Britton A, Hardy R, Kuhf D, Deanfield J, Charakida M, Bell S. Twenty-year trajectories of alcohol consumption during midlife and atherosclerotic thickening in early old age: findings from two British population cohort studies. BMC Medicine 2016l;14:111. DOI 10.1186/s12916-016-0656-9.
Background: Epidemiological evidence indicates a protective effect of light-moderate drinking on cardiovascular disease and an increased risk for heavier drinking. Nevertheless, the effect of alcohol on atherosclerotic changes in vessel walls is disputed. Most previous studies have only looked at the cross-sectional relationship between alcohol and carotid intima media thickness (cIMT) – a surrogate marker of atherosclerosis. Single measurements of alcohol assume that alcohol exposure is stable and ignore the possible cumulative effects of harm, leading to possibly incorrect inferences.
Methods: Data were retrieved from two UK population based cohort studies: the Whitehall II cohort of civil servants and the MRC National Survey of Health and Development (combined sample size of 5403 men and women). Twenty year-drinking trajectories during midlife were linked to measures of cIMT when participants were in early old age, and adjusted for age, sex, socioeconomic position, ethnicity and smoking.
Results: Those who consistently drank heavily had an increased cIMT compared to stable moderate drinkers (pooled difference in cIMT 0.021 mm; 95 % CI 0.002 to 0.039), after adjustment for covariates. This was not detected in cross-sectional analyses. Former drinkers also had an increased cIMT compared to moderate drinkers (pooled difference in cIMT 0.021; 95 % CI 0.005 to 0.037). There were no appreciable differences in cIMT between non-drinkers and consistent moderate drinkers.
Conclusion: The drinking habits among adults during midlife affect the atherosclerotic process and sustained heavy drinking is associated with an increased cIMT compared to stable moderate drinkers. This finding was not seen when only using cross-sectional analyses, thus highlighting the importance of taking a life course approach. There was no evidence of a favourable atherosclerotic profile from stable moderate drinking compared to stable non-drinking.
Carotid artery disease can be estimated from the thickness of the wall of the arteries measured with ultrasound (recorded as carotid artery intima/medial thickness, cIMT), and by evidence of atherosclerotic plaques within the carotid arteries. The association between alcohol intake and “carotid lesions” is unclear, as some studies show a positive association with cIMT and/or plaques while others show no association. Given that carotid disease relates to the subsequent risk of coronary artery disease, there is increasing use of carotid ultrasound measurements to help determine long-term risk of cardiovascular disease, especially among subjects with intermediate levels of risk as determined by usual risk factor assessment.
The present study provides valuable information by reporting the cross-sectional relation between alcohol consumption and cIMT as well as how the drinking pattern over 20 years in middle age may relate to cIMT later in life. Specifically, it combines data from two cohort studies (the Whitehall II cohort of civil servants and the MRC National Survey of Health and Development), each of which had repeated assessments of alcohol intake during middle age. The investigators were able to evaluate the effect on cIMT at the time of alcohol assessment (at age 50-74 in the Whitehall study and age 60-64 in the MRC study) as well as the effect on cIMT of the pattern of drinking over earlier periods in life. Thus, the investigators were able to report both cross-sectional associations between alcohol and cIMT and a trajectory of the pattern (always none, always moderate, always heavy, etc.) for estimating the effect of alcohol intake on cIMT. The analyses are well done and appropriate.
Specific comments on the study by Forum members: Reviewer Puddey carried out an extensive review of the methodology of this study: “This paper utilises a meta-analytic approach to 2 prospective British studies to link 20 year drinking pattern trajectories, characterised as either ‘stable heavy’ drinkers or ‘former drinkers’, to a finding of increased carotid artery IMT. In addition, a ‘mostly moderate’ 20 yr drinking trajectory was not linked to either anti-atherogenic or pro-atherogenic effects. These findings are of interest given the ongoing controversy with respect to potential anti-atherogenic and pro-atherogenic effects of alcohol consumption and the relative paucity of literature that has taken long term patterns of consumption into account.
“The meta-analysis utilises the difference from median IMT for each of the 2 cohorts as the dependent variable. Were median IMT and IMT distributions similar in the 2 cohorts? The reported values look significantly lower and with much wider confidence intervals in the NHSD cohort compared to the Whitehall II cohort. Should IMT have been standardised before meta-analysis? In the meta-analysis ‘stable moderate’ drinkers were used as the reference category. Most previous cross-sectional and longitudinal studies have used non-drinkers as the reference group. What is the result if the ‘stable non-drinkers’ are used as the reference category? Given the wide 95% CI for IMT in ‘stable non-drinkers’, it is probable that no significant differences would have been seen.
