Differential therapeutic effects of atomoxetine and methylphenidate in childhood attention deficit/hyperactivity disorder as measured by near-infrared spectroscopy

Attention-deficit/hyperactivity disorder (ADHD) is one of the most commonly diagnosed neurodevelopmental disorders in children with lifelong deficits in a wide range of executive functions [1]. ADHD symptoms are thought to arise from dysregulation of prefrontal and subcortical catecholamine neurotransmission [2, 3]. The stimulant methylphenidate (MPH) and the nonstimulant atomoxetine (ATX) are the most frequently prescribed drugs for the treatment of ADHD. Both drugs are known to reduce clinical ADHD symptoms. The common mechanism of both drugs is that they modulate dopamine (DA) and norepinephrine (NE) neurotransmission [4]. Small changes in DA or NE concentration affect networks of pyramidal cells in the prefrontal cortex (PFC), which regulates and sustains attention [5]. It is believed that the therapeutic effects of both medications occur primarily in the PFC [5], although the exact mechanisms of their actions are unclear.

Methylphenidate acts as an indirect DA agonist, inhibiting DA reuptake by occupying the DA transporter [6]. MPH has also shown to block the norepinephrine (NE) transporter in NE transporter-rich regions, including the PFC [7]. In rodent studies, MPH has been shown to enhance the extracellular levels of both DA and NE [8]. In contrast, although ATX is a selective NE reuptake inhibitor, it also inhibits DA reuptake in the PFC. Therefore, while it does not increase DA in the striatum, it increases both DA and NE in the prefrontal cortex [8]. The partially overlapping pharmacologic profiles of these medications suggest both similarities and differences in their therapeutic mechanisms of action. In the meta-analysis focused on the comparison between MPH and ATX, MPH showed a higher response rate compared to ATX [9]. In a randomized study directly comparing MPH and ATX in adults with ADHD, the effects on executive functions were generally similar, although there was a suggestion that ATX might show a slight benefit to the immediate-release MPH in terms of improving spatial planning [10]. However, another head-to-head study comparing the two drugs found that only osmotically-released MPH improved set-shifting and verbal fluency, although osmotically-released MPH and ATX both improved executive function generally in children and adolescents with ADHD [11]. Distinct underlying pharmacological mechanisms may cause these practical differences. There are few neuroimaging studies that examined these differences [12, 13]. Cubillo et al. showed that ATX upregulated and normalized right dorsolateral prefrontal cortex under activation measured by functional magnetic resonance imaging (fMRI), while MPH upregulated left inferior frontal cortex activation [13].

Near-infrared spectroscopy (NIRS) enables the noninvasive detection of neural activity near the surface of the brain using near-infrared light [14, 15]. It measures alterations in oxygenated hemoglobin ([oxy-Hb]) and deoxygenated.

Hemoglobin ([deoxy-Hb]) concentrations in micro-blood vessels on the brain surface. Local increases in [oxy-Hb] and decreases in [deoxy-Hb] are indicators of cortical activity [15, 16]. In animal studies, [oxy-Hb] is the most sensitive indicator of regional cerebral blood flow because the direction of change in [deoxy-Hb] is determined by the degree of changes in venous blood oxygenation and volume [17]. Therefore, we decided to focus on changes in [oxy-Hb]. Furthermore, changes in [oxy-Hb] have been associated with changes in regional cerebral blood volume, using a combination of NIRS and positron emission tomography (PET) measurements [18, 19]. NIRS is a neuroimaging modality that is especially suitable for psychiatric patients for the following reasons [20]. First, because NIRS is relatively insensitive to motion artifact, it can be used in experimental scenarios in which motion may occur, such as while assessing participants who are prone to vocalization. Second, participants can be examined in a natural sitting position, without any surrounding distractions such as fMRI. Third, the cost is much lower than that of other neuroimaging modalities and the setup is very easy. Fourth, the high temporal resolution of NIRS is useful in characterizing the time course of prefrontal activity in people with psychiatric disorders [21, 22]. Fifth, functional studies of pediatric patients using single-photon emission computed tomography (SPECT) and PET are rare due to restrictions regarding the use of radioactive materials in young individuals. Accordingly, NIRS has been used to assess brain function in people with many types of psychiatric disorders, including schizophrenia, bipolar disorder, post traumatic disorder, obsessive–compulsive disorder, and ADHD [20‐28].

