International Centre for Health and Society, Department of Epidemiology and Public Health, University College London, 1-19 Torrington Place, London WCIE 6BT UK, 1 Institute of Internal Medicine, Novosibirsk, Russia, 2 Jagiellonian University, Krakow, Poland, 3 National Institute of Public Health, Prague, Czech Republic and 4 Institute of Clinical and Experimental Medicine, Prague, Czech Republic
* Author to whom correspondence should be addressed at: International Centre for Health and Society, Department of Epidemiology and Public Health, University College London, 1-19 Torrington Place, London WCIE 6BT UK. Tel.: +44 207 679 1706; Fax: +44 207 813 0280; E-mail: a.peasey{at}ucl.ac.uk
(Received 11 November 2004; first review notified 16 February 2005; in revised form and accepted 16 March 2005)
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
However, in addition to average volume of alcohol intake, there is a growing interest in the effects of pattern of drinking on coronary heart disease (Rehm et al., 1996). Several studies indicated that binge drinking is not associated with a reduced risk of cardiovascular disease and that it may, in fact, increase the risk of cardiac death or stroke (Kauhanen et al., 1997
; Mazzaglia et al., 2001
; Murray et al., 2002
; Rehm et al., 2003a
). These studies raise questions about the effect of binge drinking on lipids; it has been suggested that, contrary to regular moderate drinking, binge drinking is associated with an unfavourable lipid profile (McKee and Britton, 1998
). So far, however, the evidence on binge drinking and blood lipid concentrations is sparse and inconsistent (Puddey et al., 1999
).
The aim of this paper is to contribute to the understanding of the biological effects of binge drinking. To do so, we examined the association between average drinking volume, binge drinking behaviour and lipid profile in population samples from the Czech Republic, Poland and the Russian Federation.
![]() |
SUBJECTS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Measurements
Participants completed a questionnaire and attended a medical examination. Annual alcohol consumption and drinking pattern were estimated from a graduated frequency questionnaire, which assessed the frequency of consuming different amounts of alcohol per day in the past year (Rehm, 1998). For the present analyses, annual alcohol intake was classified into five categories; the following cut-off points were selected to make sure that sufficient numbers of men were in each group: non-drinkers; 1200; 2012000; 20018000 and >8000 g of ethanol per year. We used two measures of binge drinking. First, (any) binge drinking was defined as consumption of
100 g ethanol per occasion; second, heavy binge drinking was defined as consuming
140 g ethanol per occasion. The frequency of any or heavy binge drinking in the past year was defined as never, <1 per month, 13 per month and
1 per week.
Serum samples were analysed centrally in one batch in the WHO Regional Lipid Reference Centre, Institute of Clinical and Experimental Medicine, Prague. Lipid concentrations in serum were measured on a Roche COBAS MIRA auto-analyser, using a conventional enzymatic method with reagents from Boehringer Mannheim Diagnostics and Hoffman-La Roche. LDL cholesterol (LDL-C) was calculated by the Friedewald formula (Friedewald et al., 1972).
Statistical analysis
To investigate the association between lipid concentrations and alcohol drinking patterns, we estimated the differences in the mean concentrations of total, HDL-C and LDL-C and triglycerides (TG), and in the total/HDL-C and LDL-C/HDL-C ratios, between categories of annual alcohol intake binge drinking, using linear regression and controlling for age, country, body mass index (BMI) and smoking status. We initially conducted country-specific analyses; since the associations between blood lipids and alcohol variables were similar between countries, data from all three countries were pooled and were adjusted for country in the analyses. There were some differences in the concentrations of serum lipids between countries; levels of HDL-C were somewhat higher (reflected in a slightly lower total/HDL-C ratio) and concentrations of TG were considerably lower in Russia than in the other two countries. To avoid the problem of different absolute concentrations, subsequent results are reported as differences in concentrations and ratios between categories, rather than adjusted means and ratios, and all analyses are adjusted for country.
