The long-standing romanticization of moderate alcohol consumption has spent nearly half a century weaving itself into clinical dogma. In 1979, the British medical journal The Lancet published St Leger’s provocative observation linking wine consumption to reduced ischemic heart disease mortality. Over the subsequent four decades, this observation crystallized into the ubiquitous “J-shaped curve,” which posited that while heavy drinking accelerates mortality, modest indulgence acts as a cardioprotective shield. Popularized under the banner of the “French Paradox” in the 1990s, moderate drinking was credited with elevating high-density lipoprotein cholesterol, dampening inflammatory markers like C-reactive protein and interleukin-6, and improving insulin sensitivity.
Yet as analytical epidemiology matured into the genomic era, cracks spread across this architectural facade. Mendelian randomization studies leveraging vast biobanks uncovered that many purported benefits stemmed from methodological artifacts, most notably the “sick quitter” phenomenon, where individuals in declining health abstain from drinking, artificially inflating the baseline risk of non-drinkers. Despite these paradigm shifts, an essential biological question remained unanswered: does cumulative, long-term moderate drinking actually retard or accelerate the subclinical pathological architecture of arterial disease across human midlife? A groundbreaking prospective investigation from Umeå University, published in Frontiers in Cardiovascular Medicine, leveraged two full decades of prospective human tracking to address this fundamental question.
The cohort was derived from the Northern Sweden Västerbotten Intervention Programme and the VIPVIZA trial. Researchers identified 1,014 individuals who completed validated food frequency questionnaires at age 40, age 50, and age 60, followed by bilateral high-resolution 2D B-mode ultrasound examinations of the carotid arteries at age 60. By deliberately excluding consistent lifelong abstainers, individuals meeting criteria for alcohol use disorders (CAGE score of 2 or higher), and heavy drinkers exceeding the risk ceiling of 100 grams per week, the study avoided classical confounding and zeroed in on below-risk exposure.
The primary surrogate endpoints were bilateral carotid atherosclerotic plaque presence and carotid intima-media thickness (CIMT). Over the twenty-year observation window, mean weekly intake rose modestly from midlife to late-midlife, averaging 26.0 grams per week across the entire cohort (37.5 grams in men and 19.2 grams in women). At age 60, subclinical carotid plaques were detected in 49.6 percent of participants (57.7 percent of men and 46.1 percent of women), with a mean cohort CIMT of 0.77 millimeters.
When evaluating cumulative alcohol exposure against vascular morphology, binary logistic regression revealed an adjusted odds ratio of 1.00 (95 percent confidence interval: 0.99 to 1.01) for carotid plaque presence per gram increase of weekly alcohol intake in the total cohort. Sex-stratified analyses yielded identical neutrality: men displayed an odds ratio of 1.00 (95 percent confidence interval: 0.99 to 1.01), while women displayed an odds ratio of 0.99 (95 percent confidence interval: 0.99 to 1.00). Categorical exposure stratification—comparing intake brackets of >25 to 50 grams, >50 to 75 grams, and >75 to 100 grams per week against the baseline category of >0 to 25 grams per week—showed no protective deviation (odds ratios of 0.94, 0.99, and 0.73, respectively, with all confidence intervals encompassing unity).
For continuous arterial wall hypertrophy measured via CIMT, ordinary least squares regression combined with four-knot restricted cubic splines yielded a non-significant association in fully adjusted models (p = 0.48, adjusted R-squared = 0.04). Tests for non-linear inflection confirmed the complete absence of any threshold-dependent or curved dynamic across sexes (p = 0.21 in men; p = 0.22 in women).
The biological and clinical implications of these data fundamentally challenge prevailing dogma. Because subclinical carotid plaque and intima-media thickening represent direct, tangible morphological substrates of vascular aging and future myocardial infarction or ischemic stroke, the complete absence of an inverse relationship disproves the premise of primary anti-atherosclerotic cardioprotection. Modest biochemical shifts—such as modest bumps in circulating HDL-C or transient postprandial vasodilation—fail to translate into structural remodeling or plaque arrest in the arterial lumen over twenty years of exposure. The J-shaped curve does not reflect vascular wall preservation.
Nevertheless, translation of these findings into clinical practice warrants careful contextualization. In Northern Sweden, alcohol consumption exhibits a pattern heavily concentrated during weekend leisure hours rather than distributed daily alongside meals, and less than half of reported intake stems from red wine, unlike Mediterranean cohorts where polyphenol-rich wines predominate. Furthermore, nutritional epidemiology carries inherent vulnerabilities: self-reported intake on food frequency questionnaires underestimates absolute ethanol consumption relative to objective metabolic biomarkers like phosphatidylethanol (PEth), and the small number of subjects consuming between 75 and 100 grams per week limited statistical power at the upper margin of the cohort. Finally, while cumulative midlife consumption below 100 grams per week demonstrated neither benefit nor structural harm at age 60, alcohol remains an established cellular toxin, oxidative stressor, and carcinogen whose biological liabilities accumulate across organ systems.
The myth that moderate alcohol serves as an elixir for the vasculature collapses under longitudinal morphological scrutiny. When tracked across twenty vital years of human midlife, ethanol neither protects the endothelial lining nor prevents the advancement of arterial calcification and plaque. Primary prevention of cardiovascular aging cannot be outsourced to a cocktail glass; it demands the unglamorous, foundational work of metabolic stewardship, physical resilience, and sound vascular mechanics.
- Wood AM, Kaptoge S, Butterworth AS, et al. Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet. 2018;391(10129):1513-1523. https://doi.org/10.1016/S0140-6736(18)30134-X
- Biddinger KJ, Emdin CA, Haas ME, et al. Association of habitual alcohol intake with risk of cardiovascular disease. JAMA Netw Open. 2022;5(3):e223849. https://doi.org/10.1001/jamanetworkopen.2022.3849
- Dahlin Almevall A, Wennberg P, Nyman E, et al. Low to moderate alcohol consumption across two decades and subclinical atherosclerosis at age 60: findings from the Northern Sweden Västerbotten Intervention Programme-visualisation of atherosclerosis (VIPVIZA) study. Front Cardiovasc Med. 2026;12:1710165. https://doi.org/10.3389/fcvm.2025.1710165
- Touboul PJ, Hennerici MG, Meairs S, et al. Mannheim carotid intima-media thickness and plaque consensus (2004-2006-2011). Cerebrovasc Dis. 2012;34(4):290-296. https://doi.org/10.1159/000343145
- Scholten K, Twohig P, Samson K, et al. You can’t handle the truth! Comparing serum phosphatidylethanol to self-reported alcohol intake in chronic liver disease patients. Dig Liver Dis. 2024;56(7):1215-1219. https://doi.org/10.1016/j.dld.2024.01.195