Documents

Konrad_et_al-2010-Human_Brain_Mapping.pdf

Description
r Human Brain Mapping 31:904–916 (2010) r Is the ADHD Brain Wired Differently? A Review on Structural and Functional Connectivity in Attention Deficit Hyperactivity Disorder Kerstin Konrad,1,2,3* and Simon B. Eickhoff2,3,4 1 Child Neuropsychology Section, Department of Child and Adolescent Psychiatry and Psychot
Categories
Published
of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  r  H uman  B rain  M apping 31:904–916 (2010)  r Is the ADHD Brain Wired Differently? A Review onStructural and Functional Connectivity inAttention Deficit Hyperactivity Disorder  Kerstin Konrad, 1,2,3 *  and Simon B. Eickhoff  2,3,4 1 Child Neuropsychology Section, Department of Child and Adolescent Psychiatry and Psychotherapy,University Hospital of the RWTH Aachen, Germany 2 Institute of Neuroscience and Medicine (INM-2, INM-3), Research Center Juelich, Germany 3  JARA – Translational Brain Medicine, Germany 4 Department of Psychiatry and Psychotherapy, University Hospital of the RWTH  Aachen University, Germany r r Abstract:  In recent years, a change in perspective in etiological models of attention deficit hyperactivity dis-order (ADHD) has occurred in concordance with emerging concepts in other neuropsychiatric disorderssuch as schizophrenia and autism. These models shift the focus of the assumed pathology from regional brain abnormalities to dysfunction in distributed network organization. In the current contribution, wereport findings from functional connectivity studies during resting and task states, as well as from studies onstructural connectivity using diffusion tensor imaging, in subjects with ADHD. Although major methodolog-ical limitations in analyzing connectivity measures derived from noninvasive in vivo neuroimaging still exist,there is convergent evidence for white matter pathology and disrupted anatomical connectivity in ADHD. Inaddition, dysfunctional connectivity during rest and during cognitive tasks has been demonstrated. How-ever, the causality between disturbed white matter architecture and cortical dysfunction remains to be eval-uated. Both genetic and environmental factors might contribute to disruptions in interactions betweendifferent brain regions. Stimulant medication not only modulates regionally specific activation strength butalso normalizes dysfunctional connectivity, pointing to a predominant network dysfunction in ADHD. Bycombining a longitudinal approach with a systems perspective in ADHD in the future, it might be possibleto identify at which stage during development disruptions in neural networks emerge and to delineate possi- ble new endophenotypes of ADHD.  Hum Brain Mapp 31:904–916, 2010 .  V C  2010 W iley- L iss, I nc. Keywords:  connectivity; ADHD; fMRI; DTI r r INTRODUCTION Attention deficit hyperactivity disorder (ADHD) is oneof the most common childhood neuropsychiatric disorders,and it often persists into adulthood. The psychopathologyof this disorder is marked by developmentally inappropri-ate and pervasive expressions of inattention, overactivity,and impulsiveness. ADHD is also associated with func-tional impairments across multiple academic and socialdomains and is commonly accompanied by a range of  Contract grant sponsor: German Federal Ministry of Education andScience (for K.K.); Contract grant numbers: BMBF-EDNET-01GV0602,BMBF-ANAC-01GJ0808; Contract grant sponsor: Human BrainProject (for S.B.E.); Contract grant number: NIH R01-MH074457-01A1;Contract grant sponsors: Excellence Initiative of the German federaland state governments (JARA-Seed fund) for K.K.; the HelmholzInitiative on Systems-Biology ‘‘The Human Brain Model’’for S.B.E.*Correspondence to: Kerstin Konrad, Child Neuropsychology Sec-tion, Department of Child and Adolescent Psychiatry and Psycho-therapy, University Hospital of the RWTH Aachen, Germany.E-mail: kkonrad@ukaachen.deReceived for publication 31 October 2009; Revised 17 February2010; Accepted 4 March 2010DOI: 10.1002/hbm.21058Published online 3 May 2010 in Wiley InterScience (www.interscience.wiley.com). V C  2010  W iley- L iss,  I nc.  externalizing and internalizing disorders [Biederman andFaraone, 2006]. Given the associated burden to society,family and the individual child, understanding the causesof ADHD and developing new and more effective treat-ments targeting these underlying causes is an importantgoal for neuroscience research.Although neuroimaging studies clearly point to a neuro- biological basis for the disorder, the pathophysiologicalmechanisms of ADHD and the specific nature of the atypi-cal brain development underlying it remain poorly under-stood. Recently, in concordance with emerging concepts inother neuropsychiatric disorders such as schizophreniaand autism, a change in perspective in etiological modelsof ADHD has occurred. These models shift the focus of the assumed pathology from regional brain abnormalitiesto dysfunctions in distributed network organization [Ser-geant et al., 2006]. While the assessment of functional seg-regation in the human brain, i.e., the localization of regionally specific functions, has been the predominantconcept in imaging neuroscience for many years, thepathophysiology of neuropsychiatric disorders is now being increasingly treated from a systems perspective inwhich function emerges from an interaction of regionallyspecialized elements. As a result, the analysis of brain con-nectivity has become more and more critical. However,the analysis of each of the three fundamental aspects of  brain connectivity, namely, anatomical, functional