Saturday Morning, May 17, 1958

8:30 A.M. Scientific Session: THORACIC SURGERY FORUM - Imperial Ballroom

12. The Simplified Stationary Screen Pump Oxygenator.

Jerome Harold Kay (by invitation) and Robert M. Anderson

(by invitation), Los Angeles, Calif.

In 1956 at The Surgical Forum of The American College of Surgeons, one of us (JHK) described a simplified screen-type pump oxygenator with flow equal to normal cardiac output. This apparatus oxygenated blood by passing a film of blood on stationary screens in an atmosphere of oxygen. The blood was pumped by two roller pumps. The results with this apparatus have been very good. Recently we have greatly simplified and improved this apparatus so that the entire oxygenator, coronary sinus reservoir, venous reservoir, air trap and filter are combined into one chamber. This eliminates the connections on the arterial side and decreases the priming volume. One of the most important improvements consists of an automatic filmer. This completely eliminates saline to start a film. The apparatus is far simpler to use than a bubbler and is less traumatic to the blood elements. It can be run by one person. It is now being used clinically.

13. Comparison of Relative Merits of Occlusive and Non-Occlusive Pumps for Open Heart Surgery.

Paul C. Hodges (by invitation), C. Walton Lillehei, Richard Cardozo

(by invitation), Andre Thevenet (by invitation), Minneapolis, Minn.

Both occlusive and non-occlusive types of pumps have been recommended for open heart surgery. In general these recommendations have been based upon personal preference or sentiment rather than on factual comparative data.

At the University of Minnesota Heart Hospitals more than 450 patients have had open heart surgery utilizing a pump perfusion system. More than 50 of these clinical perfusions have exceeded a duration of 50 minutes. This clinical experience has been obtained using either an occlusive or non-occlusive pump of several designs. The clinical experience has been supplemented by an extensive laboratory investigation encompassing controlled observations on variables.

Analysis of this experimental and clinical data makes possible some definite conclusions as to the relative merits of occlusive versus non-occlusive pumps. These conclusions are based upon analysis of (1) ability to maintain blood pressure in the perfused subject, (2) evidence of blood trauma (hemolysis, protein denaturation, bleeding, and platelet destruction), (3) ease of flow regulation, and (4) operative risk.

14. Cardiac By-Pass Without Artificial Oxygenator.

W. T. Mustard (by invitation), W. Sapirstein (by invitation)

and Denise Pay (by invitation).

Sponsored by F. G. Kergin, Toronto, Canada

Cardiac by-pass is accomplished without the use of an oxygenator by perfusing the aerated lung fields. Avoiding an artificial oxygenator reduces extra-corporeal handling of the blood and consequently the effects of blood damage. An extended period of cardiac by-pass is thus permitted enabling intracardiac surgical correction of some of the more complicated defects.

An operative technique is described together with experimental results. Using conventional cardiac pumps, vena caval return is perfused through the pulmonary circuit. Oxygenated blood returning to the left atrium is cannulated into a reservoir from where it is further pumped into the systemic system. With the heart in asystole, a distended rubber balloon has been developed to occlude the mitral valve orifice, ensuring dry right and left ventricles even in the presence of septal defects.

The results of animal experiments suggest the superiority of this by-pass procedure for the treatment of selected cardiac anomalies over presently used systems with extracorporeal oxygenation.

15. Monitor and Control of Blood Oxygen Tensions and pH During Total Body Perfusion.

Leland C. Clark, Jr. (by invitation), Samuel Kaplan (by invitation),

Edward C. Matthews (by invitation), F. Kathryn Edwards (by invitation)

and james A. helmsworth, Cincinnati, Ohio

There is a need for a relatively simple method to continuously monitor the oxygenation of blood and carbon dioxide exchange during total body perfusion for direct vision intracardiac surgery. Measurements of blood saturation are unsatisfactory because during perfusion the oxygen tension of blood is frequently considerably above normal arterial tensions.

For the last several years we have used polarographic methods to record arterial and venous blood oxygen tensions during cardiopulmonary bypass. This method which is applicable to any extracorporeal system has the following advantages: (1) Continuous records of absolute oxygen tension over long periods of time are obtained. (2) The oxygen electrode adds only a few square millimeters of polyethelene to the system. (3) Neither the cathode nor the anode is in contact with the blood, and no voltage is impressed in the blood. (4) The electrode gives a linear response in current flow to a linear increase in oxygen content. (5) The speed of response is full scale in approximately five seconds. (6) The electrode is insensitive to changes in pH.

