The influence of methylphenidate on resting-state
fMRI connectivity in ADHD patients: A review
Author
Caroline Broeder
Supervisor
Antonia Kaiser
Examiner
Bianca Boyer
A thesis submitted to the Institute of Interdisciplinary Studies,
University of Amsterdam
In partial fulfillment of the requirements for the degree of
Master in Brain and Cognitive Sciences
Abstract
Methylphenidate is the first-choice pharmacological treatment for Attention- Deficit/Hyperactivity Disorder (ADHD). Although resting-state fMRI (rs-fMRI) research has provided better understanding of the network pathology of ADHD, knowledge of the effects of methylphenidate on rs-fMRI is lagging behind. This paper reviews the effects of methylphenidate on rs-fMRI connectivity in ADHD patients in comparison to healthy controls. Although findings of rs-fMRI in ADHD are divergent, overarching hypotheses suggest abnormalities in several resting-state networks, including the default mode network (DMN), cognitive control network, salience network and affective-motivational network. Methylphenidate has been shown to normalize altered resting-state connectivity in children with ADHD, which was mainly found in fronto-parietal-cerebellar circuits. Preliminary evidence suggests similar normalization in adults with ADHD. In healthy adults, these effects appear more divergent, with evidence suggesting connectivity alterations in temporal and lingual gyri, thalamus, striatum, locus coeruleus, basal nucleus of Meynert, ventral tegmental area and in several resting-state networks, including the DMN, sensory-motor, visual, frontoparietal and executive control network and cortico-striato-thalamo-cortical circuits. Remarkable are the effects found in multiple subcortical areas, as this has not been investigated in ADHD patients yet. Additionally, the differences in effects between ADHD patients and controls suggest that the effects of methylphenidate might depend on resting-state connectivity at baseline, which highlights the importance of dimensional analyses rather than categorical comparisons in further research. Additionally, the use of suitable methods for motion correction, large-scale data-sharing initiative and inclusion of medication-naïve participants seem the way forward in this growing field of research. Taken together, these findings highlight possible causal mechanisms by which alterations in resting-state connectivity might facilitate symptoms in ADHD.
Keywords: Attention- Deficit/Hyperactivity Disorder, methylphenidate, resting-state functional connectivity, functional magnetic resonance imaging
Introduction
Attention-deficit/hyperactivity disorder (ADHD) is one of the most prevalent neurodevelopmental disorders in school-aged children (Wolraich, Hagan, Allan, & Chan, 2015), with a worldwide-pooled prevalence of 7.2% in children and adolescents (Thomas, Sanders, Doust, Beller, & Glasziou, 2015). It is mainly characterized by developmentally inappropriate inattention, hyperactivity and impulsivity (American Psychiatric Association, 2013). ADHD is typically diagnosed in childhood, with symptoms persisting into adulthood in 60% of cases (Sibley et al., 2017). Most adults do not continue to meet the full set of diagnostic criteria: symptoms usually become more covert in adulthood and hyperactivity and impulsivity often decline, whereas inattention persists (Faraone et al., 2015; Hart, Lahey, Loeber, Applegate, & Frick, 1995; Holbrook et al., 2016; Larsson, Dilshad, Lichtenstein, & Barker, 2011). Nevertheless, ADHD is associated with a severe social burden, as the severity of ADHD symptoms is linked to the number of adverse life events, even when excluding the influence of comorbid psychiatric disorders (Garcia et al., 2012). Ultimately, ADHD symptoms seriously affect the quality of life in
both children (Danckaerts et al., 2010) and adults (Agarwal, Goldenberg, Perry, & Ishak, 2012).Therefore, ADHD patients of all ages would greatly benefit from effective treatment.
Treatment options for ADHD include pharmacological and non-pharmacological interventions. International guidelines on the treatment of ADHD recommend a combination of these treatment types, as medication might be required in order to obtain maximal symptom improvement. Various types of ADHD medication exist, including stimulant and nonstimulant medications. Although these types of medication have both been proven effective, evidence is the strongest for psychostimulant drugs. Therefore, these have remained a first-line treatment (Van der Oord, Prins, Oosterlaan, & Emmelkamp, 2008; Wolraich et al., 2015). Worldwide, one of the most frequently prescribed psychostimulant treatments for ADHD is methylphenidate, a stimulant that inhibits the reuptake of dopamine and noradrenaline (Cortese, D’Acunto, Konofal, Masi, & Vitiello, 2017). This is no surprise, as its safety and effectiveness in controlling ADHD
symptoms have been demonstrated by a large body of research. Several meta-analyses summarizing these findings showed that methylphenidate is highly effective in controlling symptoms in children (Faraone & Buitelaar, 2010; Maia et al., 2017; Schachter, Pham, King, Langford, & Moher, 2001), adolescents (Faraone & Buitelaar, 2010; Schachter et al., 2001) and adults with ADHD (Castells et al., 2011).
The high level of effectiveness of methylphenidate in treating ADHD symptoms is remarkable, as ADHD is a disorder with a strongly heterogeneous pattern of clinical manifestation (Wåhlstedt, Thorell, & Bohlin, 2009). For example, symptom expression depends on age and sex (Krain & Castellanos, 2006) as well as on the social and academic setting (Faraone et al., 2015). Recent research has made clear that this heterogeneity of ADHD also extends to its underlying causes. Initially, the general idea of the pathophysiology of ADHD was a deficit in prefrontal-striatal circuits, which are involved in executive functions (Castellanos, Sonuga-Barke, Milham, & Tannock, 2006). This hypothesis was later expanded to include the cerebellum as well (Krain & Castellanos, 2006). Subsequently, this model was tested in a large number of structural and functional imaging studies, yielding divergent results. Although some evidence was found for the prefrontal-striatal model of ADHD, the diversity of findings suggests that this model should be extended to include other networks as well (Castellanos & Proal, 2012; Konrad & Eickhoff, 2010). Hence, the idea that ADHD does not rely on abnormalities within specific brain areas, but rather in network disturbances throughout the entire brain has gained more support in recent years. A recent large-scale meta-analysis intended to settle the ongoing debate on brain alterations in ADHD while addressing heterogeneity and statistical power issues of previous meta-analyses (Samea et al., 2019). Again, their findings were divergent, supporting the notion that ADHD is a dysconnectivity disorder, rather than a disorder that stems from regional impairments. According to Samea et al. (2019), such a distributed network pathology would also explain the impairment on several cognitive and emotional domains that occurs in ADHD.
A neuroimaging method that has significantly contributed to the understanding of ADHD as a dysconnectivity disorder is resting-state functional Magnetic Resonance Imaging (rs-fMRI). Rs-fMRI is a technique in which fMRI is used to examine temporal correlations in neural activity between brain regions at
rest (i.e., in the absence of a specific task or stimulus). This allows the analysis of interactions between brain regions, which makes rs-fMRI is a highly suitable method for uncovering the characteristics of ADHD as a dysconnectivity disorder. In addition, as this non-invasive method does not depend on participants’ abilities to complete specific cognitive tasks, this method is also applicable to clinical and pediatric populations who might normally be limited in their abilities to complete such tasks (Fox & Greicius, 2010). Moreover, the absence of specific tasks allows for easier data sharing across studies, for example for reanalysis or inclusion in pooled datasets. Taken together, these advantages contribute to better reliability of results. Unsurprisingly, both the quantity and statistical power of studies using rs-fMRI to investigate the pathophysiology of ADHD have increased in recent years (Castellanos & Aoki, 2016).
