t ain’t so!)

The Study: No safe level of alcohol consumption for brain health: observational cohort study of 25,378 UK Biobank participants

Editor’s Note: This study is not yet out of peer review but was published in the “pre print” publication MedRxiv. We do not normally comment on non peer reviewed papers but I thought this one was interesting and relevant to a large percentage of the patients we see in our offices.

Overview: While it is well established that heavy alcohol consumption is damaging to the brain, the effects of moderate levels of consumption are less clear in the literature.  Moderate levels of alcohol consumption are very common in modern society and may possibly represent a modifiable risk factor for neurodegenerative disease. This paper is an observational cohort study designed to “estimate the relationship between moderate alcohol consumption and brain health”. 

Key Points: 

• The authors began the study with four hypotheses: 1) the threshold intake for “safe’ levels of alcohol intake were lower than those currently defined as “low risk” drinking,  2) certain comorbidities would likely increase risk, 3) binge drinking would lead to worse outcomes, and 4) type of beverage consumed was unlikely to matter much in terms of outcomes.

•  This is a large study which looked 25,378 participants with a mean age of 54.9 years. 

• Brain health was measured by structural and functional parameters using multimodal MRI scans from the UK Biobank database, the largest neuro imaging resource in the world.   

• The authors cite the advantage of using MRI to determine preservation of optimal brain integrity because of the high sensitivity to measurement of changes, often years before clinical symptoms present themselves. 

• Additionally the authors pulled other clinical data including prior diagnoses, brain imaging, cognitive testing and genotyping from linked Hospital Episode Statistics. 

• Cognitive function was assessed at the time of imaging using a number of standardized tests.  

• Other variables taken into account at baseline included smoking status, educational qualifications, blood pressure, body mass index, cholesterol, body mass index (BMI), Townsend Deprivation Index ( a measure of material deprivation ) and weekly exercise habits. 

• Average weekly quantity and type of alcohol consumed was determined at baseline assessment visit using touch screen questionnaire and subjects were divided into quintiles for analysis. 

• Non drinkers were further divided into “former and never” drinkers.  

• Median alcohol intake was 13.5 units weekly (just barely below the UK current “low risk” guidelines) but few subjects drank very heavily. 

Author’s Conclusions: 

• Alcohol consumption was linearly associated with lower indices of brain health, including grey matter density.  

• Individuals in the highest quintiles of consumption, particularly those who were binge drinking, had significantly lower grey matter density, however no significant differences were observed whether the weekly units consumed were wine, beer, or distilled spirits. 

• The significance of the study is that “alcohol made a larger contribution than any other modifiable risk factor tested, including smoking” however it needs to also be emphasized that the effect was very small (0.8%) compared to the negative effects of Age which accounted for 54% of observed grey matter loss.

• Both high blood pressure and high Body Mass Index were found to steepen the negative impact of alcohol consumption. In contrast, no significant interactions were found between alcohol and age, sex, GGT, heart disease or ApoE4 genotype.

 Reviewer's Comments:  

This is a large study but possibly problematic because it relies on self reported levels of alcohol consumption.  It is interesting in that it leads to the conclusion that there may be no truly safe level of alcohol consumption.  Because alcohol consumption is certainly a modifiable activity, this may be cause for consideration when managing cases with existing neurodegenerative disease and/or additive factors such as obesity and high blood pressure. However, the practical significance for most patients is probably muted by the fact that the effect is so small (0.8%) compared with the overwhelming effects of age (54%).

 Reviewer:  Mark R. Payne DC 

Link to Full Text: https://www.medrxiv.org/content/10.1101/2021.05.10.21256931v1.full-text