“The multivariate model has not included a measure of adiposity, and BMI is not listed in the table of characteristics of participants. BMI has been a predictor of carotid IMT in previous studies and could be a confounder of the reported 20 yr alcohol drinking pattern trajectory association with IMT. Further, increased BMI is associated with impaired glucose tolerance and it has been previously reported that there is an interaction between alcohol intake and glucose tolerance, with alcohol predictive of reduced carotid IMT in those with normal glucose tolerance but increased carotid IMT in those with impaired glucose tolerance (Cooper et al). A long term pattern of heavy drinking has been linked to not just increased prevalence of smoking and ex-smoking but also to poorer dietary choices and less physical activity. It has also been linked to type of alcoholic beverage (spirits vs beer or wine) which hasn’t been measured in the current study. The study has therefore not convincingly demonstrated that unmeasured confounding could have influenced the final result.”
Puddey concluded: “Several previous studies have indicated a potentially favourable association of increasing alcohol intake with increased carotid artery lumen diameter, despite simultaneously demonstrating a link to increased IMT. We do not know if carotid artery lumen diameter was measured in the current study and we are also not told whether carotid IMT was related to incidence of cerebrovascular disease events, an outcome which has been previously measured in the Whitehall study.”
Reviewer Stockley pointed out that the pattern of drinking could not be assessed in the present analyses. It is clear that subjects who binge drink do not get the protection against cardiovascular disease apparent among regular moderate drinkers. Further, consumption with meals provides additional protection, as shown in a clinical trial by Hendriks et al. That trial revealed that moderate alcohol consumption with dinner affects plasminogen activator inhibitor activity, plasminogen activator antigen level, and tissue type plasminogen activator activity temporarily, all of which lower cardiovascular risk through favorable effects on fibrinolysis.”
Forum member Finkel noted: “I am never sanguine about studies that depend on avatars; it would be more helpful to determine the frequency of cardiovascular events over time. There are many potential confounding variables (obesity, diet, physical activity, etc.) not taken into consideration in these analyses. How do we know that the group with thick carotids hadn’t spent most of their time away from the office at their bowling alley drinking ale and eating crisps?”
Does cIMT predict future cardiovascular disease? In 1997, Bots et al reported from the Rotterdam Study that “an increased common carotid IMT is associated with future cerebrovascular and cardiovascular events.” In 1999, O’Leary et al reported that “Increases in the thickness of the intima and media of the carotid artery, as measured noninvasively by ultrasonography, are directly associated with an increased risk of myocardial infarction and stroke in older adults without a history of cardiovascular disease.” Numerous studies since these have generally shown that cIMT is a risk factor for future disease. But there are still divided opinions as to whether carotid artery ultrasound studies should be offered to all patients in a clinical practice.
Measuring both cIMT and carotid atherosclerotic plaques: Previous studies have evaluated not only the carotid wall thickness (cIMT) but identified atherosclerotic plaques within the carotid arteries. The present study focuses only on the cIMT, and does not comment on the presence or absence of atherosclerotic plaques. Forum reviewer Ellison states: “Previous studies from our group have found no relation between alcohol intake and IMT or the presence or absence of atherosclerotic lesions [Djousse et al (a)]; other such studies have also shown a lack of effect on carotid atherosclerosis from albumin levels [Djousse et al (b)] or from measures of hostility or social support (Knox et al). However, both carotid atherosclerotic plaques and cIMT levels have been found to relate inversely to the intake of linolenic acid [Djousse et al (c)] and positively (at least in men) to serum uric acid levels (Neogi et al). In any case, it would seem prudent that both cIMT and the presence or absence of atherosclerotic plaques within the carotid arteries be evaluated in estimating later cardiovascular risk.”
Alcohol intake, atherosclerosis and coagulopathy as risk factors for myocardial infarction (MI): Acute MI is known to relate especially to the rupture of an atherosclerotic plaque and subsequent clot formation. The degree of atherosclerosis of a coronary vessel is not a particular good predictor of whether or not a MI will occur from obstruction of that artery. Data indicate that the effects of wine and other alcoholic beverages on coagulation may be more relevant for the development of MI than from its effects on atherosclerosis. For example, Renaud proposed the hypothesis of a hemostatic mechanism, rather than an interaction with the atherosclerosis process, as a key factor for this paradox (Renaud & de Lorgeril).
