Alcohol has complex effects on the cardiovascular system. The purpose of this article is to review physio-pathological effects of alcohol on cardiovascular and related systems and to describe its role in hypertension and cardiovascular disease. The relationship between alcohol and hypertension is well known, and a reduction in the alcohol intake is widely recommended in the management of hypertension. Moreover, alcohol has both pressor and depressor actions. The latter actions are clear in Oriental subjects, especially in those who show alcohol flush because of the genetic variation in aldehyde dehydrogenase activity. Repeated alcohol intake in the evening causes an elevation in daytime and a reduction in nighttime blood pressure (BP), with little change in the average 24-h BP in Japanese men. Thus, the hypertensive effect of alcohol seems to be overestimated by the measurement of casual BP during the day. Heavy alcohol intake seems to increase the risk of several cardiovascular diseases, such as hemorrhagic stroke, arrhythmia and heart failure. On the other hand, alcohol may act to prevent atherosclerosis and to decrease the risk of ischemic heart disease, mainly by increasing HDL cholesterol and inhibiting thrombus formation. A J- or U-shaped relationship has been observed between the level of alcohol intake and risk of cardiovascular mortality and total mortality. It is reasonable to reduce the alcohol intake to less than 30 ml per day for men and 15 ml per day for women in the management of hypertension. As a small amount of alcohol seems to be beneficial, abstinence from alcohol is not recommended to prevent cardiovascular disease.
Alcohol has complex effects on the cardiovascular system. The relationship between alcohol and hypertension is well known, and a restriction of alcohol intake is widely recommended as a part of lifestyle modifications in the management of hypertension.1, 2, 3, 4, 5, 6 Alcohol has both pressor and depressor actions, however, and the genetic susceptibility regarding alcohol metabolism influences the cardiovascular effect of alcohol.3, 7 The effect of alcohol on blood pressure (BP) is also modified by several factors, such as the level of consumption, time period after the last drink and overall drinking behavior.
Alcohol consumption is associated with several cardiovascular diseases, such as brain hemorrhage, heart failure and arrhythmia, as well as with other disorders.3, 8, 9, 10, 11 Heavy drinking and alcoholism not only lead to medical problems but are also serious social concerns. However, alcohol also seems to have beneficial effects, including the prevention of ischemic heart disease. It has been shown that cardiovascular and all-cause mortality is lower in light drinkers compared with nondrinkers.3, 8, 9, 10, 11, 12
The purpose of this article is to review physio-pathological effects of alcohol on cardiovascular and related systems and to describe its role in hypertension and cardiovascular disease. I will outline the cardiovascular actions of alcohol, the effects of alcohol on BP and hypertension, including changes in 24-h BP, and the relationship between alcohol and cardiovascular diseases.
Cardiovascular actions of alcohol
Acute effect on BP
The effect of a single intake of alcohol on BP in normal subjects is not consistent among studies. Some studies have shown an increase in BP,13, 14 whereas it decreased15, 16 or remained unchanged17, 18 in others. In hypertensive patients, the BP also became elevated19 or fell20 after a single ingestion of alcohol.
In studies showing the pressor effect of alcohol, a BP increase was observed within 1 h after drinking; however, the increase was not sustained.3 On the other hand, BP usually fell or remained unchanged after alcohol consumption in studies with prolonged observation periods. Stott et al.21 reported that BP levels increased at 1 h after drinking but tended to decrease over the next 7 h in normotensive subjects.
We examined the effect of a single intake of alcohol (1 ml kg−1) on BP using ambulatory BP monitoring (ABPM) in hypertensive Japanese men.7 As shown in Figure 1, the BP decreased and the heart rate increased for several hours after drinking alcohol. This alcohol-induced hypotension was marked in subjects showing facial flush identified by visual inspection after drinking, and was mild in those who did not show such flush. A transient pressor response to alcohol consumption was not observed in our study, and BP values the day after drinking were comparable to those on the control day.
Minami et al.22 also studied the effect of a single intake of alcohol on ambulatory BP in normotensive and hypertensive Japanese men in relation to the genotype of aldehyde dehydrogenase 2 (ALDH2). In their study, BP decreased significantly after alcohol ingestion in the inactive ALDH2 group (heterozygotes and a homozygote for ALDH2*2), whereas the reduction in BP was small but significant only for diastolic BP in the active ALDH2 group (homozygotes for ALDH2*1).