Continuous records of the pH of arterial blood are routinely obtained during cardio-pulmonary bypass. This information has been very helpful in the monitoring of carbon dioxide exchange in bubble oxygenators. Respiratory acidosis or alkalosis can be prevented by varying the gas flow and size of the oxygen bubbles in the extracorporeal circuit.

The details of these two methods will be described and their value will be demonstrated in the monitoring of human perfusions.

16. Experimental and Clinical Studies of Controlled Hypothermia Rapidly Produced and Corrected by a Blood Heat Exchanger During Extracorporeal Circulation.

Ivan W. Brown, Jr. (by invitation), Wirt W. Smith (by invitation),

W. Glenn Young (by invitation) and W. C. Sealy,

Durham, N. C.

A simple blood heat exchanger capable of use with any type of heart-lung machine has been developed by our laboratory. Experimental studies with the exchanger of the cooling and rewarming velocities of various tissues such as the brain, heart, kidney, liver, intestinal tract, and muscle masses have been determined for various body flow rates, inflow blood temperatures and arterial inflow sites. The exchanger permits a lowering of vital organ temperatures as rapid as 1.0° C. per minute for the range from 37° C. to 30° C. with a slightly slower rewarming velocity. In most instances, the mid-esophagus temperature closely follows the intracardiac temperature curve. Changes of heart, kidney, intestinal, cerebral, and mixed body A-V oxygen differences during cooling and rewarming were determined for various total body inflow rates. No toxic or deleterious effects on the blood have been noted on rapidly removing or adding heat to the inflowing blood over a wide range of temperatures from 0.5° C. to 42° C.A specially designed bubble trap used with the exchanger for excluding oxygen bubbles which might be released from the oxygen saturated rewarming plasma will also be described. A summary of our clinical data from more than 12 open heart operations will be presented which indicates the many complementary advantages of combining hypothermia and extracorporeal circulation particularly when controlled hypothermia can be achieved rapidly at any time during operation and similarly corrected.

17. Cerebral Blood Flow and Brain Volume Changes in Extra-corporeal Circulation.

Max M. Halley (by invitation), Keith Reemtsma (by invitation),

Manuel Bresler (by invitation) and Oscar Creech, Jr.,

New Orleans, La.

Cerebral Damage may be evident in the early postoperative period following extracorporeal circulation. This phenomenon is sometimes suggestive of cerebral edema, and may be completely reversible.

Two groups of animal experiments are being conducted in order to clarify the pathogenesis of the syndrome: The first group involves direct measurement of cerebral blood flow during extracorporeal circulation by means of a magnetic rotameter. It includes determination of cerebral oxygen consumption at different rates of flow; determination of the arterial-venous difference of glucose, pyruvate, and lactate in the cerebral circulation; and measurement of arterial and venous pO₂ and pCO₂.

The second group involves the measurement of central venous pressures and cerebro-spinal fluid pressure during extracorporeal circulation, and subsequent determination of brain volume changes by the method of White, immediately and at intervals after extracorporeal circulation. Correlation between brain volume changes and the variables of both groups is being attempted.

Results to date in the first group indicate that (a) Cerebral flow during extra-corporeal circulation is directly proportional to systemic blood pressure and is reduced to very low levels during hypotension (b) Perfusion rate is not a good index of cerebral flow (c) Cerebral oxygen consumption falls markedly at low levels of cerebral blood flow (d) Vasopressors, such as nor-adrenalin increase cerebral flow by elevating the systemic blood pressure (e) Glucose levels of venous and arterial cerebral blood show no definite variation during extracorporeal circulation (f) Increase in pCO₂ does not appear to increase cerebral flow.

Results in the second group to date indicate that (a) Brain volume may be significantly increased after extracorporeal circulation (b) Spinal fluid pressure may be elevated to various degrees during extracorporeal circulation, even in the presence of normal venous pressures, particularly at high rates of perfusion (c) Marked spinal fluid pressure increases occur in the presence of elevated venous pressures, particularly in the superior vena cava (d) No definite correlation has yet been established between venous or spinal fluid pressure changes and post-operative increase in brain volume.

18. Clotting Deviations in Man During Cardiac By-Pass: Fibrinolysis and Circulating Anticoagulant.

Kurt Von Kaulla (by invitation) and Henry Swan, Denver, Colo.

