalamus. In a subsequent study of adolescents with ADHD (Zametkin et al., 1993), no significant group differences were found for global or absolute regional cerebral glucose metabolism. However, reduced normalized (regional/global) metabolism in ADHD subjects was found in six regions including the left anterior frontal cortex, where significantly reduced glucose metabolism was also found in adults with ADHD in the previous study (Zametkin et al., 1990). Left anterior frontal metabolism was significantly correlated with ADHD symptom severity in the adolescent sample, providing further evidence of a link between reduced frontal metabolism and deficits in motor and attentional control (Zametkin et al., 1993).
In both the adult (Zametkin et al., 1990) and adolescent (Zametkin et al., 1993) studies, a trend toward stronger group differences in metabolism in female subjects than in males was observed (Ernst et al., 1994a; Zametkin et al., 1993). When the adolescent sample was expanded in a subsequent study, no group differences in global or absolute regional metabolism were found across the entire groups (Ernst et al., 1994a). However, global glucose metabolism was 15% lower in girls with ADHD than in normal girls, and significant regional metabolism reductions occurred in premotor, orbito-frontal and temporal cortex in girls with ADHD. No significant differences were found between boys with ADHD and normal boys. A further study which included a larger independent sample of adolescent girls failed to replicate this finding, as no differences in global or regional metabolism between girls with ADHD and normal girls were found (Ernst et al., 1997). Differences in sample characteristics and data analysis techniques were discussed as possible reasons for the conflicting results (Ernst et al., 1997). Lateralization of normalized metabolism was significantly different between groups in the later study, with lower metabolism in the left hemisphere in girls with ADHD and in the right hemisphere in controls (Ernst et al., 1997).
Two recent PET studies by Ernst and colleagues used the tracer [flourine-18]flourodopa to examine presynaptic dopaminergic function in adults and children with ADHD. In the first study with adults, prefrontal cortex, striatum and midbrain regions were examined, but only the prefrontal cortex showed significantly lower DOPA decarboxylase activity in ADHD adults, with medial and left prefrontal areas showing the largest differences (Ernst et al., 1998). Given this and previous findings of discrepancies between adults and adolescents with ADHD, the authors hypothesized that a prefrontal dopaminergic dysfunction underlies ADHD symptoms in adults and that this dysfunction may be secondary to subcortical dopaminergic deficits and their interactions with maturational processes (Ernst et al., 1998). This hypothesis is supported by their subsequent findings that children with ADHD had higher midbrain [18F]flourodopa accumulation than normal controls and that right midbrain [18F]flourodopa accumulation was correlated with symptom severity (Ernst et al., 1999). There were no differences in prefrontal and striatal measures in these children.
A study examining regional cerebral blood flow changes associated with working memory found that rCBF changes in men with ADHD were prominent in occipital regions, while those in healthy controls were prominent in frontal and temporal regions (Schweitzer et al., 2000). The authors concluded that their results suggest compensatory mechanisms in subjects with ADHD in response to disrupted inhibition and internally guided behavior.
Neither methylphenidate nor dexamphetamine have been shown to alter global glucose metabolism in adults with ADHD (Ernst et al., 1994b; Matochik et al., 1993, 1994). In addition, both drugs produced inconsistent patterns of increases and decreases in regional metabolism (Matochik et al., 1993, 1994).