A single intake of alcohol therefore mainly acts to lower the BP, at least in Japanese men. As this depressor effect varies according to race and the presence and absence of alcohol flush, genetic variation in ALDH2 activity seems to have a major function. It has been shown that subjects with the ALDH2*2 genotype, which is common in Mongoloids but rare in Caucasians and Africans, show facial flush after drinking alcohol because of the accumulation of vasodilative acetaldehyde.3, 22, 23, 24 The mechanisms causing a transient pressor response to alcohol are not clear. An emotional change or gastric irritation rather than a direct effect of alcohol, however, may be involved in the pressor response because it takes several hours for blood alcohol to disappear after a single intake.3, 21
The action of alcohol on the vasculature is variable according to its concentration and the kind of blood vessel.25, 26 High concentrations of alcohol constrict most blood vessels. This vasoconstriction depends on calcium ions and is inhibited by calcium channel blockers. Alcohol also acts to augment the vasoconstriction caused by catecholamines and vasopressin and inhibits endothelium-dependent vasodilation.27, 28 It has been suggested that endothelin and nitric oxide are involved in alcohol-induced vasoconstriction.29 Soardo et al.30 observed that alcohol increased the levels of endothelin-1, nitric oxide, plasminogen activator inhibitor-1 and oxidative stress both in vivo and in vitro. As the scavengers of oxidants prevented those changes, oxidative stress may have a role in the alcohol-induced endothelial dysfunction.30 It was, however, reported that the flow-mediated dilation of the brachial artery and blood markers of endothelial function were similar between the usual drinking period and the alcohol restriction period in healthy men.31
On the other hand, low concentrations of alcohol usually dilate blood vessels.25, 26 This effect also seems to be mediated by calcium ions and endothelium-derived nitric oxide. It has been shown that low doses of alcohol increase the release of nitric oxide and augment endothelium-dependent vasodilation.32 Criscione et al.27 reported that ethanol inhibits norepinephrine-induced vasoconstriction in the rat mesenteric artery. They also observed that norepinephrine-induced vasoconstriction is enhanced after the withdrawal of alcohol. These results seem to be consistent with the time-dependent BP changes after alcohol consumption in humans.
Acetaldehyde, a metabolite of alcohol, acts as a vasodilator.17 Subjects with low-active aldehyde dehydrogenase (ALDH2*2) show facial flush after alcohol ingestion because of the accumulation of acetaldehyde in the blood. Such subjects, especially those homozygous for the ALDH2*2 genotype, show marked tachycardia and hypotension after alcohol consumption.3, 23
In our study, the alcohol-induced BP reduction in hypertensive patients was due to a decrease in peripheral vascular resistance (Table 1).7 We also observed that the intracellular sodium concentration in red blood cells decreases after alcohol ingestion.33 This change may also act to dilate blood vessels through a decrease in the intracellular calcium concentration.
Taken together, alcohol has both constrictive and dilative actions on blood vessels, and these effects may be dependent on race, the dose and timing of alcohol consumption.
The effects of alcohol on the heart are also complex.3, 8, 34 It has been shown that alcohol directly inhibits the contractility of cardiac muscle in a dose-dependent manner.18, 26 This negative inotropic action is apparent in the isolated heart or after blocking of the autonomic nervous system.18
Cardiac function, however, does not often change or even increase after the administration of alcohol in normal humans and animals. Kupari et al.23 reported that both the heart rate and cardiac output increased whereas systemic vascular resistance decreased after alcohol ingestion in healthy volunteers.15 They also observed that those changes were small in subjects who did not show facial flush, but were marked in subjects who showed flush after drinking. In our study, the heart rate and cardiac output also increased significantly after alcohol ingestion in hypertensive patients7 (Table 1). The activation of the sympathetic nervous system seems to mask the direct inhibitory action of alcohol on the heart. Indeed, we have observed that the alcohol-induced increases in heart rate and cardiac output are attenuated after the administration of a beta blocker.35
The adverse influence of alcohol on the heart is clear after the consumption of large amounts for many years. It has been shown that the total alcohol intake is positively related to the left ventricular mass and negatively related to the left ventricular ejection fraction.36 Structural changes in cardiac muscle were also observed in heavy drinkers 37 as well as in ethanol-fed rats.38 These changes may be involved in alcohol-induced cardiomyopathy, heart failure and arrhythmia.
Alcohol withdrawal syndrome
Chronic heavy drinkers, such as alcoholic patients, show alcohol withdrawal syndrome, which is characterized by psycho-neurological symptoms and signs after the sudden cessation of alcohol consumption. This syndrome includes elevation of the BP and heart rate because of activation of the sympathetic nervous system.26 The pressor response during alcohol withdrawal reaches a peak the day after cessation.39 In this case, the BP decreased to a lower level compared with baseline within several days after alcohol withdrawal in heavy drinkers.40 As habitual drinkers experience a mild degree of repeated alcohol withdrawal in daily life, it is possible that this withdrawal phenomenon contributes to alcohol-related hypertension.