Serial coagulation studies before, during, and after heart surgery using a fixed screen oxygenator with DeBakey pumps reveal that two major deviations from the normal coagulation process may occur. One is a very marked increase in the fibrinolytic activity of the plasma reaching a peak during or immediately following the by-pass procedure. This was a persistent finding in patients with a perfusion rate of 55-70 ml/kg/min. (9 out of 9 patients). Fibrinolysis was less pronounced with a rate of 100-110 ml/kg/min. (1 out of 1), and was absent with a rate of 190-200 ml/kg/min. (2 out of 2). The euglobulin technique, permitting early recognition of fibrinolytic activity in heparinized blood, was used. The plasminogen activator in the urine rose to high levels shortly before or during the fibrinolytic phase; dropped to zero in the following hours; and returned to normal after several days. It is suggested that a tissue activator is released during the operative trauma, particularly during the pump-dependent phase, which subsequently activates the blood plasminogen. The intensity of the reaction may be related to the COa tension.

The second phenomenon, independent of the perfusion speed, is the appearance of a powerful anticoagulant as measured by the thrombin time. It develops after protamin neutralization of the administered heparin and may persist from minutes to hours. Neither fibrinolysis nor the circulating anticoagulant per se necessarily create clinical hemorrhage, but hemorrhage can occur if both phenomena overlap. In 2 patients out of 20, clinical hemorrhage was evident.

19. Elective Cardiac Arrest: The Relationship of Elevated Intra-cardiac Pressures to Subsequent Myocardial Function and Pathologic Pulmonary Changes.

John Ross, Jr. (by invitation), Joseph W. Gilbert, Jr. (by invitation),

Edward H. Sharp (by invitation) and Andrew G. Morrow,

Bethesda, Md.

Cardiac failure, ventricular fibrillation and pulmonary parenchymal changes sometimes follow elective cardiac arrest during cardiopulmonary bypass. A study was undertaken to define more precisely the central hemodynamic effects of perfusion with cardiac arrest and to determine the relationship of these changes to the above complications.

With a Melrose pump-oxygenator, 31 dogs were subjected to total cardio-pulmonary bypass. Pressures in the left atrium, right ventricle, and aorta were recorded throughout perfusion and recovery. Control and post-perfusion lung biopsies were obtained.

Seven of the 31 dogs underwent perfusion without cardiac arrest or cardiotomy. Left and right heart pressures remained low during perfusion, cardiac failure following bypass was not encountered, and post-perfusion lung biopsies were normal. In the remaining 24 dogs, cardiac arrest was induced with potassium citrate. In seven of these animals, right atriotomy was maintained throughout the arrest and recovery periods. Intracardiac pressures did not arise, and cardiac failure occurred in only one dog. In four dogs in which cardiac arrest was induced without atriotomy, left and right heart pressures showed progressive elevations during arrest. Following arrest, ventricular fibrillation occurred in two dogs, and three animals developed cardiac failure. Lung biopsies in two of these showed perivascular edema and hemorrhage. In the remaining 13 dogs, right atriotomy was performed immediately before arrest was terminated. Although the elevated intracardiac pressures returned to normal with right atriotomy, complications were no less frequent. Clamping the pulmonary artery during arrest prevented right ventricular pressure elevation, and augmented left atrial pressure increase.

During cardiac arrest bronchial arterial blood flows into the left heart, and also reaches the right heart through a collapsed pulmonary valve. Without cardiotomy, the resulting change in diastolic myocardial fiber length imposed by dilatation may often result in cardiac failure, sometimes associated with lung damage. Both left and right heart distension can be prevented by right atriotomy alone, provided the pulmonary valve remains incompetent. The importance of these findings in the management of patients undergoing elective cardiac arrest will be discussed.

20. Lung Lymph in Experimental Pulmonary Edema.

A. Robert Cordell (by invitation}, Winston-Salem, N. C., Richard A.

Bahn (by invitation), Buffalo, N. Y., James C. Stephens (by invitation),

Buffalo, N. Y. and H. H. Bradshaw, Winston-Salem, N. C.

A technique has been developed whereby lymph can be collected from the right lymphatic duct in the neck of dogs. It will be shown by means of dye injection studies that in approximately fifty per cent of dogs this lymph comes directly from the lungs without mixing with chyle from the thoracic duct. Thirty-five animals have undergone successful right lymphatic duct cannulation.