andeffective connectivity, is associated with its own technicaland conceptual challenges.In the current review, we aim to integrate findings fromfunctional connectivity studies during resting and taskstates with those from studies on structural connectivity.After a description of different connectivity methods and acritical discussion of methodological challenges and limita-tions associated with each of these measures, we will briefly summarize the typical development of cortical con-nectivity patterns across the human lifespan and then dis-cuss the findings of atypical brain connectivity in subjectswith ADHD. Finally, we will discuss how a systems per-spective might impact our understanding of the develop-ment of ADHD pathology and how such a systemsperspective can be addressed in future connectivity studiesof this disorder. Connectivity Measures Functional connectivity during resting state and during tasks using functional magneticresonance imaging  Functional connectivity is defined as the temporal corre-lation or coherence of spatially remote neurophysiologicalevents. In resting-state fMRI, which is defined by the ab-sence of external perturbations, stimulus-locked averagingof responses is not applicable. This leads to the develop-ment of new analytical approaches for the assessment of functional connectivity.Using resting state fMRI, the so-called default-mode net-work (DMN), a large and robustly replicable network of  brain regions that is associated with task-irrelevant mentalprocesses and mind wandering, has been identified. TheDMN comprises the precuneus/posterior cingulate cortex(PCC), the medial prefrontal cortex (MPFC) and themedial, lateral, and inferior parietal cortex [Laird et al., inpress; Schilbach et al., 2008]. The DMN shows higheractivity and stronger functional connectivity during restthan during externally driven tasks [Raichle et al., 2001].Activity in the DMN is attenuated, although not extin-guished, during the transition from rest-to-task states(Eichele et al., 2008; Greicius and Menon, 2004], andstronger deactivation is associated with increased taskdifficulty [Singh and Fawcett, 2008]. Persistence of DMNactivity during tasks has been shown to predict errors inthe flanker [Eichele et al., 2008] and the stop signal task[Li et al., 2007]. Moreover, unsuccessful attenuation of the DMN has also been reported to be associated withmomentary lapses in attention denoted by longer reactiontimes (RTs) and less accurate performance in an atten-tional control task [Weissman et al., 2006]. Hence, a failureto sufficiently suppress DMN activity may result inexcessive DMN activity that interferes with performanceon tasks [Li et al., 2007].Regions of the DMN show strong functional (temporalcoherence of BOLD signal in fMRI; Fox and Raichle, 2007]and structural connectivity (fiber tracking based on DTI;Grecius et al., 2008]. Consequently, there is considerableevidence, though no unequivocal proof, for structuredexchange of information between the different DMNregions [Buzsa´ki and Draguhn, 2004; Laird et al., in press].On the basis of these findings and on the apparent antago-nism between its activation and task performance, theDMN can be conceptualized as a strongly interconnectedtask-negative network.In addition to the DMN, a second prominent networkhas been characterized by spontaneous low frequency ac-tivity. Unlike the DMN, this network, which includes thedorsolateral prefrontal cortex (DLPFC), the intraparietalsulcus (IPS), and the supplementary motor area (SMA),has been described as task-positive, i.e., showing more ac-tivity during tasks that require active allocation of atten-tional resources than during rest. These regions thereforeappear to be associated with increased alertness, responsepreparation and selective attention in a manner that islargely independent of the specific task at hand [Fox et al.,2005; Sonuga-Barke and Castellanos, 2007]. Interestingly,the task-positive fronto-parietal network and the DMN aretemporally anti-correlated, such that task-specific activa-tion of the task-positive network is associated with attenu-ation of the DMN and vice versa.Despite recent advances in resting-state fMRI data anal-ysis, it should be noted that an important limitation of thewidely applied correlation approach is the inherently r  S tructural and  F unctional  C onnectivity in  ADHD  rr  905  r  subjective choice of the seed region of interest (ROI) bythe investigator. Furthermore, in the correlation approachthe global signal is usually regressed out. The use of sucha preprocessing step can induce false negative correlations between brain regions [Murphy et al., 2009]. To avoidthese issues, data-driven approaches such as independentcomponent analysis (ICA) have become increasingly prev-alent in the analysis of resting-state data. ICA decomposesthe four-dimensional (i.e., brain volume over time) bloodoxygen level dependent (BOLD) signal into a set of spa-tially distinct maps and their associated time courses.Among these independent components are several reliablyidentified functional brain networks, but also artifactsrelated to movement and physiological noise [Beckmannet al., 2005]. A general limitation of resting-state fMRI isthat it is very difficult to separate physiological noise fromthe BOLD signal of interest. Independent component anal-ysis largely separates these signals; however, residualnoise may still be present in the components of interest[Birn et al., 2008]. A possible solution to this problem is tocollect physiological measurements, model the evoked sig-nal