Cardiopulmonary Bypass: Principles and Practice

Next Chapter >



Quick Links to Sections in this Chapter








C. W. Lillehei: Deceased, July 1999.

A physician at the bedside of a child dying of an intracardiac malformation as recently as 1952 could only pray for a recovery. Today, with the heart–lung machine, correction is routine. As a result, open heart surgery has been widely regarded as one of the most important medical advances of the 20th century. Its application is so widespread (2,000 such surgeries performed every 24 hours worldwide), performed so effortlessly, and carries such low risk at all ages that it may be difficult for the current generation of cardiologists and cardiac surgeons, much less the lay public, to appreciate that just 40 years ago the outer wall of the living human heart presented an impenetrable anatomic barrier to the surgeon's knife and to the truly incredible therapeutic accomplishments that are so commonplace today.

The keystone to this astonishing progress has been cardiopulmonary bypass (CPB) by extracorporeal circulation (ECC). These methods for ECC have allowed surgeons to empty the heart of blood, stop its beat as necessary, open any desired chamber, and safely carry out reparative procedures or even total replacement in an unhurried manner.

Beginning in 1951, a number of the developments that made the transition from the research laboratory to clinical open heart surgery possible and successful occurred in the Department of Surgery at the University of Minnesota (Table 1.1). This institution boasted two unequaled assets. One was the world's first heart hospital devoted entirely to the medical and surgical treatment of heart diseases. This 80-bed facility for pediatric and adult patients was donated to the University of Minnesota by the Variety Club of the Northwest and opened its doors to patients on July 1, 1951. The second, and perhaps even more important, advantage was the presence of Owen H. Wangensteen, a truly visionary surgeon, as Chairman of the Department of Surgery. He was not a cardiac surgeon but had made immense contributions in the field of general surgery by his innovative work in the treatment and prevention of bowel obstruction.

Over the years, beginning in 1930, he had evolved the unique "Wangensteen system" for the training of young surgeons. He placed a heavy emphasis on in-depth knowledge of the basic sciences and in research. He believed that this combination of a thorough grounding in the basic sciences and the insights gained by research gave young surgeons the confidence to disregard or abandon previously held ideas and traditions and to go forward on the basis of their own judgment and knowledge.

Proverbial also was his ability to spot talent and capabilities in his younger colleagues, whose aptitudes were not at all obvious to others—often not even to themselves. He would then proceed to develop that student using a combination of intellectual stimulation and material assistance.


Back to Quick Links


In this stimulating milieu, major accomplishments were soon forthcoming. The first of these occurred on September 2, 1952 when Dr. F. John Lewis, a medical school classmate and close personal friend, after a period of laboratory research on dogs successfully closed a secundum atrial septal defect (ASD) (1) in a 5-year-old girl under direct vision using inflow stasis and moderate total body hypothermia (Fig. 1.1).1 The date has considerable historical significance because that was the world's first successful operation within the open human heart under direct vision. Dr. Lewis had been inspired by Bigelow et al.'s experimental studies (2) on general body hypothermia as a technique for open heart repairs. Such operations became routine at the University of Minnesota Hospital, and news of these successes spread rapidly throughout the medical world. Swan et al. (3) were next to report successful direct-vision intracardiac operations in humans using general hypothermia and inflow stasis.

1This first patient had a normal postoperative heart catheterization. She is now the mother of two healthy children and remains entirely well, nearly 50 years after her operation.

FIG 1.1. Scene in the University of Minnesota Hospital operating room on September 2, 1952 near the end of the first successful open heart operation in medical history. On that date, Dr. F. John Lewis closed by suture an atrial secundum defect (2 cm in diameter) under direct visualization using inflow stasis and moderate total body hypothermia (26°C) in a 5-year-old girl who remains alive and well today. Postoperative heart catheterization confirmed a complete closure. She is the mother of two normal children.

Hypothermia with inflow stasis proved to be an excellent method for simple atrial defects. Lewis (4) reported in 1954 that eight of nine patients had their atrial septal defects successfully closed, with only one death. Later in 1954, Lewis et al. (5) reported on closure of ASD in 11 patients, with a mortality of only 18% (Table 1.2). By 1955, Lewis (6) reported 33 atrial septal defects closed at a 12.1% mortality rate compared with Gross' 30.2% mortality using blind techniques (atrial well). Also, the blind atrial well provided significantly fewer complete corrections (6).

Hypothermia also proved excellent for isolated congenital pulmonic or aortic stenoses (3). However, failure was uniform when this technique was applied to more complex lesions such as ostium primum, atrioventricularis communis, or ventricular septal defect (VSD). These experiences reconfirmed the oft-predicted need for a perfusion method for the more complex intracardiac lesions.


