Abstract
Alcohol is the most consumed drug worldwide. Both acute and chronic alcohol use have been associated with cardiac arrhythmias, in particular atrial fibrillation, or so-called 'holiday heart syndrome'. Epidemiological, clinical and experimental studies have attempted to elucidate the mechanisms involved in this association. However, because most of these studies have shown conflicting results, the connection between ethanol and atrial arrhythmias remains controversial. Historical, epidemiological and pharmacological aspects of alcohol, as well as recent concepts on atrial fibrillation are reviewed. We then examine the literature and provide a critical point of view on the still elusive association between alcohol and atrial fibrillation.
Historic Background
The use of alcoholic beverages dates back to prehistoric times. Their medical use was recorded as early as the cuneiform writings in the Mesopotamia of 2200 BC. Out of the approximate 800 medications in ancient Egypt, some 15% included different kinds of beer or wine in their formulation. References to wine are also commonly found in the Old Testament. The Book of Genesis tells that Noah planted a vine after the Great Flood and got drunk. Interestingly, wine and beer could be mixed with other drugs. Thus, highly powerful beverages were produced at a time when distillation was still unknown.
The word alcohol comes from the Arabic word alkuhl, which means essence. Distillation techniques were developed by the Arabs during the Middle Ages, around 800 AD. Alchemists were fascinated by the invisible 'spirit' distilled from wine. They believed alcohol was the elixir of life and used it therapeutically. Worldwide, alcoholic beverages are consumed in multiple circumstances. Alcohol has been consumed for a long time in many different cultures, and consumption may vary within the same culture. As a consequence of such cultural diversity, attitudes towards alcohol have varied from tolerance to censure.
Epidemiological Aspects of Alcohol
Alcohol is currently the number one drug in the world in consumption levels. It has become a serious public health problem. Alcohol addiction affects from 10% to 12% of the population worldwide [WHO, 1999]. In the US, the National Household Survey on Drug Abuse published in 1996 showed that in the month previous to the research, 23.8% of men and 8.5% of women referred an episode of massive alcohol intake, which means five or more drinks at the same occasion at least once in that month. Heavy intake of ethanol is generally defined as the consumption of seven or more drinks per week by women, or over three doses at one time, and over fourteen drinks per week for men, or over four drinks at one time [US Department of Health and Human Service, 1995]. The financial burden from drug abuse is astonishing. Total annual economic cost resulting from the abuse of tobacco, alcohol and other drugs exceeds 238 billion dollars in the US. From that amount, alcohol alone accounts for 98.6 billion every year [Institute for Health Policy, 1993]. In the same period, at least 14 million Americans were reported as alcohol abusers or alcoholics. Those figures are most likely higher today and they reflect a world trend. It should be pointed out that clinical diagnosis and treatment are usually postponed to the point when the disease reaches an advanced stage and is already associated with a number of social and clinical complications that make treatment much more difficult.
Ethanol Pharmacology
Ethanol or ethyl alcohol (CH3CH2OH) is quickly absorbed in the stomach (20%) and in the small intestines (80%). After intake, maximum plasma concentration is reached between 30 and 90 min. As a result of ethanol first-pass metabolism through gastric and hepatic alcohol dehydrogenase (ADH), oral intake of ethanol results in lower blood levels when compared with IV administration. Following zero-order kinetics, which means being constant along time and independent of plasma concentrations, alcohol distribution over all tissues and body liquids is fast and uniform, and crosses the haematoencephalic and placental barriers. Redistribution velocity to specific tissues – as occurs with volatile anaesthetics – depends mostly on blood flow. Ethanol is highly liposoluble: 90% is metabolized in the liver, and from 5% to 10% is excreted without any change via exhaled air and urine. Ethanol oxidation occurs first into acetaldehyde and then into acetic acid. The intermediate metabolite acetaldehyde (CH3COH) is a reactive, toxic component and may be a contributor to hepatotoxicity.
