Binge Drinking and Mortality After Acute Myocardial Infarction
http://www.100md.com
《循环学杂志》
the Department of Medicine (K.J.M., M.A.M.), Beth Israel Deaconess Medical Center, and the Department of Epidemiology (M.M., M.A.M.), Harvard School of Public Heath, Boston, Mass
the School of Health Information Science (M.M.), University of Victoria, Victoria, British Columbia, Canada
the Division of Cardiology (J.E.M.), Massachusetts General Hospital, Boston, Mass.
Abstract
Background— Moderate drinkers have a lower risk of mortality after myocardial infarction (MI). Although binge drinking has been associated with a higher risk of MI in some studies, its relation to prognosis after MI is uncertain.
Methods and Results— In a prospective, inception cohort study conducted at 45 US hospitals, 1935 patients hospitalized with a confirmed MI between 1989 and 1994 underwent detailed personal interviews. Patients reported their usual frequency of binge drinking of beer, wine, and liquor, defined as intake of 3 or more drinks within 1 to 2 hours, and were followed up for mortality for a median of 3.8 years. Of 1919 eligible patients, 250 (94% men) reported binge drinking during the prior year, and a total of 318 patients died during follow-up. Binge drinkers had a 2-fold higher risk of mortality than drinkers who did not binge (hazard ratio, 2.0; 95% confidence interval, 1.3 to 3.0). A comparison of 192 binge drinkers and 192 other patients matched on propensity scores yielded a similar result. The association between binge drinking and total mortality tended to be similar among patients whose usual alcohol intake was light or heavier and for binge drinkers who consumed beer, wine, or liquor. Usual alcohol intake was inversely associated with mortality, but binge drinking completely attenuated this relation.
Conclusions— Our results suggest that alcohol consumption may be linked to potential hazards among patients who survive acute MI. Although moderate intake has been associated with lower mortality, binge drinking, even among light drinkers, appears to be associated with 2-fold higher mortality.
Key Words: alcohol epidemiology myocardial infarction prognosis
Introduction
In the general population, individuals who consume 1 alcoholic beverage every 1 to 2 days have lower rates of myocardial infarction (MI) than do abstainers or heavier drinkers.1 At the same time, heavy episodic (or binge) drinking appears to be associated with a higher risk,2,3 although the relation is not as well defined. For example, a Swedish case-control study found that drinking to intoxication or drinking at least a half bottle of spirits on occasion was associated with a slightly lower risk of MI among men but a higher risk among women.4 In a Canadian cohort, Murray and colleagues5 found that drinking 8 or more drinks at 1 occasion was associated with an 8-year relative risk for coronary heart disease (CHD) of 2.26 (95% confidence interval [CI], 1.22 to 4.20) among men, but the corresponding hazard ratio (HR) among men enrolled in the Lung Health Study was 0.83 (95% CI, 0.62 to 1.11).6
Editorial p 3818
Moderate drinking may be associated with a better prognosis among patients with established CHD. Some studies of patients with CHD, including recent MI, distant MI, and ischemic cardiomyopathy, have reported lower mortality rates among patients who consumed alcohol in moderation,7–9 although others have not.10,11 Likewise, we previously reported that average alcohol intake is inversely associated with mortality among patients with acute MI enrolled in the Determinants of Myocardial Infarction Onset Study (Onset Study),12 but none of these earlier studies examined the potentially detrimental effects of binge drinking on patients with CHD.
For these reasons, we studied binge drinking and mortality after acute MI among Onset Study participants. This multicenter, prospective cohort study included chart reviews and face-to-face interviews with hospitalized patients with confirmed acute MI.12–14
Methods
Onset Study Enrollment and Data Collection
The Onset Study was conducted in 45 community and tertiary-care medical centers. Between August 1989 and September 1994, 1935 patients (601 women and 1334 men) were interviewed a median of 4 days after sustaining a confirmed MI. Trained research interviewers identified eligible patients by reviewing coronary care unit admission logs and patient charts. For inclusion, patients were required to have a creatine kinase level above the upper limit of normal, positive MB isoenzymes, an identifiable onset of symptoms of infarction, and the ability to complete a structured interview. For these analyses, we excluded patients with missing information about usual alcohol consumption or binge drinking (n=16), leaving 1919 eligible patients for analysis. The institutional review board of each center approved this protocol, and each participant gave informed consent.
Interviewers used a structured data abstraction and questionnaire form. Participants reported their usual frequency of consumption (and corresponding numbers of drinks) during the preceding year of wine, beer, and liquor individually. As in previous analyses,12 we computed each patient’s average weekly intake of ethanol based on their reported consumption and the estimated ethanol content for a serving of each beverage (13.2 g for beer, 10.8 g for wine, and 15.1 g for liquor15) and categorized usual weekly alcohol consumption as none, light (<105 g of ethanol; &7 drinks), or heavier (105 g of ethanol). In a validation analysis of 115 participants with measured levels of HDL cholesterol, alcohol intake was correlated with these levels to the expected degree (Pearson r=0.21, P=0.02).16,17
As a measure of binge drinking during the preceding year, patients reported the usual frequency with which they consumed "3 drinks or more (drinks) in 1 to 2 hours" for each of the beverage types. We summed the number of such episodes to determine the usual frequency of binge drinking. In beverage-specific analyses, we examined patients who reported heavy episodic intake of beer only, wine only, liquor only, or multiple beverages.
Other information collected from each interview and chart review included demographics, medical history, and medication use (prescription and nonprescription) at the time of hospitalization. During chart review, interviewers recorded complications of congestive heart failure (CHF) or ventricular arrhythmias based on clinical diagnoses documented in the medical record. Interviewers also collected initial blood pressure on admission and all creatine kinase values available at the time of chart review (median=4).
We defined initial hypotension as a presenting systolic blood pressure of <90 mm Hg, and current aspirin use, as reported use of any aspirin-containing product in the 4 days before the index MI; for other medications, regular use included active use up to the time of admission. We used 1990 US census data to derive median household income from US Postal Service zip codes (n=1857). Participants reported their educational attainment and marital status during interviews. We derived body mass index from self-reported height and weight. We defined noncardiac comorbidity as any diagnosis of cancer, respiratory disease, renal failure, or stroke recorded in the medical record.
We searched the National Death Index for deaths of Onset Study participants through January 1, 1996 and requested death certificates from state offices of vital records for all probable matches according to a previously validated algorithm.18 Three physicians independently verified the determination of each death. Two physicians categorized the cause of each death as due to cardiovascular disease or noncardiovascular disease. Disagreements among raters were resolved by discussion. All-cause mortality was the primary outcome in all analyses, with cardiovascular and noncardiovascular mortality as secondary outcomes.
Statistical Analysis
We used Cox proportional-hazards models to examine the relation of binge drinking to mortality after adjustment. We first adjusted for age, sex, and their interaction.14 In subsequent models, we further adjusted for race, body mass index (as linear and quadratic terms), marital status (married versus other), current smoking, previous smoking, usual frequency of exertion (in 3 categories), household income (in quartiles), educational attainment (in 3 categories), previous MI, previous CHF, diabetes mellitus, hypertension, noncardiac comorbidity, previous medication use (aspirin, -adrenergic antagonists, calcium channel blockers, digoxin, diuretics, hypolipidemic agents, and angiotensin-converting-enzyme inhibitors individually), and receipt of thrombolytic therapy. We assigned indicator variables for patients with missing information on educational attainment (n=58), household income (n=56), smoking (n=20), and marital status (n=22) and set body mass index to the mean for 22 patients with missing information. Models that excluded individuals with any missing covariate information yielded similar results and are not shown here.
