[Atrial fibrillation and cognitive function].

Atrial fibrillation (AF), which prevalence increases with age, is a growing public health problem and a well known risk factor for stroke. On the other hand, dementia is one of the most important neurological disorders in the elderly, and with aging of the population in developed countries, the number of demented patients will increase in absence of prevention. In the past decade, several vascular risk factors (hypertension, obesity and metabolic syndrome, hypercholesterolemia) have been found, with various degree of evidence, to be associated with vascular dementia but also, surprisingly, with Alzheimer's disease. This review is devoted to the links between atrial fibrillation, cognitive decline and dementia. Globally, transversal studies showed a significant association between atrial fibrillation, cognitive decline and dementia. However, these studies are particularly sensitive to various biases. In this context, recent longitudinal studies of higher level of evidence have been conducted to assess the link between AF and dementia. One study disclosed a high incidence of dementia among patients suffering from atrial fibrillation during a 4.6 years follow-up. Similarly another study showed that atrial fibrillation was significantly associated with conversion from mild cognitive impairment to dementia during a 3 years follow-up. Nevertheless two other longitudinal studies did not find any significant association between AF and dementia, but this discrepancy should be interpreted taking into account that the comparability of all these studies is moderate because they were using different methodologies (population, cognitive testing, and mean follow-up). Possible explanatory mechanisms for the association between AF and the risk of dementia are proposed, such as thrombo-embolic ischemic damage and cerebral hypo perfusion due to fluctuations in the cardiac output. Thus, there is some evidence that FA could be associated with cognitive decline and dementia but this link should be supported by more powerful long term longitudinal studies.