Research Article Connectome-Scale Assessments of...

Research ArticleConnectome-Scale Assessments of Functional Connectivity inChildren with Primary Monosymptomatic Nocturnal Enuresis

Du Lei12 Jun Ma3 Jilei Zhang1 Mengxing Wang1 Kaihua Zhang1

Fuqin Chen2 Xueling Suo2 Qiyong Gong2 and Xiaoxia Du1

1Shanghai Key Laboratory of Magnetic Resonance Department of Physics East China Normal University3663 North Zhongshan Road Shanghai 200062 China2MR Research Center Department of Radiology West China Hospital of Sichuan University Chengdu Sichuan 610044 China3Department of Developmental and Behavioral Pediatrics of Shanghai Childrenrsquos Medical CenterXinHua Hospital Affiliated to Shanghai Jiao Tong University School of MedicineShanghai Key Laboratory of Childrenrsquos Environmental Health Shanghai 200127 China

Correspondence should be addressed to Xiaoxia Du xxduphyecnueducn

Received 6 August 2014 Revised 10 November 2014 Accepted 5 December 2014

Academic Editor Xi-Nian Zuo

Copyright copy 2015 Du Lei et al This is an open access article distributed under the Creative Commons Attribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Primary monosymptomatic nocturnal enuresis (PMNE) is a common developmental disorder in children Previous literaturehas suggested that PMNE not only is a micturition disorder but also is characterized by cerebral structure abnormalities anddysfunction However the biological mechanisms underlying the disease are not thoroughly understood Graph theoretical analysishas provided a unique tool to reveal the intrinsic attributes of the connectivity patterns of a complex network from a globalperspective Resting-state fMRIwas performed in 20 childrenwith PMNEand 20 healthy controls Brain networkswere constructedby computing Pearsonrsquos correlations for blood oxygenation level-dependent temporal fluctuations among the 2 groups followedby graph-based network analyses The functional brain networks in the PMNE patients were characterized by a significantly lowerclustering coefficient global and local efficiency and higher characteristic path length compared with controls PMNE patientsalso showed a reduced nodal efficiency in the bilateral calcarine sulcus bilateral cuneus bilateral lingual gyri and right superiortemporal gyrus Our findings suggest that PMNE includes brain network alterations that may affect global communication andintegration

1 Introduction

Nocturnal enuresis is a common developmental disorder thataffects 15ndash20 of 5-year-old children [1] This disorder canpersist in adolescence and has important negative effects onthe self-image and performance of these children [2] Thecondition is defined as primarymonosymptomatic nocturnalenuresis (PMNE) when a child has enuresis without addi-tional lower urinary tract symptoms (excluding nocturia) ora history of bladder dysfunction and has never had a periodof established urinary continence for more than six months[3]

Several factors are associated with and contribute tonocturnal enuresis including heredity polyuria detrusor

overactivity sleep and central nervous system mechanisms[4] Recently maturational delays of the central nervoussystem have been indicators of PMNE pathogenesis Toroset al reported an increased frequency of a high-levelhyperventilation response in recordings of a resting-stateelectroencephalogram suggesting the existence of delayedcortical maturity in PMNE [5] Event-related brain potentialshave also been used to study enuresis results have shownlonger P300 latency in primary enuretics compared withnonenuretics [6] and P300 amplitude is decreased in theparietal recordings of enuretics when compared with thecontrols [7] which is evidence of a maturational delay inthe central nervous system function [6 7] Freitag et al alsoreported the existence of increased I-III and I-V interpeak

Hindawi Publishing CorporationBioMed Research InternationalVolume 2015 Article ID 463708 8 pageshttpdxdoiorg1011552015463708

2 BioMed Research International

latencies of the brainstem auditory evoked potential sug-gesting maturational deficits in the brain stems of nocturnalenuretic children [8] Desamino-arginine vasopressin andalarm therapy for nocturnal enuresis were shown to increasethe prepulse inhibition of startle reflexes thus supportingthe hypothesis of a maturational delay of reflex inhibition innocturnal enuresis [9 10]

In recent years magnetic resonance imaging (MRI) tech-niques such as structuralMRI functionalMRI and diffusionMRI have provided an efficient feasible and noninvasivemethod to investigate the biological mechanisms of incon-tinence Using functional MRI a number of studies havereported alterations in several brain functions in patientswith urgency and urge incontinence [11 12] One studyutilized event-related fMRI in PMNE subjects and revealedthat children with PMNE had deficits in working memory[13] Abnormal functional connectivity has also been foundin children with PMNE including the cerebellum frontallobe and thalamus in the cerebello-thalamo-frontal circuit[14] We have performed a series of MRI experiments toinvestigate the functional and structural abnormalities thatare associated with PMNE We previously reported thatforebrain activation was altered during a response inhibitiontask [15] and that the nature of the local intrinsic activ-ity (ie amplitude low frequency fluctuation and regionalhomogeneity) changed in the prefrontal cortex during theresting state in children with PMNE [16] We also identifiedmicrostructural abnormalities in the thalamus the medialfrontal gyrus the anterior cingulated cortex and the insularcortex of children with PMNE using diffusion tensor imag-ing [17] and neurochemical abnormalities in the prefrontalcortex and the pons of children with PMNE using protonmagnetic resonance spectroscopy [18] These studies showedthat children with PMNE have structural functional andneurochemical abnormalities in the brain

Prior studies mostly focused on several regions whichinvolved cerebral micturition control network In fact chil-dren with PMNE probably have more serious problemsbeyond the micturition control they probably have poten-tial cognitive problems such as working memory [14] andresponse inhibition [15] Actually the whole brain can bemodeled as a large-scale complex network its function canbe fulfilled through both segregated and integrated specificfunctional connections patterns with optimized efficiency[19 20] The investigation of PMNE-related alterations inwhole brain functional networks instead of specific brainregions or local networks may give further network-levelinformation about the children with PMNE Up to now littleis known about the PMNE-related alterations in topologicalproperties especially the topological efficiency of the wholebrain functional networks during resting stateThe advantageof a graph theory-based network analysis is that it providesmeasures for both global and local connectivity

The topological organization of brain networks hasrecently been studied with graph theory [19 21 22] Graphtheory-based approaches model the brain as a complexnetwork representing it graphically using a collection ofnodes and edges Generally speaking a network consists of119873nodes that are linked by119870 edges Networks can be described

by an adjacent matrix 119860(119899 119899) in which 119899 is the number ofnodes and the value of 119860

119894119895refers to the edge linking node

119894 and node 119895 There are many graph metrics that can beused to describe the topological properties of a networkincluding the clustering coefficient (119862

119901) characteristic path

length (119871119901) normalized clustering coefficient (120574) normalized

characteristic path length (120582) small-worldness (120590) globalefficiency (119864glob) local efficiency (119864loc) nodal betweennessnodal degree and nodal efficiency After the network mod-eling procedure various graph theoretical metrics can beused to investigate the organizational mechanism underlyingthe relevant networks The graph-based network analysesenable us not only to visualize the overall connectivity patternamong all brain regions but also to quantitatively characterizethe global organization Graph-based techniques used tostudy brain networks including normal development agingand various brain disorders have increased [21ndash25] Previousstudies showed that the brainrsquos intrinsic activity is organizedas a small-world highly efficient network with significantmodularity and highly connected hub regions [19] whichhave also been found to change throughout normal devel-opment and aging and in various pathological conditions[21ndash25] In the present paper we propose a connectome-scale assessment of functional connectivity for children withPMNE via resting-state fMRI data

2 Methods

21 Subjects We studied 24 children with PMNE and 29healthy children with the consent of the children and theirguardians All children were right-handed with an IQ greaterthan 75 and the presence of neurological and psychiatricdiseases was excluded based on both a clinical examination(theDSM-IV criteria) and a structured interviewAll childrenwith PMNE were outpatients of the Shanghai ChildrenrsquosMedical Center This study was approved by the IRB of theShanghai Childrenrsquos Medical Center in affiliation with theShanghai Jiao Tong University School of Medicine (numberSCMC-201014)

The inclusion criteria for all participants were listed as fol-lows (1) a physical psychiatric and neurological evaluationconducted by at least 3 members of a team of certified andexperienced developmental and behavioral pediatricians (2)age 7ndash15 years (3) right-handedness (4) children diagnosedwith PMNE based on the criteria a child has enuresiswithout additional lower urinary tract symptoms (excludingnocturia) or history of bladder dysfunction and has neverhad a period of established urinary continence for morethan six months (5) the Wechsler Intelligence Scale forChildren-Revised (WISC-R) test employed to determine theintelligence quotient (IQ) of all subjects and (6) the numberand gender of each subgroup being matched

The exclusion criteria for this study were listed as follows(1) attention deficithyperactivity disorder autism or anypsychiatric comorbid disorders (2) previous head traumaneurologic disorders psychosurgery or substantial physicalillness (3) fMRI data with obvious artifacts and distor-tions (4) left-handedness as assessed with the Annett Hand

BioMed Research International 3

Preference Questionnaire and (5) a full-scale IQ less than 80according to an age appropriate WISC-Chinese Revision

Due to excessive head motion in some cases functionalimages of 20 PMNE patients and 20 healthy children wereavailable for further analysisThere were 2 groups of childrenthe average age was 108 plusmn 20 years for the PMNE group(14 males 6 females) and 99 plusmn 20 years for the normalcontrol group (14 males 6 females) Additional clinical dataregarding the patient groups are shown in SupplementaryMaterial 1 in Supplementary Material available online athttpdxdoiorg1011552015463708

22 Data Acquisition All subjects underwent a resting-statefunctional MRI scan using a 3T magnetic resonance system(Siemens Magnetom Trio Tim) with a 12-channel phasedarray head coil The sequence parameters were as followsrepetition timeecho time (119879

119877119879119864) = 200030ms flip angle

= 90∘ 32 axial slices per volume 3 mm slice thickness (33dist factor) matrix = 64 times 64 and FOV = 220 times 220mm2Each functional run contained 210 image volumes resultingin a total scanning time of 420 s for each participant Thefirst ten scans were discarded before the preprocessing of thedata to remove the impact of magnetization stabilization Allparticipants were instructed not to focus their thoughts onanything in particular and to keep their eyes closed duringthe acquisition

23 Data Preprocessing Image preprocessing was performedusing the SPM8 package (httpwwwfilionuclacukspmWellcomeTrust Centre forNeuroimagingUniversityCollegeLondon United Kingdom) and Graph Theoretical NetworkAnalysis (GRETNA httpwwwnitrcorgprojectsgretna)First for each participant the first 10 time points were dis-carded to avoid the instability of the initial MRI signal and tofamiliarize the subjects with the fMRI scanning noise Nextthe remaining fMRI data were corrected for the intravolumeacquisition time delay and head motion The head motionparameters of all participants were determined and theextendable inclusion criteria for translationalmovementwerelt30mm and lt30∘ rotation After these corrections theimages were spatially normalized to the standard space ofthe standardMontreal Neurological Institute (MNI) templateby applying the EPI template at a 3 times 3 times 3mm3 resolutionFinally the resulting data were further filtered through atemporal band pass (00ndash008Hz) to reduce the effects of low-frequency drift and high-frequency physiological noises

24 Network Construction and Analysis We used GRETNAto construct the network Todefine the brain nodes an atlas ofAutomatedAnatomical Labeling [26] was employed to dividethe entire brain into 90 cortical and subcortical regions ofinterest each representing a node of the network To definethe network edges first the representative mean time seriesof each region was acquired by averaging the time series ofall voxels within that region followed by a correction of headmotion effects by regressing out the head motion profilesestimated in the image realignment from the mean time-course Then the residuals of the regression analyses were

used to compute the partial correlation in this study resultingin a 90times 90 partial correlationmatrix for each subject Finallyindividual partial correlation matrices were converted intoweighted matrices this method has been used in previousbrain network studies [27ndash31]