As our understanding of the network pathology of ADHD grows, knowledge on the effects of its most frequently prescribed pharmacological treatment appears to be lagging behind. Although the cellular and regional working mechanisms of methylphenidate are relatively well understood, very little is known about its effects on brain-wide networks. The effects of methylphenidate on resting-state connectivity might have implications for methylphenidate-induced task-related effects, as resting-state properties can predict individual differences in task-related fMRI activity (Mennes et al., 2010, 2011). Moreover, this would also contribute to better understanding of the network pathology of ADHD in general. After all, although neuroimaging studies have contributed significantly to the understanding of ADHD pathology so far, these methods remain correlational and therefore unable to inform on causal mechanisms (Castellanos & Proal, 2012). Studying these mechanisms with a manipulation, such as the administration of methylphenidate, would shed light on the causal mechanisms underlying ADHD. Therefore, this narrative literature review aims to investigate the local and large-scale effects of methylphenidate on rs-fMRI connectivity in ADHD patients in comparison to healthy controls. Firstly, the principle of rs-fMRI will be explained, followed by some frequently used methods to analyze rs-fMRI data. Additionally, a brief overview of recent findings on rs-fMRI connectivity in ADHD will be provided as background information on the network pathology of ADHD. Next, studies examining the effects of methylphenidate on rs-fMRI connectivity in healthy
controls and persons with ADHD will be discussed. Finally, factors that influence the effects of methylphenidate on rs-fMRI connectivity, such as participant demographics and methylphenidate pharmacokinetics, will be discussed. To our knowledge, this is the first article to examine the influence of methylphenidate on resting-state connectivity in ADHD patients.
The methodology of this review was as follows. Studies were selected through an extensive search on PubMed based on the Medical Subject Headings
(MeSH) terms “methylphenidate” and “rest”. Given the limited number of studies on this subject, all studies investigating the influence of methylphenidate on rs-fMRI connectivity in ADHD patients or healthy controls were included, regardless of their methodology, year of publication or number of participants. The selection procedure is displayed in figure 1. This resulted in a total of five studies investigating the effects of methylphenidate in ADHD patients and five studies investigating its effects in healthy controls, which are summarized in tables 1 and 2, respectively.
Figure 1: Study selection procedure
Resting-state fMRI connectivity in ADHD
Resting-state fMRI
Rs-fMRI is a neuroimaging method that examines temporal correlations in spontaneous low-frequency blood oxygenation level dependent (BOLD) signals between brain regions at rest. Although rs-fMRI is a rapidly developing field, the technique itself is not new. The use of fMRI during resting-state was first described by Biswal, Yetkin, Haughton & Hyde (1995), who discovered that anatomically separate brain regions show strong correlations in their low-frequency BOLD signal fluctuations. The relevance of these findings was highlighted by the discovery of the Default Mode Network (DMN), a brain-wide resting-state network
consisting of the precuneus/posterior cingulate cortex (PCC), medial prefrontal cortex (mPFC) and lateral parietal (LP) regions (Fox et al., 2005; Raichle et al., 2001). Remarkable about this network is that it shows greater activity at rest than when engaging in a variety of goal-directed tasks. This “default mode” of brain activity is considered to have wide implications for brain function and has thus become a key focus in neuroscientific research.
The most frequently used technique to analyze resting-state networks such as the DMN is whole-brain regression analysis. In this type of analysis, the
identification of networks in rs-fMRI data can either be based on data-driven independent component analysis (ICA) or on prespecified regions of interest (ROIs) in seed-based correlation analysis (SCA). Both of these methods calculate correlations in BOLD signals between voxels throughout the entire brain, but in ICA all voxels in the brain are considered simultaneously, whereas SCA correlates the BOLD signal from a prespecified seed region with that of all other voxels in the brain to obtain a map of functional connectivity. The latter has the advantage of simpler computations and more intuitive interpretations, but it does require prior assumptions about ROIs, which makes this method more susceptible to bias (Lv et al., 2018).
Besides these methods addressing functional connectivity throughout the entire brain, there are also measures for quantifying local brain functions in rs-fMRI data. Two of these measures can be distinguished: regional homogeneity (ReHo) and the amplitude of low frequency fluctuations (ALFF). Both of these measures give insight into regional brain activity without requiring pre-specified ROIs. The difference between them is in the aspects of brain activity they address: ReHo measures the synchrony between a given voxel and its nearest neighbors, whereas ALFF measures the total power of oscillations in the low-frequency range within a given brain area (Lv et al., 2018; Margulies et al., 2010). As the brain is increasingly viewed from a network perspective rather than as a set of functionally segregated clusters, the interest in these local measures of rs-fMRI connectivity has gradually decreased in favor of whole-brain methods. Nevertheless, ReHo and ALFF are commonly used for defining ROIs for SCA. As these different measures reflect different aspects of brain activity, the best results may be obtained by applying several of them to the same dataset (Lv et al., 2018).
Resting-state fMRI connectivity alterations in
ADHD patients
As ADHD has been increasingly viewed from a network perspective, the number of studies using rs-fMRI to investigate the neurophysiological underpinnings of ADHD is increasing. Although the findings of these studies are still heterogenous, some overarching hypotheses about the pathophysiology of ADHD have emerged (Castellanos & Aoki, 2016).
A particular target in rs-fMRI studies investigating the neurophysiological underpinnings of
ADHD is the DMN. As the DMN has been linked to attention, executive functions and reward processing, all of which are considered to play a role in ADHD, the number of studies investigating the DMN in search of the pathophysiology of ADHD has increased over the past years (Posner, Park, & Wang, 2014). Indeed, a large number of studies demonstrated involvement of the DMN in ADHD pathology, as they report reduced synchrony within the DMN in people with ADHD in comparison to healthy controls (Posner et al., 2014). More specifically, weaker functional connectivity between anterior and posterior nodes of the DMN has been observed in children and adults with ADHD (Castellanos & Aoki, 2016). This pattern of weaker within-DMN connectivity was recently confirmed by two seed-based rs-fMRI meta-analyses in persons with ADHD, which suggest reduced connectivity between DMN seed regions and the mPFC (Gao et al., 2019) and the dPCC (Sutcubasi et al., 2020), two core regions of the DMN.
Besides dysconnectivity within the DMN, a dysfunction in the interplay between the DMN and the cognitive control network (CCN) has also frequently been linked to ADHD psychopathology. The CCN is a fronto-parietal network responsible for top-down regulation of emotion and attention that is activated during goal-directed tasks (Sutcubasi et al., 2020). A leading hypothesis about the interplay between the DMN and the CCN in ADHD is the “DMN interference hypothesis”, which suggests that inattentiveness in ADHD may be caused by an inability to down-regulate DMN activity during goal-directed tasks (Sonuga-Barke & Castellanos, 2007). As a result, goal-directed attention mediated by the CCN may become temporarily disrupted, leading to fluctuations in attentional performance (Castellanos et al., 2008). This hypothesis is supported by a large body of evidence suggesting that the negative correlations between the DMN and the CCN were disrupted in children, adolescents and adults with ADHD compared to healthy controls (Posner et al., 2014; Sutcubasi et al., 2020).
