New online tool designed to predict risk of stroke recurrence

Up to a quarter of strokes are second strokes, and people who suffer two strokes in quick succession are more likely to die or be severely disabled. It is a serious problem affecting about 5 percent of stroke patients, but because doctors don't know which of their patients are at highest risk of recurrence, they typically prescribe medications -- some with troubling side effects -- for all patients. In a study published online today in Neurology,  Boston researchers report they have developed a free, Web-based tool designed to predict which people will go on to have a second stroke within 90 days, based on information gathered when they had their first stroke.

Dr. A. Gregory Sorensen and colleagues from Massachusetts General Hospital, the Broad Institute, and the National Institute of Neurological Disorders and Stroke analyzed the medical records of almost 1,500 patients who were admitted to the hospital within 72 hours of suffering ischemic strokes, which are caused by a blockage of blood flow in the brain. After 90 days, 806 patients were reached by phone. Of those patients, 60 had suffered second strokes, and half of those happened in the first two weeks. The researchers calculated the risk of recurrence among these 806 patients to be 2.6 percent at two weeks and 6 percent at 90 days.

After reviewing clinical records and brain scans from these patients' first strokes, the researchers developed a tool that took into account a number of risk factors, including a history of many small strokes in the month before the first major stroke; multiple areas of brain tissue death that occurred at different ages; obstructed small arteries; and large-artery
atherosclerosis. People who had four or more risk factors had a risk of a second stroke that was 40 times higher than people with no risk factors. More than 96 percent of second-stroke patients had at least one risk factor.

"We've suspected some patients are much more likely to have stroke, but ... to treat all 100 percent to benefit those 5 percent sometimes has a downside," Sorensen said in an interview. "This is a group of patients who have already been seen once by a doctor. It's kind of a shame we haven't figured out how to stop strokes once they've had that first stroke."

Long-term risk factors for stroke such as smoking, diabetes, or high blood pressure were not associated with a higher risk of stroke recurrence within 90 days, confirming previous research. Sorensen compared the person who continues to have many small strokes, also called transient ischemic attacks, to a person with chest pain that may signal an impending heart attack. These are the patients who need to be followed closely and urged to take their medications, including blood thinners that may limit activities, he said.

The study's results need to be validated in a larger, multi-center trial, both the researchers and the authors of an accompanying editorial said. Dr. Tobias Kurth and Dr. Christian Stapf of the Institut National de la Sante et de la Recherche Medicale in Paris also noted that follow-up information was available for only about half of the patients and that the number of strokes was relatively small, possibly limiting the power of the prediction tool.

Despite these methodological considerations, they write, "A refined forecast may guide stroke physicians to the good weather zones, allow early identification of high-risk patients, lead to adjusted treatment  plans in the postacute period, and help improve stroke recurrence in specific patient subgroups at risk."