ACES: Transcranial Doppler Flags Risk for Stroke or TIA in Patients With Asymptomatic Carotid Stenosis

Susan Jeffrey

May 28, 2010

May 28, 2010 (Barcelona, Spain) — Results of a prospective observational study suggest that use of transcranial Doppler (TCD) to detect embolic signals may be used to help with risk stratification of patients with high grade but asymptomatic carotid stenosis.

"Detection of asymptomatic emboli with TCD can identify patients with asymptomatic carotid stenosis who are at low risk, and also patients who are at high risk, and it may be useful in selection of patients for carotid endarterectomy," Hugh Markus, FRCP, from Clinical Neuroscience, St. George's University of London in the United Kingdom, concluded.

He cautioned, though, that in their study, TCD recordings were analyzed manually in a central reading office. For this technique to be useful on a more routine basis, he noted, "there needs to be really robust systems for detecting these emboli in real time, and at the moment there aren't validated systems available."

Dr. Markus presented the findings from the Asymptomatic Carotid Emboli Study (ACES) here at the XIX European Stroke Conference to coincide with their publication online May 28 in the Lancet Neurology.

Asymptomatic Management Controversial

Although intervention in patients with symptomatic carotid disease is generally accepted as beneficial, the management of asymptomatic disease has become controversial, the study authors write. Asymptomatic disease is associated with a lower stroke risk than symptomatic disease, at about 2% per year or less. Surgical trials, including the Asymptomatic Carotid Atherosclerosis Study (ACAS) and the Asymptomatic Carotid Surgery Trial (ACST), suggested about 32 patients would need to undergo endarterectomy to prevent a disabling stroke or death in 1 patient during 5 years.

The cost-effectiveness of surgery in patients with asymptomatic disease has been recently called into question and the benefit suggested to be even less now with best medical therapies introduced since ACAS and ACST were reported. "Recent analyses have reported that, with improved medical treatment, the annual risk of stroke patients with asymptomatic carotid stenosis is lower than that reported in the carotid endarterectomy trials (nearer to 1% in these trials, compared with 2%–3% in ACAS), which further reduces the benefit of surgical intervention," the study authors write.

Still, only about 15% of strokes are preceded by a transient ischemic attack (TIA), they point out, so "waiting for stenoses to become symptomatic fails to prevent most strokes caused by carotid stenosis."

"So we need better ways to select out patients who may benefit but also spare those patients any operation at all, and one of the possible ways is detection of emboli," Dr. Markus said. Stroke is usually embolic in these patients, and emboli detection is possible using TCD.

ACES, then, was a prospective, observational study investigating the hypothesis that detection of asymptomatic embolic signals using TCD could help predict the risk for subsequent stroke among patients with high-grade (≥70%) stenosis who were as yet asymptomatic.

Patients had two 1-hour TCD recordings from the ipsilateral middle cerebral artery at baseline and then recordings at 6, 12, and 18 months. As noted, all recordings were analyzed centrally by investigators who were masked to patient identity.

The primary endpoint was ipsilateral stroke and TIA, and patients were followed up for 2 years.

Among the 482 patients recruited, 467 had evaluable recordings. Of these, 77 had embolic signals on TCD at baseline. During follow-up, there were 26 ipsilateral TIAs and 6 ipsilateral strokes; 4 of those with TIAs went on to have strokes, the study authors note, for a total of 10 strokes during follow-up.

They found that the presence of embolic signals at baseline was significantly correlated with risk for subsequent ipsilateral stroke and TIA. The hazard ratio was higher for ipsilateral stroke alone, he noted, but the confidence intervals were wider.

Table. Risk for Ipsilateral Stroke and TIA and Ipsilateral Stroke Alone for ACS Patients With and Without Embolic Signals on TCD

Endpoint Hazard Ratio (95% CI) P Value
Ipsilateral stroke and TIA 2.54 (1.20 – 5.30) .015
Ipsilateral stroke 5.57 (1.61 – 19.32) .007

ACS = Asymptomatic Carotid Emboli Study; CI = confidence interval; TCD = transcranial Doppler; TIA = transient ischemic attack

After adjustment for other factors, including antiplatelet therapy at baseline, which is known to affect emboli counts, he noted, the odds ratios were not significantly altered, "so it seems to be an independent association."

In an additional analysis, they looked at the risk for ipsilateral stroke and TIA for patients who had embolic signals on the recording preceding the next 6-month follow-up compared with those who did not was 2.63 (95% confidence interval [CI], 1.01 – 6.88; P = .049); for ipsilateral stroke alone, the hazard ratio was 6.37 (95% CI, 1.59 – 25.57; P = .009).

"Perhaps a helpful way to look at the data is to say what's the absolute risk of stroke in patients who do and don't have embolic signals?" Dr. Markus concluded. In this study, the absolute annual risk of ipsilateral stroke and TIA between baseline and 2 years was 7.13% for those with embolic signals and 3.04% for those without. For ipsilateral stroke alone, the risk was 3.62% for those with emboli and 0.70% for those without, a very low risk.

