HDL cholesterol may account for about 45 percent of the decrease in CAD associated with moderate drinking. However, Seppa et al. (1992) found no significant difference in the HDL/total cholesterol ratio between abstainers and moderate drinkers. More studies are therefore needed to determine the role, if any, of lipoproteins in protecting from CAD in moderate drinking.

Recently, attention has been focused on the role that acetate, a metabolite of alcohol (see chapter 7) may play in the protective effect of alcohol against coronary heart disease and myocardial ischemia. Acetate infusion into the heart resulted in a significant increase in coronary blood flow (Blaise et al. 1989). Acetate produced from alcohol metabolism can result in the production of adenosine (see chapter 7) (Vinay et al. 1987). Adenosine mediates a wide range of physiological functions in the heart (including coronary vasodilation and decrease in myocardial energy) (Ely and Berne 1992), liver, platelets, and brain (Stiles 1992). In the heart, adenosine has been associated with dilation of small coronary arteries and a reduction in the heart rate (Pelleg and Porter 1990).

Another mechanism by which light to moderate alcohol consumption may decrease risk for coronary heart disease is by interfering with thrombosis (Langer et al. 1992). Blood platelets (thrombocytes) are involved in the proliferation of smooth vascular cells and in the formation of thrombi, which ultimately results in impeding blood flow in small coronary arteries and in precipitating acute myocardial infarction. The aggregation of blood platelets and the secretion of their contents depend on the formation and function of a hormone-like substance known as thromboxane (Packham et al. 1987). Blood alcohol concentrations of 100 to 200 mg/dL have been observed to reduce thromboxane production by isolated platelets (Jakubowski et al. 1988; Rand et al. 1988) and to diminish significantly the formation of thrombi (Rand et al. 1991) and the aggregation of blood platelets (Mikhaildis et al. 1990), even those from cholesterol-fed animals (Rand et al. 1992). Another animal model for thrombus formation showed that alcohol inhibited thrombus formation in vivo and potentiated the inhibitory effect of aspirin on platelet thrombus formation (Keller and Folts 1990). Inhibition of platelet activity following chronic alcohol consumption has been proposed as a mechanism for the apparent ability of alcohol to protect against coronary heart disease (Lands and Zakhari 1990; Renaud and DeLorgeril 1992;

Renaud et al. 1992). This inhibition of platelet activity was reported to be connected to calciummediated events (Benistant and Rubin 1990) and cyclic adenosine monophosphate levels in the platelets (Rubin and Hoek 1988). However, the same mechanism may be responsible for hemorrhagic strokes induced by alcohol (see later section on cerebrovascular disease).

In postmenopausal women, the apparent protective effect of alcohol may be explained, in part, by an alcohol-induced increase in estrogen levels (Gavaler et al. 1991). Other plausible mechanisms have been discussed elsewhere (Zakhari 1991).

The following section focuses on deleterious cardiovascular effects associated with chronic alcohol consumption.

Hypertension

Hypertension is a major risk factor for cerebrovascular hemorrhage (or hemorrhagic stroke) as well as myocardial infarction (heart attack) (Buckley and Miller 1988). Numerous studies (reviewed by MacMahon 1987) relate alcohol consumption to a greater risk for hypertension. The risk for hypertension (see review by Pohorecky 1990) among individuals drinking three to four drinks per day was 50 percent higher than among nondrinkers; the risk among individuals consuming six to seven drinks per day was 100 percent higher (MacMahon 1987). In men, 11 to 30 percent of all hypertension can be attributed to heavy alcohol consumption (see review by MacMahon 1987). Some studies have shown that the risk for hypertension may increase with consumption of more than three drinks per day (Moore et al. 1990). In a followup study of women who had normal blood pressure at baseline, Witteman et al. (1990) found that two to three drinks per day were associated with a 40 percent greater risk for hypertension. Alcoholinduced hypertension was reported to be reversible after abstinence from drinking (Klatsky 1990; Puddey et al. 1987).

Many epidemiological studies investigating the relationship between hypertension and alcohol consumption have measured the average quantity of drinking. A recent study demonstrated the importance of measuring frequency as well as amount of drinking (Russell et al. 1991). In this study, the positive relationship between alcohol consumed and hypertension was discussed in terms of frequency. Mean systolic and diastolic blood pressure was 6.6 and

4.7 mmHg higher, respectively, among daily drinkers than among individuals who drank less than once a week.

