Contact

Duke Health News

919-660-1306

DURHAM, N.C. -- After comparing two commonly used non-invasive imaging techniques to the older and more invasive "gold standard" for detecting potentially stroke-causing blockages in arteries in the neck, a Duke University Medical Center neurologist cautions against basing a decision to operate solely on the results of these non-invasive tests.

In question is a surgical procedure known as carotid endarterectomy, in which surgeons clear obstructions from the carotid artery, the major artery in the neck that supplies oxygen-rich blood to the brain. Past studies have shown that carotid endarterectomy -- which does have its own risks -- is most effective in reducing the risk for future strokes in patients with symptoms whose arteries are at least 50 percent to 70 percent blocked.

The key first step for physicians, therefore, is determining the degree and scope of the blockage, since the benefits of the surgery rise in relation to the degree of blockage, Duke researchers say. The gold standard for many years has been contrast angiography, in which a contrast dye is injected into the blood vessels leading to the brain. Subsequent X-rays of the head and neck provide detailed pictures of the arteries. However, there is a risk that patients undergoing this test will suffer an adverse event, such as stroke or even death.

"Many surgeons, in an attempt to avoid these potential complications, will base their decision to operate on the results of a non-invasive test," said Dr. Larry Goldstein, director of Duke's Center for Cerebrovascular Disease. "Based on the results of our study, however, we would recommend caution in following this approach."

The results of Goldstein's study are published in Tuesday's issue of the journal Neurology.

The non-invasive procedures studied were duplex ultrasound (DU), which operates on the same principle as ultrasound studies of unborn babies, and magnetic resonance angiography (MRA), which uses standard MRI technology. Both tests are virtually risk-free and can be performed relatively quickly. However, according to Goldstein, they do not provide the same level of detail as does angiography.

In the Duke study, the charts of 569 consecutive patients during a three-year period who received all three tests at an academic medical center and a community hospital were reviewed. After reviewing the charts, the researchers found that if the decision to operate were based solely on DU, about 29 percent of the patients would be misclassified, while 18 percent with MRA were misclassified. Misclassification rates dropped to 8 percent if both tests were used and showed the same degree of blockage.

"Misclassification based on non-invasive studies alone went both ways -- some patients could receive the surgery when they shouldn't have (false-positive), or they might not be offered the operation when they should have (false-negative)," Goldstein explained. "The majority of the errors reported were related to the tests' ability to help physicians determine the degree and extant of blockage in the artery."

Specifically, DU alone had a false-positive rate of 23 percent and a false-negative rate of 6 percent. MRA alone had a 9 percent rate for both false-positive and false-negative outcomes. If both tests were done, the false-positive rate dropped to 5 percent, and the false-negative rate dropped to just under 3 percent.

"At minimum, we'd suggest that both tests be performed," Goldstein said. "Endarterectomy is an effective operation when performed in selected patients by skilled surgeons. If the patient's primary physician is concerned about the risks of angiography or the surgery, the patient should be referred to a center that does many of these procedures, since studies show high-volume centers have the lowest complication rates."

In performing a carotid endarterectomy, surgeons make an incision in the neck, and the carotid artery is split open lengthwise. The surgeon removes the atherosclerotic plaque or blood clots impeding the flow of blood to the brain. The surgeon then closes the artery with or without the addition of a "patch," depending on the size of the obstruction.

These blockages can lead to stroke or transient ischemic attacks (TIA or mini-strokes), which can be a precursor to a major stroke. Patients experiencing a TIA from a narrowing of the carotid artery may experience temporary loss or blurring of vision in one eye, numbness or weakness in the arm or leg on one side of the body, or have trouble speaking or understanding.

Goldstein's analysis was supported by the Duke Center for Cerebrovascular Disease. Dr. Dean C.C. Johnston, who also contributed to the study during a fellowship at Duke, was supported by Canada's health system.