functional connectivity analysis and prediction of

CLINICAL ARTICLEJ Neurosurg 131:1709–1715, 2019

While the goal of carotid artery stenting (CAS) is to prevent stroke in patients with severe ca-rotid artery stenosis, as emphasized in large

randomized trials,19,33 carotid revascularization sometimes improves cognitive function.12 The cognitive function of patients with internal carotid artery (ICA) stenosis, even

those who are asymptomatic, has been shown to be im-paired.8,21,31 Cognitive impairment is one of the most seri-ous problems elderly persons can face, and it stands as a social problem that is increasing along with the general aging of the population. The neurocognitive course of pa-tients who have undergone cerebral revascularization has

ABBREVIATIONS aDMN = anterior part of DMN; CAS = carotid artery stenting; DMN = default mode network; FC = functional connectivity; FIQ = full-scale IQ; ICA = inter-nal carotid artery; MFG = middle frontal gyrus; MNI = Montreal Neurological Institute; NASCET = North American Symptomatic Carotid Endarterectomy; pDMN = posterior part of DMN; PIQ = performance IQ; rs-fMRI = resting-state functional MRI; SFG = superior frontal gyrus; VIQ = verbal IQ; WAIS-III = Wechsler Adult Intelligence Scale III; WMS-R = Wechsler Memory Scale–Revised.SUBMITTED February 16, 2018. ACCEPTED July 19, 2018.INCLUDE WHEN CITING Published online December 14, 2018; DOI: 10.3171/2018.7.JNS18404.

Functional connectivity analysis and prediction of cognitive change after carotid artery stentingNaoki Tani, MD, PhD,1,2 Takahide Yaegaki, MD,1 Akio Nishino, MD, PhD,3 Kenta Fujimoto, MD, PhD,1 Hiroyuki Hashimoto, MD, PhD,1 Kaoru Horiuchi, MD,1 Mitsuhisa Nishiguchi, MD, PhD,1 and Haruhiko Kishima, MD, PhD2

1Department of Neurosurgery, Osaka General Medical Center, Osaka; 2Department of Neurosurgery, Osaka University Graduate School of Medicine, Suita; and 3Department of Neurosurgery, Hanwa Memorial Hospital, Osaka, Japan

OBJECTIVE The neurocognitive course of patients who have undergone cerebral revascularization has been the subject of many studies, and the reported effects of carotid artery stenting (CAS) on cognitive function have varied from study to study. The authors hypothesized that cognitive amelioration after CAS is associated with alteration of the default mode network (DMN) connectivity, and they investigated the correlation between functional connectivity (FC) of the DMN and post-CAS changes in cognitive function in order to find a clinical marker that can be used to predict the effect of cerebral revascularization on patients’ cognitive function in this preliminary exploratory study.METHODS The authors examined post-CAS changes in cognitive function in relation to FC in patients treated for uni-lateral carotid artery stenosis. Resting-state functional MRI (rs-fMRI) was performed with a 3-T scanner before and 6 months after CAS in 8 patients. Neuropsychological tests (Wechsler Adult Intelligence Scale III and Wechsler Memory Scale–Revised) were administered to each patient before and 6 months after CAS. The DMN was mapped for each patient through independent component analysis of the rs-fMR images, and the correlation between FC of the DMN and post-CAS change in cognitive function was analyzed on a voxel level. Multivariable regression analysis was performed to identify preoperative factors associated with a post-CAS change in cognitive function.RESULTS Post-CAS cognitive function varied between patients and between categories of neuropsychological tests. Although there was no significant overall improvement in Working Memory scores after CAS, post-CAS Working Memo-ry scores changed in negative correlation with changes in FC between the DMN and the precentral/superior frontal gy-rus and between the DMN and the middle frontal gyrus. In addition, the preoperative FC between those areas correlated positively with the post-CAS improvement in working memory.CONCLUSIONS FC between the DMN and working memory–related areas is closely associated with improvement in working memory after CAS. Preoperative analysis of FC of the DMN may be useful for predicting postoperative improve-ment in the working memory of patients being treated for unilateral stenosis of the extracranial internal carotid artery.Clinical trial registration no.: UMIN000020045 (www.umin.ac.jp/ctr/index.htm)https://thejns.org/doi/abs/10.3171/2018.7.JNS18404KEYWORDS carotid artery stenting; cognitive change; working memory; default mode network; functional connectivity; vascular disorders

J Neurosurg Volume 131 • December 2019 1709©AANS 2019, except where prohibited by US copyright law

Unauthenticated | Downloaded 10/19/21 01:51 AM UTC

Tani et al.