changes and remove these confounds from the fMRIdata. Another potential difficulty with the analysis of rest-ing-state fMRI is that resting-state connectivity showsprominent very low frequency ( < 0.1 Hz) oscillations thatare hypothesized to provide temporal synchrony between brain regions (Fox et al., 2006; Sonuga-Barke and Castella-nos, 2007]. The physiology of these BOLD fluctuations andtheir relationship to neuronal activity, however, are a mat-ter of conjecture. These conceptual problems haveprompted the need for additional cross-validation of rest-ing-state connectivity. One approach that has providedsupport for the validity of the ensuing networks [Smithet al., 2009] is the delineation of task-related functionalconnectivity networks using coordinate-based meta-analy-sis [Eickhoff et al., 2009a; Laird et al., 2009]. While the con-gruence between the functional architecture of the brainduring the task and at rest is reassuring, further researchinto the nature and physiology of resting-state networksshould add to the validity of these approaches.Functional connectivity refers to the temporal correlationof spatially remote neurophysiological events. It has beeninvestigated using several different approaches, includingseed-voxel correlation analyses, ICAs and meta-analyticalconnectivity modeling. Interestingly, all of these methodsseem to provide evidence for several functional networksin the brain, most prominently the DMN and a fronto-pa-rietal task-positive network. There are, however, also dis-crepancies between the results obtained using the differentmethods. These may reflect differential confounds or pointto different aspects of brain connectivity being captured.The currently used approaches to functional connectivityshare some important drawbacks. These include the dan-ger of the induction of spurious correlations by indirecteffects and the fact that type and directionality of interac-tions cannot be delineated. Another major drawback isthat functional connectivity analyses usually do not allowinferences about the context-dependent dynamics of inter-regional interactions. In summary, functional connectivitysubsumes a variety of conceptually and technically differ-ent approaches that enable the delineation of functional brain networks defined by correlated neuronal activity butdo not allow inferences to be made about the causal na-ture of interactions between individual areas.Effective connectivity refers to the influence that a par-ticular brain region exerts over another, spatially distantregion. Because these effects cannot be directly assessed,inference on effective connectivity always relies on modelsof interactions; such models are usually informed byknowledge of the structure of the experiment, physiologi-cal constraints and other a priori assumptions. After fittingthe model to the experimental data, inference is thensought on the parameters capturing the influences betweendifferent brain regions. It becomes evident that effectiveconnectivity analysis can provide information about direc-tionality and context-dependent dynamics of interactionsthat cannot be derived from any other measure of connec-tivity. As these approaches are highly driven by hypothe-ses and prior assumptions, however, the validity of theinferences is also crucially dependent on these assump-tions. In a complex and yet poorly understood systemsuch as the brain, unequivocally acceptable assumptionsare sparse, resulting in competing approaches and apotential dependency of the results on the chosen model.In summary, effective connectivity provides models of  brain function that allow mechanistic insight into thecausal nature of inter-regional interactions but can do soonly by reference to several assumptions and a potential(over)simplification of the network at hand. Structural connectivity  Anatomical connectivity refers to the presence of an axo-nal connection between two brain regions. It was histori-cally investigated exclusively in non-human primatesusing invasive tracing; however, with the advent of diffu-sion tensor imaging (DTI), it has been the focus of manyin vivo imaging studies.DTI is an imaging technique based on the random diffu-sion of water molecules [Le Bihan, 2003; Moseley et al.,1990]. In an unrestricted environment, water moleculesdiffuse freely in any direction. In the white matter, how-ever, diffusion is restricted by the cell membranes andmyelin sheath. Consequently, water diffuses more readilyalong the orientation of axons, i.e., fiber tracts, than per-pendicular to the axons. Measuring the direction of diffu-sivity can therefore be used to infer the orientation of white matter tracts in the brain. Several measures have been developed and can be used to quantify white matterintegrity using DTI, the most common being fractional ani-sotropy (FA), mean diffusivity (MD), fiber count, andprobabilistic tractography.The development of DTI has been an important contri- bution to the field of neuroimaging, allowing inference r  K  onrad and  E ickhoff  rr  906  r  about structural connectivity in vivo. Despite its potential,however, DTI also has several important limitations,including susceptibility-induced signal loss, limited resolu-tion and dependency on the mathematical models used toinfer fiber orientation. In particular, due to the resolutionof the DTI signal, a given voxel may often include severalfiber tracts coursing in multiple directions. This can leadto an incorrect measure of the principal diffusion directionfor a particular tract. Moreover, it should be pointed outthat, in contrast to