Premium
IC‐P‐153: Whole cortical and DMN mean functional connectivity as potential biomarkers for mild Alzheimer's disease
Author(s) -
Balthazar Marcio,
Campos Brunno,
Franco Alexandre,
Damasceno Benito,
Cendes Fernando
Publication year - 2013
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2013.05.150
Subject(s) - default mode network , posterior cingulate , audiology , dementia , psychology , neuropsychology , clinical dementia rating , alzheimer's disease , cognition , medicine , disease , neuroscience
Background: The search for an Alzheimer’s disease (AD) biomarker is one of the most relevant contemporary research topics due to high prevalence and social costs of the disease. Functional connectivity alterations of the Default Mode Network (DMN) are a plausible candidate for such task, since AD dementia is among the main conditions that can disrupt the functional organization of cognitive functions. Methods: We evaluated 22 patients with mild AD (mean age: 73.4 years-old; 15 women) and 26 healthy controls (mean age: 71.03 years-old; 19 women) matched for gender and age. All subjects underwent a 10 minutes task-free fMRI at 3.0T MRI PHILIPS InteraAchieva scanner. Images were pre-processed by applying slice-time and motion corrections algorithms and removing linear trends in these data. Data pre-processing also included smoothing with a 6 mm FWHM Gaussian kernel, bandpass filtering (0.008 to 0.1 Hz) and spatially normalized to standard space (MNI152). Six parameters of head motion as well as cerebrospinal fluid and white matter time series were regressed as nuisance variables. To identify the DMN, seed-based functional connectivity was calculated by placing a seed in the posterior cingulate cortex (PCC; 0,-51,15; seed radius 1⁄4 3 mm), and later calculated a Fisher’s r-to-z transformation to obtain whole cortical statistical z-scoremaps. Then, we calculated an average z-scorefor each subject, and compared the results between the groups by using an independent two-sample t-test. We also calculated the sensitivity and specificity of the method, as well as a ROC curve analysis. Results: We found a significant statistical difference between mild AD and controls considering positive average z scores (z score > 0, p 1⁄4 0.005) and absolute z-scores values (p1⁄4 0.01). Considering individual values, we found a sensitivity1⁄4 76.2% and a specificity1⁄4 72%, when using a cutoff value1⁄4 0.195. Area under ROC curve 1⁄4 0.794, standard error 1⁄4 0.67. Conclusions: We showed that individual measures of whole cortex connectivity in relation to DMN’s PCC could be considered a promising method to differentiate AD, even at an early phase, from normal aging. Further studies with large sample size as well as validation of normal values are needed for safer conclusions.