Functional connectivity in mild cognitive impairment during a memory task: Implications for the disconnection hypothesis

Please download to get full document.

View again

of 11
4 views
PDF
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.
Document Description
Functional connectivity in mild cognitive impairment during a memory task: Implications for the disconnection hypothesis
Document Share
Document Transcript
  Functional Connectivity  in  Mild Cognitive Impairment During  a  Memory Task: Implications  for  the Disconnection Hypothesis Ricardo Bajo a   1  *, Fernando Maestú a   b   1 , Angel Nevado a   b , Miguel Sancho 0 , Ricardo Gutiérrez a , Pablo Campo a , Nazareth P. Castellanos a , Pedro Gil d , Stephan Moratti a   e , Ernesto Pereda f  and Francisco del-Pozo a a  Laboratory ofCognitive and Computational  Neuroscience,  Center of Biomedical  Technology, Madrid,  Spain b  Department ofBasic Psychology II, Complutense University of  Madrid,  Spain c  Department of Applied Physics III of Complutense University of  Madrid,  Spain d  Department of  Geriatrics  Memory Unii), San Carlos University  Hospital, Madrid,  Spain e  Department of Basic Psychology I, Complutense University of  Madrid,  Spain f  Electrical Engineering and Bioengineering Group, Dept. of Basic Physics, University of La Laguna, Tenerife, Spain Abstract.  Mild cognitive impairment (MCI) has been considered an intermediate state between healthy aging and dementia.  The early damage in anatomical connectivity and progressive loss of synapses that characterize early Alzheimer's disease suggest that MCI could also  be a  disconnection syndrome. Here,  we  compare  the  degree  of  synchronization  of  brain signals recorded with magnetoencephalography from patients (22) with MCI with that  of  healthy controls (19) during  a  memory task. Synchronization Likelihood,  an  index based  on  the theory  of  nonlinear dynamical systems, was used  to  measure functional connectivity. During the memory task patients showed higher interhemispheric synchronization than healthy controls between left and right -anterior temporo-frontal regions (in all studied frequency bands) and in posterior regions in the  7  band. On the other hand, the connectivity pattern from healthy controls indicated two clusters  of  higher synchronization, one among left temporal sensors and another  one among central channels. Both  of  them were found in all frequency bands. In the  7  band, controls showed higher Synchronization Likelihood values than  MCI  patients between central-posterior  and  frontal-posterior channels  and a  high synchronization  in posterior regions. The inter-hemispheric increased synchronization values could reflect  a  compensatory mechanism  for  the lack of efficiency  of  the memory networks  in  MCI patients. Therefore, these connectivity profiles support only partially the idea  of MCI  as a  disconnection syndrome,  as  patients showed increased long distance inter-hemispheric connections  but a  decrease  in anteroposterior functional connectivity. Keywords: Disconnection syndrome, functional connectivity, magnetoencephalography, memory, mild cognitive impairment, synchronization likelihood 1  These  two  authors have contributed equally to this work. Correspondence  to:  Ricardo Bajo, Laboratory  of  Cognitive and Computational Neuroscience Center for Biomedical Technology (CBT) Campus  de  Montegancedo 28660, Universidad Politécnica de Madrid, Spain. Tel.: +34 628909785; E-mail: ricbajo@gmail.com.   INTRODUCTION The disruption of anatomic and functional connectivity in the brain of Alzheimer's disease (AD) patients due to neurofibrillary pathology [1] has led to the idea of conceptualizing the cognitive symptomatology as a disconnection syndrome [2]. This connectivity impairment suggests the existence of abnormal interactions between neuronal systems [3]. Because the pathophysiological characteristics of the disease seem to begin even decades before the cognitive symptomatology, it is of interest to evaluate whether functional connectivity profiles are affected in early clinical conditions such as Mild cognitive impairment (MCI), as patients suffering from MCI present  a  higher risk of developing dementia (rate of conversion of 10-15% per year [4]). Whether the early anatomical connectivity impairment modulates the profiles of functional connectivity in MCI patients is still a matter of debate [5]. To describe how brain regions are coordinated to support higher cognitive functions, the term functional connectivity has been coined  ([6,7],  see [8], for a classification of different types of brain connectivity). Functional connectivity reflects the statistical interde-pendencies between two physiological signals, providing information about functional interactions between the corresponding brain regions. Long range synchronization between signals (oscillatory activity) srcinated in relatively distant neuronal populations, has been proposed as the mechanism for communication and integration of information in the brain  [9-11].  