Although disrupted DMN-connectivity is the most widely replicated and robust finding regarding the neurophysiological underpinnings of ADHD (Castellanos & Proal, 2012; Sutcubasi et al., 2020), a significant amount of evidence suggests that the pathology of ADHD is not limited to the DMN and more complex models of have emerged. A model that has recently gained evidence is the triple network model, which suggests that many psychiatric conditions,
including ADHD, are characterized by deficits in the engagement and disengagement of three neurocognitive networks: the DMN, the CCN and the salience network (SN; Menon, 2011). The SN consists of the dorsal anterior cingulate cortex (dACC) and the anterior insula (AI), and is involved in detecting and filtering salient stimuli and engaging relevant functional networks to react accordingly (Seeley et al., 2007). Evidence for the involvement of the SN in the pathophysiology of ADHD was found by multiple studies reporting decreased negative connectivity between the DMN and the dACC in persons with ADHD (Castellanos & Aoki, 2016; Sutcubasi et al., 2020). Additionally, the triple network model is supported by findings from a recent meta-analysis of resting-state fMRI studies in ADHD, reporting imbalanced connectivity between the CCN and the DMN and SN (Gao et al., 2019).
In addition to supporting the triple network model, Gao et al. (2019) suggest extending this model by underscoring the involvement of the affective-motivational network (AMN) and the somatosensory network (SSN). The AMN is involved in processing the valence of emotional stimuli, and consists of the ventral striatum, amygdala and ACC (Etkin, Egner, & Kalisch, 2011). Its involvement in ADHD pathology was also suggested by two earlier reviews that found altered connectivity in reward-related and affective circuits in ADHD (Castellanos & Aoki, 2016; Posner et al., 2014). A similar finding was observed in a more recent meta-analysis, which found reduced connectivity between the DMN and the AMN in children with ADHD (Sutcubasi et al., 2020). These disruptions in AMN connectivity are suggested to be involved in ADHD-related symptoms of hyperactivity (Castellanos & Aoki, 2016), impulsivity (Castellanos & Aoki, 2016; Gao et al., 2019) and emotional dysregulation (Gao et al., 2019).
All in all, resting-state fMRI has contributed greatly to our understanding of the neurophysiological underpinnings of ADHD and has revealed a series of networks involved. Although findings are somewhat divergent, a growing body of evidence suggests altered connectivity of the DMN, CCN, SN and AMN in persons with ADHD in comparison to healthy controls.
The influence of methylphenidate on resting-state
fMRI connectivity
The influence of methylphenidate on resting-state
fMRI connectivity in ADHD patients
The first study to investigate the influence of methylphenidate on rs-fMRI connectivity in ADHD patients suggested that methylphenidate might cause a more control-like state. Using 3T fMRI, An et al. (2013) demonstrated increased ReHo in bilateral sensorimotor and parieto-visual cortices and decreased ReHo in bilateral DLPFC in 23 boys with ADHD under placebo relative to controls, whereas these differences were no longer observed one hour after administration of a single 10 mg dose of methylphenidate. Additionally, within-patient comparisons demonstrated that methylphenidate significantly increased ReHo in the left inferior frontal cortex (IFC), right orbital frontal cortex (OFC) and the cerebellum and decreased ReHo in the right parietal and visual cortex in comparison to placebo. These findings suggest that methylphenidate may promote symptom reduction in ADHD by acutely normalizing regional activity as well as functional connectivity within the fronto-parietal cerebellar circuit.
Further supporting this notion, Silk, Malpas, Vance, & Bellgrove (2016) demonstrated that a single dose of methylphenidate normalized resting-state fMRI connectivity in a large-scale network in 16 male adolescents with ADHD using a relatively new method: the network-based statistic (NBS). This statistical approach identifies connections from a connectivity matrix while dealing with the multiple comparisons problem (Zalesky, Fornito, & Bullmore, 2010). Using NBS, a significant increase in resting-state connectivity in a network primarily consisting of occipital, temporal and cerebellar regions was found in ADHD patients under placebo in comparison to healthy controls, whereas these differences were no longer observed 90 minutes after administration of a single 20 mg dose of methylphenidate. The same was found for resting-state connectivity between visual, executive and default mode networks. Similar to An et al. (2013), these findings suggest that methylphenidate normalizes increased resting-state connectivity in visual and parietal regions in young ADHD patients. As An et al. (2013) used ReHo, a local method to analyze resting-state connectivity, Silk et al. (2017) extended these findings by using a network-based approach that examines global connectivity changes. This difference in methodology is important to consider when interpreting the findings of
cerebellar involvement. An et al. (2013) observed methylphenidate-induced upregulation of within-cerebellar activity, whereas Silk et al. (2017) observed methylphenidate-induced downregulation of connectivity between the cerebellum and other brain areas. These findings might seem contradictory, but essentially, they point out differences in local and large-scale effects of methylphenidate on resting-state connectivity.
In line with this idea, Yoo, Kim, Choi, & Jeong (2018) specifically explored both local and large-scale effects of continuous 12-week methylphenidate treatment in 20 children (boys and girls) with ADHD
using ALFF and ICA, respectively. Local analyses demonstrated a methylphenidate-induced decrease in ALFF in the superior parietal cortex (SPC), and an increase in ALFF in the right inferior fronto-temporal area, which is in accordance with previous findings of increased ReHo in the OFC and IFC and decreased ReHo in the SPC (An et al., 2013). On a large-scale, a treatment-induced increase in connectivity was observed between the left fronto-parietal network and the left insular cortex and the right fronto-parietal network and the brainstem. Additionally, a reduced clustering coefficient was found in the visual and left fronto-parietal network and the cerebellum, reflecting reduced
Table 1: Studies investigating the effects of methylphenidate on rs-fMRI connectivity in ADHD patients.
Study Sample Methods Medication status Results
An et al. (2013)
n = 55 with 23 ADHD boys and 32 controls.
Age 8–15 years.
ReHo: voxel-wise analysis, 3T MRI
Medication discontinued 1 month before scan,
scanning 1h after 10mg dose of immediate-release MPH/placebo
ReHo:
sensorimotor, parieto-visual cortices, left IFC, right OFC, cerebellum ↑
bilateral DLPFC, right parietal and visual ↓ Silk et al. (2016) n = 36 ADHD males. Age 9 – 18 years. NBS, 3T MRI Medication naïve (n = 10) or withdrawn for at least 48h (n = 6),
scanning 1.5h after 20mg dose of MPH/placebo.
connectivity between occipital, temporal and cerebellar regions and between visual, executive and default mode networks ↓
Yoo et al. (2018)
n = 47 with 16 ADHD boys, 4 ADHD girls, 19 control boys and 8 control girls.
Age 7 – 16 years.
ALFF and fALFF, ICA with dual regression, graph theoretical analysis, 1.5T MRI
Medication-naïve, followed by 12-week extended release MPH treatment. Doses adjusted to 18-72mg according to symptom improvement, scanning 2h after last dose.