Several mechanisms have been postulated as underlying the hypertensive effects of chronic alcohol consumption (Zakhari 1991). Chronic alcohol use may induce hypertension by activating the sympathoadrenal system, by inducing vasospasm caused by magnesium loss (Altura and Altura 1989), or by affecting hormonal systems such as aldosterone and the renin-angiotensin system (Zakhari 1991).

Several mechanisms have been postulated as underlying the hypertensive effects of chronic alcohol consumption. Chronic alcohol use may induce hypertension by activating the sympathoadrenal system, by inducing vasospasm caused by magnesium loss, or by affecting hormonal systems such as aldosterone and the renin-angiotensin system.

Alcoholic Cardiomyopathy

The cardinal feature of cardiomyopathy is a dilated, weakened heart muscle, which can be caused by chronic alcohol abuse, infections, autoimmune diseases, or other toxic substances (Moushmoush and Abi-Mansour 1991) (see figure 2). In the end stages of the disease, which affects both chambers of the heart, dysfunction of the left ventricular chamber of the heart may lead to pulmonary hypertension (increase in blood pressure in the lung vessels) and subsequently may result in right-sided heart failure (Regan 1990). Alcoholic cardiomyopathy is clinically and pathologically indistinguishable from other forms of congestive cardiomyopathy; its diagnosis depends on documenting a clinical history of alcohol abuse. It is estimated that 20 to 30 percent of cardiomyopathy cases can be attributed to alcohol abuse (Regan 1990). Lifetime alcohol consumption was found to be significantly correlated with a decrease in ventricular efficiency, dilation of the left ventricle, and a greater left ventricular mass (Urbano-Marquez 1989). In addition, preclinical cardiomyopathy has been observed in a large proportion of alcoholics at autopsy (Davidson 1989).

Alcoholic cardiomyopathy could be attributed to the attenuation of the contractile capacity of heart muscle induced by disturbing electrolyte balances, for instance, by decreasing calcium uptake and impairing sodium flux (Thomas et al. 1989). Further, alcohol-induced mitochondrial damage by impaired protein metabolism and the formation in the heart of FAEE (Bora et al. 1989; Lange and Kinnunen 1987) may contribute to cellular energy depression (Preedy and Peters 1990; Schreiber 1989; Zakhari 1991).

The pathogenesis of myocardial disturbances caused by chronic alcohol consumption may be linked to lipid peroxidation and free radical formation, as discussed earlier in this chapter. A study on cardiac cells (myocytes) from rats revealed a relationship between lipofuscin accumulation (a measure of free radical damage) and alcohol concentration, thus suggesting that lipid peroxidation is involved in cellular damage to heart muscle cells (Sohal et al. 1989). Several studies have reported on the occurrence in the myocardium of enhanced lipid peroxidation following chronic alcohol administration (Antonenkov et al. 1989). Acetaldehyde generation may also play a role in myocardial damage (Reinke et al. 1988).

Abstinence from alcohol may improve the clinical course of cardiomyopathy in approximately 30 percent of patients if the condition has not advanced to congestive heart failure or heart muscle degeneration (Rubin and Doria 1990).

Alcohol-Related Arrhythmias

Arrhythmias, or disturbances of the normal heart rhythm, are commonly observed in association with acute alcohol intoxication and prolonged alcohol consumption. Atrial fibrillation and ventricular dysrhythmia are the most common disturbances associated with alcohol use (Regan 1990). In one study, 63 percent of patients with acute atrial fibrillation were found to have a history of heavy alcohol use, as compared with 33 percent of normal controls (Rich 1985). In another study, consumption of six or more drinks per day was associated with twice the risk for atrial fibrillation, atrial flutter, tachycardia, or premature atrial complexes (Cohen et al. 1988). The high incidence of sudden death in alcoholics may be explained in part by arrhythmias (Beard et al. 1986; Suhonen et al. 1987).

Disturbances in the normal heart rhythm induced by alcohol may be linked to disruptions

in the generation of heart impulses because alcohol may affect the automatic function of the sinoatrial node (special tissue in the right atrium of the heart that generates impulses that cause heart contraction) or may prolong impulse conduction (Zakhari 1991). In addition, by stimulating catecholamine release, alcohol may provoke tachycardia (rapid heart rate) (Greenspon and Schaal 1983; Zakhari 1991). Arrhythmias secondary to withdrawal have been reported (Zakhari 1991).