In fact, the binding phenomena in perception [12] or the formation of new memories [13] seems to be based on synchronization, at specific frequency bands, between two oscillatory time series which reflect activity from two brain regions. Previous fMRI studies using functional connectivity in MCI patients have shown decreases [14-16] and increases in functional connectivity values [15,16] in MCI patients as compared with healthy age-matched participants. However, although fMRI connectivity measures provide spatially-resolved information about connectivity patterns between brain regions, they do not directly reflect coupling between neuronal oscillators in different frequency bands known to play distinct roles in cognition (see [17] for an example on MCI). For this purpose various connectivity measures for magnetoen-cephalography (MEG)/electroencephalography (EEC) have been developed [18-20], as MEG/EEG signals provide a direct measure of neuronal activity with high temporal resolution. Here we use Synchronization Likelihood (SL) [21], an index which provides a nonlinear characterization of functional connectivity, to: 1) describe the synchronization topologies that support memory success in MCI patients and healthy aging participants and thus evaluate the disconnection hypothesis and 2) evaluate whether these synchronization topologies allow to  dif- ferentiate between MCI and age matched elderly controls. Moreover, the rationale for using SL in this study is two-fold: First, SL has been widely used as a functional connectivity measure in AD patients with both EEC [22,23] and MEG [24]. Second, it is a robust and nonlinear algorithm which overcomes the limitations of linear approaches. SL could be complementary  to  other measures of functional connectivity such as mutual information and phase synchronization. More importantly, functional connectivity measures could add  signif- icant diagnostic information to other biomarkers such as anatomical connectivity or measures of amyloid-/? (A/3) deposition and neurofibrillary tangles that probably represent the srcin of anatomical disconnection in AD.  Disconnection in  AD  is associated with an increase in local synchronization  [24].  The same finding in MCI would suggest that damage to anatomical connectivity already occurs at this stage. On the other hand, an increase in long distance synchronization would be compatible with the use of alternative networks. MATERIALS AND METHODS Participants Forty-one right handed, elderly participants recruited from the Geriatric Unit of  the  Hospital Universitario San Carlos, Madrid, participated in the study. Participants were divided into two groups based on their clinical profiles: twenty-two participants were classified as multi-domain MCI patients, and the other nineteen as healthy control volunteers. Three MCI recordings were excluded from further analysis due to an excessive noise level. MCI diagnosis was established according to the criteria proposed by Petersen ([4], see also [25]). To be diagnosed as having MCI, patients had to fulfill the following criteria: 1) report cognitive complaints corroborated by an informant (a person who stays with the patient for half a day at least 4 days a week); 2) objective cognitive impairment, documented by delayed recall as measured by the Logical Memory II subtest of the Wechsler Memory Scale Revised (score below 17  for 16 or more years of education; score below 9 for 8 to 15 years of education) and 1.5 Standard Deviations (SD) below mean in test of executive functions such as WCST or Stroop; 3) normal general cognitive function, as determined by the clinician's judgment based on a structured interview with the patient and an informant; 4) a MMSE score greater than 24; 5) relatively preserved daily living activities as measured by the Law-ton scale; and 6) not sufficiently impaired, cognitively and functionally to meet