Memory is already impaired years before a stroke occurs, declining most rapidly in people destined to suffer a fatal stroke, according to a population-based study of more than 11,000 people.

“For every year that passed [after age 50 years], those who survived a stroke lost their memory at twice the rate as controls. And for those who didn’t survive, their annual memory loss was 3 to 4 years faster than the controls,” said Maria Glymour, Sc.D., of the Harvard University School of Public Health, Boston.*

Dr. Glymour and her colleagues examined the relationship between memory and stroke using data from the national Health and Retirement Study (HRS) cohort of people aged 50 years or older.

Interviews with the study’s participants occurred every 2 years, starting in 1998. In their 10-year follow-up study, the investigators focused on 1,456 people who had survived a stroke, 364 who died from a stroke, and 9,994 who did not have a stroke. An immediate and delayed word list recall task, included in the HRS, served as the proxy for working memory among the group.

As expected, Dr. Glymour and her associates found trajectories showing age-related memory decline among all subjects after their baseline interview.

There were also baseline differences between the groups. People who had a stroke – both survivors and nonsurvivors – had significantly poorer memory function at the beginning of the follow-up period.

“More than half of the memory gap between recent stroke survivors and stroke-free adults of similar age was evident before the stroke,” she said during a press briefing at the International Stroke Conference.

The groups continued to separate as the years progressed, although individuals in all three groups had trajectories of at least some age-related memory decline during the 10-year follow-up period.

The study did not assess total cognitive ability, so Dr. Glymour could not say how many in each group already had some basis for memory impairment. But even among the stroke groups, many subjects started out with high function, she noted. “It was the rate of decline that was the key factor.”

“We are really interested to know what’s going on in the brains of these people that seems to be related to having a stroke. My guess is that there is a physiologic basis – that there might be some subclinical strokes that have already occurred that affect memory.”

Another possibility could be memory-related behavioral issues, she said. “If people are already having trouble with memory, they might have a harder time controlling their stroke risks – for example, taking their blood pressure medications.”

Dr. Steven Greenberg, director of the Hemorrhagic Stroke Research Program at Massachusetts General Hospital, Boston, and an American Heart Association spokesman, agreed.

“The cause and effect are still a little unclear. But one possibility is this idea of unnoticed strokes going on over time that cause people to have such a bad trajectory of memory loss. When they do have a symptomatic stroke, they have very little reserve left.”

The study was sponsored by the National Institutes of Health. None of the investigators or Dr. Greenberg had any relevant financial disclosures.

*This story was updated 2/2/2012.