New Tool Improves Prediction Of Stroke Risk

The lead author of the study is S. Claiborne Johnston, MD, a neurologist at UCSF Medical Center and associate professor in the UCSF School of Medicine.

About 240,000 TIAs are diagnosed every year in the United States and about 70,000 in the United Kingdom. A TIA is a serious condition caused by temporary reduction in blood and oxygen supply to part of the brain that can cause acute symptoms such as loss of vision, leg and arm weakness, slurring of speech and loss of consciousness. A TIA is sometimes called a mini-stroke because symptoms are the same as a stroke. The severe symptoms of TIA normally last up to a few hours, and all symptoms disappear within 24 hours. Because of the short duration, many people who experience TIAs do not go to see a doctor, or if they do, they are not always treated with urgency.

Recent studies have shown that 4--20 percent of these patients will have a stroke within 90 days of a TIA, and half within the first two days. Identification of those at highest and lowest risk of stroke after a TIA would help physicians decide who needs to come into the hospital right away and who can be treated as an outpatient, according to Johnston. However, it has not been possible to estimate individual risk with sufficient accuracy to guide these clinical decisions.

Two current prognostic scores have been proposed: the California score, which estimates the risk of stroke within 90 days after presentation of TIA, and the ABCD score, which estimates the risk within seven days. However, Johnston emphasized, estimating the risk of stroke within two days after TIA is often most relevant for decisions about necessity of urgent evaluation and observation.

Johnston and colleagues tested the two existing prognostic scores in large independent groups from different populations from the U.S. and the U.K., comparing predictions of stroke risk at two, seven and 90 days after TIA. Since both previous prognostic scores predicted stroke risk reliably across a wide range of populations and contained several similar components, the researchers generated a new unified score for optimum two-day risk. The new score, ABCD-squared, was a more accurate predictor of risk of stroke than either of the two previous scores, creating a single standard for use in clinical care and public education.

"Right now, there is no consensus as to who gets admitted to the hospital or has other medical intervention after a TIA," Johnston said. "We are hoping this will be a useful tool for emergency department physicians as well as those in clinics." He added that this would also be useful for patients, especially those with diabetes and the elderly.

This new scoring method assigns points to each patient on the basis of five clinical features: high blood pressure, unilateral weakness, speech impairment without weakness, diabetes and being older. The sum of the points delineate groups at high, moderate, or low-risk.

Johnston noted that while intervention such as hospitalization is costly, in the long run it could result in a cost savings because follow-up to a stroke usually involves lengthy and costly inpatient rehabilitation. Also, if the patient does have a stroke while he or she is being observed, the stroke can be treated with clot-busting drugs immediately.

In the "Lancet" report, the authors conclude: "Findings of many studies have confirmed that the short-term risk of stroke is raised after TIA. The ABCD2 score allows identification of groups at especially high risk, in whom aggressive evaluation and urgent intervention is clearly justified."

Along with Johnston, researchers included Mai Nguyen-Huynh, MD and Jacob Elkins, MD, assistant adjunct professors at UCSF; Peter Rothwell, MD, and Matthew Giles, MD, Department of Clinical Neurology, Radcliffe Infirmary, Oxford England; and Allan Bernstein, MD, and Stephen Sidney, MD, Division of Research, Kaiser-Permanente, Northern California.

The research was funded by the NIH National Institute of Neurological Disorders and Stroke, the American Heart Association, UK Stroke Association, BUPA Foundation, and the UK Medical Research Council.