AF Linked to Increased Dementia Risk in Stroke Patients

Susan Jeffrey

March 08, 2011

March 8, 2011 — Results of a new meta-analysis provide "consistent evidence" to support an association between atrial fibrillation (AF) and increased risk for incident dementia in patients with a history of stroke. However, the association was not so clear in other populations.

"I think now we can say that we have convincing evidence that people who have atrial fibrillation are more likely to develop dementia, particularly in the stroke patient population," Phyo Kyaw Myint, MD, of the University of East Anglia in Norfolk, United Kingdom, told Medscape Medical News. "There's still a bit of uncertainty about whether this is also true for the general population," he added.

Their findings are published in the March 8 issue of Neurology.

Strict Criteria

Dr. Phyo Kyaw Myint

There is a substantial body of evidence examining the relationship between AF and dementia in the general population and among those with a history of stroke, but previous reviews have included evidence of varying quality, including retrospective studies, the study authors write. In addition, no previous studies have attempted to quantify the risk using meta-analysis.

"In this article we used fairly strict criteria," Dr. Myint said. "We only included papers that examined the prospective relationship."

For this study, they searched MEDLINE, Embase, and PsychINFO databases for published prospective studies looking at the relationship between baseline AF and incident dementia. They identified 15 such studies, including a total of 46,637 participants with a mean age of 71.7 years.

One of these studies, which reported no significant difference in Mini-Mental State Examination scores between patients with and without AF, could not be pooled and was excluded from the meta-analysis. They then calculated pooled odds ratios for the risk for dementia associated with AF and also assessed heterogeneity among studies (I2 ).

For the remaining 14 studies, meta-analysis showed that AF was associated with a significant increase in dementia overall, but with substantial heterogeneity.

When they stratified studies by whether participants had had a prior stroke, the relationship became highly significant with little heterogeneity. For other populations, the association between AF and dementia was only of borderline significance again with substantial heterogeneity.

Table. Risk for Incident Dementia With AF vs No AF

Comparison Odds Ratio (95% CI) P Heterogeneity, % (I2 )
AF vs no AF overall (14 studies) 2.0 (1.4 – 2.7) <.0001 75
AF vs no AF in stroke patients (7 studies) 2.4 (1.7 – 3.5) <.001 10
AF vs no AF in other populations (7 studies) 1.6 (1.0 – 2.7) .05 87

AF = atrial fibrillation; CI = confidence interval

One study showed a significant relationship between AF and conversion from mild cognitive impairment to dementia (odds ratio, 4.6; 95% confidence interval, 1.7 – 12.5).

A limitation of this study is that they are not able to look at specific types of dementia because most studies did not include this information, Dr. Myint noted. "Intuitively you would think that vascular dementia would be more likely to be linked to atrial fibrillation, and maybe that's why the evidence is stronger in stroke patients with atrial fibrillation compared to the general population."

To look further at the association prospectively would be costly, he added, and so it's likely that meta-analysis is the most reliable and cost-effective way to look at it with the current evidence. "But having said that, we need to know more about whether this link exists for both major types of dementia, Alzheimer's disease and vascular dementia, that will need more rigorously designed study, including neuroimaging, etc, in the future," he said.

If there is a direct link, it may be possible to reduce the risk by treating AF aggressively, he added. In the meantime, Dr. Myint concluded, "people with stroke and AF need to be aware that their risk of dementia is increased."

Dr. Myint has disclosed no relevant financial relationships. Disclosures for coauthors appear in the original article.

Neurology. 2011;76:914-922.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: