Functional neuroimaging of healthy elderlies: correlation between resting-state functional magnetic resonance imaging and cognitive and demographic variables, and amyloid peptide deposition

Author(s)

Luís Gustavo Baptista e Ribeiro

• Document type: Master's Dissertation
• Press: São Paulo.
• Institution: Universidade de São Paulo (USP). Faculdade de Medicina (FM/SBD)
• Defense date: 2021-09-16

Examining board members

Geraldo Busatto Filho; Marcio Luiz Figueiredo Balthazar; Marcelo Camargo Batistuzzo

• Advisor: Geraldo Busatto Filho

Abstract

The comprehension of brain alterations due to aging becomes imperative in the global context of life expectancy increase. Available literature on brain functioning reports important alterations in functional connectivity (FC) during aging. Several studies, however, are restricted to certain brain areas when looking for those changes. The present study investigated associations between the whole-brain functional connectivity (FC) of a sample of healthy elderlies measured by resting-state magnetic resonance imaging (rs-fMRI) and -amyloid deposition indexes evaluated by the positron emission tomography after venous injection of the Pittsburgh compound B (11C-PiB PET). Furthermore, we investigated association to other relevant variables to the FC alterations process according to the literature, such as: the participants age and years of education; their scores on the Rey Auditory-Verbal Learning Test (RAVLT) as measure of episodic memory; and scores on the Mini Mental State Examination (MMSE) as measure of general cognition. The 26 healthy elderlies enrolled in this study underwent a thorough neuropsychological test battery to exclude subjects presenting neurological and psychiatric diseases, and structural magnetic resonance imaging to exclude any patients with gross brain lesions. A visual inspection of our samples 11C-PiB PET was carried out by two expert nuclear medicine physicians dichotomizing it by amyloid profile, thus dividing A- and A+ subgroups. We also obtained a semi-quantitative measure of amyloid deposition by calculating standard uptake values (SUV) of 11C-PiB PET in several brain regions, averaging them and normalizing the result by the cerebellar SUV (SUVr-PiB). After rs-fMRI acquisition and pre-processing, the participants FC data were parceled into 278 whole-brain contiguous regions that had their mean connectivity calculated. The mean connectivity of each region was then analysed in pairs, indicating whether the correlation was positive or negative in each pair, thus providing a FC panorama for each subject. Using the Partial Least Squares (PLS) method, we verified whether the FC differences between our subjects were associated with the subjects cognitive and demographic variables, and SUVr-PiB. In the case of statistically significant PLS analyses, we performed a new dichotomization of the sample to compare the mean connectivity of each subgroup, dividing our sample into younger and older elderlies. We categorized the differences between subgroups by the nature of the FC change (increases or decreases of positive or negative correlations) and mapped the results according to their location in the brain in one of seven functional networks. The visual inspection led to the identification of 5 subjects presenting important amyloid deposition (the A+ subgroup), placing them on the Alzheimers Disease (AD) continuum. We did not find statistically significant results in the analyses searching associations between whole-brain FC and SUVr-PiB, education, RAVLT or MMSE scores. We found significant association between FC and age. After dichotomization into younger and older elderlies subgroups, we found a complex pattern of FC alterations widespread across the whole brain. Both between- and intra-networks results were mostly categorized as FC decreases, but with very different patterns in each case. To a lesser extent, FC increases due to aging were also noted. Even though we did not find association between FC and amyloid deposition, it is notable that all but our five A+ subjects had close SUVr-PiB values. This indicates that bigger samples are warranted to augment statistical power and confirm this result. On the other hand, our results from the FC and age analysis are backed by the available literature, showing significant association even in a sample with small age difference between subjects, and demonstrating both FC decreases during aging and FC increases because of compensatory mechanisms (AU)

• FAPESP's process: 18/16610-6 - Functional neuroimaging in healthy elderlies: correlation between resting state functional magnetic resonance imaging and amyloid accumulation mapping via PiB-PET
• Grantee: Luís Gustavo Baptista e Ribeiro
• Support Opportunities: Scholarships in Brazil - Master