- Split View
-
Views
-
Cite
Cite
Poggesi Anna, Cognitive decline in patients with atrial fibrillation: can we do something about it?, Age and Ageing, Volume 45, Issue 6, 2 November 2016, Pages 749–751, https://doi.org/10.1093/ageing/afw158
- Share Icon Share
Cognitive impairment and atrial fibrillation (AF) are very common conditions, especially in elderly subjects. Available evidence suggests that AF may increase the risk of dementias, but there is still a need to better elucidate the exact mechanisms of the links between these two conditions, and the topic is matter of investigations [1]. The potential association between AF and dementia has also important therapeutic implications. Thromboprophylaxis with oral anticoagulation has been proven effective in reducing stroke risk, although benefits have to be balanced against the risk of bleeding. In this scenario, Moffitt et al. [2] performed a systematic review with the aim of assessing the possible effect on cognition and/or dementia of any treatments designed to reduce cardioembolism in AF. Treatments under investigation mainly focused on anticoagulation and antiplatelet therapy but also included mechanical interventions. Controls were “placebo” or different kind of treatments as for example anticoagulant versus antiplatelet drugs. Considered outcomes were any quantitative measure of cognition or clinical diagnosis of dementia with pre-specified analyses set to compare longitudinal temporal change in cognitive scores or rates of incident dementia/cognitive decline between treatment arms. Nineteen studies were included in the review (n = 15,876 patients) out of which ten could be used for quantitative analyses (n = 8,360 patients). Prospective data were suggestive of a modest protective cognitive effect from anticoagulation in AF in the longer term but no decrease in rates of incident dementia was found. Included studies did not contain sufficient information to allow for any of the pre-specified subgroup analyses, and there was substantial risk of bias. Results of this elegant and nicely conducted review might appear irrelevant at a first sight, as no clear effect on cognitive outcomes of anticoagulation seems to appear, but indeed they offer an incentive for some thoughts and considerations.
Few studies on AF patients assessed cognitive outcomes. Among randomised clinical trials, only one had cognitive data suited for the proposed analysis. The Birmingham Atrial Fibrillation Treatment of the Aged Study randomised 973 AF patients aged >75 years with a CHA2DS2-VASc of at least 2 to warfarin (INR range 2–3) versus aspirin (75 mg daily). The study was conducted between 2001 and 2004, at a time when there still was uncertainty over the optimum treatment of elderly people with AF, and main results confirmed the superiority of warfarin versus aspirin on stroke prevention [3]. Patients were also evaluated by means of MMSE at 9, 21 and 33 months form enrolment. No major differences in cognitive functions and decline were found between the two treatment arms, only a non-statistically significant trend toward decreased cognitive decline at 33 months within the warfarin group emerged [4]. Of note, enrolled patients had normal cognitive functions and demented patients were excluded, more than 80% did not have a history of stroke at baseline, and those suffering from stroke during follow-up were excluded from further analyses related to cognitive decline. Such “negative” result should thus be interpreted with caution, as the main mechanism on which treatment with anticoagulants has an effect, i.e. embolic insult to the brain, was not taken into account. Nowadays, it would be impossible to replicate such a trial, as no doubts remain on the benefit of anticoagulation in AF patients. AF is the main cause of embolic strokes, accounts for nearly 20% of all stroke events, and increases stroke risk up to 5 times. AF is indeed considered one of the independent determinants of post-stroke dementia. The presence of AF may be directly responsible for blood stasis in the left atrium, altering blood flow and thus favoring the formation of thrombus, which may finally detach and enter the circulation as emboli. When the embolus reaches the cerebral circulation stroke may happen, and, by means of the same mechanism, may recur. The type of mechanism subsiding post-stroke cognitive decline in AF patients may thus be multiple infarcts or strategic infarcts caused by a single strategically located cortico-subcortical lesion. An embolic insult to the brain may also be clinically silent, and silent infarcts are well-known lesions associated with cognitive impairment and dementia [1].