Typically, alcohol content ranges from 4% to 6% in beer, 10–15% in wine and 40% or more in distilled beverages. The alcoholic degree on beverage labels corresponds to alcoholic percentage multiplied by two (e.g. 40% means that alcoholic degree is 80). Curiously enough, and as opposed to what lay individuals think, the size of alcoholic beverages containers is designed to contain approximately 14 g of alcohol per serving (which means to say 0.3 mol of ethanol), whether in a glass of beer (350 ml), in a glass of wine (150 ml) or in a glass of whisky or any other distilled beverage (45 ml). When consumed by a 70 kg individual, any of those beverages will result in a 30 mg% blood alcohol concentration. Alcohol consumption is measured per unit. One unit corresponds to 10 g of alcohol. In order to obtain the units equivalent to one given beverage, the amount taken must be multiplied by alcoholic concentration. That way, the absolute amount of alcohol in that beverage is calculated. Then, the following conversion is applied: one unit for every 10 g of alcohol contained in that beverage.
Clinical Effects of Ethanol
The acute clinical effects of ethanol are correlated to plasma concentration. When alcohol blood concentration reaches approximately 30 mg%, patients exhibit euphoria, excitement and slight changes in attention span; at 50 mg%, patients exhibit discreet motor discoordination and personality and behaviour changes; at 100 mg%, patients exhibit pronounced motor discoordination with ataxia, reduced concentration, worsening of sensitive reflexes and compromised humour; at 200 mg%, ataxia is aggravated and nausea and vomiting may occur; at 300 mg%, dysartria, amnesia, hypothermia and anaesthesia (stage I) may occur; at 400 mg%, coma and death can occur as a result of changes in central respiratory drive and/or systemic arterial hypotension. Blood levels of ethanol may vary depending on a number of factors; for instance, the time taken for alcohol intake, gender, body weight and water content in the body, as well as gastric emptying levels and the metabolism of individuals. On average, three conventional doses of alcohol (approximately 42 g of ethanol) while fasting leads to a maximum serum concentration ranging from 67 to 92 mg/dl in men. After a meal, the same dose will result in serum concentration ranging from 30 to 53 mg/dl. Serum concentrations are higher in women when compared with men when the same dose is considered. That may be explained by their lower average stature, their lower water content per body weight and their reduced gastric dehydrogenase activity when compared with men. Ethanol is metabolized at the speed of one conventional dose every 60–90 min in individuals who have normal hepatic function.
At low and moderate doses (one to three doses a day, ≤15 g/day for women and ≤30 g/day for men) ethanol has a protective effect in coronary cerebrovascular and peripheral vascular artery diseases and in metabolic syndrome, as demonstrated by large epidemiological studies such as the Framingham [Friedman and Kimball, 1986] and Physician's Health Study [Albert et al. 1999; Camargo et al. 1997], and also in a meta-analysis [Di Castelnuovo et al. 2002].
Ethanol has toxic effects on the myocardium, which are dose related and most prominent in long-term users [Patel et al. 1997; Preedy et al. 1994]. Fatty acids esters formed from ethanol enzyme reaction with free fatty acids and acetaldehyde seem to play a key role in the development of diffuse myocardial hypokinesia. This alcoholic cardiomyopathy is more prevalent among men and is characterized by cardiomegaly, diffuse hypokinesia and pathologic changes (dilation, fibrosis) in the myocardium of both ventricles. That applies to patients whose intake is at least 80 g of ethanol daily for a period of at least ten years. Electron microscopy studies have revealed ultrastructural myocardial damage including separation of filaments and loss of striation in addition to injuries in myofibril Z lines, with contractility loss at advanced stages of the disease, and dilation and rupture of mitochondrial crests. Other characteristics include lipid deposits (in particular triglycerides), fibrosis, widening of junctions, and damage to sarcoplasmic reticulum. At the final stages of the disease, inflammatory infiltrates can be observed. Acute alcoholic intoxication may also reduce myocardial contractility through direct effects of ethanol or acetaldehyde on troponin–tropomyosin coupling that is mediated by calcium inhibition [Horton and White, 1996; Guarnieri and Lakatta, 1990], and/or strong reduction of protein synthesis, particularly by acetaldehyde [Vary et al. 2005] and/or release of toxic free radicals to cardiac muscle [Atkinson et al. 1992].
