Evaluation of contrast densities in the diagnosis of carotid stenosis by CT angiography

AJR Am J Roentgenol. 1997 Aug;169(2):569-73. doi: 10.2214/ajr.169.2.9242779.


Objective: This study assesses intravascular density produced by ionic and nonionic contrast material and its effect on visualization of stenoses by CT angiography.

Materials and methods: CT angiography was performed using a 32-vessel phantom to study grades of luminal stenoses (0-100%), three lengths of stenoses (1, 3, and 5 mm), and two angles of inclination into the stenoses (45 degrees and 75 degrees). Scans were obtained with a slice thickness of 2 mm, a slice interval of 1.5 mm, a pitch of 1.0, a voltage of 120 kV, and a current of both 100 and 200 mA. Vessels were oriented parallel to the z-axis, and opacified with ionic and nonionic contrast material that had densities of 100, 150, 200, 250, 300, and 350 H. Cross-sectional luminal diameters were measured in and out of the stenoses. Edge definition and halo artifact for each vessel were graded by an investigator who was unaware of the contrast material density used.

Results: A contrast density of 150 H as revealed by CT angiography yielded the most accurate stenosis measurements with ionic contrast material. For nonionic contrast material, attenuation values of 150 and 200 H produced the best results on CT angiography. A density of 100 H or greater than 250 H significantly increased the error of vessel measurement (p = .001) for both types of contrast material. For the two current levels tested (100 and 200 mA), no statistical difference was found.

Conclusion: The accuracy of CT angiography in measuring carotid stenosis depends on the luminal attenuation value. An optimum contrast density is 150 H for ionic contrast material; for nonionic contrast material, 150-200 H (at the window and level settings of 300 H and 40 H).

MeSH terms

  • Carotid Arteries / diagnostic imaging
  • Carotid Stenosis / diagnostic imaging*
  • Contrast Media*
  • Humans
  • Phantoms, Imaging
  • Tomography, X-Ray Computed*


  • Contrast Media