Among women aged 70-89 years, a history of cardiac disease was strongly associated with an increased risk of developing nonamnestic mild cognitive impairment in a large population-based study published online Jan. 28 in JAMA Neurology.
This association was not found among elderly men who participated in the study, said Dr. Rosebud O. Roberts of the department of Health Sciences Research and the division of epidemiology, Mayo Clinic, Rochester, Minn., and her associates.
A link between cardiovascular disease and mild cognitive impairment (MCI) and dementia has been reported in several previous studies, and association is stronger among women than men, the investigators said.
Dr. Roberts and her colleagues analyzed data from the Mayo Clinic Study of Aging, which assessed risk factors for MCI in residents of Olmsted County, Minn., who were aged 70-89 years at baseline in 2004. These study subjects underwent physical examinations and extensive cognitive testing every 15 months, and at each follow-up visit, they were diagnosed as having normal cognition, MCI, or dementia by a team of clinicians including a physician, a nurse, and a neuropsychologist.
For this study, Dr. Roberts and her associates assessed a random sample of 2,719 participants who were followed for a median of 4 years.
Cardiac disease was defined as a diagnosis of atrial fibrillation, coronary heart disease, or congestive heart failure. The physical and neuropsychological evaluations included assessment using the Clinical Dementia Rating Scale, Functional Activities Questionnaire, Short Test of Mental Status, and nine tests examining memory, executive function, language, and visuospatial skills.
A total of 1,450 of these subjects had at least one follow-up assessment; 348 of them developed incident MCI and 18 developed incident dementia. About 66% of those with MCI had amnestic MCI, 27% had nonamnestic MCI, and the remaining 7% had MCI of unknown subtype because they were followed up by telephone interview only.
In the study cohort as a whole, the incidence of MCI was higher among subjects with cardiac disease (73.2/1,000 person-years) than in those without it (62.1/1,000 person-years).
When the data were analyzed by type of MCI, cardiac disease was found to have no correlation with amnestic MCI but was linked to a doubling of the rate of nonamnestic MCI. Subjects with cardiac disease had a rate of nonamnestic MCI of 23.4/1,000 person-years, compared with 12.2/1,000 person-years for subjects without cardiac disease.
When the data were analyzed by gender, cardiac disease was found to have no correlation with nonamnestic MCI among men: The incidence of this form of cognitive impairment was similar between men with cardiac disease (22.3/1,000 person-years) and men without cardiac disease (18.9/1,000 person-years).
However, the rate of nonamnestic MCI was significantly higher among women with cardiac disease (24.3/1,000 person-years) than in those without (7.4/1,000 person-years), the investigators said (JAMA Neurol. 2013 Jan. 28 [doi:10.1001/jamaneurol.2013.607]).
This association in women remained robust in further analyses that adjusted for a history of stroke, age, education, and APOE genotype status.
For women with atrial fibrillation, the risk of nonamnestic MCI had a hazard ratio of 2.25. That risk had an HR of 2.05 for women with coronary artery disease and an HR of 2.82 for women with heart failure.
These findings suggest that cardiac disease is an independent risk factor for nonamnestic MCI in elderly women and support the hypothesis that such MCI has a vascular etiology, Dr. Roberts and her associates said.
Several possible mechanisms may contribute to the association between cardiac disease and nonamnestic MCI. Cardiac sympathetic nerve dysfunction and obstructive sleep apnea both have been linked to cardiac disease and could contribute to dementia as well. And cardiac disease could precipitate cerebral disease through the release of microemboli that travel to the brain or through hypoperfusion of the brain from impaired cardiac function, they added.
This study was limited in that the population of Olmsted County is predominantly of northern European extraction, so the findings may not be applicable to other ethnic groups.