We applied a wide range sparsity threshold 119878 to allcorrelation matrices which is determined in this procedureto guarantee that the threshold networks were estimable forsmall-worldness (scalar 120590 was larger than 11) and had sparseproperties with as few spurious edges as possible and theaverage degree over all nodes of each threshold network waslarger than 2 times log(90) [29 32] Our generated thresholdrange was 010 lt 119878 lt 034 with an interval of 001 For brainnetworks at each sparsity threshold we calculated both globaland node network metrics

The global metrics included (1) small-world parame-ters [32] (ie clustering coefficient 119862

119901 characteristic path

length 119871119901 normalized clustering coefficient 120574 normalized

characteristic path length 120582 and small-worldness 120590) and(2) network efficiency [33] (ie local efficiency 119864loc andglobal efficiency119864glob)The nodemetrics induced three nodalcentrality metrics the degree efficiency and betweennessFurthermore we calculated the area under the curve (AUC)for each network metric which provides a summarizedscalar for the topological characterization of brain networksindependent of a single threshold selection The AUC fora network metric 119884 which was calculated over the sparsitythreshold range of 119878

1to 119878119899with interval of Δ119878 was computed

as119884AUC = sum119899minus1119896=1[119884(119878119896)+119884(119878

119896+1)]timesΔ1198782 In the current study

1198781= 010 119878

119899= 034 andΔ119878 = 001 (supplementarymaterials

2) The AUCmetric has been used in previous brain networkstudies and is sensitive in detecting topological alterations ofbrain disorders [29 34ndash36]

Moreover to further localize the specific pairs of brainregions in which functional connectivity was altered inthe PMNE patients we identified region pairs that exhib-ited between-group differences in nodal characteristicsand utilized the network based statistics (NBS) method(httpwwwnitrcorgprojectsnbs) [37] to localize the con-nected networks that exhibited significant changes in thePMNE patients

Specially we firstly choose the nodes which exhibitedbetween-group differences in at least one of the three nodalcentralities (the node degree efficiency and betweenness)Then a subset of connections matrix for each participant wasconducted according to these altered nodes Subsequentlythe NBS approach was applied to define a set of suprathresh-old links that included any connected components (thresh-old 119879 = 16 119875 lt 005) To estimate the significance foreach component the nonparametric permutation approach(10000 permutations) was also conducted For a detaileddescription see [37]

25 Statistical Analysis Weused nonparametric permutationtests [29] for each network metricrsquos AUC We comparedthe overall topologies (ie small-world properties weightclustering coefficient and weight characteristic shortest pathlength) and the nodal characteristics (ie nodal degree nodal


125

175

225

275

325

007 017 027 037Sparsity

HCPMNE

105

11

115

12

125

007 017 027 037Sparsity

HCPMNE

Normalized

Cp

Normalized

Lp

Figure 1The key small-world parameters of functional network as a function of sparsity threshold Both PMNE group and non-PMNE groupshowed normalized 119862

119901larger than 1 and normalized 119871

119901approximately equal to 1 indicating both groups exhibited a small-world topology

PMNE children with primary monosymptomatic nocturnal enuresis HC healthy children 119862119901 clustering coefficient 119871


length

efficiency and betweenness centrality) of the functionalconnectivity network between the patients and controlsBriefly we first calculated the between-group difference inthe mean value of each network metric To test the nullhypothesis that the observed group differences could occurby chance for each network metric we then randomlyreallocated all the values into two groups and recomputed themean differences between the two randomized groups Thisrandomization procedure was repeated 10000 times and the95th percentile points of each distribution were used as thecritical values for a two-tailed test of the null hypothesis witha probability of type I error of 005 Additionally to addressthe problem of multiple comparisons the nodal centralitieswere tested on whether they survived a false discovery rate(FDR) threshold of 119902 = 005

3 Results

31 Global Topological Organization of the Functional Con-nectome The whole-brain connectome of both the PMNEand control groups exhibited typical features of small-worldtopology that is compared with matched random net-works the functional brain networks had higher clusteringcoefficients (119862

119901) but similar characteristic path length (119871

119901)

(Figure 1)The patient group showed significantly lower values for119862119901(119875 = 0007) and higher values for 119871

119901(119875 = 00008) No

significant (119875 gt 005) differences were found in the 120582 120574 or120590 In terms of network efficiency the comparisons revealedsignificant decreases in both 119864glob (119875 = 0001) and 119864loc(119875 = 0001) in the functional brain networks of the patientscompared with the healthy controls (Figure 2)

00102030405060708091

HCPMNE

Cp Lp120574 120590120582

(P = 00013)

(P = 00011)(P = 00066)

(P = 00008)

(P = 00147)

Eglob Eloc

Figure 2 Differences in topological properties of functional brainnetworks between pediatric PMNE patients and trauma exposednon-PMNE controls Significant differences were found in 119862

119901

(119875 = 00066) 120582 (119875 = 00147) 119864glob (119875 = 00013) and 119864loc(119875 = 00011) between pediatric PMNE patients and non-PMNEcontrols 995333 the black stars indicate the significantly statisticaldifference between the two groups (119875 lt 005 uncorrected) Errorbars denote standard deviations PMNE children with primarymonosymptomatic nocturnal enuresis HC healthy children 119864globglobal efficiency 119864loc local efficiency 119862

119901 clustering coefficient 120574

normalized clustering coefficient 120582 normalized characteristic pathlength 119871

119901 characteristic path length 120590 small-worldness

32 Regional Topological Organization of the Functional Con-nectome We identified the brain regions showing significantbetween-group differences in at least one nodal metric(119875 lt 005 FDR corrected) Compared with normal controlsubjects the patients showed decreased nodal centralities in


Table 1 Regions showing decreased nodal centralities in PMNE patients as compared with control subjects

Brain regions 119875 valuesNodal betweenness Nodal degree Nodal efficiency

Left calcarine sulcus 0460 0016 0001Right calcarine sulcus 0383 0023 0002Left cuneus 0124 0022 0001Right cuneus 0502 0017 0002Left lingual gyrus 0351 0008 0000Right lingual gyrus 0128 0000 0000Right superior temporal gyrus 0020 0017 0003Regions were considered abnormal in PMNE patients if they exhibited significant between-group differences (FDR corrected 119875 lt 005 shown in bold font) inat least one of the three nodal centralities