J Neurosurg Volume 131 • December 20191710

been the focus of many studies, and the reported effects of CAS on cognitive function have varied from study to study.12,26,37 The discrepancy is attributable to differences in several patient-related variables, e.g., age, cerebral he-modynamics, side of treatment, previous stroke, and de-gree of stenosis.15,18,24,27 Currently, it is not possible to pre-dict the effect of CAS on cognitive function.12

Resting-state functional MRI (rs-fMRI) is increasingly used to examine brain connectivity, so-called functional connectivity (FC), and its relation to cognition and behav-ior. In particular, rs-fMRI studies have elucidated changes in the default mode network (DMN) associated with cogni-tive impairment in several disorders, e.g., Alzheimer’s dis-ease, multiple sclerosis, traumatic brain injury, idiopathic normal pressure hydrocephalus, and attention deficit hy-peractivity disorder.5,6,17,20,23,32 Furthermore, recent studies have pinpointed the DMN alteration involved in cognitive decline in patients with cerebral vascular disease.8,10

We hypothesized that post-CAS improvement in cogni-tive function is associated with DMN alteration and that preoperative DMN connectivity is predictive of such im-provement in cognitive function. This preliminary, small-sample, prospective exploratory study was performed to investigate the correlation between a post-CAS change in cognitive function and the FC of the DMN in order to find a clinical marker that can be used to predict the effect of cerebral revascularization on patients’ cognitive function.

MethodsStudy Patients

The study group comprised 8 male patients ranging in age from 59 to 78 years (mean 69.3 ± 6.2 years) and in ed-ucation level from 9 to 12.5 years (mean 11.2 ± 1.4 years). These were testable patients with unilateral stenosis of the extracranial ICA who were being treated at the Osaka General Medical Center or Hanwa Memorial Hospital between August 2014 and April 2016. The patients were consecutive, except that no patient with a neurological co-morbidity, severe medical disorder, or functional disability was included in the study. The degree of patients’ carotid stenosis was assessed angiographically and recorded as described in the North American Symptomatic Carotid Endarterectomy Trial (NASCET).3 All 8 patients were right-handed, native Japanese speakers. Seven of the pa-tients had suffered a stroke more than 3 months before CAS and had only minor neurological sequelae; the other patient had no history of cerebral infarction.

The study was approved by the institutional review boards of Osaka General Medical Center and Hanwa Me-morial Hospital, and written informed consent for par-ticipation in the study was obtained from all 8 patients included. This study was registered with the UMIN Clini-cal Trials Registry, and its clinical trial registration no. is UMIN000020045 (www.umin.ac.jp/ctr/index.htm).

Neuropsychological TestingEach patient was seen by a certified, experienced neu-

ropsychologist who used the Japanese versions of both the Wechsler Adult Intelligence Scale III (WAIS-III)34 and the Wechsler Memory Scale–Revised (WMS-R)35 for evalua-

tion. These neuropsychological tests have been standard-ized for the Japanese population. The WAIS-III measures general intellectual functioning and yields a performance intelligence quotient (PIQ), verbal intelligence quotient (VIQ), and full-scale intelligence quotient (FIQ), along with 4 secondary indices, namely, the Verbal Comprehen-sion Index, Working Memory Index, Perceptual Organiza-tion Index, and Processing Speed Index. The WMS-R is used to assess general memory, verbal memory, delayed memory, visual memory, attention, and concentration. These neuropsychological tests were administered to each patient before and 6 months after CAS.

rs-fMRIrs-fMRI imaging was performed before and 6 months

after CAS at Hanwa Memorial Hospital. A 3-T scanner (Vantage Titan 3T, Toshiba Medical Systems Corp.) was used, and for the functional images, axial slices were ac-quired with a single-shot gradient echo-planar imaging sequence (TR 4000 msec, TE 25 msec, flip angle 90°, FOV 256 mm, voxel size 2 × 2 × 4 mm, 92 repetitions). Resting-state functional images were acquired for each patient, who was instructed to relax, stay awake, lie still, and think of nothing in particular while keeping his eyes open. The first 5 images in each run were discarded to al-low magnetization to reach equilibrium. A high-resolution T1-weighted structural image (TR 400 msec, TE 16.5 msec, flip angle 90°, FOV 256 mm, isotropic voxel size 0.5 mm) was acquired during the same session.

