RSNA News - September 2004

Radiologists Play More Prominent Role in Stroke Triage

Radiologists need to be comfortable in quickly obtaining and interpreting the information needed for stroke management decisions.
— Vincent P. Mathews, M.D.

Vincent P. Mathews, M.D.

Northwest Radiology Network

Advanced CT and MR imaging technology, along with new clinical research, has made the radiologist an integral part of the medical team, diagnosing and treating stroke—the third leading cause of death and the leading cause of serious, long-term disability in the United States.

A growing number of medical authorities want to expand the radiologist's role on the stroke team. Two radiologists who will teach courses on the subject at RSNA 2004 agree.

Vincent P. Mathews, M.D., and Howard A. Rowley, M.D., point out that sophisticated imaging can improve the selection of patients eligible for thrombolysis, as well as expand the window of treatment opportunity beyond the current tissue plasminogen activator (tPA) limit of three hours. tPA, a safe and effective clot-buster that has been available for more than a decade, is being used in fewer than four percent of the 750,000 new stroke patients each year.

Dr. Rowley, chief of neuroradiology and the Joseph Sackett Professor of Radiology at the University of Wisconsin in Madison, says radiologists can play a more prominent role in the triage of stroke patients. "One of the major directions of stroke therapy is the triage of stroke patients using CT and MR," he says.

Dr. Mathews, a neuroradiologist with Northwest Radiology Network and a professor of radiology at the Indiana University School of Medicine in Indianapolis, says that while imaging has been helping to improve stroke survival and reduce disability, there's a lot of room for improvement. "One thing we learned from the European Cooperative Acute Stroke Study was that if you saw low density in more than a third of a vascular territory on an initial CT scan, that patient was at increased risk to have hemorrhage and a potentially bad outcome," he says. "Clinical researchers have extrapolated from those findings when evaluating MR images. For example, when MR shows that there's not a perfusion-diffusion mismatch, we may conclude that we don't have anything to gain because the diffusion defect is already as big as the perfusion defect. This indicates that there may be no brain tissue to salvage, so we would opt not to give such a patient thrombolytic therapy."

Dr. Mathews says the field is in a state of flux as far as determining the best imaging modalities to use in the acute stroke setting. "At St. Vincent Hospital in Indianapolis, we rely primarily on CT to look at the brain parenchyma of the acute stroke patient," says Dr. Mathews. "Then we perform CT angiography of the head and neck, and a CT perfusion scan, which is faster and more quantitative than MR perfusion." He adds that whether radiologists use CT or MR imaging, it's important that they learn how to use the more advanced imaging tools to manage stroke patients. "Radiologists need to be comfortable in quickly obtaining and interpreting the information needed for stroke management decisions, because clinicians rely on us to tell them when patients may benefit from appropriate treatment," Dr. Mathews says.

DIAS and DEDAS

Recent clinical trials show encouraging evidence that better neurologic outcome can be achieved when treatment protocols are based on physiologic imaging criteria, not just clinical features or arbitrary time cut-offs, according to Dr. Rowley. The recently completed Desmoteplase in Acute Ischemic Stroke (DIAS) trial used desmoteplase, a new type of intravenous clot-busting drug derived from bat saliva.

"This trial was novel, not only because it used this very fibrin-specific clot-buster, but also because it was set up so that patients could not be randomized unless they had appropriate MR imaging findings and unless they had a perfusion-diffusion mismatch of at least 20 percent (mean transit time vs. trace diffusion-weighted imaging)," he says. DIAS was also novel because the treatment window was extended from three hours to nine hours.

In the DIAS trial, the bleeding complication rate was less than five percent, which Dr. Rowley says is another improvement over tPA. "What's key here is not only that desmoteplase looks promising, but also that, in the first moments of stroke, radiologists are being asked to step in with advanced imaging to provide risk stratification and patient selection," he says. "To me, this is where a lot of future trials should go."

Howard A. Rowley, M.D.

University of Wisconsin in Madison

Dr. Rowley was among those presenting information from Phase II of the DIAS trial at the American Stroke Association meeting this past spring. "We've just finished the North American Dose Escalation Study of Desmoteplase in Acute Ischemic Stroke (DEDAS) trial, which hopefully will confirm DIAS," he says. "An even larger trial is planned for this fall."

In addition to MR imaging used in these studies, other researchers are looking at the use of CT in trials of IV tPA and IV desmoteplase with patients triaged by CT angiography, CT perfusion and standard CT. Dr. Rowley believes that through research, education and the will to improve, clot-busting therapy can be made available to more acute stroke patients. As evidence, he points to sophisticated stroke programs in Calgary, Cleveland and Houston, where centers are reporting in excess of 10 percent of patients getting thrombolysis.

Stroke Prevention

Researchers at Massachusetts General Hospital (MGH) are using positron emission tomography (PET) in an attempt to identify vessels that are at high risk to cause stroke. "We want to stop stroke before it occurs," says Ahmed Tawakol, M.D., associate director of nuclear cardiology at MGH. "We believe the highly inflamed plaques—the metabolically active plaques—are the ones that go on to progress rapidly or to rupture altogether and cause thrombosis. PET imaging identifies these highly inflamed plaques and we hope that, in the future, we can tailor therapy for those who have these vulnerable plaques compared to those who have fibrous, stable plaques."

Dr. Tawakol and his colleagues assess the biology of plaques using fluorodeoxyglucose (FDG) radio-labeled with the positron emitter F-18. "We've looked at a population of patients undergoing carotid endarterectomy. All the patients that entered the study had been identified by clinicians as having very tight narrowings of their carotid arteries and therefore were deemed appropriate for surgery," he explains. "We used either CT or MR imaging to help with localization. At the end of the study we were able to compare the amount of FDG uptake that we were able to see noninvasively with what was actually removed surgically. We found that the higher the FDG uptake, the more inflammation we saw. The lower the uptake, the more fibrosis we saw."

Dr. Tawakol hopes that this technology, once validated, will allow physicians to ferret out those who truly need surgery from those who do not need surgery, and, more important, to noninvasively identify dangerous plaques before they rupture and cause a heart attack or stroke.