Atrial Fibrillation Mechanisms
Atrial fibrillation is the most common sustained arrhythmia in clinical practice. Incidence increases with age, advancing from 0.5% in the sixth decade in life to approximately 10% of individuals over 80 years old. It is a significant risk factor for cerebral thromboembolism and is associated with an increased mortality [Abusaada et al. 2004; Friberg et al. 2003; Benjamin et al. 1994; Wolf et al. 1987, 1991, 1996]. The understanding of the mechanisms implicated in the genesis of atrial fibrillation has grown substantially in the last decade, which has led to the development of more effective clinical and interventional treatments, particularly catheter ablation techniques. In spite of this, its basic mechanisms have not been fully understood [Nattel and Opie, 2006]. Clinical and experimental evidence has suggested that atrial fibrillation patients may have atrial histological and/or electrophysiological changes, thus triggering the onset of arrhythmia as well as its perpetuation. However, in some situations, it may not be clear whether those changes are the cause of or the consequence from atrial fibrillation. Nevertheless, some cardiac and non-cardiac factors have been consistently associated with this kind of arrhythmia. Among the cardiac factors deserves mention any cause that leads to increases in left atrium dimensions (e.g. mitral valve disease, aortic valve disease, systemic arterial hypertension, pericarditis, myocarditis), left ventricular dysfunction, and the post-operative period following cardiac surgery. Among the non-cardiac factors: thyreotoxicosis, electrolyte disorders, drugs (both legal and illegal drugs), and alcohol abuse, especially acute alcohol abuse, but chronic as well.
Atrial tissue changes are involved in atrial fibrillation genesis, affecting the refractory periods and dispersion, as well as conduction velocity and triggering factors [West and Landa, 1962]. Solid evidence has been available to show that atrial fibrillation is based on multiple, continuous inter-atrial re-entries [Rensma et al. 1988; Moe and Abildskov, 1959], but recently the concept of focal atrial fibrillation has gained renewed interest [Nattel and Opie, 2006]. It has been consistently shown that pulmonary veins foci are the most relevant triggers of atrial fibrillation, but foci originating in the left atrial posterior wall, crista terminalis and caval veins have also been described. The intrinsic mechanisms of these foci remain uncertain, and rotors, anisotropic re-entry and automatic or triggered activity could play a role.
Given the major role of re-entry in the development and maintenance of atrial fibrillation, it is important to briefly review the conditions required to set up a re-entrant circuit. Re-entry requires that (1) the impulse blocks in a unidirectional fashion; and (2) the recirculation time of the impulse to the original site has to be longer than the refractory period of the proximal segment of the circuit. Should recirculation time be shorter, the impulse will reach the original site during refractoriness of the proximal segment of the circuit and re-entry will not be completed. In other words, the anatomic length of the circuit must be identical or longer than the distance travelled by the activation wave during the refractory period. This seminal concept defines the so-called wavelength, which corresponds to the product of the refractory period and conduction velocity [Wiener and Rosenblueth, 1946]. In the same tissue mass, shorter wavelengths – whether resulting from short refractory periods, slow conduction velocity, or both – are more bound to develop re-entry circuits than longer wavelengths. Wavelength is the major determining factor for atrial fibrillation inducibility via re-entry [Rensma et al. 1988].
Finally, in vivo monophasic action potential recordings provide additional key information for the electrophysiologic evaluation of cardiac arrhythmia mechanisms, since it keeps reasonable correlation with transmembrane action potential [Franz, 1991]. Thus, they are useful in determining refractory period changes. However, it is important to remark that the autonomic nervous system may cause major variations in the electrophysiologic parameters mentioned earlier. Therefore, accurate evaluation of those parameters should be undertaken in the presence and absence of complete autonomic blockade. These basic concepts are necessary for appreciation of the mechanisms by which alcohol could promote the genesis of atrial fibrillation.
Alcohol and Cardiac Arrhythmias
The association between alcohol use – whether acute or chronic – and cardiac arrhythmias has been widely described in the literature. In 1978, Ettinger et al. described a clinical syndrome they defined as 'acute changes in cardiac conduction or rhythm, associated to the ingestion of high amounts of alcohol in individuals with no other evidence of cardiac diseases, and which disappear without sequelae under abstinence'. As assistance to those patients was higher at certain week days (from Saturday to Tuesday), and on holiday season (between December 24 and January 1), the condition was named 'holiday heart syndrome'. The major electrocardiographic change found in that syndrome was the onset of supraventricular arrhythmias, particularly atrial fibrillation, which led the authors to infer that those patients had reduced atrial fibrillation threshold. The description was based on the observation of 24 alcoholics, with a total of 32 hospital admissions. The patients showed premature beats or tachyarrhythmias, especially atrial fibrillation. From those 32 hospital admissions, 19 occurred between Saturday and Tuesday. The remaining six episodes occurred between Christmas and New Year.
