Wardlaw JM et al. (2006) Non-invasive imaging compared with intra-arterial angiography in the diagnosis of symptomatic carotid stenosis: a meta-analysis. Lancet 367: 1503–1512

Safer, noninvasive imaging techniques are replacing intra-arterial angiography for the assessment of carotid artery stenosis. To assess the accuracy of the various noninvasive methods, Wardlaw et al. conducted a meta-analysis of trials comparing these imaging techniques with intra-arterial angiography.

Of 672 studies published from January 1980 to April 2004, 41 met the inclusion criteria for the study; papers were excluded because of subsequent technological advances, incomplete data, or poor methodology. For the detection of 70–99% stenosis, contrast-enhanced magnetic resonance angiography (MRA) had the highest sensitivity (94%) of the techniques examined, followed by Doppler ultrasound (89%), non-contrast-enhanced MRA (88%) and CT angiography (77%). CT angiography, however, was found to have the highest specificity (95%), followed by contrast-enhanced MRA (93%), Doppler ultrasound (84%) and MRA (84%). For 50–69% stenosis, reliable comparisons could not be made because of insufficient data, but the accuracy of all methods was considerably reduced.

The authors state that noninvasive imaging techniques are accurate for the diagnosis of 70–99% stenosis, but additional carefully designed studies are required to determine the accuracy of these techniques in moderate stenosis. Pre-screening of patients with Doppler ultrasound introduced verification bias to the studies. Most trials involved patients with an average age in the 60s, whereas trials in older patients would be more relevant to stroke prevention. The authors conclude that prospective, properly blinded studies in older patients are necessary to determine the best procedure for the diagnosis of carotid artery stenosis.