ARTICLE IN BRIEF
The Asymptomatic Carotid Emboli Study, an eight-year multicenter, international, prospective observational study enrolling 482 subjects, found that detection of embolic signals using TCD reliably predicted risk of ipsilateral stroke and transient ischemic attack after two years in patients with asymptomatic stenosis.
Can patients with asymptomatic carotid stenosis who are at high risk of stroke be identified using transcranial Doppler (TCD) ultrasound? An Aug. 9 study in The Lancet Neurology suggests they can, promising to improve the selection of patients who might benefit from surgery, as well as those who can safely be treated with medical interventions alone.
The Asymptomatic Carotid Emboli Study (ACES), an eight-year multicenter, international, prospective observational study enrolling 482 subjects, found that detection of embolic signals using TCD reliably predicted risk of ipsilateral stroke and transient ischemic attack (TIA) after two years in patients with asymptomatic stenosis.
“Previous smaller studies have produced conflicting results as to whether this technique could identify patients with asymptomatic carotid stenosis at high risk of stroke,” study author Hugh S. Markus, MD, told Neurology Today. “However, previous studies did not have sufficient patient numbers to answer the question definitively. Our study had sufficient power to get a statistically reliable result and demonstrate that the presence of embolic signals on ultrasound does identify a group who are at higher risk of stroke and conversely a group who are at lower risk of stroke.
“This study provides convincing data that the technique could be used to identify individuals in whom one should operate and also a group in whom operation should not be performed as their risk of stroke is very low,” said Dr. Markus, professor in the department of clinical neuroscience at St. George's University of London.
Optimal management of asymptomatic carotid stenosis has been controversial, and practices vary between different clinicians and in different countries. The cost and risks associated with surgery, and the fact that non-invasive strategies such as education, support of healthy lifestyle practices, and antiplatelet therapy have improved, make choice of therapy — medical management or surgery — difficult.
In the study, patients had two one-hour TCD recordings from the ipsilateral middle cerebral artery at baseline and one one-hour recording at six, 12, and 18 months. Patients were followed up for two years and the primary endpoint was ipsilateral stroke and TIA.
Of 482 patients who were recruited, 467 had evaluable recordings. Embolic signals were present in 77 of 467 patients at baseline.
The hazard ratio for the risk of ipsilateral stroke and TIA from baseline to two years in patients with embolic signals was 2.54 compared to those without a signal. For ipsilateral stroke alone, the hazard ratio was 5.57. The absolute annual risk of ipsilateral stroke or transient ischemic attack between baseline and two years was 7.13 percent in patients with embolic signals and 3.04 percent in those without; for ipsilateral stroke the absolute annual risk was 3.62 percent in patients with embolic signals and 0.70 percent in those without.
Controlling for antiplatelet therapy, degree of stenosis, and other risk factors did not alter the results, according to the study.
Experts in stroke and vascular disease who reviewed the ACES study for Neurology Today said the results are promising and offer an indication that embolic signals detected by TCD might help with clinical decision-making.
Yet Dr. Markus and colleagues acknowledge that review of TCD recordings and identification of embolic signals is a labor-intensive task not easily performed in routine clinical settings. And at least one reviewer said that results of the ACES study still do not justify routine use of TCD to identify candidates for surgery, especially given the steady improvement in medical management of carotid stenosis.
Neurologist Seemant Chaturvedi, MD, professor of neurology at Wayne Statue University School of Medicine, called the results “somewhat convincing” and noted that the subject is of great interest to clinicians trying to weigh risks and benefits of surgery for individual patients. Two prior randomized trials — the Asymptomatic Carotid Atherosclerosis Study (ACAS)3 and the Asymptomatic Carotid Surgery Trial (ACST) — reported that about 32 patients needed to have carotid endarterectomy to prevent disabling stroke or death in one patient over a five-year period.
But a 2005 report in Stroke by J. David Spence, MD, at the University of Western Ontario, reported that among patients with carotid stenosis of at least 60 percent, absence of microemboli as detected by TCD identified those who could be managed medically without endarterectomy.
“TCD has not been widely used in part because clinicians have been waiting for confirmation of its utility,” Dr. Chaturvedi told Neurology Today. “It's not invasive so there is no real downside to the procedure itself for the patients.” But he added that there are technical issues that can make the routine use of TCD in clinical practice problematic.
Dr. Chaturvedi noted, for instance, that asymptomatic patients are likely not to have many emboli, and he likened the problem to that of bird watchers scouting the skies for a single rare species. “You can spend a lot of time looking for that one bird,” he said. “Similarly, clinicians need to know that asymptomatic patients aren't going to have that many emboli, so you may need to record for an hour to get one signal.”
Dr. Markus acknowledged the problem in remarks to Neurology Today. “Before its widespread application, reliable techniques for reviewing the recordings and identifying these embolic signals are required,” he said. “In the current study, we did this all using an experienced observer. This approach is possible in a research study but not in clinical practice.”
And one other reviewer said she continues to believe that medical intervention alone is still the best course for asymptomatic patients.
“Rates of embolic signal detection rates and stroke symptoms are highly dependent on patient risk factor load and the efficacy of the medical intervention given,” said Anne L. Abbott, PhD, of the Baker IDI Heart and Diabetes Institute in Melbourne, Australia. “We know reported average annual stroke rates in patients given medical intervention alone have been falling over the last 20-30 years and are expected to continue falling, because of continued improvements in medical intervention and its implementation. Recently reported stroke rates match or are lower than those of patients who received surgery in the randomized trials.”
Noting that ACES patients were recruited between 1999 and 2007, she added that if the study was repeated now, “it is likely that the embolic signal detection rates and stroke /TIA rates would be even lower and a statistically significant correlation between detecting one embolic signal and stroke symptoms might not have been shown. We also need a clear picture of patient risk factor load and of the medical intervention given leading up to and during an entire risk stratification study.”
Still, the results appear to advance the possibility — with further replication of results and refinement of TCD techniques — for a better risk-and cost-benefit profile in the selection of asymptomatic patients for endarterectomy.
Said Dr. Chaturvedi: “Most neurologists feel there are too many revascularization procedures on asymptomatic patients, so this tool has the potential to narrow it down to the people who will benefit from it the most.”