Wine and alcohol intake have repeatedly been shown to relate to improvements in blood clotting, fibrinolysis, and other aspects of coagulation. Thus, any study evaluating alcohol intake for predicting the risk of MI must evaluate not only the development of atherosclerosis but also the effects on coagulopathy. While this study did not show that moderate drinking results in less thickness of the carotid arteries later in life, most studies show that such drinking is associated with a lower risk of MI, stroke, and total mortality. If indeed cIMT reflects coronary disease, the results of this study suggest that the protection against MI generally seen among moderate drinkers may be due more to improvements in coagulation than it is to the degree of atherosclerosis.
Forum member De Gaetano commented: “This is an interesting paper indeed. The apparent discrepancy between their results (lack of effect on cIMT later in life) and the well-known preventive effect of moderate alcohol consumption on fatal and non-fatal cardiovascular clinical events suggests that atherosclerosis is a benign disease unless thrombosis (activation of platelets and coagulation) occurs on an atherosclerotic vessel wall. Together with Ramon Estruch we showed, years ago, that wine polyphenols did not modify the lipid pattern of healthy volunteers, but inhibited several plasma and cell functions related to inflammation and ischemic disease (Badia et al, Estruch et al). The data on heavy or moderate drinkers on cIMT might suggest that both abstainers and moderate drinkers have a very slow progression of atherosclerosis while heavy drinking accelerates it. If moderate drinking is not preventive, heavy drinking is promoting the atherosclerotic process.”
Reviewer Waterhouse noted: “This study seems to confirm that the mechanism for the protective effect of moderate alcohol consumption is not so much related to preventing atherogenesis, but due primarily to its effects on coagulopathy. This is a very useful distinction for future investigations. If this is the case, it also suggests another line of investigation, and that is, how quickly is one’s risk of cardiovascular disease reduced after starting moderate alcohol consumption?”
Reviewer Skovenborg agreed with other reviewers that carotid artery wall thickness is a proxy variable with dubious value for the future risk of stroke. He adds: “Even so, recent data indicate that light to moderate alcohol consumption is associated with lower atherosclerotic burden in the proximal aortic arch (Kohsaka et al). Further, regarding the proposition that the largest effect of alcohol on cardiovascular disease is from its effects on coagulopathy, two conditions are of interest as they are associated with atherosclerosis and not with coagulopathy: stable angina pectoris and peripheral arterial disease (PAD). The Physicians’ Health Study demonstrated that moderate drinking decreases the risk of angina pectoris (RR 0.69, 95% CI 0.59-0.81) and myocardial infarction (RR 0.65, 95% CI 0.52-0.81) and also that, after control for confounding by smoking, moderate alcohol consumption decreases the risk of PAD (RR 0.74, 95% CI 0.57-0.97) [Camargo CA et al (a); Camargo et al (b)].”
Forum member Lanzmann-Petithory commented: “I agree that the role of coagulation rather than atherosclerosis is most important in the protection against cardiovascular disease for wine drinkers. The ‘heavy drinkers’ in this study, including binge drinkers, are probably not wine drinkers and are likely to have higher blood pressure. And blood pressure is directly related to cIMT, as shown by numerous studies (e.g., Ferreira et al). The positive relation of heavy drinking with cIMT may be reflecting partially the effect of higher blood pressure in binge drinkers, especially of beer and spirits.”
Overall, current scientific data provide strong evidence that moderate alcohol consumption lowers the risk of essentially all manifestations of cardiovascular disease; it appears to work through a combination of effects on lipids, inflammation, coagulation, fibrinolysis, glucose metabolism, and other paths. In most studies it decreases the risk of the development and progression of atherosclerosis, but also helps prevent clot formation within arteries, often the precipitating event for an acute cardiovascular event.
Is carotid artery screening appropriate for most patients in clinical practice? Robertson et al stated that “It is increasingly recognized that in terms of assessing disease burden and risk of subsequent vascular events, both IMT and plaque should be considered separately. Carotid plaque itself has been shown to be associated with concomitant CVD, and indeed is considered to be an indicator of high vascular risk (as shown by Ebrahim et al).” The value of adding cIMT to already available conventional risk factors in predicting future disease remains unclear. As Robertson et al concluded: “Many large epidemiological studies have shown a strong relationship between IMT and incident CVD, but the evidence for the use of cIMT in clinical practice is incomplete. Current evidence suggests that for people at intermediate risk according to Framingham risk scoring, IMT may add useful information on vascular risk, but findings are inconsistent between studies and, in some cases, the improvement in classification is modest.”
