Background and purpose: The relationship between alcohol consumption and subclinical findings on magnetic resonance imaging (MRI) remains uncertain. We examined the relationship between light to moderate alcohol intake and silent brain infarction (SBI), white matter lesions (WMLs), and cerebral atrophy.
Methods: Cranial MRI was performed on subjects>or=40 years residing in a rural community in Japan (n=385; mean age, 67.2). Alcohol intake and type was determined using a detailed questionnaire; subjects were categorized into three groups: non-drinkers, light drinkers (<7 drinks per week), and moderate drinkers (>or=7 drinks per week). Former drinkers were considered non-drinkers. Periventricular WMLs, deep WMLs and cerebral atrophy were measured quantitatively using a computer-assisted processing system (%PVWML, %DWML, and %Brain, respectively).
Results: Compared with non-drinkers, the prevalence odds ratios for SBI were significantly higher in light and moderate drinkers, after multivariate adjustment. After adjusting for age, sex, and other related factors, the geometric mean %PVWML volumes in light and moderate drinkers were 1.27% and 1.52%, respectively, significantly larger than those for non-drinkers (0.95%). The geometric mean %DWML volume in light drinkers was 0.10%, which was larger than the value for non-drinkers (0.06%); the value for moderate drinkers (0.13%) was significantly larger than that for non-drinkers. The geometric mean %Brain values for non-, light, and moderate drinkers were 92.1, 91.9 and 90.8%, respectively; a statistically significant difference was found between non-drinkers and moderate drinkers.
Conclusions: The present study indicates that regular drinking, including even low levels of consumption, may be a risk factor for subclinical findings detected on MRI in community-dwelling Japanese people.