Implantation of the VentrAssist Implantable Rotary Blood Pump in sheep.

The VentrAssist Implantable Rotary Blood Pump (IRBP) is a hydrodynamically suspended, electromagnetically driven, centrifugal blood pump that provides continuous flow of up to 10 L/min at 3,000 rpm. In vivo studies in sheep were conducted to assess system design and performance. Surgery involved thoracotomy with subdiaphragmatic pump placement. Cannulae were transdiaphragmatic, with inflow in the left ventricular apex and outflow anastomosed to the descending aorta. Animals had no anticoagulation or antiplatelet therapy after surgery and no prophylactic antibiotics after recovery. Twelve sheep were supported for 622 pump days. Estimated pump flow ranged from 1 to 5.5 L/min at 1,800 to 2,000 rpm using 2.5 to 4.5 W. There was no clinical evidence of hemolysis or cardiovascular, renal, or hepatic dysfunction. Adverse outcomes included kinking/disconnection of the outflow cannula caused by the graft bend relief (n = 4), which was addressed through cannula redesign. Pump electrical malfunction (n = 4), caused by a silicone potting compound, was corrected using a neutral curing potting material. Surgical/husbandry issues (n = 2) also were addressed. The VentrAssist IRBP provides high flow at low rotational speed and power consumption. Further trials are in progress in advance of in vivo studies of the safety and efficacy of the final system.

Initial in vivo experience of the VentrAssist implantable rotary blood pump in sheep.

VentrAssist (VentrAssist Division, Ventracor Ltd., Chatswood, NSW, Australia) has developed an implantable centrifugal blood pump with an integrated rotor and impeller that is hydrodynamically suspended. Bench testing has been used to assess the performance of the pump under a broad range of operating conditions. This study examined the performance of the pump in vivo up to 90 days implantation. Pumps were implanted via a left lateral thoracotomy. The inflow cannula was inserted at the apex of the left ventricle. The outflow cannula was anastomosed to the descending thoracic aorta. Eighteen implants were performed. Poor recovery from surgery was the main cause of early study termination. These studies demonstrate the suitability of the animal model for evaluation of the VentrAssist rotary blood pump. Further in vivo studies prior to preclinical trials are in progress.