Alcohol may act proarrhythmogenic in many mechanisms. At the cellular level, alcohol can damage intercellular junctions and cells, trigger inflammation and oxidative stress, and disrupt the regulation of ion channels in the myocardium [14, 15]. Within the autonomic nervous system, alcohol increases the activity of the sympathetic component and reduces heart rate variability (HRV) [14, 15].

Moreover, alcohol promotes dilatation and fibrosis of the left atrium and increases its pressure [16, 17]. Long-term alcohol consumption promotes atrial cardiomyopathy (alcoholic cardiomyopathy) associated with structural, functional and electrical remodeling of the atria, thus stabilizing AF episodes and contributing to AF progression [16, 17]. In addition, alcohol consumption often leads to other risk factors for AF, such as being overweight, obstructive sleep apnea, and arterial hypertension [16]. Myocardial hypertrophy resulting from chronic alcohol consumption by increasing the reactivity of transient receptor potential cation channels (TRPC) leads to electrical instability of cardiomyocytes, which directly contributes to the occurrence of arrhythmias [18, 19].

Alcohol directly causes a number of disorders in the electrophysiology of the heart. This effect is related to the dysregulation of ion channels. One in silico study found that alcohol reduced I${}_{\text{Na}}$, I${}_{\text{Ca,L}}$, I${}_{\text{to}}$, I${}_{\text{Kr}}$ and I${}_{\text{Kur}}$, dual effects on I${}_{\text{K1}}$ and I${}_{\text{K,ACh}}$ (inhibition at low and augmentation at high concentrations), and increased I${}_{\text{NCX}}$ and sarcoplasmatic reticulum (SR) Ca${}^{2+}$ leak and SERCA uptake in cardiomyocytes. Alcohol may also damage the gap junctions between cardiomyocytes [20]. Finally, alcohol consumption leads to decreased biatrial conduction velocities (CV), action potential duration and refractoriness, and shortened right atrial effective refractory periods (AERP) while increasing the dispersion of atrial refractoriness [20].

Heavy alcohol consumption can result in the holiday heart syndrome, which was first described in 1978 and occurs when healthy subjects without heart disease known to cause arrhythmia experience AF after excessive alcohol consumption, which was observed mainly after weekends or holidays [21]. This can probably be attributed to such effects of alcohol as: inhibition of sodium channels in myocardial cells, stimulation of the sympathetic and parasympathetic nervous system, increase in the concentration of free fatty acids in the blood and stimulation of c-Jun N-terminal kinase 2 (JNK2) kinase [secondarily leads to the stimulation of calmodulin II kinase (CaMKII) and the release of calcium ions from the SR, which ultimately contributes to the development of AF], and above all - electrolyte changes, including hypokalemia and hypomagnesemia, after taking large doses of alcohol in a short time [21, 22, 23].

In a study by Voskoboinik et al. [24] including 75 patients undergoing AF ablation using high density electroanatomical mapping, moderate alcohol consumption (8–21 drinks/week) was found to be associated with lower bipolar voltage and slower atrial conduction velocity. In a randomized and placebo-controlled study by Marcus et al. [25], covering 100 subjects, acute exposure to alcohol reduces atrial effective refractory periods (AERP), particularly in the pulmonary veins, which is substrate for AF. In a study by Sha et al. [26], including 134 patients with AF, alcohol causes echocardiographic and electrophysiological changes such as impaired peak left atrial longitudinal strain, obvious inter-atrial conduction delay and increasing ERP dispersion. Intra-atrial conduction delay and ERP dispersion also increased with increasing levels of alcohol consumption [26]. The most important mechanisms of the influence of alcohol consumption on the pathogenesis of AF are summarized in Fig. 1 (Ref. [14, 15, 27, 28]).

Fig. 1.