Back to Quick Links

Beginning efforts and then discouragement

The first attempts to use a heart–lung machine for total CPB to permit intracardiac surgery in humans were also carried out at the University of Minnesota Hospital by Dennis et al. on April 5, 1951 (7). Two patients were operated on within a month's time, but both died in the operating room. The first patient had an erroneous preoperative diagnosis despite two heart catheterizations and finger exploration of the heart's interior 5 months earlier. Instead of the anticipated ASD, she had an unexpected partial atrioventricularis communis lesion. This pathology was baffling at the time. The second patient, operated on 2 weeks later, had an ASD repaired but died intraoperatively from massive air embolism (8).

In both operations, the failures were related to the high perfusion rates that were considered necessary. In the first patient, in addition to the unfamiliar pathology, Dennis et al. stated that they were visually handicapped by "an amazing amount of blood" lost from the coronary sinus and thebesian vein and that "adjacent tissue anteriorly was employed to attempt closure in spite of the recognition of a good deal of encroachment upon the tricuspid orifice" (7). The cardiac specimen was studied by Dr. Jesse Edwards (9), the world-renowned cardiac pathologist, who found the tricuspid orifice had been severely stenosed in the attempt to close the ostium primum defect. In the second patient, the arterial reservoir was emptied suddenly (by the high flow), resulting in air being pumped into the patient's systemic circuit (8). Later in 1951, Dr. Dennis and many of his team moved to the University of New York (Brooklyn) to continue their work.

The next milestone was reached in May 1953 by Dr. John Gibbon, Jr. (10), who had started working on a pump oxygenator in the 1930s. He had developed his apparatus and techniques to the point where 12 of 20 dogs survived the closure of a surgically created VSD for 1 week to 6 months (11). By 1952 he believed he was ready to venture into the clinical area. His first patient had died,2 but the second case, with an atrial (secundum) defect, was operated on May 6, 1953 and was a complete success (12). This success was well received in a report in the lay press 12 days later (13) but aroused surprisingly little enthusiasm or interest among cardiologists and cardiac surgeons at the time, for several reasons. First, Dr. Gibbon duplicated only once Dr. Lewis' successes beginning 8 months earlier and could not repeat or extend his one success (12,14). Second, Dr. Lewis was regularly closing ASDs under direct vision using inflow stasis and moderate hypothermia with excellent results (1,4–6). Swan and colleagues (3,15) were also duplicating these excellent results with ASD. Third, and perhaps most important, was the fact that Dr. Gibbon, not repeating his one success with ASD or achieving success with the more complex VSD, became so discouraged after five failures (Table 1.3) that he abandoned open heart surgery as a means of repair of human heart lesions. That decision by the dean of the pioneering surgeons at that time had a profound effect in the minds of many investigators on the future of open heart surgery.

2Similar to the experience of Dennis, Gibbon's first patient selected for intracardiac surgery to close an atrial septal defect was a 15-month-old infant with an erroneous preoperative diagnosis. At operation, no septal defect was found. Autopsy disclosed a large unrecognized patent ductus arteriosus. Patient 2 was the 18-year-old girl with the successful atrial defect closure. Patients 3 and 4 were both 5.5-year-old girls operated on in July 1953, and both died intraoperatively. Patient 3 had an ASD, and repair was attempted. Patient 4 had been diagnosed preoperatively as having an atrial septal defect but also had a VSD and small patent ductus. Dr. Gibbon stated that "none of the defects could be repaired because of the flooding of the intracardiac field by blood" inside the bypassed heart (12). Kirklin (14) wrote (and also confirmed in a letter to me) that Gibbon operated on four patients after his May 1953 success and none survived. These last two patients were never reported by Gibbon or associates, and details are not available.

From 1951 to early 1954, there were many reported—and many more unreported—attempts to use CPB for intracardiac operations (Table 1.4). In all these reported clinical attempts at open heart operations, there was a common scenario: good-to-acceptable survival in the experimental animals but universal failure when the same apparatus and techniques were applied to humans. Thus, virtually all of the most experienced investigators of that era concluded with seemingly impeccable logic that the problems were not with the perfusion techniques or the heart–lung machines but that the "sick human heart," ravaged by failure, could not possibly be expected to tolerate the magnitude of the operations required and then recover immediately, with adequate output as occurred when the same machines and techniques were applied to dogs with healthy hearts. Thus, discouragement was rampant, and pessimism about the future of open heart surgery became widespread.

The prevalent belief was that the concept of open heart repair, however attractive, was doomed for patients with the more complex pathologic conditions who urgently required and would benefit the most from corrective procedures. What was necessary, many thought, was a means of mechanical support for the heart during its recovery period. Even today, more than three decades later, prolonged mechanical support of the failing heart during recovery still presents many unsolved problems.

