Published in the May 2007 issue of Today’s Hospitalist.
A 68-year-old woman comes into the hospital after having suffered daily episodes of temporary blindness in her right eye. Those episodes last only minutes and then go away.
Suspecting amaurosis fugax, an ocular transient ischemic attack (TIA) or stroke, you order an ultrasound to look for stenosis in her right carotid artery. The vessel, which turns out to be 90% stenosed, clearly needs to be fixed to reduce the patient’s risk of having a second stroke.
But should you recommend endarterectomy, the gold standard of carotid revascularization for the last several decades? Or should you push the popular, much newer and less invasive stenting approach?
"Patients are much more enthusiastic about a procedure with a single puncture in the femoral artery that fixes things endovascularly than having a big neck surgery," said S. Andrew Josephson, MD, assistant clinical professor of neurology at the University of California, San Francisco (UCSF). "But in its current form, carotid stenting is probably used nationally more than current evidence supports."
Dr. Josephson, who discussed his recommendations at UCSF’s annual meeting on managing hospitalized patients last fall, backed up this view by citing the results of new European studies with surprising findings. Two "noninferiority" randomized trials found that rates of short-term death and recurrent stroke following stenting for treatment of carotid artery stenoses were not equivalent to rates for endarterectomy. In fact, those rates were worse.
"It turned out that both trials failed to demonstrate that stenting was noninferior to endarterectomy in patients with recent stroke or TIA, and that is really news," said Dr. Josephson. Experts are now waiting for the results of a third large trial “this one in North America, called the CREST study “to answer the question definitively.
"If this trial also shows that endarterectomy is better, I think that stenting will really be limited in its use," Dr. Josephson said. (Information about CREST, which is being funded by the National Institutes of Health, is online.)
However, an earlier trial did show that stenting could play a role for a subset of patients at very high risk for endarterectomy, according to Dr. Josephson.
The high-risk patients identified by the SAPPHIRE trial, which was published in the Oct. 7, 2004, New England Journal of Medicine, include those over age 80 or patients with active coronary disease. These individuals could be having a heart attack at the same time they are having a stroke.
Other patients considered high risk include those whose carotid artery is stenotic again, despite a previous endarterectomy; patients who have stenotic vessels due to previous radiation therapy to the neck; and patients with severe pulmonary disease.
"These are the patients about whom the surgeon says, ‘They are so sick, I’m not sure they are going to make it through the [endarterectomy] surgery. Is there another alternative?’ " Dr. Josephson explained.
The SAPPHIRE trial found that stenting was "noninferior" compared with endarterectomy for these very high-risk patients. The study created a lot of excitement at the time, said Dr. Josephson, and stenting took off in popularity “even for patients not at high surgical risk.
Two trials published last fall, however, compared the two procedures for patients without these high-risk features and came to a different conclusion. The SPACE trial results, published in the Oct. 7, 2006, The Lancet, "could not establish noninferiority" for stenting vs. endarterectomy at 30 days, with an endpoint of ipsilateral ischemic stroke or death.
The EVA-3S findings, which appeared in the Oct. 19, 2006, New England Journal of Medicine, were similar. Researchers stopped the trial early after finding that the 30-day incidence of any stroke or death after stenting was 9.6% vs. 3.9% for endarterectomy. After six months, those incidence rates were 11.7% and 6.1% respectively.
Look for experience
Since the publication of the two studies last year, Dr. Josephson said that he recommends endarterectomy to anyone needing revascularization “that is, whose ipsilateral carotid artery shows 70% to 99% stenosis following a stroke or TIA “except those for whom the procedure is of high risk, as shown in the SAPPHIRE trial.
One other important piece of advice is that hospitalists should refer patients to surgeons or interventionalists "who have the most experience" performing the procedures. "The people with the most experience have the best surgical and endovascular results," he pointed out. Endarterectomies are usually done by vascular surgeons or neurosurgeons, while stenting is performed by interventional radiologists or sometimes by cardiologists or vascular surgeons.
"We know that the more experienced the surgeon, the better the outcome, as a general rule in most trials in medicine," said Dr. Josephson. Endarterectomy is major surgery, he added, and there is some risk that a patient will have a stroke during or shortly after it.
As far as determining the extent of stenosis, Dr. Josephson said, studies continue to compare ultrasound, CT angiography and MR angiography to the gold standard of conventional angiography.
At UCSF, Dr. Josephson said, patients are screened with CT angiography as part of their initial stroke workup.
Those patients whose vessels are less than 50% stenosed do not need another evaluation, due to the high negative predictive value of CT angiography. Despite recent small trials, he added, the most important finding of any imaging study is still the degree of occlusion, rather than any other characteristics such as plaque morphology.
Deborah Gesensway is a freelance writer who reports on U.S. health care from Toronto, Canada.