In pediatric ADHD, reduced prefrontal hemodynamic response has been measured by NIRS [23, 29, 30]. Negoro et al. examined prefrontal hemodynamic response during the Stroop color-word task in 20 children with ADHD and 20 healthy age- and sex-matched controls. They found that the oxy-Hb changes in the inferior prefrontal cortex in the control group were significantly larger than those in the ADHD group during the Stroop color-word task [23]. In an NIRS study of medication, Ota et al. examined the effects of a clinical dose of ATX on changes in prefrontal hemodynamic response during the Stroop color-word task in pediatric ADHD. They found that ATX induced an intensified prefrontal hemodynamic response [31]. In another NIRS study, Araki et al. found that the oxy-Hb concentration in the right dorsolateral PFC in the post-ATX condition was significantly increased compared to the pre-ATX condition during a continuous performance task [32]. Despite several NIRS studies with ADHD, only a few studies have examined the therapeutic effects of medication. Moreover, no studies have compared MPH with ATX directly. In this study, we examined the drug-specific effects of a clinical dose of either MPH or ATX on frontal activation as measured by NIRS in a cohort of medication-naïve pediatric ADHD subjects. We used the Stroop color-word task to assess inhibitory control and selective attention. As outlined above, there are distinct underlying pharmacological mechanisms associated with MPH and ATX. Therefore, we hypothesized that there might be a differential hemodynamic response across MPH and ATX.

To our knowledge, this is the first NIRS study to compare the effectiveness of MPH with ATX directly in children with ADHD by measuring hemodynamic responses during the Stroop color-word task. ATX significantly increased activation in the prefrontal cortex, especially in left lateral frontal pole cortex (FPC), after 12 weeks of administration. MPH did not increase activation in the prefrontal cortex, but it did make comparable improvements in terms of ADHD symptoms and Stroop color-word task performance to those seen in ATX.

Some studies have referred indirectly to differences in the neurobiological actions between MPH and ATX. Event-related potential studies of oddball tasks in pediatric ADHD have shown that MPH can normalize low P300 or mismatch negativity amplitudes [41], while ATX can normalize long P300 latencies and low MMN amplitudes, at least partially [42]. In a fMRI study of adult ADHD using a multi-source interference task, ATX did not activate dorsal anterior midcingulate cortex [43], as MPH has been shown to do [44]. However, few studies have directly investigated how the pharmacological mechanisms of action differ between the two drugs, and little is known about the mechanisms by which they exert their therapeutic effects. In one fMRI study that used a go/no-go task with 36 participants with pediatric ADHD, comparable improvements in response inhibition and ADHD symptoms were seen after 6 to 8 weeks of daily treatment with MPH vs ATX. Symptomatic improvement was associated with gains in task-related activation for ATX and reductions in activation for MPH in the right inferior frontal gyrus, left anterior cingulate/supplementary motor area, and bilateral posterior cingulate cortex [45]. In another fMRI study using a counting Stroop paradigm, 12 weeks of ATX pharmacotherapy decreased activity in the dorsal anterior cingulate cortex and dorsolateral prefrontal cortex in 42 participants with pediatric ADHD, which correlated with improvement in focused attention. In contrast, MPH increased activity in the inferior frontal gyrus, which correlated with decreasing severity of impulsivity [46]. Comparing effects of acute doses of both drugs and a placebo with boys with ADHD during a stop task, MPH had a drug-specific effect of normalizing the right ventrolateral prefrontal and cerebellar under-activation observed under both placebo and ATX [47]. Taken together, these reports indicate that the mechanisms by which MPH and ATX exert their therapeutic effects are different: this is consistent with the findings from the present study. Nevertheless, the concept of drug-specific laterality effects on prefrontal regions is still controversial. Our data showed that ATX upregulated the frontal cortex during Stroop interference, at least partially. The present findings suggest that frontal mechanisms serve an important role in the therapeutic actions of ATX. However, despite the fact that MPH did not increase activation in the PFC, there were still comparable improvements in terms of ADHD symptoms for those taking this medication. One parsimonious explanation is that MPH increases activation in other brain regions, which might contribute to the improvement in ADHD symptoms.