A new outlook

During some canine experiments in which the cavae were temporarily occluded to test tolerance limits of the brain and heart to ischemia (20), it was discovered that if the azygos vein was not clamped (but all other inflow to the heart was), the resulting very small cardiac output (measured at 8 to 14 mL/kg body weight/min) (21) was sufficient to sustain the vital organs safely in every animal for a minimum of 30 minutes at normothermia. To even mention at that time that such a low flow might be adequate for perfusions was heresy; thus, we were pleased to learn of a similar observation in 1952 in England by Andreason and Watson (22). Both studies agreed that only about 10% of the so-called basal cardiac output was needed to sustain animals unimpaired physiologically for a reasonable period of time at normothermia.

From the earliest days, the universally accepted minimum flow for CPB (at normothermia) was considered by the authorities at that time to be 100 to 165 mL/kg per body weight/min in animals and humans (7,12,16,19,23). Our findings of this remarkable tolerance to drastically lower flows of only 8 to 14 mL/kg/min was very surprising, but the animal (dog) results were unmistakably clear. In analyzing these findings, we described at that time (1954) at least three identifiable important physiologic adjustments that were occurring in response to lowered blood flow (21). These compensating readjustments were additive and in their entirety at normal body temperatures accounted very well for the fact that these animals survived for 30 minutes or longer with their vital organs (brain, liver, heart, and kidneys) well protected.

At that time in our studies, we quickly found that "low flow" was a pejorative term and that advocacy of systemic flows much lower than the so-called basal cardiac output of 100 to 120 mL/kg/min was considered "totally wrong." What most clinicians and even physiologists did not appreciate was the simple fact that with the basal cardiac output, venous blood was returning with 65% to 75% of its oxygen content unused. There was no physiologic harm whatsoever in fully using the oxygen contained in the blood. Thus, the "azygos flow was really not low flow, but physiologic flow (21)."

Reducing the volume of blood necessary to be pumped had immediate and immense benefits. It has been observed repeatedly that one of the universal problems responsible in a very large part for the early failures with ECC by Dennis et al. (7), Gibbon (12), and Helmsworth et al. (16) was the enormous and unexpected blood return out of the open hearts due to well-developed systemic-to-pulmonary collaterals that made accurate vision almost impossible. Also, these unanticipated losses often made the perfusions physiologically precarious.

We immediately appreciated that the discovery of the azygos flow concept represented the sword that would eventually sever the Gordian knot of complexity that had garroted perfusion technology. I was convinced that some simple way could be found to successfully perfuse at only 20 to 25 mL/kg/min, which we set as a desirable flow rate with a comfortable safety margin. This low flow or physiologic flow quantity was only 10% to 20% of what others deemed necessary. Consequently, armed with this information in 1952, I believed that successful open heart surgery was not only possible but inevitable in the near future.

Autogenous lung for cardiopulmonary bypass

The low-flow principle made autogenous lung oxygenation much simpler and thus attractive. However, we found that the extra cannulas and tubing in the operative field were sensitive to even slight displacements, with the subsequent rapid onset of pulmonary edema. This rather frequent complication in our animal studies dampened our enthusiasm for potential clinical use (24). However, these venous drainage kinking problems led directly to the idea of moving these extrapulmonary cannulas completely out of the operative field by using a separate donor animal for oxygenation (cross-circulation). These experimental studies (24) convinced us that the autogenous lung was not a feasible route to pursue clinically, even with the significant advantages offered by the azygos flow concept.

The Dodrill experience with autogenous lung pump bypass

Dodrill et al. (25), in collaboration with the General Motors Corporation engineers in Detroit, developed a blood pump for animal and clinical use as a right, left, or combined heart bypass with autogenous lung oxygenator. All their reported clinical experiences are summarized in Table 1.5. In their series of four patients, three had partial heart bypasses (two left sides, one right side). All three lived but in only one (patient 2) was a therapeutic procedure (pulmonary valvuloplasty) carried out (17). The fourth (patient 3 in Table 1.5) had bypass of both sides of the heart but did not survive pulmonary valvuloplasty.

In their patients, Dodrill et al. used high flow rates (4,500 mL/min or about 56 to 64 mL/kg body weight. Those perfusion rates led to a significant amount of collateral flow within the bypassed heart, making it difficult to open the heart without sizable blood losses, which made therapeutic maneuvers difficult or impossible. For this and various other reasons, they did not report any further clinical work.

Controlled cross-circulation for cardiopulmonary bypass

Initially, our extracorporeal perfusions using cross-circulation in dogs had been intended only as an interim method to permit some open heart experience in animals without the need for a complex conventional pump-oxygenator, which was unavailable to us at the time. The term "controlled" refers to the use of a pump to precisely control the balance of the volume of blood flowing into and out of the donor and the patient. However, as the experiments progressed, it became apparent that the dogs undergoing a 30-minute open heart interval using physiologic flow with cross-circulation not only survived at a far higher rate but recovered far more rapidly when compared with the dogs we had observed undergoing a similar period of high-flow pump-oxygenator perfusions.3 The differences were truly astonishing, and for the first time we realized that this might be the simple and effective clinical method for intracardiac operations for which we were searching. The experimental and clinical data on cross-circulation perfusions and the reduced or physiologic perfusion flow rates based on the azygos flow studies have been documented elsewhere (27,28).