The assumption of the current review deals with data describing a possible cognitive effect of treatments to reduce cardioembolism in AF, which certainly relates to stroke (and probably silent infarcts) prevention. At present, there are little doubts about the fact that by preventing stroke, it is possible to prevent cognitive decline, at least post-stroke cognitive impairment. Interestingly, there is some evidence pointing to an effect of the quality of anticoagulation with warfarin (i.e. time in therapeutical range), on cognitive functions [5, 6]. A number of non-vitamin K antagonist oral anticoagulants are now available for stroke prevention in nonvalvular AF, including dabigatran (direct thrombin inhibitor), rivaroxaban, apixaban and edoxaban (factor Xa inhibitors). These anticoagulants have been proven effective as warfarin in stroke prevention, but associated with a decreased risk of intracranial haemorrhage [7]. For this reason, and probably for the fact that they mitigate the challenges of time in the therapeutic range, one may speculate that newer oral anticoagulants might have a different impact on cognition compared with warfarin, but such differences still need to be explored.
Apart from cardioembolism, either manifesting with clinical overt stroke or silent infarcts, other mechanisms are hypothesised to explain the link between AF and cognition. Available evidence suggests that dementia, and its different subtypes, may be related to AF, even in stroke-free patients [1]. Among putative mechanisms, brain hypoperfusion related to reduced cardiac output and to the beat-to-beat variability in the length of the cardiac cycle, and pro-thrombotic and inflammatory state related to AF are considered important. Beside anticoagulation, other pharmacological and non-pharmacological interventions targeting rate and rhythm control might have a positive effect on cognition through the improvement of cerebral perfusion. Circulating biomarkers, particularly those related to endothelial function and haemostatic functions, have been proposed as useful tools for a better comprehension of AF pathophysiology [8]. Interestingly, the same circulating markers seem to be related with dementia and with small vessel disease (SVD) [9, 10].
When dealing with dementia, it is important to consider the underlying pathological brain changes. The most frequent type of dementia is Alzheimer's disease, followed by vascular dementia, a category encompassing different subtypes among which cerebral SVD and multi-infarct dementia are the most common. Available evidence suggests an association between AF and all the above-mentioned subtypes of dementia [1]. AF and dementias share common vascular risk factors. As these same risk factors seem to play an important role in the progression of cognitive impairment, from mild cognitive impairment to dementia, it is plausible that treating risk factors might have a favourable effect on the onset and progression of cognitive decline [11].
Finally, neuroimaging, particularly magnetic resonance imaging, has led to increased detection of “asymptomatic” brain changes, mainly those related to SVD such as microbleeds, which also represent the pathologic substrate of intracranial haemorrhage. The clinical dilemma in AF patients is now which kind of preventive therapy should be used in patients with previous stroke and microbleeds [12]. In this setting, two large prospective multicenter observational studies are under way: CROMIS-2 (Clinical Relevance of Microbleeds in Stroke) in the United Kingdom, and HERO (Intracerebral Hemorrhage Due to Oral Anticoagulants: Prediction of the Risk by Magnetic Resonance) in Spain [13].
This month's Age and Ageing adds another piece of information on the relationship between AF and cognitive decline and surely stimulates more research. While waiting for more data to come, thromboprophylaxis with oral anticoagulation should be used in patients with AF, as recommended by guidelines for patients with moderate/high risk of stroke, and the best vascular risk factor control should be achieved [14]. For the future, AF patients should be prospectively evaluated for cognitive and functional decline, and the effects of different kind of treatments, anticoagulants (vitamin K versus newer anticoagulants), rhythm and rate control, and non-pharmacological interventions such as left atrial appendage occlusion should be assessed on long-term cognitive function, in parallel to stroke prevention.
There is a relationship between AF and cognitive impairment/dementia.
AF is a risk factor for cognitive impairment/dementia in patients who suffered a stroke but also in patients who did not.
Treatments reducing cardioembolism prevent cognitive impairment/dementia related to stroke (or silent infarcts).
hypoperfusion, endothelial dysfunction, pro-inflammatory and pro-thrombotic states related to AF are other contributing factors.
Comments