Finally, they also did a meta-analysis, combining results from previously published studies with their own, for a total of 1144 patients, although they point out that the cutoff for being defined as having or not having embolic signals was slightly different between studies. "When you put all the data together," he said, "you can see there is a highly significant association of 6.6 (6.63 [95% CI, 2.85 – 15.44]; P < .0001)."

The overall improvement in the natural history of asymptomatic carotid stenosis with better medical therapy "might make surgical intervention hazardous," the study authors write. "Carotid stenting has been suggested as an alternative to endarterectomy, but as yet there are no data showing it is safer than endarterectomy for asymptomatic carotid stenosis."

Low Risk in Asymptomatic Disease

J. David Spence, MD, from the Robarts Research Institute in London, Ontario, Canada, and colleagues recently published a similar paper using TCD to define stroke risk among patients with asymptomatic carotid stenosis. The report from his group, published in the Archives of Neurology earlier this year, showed that intensive medical therapy instituted after 2003 in their institution reduced both the frequency of microemboli on TCD and the risk for hard endpoint events, including stroke, myocardial infarction, and death, compared with the period before 2003 (Arch Neurol. 2010;67:180-186).

Asked for comment on the ACES findings, Dr. Spence pointed out the stroke risk for asymptomatic patients was still higher in this current study by Markus and colleagues than in their recent report.

"I suspect the reason the Markus study showed less impressive differences than we did was that their medical therapy was less intensive," Dr. Spence concludes. "Even so, their findings corroborate ours."

He feels strongly that intervention is overused in patients with asymptomatic disease, and he has been an outspoken critic on this point. "The real story here is that most patients with asymptomatic carotid stenosis cannot benefit from stenting or endarterectomy, but in the US, about 70% of stenting and endarterectomy is for asymptomatic stenosis."

This situation, he says, will probably not improve with publication of the Carotid Revascularization Endarterectomy versus Stenting Trial (CREST) results earlier this week. CREST compared carotid stenting with endarterectomy in patients with both symptomatic and asymptomatic carotid disease and concluded that the net benefit for the 2 procedures was similar, and this was true for both groups of patients (N Engl J Med. Published online May 26, 2010).

"In CREST, the risk for either endarterectomy or stenting is higher than the risk with intensive medical therapy in our recent study," Dr. Spence points out. Data from a study published by the Robarts group in 2005 (Spence JD, et al. Stroke. 2005;36:2373-2378), included in the meta-analysis in the Markus paper, showed that only 10% of asymptomatic patients could benefit from intervention. "By 2010," he added, "with more intensive therapy, it is less than 5%."

Others have been swayed to this view as well. Dr. Spence pointed to recent publications, one a systematic review (Abbott AL. Stroke. 2009;40:e573-e578) and the other a population-based study (Marquardt L, et al. Stroke. 2010;41:e11-e17), in which authors concluded that the risk associated with asymptomatic carotid stenosis is so low with current standard therapies that medical therapy should be considered the treatment of choice for asymptomatic stenosis.

Useful Approach to Risk Stratification

Asked for comment on these findings, Philip B. Gorelick, MD, MPH, professor and head of the Department of Neurology and Rehabilitation and director of the Center for Stroke Research at the University of Illinois College of Medicine at Chicago, said the researchers provide the results of a "carefully implemented" TCD study designed to determine whether embolic signals detected by TCD could predict stroke risk in a cohort of asymptomatic patients with at least 70% stenosis and find that overall patients with embolic signals have about 2.5 times the risk for ipsilateral or TIA over 2 years compared with those without such signals. The study, he said, "provides a useful approach to risk stratification with TCD in patients with higher-grade asymptomatic carotid artery stenosis."

Interestingly, Dr. Gorelick points out, of the 32 primary outcome events, 26, or about 81%, were TIAs and only 6, or about 19%, were strokes. "These findings are concordant with some prior studies which have suggested that the occurrence of stroke in asymptomatic carotid stenosis, though not guaranteed, may be heralded by a TIA rather than a frank stroke, thus providing a warning symptom, which, if recognized, can possibly be diagnosed and treated to prevent a subsequent stroke event," he said. "This study does nicely show that TCD detection of embolic signals may be used to identify asymptomatic higher-grade carotid artery stenosis patients who are at a higher risk for stroke or TIA and those who have a lower absolute risk."

Information that seemed to be missing, though, was data on lipid markers and use of statin therapy, Dr. Gorelick noted. It is possible that more aggressive statin therapy or tighter blood pressure control might have further reduced stroke and TIA risk.

"If this were the case, it might provide additional support for the use of medical therapy as opposed to carotid endarterectomy or angioplasty and stenting in this contentious group of patients with asymptomatic carotid stenosis," he said. "In these patients, the latter interventions provide only modest benefit over medical therapy alone and are highly dependent on operator skill, as it does not take much in the way of perioperative morbidity and mortality complications to tip the balance against these procedures."

The study was funded by the British Heart Foundation. The study authors have disclosed no relevant financial relationships.

Lancet Neurol. Published online May 28, 2010.

XIX European Stroke Conference (ESC): Large Clinical Trials B. Presented Friday, May 28, 2010.