As a sensitivity analysis, we matched binge drinkers with an equal number of other patients by propensity scores.19 Each patient’s score is that individual’s probability that he or she would report binge drinking, based on demographic, behavioral, and clinical characteristics. To create propensity scores, we used multivariable logistic regression with a dependent variable of binge drinking, and the independent variables were all covariates used in the Cox models noted earlier, except for receipt of thrombolytic therapy and with the addition of linear and quadratic terms for age and alcohol use and interaction terms of sex with age and race. The area under the receiver operating characteristics curve was 0.90, indicating excellent discrimination for the logistic model. We then individually matched binge drinkers to unique non–binge-drinking patients with a nearest-available-pair greedy matching method.20 We were successfully able to match 192 binge drinkers to 192 unique patients who did not report binge drinking; the remaining 58 had disjoint ranges of propensity scores. We compared Kaplan-Meier survival plots of these 384 patients by the log-rank test.
We tested the proportionality of hazards by time-varying covariates21 and found no significant violations. We used standard methods to estimate the fraction of deaths attributable to binge drinking among current drinkers.22 In beverage-type analyses, we compared patients who reported binge consumption of only 1 beverage or mixed beverages to a single reference category of all nonbinge drinkers. We conducted statistical analyses with the SAS System for Windows, release 8.01 (SAS Institute, Inc).
Results
Patient Characteristics
Table 1 shows the characteristics of the Onset Study participants according to their alcohol consumption. As expected,23 binge drinkers tended to be younger than other patients, had heavier usual alcohol consumption, and were more likely to be male, current smokers, and divorced or separated. Only 8 binge drinkers (3%) had a history of alcohol abuse documented in the medical record. Among binge drinkers, the median frequency of binge drinking episodes was 1 per week (interquartile range, 1/month to 5/week).
Binge Drinking and Mortality
A total of 318 patients died during a median of 3.8 years of follow-up. Table 2 shows HRs for mortality by comparing binge drinkers with other current drinkers. Binge drinking was associated with 2-fold greater total mortality among current drinkers, both in age- and sex-adjusted models and in more fully adjusted models. When compared with abstainers rather than other drinkers, binge drinking was associated with an HR of 1.4 (95% CI, 0.9 to 2.2; P=0.09).
As shown in Table 2, binge drinking was associated with 2-fold greater post–acute MI mortality from both all causes and cardiovascular causes. In addition, the risk of noncardiovascular mortality was &80% higher but was not statistically significant. Given that 25% of current drinkers reported binge drinking and the 2-fold greater total mortality associated with it, we estimate that 20% of deaths (95% CI, 7% to 33%) among current drinkers in this study were attributable to binge drinking.
Table 3 shows a series of factors that could alter the positive association between binge drinking and total mortality. As expected, adjustment for usual alcohol intake strengthened the association between binge drinking and mortality, because usual intake is inversely associated with mortality in the Onset Study.12 When stratified, there were similar associations between binge drinking and mortality among light and heavier drinkers (based on usual alcohol intake) and among patients who reported episodes up to once per week or more frequent episodes.
The risk associated with binge drinking was similar among patients who reported episodes of heavy drinking of beer, wine, liquor, or multiple beverages. Analyses that assessed the risk associated with binge drinking of any given beverage, simultaneously adjusting for binge intake of the other 2 beverages, yielded similar results (P for interaction, 0.27).
We also performed additional sensitivity analyses. To minimize possible overfitting, we performed stepwise models (with entry and stay criteria of P=0.20) that gave very similar results, as did a model adjusted for propensity score with no other covariates. Models that excluded patients who died during hospitalization did not materially change our results, nor did further adjustment for tea intake,13 measures of infarct size and severity (initial hypotension, complications of CHF or ventricular arrhythmias during hospitalization, and peak creatine kinase levels), the timing of last alcohol consumption before the index MI (<6, 6 to <24, or 24 hours), or current smoking in 4 categories (70, 71 to 140, 141 to 280, and 280 cigarettes/week). We also repeated our analyses, further adjusting for use of coffee, marijuana, and cocaine or heroin among the 1857 participants with complete information on these covariates; the HR associated with binge drinking was 1.8 (95% CI, 1.2 to 2.8). Finally, we explored the dose-response relation between frequency of binge episodes and risk. Although a positive trend among all binge drinkers was not statistically significant (P for trend, 0.15), this appeared disproportionately influenced by the 20 individuals at the extreme of binge frequency. Among the other 92% of binge drinkers, there appeared to be a positive relation between binge frequency and mortality (P for trend, 0.03).
Stratified Analyses
We conducted analyses of participants in subgroups of age and sex to assess the homogeneity of our findings further. Although few women reported binge drinking, the risks associated with binge drinking among current drinkers appeared to be similar among men (HR, 1.7; 95% CI, 1.1 to 2.7) and women (HR 1.9, 95% CI, 0.2 to 15.1). The HR associated with binge drinking was 1.5 (95% CI, 0.8 to 2.6) among participants younger than 65 years of age and 2.1 (95% CI, 1.1 to 4.1) among participants aged 65 years and older.
Matched-Pair Analysis
Figure 1 shows the estimated survival of 192 binge drinkers and 192 other patients matched on propensity scores. These groups were similar in baseline characteristics, including mean age (52.9 versus 53.1 years), mean body mass index (27.4 versus 27.8 kg/m2), and the proportions of women (6% versus 7%), nonwhites (11% versus 10%), and current smokers (45% versus 42%). Clinical characteristics of the index infarction were comparable, with similar proportions receiving thrombolytic therapy (45% versus 41%) and peak creatine kinase levels (1730 versus 1745 IU/L). Among these 384 matched patients, binge drinkers had significantly worse survival than other patients (P=0.05); the unadjusted HR was 1.8 (95% CI, 1.0 to 3.3). When we excluded 36 abstainers and their matched binge drinkers, the corresponding HR was 2.1 (95% CI, 1.0 to 4.2).
Usual Drinking, Binge Drinking, and Mortality
To illustrate the joint effects of usual alcohol intake and binge drinking, we examined the risks among light and heavier drinkers who did and did not report binge drinking (Figure 2). Alcohol intake was inversely associated with mortality among patients who did not report binge drinking (P for trend, 0.009), with HRs of 0.75 (95% CI, 0.57 to 1.00) among nonbinging light drinkers and of 0.59 (95% CI, 0.33 to 1.04) among nonbinging heavier drinkers. Binge drinking completely attenuated this relation (P for trend, 0.83), with HRs of 1.58 (95% CI, 0.91 to 2.75) among binging light drinkers and 1.30 (95% CI, 0.76 to 2.22) among binging heavier drinkers.
Discussion
In this inception cohort study of early survivors of acute MI, binge drinking, measured at the time of hospitalization, was associated with 2-fold higher subsequent mortality. The magnitude of this association was similar for total, cardiovascular, and noncardiovascular mortality and among light and heavier drinkers.
Although several reports,7–9 including one from the Onset Study,12 have suggested that light to moderate alcohol intake may be associated with a better prognosis than abstention among patients with CHD, our findings suggest that any potential benefits of moderate drinking must be weighed against the risks of even occasional binge drinking. Three fourths of all Americans who report at least occasional binge drinking are otherwise light or moderate drinkers.23 In our analyses, the apparent benefit associated with otherwise light drinking among patients with acute MI was completely eliminated by episodes of binge drinking.
One difficulty in generalizing our results is that no single uniform measure of binge drinking exists.4,5,24–28 Some studies have used measures of regular heavy drinking,29 whereas others have used episodic heavy drinking.5 Even previous definitions of episodic heavy drinking show considerable heterogeneity, with variability in the quantity of alcohol that defines a binge and the time period during which such drinking occurs. It seems likely that all of these measures capture some portion of the same underlying construct, but no specific measure has an unequivocal advantage over others.30 In support of their similarity, the Onset Study definition (3 drinks within 2 hours) was correlated with the same sociodemographic characteristics linked to binge drinking according to other definitions,5,23 including age, sex, smoking, marital status, and usual alcohol consumption. However, in the absence of studies with multiple assessment methods, we cannot determine the magnitude and speed of drinking that is most strongly associated with postinfarction mortality. Given the relatively conservative nature of the Onset Study definition, our results highlight the serious potential risks of drinking to excess or intoxication.
