Cocarboxylase

SYNOPSIS.

The substantial medical risks of heavy alcohol drinking as well as the existence of a safe drinking limit have been evident for centuries. Modern epidemiologic studies also show lower risk of both morbidity and mortality among lighter drinkers. Defining "heavy" as > or =3 standard drinks per day, the alcohol-mortality relationship is a J-curve with risk highest for heavy drinkers, lowest for light drinkers and intermediate for abstainers. A number of non-cardiovascular and cardiovascular problems contribute to the increased mortality risk of heavier drinkers. The lower risk of light drinkers is due mostly to lower risk of the most common cardiovascular condition, coronary heart disease (CHD). Thus, disparate relationships of alcoholic drinking to various cardiovascular and non-cardiovascular conditions constitute a modern concept of alcohol and health. Increased cardiovascular risks of heavy drinking include: 1) alcoholic cardiomyopathy, 2) systemic hypertension (high blood pressure), 3) heart rhythm disturbances in binge drinkers, and 4) hemorrhagic stroke. Lighter drinking is unrelated to increased risk of any cardiovascular condition and, in observational studies, is consistently related to lower risk of CHD and ischemic stroke. A protective hypothesis for CHD is robustly supported by evidence for plausible biological mechanisms attributable to ethyl alcohol. International comparisons and some prospective study data suggest that wine is more protective against CHD than liquor or beer. Possible non-alcohol beneficial components in wine (especially red) support possible extra protection by wine, but a healthier pattern of drinking or more favorable risk traits in wine drinkers may also be involved.

INTRODUCTION

Few areas of study of life-style habits and cardiovascular diseases have been as exciting in recent decades as the relationships of alcohol drinking to these conditions. Because "modern" can be variously defined and historical perspective should never be ignored, this brief review includes discussion of concepts which have developed over the past two centuries. Disparities with respect to various cardiovascular conditions are increasingly evident (Klatsky, 1995). Some previous attempts to simplify the subject have slowed progress in understanding this area. Conditions needing separate consideration include cardiomyopathy, arsenic and cobalt beer drinkers' disease, cardiovascular beri-beri, systemic hypertension, cardiac arrhythmias, stroke, and atherosclerotic coronary heart disease (CHD). Total mortality in relation to alcohol drinking and definitions of safe drinking limits will also be briefly considered.

Basic disparity between effects of lighter and heavier drinking underlies all alcohol-health relationships. Preceding other reports of the J-shaped alcohol-mortality curve by half a century, Pearl (1926) described this relationship in a Baltimore, Maryland study. In his data, mortality was highest in "heavy/steady" drinkers; "abstainers" were next; and "moderate" drinkers had the lowest mortality. Recent studies have confirmed this, and have shown that the lower risk of lighter drinkers is due largely to lower CHD risk. Pearl did not know about the favorable CHD mortality of moderate drinkers. His report was presented during the U.S. prohibition era, a probable factor in his cautious interpretation that moderate drinking was "not harmful." Cultural context influences what research gets done and how it is interpreted, especially if the subject arouses strong feelings. Perhaps Pearl's major contribution was to realize the fallacy in comparing health risks of all drinkers to abstainers. Such comparison masks differences between the risks of heavy and light/moderate drinkers. As he memorably said: "one cannot judge the role of diet by starvation or excess."

DEFINITION OF MODERATE AND HEAVY DRINKING

Since definitions of moderate and heavy drinking are arbitrary, the operational boundary used here represents the level of drinking above which net harm is evident in epidemiologic studies. Thus, 3 or more drinks per day is called "heavy," and lesser amounts "light," "lighter," or "moderate" drinking. Individual factors, including sex and age, lower the boundary of safe drinking for some persons and raise it for others. In survey-based alcohol data there is always systematic underestimation, with some heavy drinkers alleging lighter intake. The result of this underestimation is usually a lower apparent threshold for harm.

Most persons think in terms of "drinks," not milliliters or grams of alcohol. Thus it is fortunate that a standard-sized drink of wine, liquor, or beer contains approximately the same amount of alcohol (12-14 g). It seems best to describe relationships of drinking to health in terms of drinks per day or week. When talking with clients, health professionals should always remember the importance of defining the size of drinks.