3In the years 1950 and 1951, Drs. C. Dennis and C.W. Lillehei had experimental laboratories next door to each other in the attic of the physiology building (Millard Hall) at the University of Minnesota Medical School.

Clinical application

Cross-circulation for clinical intracardiac operations was an immense departure from established surgical practice at the time (1954). The thought of taking a normal human being to the operating room to provide a donor circulation (with potential risks, however small), even temporarily, was considered unacceptable and even "immoral" by some critics. However, we had begun to suspect massive physiologic disturbances evoked by total body perfusion and open cardiotomy about which we knew very little and that by temporarily instituting a "placental" circulation we might minimize or even correct those to permit successful operations that would have otherwise been impossible (Fig. 1.2).

FIG 1.2. Direct-vision intracardiac surgery using extracorporeal circulation by means of controlled cross-circulation. A: The patient, showing sites of arterial and venous cannulations. B: The donor, showing sites of arterial and venous (superficial femoral and great saphenous) cannulations. C: The single Sigmamotor pump controlling precisely the reciprocal exchange of blood between the patient and donor. D: Close-up of the patient's heart, showing the vena caval catheter positioned to draw venous blood from both the superior and inferior venae cavae during the cardiac bypass interval. The arterial blood from the donor was circulated to the patient's body through the catheter that was inserted into the left subclavian artery.

The continued lack of any success in the other centers around the world that were working actively on heart–lung bypass (Table 1.4) and the widespread doubt about the feasibility of open heart surgery in humans contributed to our decision to go ahead clinically on March 26, 1954 (28) (Fig. 1.3). The cross-circulation technique was a dramatic success in humans (28–36). In the months that followed its first use to close a VSD, a rapid succession of surgical firsts occurred for correction of congenital heart defects that previously had been inoperable (Table 1.1). Cross-circulation as the means for ECC to permit work inside the human heart was used for 45 operations (Table 1.6). There was no donor mortality and no long-lasting donor sequelae (36).

FIG 1.3. The scene on March 26, 1954 in Operating Room B, University of Minnesota Medical Center, during the first controlled cross-circulation operation. At that time, a ventricular septal defect (VSD) was successfully visualized by ventricular cardiotomy and closed in a 12-month-old infant. The lightly anesthetized donor (the patient's father), with the groin cannulations serving as the extracorporeal oxygenator, may be seen to the far right (the patient is in the left foreground). Dr. C. W. Lillehei is immediately to the right of the scrub nurse, and opposite him is Dr. R. L. Varco. Behind Dr. Lillehei is Dr. H. E. Warden, and next to him, looking over the shoulder of the scrub nurse, is Dr. M. Cohen. Drs. Cohen and Warden are the two residents who had perfected this technique in the experimental dog laboratory. To Dr. Varco's right is Dr. J. B. Aust, an assistant resident. Behind Dr. Varco, at the left upper corner, is Dr. V. L. Gott, the surgical intern. Also, behind Dr. Varco is an observer, Dr. Norman Shumway, who was an assistant resident at the time. The VSD was closed by direct suture during a bypass interval of 19 minutes. The average flow rate was 40 mL/kg body weight/min at normothermia. The Sigmamotor pump that served to control the interchange of blood is located on Dr. Warden's right between donor and patient, but it is not visible in this photo.

Almost overnight, the "sick human heart theory" was refuted because the patients operated on with cross-circulation, mostly infants in terminal congestive failure, could not have been worse operative risks. Thus, after 15 years in the experimental laboratory, open heart surgery moved permanently into the clinical arena.

A three-decade follow-up

The follow-up of a minimum of 30 years on these first patients—with VSD, atrioventricularis communis, infundibular pulmonic stenosis, and tetralogy of Fallot—to have successful intracardiac corrections has been particularly informative and impressively sanguine (36). Twenty-eight of 45 (62%) patients undergoing ECC survived their operations and were discharged from the hospital. Even more impressive was the finding that only six of these survivors had died in 30 years. Thus, 22 of 45 patients (49%) initially operated on were alive 30 years later, and all were in good health.

The 27 patients with VSDs constituted the largest category to have repair, and 17 (63%) were living and well 30 or more years later. There were only two later VSD patient deaths in all of the years after hospital discharge. Both occurred in patients with closed defects but inexorable progression of their pulmonary vascular disease. Similarly, the late follow-up on the more complex tetralogy patients has been equally rewarding (35–38).

