News from the American Stroke Association's International Stroke Conference 2008, New Orleans, La, February 20-22, 2008

About 5000 attendees gathered in New Orleans to hear the latest research in stroke and cerebrovascular disease prevention, risk factor burden, and treatment at a two and a half day conference featuring more than 850 abstract presentations and lectures. A selection of some of the breaking news from that conference follows.

Gender differences in stroke: Women stroke sufferers get heavier in middle age, have greater disability than men in old age

The incidence of stroke over the past 2 decades has tripled in middle-aged women in the United States, and this increase is accompanied by significant increases in waist circumference and obesity, said Amytis Towfighi, MD.

"In the United States, there are 1 million stroke survivors 20 to 59 years old, and these people are understudied," she said. Most stroke epidemiologic studies focus on the 60-and-older age group.

Her findings were derived from an examination of 2 National Health and Nutrition Examination Surveys (NHANES), 1 conducted from 1988 to 1994 and the other from 1999 to 2004, looking specifically at stroke prevalence, medical histories, and biomarkers among 35- to 54-year-old men and women.

Stroke prevalence in this age group increased from 0.6% to 1.8% among women from the first survey to the second survey, whereas it remained essentially flat among similarly aged men (0.9% during the 1988-1994 survey and 1.0% during the 1999-2004 survey).

The percentage of middle-aged women with abdominal obesity, defined as a waist circumference exceeding 88 cm, increased from 47% to 59% over the 2 survey periods. Among middle-aged men, the prevalence of abdominal obesity increased from 29% to 41%.

"Women in NHANES 1999-2004 had an average waist circumference of nearly 4 cm more than women in the earlier study," said Dr. Towfighi, assistant professor, department of neurology, University of Southern California, Los Angeles.

Mean body mass index increased in both genders across the surveys—from 27.0 kg/m² to 28.7 kg/m² among the middle-aged women, and from 27.2 kg/m² to 28.4 kg/m² among the middle-aged men.

Many levels of key biomarkers and traditional vascular risk conditions were either unchanged or showed slight improvement among the women from the earlier to the later survey, possibly due to increased use of vascular risk-reduction pharmacotherapy, said Dr Towfighi. For example, levels of high-density lipoprotein cholesterol and homocysteine actually improved over this time. Use of antihypertensive medications increased from 8.9% of 35-to-54-year-old women in the earlier survey to 14.8% in the later survey, and the percentage taking drugs to treat dyslipdemia increased from 1.4% to 4.0%.

However, glycemic markers and luteinizing hormone levels did increase significantly in women over the 2 surveys, which supports concerns over the untoward role of a growing obesity epidemic, she said.

Waist circumference is one of the components of the metabolic syndrome, and the findings suggest that "the effects of the metabolic syndrome on stroke risk is greater in women vs men," she said.

Other research shows that women with stroke suffer greater disability than men, reported F. Rodica Petrea, MD, in the department of neurology at Boston University School of Medicine.

Using participants from the Framingham Heart Study, researchers compared sex-specific stroke incidence, age at first stroke, stroke severity, and post-stroke disability, including institutionalization. With follow-up of up to 50 years, they found that the average age at the initial stroke was 76.1 years for women and 71.1 years for men. "At age 85, women have a higher incidence of stroke," she said.

Thirty-five percent of women and 10% of men were placed in nursing homes 3 to 6 months post-stroke, which corresponded to a 4.65-fold increase after adjustment. Men who suffered a stroke were more than 3 times as likely to be married, she noted, which may have contributed to their lower rate of institutionalization.

The Reasons for Geographic and Racial Differences in Stroke (REGARDS) study is a large-scale, population-based study of stroke in the United States. Researchers intend to follow 30,000 individuals to better understand the underlying causes for the geographic and racial differences in stroke mortality. The following 2 presentations are drawn from that study.

Stroke patients often unaware of having vascular risk factors

Vascular risk factors are often unrecognized and undertreated in patients who have suffered a stroke or transient ischemic attack (TIA), according to recent data from REGARDS.

"Stroke patients had a high prevalence of undetected vascular risk factors and poor control when treated for them," said David Brenner, MD, lead investigator of the study and assistant professor of neurology at the University of Alabama, Birmingham. "Rates of hypertension unawareness, diabetes unawareness, and uncontrolled hypertension despite treatment were significantly higher in stroke patients than in stroke-free controls."

In REGARDS, 30,201 black and white adults 45 years or older were recruited by telephone. Approximately 30% of the subjects were recruited from states that constitute the "Stroke Belt," the rest were from the contiguous states. Participants were asked if a health care professional ever told them that they had a stroke or TIA, to which 2830 answered "yes." All participants were then visited at home and had their blood pressures, fasting and nonfasting glucose, and lipids measured.