Main acute and chronic pathophysiological mechanisms triggered by alcohol consumption and leading to the development of atrial fibrillation. Based on [14, 15, 27, 28]. Abbreviations: AF, atrial fibrillation; ERP, effective refractory period; PV, pulmonary vein; HRV, heart rate variability; JNK2, c-Jun N-terminal kinase 2; CaMKII, calmodulin II kinase.

The safer effect of drinking wine, observed in some studies [17], in contrast to other alcohols, may be due to the richness of polyphenols contained in this red and white beverage. Of particular interest is resveratrol, a polyphenol found in large amounts in red wine [29]. Resveratrol has been shown to reduce the inducibility of AF, potentially by reducing left atrial fibrosis and regulating ion channel function [30]. The overall effect of polyphenols on reducing the risk of AF is associated with, among others, anti-inflammatory, antioxidant and anti-fibrotic effects, as is the case with the consumption of polyphenol-rich coffee [31].

In summary, alcohol consumption triggers both acute and chronic mechanisms that lead to the onset of AF. Regular consumption of even ESC-acceptable alcohol doses (100 g/week) leads to electrical and structural remodeling of the atria and increases the risk of AF. The polyphenols contained in wine can to some extent mitigate the negative effects of alcohol itself.

The most important results relating alcohol consumption and AF are summarized in Table 1 (Ref. [17, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]).

In a meta-analysis of 6 studies the risk of AF was significantly increased from the consumption of 12 g or more of alcohol/day. A linear dose-response relationship has been demonstrated between alcohol consumption and the risk of AF [32]. A meta-analysis of 14 studies found that higher (versus lower) alcohol consumption was associated with a significant increase in the risk of AF (RR = 1.51; 95% CI: 1.31–1.74). A linear dose-response relationship has been demonstrated between alcohol consumption and the risk of AF [33]. A meta-analysis of 9 studies (249,496 subjects) found that heavy to moderate alcohol consumption was significantly associated with an increased risk of AF (by 34% and 11%, respectively). Low alcohol consumption was not associated with the risk of AF [34]. In a meta-analysis of 7 prospective studies (205,073 participants), it was found that each one drink/day increase in alcohol consumption was significantly associated with an 8% increase in the risk of AF (RR = 1.08; 95% CI: 1.06–1.10) [35]. One meta-analysis of 13 studies (10,266,315 subjects) found that regular moderate (about 2 units of alcohol/day) alcohol consumption increased the risk of AF (HR = 1.14; 95% CI: 1.07–1.21), while low alcohol consumption (about 1 unit of alcohol/day) had no effect on this risk [36]. Another meta-analysis of 13 studies (668,905 subjects) found that heavy alcohol consumption increased the risk of AF by 30% (HR = 1.30; 95% CI: 1.20–1.41), moderate by 12% (HR = 1.12; 95% CI: 1.06–1.18), while low did not affect the risk (HR = 1.00; 95% CI: 0.96–1.05) [37].

Nonetheless, a meta-analysis of 16 studies (13,044,007 subjects) showed a slightly different relationship between alcohol consumption and the risk of AF. A relationship in the form of a “J” curve was found, as high alcohol consumption ($>$168 g/week) was significantly associated with an increased risk of AF, while moderate consumption (168 g/week) was characterized by a protective effect [38]. However, the results of this meta-analysis were not confirmed in meta-analysis of 13 prospective studies, covering over 10 million subjects, conducted by Jiang et al. [39]. This meta-analysis found that each increase in alcohol consumption by one drink/day is significantly associated with an increased risk of AF (RR = 1.06; 95% CI: 1.03–1.08) [39].

