The preliminary information was gathered from a study of 34 academic medical institutions.
Clinical studies have shown that the clot-buster tPA (tissue plasminogen activator) can reduce the debilitating and crippling effects of stroke. However, it must be given within three hours of symptom onset.
In 2000, stroke experts comprising the Brain Attack Coalition (BAC) suggested that defining and establishing primary stroke centers should improve patient care and outcomes. They listed 11 major criteria that every primary stroke center should meet. The Coalition consists of 13 health organizations, including the American Stroke Association, a division of the American Heart Association; and the American Academy of Neurology.
However, little evidence existed in medical literature at the time to indicate which specific factors improved patient care at stroke centers, said senior author S. Claiborne Johnston, M.D., director of Stroke Services at the University of California San Francisco Medical Center.
"We wanted to see how well the Brain Attack Coalition recommendations actually predicted patient outcome and better quality of care in stroke centers," Johnston said.
"We found that seven of 11 recommended criteria increased the use of tPA. The more of these seven criteria that a center followed, the greater the use of the drug," he said.
Having written procedures for treating stroke was the strongest predictor. Three other factors significantly influenced use of tPA: integrating emergency medical personnel into the treatment effort; an emergency room staff well trained in recognizing stroke; and continuing medical education in stroke for all members of the stroke team.
Three more criteria showed a positive trend in determining the use of tPA. They were: a stroke team on call around the clock, seven days a week; a formal stroke unit that provides specialized monitoring and care; and the rapid availability of CT scans to assess whether the stroke was caused by a blood clot or a hemorrhaging blood vessel. While these three didn't have a statistically significant impact individually, they were important contributors to the overall goal.
The CT scan is critical, since giving tPA to patients with hemorrhagic stroke increases the risk of death or complications because it slows or prevents the clotting needed to seal a ruptured vessel.
At hospitals that met all seven criteria, patients with ischemic strokes – those caused by blood clots – had a 4.7 times greater chance of getting tPA.
"The seven criteria influence how rapidly treatment can begin," Johnston said.
Johnston and his colleagues focused on the impact of the 11 BAC criteria on acute stroke treatment with tPA. The researchers sent a questionnaire to stroke specialists at 34 academic medical centers. The team compared the answers they received to the BAC criteria to assess which of the 11 standards each medical center met. They then statistically compared the number of people with ischemic strokes treated with tPA at each facility with the 11 recommended criteria.
"Although patient outcomes and whether tPA was being used appropriately were not measured, studies like this can help hospitals target procedures that can be improved," said Larry B. Goldstein, M.D., Chair of the American Stroke Association Advisory Committee.
Of 16,853 stroke patients, 399 (2.4 percent) received tPA.
Four of the 11 criteria did not seem to affect tPA use. But three seemed to fail. There were only a small number of centers that did not meet three of the criteria, Johnston said. For example, 82 percent of the medical centers met the medical laboratory criteria. "So, with only 18 percent not meeting that criterion, we couldn't really assess in a study of 34 hospitals if that made a difference," he said.
The fourth nonpredictive criterion proved more puzzling. It calls for a system to track the number and type of stroke patients seen, their treatment and their outcomes. Forty-five percent of the medical centers in the study had such a system.
"It looks like it should make a difference but it didn't predict tPA use; it didn't predict anything," Johnston said. "So we are not sure what that means."
However, the ability of such a system to impact care may only be seen over time.
He cautioned against dismissing the value of the four nonpredictive criteria at this time. His group looked only at a small number of academic medical centers. Johnston called for a much larger study encompassing different types of medical facilities, including community hospitals.
"Our study was not statistically powerful enough to fully assess the criteria," he said. "We think based on logic and on other studies that the four factors that weren't significant predictors of tPA use in our study actually could be important in improving the quality of care."
Co-authors are Vanja C. Douglas, B.S.; David C. Tong, M.D.; and Shoujun Zhao, M.D. Editor's Note: Based on the Brain Attack Coalition recommendations and recently developed stroke performance measures, the American Stroke Association and the Joint Commission on Accreditation of Healthcare Organizations is launching the Primary Stroke Center Certification program. For more information visit strokeassociation.org.
NR04-1201 (ISC04/Johnston)
Note: Presentation time is 3:12 p.m. PST Friday, Feb. 6, 2004