ALCOHOLIC CARDIOMYOPATHY

A relationship of chronic heavy drinking to heart disease was repeatedly noted in the 19lh century (Friedreich, 1961; Walsche, 1873; Strümpel, 1890; Steell, 1893). Noting an average yearly consumption of beer of 432 liters in Munich vs. 82 liters per year elsewhere in Germany Bollinger (1884) described the "Münchener bierherz," characterized by cardiac dilatation and hypertrophy. Graham Steell (1893) stated "Not only do I recognize alcoholism as one of the causes of muscle failure of the heart but I find it a comparatively common one." However, after a heart disease epidemic due to arsenic contaminated beer occurred in Manchester, England in 1900, he wrote "in the production of the combined affection of the peripheral nerves and the heart met with in beer drinkers, arsenic has been shown to play a conspicuous part" (Steell, 1906). In "The Study of the Pulse," William MacKenzie (1902) described heart failure from alcohol and used the term "alcoholic heart disease." After early descriptions of beri-beri (Aalsmeer and Wenckebach, 1929; Keefer, 1930), which later became known to represent vitamin B1 (cocarboxylase) deficiency, the concept of "beriberi heart disease" dominated thinking about alcohol and the heart. Despite widespread doubt in the early twentieth century that alcohol had a direct role in heart muscle disease, but some (Vaquez, 1921) took a strong view in favor of such a relationship.

Recent decades have seen renewed interest in possible direct toxicity of alcohol upon the myocardial cells. The concept of alcoholic cardiomyopathy is now solidly established by clinical observations, evidence of decreased myocardial function in many heavy chronic drinkers, and a few good controlled studies. Since the entity is indistinguishable from dilated cardiomyopathy of other causes, epidemiologic study is impaired by the absence of diagnostic tests. Most cases of dilated cardiomyopathy in 2004 remain of unknown cause, with post-viral auto-immune processes and genetic predisposition the leading etiologic hypotheses. Lifetime alcohol consumption is quantitatively related to structural and functional myocardial abnormalities in alcoholics (Urbano-Marquez et al, 1989). The equivalent of 120 grams alcohol/day for 20 years is needed to see an effect, making the term "cirrhosis of the heart" (Walsche, 1873) appropriate.

Only a small proportion of alcoholics develop cardiomyopathy, a fact leading to interest in predisposing traits which might increase risk. In this context, it is appropriate to consider further the arsenic and cobalt beer drinker episodes and thiamine (cocarboxylase) deficiency-or beri-beri heart disease.

ARSENIC-BEER DRINKERS' DISEASE

In 1900 an epidemic (>6,000 cases with >70 deaths) in and near Manchester, England, was proved due to accidental contamination of beer by arsenic. Manifestations involved the skin, nervous system, gastrointestinal system and, prominently, the cardiovascular system. A superb clinical description (Reynolds, 1901) included: (a) "cases were associated with so much heart failure and so little pigmentation that they were diagnosed as beri-beri-"; (b) "the principal cause of death has been cardiac failure," and (c) "at post-mortem-the only prominent signs were interstitial nephritis and the dilated flabby heart-."

It was estimated that the affected beer had 2-4 parts per million of arsenic, an amount not-in itself-likely to cause serious toxicity. Gowers (1901) mentioned prescribing for epilepsy 10 times the amount of arsenic involved over long periods of time without toxicity ("the amount of arsenic-was not sufficient to explain the poisoning"). Some seemed to have a "peculiar idiosyncrasy," in that "many persons became ill who drank less beer than others who were not affected." An appointed committee's report (Royal Commission, 1903) suggested that "alcohol predisposed people to arsenic poisoning." As best one can determine, no one suggested the converse.

COBALT-BEER DRINKERS' DISEASE

Recognized sixty-five years after the arsenic-beer episode, this condition was similar in some respects. Reports appeared of heart failure epidemics among beer drinkers in Omaha, Minneapolis, Quebec, and in Leuven, Belgium. The symptoms developed fairly abruptly in chronic heavy beer drinkers. The North American patients suffered a high mortality rate, but those who recovered did well despite return, by many, to previous beer habits.

The explanation proved to be the addition of small amounts of cobalt chloride by certain breweries to improve the foaming qualities of beer. Widespread use of detergents (new at that time) in taverns had a depressant effect upon foaming. Quebec investigators (Morin and Daniel, 1967) tracked down the etiology; the condition became justly known as Quebec beer-drinkers cardiomyopathy. Removal of the cobalt additive ended the epidemic in all locations.