References from Forum critique
Badía E, Sacanella E, Fernández-Solá J, Nicolás JM, Antúnez E, Rotilio D, de Gaetano G, Urbano-Márquez A, Estruch R. Decreased tumor necrosis factor-induced adhesion of human monocytes to endothelial cells after moderate alcohol consumption. Am J Clin Nutr 2004;80:225-230.
Bots ML, Hoes AW, Koudstaal PJ, Hofman A, Grobbee DE. Common Carotid Intima-Media Thickness and Risk of Stroke and Myocardial Infarction: The Rotterdam Study. Circulation 1997;96:1432-1437.
Camargo CA (a), Stampfer MJ, Glynn RJ, Grodstein F, Gaziano JM, Manson JE, Buring JE, Hennekens CH. Moderate Alcohol Consumption and Risk for Angina Pectoris or Myocardial Infarction in U.S. Male Physicians. Ann Intern Med 1997;126:372-375.
Camargo CA (b), Stampfer MJ, Glynn RJ, Gaziano JM, Manson JE, Goldhaber SZ, Hennekens CH. Prospective Study of Moderate Alcohol Consumption and Risk of Peripheral Arterial Disease in US Male Physicians. Circulation 1997;95:577-580.
Cooper DE, Goff DC, Jr., Bell RA, Zaccaro D, Mayer-Davis EJ, Karter AJ. Is insulin sensitivity a causal intermediate in the relationship between alcohol consumption and carotid atherosclerosis?: the insulin resistance and atherosclerosis study. Diabetes Care 2002; 25:1425-1431.
Djoussé L (a), Myers RH, Province MA, Hunt SC, Eckfeldt JH, Evans G, Peacock JM, Ellison RC. Influence of apolipoprotein E, smoking, and alcohol intake on carotid atherosclerosis: National Heart, Lung, and Blood Institute Family Heart Study. Stroke 2002;33:1357-1361.
Djoussé L (b), Rothman KJ, Cupples LA, Arnett DK, Ellison RC. Relation between serum albumin and carotid atherosclerosis: the NHLBI Family Heart Study. Stroke 2003;34:53-57.
Djoussé L (c), Folsom AR, Province MA, Hunt SC, Ellison RC. Dietary linolenic acid and carotid atherosclerosis: the National Heart, Lung, and Blood Institute Family Heart Study. Am J Clin Nutr 2003;77:819-825.
Ebrahim S, Papacosta O, Whincup P, et al. Carotid plaque, intima media thickness, cardiovascular risk factors, and prevalent cardiovascular disease in men and women: the British Regional Heart Study. Stroke 1999;30:841–850.
Estruch R, Sacanella E, Badia E, Antúnez E, Nicolás JM, Fernández-Solá J, Rotilio D, de Gaetano G, Rubin E, Urbano-Márquez A. Different effects of red wine and gin consumption on inflammatory biomarkers of atherosclerosis: a prospective randomized crossover trial. Effects of wine on inflammatory markers. Atherosclerosis 2004;175:117-123.
Ferreira JP, Girerd N, Bozec E, Machu JL, Boivin JM, London GM, Zannad F, Rossignol P. Intima-Media Thickness Is Linearly and Continuously Associated With Systolic Blood Pressure in a Population-Based Cohort (STANISLAS Cohort Study). J Am Heart Assoc 2016;16;5. pii: e003529. doi: 10.1161/JAHA.116.003529.
Hendriks HF, Veenstra J, Velthuis-te Wierik EJ, Schaafsma G, Kluft C. Effect of moderate dose of alcohol with evening meal on fibrinolytic factors. BMJ 1994;308:1003-1006.
Knox SS, Adelman A, Ellison RC, Arnett DK, Siegmund K, Weidner G, Province MA. Hostility, social support, and carotid artery atherosclerosis in the National Heart, Lung, and Blood Institute Family Heart Study. Am J Card 2000;86:1086-1089.
Kohsaka S et al. Alcohol consumption and atherosclerotic burden in the proximal thoracic aorta. Atherosclerosis 2011;19:794–798.
Neogi T, Ellison RC, Hunt S, Terkeltaub R, Felson DT, Zhang Y. Serum uric acid is associated with carotid plaques: The National Heart, Lung, and Blood Institute Family Heart Study. J Rheumatol 2009;36:378-384.