ALFF:
SPC ↓ right inferior fronto-temporal area ↑
Cary et al. (2017)
n = 22 ADHD patients with 14 females.
Age 18 – 35 years.
NDI, 3T MRI
Treated with methylphenidate
(n = 7) or (dextro)amphetamine (n = 15) at prescribed dosage.
One scan 60-180min after
administration, one scan after 48-72h abstention (7 half-life washout).
Altered organization of resting-state networks, including DMN and FPN.
Picon et al. (2020)
n = 18 ADHD males, mean age 30.67 years (SD = 7.41).
Seed-based analysis with ROIs: MPFC, PCC, RLP, and LLP, 1.5T MRI
Medication-naïve at baseline, followed by 12-month treatment with immediate release MPH, beginning with 5mg once a day, increased according to symptom improvement and side effects. Scanning 72h after last dose.
Connectivity PCC – LLP ↑
Abbreviations: ADHD = Attention-deficit/hyperactivity disorder; MPH = methylphenidate; ReHo = regional homogeneity; NBS = network-based statistic; (f)ALFF = (fractional) amplitude of low frequency fluctuations; ICA = independent component analysis; NDI = node dissociation index; ROIs = regions of interest; MPFC = medial prefrontal cortex, PCC = posterior cingulate cortex; RLP = right lateral parietal cortex, LLP = left lateral parietal cortex, IFC = inferior frontal cortex, OFC = orbitofrontal cortex, DLPFC = dorsolateral prefrontal cortex, SPC = superior parietal cortex, DMN = default mode network, FPN = frontoparietal network.
connectivity within these areas. Again, these findings support findings from previous studies suggesting that methylphenidate treatment normalizes resting-state connectivity in fronto-parietal-cerebellar circuits. This replication of previous findings is particularly valuable as this was the first study to specifically recruit drug-naïve subjects, thus excluding the possibly confounding role of prior psychotropic medication use. A limitation to this study, however, is that most analyses targeted cortical structures, leaving the effects of methylphenidate on subcortical structures such as the striatum largely unexplored.
In sum, four studies using different methods reported analogous findings regarding the effects of methylphenidate on resting-state fMRI connectivity in children with ADHD, suggesting a normalizing effect on fronto-parietal-cerebellar networks (An et al., 2013; Silk et al., 2016). These studies were all based on children with ADHD. Recently, preliminary evidence suggested that psychostimulants might also have a normalizing effect on resting-state connectivity in adults with ADHD. Cary et al. (2017) observed differences in the organization of various resting-state networks including the DMN and fronto-parietal network in 22 ADHD patients compared to controls that were significantly reduced by dextroamphetamine/amphetamine (n = 15) or methylphenidate (n = 7) treatment. These findings suggest that psychostimulants cause a more control-like state in these resting-state networks, although this study did not take prior medication use into account. Subsequently, this notion was supported by a study specifically investigating the effects of methylphenidate on within-DMN resting-state fMRI connectivity in drug-naïve ADHD patients (Picon et al., 2020). In this study, 18 males were scanned once before and once after 72 hours of treatment. The results demonstrated methylphenidate-induced strengthening of functional connectivity between the PCC and left lateral parietal cortex, two nodes of the DMN. The homogenous and relatively small sample used in this study might limit generalizability of these results, but together with findings from Cary et al. (2017) they can be considered preliminary evidence that the normalizing effects of methylphenidate on resting-state networks in ADHD patients also apply to adults.
All in all, an emerging body of research has provided insight into the local and brain-wide effects of methylphenidate on resting-state connectivity in ADHD patients. Using different methods, including ReHo, NBS, ALFF and ICA, three separate studies found
similar results, suggesting that methylphenidate might facilitate symptom reduction by normalizing resting-state connectivity in fronto-parietal-cerebellar circuits in children with ADHD (An et al., 2013; Silk et al., 2016; Yoo et al., 2018). Two recent studies have provided preliminary evidence that these normalizing effects might also apply to adults with ADHD (Cary et al., 2017; Picon et al., 2020).
The influence of methylphenidate on resting-state
fMRI connectivity in healthy controls
In addition to studies investigating the effects of methylphenidate in ADHD patients, the number of studies investigating its effects in healthy controls is growing. These studies can make a valuable contribution to understanding the effects of methylphenidate on rs-fMRI connectivity, as samples of healthy controls generally contain less confounding factors, yielding more reliable results (Zhu et al., 2013).
One of the first studies to explore this demonstrated increased ReHo in the left middle and superior temporal gyri and reduced ReHo in the left lingual gyrus 1 hour after administration of a 20 mg dose of methylphenidate compared to placebo in 18 adult males (Zhu et al., 2013). Previous research has related abnormalities in these areas to ADHD pathology, so it is likely that methylphenidate-induced modulation of these areas may induce symptom improvement in ADHD. However, no notable ReHo differences were found in the basal ganglia and prefrontal cortex, two other areas frequently related to ADHD pathology. According to the authors, the absence of these findings might be due to the relatively slight signal in the BG or the threshold of the statistics. Additionally, the 20 mg dose used in this study is a relatively low dose for adults, which might limit its effectiveness. Interestingly, no significant behavioral improvement on a Go/No-go task carried out after the resting-state scan was observed either, suggesting that either the dose of methylphenidate was too low to cause significant improvement, or its effects might be different in healthy controls compared to ADHD patients.
A subsequent study using a higher dose of methylphenidate (45mg) and a slightly larger sample size (16 females, 8 males) suggested that the latter might be the case, as they also observed methylphenidate-induced effects in healthy adults that were in contrast to ADHD findings (Farr et al., 2014). This study conducted a seed-based analysis with the thalamus and dorsal
striatum as seed ROIs because these areas are part of catecholaminergic systems, the neurotransmitter systems targeted by methylphenidate. Findings demonstrated methylphenidate-induced increase in thalamic/striatal connectivity to the hippocampus, amygdala and primary motor cortex, whereas connectivity to frontal executive regions was decreased. This is in contrast to findings for clinical populations, in whom methylphenidate has been shown to increase connectivity to frontal executive areas. Farr et al. (2014) speculate that this might reflect catecholaminergic signaling exceeding the optimal level, causing suboptimal cognitive performance. However, as cognitive performance itself was not assessed in this study, these functional implications remain speculative. Additionally, it should be noted that the resting scan was conducted 40 minutes after administration of methylphenidate, which is relatively short given that
plasma levels usually peak around 90 minutes after administration (Müller et al., 2005). Nevertheless, taken together with findings from Zhu et al. (2013), these findings suggest that the effects of methylphenidate might be different in healthy controls compared to ADHD patients.
Expanding upon this work, another seed-based study demonstrated diverse methylphenidate-induced effects in other catecholaminergic areas: the locus coeruleus (LC), basal nucleus of Meynert (BNM) and ventral tegmental area (VTA)/substantia nigra pars compacta (SNc; Kline et al., 2016). For the BNM, negative connectivity with precentral gyri, including the primary motor cortex, was reversed. For the LC, positive connectivity to the cerebellum was reduced and positive connectivity to the hippocampus was induced. For the VTA/SNc, positive connectivity to the cerebellum and
Table 2: Studies investigating the effects of methylphenidate on rs-fMRI connectivity in healthy controls.