Volkow et al. [48, 49] found that in healthy adults, MPH enhanced the salience of a reward task, increased levels of extra-cellular dopamine, and induced reductions in glucose metabolism within the default mode network (DMN). The DMN is a distributed brain system, comprising medial pre-frontal cortex and medial and lateral parietal regions. It is anti correlated with the attentional networks activated by goal-directed behavior, and is thought to reflect intrinsic activity [50]. Recently, influential new brain network models [51, 52] have proposed that proper sustained attention functioning requires both engagement of task-positive networks (TPNs), including a frontoparietal control network and dorsal and ventral attention networks, and suppression of the DMN [50, 53, 54]. A failure of the anti-phase synchronization between DMN and TPN may be involved in the manifestation of ADHD. There is evidence suggesting that the striatal DA system plays a role in the modulation of the DMN [55, 56]. MPH produces robust increases in extracellular dopamine levels [57], which potentiate corticostriatal inputs [58] and have been found to enhance striatal activation in child ADHD [59, 60]. Furthermore, some studies have shown that MPH may normalize DMN deactivation patterns [61, 62]. Therefore, we speculate that MPH might tend to activate DMN regions rather than TPN during task-related activation. In contrast, an increase of prefrontal activation has been reported after MPH treatment in several studies using different neuroimaging modalities, including NIRS [44, 59, 63, 64]. The variability in findings across studies is likely related to different cognitive tasks, dosage, patients’ ages, and treatment duration.

Increases in left lateral FPC activity were observed after ATX treatment in our study. However, we found no significant correlations between the hemodynamic changes in this area and degree of the clinical improvements. The FPC is the most anterior part of the cerebral cortex, and has reciprocal connections with most prefrontal areas [65, 66]. Tsujimoto et al. suggested that the FPC has a role in monitoring and evaluating decisions, especially those with a self-generational component [67]. Arai et al. found that children with ADHD show abnormalities in functional maturation of the frontal pole [68]. Based on these findings, a direction for future research will be to assess participants using another battery associated with self-generated behavior, separate to NIRS recordings.

The results of the present study suggest that multi-channel NIRS systems may have potential in the pharmacotherapeutic evaluation in children with ADHD for clinical practice. It is very significant for patients that an effect of the pharmacotherapy is visualized. In the future, it is need to predict the effect of the pharmacotherapy using the NIRS for clinical practice.

The present study has several potential limitations. First, methodological limitations include the relatively small number of participants, non-randomised study, and lack of a double-blind, placebo-controlled design. At baseline, the ATX group had higher mean SCWC1 and SCWC2 scores than the MPH group. Although scores were not correlated with degree of clinical severity with ADHD-RS, the two groups were not quite entirely equivalent in their characteristics. Future work seeking to compare MPH, ATX and/or placebo should consider a double-blind randomized or crossover design with larger samples. Second, we had no healthy control as a comparison cohort. Our study showed that ATX significantly increased activation in channel 21. In one previous NIRS study using the Stroop, Negoro et al. reported a lower increase of oxy-changes in channels 8, 18, 19, 21, and 22 in individuals with child ADHD compared with controls [23]. Considering the above findings, we predicted that improvement of ADHD symptoms with ATX treatment would be associated with increased activation in those regions; our findings were consistent with these predictions. Third, NIRS does not detect activity in deeper cortical structures, such as the medial pre-frontal cortex, which is part of the DMN. Fourth, the spatial resolution for the detection of hemodynamic responses from the scalp surface using NIRS is lower than that of fMRI, SPECT, or PET. However, the spatial resolution may be within an acceptable range because previous NIRS studies have also found clear distinctions in hemodynamic responses between diagnostic groups [23‐26, 28, 69].

In conclusion, this is the first NIRS study using the Stroop interference task to examine how the pharmacological mechanisms of action differ between MPH and ATX. Findings suggest that effective treatment with MPH and ATX is produced by distinct mechanisms in frontal regions.