An Accelerated Thrombosis Model for Computational Fluid Dynamics Simulations in Rotary Blood Pumps.

PURPOSE: Thrombosis ranks among the major complications in blood-carrying medical devices and a better understanding to influence the design related contribution to thrombosis is desirable. Over the past years many computational models of thrombosis have been developed. However, numerically cheap models able to predict localized thrombus risk in complex geometries are still lacking. The aim of the study was to develop and test a computationally efficient model for thrombus risk prediction in rotary blood pumps. METHODS: We used a two-stage approach to calculate thrombus risk. The first stage involves the computation of velocity and pressure fields by computational fluid dynamic simulations. At the second stage, platelet activation by mechanical and chemical stimuli was determined through species transport with an Eulerian approach. The model was compared with existing clinical data on thrombus deposition within the HeartMate II. Furthermore, an operating point and model parameter sensitivity analysis was performed. RESULTS: Our model shows good correlation (R2 > 0.93) with clinical data and identifies the bearing and outlet stator region of the HeartMate II as the location most prone to thrombus formation. The calculation of thrombus risk requires an additional 10-20 core hours of computation time. CONCLUSION: The concentration of activated platelets can be used as a surrogate and computationally low-cost marker to determine potential risk regions of thrombus deposition in a blood pump. Relative comparisons of thrombus risk are possible even considering the intrinsic uncertainty in model parameters and operating conditions.

In-vitro performance of a single-chambered total artificial heart in a Fontan circulation.

An in-vitro study was conducted to investigate the general feasibility of using only one pumping chamber of the SynCardia total artificial heart (TAH) as a replacement of the single ventricle palliated by Fontan circulation. A mock circulation loop was used to mimic a Fontan circulation. The combination of both ventricle sizes (50 and 70 cc) and driver (Freedom Driver and Companion C2 Driver) was investigated. Two clinical relevant scenarios (early Fontan; late Fontan) as derived from literature data were set up in the mock loop. The impact of increased transpulmonary pressure gradient, low atrial pressure, and raised central venous pressure on cardiac output was studied. From a hemodynamic point, the single-chambered TAH performed sufficiently in the setting of the Fontan circulation. Increased transpulmonary pressure gradient, from ideal to pulmonary hypertension, decreased the blood flow in combinations by almost 2 L/min. In the early Fontan scenario, a cardiac output of 3-3.5 L/min was achieved using the 50 cc ventricle, driven by the Companion C2 Driver. Even under pulmonary hypertension, cardiac outputs greater than 4 L/min could be obtained with the 70 cc pump chamber in the late Fontan scenario. In the clinically relevant Fontan scenarios, implementation of the single chambered TAH performed successfully from a hemodynamic point of view. The replacement of the failing univentricular heart by a single chamber of the SynCardia TAH may provide an alternative to a complex biventricular repair procedure or ventricular support in Fontan patients.

Intraventricular Flow Simulations in Singular Right Ventricles Reveal Deteriorated Washout and Low Vortex Formation.

PURPOSE: Patients with a functionally univentricular heart represent one of the most common severe cardiac lesions with a prevalence of 3 per 10,000 live births. Hemodynamics of the singular ventricle is a major research topic in cardiology and there exists a relationship between fluid dynamical features and cardiac behavior in health and disease. The aim of the present work was to compare intraventricular flow in single right ventricle (SRV) patients and subjects with healthy left hearts (LV) through patient-specific CFD simulations. METHODS: Three-dimensional real-time echocardiographic images were obtained for five SRV patients and two healthy subjects and CFD simulations with a moving mesh methodology were performed. Intraventricular vortex formation and vortex formation time (VFT) as well as the turbulent kinetic energy (TKE) and ventricular washout were evaluated. RESULTS: The results show significantly lower values for the VFT and the TKE in SRV patients compared with healthy LV subjects. Furthermore, vortex formation does not progress to the apex in SRV patients. These findings were confirmed by a significantly lower washout in SRV patients. CONCLUSIONS: The study pinpoints the intriguing role of intraventricular flows to characterize performance of SRVs that goes beyond standard clinical metrics such as ejection fraction.