In the losartan intervention for endpoint Reduction in Hypertension (LIFE) randomized clinical trial, including 9193 patients with electrocardiography (ECG) signs of left ventricular hypertrophy and arterial hypertension, the consumption of $>$10 units of alcohol/week increased the risk of AF by 60% (HR = 1.60; 95% CI: 1.02–2.51) [40]. In a prospective study including 79,019 subjects who were followed for 11 years, the consumption of 1–6 alcohol units/week was not associated with the risk of AF (RR = 1.06; 95% CI: 0.98–1.15). Consumption of 7–14, 15–21 and $>$21 alcohol units/week significantly increased the risk of AF by 12%, 18% and 43%, respectively [35]. In a study by Cha et al. [41], including 19,634 subjects who were followed for 7.0 $\pm$ 2.8 years, the risk of AF was significantly higher in alcohol drinkers versus nondrinkers (HR = 2.21; 95% CI: 1.55–3.14) and higher in subjects drinking more versus less alcohol (HR = 3.15; 95% CI: 1.98–4.99). In a cohort study by Csengeri et al. [42] including 107,845 subjects from Europe, the effect of the amount of alcohol consumed on the risk of AF was analyzed during a 13.9-year follow-up, whereby regular consumption of one drink/day was associated with a 16% increased risk of AF (HR = 1.16; 95% CI: 1.11–1.22). Indeed, a statistically significantly increased risk of AF was found with regular alcohol consumption at the level of 2 g/day [42]. From a methodological point of view, the most convincing results come from a prospective study by Tu et al. [17], involving 403,281 subjects from UK Biobank. In this study, the number of women and men was comparable, and the follow-up period was 11.4 years. A J-shaped relationship was found between the amount of alcohol consumed and the risk of AF. The lowest risk of AF was observed with less than 7 drinks/week (56 g/week). Above this amount, the risk of AF started to increase (significant increase risk of AF: $>$14 standard drink/week; 112 g pure ethanol/week) (Fig. 2, Ref. [17]). In a cohort study by Han et al. [43] including 1,537,836 adults aged 20 to 39 years, it was found that regular moderate (105–210 g/week) to heavy ($\ge$210 g/week) alcohol consumption significantly increases the risk of AF by 25% and 47%, respectively (HR = 1.25; 95% CI: 1.12–1.40 and HR = 1.47; 95% CI: 1.18–1.83).

Fig. 2.

Effect of alcohol consumption and different types of alcoholic beverages on the risk of developing atrial fibrillation. Data adapted from [17]. Abbreviations: AF, atrial fibrillation; HR, hazard ratio.

In an observational study by Frederiksen et al. [47] of 43,758 participants without AF at baseline, increasing alcohol consumption over a 5-year follow-up was associated with a higher risk of AF [from $\le$6.9 drinks/week to $\ge$21 drinks/week (HR = 1.38; 95% CI: 1.09–1.72) or from 14–20.9 drinks/week to $\ge$21 drinks/week (HR = 1.27; 95% CI: 1.01–1.59)]. Biddinger et al. [48] reported an analysis of 371,463 subjects from UK Biobank, where any amount of alcohol (from 1 drink/week) was significantly associated with an increased risk of AF.

In the I-STOP-AFib randomized clinical trial of 466 patients, acute exposure to alcohol increased AF risk (OR = 2.15; 95% CI: 1.17–3.61), with no evidence that other exposures, including caffeine, triggered AF [44]. In another study by Marcus et al. [45], including 100 subjects, 56 of whom had at least 1 episode of AF, within 4 hours of consuming one drink, the risk of AF (real-time documentation of each alcoholic drink consumed was self-recorded using a button on the ECG recording device) was doubled (OR = 2.02; 95% CI: 1.38–3.17), with a greater than 3-fold higher odds of at least 2 drinks (OR = 3.58; 95% CI: 1.63–7.89).

In a study covering 36,158 individuals, occasional (holiday, etc.) increases in alcohol consumption were associated with a statistically significant increase in ED visits for AF and for new-onset AF, with 1757 additional visits for new-onset, incident AF ED visits/100,000 person-years (95% CI: 945–2569 visits, p 0.001) during and shortly after periods of increased alcohol consumption [46]. The increased occurrence of AF after heavy alcohol consumption during weekend or holidays is known as holiday heart syndrome (HHS) [49].