CUNL CUNR

STGR

LINGRLINGL

CALLCALR

Figure 3 The region pairs showing altered nodal centralities brainregions and functional connections in the PMNE patients Theseconnections formed a single connected network with 7 nodes and 10connections whichwas significantly (119875 lt 005 corrected) abnormalin the patients Edge in cyan increased functional connectionsin the PMNE patients edge in magenta decreased functionalconnections in the PMNE patients CUN cuneus LING lingualgyrus CAL calcarine sulcus STG superior temporal gyrus R righthemisphere P posterior The nodes and connections were mappedonto the cortical surfaces using the BrainNet Viewer package(httpwwwnitrcorgprojectsbnv)

several brain regions including the bilateral calcarine sulcusthe bilateral cuneus the bilateral lingual gyri and the rightsuperior temporal gyrus (Figure 3 Table 1) There were nosignificantly increased nodal centralities

33 Disrupted Functional Network Connectivity in thePatients Using NBS [37] we identified a connected networkwith 7 nodes and 10 connections this network was signif-icantly altered in the PMNE group (119875 lt 005 corrected)(Figure 3 Table 2) Within this network both decreasedand increased connections were detected in the patientscompared with the control subjects The connections of theleft cuneus in the bilateral calcarine sulcus the left cuneus inthe right cuneus and the left cuneus in the right lingual gyrus

were decreased Furthermore the connections of the rightlingual gyrus in the left lingual gyrus and the right lingualgyrus in bilateral calcarine sulcus were also decreased Onthe other side the increased connections included the rightsuperior temporal gyrus in the bilateral calcarine sulcus andthe right superior temporal gyrus in the left cuneus

4 Discussion

41 Global Topological Organization of the Functional Con-nectome In our study we computed small-word metrics of90 node networks based on the ALL template and found that120582 120574 and 120590 of the functional networks were not significantlychanged in the PMNE group however the patient groupshowed significantly lower values for 119862

119901and higher values

for 119871119901 and the network efficiency parameters 119864glob and 119864loc

were significantly decreased in the functional brain networksof the patients

According to graph theory a high 119862119901indicates that the

nodes tend to form dense regional cliques implying that theefficiency in local information transfer and processing is high[38] a low 119871

119901indicates a high transfer speed through the

overall network implying that the network has a high globalefficiency [38] Thus the 119862

119901and 119864loc were reduced in PMNE

group which demonstrated that the efficiency in local infor-mation transfer and processing is low in childrenwith PMNEThe higher 119871

119901and lower 119864glob in the PMNE group suggested

that the brain network has a low global efficiency and a lowability of integrating information Our present results arecorrelated with the previous findings that the efficiency ofregional information processing and the efficiency of overallinformation transfer across the entire network were reducedin children with PMNE [38] Therefore our results suggestthat intrinsic brain functional organization was disrupted inPMNE and that there was reduced efficiency in informationexchange and integration in both local and global regions

42 Regional Topological Organization of the Functional Con-nectome Childrenwith PMNE showed significant decreasednodal efficiency in the bilateral calcarine sulcus bilateralcuneus bilateral lingual gyri and right superior temporalgyrus The calcarine sulcus cuneus and lingual gyrus are inthe occipital lobe which is part of visual network The right


Table 2 Altered functional connections in the PMNE patients group as compared to the control group

Region 1 Region 2 119905-score IncreasedecreaseRight superior temporal gyrus Left cuneus 395 IncreaseRight superior temporal gyrus Left calcarine sulcus 229 IncreaseRight superior temporal gyrus Right calcarine sulcus 221 IncreaseLeft cuneus Left calcarine sulcus 225 DecreaseLeft cuneus Right cuneus 175 DecreaseLeft cuneus Right calcarine sulcus 170 DecreaseLeft cuneus Right lingual gyrus 165 DecreaseRight lingual gyrus Left lingual gyrus 336 DecreaseRight lingual gyrus Left calcarine sulcus 206 DecreaseRight lingual gyrus Right calcarine sulcus 172 DecreaseConnections are listed in descending order of statistical significance (119875 lt 005) These connections formed a connected network that was identified using anetwork-based statistical approach See Figure 2 for a graphical representation of these connections

superior temporal gyrus is an essential structure involved inauditory processing and part of the auditory network

Tomasi and Volkow reported that the cuneus is oneof 4 major cortical hubs and the cuneus hub networkis both correlated with the somatosensory network andanticorrelated with the default mode network activity ofother networks [39] The calcarine cortex lingual gyrifusiform occipital gyri and paracentral lobule were thesecondary hubs identified in this network which suggeststhat the visual processing performed by the secondary hubsis integrated in the cuneus [39] A previous study suggeststhat the cuneus and lingual gyrus take part in stop-downprocesses of visuospatial attention [40] The cuneus is animportant node of the default mode network and is alsoaffected by attention-deficithyperactivity disorder [41 42]Decreased connectivity in the left cuneus has also beenfound in individuals with borderline personality disorder[43] Additionally gray matter volume in the cuneus hasbeen suggested to be associated with better inhibitory con-trol in bipolar depression patients [44] and dysfunctionhas been reported in the response inhibition in childrenwith PMNE [15] Thus reduced nodal efficiency in thecuneus may point to insufficiencies in communicationsbetween executive control regions and visual processingregions

In our results the nodal efficiency was reduced in thebilateral lingual gyri and the degree was reduced in theright lingual gyrus Interestingly the lingual gyrus has beenassociated with psychopathology such as major depression[45] posttraumatic stress disorder [46 47] and childhoodmaltreatment [48] A previous task related neuroimagingstudy reported that the lingual gyrus was associated withthe identification of facial expressions of emotion [49]Equit et al reported that children with nocturnal enuresisprocessed emotions differently from children with attention-deficithyperactivity disorder and controls [50] Thus wespeculated that the abnormality in the lingual gyri would berelated to the burden of disease and negative psychologicalfactors