In the next step, we first examined the differences in serum lipids by categories of annual intake. Second, we investigated the differences in serum lipids by frequency of drinking (the pattern of results mirrored those for annual intake). Third, we analysed differences in serum lipids by the frequency of any and heavy binge drinking; we have additionally controlled for annual alcohol intake in these analyses. Finally, we analysed the differences in serum lipids between categories of annual intake after stratifying for heavy binge drinking behaviour. Likelihood ratio test P-values for linear trend were presented. All statistical analyses were performed using Stata 6.0.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Several limitations of this study should be considered. First, the cross-sectional design complicates the assessment of causality. It is likely that the abstainer group contains some former heavy drinkers; if these drinkers had different levels of blood lipids than other abstainers, the difference between drinkers and non-drinkers would be inaccurate. We do have some data on former drinkers in the Russian sample, and we found no significant differences between current non-drinkers and non-drinkers who used to drink 6 years before. However, the sample size was small (57 persons only), and the possibility of reverse causation cannot be entirely excluded.
Second, the size of the study was modest and the numbers of men in the high alcohol intake categories were relatively small. The statistical power of the study was therefore limited and some of the confidence limits around the point estimates are wide, although associations were found.
Third, the inaccuracy with which individuals report drinking could result in an under- or over-estimate of drinking patterns. Graduated frequency measures seem to underestimate alcohol consumption to a lesser extent than the average quantity and frequency measures commonly used in epidemiology (Rehm and Gmel, 2003). Self-reported drinking correlated well with the serum GGT concentrations in a subset of the Russian subjects (Malyutina et al., 2002
) and with self-reported problem drinking (Bobak et al., 2004
). Repeated interviews with a subsample added further support to the reliability of the alcohol measurement. Nevertheless, some misclassification of alcohol intake certainly occurred which, if random, would tend to underestimate the association between drinking and lipids.
Finally, there were differences between the three countries in drinking patterns, serum TG and BMI. The differences between populations, both in the high frequency of binge drinking and in the low levels of triglycerides in Russia, are consistent with other reports (Malyutina et al., 2002; Averina et al., 2003
; Bobak et al., 2004
). Laboratory bias is unlikely, as all lipid concentrations were determined in a central laboratory using a single methodology. The associations between alcohol and lipids were similar between countries and were not affected by the absolute levels of lipid concentrations, drinking rates or BMI. Inclusion of waist-to-hip ratio, a measure of central adiposity, did not influence the association between alcohol and lipids. This strongly suggests that the pattern of results is not specific or due to one particular country.
The association between overall alcohol intake and the lipid profile has been investigated in many studies (for reviews see, for example: McKee and Britton, 1998; Puddey et al., 1999
; Rehm et al., 2003b
). The main finding of these studies is that a moderate intake of alcohol is associated with a beneficial effect on the lipid profile, in particular with elevated concentrations of HDL-C (Rimm et al., 1999
). Some investigators found a reduction in LDL-C (Nakanishi et al., 2001
) while others found an increase in TG levels (Chrysohoou et al., 2003
) in moderate drinkers. Our results are consistent with the literature with respect to HDL-C but we did not observe any reduction in LDL-C among non-binge drinkers.
Only recently have drinking patterns been included in the studies of alcohol intake and lipid concentrations (Puddey et al., 1999; Rehm et al., 2003a
). So far, the results are inconclusive. Some of the early studies on binge drinking found that binge drinkers had lower HDL-C and higher LDL-C than non-bingers (Gruchow et al., 1982
). A study on squirrel monkeys revealed no change in HDL-C and an increase in LDL-C with increased alcohol intake (Hojnacki et al., 1991
) but interpolation to humans is not straightforward. An intervention study by Rakic et al. (1998)
found an increase in HDL-C and a reduction in LDL-C in binge drinkers and an increase in HDL-C in moderate drinkers. The increase in HDL in both groups of drinkers in our results are consistent with the intervention study; however, we also observed that drinking alcohol was related to an increase in LDL-C in heavy binge drinkers.