Neurohormonal actions of alcohol
Actions on the autonomic nervous system
It has been shown that alcohol activates the sympathetic nervous system.3, 26 Van de Borne et al.41 observed an increase in muscle sympathetic nerve activity after a single intake of alcohol in normal men. In their study, BP did not change, although the heart rate increased significantly. In our study, plasma catecholamines increased but BP decreased after alcohol ingestion in hypertensive patients, and the increase in plasma catecholamines was more pronounced in subjects with a large BP reduction.7, 42 These results suggest that the activation of the sympathetic nervous system occurs in response to BP change and acts to compensate for any further BP reduction.
Experimental studies have shown that alcohol suppresses the baroreceptor reflex.43 Narkiewicz et al.44 reported that alcohol enhances the hypotension induced by lower body negative pressure. The combination of impairment of the baroreceptor reflex and systemic vasodilation acts to potentiate orthostatic hypotension and may induce syncope after drinking in susceptible subjects.
Actions on the endocrine system
It is known that alcohol increases plasma renin activity.7, 26, 42 As the increase in renin activity was suppressed by pretreatment with propranolol in our study,35 it seems to be mediated by the sympathetic nervous system. Alcohol also stimulated the release of adrenocorticotropic hormone, and increases in plasma cortisol and aldosterone were observed after drinking.26 It has been reported that dexamethasone inhibited the BP elevation and sympathetic activation after alcohol ingestion.45 We and others, however, have failed to observe significant changes in adrenocorticotropic hormone, cortisol or aldosterone.21, 42
Alcohol suppresses the release of vasopressin; however, this change does not seem to mediate the acute depressor effect of alcohol.21, 26, 42 Several depressor hormones and substances, such as atrial natriuretic peptide, prostaglandin E2, beta endorphin and cyclic GMP, did not change after alcohol ingestion.42 Although data relating alcohol intake to plasma atrial natriuretic peptide have been inconsistent, Djousse et al.46 observed a positive relationship after adjusting for several confounding factors.
The level of plasma insulin increases after alcohol intake; however, the change is less than that induced by an isocaloric control drink.42 It has been shown that a light-to-moderate intake of alcohol enhances insulin sensitivity47 and reduces the risk of type 2 diabetes mellitus.48 Alcohol therefore seems to have a beneficial effect on insulin and glucose metabolism.
Taken together, alcohol causes various changes in the autonomic nervous system and endocrine system, but these changes do not seem to have a major role in the pressor or depressor effect of alcohol except in the case of alcohol withdrawal syndrome.
Actions of alcohol on water and electrolyte metabolism
Alcohol also exhibits actions on water and electrolyte metabolism. Urinary excretion increases after alcohol ingestion.49, 50 The increase in urine volume seems to be caused by fluid intake and the suppression of vasopressin.21, 27
We have studied the effect of repeated alcohol intake for 7 days on the urine volume and sodium excretion in hypertensive patients.51 Urine volume increased on days 3–5 but not on day 1. Urinary sodium excretion decreased in the early phase but increased in the late phase. The average BP also decreased in the early phase and then returned toward the baseline levels. The initial BP reduction may mask the alcohol-induced diuresis and causes sodium retention, which may be involved in subsequent BP elevation.
It has been shown that urinary potassium excretion decreases after alcohol ingestion.48, 49 In our study, the serum potassium level decreased after a single intake of alcohol.7, 35 This change in serum potassium seems to be mediated by the sympathetic nervous system, as propranolol attenuated alcohol-induced hypokalemia.35 Conversely, alcohol increases the urinary excretion of magnesium and calcium.49, 50, 52 It is possible that magnesium and calcium are depleted in habitual drinkers, and the alcohol-induced changes in these minerals may contribute to BP elevation and arrhythmia.3, 53
Alcohol and hypertension
Numerous epidemiological observational studies have examined the relationship between alcohol consumption and BP or hypertension. Almost all of them have shown that habitual drinkers have a higher BP and higher prevalence of hypertension than nondrinkers.1, 2, 3, 54, 55, 56, 57, 58 These associations have been observed regardless of race, gender, age and the type of alcohol (Figure 2). Although some studies suggest the presence of a threshold regarding the pressor effect of alcohol,59, 60 the relationship between the level of alcohol consumption and BP is usually linear. In cross-sectional studies, the systolic BP increased by 3–4 mm Hg and diastolic BP increased by 1–2 mm Hg per three drinks per day (one drink contained 10–15 ml, or 8–12 g of alcohol).1 Intake of 10 ml per day of alcohol therefore seems to elevate the systolic BP by about 1 mm Hg in humans. It has been estimated that about 10% of hypertension in the general population can be attributed to alcohol.1, 2, 3
The relationship between alcohol and BP seems to be independent of confounding variables. Increases in the body weight and abdominal fat associated with alcohol consumption, however, may have a role in alcohol-related hypertension. Suter et al.61 observed that both the BP and waist/hip ratio increased with the level of alcohol intake, and there was a positive association between changes in body weight and alcohol consumption.