Studies using Single Photon Emission Computed Tomography (SPECT) have revealed reduced cerebral blood flow in the frontal lobes and in the caudate nucleus in children with ADHD (Amen & Carmichael, 1997; Lou et al., 1984, 1989; Sieg et al., 1995). In the first, small study using Xenon-133 SPECT, cerebral blood flow was reduced in frontal lobe white matter regions and in the caudate nuclei in ADHD subjects (Lou et al., 1984). These findings were replicated in subsequent studies which found hypoperfusion in the striatum, especially on the right, and hyperperfusion in the occipital lobes and left sensori-motor and primary auditory regions (Lou et al., 1989, 1990a). The reduced activity in frontal and striatal regions and increased activity in primary sensory regions were partly reversible with administration of methylphenidate (Lou et al., 1984,1989). The authors postulated a primary dysfunction of striatal structures in ADHD, which leads to disinhibition and hyperfunction of primary sensory and sensori-motor cotices (Lou et al., 1989). Lower striatal activity was also found in pre-school children compared to older children in a normal developmental study (Lou et al., 1990b), suggesting a neu-robiological correlate to the behavioural immaturity seen in children with ADHD (Lou, 1992). These findings of reduced striatal activity in children with ADHD are consistent with the known anatomical connections between the caudate nuclei and prefrontal cortex (O'Tuama & Treves, 1993) and suggest that dysfunction in these pathways may be associated with the symptoms of ADHD (Dinklage & Barkley, 1992). One problem with the Lou et al. (1984, 1989) studies is that many of their subjects suffered from some type of early neurological insult such as hypoxia or encephalitis, raising the question of whether the findings are applicable to more typical groups of children with ADHD whose development was free of potentially damaging traumas (O'Tuama & Treves, 1993).
In a more recent study using N-Isopropyl I-123 IMP SPECT, Sieg et al. (1995) found greater hemispheric I-123 IMP uptake asymmetry in ADHD subjects, with reduced blood flow in left frontal and left parietal regions, in comparison to psychiatric controls. The authors suggested their results, in addition to other PET and SPECT findings, might reflect a maturational lag resulting from delayed myelin-ization in ADHD, especially in the frontal lobes (Sieg et al., 1995). Reduced frontal activation was also found by Amen and Carmichael (1997) using high-resolution
SPECT in children and adolescents with ADHD under resting and "intellectual stress" conditions. Under intellectual stress, i.e. when performing a concentration task, 65% of ADHD subjects showed reduced perfusion compared to the resting condition in the prefrontal cortex. Only 5% of control subjects showed the same reduction in prefrontal activation.
There have been only a few preliminary studies of ADHD using functional magnetic resonance imaging (fMRI) to date. A study of adolescents with ADHD found that, compared with normal peers, they had reduced activation in right hemisphere cortical regions including the anterior cingulate, pre- and post-central gyrus and posterior parietal cortex during a visual stop signal task (Rubia et al., 1999). However the ADHD group had increased activation in subcortical areas including the right and left insula and left caudate. The authors concluded that ADHD is associated with dysfunction of right hemisphere inferior frontal and striatal regions during motor inhibition (Rubia et al., 1999). Another study of 10 ADHD subjects aged from 14 to 51 years found that activation was predominant in the right middle frontal gyrus during a visual vigilance task (Sunshine et al.,
1997). No comparison group was included in this study, but the authors reported that similar areas of activation were found in a previous study of normal subjects using the same task (Lewin et al., 1996). Some additional areas of activation not seen in normal subjects were found in the ADHD group in right and left frontal, left precentral and left parietal regions. The authors concluded that this result might represent true regions of abnormality in the ADHD subjects during visual vigilance, perhaps related to attempted compensation for their disorder, or alternatively may be due to artifacts (Sunshine et al., 1997).
A recent fMRI study examined fronto-striatal function in children with ADHD and matched controls during the performance of go/no-go tasks (Vaidya et al.,
1998). Compared to the controls, the ADHD group had significantly reduced stri-atal activation during a stimulus-controlled (more difficult) task and significantly increased frontal activation during a response-controlled (easier) task. The authors suggested that the finding of increased frontal activation, which differs from previous reports of frontal hypometabolism in ADHD (Sieg et al., 1995; Zametkin et al., 1990), might reflect greater inhibitory effort. The finding of reduced striatal activation is consistent with other functional imaging studies (Lou et al., 1984, 1989) and with structural imaging studies that have reported associations between anatomical abnormalities of the striatum and poor performance on inhibitory tasks in ADHD subjects (Casey et al., 1997; Mataro et al., 1997). The authors concluded that activation in children with ADHD may be abnormally high or low depending on the specific demands of inhibitory control imposed by the task (Vaidya et al., 1998).
Was this article helpful?