Control studies of lung lymph have been carried out on sixteen dogs with the following results: Flow rate ranges from 0.1 to 1.3 cc. per hour. Positive pressure respiration doubles this output in the average dog. Total protein content is approximately one-half that of circulating plasma. Sodium level averages 145 milli-equivalents per liter and chloride content 140 milli-equivalents per liter. Potassium level varies from 2.8 to 4.2 milligrams per liter.

Experimental pulmonary edema has been produced by the intravenous injection of alpha naphthyl thiourea in the dosage of 50 milligrams per kilogram.

Studies of pulmonary lymph during development of this fatal condition have shown the following results: Pulmonary lymphatic flow rises sharply to levels of 12 to 15 cc. per hour. Protein content shows a rise from 3.0 to 4.0 grams per cent. Sodium, potassium, and chloride concentrations show relatively little change.

Pulmonary edema has also been produced by means of creating aortopulmonary artery fistulae. Studies of lung lymph in these animals have shown close correlation with those from dogs in which pulmonary edema was drug-induced.

This technique is presently being used in the laboratory in an attempt to increase our knowledge of pulmonary lymphatic physio-pathology.

21. The Effect of Position on Pulmonary Ventilation.

Thomas F. Nealon, Jr. (by invitation) and

John H. Gibbon, Jr., Philadelphia, Pa.

It has been stated that it is more difficult to provide adequate pulmonary ventilation for anesthetized patients in the lateral position than in the supine position. As the lateral position is frequently used in thoracic surgery, we undertook to determine whether a greater minute volume ventilation was required in this position than in the supine position.

For the purposes of this study, adequate pulmonary ventilation was taken to be the minute volume ventilation required to maintain the pCO₂ of arterial blood at its preoperative level. The pCO₂ of the expired gas was measured continuously by an infra-red absorption carbon dioxide analyzer. Periodic checks were made by gas analysis of arterial blood samples. Ventilation was provided by an intermittent positive and negative pressure ventilator attached to a closed-circle anesthetic machine with a cuffed endotracheal tube. The minute volume ventilation was measured by a gas flow meter in the expiratory line. Determinations were made on the same patients in the supine and the lateral positions before and after the intrathoracic portion of the operation.

No increase in ventilation was found to be required when the patient was turned to the lateral position from the supine position. An open thoracotomy did not alter these findings when the lung was not retracted or compressed.

22. Aortic Homograft Replacement of the Main Pulmonary Artery.

George Robinson (by invitation), Philip Glotzer (by invitation),

Marvin Gilbert (by invitation) and Elliott S. Hurwitt,

New York, N. Y.

Two techniques have been developed in dogs for replacement of the main pulmonary artery and its bifurcation by an aortic homograft. One method includes the use of cardiopulmonary by-pass and elective cardioplegia. The second procedure may be accomplished without temporarily replacing the heart and lungs or interrupting the circulation, by employing a side-to-side anastomosis between the anterior wall of the main trunk of the pulmonary artery and the posterior wall of the mid-portion of the graft. Evaluation of the fate of the grafts in long-term surviving animals includes angiocardiography and histologic study. The potential clinical applications for pulmonary arterial replacement include atresia of the pulmonary artery (either as a solitary lesion or as a component of the Tetralogy of Fallot); chronic thrombotic occlusion of the pulmonary artery; coarctation of the pulmonary artery; and involvement of the pulmonary artery by aneurysms, tumors, or trauma.

23. Experimental Results with a Prosthetic Aortic Valve.

Benson B. Roe, John W. Owsley (by invitation) and Peter C.

Boudoures (by invitation), San Francisco, Calif.

A collapsible plastic intraluminal valve has been constructed for intravascular implantation and a suitable technique has been developed for placing this valve into the ascending aorta of experimental animals without total circulatory arrest or bypass. Three years' experience with various modifications of a tubular flexible tricuspid prosthetic valve has resulted in many improvements to reduce resistance and clotting. The physical properties of the silicone material are excellent in strength and resilience; stress studies reveal no significant fatigue, and no surface clotting or loss of flexibility has occurred after several months of function. Problems of adjacent clotting and embolization will be discussed and preventive measures presented.

Satisfactory coronary artery flow is maintained with a completely competent valve distal to the ostia. At least three dogs are living with valves functioning for six months and one survived secondary avulsion of the anatomical cusps. A large number of shorter survivors have demonstrated tolerance to the position of the prosthesis, and pressure studies indicate minimal resistance across the valve and total competence to pressures well above 300 mm. Hg.