O’Leary DH, Polak JF, Kronmal RA, Manolio TA, Burke GL, Wolfson SK Jr. Carotid-artery intima and media thickness as a risk factor for myocardial infarction and stroke in older adults. Cardiovascular Health Study. N Engl J Med 1999;340:14–22.
Renaud S, de Lorgeril M. Wine, alcohol, platelets, and the French Paradox for coronary heart disease. Lancet 1992;339:1523–1526.
Robertson CM, Fowkes FGR, Price JF. Carotid intima–media thickness and the prediction of vascular events. Vascular Medicine 2012;17:239-248. DOI: 10.1177/1358863X12445103.
Carotid artery disease can be estimated by ultrasound from the thickness of the wall of the arteries (recorded as carotid artery intima/medial thickness, cIMT) and by evidence of atherosclerotic plaques within the carotid arteries. The association between alcohol intake and such lesions is unclear, as some studies show a positive association with cIMT and/or plaques while others show no association. Given that carotid disease relates to the subsequent risk of coronary artery disease, there is increasing use of carotid ultrasound measurements to help determine long-term risk.
The present large study from the UK provides valuable information by reporting the cross-sectional relation between alcohol consumption and cIMT as well as how the drinking pattern over 20 years in middle age may relate to cIMT later in life. While heavy alcohol intake was found to increase later cIMT measures, no clear differences were noted between subjects reporting abstinence and those reporting moderate drinking in middle age.
Forum reviewers considered this to be a well-done study. There were some concerns that only total average alcohol intake was considered, as there were no data on the type of beverage consumed, the pattern of drinking (binge versus regular moderate), whether the alcohol was consumed with food, etc., and there was an absence of data on diet, physical activity, adiposity, and other factors related to atherosclerosis. Further, only the thickness of the carotid artery was evaluated, and not the presence or absence of atherosclerotic plaques on the ultrasound readings.
Overall, current scientific data provide strong evidence that moderate alcohol consumption lowers the risk of most manifestations of cardiovascular disease; it appears to work through a combination of effects on lipids, inflammation, coagulation, fibrinolysis, glucose metabolism, and other paths. Further, heavy drinking is known to increase blood pressure, and is an important factor for developing hypertension (and a strong determinant of cIMT). The failure of the present study to find a significant association between moderate alcohol intake and later carotid thickness supports what has been shown in some, but not all, previous studies.
If indeed chronic alcohol use has little effect on atherosclerosis (using images from the carotids as an index of atherosclerosis in the coronary arteries and elsewhere in the body), it may indicate that the mechanisms for the protective effect of moderate alcohol consumption on cardiovascular disease are not so much related to preventing atherogenesis, but due primarily to the effects on coagulopathy. And it is known that clot formation within the arterial wall is often the precipitating event for an acute myocardial infarction or other cardiovascular event.
* * * * * *
Comments for this critique by the International Scientific Forum on Alcohol Research were provided by the following members:
Giovanni de Gaetano, MD, PhD, Department of Epidemiology and Prevention, IRCCS Istituto Neurologico Mediterraneo NEUROMED, Pozzilli, Italy
R. Curtis Ellison, MD, Professor of Medicine & Public Health, Boston University School of Medicine, Boston, MA, USA
Ramon Estruch, MD, PhD, Hospital Clinic, IDIBAPS, Associate Professor of Medicine, University of Barcelona, Spain
Creina Stockley, PhD, MSc Clinical Pharmacology, MBA; Health and Regulatory Information Manager, Australian Wine Research Institute, Glen Osmond, South Australia, Australia
Arne Svilaas, MD, PhD, general practice and lipidology, Oslo University Hospital, Oslo, Norway
Pierre-Louis Teissedre, PhD, Faculty of Oenology–ISVV, University Victor Segalen Bordeaux 2, Bordeaux, France
David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South Africa
Andrew L. Waterhouse, PhD, Department of Viticulture and Enology, University of California, Davis, USA
Ian Puddey, MD, Dean, Faculty of Medicine, Dentistry & Health Sciences, University of Western Australia, Nedlands, Australia
Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark
Harvey Finkel, MD, Hematology/Oncology, Boston University Medical Center, Boston, MA, USA
Dominique Lanzmann-Petithory, MD, PhD, Nutrition Geriatrics, Hôpital Emile Roux, APHP Paris, Limeil-Brévannes, France