Image AnalysisImage preprocessing was performed with SPM12

(Wellcome Centre for Human Neuroimaging; http://www.fil.ion.ucl.ac.uk/spm/software/spm12/), which was de-signed to aid in the analysis of functional neuroimaging data. Functional images were realigned to the first echo-planar image. Rigid body transformation described by 6 estimated parameters was applied to correct for within-run head motion. The images were then spatially normal-ized to a standard template (Montreal Neurological Insti-tute [MNI]), resampled to 3-mm cubic voxels, and spa-tially smoothed by convolution with an isotropic Gaussian kernel (full width at half maximum = 8 mm). Structural images were segmented into individual white matter, gray matter, and cerebrospinal fluid masks. Data were bandpass filtered (0.008 Hz < f < 0.09 Hz), further processed, and corrected by means of the CONN functional connectivity toolbox (www.nitrc.org/projects/conn/). Sources of spu-rious variance, such as signals from white matter, cere-brospinal fluid (5 dimensions), and movement parameters extracted from the realignment process, were removed by linear regression. CONN incorporates an anatomical component-based noise correction method (aCompCor),4 which has been shown to effectively reduce physiological noise and noise from other sources in the blood oxygen level–dependent (BOLD) contrast signal.

To examine change in brain connectivity subsequent to CAS, we used the independent component analysis feature of CONN.36 Independent component analysis is a data-driven computational method that can be used to


J Neurosurg Volume 131 • December 2019 1711

Tani et al.

identify intrinsic brain networks7 and has been widely ap-plied to rs-fMRI data. We used independent component analysis to extract the DMN, which is thought to be the key brain network for cognitive function.2,25 We identified DMN components using group-level independent compo-nent analysis. Group-level DMN components (both spatial maps and time courses) were then back reconstructed for each subject. The time course of the DMN represents the average pattern of coherent brain activity in the network, and the intensity of this pattern of brain activity across the voxels is expressed in the associated spatial map.

Statistical AnalysisResults of the WAIS-III and WMS-R tests are shown

as mean ± standard deviation scores. Differences between pre- and post-CAS scores were analyzed by means of the Wilcoxon signed-rank test for paired samples. Statistical analyses were performed using JMP software (version 12.0, SAS Institute Inc.).

We then examined the correlation between change in the secondary WAIS-III indices (Verbal Comprehension, Working Memory, Perceptual Organization, and Process-ing Speed) and voxel-wise change in FC of the DMN. The statistical threshold was set to p < 0.005 uncorrected at the voxel level and p < 0.05 false discovery rate (FDR)-corrected at the cluster level. When a significant cluster was found, strengths of the correlation between the pre-operative FC and preoperative cognitive performance and the postoperative change in cognitive performance were assessed on the basis of Pearson’s correlation coefficient.

Correlates of post-CAS improvement in cognitive function were determined by means of logistic regression modeling. Clinical variables were first evaluated for mar-ginal association with post-CAS improvement in cogni-tive function (as assessed with the WAIS-III and WMS-R) by means of univariate logistic regression. Factors for which a significant (p ≤ 0.20) marginal association was found were then entered into a single multivariate logistic regression model.

Additionally, to clarify the effect of cerebral blood flow on cognitive function, we analyzed the relation between pre-CAS NASCET% stenosis and baseline cognitive function and between post-CAS change in NASCET% stenosis (pre-CAS NASCET% stenosis minus post-CAS NASCET% stenosis) and post-CAS improvement in cog-nitive function.

ResultsCAS Procedure and Postoperative Follow-Up

Starting 7 days before CAS, patients received oral an-tiplatelet drugs (aspirin 100 mg and clopidogrel 75 mg). After a local anesthetic was applied, CAS was performed with the aid of an embolic protection device and was suc-cessful in all 8 patients. Seven patients received a closed-cell stent (WALLSTENT, Boston Scientific), and 1 patient received an open-cell stent (Precise, Cordis). All 8 patients completed the 6-month WAIS-III testing, and 7 of the 8 patients completed the WMS-R testing.

