Which Non-Invasive Test is Best for the Diagnosis of Clinically Significant Carotid Artery Stenosis?

Abstract & Commentary

By Matthew E. Fink, MD, Vice Chairman, Professor of Clinical Neurology, Weill Cornell Medical College, Chief, Division of Stroke and Critical Care Neurology, NewYork-Presbyterian Hospital. Dr. Fink reports no financial relationship relevant to this field of study.

Synopsis: All of the currently available non-invasive techniques for carotid artery imaging give accurate results when there is a high-grade stenosis (70-99%), but contrast-enhanced MRA is slightly more sensitive.

Source: Wardlaw JM, et al. Non-Invasive Imaging Compared with Intra-Arterial Angiography in the Diagnosis of Symptomatic Carotid Stenosis: A Meta-Analysis. Lancet. 2006;367:1503-1512.

With improvements in the treatment of extracranial carotid artery stenosis using carotid endarterectomy (CEA) and carotid artery stenting (CAS), it has become important for neurologists to understand the relative accuracy of various non-invasive tests for carotid stenosis that can be used as a screening procedure or as a definitive study prior to referral for treatment. There have been few head-to-head studies that have directly compared Doppler ultrasound (DUS) with MR angiography, with (CEMRA) and without (MRA) contrast enhancement, and CT angiography (CTA). Therefore, Wardlaw and colleagues performed a meta-analysis of all studies in the English language literature from 1980-2004 that evaluated the sensitivity and specificity of various non-invasive carotid studies compared to either NASCET or ECST criteria for angiography. To be included in the analysis, the published studies had to meet the Standards for Reporting of Diagnostic Accuracy, as well as the methods of the Cochrane Database of Systematic Reviews. After excluding all papers from 1980-1986 because of obsolete technology, and eliminating those papers that did not define what proportion of patients or arteries were symptomatic, there were 41 original papers available for inclusion. For each paper, the authors computed an estimate for sensitivity and specificity with 95% confidence intervals (CIs) for each non-invasive imaging technique compared with intra-arterial angiography.

In the 41 studies, which represented 2541 patients and 4876 arteries, contrast-enhanced MR angiography was more sensitive (0.94, 95%; CI = 0.88-0.97) and specific (0.93, 95%; CI = 0.89-0.96) for 70%-99% stenosis than Doppler ultrasound, MR angiography without contrast, or CT angiography (sensitivities = 0.89, 0.88, 0.76; specificities 0.84, 0.84, 0.94, respectively). For any degree of stenosis less than 70%, none of the non-invasive tests were deemed reliable when compared to intra-arterial angiography, and data regarding any combination of tests was too sparse to draw any conclusions.


Non-invasive testing for carotid artery disease is widely available and liberally utilized as part of the clinical identification of stroke risk factors. In symptomatic patients who have had a TIA or stroke, where treatment decisions are urgent, we often combine tests, commonly DUS with MRA, or DUS with CTA, prior to CEA or CAS. The study by Wardlaw et al seems to indicate that all of the available, non-invasive tests are reasonably sensitive and specific when the degree of stenosis is 70%-99% and, probably, it is not necessary to proceed with intra-arterial angiography before surgery or stenting if a high-grade stenosis in found. CEMRA is slightly more sensitive than the other tests, and CTA is slightly more specific. In those patients who are symptomatic with lesser degrees of stenosis (ie, 50%-69%), the sensitivity and specificity of the non-invasive studies is not very high, but there is a lack of good data to compare the various studies. Because the indications for surgical or endovascular therapy in this group with less than 70% stenosis is not as clear and the therapeutic benefit is small, we would continue to recommend intra-arterial angiography if CEA or CAS is being considered. In addition, we need to continue to collect data in a prospective and blinded fashion in the moderate stenosis group and compare the results with intra-arterial angiography whenever possible. As morbidity and mortality from CEA and CAS continues to improve, there will be a desire to treat more patients with moderate stenosis, and more accurate non-invasive carotid imaging would be helpful.