Search

The most common technical complication during ECMO is clot formation. A large clot inside a membrane oxygenator reduces effective membrane surface area and therefore gas transfer capabilities, and restricts blood flow through the device, resulting in an increased membrane oxygenator pressure drop (dpMO). The reasons for thrombotic events are manifold and highly patient specific. Thrombus formation inside the oxygenator during ECMO is usually unpredictable and remains an unsolved problem. Clot sizes and positions are well documented in literature for the Maquet Quadrox-i Adult oxygenator based on CT data extracted from devices after patient treatment. Based on this data, the present study was designed to investigate the effects of large clots on purely technical parameters, for example, dpMO and gas transfer. Therefore, medical grade silicone was injected into the fiber bundle of the devices to replicate large clot positions and sizes. A total of six devices were tested in vitro with silicone clot volumes of 0, 30, 40, 50, 65, and 85 mL in accordance with ISO 7199. Gas transfer was measured by sampling blood pre and post device, as well as by sampling the exhaust gas at the devices' outlet at blood flow rates of 0.5, 2.5, and 5.0 L/min. Pre and post device pressure was monitored to calculate the dpMO at the different blood flow rates. The dpMO was found to be a reliable parameter to indicate a large clot only in already advanced "clotting stages." The CO2 concentration in the exhaust gas, however, was found to be sensitive to even small clot sizes and at low blood flows. Exhaust gas CO2 concentration can be monitored continuously and without any risks for the patient during ECMO therapy to provide additional information on the endurance of the oxygenator. This may help detect a clot formation and growth inside a membrane oxygenator during ECMO even if the increase in dpMO remains moderate.

Flow distribution is key in artificial lungs, as it directly influences gas exchange performance as well as clot forming and blood damaging potential. The current state of computational fluid dynamics (CFD) in artificial lungs can only give insight on a macroscopic level due to model simplification applied to the fiber bundle. Based on our recent work on wound fiber bundles, we applied particle image velocimetry (PIV) to the model of an artificial lung prototype intended for neonatal use to visualize flow distribution in a stacked fiber bundle configuration to (i) evaluate the feasibility of PIV for artificial lungs, (ii) validate CFD in the fiber bundle of artificial lungs, and (iii) give a suggestion how to incorporate microscopic aspects into mainly macroscopic CFD studies. To this end, we built a fully transparent model of an artificial lung prototype. To increase spatial resolution, we scaled up the model by a factor of 5.8 compared with the original size. Similitude theory was applied to ensure comparability of the flow distribution between the device of original size and the scaled-up model. We focused our flow investigation on an area (20 × 70 × 43 mm) in a corner of the model with a Stereo-PIV setup. PIV data was compared to CFD data of the original sized artificial lung. From experimental PIV data, we were able to show local flow acceleration and declaration in the fiber bundle and meandering flow around individual fibers, which is not possible using state-of-the-art macroscopic CFD simulations. Our findings are applicable to clinically used artificial lungs with a similar stacked fiber arrangement (e.g., Novalung iLa and Maquet QUADROX-I). With respect to some limitations, we found PIV to be a feasible experimental flow visualization technique to investigate blood-sided flow in the stacked fiber arrangement of artificial lungs.

The jet of the outflow cannula is a potential risk for patients undergoing cardiopulmonary bypass (CPB), because increased jet velocities lead to altered flow conditions and might furthermore mobilize atherosclerotic plaques from calcified aortas. The cannula jet is therefore among the main reasons for cerebral hypoxia and stroke in CPB patients. In the past, we developed a validated computational fluid dynamics (CFD) model to analyze flow conditions during CPB as dependent on cannulation and support modalities. This model is now applied to develop a novel CPB outflow cannula to reduce the jet effect and increase cerebral blood flow. The Multi-Module Cannula (MMC) is based on a generic elbow cannula that was iteratively improved. It features an inner wall to smoothly guide the blood as well as an elliptically shaped outlet diffuser. During standard CPB conditions of 5 L/min, the pressure drop over the MMC is 61 mm Hg, compared with 68 mm Hg with a standard cannula. The maximum velocities are decreased from 3.7 m/s to 3.3 m/s. In the cannula jet of the MMC, the velocities are reduced further, down to 1.6 m/s. The cerebral blood flow is typically reduced during CPB. Using the MMC, however, it reaches almost physiological values at 715 mL/min. These results suggest that the MMC outperforms standard CPB cannulas. Further design improvements and improved insertion techniques are under consideration.