Binge drinking may have adverse consequences both on cardiovascular physiology and thrombosis/fibrinolysis. Canine models show acute coronary vasoconstriction with alcohol exposure.31 Single episodes of heavy drinking increase both blood pressure and heart rate,32 and trials with stress tests have confirmed greater ischemia after short-term heavy alcohol intake.33 Large amounts of alcohol given to nonalcoholic men increase platelet reactivity, thromboxane B2 formation, and plasminogen activator inhibitor-1 activity.34–36 In one controlled trial, heavy but not moderate drinking increased plasminogen activator inhibitor-1 activity, decreased tissue plasminogen activator activity, and prolonged clot lysis time, in some cases for >12 hours.37 These effects could clearly interact to trigger recurrent MI and thus, to increase mortality among those with a recent MI.
Binge alcohol use could also contribute to sudden cardiac death (SCD) related to ventricular arrhythmias. Heavy episodic consumption appears to lower the ventricular fibrillation threshold in response to both electric38 and mechanical39 provocation and increases the risk of SCD.40 This is likely to be a particularly important factor for MI survivors, who have a 4- to 6-fold increased baseline risk of SCD.41
Although the ethanol content of a typical serving of beer, wine, or liquor is relatively similar, the amount of ethanol in a standard serving can vary substantially between countries. For example, standard drinks (12 oz of beer, 5 oz of wine, or 1.5 oz of liquor) contain &13 to 14 g in the United States, but 1 unit of alcohol in the United Kingdom contains 8 g; despite this discrepancy, recommended limits for drinking (in grams of ethanol) are similar in the 2 regions.42,43 However, discrepancies such as these complicate cross-national comparisons of binge drinking. The definition used here (3 drinks in 1 to 2 hours) represents &40 to 45 g of ethanol, but appropriate scaling would be needed to replicate our findings elsewhere.
The Onset Study is subject to limitations. As with any observational study, we cannot prove cause-and-effect relations, and there may be unmeasured aspects of lifestyle that are responsible in part for our findings.44 For example, we adjusted for educational attainment and income associated with zip code of residence (a measure at the community level), but we did not have information on current personal income. Comparisons of binge drinkers with other drinkers, rather than abstainers, may be somewhat less susceptible to confounding because of the greater comparability of these groups in sociodemographic features.
Participants reported their usual drinking habits during the year before hospitalization, and we do not have information on posthospitalization drinking patterns. Limited evidence suggests that preinfarction and postinfarction alcohol consumption patterns are strongly correlated.12,45 If some binge drinkers cease such drinking after hospitalization, we may have underestimated the true effect of postinfarction binge drinking on survival, but variation over time appears unlikely to have produced an overestimate of risk.46 On the other hand, by determining each patient’s usual alcohol consumption before hospitalization and the onset of follow-up, we minimized the possibility of differential misclassification caused by sicker patients giving up binge drinking more often than healthier patients during the follow-up period. For the most accurate results, future studies will need to assess both pre-MI and post-MI drinking patterns and correct for changes in intake among sicker patients.
We do not have information on how Onset Study participants changed their lifestyle or complied with medical therapy after hospitalization. Numerous studies suggest that heavier alcohol consumption directly affects smoking cessation, likelihood of relapse among ex-smokers, and medication compliance.47–49 Because binge drinking is likely to affect these factors directly, lack of adjustment for differences in posthospitalization behaviors is appropriate to estimate the full effect of binge drinking. However, this limitation precludes determination of whether our findings may relate to cardiovascular or neurobehavioral effects of alcohol.
Our patients were hospitalized during the thrombolytic era, before widespread use of primary angioplasty and coronary stenting. Additional studies are needed to determine whether current treatment approaches modify the hazards associated with binge drinking.
Despite the size of the Onset Study, the power of our analyses was limited in some cases, particularly in analyses of subgroups and interactions. Larger studies with different measures of binge drinking will be needed to describe possible dose-response relations in the quantity of alcohol consumed per binge and the consistency of the association among women and across beverage types and usual levels of intake.
In conclusion, binge drinking appears to be associated with 2-fold greater mortality after acute MI in the Onset Study. Because binge drinking occurs frequently even in otherwise light drinkers, this finding highlights the potential hazards of inappropriate alcohol consumption at all levels of usual intake. Although determination of the risks and benefits of regular drinking for an individual requires consideration of numerous personal, clinical, and social factors best addressed by a physician, we encourage physicians and patients to consider the detrimental consequences that accompany alcohol use when not consumed with care.
Acknowledgments
This study was supported by National Institute of Health grants R01HL41016 and R21AA014900 and American Heart Association grant 9630115N for the Determinants of Myocardial Infarction Onset Study.
Disclosures
None.
Footnotes
Presented in part at the 2004 American Heart Association Scientific Sessions, New Orleans, La, November 7–10, and published in abstract form (Circulation. 2004;110[suppl III]:III-743).
References
Corrao G, Rubbiati L, Bagnardi V, Zambon A, Poikolainen K. Alcohol and coronary heart disease: a meta-analysis. Addiction. 2000; 95: 1505–1523.
Trevisan M, Dorn J, Falkner K, Russell M, Ram M, Muti P, Freudenheim JL, Nochajaski T, Hovey K. Drinking pattern and risk of non-fatal myocardial infarction: a population-based case-control study. Addiction. 2004; 99: 313–322.
McElduff P, Dobson AJ. How much alcohol and how often population based case-control study of alcohol consumption and risk of a major coronary event. BMJ. 1997; 314: 1159–1164.
Hammar N, Romelsjo A, Alfredsson L. Alcohol consumption, drinking pattern and acute myocardial infarction: a case referent study based on the Swedish Twin Register. J Intern Med. 1997; 241: 125–131.
Murray RP, Connett JE, Tyas SL, Bond R, Ekuma O, Silversides CK, Barnes GE. Alcohol volume, drinking pattern, and cardiovascular disease morbidity and mortality: is there a U-shaped function Am J Epidemiol. 2002; 155: 242–248.
Murray RP, Connett JE, Makela P, Rehm J. Difficulty in demonstrating a risk from drinking pattern in fourteen years of coronary heart disease morbidity and mortality: the Lung Health Study. Addict Behav. 2005; 30: 875–887.
Muntwyler J, Hennekens CH, Buring JE, Gaziano JM. Mortality and light to moderate alcohol consumption after myocardial infarction. Lancet. 1998; 352: 1882–1885.
Cooper HA, Exner DV, Domanski MJ. Light-to-moderate alcohol consumption and prognosis in patients with left ventricular systolic dysfunction. J Am Coll Cardiol. 2000; 35: 1753–1759.
de Lorgeril M, Salen P, Martin JL, Boucher F, Paillard F, de Leiris J. Wine drinking and risks of cardiovascular complications after recent acute myocardial infarction. Circulation. 2002; 106: 1465–1469.
Shaper AG, Wannamethee SG. Alcohol intake and mortality in middle aged men with diagnosed coronary heart disease. Heart. 2000; 83: 394–399.
Aguilar D, Skali H, Moye LA, Lewis EF, Gaziano JM, Rutherford JD, Hartley LH, Randall OS, Geltman EM, Lamas GA, Rouleau JL, Pfeffer MA, Solomon SD. Alcohol consumption and prognosis in patients with left ventricular systolic dysfunction after a myocardial infarction. J Am Coll Cardiol. 2004; 43: 2015–2021.