criteria for dementia. As a result twenty-two participants were included in the MCI group. According to their clinical and neuropsychological profile, all participants in this group were considered multi-domain MCI patients [4]. Nineteen age-matched, healthy elderly volunteers, without memory complaints, recruited for a project called "Aging with Health", at the San Carlos Hospital in Madrid consented to participate in the study. This group undergoes a complete medical revision every year. Patients and controls underwent a neuropsychological assessment, in order to establish their cognitive status in multiple cognitive functions. Specifically, memory impairment was assessed by the Logical Memory (immediate and delayed) from the Wechsler Memory Scale-III-R. Two scales of cognitive and functional status were also applied : the Spanish version of  the  Mini Mental State Examination (MMSE) [26], and the Global Deterioration Scale/Functional Assessment Staging CDS/FAST [27]. In order to avoid possible differences due to the years of education, patients and controls were chosen so that the resulting average number of years of education was similar:  10  years for patients and  11  years for controls. Table 1 summarizes the demographic and clinical information for both groups. The fact that our sample of MCI patients came from a memory clinic rather than from a population based sample could explain the high proportion of amnestic multidomian MCI. Before the MEG recording, all participants or legal representatives gave informed consent to participate in the study. The study was approved by the local ethics committee. Stimuli and task A modified version of the Sternberg's letter-probe task ([28,29]) was used. A set of five letters was presented and the participants were asked to keep the letters in mind. After the presentation of the five-letter set, a series of single letters (1000 ms in duration with a random ISI between 2-3 s) was presented one at a time, and the participants were asked to press a button Table 1 Age MMSE GPS LM1 LM2 Control 71.6 ±8 29.5 ±0.7 1 42.5 ± 8* 26.7 ± 7* MCI 74.8 ±3 27.7 ±1 3 19.1 ±  5  13.1 ± 6 MMSE, Mini-Mental State Examination; CDS, global deterioration scale; LM, logical memory. *p  < 0.0001 showing the differences in LM1 and LM2 between control and MCI cases. with their right hand when  a  member of the previous set was detected. The list consisted of 250 letters in which half were targets (previously presented letters), and the remaining letters were distracters (different from the previously presented letters). All participants completed a training session before the actual test, which did not start until the participant demonstrated that he/she could remember the five-letter set. Letters were projected through a LCD video-projector (SONY VPL-X600E), situated outside of the shielded-room onto a series of in-room mirrors, the last of which was suspended approximately 1 meter above the participant's face. The letters subtended 1.8 and 3 degrees of horizontal and vertical visual angle respectively. MEG data collection The MEG signal was recorded with a 254 Hz sampling frequency and a band pass of 0.5 to 50 Hz, using a 148-channel whole-head magnetometer (MAGNES® 2500 WH, 4-D Neuroimaging) confined in a magnetically shielded room. An environmental noise reduction algorithm using reference channels at a distance from the MEG sensors was applied to the data. Thereafter, single trial epochs were visually inspected by an experienced investigator, and epochs containing visible blinks, eye movements or muscular artifacts were excluded from further analysis. Artifact-free epochs from each channel were then classified into four different categories, according to the subject's performance in the experiment: hits, false alarms, correct rejections, and omissions. Only hits were considered for further analysis because we were interested in evaluating the functional connectivity patterns which support recognition success. 35 epochs were used to calculate SL values. This lower bound was determined by the participant with least epochs. To have an equal number of epochs across participants, 35 epochs were randomly chosen from each of the other participants. In the only previous SL-memory study we are aware of [30], a  1  minute time-window was used for SL anal ysis.  Such a long time-window makes it difficult to ensure the homogeneity of the cognitive processes in-  MCI  >  CTRL 1 AIP11AIH.1I 11,1 1 1 '1 • • BETAKU-25II2) . -<—Vi-Vfi i ^ i§i 7 ír-«  JL^*»** §V «in:  (23 - 33 Hi) I IUMU1 •?