Cross-circulation was so successful because the donor automatically corrected all the various hematologic and metabolic derangements. At that time, we had no idea what these physiologic aberrations were and thus no knowledge about measuring them, much less treating them. In 1954 to 1955, pH and blood gases were not available clinically. Even emergency plasma electrolytes took 4 to 6 hours. There was no respiratory assistance equipment for infants or children, and there were no intensive care units, much less any monitors, pacemakers, or external defibrillators. In temporarily reconstituting the placental circulation with cross-circulation, we rediscovered the world's greatest intensive care unit: the intrauterine environment. It was some years before we could duplicate these remarkable results in equally sick patients using the pump oxygenator.


Back to Quick Links

For open intracardiac operations in humans to be regularly successful, workable solutions had to be identified for the three major obstacles that had stalled progress for so long. First, an effective method was needed for safely emptying the heart of blood for a reasonable length of time. ECC by cross-circulation fulfilled that need. Next, having gained access to the interior of the living human heart, it was soon evident that these malformations existed in a very broad spectrum and in many forms not yet described or even recognized by clinicians or pathologists. Surgical methods for dealing with these unfamiliar lesions required rapid technical development, often improvised on the spot, and sometimes with poignant failures. Moreover, given the existing state of technology, the preoperative diagnoses were often wrong or incomplete. Finally, these patients, often critically ill preoperatively, required postoperative care on a much higher level of sophistication than was known or available at the time.

Knowing now what we did not know in 1951 through 1954, it seems very probable that the only method for ECC that could possibly have succeeded so rapidly in the face of such formidable problems, in the face of such limited knowledge, and in the many high-risk infants and children with complex anatomic lesions was cross-circulation. The homeostatic mechanisms of the donor automatically corrected the untold number of mostly unknown physiologic aberrations evoked by total body perfusion.

Thirty years ago we wrote that "clinical experience with cross-circulation has made it apparent that it is unlikely that a technique for total cardiopulmonary bypass will be developed which excels this one for the patients' safety" (32). The spectacular success of clinical cross-circulation operations stimulated intensive laboratory work on alternative methods for CPB without the need for a living human donor.

Heterologous biologic oxygenators

Beginning on March 1, 1955, a series of clinical open heart operations was started at the University of Minnesota using a pair of canine lungs as oxygenators. Twelve patients were operated on, with 4 long-term survivors (39,40). Subsequent to those two reports, two more patients were operated on, for a total of 14, with 5 long-term survivors. In no patient was death attributable to oxygenator dysfunction. The only other attempt to use heterologous lungs at that time was the report of Mustard and associates (18,41) using monkey lungs. In their series of seven patients, there were no survivors. Mustard and Thomson (18) subsequently reported on surgery using monkey lungs in 21 infants and children having ECC between 1952 and 1956; there were three survivors in this series.

Extracorporeal circulation from a reservoir of oxygenated blood

Beginning March 3, 1955, the first of a series of five patients were operated on at the University of Minnesota for intracardiac repairs of VSD or transposition of the great vessels by continuous perfusion from a reservoir of oxygenated blood (29,42,43). This very simple technique was particularly applicable to infants needing relatively simple intracardiac repairs, thereby requiring lesser blood requirements.4 The arterialized venous blood for perfusion was drawn in the blood bank a few hours preoperatively using an ordinary venipuncture in donors whose arms had been immersed in water heated to 45°C for 15 minutes before collection, which effectively oxygenated the venous blood.

4On March 29, 1955, the patient, at age 6 months, weighing 4.7 kg, was operated on using reservoir perfusion for closure of a VSD. This patient reported on in 1955 (29,43) has now been followed for 37 years. He was recatheterized in 1964 with findings of normal pulmonary pressures and a completely closed defect. He graduated from college and was a Federal Bureau of Investigation agent for 20 years. He presently heads a private security company.


Back to Quick Links

In two publications in 1955 from the Mayo Clinic, Jones et al. (44) and Donald et al. (45) described their experimental results using the design of the Gibbon-type pump oxygenator as originally built by IBM and modified by the Mayo Clinic (Fig. 1.4). Their first clinical application was March 22, 1955, and a report followed on their first eight patients undergoing open heart surgery (46). Two of their four survivors had VSDs, one had an atrial septal defect, and one had an atrioventricularis communis canal. The four deaths occurred in patients with VSD (two), tetralogy (one), or atrioventricularis communis canal (one) (Table 1.7). Their flow rates varied from 100 to 200 mL/kg body weight. By September 1958, 245 patients had been operated on at the Mayo Clinic by Kirklin et al. (47). In their skillful and thoughtful hands, the initial high mortality declined rapidly.

FIG 1.4. The Mayo Clinic-Gibbon screen oxygenator. This model was used in 1955 during the first series of open heart operations performed by Dr. John Kirklin and associates at the Mayo Clinic, Rochester, MN. (Photo courtesy of J. W. Kirklin).

