Measurement of carotid stenosis: comparisons between duplex and different angiographic grading methods

J Neuroimaging. 2003 Apr;13(2):133-9.


Background and purpose: Endarterectomy has been proved to be an effective stroke prevention procedure. However, there are still inconsistencies between the results of different preoperative evaluation methods, which may sometimes complicate treatment plans. This study measured the discrepancies between different angiographic grading methods and attempted to further assess the accuracy of the carotid duplex examination according to these different angiographic grading methods.

Methods: One hundred seventy-one preendarterectomy carotid duplex examinations and angiograms were reviewed. All angiograms were measured blindly by one of the authors using the North American Symptomatic Carotid Endarterectomy Trial (N), the European Carotid Surgery Trial (E), and the common carotid (C) methods. The measurement results were further converted into the area of stenosis indices (N2, E2, and C2, respectively). By using regression testing, all results could be compared. The duplex examination data were then compared with the results of different angiographic measurement methods to evaluate their accuracy.

Results: The measurement results of all angiographic grading methods were well correlated. Using different angiographic grading systems as the gold standard, duplex examination for screening endarterectomy candidates produced the following results: According to the N method, accuracy was 74%; according to the E method, accuracy was 90%; and according to the C method, the accuracy rate was 92%. According to N2, accuracy was 88%; according to E2, accuracy was 94%; and according to C2, the accuracy rate was 93%.

Conclusions: The measurement results of all 3 commonly used angiographic grading methods were linearly correlated with one another. (It is important to note that none of those standards should be treated as the only gold standard.) In this study, the duplex criteria have a greater accuracy rate according to E or C rather than N. However, this study also demonstrated that the cut point of the Doppler criteria is the determinant factor for accuracy rather than which gold standard was compared. Through careful adjustment of the cutoff criteria, carotid duplex can be highly accurate, despite the use of different reference standards.

Publication types

  • Comparative Study

MeSH terms

  • Aged
  • Aged, 80 and over
  • Angiography* / methods*
  • Angiography* / standards
  • Angiography, Digital Subtraction
  • Carotid Stenosis / diagnostic imaging*
  • Carotid Stenosis / surgery
  • Endarterectomy, Carotid
  • Female
  • Humans
  • Linear Models
  • Male
  • Middle Aged
  • Reproducibility of Results
  • Ultrasonography, Doppler, Duplex / methods*
  • Ultrasonography, Doppler, Duplex / standards