Serum carotenoids and cerebral white matter lesions: The Rotterdam Scan Study

OBJECTIVES: To study the relation between serum levels of carotenoids and white matter lesions (WMLs) on magnetic resonance imaging (MRI). DESIGN: Evaluation of cross-sectional data from a cohort study. SETTING: The Rotterdam Scan Study. PARTICIPANTS: Two hundred and three nondemented older persons, age 60 to 90, from the Rotterdam Scan Study. MEASUREMENTS: Serum levels of carotenoids were determined. WMLs on MRIs were rated separately into periventricular and subcortical WMLs. Odds ratios (ORs) for the presence of severe WMLs (upper decile) were calculated per standard deviation (SD) increase in serum carotenoid level and per SD increase in overall carotenoid serum level. Effect modification by smoking status was studied through stratified analyses. RESULTS: Increasing levels of all the separate carotenoids were associated with less severe periventricular WMLs, which reached statistical significance for the overall carotenoid serum level (OR 0.4 per SD; 95% confidence interval (CI) = 0.2-0.9). We found no association between carotenoid levels and the presence of severe subcortical WMLs (OR 1.2 per SD; 95% CI = 0.7-2.0). The association of carotenoid levels with severe periventricular WMLs was more marked in those who ever smoked (OR 0.1 per SD; 95% CI = 0.0-0.9) than in those who had never smoked (OR 0.9 per SD; 95% CI = 0.4-2.1). CONCLUSIONS: These findings are compatible with the view that high levels of carotenoids may protect against WMLs in the periventricular region, in particular in smokers. Longitudinal studies with repeated measurements of both carotenoids and WMLs are necessary to explore this hypothesis further