Advent of the DeWall-Lillehei bubble oxygenator

Before 1955, there was universal agreement among the world's authorities on ECC that the one way that blood could not be arterialized for clinical CPB was by a bubble oxygenator because of potential problems with air embolism. On May 13, 1955, DeWall and Lillehei, based on their dog laboratory research, began routine clinical use of a simple disposable bubble oxygenator (Fig. 1.5A). In their first report, Lillehei et al. (48) described surgery for seven patients with closure VSD, five of whom were long-time survivors. The operations took place at normothermia with perfusion rates (using a Sigmamotor pump, SIGMA-MOTOR, INC., Middleport, NY) of 25 to 30 mL/kg body weight. All seven patients awoke postoperatively, and there was no evidence of neurologic, hepatic, or renal impairment of even a temporary nature.

FIG 1.5. Evolution of the simple disposable DeWall-Lillehei bubble oxygenator for open heart surgery. A: The first 1955 clinical model; it was successful in infants and small children. B: Later in 1955, helix reservoir model with adult capacity was developed. C: A 1956 commercially manufactured model, shipped sterile in a package (left upper inset), ready to hang up and use.

In an addendum to that study (48), DeWall et al. (49) reported that 36 patients ranging in age from 16 weeks to 21 years had their hearts and lungs totally bypassed for intracardiac correction using their bubble oxygenator with similar excellent results. The congenital defects successfully corrected were VSD, tetralogy of Fallot, atrioventricularis communis canal, complete transposition, and ASDs. As the number of patients having open heart surgery by ECC increased rapidly, the bubble oxygenator was refined to increase capacity for adult patients (Figs. 1.5, B and C, and 1.6).

FIG 1.6. Top: Diagram of the 1955 DeWall-Lillehei helix reservoir, disposable bubble oxygenator with adult capacity. The upright oxygenating column with the venous blood mixing with oxygen bubbles formed at the base, transverse debubbling chamber, and the spiral (helix) debubbling reservoir immersed in a water bath are evident. The two insets show the wavelike pattern of the Sigmamotor pump's 12 metallic "fingers" as they stroke the blood through the plastic tubing without direct contact. Bottom: An open heart operation in an adult patient at the University of Minnesota Hospital in 1956.

In the early clinical open heart operations, considerable physiologic and biochemical data (50) were collected, analyzed, and compared with the earlier animal studies (51). This information confirmed the excellence of the patients' physiologic status while undergoing perfusions from the bubble oxygenator at the lower (more physiologic) flow rates based on the azygos flow concept. Tests done by psychologists and neurologists on these patients before and after perfusion detected no significant abnormalities in cerebral function attributable to the perfusions (52).

In a 26- to 31-year follow-up of 106 patients operated on for correction of tetralogy (37), 34 (32%) had college or graduate degrees, including two MDs, two PhDs, and one lawyer (LLB). Obviously, putting people on the bubble oxygenator could not be expected to increase intelligence, but these figures were far beyond the average for a random group from the general population and at the very least confirmed the absence of any significant cerebral dysfunction.

The DeWall-Lillehei bubble oxygenator was an instant success because it had so many practical advantages. It was efficient, inexpensive, heat sterilizable, easy to assemble and check, and had no moving parts. Because it could be assembled from commercially available materials at a small material cost, it was also disposable (Figs. 1.5 and 1.6). The development of the self-contained unitized plastic sheet oxygenator (Figs. 1.5C and 1.7) in 1956 by Gott et al. (53,54) further improved this system and played an important role in the tremendous expansion of open heart surgery that occurred after 1956. The revelation that safe perfusion of the body could be maintained with several lengths of plastic tubing, a few clamps, and some oxygen had an explosive effect on the worldwide development of cardiac surgery (Fig. 1.8). The surgeon's dream of routinely performing intracardiac correction in the open heart had become a reality (55) (Table 1.8).

FIG 1.7. A: A DeWall-Lillehei unitized plastic sheet oxygenator, commercially manufactured and shipped sterile ready to hang up, prime, and use as shown here. (Courtesy of D. A. Cooley, Texas Heart Institute, Houston, TX.) B: The Temptrol disposable bubble oxygenator with self-contained heat exchanger during a perfusion. In this unit, Dr. DeWall introduced the rigid presterilized plastic outer shell, which has been the basis of all subsequent oxygenator designs for both bubble and membrane units.

FIG 1.8. The ready availability of the simple and effective disposable Helix Reservoir Bubble Oxygenator had an explosive effect on worldwide growth of open heart surgery. Top: Dr. Denton Cooley with a perfusionist after an atrial septal defect closure, September 12, 1957, in Caracas, Venezuela. Bottom: The equipment used by Professor Pan Chih and associates for many successful open heart operations at the Shanghai Chest Hospital in China, 1957 to 1958. The first successful cardiac operations using cardiopulmonary bypass in both China and Japan were done with the DeWall-Lillehei bubble oxygenator.