The patient and clinical characteristics at the time of CAS are summarized in Table 1. CAS was performed on the left side in 6 patients and on the right side in 2. There were no significant differences in preoperative cogni-tive function between patients who underwent treatment on the left and those treated on the right side (Table 2). For all 8 patients, the postoperative course was unevent-ful, and none suffered a further cerebral ischemic event. Hyperperfusion syndrome did not develop in any patient.

TABLE 1. Clinical characteristics of study patients at the time of CAS

Pt Age (yrs), Sex Side of Tx Previous Stroke Previous CAS/CEA Education Level (no. of yrs) Pre-CAS Stenosis Post-CAS Stenosis

69, M Rt Bilat None 11 51% 8%67, M Lt None None 12 84% 0%63, M Rt Lt None 12 61% 6%70, M Lt Brainstem None 12 99% 24%74, M Lt Lt None 9 50% 26%78, M Lt Lt None 9 66% 26%59, M Lt Rt Rt CAS 12.5 77% 5%74, M Lt Lt None 12 71% 15%

CEA = carotid endarterectomy; Pt = patient; Tx = treatment.

TABLE 2. Preoperative cognitive performance per treatment side

Test Lt Side Treated

Rt Side Treated

p Value

WAIS-III FIQ 93.2 ± 17.6 93.5 ± 23.3 1.00 VIQ 94.7 ± 13.9 96.5 ± 20.5 1.00 Verbal Comprehension Index 89.3 ± 11.8 96.5 ± 14.8 0.62 Working Memory Index 101.2 ± 20.5 92.5 ± 13.4 0.74 PIQ 93.0 ± 19.7 92.0 ± 22.6 1.00 Perceptual Organization Index 95.2 ± 16.4 106.0 ± 24.0 0.50 Processing Speed Index 91.3 ± 18.0 78.0 ± 4.2 0.24WMS-R Verbal memory 53.8 ± 16.4 43.5 ± 36.1 1.00 Visual memory 45.8 ± 12.9 43.0 ± 18.4 0.74 General memory 99.07 ± 28.2 86.5 ± 54.4 0.87 Attention & concentration 58.7 ± 14.9 55.0 ± 9.9 1.00 Delayed recall 59.3 ± 10.8 51.0 ± 52.3 1.00

Values expressed as the mean ± standard deviation, unless indicated otherwise. WAIS-III results are expressed as indices, and WMS-R results are expressed as raw scores.


Tani et al.


Angiography performed at 6 months confirmed that no restenosis had occurred.

Postoperative Cognitive FunctionFIQ, VIQ, and Verbal Comprehension scores for all

patients increased significantly after CAS. Because the in-crease in scores varied widely between patients, the stand-ard deviations of the increases in scores were larger than the mean increases in scores (Table 3).

Correlation Between Change in Cognitive Function and Change in FC of the DMN

From the 20 independent components computed by means of independent component analysis, we extracted the anterior part of the DMN (aDMN) and posterior part of the DMN (pDMN; Fig. 1). Analysis of the correlation

between the change in FC of the pDMN at the voxel level and the increase in Working Memory scores revealed 2 suprathreshold clusters in the left frontal lobe (Fig. 2). One cluster was located in the precentral/superior frontal gyrus (SFG; MNI coordinates xyz = [-28, 10, 72], Brodmann areas 4 and 6) and comprised 241 voxels, and the other cluster was located in the middle frontal gyrus (MFG; MNI coordinates xyz = [-30, 28, 38], Brodmann area 8) and comprised 176 voxels. Two regions of interest from the suprathreshold clusters were identified for further sta-tistical analysis. The change in FC of the clusters in the

TABLE 3. Pre- and postoperative WAIS-III and WMS-R scores and change in scores

Test Before CAS 6 Mos After CAS Change p Value*

WAIS-III FIQ 93.3 ± 17.3 98.6 ± 20.6 5.38 ± 5.90 0.02 VIQ 95.1 ± 14.1 101.0 ± 16.9 5.88 ± 5.33 0.02 Verbal Comprehension Index 91.0 ± 11.9 97.0 ± 15.0 6.13 ± 5.22 0.01 Working Memory Index 99.0 ± 18.5 101.0 ± 19.5 2.00 ± 4.81 0.14 PIQ 92.8 ± 18.8 96.0 ± 20.5 3.25 ± 6.18 0.09 Perceptual Organization Index 97.9 ± 17.3 101.8 ± 20.5 3.88 ± 7.99 0.11 Processing Speed Index 88.0 ± 16.5 91.5 ± 13.9 3.50 ± 10.5 0.19WMS-R Verbal memory 53.3 ± 20.7† 64.6 ± 16.5† NA NA Visual memory 47.3 ± 12.4† 50.4 ± 14.4† NA NA General memory 100.6 ± 32.2† 114.4 ± 26.0† NA NA Attention & concentration 60.3 ± 12.0† 59.7 ± 13.8† NA NA Delayed recall 59.7 ± 22.7† 64.4 ± 25.5† NA NA

