Hemodynamic
Measurements
and
Dynamics
of
Assisted
Circulation
243
essential. Dynamic control of the assisted circulation can be performed
by, for instance, the utilization of state variable analysis and a conjugate
gradient optimization method. The chosen state variables can be the
fundamental components and phases of aortic pressure and flow with a
constraint for pressure and flow to return to normal. It
is
obvious that the
constraint necessity is for the fundamental components of aortic pressure
and flow to be 180 degrees out of phase. This allows maximal coronary
perfusion due to aortic diastolic pressure augmentation.
A
performance
index of normalized left ventricular power can then be used to assess the
efficacy of counterpulsation.
IABP
efficacy is dependent on selected performance indices that are
implemented based on the control algorithms. These also need to operate
in real-time for practical purposes. These indices should be evaluated on
a beat to beat basis and attempts made to either minimize or maximize
each index by adjusting balloon timing. Clark et al.
(1
973)
developed a
closed-loop control scheme to maximize a weighted performance index
comprised of mean systolic pressure, mean diastolic pressure, and end
diastolic pressure to optimize
IABP
performance. The algorithm was
implemented on both mini- and microcomputers. Timing was controlled
using a regression equation to predict the systolic time interval (STI)
from the measured heart rate. The algorithm gave reasonable results.
Other schemes and computer-based controller have also been applied,
based on weighted indices, such as coronary blood flow, tension time
index, end diastolic pressure, left ventricular stroke work, and cardiac
output, as well as left ventricular viability ratio and endocardia1 viability
ratio, to assess pumping efficacy.
Earlier, we developed a closed loop model of the cardiovascular
system in order to evaluate the most suitable control variables for
mechanical assistance (Puri et al.,
1982).
The model assumes that the
heart operates on the Frank-Starling mechanism and includes the carotid
sinus baroreceptor control. Their results showed that heart rate and left
atrial pressure
(LAP)
were the most sensitive parameters for the control
of assistance, and that the extent of assistance is dependent on the
severity of heart failure. The choice of the performance index, as well as
the weighting factors, is critical in determining
IABP
efficacy. There is
no totally automated system that presently exists clinically for closed-