In Belgium, the cobalt concentrations were less and the cardiac manifestations less severe, with more of the usual findings of chronic cobalt use (polycythemia and goiter). However, even in Quebec, where cobalt doses were greatest, 12 liters of contaminated beer provided only about 8 mg of cobalt, less than 20% of the dose sometimes used as a hematinic. The hematinic use had not been implicated as a cause of heart disease, whereas the first cases of this dramatic heart condition occurred 4-8 weeks after cobalt was added to beer,

It seems almost certain that both cobalt and substantial amounts of alcohol were needed to produce this condition. There must have been other factors, since most exposed persons did not develop the condition. Biochemical mechanisms were not established. Viewing the arsenic and cobalt episodes one observer (Alexander, 1969) commented: "This is the second known metal induced cardiotoxic syndrome produced by contaminated beer."

Speculative possibilities which have been mentioned as possible cofactors for alcoholic cardiomyopathy include cardiotropic viruses, drugs, selenium, copper, and iron. Deficiencies (zinc, magnesium, protein, and vitamins) have also been suggested as cofactors, but only thiamine deficiency is proven.

CARDIOVASCULAR BERI-BERI Aalsmeer and Wenckebach (1929) defined, in Javanese polished-rice eaters, high-output heart failure resulting from decreased peripheral vascular resistance. Many assumed that heart failure among Western heavy alcohol drinkers was due to associated nutritional deficiency states (Keefer, 1930). A few heart failure cases in North American and European alcoholics suited this hypothesis. Most did not, however, as they had low output heart failure, were well-nourished, and responded poorly to thiamine. Chronicity and ultimate irreversibility of beri-beri, was used by some to explain the situation. Blacket and Palmer (1960) sorted the conditions out: "It (beri-beri) responds completely to thiamine, but merges imperceptibly into another disease, called alcoholic cardiomyopathy, which doesn't respond to thiamine." Jn beriberi there is generalized peripheral arteriolar dilatation creating a large arteriovenous shunt and high resting cardiac outputs. A few cases of complete recovery with thiamine within 1-2 weeks have been documented.

Thus, many cases earlier called "cardiovascular beriberi" would now be called "alcoholic heart disease." Does chronic thiamine deficiency play a role in some cases of alcoholic cardiomyopathy? Currently unpopular, this hypothesis has not been proved or disproved.

HYPERTENSION (HIGH BLOOD PRESSURE)

Almost 90 years ago Lian (1916) reported a relationship between heavy drinking (mostly as wine) and hypertension (HTN) in WWl French servicemen. Unless these soldiers exaggerated, they were prodigious drinkers, as the HTN threshold appeared at >2 liters/ wine per day. No further attention was given to this subject for almost 60 years. Since the mid 1970s, dozens of cross-sectional and prospective epidemiologic studies (Beilin and Puddey, 1992; Klatsky, 2000) have solidly established an empiric alcohol-HTN link. The apparent threshold for this relationship is approximately 3 drinks per day. The studies involve both sexes, various ages and populations in North American, Europe, Australia, and Japan. Importantly, most studies show no increased HTN with lighter drinking; in fact, several show an unexplained J-shaped curve in women with lowest pressures in lighter drinkers. The relationship is independent from several potential confounders such as adiposity, salt intake, education, smoking, and beverage type (wine, liquor, or beer).

An early clinical experiment in hospitalized hypertensive men showed blood pressure increases with 34 days of drinking 4 pints of beer and decreases with 3-4 days of abstinence (Potter and Beevers, 1984). Similar results were later seen in ambulatory persons with both normal and high pressures (Puddey et al., 1985, 1987). Heavier drinking also impairs drug treatment of HTN and moderation or avoidance of alcohol supplements or betters other nonpharmacologic interventions such as weight reduction, exercise, or sodium restriction (Keil et al., 1993; Klatsky, 2000). Experiments do not show evidence that acute alcohol withdrawal causes alcohol-associated HTN. Even without establishing a mechanism, intervention studies strongly support a causal hypothesis. Alcohol restriction plays a role in management and prevention of this important very common condition.