In summary, most studies provide evidence of a strong association between alcohol consumption and the risk of AF. It cannot be arbitrarily concluded that light or moderate alcohol consumption (up to 100 g/week) does not affect the risk of AF, as even small amounts of alcohol regularly consumed increase the risk of AF.

A prospective study by Larsson et al. [35] found that for liquor, the risk of AF was increased when drinking 7–14 and $>$14 units of alcohol/week; while in the case of wine, consumption $>$14 alcohol units/week was associated with an increased risk of AF, and for beer consumption, 7–14 alcohol units/week. No significant differences in the effects of spirits or beer on AF risk were found in a prospective study by Csengeri et al. [42], where the risk of AF associated with wine consumption was the least pronounced or insignificant. In the meta-analysis by Yang et al. [36], analyzing the same types of alcohol, no significant differences in their impact on the risk of AF were found. In a prospective study by Tu et al. [17], beverage-specific analyzes have shown detrimental associations between any beer/cider consumption and the risk of AF, while for consumption of red and white wine and spirits up to 10, 8 and 3 drinks/week, respectively, no harmful effects were observed (Fig. 2). In an observational study by Frederiksen et al. [47] of 43,758 participants without AF at baseline no differences in the effect of different types of alcohol (beer, wine, spirit or liqueur) on the risk of AF were found.

In summary, the risk of AF depends on the type of alcohol consumed (highest: beer, cider and spirits; lowest: white and red wine).

Women consume significantly less alcohol compared to men. In a meta-analysis by Kodama et al. [33], heavy alcohol consumption increased the risk of AF in both women and men (37% and 32%, respectively). A similar finding was made in the meta-analysis by Samokhvalov et al. [32] showing an increased risk of AF in both women and men with 12 g of alcohol/day and more). In a prospective study by Tu et al. [17] involving an equal number of women and men (52.4% versus 47.6%), no significant sex differences in the risk of AF associated with alcohol consumption were found.

A meta-analysis by Gallagher et al. [34] found that moderate alcohol consumption significantly increased the risk of AF in men (HR = 1.26; 95% CI: 1.04–1.54) but not in women (HR = 1.03; 95% CI: 0.85–1.24). In a study by Cha et al. [41], men were also more predisposed to a higher risk of AF related to alcohol consumption compared to women. A greater susceptibility of men to develop AF in relation to alcohol was also found in the meta-analysis by Yang et al. [36] who found that although low alcohol consumption did not affect the risk of AF, moderate alcohol consumption increased it among men (HR = 1.09; 95% CI: 1.07–1.11) but not among women. A meta-analysis by Zhang et al. [37] found that moderate and heavy alcohol consumption was significantly associated with the risk of AF among men (HR = 1.21; 95% CI: 1.10–1.33 and HR = 1, respectively, 54; 95% CI: 1.26–1.89), while among women this was only evident with high alcohol consumption (HR = 1.32; 95% CI: 1.10–1.60). In a meta-analysis by Jiang et al. [39], the relationship between alcohol consumption and the risk of AF is linear among men, while among women it takes the shape of a “J” curve. In an observational study by Frederiksen et al. [47] of 43,758 participants without AF at baseline was found that increasing alcohol consumption in a 5-year follow-up significantly increased the risk of AF only among women.

Some [29, 31, 32, 34, 36], but not all [15, 27, 28, 47], studies have found a higher risk of AF in men who consume alcohol. Several studies have failed to show an increase in the risk of AF at all levels of alcohol consumption in women. Most of these earlier studies included only a few women who consumed large amounts of alcohol.

In summary, sex does not seem to have a significant effect on the risk of AF associated with alcohol consumption.

An important issue is whether certain gene polymorphisms of enzymes involved in alcohol metabolism (alcohol dehydrogenase—ADH; aldehyde dehydrogenase—ALDH) [50] may affect the risk of AF associated with alcohol consumption.

