Association of Sickle Cell Trait with Measures of Cognitive Function and Dementia in African Americans

Background: The prevalence of cerebral small vessel disease (CSVD) and its associated complication of cognitive decline is significantly higher amongst African Americans than non-Hispanic Whites. Sickle cell anemia or sickle cell disease has been associated with a 30-45% increased prevalence of CSVD which presents as silent cerebral infarcts and impaired cognitive function. However, the association between sickle cell trait (heterozygosity for the sickle cell mutation) and cognitive decline or dementia has not been reported.

Hypothesis: African Americans with SCT will have a significantly higher incidence and prevalence of cognitive impairment and dementia compared to those without SCT.

Methods: We studied African Americans participants enrolled in the community-based prospective Atherosclerosis Risk in Communities (ARIC) study. SCT genotype status was determined using Taqman® genotyping from blood samples collected at baseline. Data from cognitive assessments at visits 2, 4 and 5, and an MRI performed at visit 5 were used for analysis. Using linear regression models for visit 2 cognitive measures and visit 5 brain MRI outcomes, a generalized estimating equation (GEE) for cognitive change, and Cox models for the incidence of dementia, we determined whether SCT was associated with a higher risk for cognitive dysfunction, global and regional brain volumes, and dementia.

Results: Distribution of traditional risk factors for cognitive decline were not significantly different between participants with SCT (N = 176) and those without SCT (N = 2,532). In multivariable, cross-sectional analyses of 2,708 participants, those participants with SCT compared to those without SCT did not show a statistically significant difference in the global or domain-specific cognitive function scores at baseline. Participants with SCT did not experience a faster 20-year cognitive decline compared to participants without SCT. Also, participants with SCT had larger parietal cortical volume (100.5 cm³ vs. 97.9 cm³, diff. = 2.67 (0.24, 5.11) cm³, p = 0.03), and lower incidence of dementia (HR = 0.63 95% CI = 0.38, 1.05) compared to those without SCT. Participants with a co-inheritance of the apolipoprotein E (APOE) ε4 risk allele and SCT (N = 63) had worse scores on the digit symbol substitution test (DSST) at baseline (z-score = -0.08 (-0.26, 0.09), P_interaction = 0.05) and over time (z-score = -0.12 (-0.38, 0.14), P_interaction = 0.04), compared to those with the APOE ε4 risk allele who do not have SCT (N = 113). SCT was associated with 2-fold increased risk of dementia among participants with diabetes mellitus and a 55% reduction in risk of dementia among those without diabetes mellitus (P_interaction = 0.01).

Conclusions: SCT was not an independent risk factor for prevalent or incident cognitive decline, but it could potentially interact with and modify other risk factors for dementia and cognitive dysfunction.