ARRHYTHMIAS

Based on the observation that heart rhythm disturbances were more frequent on Mondays and in the Christmas-New Year's Day period, the term "holiday heart syndrome" was born (Ettinger et al., 1978). An association of heavier drinking with atrial arrhythmias had been suspected for decades, perhaps occurring especially after a large alcohol-accompanied meal. Atrial fibrillation is the commonest manifestation, typically resolving with abstinence, with or without specific treatment. In 1,322 persons reporting intake of > or =6 drinks per day a Kaiser Permanente study found a doubled relative risk of atrial fibrillation, atrial flutter, supraventricular tachycardia, and atrial premature complexes (Cohen et al., 1988).

STROKE

Prior to modern imaging techniques imprecise diagnosis of type of stroke was a limitation in analysis of alcohol-stroke relations. All studies of alcohol and stroke also have to deal with the complex disparate inter-relationships of stroke, alcohol, and other cardiovascular conditions. Some reports deal only with drinking sprees; others fail to differentiate hemorrhagic and ischemic strokes. Several of these reports suggested that alcohol drinking (heavier especially) carried increased risk of stroke (Van Gign et al., 1993). HTN is an important risk factor for all types of stroke and could be an intermediary between heavy alcohol drinking and increased risk of all stroke types. Antithrombotic effects of alcohol might increase hemorrhagic, but lower ischemic stroke risk. Blood lipid effects of alcohol (see CHD discussion below) might favorably affect ischemic stroke risk.

The importance of these deficiencies is highlighted by several recent studies suggesting that regular lighter drinkers may be at higher risk of hemorrhagic stroke types, but at lower risk of several types of ischemic stroke. For example, The Nurse's Health Study (Stampfer et al., 1988) showed drinkers to be at higher risk of subarachnoid hemorrhage, but lower risk of occlusive stroke. Kaiser Permanente studies (Klatsky et al., 1989, 2002a) found that daily consumption of >3 drinks, but not lighter drinking, was related to higher hospitalisation rates for hemorrhagic stroke with higher blood pressure a partial mediator of this relationship. However, alcohol use was associated with lower hospitalization rates for ischemic stroke, a relationship present in both sexes, whites and blacks, and for extracranial and intracerebral occlusive lesions (Klatsky et al, 1989, 2001a). Except for the effect of binges, the relationship of heavier drinking to ischemic stroke risk remains unclear.

Thus, there is increased hemorrhagic stroke risk among heavy drinkers, but no current consensus about relationships of alcohol drinking to various types of stroke (Klatsky, 2002b).

CORONARY HFART DISEASE (CHD)

Although incidence is decreasing in developed countries, CHD remains the leading cause of death in men and women. Since it causes a majority of all cardiovascular deaths, CHD dominates statistics for cardiovascular mortality and has substantial impact upon total mortality. Population studies have uncovered several, probably causal, CHD risk factors, including cigarette smoking, HTN, diabetes mellitus, high low-density lipoprotein (LDL) cholesterol, and low high-density lipoprotein (HDL) cholesterol. Sometimes, the LDL is called the "bad" cholesterol and HDL the "good" cholesterol. Atherosclerotic narrowing of major epicardial vessels is the usual basis, with clot formation in narrowed vessels playing a critical role in major events, such as acute myocardial infarction ("heart attack") or sudden death. Angina pectoris is a common symptom of CHD. Early studies of alcoholics and problem drinkers suggested a high CHD rate, but these studies did not allow for the role of traits associated with alcoholism, such as cigarette smoking. Studies of heavy drinkers can tell nothing about the role of light-moderate drinking.

The classic description of angina pectoris (Heberden, 1786) included: "Wine and spirituous Liquors and opium-afford considerable relief." Thus, it was widely presumed that alcohol is a coronary vasodilator (Osier, 1899; White, 1931; Levine, 1951). However, exercise ECG test data (Russek et al., 1950; Orlando et al., 1976) suggest that alcohol does not improve myocardial oxygen deficiency, and may mask symptoms by a sedative/analgesic effect. Thus, symptomatic benefit is likely to be dangerously misleading in patients with angina who drink before exercise, since available data suggest no major immediate effect of alcohol upon coronary blood flow (Renaud et al., 1993; Klatsky, 1994). In the early 1900s an inverse relationship between alcohol consumption and atherosclerotic disease (including CHD) was reported (Cabot, 1904; Hultgen, 1910; Leary, 1931). A "solvent" action of alcohol was suggested, another explanation offered was that premature deaths in heavier drinkers precluded development of CHD (Wilens, 1947; Ruebner et al., 1961; Parrish and Eberly, 1961). Population and case-control studies in the past 28 years have solidly established an inverse relation between alcohol drinking and fatal or nonfatal CHD (Renaud et al, 1993; Klatsky, 1994). Since data supporting plausible protective mechanisms against CHD by alcohol have also appeared (Renaud et al., 1993; Klatsky, 1994; Klatsky, 2001b), it now seems likely that alcohol drinking protects against CHD.