Other reports on acute atrial fibrillation and alcohol that are worth pointing out are as follows. Thornton et al. (1984) described four cases of acute atrial fibrillation and alcohol. Loewenstein et al. (1983) and Rich et al. (1985) considered alcohol as the cause for atrial fibrillation in 30–60% of patients presenting no cardiac condition, especially in those under 60 year old. Cohen et al. (1988) showed relative risk to be two-fold among those consuming high doses of ethanol (>6 doses/day) as compared to those consuming little ethanol (<1>et al. (1990, 1987) and Kupari and Koskinen (1991), while assessing ethylic populations through a questionnaire, observed that 42% of all cases of isolated atrial fibrillation among middle-aged men consuming over 150 g of ethanol/week was due to alcohol. Whyte et al. (2004) described one case of a freestyle skier who presented an atrial fibrillation episode during an exercise test. Later, the skier informed that he had ingested 12 units of alcohol on the previous day. After four weeks of abstinence, the test was repeated and no atrial fibrillation was observed. Finally, Koul et al. (2005) described one case of alcohol-induced atrial fibrillation in a 16-year-old male.
Controversies abound, starting with the term 'holiday heart syndrome'. Koskinen et al. (1987) and Kupari and Koskinen (1991) found correlation between ethanol consumption and atrial fibrillation. However, as opposed to the report by Ettinger et al. (1978), the incidence of that arrhythmia was higher on weekdays. Another conflicting aspect is whether the development of atrial fibrillation would occur during acute ingestion of alcohol, some hours later, or in the hangover period.
Epidemiological Studies
No correlation between ethanol ingestion and atrial fibrillation could be found by some major epidemiological studies, such as the one conducted in Framingham [Benjamin et al. 1994], the Manitoba study [Krahn et al. 1995], the Multifactor Primary Prevention study [Wilhelmsen et al. 2001], and the Renfrew/ Paisley study [Stewart, 2001]. The Cardiovascular Health study [Psaty et al. 1997] reported a reduced risk of atrial fibrillation depending on the alcohol dose ingested. A Danish study comprising 47,949 participants, in turn [Frost and Vestergaard, 2004], reported increased risk of atrial fibrillation or atrial flutter in males ingesting alcohol at least twice a week. The same authors could not, however, correlate alcoholic acute intoxication episodes to atrial fibrillation episodes in women, which was explained by the lower consumption of ethanol by females. In an analysis of the Framingham study [Djousse et al. 2004], comparing alcohol ingestion and counting on a control group, the follow-up of 10,333 patients for a longer-than-50-year period, with 1,055 cases of atrial fibrillation in that time span (544 males and 511 females), the association between moderate alcohol consumption and atrial fibrillation was low, but significant among individuals ingesting over 36 g/day of ethanol, which means to say more than 3 drinks per day (34% increased risk of AF – 95% CI 1–78%). Planas et al. (2006) investigated 115 patients reporting their first atrial fibrillation episode in Catalonia, Spain. In the six-month follow-up, 32 patients (27.8%) reported relapses. The authors concluded that the risk of idiopathic atrial fibrillation recurrence was high, and was enhanced by moderate alcohol consumption and increased left ventricular ectopic activity, probably of sympathetic origin. This trend was less marked in paroxysmal atrial fibrillation of vagal origin. In the Copenhagen City Heart study, Mukamal et al. (2005) described increased risk of atrial fibrillation (RR 1.45, 95% CI 1.02– 2.04) in males consuming more than 35 drinks per week. While assessing 1,232 atrial fibrillation cases in the Cardiovascular Health study on an average follow-up time of 9.1 years, Mukamal et al. (2007) also concluded that current moderate alcohol consumption is not associated with the risk of atrial fibrillation or with risk of death after diagnosis of atrial fibrillation, but former drinking identifies individuals at higher risk. Guize et al. (2007) studied atrial fibrillation prevalence in France. In a large population (98,961 males and 55,109 females), with average follow-up time of 15.2 years, alcohol consumption showed to be associated with atrial fibrillation only in men [OR = 1.7 (1.2–4)]. In a prospective study with a control group, Marcus et al. (2008) recently evaluated 195 consecutive patients that had been referred for atrial fibrillation ablation or atrial flutter in a period of two years. A significant, positive association between alcohol use and atrial flutter was reported in younger patients. The authors inferred that a possible mechanism for alcohol action could have been linked to high right atrium effective refractory period reduction in those patients.