43 Disrupted Functional Network Connectivity in thePatients The NBS method analysis showed increased con-nections between auditory system (right superior temporalgyrus) and visual system and decreased connections withinthe visual network including the bilateral calcarine sulcusthe bilateral cuneus and the bilateral lingual gyri Thesefindings suggest that connectivity of the auditory and visualnetworks is unbalanced in children with PMNE

Visual information is the main source of informationand the visual network is closely linked to other brainnetworks When the visual network functions abnormallya great potential for harm is induced possibly resultingin damage to cognitive function Abnormal visual networkfunctionality can also result in reduced efficiency of theglobal network and can affect global communication andintegration The network efficiency of both 119864glob and 119864loc inthe functional brain networks of the patients was significantlydecreased This observation suggests that the disruptionof brain communication and integration may also be animportant factor in PMNE disorders

Several issues must be addressed further First the brainnetworks were constructed by parcellating the entire braininto 90 brain regions at a coarsely regional level Futurestudies should use a more precise parcellation strategy orspatial scale to overcome this challenge Second the func-tional brain networks constructed from the r-fMRI data werelargely constrained by anatomical pathways [51 52] Thuscombining multimodal neuroimaging data could facilitatethe uncovering of the structure-function relationships inPMNE patients Finally the results in this paper are differentfrom those of our previous resting-state fMRI study [16]which focused on local intrinsic activity We did not identifya direct relationship between the results in this study and thepathology of PMNE However this study sheds light on thefunctional connectivity of children with PMNE and suggeststhat there may be potential cognition dysfunction in childrenwith PMNE

It is important to note some limitations in this study Firstfunctional brain networkswere constructed at a regional level


by parcellating the whole brain into 90 regions based ona previously published atlas Brain networks derived usingdifferent parcellation schemes or at different spatial scalesexhibit distinct topological architectures [53 54] Furtherstudies are needed to determine which brain parcellationstrategy or spatial scale is more appropriate for the charac-terization of network topology in PMNE Second we havenot found the direct relationship between our results andthe pathology of PMNE The regions showing abnormalitynodal centralities in PMNE patients were not involved inthe micturition neural control network However our resultsgive direct evidence that the PMNE probably has cognitiveproblem

5 Conclusion

This is the first study to investigate the characteristics ofPMNE patients with network-based graph theory usingresting-state fMRI In the PMNE group there were reducedlocal and global efficiency in the brain as well as disturbancesin connectivityThe alterations in network topologies primar-ily occurred in the right superior temporal gyrus (auditorynetwork) and decreased connectivity was observed in thevisual network including the bilateral calcarine sulcus thebilateral cuneus and the bilateral lingual gyri Our findingssuggest that PMNE includes brain network alterations whichmay affect both local and global communication and integra-tion

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

This research was supported by grants from the NationalNatural Science Foundation of China 81201082 (X Du) theScience and Technology Commission of Shanghai Munic-ipality (14411969200) and the Shanghai Key Laboratory ofEnvironment and Children Health (14DZ2271400)

References

[1] M Riccabona ldquoEvaluation und management of enuresis anupdaterdquo UrologemdashAusgabe A vol 49 no 7 pp 861ndash870 2010

[2] M Theunis E van Hoecke S Paesbrugge P Hoebeke andJ Vande Walle ldquoSelf-image and performance in children withnocturnal enuresisrdquo European Urology vol 41 no 6 pp 660ndash667 2002

[3] T Neveus A von Gontard P Hoebeke et al ldquoThe standardiza-tion of terminology of lower urinary tract function in childrenand adolescents report from the Standardisation Committeeof the International Childrenrsquos Continence Societyrdquo Journal ofUrology vol 176 no 1 pp 314ndash324 2006

[4] T Neveus ldquoDiagnosis and management of nocturnal enuresisrdquoCurrent Opinion in Pediatrics vol 21 no 2 pp 199ndash202 2009

[5] F Toros A Ozge M Bozlu and S Cayan ldquoHyperventilationresponse in the electroencephalogramandpsychiatric problems

in children with primary monosymptomatic nocturnal enure-sisrdquo Scandinavian Journal of Urology andNephrology vol 37 no6 pp 471ndash476 2003

[6] A Iscan YOzkul DUnal et al ldquoAbnormalities in event-relatedpotential and brainstem auditory evoked response in childrenwith nocturnal enuresisrdquo Brain and Development vol 24 no 7pp 681ndash687 2002

[7] R Karlidag H I Ozisik A Soylu et al ldquoTopographic abnor-malities in event-related potentials in children with monosyp-tomatic nocturnal enuresisrdquo Neurourology and Urodynamicsvol 23 no 3 pp 237ndash240 2004

[8] C M Freitag D Rohling S Seifen R Pukrop and Avon Gontard ldquoNeurophysiology of nocturnal enuresis evokedpotentials and prepulse inhibition of the startle reflexrdquoDevelop-mental Medicine amp Child Neurology vol 48 no 4 pp 278ndash2842006

[9] S Schulz-Juergensen A Langguth and P Eggert ldquoEffect ofalarm therapy on conditioning of central reflex control in noc-turnal enuresis pilot study on changes in prepulse inhibition(PPI)rdquo Pediatric Nephrology vol 29 no 7 pp 1209ndash1213 2014

[10] S Schulz-Juergensen M Rieger J Schaefer A Neusuess andP Eggert ldquoEffect of 1-desamino-8-D-arginine vasopressin onprepulse inhibition of startle supports a central etiology ofprimary monosymptomatic enuresisrdquo Journal of Pediatrics vol151 no 6 pp 571ndash574 2007

[11] C J Fowler and D J Griffiths ldquoA decade of functionalbrain imaging applied to bladder controlrdquo Neurourology andUrodynamics vol 29 no 1 pp 49ndash55 2010

[12] D J Griffiths and S D Tadic ldquoBladder control urgency andurge incontinence evidence from functional brain imagingrdquoNeurourology and Urodynamics vol 27 no 6 pp 466ndash4742008

