A retrospective national Swedish registry study by Leif Friberg and Mårten Rosenqvist (Karolinska Institutet, Stockholm, Sweden) has found that the risk of dementia is higher in atrial fibrillation (AF) patients without oral anticoagulant (OAC) treatment. The paper was presented at Europe AF (7-8 November 2017, London, UK) and published in the European Heart Journal. Leif Friberg spoke to Cardiac Rhythm News about the study outcomes.
Although AF has previously been observed to be independently associated with dementia, any clear data on the protective qualities of OAC treatment for dementia have not previously been available. The study by Friberg and Rosenqvist compared the incidence of dementia developed in AF patients with and without ongoing OAC treatment.
Retrospective registry data from the Swedish Patient register and the Swedish Dispensed Drug register was collected on patients diagnosed with and treated for AF in Sweden between 2006-2014. This included 444,106 patients, and over 1.5 million patient-years.
Friberg and Rosenqvist found that OAC treatment at baseline was associated with a 29% reduced risk of dementia (hazard ratio 0.71, 95% CI 0.68-0.74), and in fact, absence of OAC treatment was among the strongest predictors for dementia along with age, Parkinson’s Disease, and alcohol abuse.
Although the study also sought to investigate and compare vitamin K antagonist OAC warfarin with novel oral anticoagulants (NOACs), direct comparison between the two medical therapies showed no difference (hazard ratio 0.97, 95% CI 0.67-1.40).
Cardiac Rhythm News spoke to study author Leif Friberg about the outcomes and their implications:
What are the most important things to note from this study?
“To me, the most important thing is that we may have a possibility to prevent some patients from developing dementia. Not all, of course, since AF is just one of many reasons why you can get dementia. We have to find out if we should start anticoagulant treatment earlier than the current guidelines tell us to do in order to prevent dementia. A 50-year-old person with AF but no risk factors would have to wait 25 years until she or he turns ≥75 years, receiving two more points on the CHA2DS2-VASc scheme and thus indication for treatment. Yet, we know that performance on cognitive tests decline with duration of AF. There are many new studies needed to ascertain the optimal time for initiation of anticoagulant therapy where you weigh together risks of dementia and stroke against bleeding risks and costs for individual and society.”
Why do you think you were unable to observe that risk of dementia was lower with NOACs than with warfarin?
“Warfarin is a very good drug if it is managed correctly, in which case both bleedings and thrombotic events are minimised. Warfarin treatment in Sweden is of very high standard in comparison with almost any other country in the world. This has been shown in clinical trials as well as in general clinical practice. The mean time in therapeutic range is consistently around 73-75%. This means that warfarin in Sweden is as tough competition as the NOAC can meet anywhere. It would be very interesting to see what the results would be if the study could be reproduced in some other part of the world.”
How do you hope these findings will affect treatment plans for AF?
“For the moment we should not change our treatment practices. We need more confirmatory information, and I hope our report will stimulate such studies. Demographic studies indicate that dementia will increase rapidly with aging populations all over the world. This makes it urgent to produce the evidence we need to change current guideline recommendations.”
What are the next steps to investigate the benefits of OACs in relation to dementia?
“Firstly, we must try to reproduce our findings in other populations. Secondly, we need to reassess risk stratification tools with recognition of dementia as a complication of AF.
Finally, the finding of increased risks for all types of dementia, not just vascular, in AF is interesting. Is oral anticoagulation equally protective for dementia of non-embolic type? Is AF just an innocent bystander and marker for dementia risk as well as for stroke risk—as some pacemaker studies showing poor temporal relationship between AF episodes and stroke seem to indicate? If so, could individuals with high risk scores but without diagnosed AF also benefit from anticoagulant treatment?”
In the published paper, Friberg and Rosenqvist consider the benefits of early initiation of OAC therapy, as they also noted “an interaction between OAC and the time since the first diagnosis of AF suggesting that the benefit of treatment may be larger if initiated early rather than late”, but acknowledge that more research is needed to weigh up the risk factors.
They also highlighted the limitations faced by further research, as randomised placebo trials would be unethical. “It is not possible to give placebo to AF patients and then wait for dementia or stroke to occur”, the authors write, but that information provided by population-wide health databases used for retrospective studies provide a “second best” alternative.