Despite these controversial results, it seems reasonable to conclude that chronic overconsumption of ethanol is a common risk factor for atrial fibrillation in an otherwise healthy individual ( Table 1 ).
Electrophysiological Effects of Ethanol in Humans
Several studies with limited number of patients have been conducted attempting to elucidate how acute alcoholic promotes atrial arrhythmias, particularly atrial fibrillation. Greenspon and Schaal (1983) investigated 14 patients after ethanol infusion. They demonstrated that ethanol increased heart rate and reduced corrected sinus node recovery time. However, noteworthy is the fact that it did not change refractory periods. Gould et al. (1978) also investigated 14 patients. They observed that after alcohol infusion only the ventricular refractory period was shortened, as opposed to Engel and Luck (1983), who did not detect any change in atrial effective refractory period in the eleven patients investigated. The latter authors suggested that alcohol-related atrial arrhythmias might be attributed to intra-myocardial catecholamine release or to toxic direct effect of the metabolite acetaldehyde. Steinbigler et al. (2003) investigated 40 patients with a history of atrial fibrillation related to alcohol consumption with signal averaged ECG and demonstrated that those individuals had their P wave duration significantly prolonged by ethanol – a predisposing factor for atrial fibrillation. Maki et al. (1998) studied heart rate variability in six male patients with a previous history of ethanol-induced atrial fibrillation. The authors demonstrated that acute alcoholic intoxication leads to an increased sympathetic drive. Another pending issue is whether alcohol could cause arrhythmia upon withdrawal. This could be associated with high adrenergic responses and/or electrolyte disorders, particularly low potassium and magnesium levels. However, other authors, as Buckingham et al. (1985), Gribaldo et al. (1985) and Denison et al. (1994) have not observed an increased incidence of atrial or ventricular arrhythmias in the period through electrocardiographic monitoring.
In regard to electrocardiographic parameters in the literature, the few occurrences on ECG changes and alcohol refer to the chronic use of the drug. Uyarel et al. (2005) studied ten young, healthy volunteers and the changes in ECG after oral consumption of ethanol. They observed P wave duration increase. Lorsheyd et al. (2005) also studied ten healthy, young volunteers. After acute ethanol intake, the authors found PR and QTc interval increase. Those studies were conducted with a small number of patients who had an intact autonomic nervous system. Only surface ECG was evaluated. No other electrophysiologic parameters were investigated.
A few studies [Engel and Luck, 1983; Greenspon and Schaal, 1983] evaluating the effects of ethanol on inducibility of atrial arrhythmias were undertaken in alcoholics, and suggested that chronic alcohol ingestion may increase vulnerability to atrial arrhythmias. However, whether these findings are applicable to nonalcoholic subjects with normal hearts remain uncertain. In fact, one of the drawbacks in some of the above-mentioned clinical studies is that the patients themselves, through a questionnaire, provided the information on alcohol consumption. The most commonly used questionnaire was CAGE, with 81–91% sensitivity, and 77– 89% specificity for the discrimination of alcoholic and non-alcoholic patients.
Based on previous studies, it is fair to state that chronic consumption of alcohol may create a substrate that eventually increases vulnerability to atrial arrhythmias ( Table 1 ). However, the mechanisms responsible for this proarrhythmic response remain obscure.
Electrophysiological Effects of Ethanol in Experimental Studies
Most studies conducted to evaluate the electrophysiologic properties of alcohol were based on in vitro isolated preparations of heart cells. However, notwithstanding the relevance of these studies, one has to keep in mind the limitations of this kind of preparations, such as the high concentrations of ethanol used in the perfusate, when compared with in vivo models. Thus, in vitro findings should not be directly extrapolated to the clinical arena. Some of the in vitro studies that should be mentioned include the study by Williams et al. (1980), using isolated canine and swine cells. The authors demonstrated a reduction (around 8%) of transmembrane action potential duration and inferred that such alteration was secondary to decreased calcium currents. The study by Habuchi et al. (1995) using ventricular cells of swine demonstrated that the observed calcium channel inhibition – both under acute and chronic alcoholic intoxication – was responsible for the negative inotropic effect through action potential reduction and the development of arrhythmia. The study conducted by Snoy et al. (1980) showed the same results. Carpentier and Gallardo-Carpentier (1987) used rat cells, and demonstrated increased automaticity of sinoatrial cells. Opposite results also related to the automaticity of rat atrial cells were reported by Jain and Carpentier (1998). Of note, while studying cardiomyocytes from the pulmonary veins of rabbits and counting on a control group (with no perfusion with ethanol), Chen et al. (2004, 2002, 2001) demonstrated that although alcohol reduces action potential duration, it did not increase the incidence of delayed afterdepolarizations of pulmonary veins cardiomyocytes, in contrast to what had been shown by the same group during the perfusion of the same cells with thyroid hormone or after fast stimulation. This finding suggests that ethanol has no direct effects on the arrhythmogenic potential of these cells.
