Distinct network topology in Alzheimer's disease and behavioral variant frontotemporal dementia

Background

Alzheimer’s disease (AD) and behavioral variant frontotemporal dementia (bvFTD) cause distinct atrophy and functional disruptions within two major intrinsic brain networks, namely the default network and the salience network, respectively. It remains unclear if inter-network relationships and whole-brain network topology are also altered and underpin cognitive and social–emotional functional deficits.

Methods

In total, 111 participants (50 AD, 14 bvFTD, and 47 age- and gender-matched healthy controls) underwent resting-state functional magnetic resonance imaging (fMRI) and neuropsychological assessments. Functional connectivity was derived among 144 brain regions of interest. Graph theoretical analysis was applied to characterize network integration, segregation, and module distinctiveness (degree centrality, nodal efficiency, within-module degree, and participation coefficient) in AD, bvFTD, and healthy participants. Group differences in graph theoretical measures and empirically derived network community structures, as well as the associations between these indices and cognitive performance and neuropsychiatric symptoms, were subject to general linear models, with age, gender, education, motion, and scanner type controlled.

Results

Our results suggested that AD had lower integration in the default and control networks, while bvFTD exhibited disrupted integration in the salience network. Interestingly, AD and bvFTD had the highest and lowest degree of integration in the thalamus, respectively. Such divergence in topological aberration was recapitulated in network segregation and module distinctiveness loss, with AD showing poorer modular structure between the default and control networks, and bvFTD having more fragmented modules in the salience network and subcortical regions. Importantly, aberrations in network topology were related to worse attention deficits and greater severity in neuropsychiatric symptoms across syndromes.

Conclusions

Our findings underscore the reciprocal relationships between the default, control, and salience networks that may account for the cognitive decline and neuropsychiatric symptoms in dementia.

Quantitative analysis of grey matter degeneration in FTD patients using fractal dimension analysis

Fractal dimension (FD) is a quantitative parameter that can characterizes the complexity of human brain tissue. Extensive grey matter (GM) pathology has been previously identified in Frontotemporal dementia (FTD) and its variants. The aim of the present study was to investigate the GM morphometric abnormalities in the behavioral variant FTD (bvFTD) and primary progressive aphasia (PPA) using FD analysis. Twenty-seven bvFTD, 12 PPA and 20 controls were studied. SPM8 was used to segment the brain into GM tissue. Then the FD values were estimated for the GM skeleton, surface and general structure in patients and controls using our previously published algorithm. We found that patients with bvFTD had significant reduction in FD values of skeleton and general structure when compared to controls. In PPA, more significant decrease in FD was noted in the whole brain and left hemisphere skeleton along with left hemisphere general structure. Only the right hemisphere skeleton had a significant correlation with total score of Frontal Systems Behavior Scale (FrSBe). The results showed that the variants of FTD are associated with disease specific morphometric complexity patterns. These results indicate that FD can be used as a biomarker for the structural changes associated with neurodegenerative diseases.

Early stage of behavioral variant frontotemporal dementia: clinical and neuroimaging correlates

The early stages of behavioral variant frontotemporal dementia (bvFTD) are still not completely characterized. In a consecutive series of patients with probable bvFTD diagnosis confirmed by follow-up, we retrospectively evaluated the features at onset. Patients were reclassified according to presenting features and current diagnostic criteria into probable and possible bvFTD. The term "pre-bvFTD" was adopted for patients with cognitive and/or behavioral impairment not fulfilling bvFTD criteria and no deficits in activities of daily living. One hundred ninety-four subjects were included; at first visit, 70% (n = 136) patients were already classified as probable bvFTD. Of the remaining 30% (n = 58), 60% fulfilled criteria for possible bvFTD, while 40% did not, and were classified as pre-FTD. The neuropsychological pattern in possible bvFTD and pre-bvFTD was similar, although possible bvFTD showed more behavioral abnormalities. Pre-bvFTD subjects had frontotemporal gray matter atrophy, although less extensive than possible bvFTD. Conclusively, most bvFTD patients fulfill current diagnostic criteria at first admission, whereas a relatively small group is characterized by mild behavioral and/or cognitive abnormalities in spite of frontotemporal gray matter atrophy. Our preliminary findings will require a validation in prospective studies involving larger samples of patients.