NA = not applicable.Boldface type indicates statistical significance.* Per the Wilcoxon signed-rank test.† Mean score for 7 of the 8 study patients.

FIG. 1. Anterior (left) and posterior (right) components of the DMN ex-tracted by means of independent component analysis. Both components included the precuneus, posterior cingulate cortex, bilateral lateral oc-cipital cortices, and medial prefrontal cortex.

FIG. 2. Areas in which the changes in the pDMN significantly correlated with the changes in working memory. One cluster was located in the precentral/SFG, and the other cluster was located in the MFG.



Tani et al.

left precentral/SFG and in the MFG correlated negatively with the change in the Working Memory score (p = 0.02, r = -0.78 and p < 0.05, r = -0.71, respectively). A signifi-cant correlation was not found between change in FC of the pDMN and any other test score, and neither was there a significant correlation between change in FC of the aDMN and change in any test score.

Correlation Between Preoperative FC, Preoperative Working Memory Scores, and Increases in Working Memory Scores

Preoperative FC between the pDMN and left precen-tral/SFG and between the pDMN and left MFG did not correlate with preoperative Working Memory scores but did correlate with postoperative increases in Working Memory scores (r = 0.93, p = 0.0008 and r = 0.81, p = 0.01, respectively; Fig. 3).

Preoperative FC of the pDMN–left precentral/SFG and of the pDMN–left MFG and age were associated at the p ≤ 0.20 level with the postoperative increase in Work-ing Memory scores (Table 4). When these variables were entered into a multivariate model, only the preoperative FC of the pDMN–precentral/SFG was shown to have an influence on the change in Working Memory scores. Be-cause of the strong correlation between preoperative FC of the pDMN-MFG and preoperative FC of the pDMN–pre-central/SFG (r = 0.91, p < 0.01), we excluded preoperative FC of the pDMN-MFG as a covariate in the multivariate analysis.

There was, however, no significant correlation between preoperative NASCET% stenosis and baseline Working Memory score (p = 0.37) and no significant correlation between change in NASCET% stenosis and improvement in Working Memory score (p = 0.72).

DiscussionIn this preliminary exploratory study utilizing WAIS-

III, WMS-R, and rs-fMRI, we examined the change in cognitive function after CAS in relation to FC of the DMN in patients with unilateral carotid artery stenosis. Improvement in cognitive function after CAS varied be-tween patients and between the types of cognitive func-tion tested (between subtests). Our main finding was that patients’ post-CAS working memory improved along with post-CAS decreases in FC of the pDMN–precentral/SFG and pDMN-MFG. Furthermore, preoperative FC of the pDMN–precentral/SFG and pDMN-MFG correlated posi-tively with the post-CAS improvement in working mem-ory.

Like other investigators,12,26,37 we found variations in the post-CAS improvement in cognitive function, both per patient and per type of function being tested. These varia-tions may be attributable to the different types of neuro-psychological tests used in the studies and to differences in background clinical factors, including patients’ cerebral hemodynamics.

In our patients, the post-CAS changes in FC of the

FIG. 3. Scatterplots for analysis of the correlation between preoperative working memory (WM) and preoperative FC in the precentral/SFG (A), between preoperative WM and preoperative FC in the MFG (B), between change in WM and preoperative FC in the precentral/SFG (C), and between change in WM and preoperative FC in the MFG (D). Working memory was scored on the basis of the WAIS-III Working Memory Index. A significant correlation was found between change in WM and preoperative FC in the precentral/SFG, and between change in WM and preoperative FC in the MFG.


Tani et al.