invasive techniques, DTI tractographyreveals only the likely presence of a fiber tract betweentwo regions, not axonal (i.e., synaptic) connectivity. Like-wise, DTI does not allow inference regarding the direction-ality of a particular tract or the laminar specificity of theinput/output neurons, which in turn provide critical con-straints for cortical information processing. DTI can thusdelineate the task-invariant anatomy of the WM providingthe scaffold for any sort of functional interaction but doesnot allow any inference on the dynamic nature of theseinteractions.The relationship between anatomical and functional con-nectivity is still disputed, and the number of studiesdirectly comparing resting-state functional connectivity toDTI results is still relatively small. Honey et al. [2007]used a computational approach to study spontaneous rest-ing-state fluctuations and their associations with structuralconnectivity based on invasive tract tracing data frommacaque monkeys. They demonstrated that the anatomicalconfiguration of neuronal networks can predict functionalconnectivity within these networks on multiple time-scales. Their work provided the first demonstration thatthe dynamics of resting-state networks may be deter-mined, at least partially, by anatomical constraints.Recently, Damoiseaux and Greicius [2009] reviewedeight articles that directly compare resting-state functionalconnectivity with structural connectivity and three clinicalcase studies of patients with limited white matter connec-tions between the cerebral hemispheres. The reviewedstudies showed largely convergent results, indicating thatthe strength of resting-state functional connectivity is posi-tively correlated with structural connectivity strength.Functional connectivity was also observed, however, between regions where there is little or no structural con-nectivity and vice versa. Such divergences between func-tional and anatomical connectivity may be interpreted inseveral ways: (i) coactivation of two regions may not bemediated by direct anatomical connections but via addi-tional structures. Such relays could either consist of a sin-gle area or, e.g., involve cascades of several intermediateareas or cortical-subcortical loops [Eickhoff et al., 2009b;Grefkes et al., 2008]; (ii) A third area could project(directly) to two regions, inducing a correlation of func-tional activation between them without a direct anatomicalconnection. That is, functional connectivity may be driven by an external source that induces concurrent activity in both areas; and (iii) A very weak anatomical connection between two regions may still hold a high functional sig-nificance (Eickhoff et al., 2008; Friston, 2002]. One examplewould be a case in which activity in one area depends ona ‘‘go signal’’ from another region. Functional connectivityis hence strongly influenced not only by the strength of ananatomical connection but also by the information con-veyed through it.The limitations inherent to each of the different methodscommonly used to assess brain connectivity, as well as thediscrepancies in the inferences that can be derived fromeach, likely prohibits a full understanding of physiologicalor pathological networks based on any one approach. Onthe other hand, information about different aspects of  brain connectivity may provide converging or complemen-tary evidence regarding network properties. A deeperunderstanding of brain connectivity and its ensuing func-tional networks should thus rely on a combination of dif-ferent but complementary approaches.In the following, we will briefly summarize the typicaldevelopment of brain connectivity before discussing theatypical development of neural network integration in sub- jects with ADHD. Typical Development of Brain Connectivity Fair et al. [2008] showed that the default network struc-ture in children deviates significantly from that seen inhealthy adults. Comparable to adults, interhemisphericfunctional connections between homotopic regions appearto be relatively strong in children. As a whole, however,the DMN is only sparsely connected, i.e., it is more frag-mented. During the course of normal brain development,the default network then becomes significantly more inte-grated until the adult configuration is reached. This sug-gests a more predominant functional segregation inchildren and greater functional integration in youngadults.Recent advances in graph theoretical approaches allowthe characterization of topological properties of complexnetworks. Using these approaches, Watts and Strogatz[1998] have shown that graphs with dense local connec-tions and few long connections can be characterized assmall-world networks. Small-world network topology has been demonstrated in many complex networks, includingsocial, economical, and biological networks (see Boccalettiet al., 2006 for a review); more recently, small-world topol-ogy has also been demonstrated in large-scale structural brain networks in humans and nonhuman primates [Hag-mann et al., 2007].Using this small-world approach, Supekar et al. [2008]recently reported that normal development of functionalconnectivity is characterized by simultaneous reduction of local circuitry and strengthening of long-range connectiv-ity. Importantly, this study showed that this is a generaldevelopmental principle that operates at the level of theentire brain and not just in circumscribed nodes of theattentional control and default mode networks, as r  S tructural and  F unctional  C onnectivity in  ADHD  rr  907  r
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks