LEFT-HEART BYPASS
by

Clarence Dennis

Reprinted from

MECHANICAL DEVICES TO ASSIST THE FAILING HEART
NAS-NRC Publication 1283
17

Left-Heart Bypass

 

CLARENCE DENNIS

I SPEAK for Professor Ake Senning, Dr. Eric Carlens, Dr. S. A. Wesolowski,
Dr. D. P. Hall, and our many collaborators both in Stockholm and in
Brooklyn.

Mortality in patients with myocardial infarction and shock has been
reported to be in excess of 84 percent by many authors, among them
Biérck,? Wahlberg,!® and Malach and Rosenberg.’ The latter authors found
25 patients in 1 year at Kings County Hospital with myocardial infarction
and blood pressures of 70 mm Hg (systolic) or lower, or with blood pressures
more than 70 mm Hg below those known to exist before the coronary oc-
clusion, accompanied by signs of peripheral vascular collapse. These patients
all died, regardless of the type of therapy employed.

Upon reflection, the course of events in these patients is seen to be local
ischemia, leading to poor myocardial function, giving rise in turn to a
reduction in cardiac output and, secondarily, in systemic blood pressure.
Secondary to this, in turn, is poor perfusion of the still-unoccluded portions
of the coronary arterial tree and therefore a reduced possibility of survival
of marginally arterialized muscle. Quite apart from the cardiac aspects of
this chain of events, there is poor perfusion of the viscera generally, with
secondary development of irreversible shock. Any measures likely to lead
to survival in these patients not only must provide a greater head of pressure
for myocardial perfusion through the remaining patent arterial channels,
but must perfuse the viscera sufficiently to permit recovery from the endo-
toxin shock that accompanies the cardiac disaster. The presence of endo-
toxin in the peripheral blood when the blood pressure is low has been
established by Nelson and Noyes,!2 and the enhancement of survival in
endotoxin shock by left-heart bypass has been demonstrated by D. P. Hall
and associates.®

213
214 CARDIAC ASSISTANCE

Fic. 1. The transseptal cannula for
use in dog or man. Top: the entire
cannula. Bottom: left, closeup of
smooth tip of cannula; left center,
closeup of tip with aspirating needle
advanced as for confirmation of posi-
tion; right center, closeup of tip with
cutting lip and needle advanced; right,
closeup of lever mechanism for finger-
tip advancement of cutting lip. (In
part from Dennis, C., E. Carlens, A.
Senning, D. P. Hall, J. R. Moreno,
R. R. Cappelletti, and S. A. Wesolow-
ski. Clinical use of a cannula for left
heart bypass without thoracotomy.
Ann, Surg. 156:623-637, 1962. Re-
printed by permission of J. B. Lippin-
cott Co.)

 

When Senning, Hall, Moreno, and I began to work in Stockholm in 1960,
Senning and I had already resolved to pursue the physiological and tech-
nical problems of left-heart bypass. Our reasons lay, first, in the excessive
blood trauma involved in prolonged use of an oxygenator if heart-lung
bypass were to be undertaken, and second, in our being impressed by the
limited volume flows of veno-arterial pumping without oxygenation (Con-
nolly et al.*) and the disastrous cardiac response to the entrance of un-
saturated blood into the coronary system in the presence of heart failure
(Patt and associates'*). Finally, we were not yet ready to pursue counter-
pulsation as Clauss et al.3 had done. We were aware of Salisbury’s suggestion
that left-heart bypass might surpass heart-lung bypass for this purpose.™

Senning had observed benefit from partial left-heart bypass during the
15 to 45 min after open-heart repair of certain cardiac defects with the use
of coronary flow interruption. Inasmuch as Bevegard and his associates had
shown the feasibility of atrial septal puncture via the jugular vein for
diagnostic purposes,! Senning proposed that we use the welcome sabbatical
leave of one of us (C. D.) to work out jointly the problems of jugular, trans-
septal, left atrial cannulation without thoracotomy and to settle the question
of diminution of metabolic work of the heart by such bypass.

Exploration of the interior of the atria in a series of cadavers showed
that the anterior superior corner of the fossa ovalis provides the one con-
sistently safe point of puncture into the left atrium. Fortuitously, the
limbus of the fossa ovalis is one of the most striking landmarks to palpation,
and in appreciation of this our group was joined by Carlens in perfecting
the cannula (AB Stille-Werner, Stockholm, Sweden*) by which consistent
puncture at the desired site has been possible without fluoroscopy (Fig. 1).

* In the U.S., available from Ohio Chemical Co., Madison, Wisconsin.
LEFT-HEART BYPASS 215

The placement of the cannula in a patient requires him to be supine
with the head thrown slightly back and to the left. General anesthesia with
tracheal intubation is most commonly used. ‘The jugular vein is most easily
exposed by a vertical incision between the sternal and clavicular heads
of the sternomastoid muscle just above the inferior belly of the omohyoid,
well above the entrance of the numerous low cervical tributaries. The can-
nula is filled with isotonic saline solution, lubricated with a small amount
of thin mineral oil, and passed downward into the chest. When the tip has
been advanced to about 10 cm below the top margin of the clavicle, the
cannula is rotated so the tip is directed posteriorly and perhaps 15° medially,
following which, with light pressure in that direction, the tip is passed
caudally until it is felt to fall into the fossa ovalis. This occurs at a mean
distance of 13 cm below the top margin of the clavicle (standard deviation
from the mean of 13 cm is 1.3 cm, regardless of adult stature). In case of
doubt, the cannula can be passed farther down and withdrawn until the
arrest of the cannula tip by the limbus of the fossa ovalis is appreciated. In
the dog the relative distance below the clavicle is usually greater than in
man, and the site of puncture is usually found to lie almost precisely at the
midsternal level. Fluoroscopy is no longer used in either man or dog.

The external circuit is of 14-in. tubing and includes a flexible siphonage
chamber to avoid blood-air contact and a pump for return to the femoral
artery (Fig. 2). In the clinical circuit, a heat exchanger constitutes the
tubing leading to and from the patient and a flowmeter is included. Priming
volume is 650 ml, and maximum flow is 5 liters/min.

Fic. 2. The extracorporeal circuit. (Figures
2 and 3 in this presentation from Dennis,
C., D. P. Hall, J. R. Moreno, and A. Senning.
Left atrial cannulation without thoracotomy
for total left heart bypass. Acta Chir. Scand.
123:267-279, 1962. Reprinted by permission
of the publisher.)

 
216 CARDIAC ASSISTANCE

Fic. 3 (a) and (b). Pressure trac-
ings during institution of closed-
chest left-heart bypass.

 

pump. :

Totality of diversion of blood from the left atrium has been demonstrated
with pressure recordings and angiographic studies. Figure 3 shows a set
of pressure tracings recorded while a dog was being placed upon left-heart
bypass without thoracotomy. In Fig. 3a, aortic pressure rose within a few
seconds to levels above the peak left ventricular pressure. In Fig. 3b, 32 sec
after the start of bypass, an extrasystole emptied for the left ventricle, after
which left ventricular pressure remained far below aortic. Figure 4 shows
such a tracing upon termination of a period of total bypass by this circuitry,
but with the chest open to permit the recording of pressures in several
chambers. Here it is obvious that the left atrial pressure was lowered by left-
heart bypass and that, upon cessation of bypass, ventricular filling became
possible only as the left atrial pressure rose, secondary to termination of
siphonage from this chamber. Note the small pulse imposed by the pump
in the aortic tracing and the absence of the steep rise mentioned by Dr.
Rainer.

The angiocardiographic evidence of totality of left-heart bypass without
thoracotomy is presented in Fig. 5. As the figure indicates, after injection
of dye into the pulmonary artery, returning blood was totally withdrawn
from the left atrium, so that none entered the ventricle. Good perfusion
occurred in retrograde fashion through the aorta (d, e, f), so that both
coronary and cephalic vessels were well opacified (f).

