The purpose of the study was to evaluate the diagnostic performance of non‐invasive imaging methods of detecting carotid artery stenosis in comparison to digital subtraction angiography (DSA) by undertaking a systematic review of the literature. The non‐invasive methods reviewed include multi‐slice computed tomography (MSCT) angiography, duplex ultrasound (US) and magnetic resonance (MR) angiography.
A search of medical databases (PubMed, Medline, ScienceDirect, ProQuest) of the English literature was performed and 22 studies with 68 comparisons were found to meet the selection criteria and included in our study. Five studies were performed with MSCT angiography, 11 with duplex US and 13 with MR angiography. Both duplex US and MR angiography were studied in seven out of 24 US and MR studies. The diagnostic value of each imaging modality was reviewed in terms of the sensitivity and specificity compared to DSA, which is regarded as the standard method.
The mean sensitivity and specificity for MSCT angiography, duplex US and MR angiography were 88% (95% CI; 83%–92%) and 90% (95% CI; 85%–94%), 88% (95% CI; 81%–94%) and 89% (95% CI; 85%–94%), 94% (95% CI; 90%–97%) and 89% (95% CI; 85%–92%), respectively based on overall assessment. The evaluation showed that contrast‐enhanced MR angiography has high diagnostic value for detection of more than 50% stenosis of carotid artery stenosis with mean sensitivity and specificity being 95% (95% CI; 92%–98%) and 91% (95% CI; 86%–95%). When assessment was based on a combination of MR angiography and duplex US, the sensitivity reached the highest value of 98% (95% CI; 96%–100%). This analysis indicates that MR angiography, especially contrast‐enhanced MR angiography could be used as a reliable alternative modality to DSA in the detection of carotid artery stenosis.