Optimizing cerebral perfusion and hemodynamics during cardiopulmonary bypass through cannula design combining in silico, in vitro and in vivo input.

Cardiopulmonary bypass (CPB) is a standard technique for cardiac surgery, but comes with the risk of severe neurological complications (e.g. stroke) caused by embolisms and/or reduced cerebral perfusion. We report on an aortic cannula prototype design (optiCAN) with helical outflow and jet-splitting dispersion tip that could reduce the risk of embolic events and restores cerebral perfusion to 97.5% of physiological flow during CPB in vivo, whereas a commercial curved-tip cannula yields 74.6%. In further in vitro comparison, pressure loss and hemolysis parameters of optiCAN remain unaffected. Results are reproducibly confirmed in silico for an exemplary human aortic anatomy via computational fluid dynamics (CFD) simulations. Based on CFD simulations, we firstly show that optiCAN design improves aortic root washout, which reduces the risk of thromboembolism. Secondly, we identify regions of the aortic intima with increased risk of plaque release by correlating areas of enhanced plaque growth and high wall shear stresses (WSS). From this we propose another easy-to-manufacture cannula design (opti2CAN) that decreases areas burdened by high WSS, while preserving physiological cerebral flow and favorable hemodynamics. With this novel cannula design, we propose a cannulation option to reduce neurological complications and the prevalence of stroke in high-risk patients after CPB.

Hemolysis at low blood flow rates: in-vitro and in-silico evaluation of a centrifugal blood pump.

BACKGROUND: Treating severe forms of the acute respiratory distress syndrome and cardiac failure, extracorporeal membrane oxygenation (ECMO) has become an established therapeutic option. Neonatal or pediatric patients receiving ECMO, and patients undergoing extracorporeal CO2 removal (ECCO2R) represent low-flow applications of the technology, requiring lower blood flow than conventional ECMO. Centrifugal blood pumps as a core element of modern ECMO therapy present favorable operating characteristics in the high blood flow range (4 L/min-8 L/min). However, during low-flow applications in the range of 0.5 L/min-2 L/min, adverse events such as increased hemolysis, platelet activation and bleeding complications are reported frequently. METHODS: In this study, the hemolysis of the centrifugal pump DP3 is evaluated both in vitro and in silico, comparing the low-flow operation at 1 L/min to the high-flow operation at 4 L/min. RESULTS: Increased hemolysis occurs at low-flow, both in vitro and in silico. The in-vitro experiments present a sixfold higher relative increased hemolysis at low-flow. Compared to high-flow operation, a more than 3.5-fold increase in blood recirculation within the pump head can be observed in the low-flow range in silico. CONCLUSIONS: This study highlights the underappreciated hemolysis in centrifugal pumps within the low-flow range, i.e. during pediatric ECMO or ECCO2R treatment. The in-vitro results of hemolysis and the in-silico computational fluid dynamic simulations of flow paths within the pumps raise awareness about blood damage that occurs when using centrifugal pumps at low-flow operating points. These findings underline the urgent need for a specific pump optimized for low-flow treatment. Due to the inherent problems of available centrifugal pumps in the low-flow range, clinicians should use the current centrifugal pumps with caution, alternatively other pumping principles such as positive displacement pumps may be discussed in the future.

Estimation of left ventricular stroke work based on a large cohort of healthy children.

Left ventricular stroke work is an important prognostic marker to analyze cardiac function. Standard values for children are, however, missing. For clinicians, standards can help to improve the treatment decision of heart failures. For engineers, they can help to optimize medical devices. In this study, we estimated the left ventricular stroke work for children based on modeled pressure-volume loops. A lumped parameter model was fitted to clinical data of 340 healthy children. Reference curves for standard values were created over age, weight, and height. Left ventricular volume was measured with 3D echocardiography, while maximal ventricular pressure was approximated with a regression model from the literature. For validation of this method, we used 18 measurements acquired by a conductance catheter in 11 patients. The method demonstrated a low absolute mean difference of 0.033 J (SD: 0.031 J) for stroke work between measurement and estimation, while the percentage error was 21.66 %. According to the resulting reference curves, left ventricular stroke work of newborns has a median of 0.06 J and increases to 1.15 J at the age of 18 years. Stroke work increases over weight and height in a similar trend. The percentile curves depict the distribution. We demonstrate how reference curves can be used for quantification of differences and comparison in patients.