The hypertensive effect of alcohol has also been shown in longitudinal studies.62, 63, 64 Tsuruta et al.62 reported that the probability of the development of hypertension in heavy drinkers (alcohol consumption ≧46 g per day) was about twice that of the rest of the population after a 12-year follow-up among normotensive men. Fuchs et al.63 showed that the consumption of alcohol at ≧30 g per day was an independent risk factor among participants in the Atherosclerosis Risk in Communities (ARIC) study. Nakanishi et al.64 also observed that the risk for hypertension increased in a dose-dependent manner with increases in alcohol intake among Japanese men in a longitudinal study.
Although epidemiological studies have clearly shown the hypertensive effect of alcohol, most studies did not consider the time-related effect of alcohol on BP. This fact may be important because BP measurement has been carried out during the daytime, whereas alcohol is usually consumed at night. Moreira et al.65 reported that the BP in habitual drinkers was higher at 13–24 h after the last drink compared with that within 3 h or at more than 24 h. Kawabe et al.66 observed that the evening home BP was lower but the morning home BP was higher on drinking compared with nondrinking days in Japanese volunteers. These findings suggest that the BP in habitual drinkers is overestimated by casual BP measurement taken during the day.
Clinical intervention studies have also revealed an increase in BP with alcohol consumption and a BP decrease with alcohol restriction. Howes and Reid,67 observed a BP increase after repeated alcohol consumption for 7 days in normotensive subjects. Potter and Beevers,68 reported a BP increase after alcohol intake for 4 days in hypertensive patients. Using a crossover design, Puddey et al.69, 70 compared BP values during a 6-week period of unrestricted alcohol consumption and that of alcohol restriction in normotensive and hypertensive subjects. The average level of alcohol consumption was 50 ml per day during the unrestricted period and 10 ml per day during the restricted period, and the BP was 3/2 mm Hg higher in the former than in the latter period in normotensive subjects.69 In hypertensive subjects, the BP was 5/3 mm Hg higher in the unrestricted than during the restricted period.70 Ueshima et al.71 also examined the effect of a 2-week period of alcohol intake and restriction using a crossover method in Japanese hypertensive patients and observed a similar BP elevation with alcohol consumption.
According to a meta-analysis of 15 randomized controlled trials, BP significantly decreased with alcohol restriction72 (Figure 3). The mean BP reduction was 3.3/2.0 mm Hg, and there was a dose–response relationship between alcohol reduction and decrease in BP. The effects of intervention were enhanced in studies with a higher baseline BP. The results of this meta-analysis support the importance of alcohol reduction in the management of hypertension among heavy drinkers.
Most clinical studies have not considered the timing of alcohol intake and BP measurement. We studied the effect of repeated episodes of alcohol consumption on BP with ABPM under standardized conditions in Japanese men with hypertension.73 After several days of the control period, the subjects consumed 1 ml kg−1 of alcohol with an evening meal for 7 days. Evening BP values decreased for several hours after alcohol consumption on both days 1 and 7, whereas morning BP was unchanged on day 1 but increased on day 7. The average 24-h BP was lower on day 1 and was the same on day 7 compared with the control period. A short-term repeated intake of alcohol therefore causes biphasic changes in BP without altering the average 24-h BP, at least in Japanese men.
We also examined the effect of a 4-week period of unrestricted alcohol consumption and that of alcohol restriction on the 24-h BP in hypertensive patients in a randomized crossover study.74 The average level of daily alcohol intake was 66 ml in the unrestricted period and 11 ml in the restricted period. The daytime BP was 3/2 mm Hg higher in the unrestricted period than in the restricted period, but the nighttime BP was 4/2 mm Hg lower in the former (Figure 4). The average 24-h BP was comparable between the two periods. These effects of alcohol resulted in changes in the dipping pattern of the 24-h BP. Half of those who did not show a dip in BP during the restricted period changed and showed a dip in the unrestricted period, and half of those that showed a dip showed an extreme dip during this period.
On the other hand, Minami et al.75 observed reductions in the daytime (−3.4 mm Hg) and 24-h (−3.2 mm Hg) systolic BPs after 3 weeks of alcohol restriction in Japanese men. In their study, daytime, nighttime and 24-h diastolic BPs did not change with alcohol reduction (−1.1, +2.1 and −0.3 mm Hg, respectively).