Authors Sample Methods Medication status Results
Zhu et al. (2013)
n = 18 healthy males. Age 19 – 24 years.
ReHo, 1.5T MRI Scanning 1h after 20mg dose MPH
ReHo:
left middle + superior temporal gyrus ↑
left lingual gyrus ↓ Farr et al.
(2014)
n = 48 healthy adults with 32 females. Age 20 – 28 years.
Global connectivity analysis and seed-based analysis with ROIs: right and left caudate, putamen, pallidum and thalamus, 3T MRI
Scanning 40min after 45mg dose MPH/ placebo
Seed-based connectivity thalamus/ striatum to:
hippocampus, amygdala, M1 ↑ frontal executive regions ↓
Kline et al. (2016)
n = 24 healthy adults with 16 females. Age 19 – 31 years.
Seed-based analysis with ROIs: BNM, LC and VTA/SNc,
3T MRI
45mg dose MPH. Scanning time after administration unknown.
Connectivity:
BNM - precentral gyri ↓ LC- cerebellum ↓ LC - hippocampus ↑
VTA/SNc – cerebellum, putamen, left occipital gyrus ↓ Sripada et al. (2013) n = 32 healthy adults with 16 females. Age 18 – 27 years. Connectomic imaging coupled with multivariate pattern classification
Scanning 1h after 40mg dose MPH/ placebo.
Coupling within somatosensory and visual network ↓ Coupling DMN - task-positive networks ↓ Mueller et al. (2014) n = 54 healthy males. Mean age 23.65 years, sd = 2.97.
ICA Scanning 1h after 40mg dose MPH/ placebo.
Connectivity: DAN – thalamus↑
FPN/ECN – sensory-motor/ visual cortex ↑
FPN/ECN – CSTs ↓
Sensory-motor areas – CSTs ↑ Sensory-motor areas ↓
Abbreviations: ADHD = Attention-deficit/hyperactivity disorder; MPH = methylphenidate; ReHo = regional homogeneity; ROIs = regions of interest; BNM =basal nucleus of Meynert; LC = Locus Coeruleus; VTA/SNc = ventral tegmental area/substantia nigra, pars compacta; ICA = independent component analysis.
putamen and negative connectivity to the left occipital gyrus was reduced. These findings support the finding of increased connectivity to the primary motor cortex and hippocampus by Farr et al. (2014) and extend them by suggesting alterations in connectivity to the cerebellum, putamen and occipital gyrus in healthy adults. An important note to these findings, however, is that this study did not contain a placebo control, limiting its reliability.
In sum, three studies using seed-based analyses or ReHo have elucidated specific areas throughout the brain where methylphenidate affects rs-fMRI connectivity in healthy adults. In addition to these findings, some larger-scale, data-driven studies have also demonstrated the effects of methylphenidate within and between several resting-state networks. Using connectomic imaging coupled with multivariate pattern classification, a relatively new approach that aims to maximize computational feasibility while minimizing the restrictions of pre-specified ROIs, Sripada et al. (2013) examined the effects of a single 40mg dose of methylphenidate 1 hour after administration in 32 healthy adults (16 females). Their findings demonstrate reduced coupling within the somatosensory and visual network and between the DMN and task-positive networks (the attention and sensory-motor network) after methylphenidate administration compared to placebo. Further extending these findings, Mueller et al. (2014) demonstrated connectivity changes in various resting-state networks, including attention, sensory-motor and visual networks, one hour after administration of a single 40mg dose of methylphenidate relative to placebo in 44 healthy adults using ICA. Their findings suggest increased connectivity strength between the dorsal attention network and the thalamus. Additionally, connectivity from resting-state networks located in association cortices (the fronto-parietal network and executive control network) to sensory-motor and visual cortex regions was increased and connectivity to cortico-striato-thalamo-cortical circuits (CSTs, consisting of the thalamus, striatum and cerebellar components) was decreased. These effects were opposite for resting-state networks located in sensory-motor areas (somatosensory and visual network). Taken together, these findings suggest a methylphenidate-induced shift in connectivity: Sripada et al. (2013) suggest reduced coupling between the DMN and sensory-motor and attention networks, whereas Mueller et al. (2014) suggest increased coupling between the sensory-motor network and the frontoparietal and executive network and between the
attention network and the thalamus. Remarkable, however, is that despite their highly similar methodology, Mueller et al. (2014) were unable to replicate findings of altered DMN connectivity as described by Sripada et al. (2013). A possible explanation for this is a slightly different definition of the DMN in these studies, as Mueller et al. (2014) defined resting-state networks by visually comparing their findings to existing maps, whereas Sripada et al. (2013)defined networks according to a template prior to their measurements. Although the effects on the DMN require further investigation, these studies clearly demonstrate a reliably detectable methylphenidate-induced neural signature throughout several resting-state networks in healthy adults.
All in all, a growing number of studies has demonstrated divergent effects of methylphenidate on rs-fMRI connectivity in healthy adults. ReHo and seed-based studies have revealed specific brain areas where alterations in connectivity occur, including the temporal and lingual gyrus (Zhu et al., 2013), thalamus, striatum (Farr et al., 2014), LC, BNM and VTA/SNc (Kline et al., 2016). Alterations in large-scale resting-state networks have also been demonstrated, including the attention, sensory-motor, visual and default mode network (Sripada et al., 2013) and the fronto-parietal and executive control network and CSTs (Mueller et al., 2014).
Comparison of the effects of methylphenidate on
resting-state fMRI connectivity in ADHD patients
and healthy controls
Brain areas displaying methylphenidate-induced alterations in rs-fMRI connectivity in ADHD patients and healthy controls are displayed in figure 2. On a brain-wide scale, the effects in these populations seem analogous. Methylphenidate produces a clear neural signature across multiple resting-state networks in both ADHD patients and healthy controls, including the frontoparietal network and cerebellar circuits. When comparing these effects on a more specific level, however, some notable differences can be observed. In ADHD patients, findings are generally consistent, suggesting altered connectivity within fronto-parietal-cerebellar circuits. In healthy subjects, findings are much more divergent, suggesting specific effects on distinct brain areas, including multiple subcortical areas, as well as effects on large-scale resting-state networks,
Figure 2: Brain areas displaying methylphenidate-induced alterations in resting-state fMRI connectivity in ADHD patients and healthy controls. (A) Fronto-parietal-cerebellar networks. (B) Thalamus, striatum, locus coeruleus, basal nucleus of Meynert, ventral tegmental area/substantia nigra pars compacta, attention network, sensory-motor network, visual network, default mode network, fronto-parietal network, executive control network and cortico-striato-thalamo-cortical circuits. Not visible in this image, but also involved are lingual and temporal gyrus.
including the attention, sensory-motor, visual and executive control network.
These different findings can be partially explained by the fact that current studies in ADHD patients and healthy controls are focused on different brain areas. Studies investigating the effects of methylphenidate in ADHD patients almost exclusively investigated cortical areas associated with ADHD pathology, such as the prefrontal cortex, whereas studies in healthy controls mainly examined subcortical areas associated with catecholaminergic signaling, such as the basal ganglia. This is important to keep in mind, as it complicates the comparison of the effects of methylphenidate between these populations.
