Eric K. Shang, Alison M. Pouch, Chun Xu, Melissa M. Levack, Robert C. Gorman, Clyde F. Barker, Chandra M. Sehgal, Benjamin M. Jackson
Surgery, University of Pennsylvania, Philadelphia, PA.
OBJECTIVES: Carotid wall thickness (CWT) is important for cardiovascular risk stratification and for biomechanical modeling of carotid atherosclerotic disease. Surface ultrasound (US) is the standard means of measuring carotid intimal-medial thickness (IMT). We hypothesized that semi-automated segmentation algorithms applied to computed tomographic angiography (CTA) can accurately measure CWT.
METHODS: Patients (n=5) with carotid disease having undergone both CTA and US were identified retrospectively. CTA DICOM images were segmented with custom algorithms utilizing active contour generation for luminal surfaces, and isoline contour detection to delineate the outer (adventitial) carotid wall and both calcified and soft plaque. CWT was computed in areas absent of plaque as the minimal distance between the luminal and adventitial walls. Six corresponding locations in the common and internal carotid in each patient were identified on duplex and wall thickness measurements taken to validate CTA results (n=30 comparisons).
RESULTS: CWT by CTA ranged from 0.78 to 1.41 mm. There was 3.8%±2.4% absolute difference between CTA and US. The measurements correlated well with R=0.955. There was no detectable bias by Bland-Altman analysis.
CONCLUSIONS: CTA can be used to accurately measure CWT. 3D models of the carotid bifurcation which incorporate locally-resolved wall thickness are possible and will allow more realistic biomechanical engineering analyses.
AUTHOR DISCLOSURES: C. F. Barker, Nothing to disclose; R. C. Gorman, Nothing to disclose; B. M. Jackson, Nothing to disclose; M. M. Levack, Nothing to disclose; A. M. Pouch, Nothing to disclose; C. M. Sehgal, Nothing to disclose; E. K. Shang, Nothing to disclose; C. Xu, Nothing to disclose.
Posted April 2012