DMN–precentral/SFG and DMN-MFG correlated in-versely with post-CAS improvement in working memory. The precentral cortex, SFG, and MFG are activated dur-ing assessments that are relatively high in working memory load.9,11 The DMN appears to be activated when individuals are engaged in stimulus-independent thought, such as auto-biographical memory and imagining the future,1,25,30 and ac-tivation of the DMN typically decreases during external or attention-demanding tasks involving mental control.13,14,16,28 Therefore, the DMN is optimally inversely correlated with brain networks commonly activated for external tasks that demand attention and mental control, regions such as the precentral cortex, SFG, and MFG.13,22,30 The magnitude of the inverse correlation between the DMN and the working memory network has been shown to correlate with working memory performance.22,29 In our patients in whom FC of the DMN decreased and working memory improved after CAS, DMN FC with the precentral/SFG and MFG may have returned to its normal state along with the improve-ment in cerebral blood flow. As far as we know, this is the first report of such restoration of FC in conjunction with intervention-based recovery of working memory.

Importantly, we found that Working Memory scores improved in significant positive correlation with preopera-tive FC of the pDMN–precentral/SFG and pDMN-MFG; that is, working memory improved only in the patients in whom a positive correlation between FC of the DMN and that of the working memory–related area was seen before CAS. This correlation was independent of the preoperative Working Memory score as well as the side of treatment, patient age, and NASCET% stenosis. Therefore, an inverse correlation between FC of the DMN and the precentral/SFG and MFG could be an important predictive factor for improvement in working memory after CAS.

In this study, the pre-CAS NASCET% stenosis did not correlate with patients’ baseline Working Memory scores, and the change in NASCET% stenosis did not correlate with the post-CAS improvement in Working Memory scores. The small patient group may explain why we did not find a correlation between these variables, and we as-sume that background factors in addition to the severity of carotid artery stenosis, factors such as collateral blood flow

and peripheral arteriosclerosis, affected patients’ cognitive function.

We note several important study limitations. The first is that a power analysis was not performed to determine the precise sample size needed for testing our hypothesis. Rather, our study was conducted as a preliminary investi-gation in a small group of patients. The second limitation is that the size of the study might have prevented elucida-tion of any effect that the treatment side might have had on patients’ postoperative cognitive function. Further studies involving larger patient groups (e.g., multicenter studies) are needed to confirm our results. The third limitation is that our study did not include assessment of brain perfu-sion. The relation among FC, regional cerebral blood flow change, and cognitive function is very interesting, and re-sults of brain perfusion scanning should be included in a future study. In addition, whether a learning effect played a role in the increase in patients’ test scores after carotid revascularization was not ruled out. However, more than 6 months passed between CAS and reevaluation, so a learn-ing effect, if present, was probably minimal. Despite these limitations, we believe that our study results are impressive enough to substantiate the argument that rs-fMRI study has the potential to predict cognitive change after CAS.

In this preliminary study, we did not explore in depth the mechanism responsible for the cognitive decline or the mechanism responsible for the observed restoration of cognitive function. Longitudinal investigation of working memory that includes both preoperative and postoperative working memory and rs-fMRI assessment of FC in a large number of patients is warranted to confirm and expand upon our findings.

ConclusionsAlthough wide variation exists between individual pa-

tients in the degree to which cognitive function improves after CAS, FC between the DMN and the working memo-ry–related area correlates closely with the improvement in working memory function after CAS. Preoperative FC of the DMN could prove useful for the prediction of postop-erative improvement in working memory.

AcknowledgmentsThis work was supported by a grant for medical research from

Mitsui Life Social Welfare Foundation.

References 1. Andrews-Hanna JR, Saxe R, Yarkoni T: Contributions of epi-

sodic retrieval and mentalizing to autobiographical thought: evidence from functional neuroimaging, resting-state connec-tivity, and fMRI meta-analyses. Neuroimage 91:324–335, 2014

2. Anticevic A, Cole MW, Murray JD, Corlett PR, Wang XJ, Krystal JH: The role of default network deactivation in cog-nition and disease. Trends Cogn Sci 16:584–592, 2012

3. Barnett HJM, Taylor DW, Haynes RB, Sackett DL, Peerless SJ, Ferguson GG, et al: Beneficial effect of carotid endar-terectomy in symptomatic patients with high-grade carotid stenosis. N Engl J Med 325:445–453, 1991

4. Behzadi Y, Restom K, Liau J, Liu TT: A component based noise correction method (CompCor) for BOLD and perfusion based fMRI. Neuroimage 37:90–101, 2007