The group of us has reported reduction of oxygen utilization on total
left-heart bypass by approximately 50 percent, with proportional decreases
on lesser degrees of bypass.7? We have also reported long-term survival of
nine out of 10 dogs after near-total bypass lasting 24 hr.* The sites of
LEFT-HEART BYPASS 217

puncture had healed completely in both of two dogs later held for 3 months
before sacrifice. In these dogs there was an average plasma hemoglobin level
below 80 mg percent at the end of bypass, and, in contrast to Dr. Rainer’s
observations, no drop in pH.

It is appropriate to mention here an experiment performed by the group
of us in Stockholm, which has not been published in full. We undertook to
see how long we could maintain a dog on nearly total left-heart bypass.
As had been the case in the 10 dogs subjected to 24-hr total or nearly total
left-heart bypass without thoracotomy, this dog did nicely for the first 24
hr. Thereafter, beginning at about 36 hr, he began to gain weight at the
expense of the blood transfusion required to maintain a satisfactory blood
volume. By the end of 83 hr of continuous bypass, he had developed enough
edema to increase his body weight by 33 percent. The edema was largely
centered about the shoulders, neck, and jowls, so that he more nearly re-
sembled a Cheshire pig than a shepherd dog. In the light of this observation,
our laboratory is beginning experiments to determine the role of pulsatile
aortic pressure in lymphatic flow. Wesolowski et al. demonstrated this to be
negligible for 4-hr perfusions a decade ago,’” but the more prolonged sup-
port we are now considering would appear to require re-evaluation of
this factor.

On the basis of the observations I have mentioned, our thesis is that both
shock secondary to coronary arterial occlusion and left-heart failure due
to valvular disease other than aortic insufficiency may well be ameliorated
by left-heart bypass without thoracotomy. The following established effects
of left-heart bypass are factors in this thesis: (1) maintenance of normal
systemic blood pressure, and therefore proper perfusion of all tissues except
that portion of the myocardium rendered ischemic by coronary occlusion;
(2) relief of pulmonary edema and attendant arterial unsaturation by
reduction of left atrial pressure to normal or below normal, repeatedly
demonstrated in clinical patients; (3) reduction of the oxygen need of the
myocardium, including the marginal myocardium rendered only partially
ischemic by the coronary block, and therefore expected salvage of more
muscle than without support; and (4) elevation of the oxygen tension in

  
  
   

. : bebop hed
Fic. 4. Pressure tracings in a dog as poppy

Dob bebe
open-chest left-heart bypass is termi- PN NNN
nated. (From Dennis, C., D. P. Hall, 200;—
J. R. Moreno, and A. Senning. Reduc- 150} Aoria
tion of the oxygen utilization of the a
heart by left-heart bypass. Circulat.
Res. 10:298-305, 1962. Reprinted by
permission of the American Heart As-
sociation, Inc.)

 
218 CARDIAC ASSISTANCE

 

Fic. 5. Angiographic evidence of totality of left-heart bypass without thoracotomy—a series of
films taken during 24 sec after injection of 50 ml of Urografin by catheter into the pulmonary
artery. (From Moreno, J. R., A. Senning, C. O. Ovenfors, C. Dennis, and D. P. Hall. Angio-
graphic studies during total left-heart bypass without thoracotomy. Acta Radiol. 57:17-23, 1962.
Reprinted by permission of the publisher.)

the myocardium, indicated by a rise in coronary sinus oxygen saturation on
institution of either partial or total left-heart bypass.

Our clinical experience is summarized in Table 1. The developmental
errors of the first four cases listed and the tardy arrival of the next four need
little comment and are failures not likely to recur as more experience and
collaboration are gained (Table 2). Among the last four listed in the table,
Case 12 was a man with a 10-day-old infarction, a duration not recognized
until autopsy, in whom massive hemopericardium apparently began
some 12 hr after institution of bypass. Improving circulation and urine
formation and occasional disappearance of heart block had provided real
encouragement until that time, when a rising venous pressure and per-
LEFT-HEART BYPASS 219

Tase |: Left-Heart Bypass

 

 

aie Lesion Duration of Immediate Comments
no. bypass, hr result
9 0 5.5 Died Collapse 4 days after fem. hernia c.
gangrene of gut; stress ulcer, no
infarct
10 0 4 Died Dx error; massive pulmon. embolism
2 ? 2 Died False passage with primitive cannula
4 ? 1 Died Attempted under local anesthesia;
vomited, drowned
7 ? 0 Died Cardiac massage 2 hr before arrival
6 ? . 0 Died Cardiac massage | hr before arrival
ll Lape and Circ. aa. 5 Died 4x vent. fibrill. before arrival; massage
during cannulation
5 LMG 4 Died Fibrill. before and during cannulation
(open massage)
12 LADG and PpCcA 17 Died Hemopericardium; path. 10-day-old
infarct with early aneurysm
3 Mitr. insuff. 16 Good Congestion cleared and circ. stable
pulm. edema without pump; Starr valve placed
later
1 Infarct sept. and 16 Imp. After 12 hr pt. cooled and vsp
post. wall sutured; died of failure of septal
closure
8 Infarct sept. and 7 Imp. Awoke stabilized; cannulas removed
post. wall 2 hr later; died of arrest 14 hr
thereafter

 

* Case 1: Dennis, Senning, Hall, and Moreno. Case 2: Dennis, Hall, and Moreno. Case 3: Hall
and Ellison, Augusta, Ga. Cases 6, 7, and 8: Carlens and Ekestrom, Stockholm. Cases 4, 5, 9, 10,
11, and 12: Brooklyn.

sistence of complete atrioventricular dissociation led us to continue bypass.
Hemopericardium was not recognized, and he was given up after 17 hr of
bypass. An already developing ventricular aneurysm might have made
pericardiocentesis hazardous. Case 3 suggests that left-heart bypass should
be thoroughly evaluated in patients with mitral disease who cannot securely
be prepared for definitive repair in other ways, for she regained cardio-
vascular stability and enjoyed several hours of respite between cessation of
bypass and beginning of thoracotomy for placement of the Starr valve, only
to die later of pulmonary hypertension.’ Our near success in Case 1, also,
offers hope that definitive repair of some complications of myocardial in-
farction may be feasible in the acute state. Finally, Case 8 suggests the hope
that some patients with myocardial infarction may respond sufficiently to
nothing more than temporary support, and survive, as did our first myo-
cardial-infarction patient placed on heart-lung bypass, presented by Dr.
Stuckey. Continuous monitoring with electrodes and direct-current con-
verter in place might have been successful in this instance.
220 CARDIAC ASSISTANCE

TABLE 2: Diagnostic and Developmental Errors in Left-Heart Bypass Cases

 

 

Serial no. Diagnostic errors
9 Collapse from postoperative bleeding; stress ulcer 1
10 Massive pulmonary embolism l

 

Developmental errors

 

2 False passage with primitive cannula 1
4 Local anesthesia, vomiting and drowning l

5, 6, 7, 11 Arrived too late and cannulation done during and after prolonged
cardiac massage 4
12 Hemopericardium after 10-hr bypass 1

3 Mitral insufficiency and congestion cleared by 16-hr bypass; died of
pulmonary hypertension after uneventful placement of Starr valve 1

1 Good circulation restored and postinfarction vsp repaired; died of
breakdown of repair 1
8 Stabilized on 7 hr of bypass; died 214 hr later of sudden arrest 1

 

At least three of the cases had sufficiently localized coronary occlusion
that hypothermia, coupled with addition of right-heart bypass, such as that
employed in Case | for repair of a ventricular septal defect, might have been
used for definitive plastic repair of the coronary arterial tree. Proper diag-
nostic methods should make this feasible.

In conclusion, left-heart bypass without thoracotomy is possible within
an hour of the decision to use it, and is theoretically sound. With the in-
creased collaboration now developing with our confreres in cardiology,
anesthesiology, and radiology, it should be possible either (1) to provide
temporary support for those in whom it would suffice; (2) to provide coro-
nary arterial plastic repair for those with sufficiently localized occlusions
and early enough in their attacks to have still viable muscle; (3) to excise
the infarct where it is necessary in later cases with localized infarction; or
(4) to exclude those for whom stage and extent of disease make survival im-
possible.

