Predictors of Memory in Healthy Aging: Polyunsaturated Fatty Acid Balance and Fornix White Matter Integrity

Abstract

Recent evidence demonstrates that age and disease-related decline in cognition depends not only upon degeneration in brain structure and function, but also on dietary intake and nutritional status. Memory, a potential preclinical marker of Alzheimer's disease, is supported by white matter integrity in the brain and dietary patterns high in omega-3 and omega-6 polyunsaturated fatty acids. However, the extent to which memory is supported by specific omega-3 and omega-6 polyunsaturated fatty acids, and the degree to which this relationship is reliant upon microstructure of particular white matter regions is not known. This study therefore examined the cross-sectional relationship between empirically-derived patterns of omega-3 and omega-6 polyunsaturated fatty acids (represented by nutrient biomarker patterns), memory, and regional white matter microstructure in healthy, older adults. We measured thirteen plasma phospholipid omega-3 and omega-6 polyunsaturated fatty acids, memory, and regional white matter microstructure in 94 cognitively intact older adults (65 to 75 years old). A three-step mediation analysis was implemented using multivariate linear regressions, adjusted for age, gender, education, income, depression status, and body mass index. The mediation analysis revealed that a mixture of plasma phospholipid omega-3 and omega-6 polyunsaturated fatty acids is linked to memory and that white matter microstructure of the fornix fully mediates the relationship between this pattern of plasma phospholipid polyunsaturated fatty acids and memory. These results suggest that memory may be optimally supported by a balance of plasma phospholipid omega-3 and omega-6 polyunsaturated fatty acids through the preservation of fornix white matter microstructure in cognitively intact older adults. This report provides novel evidence for the benefits of plasma phospholipid omega-3 and omega-6 polyunsaturated fatty acid balance on memory and underlying white matter microstructure.

Keywords: healthy aging; memory; nutritional cognitive neuroscience; polyunsaturated fatty acids; white matter integrity.

Figure 1.. Proposed mediation model

The primary requirement for mediation is a significant indirect mediation effect, defined as the effect of the independent variable (nutrient biomarker pattern) through the mediation (fractional anisotropy in white matter regions) on the dependent variable (memory).

Figure 2.

Scree plot: inspection of the scree plot visually indicates which nutrient biomarker patterns explain the most variability in the data. A change in curvature, or inflection point, occurred after the third component, or nutrient biomarker pattern, was extracted. Thus, three components explained most variability in the data.

Figure 3.

Mediation path a: linear regression modeling showed that nutrient biomarker pattern 1 (LCPUFA) positively and reliably associated with fornix fractional anisotropy (=0.042, p<0.001).

Figure 4.

Mediation path c: linear regression modeling showed that nutrient biomarker pattern 1 (LCPUFA) positively and reliably associated with memory (=0.320, p=0.003).

Figure 5.

Mediation model statistics: nutrient biomarker pattern 1 (LCPUFA) positively associated with fractional anisotropy of the fornix (path a). LCPUFA positively associated with memory (path c). The indirect pathway of mediation (i.e., the effect of LCPUFA through fornix fractional anisotropy on memory; path a-b) was statistically significant. The direct pathway of mediation (i.e., the effect of LCPUFA on memory, accounting for fornix fractional anisotropy; path c’) was not significant. Therefore, fornix fractional anisotropy fully mediated the relationship between LCPUFA and memory.