Abstract
Background
Current models of preclinical Alzheimer's disease (AD) have been developed from the neuropsychological perspective for detecting dementia. However, as preclinical AD occurs in cognitively normal older individuals the magnitude and importance of changes detected is difficult to determine and differentiate from normal cognitive aging. The aim of this study was to examine the nature of amyloid-β (Aβ) related cognitive change using models of cognition used to investigate cognitive aging, so as to provide a context for understanding cognitive changes in preclinical AD.
Methods
Older adults (n = 335) from the Australian Imaging Biomarkers and Lifestyle (AIBL) study, who were CN, had no severe or uncontrolled systemic, cardiovascular, or psychiatric conditions at enrolment and did not progress to MCI or dementia over 72 months were included in the present study. All participants underwent PET amyloid imaging and were classified as either Aβ+ (SUVR ≥ 1.4; n = 58) or Aβ− (n = 277). All participants underwent comprehensive clinical and cognitive assessment at least twice across 72-months.
Results
Principal Axis Factoring identified a fluid intelligence factor from the neuropsychological test battery. The Wechsler Test of Adult Reading – Revised (WTAR) was also included as an indicator of crystallized intelligence. Linear Mixed Model (LMM) analyses indicated that the Aβ+ group declined over 72 months while Aβ- group remained stable on the fluid intelligence factor. This difference in rate of change between groups was small in magnitude (Cohen's d = -0.17). Both groups did not show decline and there was no difference between groups for rate of change for crystallized intelligence.
Conclusions
In cognitively normal older adults who do not progress to MCI or dementia over 72 months, Aβ+ is associated with subtle decline in fluid intelligence, whereas Aβ- is associated with stability in fluid intelligence. Thus, decline in fluid intelligence in cognitively normal older adults may be a marker of early preclinical AD rather than representing normal cognitive aging as has been previously hypothesized.