Neurology 2014 Nov 15;83(20):1844-52. Epub 2014 Oct 15.
From the University of Bordeaux Ségalen (S.S., C.T., C.D., Y.Z., A.A., S.D.), INSERM U897 Neuroepidemiology, Bordeaux, France; Pekin Union Medical College Hospital (Y.Z.), China; Inserm U1061(C.B.), Montpellier; University Montpellier I (C.B.); University Pierre et Marie Curie-Paris 6 (A.A.); CNRS-CEA UMR5296 (F.C., B.M.), Université Bordeaux Segalen, Bordeaux; University of Versailles Saint-Quentin-en-Yvelines (S.D.); Department of Neurology (S.D.), Lariboisière Hospital, Paris; INSERM UMR S-1161 (S.D.), Paris 7 University, France; Department of Neurology (S.D.), Bordeaux University Hospital, Bordeaux; and Department of Neurology (S.D.), Boston University School of Medicine, Framingham Heart Study, Boston, MA.
Objectives: We examined the cross-sectional association between lipid fractions and 2 MRI markers of cerebral small vessel disease, white matter hyperintensity volume (WMHV) and lacunes, representing powerful predictors of stroke and dementia.
Methods: The study sample comprised 2,608 participants from the 3C-Dijon Study (n = 1,842) and the Epidemiology of Vascular Aging Study (EVA) (n = 766), 2 large French population-based cohorts (72.8 ± 4.1 and 68.9 ± 3.0 years; 60.1% and 58.4% women, respectively). Analyses were performed separately in each study and combined using inverse variance meta-analysis. Lipid fractions (triglycerides, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol) were studied as continuous variables. WMHV was studied both in a continuous and dichotomous manner, the latter reflecting the age-specific top quartile of WMHV (EXT-WMHV). Analyses were adjusted for age and sex.
Results: Increasing triglycerides were associated with larger WMHV in the 3C-Dijon Study (β ± SE = 0.0882 ± 0.0302, p = 0.0035), in the EVA Study (β ± SE = 0.1062 ± 0.0461, p = 0.021), and in the combined analysis (β ± SE = 0.0936 ± 0.0252, p = 0.0002) and with higher frequency of lacunes in the 3C-Dijon Study (odds ratio [OR] = 1.65 [95% confidence interval 1.10-2.48], p = 0.015), in the EVA Study (OR = 1.58 [95% confidence interval 0.93-2.70], p = 0.09), and in the combined analysis (OR = 1.63 [95% confidence interval 1.18-2.25], p = 0.003). Associations were attenuated but maintained after adjusting for other vascular risk factors or for inflammatory markers. Associations were present and in the same direction both in participants taking and those not taking lipid-lowering drugs but tended to be stronger in the former for EXT-WMHV. Increasing low-density lipoprotein cholesterol tended to be associated with a decreased frequency and severity of all MRI markers of cerebral small vessel disease in both studies.
Conclusions: Increasing triglycerides but not other lipid fractions were associated with MRI markers of cerebral small vessel disease in older community persons.