5. Bonnelle V, Leech R, Kinnunen KM, Ham TE, Beckmann

TABLE 4. Results of univariate and multivariate analysis of 8 factors with potential influence on working memory after CAS

Univariate Analyses

Multivariate Analysis

Preop FC of pDMN–pre-central/SFG

F = 39.30, p < 0.01 F = 48.68, p < 0.01

Preop FC of pDMN-MFG F = 11.53, p = 0.015 ExcludedAge F = 2.18, p = 0.19 F = 4.69, p = 0.08Side of Tx F = 1.20, p = 0.32Preop Working Memory

scoreF = 0.04, p = 0.85

NASCET% stenosis F = 0.03, p = 0.87NASCET% stenosis change F = 0.14, p = 0.72Education level F = 0.55, p = 0.49

Boldface type indicates statistical significance.



Tani et al.

CF, De Boissezon X, et al: Default mode network connectiv-ity predicts sustained attention deficits after traumatic brain injury. J Neurosci 31:13442–13451, 2011

6. Broyd SJ, Demanuele C, Debener S, Helps SK, James CJ, Sonuga-Barke EJ: Default-mode brain dysfunction in men-tal disorders: a systematic review. Neurosci Biobehav Rev 33:279–296, 2009

7. Calhoun VD, Adali T, Pearlson GD, Pekar JJ: A method for making group inferences from functional MRI data using independent component analysis. Hum Brain Mapp 14:140–151, 2001

8. Cheng HL, Lin CJ, Soong BW, Wang PN, Chang FC, Wu YT, et al: Impairments in cognitive function and brain con-nectivity in severe asymptomatic carotid stenosis. Stroke 43:2567–2573, 2012

9. Courtney SM, Petit L, Maisog JM, Ungerleider LG, Haxby JV: An area specialized for spatial working memory in hu-man frontal cortex. Science 279:1347–1351, 1998

10. Dacosta-Aguayo R, Graña M, Iturria-Medina Y, Fernández-Andújar M, López-Cancio E, Cáceres C, et al: Impairment of functional integration of the default mode network correlates with cognitive outcome at three months after stroke. Hum Brain Mapp 36:577–590, 2015

11. de Fockert JW, Rees G, Frith CD, Lavie N: The role of work-ing memory in visual selective attention. Science 291:1803–1806, 2001

12. De Rango P, Caso V, Leys D, Paciaroni M, Lenti M, Cao P: The role of carotid artery stenting and carotid endarterec-tomy in cognitive performance: a systematic review. Stroke 39:3116–3127, 2008

13. Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME: The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci U S A 102:9673–9678, 2005

14. Fransson P: Spontaneous low-frequency BOLD signal fluc-tuations: an fMRI investigation of the resting-state default mode of brain function hypothesis. Hum Brain Mapp 26:15–29, 2005

15. Fukunaga S, Okada Y, Inoue T, Hattori F, Hirata K: Neuro-psychological changes in patients with carotid stenosis after carotid endarterectomy. Eur Neurol 55:145–150, 2006

16. Greicius MD, Krasnow B, Reiss AL, Menon V: Functional connectivity in the resting brain: a network analysis of the default mode hypothesis. Proc Natl Acad Sci U S A 100:253–258, 2003

17. Greicius MD, Srivastava G, Reiss AL, Menon V: Default-mode network activity distinguishes Alzheimer’s disease from healthy aging: evidence from functional MRI. Proc Natl Acad Sci U S A 101:4637–4642, 2004

18. Grunwald IQ, Supprian T, Politi M, Struffert T, Falkai P, Krick C, et al: Cognitive changes after carotid artery stent-ing. Neuroradiology 48:319–323, 2006

19. Gurm HS, Yadav JS, Fayad P, Katzen BT, Mishkel GJ, Bajwa TK, et al: Long-term results of carotid stenting versus end-arterectomy in high-risk patients. N Engl J Med 358:1572–1579, 2008

20. Hawellek DJ, Hipp JF, Lewis CM, Corbetta M, Engel AK: Increased functional connectivity indicates the severity of cognitive impairment in multiple sclerosis. Proc Natl Acad Sci U S A 108:19066–19071, 2011