Mukamal KJ, Maclure M, Muller JE, Sherwood JB, Mittleman MA. Prior alcohol consumption and mortality following acute myocardial infarction. JAMA. 2001; 285: 1965–1970.
Mukamal KJ, Maclure M, Muller JE, Sherwood JB, Mittleman MA. Tea consumption and mortality after acute myocardial infarction. Circulation. 2002; 105: 2476–2481.
Mukamal KJ, Muller JE, Maclure M, Sherwood JB, Mittleman MA. Evaluation of sex-related differences in survival after hospitalization for acute myocardial infarction. Am J Cardiol. 2001; 88: 768–771.
Consumer and Food Economics Institute. Composition of Foods: Handbook 8 series, 1976–1989. Washington, DC: US Department of Agriculture, GPO; 1989.
Linn S, Carroll M, Johnson C, Fulwood R, Kalsbeek W, Briefel R. High-density lipoprotein cholesterol and alcohol consumption in US white and black adults: data from NHANES II. Am J Public Health. 1993; 83: 811–816.
Ellison RC, Myers RH, Zhang Y, Djousse L, Knox S, Williams RR, Province MA. Effects of similarities in lifestyle habits on familial aggregation of high density lipoprotein and low density lipoprotein cholesterol: the NHLBI Family Heart Study. Am J Epidemiol. 1999; 150: 910–918.
Stampfer MJ, Willett WC, Speizer FE, Dysert DC, Lipnick R, Rosner B, Hennekens CH. Test of the National Death Index. Am J Epidemiol. 1984; 119: 837–839.
Joffe MM, Rosenbaum PR. Invited commentary: propensity scores. Am J Epidemiol. 1999; 150: 327–333.
Parsons LS. Reducing bias in a propensity score matched-pair sample using greedy matching techniques. Proceedings of the Twenty-Sixth Annual SAS Users Group International Conference. Cary, NC: SAS Institute, Inc; 2001.
Ng’andu NH. An empirical comparison of statistical tests for assessing the proportional hazards assumption of Cox’s model. Stat Med. 1997; 16: 611–626.
Wacholder S, Benichou J, Heineman EF, Hartge P, Hoover RN. Attributable risk: advantages of a broad definition of exposure. Am J Epidemiol. 1994; 140: 303–309.
Naimi TS, Brewer RD, Mokdad A, Denny C, Serdula MK, Marks JS. Binge drinking among US adults. JAMA. 2003; 289: 70–75.
Midanik LT. Drunkenness, feeling the effects and 5+ measures. Addiction. 1999; 94: 887–897.
Wechsler H, Austin SB. Binge drinking: the five/four measure. J Stud Alcohol. 1998; 59: 122–124.
Rehm J, Greenfield TK, Rogers JD. Average volume of alcohol consumption, patterns of drinking, and all-cause mortality: results from the US National Alcohol Survey. Am J Epidemiol. 2001; 153: 64–71.
Kauhanen J, Kaplan GA, Goldberg DE, Salonen JT. Beer binging and mortality: results from the Kuopio ischaemic heart disease risk factor study, a prospective population based study. BMJ. 1997; 315: 846–851.
Makela P, Fonager K, Hibell B, Nordlund S, Sabroe S, Simpura J. Episodic heavy drinking in four Nordic countries: a comparative survey. Addiction. 2001; 96: 1575–1588.
Kauhanen J, Kaplan GA, Goldberg DE, Salonen R, Salonen JT. Pattern of alcohol drinking and progression of atherosclerosis. Arterioscler Thromb Vasc Biol. 1999; 19: 3001–3006.
Borsari B, Neal DJ, Collins SE, Carey KB. Differential utility of three indexes of risky drinking for predicting alcohol problems in college students. Psychol Addict Behav. 2001; 15: 321–324.
Rogers PJ, Bove AA. Epicardial coronary artery constriction with intravenous ethanol. Int J Cardiol. 1989; 22: 301–310.
Seppa K, Sillanaukee P. Binge drinking and ambulatory blood pressure. Hypertension. 1999; 33: 79–82.
Rossinen J, Partanen J, Koskinen P, Toivonen L, Kupari M, Nieminen MS. Acute heavy alcohol intake increases silent myocardial ischaemia in patients with stable angina pectoris. Heart. 1996; 75: 563–567.
Hillbom M, Kangasaho M, Lowbeer C, Kaste M, Muuronen A, Numminen H. Effects of ethanol on platelet function. Alcohol. 1985; 2: 429–432.
Numminen H, Hillbom M, Vapaatalo H, Seppala E, Laustiola K, Benthin G, Muuronen A, Kaste M. Effects of exercise and ethanol ingestion on platelet thromboxane release in healthy men. Metabolism. 1991; 40: 695–701.
Numminen H, Syrjala M, Benthin G, Kaste M, Hillbom M. The effect of acute ingestion of a large dose of alcohol on the hemostatic system and its circadian variation. Stroke. 2000; 31: 1269–1273.
van de Wiel A, van Golde PM, Kraaijenhagen RJ, von dem Borne PA, Bouma BN, Hart HC. Acute inhibitory effect of alcohol on fibrinolysis. Eur J Clin Invest. 2001; 31: 164–170.
Khedun SM, Maharaj B, Lockett CJ, Leary WP. Influence of ethyl alcohol when given alone or in combination with potassium and/or magnesium supplements on ventricular fibrillation threshold levels in laboratory rats. Magnes Res. 1992; 5: 115–120.
Stein PD, Sabbah HN, Przybylski J, Goldberg DA, Hamid MS, Viano DC. Effect of alcohol upon arrhythmias following nonpenetrating cardiac impact. J Trauma. 1988; 28: 465–471.
McKee M, Britton A. The positive relationship between alcohol and heart disease in eastern Europe: potential physiological mechanisms. J R Soc Med. 1998; 91: 402–407.
American Heart Association. Heart Disease and Stroke Statistics—2004 Update. Dallas, Tex; American Heart Association; 2003.
US Department of Agriculture, US Department of Health and Human Services. Nutrition and Your Health: Dietary Guidelines for Americans, 5th ed. Washington, DC: US. Government Printing Office; 2000.
Inter-Departmental Working Group. Sensible Drinking. Wetherby, England: Department of Health; 1995.
Naimi TS, Brown DW, Brewer RD, Giles WH, Mensah G, Serdula MK, Mokdad AH, Hungerford DW, Lando J, Naimi S, Stroup DF. Cardiovascular risk factors and confounders among nondrinking and moderate-drinking U.S. adults. Am J Prev Med. 2005; 28: 369–373.
Cleophas TJ, Tuinenberg E, van der Meulen J, Zwinderman KH. Wine consumption and other dietary variables in males under 60 before and after acute myocardial infarction. Angiology. 1996; 47: 789–796.
Emberson JR, Shaper AG, Wannamethee SG, Morris RW, Whincup PH. Alcohol intake in middle age and risk of cardiovascular disease and mortality: accounting for intake variation over time. Am J Epidemiol. 2005; 161: 856–863.
Kim MT, Dennison CR, Hill MN, Bone LR, Levine DM. Relationship of alcohol and illicit drug use with high blood pressure care and control among urban hypertensive Black men. Ethn Dis. 2000; 10: 175–183.
Johnson KH, Bazargan M, Bing EG. Alcohol consumption and compliance among inner-city minority patients with type 2 diabetes mellitus. Arch Fam Med. 2000; 9: 964–970.