[13] B Yu Q Guo G Fan H Ma L Wang and N Liu ldquoEvaluationof working memory impairment in children with primarynocturnal enuresis evidence from event-related functionalmagnetic resonance imagingrdquo Journal of Paediatrics and ChildHealth vol 47 no 7 pp 429ndash435 2011

[14] B Yu H Sun H Ma et al ldquoAberrant whole-brain functionalconnectivity and intelligence structure in childrenwith primarynocturnal enuresisrdquo PLoS ONE vol 8 no 1 Article ID e519242013

[15] D Lei J Ma X Du G Shen M Tian and G Li ldquoAltered brainactivation during response inhibition in children with primarynocturnal enuresis an fMRI studyrdquoHuman BrainMapping vol33 no 12 pp 2913ndash2919 2012

[16] D Lei J Ma X Du G Shen M Tian and G Li ldquoSpontaneousbrain activity changes in children with primary monosymp-tomatic nocturnal enuresis a resting-state fMRI studyrdquo Neu-rourology and Urodynamics vol 31 no 1 pp 99ndash104 2012

[17] D Lei JMa X Shen et al ldquoChanges in the brainmicrostructureof children with primary monosymptomatic nocturnal enure-sis a diffusion tensor imaging studyrdquo PLoS ONE vol 7 no 2Article ID e31023 2012

[18] J Zhang D Lei J Ma et al ldquoBrain metabolite alterationsin children with primary nocturnal enuresis using protonmagnetic resonance spectroscopyrdquoNeurochemical Research vol39 no 7 pp 1355ndash1362 2014

[19] Y He and A Evans ldquoGraph theoretical modeling of brainconnectivityrdquo Current Opinion in Neurology vol 23 pp 341ndash350 2010

[20] D S Bassett and E Bullmore ldquoSmall-world brain networksrdquoNeuroscientist vol 12 no 6 pp 512ndash523 2006


[21] J Wang X Zuo and Y He ldquoGraph-based network analysis ofresting-state functionalMRIrdquo Frontiers in SystemsNeurosciencevol 4 article 16 2010

[22] Y Iturria-Medina ldquoAnatomical brain networks on the predic-tion of abnormal brain statesrdquo Brain Connectivity vol 3 no 1pp 1ndash21 2013

[23] K Wu Y Taki K Sato et al ldquoTopological organization offunctional brain networks in healthy children differences inrelation to age sex and intelligencerdquo PLoS ONE vol 8 no 2Article ID e55347 2013

[24] M Cao J-H Wang Z-J Dai et al ldquoTopological organizationof the human brain functional connectome across the lifespanrdquoDevelopmental Cognitive Neuroscience vol 7 pp 76ndash93 2014

[25] BM Tijms AMWinkW deHaan et al ldquoAlzheimerrsquos diseaseconnecting findings from graph theoretical studies of brainnetworksrdquo Neurobiology of Aging vol 34 no 8 pp 2023ndash20362013

[26] N Tzourio-Mazoyer B Landeau D Papathanassiou et alldquoAutomated anatomical labeling of activations in SPM using amacroscopic anatomical parcellation of the MNI MRI single-subject brainrdquo NeuroImage vol 15 no 1 pp 273ndash289 2002

[27] Y Liu M Liang Y Zhou et al ldquoDisrupted small-worldnetworks in schizophreniardquo Brain vol 131 no 4 pp 945ndash9612008

[28] L Ferrarini I M Veer E Baerends et al ldquoHierarchical func-tional modularity in the resting-state human brainrdquo HumanBrain Mapping vol 30 no 7 pp 2220ndash2231 2009

[29] J Zhang J Wang Q Wu et al ldquoDisrupted brain connectivitynetworks in drug-naive first-episode major depressive disor-derrdquo Biological Psychiatry vol 70 no 4 pp 334ndash342 2011

[30] R Salvador J SucklingMRColeman JD PickardDMenonandE Bullmore ldquoNeurophysiological architecture of functionalmagnetic resonance images of human brainrdquo Cerebral Cortexvol 15 no 9 pp 1332ndash2342 2005

[31] Y He Z Chen and A Evans ldquoStructural insights into aber-rant topological patterns of large-scale cortical networks inAlzheimerrsquos diseaserdquoThe Journal of Neuroscience vol 28 no 18pp 4756ndash4766 2008

[32] D J Watts and S H Strogatz ldquoCollective dynamics of ldquosmall-worldrdquo networksrdquoNature vol 393 no 6684 pp 440ndash442 1998

[33] V Latora and M Marchiori ldquoEfficient behavior of small-worldnetworksrdquo Physical Review Letters vol 87 no 19 Article ID198701 2001

[34] S Achard and E Bullmore ldquoEfficiency and cost of economicalbrain functional networksrdquo PLoS Computational Biology vol 3no 2 article e17 2007

[35] Y He A Dagher Z Chen et al ldquoImpaired small-worldefficiency in structural cortical networks in multiple sclerosisassociated with white matter lesion loadrdquo Brain vol 132 no 12pp 3366ndash3379 2009

[36] JWang LWang Y Zang et al ldquoParcellation-dependent small-world brain functional networks a resting-state fmri studyrdquoHuman Brain Mapping vol 30 no 5 pp 1511ndash1523 2009

[37] A Zalesky A Fornito and E T Bullmore ldquoNetwork-basedstatistic identifying differences in brain networksrdquoNeuroImagevol 53 no 4 pp 1197ndash1207 2010

[38] T Xie and Y He ldquoMapping the alzheimerrsquos brain with connec-tomicsrdquo Frontiers in Psychiatry vol 2 article 77 2012

[39] D Tomasi and N D Volkow ldquoAssociation between functionalconnectivity hubs and brain networksrdquo Cerebral Cortex vol 21no 9 pp 2003ndash2013 2011

[40] B Hahn T J Ross and E A Stein ldquoNeuroanatomicaldissociation between bottom-up and top-down processes ofvisuospatial selective attentionrdquo NeuroImage vol 32 no 2 pp842ndash853 2006

