This post is also available in: हिन्दी (Hindi)

The first thing I want to say is that unfortunately the practice of modern day medicine is largely servile to clinical guidelines. We decide how to treat a patient based on how someone else (i.e whoever wrote the guideline) tells us how to manage the patient. If we stick to the guideline, we feel that we are offering top quality care and we can defend ourselves in court. If we don’t stick to the guideline then we feel vulnerable to criticism by our own colleagues and also medico-legal lawyers. 

Much of the guidelines that are published by august bodies like NICE (National Institute of Clinical Excellence), are based around an examination of the evidence base to see how beneficial an intervention may be but also how cost effective it may be. If it is not deemed beneficial or cost effective, it is not recommended and unfortunately everyone then believes that it is not worth considering in the patient’s management. 

The problem is that guidelines lag several years behind research and therefore when scientists find something important through observation, the first thing they have to do is conduct rigorous experiments to confirm or refute their suspicion, then they have to further confirm this by doing large scale human trials, then they have to publish the data, then if the data are very persuasive a bunch of ‘experts’ get together and decide whether the findings are worthy of changing guidance and then that guidance is published and it can take a few years for that change in guidelines to be adopted as a change in medical practice at grassroots level. The whole process can easily take 10-15 years and therefore patients may continue to be managed in sub-optimal manner for this duration of time even when there is available research to say that things could be done better. As doctors are generally very defensive and not particularly reflective, it is important for patients to be aware of the latest research so that they can advocate for themselves and use any available new research in their decision making process. This is the foundation of patient empowerment. This is why I started my youtube channel. I believe that patients should be equipped with all the information that is out there to allow them to work out the best way they would like their condition to be managed and their doctor’s role is that of the educator and enabler rather than the doctor acting as if they know everything and the patient knows nothing. 

Today I wanted to talk about some troubling research in the field of AF. 

AF stands for atrial fibrillation. AF is one of the most common heart rhythm disturbances and can affect up to 2% of the population. The big risk of AF is thought to be an increased risk of strokes and therefore when we see patients above the age of 65, or patients who carry comorbidities we recommend lifelong anticoagulation and as long as the patient is anticoagulated, we feel that the patient is safe.  We never really think beyond the risk of stroke. If the patient is younger than 65 and does not carry comorbidities we don’t anticoagulate them because we believe that the risk of stroke is very low.

However over the past few years there have been several studies which have studied AF patients and discovered patients with AF have a significantly higher prevalence of cognitive impairment and dementia and as we do not have any clinical guidelines as yet, we have to try and work out for ourselves as to why there is this association.

  1. Could it just be a coincidence?

It is very unlikely that this is just a coincidence because there have been several large scale studies that have confirmed this. There was a study in 2009 called the Intermountain Heart collaborative study which evaluated 37000 patients and followed them for development of AF and dementia and found that patients with AF were 44% more likely to develop dementia compared to patients without AF. In addition younger patients (<70y – this is important because these patients are traditionally considered lower risk of strokes)were at a higher risk of developing all forms of dementia and particularly Alzheimer’s (130% increased risk) and patients with AF and dementia had a significantly higher mortality compared to patients with dementia who did not have AF.

Another analysis of studies called ONTARGET and TRANSCEND showed that AF was associated with increased risk of cognitive decline, new onset dementia, loss of independence and performing activities of daily living and admission to long term care facilities.

There was another study known as the Rotterdam study which again showed that dementia was commoner in AF patients and in particular in younger patients. 

All in all there have been more than 14 studies which have looked at this association and most have confirmed this finding so it is highly likely that this is not just coincidental


  1. Could it be because AF and dementia share the same risk factors rather than specifically because the AF is causing the dementia?

AF and dementia both affect older people and sicker people esp those that have vascular risk factors such as diabetes and high blood pressure and I am sure that is true to an extent and therefore it is also way crucial when we manage patients with AF, that we tackle additional risk factors. In some ways just by anti-coagulating the patient we are not making the patient healthier, we are just trying to reduce the stroke risk. However, educating the patient to improve their lifestyle will make the patient a healthier person and therefore should be encouraged in every patient.