Shiffman S, Balabanis M. Associations between alcohol and tobacco. In: Fertig JB, Allen JP, eds. Alcohol and Tobacco: From Basic Science to Clinical Practice. Bethesda, Md: National Institute on Alcohol Abuse and Alcoholism; 1995.(Kenneth J. Mukamal, MD; Malcolm Maclure,)
the School of Health Information Science (M.M.), University of Victoria, Victoria, British Columbia, Canada
the Division of Cardiology (J.E.M.), Massachusetts General Hospital, Boston, Mass.
Abstract
Background— Moderate drinkers have a lower risk of mortality after myocardial infarction (MI). Although binge drinking has been associated with a higher risk of MI in some studies, its relation to prognosis after MI is uncertain.
Methods and Results— In a prospective, inception cohort study conducted at 45 US hospitals, 1935 patients hospitalized with a confirmed MI between 1989 and 1994 underwent detailed personal interviews. Patients reported their usual frequency of binge drinking of beer, wine, and liquor, defined as intake of 3 or more drinks within 1 to 2 hours, and were followed up for mortality for a median of 3.8 years. Of 1919 eligible patients, 250 (94% men) reported binge drinking during the prior year, and a total of 318 patients died during follow-up. Binge drinkers had a 2-fold higher risk of mortality than drinkers who did not binge (hazard ratio, 2.0; 95% confidence interval, 1.3 to 3.0). A comparison of 192 binge drinkers and 192 other patients matched on propensity scores yielded a similar result. The association between binge drinking and total mortality tended to be similar among patients whose usual alcohol intake was light or heavier and for binge drinkers who consumed beer, wine, or liquor. Usual alcohol intake was inversely associated with mortality, but binge drinking completely attenuated this relation.
Conclusions— Our results suggest that alcohol consumption may be linked to potential hazards among patients who survive acute MI. Although moderate intake has been associated with lower mortality, binge drinking, even among light drinkers, appears to be associated with 2-fold higher mortality.
Key Words: alcohol epidemiology myocardial infarction prognosis
Introduction
In the general population, individuals who consume 1 alcoholic beverage every 1 to 2 days have lower rates of myocardial infarction (MI) than do abstainers or heavier drinkers.1 At the same time, heavy episodic (or binge) drinking appears to be associated with a higher risk,2,3 although the relation is not as well defined. For example, a Swedish case-control study found that drinking to intoxication or drinking at least a half bottle of spirits on occasion was associated with a slightly lower risk of MI among men but a higher risk among women.4 In a Canadian cohort, Murray and colleagues5 found that drinking 8 or more drinks at 1 occasion was associated with an 8-year relative risk for coronary heart disease (CHD) of 2.26 (95% confidence interval [CI], 1.22 to 4.20) among men, but the corresponding hazard ratio (HR) among men enrolled in the Lung Health Study was 0.83 (95% CI, 0.62 to 1.11).6
Editorial p 3818
Moderate drinking may be associated with a better prognosis among patients with established CHD. Some studies of patients with CHD, including recent MI, distant MI, and ischemic cardiomyopathy, have reported lower mortality rates among patients who consumed alcohol in moderation,7–9 although others have not.10,11 Likewise, we previously reported that average alcohol intake is inversely associated with mortality among patients with acute MI enrolled in the Determinants of Myocardial Infarction Onset Study (Onset Study),12 but none of these earlier studies examined the potentially detrimental effects of binge drinking on patients with CHD.
For these reasons, we studied binge drinking and mortality after acute MI among Onset Study participants. This multicenter, prospective cohort study included chart reviews and face-to-face interviews with hospitalized patients with confirmed acute MI.12–14
Methods
Onset Study Enrollment and Data Collection
The Onset Study was conducted in 45 community and tertiary-care medical centers. Between August 1989 and September 1994, 1935 patients (601 women and 1334 men) were interviewed a median of 4 days after sustaining a confirmed MI. Trained research interviewers identified eligible patients by reviewing coronary care unit admission logs and patient charts. For inclusion, patients were required to have a creatine kinase level above the upper limit of normal, positive MB isoenzymes, an identifiable onset of symptoms of infarction, and the ability to complete a structured interview. For these analyses, we excluded patients with missing information about usual alcohol consumption or binge drinking (n=16), leaving 1919 eligible patients for analysis. The institutional review board of each center approved this protocol, and each participant gave informed consent.
Interviewers used a structured data abstraction and questionnaire form. Participants reported their usual frequency of consumption (and corresponding numbers of drinks) during the preceding year of wine, beer, and liquor individually. As in previous analyses,12 we computed each patient’s average weekly intake of ethanol based on their reported consumption and the estimated ethanol content for a serving of each beverage (13.2 g for beer, 10.8 g for wine, and 15.1 g for liquor15) and categorized usual weekly alcohol consumption as none, light (<105 g of ethanol; &7 drinks), or heavier (105 g of ethanol). In a validation analysis of 115 participants with measured levels of HDL cholesterol, alcohol intake was correlated with these levels to the expected degree (Pearson r=0.21, P=0.02).16,17
As a measure of binge drinking during the preceding year, patients reported the usual frequency with which they consumed "3 drinks or more (drinks) in 1 to 2 hours" for each of the beverage types. We summed the number of such episodes to determine the usual frequency of binge drinking. In beverage-specific analyses, we examined patients who reported heavy episodic intake of beer only, wine only, liquor only, or multiple beverages.
Other information collected from each interview and chart review included demographics, medical history, and medication use (prescription and nonprescription) at the time of hospitalization. During chart review, interviewers recorded complications of congestive heart failure (CHF) or ventricular arrhythmias based on clinical diagnoses documented in the medical record. Interviewers also collected initial blood pressure on admission and all creatine kinase values available at the time of chart review (median=4).
We defined initial hypotension as a presenting systolic blood pressure of <90 mm Hg, and current aspirin use, as reported use of any aspirin-containing product in the 4 days before the index MI; for other medications, regular use included active use up to the time of admission. We used 1990 US census data to derive median household income from US Postal Service zip codes (n=1857). Participants reported their educational attainment and marital status during interviews. We derived body mass index from self-reported height and weight. We defined noncardiac comorbidity as any diagnosis of cancer, respiratory disease, renal failure, or stroke recorded in the medical record.
We searched the National Death Index for deaths of Onset Study participants through January 1, 1996 and requested death certificates from state offices of vital records for all probable matches according to a previously validated algorithm.18 Three physicians independently verified the determination of each death. Two physicians categorized the cause of each death as due to cardiovascular disease or noncardiovascular disease. Disagreements among raters were resolved by discussion. All-cause mortality was the primary outcome in all analyses, with cardiovascular and noncardiovascular mortality as secondary outcomes.
Statistical Analysis
We used Cox proportional-hazards models to examine the relation of binge drinking to mortality after adjustment. We first adjusted for age, sex, and their interaction.14 In subsequent models, we further adjusted for race, body mass index (as linear and quadratic terms), marital status (married versus other), current smoking, previous smoking, usual frequency of exertion (in 3 categories), household income (in quartiles), educational attainment (in 3 categories), previous MI, previous CHF, diabetes mellitus, hypertension, noncardiac comorbidity, previous medication use (aspirin, -adrenergic antagonists, calcium channel blockers, digoxin, diuretics, hypolipidemic agents, and angiotensin-converting-enzyme inhibitors individually), and receipt of thrombolytic therapy. We assigned indicator variables for patients with missing information on educational attainment (n=58), household income (n=56), smoking (n=20), and marital status (n=22) and set body mass index to the mean for 22 patients with missing information. Models that excluded individuals with any missing covariate information yielded similar results and are not shown here.
