Background: Individual differences in the risk to develop dementia remain poorly understood. These differences may partly be explained through reserve, which is the ability to buffer cognitive decline due to neuropathology and age.

Objective: To determine how much early and late–life cognitive reserve (CR) and brain reserve (BR) contribute to the risk of dementia.

Methods: 4,112 dementia-free participants (mean age = 66.3 years) from the Rotterdam Study were followed up for on average 6.0 years. Early-life CR and BR were defined as attained education and intracranial volume, respectively. Late-life CR was derived through variance decomposition based on cognition. Late-life BR was set as the total non-lesioned brain volume divided by intracranial volume.

Results: Higher early-life CR (hazard ratio = 0.48, 95% CI = [0.21; 1.06]) but not early-life BR associated with a lower risk of incident dementia. Higher late-life CR (hazard ratio = 0.57, 95% CI = [0.48; 0.68]) and late-life BR (hazard ratio = 0.54, 95% CI = [0.43; 0.68]) also showed lower levels of dementia. Combining all proxies into one model attenuated the association between early-life CR and dementia (hazard ratio = 0.56, 95% CI = [0.25; 1.25]) whereas the other associations were unaffected. These findings were stable upon stratification for sex, age, and APOE ɛ4. Finally, high levels of late-life CR and BR provided additive protection against dementia.

Conclusion: The findings illustrate the importance of late-life over early-life reserve in understanding the risk of dementia, and show the need to study CR and BR conjointly.

, , , ,,
Journal of Alzheimer's Disease
Department of Radiology