Résumé
While statins are routinely prescribed to prevent cardiovascular diseases, their effects on brain alterations remain largely unknown. Very few studies have examined the differences in brain volumes between statin users and non-users, and existing research has yielded inconsistent results. This cross-sectional study aims to investigate the association between statin use at baseline and global and specific brain volumes measured 9 years later in a large population-based sample of middle-aged and older adults. Participants from the UK Biobank without neurological and psychiatric disorders consisted of 3285 statin users (mean 60 years and 69% males) and 36 229 non-users (mean 55 years and 46% males). We used linear models to estimate the mean volumetric differences between statin users and non-users while adjusting for UK Biobank assessment centre, age, sex, ethnicity, education, apolipoprotein E epsilon 4 status, Townsend deprivation index, antidepressant use, intracranial volume, lifestyle factors (alcohol intake frequency, smoking and physical activity) and health-related conditions (body mass index, blood pressure, diabetes, coronary heart disease, stroke, head injury, depression and insomnia). Moreover, mediation analysis was performed to evaluate whether the association between statin use and global brain volumes was mediated by total serum cholesterol concentration. Statin use was associated with lower grey matter volume [(beta = -1575 mm(3) (-2358, -791)], with 20% of this association mediated by total serum cholesterol concentration. Statin use was also associated with lower peripheral cortical grey matter volumes [(3 = -1448 mm3 (-2227, -668)] and higher white matter hyper- intensity [(beta = 0.11 mm(3) (0.07, 0.15)]. However, white matter volume did not differ significantly between statin users and non-users. Further analyses revealed that volumes of thalamus, pallidum, hippocampus, nucleus accumbens and other regions of the temporal lobe were smaller among statin users compared with non-users. This study showed that statin use is associated with higher white matter hyperintensity volumes and lower total and peripheral cortical grey matter volumes 9 years later, indicative of the brain's ageing process. Moreover, the observed grey matter alterations were partially explained by statin-induced total serum cholesterol reduction. This study emphasizes the potential direct and indirect effects of statins on brain volume.