Nonetheless, studies employing a data-driven approach to investigate the effects of methylphenidate still yielded different results: in ADHD patients, these studies mainly reported alterations in fronto-parietal-cerebellar circuitry, whereas studies in healthy controls
reported alterations in several resting-state networks, including sensory-motor, attention and visual networks and CSTs.
These differences raise the question whether the effects of methylphenidate on rs-fMRI connectivity might depend on ‘baseline’ rs-fMRI connectivity (or connectivity in an unmedicated state), which would explain why they differ between ADHD patients and healthy controls. This rate-dependency hypothesis has been proposed in earlier research, and suggests that the effects of methylphenidate are inversely correlated to the basal rates of brain activity or behavior (Andersen, 2005). In other words, it decreases activity that normally occurs at high rates and increases activity that normally occurs at low rates. As most studies investigating the effects of methylphenidate in ADHD patients suggest “normalization” of resting-state connectivity, indicating a connectivity pattern more comparable to healthy controls, it seems reasonable to expect that the effects of
methylphenidate depend on how strongly the resting-state connectivity differs from healthy controls at baseline. This is supported by two studies in healthy controls that reported results contrasting to what would be expected based on findings in ADHD patients (Farr et al., 2014; Zhu et al., 2013).
Interestingly, the idea of psychostimulants normalizing altered rs-fMRI connectivity has also been suggested for amphetamine, another commonly prescribed stimulant for ADHD that inhibits the reuptake of dopamine and noradrenaline. Using a data-driven approach, Yang et al. (2016) demonstrated that 3-week amphetamine treatment normalized abnormally decreased functional segregation between the CCN, DMN and SN in adults with ADHD. Importantly, this effect was directly associated with symptom reduction, suggesting that the normalization of resting-state connectivity may underlie the clinical benefits of psychostimulants. Additionally, such a clinical improvement facilitated by normalized rs-fMRI connectivity has also been suggested in a different clinical population. A seed-based analysis by Konova, Moeller, Tomasi, Volkow, & Goldstein (2013) demonstrated that 120 minutes after intake of a single 20mg dose of methylphenidate, abnormally strong connectivity between the ventral and dorsal striatum was reduced in 18 cocaine addicts compared to placebo, whereas several corticolimbic and cortico-cortical connections were strengthened. This resulted in a connectivity pattern resembling that of healthy controls, a connectivity alteration correlated with less severe addiction. Taken together, these findings suggest that psychostimulants normalize altered rs-fMRI connectivity in clinical populations.
Although these findings in clinical populations clarify how methylphenidate might facilitate symptom reduction, studies in healthy controls have elucidated other aspects of methylphenidate-related effects. Firstly, as these studies were mainly focused on catecholaminergic systems, they have demonstrated clear methylphenidate-induced effects in several subcortical areas, whereas in ADHD patients, this has not been investigated yet. Secondly, multiple studies in healthy controls demonstrated a clear methylphenidate-induced neural signature in adults, whereas ADHD research has mainly investigated effects in children. The difference in populations included in these studies is most likely due to ethical concerns limiting the possibility to study the effects of methylphenidate in healthy children. Nevertheless, these findings provide a
good basis for further research into methylphenidate-induced effects in adults with ADHD and in subcortical areas in ADHD patients.
Factors that influence the effects of
methylphenidate on resting-state connectivity
Participant
demographics
An extensive body of research has made clear that the effects of methylphenidate on rs-fMRI connectivity are complex and heterogenous. In order to provide some understanding of the heterogeneity of these effects, some confounding factors need to be considered. One of these factors is the age of participants, as the dysconnectivity pattern found in ADHD might be age-dependent. This was suggested by Mostert et al. (2016), whose rs-fMRI findings in adults with ADHD were similar to those in children with ADHD, but with smaller effect sizes than initially thought, despite using a relatively large sample size. This was supported by a recent large-scale meta-analysis by Sutcubasi et al. (2020), who reported more apparent and extensive alterations in resting-state connectivity in ADHD patients when their analyses were restricted to studies including children than when they also included adults. For example, they found reduced connectivity within core nodes of the DMN in a sample of ADHD patients of all ages, but when excluding adult studies this dysconnectivity also included the medial temporal lobe (MTL) and PFC. These observations are in line with models of maturational delays in network connectivity in ADHD, which are supported by rs-fMRI as well as structural imaging studies (Posner et al., 2014). All in all, baseline connectivity patterns might differ between children and adults with ADHD, so it is important to take this into account when interpreting the effects of methylphenidate on resting-state connectivity. In addition, the catecholaminergic system targeted by methylphenidate is also in continuous development across childhood (Andersen, 2005), and so are other parts of the brain. As different brain regions each have individual developmental processes, this leads to a complex interplay of brain regions with varying degrees of sensitivity to methylphenidate as people get older (Rapoport & Gogtay, 2008), likely causing age-related differences in methylphenidate-induced effects. For example, a task-based study demonstrated increased fronto-striatal and cerebellar activation in children after administration of methylphenidate, whereas increased striatal activation and decreased frontoparietal activation was demonstrated in adults (Epstein et al., 2007). More
recently, Schrantee et al. (2017) investigated these age-related differences by directly comparing the effects of a single dose of methylphenidate on children and adults with ADHD. In accordance with Epstein et al. (2007), their findings suggest age-related differences in the effects of methylphenidate on cerebral blood flow (CBF). In both populations, decreased CBF was observed in cortical areas, but this effect was stronger in adults. Additionally, a methylphenidate-induced decrease in thalamic CBF was observed in children, but not in adults. These findings underscore the importance of taking age into account when interpreting methylphenidate-induced effects.
Another participant-related factor important to keep in mind is sex. Generally, males are more likely to be diagnosed with ADHD than females and they are included in ADHD research more frequently (Slobodin & Davidovitch, 2019). However, the generalizability of these studies might be limited, as sex differences have been reported in the clinical manifestation of ADHD (Cortese, Faraone, Bernardi, Wang, & Blanco, 2016) and in functional connectivity in the brain in general (Ingalhalikar et al., 2014; Satterthwaite et al., 2015). Recently, sex-related differences in functional connectivity have also been suggested for ADHD patients specifically, as disruptions in fronto-subcortical resting-state connectivity appeared stronger in girls than in boys with ADHD (Rosch, Mostofsky, & Nebel, 2018). Although the exact sex-related differences in clinical manifestation of ADHD are still poorly understood, they are important to keep in mind when comparing results across studies, as the effects of methylphenidate might also differ between males and females.