How Computational Modeling can Help to Predict Gas Transfer in Artificial Lungs Early in the Design Process.

Wearable extracorporeal membrane oxygenation (ECMO) circuits may soon become a viable alternative to conventional ECMO treatment. Common device-induced complications, however, such as blood trauma and oxygenator thrombosis, must first be addressed to improve long-term reliability, since ambulatory patients cannot be monitored as closely as intensive care patients. Additionally, an efficient use of the membrane surface can reduce the size of the devices, priming volume, and weight to achieve portability. Both challenges are linked to the hemodynamics in the fiber bundle. While experimental test methods can often only provide global and time-averaged information, computational fluid dynamics (CFD) can give insight into local flow dynamics and gas transfer before building the first laboratory prototype. In this study, we applied our previously introduced micro-scale CFD model to the full fiber bundle of a small oxygenator for gas transfer prediction. Three randomized geometries as well as a staggered and in-line configuration were modeled and simulated with Ansys CFX. Three small laboratory oxygenator prototypes were built by stacking fiber segments unidirectionally with spacers between consecutive segments. The devices were tested in vitro for gas transfer with porcine blood in accordance with ISO 7199. The error of the predicted averaged CFD oxygen saturations of the random 1, 2, and 3 configurations relative to the averaged in-vitro data (over all samples and devices) was 2.4%, 4.6%, 3.1%, and 3.0% for blood flow rates of 100, 200, 300, and 400 ml/min, respectively. While our micro-scale CFD model was successfully applied to a small oxygenator with unidirectional fibers, the application to clinically relevant oxygenators will remain challenging due to the complex flow distribution in the fiber bundle and high computational costs. However, we will outline our future research priorities and discuss how an extended mass transfer correlation model implemented into CFD might enable an a priori prediction of gas transfer in full size oxygenators.

An energy-dissipation-based power-law formulation for estimating hemolysis.

Hemolysis is a major concern in blood-circulating devices, which arises due to non-physiological stresses on red blood cells from ambient flow environment or moving mechanical structures. Computational fluid dynamics (CFD) and empirical hemolysis prediction models have been increasingly used for the design and optimization of blood-circulating devices. The commonly used power-law models for hemolysis prediction often use Reynolds stress to represent effective stress, tend to over-predict hemolysis and fail to capture trends of flow-related hemolysis. This study proposed a new power-law formulation for the numerical hemolysis prediction. The new formulation related hemolysis to the energy dissipation rate, which could be readily obtained from CFD simulations. The model constants were regressed from existing hemolysis data. The new formulation was tested for three benchmark cases and compared to conventional power-law models. The results showed that the new formulation improved prediction of hemolysis for a broad range of flow regimes. The deviations of the predicted hemolysis from experimental results were within one order, and better correlated with experimental results. This study confirmed that Reynolds stress is the main cause of over-prediction of hemolysis for conventional power-law models. Proportionally, the blood damage predicted with Reynolds stresses is more than one order higher than viscous stress, in terms of energy dissipation.

Low-flow assessment of current ECMO/ECCO2R rotary blood pumps and the potential effect on hemocompatibility.

