Atrial fibrillation entails an increased risk of dementia, even among people who never have a stroke, according to a prospective population-based cohort study published November 1, 2015, in JAMA Neurology. After adjusting for confounders such as age, sex, and cardiovascular risk factors, the investigators found that the 34% increased risk of dementia in individuals who develop atrial fibrillation, relative to those who do not, was essentially unchanged in people with or without a history of stroke.
The association between atrial fibrillation and risk of dementia in this study, which included more than 81,000 person-years of follow-up, was stronger in participants whose rhythm disturbance had an onset before age 67, compared with onset at a later age. Moreover, a longer duration of atrial fibrillation in younger individuals was associated with an increasing risk of dementia, implying a dose response, according to the study authors, who were led by M. Arfan Ikram, MD, PhD, of the Departments of Epidemiology and Radiology at Erasmus Medical Center in Rotterdam, the Netherlands.
This association is consistent with other risk factors linked with dementia, such as hypertension, hypercholesterolemia, and obesity, for which the association strengthens with earlier onset and longer duration, according to the authors. The clinical implication is that control of any or all of these risk factors might reduce the risk of subsequent dementia, even though the authors acknowledged that no protective effect from the control of atrial fibrillation could be derived from this study.
An Analysis of the Rotterdam Study
This dementia analysis was embedded in the Rotterdam Study, which investigates chronic diseases in the elderly and began enrolling individuals age 55 or older in 1989. Participants were screened for rhythm abnormalities and dementia at baseline. Follow-up examinations have been conducted every three to four years. Participants with atrial flutter were grouped with those who had atrial fibrillation because, according to the authors, the risk factors and consequences of these arrhythmias are similar.
Of the 6,514 participants evaluated, 318 had atrial fibrillation at baseline, which was classified as prevalent atrial fibrillation, and 6,196 did not. Of those without atrial fibrillation, 723 (11.7%) developed atrial fibrillation during follow-up, which was classified as incident atrial fibrillation. The relative rates of dementia were assessed for prevalent and incident atrial fibrillation, relative to the absence of atrial fibrillation.
A history of stroke at baseline was an exclusion factor, but the analysis included participants who had a stroke during the course of follow-up. Stroke occurred in approximately 3% of the study population. The relative risk of dementia was then calculated for participants with and without a history of stroke. This study also analyzed the association between atrial fibrillation and risk of Alzheimer’s disease as a specific subset of dementia.
Association With Alzheimer’s Disease Was Not Significant
Among cases of prevalent atrial fibrillation, the association of atrial fibrillation with dementia reached significance for the study population overall (hazard ratio [HR], 1.34) and for participation censored for stroke (HR, 1.35). Among cases of incident atrial fibrillation, the same patterns were observed, but fell short of statistical significance (HR, 1.13 including stroke and HR, 1.14 censored for stroke).
For the risk of Alzheimer’s disease in participants with prevalent atrial fibrillation, the trend for an association, although not significant, was strong and similar with or without the inclusion of stroke cases (HR, 1.29 including stroke and HR, 1.28 censored for stroke). Among patients with incident atrial fibrillation, the same nonsignificant trends were seen.
When the analysis was stratified by age, strong but nonsignificant trends were observed for participants age 67 or older, particularly for those with prevalent atrial fibrillation (HR, 1.28). For people younger than 67, the risk of dementia also fell short of significance for people with prevalent (HR, 1.91) and incident atrial fibrillation (HR, 1.81), but the risk of dementia climbed incrementally with longer duration of follow-up. The strongest association between atrial fibrillation and risk of dementia was in the younger group with incident atrial fibrillation followed more than 12 years. In these participants, the dementia risk was more than threefold higher, relative to those without atrial fibrillation (HR, 3.30). The correlation between duration of atrial fibrillation and dementia risk was highly significant.
Previous Findings Were Inconsistent
Previous studies have looked for an association between atrial fibrillation and dementia, but findings have been inconsistent. Methodologic problems, such as small samples sizes, may be the reason for this inconsistency, according to the authors of this most recent study. Although this study has the advantages of a prospective design, a larger sample size, and adjustment for potential confounders, the authors acknowledged potential weaknesses. In particular, they expressed concern about their inability to control for the impact of treatment of atrial fibrillation on dementia risk.