In 1954, there was only one place in the world doing regularly scheduled open heart surgery by ECC: the University of Minnesota Hospital in Minneapolis (using cross-circulation). Beginning in May 1955 and well into 1956, there were only two places in the world performing these operations, the University of Minnesota in Minneapolis and the Mayo Clinic in Rochester, Minnesota, only 90 miles apart. Visitors from all parts of the world traveled to these two places to observe open heart surgery. On the one hand, there was the Mayo Clinic-Gibbon apparatus, which was very expensive, handcrafted, and very impressive in appearance but difficult to use and maintain. On the other hand, there was the unbelievably simple, disposable, heat-sterilized bubble oxygenator of DeWall and Lillehei, costing only a few dollars to assemble. It is no wonder, as Professor Naef (56) wrote, that "many surgeons left these two clinics with their minds in a totally confused state as to which method they should seek to pursue."

Dr. Denton Cooley (57), an early visitor and observer in June 1955, was to later write

The contrast between the two institutions and the two surgeons was striking. We observed Lillehei and a team composed mostly of house staff correct a ventricular septal defect using cross-circulation. During the visit, we also saw an oxygenator developed by Richard DeWall at the University of Minnesota. The next day we observed John Kirklin and his impressive team in Rochester that was made up of physiologists, biochemists, cardiologists, and others as they performed operations using the Mayo-Gibbon apparatus. Such a device was beyond my organizational capacity and financial reach. Thus I was deeply disappointed on our return to Houston when Dr. McNamara stated that he would not permit me to operate on his patients unless I had a Mayo-Gibbon apparatus.

However, Cooley succeeded in convincing some of his cardiologists that "the era of open heart surgery had arrived" and in 1956 began to perform open heart surgery using the DeWall-Lillehei bubble oxygenator with considerable success.

Professor Naef (56) also later wrote "the homemade helix reservoir bubble oxygenator of DeWall and Lillehei, first used clinically on May 13, 1955, went on to conquer the world and helped many teams to embark on the correction of malformations inside the heart in a precise and unhurried manner. The road to open-heart surgery had been opened."

Rotating disc film oxygenator (Kay-Cross)

After finishing his surgical training at the University of Minnesota in 1953, Dr. Frederick Cross5 moved to Cleveland, where he developed with Earl Kay in 1956 a rotating disc oxygenator that had wide use in the later 1950s, particularly in the United States. This oxygenator, called the Kay-Cross apparatus (58,59), was based on the earlier experimental work of Bjork (60). It had multiple vertical discs placed on a horizontal axis that rotated, with the discs dipping into a pool of venous blood, creating a film on the discs in an atmosphere of oxygen. This filming unit, like the Mayo Clinic-Gibbon film oxygenator, was capable of good oxygenation, but both, being nondisposable, shared similar problems: cumbersome to use, large priming volumes, very difficult and tedious to clean and sterilize, and rapid loss of efficiency if hemodilution was attempted. However, the Kay-Cross unit became commercially available in contrast to the Mayo Clinic-Gibbon machine, which was extremely expensive to handcraft and was not available commercially in those early years. Although this oxygenator certainly accomplished its purpose, the rotating disc mechanism had clinical limitations, in addition to its cumbersome assembly and maintenance. Disc oxygenation was later supplanted everywhere by bubble oxygenation, and perfusionists no longer had to spend all day setting up for the next surgery. The blood bank personnel were equally grateful at its passing.

5Dr. Cross was a surgery resident at the University of Minnesota who assisted in the world's first successful open heart operation, closure of an ASD under hypothermia, on September 2, 1952 (1).

Robert E. Gross, who pioneered cardiac surgery in 1938 with his brilliantly conceived ligation of patent ductus, began open heart surgery with CPB at the Boston Children's Hospital in 1956 (Table 1.9). For his first patient, who died after a pulmonary valvuloplasty, he had used a screen pump oxygenator modeled after the Gibbon machine but constructed in the Harvard laboratories by an engineer, Mr. Savage. Dr. Dwight Harken, who also used this oxygenator, stated that the unit was cumbersome, not heat sterilizable, required a very large priming volume, was inefficient as an oxygenator, and was extremely noisy (61).

Gross was also disturbed by the performance of this device and instituted a moratorium on open heart surgery in his unit for almost 8 months. He restarted CPB in January 1957 using the Kay-Cross oxygenator with high flow (2.3 to 2.5 L/m2/min) at normothermia and without hemodilution. In the next 3.5 months, 11 patients had been operated on with a disastrously high mortality rate of 67% (Table 1.9). Gross persisted with the same equipment and methods, and even at the time of his Shattuck Lecture to the Massachusetts Medical Society in Boston on May 20, 1959, his operative mortality remained very high. He reported a 13% early mortality for pulmonary valvuloplasty, a 40% mortality for VSD closure, and recommended against intracardiac correction in cyanotic tetralogies (63). Looking at these results is a clear reiteration of a lesson that many pioneer cardiac surgeons had to learn, "that even consummate surgical skill could not compensate for deficiencies in perfusion physiology."