Technical and physiological material will be presented to demonstrate the valve action and competence under mechanical stress, while functioning in the ascending aorta, and after long-term implantation.

24. Postmortem Perfusion Studies in the Evaluation of Techniques of Aortic Valvulotomy.

W. Gerald Austen (by invitation), Robert S. Shaw (by invitation),

W. M. Thurlbeck (by invitation) and J. Gordon Scannell,

Boston, Mass.

During the past two years the hearts of 25 patients dying with severe aortic stenosis have been studied by perfusion of the aortic valve so as to determine quantitatively the degree of aortic stenosis and regurgitation. The valves were subsequently restudied after the performance of reconstructive procedures. The following observations were made. (1) Closed or "blind" commissurotomy or dilatation usually results in a forbidding degree of regurgitation and inconsistent relief of stenosis, often with release of gross calcareous emboli. (2) Debridement and commissurotomy under direct vision consistently allows adequate relief of stenosis without producing significant regurgitation and without the uncontrolled release of emboli. (3) Anatomic identification of the commissure is the key to successful commissurotomy. After debridement accurate definition of the valvular landmarks is possible. (4) Inspection of normally functioning valves and perfusion studies have shown that valve leaflets may be normally fused to a point 2 mm. from the aortic wall and that division of adherent leaflets beyond this point affords no decrease in valvular resistance to forward flow but may increase the degree of regurgitation.

The application of these principles in a limited number of operative cases is described.

25. Intercoronary Collaterals in Normal Hearts.

Sven Bellman (by invitation) and Howard A. Frank, Boston, Mass.

For an understanding of the vascular response to the narrowing or occlusion of coronary arteries and the variation in the adequacy of this response from individual to individual, an accurate knowledge of the numbers, dimensions, locations, pathways, and connections of intercoronary collateral channels in normal hearts seems essential. This information should also be helpful in the planning of surgical efforts to increase coronary collaterals.

Most current knowledge of intercoronary communications in normal hearts has been gained by indirect means, such as the measurement of retrograde pressures and flows, or the demonstration of the interarterial passage of test substances. The connecting vessels themselves cannot usually be demonstrated in normal hearts by standard dissection, histological or radiological technics. Microradiography, applied stereoscopically following the injection of suitable contrast media, has been used successfully to demonstrate the finest vascular elements of many tissues and is well suited to an investigation of the intercoronary vessels.

The present paper presents the results of a post-mortem stereomicroradiographic study of the intercoronary vessels in normal human hearts, and, for correlation of experimental with clinical observations, in the hearts of commonly used laboratory animals as well.

26. Changes in Pulmonary Artery Pressure During Cardiopul-monary By-Pass: An Experimental Study.

James B. Littlefield (by invitation), J. Francis Dammann, Jr.,

Phyllis R. Ingram (by invitation) and William H. Muller, Jr.,

Charlottesville, Va.

Pulmonary complications associated with cardiopulmonary by-pass may be transient or result in fatal hemorrhage and edema. This experimental study was performed to evaluate pulmonary artery pressure changes during cardiopulmonary by-pass in dogs with a normal and increased collateral pulmonary circulation.

Acute experiments performed on seventeen dogs employed cardiopulmonary by-pass at intervals up to two hours, using a high flow, bubble, pump-oxygenator. The pulmonary artery and aorta were clamped simultaneously and cardiac standstill induced. Left auricular, pulmonary and femoral artery pressures were monitored and lung biopsies obtained. A right cardiotomy was employed. Left pulmonary artery ligation (termed "physiological pneumonec-tomy") performed in five dogs, several months before evaluation, simulated increased collateral pulmonary circulation seen in certain congenital cardiac patients.

SUMMARY:

1) Pulmonary hypertension was not a problem before standstill but persisted in the "physiological pneumonectomy" dogs during standstill, accompanied by severe hemorrhage.

2) Both groups of dogs showed intermittent pulmonary hypertension during cardiac recovery until the left ventricle functioned efficiently.

3) Accumulative periods of pulmonary hypertension may produce lung complications at any time during cardiopulmonary by-pass.

4) Pulmonary hypertension during cardiopulmonary by-pass may be reduced or prevented by: an unoccluded pulmonary artery, left auricular decompression, prompt return of left ventricular function, maintenance of normal systemic blood pressure and delayed cardiotomy closure.