As a sensitivity analysis, we matched binge drinkers with an equal number of other patients by propensity scores.19 Each patient’s score is that individual’s probability that he or she would report binge drinking, based on demographic, behavioral, and clinical characteristics. To create propensity scores, we used multivariable logistic regression with a dependent variable of binge drinking, and the independent variables were all covariates used in the Cox models noted earlier, except for receipt of thrombolytic therapy and with the addition of linear and quadratic terms for age and alcohol use and interaction terms of sex with age and race. The area under the receiver operating characteristics curve was 0.90, indicating excellent discrimination for the logistic model. We then individually matched binge drinkers to unique non–binge-drinking patients with a nearest-available-pair greedy matching method.20 We were successfully able to match 192 binge drinkers to 192 unique patients who did not report binge drinking; the remaining 58 had disjoint ranges of propensity scores. We compared Kaplan-Meier survival plots of these 384 patients by the log-rank test.
We tested the proportionality of hazards by time-varying covariates21 and found no significant violations. We used standard methods to estimate the fraction of deaths attributable to binge drinking among current drinkers.22 In beverage-type analyses, we compared patients who reported binge consumption of only 1 beverage or mixed beverages to a single reference category of all nonbinge drinkers. We conducted statistical analyses with the SAS System for Windows, release 8.01 (SAS Institute, Inc).
Results
Patient Characteristics
Table 1 shows the characteristics of the Onset Study participants according to their alcohol consumption. As expected,23 binge drinkers tended to be younger than other patients, had heavier usual alcohol consumption, and were more likely to be male, current smokers, and divorced or separated. Only 8 binge drinkers (3%) had a history of alcohol abuse documented in the medical record. Among binge drinkers, the median frequency of binge drinking episodes was 1 per week (interquartile range, 1/month to 5/week).
Binge Drinking and Mortality
A total of 318 patients died during a median of 3.8 years of follow-up. Table 2 shows HRs for mortality by comparing binge drinkers with other current drinkers. Binge drinking was associated with 2-fold greater total mortality among current drinkers, both in age- and sex-adjusted models and in more fully adjusted models. When compared with abstainers rather than other drinkers, binge drinking was associated with an HR of 1.4 (95% CI, 0.9 to 2.2; P=0.09).
As shown in Table 2, binge drinking was associated with 2-fold greater post–acute MI mortality from both all causes and cardiovascular causes. In addition, the risk of noncardiovascular mortality was &80% higher but was not statistically significant. Given that 25% of current drinkers reported binge drinking and the 2-fold greater total mortality associated with it, we estimate that 20% of deaths (95% CI, 7% to 33%) among current drinkers in this study were attributable to binge drinking.
Table 3 shows a series of factors that could alter the positive association between binge drinking and total mortality. As expected, adjustment for usual alcohol intake strengthened the association between binge drinking and mortality, because usual intake is inversely associated with mortality in the Onset Study.12 When stratified, there were similar associations between binge drinking and mortality among light and heavier drinkers (based on usual alcohol intake) and among patients who reported episodes up to once per week or more frequent episodes.
The risk associated with binge drinking was similar among patients who reported episodes of heavy drinking of beer, wine, liquor, or multiple beverages. Analyses that assessed the risk associated with binge drinking of any given beverage, simultaneously adjusting for binge intake of the other 2 beverages, yielded similar results (P for interaction, 0.27).
We also performed additional sensitivity analyses. To minimize possible overfitting, we performed stepwise models (with entry and stay criteria of P=0.20) that gave very similar results, as did a model adjusted for propensity score with no other covariates. Models that excluded patients who died during hospitalization did not materially change our results, nor did further adjustment for tea intake,13 measures of infarct size and severity (initial hypotension, complications of CHF or ventricular arrhythmias during hospitalization, and peak creatine kinase levels), the timing of last alcohol consumption before the index MI (<6, 6 to <24, or 24 hours), or current smoking in 4 categories (70, 71 to 140, 141 to 280, and 280 cigarettes/week). We also repeated our analyses, further adjusting for use of coffee, marijuana, and cocaine or heroin among the 1857 participants with complete information on these covariates; the HR associated with binge drinking was 1.8 (95% CI, 1.2 to 2.8). Finally, we explored the dose-response relation between frequency of binge episodes and risk. Although a positive trend among all binge drinkers was not statistically significant (P for trend, 0.15), this appeared disproportionately influenced by the 20 individuals at the extreme of binge frequency. Among the other 92% of binge drinkers, there appeared to be a positive relation between binge frequency and mortality (P for trend, 0.03).
Stratified Analyses
We conducted analyses of participants in subgroups of age and sex to assess the homogeneity of our findings further. Although few women reported binge drinking, the risks associated with binge drinking among current drinkers appeared to be similar among men (HR, 1.7; 95% CI, 1.1 to 2.7) and women (HR 1.9, 95% CI, 0.2 to 15.1). The HR associated with binge drinking was 1.5 (95% CI, 0.8 to 2.6) among participants younger than 65 years of age and 2.1 (95% CI, 1.1 to 4.1) among participants aged 65 years and older.
Matched-Pair Analysis
Figure 1 shows the estimated survival of 192 binge drinkers and 192 other patients matched on propensity scores. These groups were similar in baseline characteristics, including mean age (52.9 versus 53.1 years), mean body mass index (27.4 versus 27.8 kg/m2), and the proportions of women (6% versus 7%), nonwhites (11% versus 10%), and current smokers (45% versus 42%). Clinical characteristics of the index infarction were comparable, with similar proportions receiving thrombolytic therapy (45% versus 41%) and peak creatine kinase levels (1730 versus 1745 IU/L). Among these 384 matched patients, binge drinkers had significantly worse survival than other patients (P=0.05); the unadjusted HR was 1.8 (95% CI, 1.0 to 3.3). When we excluded 36 abstainers and their matched binge drinkers, the corresponding HR was 2.1 (95% CI, 1.0 to 4.2).
Usual Drinking, Binge Drinking, and Mortality
To illustrate the joint effects of usual alcohol intake and binge drinking, we examined the risks among light and heavier drinkers who did and did not report binge drinking (Figure 2). Alcohol intake was inversely associated with mortality among patients who did not report binge drinking (P for trend, 0.009), with HRs of 0.75 (95% CI, 0.57 to 1.00) among nonbinging light drinkers and of 0.59 (95% CI, 0.33 to 1.04) among nonbinging heavier drinkers. Binge drinking completely attenuated this relation (P for trend, 0.83), with HRs of 1.58 (95% CI, 0.91 to 2.75) among binging light drinkers and 1.30 (95% CI, 0.76 to 2.22) among binging heavier drinkers.
Discussion
In this inception cohort study of early survivors of acute MI, binge drinking, measured at the time of hospitalization, was associated with 2-fold higher subsequent mortality. The magnitude of this association was similar for total, cardiovascular, and noncardiovascular mortality and among light and heavier drinkers.
Although several reports,7–9 including one from the Onset Study,12 have suggested that light to moderate alcohol intake may be associated with a better prognosis than abstention among patients with CHD, our findings suggest that any potential benefits of moderate drinking must be weighed against the risks of even occasional binge drinking. Three fourths of all Americans who report at least occasional binge drinking are otherwise light or moderate drinkers.23 In our analyses, the apparent benefit associated with otherwise light drinking among patients with acute MI was completely eliminated by episodes of binge drinking.
One difficulty in generalizing our results is that no single uniform measure of binge drinking exists.4,5,24–28 Some studies have used measures of regular heavy drinking,29 whereas others have used episodic heavy drinking.5 Even previous definitions of episodic heavy drinking show considerable heterogeneity, with variability in the quantity of alcohol that defines a binge and the time period during which such drinking occurs. It seems likely that all of these measures capture some portion of the same underlying construct, but no specific measure has an unequivocal advantage over others.30 In support of their similarity, the Onset Study definition (3 drinks within 2 hours) was correlated with the same sociodemographic characteristics linked to binge drinking according to other definitions,5,23 including age, sex, smoking, marital status, and usual alcohol consumption. However, in the absence of studies with multiple assessment methods, we cannot determine the magnitude and speed of drinking that is most strongly associated with postinfarction mortality. Given the relatively conservative nature of the Onset Study definition, our results highlight the serious potential risks of drinking to excess or intoxication.
