Computational Fluid Dynamics-Based Blood Flow Assessment Facilitates Optimal Management of Portal Vein Stenosis After Liver Transplantation.

BACKGROUND: Portal vein stenosis develops in 3.4-14% of split liver transplantation1-3 and its early detection and treatment are essential to achieve long-term graft survival,2-5 although the diagnostic capability of conventional modalities such as Doppler ultrasound and computed tomography is limited.1,4,5 METHODS: This study used computational fluid dynamics to analyze portal vein hemodynamics in the management of post-transplant portal vein stenosis. To perform computational fluid dynamics analyses, three-dimensional portal vein model was created using computed tomographic DICOM data. The inlet flow condition was set according the flow velocity measured on Doppler ultrasonography. Finally, portal vein flow was simulated on a fluid analysis software (Software Cradle, Japan). RESULTS: An 18-month-old girl underwent liver transplantation using a left lateral graft for biliary atresia. At the post-transplant 1-week evaluation, the computational fluid dynamics streamline analysis visualized vortices and an accelerated flow with a velocity ratio < 2 around the anastomotic site. The wall shear stress analysis revealed a high wall shear stress area within the post-anastomotic portal vein. At the post-transplant 6-month evaluation, the streamline analysis illustrated the increased vortices and worsening flow acceleration to reach the proposed diagnostic criteria (velocity ratio > 3:1).3,5 The pressure analysis revealed a positive pressure gradient of 3.8 mmHg across the stenotic site. Based on the findings, the patient underwent percutaneous transhepatic portal venoplasty with balloon dilation. The post-treatment analyses confirmed the improvement of a jet flow, vortices, a high wall shear stress, and a pressure gradient. DISCUSSION: The computational fluid dynamics analyses are useful for prediction, early detection, and follow-up of post-transplant portal vein stenosis and would be a promising technology in post-transplant management.

Blood flow competition after aortic valve bypass: an evaluation using computational fluid dynamics.

OBJECTIVES: Aortic valve bypass (AVB) (apico-aortic conduit) remains an effective surgical alternative for patients in whom surgical aortic valve replacement or transcatheter aortic valve implantation is not feasible. However, specific complications include thrombus formation, possibly caused by stagnation arising from flow competition between the antegrade and retrograde flow, but this has not been fully investigated. The aim of this study was to analyse flow characteristics after AVB and to elucidate mechanisms of intra-aortic thrombus using computational fluid dynamics (CFD). METHODS: Flow simulation was performed on data obtained from a 73-year-old postoperative AVB patient. Three-dimensional cine phase-contrast magnetic resonance imaging at 3 Tesla was used to acquire flow data and to set up the simulation. The vascular geometry was reconstructed using computed tomography angiograms. Flow simulations were implemented at various ratios of the flow rate between the ascending aorta and the graft. Results were visualized by streamline and particle tracing. RESULTS: CFD demonstrated stagnation in the ascending aorta-arch when retrograde flow was dominant, indicating that the risk of thrombus formation exists in the ascending arch in cases with severe aortic stenosis and/or poor left ventricular function. Meanwhile, stagnation was observed in the proximal descending aorta when the antegrade and retrograde flow were equivalent, suggesting that the descending aorta is critical when aortic stenosis is not severe. CONCLUSIONS: Flow stagnation in the aorta which may cause thrombus was observed when retrograde flow was dominant and antegrade/retrograde flows were equivalent. Our results suggest that anticoagulants might be recommended even in patients who receive biological valves.

Patient-specific assessment of hemodynamics by computational fluid dynamics in patients with bicuspid aortopathy.

OBJECTIVE: Hemodynamics related to eccentric blood flow may factor into the development of bicuspid aortic valve aortopathy. We investigated wall shear stress distribution by means of magnetic resonance imaging-based computational fluid dynamics in patients with a bicuspid aortic valve. METHODS: Included were 12 patients with a bicuspid aortic valve (aortic stenosis, n = 11; root enlargement, n = 1). Three patients with a normal tricuspid aortic valve (arch aneurysm, n = 1; descending aortic aneurysm, n = 2) were included for comparison. The thoracic aorta geometry was reconstructed by means of 3-dimensional computed tomography angiography, and the bicuspid aortic valve orifice was modeled. Flow rates at the sinotubular junction and 3 aortic branches were measured at various time points by cine phase-contrast magnetic resonance imaging to define boundary conditions for computational fluid dynamics, and the flow was simulated. RESULTS: Bicuspid aortic valve cusp configurations were type 0 lateral (n = 4), type 0 anterior-posterior (n = 2), type 1 L-R (n = 4), and type 1 R-N (n = 2). Abnormal aortic helical flow was seen in the ascending aorta and transverse arch in all patients with bicuspid aortic valves and was right handed in 11 patients (91%). No such flow was seen in the patients with tricuspid aortic valves. The patients with bicuspid aortic valves were likely to have jet flow/wall impingement against the greater curvature of the proximal ascending aorta, resulting in remarkably increased wall shear stress around the impingement area. CONCLUSIONS: Computational fluid dynamics simulation is useful for precise evaluation of hemodynamics related to bicuspid aortic valve aortopathy. Such evaluation will advance our understanding of the disease pathophysiology and may facilitate molecular biological investigation.