MULTISTAGE GRADING OF AMCI: THE ASSOCIATED BRAIN GRAY MATTER VOLUME AND COGNITIVE BEHAVIOR CHARACTERIZATION

Caishui Yang,Xuan Sun,Wuhai Tao,Xin Li,Junying Zhang,Jianjun Jia,Kewei Chen,Zhanjun Zhang
DOI: https://doi.org/10.1016/j.jalz.2016.07.100
2016-01-01
Alzheimer s & Dementia
Abstract:Amnestic mild cognitive impairment (aMCI) has been widely viewed as the transitional preclinical stage between normal aging and the onset of Alzheimer’s disease (AD). Increasing evidence points to the fact that current aMCI inclusion criteria actually cover a wide spectrum of different pathological stages, which might result in the imprecise depiction of biomarker trajectories and even failure of disease-modifying interventions. Further, refinement of the clinical course of aMCI is of great importance to determine the effective time window for early diagnosis and intervention evaluations. This study attempts to refine aMCI according to episodic memory function and explores the cognitive and neuroimaging profiles of these stages. In this cross-sectional study, 125 aMCI patients were classified into declined progressively three-stages of mild, moderate and severe (26.4%, 45.6% and 28%, respectively), utilizing the extreme groups approach based on episodic memory function. Fifty-two aMCI patients, in addition to 24 cognitively normal subjects, were included in subsequent structural MRI analyses. Characteristics of cognitive functions (including general cognitive function, episodic memory, processing speed, executive function and language function) and grey matter (GM) volume across these stages were explored through general linear models. Almost all of the non-memory cognitive functions showed progressive decline as memory function deteriorated. In addition, medial structures including the right hippocampus, right lingual and left fusiform gyrus, presented with greater GM atrophy during the later stages of aMCI (corrected p < 0.05). Moreover, non-linear trajectories over stages were found in execution function (sigmoidal model, p = 0.002), the fusiform gyrus volume (exponential model, p < 0.001), and the lingual gyrus and hippocampus volume (both quadratic models, p < 0.001). Correlations were identified between GM volume of the lingual gyrus and processing speed (r = 0.419, p = 0.003), and between the fusiform gyrus and general cognitive function (r = 0.281, p = 0.046). Our study characterized the cognitive and brain structures profiles along with the degree of episodic memory impairment, as defined by the three-part staging for aMCI patients. The non-linear progressive decline in cognitive functions and GM volumes both showed that our newly introduced three-stage definition for aMCI was informative and meaningful.
What problem does this paper attempt to address?