Biomedical Engineering Reference
In-Depth Information
FIGURE 8-40
VentrAssist VAD.
(a) Schematic
diagram showing
installation into a
human patient. (b)
Schematic diagram
of the pump.
[Adapted from
(Gosline 2004).]
Its strengths, compared with other similar designs, are that because of the large im-
peller size it rotates relatively slowly and the impeller has no shaft seals or bearings and has
clean flow lines with no stagnant zones. Initial tests made at a flow rate of 5 L/min and 100
mmHg differential pressure showed a normalized index of hemolysis of 0.002-.005 g/100L
(Watterson, Woodard et al., 1999). It was found that improving the surface finish to 0.2
m
improved this significantly, with a measured normalized index of hemolysis of 0.000167
+
/
−
0.00007 g/100L in whole human blood (James, Wilkinson et al., 2003).
The system consists of five components, shown in Figure 8-40:
μ
• The natural heart remains in place.
• A short inflow conduit is attached to left ventricle, which delivers blood from the
heart to the device.
• The outflow conduit from the pump delivers blood from the device to the ascending
aorta.
• The device is implanted in the “pump pocket,” which is located on the left side of
the body, behind the muscles of the abdominal wall and below the rib cage. The
drive line from the pump exits from the right side of the abdomen below the ribs.
• This connects the pump to the controller and batteries worn on an external belt or
backpack.
Rotating magnetic fields generated by the six copper coils in the base and walls of
the unit interact with the permanent magnets mounted within the rotor and cause it to spin
rapidly. Hydrodynamic forces, which result from the small clearances between the outside
surfaces of the impeller and the pump walls, support it. These small clearances range from
approximately 50 to 230
m (Chung, Zhang et al., 2004). Blood enters the center of the
pump and is spun up by the rotors and forced outward by centrifugal force, where it exits
through a pipe on the outside edge. A photograph of the device is shown in Figure 8-41.
Ideally, pump operating parameters are measured directly with pressure and flow
sensors, but in the case of LVADs the rationale is that additional complexity leads to
additional points of failure and a lower overall system reliability. Therefore, the controllers
of the VentrAssist and most other LVADs estimate them using the impeller speed and input
power for an assumed blood viscosity.
The VentrAssist device was first patented in 2001. The first human implant was made
in June 2003 at The Alfred Hospital in Melbourne. By 2008, more than 250 implants had
been made in 36 centers across 10 countries. At this time the product had achieved CE
μ