The effect of alcohol on the 24-h BP may differ between Orientals and Caucasians. Howes et al.76 observed that short-term alcohol intake increased BP variability without changing the average BP in Australian subjects. However, Rakic et al.77 showed that the average 24-h BP increased significantly after 4 weeks of alcohol consumption in Australian men. In their study, the average 24-h systolic BP was 2–3 mm Hg higher and the nighttime BP was not lower in the unrestricted compared with the restricted period. It was also reported that the average 24-h BP decreased after abstinence in alcoholic patients.78
In a systematic review, McFadden et al.79 analyzed clinical trials that examined the BP after a period of sustained alcohol intake. In this review, the pressor effect of alcohol was evident in non-ABPM studies, but not in ABPM studies. An early effect of reducing the BP and a later effect of raising the BP led to smaller differences in the net effect of alcohol on BP values in ABPM studies.
We also studied changes in morning and evening home BP measurements during each of the 4 weeks of unrestricted consumption and restriction in hypertensive patients.80 In this study, the morning BP increased by 4.4 mm Hg but late evening BP decreased by 7.4 mm Hg at the end of the unrestricted alcohol intake period. The pressor effect was significant from week 2, whereas the depressor effect was evident from day 1. These results indicate that the status of alcohol intake influences the morning–evening BP difference, and that slow pressor mechanism(s) are involved in alcohol-induced BP elevation.
It is therefore clear that alcohol consumption contributes to hypertension, and alcohol restriction decreases the daytime BP. It should be noted, however, that the effects of alcohol on BP vary according to the level and duration of consumption and the time from the last drink. Alcohol seems to exert a marked influence on circadian BP variation, whereas its influence on the average 24-h BP seems to be small. The mechanisms of the pressor effects of alcohol have not been fully clarified; however, changes in vascular reactivity and sympathetic nerve activity related to intermittent alcohol withdrawal seem to be more important than the direct actions of alcohol. Deficiencies in magnesium and calcium may contribute to BP elevation after chronic alcohol consumption. Increases in the caloric intake through consuming alcoholic beverages and elevated salt intake associated with drinking may also be involved in alcohol-related hypertension.
Many experimental studies have examined the effect of chronic administration of alcohol on BP; however, the results have been inconsistent.
Strickland and Wooles,81 reported an elevation of BP after ethanol administration (5–20% in drinking water) for 4 weeks in rats. In their study, the plasma level of norepinephrine was decreased in ethanol-fed animals. Vasdev et al.82 observed BP elevation after 1 week during the administration of ethanol (5–10%) to Wister Kyoto rats. They noted increases in the concentration of platelet intracellular calcium ions and the uptake of calcium in the aorta. Hsieh et al.53 identified increases in BP and intracellular calcium ions and a decrease in intracellular magnesium ions after 4 weeks of ethanol administration (15%) to rats. They suggested a role of magnesium deficiency in ethanol-induced hypertension as the BP elevation was attenuated by magnesium supplementation. Puddey et al.83 reported a BP elevation of ∼10 mm Hg and decreases in the level of phospholipids and the ratio of unsaturated/saturated fatty acids in the aorta and kidney after the chronic administration of alcohol to rats. Harada et al.84 also observed increases in the BP and platelet-free calcium concentration with ethanol consumption (15%) in Wister Kyoto rats.
In some studies, the BP did not change after the chronic administration of alcohol to animals. Abdel-Rahman85 reported that the BP increase was not different between ethanol-fed (5–20% in drinking water) spontaneously hypertensive (SH) rats and control SH rats during a 13-week observation period. The depressor effect of clonidine, however, was reduced in the ethanol-fed SH rats, suggesting a change in the neural regulation of BP.
Several studies have shown that chronic alcohol intake decreases BP in animals. Howe et al.86 reported that BP values in alcohol-fed (5–20% in drinking water) Wister Kyoto, SH and stroke-prone SH rats were lower than those of respective control rats during a 6-month observation period. Hatton et al.87 observed a BP decrease during chronic ethanol administration (36% in a liquid diet) for 18 weeks in Wistar rats. The vasoconstrictor response of resistant arteries to norepinephrine was enhanced and the vasodilator response to alcohol was attenuated in their study. Beilin et al.88 reported a decrease in the resting BP of Wister Kyoto and SH rats after 12 weeks of ethanol administration (20% in drinking water), although cardiovascular reactivity to noise-related stress was augmented in the ethanol-fed SH rats. El-Mas and Abdel-Rahman89 also observed a lower BP in freely moving, ethanol-fed (5% in a liquid diet) rats compared with control rats based on telemonitoring of the BP.