^|'*J' *»»•— JM'E SfimL •v. .15 nil,] l_ tfT tr^t^/  S \  ^ Fig. 1. Significant differences in SL between electrode pairs for different frequency bands. (MCI > Control). False Discovery Rate type I was applied. volved. Thus, it seems convenient to apply the traditional SL algorithm to a shorter time window. This will lead to an Event Related-SL (ER-SL) that achieves the stability of functional connectivity patterns across participants during the memory task. In-house Fortran code was used to implement the SL algorithm as described by [21]. The SL algorithm was applied to the 35 extracted artifact-free one second epochs for each subject. For each frequency band optimal SL parameter values were chosen according to [31 ] for each frequency band and one second length: Lag:L=f s /(3*HF), Embedding dimension:  M Theiler window:  Wl = 2 0.05, Window length:  W2 > 10/Pref 3 * HF / LF, L * (Ml), Pre/below Wl- 1. Where  f s  sampling rate, and HF and LF are the high and low frequency bound, respectively. The following frequency bands were considered: alpha 1 (al, 8-11 Hz), alpha2 («2, 11-14 Hz), betal 031,  14-25 Hz), beta2 (/3 2, 25-35 Hz), gamma (7, 35-45 Hz). The SL index was not computed for bands under 8 (Hz) as the epoch length and sampling rate do not allow an accurate enough estimation [31]. All epochs were digitally filtered off-line at the above frequency bands. Subsequently, the SL was calculated for each of the 35 one-second epochs with 148*147/2 channel pairs for each frequency band, un-filtered epochs, and each subject (19 controls and 19 patients). Statistical analysis To  compare the level of SL between the 2 groups, SL values were  first  averaged across epochs for each participant and channel  pair.  Then, False Discovery Rate [32, 33] was applied to find channel pairs with significant differences between groups. For each channel pair a between-groups Kruskal-Wallis (non-parametric) test was calculated. From the resulting p-values a  signif- icance threshold was calculated with a corresponding q =  0.2  (q =  0.4 in  a  band) using the type I false discovery rate implementation from [33]. Additionally, we analyzed similarities in the connectivity patterns of MCI patients and controls at the in-  MCI  <  CTRL ALPHA1(8-1I Hz) BETA (14-25 Hz) ALPHA2(ll-14Hz) BETA2(25-35Hz) GAMMA (35-45 Hz) Fig. 2. Significant differences in SL between electrode pairs for different frequency bands. (MCI < Control). False Discovery Rate type I was applied. dividual level. For each individual we report connectivity values 2 standard deviations above or below the average value across all channel pairs. RESULTS Beha vioral performance Behavioral performance during the memory task revealed no significant differences between groups, either with respect to the targets (total number of  hits,  misses and reaction time), or to the distracters (correct rejections, false alarms and reaction time). The percentage of hits (80% control group and 84% MCI group) and correct rejections (92% control group and 89% MCI group) was high enough in both groups, indicating that participants actively engaged in the task. MEG profiles of functional connectivity MCI> Control participants Comparing both groups (see Fig. 1), MCI patients showed a clear cluster of higher synchronization values over the anterior and central regions in all frequency bands. Additionally, MCI patients showed higher inter-hemispheric SL values than the control group between left and right temporo-frontal sensors in all frequency bands. Finally, only in 7 band, MCI subjects present a pattern of higher inter-hemispheric SL than controls in posterior regions. MCI  <  Control participants Two non-functionally related clusters of local interactions showed higher SL values in the control group: one among left temporal sensors and another one in central-posterior channels. Both of them were found in all frequency bands. Additionally, controls showed higher SL values, in 7 and  ¡32  bands, between central and posterior channels and between frontal and posterior regions. Finally, also in 7 band, there is a higher posterior synchronization in controls when compared to patients (see Fig. 2). Single-patient analysis: MCI and control participants We calculated the average SL for each participant. SL values above and below two standard deviation from
Similar documents
View more...
Search Related
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