It is not clear at what level of binge drinking changes in blood lipids occur. The usual definition of binge drinking is 80 g of alcohol in one session but this is an arbitrary cut-off point. It ignores the frequency of bingeing. We found no changes in the lipid profile among drinkers who consumed
100 g of ethanol per session at least once per month. In contrast, drinkers consuming
140 g of ethanol per session at least once per month had elevated concentrations of HDL and LDL. While elevation of HDL among drinkers is thought to be a consequence of increased hepatic production and secretion of apolipoproteins and lipoproteins (Rimm et al., 1999
; Sierksma et al., 2004
), the increase in LDL cholesterol in heavy drinking could also be confounded by dietary fat intake (Rumpler et al., 1999
; Richter et al., 2004
). We assessed this possibility, using data from a food frequency questionnaire, but fat intake was similar in non-binge and binge drinkers. This suggests that confounding by dietary fat was unlikely.
A possible confounder is whether drinking occurred with a meal, since drinking with meals has been associated with a reduction in cardiovascular risk compared with those drinking without food (Trevisan et al., 2004). Unfortunately, in this study we did not assess the relation between dietary and drinking patterns.
The interpretation of our results in terms of cardiovascular risk is not unequivocal. On the one hand, the finding of elevated LDL-C in heavy binge drinkers may help explain reports of increased cardiovascular mortality in heavy or binge drinkers (Kauhanen et al., 1997; Mazzaglia et al., 2001
; Murray et al., 2002
; Rehm et al., 2003a
). In this study, the group of heavy binge drinkers with high annual intake was relatively small, accounting for
7% of the men. This appears consistent with a cohort study in Russia in which cardiovascular mortality in frequent heavy drinkers was about double of that in moderate drinkers but this group was also relatively small,
5% of the cohort (Malyutina et al., 2002
). On the other hand, the ratios of total and LDL to HDL-C were lower in heavy drinkers. Although the Apo B/Apo A-I ratio may be a more sensitive measure of future cardiac risk than LDL/HDL-C ratio (Yusuf et al., 2004
), the total and LDL to HDL-C ratios are well established predictors of cardiovascular risk (Kinosian et al., 1994
; Meisinger et al., 2005
). On the basis of our data on total, HDL and LDL-C, the results do not indicate a more atherogenic lipid profile in heavy binge drinkers.
In summary, our findings suggest that both HDL and LDL concentrations are affected by heavy binge drinking but the role of lipids in potentially mediating the relation between binge drinking and cardiovascular disease remains to be clarified.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bobak, M., Room, R., Pikhart, H. et al. (2004) Contribution of drinking patterns to differences in rates of alcohol related problems between three urban populations. Journal of Epidemiology and Community Health 58, 238242.
Chrysohoou, C., Panagiotakos, D. B., Pitsavos, C. et al. (2003) Effects of chronic alcohol consumption on lipid levels, inflammatory and haemostatic factors in the general population: the ATTICA Study. European Journal of Cardiovascular Prevention and Rehabilitation 10, 355361.
Corrao, G., Rubbiati, L., Bagnardi, V. et al. (2000) Alcohol and coronary heart disease: a meta-analysis. Addiction 95, 15051523.[CrossRef][ISI][Medline]
Friedewald, W. T., Levy, R. I. and Fredrickson, D. S. (1972) Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clinical Chemistry 18, 499502.
Gruchow, H. W., Hoffmann, R. G., Anderson, A. J. et al. (1982) Effects of drinking patterns on the relationship between alcohol and coronary occlusion. Atherosclerosis 43, 393404.[ISI][Medline]
Hojnacki, J. L., Deschenes, R. N., Cluette-Brown, J. E. et al. (1991) Effect of drinking pattern on plasma lipoproteins and body weight. Atherosclerosis 88, 4959.[CrossRef][ISI][Medline]
Kauhanen, J., Kaplan, G. A., Goldberg, D. E. et al. (1997) Beer binging and mortality: results from the Kuopio ischaemic heart disease risk factor study, a prospective population based study. British Medical Journal 315, 846851.
Kinosian, B., Glick, H. and Garland, G. (1994) Cholesterol and coronary heart disease: predicting risks by levels and ratios. Annals of Internal Medicine 121, 641647.
Malyutina, S., Bobak, M., Kurilovitch, S. et al. (2002) Relation between heavy and binge drinking and all-cause and cardiovascular mortality in Novosibirsk, Russia: a prospective cohort study. Lancet 360, 14481454.[CrossRef][ISI][Medline]
Mazzaglia, G., Britton, A. R., Altmann, D. R. et al. (2001) Exploring the relationship between alcohol consumption and non-fatal or fatal stroke: a systematic review. Addiction 96, 17431756.[CrossRef][ISI][Medline]
Meisinger, C., Loewel, H., Mraz, W. et al. (2005) Prognostic value of apolipoprotein B and A-I in the prediction of myocardial infarction in middle-aged men and women: results from the MONICA/KORA Augsburg cohort study. European Heart Journal 26, 271278.