The reasons for the inconsistent results in experimental studies are not clear, but cannot be explained by differences in daily doses of alcohol administration. The periods of alcohol administration, however, are generally longer in studies showing BP reduction than those showing BP elevation. The timing of BP measurement may be important, such as in clinical studies. Crandall et al.90 administered 30% alcohol twice daily (7–8 g kg−1) for 10 weeks to rats and examined the levels of BP and blood alcohol. In their study, BP was normal at the time of the peak blood alcohol level but was elevated at 24 h after alcohol consumption, when alcohol was not detected in the plasma. Their results suggest that alcohol-induced hypertension is not because of its direct action but to alcohol withdrawal.
The harmful effects of large doses of alcohol, such as cardiac dysfunction, have been shown in experimental studies.91 Schlicht et al.92 however, reported that the lifespan of SH rats was prolonged by the chronic administration of ethanol. These observations are important, as both the adverse effects of a large amount of alcohol and the beneficial effects of a moderate amount on cardiovascular disease and total mortality have been shown in large-scale epidemiological studies.8, 9, 10, 11, 12
Interaction with antihypertensive drugs
Alcohol interacts with several antihypertensive agents. Experimental studies have shown that alcohol attenuates the effect of centrally acting antihypertensive drugs such as clonidine.86 Heavy drinking is recognized as one of the factors responsible for resistant hypertension. The interaction between alcohol and antihypertensive drugs and the hypertensive effect of alcohol may have a role in alcohol-related resistant hypertension. In addition, heavy drinkers often show poor adherence to both pharmacological treatment and lifestyle modifications. Habitual drinkers taking antihypertensive drugs are also prone to morning hypertension.93
The combination of alcohol and antihypertensive drugs may also lead to a marked BP reduction. This phenomenon has been known; however, few clinical studies have addressed this interaction. It is possible that sympatholytic drugs augment the depressor effect of alcohol, as alcohol-induced hypotension is associated with the reflex activation of the sympathetic nervous system.7 In our studies, alcohol and a beta blocker, propranolol, additively lowered the nighttime BP,35 whereas alcohol and an alpha blocker, prazosin, synergistically acted to lower the BP in hypertensive patients.94 It has also been reported that alcohol enhances the depressor effect of the calcium antagonist felodipine.95 Changes in the type and timing of antihypertensive medication along with the moderation of alcohol consumption should be considered to treat hypertensive patients with a drinking habit.
Hypertension treatment guidelines
As an excess consumption of alcohol is a risk factor for hypertension, all hypertension treatment guidelines recommend the moderation of alcohol intake as a part of lifestyle modifications for the management of hypertension (Table 2). The 7th report of the Joint National Committee in the United States (JNC-7) recommends the limitation of daily alcohol consumption to no more than two drinks (30 ml) for most men and to no more than one drink (15 ml) for women and light-weight men.4 According to the European guidelines (European Society of Hypertension–European Society of Cardiology guidelines, ESH–ESC 2007), the upper limit is 20–30 g per day for men and 10–20 g per day for women. The Japanese guidelines provide similar recommendations (20–30 ml per day for men and 10–20 ml per day for women). Of note, 600 ml of beer or 250 ml of wine contains about 30 ml of ethanol.
These recommendations put forward by the guidelines are appropriate because small doses of alcohol exert little adverse effects on BP and the cardiovascular system. There are, however, some concerns regarding the efficacy of alcohol restriction on BP because the effect of alcohol on average 24-h BP levels seems to be very small. In our studies, salt restriction and weight reduction substantially decreased the BP for 24 h, but the effect of alcohol restriction on average 24-h BP was not significant.96, 97, 98 As light drinking has beneficial effects on the cardiovascular system, as described later, abstinence from alcohol should not be imposed on hypertensive individuals except for patients with special conditions.
Alcohol and cardiovascular disease
A heavy alcohol intake is associated with cardiac hypertrophy and the risk of cardiomyopathy and heart failure.8, 36, 37 It has been shown that the total consumption of alcohol is positively related to a left ventricular mass and is negatively associated with the ejection fraction in asymptomatic alcoholic subjects.37 Recent epidemiological studies, however, have shown that moderate alcohol consumption is associated with a lower risk of heart failure.99, 100 Klatsky et al.101 reported that heavy drinkers had an increased risk of heart failure because of noncoronary artery disease, whereas alcohol drinking was inversely related to the risk of heart failure because of coronary artery disease. Heavy drinking therefore seems to increase the risk of heart failure but light-to-moderate drinking may decrease the risk, probably because of its favorable association with coronary artery disease.
