An in vitro fluid dynamic study of pediatric cannulae: the value of animal studies to predict human flow.

A challenge to the development of pediatric ventricular assist devices (PVADs) is the use of the aortic cannulae attached to the devices. Cannulae used for pediatric application have small diameters and large pressure drops. Furthermore, during the development of the 12cc Penn State pediatric PVAD, particle image velocimetry (PIV) illustrated that hematocrit levels, through changes in blood viscoelasticity, affected the fluid dynamics. The objective of this study is to compare the fluid dynamics of a pediatric viscoelastic blood analog and a goat viscoelastic blood analog within the PVAD aortic cannula. Two acrylic models were manufactured to model the aortic cannula (6 mm and 8 mm diameters). PIV data was collected to examine the flow at the outlet of the VAD and in the aortic cannula at heart rates of 50 and 75 beats per minute (bpm). Three planes of data were taken, one at the centerline and two 1.5 mm above and below the centerline. Three more planes of data were taken orthogonal to the original planes. While a 75 bpm heart rate was used to represent normal operating conditions, a 50 bpm heart rate represented use of the PVAD during weaning. At 75 bpm, differences were evident between the two different fluids and the two models. Separation zones developed in the plane below the centerline for the higher hematocrit pediatric blood analog. This study raises question to the usefulness of animal testing results in regard to how well they predict the outcome of pediatric patients.