21. Johnston SC, O’Meara ES, Manolio TA, Lefkowitz D, O’Leary DH, Goldstein S, et al: Cognitive impairment and decline are associated with carotid artery disease in patients without clinically evident cerebrovascular disease. Ann In-tern Med 140:237–247, 2004

22. Keller JB, Hedden T, Thompson TW, Anteraper SA, Gabrieli JDE, Whitfield-Gabrieli S: Resting-state anticorrelations be-tween medial and lateral prefrontal cortex: association with working memory, aging, and individual differences. Cortex 64:271–280, 2015

23. Khoo HM, Kishima H, Tani N, Oshino S, Maruo T, Hosomi K, et al: Default mode network connectivity in patients with idiopathic normal pressure hydrocephalus. J Neurosurg 124:350–358, 2016

24. Kishikawa K, Kamouchi M, Okada Y, Inoue T, Ibayashi S, Iida M: Effects of carotid endarterectomy on cerebral blood flow and neuropsychological test performance in patients with high-grade carotid stenosis. J Neurol Sci 213:19–24, 2003

25. Leech R, Sharp DJ: The role of the posterior cingulate cortex in cognition and disease. Brain 137:12–32, 2014

26. Lin MS, Chiu MJ, Wu YW, Huang CC, Chao CC, Chen YH, et al: Neurocognitive improvement after carotid artery stent-ing in patients with chronic internal carotid artery occlusion and cerebral ischemia. Stroke 42:2850–2854, 2011

27. Mononen H, Lepojärvi M, Kallanranta T: Early neuropsycho-logical outcome after carotid endarterectomy. Eur Neurol 30:328–333, 1990

28. Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL: A default mode of brain function. Proc Natl Acad Sci U S A 98:676–682, 2001

29. Sala-Llonch R, Peña-Gómez C, Arenaza-Urquijo EM, Vidal-Piñeiro D, Bargalló N, Junqué C, et al: Brain connectivity dur-ing resting state and subsequent working memory task predicts behavioural performance. Cortex 48:1187–1196, 2012

30. Spreng RN, Grady CL: Patterns of brain activity supporting autobiographical memory, prospection, and theory of mind, and their relationship to the default mode network. J Cogn Neurosci 22:1112–1123, 2010

31. Sztriha LK, Nemeth D, Sefcsik T, Vécsei L: Carotid stenosis and the cognitive function. J Neurol Sci 283:36–40, 2009

32. Tian L, Jiang T, Wang Y, Zang Y, He Y, Liang M, et al: Al-tered resting-state functional connectivity patterns of anterior cingulate cortex in adolescents with attention deficit hyperac-tivity disorder. Neurosci Lett 400:39–43, 2006

33. Vilain KR, Magnuson EA, Li H, Clark WM, Begg RJ, Sam AD II, et al: Costs and cost-effectiveness of carotid stenting versus endarterectomy for patients at standard surgical risk: results from the Carotid Revascularization Endarterectomy Versus Stenting Trial (CREST). Stroke 43:2408–2416, 2012

34. Wechsler D (ed): Wechsler Adult Intelligence Scale, ed 3 (WAIS-III). San Antonio, TX: Psychological Corp, 1997

35. Wechsler D (ed): Wechsler Memory Scale–Revised Manu-al. San Antonio, TX: Psychological Corp, 1987

36. Whitfield-Gabrieli S, Nieto-Castanon A: Conn: a functional connectivity toolbox for correlated and anticorrelated brain networks. Brain Connect 2:125–141, 2012

37. Yoon BA, Sohn SW, Cheon SM, Kim DH, Cha JK, Yi S, et al: Effect of carotid artery stenting on cognitive function in patients with carotid artery stenosis: a prospective, 3-month follow-up study. J Clin Neurol 11:149–156, 2015

DisclosuresThe authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author ContributionsConception and design: Tani. Acquisition of data: Tani, Yaegaki, Nishino, Fujimoto, Hashimoto, Horiuchi, Nishiguchi. Analysis and interpretation of data: Tani, Yaegaki. Drafting the article: Tani. Critically revising the article: Kishima. Reviewed submitted version of manuscript: Yaegaki, Nishino, Fujimoto, Hashimoto, Horiuchi, Nishiguchi, Kishima. Administrative/technical/material support: Nishino. Study supervision: Kishima.

CorrespondenceNaoki Tani: Osaka University Graduate School of Medicine, Osaka, Japan. [email protected].


functional connectivity analysis and prediction of

Documents