In 1819 Dr. Samuel Black, an Irish physician interested in angina pectoris perceptive about epidemiologic aspects, wrote (Black, 1819) what is probably the first commentary about the "French Paradox." Noting much angina in Ireland but observing little discussion in the writings of French physicians, his explanation of the presumed disparity was "the French habits and modes of living, coinciding with the benignity of their climate and the peculiar character of their moral affections." It was to be 160 years until presentation of international comparison data showing less CHD mortality in wine drinking countries than in countries where beer or liquor drinking predominate (St. Leger et al., 1979). Confirmatory international comparisons plus reports of nonalcoholic antioxidant phenolic compounds and antithrombotic substances in wine, especially red wine (Renaud, 1993; Rimm, 1996; Klatsky, 1994, 1997), have created great interest in this area. However, prospective population studies show no consensus about the wine/liquor/beer issue (Rimm, 1996; Klatsky, 1997). Kaiser Permanente studies show that both beer and wine drinkers have lower CHD hospitalization rates than liquor drinkers (Klatsky, 1997), but that wine drinkers clearly have the lowest CHD mortality (Klatsky, 2003a). For both CHD endpoints (fatal and non-fatal) the Kaiser Permanente studies show no difference between drinkers of red wine and white wine. The beverage choice issue remains unresolved at this time.

THE "SENSIBLE DRINKING LIMIT"

Attempts to define a safe limit are hardly new, since the medical risks of heavier drinking and the relative safety of lighter drinking have long been evident. Probably the most famous such limit has been known for more than 100 years as "Anstie's Rule (Anstie, 1870). The rule suggested an upper limit of approximately three standard drinks daily. In the mid-19th century, the limit was intended to apply primarily to mature men, but Sir Anstie was a distinguished neurologist and public health activist who emphasized individual variability in the ability to handle alcohol. Modern scientific advances have added little; the threshold for net harm in most population studies is exactly where Anstie, using common-sense observation, placed his limit. Now, as then, considerations of age, sex, and individual risks and benefits become the foci of any discussion (Friedman and Klatsky 1993; Klatsky, 2001b, 2003b) in which a health practitioner advises his or her client about alcohol drinking.

1 From the Symposium In Vino Veritax: The Comparative Biology of Alcohol Consumption presented at the Annual Meeting of the Society for Inegratve and Comparative Biology, 5-9 January 2004, at New Orleans, Louisiana.

REFERENCES

Aalsmeer, W. C. and K. E Wenckebach, 1929. Herz und Kreislauf bei der Beri- Beri Krankheit. Wien. Arch. Inn. Med. 16:193272.

Anstie, R E. 1870. On the uses of wines in health, and disease. J S Redfield, New York, New york:11-13.

Alexander, C. S. 1969. Cobalt and the heart. Ann. Intern. Med. 70: 411-413.

Beilin, L. J. and I. B. Puddey. 1992. Alcohol and hypertension. Clin Exp. Hypertens. Theory Pract. A14(1 and 2):119-138.

Black, S. 1819. Clinical and pathological reports. Alex Wilkinson, Newry, England: 1-47.

Blacket, R. B. and A. J. Palmer. 1960. Haemodynarnic studies in high output beri-beri. Br. Heart J. 22:483-501.

Bollinger, O. 1884. Ueber die Haussigkeit und Ursachen der idiopathischen Herzhypertrophie in Munchen. Disch. Med. Wochensch. (Stuttgart) 10:180.

Cabot, R. C. 1904. The relation of alcohol to arteriosclerosis. JAMA 43:774-775.

Cohen, E. J., A. L. Klatsky, and M. A. Armstrong. 1988. Alcohol use and supraventricular arrhythmia. Am. J. Cardiol. 62:971973.