Heterogeneity of ADHD
As ADHD is a disorder with a highly heterogenous pattern of clinical manifestation, it is important to note that the effects of methylphenidate in ADHD patients might also differ between individuals. Based on the core symptoms of ADHD, DSM-5 describes three presentations: mainly inattentive, mainly hyperactive/impulsive and a combined presentation. This heterogeneity appears to be reflected in the underlying pathophysiology as well, as distinct resting-state connectivity patterns have been demonstrated for the inattentive and hyperactive/impulsive presentations (Fair et al., 2013). Some studies investigating resting-state connectivity in ADHD have taken these different
manifestations into account by performing dimensional analyses with symptom severity instead of categorical comparisons between ADHD patients and healthy controls. Using such an approach, Mostert et al. (2016) demonstrated a positive association between inattention symptoms and connectivity in the executive control network. These differences in baseline connectivity might affect findings of methylphenidate-induced effects. However, current studies into the effects of methylphenidate on resting-state connectivity in ADHD have remained categorical, leaving the heterogenous manifestation of ADHD a confounding factor.
Prior exposure, dose and pharmacokinetic profile
of methylphenidate
Another factor that might seriously affect the influence of methylphenidate on resting-state connectivity is prior exposure. A growing body of research suggests that stimulants like methylphenidate might cause neuronal imprinting, a process in which drug exposure causes developmental alterations during specific sensitive stages in maturation (Andersen, 2005). This is especially relevant since methylphenidate is often prescribed from a young age. Animal studies suggest that the age at initiation of methylphenidate treatment can influence its effects on development in a highly specific manner (Canese et al., 2009). Such age-dependent effects of methylphenidate have also been suggested in humans, where a significant increase in CBF was found in the striatum and thalamus after 16-week treatment in children, but not in adults (Schrantee et al., 2016). Thus, findings of methylphenidate-induced effects might be different for ADHD patients who have a history of methylphenidate treatment compared to those who have never been exposed to psychostimulants before. Unfortunately, most studies discussed in the current review did not take this into account.
In addition, the dose of methylphenidate might influence its effects. Previous research suggested average effective doses of 0.7-0.9 mg/kg per day for children (Vitiello et al., 2007) and 1.1 mg/kg per day for adults (Spencer et al., 2005). As immediate-release formulations of methylphenidate are eliminated relatively quickly, they are usually given twice a day, corresponding to 0.35 – 0.55 mg/kg per dose. Keeping these effective doses in mind, the 40-45 mg doses used in most of the reviewed studies seem reasonable for adults, but relatively high for children. This might lead
to different results, as previous research suggested that the effects of methylphenidate might be dose-dependent (see, for example, Linssen, Sambeth, Vuurman, & Riedel, 2014; Stein et al., 2003). Additionally, as most of these studies used standardized doses, rather than adjusting for body weight, this might have caused individual differences in responsiveness.
Finally, the fraction of the administered methylphenidate dose that reaches the systemic circulation, also known as bioavailability, might impact its effects. Although it is impossible to control for this entirely, as the rate of absorption also depends on individual differences and diet, it is important to keep in mind that a number of different formulations of methylphenidate exist that each have different pharmacokinetic profiles (Cortese et al., 2017). This may lead to different absorption rates, which is important to take into account when determining the time between methylphenidate administration and scanning session. Remarkably, multiple studies reviewed in the current paper conducted fMRI scanning approximately 60 minutes after administration of a single dose of methylphenidate, although previous research suggested that peak plasma concentrations are attained at approximately 90 minutes after administration (Müller et al., 2005). Quite possibly, these studies measured the effects of methylphenidate before peak effects were reached, which could have led to less significant results. Therefore, the pharmacokinetic profile and time after administration of methylphenidate may underlie variability of the effects found in different studies.
Discussion
The aim of this review was to provide a state-of-the-art overview of the effects of methylphenidate on rs-fMRI connectivity in ADHD patients in comparison to healthy controls. Findings of resting-state connectivity in ADHD are still somewhat divergent, but overarching hypotheses suggest altered connectivity of the DMN, CCN, SN and AMN in persons with ADHD in comparison to healthy controls. Although the number of studies investigating the effects of methylphenidate on these altered connectivity patterns is still limited, their findings are largely analogous, suggesting normalization of fronto-parietal-cerebellar resting-state networks. In light of ADHD physiopathology, this is not surprising: abnormalities in the cerebellum and the CCN, a frontoparietal network, have been related to ADHD.
These findings suggest that normalization of these fronto-parietal-cerebellar abnormalities facilitates symptom reduction in ADHD. Although this notion is mainly based on findings in children, preliminary evidence suggested similar normalization in adults with ADHD. Furthermore, studies in healthy adults revealed that the effects of methylphenidate are not limited to normalizing altered connectivity patterns, as they also demonstrated specific effects on resting-state connectivity in these populations. These findings are more divergent than those in ADHD patients, suggesting methylphenidate-induced alterations within temporal and lingual gyri and several subcortical areas as well as altered connectivity in several resting-state networks, including the DMN, sensory-motor, visual, frontoparietal and executive control network and CSTs. A major concern in this field of research is the vulnerability of fMRI to head motion. Generally, head motion occurs at frequencies similar to the BOLD signal, increasing the chance of false positives. This is especially a concern in rs-fMRI due to the lack of task-related temporal structure of the signal (Castellanos & Aoki, 2016). Moreover, head motion is a particularly challenging issue in participants with ADHD, a disorder characterized by hyperactivity. Fortunately, awareness of this issue has grown over the past years and studies investigating rs-fMRI in ADHD patients have incorporated different approaches to control for head motion. In addition to “scrubbing”, the removal of volumes affected by head motion, the application of scanning techniques (e.g. the use of fitted pillows and taping the head) as well as post-processing techniques has increased. With regard to post-processing techniques, many rs-fMRI studies in ADHD patients have incorporated motion parameters to covary for head motion. A frequently used technique for this is the component-based noise correction method (CompCor), a regression strategy that uses signal components from white matter and cerebrospinal fluid to estimate noise components (Behzadi, Restom, Liau, & Liu, 2007). However, as head motion may be inherent to ADHD, such approaches come with the risk of covarying for the effects of ADHD itself (Posner et al., 2014). This is supported by research showing that head motion is not simply a random factor, but it is correlated with impulsivity ratings (Kong et al., 2014). An analytic technique to control for head motion that does not reduce sensitivity towards effects of interest is ICA-based Automatic Removal of Motion Artifacts (ICA-AROMA), a data-driven approach for removing
motion-related signal components. This method was shown to improve reproducibility of resting-state networks compared to other motion correction approaches in several datasets, including a sample of ADHD patients (Pruim, Mennes, Buitelaar, & Beckmann, 2015). This might provide a valuable alternative to regression strategies like CompCor, but has not been applied to rs-fMRI research into the effects of methylphenidate so far. Head motion remains an issue for this type of studies and the interpretation of their results requires caution, as it may seriously affect data quality.
A valuable development is the initiative of large-scale open datasets. An example of this is the ADHD-200 dataset, consisting of rs-fMRI data of 285 children and adolescents with ADHD and 491 age-matched controls (Milham, Damien, Maarten, & Stewart, 2012). A similar data-sharing initiative is the ENIGMA-ADHD consortium, consisting of MRI data of over 4000 subjects, including children and adults with ADHD and healthy controls. This initiative aims to include multiple imaging modalities, including rs-fMRI data (Hoogman et al., 2020). Both of these data sharing initiatives aim to overcome the shortcomings of small sample sizes and heterogeneity of methods used in ADHD research. The need for such data sharing is highlighted by a recent meta-analysis underscoring the heterogeneity of rs-fMRI findings in ADHD (Cortese, Aoki, Itahashi, Castellanos, & Eickhoff, 2020). These authors suggest that the lack of convergent findings in this field of research might be due to heterogeneity in terms of participant characteristics or manifestation of ADHD or due to individual differences in connectivity patterns. Large-scale shared datasets might provide sufficient statistical power to clarify this.