McKee, M. and Britton, A. (1998) The positive relationship between alcohol and heart disease in eastern Europe: potential physiological mechanisms. Journal of the Royal Society of Medicine 91, 402407.[Abstract]
Murray, R. P., Connett, J. E., Tyas, S. L. et al. (2002) Alcohol volume, drinking pattern, and cardiovascular disease morbidity and mortality: is there a U-shaped function? American Journal of Epidemiology 155, 242248.
Nakanishi, N., Yoshida, H., Nakamura, K. et al. (2001) Influence of alcohol intake on risk for increased low-density lipoprotein cholesterol in middle-aged Japanese men. Alcoholism: Clinical and Experimental Research 25, 10461050.[CrossRef][ISI][Medline]
Puddey, I. B., Rakic, V., Dimmitt, S. B. et al. (1999) Influence of pattern of drinking on cardiovascular disease and cardiovascular risk factorsa review. Addiction 94, 649663.[CrossRef][ISI][Medline]
Rakic, V., Puddey, I. B., Dimmitt, S. B. et al. (1998) A controlled trial of the effects of pattern of alcohol intake on serum lipid levels in regular drinkers. Atherosclerosis 137, 243252.[CrossRef][ISI][Medline]
Rehm, J. (1998) Measuring quantity, frequency, and volume of drinking. Alcoholism: Clinical and Experimental Research 22 (Suppl.), 4S14S.[ISI][Medline]
Rehm, J. and Gmel, G. (2003) Alcohol consumption and total mortality/morbidity-definitions and methodological implications. Best Practice & Research. Clinical Gastroenterology 17, 497505.[CrossRef][ISI][Medline]
Rehm, J., Ashley, M. J., Room, R. et al. (1996) On the emerging paradigm of drinking patterns and their social and health consequences. Addiction 91, 16151621.[CrossRef][ISI][Medline]
Rehm, J., Sempos, C. T. and Trevisan, M. (2003a) Alcohol and cardiovascular diseasemore than one paradox to consider. Average volume of alcohol consumption, patterns of drinking and risk of coronary heart diseasea review. Journal of Cardiovascular Risk 10, 1520.[CrossRef][ISI][Medline]
Rehm, J., Room, R., Graham, K. et al. (2003b) The relationship of average volume of alcohol consumption and patterns of drinking to burden of disease: an overview. Addiction 98, 12091228.[CrossRef][ISI][Medline]
Richter, V., Rassoul, F., Hentschel, B. et al. (2004) Age-dependence of lipid parameters in the general population and vegetarians. Zeitschrift fur Gerontologie und Geriatrie 37, 207213.[CrossRef][ISI]
Rimm, E. B., Williams, P., Fosher, K. et al. (1999) Moderate alcohol intake and lower risk of coronary heart disease: meta-analysis of effects on lipids and haemostatic factors. British Medical Journal 319, 15231528.
Rumpler, W. V., Clevidence, B. A., Muesing, R. A. et al. (1999) Changes in women's plasma lipid and lipoprotein concentrations due to moderate consumption of alcohol are affected by dietary fat level. Journal of Nutrition 129, 17131717.
Sierksma, A., Vermunt, S. H., Lankhuizen, I. M. et al. (2004) Effect of moderate alcohol consumption on parameters of reverse cholesterol transport in postmenopausal women. Alcoholism: Clinical and Experimental Research 28, 662666.[ISI][Medline]
Trevisan, M., Dorn, J., Falkner, K. et al. (2004) Drinking pattern and risk of non-fatal myocardial infarction: a population-based case-control study. Addiction 99, 31322.[CrossRef][ISI][Medline]
Yusuf, S., Hawken, S., Ounpuu, S. et al. (2004) Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): casecontrol study. Lancet 364, 937952.[CrossRef][ISI][Medline]
|