Among stroke participants aware of hypertension, 7.6% were untreated and 33.3% were uncontrolled despite treatment (9.1% had stage 2 hypertension despite treatment). For those aware of diabetes, 11.7% were untreated and 40.8% were uncontrolled despite treatment. For those aware of dyslipidemia, 23.7% were untreated and 53.8% were uncontrolled despite treatment.

Some 18.7% of the stroke/TIA cohort with hypertension denied having this condition, compared with only 13.5% (P = .003) of those without stroke or TIA (controls). Among those with stage 2 hypertension, 4.4% of the stroke/TIA subset denied having it, twice the percentage of the controls (P = .006). Denial of the presence of diabetes was also more common in the stroke/TIA subset than in the controls (4.2% vs 3.2%, respectively; P = .026). Almost 60% (59.1%) of participants with stroke or TIA denied having dyslipidemia despite having low-density lipoprotein cholesterol levels greater than 100 mg/dL, although this percentage was less than that of the controls (65.5%).

Among the subjects aware of their condition, significantly fewer in the stroke/TIA subpopulation were receiving treatment for hypertension, diabetes, or dyslipidemia than those without stroke or TIA (P < .001 for all).

Further, among the subjects who were aware and treated for stage 1 or stage 2 hypertension, fewer of the stroke/TIA subset was treated to control than those without stroke or TIA.

"Stroke survivors and their physicians are not aggressively recognizing and controlling risk factors to prevent subsequent strokes," said Dr Brenner.

Obese participants and African Americans were more likely to be unaware of hypertension and had poorer control when treated for it. There were no substantial geographical differences in awareness, treatment, and control of any risk factor.

More data from REGARDS: Stroke risk factors speed cognitive decline

Even among those who have not suffered a stroke, persons who have vascular risk factors experience a faster rate of cognitive decline than persons without stroke risk factors, said George Howard, PhD.

The decline in cognitive function was twice as rapid in persons at high risk of stroke based on their Framingham Stroke Risk Function (FSRF) score compared with persons considered to be at average risk, he said.

The FSRF estimates the 10-year chance of a stroke based on risk factors. The components of the FSRF are systolic blood pressure, antihypertensive medication use, diabetes, coronary heart disease (CHD), cigarette smoking, atrial fibrillation, and left ventricular hypertrophy (LVH).

A simple 6-item cognitive test was administered yearly by telephone to 17,626 randomly selected stroke-free participants (African Americans and whites) as part of REGARDS. The test was derived from the Mini-Mental State Exam. Scores on the test range from 0 to 6. All subjects had at least 2 cognitive assessments. The mean age of the participants was 65.9 years, and their average systolic blood pressure was 127.9 mm Hg. Fifty-six percent had hypertension, 19.3% had diabetes, 21.9% had established CHD, 6.5% had LVH, and 13.1% were smokers.

The differences in the annual rate of decline of the mean cognitive score associated with differences in stroke risk factors was assessed. Cognitive assessments were performed until the last follow-up or subjects had a stroke (if a person had a stroke, all following cognitive assessments were censored).

The mean FSRF among participants was about 10%; about 15% have a risk of stroke of 20% or more over the following 10 years, and about 5% have a 30% or greater risk.

In subjects with a 0% risk of stroke over the following 10 years (FSRF of 0), the annual rate of decline in the cognitive score was 0.02 points. Subjects with a FSRF that indicated an average risk of stroke (10.5% risk), had an annual rate of decline approaching 0.6 points. Those with a 31.5% risk had an annual rate of decline in excess of 0.11 points.

"Someone with an FSRF of 30 lost cognitive ability more than twice as fast as someone with an FSRF of 10," said Dr Howard, professor and chair of biostatistics at the University of Alabama in Birmingham.

There were 3 components of the FSRF that were responsible for cognitive decline, one being systolic blood pressure. A 31-mm Hg higher systolic blood pressure was associated with a 29% increase in the rate of cognitive decline.

"It's the blood pressure that makes the difference, not taking the [antihypertensive] medication," he said. "It suggests that if you're hypertensive and controlled, then it's not strongly associated with cognitive decline, but if your blood pressure levels are high, then it's associated with the decline. To prove this, though, would take a clinical trial."

The other 2 components associated with cognitive decline were diabetes and LVH, with LVH being a "particularly potent" risk factor for cognitive decline, he said. Diabetes was associated with a 56% increase in the rate of cognitive decline and LVH was associated with a 60% increased rate of cognitive decline.

"Surprisingly, cigarette smoking was not associated with cognitive decline," he added.