ACKNOWLEDGMENTS

The work reported here was supported by the State University of New York, United
States Public Health Service grants HE-1011 and HE-7028, American Heart Association
grant 60 G 70, and the Gelb-Strauss Fund for Medical Research.

REFERENCES

1. Bevegard, S., E. Carlens, B. Jonsson, and I. Karléf. A technique for transeptal left
heart catheterization via the right external jugular vein. Thorax 15:299-302, 1960.

2. Bidrck, G. Some fundamental problems in coronary heart disease. Acta Med. Scand.
168:245-268, 1960.
11.

12.

13.

14.

15.

16.

17.

LEFT-HEART BYPASS 221

Clauss, R. H., W. C. Birtwell, G. Albetal, $. Lunzer, W. J. Taylor, A. M. Fosberg,
and D. E. Harken. Assisted circulation. I. The arterial counterpulsator. J. Thor.
Cardiov. Surg. 41:447-458, 1961.

Connolly, J. E., M. B. Bacaner, D. L. Bruns, J. M. Lowenstein, and E. Storli.
Mechanical support of the circulation in acute heart failure. Surgery 44:255-262, 1958.
Dennis, C., E. Carlens, A. Senning, D. P. Hall, J. R. Moreno, R. R. Cappelletti, and
S. A. Wesolowski. Clinical use of a cannula for left heart bypass without thoracotomy:
experimental protection against fibrillation by left heart bypass. Ann. Surg. 156:623-
637, 1962.

Dennis, C., D. P. Hall, J. R. Moreno, and A. Senning. Left atrial cannulation without
thoracotomy for total left heart bypass. Acta Chir. Scand. 123:267-279, 1962.

Dennis, C., D. P. Hall, J. R. Moreno, and A. Senning. Reduction of the oxygen
utilization of the heart by left heart bypass. Circulat. Res. 10:298-305, 1962.

Hall, D. P., J. R. Moreno, C. Dennis, and A. Senning. An experimental study of
prolonged left heart bypass without thoracotomy. Ann. Surg. 156:190-196, 1962.
Hall, D. P. Personal communication.

Malach, M., and B. A. Rosenberg. Acute myocardial infarction in a city hospital. [II.
Experience with shock. Amer. J. Cardiol. 5:487-492, 1960.

Moreno, J. R., A. Senning, C. O. Ovenfors, C. Dennis, and D. P. Hall. Angiographic
studies during total left heart bypass without thoracotomy. Acta Radiol. 57:17-23,
1962.

Nelson, R. M., and H. E. Noyes. Permeability of the intestine to bacterial toxins in
hemorrhagic shock. Surg. Forum 3:474-480, 1952.

Patt, H. H., J. V. Clift, P. B. Loh, J. C. Roa, J. Wexler, and A. M. Seligman. Veno-
arterial pumping in normal dogs and dogs with coronary occlusion. J. Thor. Cardiov.
Surg. 39:464-467, 1960.

Salisbury, P. F., C. E. Cross, P. A. Rieben, and R. J. Lewin. Comparison of two types
of mechanical assistance in experimental heart failure. Circulat. Res. 8:431-439, 1960.
Stuckey, J. H., M. M. Newman, C. Dennis, E. H. Berg, S. E. Goodman, C. C. Fries,
K. E. Karlson, M. Blumenfeld, S. W. Weitzner, L. S. Binder, and A. Winston. The
use of the heart-lung machine in selected cases of acute myocardial infarction. Surg.
Forum 8:342-344, 1957.

Wahlberg, F. A study of acute myocardial infarction at the Seraphimer Hospital
during 1950-1959. Amer. Heart J. 65:749-757, 1963.

Wesolowski, S. A., L. R. Sauvage, and R. D. Pinc. Extracorporeal circulation: the
role of the pulse in maintenance of the systemic circulation during heart-lung by-
pass. Surgery 37:663-682, 1955.