Paradoxical Effects of Viscosity on the VentrAssist Rotary Blood Pump *S. Vidakovic, *P. Ayre, *J. Woodard, *N. Lingard, †G. Tansley, and †J. Reizes *VentrAssist P/L and †University of Technology, Sydney, Australia Abstract: The ability of the VentrAssist blood pump to perform at its optimum design point is determined by a number of factors such as geometry of the pump, surface roughness, and fluid properties. Once the fluid properties are known, the performance characteristics of the pump can be optimized for that fluid. It is important to under- stand the effects of dynamic viscosity  (called simply vis- cosity hereafter) on the performance characteristics and stability of the pump. The performance envelope of the pump and the needs of the patient must be matched. The VentrAssist pump has no shaft, seals, or fixed bearings and relies on the fluid-dynamic forces to maintain its effective performance. A number of different fluids have been tested to determine the effects of viscosity and density on pump performance. These include aqueous glycerol, red blood cells (RBCs) suspended in phosphate buffered sa- line solution (PBS), and Haemaccel. The effects of viscos- ity on the bearing stiffness, stage efficiency, and the pres- sure-flow rate (HQ) are characterized. The experimental results show a slight increase in the pressure rise across the pump shown as a positive upward shift of the H-Q curves with a decrease in viscosity; however, this is rela- tively small. A paradox in system efficiency exists: for a given fluid asymptotic viscosity, the system efficiency (product of magnetic and stage efficiency) using Haemac- cel or PBS is greater than for the same viscosity of aqueous glycerol. Key Words: Rotary blood pump—VentrAs- sist—Viscosity. Viscosity has been identified as an important pa- rameter for the stable operation (1) of the VentrAs- sist implantable rotary blood pumps (IRBP) (Fig. 1). Stability is not a concern in shafted pumps with fixed bearings (2) because the rotor is restrained mechani- cally, or in magnetically suspended rotors because they do not depend on the fluid properties to main- tain their position within the housing (3). The hydro- dynamic surfaces in a VentrAssist IRBP produce lift on top and bottom surfaces and balance the rotor in the housing. Rotor stiffness is determined by the vis- cosity of the fluid tested (Fig. 2), the gap between the moving and stationary surfaces, surface roughness, and the taper on the hydrodynamic wedge (Fig. 3). A shafted version of the VentrAssist IRBP was used to measure the stage efficiency of a number of rotor designs to determine the optimum stage effi- ciency. The loop in this circuit was kept to a mini- mum to maintain a minimum pressure drop across the circuit to extend the range of useful data. Only nonbiological fluids were tested in this loop. The mock loop built to ASTM requirements was used to evaluate all fluids, including biological fluids using the VentrAssist IRBP. MATERIALS AND METHODS To determine the effects of viscosity on the system efficiency and stage efficiency of the VentrAssist IRBP, steady flow tests were carried out using aque- ous glycerol and red cell suspension of human blood with a range of HCT in phosphate buffered saline solution (PBS) and Haemaccel (Fig. 4). The ASTM mock loop was used for testing all fluids whereas the shorter loops were used specifically for the nonbio- logical fluids and with the shafted version of the VentrAssist IRBP. The shafted version of the pump consists of a pump machined from an acrylic block such that the flow inside the pump can be visualized for particle tracing (PT) and particle image velocimetry (PIV). The rotor is mounted on a shaft and fitted with bear- Received December 1999. Presented in part at the 7th Congress of the International So- ciety for Rotary Blood Pumps, held August 26–27, 1999, in Tokyo, Japan. Address correspondence and reprint requests to Dr. Geoff Tansley, VentrAssist, 126 Greville Street, Chatswood 2067 NSW, Australia. E-mail: geoff.tansley@ventrassist.com Artificial Organs 24(6):478–482, Blackwell Science, Inc. © 2000 International Society for Artificial Organs 478