In Vitro Benchmarking Study of Ventricular Assist Devices in Current Clinical Use.

BACKGROUND: Left ventricular assist devices (LVADs) offer live-saving therapy to transplant-ineligible heart failure patients. A major limitation of the technology includes pump thrombosis, bleeding, and recurrent infection that prove difficult to predict from in vivo animal testing. Shear stress introduced by the LVAD affects more than just hemolysis because platelets, leukocytes, and plasma proteins all contribute to the propensity for complications. It is important to assess overall damage by a new device against a baseline as early as possible in the development process so that design iterations can be made if required. METHODS: Explanted VADs currently in clinical use (HeartMate 2 and HVAD) were carefully cleaned, inspected, and run at 5 L/min and pressure at 100 mmHg in a standard 500 mL mock circulatory loop using bovine blood. The CentriMag was used as a control pump because of its low blood damage profile. Samples were collected at regular intervals and the following were analyzed: complete cell counts, hemolysis, platelet activation, leukocyte-derived microparticles (LMPs), and von Willebrand factor (vWF) degradation. RESULTS: The HeartMate 2 had the highest levels of hemolysis and platelet activation after 6 hours compared with the HVAD and CentriMag. A decreased granulocyte count, high numbers of LMPs and CD11bBrightHLADR- LMPs, and decreased vWF collagen binding activity was most evident in the HVAD. CONCLUSIONS: The results indicate that it is possible to observe differences between different pump designs during in vitro testing that might translate to clinical performance. This study demonstrates the importance of developing standard in vitro total blood damage methods against which device developers could use to modify design to reduce complication risk long before implantation.

Shear Stress-Induced Total Blood Trauma in Multiple Species.

The common complications in heart failure patients with implanted ventricular assist devices (VADs) include hemolysis, thrombosis, and bleeding. These are linked to shear stress-induced trauma to erythrocytes, platelets, and von Willebrand factor (vWF). Novel device designs are being developed to reduce the blood trauma, which will need to undergo in vitro and in vivo preclinical testing in large animal models such as cattle, sheep, and pig. To fully understand the impact of device design and enable translation of preclinical results, it is important to identify any potential species-specific differences in the VAD-associated common complications. Therefore, the purpose of this study was to evaluate the effects of shear stress on cells and proteins in bovine, ovine, and porcine blood compared to human. Blood from different species was subjected to various shear rates (0-8000/s) using a rheometer. It was then analyzed for complete blood counts, hemolysis by the Harboe assay, platelet activation by flow cytometry, vWF structure by immunoblotting, and function by collagen binding activity ELISA (vWF : CBA). Overall, increasing shear rate caused increased total blood trauma in all tested species. This analysis revealed species-specific differences in shear-induced hemolysis, platelet activation, and vWF structure and function. Compared to human blood, porcine blood was the most resilient and showed less hemolysis, similar blood counts, but less platelet activation and less vWF damage in response to shear. Compared to human blood, sheared bovine blood showed less hemolysis, similar blood cell counts, greater platelet activation, and similar degradation of vWF structure, but less impact on its activity in response to shear. The shear-induced effect on ovine blood depended on whether the blood was collected via gravity at the abattoir or by venepuncture from live sheep. Overall, ovine abattoir blood was the least resilient in response to shear and bovine blood was the most similar to human blood. These results lay the foundations for developing blood trauma evaluation standards to enable the extrapolation of in vitro and in vivo animal data to predict safety and biocompatibility of blood-handling medical devices in humans. We advise using ovine venepuncture blood instead of ovine abattoir blood due to the greater overall damage in the latter. We propose using bovine blood for total blood damage in vitro device evaluation but multiple species could be used to create a full understanding of the complication risk profile of new devices. Further, this study highlights that choice of antibody clone for evaluating platelet activation in bovine blood can influence the interpretation of results from different studies.