Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Alcohol Consumption and AF

Framingham study [2]
Danish study [3]
Copenhagen study [4]
Comment

Bandolier loves a mystery, or the appearance of a mystery. Several readers have asked about links between behaviour and atrial fibrillation (AF), specifically about exercise and alcohol. For exercise there seems little direct linkage of prior exercise leading to future AF (though others may know better). For alcohol, the situation is different, in those who imbibe too freely or too well.

The first references linking high alcohol consumption and AF were from Poland in the late 1960s, but the holiday heart syndrome was defined in 1978 [1] as:

An acute cardiac rhythm and/or conduction disturbance associated with heavy ethanol consumption in a person without other clinical evidence of heart disease, and disappearing, without evidence residual, with abstinence.

That may be a short-term poisoning effect, and is not quite the same as effects over the longer term related to habitual high alcohol consumption. Bandolier has examined some of the recent literature found in a quick search. Three observational studies come to much the same conclusion. In what follows Bandolier is giving alcohol consumption as grams per day; 10 grams is about what is found in a 330 mL bottle of beer, which is about one unit.

Framingham study [2]


This detailed, prospective, long-term study of a cohort of 10,333 Americans had 1,055 cases of AF and 4,672 controls free of AF. The average age was 45 years at baseline. Patients with AF had significantly greater rates of hypertension, myocardial infarction, heart failure, hypertrophy, valvular disease and diabetes.

Rates of AF were calculated at different levels of daily alcohol consumption, with relative risk compared to no alcohol consumption. For women and men separately there was no significant increased risk at any level of consumption, but there was a trend to greater risk at higher consumption (Figure 1 shows this for men). Combining data for both men and women, the relative risk was significantly higher at highest alcohol consumption, with a relative risk of 1.3 (95% CI 1.01 to 1.78). Risk calculations were adjusted for various possible confounding factors.


Figure 1: Relative risk of AF in men by daily alcohol consumption (Framingham)






Danish study [3]


This Danish study was large, with 50,000 people aged 50-64 years at baseline contributing information, and long, with an average duration of follow up of about 5.8 years. Over this time the incidence of AF was 1.7% in men (29 per 10,000 person years; 3 per 1,000 per year) and 0.7% in women (12 per 10,000 person years; 1 per 1,000 per year). Almost 80% of women consumed 20 grams of alcohol per day or less, while 50% of men consumed 20 grams of alcohol per day or more.

This analysis used quintiles of alcohol consumption. For women the highest quintile consumed an average of 39 grams of alcohol per day. There was no significant association between alcohol consumption and AF for women. For men the highest quintile consumed an average of 69 grams of alcohol per day. There was a significant association between alcohol consumption and AF for men (Figure 2). At the highest level of consumption the adjusted relative risk was 1.5 (1.1 to 2.0).


Figure 2: Relative risk of AF in men by daily alcohol consumption (Denmark)






Copenhagen study [4]


Another, but different, study from Denmark, used a randomly drawn sample of the population of Copenhagen, and analysed 16,400 individuals. The average age was 50 years. Almost all women (90%) consumed 20 grams or less of alcohol daily (two drinks a day), while two-thirds of men consumed more than 20 grams a day. This long-term study had an average of 17 years of follow up.

The analysis here was by different levels of alcohol intake. For women, the highest level of intake was 29 grams daily or above. There was no significant association between alcohol consumption and AF for women. For men, there were six groups; the highest was 50 grams daily or more. There was a significant association between alcohol consumption and AF for men (Figure 3). At the highest level of consumption the adjusted relative risk was 1.5 (1.02 to 2.0).


Figure 3: Relative risk of AF in men by daily alcohol consumption (Copenhagen)






Comment


AF has been associated previously with alcohol abuse [1,5], with a doubling of risk at alcohol consumption above 42 units a week, or about 60 grams daily (more than six bottles of beer a day). These three studies would seem to confirm that in men, long-term consumption of about 40 grams of alcohol a day (four units or more, four bottles of beer or more) is associated with an increase in risk by about 30% to 50%. They also show a dose response relationship.

Demonstrating any effect in women is more difficult because few women drink enough. In the studies, most women consumed two drinks a day (bottles of beer, moderate glasses of wine) or less. In consequence there were few drinking sufficient to demonstrate an effect. That does not mean, though, that the laws of physiology do not apply to women. The absence of any evidence is not evidence of no effect at higher alcohol intakes.

Having a holiday heart flutter of the non-romantic sort after drinking more than usual is probably signalling that too much alcohol is damaging that particular heart. What we are seeing is a dose response relationship, because moderate alcohol intake has been shown to produce less heart failure than no alcohol [6]. The message is to cut down, probably to no more than one or two drinks a day. Quality rather than quantity, perhaps.

References:


  1. PO Ettinger et al. Arrhythmias and the “Holiday Heart”: alcohol-associated cardiac rhythm disorders. American Heart Journal 1978 95: 555-562.
  2. L Djoussé et al. Long-term alcohol consumption and the risk of atrial fibrillation in the Framingham study. American Journal of Cardiology 2004 93: 710-713.
  3. L Frost, P Vestergaard. Alcohol and risk of atrial fibrillation or flutter. Archives of Internal Medicine 2004 164: 1993-1998.
  4. KJ Mukamal et al. Alcohol consumption and risk of atrial fibrillation in men and women. Circulation 2005 114: 1736-1742.
  5. A Ruigomez et al. Incidence of chronic atrial fibrillation in general practice and its treatment pattern. Journal of Clinical Epidemiology 2002 55: 468-473.
  6. JL Abramson et al. Moderate alcohol consumption and risk of heart failure among older persons. JAMA 2001 285: 1971-1977.

previous or next story