New Data on Alcohol and the Brain

Lauri L. Barclay, MD


July 06, 2017

Moderate Alcohol Consumption as Risk Factor for Adverse Brain Outcomes and Cognitive Decline: Longitudinal Cohort Study

Topiwala A, Allan CL, Valkanova V, et al
BMJ. 2017;357:j2353


The reported health effects of drinking alcohol are varied, with clear deleterious effects of heavy drinking on the brain, liver, and other organ systems, and some suggestion of benefit at low levels of consumption.

Little previous research has targeted the impact of moderate drinking on the brain, and findings have been inconsistent. The goal of this observational cohort study was to examine the association of moderate alcohol consumption with brain structure and function.

From 1985 to 2015, 550 community dwelling adult participants in the UK Whitehall II cohort imaging substudy underwent repeated measurements of cognitive performance and weekly alcohol intake, as well as multimodal MRI at study endpoint. At baseline, mean age was 43.0 ± 5.4 years, and no participants were alcohol-dependent, on the basis of the CAGE screening questionnaire. Incomplete or poor-quality MRI or clinical data or gross structural abnormality led to exclusion of 23 participants.

In a dose-dependent manner, higher alcohol consumption was associated with increased risk for hippocampal atrophy during the 30-year follow-up, after adjustment for confounding factors that included age, sex, education, social class, physical and social activity, smoking, stroke risk, and medical history.

Compared with abstainers, participants who drank more than 30 units/week had the highest risk (odds ratio [OR], 5.8; 95% confidence interval, 1.8 to 18.6; P ≤ .001). Even moderate drinkers (14 to < 21 units/week in men; 14 units was equivalent to four pints of strong beer or five large glasses of wine) had more than threefold the risk for right-sided hippocampal atrophy compared with abstainers (OR, 3.4; 95% CI, 1.4 to 8.1; P = .007), Light drinking (1 to < 7 units/week) did not protect against hippocampal atrophy.

Other risks associated with higher alcohol consumption were differences in corpus callosum microstructure, reduced gray-matter density, reduced white-matter microstructural integrity, and faster decline in lexical fluency, but there were no associations with cross-sectional cognitive performance at the time of MRI or longitudinal changes in semantic fluency or word recall.


Study limitations include the observational design, which precludes causal inferences; a possible lack of generalizability; and possible bias. However, these limitations are offset by major strengths, including long-term data on alcohol intake, detailed information on confounding variables, a large amount of MRI data, and advanced methods of imaging analysis.

The findings suggest that even moderate alcohol drinking is associated with threefold risk for atrophy in the hippocampus (a key region for memory and spatial navigation) and other adverse brain outcomes. Moderate alcohol drinking has not previously been linked to hippocampal atrophy. Associations of alcohol intake with compromised white-matter integrity in this study also suggest potential threats to cognitive efficiency.

These findings have important potential public health implications, supporting recent UK guidelines recommending lower alcohol intake and raising concerns regarding current limits recommended in US guidelines. The latter suggest that up to 24.5 units/week is safe for men, but this study showed a threefold increase in risk for hippocampal atrophy at only 14-21 units/week. In this study, nearly one half of the men and one quarter of the women were "moderate" drinkers.

This study showed no protective effect of light drinking over abstinence. Earlier reports claiming such a protective effect might have been limited by confounding if light drinking was associated with higher social class or IQ.

As life span increases, preserving quality of life will depend on maintaining cognitive function, which declines with increasing age. Alcohol drinking may be a modifiable risk factor for cognitive impairment, if primary prevention efforts begin in early adulthood or sooner. In light of these findings, justifying drinking habits that are regarded as normal or even as beneficial may no longer be rational.



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.