Cerebrovascular Disease

Stroke is a major cause of mortality as well as a significant contributor to long-term disability in the United States. Stroke refers to an impaired flow of blood in the brain caused by blockage of blood flow (ischemic stroke) or by loss of blood through the vessel wall (hemorrhagic stroke). Males typically have higher rates of stroke than females, and blacks have a higher incidence of stroke than whites (Camargo 1989). Heavy drinking (about five drinks per day) was associated with a fourfold increase in the risk of hemorrhagic stroke, but the relationship between alcohol consumption and ischemic stroke may be U-shaped, that is, light drinking may decrease risk, whereas heavy drinking is associated with increased risk (Gorelick 1989; Klatsky et al. 1989). As discussed earlier, alcohol consumption may decrease the thrombogenic activity of platelets, thereby increasing the likelihood of hemorrhage. Other mechanisms by which alcohol influences the risk for stroke include increased blood pressure and increased arteriospasm (Altura and Altura 1989; Zhang et al. 1992).

Thus, alcohol abuse confers risk for hypertension, cardiomyopathy, sudden death, and hemorrhagic stroke, whereas light to moderate alcohol consumption is associated with a decreased risk for heart attack and ischemic stroke.

Alcohol-Induced Immune

System Impairment

The immune system is a complex network of specialized cells (such as lymphocytes and macrophages) and tissues (such as the thymus gland, spleen, lymph nodes, and bone marrow) designed to defend the body against such foreign substances as viruses, bacteria, or parasites. These foreign substances, which are capable of triggering an immune response, are called antigens. A substantial fraction of lymphocytes and

macrophages comprise the pool of cells found in blood and the lymph, a clear fluid that bathes body tissues (Grossman and Wilson 1992).

Lymphocytes are white blood cells that play the major role in defending the body against antigens. There are two broad classes of lymphocytes: the B lymphocytes (which mature, or become immunocompetent, in the Bone marrow) and the T lymphocytes (which migrate to the Thymus gland to mature). B cells work primarily by secreting substances called antibodies into body fluids, or humors (hence the term "humoral immunity"), whereas T cells interact directly with body cells that have been invaded by antigens or deformed by malignancy (hence the term "cellular immunity").

Antibodies produced by B cells interlock with the antigens and thereby mark them for destruction. Antibodies belong to a family of large molecules known as immunoglobulins. T lymphocytes consist of a series of subtypes, some of which mediate such regulatory functions as the ability to help other cells (for instance, as B cells and macrophages) to develop immune responses (these T cells are known as helper, T4, or CD4 cells) or the ability to suppress other immune cells (these T cells are known as suppressor cells). Cytotoxic T cells (T8 or CD8 cells) are directly involved in the destruction of agents bearing antigenic substances. Figure 3 shows cytotoxic T cells attacking a cancer cell. T cells work primarily by secreting substances known as lymphokines (such as interferon or interleukin-2), which are potent chemical messengers. In addition to the two major lymphocytes

Figure 3. Cytotoxic T lymphocytes attack a cancer cell.

Electron micrograph by Lennart Nilsson. Boehringer Ingelheim International GmbH. Reproduced by permission.

(B and T), other lymphoid cells known as natural killer (NK) cells provide important defense against spontaneously arising tumor cells, in particular blood-borne tumor cells, and certain forms of infectious diseases.

Chronic alcohol abuse depresses the immune system and results in a predisposition to infectious diseases, including respiratory infections, pneumonia, and tuberculosis, and even cancer (Adams and Jordan 1984; Bautista, D'Souza, Bagby et al. 1991; Bautista, D'Souza, Lang et al. 1991; Jerrells 1991 a, 1991b). Chronic alcohol consumption not only suppresses the proliferation of lymphocytes in blood, spleen, and thymus (Jerrells et al. 1986; Mutchnick and Lee 1988) but also decreases the ability of the lymphocytes to perform their function. For example, the ability of lymphocytes to produce antibodies against foreign antigens has been shown to be compromised by alcohol (Bagasra et al. 1987; Jerrells et al. 1988, 1989; Stefanini et al. 1989; Watson et al. 1988). This effect was ascribed to the inability of T cells to use interleukin-2, an important growth factor for T cells (Jerrells 1991a). Furthermore, alcohol suppressed the activity of NK cells, which provide an important defense against metastasizing tumor cells (Irwin, Caldwell et al. 1990; Irwin, Hauger et al. 1990; Meadows et al. 1989, 1990). Indeed, Yirmiya et al. (1991) found that chronic administration of alcohol to rats significantly increased the number of lung metastases of mammary adenocarcinoma.