Ettinger, P. O., C. E, Wu, C. De La Cruz, Jr, A. B. Weisse, S. S, Ahmed, and T. J. Regan. 1978. Arrhythmias and the "holiday heart": Alcohol-associated cardiac rhythm disorders. Am. Heart J. 95:555-562.

Friedman, G. D. and Klatsky, A. L. 1993. Is alcohol good for your health? N. Engl. J. Med. 329:1882-3.

Friedreich, N. 1861. Handbuch der speziellen Pathologic und Therapie. 5th Sect. Krankheiten des Herzens, Ferdinand Enke, Erlangen, The Netherlands.

Gowers, W. R. 1901. In Royal Medical and Chiururgical Society. Epidemic of arsenical poisoning in beer-drinkers in the north of England during the year 1900. Lancet 1:98-100.

Heberden, W. 1786. Some account of a disorder of the breast. Med. Trans. R. Col. Physicians (London) 2:59-67.

Hultgen, J. F. 1910. Alcohol and nephritis: Clinical study of 460 cases of chronic alcoholism. JAMA 55:279-281.

Keefer, C. S. 1930. Archives of Internal Medicine 45:1-22.

Keil, U., J. D, Swales, and D. E. Grobbee. 1993. Alcohol intake and its relation to hypertension. In P. M. Verschuren (ed.), Health issues related to alcohol consumption, pp. 17-42. ILSI Press, Washington DC.

Klatsky, A. L. 1994. Epidemiology of coronary heart disease-influence of alcohol. Alcohol Clin. Exp. Res. 18:88-96.

Klatsky, A. L. 1995. Cardiovascular effects of alcohol. Scientific American Sci. and Med. 2:28-37.

Klatsky, A. L. 2000. Alcohol and hypertension. In S. Operil and M. Weber (eds.), Hypertension. 2nd ed., pp. 211-220. W. B. Saunders Co., Philadelphia, Pennsylvania.

Klatsky, A. L. 20016. Should patients with heart disease drink alcohol? (Editorial) JAMA 285:2004-5.

Klatsky, A. L. 20026. Alcohol consumption and stroke-the difficulties in giving responsible advice. Addiction 97:103-104.

Klatsky, A. L. 20036. Drink to your health? Sci. Am. 288:74-81.

Klatsky, A. L., M. A. Armstrong, and G. D. Friedman. 1989. Alcohol use and subsequent cerebrovascular disease hospitalizations. Stroke 20:741-6.

Klatsky, A. L., M. A. Armstrong, and D. G. Friedman. 1997. Red wine, white wine, liquor, beer, and risk for coronary artery disease hospitalization. Am. J. Cardiol. 80:416-420.

Klatsky, A. L., M. A. Armstrong, S. Sidney, and G. D. Friedman. 2001a. Alcohol drinking and risk of ischemic stroke. Am. J. Cardiol 88:703-6.

Klatsky, A. L., M. A. Armstrong, S. Sidney, and G. D. Friedman. 2002a. Alcohol drinking and risk of hemorrhagic stroke. Neuroepidemiology 21:115-122.

Klatsky, A. L., G. D. Friedman, M. A. Armstrong, and H. Kipp, 2003[alpha]. Wine, liquor, beer and mortality. Am. J. Epidemiol. 158: 585-95.

Leary, T. 1931. Therapeutic value of alcohol, with special consideration of relations of alcohol to cholesterol, and thus to diabetes, to arteriosclerosis, and to gallstones. N. Eng. J. Med. 205: 231-242.

Levine, S. A. 1951. Clinical heart disease. 4th ed. Saunders, Philadelphia, Pennsylvania, p. 98.

Lian, C. 1915. L'alcoholisme cause d'hypertension arterielle. Bull. Acad. Med. (Paris) 74:525-28.

MacKenzie, J. 1902. The study of the pulse. Y. J. Pentland, Edinburgh and London, United Kingdom, p. 237.

Morin, Y. and P. Daniel. 1967. Quebec beer-drinkers' cardiomyopathy: Etiologic considerations. Can. Med. Assoc. J. 97:926-928.

Orlando, J., W. S. Aronow, J. Cassidy, and P. Prakash. 1976. Effect of ethanol on angina pectoris. Ann. Intern. Med. 84:652-655.