Extracorporeal membrane oxygenation (ECMO) is a well-established therapy for several lung and heart diseases in the field of neonatal and pediatric medicine (e.g., acute respiratory distress syndrome, congenital heart failure, cardiomyopathy). Current ECMO systems are typically composed of an oxygenator and a separate nonpulsatile blood pump. An oxygenator with an integrated pulsatile blood pump for small infant ECMO was developed, and this novel concept was tested regarding functionality and gas exchange rate. Pulsating silicone tubes (STs) were driven by air pressure and placed inside the cylindrical fiber bundle of an oxygenator to be used as a pump module. The findings of this study confirm that pumping blood with STs is a viable option for the future. The maximum gas exchange rate for oxygen is 48mL/min/L(blood) at a medium blood flow rate of about 300mL/min. Future design steps were identified to optimize the flow field through the fiber bundle to achieve a higher gas exchange rate. First, the packing density of the hollow-fiber bundle was lower than commercial oxygenators due to the manual manufacturing. By increasing this packing density, the gas exchange rate would increase accordingly. Second, distribution plates for a more uniform blood flow can be placed at the inlet and outlet of the oxygenator. Third, the hollow-fiber membranes can be individually placed to ensure equal distances between the surrounding hollow fibers.

Current goals in the development of oxygenators are to reduce extrinsic surface contact area, thrombus formation, hemolysis, and priming volume. To achieve these goals and provide a favorable concentration gradient for the gas exchange throughout the fiber bundle, this study attempts to find an optimized inlet and outlet port geometry to guide the flow of a hexagonal-shaped oxygenator currently under development. Parameters derived from numerical flow simulations allowed an automated quantitative evaluation of geometry changes of flow distribution plates. This led to a practical assessment of the quality of the flow. The results were validated qualitatively by comparison to flow visualization results. Two parameters were investigated, the first based on the velocity distribution and the second calculated from the residence time of massless particles representing erythrocytes. Both approaches showed significant potential to improve the flow pattern in the fiber bundle, based on one of the parameters of up to 66%. Computational fluid dynamics combined with a parameterization proved to be a powerful tool to quickly improve oxygenator designs.

Tip geometry and placement of rotary blood pump inflow and outflow cannulae influence the dynamics of flow within the ventricle and aortic branch. Cannulation, therefore, directly influences the potential for thrombus formation and end-organ perfusion during ventricular assist device (VAD) support or cardiopulmonary bypass (CPB). The purpose of this study was to investigate the effect of various inflow/outflow cannula tip geometries and positions on ventricular and greater vessel flow patterns to evaluate ventricular washout and impact on cerebral perfusion. Transparent models of a dilated cardiomyopathic ventricle and an aortic branch were reconstructed from magnetic resonance imaging data to allow flow measurements using particle image velocimetry (PIV). The contractile function of the failing ventricle was reproduced pneumatically, and supported with a rotary pump. Flow patterns were visualized around VAD inflow cannulae, with various tip geometries placed in three positions in the ventricle. The outflow cannula was placed in the subclavian artery and at several positions in the aorta. Flow patterns were measured using PIV and used to validate an aortic flow computational fluid dynamic study. The PIV technique indicated that locating the inflow tip in the left ventricular outflow tract improved complete ventricular washout while the tip geometry had a smaller influence. However, side holes in the inflow cannula improved washout in all cases. The PIV results confirmed that the positioning and orientation of the outflow cannula in the aortic branch had a high impact on the flow pattern in the vessels, with a negative blood flow in the right carotid artery observed in some cases. Cannula placement within the ventricle had a high influence on chamber washout. The positioning of the outflow cannula directly influences the flow through the greater vessels, and may be responsible for the occasional reduction in cerebral perfusion seen in clinical CPB.

The influence of heat dissipating systems, such as rotary blood pumps, was investigated. Titanium cylinders as rotary blood pump housing dummies were immersed in porcine blood and constantly tempered at specific temperatures (37-60 degrees C) over a defined period of time. The porcine blood was anticoagulated either by low heparin dosage or citrate. At frequent intervals, samples were taken for blood analysis and the determination of the plasmatic coagulation cascade. Blood parameters do not alter at surface temperatures below 50 degrees C. Hyperthermia-induced hemolysis could be confirmed. The plasmatic coagulation cascade is terminated at surface temperatures exceeding 55 degrees C. The adhesion of blood constituents on surfaces is temperature and time dependent, and structural changes of adhesions and blood itself were detected.