Alcohol intake is associated with the risk of tachyarrhythmia, such as ventricular and supraventricular premature contractions and atrial fibrillation.8, 102, 103, 104 In the Danish Diet, Cancer and Health study, moderate-to-heavy consumption of alcohol was associated with an increased risk of atrial fibrillation.103, 104 Such alcohol-induced arrhythmia often occurs after binge drinking. Activation of the sympathetic nervous system and a decrease in the serum potassium level after drinking may trigger this arrhythmia.3 Cardiac functional and structural changes because of chronic alcohol consumption also seem to have a role in arrhythmia.37
Coronary heart disease
Alcohol seems to have a beneficial effect on coronary heart disease.8, 9 It has been shown that the risk of myocardial infarction is 20–50% lower in habitual compared with nondrinkers.9, 105, 106, 107, 108 This risk reduction is dose-dependent up to the level of moderate drinking, but further risk reduction has not been observed in heavy drinkers (Figure 5). A U-shaped relationship has been observed between the level of alcohol consumption and degree of coronary calcification in a general population.109 In some studies, such as the Japan Collaborative Cohort Study,110 the beneficial effect of alcohol on coronary heart disease was modest and not significant (Table 3).
The mechanisms behind the inverse association of alcohol with coronary heart disease have not been fully clarified. The alcohol-induced increase in HDL cholesterol, however, seems to be the most important mechanism.107, 111 The inhibitory effect of alcohol on blood coagulation also contributes to the lower risk of myocardial infarction.107 In addition, it has been shown that moderate alcohol consumption is associated with a lower plasma level of C-reactive protein, suggesting an anti-inflammatory action of alcohol.112 The weak effect of alcohol on the average 24-h BP may also have a role as a high BP is a strong risk factor for coronary heart disease.74, 79
Red wine contains polyphenols that act to prevent atherosclerosis because of their antioxidant effect. Several studies have shown that people who mainly drink red wine have a lower risk of cardiovascular disease than those who drink other kinds of alcoholic beverage.113, 114 It is suggested, however, that the low incidence of myocardial infarction in habitual drinkers is largely attributed to the effect of alcohol itself.115
The relationship between alcohol consumption and total cerebrovascular disease is generally J-shaped, although it differs according to subtypes of stroke 105, 116 (Figure 5). It is clear that alcohol is a risk factor for hemorrhagic stroke. A positive linear relationship has been observed between the level of alcohol consumption and risk of brain or subarachnoid hemorrhage.8, 110, 116, 117 Actions on the BP and blood coagulation system seem to be underlying mechanisms for this adverse influence of alcohol.
On the other hand, the relationship between alcohol consumption and the risk of ischemic stroke has been found to be J- or U-shaped.110, 116, 117, 118 The low risk in light drinkers seems to be due to the lower degree of atherosclerosis and the inhibition of blood coagulation, as in the case of ischemic heart disease. The increased risk in heavy drinkers is probably related to increases in the level and variability of the BP, hemoconcentration because of dehydration and thromboembolism associated with alcohol-induced atrial fibrillation. Regarding alcoholic beverage types, wine drinkers seem to have a lower risk of ischemic stroke.119, 120
The favorable association of light-to-moderate drinking with the risk of ischemic stroke seems to be more apparent in Caucasians than in Japanese, although the results of epidemiological studies have been inconsistent in both populations. The racial differences may be related to variation in the frequencies of stroke subtypes. Atherothrombotic brain infarction is common in Caucasians, whereas lacunar stroke is more common in Japanese.
Several studies have examined the relationship between alcohol intake and subclinical findings on magnetic resonance imaging of the brain in general populations. In the Cardiovascular Study, a U-shaped relationship was observed between alcohol consumption and white matter abnormalities. Moreover, moderate drinking was also associated with a lower risk of lacunar infarction compared with abstainers.121 Such risk reduction with moderate drinking, however, was not observed in the ARIC study, and an increased level of alcohol intake was associated with brain atrophy.122
Peripheral arterial disease and atherosclerosis
As light-to-moderate consumption of alcohol seems to act to suppress the progression of atherosclerosis, it may also have a favorable influence on peripheral arterial disease. The Edinburgh Artery Study supported the protective effect of alcohol, as there was a positive association between the level of alcohol intake and the ankle brachial index.123 In the Physicians’ Health Study, habitual drinkers showed a 26% lower incidence of peripheral arterial disease compared with nondrinkers after adjustment for confounding factors.124 Similar results were also shown in the Framingham Heart Study and the ARIC study.125, 126
Regarding the association of alcohol and carotid atherosclerosis, an inverse relationship was noted in the Lausanne Stroke Registry.127 On the other hand, there was no significant association between alcohol intake and the carotid artery thickness in the ARIC study.128 A U- or J-shaped relationship was observed between the level of alcohol intake and severity of carotid atherosclerosis in the Bruneck Study and the Study of Health in Pomerania.129, 130 Although the results of epidemiological studies have been inconsistent, light-to-moderate consumption seems to inhibit the development of carotid atherosclerosis.
