Risk Factors, Neuroanatomical Correlates, and Outcome of Neuropsychiatric Symptoms in Alzheimer's Disease

Background: An integrative model of neuropsychiatric symptoms (NPS) in Alzheimer's disease (AD) is lacking. Objective: In this study, we investigated the risk factors, anatomy, biology, and outcomes of NPS in AD. Methods: 181 subjects were included from the Alzheimer's Disease Neuroimaging Study (ADNI). NPS were assessed with the Neuropsychiatric Inventory Questionnaire at baseline and 6 months. NPI > 3 was used as a threshold for NPS positivity. Three NPS courses were characterized: 1) minimal/absent (negative at 0 and 6 months, n = 77); 2) fluctuating (positive only at one time point, n = 53); 3) persistent (positive at both time points, n = 51). We examined the association between NPS course and family history of dementia, personal history of psychiatric disorders, cerebrospinal fluid biomarkers, atrophy patterns, as well as longitudinal cognitive and functional measures at 12 and 24 months (MMSE, CDR-SOB, FAQ). Results: AD subjects with absent, fluctuating, or persistent NPS had similar CSF amyloid-beta and tau levels. AD subjects with minimal/absent NPS had less personal history of psychiatric disorders (35%) than those with fluctuating (57%; p = 0.015) or persistent NPS (47%, not significant). At 24 months, AD subjects with persistent NPS had worse cognitive (MMSE; p = 0.05) and functional (CDR-SOB; p = 0.016) outcomes. Dorsolateral prefrontal atrophy was seen in persistent NPS, but not in fluctuating NPS. Conclusions: Our results suggest that individuals with personal history of psychiatric disorders might be more vulnerable to develop NPS throughout the course of AD. The worst cognitive and functional outcomes associated with NPS in AD underscores the importance of monitoring NPS early in the disease course.