Alcohol consumption is associated with structural changes in various organ systems: A population-based study in UK Biobank

Εvangelos Εvangelou, Hideaki Suzuki, Wenjia Bai, Raha Pazoki, He Gao, Paul M. Matthews, Paul ElliottmedRxiv 2021 · DOI

Abstract

Excessive alcohol consumption is associated with damage to various organs, but its multi-organ effects have not been characterised across the usual range of alcohol drinking in a large general population sample. We assessed global effects of alcohol consumption on quantitative magnetic resonance imaging phenotypic measures of the brain, heart, aorta and liver of UK-Biobank participants who reported drinking alcohol. We found a monotonic association of higher alcohol consumption with lower normalised brain volume across the range of alcohol intakes (–1.7×10⁻³ ± 0.76×10⁻³ per doubling of alcohol consumption, P = 3.0×10⁻¹⁴). Alcohol consumption also was associated directly with measures of left ventricular mass index and left ventricular and atrial volume indices. Liver fat increased by a mean of 0.15% per doubling of alcohol consumption. Our results imply that there is not a “safe threshold” below which there are no toxic effects of alcohol. Current public health guidelines concerning alcohol consumption may need to be revisited.

Introduction

While the deleterious effects of heavy alcohol consumption are well-established, the impact of "moderate" drinking remains a subject of significant scientific and public debate. Previous epidemiological studies have occasionally suggested a "U-shaped" relationship, implying that low levels of alcohol might offer cardiovascular benefits. However, many of these studies were limited by small sample sizes or a narrow focus on single organ systems.

This study leverages the massive UK Biobank dataset to clarify these uncertainties. By examining the structural integrity of the brain, heart, aorta, and liver simultaneously through quantitative Magnetic Resonance Imaging (MRI), researchers sought to determine if a "safe" level of consumption exists or if structural damage begins at even low levels of intake.

Methodology

The researchers utilized quantitative MRI to extract Image Derived Phenotypes (IDPs)—specific, objective measurements of organ structure and function—from thousands of participants.

Statistical Modeling

To capture the dose-response relationship, the team regressed these IDPs against $log_2$-transformed alcohol consumption (grams per day). This approach is particularly effective because a one-unit increase in the predictor represents a doubling of alcohol intake (e.g., moving from 10g/day to 20g/day), allowing for a clear view of the effects across a wide range of drinking habits.

Rigorous Controls

The models were adjusted for a comprehensive set of confounders, including age, sex, ethnicity, BMI, smoking history, education, hypertension, and diabetes. For the brain, the researchers performed voxel-wise analysis to map specific regional sensitivities. For the heart, they employed fully convolutional neural networks (CNNs) to segment cardiac images and derive precise indices for ventricular and atrial volumes, as well as aortic distensibility (a measure of arterial stiffness).

Data Overview

The study analyzed specific subsets of the UK Biobank (participants aged 40–69) where high-quality MRI data was available:

  • Brain: ~10,000 participants.
  • Heart/Aorta: ~12,000 participants.
  • Liver: ~3,600 participants.

Crucially, the analysis excluded non-drinkers and individuals with pre-existing organ diseases to focus specifically on the effects of alcohol within a generally "healthy" drinking population.

Key Insights

The Myth of the "Safe Threshold"

The most significant finding is the monotonic relationship between alcohol intake and organ damage. The data showed no evidence of a "U-shaped" curve or a plateau at low levels of consumption. Instead, as alcohol intake increases, measures of organ health decrease linearly on a log scale. This suggests that even consumption levels currently considered "low-to-moderate" are associated with adverse structural changes.

Brain Atrophy and Accelerated Aging

Higher alcohol consumption was significantly linked to lower normalized brain volume ($P = 3.0 \times 10^{-14}$). The researchers noted specific atrophy in the thalami, insula, and the cingulate and orbital frontal cortices. Quantitatively, doubling one's alcohol intake was associated with a reduction in brain volume equivalent to approximately half a year of natural aging.

Pathological Cardiac Remodeling

In the cardiovascular system, alcohol consumption was associated with increased left ventricular mass and increased atrial volume indices. While aging typically causes the heart to shrink, alcohol appears to drive a form of structural remodeling—characterized by larger, heavier heart chambers—that often serves as a precursor to heart failure. Additionally, increased intake was linked to larger aortic diameters and reduced distensibility, mirroring the effects of vascular aging.

Liver Fat Accumulation

The liver showed a clear dose-response relationship with alcohol, with liver fat increasing by an average of 0.15% for every doubling of alcohol consumption. This highlights the systemic nature of alcohol's impact, even in individuals who do not meet the clinical criteria for alcoholism.

The White Matter Paradox

Interestingly, the study observed increased fractional anisotropy (a measure of white matter integrity) in certain brain tracts. While this might initially seem positive, the authors interpret it as a sign of the loss of "crossing fibers." As secondary nerve fibers degrade, the primary tracts appear more "organized" on an MRI, a phenomenon also documented in the early stages of neurodegenerative diseases like Alzheimer’s.

Limitations

As an observational, cross-sectional study, these findings show strong associations but cannot definitively prove causality. Furthermore, the reliance on self-reported alcohol intake may lead to under-reporting. Finally, as the UK Biobank is predominantly composed of individuals of European ancestry, further research is needed to confirm if these results generalize across all ethnic groups.

Conclusion

The findings suggest that current public health guidelines (such as the UK’s 14 units per week) may still be set too high to prevent subclinical structural damage to the brain and heart. The lack of a clear "safe" threshold indicates that the toxic effects of alcohol on multiple organ systems begin at much lower levels of consumption than previously assumed.

Summary updated

Abstract

Abstract Excessive alcohol consumption is associated with damage to various organs, but its multi-organ effects have not been characterised across the usual range of alcohol drinking in a large general population sample. We assessed global effects of alcohol consumption on quantitative magnetic resonance imaging phenotypic measures of the brain, heart, aorta and liver of UK-Biobank participants who reported drinking alcohol. We found a monotonic association of higher alcohol consumption with lower normalised brain volume across the range of alcohol intakes (–1.7×10 −3 ±0.76×10 −3 per doubling of alcohol consumption, P =3.0×10 −14 ). Alcohol consumption also was associated directly with measures of left ventricular mass index and left ventricular and atrial volume indices. Liver fat increased by a mean of 0.15% per doubling of alcohol consumption. Our results imply that there is not a “safe threshold” below which there are no toxic effects of alcohol. Current public health guidelines concerning alcohol consumption may need to be revisited.
BibTeX
@misc{vangelou2021alcohol,
  author = {Εvangelos Εvangelou and Hideaki Suzuki and Wenjia Bai and Raha Pazoki and He Gao and Paul M. Matthews and Paul Elliott},
  title = {{Alcohol consumption is associated with structural changes in various organ systems: A population-based study in UK Biobank}},
  year = {2021},
  doi = {10.1101/2021.01.20.21249931},
  howpublished = {medRxiv},
}
Figure