In addition to these developments, this review has underscored some gaps in current literature that should be addressed in future research into the effects of methylphenidate on resting-state connectivity in ADHD patients. Firstly, some evidence was found for the rate-dependency hypothesis of methylphenidate, suggesting that its effects might depend on baseline levels of resting-state connectivity. In order to investigate this, further research should implement dimensional approaches examining symptom severity, rather than categorically comparing often clinically heterogenous samples of ADHD patients to healthy controls. Secondly, most studies investigating the effects of methylphenidate on rs-fMRI connectivity did not take
into account prior medication use, although this might seriously affect their findings. In order to provide more reliable insight into the acute effects of methylphenidate in the future, inclusion of medication-naïve participants is highly recommended. Thirdly, studies in healthy controls have clearly demonstrated methylphenidate-induced effects on resting-state connectivity in several subcortical areas, but the subcortex has remained largely unexplored in studies investigating the effects of methylphenidate in ADHD patients. This is unfortunate, as subcortical alterations have been related to ADHD pathology and may underly cortical effects of methylphenidate, as methylphenidate is known to act on catecholaminergic systems located in the subcortex. Additionally, task-based fMRI studies have demonstrated significant methylphenidate-induced effects in fronto-striatal areas in ADHD patients (Czerniak et al., 2013; Epstein et al., 2007), suggesting that these effects might also be present at rest. Nevertheless, very little resting-state studies have looked into the effects of methylphenidate on subcortical areas such as the striatum yet. Therefore, the subcortex should be a target for further research into the effects of methylphenidate in ADHD patients. Preferably, this research should be done at high field strengths, as subcortical areas are generally small, directly adjunct and more distant from MRI head coils than cortical structures, making it more difficult to distinguish between them and visualize their activity at lower field strengths (Keuken, Isaacs, Trampel, van der Zwaag, & Forstmann, 2018). Finally, research in healthy subjects has clearly demonstrated methylphenidate-induced effects in adults, but studies in ADHD patients have mainly been focused on children. As the effects of methylphenidate might differ significantly between children and adults, this also remains a topic for further research.
All in all, our findings reveal possible causal mechanisms by which alterations in resting-state connectivity might facilitate symptoms in ADHD. Given the high societal relevance of these findings, further research into these mechanisms is necessary. The use of suitable methods for motion correction, large-scale data-sharing initiatives, dimensional analyses, inclusion of medication-naïve participants and a focus on subcortical effects and adult populations seem the way forward in this growing field of research.
Literature
Agarwal, R., Goldenberg, M., Perry, R., & Ishak, W. W. (2012). The quality of life of adults with attention deficit hyperactivity disorder: A systematic review. Innovations in Clinical Neuroscience, 9(5–6), 10–21.
An, L., Cao, X. H., Cao, Q. J., Sun, L., Yang, L., Zou, Q. H., … Wang, Y. F. (2013). Methylphenidate normalizes resting-state brain dysfunction in boys with attention deficit hyperactivity disorder.
Neuropsychopharmacology, 38(7), 1287–1295.
Andersen, S. L. (2005). Stimulants and the developing brain. Trends in Pharmacological Sciences, 26(5), 237– 243.
Behzadi, Y., Restom, K., Liau, J., & Liu, T. T. (2007). A component based noise correction method (CompCor) for BOLD and perfusion based fMRI. NeuroImage, 37(1), 90–101.
Canese, R., Adriani, W., Marco, E. M., De Pasquale, F., Lorenzini, P., De Luca, N., … Laviola, G. (2009). Peculiar response to methylphenidate in adolescent compared to adult rats: A phMRI study.
Psychopharmacology, 203(1), 143–153.
Cary, R. P., Ray, S., Grayson, D. S., Painter, J., Carpenter, S., Maron, L., … Fair, D. A. (2017). Network structure among brain systems in adult ADHD is uniquely modified by stimulant administration. Cerebral
Cortex, 27(8), 3970–3979.
Castellanos, F. X., Margulies, D. S., Kelly, A. M. C., Uddin, L. Q., Ghaffari, M., Kirsch, A., … Milham, M. P. (2008). Cingulate - Precuneus Interactions: A New Locus of Dysfunction in Adult
Attention-Deficit/Hyperactivity Disorder. Biol Psychiatry, 63(3), 332–337.
Castellanos, F. Xavier, & Proal, E. (2012). Large-scale brain systems in ADHD: Beyond the prefrontal-striatal model. Trends in Cognitive Sciences, 16(1), 17–26.
Castellanos, F. Xavier, Sonuga-Barke, E. J. S., Milham, M. P., & Tannock, R. (2006). Characterizing cognition in ADHD: Beyond executive dysfunction. Trends in Cognitive Sciences, 10(3), 117–123.
Castellanos, X., & Aoki, Y. (2016). Intrinsic Functional Connectivity in Attention-Deficit/ Hyperactivity Disorder: A Science in Development. Biol Psychiatry Cogn Neurosci Neuroimaging., 1(3), 253–261.
Castells, X., Ramos-Quiroga, J. A., Rigau, D., Bosch, R., Nogueira, M., Vidal, X., & Casas, M. (2011). Efficacy of methylphenidate for adults with attention-deficit hyperactivity disorder: a meta-regression analysis. CNS
Drugs, 25(2), 157–169.
Cortese, S., Aoki, Y. Y., Itahashi, T., Castellanos, F. X., & Eickhoff, S. B. (2020). Systematic Review and Meta-analysis: Resting State Functional Magnetic Resonance Imaging Studies of
Attention-Deficit/Hyperactivity Disorder. In Journal of the American Academy of Child & Adolescent Psychiatry. Cortese, S., D’Acunto, G., Konofal, E., Masi, G., & Vitiello, B. (2017). New Formulations of Methylphenidate
for the Treatment of Attention-Deficit/Hyperactivity Disorder: Pharmacokinetics, Efficacy, and Tolerability. CNS Drugs, 31(2), 149–160.
Cortese, S., Faraone, S. V., Bernardi, S., Wang, S., & Blanco, C. (2016). Gender Differences in Adult Attention-Deficit/Hyperactivity Disorder: Results From the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). J Clin Psychiatry, 77(4), 421–428.
Czerniak, S. M., Sikoglu, E. M., King, J. A., Kennedy, D. N., Mick, E., Frazier, J., & Moore, C. M. (2013). Areas of the brain modulated by single-dose methylphenidate treatment in youth with ADHD during task-based fMRI: A systematic review. Harvard Review of Psychiatry, 21(3), 151–162.
Danckaerts, M., Sonuga-Barke, E. J. S., Banaschewski, T., Buitelaar, J., Döpfner, M., Hollis, C., … Coghill, D. (2010). The quality of life of children with attention deficit/hyperactivity disorder: A systematic review.