Alcoholics, with or without liver disease, have been reported to have increased helper cell function, decreased suppressor cell function, or both, thereby disrupting the regulation of the immune response (Watson 1988). Furthermore, activation of hepatic phagocytes by endotoxins was compromised by alcohol (Bautista, D'Souza, Bagby et al. 1991; Bautista, D'Souza, Lang et al. 1991; Spitzer et al. 1991).

Alcohol and AIDS

The seriousness and magnitude of the acquired immunodeficiency syndrome (AIDS) epidemic has prompted investigators to examine the role of alcohol consumption and alcohol abuse on the propagation of this disease. It has become clear that patients infected with human immunodeficiency virus (HIV) are at increased risk of bacterial pneumonias and have a higher incidence of these infections compared with the general population (Schlamm and Yancovitz 1989). Alcohol consumption by these immuno

compromised patients may abrogate pulmonary defense mechanisms (bacterial clearance by ciliary activity and alveolar macrophages). Macrophages are large white blood cells that can engulf and "digest" foreign material and that play a crucial role in initiating the immune response. Alveolar macrophages in the lungs are the primary defense against bacterial infections. Several investigators have shown that alcohol consumption reduces alveolar macrophage capacity (Adams and Jordan 1984). The function of macrophages is mediated by tumor necrosis factor-a. Nelson et al. (1990) recently reported that acute alcohol administration in rats attenuated the TNF-α response to two bacterial strains, Staphylococcus aureus and Klebsiella pneumonia.

Alcohol and susceptibility to HIV infection

Although alcohol abusers are more susceptible to infections, and despite reports that 13 percent of alcoholics in New York City were HIV positive (Jacobson et al. 1988), there is no conclusive evidence that HIV infection is more likely to occur in people (exposed to the virus through the normal mechanisms of risk) whose immune systems have already been weakened. Some studies, however, reported that this may be the case. For example, in vitro studies (Saravolatz et al. 1990) have suggested that alcohol increases the susceptibility of lymphocytes to HIV infection. In a similar study performed on lymphocytes from healthy volunteers, it was reported that alcohol decreased resistance to HIV infection and that T-cell functions were compromised by alcohol (Bagasra et al. 1989). In addition, NK cells and lymphokine-activated killer (LAK) cells, which are important defenses against viral infections such as HIV, were reported to be suppressed by alcohol (Nair et al. 1990).

The relationship between alcohol intake and high-risk sexual behavior, which may be a factor in susceptibility to HIV infection, is discussed in Chapter 10, Effects of Alcohol on Behavior and Safety.

Alcohol and Cancer

Although alcohol has not been shown to cause cancer in experimental animals, its role in the causation of some human cancers has been clearly demonstrated. For instance, a strong association has been observed between alcohol

abuse and cancer in the upper airways (Hsairi et al. 1989; Talamini et al. 1990).

The risk for cancers of the mouth has been found to be directly related to the number of drinks per day (Tuyns 1990).

Although alcohol has not been shown to cause cancer in experimental animals, its role in the causation of some human cancers has been clearly demonstrated.

The role of alcohol consumption in cancers of the gastrointestinal tract remains controversial. While emerging evidence suggests little or no elevated risk for stomach, pancreatic, and colon cancer in alcohol consumers as compared with controls (Bouchardy et al. 1990; Farrow and Davis 1990; Ferraroni et al. 1989; Nomura et al. 1990; Olsen et al. 1989), there is some evidence to support a link between alcohol consumption and rectal carcinoma (Stemmerman et al. 1990). It is postulated that alcohol may undermine the protective effect of dietary fiber against rectal carcinoma or that acetaldehyde can, under certain conditions, potentiate the effects of chemically induced rectal carcinogenesis (Seitz et al. 1990).

Alcohol abuse is an important factor in the development of primary liver cancer (hepatocellular carcinoma), which frequently is superimposed on alcoholic cirrhosis. Although some studies suggest that hepatitis B virus is an important risk factor for primary liver cancer (Hadengue et al. 1990), some investigators found that alcohol consumption elevated risk for liver cancer even in the absence of hepatitis B (Ohnishi et al. 1987). However, a recent study found that hepatitis C virus, which is an important risk factor in alcoholic cirrhosis, was a significant predictor for the development of hepatocellular carcinoma (Caporaso et al. 1991).

Breast cancer is the most common form of cancer diagnosed in women. A number of risk factors have been identified for breast cancer, including nulliparity (having borne no children), family history of breast cancer, age at menarche and menopause, and body mass index. Population-based studies have been conducted to determine whether alcohol consumption increases risk for breast cancer. Although several investigators have reported that the risk may increase slightly with an intake of alcohol above one