Binge drinking may have adverse consequences both on cardiovascular physiology and thrombosis/fibrinolysis. Canine models show acute coronary vasoconstriction with alcohol exposure.31 Single episodes of heavy drinking increase both blood pressure and heart rate,32 and trials with stress tests have confirmed greater ischemia after short-term heavy alcohol intake.33 Large amounts of alcohol given to nonalcoholic men increase platelet reactivity, thromboxane B2 formation, and plasminogen activator inhibitor-1 activity.34–36 In one controlled trial, heavy but not moderate drinking increased plasminogen activator inhibitor-1 activity, decreased tissue plasminogen activator activity, and prolonged clot lysis time, in some cases for >12 hours.37 These effects could clearly interact to trigger recurrent MI and thus, to increase mortality among those with a recent MI.
Binge alcohol use could also contribute to sudden cardiac death (SCD) related to ventricular arrhythmias. Heavy episodic consumption appears to lower the ventricular fibrillation threshold in response to both electric38 and mechanical39 provocation and increases the risk of SCD.40 This is likely to be a particularly important factor for MI survivors, who have a 4- to 6-fold increased baseline risk of SCD.41
Although the ethanol content of a typical serving of beer, wine, or liquor is relatively similar, the amount of ethanol in a standard serving can vary substantially between countries. For example, standard drinks (12 oz of beer, 5 oz of wine, or 1.5 oz of liquor) contain &13 to 14 g in the United States, but 1 unit of alcohol in the United Kingdom contains 8 g; despite this discrepancy, recommended limits for drinking (in grams of ethanol) are similar in the 2 regions.42,43 However, discrepancies such as these complicate cross-national comparisons of binge drinking. The definition used here (3 drinks in 1 to 2 hours) represents &40 to 45 g of ethanol, but appropriate scaling would be needed to replicate our findings elsewhere.
The Onset Study is subject to limitations. As with any observational study, we cannot prove cause-and-effect relations, and there may be unmeasured aspects of lifestyle that are responsible in part for our findings.44 For example, we adjusted for educational attainment and income associated with zip code of residence (a measure at the community level), but we did not have information on current personal income. Comparisons of binge drinkers with other drinkers, rather than abstainers, may be somewhat less susceptible to confounding because of the greater comparability of these groups in sociodemographic features.
Participants reported their usual drinking habits during the year before hospitalization, and we do not have information on posthospitalization drinking patterns. Limited evidence suggests that preinfarction and postinfarction alcohol consumption patterns are strongly correlated.12,45 If some binge drinkers cease such drinking after hospitalization, we may have underestimated the true effect of postinfarction binge drinking on survival, but variation over time appears unlikely to have produced an overestimate of risk.46 On the other hand, by determining each patient’s usual alcohol consumption before hospitalization and the onset of follow-up, we minimized the possibility of differential misclassification caused by sicker patients giving up binge drinking more often than healthier patients during the follow-up period. For the most accurate results, future studies will need to assess both pre-MI and post-MI drinking patterns and correct for changes in intake among sicker patients.
We do not have information on how Onset Study participants changed their lifestyle or complied with medical therapy after hospitalization. Numerous studies suggest that heavier alcohol consumption directly affects smoking cessation, likelihood of relapse among ex-smokers, and medication compliance.47–49 Because binge drinking is likely to affect these factors directly, lack of adjustment for differences in posthospitalization behaviors is appropriate to estimate the full effect of binge drinking. However, this limitation precludes determination of whether our findings may relate to cardiovascular or neurobehavioral effects of alcohol.
Our patients were hospitalized during the thrombolytic era, before widespread use of primary angioplasty and coronary stenting. Additional studies are needed to determine whether current treatment approaches modify the hazards associated with binge drinking.
Despite the size of the Onset Study, the power of our analyses was limited in some cases, particularly in analyses of subgroups and interactions. Larger studies with different measures of binge drinking will be needed to describe possible dose-response relations in the quantity of alcohol consumed per binge and the consistency of the association among women and across beverage types and usual levels of intake.
In conclusion, binge drinking appears to be associated with 2-fold greater mortality after acute MI in the Onset Study. Because binge drinking occurs frequently even in otherwise light drinkers, this finding highlights the potential hazards of inappropriate alcohol consumption at all levels of usual intake. Although determination of the risks and benefits of regular drinking for an individual requires consideration of numerous personal, clinical, and social factors best addressed by a physician, we encourage physicians and patients to consider the detrimental consequences that accompany alcohol use when not consumed with care.
Acknowledgments
This study was supported by National Institute of Health grants R01HL41016 and R21AA014900 and American Heart Association grant 9630115N for the Determinants of Myocardial Infarction Onset Study.
Disclosures
None.
Footnotes
Presented in part at the 2004 American Heart Association Scientific Sessions, New Orleans, La, November 7–10, and published in abstract form (Circulation. 2004;110[suppl III]:III-743).
References
Corrao G, Rubbiati L, Bagnardi V, Zambon A, Poikolainen K. Alcohol and coronary heart disease: a meta-analysis. Addiction. 2000; 95: 1505–1523.
Trevisan M, Dorn J, Falkner K, Russell M, Ram M, Muti P, Freudenheim JL, Nochajaski T, Hovey K. Drinking pattern and risk of non-fatal myocardial infarction: a population-based case-control study. Addiction. 2004; 99: 313–322.
McElduff P, Dobson AJ. How much alcohol and how often population based case-control study of alcohol consumption and risk of a major coronary event. BMJ. 1997; 314: 1159–1164.
Hammar N, Romelsjo A, Alfredsson L. Alcohol consumption, drinking pattern and acute myocardial infarction: a case referent study based on the Swedish Twin Register. J Intern Med. 1997; 241: 125–131.
Murray RP, Connett JE, Tyas SL, Bond R, Ekuma O, Silversides CK, Barnes GE. Alcohol volume, drinking pattern, and cardiovascular disease morbidity and mortality: is there a U-shaped function Am J Epidemiol. 2002; 155: 242–248.
Murray RP, Connett JE, Makela P, Rehm J. Difficulty in demonstrating a risk from drinking pattern in fourteen years of coronary heart disease morbidity and mortality: the Lung Health Study. Addict Behav. 2005; 30: 875–887.
Muntwyler J, Hennekens CH, Buring JE, Gaziano JM. Mortality and light to moderate alcohol consumption after myocardial infarction. Lancet. 1998; 352: 1882–1885.
Cooper HA, Exner DV, Domanski MJ. Light-to-moderate alcohol consumption and prognosis in patients with left ventricular systolic dysfunction. J Am Coll Cardiol. 2000; 35: 1753–1759.
de Lorgeril M, Salen P, Martin JL, Boucher F, Paillard F, de Leiris J. Wine drinking and risks of cardiovascular complications after recent acute myocardial infarction. Circulation. 2002; 106: 1465–1469.
Shaper AG, Wannamethee SG. Alcohol intake and mortality in middle aged men with diagnosed coronary heart disease. Heart. 2000; 83: 394–399.
Aguilar D, Skali H, Moye LA, Lewis EF, Gaziano JM, Rutherford JD, Hartley LH, Randall OS, Geltman EM, Lamas GA, Rouleau JL, Pfeffer MA, Solomon SD. Alcohol consumption and prognosis in patients with left ventricular systolic dysfunction after a myocardial infarction. J Am Coll Cardiol. 2004; 43: 2015–2021.
