1. Home>
2. Specialties>
3. Neurology>
4. Summary and Comment

Serial Carotid Ultrasound May Improve Stroke Prediction

Progressing carotid stenosis may predict stroke better than statically severe stenosis.

Researchers prospectively investigated the association of carotid-stenosis progression, detected by serial ultrasound (US), with adverse vascular outcomes (including stroke) in 1004 asymptomatic patients who had had at least two carotid US examinations more than six months apart (mean number of US studies per carotid artery, 3; mean time between baseline and last US, 2.1 years).

Whereas degree of stenosis at baseline did not predict cerebrovascular accidents (CVAs), the risk for a CVA increased with progressing degrees of stenosis (e.g., after 3 years, 11.4% of those with an artery that progressed from no or mild stenosis to severe stenosis had a CVA versus 4.1% of those with severe stenosis throughout the 3 years). Adjustments for stroke risk factors did not alter these risk estimates.

Comment: This study is the largest I know of in terms of patient numbers but not the longest in terms of follow-up. More information about the indications that led these patients to undergo US would have been useful (were all patients really asymptomatic, since they had to be asymptomatic for only the six months before baseline?). Also unclear was how many patients had stenosis in both arteries, as opposed to one artery; this information would have helped clarify the role of clinical risk factors (risk behaviors, cholesterol level, etc.). If both arteries showed progression in most patients with stroke or TIA, then serial carotid stenosis measurement might be a surrogate for adequacy of risk-factor control.

Serial carotid-stenosis testing might be useful in two ways: (1) to select patients for surgery and (2) to assess how well vascular risk is being curtailed, or not, by interventions in at-risk people, especially in patients who have an inadequate history or who have not made behavior changes (e.g., smoking cessation) as they claim. I am not convinced, however, that serial carotid US is more predictive of CVAs than clinical factors. The authors report that clinical factors did not add to prediction in the multivariable model, but if clinical factors had been inputted first, would the stenosis measurement have added anything? The factors they list have been clearly and strongly related to stroke in numerous studies, so the lack of added prediction here probably reflects an inability to track risk-factor control rather than the factors' lack of impact.

Despite such minor concerns, these findings add to other data that argue for carotid surveillance, not just one-time testing. The findings show that serial testing can be used to detect high-risk subgroups and that serial carotid US may give clinicians a good indication of how well interventions are working.

— Robert J. Adams, MD, MS

Dr. Adams is Regents Professor of Neurology, Medical College of Georgia, Augusta.

Published in Journal Watch Neurology March 12, 2004

Citation(s):

Bertges DJ et al. Relevance of carotid stenosis progression as a predictor of ischemic neurological outcomes. Arch Intern Med 2003 Oct 27; 163:2285-9.

• Original article (Subscription may be required)
• Medline abstract (Free)