Wall shear stress, the force per area acting on the lumen wall due to the blood flow, is an important biomechanical parameter in the localization and progression of atherosclerosis. To calculate shear stress and relate it to atherosclerosis, a 3D description of the lumen and vessel wall is required. We present a framework to obtain the 3D reconstruction of human coronary arteries by the fusion of intravascular ultrasound (IVUS) and coronary computed tomography angiography (CT). We imaged 23 patients with IVUS and CT. The images from both modalities were registered for 35 arteries, using bifurcations as landmarks. The IVUS images together with IVUS derived lumen and wall contours were positioned on the 3D centerline, which was derived from CT. The resulting 3D lumen and wall contours were transformed to a surface for calculation of shear stress and plaque thickness. We applied variations in selection of landmarks and investigated whether these variations influenced the relation between shear stress and plaque thickness. Fusion was successfully achieved in 31 of the 35 arteries. The average length of the fused segments was 36.4 ± 15.7 mm. The length in IVUS and CT of the fused parts correlated excellently (R2= 0.98). Both for a mildly diseased and a very diseased coronary artery, shear stress was calculated and related to plaque thickness. Variations in the selection of the landmarks for these two arteries did not affect the relationship between shear stress and plaque thickness. This new framework can therefore successfully be applied for shear stress analysis in human coronary arteries.

, , , , ,
doi.org/10.1007/s10554-009-9546-y, hdl.handle.net/1765/28595
International Journal of Cardiovascular Imaging
Erasmus MC: University Medical Center Rotterdam

van der Giessen, A., Schaap, M., Gijsen, F., Groen, H., van Walsum, T., Mollet, N., … Wentzel, J. (2010). 3D fusion of intravascular ultrasound and coronary computed tomography for in-vivo wall shear stress analysis: A feasibility study. International Journal of Cardiovascular Imaging, 26(7), 781–796. doi:10.1007/s10554-009-9546-y