BACKGROUND: Extracorporeal carbon dioxide removal (ECCO2R) uses an extracorporeal circuit to directly remove carbon dioxide from the blood either in lieu of mechanical ventilation or in combination with it. While the potential benefits of the technology are leading to increasing use, there are very real risks associated with it. Several studies demonstrated major bleeding and clotting complications, often associated with hemolysis and poorer outcomes in patients receiving ECCO2R. A better understanding of the risks originating specifically from the rotary blood pump component of the circuit is urgently needed. METHODS: High-resolution computational fluid dynamics was used to calculate the hemodynamics and hemocompatibility of three current rotary blood pumps for various pump flow rates. RESULTS: The hydraulic efficiency dramatically decreases to 5-10% if operating at blood flow rates below 1 L/min, the pump internal flow recirculation rate increases 6-12-fold in these flow ranges, and adverse effects are increased due to multiple exposures to high shear stress. The deleterious consequences include a steep increase in hemolysis and destruction of platelets. CONCLUSIONS: The role of blood pumps in contributing to adverse effects at the lower blood flow rates used during ECCO2R is shown here to be significant. Current rotary blood pumps should be used with caution if operated at blood flow rates below 2 L/min, because of significant and high recirculation, shear stress, and hemolysis. There is a clear and urgent need to design dedicated blood pumps which are optimized for blood flow rates in the range of 0.5-1.5 L/min.

Crucial Aspects for Using Computational Fluid Dynamics as a Predictive Evaluation Tool for Blood Pumps.

The suitability of computational fluid dynamics (CFD) as a regulatory tool for safety assessment of medical devices is still limited: A lack of standardized validation and evaluation methods impairs the quantitative comparability and reliability of simulation studies, particularly regarding the assessment of hemocompatibility. This study investigated important aspects of validation and verification for three common turbulence modeling approaches (laminar, k-ω shear stress transport [SST] and stress-blended eddy simulation [SBES]) and three different mesh refinements. Simulation results for pressure head, characteristic velocity, and shear stress for the benchmark blood pump model of the Food and Drug Administration critical path initiative were compared with its published experimental results. For the highest mesh resolution, all three models predicted the hydraulic pump characteristics with a relative deviation averaged over six operating conditions below 6.1%. In addition, the SBES model showed an accurate agreement of the characteristic velocity field in the pump's diffusor region (relative error <2.9%), while the laminar and SST model calculated significantly elevated and deviating velocity amplitudes (>43.6%). The ability to quantify shear stress is fundamental for the prediction of blood damage. In this respect, this study demonstrated that: 1) a close agreement and validation of both pressure head and characteristic velocity was feasible and 2) the shear stress quantification demanded higher near-wall mesh resolutions, although such high resolutions were not required for the validation of only pressure heads or velocity. Hence, a mesh verification analysis for shear stresses may prove significant for the development of credible CFD blood damage predictions in the future.

Mesh Sensitivity Analysis for Quantitative Shear Stress Assessment in Blood Pumps Using Computational Fluid Dynamics.

The reduction of excessive, nonphysiologic shear stresses leading to blood trauma can be the key to overcome many of the associated complications in blood recirculating devices. In that regard, computational fluid dynamics (CFD) are gaining in importance for the hydraulic and hemocompatibility assessment. Still, direct hemolysis assessments with CFD remain inaccurate and limited to qualitative comparisons rather than quantitative predictions. An underestimated quantity for improved blood damage prediction accuracy is the influence of near-wall mesh resolution on shear stress quantification in regions of complex flows. This study investigated the necessary mesh refinement to quantify shear stress for two selected, meshing sensitive hotspots within a rotary centrifugal blood pump (the blade leading edge and tip clearance gap). The shear stress in these regions is elevated due to presence of stagnation points and the flow around a sharp edge. The nondimensional mesh characteristic number y+, which is known in the context of turbulence modeling, underestimated the maximum wall shear stress by 60% on average with the recommended value of 1, but was found to be exact below 0.1. To evaluate the meshing related error on the numerical hemolysis prediction, three-dimensional simulations of a generic centrifugal pump were performed with mesh sizes from 3 × 106 to 30 × 106 elements. The respective hemolysis was calculated using an Eulerian scalar transport model. Mesh insensitivity was found below a maximum y+ of 0.2 necessitating 18 × 106 mesh elements. A meshing related error of up to 25% was found for the coarser meshes. Further investigations need to address: (1) the transferability to other geometries and (2) potential adaptions on blood damage estimation models to allow better quantitative predictions.