In a study by Tolstrup et al. [51], covering 88,782 subjects, ADH1B/ADH1C genetic variants were not found to affect the risk of AF associated with alcohol consumption. In a Mendelian randomized study (approximating the possibility of causal inference) by Yang et al. [52] including 8964 subjects, the rs671 polymorphism of the ALDH2 gene (decreased activity resulting in greater accumulation of the highly toxic acetaldehyde) showed a significant association with the risk of AF in men (OR = 1.65; 95% CI: 1.06–2.67). In Mendelian randomization, genetically predicted daily alcohol consumption was positively associated with the risk of AF in both sexes (OR = 3.17; 95% CI: 1.18–9.24) [52]. In a study by Yamashita et al. [53], including 656 subjects, the ALDH2${}^{\mathrm{*}}$1/${}^{\mathrm{*}}$2 carriers who habitually consume alcohol had a higher risk of AF (OR = 5.07; 95% CI: 2.03–12.70), which was associated with a lower rate of alcohol metabolism.

In a study by Biddinger et al. [48] of 371,463 participants from UK Biobank, which used Mendelian randomization (promoting the possibility of causal inference), alcohol consumption was significantly associated with the risk of AF (OR = 1.24; 95% CI: 1.08–1.44).

In summary, genetic polymorphisms of the genes of enzymes involved in ethanol metabolism may modulate the risk of AF associated with alcohol consumption.

In the Atherosclerosis Risk in Communities (ARIC) Study 15,222 subjects followed for 19.7 years, current and former drinkers were found to be at higher risk of AF than those who never drank alcohol, and every decade abstinent from alcohol was associated with an approximate 20% (95% CI: 11–28%) lower rate of incident AF [54]. Conversely, every additional decade of past alcohol drinking was associated with a 13% (95% CI: 3–25%) higher rate of AF and every additional drink/day during former drinking was associated with a 4% (95% CI: 0–8%) higher rate of AF [54]. The results of this study indicate that the lowest risk of AF was among subjects who had never consumed alcohol. In a study by Choi et al. [55], covering 1,112,682 patients with type 2 diabetes who were followed for 4 years, alcohol abstinence was associated with a low risk of AF development (HR = 0.81; 95% CI: 0.68–0.97).

In summary, reducing alcohol consumption is associated with a significant decrease risk of incident AF.

In a meta-analysis of 9 studies (5436 patients) conducted by Grindal et al. [56], including patients after catheter ablation due to AF, the impact of alcohol consumption on the risk of recurrence of arrhythmia showed that compared with patients who consumed little or no alcohol, those who consumed moderate or high amounts of alcohol had a higher probability of recurring AF (OR = 1.45; 95% CI: 1.06–1.99).

In a randomized clinical trial including 140 patients with AF, the effect of reducing the amount of alcohol consumed (from 16.8 $\pm$ 7.7 to 2.1 $\pm$ 3.7 standard drinks/week; reduction by 87.5%) on the incidence of AF recurrence, with a 45% reduction in the risk of AF recurrence (HR = 0.55; 95% CI: 0.36–0.84), which was maintained for 6 months of follow-up [57]. In a prospective study by Takahashi et al. [58], including 3474 patients undergoing catheter ablation due to AF, reducing alcohol consumption (from 140 g/week to approximately 70 g/week) significantly reduced the risk of AF recurrence or atrial tachycardia by 37% (HR = 0.63; 95% CI: 0.52–0.77), compared to a lower reduction of alcohol consumption. In a similar study by Sagawa et al. [59] on 110 patients with AF without structural heart disease who underwent single ablation, the success rate after a single ablation was significantly lower in drinkers than in abstainers (79.3% versus 95.9% at 12 months; mean follow-up, 18 $\pm$ 8 months, p = 0.013).

In summary, reducing alcohol consumption significantly reduces the risk of AF recurrence in patients after catheter ablation.