Mukamal KJ, Maclure M, Muller JE, Sherwood JB, Mittleman MA. Prior alcohol consumption and mortality following acute myocardial infarction. JAMA. 2001; 285: 1965–1970.
Mukamal KJ, Maclure M, Muller JE, Sherwood JB, Mittleman MA. Tea consumption and mortality after acute myocardial infarction. Circulation. 2002; 105: 2476–2481.
Mukamal KJ, Muller JE, Maclure M, Sherwood JB, Mittleman MA. Evaluation of sex-related differences in survival after hospitalization for acute myocardial infarction. Am J Cardiol. 2001; 88: 768–771.
Consumer and Food Economics Institute. Composition of Foods: Handbook 8 series, 1976–1989. Washington, DC: US Department of Agriculture, GPO; 1989.
Linn S, Carroll M, Johnson C, Fulwood R, Kalsbeek W, Briefel R. High-density lipoprotein cholesterol and alcohol consumption in US white and black adults: data from NHANES II. Am J Public Health. 1993; 83: 811–816.
Ellison RC, Myers RH, Zhang Y, Djousse L, Knox S, Williams RR, Province MA. Effects of similarities in lifestyle habits on familial aggregation of high density lipoprotein and low density lipoprotein cholesterol: the NHLBI Family Heart Study. Am J Epidemiol. 1999; 150: 910–918.
Stampfer MJ, Willett WC, Speizer FE, Dysert DC, Lipnick R, Rosner B, Hennekens CH. Test of the National Death Index. Am J Epidemiol. 1984; 119: 837–839.
Joffe MM, Rosenbaum PR. Invited commentary: propensity scores. Am J Epidemiol. 1999; 150: 327–333.
Parsons LS. Reducing bias in a propensity score matched-pair sample using greedy matching techniques. Proceedings of the Twenty-Sixth Annual SAS Users Group International Conference. Cary, NC: SAS Institute, Inc; 2001.
Ng’andu NH. An empirical comparison of statistical tests for assessing the proportional hazards assumption of Cox’s model. Stat Med. 1997; 16: 611–626.
Wacholder S, Benichou J, Heineman EF, Hartge P, Hoover RN. Attributable risk: advantages of a broad definition of exposure. Am J Epidemiol. 1994; 140: 303–309.
Naimi TS, Brewer RD, Mokdad A, Denny C, Serdula MK, Marks JS. Binge drinking among US adults. JAMA. 2003; 289: 70–75.
Midanik LT. Drunkenness, feeling the effects and 5+ measures. Addiction. 1999; 94: 887–897.
Wechsler H, Austin SB. Binge drinking: the five/four measure. J Stud Alcohol. 1998; 59: 122–124.
Rehm J, Greenfield TK, Rogers JD. Average volume of alcohol consumption, patterns of drinking, and all-cause mortality: results from the US National Alcohol Survey. Am J Epidemiol. 2001; 153: 64–71.
Kauhanen J, Kaplan GA, Goldberg DE, Salonen JT. Beer binging and mortality: results from the Kuopio ischaemic heart disease risk factor study, a prospective population based study. BMJ. 1997; 315: 846–851.
Makela P, Fonager K, Hibell B, Nordlund S, Sabroe S, Simpura J. Episodic heavy drinking in four Nordic countries: a comparative survey. Addiction. 2001; 96: 1575–1588.
Kauhanen J, Kaplan GA, Goldberg DE, Salonen R, Salonen JT. Pattern of alcohol drinking and progression of atherosclerosis. Arterioscler Thromb Vasc Biol. 1999; 19: 3001–3006.
Borsari B, Neal DJ, Collins SE, Carey KB. Differential utility of three indexes of risky drinking for predicting alcohol problems in college students. Psychol Addict Behav. 2001; 15: 321–324.
Rogers PJ, Bove AA. Epicardial coronary artery constriction with intravenous ethanol. Int J Cardiol. 1989; 22: 301–310.
Seppa K, Sillanaukee P. Binge drinking and ambulatory blood pressure. Hypertension. 1999; 33: 79–82.
Rossinen J, Partanen J, Koskinen P, Toivonen L, Kupari M, Nieminen MS. Acute heavy alcohol intake increases silent myocardial ischaemia in patients with stable angina pectoris. Heart. 1996; 75: 563–567.
Hillbom M, Kangasaho M, Lowbeer C, Kaste M, Muuronen A, Numminen H. Effects of ethanol on platelet function. Alcohol. 1985; 2: 429–432.
Numminen H, Hillbom M, Vapaatalo H, Seppala E, Laustiola K, Benthin G, Muuronen A, Kaste M. Effects of exercise and ethanol ingestion on platelet thromboxane release in healthy men. Metabolism. 1991; 40: 695–701.
Numminen H, Syrjala M, Benthin G, Kaste M, Hillbom M. The effect of acute ingestion of a large dose of alcohol on the hemostatic system and its circadian variation. Stroke. 2000; 31: 1269–1273.
van de Wiel A, van Golde PM, Kraaijenhagen RJ, von dem Borne PA, Bouma BN, Hart HC. Acute inhibitory effect of alcohol on fibrinolysis. Eur J Clin Invest. 2001; 31: 164–170.
Khedun SM, Maharaj B, Lockett CJ, Leary WP. Influence of ethyl alcohol when given alone or in combination with potassium and/or magnesium supplements on ventricular fibrillation threshold levels in laboratory rats. Magnes Res. 1992; 5: 115–120.
Stein PD, Sabbah HN, Przybylski J, Goldberg DA, Hamid MS, Viano DC. Effect of alcohol upon arrhythmias following nonpenetrating cardiac impact. J Trauma. 1988; 28: 465–471.
McKee M, Britton A. The positive relationship between alcohol and heart disease in eastern Europe: potential physiological mechanisms. J R Soc Med. 1998; 91: 402–407.
American Heart Association. Heart Disease and Stroke Statistics—2004 Update. Dallas, Tex; American Heart Association; 2003.
US Department of Agriculture, US Department of Health and Human Services. Nutrition and Your Health: Dietary Guidelines for Americans, 5th ed. Washington, DC: US. Government Printing Office; 2000.
Inter-Departmental Working Group. Sensible Drinking. Wetherby, England: Department of Health; 1995.
Naimi TS, Brown DW, Brewer RD, Giles WH, Mensah G, Serdula MK, Mokdad AH, Hungerford DW, Lando J, Naimi S, Stroup DF. Cardiovascular risk factors and confounders among nondrinking and moderate-drinking U.S. adults. Am J Prev Med. 2005; 28: 369–373.
Cleophas TJ, Tuinenberg E, van der Meulen J, Zwinderman KH. Wine consumption and other dietary variables in males under 60 before and after acute myocardial infarction. Angiology. 1996; 47: 789–796.
Emberson JR, Shaper AG, Wannamethee SG, Morris RW, Whincup PH. Alcohol intake in middle age and risk of cardiovascular disease and mortality: accounting for intake variation over time. Am J Epidemiol. 2005; 161: 856–863.
Kim MT, Dennison CR, Hill MN, Bone LR, Levine DM. Relationship of alcohol and illicit drug use with high blood pressure care and control among urban hypertensive Black men. Ethn Dis. 2000; 10: 175–183.
Johnson KH, Bazargan M, Bing EG. Alcohol consumption and compliance among inner-city minority patients with type 2 diabetes mellitus. Arch Fam Med. 2000; 9: 964–970.
Shiffman S, Balabanis M. Associations between alcohol and tobacco. In: Fertig JB, Allen JP, eds. Alcohol and Tobacco: From Basic Science to Clinical Practice. Bethesda, Md: National Institute on Alcohol Abuse and Alcoholism; 1995.(Kenneth J. Mukamal, MD; Malcolm Maclure,)