Background – Carotid duplex ultrasound (CDU) is routinely used for detection of extracranial internal carotid artery (EICA) stenosis in stroke patients. Power M-mode transcranial Doppler (PMD-TCD) is a new technology designed for studying intracranial vessels. This prospective study was performed for validation of CDU versus digital subtraction cerebral angiography (DSA) and PMD-TCD versus CDU in detection of EICA stenosis.
Methods – Part I: Validation of CDU versus DSA in 50 stroke patients (100 carotids) admitted to Walter Mackenzie Center, Canada in 2003. All of the CDUs were performed by a sonographer with an HP Sono 5500 device (USA) and a 13-MHz linear probe. The degree of EICA stenosis was measured using DSA according to North American symptomatic carotid endarterectomy trial (NASCET) method by another neuroradiologist blinded to CDU results.
Part II: Validation of PMD-TCD versus CDU in another group of 50 patients (100 carotids) with stroke. All PMD-TCD studies were performed with a TCDM100 device (Spencer, USA) and a 2-MHz probe by a sonographer blinded to CDU results.
Results – Part I: Compared with DSA, CDU had a sensitivity of 96%, specificity of 100%, positive predictive value (PPV) of 100%, and negative predictive value (NPV) of 82% in detecting EICA stenosis of 50% and over. CDU versus DSA confirmed EICA stenosis of 70% and over with a sensitivity of 88%, specificity of 95%, PPV of 96%, and NPV of 87%.
Part II: Compared with CDU, PMD-TCD had a sensitivity of 45%, specificity of 94%, PPV of 50%, and NPV of 93% for detection of EICA-stenosis of 50% and over. PMD-TCD, when compared with CDU, confirmed EICA-stenosis of 70% and over with a sensitivity of 43%, specificity of 99%, PPV of 75% and NPV of 96%.
Conclusion – In our center, CDU is a highly reliable noninvasive diagnostic tool for detection of EICA stenosis. PMD-TCD of carotid has a moderate sensitivity for detection of EICA stenosis and is not recommended as a screening test for this purpose.