Sleep duration is associated with white matter hyperintensity volume in older adults: the Northern Manhattan Study

Self-reports of long or short sleep durations have indicated an association with cardiovascular morbidity and mortality, but there are limited data evaluating their association with white matter hyperintensity volume (WMHV), a marker of cerebral small vessel disease. We conducted a cross-sectional analysis of self-reported sleep duration to test for a correlation with white matter hyperintensities, measured by quantitative magnetic resonance imaging (MRI), in the Northern Manhattan Study. We used multivariable linear regression models to assess associations between both short (<6 h) and long (≥9 h) sleep durations and log-transformed WMHV, adjusting for demographic, behavioural and vascular risk factors. A total of 1244 participants, mean age 70 ± 9 years, 61% women and 68% Hispanics were analysed with magnetic resonance brain imaging and self-reported sleep duration. Short sleep was reported by 23% (n = 293) and long sleep by 10% (n = 121) of the sample. Long sleep (β = 0.178; P = 0.035), but not short sleep (β = -0.053; P = 0.357), was associated with greater log-WMHV in fully adjusted models. We observed an interaction between sleep duration, diabetes mellitus and log-WMHV (P = 0.07). In fully adjusted models, stratified analysis showed that long sleep duration was associated with greater WMHV only in those with diabetes (β = 0.78; P = 0.0314), but not in those without diabetes (β = 0.022; P = 0.2), whereas short sleep was not associated with white matter hyperintensities in those with or without diabetes. In conclusion, long sleep duration was associated with a greater burden of white matter lesions in this stroke-free urban sample. The association was seen mainly among those with diabetes mellitus.