  1. Could AF by itself be causing dementia to develop? We know that AF patients with AF are more likely to have clot formation within the heart and those clots going to the brain cause the strokes (which result in the patient noticing a visible loss of function). However it is very possible that there may be micro-clots also forming and being dislodged and going to the brain and because these clots are so tiny they don’t cause an overt stroke but cause tiny areas of brain death and therefore cause a progressive loss of brain function. This has indeed been confirmed and there was a study called the Atherosclerosis risk in communities study which did indeed confirm that cognitive decline in patients with AF was only observed in those who had sub clinical cerebral infarcts. 


  1. Another mechanism may be that when we are in AF, our hearts are not pumping out as much blood as they would if we were in sinus rhythm and therefore could it be that lack of blood that may be contributing If indeed this were the case then perhaps keeping the patient in sinus rhythm may result in less likelihood of dementia


  1. Could it also perhaps be that the treatment of AF with anticoagulants could be contributing to micro-bleeds and this could be another mechanism for progressive loss of brain function?


Are there any studies which can help us work out which of the possible mechanisms is the most likely to explain the relationship between AF and dementia? I have already mentioned earlier that there is some evidence that AF patients can have microinfarcts without a proper stroke. If this were the case then taking an anticoagulant could be protective. There are some studies that have suggested that there is perhaps a non-significant difference favouring anticoagulation but there is no definitive evidence of cognitive benefit or harm from anticoagulation. There are however studies in patients who have been taking warfarin which suggest that if your warfarin control was very good then you were at a lower risk of dementia compared to if your warfarin control was poor. 

There was a Swedish retrospective study which looked at 444106 patients and found that those patients who had AF but never had a stroke who were also on anticoagulants at baseline had a 29% lower risk of developing dementia compared to those who were not on anticoagulants. It also seems that the benefit of anticoagulation treatment appears to be particularly more pronounced among patients in whom anticoagulation was started early after the first diagnosed episode of AF.

In summary it does appear that maybe early anticoagulation is protective against the development of dementia although the evidence is not very robust and this is why we desperately need clinical studies to look at this.

What about ablation? With an ablation, you are taking away the AF and therefore the cardiac output will improve so do patients who have had an ablation for AF less likely to develop dementia? In the intermountain AF study, the researchers compared 4212 consecutive patients who had undergone an ablation with 16848 patients with AF but who did not have an ablation and found that Alzheimer’s dementia seemed to occur in 0.2% of ablated patients compared to 0.9% non-ablated patients. It also seemed that other forms of dementia were significantly reduced in patients with AF who were treated by an ablation. We definitely need better data but a study to look at whether rhythm control can reduce dementia will need thousands and thousands of patients and would require to run for a very long time and therefore it is unlikely to be done. 

Based on these data, the conclusions i can draw are: 

  1. AF is associated with strokes but also with dementia (and all forms of dementia)
  2. Modification of lifestyle and control of vascular risk factors is essential
  3. Anticoagulation does not increase the risk of dementia but may be protective and therefore should be started as soon as possible after the first episode of AF (esp in patients who are eligible for anticoagulation)
  4. Scanning of the brain even in the absence of a stroke may help detect micro-infarcts which may then allow you to work out if you are at a higher risk of dementia even if you dont meet current eligibility criteria for anticoagulation and perhaps may help you decide for anticoagulation rather than just accept conventional wisdom that you are ‘low risk’
  5. We desperately need studies to tell us whether anticoagulation and rate control strategies such as ablation reduce the risk of dementia and if this is proven then this may just completely change how we manage AF in the future.

I hope this is helpful. I know Dementia is a very scary subject for anyone to have to confront but I do worry that it may be another 10-20 years before we have more data and clinical guidance changes and for some patients that may be 20 years too late. This is why I wanted my viewers/readers to be aware of this troubling research.

I wish you all good health and lots of joy

Link to the video: AF and dementia

This post is also available in: हिन्दी (Hindi)