. Association Between Atrial Fibrillation and Dementia in the General Population. JAMA Neurol. 2015 Nov;72(11):1288-94. PubMed.


Please login to recommend the paper.


  1. This study is confirmatory of our work (Bunch et al., 2010). In our study we found that patients who developed atrial fibrillation were at higher risk of all forms of dementia, including Alzheimer’s disease. We also found that the highest risk was in atrial fibrillation patients who were younger than 70 years of age. In this very nice study, de Brujin and colleagues found that the duration of exposure to atrial fibrillation was of particular importance in dementia risk. Their study clearly adds to the growing body of evidence that atrial fibrillation is a significant risk factor for dementia.

    Since our initial discovery, we had tried to understand the association. So far we have found five clinical risk factors in atrial fibrillation patients that raise the risk of dementia:

    1. Type, duration, and efficacy of anticoagulation treatment. Most atrial fibrillation patients are treated with anticoagulation to lower their risk of stroke. Worldwide, the most common drug used is warfarin (coumadin). Warfarin levels can be erratic in some people, often exposing them to periods of both under- and over-anticoagulation. The most exposure to periods of under- or over-anticoagulation raises risk of dementia significantly. These findings suggest that one of the reasons people with atrial fibrillation develop dementia is that they are exposed to small clots or bleeds within the brain, or both. This dementia risk is increased in atrial fibrillation patients who need to take aspirin or plavix because they have had a prior heart attack, stent, or bypass surgery. In this latter scenario, the added anticoagulant effects of aspirin likely increase risk of microbleeds.

    2. Sleep apnea. Patients with untreated sleep apnea have a 50 percent chance of developing atrial fibrillation. Sleep apnea leads to acceleration and worsening of atrial fibrillation. Untreated sleep apnea also makes it much more difficult to treat atrial fibrillation. Sleep apnea also is associated with cognitive decline and dementia.  

    3. Heart rate. The heart often compensates for a slight loss in efficacy from atrial fibrillation by elevating the heart rate. We treat elevated heart rates to minimize risk of heart failure with atrial fibrillation. However, when the heart rate falls below 70 beats per minute (a normal level in people without atrial fibrillation), organ perfusion can be impacted. We have found an inverse association between heart rate and dementia risk, i.e., the lower the average heart rate over 24 hours, the greater the risk. The risk of dementia goes up when the average 24-hour heart rate is below 70 beats per minute in people who are in atrial fibrillation all the time. In these people, we are lowering the levels of medications used to slow the heart rate, and in some people who continue to have slow heart rates we are placing pacemakers to raise the heart rate.

    4. Genetic factors.  We have found genetic mutations associated with early atrial fibrillation and with stroke and thrombosis that predict early onset dementia. In this case, people who have first-degree relatives with early atrial fibrillation and dementia may be at increased risk. In those people, we are particularly careful in regard to the risk factors noted above to minimize any additional risk.

    5. Inactivity. Inactivity and lack of exercise is a significant risk factor for atrial fibrillation and cognitive decline. We believe that if we start increasing activity to promote weight loss early after atrial fibrillation develops, the negative effects of the arrhythmia can be significantly reduced. Activity and exercise also lead to healthier vessels of organs such as the brain, which improves its perfusion and function over a lifetime. 

    View all comments by T. Jared Bunch
  2. This paper by de Bruijn et al. settles uncertainty from a previous study that evaluated the effect of atrial fibrillation (AF) as a risk factor for dementia and Alzheimer’s disease in stroke-free subjects.

    De Bruijn demonstrated that AF is associated with an increased risk of dementia independent of previous clinical stroke. This association was strongest for people who developed AF at a younger age (less than 67 years) and who had the longest duration of AF compared to those who developed AF when elderly. This suggests that possible negative effects of AF on cerebral functioning takes time to evoke dementia, at least in subjects with no evidence of previous stroke.

    We found the results of this study of great importance because they are based on a long-term follow-up of 20 years. This is an adequate time to observe the effects of not only prevalent but also incident AF on dementia. Moreover, the longitudinal evaluation reinforces findings from the previous cross-sectional Rotterdam Study (Ott et al.,  1997). 

    We found these results “conservative” relative to what we might find in the real word because the analysis was based on permanent AF cases. The study underscores the need to consider subjects with paroxysmal AF, in whom the ping-pong from normal sinus rhythm to AF could exert a greater negative effect on cerebral perfusion and thromboembolism than that exerted by a persistent AF. Moreover, paroxysmal AF is frequently underdiagnosed and consequently not treated with anticoagulation therapies. Therefore, the observed dementia hazard ratios for prevalent AF (1.34) and incident AF (1.13), the latter without statistical significance, might be greater in studies that not only include parossistic AF but consider the differences between rhythm versus rate control strategy. These results would be of great clinical interest, especially if we considered the possible benefits of anticoagulant therapy in elderly AF patients in whom asymptomatic strokes are more frequent.

    The omission of parossistic AF could also conceal an effect in elderly patients in whom the heart rate response to AF could be lower and more difficult to diagnosis, because it is not clinically symptomatic. Moreover, in a previous study we found a different risk profile depending on ventricular rate response to AF, identifying a greater risk for dementia development in subjects who were cognitively impaired and had rate response of less than 50 or greater than 90 beats per minute. Indeed, a higher variability in heart rate response defines patients already prone to dementia, in whom even smaller changes in stroke volume could have broader effects (Cacciatore et al., 2012). Thus, de Bruijn et al.’s manuscript strengthens the link between cardiovascular and central nervous system pathophysiology.

    David Della Morte also contributed to this comment.

    In summary, the relationship between dementia and cardiovascular diseases is of particular interest and could be further characterized by considering, in an extended statistical model, the number of incident low cardiac output states in the population, including the multiple, chronic, organ failure states, such as respiratory and renal failure.


    . Atrial fibrillation and dementia in a population-based study. The Rotterdam Study. Stroke. 1997 Feb;28(2):316-21. PubMed.

    . Role of ventricular rate response on dementia in cognitively impaired elderly subjects with atrial fibrillation: a 10-year study. Dement Geriatr Cogn Disord. 2012;34(3-4):143-8. Epub 2012 Sep 12 PubMed.

    View all comments by Francesco Cacciatore

Make a Comment

To make a comment you must login or register.

This paper appears in the following:


  1. Atrial Fibrillation in Middle Age Ups Dementia Risk