Osier, W. 1899. The principles and practice of medicine. 3rd ed. Appleton, New York, New york.

Parrish, H. M. and H. L. Eberly 1961. Negative association of coronary atherosclerosis with liver cirrhosis and chronic alcoholism-a statistical fallacy. J. Indiana. State Med. Assoc. 54:341347.

Potter, J. F. and D. J. Beevers, 1984. Pressor effect of alcohol in hypertension. Lancet 1:119-22.

Puddey, I. B., I. J., Beilin, and R. Vandongen. 1987. Regular alcohol use raises blood pressure in treated hypertensives. Lancet 1:647651.

Puddey, I. B., I. J. Beilin, R. Vandongen, I. L. Rouse, and P. Rogers. 1985. Evidence of a direct effect of alcohol consumption on blood pressure in normotensive men: A randomized controlled trial. Hypertension 7:707-13.

Pearl, R. 1926. Alcohol and longevity. Knopf, New York, New York.

Renaud, S., M. H. Criqui, G. Farchi, and J. Veenstra. 1993. Alcohol drinking and coronary heart disease. In P. M. Verschuren (ed.), Health issues related to alcohol consumption, pp. 81-124. ILSI Press, Washington, DC.

Reynolds, E. S. 1901. An account of the epidemic outbreak of arsenical poisoning occurring in beer drinkers in the North of England and the Midland Counties in 1900. Lancet i: 166-170.

Rimm, E., A. L. Klatsky, D. Grobbee, and M. J. Stampfer. 1996. Review of moderate alcohol consumption and reduced risk of coronary heart disease: Is the effect due to beer, wine, or spirits? B. M. J. 312:731-6.

Royal Commission Appointed to Inquire into Arsenical Poisoning from the Consumption of Beer and other Articles of Food or Drink. 1903. "Final Report," Part I. Wyman and Sons, London, England.

Ruebner, B. H., K. J. Miyai, and H. Abbey. 1961. The low incidence of myocardial infarction in hepatic cirrhosis-a statistical artefact? Lancet ii:1435-1436.

Russek, H. I., C. F. Naegele, and E D. Regan. 1950. Alcohol in the treatment of angina pectoris. JAMA 143:355-357.

St. Leger, A. S., A. L. Cochrane, and F Moore. 1979. Factors associated with cardiac mortality in developed countries with particular reference to the consumption of wine. Lancet 1:1017-20.

Stampfer, M. J., G. A. Colditz, W. C. Willett, F. E. Speizer, and C. H. Hennekens. 1988. Prospective study of moderate alcohol consumption and the risk of coronary disease and stroke in women. N. Eng. J. Med. 319:267-73.

Steell G. 1893. Heart failure as a result of chronic alcoholism. Med. Chron. Manchester 18:1-22.

Steell, G. 1906. In Textbook on diseases of the heart. Blakiston, Philadelphia, p. 79.

Strümpel, A. 1890. A textbook of medicine. Appleton, New York. p. 294.

Urbano-Marquez, A., R. Estruch, F. Navarro-Lopez, J. M. Grau, L. Mont, and E. Rubin. 1989. The effects of alcoholism on skeletal and cardiac muscle. N. Engl. J. Med. 320:409-415.

Van Gign, J., M. J. Stampfer, C. Wolfe, and A. Algra. 1993. The association between alcohol consumption and stroke. In P. M. Verschuren (ed.), Health issues related to alcohol consumption, pp. 43-80. ILSI Press, Washington, DC.

Vaquez, H. 1921. Maladies du Coeur. Bailiiere et Fils, Paris, France, p. 308.

Walsche, W. H. 1873. Diseases of the heart and great vessels. 4th ed. Smith, Elder, London, England.

White, P. D. 1931. Heart disease. Macmillan, New York, New York, p. 436.

Wilens, S. L. 1947. The relationship of chronic alcoholism to atherosclerosis. JAMA 135:1136-38.

ARTHUR L. KLATSKY2

Senior Consultant in Cardiology, Kaiser Permanente Medical Center, 280 West Mac-Arthur Boulevard, Oakland, California 94611

2 E-mail: hartmavn@pacbell.net or alk@dor.kaiser.org

Copyright Society for Integrative and Comparative Biology Aug 2004
Provided by ProQuest Information and Learning Company. All rights Reserved