On the Accuracy of Hemolysis Models in Couette-Type Blood Shearing Devices.

Hemolysis is one of the most challenging issues faced by blood contacting devices. Empirical hemolysis models often relate hemolysis to shear stress and exposure time. These models were generally derived from the experimental results of Couette-type blood shearing devices, with assumption of uniform exposure time and shear stress. This assumption is not strictly valid since neither exposure time nor shear stress is uniform. Hence, this study evaluated the influence of the nonuniform exposure time and rotor eccentricity or run-out on the accuracy of power-law hemolysis models, using both theoretical and CFD analysis. This work first provided a systematic analysis of the flow regime in a typical Couette shearing device, and showed the axial flow component can be regarded as fully developed laminar plane Poiseuille flow. It was found that the influence of nonuniform exposure time is within 4% for several widely used power-law models, which were validated by steady CFD simulations. A theoretical relationship was then built between the rotor run-out and hemolysis. We noticed that the influence of rotor run-out on hemolysis is within 5% for a moderate rotor run-out ratio of 0.2. Next, transient CFD simulations were performed to investigate the influence of rotor run-out on hemolysis with run-out ratios of 0.1 and 0.2. The results showed negligible effects for a moderate run-out ratio of 0.1. However, a run-out ratio of 0.2 led to a significant increase of hemolysis, resulting from back flows induced by the run-out of the rotor. These findings will be of great importance for the improvement of the hemolysis estimation and blood compatibility design.

Fluid Dynamics in Rotary Piston Blood Pumps.

Mechanical circulatory support can maintain a sufficient blood circulation if the native heart is failing. The first implantable devices were displacement pumps with membranes. They were able to provide a sufficient blood flow, yet, were limited because of size and low durability. Rotary pumps have resolved these technical drawbacks, enabled a growing number of mechanical circulatory support therapy and a safer application. However, clinical complications like gastrointestinal bleeding, aortic insufficiency, thromboembolic complications, and impaired renal function are observed with their application. This is traced back to their working principle with attenuated or non-pulsatile flow and high shear stress. Rotary piston pumps potentially merge the benefits of available pump types and seem to avoid their complications. However, a profound assessment and their development requires the knowledge of the flow characteristics. This study aimed at their investigation. A functional model was manufactured and investigated with particle image velocimetry. Furthermore, a fluid-structure interaction computational simulation was established to extend the laboratory capabilities. The numerical results precisely converged with the laboratory measurements. Thus, the in silico model enabled the investigation of relevant areas like gap flows that were hardly feasible with laboratory means. Moreover, an economic method for the investigation of design variations was established.

The spiral groove bearing as a mechanism for enhancing the secondary flow in a centrifugal rotary blood pump.

The rapid evolution of rotary blood pump (RBP) technology in the last few decades was shaped by devices with increased durability, frequently employing magnetic or hydrodynamic suspension techniques. However, the potential for low flow in small gaps between the rotor and pump casing is still a problem for hemocompatibility. In this study, a spiral groove hydrodynamic bearing (SGB) is applied with two distinct objectives: first, as a mechanism to enhance the washout in the secondary flow path of a centrifugal RBP, lowering the exposure to high shear stresses and avoiding thrombus formation; and second, as a way to allow smaller gaps without compromising the washout, enhancing the overall pump efficiency. Computational fluid dynamics was applied and verified via bench-top experiments. An optimization of selected geometric parameters (groove angle, width and depth) focusing on the washout in the gap rather than generating suspension force was conducted. An optimized SGB geometry reduced the residence time of the cells in the gap from 31 to 27 ms, an improvement of 14% compared with the baseline geometry of 200 µm without grooves. When optimizing for pump performance, a 15% smaller gap yielded a slightly better rate of fluid exchange compared with the baseline, followed by a 22% reduction in the volumetric loss from the primary pathway. Finally, an improved washout can be achieved in a pulsatile environment due to the SGB ability to pump inwardly, even in the absence of a pressure head.