Abstract P063: Cigarette Smoking and Neuroimaging Indices of Cerebrovascular Health in a Population-based Cohort of Middle-aged Adults

Abstract

Introduction: Several studies report an association between cigarette smoking and dementia. One proposed hypothesis is that smoking leads to adverse cognitive and brain changes through vascular pathways. To date, few data exist on the link between smoking and vascular brain measures. We examined, in a population-based middle-aged cohort, the association of history of cigarette smoking with three indicators of cerebrovascular health: (i) white matter (WM) integrity measured using WM-fractional anisotropy, (ii) abnormal WM volume, and (iii) cerebral blood flow (CBF) in the gray matter. We focused on lobar measures of these indicators to detect whether specific regional cerebrovascular indicators are differently related to smoking. Hypothesis: We assessed the hypothesis that smoking history is associated with more adverse cerebrovascular measures (lower fractional anisotropy, larger abnormal WM volumes, and lower CBF). Methods: Data come from the Coronary Artery Risk Development in Young Adults CARDIA-brain magnetic resonance imaging sub-study (n=538). Brain measures and self-reported smoking history were obtained at the 25th year follow-up (mean age=50). We used linear (for CBF and WM-fractional anisotropy) and multinomial logistic (for abnormal WM volume) regression models, adjusted for potential confounders (including vascular and lifestyle risk factors and medical conditions). Results: Compared to never-smokers, current smokers had lower WM-fractional anisotropy in all brain lobes (-0.005 lower fractional anisotropy in the occipital lobe and -0.006 in the other lobes; p<0.05). Smoking history was also associated with larger abnormal WM volumes: Compared to never-smokers, smokers and/or former-smokers had patterns of larger abnormal WM volumes in all lobes, with pronounced associations observed in the frontal lobe wherein smokers and former-smokers had up to twice the risk of having larger abnormal frontal WM volumes as never-smokers (p<0.05). Smoking history was associated with lower CBF in the occipital (-2.89 ml/100g/min (95%CI=-5.43, -0.36); p=0.03) and parietal lobes (-3.31 (95%CI=-5.91, -0.71); p=0.01). Conclusions: In this cohort of middle-aged adults, associations between smoking and adverse WM measures were observed in all brain lobes, whereas associations of smoking with CBF were localized in the occipital and parietal lobes. This localization may reflect a link between smoking and blood flow processes related to sensory-cognitive networks. In conclusion, results suggest that smoking is associated with three early-indicators of cerebrovascular burden (WM-fractional anisotropy, WM abnormal volume, and CBF) across the brain and that it may be differentially related to different regional cerebrovascular indicators.