Although in vitro studies suggest that alcohol at high concentrations has a depressant effect on calcium currents ultimately decreasing action potential duration, these observations have not been reproduced in intact animal models ( Table 1 ). From the in vivo experimental studies, those worth mentioning are the following: in an experimental swine model, Anadon et al. (1996) demonstrated that high doses of alcohol (higher than those tolerated by humans) facilitated induction of atrial fibrillation and atrial flutter, but evaluation of atrial electrophysiological parameters and experiments in sham controls were not performed. In a canine model, Goodking et al. (1975) observed that alcohol had a depressant effect in atrioventricular conduction, but did not alter intraventricular conduction. However, the authors did not investigate the autonomic influences and had no sham control group. Kostis et al. (1977), also using dogs, reported that paradoxically to what had been previously demonstrated, alcohol showed an antiarrhythmic effect, and that the atrial fibrillation very often considered resulting from alcohol ingestion could have been due to electrolyte, autonomic or histological changes. Using canine models, Nguyen et al. (1987) observed that alcohol caused vasodilation, negative inotropic effect, and atrial antiarrhythmic effect. Madan and Gupta (1967) also observed alcohol atrial and ventricular antiarrhythmic effect in canine models.
Given the conflicting results and limitations of the previously mentioned investigations, we [Fenelon et al. 2007] have recently conducted a comprehensive study on the in vivo electrophysiological effects of acute alcoholic intoxication. We evaluated the cardiac electrophysiologic effects of ethanol in 23 anaesthetized dogs, with structurally normal hearts, at baseline and after two cumulative IV doses of alcohol or saline in the control group: first dose – 1.5 ml/kg (mean plasma level at 200 mg/dl); second dose – 1.0 ml/kg (279 mg/dl). Those doses correspond to acutely moderate and severe alcoholic intoxication, respectively. The dogs were divided into five groups: group I – ethanol group: closed chest, absence of autonomic blockade (n = 5); group II – sham control group (n = 3): closed chest, saline infused, rather than ethanol; group III – ethanol group: closed chest under complete pharmacological (atropine + propranolol) autonomic blockade (n = 5); group IV – ethanol group: closed chest, no autonomic blockade, for evaluation of left ventricular ejection fraction using 2D echocardiogram and biopsies of atrial tissue for histological and ultrastructural analysis (n = 5); and group V – ethanol group: absence of autonomic blockade, open chest, and biatrial epicardial mapping (n = 5). Haemodynamic, electrocardiographic and electrophysiologic parameters were assessed. In groups I, II and III, high right atrium monophasic action potential (MAP) recordings, measured at 90% of repolarization, were obtained with standard techniques [Franz, 1991] as previously reported [Fenelon and Brugada, 1998]. Group IV was evaluated for left ventricular function and atrial tissue was obtained in the same group for optical and electron microscopy. In group V, the chest was opened and an eight-bipole plaque was placed in Bachmann's bundle to measure interatrial conduction time, conduction velocity, and wavelength. As mentioned earlier, the wavelength is the best predictive parameter for the induction of re-entrant atrial arrhythmias [Rensma et al. 1988]. Inducibility of atrial arrhythmias was assessed with up to four extra stimuli and rapid burst pacing for 15 s duration. Atrial and ventricular tachyarrhythmias longer than 30 s in duration were considered sustained.
In groups I, II and III, ethanol was not shown to significantly alter hemodynamic variables (systolic, diastolic and mean blood pressure), electrocardiographic variables (P wave duration, QRS duration; PR and QT interval), or electrophysiologic parameters (Figure 1) (PA interval duration, HV interval, MAP duration, right atrium effective refractory period, corrected sinus node recovery time). In group V (open chest and biatrial epicardial mapping), ethanol did not affect interatrial conduction time (Figure 2), conduction velocity or wavelength. In all groups, no atrial or ventricular arrhythmias were induced in any dog. Histological and ultrastructural analysis was normal in all animals in group IV.