[41] B de Celis Alonso S H Tobon P D Suarez et al ldquoA multi-methodological MR resting state network analysis to assess thechanges in brain physiology of children with ADHDrdquo PLoSONE vol 9 no 6 Article ID e99119 2014

[42] X Wang Y Jiao T Tang H Wang and Z Lu ldquoAlteredregional homogeneity patterns in adults with attention-deficithyperactivity disorderrdquo European Journal of Radiology vol 82no 9 pp 1552ndash1557 2013

[43] R C Wolf F Sambataro N Vasic et al ldquoAberrant connectivityof resting-state networks in borderline personality disorderrdquoJournal of Psychiatry and Neuroscience vol 36 no 6 pp 402ndash411 2011

[44] M Haldane G Cunningham C Androutsos and S FrangouldquoStructural brain correlates of response inhibition in BipolarDisorder Irdquo Journal of Psychopharmacology vol 22 no 2 pp138ndash143 2008

[45] I M Veer C F Beckmann M-J van Tol et al ldquoWhole brainresting-state analysis reveals decreased functional connectivityin major depressionrdquo Frontiers in Systems Neuroscience vol 4article 41 2010

[46] L-D Qin Z Wang Y-W Sun et al ldquoA preliminary study ofalterations in default network connectivity in post-traumaticstress disorder patients following recent traumardquo BrainResearch vol 1484 pp 50ndash56 2012

[47] Y Yin C Jin L T Eyler H Jin X Hu and L Duan ldquoAlteredregional homogeneity in post-traumatic stress disorder aresting-state functional magnetic resonance imaging studyrdquoNeuroscience Bulletin vol 28 no 5 pp 541ndash549 2012

[48] S J A van der Werff J N Pannekoek I M Veer etal ldquoResilience to childhood maltreatment is associated withincreased resting-state functional connectivity of the saliencenetwork with the lingual gyrusrdquo Child Abuse and Neglect vol37 no 11 pp 1021ndash1029 2013

[49] R Kitada I S Johnsrude T Kochiyama and S J LedermanldquoBrain networks involved in haptic and visual identification offacial expressions of emotion an fMRI studyrdquo NeuroImage vol49 no 2 pp 1677ndash1689 2010

[50] M Equit A Becker D El Khatib M Rubly N Beckerand A von Gontard ldquoCentral nervous system processing ofemotions in children with nocturnal enuresis and attention-deficithyperactivity disorderrdquo Acta Paediatrica vol 103 no 8pp 868ndash878 2014

[51] J S Damoiseaux and M D Greicius ldquoGreater than the sum ofits parts a review of studies combining structural connectivityand resting-state functional connectivityrdquo Brain Structure andFunction vol 213 no 6 pp 525ndash533 2009

[52] C J Honey O Sporns L Cammoun et al ldquoPredicting humanresting-state functional connectivity from structural connectiv-ityrdquoProceedings of theNational Academy of Sciences of theUnitedStates of America vol 106 no 6 pp 2035ndash2040 2009

[53] A Fornito A Zalesky and E T Bullmore ldquoNetwork scalingeffects in graph analytic studies of human resting-state fMRIdatardquo Frontiers in Systems Neuroscience vol 4 article 22 2010

[54] A Zalesky A Fornito I H Harding et al ldquoWhole-brainanatomical networks does the choice of nodes matterrdquo Neu-roImage vol 50 no 3 pp 970ndash983 2010

Submit your manuscripts athttpwwwhindawicom

Neurology Research International

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Alzheimerrsquos DiseaseHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


BioMed Research International


Research and TreatmentSchizophrenia

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Neural Plasticity


Parkinsonrsquos Disease


Research and TreatmentAutism

Sleep DisordersHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Neuroscience Journal

Epilepsy Research and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Psychiatry Journal


Computational and Mathematical Methods in Medicine

Depression Research and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Brain ScienceInternational Journal of

StrokeResearch and TreatmentHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Neurodegenerative Diseases


Journal of

Cardiovascular Psychiatry and NeurologyHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


latencies of the brainstem auditory evoked potential sug-gesting maturational deficits in the brain stems of nocturnalenuretic children [8] Desamino-arginine vasopressin andalarm therapy for nocturnal enuresis were shown to increasethe prepulse inhibition of startle reflexes thus supportingthe hypothesis of a maturational delay of reflex inhibition innocturnal enuresis [9 10]

In recent years magnetic resonance imaging (MRI) tech-niques such as structuralMRI functionalMRI and diffusionMRI have provided an efficient feasible and noninvasivemethod to investigate the biological mechanisms of incon-tinence Using functional MRI a number of studies havereported alterations in several brain functions in patientswith urgency and urge incontinence [11 12] One studyutilized event-related fMRI in PMNE subjects and revealedthat children with PMNE had deficits in working memory[13] Abnormal functional connectivity has also been foundin children with PMNE including the cerebellum frontallobe and thalamus in the cerebello-thalamo-frontal circuit[14] We have performed a series of MRI experiments toinvestigate the functional and structural abnormalities thatare associated with PMNE We previously reported thatforebrain activation was altered during a response inhibitiontask [15] and that the nature of the local intrinsic activ-ity (ie amplitude low frequency fluctuation and regionalhomogeneity) changed in the prefrontal cortex during theresting state in children with PMNE [16] We also identifiedmicrostructural abnormalities in the thalamus the medialfrontal gyrus the anterior cingulated cortex and the insularcortex of children with PMNE using diffusion tensor imag-ing [17] and neurochemical abnormalities in the prefrontalcortex and the pons of children with PMNE using protonmagnetic resonance spectroscopy [18] These studies showedthat children with PMNE have structural functional andneurochemical abnormalities in the brain

Prior studies mostly focused on several regions whichinvolved cerebral micturition control network In fact chil-dren with PMNE probably have more serious problemsbeyond the micturition control they probably have poten-tial cognitive problems such as working memory [14] andresponse inhibition [15] Actually the whole brain can bemodeled as a large-scale complex network its function canbe fulfilled through both segregated and integrated specificfunctional connections patterns with optimized efficiency[19 20] The investigation of PMNE-related alterations inwhole brain functional networks instead of specific brainregions or local networks may give further network-levelinformation about the children with PMNE Up to now littleis known about the PMNE-related alterations in topologicalproperties especially the topological efficiency of the wholebrain functional networks during resting stateThe advantageof a graph theory-based network analysis is that it providesmeasures for both global and local connectivity

