Objectives: Neuropsychiatric symptoms (NPS) are common in mild cognitive impairment (MCI) and even more so in Alzheimer's disease (AD). The affective/vegetative NPS cluster model (sleep disorders, depression, appetite changes, anxiety, and apathy) has been asso-ciated with an increased risk of dementia in patients with MCI and these five NPS have common neuroanatomical associations. Hence, in this study, we examined how brain morphology is influenced by the severity of affective/vegetative NPS across different stages of cognitive performance. Participants: 175 AD, 367 MCI and 223 cognitively normal (CN) participants. Setting: Participants were recruited at multiple centers in North America included in the ADNI project. Design: A GLM was established to test for intergroup differences (CN -MCI, CN-AD, AD-MCI) of the effects of the five NPS on brain structures. A regression model was also per-formed to show slope directionality of the regions of interest as NPS severity increases. Measurements: 3T MRI data (cortical volumes, areas and thickness) and severity scores of the five NPS. Results: Associations within AD were predominantly stronger compared to MCI. Increased severity of sleep disorders and appetite changes were associated with a decrease in frontal surface areas in AD. Furthermore, increased severity of all NPS (except apathy) were asso-ciated with changes in the temporal regions, predominantly with decreased volumes and sur-face areas. Conclusion: These findings show the implication of fronto-temporal regions with sleep disorders, depression and appetite changes, and contribute to a better understanding of brain morphological differences between CN, MCI and AD with respect to all five NPS.

Objective Neuropsychiatric symptoms (NPS) are common in mild cognitive impairment (MCI) and even more in Alzheimer's disease (AD). The symptom‐based cluster including nighttime disturbances, depression, appetite changes, anxiety, and apathy (affective and vegetative symptoms) was associated with an increased risk of dementia in MCI and has common neuroanatomical associations. Our objective was to investigate the differences in brain morphology associations with affective and vegetative symptoms between three groups: cognitively normal older adults (CN), MCI and AD. Material and Methods Alzheimer's Disease Neuroimaging Initiative data of 223 CN, 367 MCI and 175 AD, including cortical volumes, surface areas and thicknesses and severity scores of the five NPS were analyzed. A whole‐brain vertex‐wise general linear model was performed to test for intergroup differences (CN‐MCI, CN‐AD, AD‐MCI) in brain morphology associations with five NPS. Multiple regressions were conducted to investigate cortical change as a function of NPS severity in the AD‐MCI contrast. Results We found (1) signature differences in nighttime disturbances associations with prefrontal regions in AD‐MCI, (2) signature differences in NPS associations with temporal regions in AD‐MCI for depression and in CN‐AD for anxiety, (3) decreased temporal metrics in MCI as nighttime disturbances and depression severity increased, (4) decreased pars triangularis metrics in AD as nighttime disturbances and apathy severity increased. Conclusion Each NPS seems to have a signature on brain morphology. Affective and vegetative NPS were primarily associated with prefrontal and temporal regions. These signatures open the possibility of potential future assessments of links between brain morphology and NPS on an individual level.

BackgroundThe neuroanatomy of hyperactive neuropsychiatric symptoms (NPS) is poorly understood, and it is unclear whether these symptoms result from the same pathogenic processes responsible for cognitive decline or if they have an independent etiology to the neurodegeneration due to Alzheimer’s disease (AD).ObjectiveWe aim to investigate how the severity of hyperactive NPS (i.e., agitation, disinhibition, and irritability) can impact brain structures at different stages of cognitive decline.MethodsNeuropsychiatric and 3T MRI data from 223 cognitively normal (CN) participants, 367 participants with mild cognitive impairment (MCI) and 175 participants with AD were extracted from the Alzheimer Disease Neuroimaging Initiative (ADNI) database. Statistical analyses based on the general linear model (GLM) were performed to define the effects of neuropsychiatric variables on brain structures in a two-by-two comparison (AD-MCI, CN-MCI and CN-AD). Linear regression analysis was also performed to investigate cortical changes as a function of NPS severity.ResultsIn the AD group, the frontal dorsolateral is the most influenced region receiving an impact from more severe agitation, disinhibition and irritability. In AD, agitation and irritability influence some temporal inferior and parietal superior regions. Furthermore, severe disinhibition seems to have a stronger influence on CN participants compared to the other two groups, particularly in the occipital lingual, frontal middle rostral and frontal pars triangularis regions.ConclusionOur study shows that hyperactive NPS influence differently the brain morphology at different stages of cognitive performance. This might imply that their severity should be evaluated in relation to results of cognitive evaluations.