Figure 1.Electrocardiographic and right atrial monophasic action potentials (MAP) in ethanol-treated dogs without autonomic blockade. From top to bottom: ECG lead II and right atrial MAP. The duration of MAP at 90% repolarization (MAP90) in ms is indicated. Tracings recorded before (baseline) and after the first (dose 1) and second (dose 2) doses of ethanol. No appreciable changes occur after alcohol. Paper speed at 100 mm/s. Reproduced with permission from Fenelon G et al. (2007) Alcohol Clin Exp Res 31(9): 1574–1580.
Figure 2.Epicardial mapping in open-chest dogs is shown. Interatrial conduction during continuous right atrial appendage (RAA) pacing (200 ms cycle length) recorded before (baseline) and after the first (dose 1) and second (dose 2) doses of ethanol is depicted. Plaque bipole numbers and inter-atrial conduction time in milliseconds are indicated. Asterisks differentiate local left atrial appendage (LAA) activation from far field stimulus artifact. No appreciable changes in inter-atrial impulse propagation occur after alcohol. Paper speed at 200 mm/s. Reproduced with permission from Fenelon G et al. (2007) Alcohol Clin Exp Res 31(9): 1574–1580.
Ejection fraction was also evaluated in group IV and showed a significant reduction (77% vs 73% vs 66%; p = 0.04) with cumulative ethanol doses. Based on those results, and stressing the fact that it is an experimental canine model with normal hearts, moderate and high doses of alcohol promoted discreet, progressive left ventricular systolic dysfunction, did not alter electrocardiographic or electrophysiologic cardiac parameters, did not induce histological or ultrastructural changes in atrial tissue, and did not promote atrial or ventricular arrhythmia inducibility. These findings suggest that acute alcoholic intoxication does not exert direct myocardial actions that may create a substrate for the development of arrhythmias. The authors are not aware of any other study that has evaluated all those parameters concurrently, whether experimentally or clinically.
Conclusion
The 'holiday heart syndrome' was defined in 1978 as the occurrence of alcohol-induced arrhythmias, chiefly atrial fibrillation, in otherwise healthy individuals. Such arrhythmias may occur during acute alcohol intake and withdrawal. However, as mentioned earlier in this review, the association between alcohol ingestion and atrial fibrillation has been challenged by several epidemiological studies and remains controversial.
Our own and others' data suggest that acute alcohol ingestion, per se, does not render the atrium susceptible to atrial fibrillation. The observation that ethanol does not exert significant effects on atrial electrophysiology (refractory period, conduction velocity and wavelength) and arrhythmia inducibility supports the hypothesis that the development of atrial arrhythmias in the setting of alcoholic intoxication may require additional pathological conditions, such as metabolic disturbances, autonomic imbalance or sleep apnoea [Wong, 1973; Rosenqvist, 1998]. Corroborating this premise, although it is well known that ethanol directly affect myocardial contractile function, indirect effects on the myocardium may also occur. Both ethanol and its metabolite acetaldehyde have been shown to increase levels of circulating catecholamines [Williams et al. 1980]. Further, ethanol may also induce oxidative stress and release of plasma free fatty acids [Rosenqvist, 1998]. These prominent indirect effects of ethanol may be arrhythmogenic, particularly in individuals prone to atrial fibrillation such as patients with focal atrial fibrillation. Furthermore, other patients susceptible to atrial fibrillation include those with structural heart diseases and, possibly, long-term alcohol users in whom subclinical cardiac abnormalities may occur ( Table 2 ).
There is clearly a need for further studies evaluating the relationship between alcohol and atrial fibrillation in otherwise healthy individuals. However, these studies should contemplate the recent knowledge gained on the mechanisms of atrial fibrillation genesis, foremost focal atrial fibrillation. Further, the electrophysiological and structural effects of chronic alcohol consumption should be better characterized. These studies are key to determine if and how alcohol promotes atrial fibrillation.
Table 1. Summary of Studies Evaluating Alcohol and Atrial Fibrillation
Table 2. Possible Mechanisms of Alcohol-induced Atrial Fibrillation
References
|
Nenhum comentário:
Postar um comentário