BIVENTRICULAR MECHANICAL BYPASS 
PHYSIOLOGICAL STUDIES 
DURING INDUCED VENTRICULAR FAILURE 
C. Grant LaFarge and W. F. Bernhard* 
r 
Effective reproducible ventricular failure models, with 
a low morbidity and mortality, have only recently been 
available for physiological evaluation of mechanical 
cardiac support systems. Initially, a number of methods 
were evaluated in this laboratory, and all rejected be- 
cause they failed to meet the criteria above: serial 
coronary artery ligation or carbon dioxide infusion (for 
acute left ventricular failure) ; local myocardial irra- 
diation with Strontium-90 or intoxication with cobaltous 
chloride ( for chronic failure). 
Later in our experience, two methods were developed 
and used successfully for evaluation of partial (left) 
and total (left and right) ventricular bypass systems. 
These studies employed pneumatically actuated, bivalved 
blood pumps (100 ml stroke volume) implanted and op- 
erating continuously in calves (65X90 Kg) for period up 
to 203 days. To produce acute left ventricular failure, 
a balloon cuff catheter was implanted around the main 
left coronary artery. At the time of physiological eval- 
uation (12 to 28 days postoperatively) inflation of the 
■ balloon catheter produced 5-20 minutes of stable, ven- 
tricular failure. Left ventricular end-diastolic pressure 
returned to control values following deflation of the bal- 
loon. 
For chronic irreversible ventricular failure, diffuse 
myocardial ischemia, and eventual fibrosis, was produced 
by graded infusion of latex microspheres (9 microns 
average diameter) into the main left coronary artery. 
Additionally, total cardiac failure was produced by 
electrically-induced ventricular fibrillation. The acute 
and chronic failure models proved to be effective and 
reproducible with low morbidity (balloon cufT) and 
mortality (both). In both ventricular failure and con- 
trol conditions, the ventricular blood pumps proved 
capable of maintaining cardiac output, while reducing 
atrial and ventricular systolic and diastolic pressures 
to low normal levels (or to zero mmHg). In the balloon 
cuff and microsphere-induced failure states, ventricular 
ejection fraction improved towards normal, and abnor- 
mally elevated ventricular volumes were reduced to the 
normal range. No significant hematological abnormali- 
ties were produced by pumping in any of the three 
states. 
INTRODUCTION 
Effective, reproducible ventricular failure 
models, with a low morbidity and mortality, 
* Children's Hospital Medical Center, Boston, Massachusetts. 
have long been the goal of investigators in- 
terested in the study of abnormal ventricular 
dynamics. The search for ideal models has re- 
cently been spurred by the development of a 
polyhedral family of cardiac assist and replace- 
ment devices. All of these must be evaluated 
in normal animals, and in animals with de- 
ranged ventricular function, prior to their hu- 
man clinical application. 
Heretofore, there have been few reliable 
ventricular failure models which could be ap- 
plied to artificial heart devices. For acute left 
ventricular failure (1 to 24 hours), variations 
on a coronary constriction theme have been ef- 
fective, but attended by greater than desirable 
mortality.^ 
Chronic ventricular failure models, in relia- 
ble, reproducible form, have been a real chal- 
lenge, rarely met in large animals, and based 
on a wide variety of techniques. Examples 
include the Barger dog,'* the Spann guinea pig,^ 
and the Munro and Marcus dogs." ' Unfortu- 
nately, small animals, including the dog, are not 
optimal for study of artificial heart devices : 
chest and heart size are much smaller, and clot- 
ting problems are monumentally greater than 
those of the human. 
Two problems required solution : animal size, 
and reliable, reproducible failure states suitable 
for evaluation of artificial heart devices. This 
report is concerned with some solutions to both 
problems. Between the large non-human pri- 
mates, the horse, the calf, the goat, and the 
sheep, it was quickly determined that the calf 
had the most numerous advantages : size, num- 
bers available, low cost, docility, relative free- 
dom from disease (individual and colony), and 
relative physiological similarity with human 
organ systems, including blood. 
Several methods of inducing myocardial fail- 
ure were tried and rejected: multiple coronary 
75 