The topological organization of brain networks hasrecently been studied with graph theory [19 21 22] Graphtheory-based approaches model the brain as a complexnetwork representing it graphically using a collection ofnodes and edges Generally speaking a network consists of119873nodes that are linked by119870 edges Networks can be described

by an adjacent matrix 119860(119899 119899) in which 119899 is the number ofnodes and the value of 119860

119894119895refers to the edge linking node

119894 and node 119895 There are many graph metrics that can beused to describe the topological properties of a networkincluding the clustering coefficient (119862

119901) characteristic path

length (119871119901) normalized clustering coefficient (120574) normalized

characteristic path length (120582) small-worldness (120590) globalefficiency (119864glob) local efficiency (119864loc) nodal betweennessnodal degree and nodal efficiency After the network mod-eling procedure various graph theoretical metrics can beused to investigate the organizational mechanism underlyingthe relevant networks The graph-based network analysesenable us not only to visualize the overall connectivity patternamong all brain regions but also to quantitatively characterizethe global organization Graph-based techniques used tostudy brain networks including normal development agingand various brain disorders have increased [21ndash25] Previousstudies showed that the brainrsquos intrinsic activity is organizedas a small-world highly efficient network with significantmodularity and highly connected hub regions [19] whichhave also been found to change throughout normal devel-opment and aging and in various pathological conditions[21ndash25] In the present paper we propose a connectome-scale assessment of functional connectivity for children withPMNE via resting-state fMRI data

2 Methods

21 Subjects We studied 24 children with PMNE and 29healthy children with the consent of the children and theirguardians All children were right-handed with an IQ greaterthan 75 and the presence of neurological and psychiatricdiseases was excluded based on both a clinical examination(theDSM-IV criteria) and a structured interviewAll childrenwith PMNE were outpatients of the Shanghai ChildrenrsquosMedical Center This study was approved by the IRB of theShanghai Childrenrsquos Medical Center in affiliation with theShanghai Jiao Tong University School of Medicine (numberSCMC-201014)

The inclusion criteria for all participants were listed as fol-lows (1) a physical psychiatric and neurological evaluationconducted by at least 3 members of a team of certified andexperienced developmental and behavioral pediatricians (2)age 7ndash15 years (3) right-handedness (4) children diagnosedwith PMNE based on the criteria a child has enuresiswithout additional lower urinary tract symptoms (excludingnocturia) or history of bladder dysfunction and has neverhad a period of established urinary continence for morethan six months (5) the Wechsler Intelligence Scale forChildren-Revised (WISC-R) test employed to determine theintelligence quotient (IQ) of all subjects and (6) the numberand gender of each subgroup being matched

The exclusion criteria for this study were listed as follows(1) attention deficithyperactivity disorder autism or anypsychiatric comorbid disorders (2) previous head traumaneurologic disorders psychosurgery or substantial physicalillness (3) fMRI data with obvious artifacts and distor-tions (4) left-handedness as assessed with the Annett Hand





119877119879119864) = 200030ms flip angle











as119884AUC = sum119899minus1119896=1[119884(119878119896)+119884(119878


1198781= 010 119878







125

175

225

275

325

007 017 027 037Sparsity

HCPMNE

105

11

115

12

125

007 017 027 037Sparsity

HCPMNE

Normalized

Cp

Normalized

Lp






length


3 Results



119901)


119901(119875 = 00008) No


00102030405060708091

HCPMNE

Cp Lp120574 120590120582

(P = 00013)

(P = 00011)(P = 00066)

(P = 00008)

(P = 00147)

Eglob Eloc


119901










CUNL CUNR

STGR

LINGRLINGL

CALLCALR





4 Discussion


























5 Conclusion




Acknowledgments


References





































































Neural Plasticity










Psychiatry Journal









Journal of






119877119879119864) = 200030ms flip angle











as119884AUC = sum119899minus1119896=1[119884(119878119896)+119884(119878


1198781= 010 119878







125

175

225

275

325

007 017 027 037Sparsity

HCPMNE

105

11

115

12

125

007 017 027 037Sparsity

HCPMNE

Normalized

Cp

Normalized

Lp






length


3 Results



119901)


119901(119875 = 00008) No


00102030405060708091

HCPMNE

Cp Lp120574 120590120582

(P = 00013)

(P = 00011)(P = 00066)

(P = 00008)

(P = 00147)

Eglob Eloc


119901










CUNL CUNR

STGR

LINGRLINGL

CALLCALR





4 Discussion


























5 Conclusion




Acknowledgments


References





































































Neural Plasticity










Psychiatry Journal









Journal of



125

175

225

275

325

007 017 027 037Sparsity

HCPMNE

105

11

115

12

125

007 017 027 037Sparsity

HCPMNE

Normalized

Cp

Normalized

Lp






length


3 Results



119901)


119901(119875 = 00008) No


00102030405060708091

HCPMNE

Cp Lp120574 120590120582

(P = 00013)

(P = 00011)(P = 00066)

(P = 00008)

(P = 00147)

Eglob Eloc


119901










CUNL CUNR

STGR

LINGRLINGL

CALLCALR





4 Discussion


























5 Conclusion




Acknowledgments


References





































































Neural Plasticity










Psychiatry Journal









Journal of






CUNL CUNR

STGR

LINGRLINGL

CALLCALR





4 Discussion


























5 Conclusion




Acknowledgments


References





































































Neural Plasticity










Psychiatry Journal









Journal of














5 Conclusion




Acknowledgments


References





































































Neural Plasticity










Psychiatry Journal









Journal of

















































Neural Plasticity










Psychiatry Journal









Journal of


Research Article Connectome-Scale Assessments of...

Documents

Transcript of Research Article Connectome-Scale Assessments of...