Ryan J Pewowaruk1, Jenna Racine2, J Igor Iruretagoyena2, Alejandro Roldán-Alzate3,4,5. 1. Biomedical Engineering, University of Wisconsin - Madison, Madison, USA. 2. Maternal and Fetal Medicine, University of Wisconsin - Madison, Madison, USA. 3. Biomedical Engineering, University of Wisconsin - Madison, Madison, USA. roldan@wisc.edu. 4. Mechanical Engineering, University of Wisconsin - Madison, Madison, USA. roldan@wisc.edu. 5. Radiology, University of Wisconsin - Madison, Madison, USA. roldan@wisc.edu.
Abstract
PURPOSE: Preeclampsia (PE) is a pregnancy complication of abnormally elevated blood pressure and organ damage where endothelial function is impaired. Wall shear stress (WSS) strongly effects endothelial cell morphology and function but in PE the WSS values are unknown. WSS calculations from ultrasound inaccurately assume cylindrical arteries and patient specific computational fluid dynamics (CFD) typically require time-consuming 3D imaging such as CT or MRI. METHODS: Two-dimensional (2D) B-mode ultrasound images were lofted together to create simplified three-dimensional (3D) geometries of the brachial artery (BA) that incorporate artery curvature and non-circular cross sections. This process was efficient and on average took 120 ± 10 s. Patient specific CFD was then performed to quantify BA WSS for a small cohort of PE (n = 5) and normotensive pregnant patients (n = 5) and compared against WSS calculations assuming a cylindrical artery. RESULTS: For several WSS metrics (time averaged WSS (TAWSS), peak systolic WSS, oscillatory shear index (OSI), OSI/TAWSS and relative residence time) CFD on the simplified arterial geometries calculated large spatial differences in WSS that assuming a cylindrical artery cannot calculate. Bland-Altman and intra-class correlation (ICC) analyses found assuming a cylindrical artery both underestimated (p < 0.05) and had poor agreement (ICC < 0.5) with the maximum WSS values from CFD. WSS values that were abnormal compared to the normotensive patients (OSI = 0.014 ± 0.026) appear related to the pregnancy complications fetal growth restriction (n = 2, OSI = 0.14, 0.25) and gestational diabetes (n = 1, OSI = 0.23). CONCLUSION: Creating 3D artery geometries from 2D ultrasound images can be used for CFD simulations to calculate WSS from ultrasound without assuming cylindrical arteries. This approach requires minimal time for both medical imaging and CFD analysis.
PURPOSE: Preeclampsia (PE) is a pregnancy complication of abnormally elevated blood pressure and organ damage where endothelial function is impaired. Wall shear stress (WSS) strongly effects endothelial cell morphology and function but in PE the WSS values are unknown. WSS calculations from ultrasound inaccurately assume cylindrical arteries and patient specific computational fluid dynamics (CFD) typically require time-consuming 3D imaging such as CT or MRI. METHODS: Two-dimensional (2D) B-mode ultrasound images were lofted together to create simplified three-dimensional (3D) geometries of the brachial artery (BA) that incorporate artery curvature and non-circular cross sections. This process was efficient and on average took 120 ± 10 s. Patient specific CFD was then performed to quantify BA WSS for a small cohort of PE (n = 5) and normotensive pregnant patients (n = 5) and compared against WSS calculations assuming a cylindrical artery. RESULTS: For several WSS metrics (time averaged WSS (TAWSS), peak systolic WSS, oscillatory shear index (OSI), OSI/TAWSS and relative residence time) CFD on the simplified arterial geometries calculated large spatial differences in WSS that assuming a cylindrical artery cannot calculate. Bland-Altman and intra-class correlation (ICC) analyses found assuming a cylindrical artery both underestimated (p < 0.05) and had poor agreement (ICC < 0.5) with the maximum WSS values from CFD. WSS values that were abnormal compared to the normotensive patients (OSI = 0.014 ± 0.026) appear related to the pregnancy complications fetal growth restriction (n = 2, OSI = 0.14, 0.25) and gestational diabetes (n = 1, OSI = 0.23). CONCLUSION: Creating 3D artery geometries from 2D ultrasound images can be used for CFD simulations to calculate WSS from ultrasound without assuming cylindrical arteries. This approach requires minimal time for both medical imaging and CFD analysis.
Authors: Margaret C Hogan; Kyle J Foreman; Mohsen Naghavi; Stephanie Y Ahn; Mengru Wang; Susanna M Makela; Alan D Lopez; Rafael Lozano; Christopher J L Murray Journal: Lancet Date: 2010-04-09 Impact factor: 79.321
Authors: Florian Joly; Gilles Soulez; Simon Lessard; Claude Kauffmann; Irene Vignon-Clementel Journal: Ann Biomed Eng Date: 2019-10-21 Impact factor: 3.934
Authors: Harry A Silber; Pamela Ouyang; David A Bluemke; Sandeep N Gupta; Thomas K Foo; Joao A C Lima Journal: Am J Physiol Heart Circ Physiol Date: 2004-09-02 Impact factor: 4.733
Authors: G E McVeigh; G M Brennan; G D Johnston; B J McDermott; L T McGrath; W R Henry; J W Andrews; J R Hayes Journal: Diabetologia Date: 1992-08 Impact factor: 10.122
Authors: Adam Updegrove; Nathan M Wilson; Jameson Merkow; Hongzhi Lan; Alison L Marsden; Shawn C Shadden Journal: Ann Biomed Eng Date: 2016-12-08 Impact factor: 3.934