Advertisement

Anxiety is associated with brain amyloidosis in cognitively normal and mild cognitive impairment subjects: A [18F]flutemetamol PET study

      Background

      Neuropsychiatric behaviors are prominent features of Alzheimer’s disease (AD) and mild cognitive impairment (MCI). Their association with brain amyloidosis is poorly understood.

      Methods

      65 subjects (27 NC and 38 MCI) who provided yearly neuropsychiatric inventory (NPI) data received [18F]Flutemetamol PET scan 4.6 (±1.0) year into the study. In our analyses neuropsychiatric symptoms were coded as present if present at one or more visits and absent if never reported by the caregiver. Mean standard uptake volume ratios (SUVR) using the whole cerebellum as the reference were calculated. Brain amyloidosis was defined as mean SUVR≥1.27. We used Chi-square statistics to compare neuropsychiatric symptom prevalence between the amyloid positive and amyloid negative groups. Next we examined the association between amyloid status and neuropsychiatric symptoms using logistic regression with fixed effects for baseline diagnosis, amyloid status, their interaction and cognitive decline (defined as transition from NC to MCI or dementia and transition to MCI to dementia in follow up).

      Results

      There were no significant differences in baseline diagnosis, age, sex, education, follow-up duration or time to PET scan between the amyloid positive and amyloid negative groups. Amyloid positive individuals had significantly higher rates of pre-existing anxiety (46 vs. 10%, p=0.0009) and apathy (42 vs. 20%, p=0.05) relative to amyloid negative subjects. Presence of anxiety remained significantly associated with brain amyloidosis after adjusting for baseline diagnosis and longitudinal decline (β=2.5, p=0.05). NC who tested amyloid positive in follow-up were 12 times more likely to develop anxiety compared to amyloid negative NC. MCI who tested amyloid positive in follow-up were 4 times more likely to develop anxiety compared to amyloid negative MCI.

      Conclusions

      We found a strong association between amyloid pathology and pre-existing anxiety in both our cognitively normal and MCI subjects. Our analyses suggest that NC subjects develop anxiety at a minimum at the time of but potentially even prior to reaching the threshold for amyloid positivity and suggest that anxiety might be one of the first clinical correlates of brain amyloidosis.