Neuropsychiatric symptoms and brain morphology in patients with mild cognitive impairment and Alzheimer's disease with dementia

We present associations between neuropsychiatric symptoms (NPS) and brain morphology in a large sample of patients with mild cognitive impairment (MCI) and Alzheimer's disease with dementia (AD dementia). Several studies assessed NPS factor structure in MCI and AD dementia, but we know of no study that tested for associations between NPS factors and brain morphology. The use of factor scores increases parsimony and power. For transparency, we performed an additional analysis with selected Neuropsychiatric Inventory - Questionnaire (NPI-Q) items. Including regional cortical thickness, cortical and subcortical volumes, we examined associations between NPS and brain morphology across the whole brain in an unbiased fashion. We reported both statistical significance and effect sizes, using linear models adjusted for multiple comparisons by false discovery rate (FDR). Moreover, we included an interaction term for diagnosis and could thereby compare associations of NPS and brain morphology between MCI and AD dementia. We found an association between the factor elation and thicker right anterior cingulate cortex across MCI and AD dementia. Associations between the factors depression to thickness of the banks of the left superior temporal sulcus and psychosis to the left post-central volume depended on diagnosis: in MCI these associations were positive, in AD dementia negative. Our findings indicate that NPS in MCI and AD dementia are not exclusively associated with atrophy and support previous findings of associations between NPS and mainly frontotemporal brain structures.Objectives: Neuropsychiatric symptoms (NPS) are common in mild cognitive impairment (MCI) and Alzheimer's disease with dementia (AD dementia), but their brain structural correlates are unknown. We tested for associations between NPS and MRI-based cortical and subcortical morphometry in patients with MCI and AD dementia. Design: Cross-sectional. Settings: Conducted in Norway. Participants: Patients with MCI (n = 102) and AD dementia (n = 133) from the Memory Clinic and the Geriatric Psychiatry Unit at Oslo University Hospital. Measurements: Neuropsychiatric Inventory - Questionnaire (NPI-Q) severity indices were reduced using principal component analysis (PCA) and tested for associations with 170 MRI features using linear models and false discovery rate (FDR) adjustment. We also tested for differences between groups. For transparency, we added analyses with selected NPI-Q items. Results: PCA revealed four factors: elation, psychosis, depression, and motor behavior. FDR adjustment revealed a significant positive association (B = 0.20, p(FDR) < 0.005) between elation and thickness of the right caudal anterior cingulate cortex (ACC) across groups, and significant interactions between diagnosis and psychosis (B = -0.48, p(FDR) < 0.0010) on the left post-central volume and between diagnosis and depression (B = -0.40, p(FDR) < 0.005) on the thickness of the banks of the left superior temporal sulcus. Associations of apathy, anxiety, and nighttime behavior to the left temporal lobe were replicated. Conclusions: The positive association between elation and ACC thickness suggests that mechanisms other than atrophy underly elation. Interactions between diagnosis and NPS on MRI features suggest different mechanisms of NPS in our MCI and AD dementia samples. The results contribute to a better understanding of NPS brain mechanisms in MCI and AD dementia.