Bookmark and Share

Pediatric Ventricular Assist Devices

The Artificial Organs Laboratory at the University of Maryland School of Medicine is working with Jarvik Heart Inc. (New York, NY) to develop pediatric versions of the Jarvik 2000 artificial heart. The objectives of this study are as follows:

  • Design infant- and child-size models of the Jarvik 2000 heart and perform appropriate theoretical studies, as needed, to justify its design.
  • Theoretical studies may include mathematical modeling and/or computer simulation of the proposed concepts; corrosion controls; physiological performance for normal and diseased heart conditions; a reliability model for the proposed device and failure-mode analysis.
  • Design infant and child-size Jarvik 2000 hearts to fit and provide adequate circulatory support for pediatric patients weighing between 3 and 15 kg and between 15 and 25 kg.
  • Perform in-vitro testing and evaluation to demonstrate the transient and chronic performance characteristics of the completed systems.
  • Perform in-vivo testing and evaluation to measure the physiological and hematological parameters of animals with implants and record the performance of the implanted system.
  • Evaluate the biocompatibility of component testing and whole-system testing in-vitro and in-vivo.
Pediatric Ventricular Assist Devices

Computational Fluid Dynamics

The surfaces of these pumps were developed using a design strategy based on computational fluid dynamics (CFD) that involved closely coupling a Navier-Stokes solver to a parameterized geometry modeler and advanced mesh movement techniques. CFD-based blood damage models for shear-induced hemolysis, as well as surrogate functions describing thrombosis potential, were employed to help guide design improvements. Physical prototypes of the pumps were fabricated and experimentally analyzed using particle imaging flow visualization. The CFD-predicted results correlate well with the experimental data, including pressure-flow (H-Q) performance and specific flow field features.

Validating In-Vitro Experiments

Flow-Visualization Study

We propose to develop an experimental platform, techniques and protocols and to conduct fluid dynamic experimentation to validate the CFD models of the proposed pediatric blood pumps. The flow-velocity fields and flow path under steady and simulated physiologic pulsatile flow conditions will be visualized, with the fluorescent particle tracing flow visualization methods being quantified using digital particle image velocimetry. Experimental models of the child and infant Jarvik 2000 will be constructed based on the CFD optimized geometry. The flow fields at inlet, outlet and impeller regions will be visualized and quantified using the flow visualization approaches described above.

Pulsatile mock-circulation tests will be conducted to characterize the pediatric and infant Jarvik 2000 blood pumps under conditions representative of normal left ventricular myocardial contractility and of severe impairment. The loop will be constructed to contain a pneumatic pulsatile ventricle that will be operated with peak dp/dt of 1500 mmHg/sec to represent normal contractility and of only 500 mmHg/sec to mimic severe reduction of contractility. Mean aortic pressure will be maintained at 65 mmHg via back-pressure regulation, mean filling pressure will be 10-12 mmHg, and heart rates will vary between 70-130 BPM. Futhermore, as we expect the devices to possibly be positioned within a systemic venous atrium or thin, poorly contracting right ventricle, these associated congenital heart disease scenarios will be modeled and tested on the loop.

Hemolysis Testing

We plan to use standard ex-vivo flow loops to assess hemolysis of whole ovine blood of CFD/flow-visualized optimized rotary blade geometries. The supernatent will be drawn at timed intervals. Hemoglobin will be determined as measurement of frank hemolysis. The mean RBC volume will also be measured to detect sublytic injury. We will study white blood cells and platelets in-vivo as well.

In-Vivo Evaluation

The pediatric Jarvik 2000 prototype will be tested in young sheep between 20-25 kg. The infant Jarvik 2000 prototype will be tested in young sheep around 10 to 15 kg.