Sept. 30 (HealthDay News) -- Just one of every eight strokes is preceded by a milder interruption of blood flow to the brain, called a transient ischemic attack (TIA), a new Canadian study shows.

And because of that, the researchers conclude, such an attack is not the crucial warning sign that physicians need.

"The clinical implication of this study is that we cannot rely on the TIA as a warning signal to tell us to intervene to prevent a stroke because it is seen before only one of every eight strokes," said Dr. Daniel G. Hackam, an assistant professor of medicine in neurology at the University of Western Ontario in Canada and lead author of a report in the Sept. 29 issue of Neurology.

"We need better risk profiles to predict a patient who will have that first stroke," Hackam said. "This study is highlighting a gap in our knowledge base. If we know a stroke is impending, we can intervene to prevent that stroke."

In the study, Hackam and his colleagues found that, of the 16,409 people diagnosed with stroke over a four-year period in Ontario hospitals, 2,032 -- or 12.4 percent of them -- had a TIA in the weeks before the stroke.

A TIA, he said, does not have enough predictive power to warrant intensive preventive measures.

"We need better tools," Hackam said. "That is really the main message of our paper."

A TIA, which some refer to as a mini-stroke, occurs when a clot briefly blocks a brain artery. Symptoms of a TIA are the same as those of a stroke -- sudden onset of weakness or numbness in an arm or leg, loss of vision or double vision, speech difficulty, dizziness, loss of balance -- but they go away, often in a few minutes. Many people ignore the symptoms, but they are clear signs of possible trouble, Hackam said.

The numbers in the new study are similar to those about TIA and subsequent stroke that have been reported for decades, said Dr. Larry B. Goldstein, director of the Duke Stroke Center. But he disagrees with Hackam's interpretation of the predictive importance of TIAs.

"They predict 10 to 15 percent of strokes," Goldstein said. "This is not a small number, so it is an opportunity to prevent stroke that you don't want to miss when it happens."

Better predictive tools are available, Hackam said. He prefers carotid ultrasound, an inexpensive way to listen to blood flow in the main artery to the brain. "It's fairly inexpensive, and I do it for everyone I see in the clinic," Hackam noted.

His patients have been referred to the stroke clinic because they have the risk factors for stroke, which include old age, smoking, high blood pressure, obesity and high cholesterol, Hackam said. Results of a carotid ultrasound test can confirm the need for treatment not only with medications to control blood pressure and blood fats but also with lifestyle changes such as more exercise, no smoking and a less-fatty diet, Hackam said.

Goldstein, though, said that detailed tests such as carotid ultrasound are not needed to recommend such measures for people who have the risk factors for stroke. Those tests tend to measure not the specific risk of stroke but the risk of all cardiovascular problems, including heart attacks, he said.

"The more tests you do, the more chance there is to make a mistake," Goldstein said. "The standard risk factors can lead to recommending basic lifestyle changes. People who don't smoke, who drink moderately, who keep their blood pressure low, are less likely to have strokes."

And anyone who experiences a TIA should report it to a doctor immediately, Goldstein said. "It only matters if you are going to do something about it," he said. "A TIA identifies someone who is at high risk of having a stroke in a short period of time."

SOURCES: Daniel G. Hackam, M.D., Ph.D., assistant professor, neurology, clinical pharmacology and epidemiology/biostatistics, Stroke Prevention and Atherosclerosis Research Centre, Robarts Research Institute, University of Western Ontario, London, Ontario, Canada; Larry B. Goldstein, M.D., director, Duke Stroke Center, Duke University, Durham, N.C.; Sept. 29, 2009, Neurology