Several studies examined the relationship between alcohol consumption and arterial stiffness by measuring the pulse wave velocity. Sierksma et al.131 identified a U-shaped relationship between alcohol consumption and the aortic pulse wave velocity. van den Elzen et al.132 observed an inverse relationship between the alcohol intake and pulse wave velocity in young men and women. On the other hand, Kurihara et al.133 reported that the brachial–ankle pulse wave velocity was elevated in heavy drinkers. These studies also support the favorable effect of moderate and the harmful effect of heavy drinking on large arteries.
Cardiovascular mortality and total mortality
As described earlier, alcohol seems to exert both beneficial and adverse effects on cardiovascular diseases. The relationship between alcohol consumption and total cardiovascular mortality has been shown to be J-, U- or L-shaped. A J-shaped relationship was observed in the Japan Collaborative Cohort Study; however, the beneficial effect of light-to-moderate drinking was not significant.110 On the other hand, in very large longitudinal studies conducted by the American Cancer Society, a U-shaped relationship was observed in the original study,105 and the relationship was L-shape (nondrinkers showed the highest cardiovascular mortality) in Study II.10 In a meta-analysis conducted by Di Castelnuovo et al.,114 a light-to-moderate consumption of wine or beer was associated with lower cardiovascular risk. A drinking habit, particularly wine consumption, has been shown as a part of a lifestyle associated with low cardiovascular risk.134, 135 It has also been suggested that the risk reduction associated with alcohol consumption is low in individuals without cardiovascular risk factors but is high in those with a marked cardiovascular risk. Taken together, light-to-moderate alcohol consumption seems to decrease cardiovascular mortality, whereas heavy drinking may result in poor cardiovascular outcomes compared with abstainers.
Alcohol is also related to several cancers, liver disease, psychiatric and neurological disorders and injury, and it seems to influence total mortality. A J- or U-shaped relationship has been observed between the level of alcohol intake and total mortality.10, 11, 105, 136, 137 It has been suggested that all-cause mortality is the lowest among subjects who consume about one drink per day. In the American Cancer Society Prospective Study II, total mortality was lower in drinkers than in nondrinkers.10 It has also been shown that wine drinkers have a lower mortality rate than drinkers who avoid wine.136 In a meta-analysis of 34 studies including more than one million individuals, a J-shaped relationship was found between alcohol consumption and total mortality.11 In this analysis, low levels of alcohol intake (one to two drinks per day for women and two to four drinks per day for men) were inversely associated with total mortality, although higher levels of alcohol increased mortality. Those findings suggest that a light-to-moderate intake of alcohol decreases but heavy consumption increases total mortality compared to nondrinking.
Alcohol has complex effects on the cardiovascular system. It is clear that alcohol consumption is related to hypertension, and therefore the restriction of alcohol intake is recommended in the management of hypertension. Alcohol and its metabolites, however, also exhibit a vasodilatory action, and the BP usually decreased after alcohol ingestion, especially in Orientals who show alcohol flush. Mechanisms for the pressor action of alcohol have not been completely clarified; however, an increase in the vascular sensitivity, activation of the sympathetic nervous system and depletion of magnesium and calcium may be involved. The depressor action of alcohol is due to a decrease in systemic vascular resistance that may be related to the attenuation of vascular sensitivity and production of nitric oxide. The pressor effect of alcohol consumed in the evening is apparent during the day, but its effect on average 24-h BP seems to be very small. It should be mentioned that casual BP measurement may lead to overestimating the hypertensive effect of alcohol.
Alcohol seems to exert both harmful and beneficial effects on cardiovascular disease. An excessive intake of alcohol is associated with increased risks of heart failure, arrhythmia and hemorrhagic stroke and causes an increase in total mortality. Light-to-moderate drinkers, however, show lower rates of atherosclerosis and lower risks of coronary heart disease, heart failure, ischemic stroke, peripheral artery disease and cardiovascular and total mortality compared with nondrinkers. As the aim of the management of hypertension is the prevention of cardiovascular disease and premature death, moderation of alcohol intake is to be recommended to hypertensive patients, but abstinence from alcohol should not be insisted on unless there are specific indications for it.
About this article
Current Cardiovascular Risk Reports (2011)