The Kay-Cross filming unit appealed to the cardiac surgeons who at that time could not or refused to believe that bubble oxygenation was more efficient, safer, ideally adapted for hemodilution, yet vastly simpler to use and less expensive than the filming units. Even the Mayo Clinic had by 1971 converted almost entirely to the use of the bubble oxygenator (64). By 1976 it was estimated that 90% of all open heart operations worldwide involved the use of a bubble oxygenator (65).

Other oxygenators for CPB that were publicly known and worthy of mention but had moderate to ephemeral clinical applications were those of Rygg and Kyvsgaard6 (66) (bubble), Dennis (8) (film), Clark et al. (67) (bubble), Crafoord and Senning (68) (film and bubble), Clowes et al. (19) (bubble), Clowes and Neville (69) (membrane), Melrose (70) (film), and Gerbode et al. (71) (membrane). For more information on these devices, the reader is referred to the references cited as a starting point and also to two fine review articles by DeWall et al. (72,73).

6The Rygg bubble oxygenator (66) was a replica of the DeWall-Lillehei technology that was manufactured in Denmark and was used particularly in countries where U.S. patents did not apply.

Membrane oxygenators

Kolff et al. (74) described a disposable membrane oxygenator for experimental use in 1956. Clowes and Neville (69) described their experimental studies with membrane oxygenation and a complex apparatus they considered suitable for clinical perfusions in 1958. The belief that membrane oxygenation gives a better perfusion than the bubble or film oxygenators has been clear only with perfusions exceeding 8 hours in duration. Confusion has arisen over the innumerable comparative studies of membrane versus bubble oxygenators in shorter perfusions. With perfusions lasting 6 to 8 hours, the membrane oxygenator is associated with less reduction of platelets, less complement activation, less postoperative bleeding, and fewer microemboli. Because ECC times for most cardiac procedures are 2 to 3 hours or less, it has been difficult to prove that these changes, which are for the most part readily reversible, have any permanent side effects. Some studies have failed to show the theoretic benefits of membrane over bubble oxygenation (75–77), whereas other published data demonstrate improved hematologic tolerance of CPB with membrane oxygenators (78–80). I, with Lande et al. (81), described in 1967 the first compact, disposable, commercially manufactured membrane oxygenator for clinical use.

Further developments in bubble oxygenation

In 1966, DeWall et al. (82) made a very significant advance in oxygenator design with the introduction of a hard-shell bubble oxygenator with an integrated oxygenator and omnithermic heat exchanger in a disposable, presterilized, polycarbonate unit (Fig. 1.7). The adequacy of oxygenation and acid-base balance was amply documented (83,84). The integrated hard-shell concept has been the basis of all subsequent refinements, both in the bubble and membrane oxygenators (Fig. 1.9).

FIG 1.9. The Maxima (Medtronic) hollow-fiber membrane oxygenator is a widely used state of the art device. This disposable unit and similar competitive devices, such as those of Cobe, Terumo, Sarns, Shiley, and Bard, have rigid outer shells with integrated heat exchangers, easily attached venous reservoirs and cardiotomy suction chambers, low priming volumes, and efficient gas transfer. Their ease of use and more competitive price differentials versus the bubblers have resulted in increasing use for routine open heart procedures.

In the early days of open heart surgery, postoperative cerebral dysfunction was a subject of intense interest. As the major causes were identified and resolved (52), concern over this matter decreased. However, there has been a resurgence of interest in the detection and prevention of more subtle changes in personality and intellect that may be associated with an otherwise successful CPB (85–88). The reality and frequency of these changes and the need for continuous electroencephalographic monitoring for immediate correction of problems are under study (89).


A major technologic advance that has had an astonishing effect on the growth of ECC was the knowledge that the pump oxygenators could be primed with nonblood solutions, thereby immensely reducing the need for blood donors and at the same time improving the quality of perfusions by a reduction in viscosity and the safety by reducing foreign blood. Zuhdi et al. (90–92) developed the theory and process of hemodilution in 1961; they had trained in cardiac surgery at the University of Minnesota. DeWall et al. (93,94) confirmed the benefits of hypothermic hemodilution in ECC. Other hemodilution studies were reported from the Minnesota group that confirmed the value of low molecular weight dextran (95–97). Other comparative studies demonstrated the value of hemodilution with differing perfusates for improving renal blood flow and lessening hemolysis (98,99). Further, the beneficial effects of hemodilution and antiadrenergic drugs on the prevention of renal ischemia during ECC were confirmed (100,101).

Progress in pump design

In the earliest days of ECC, the multicam-activated Sigmamotor pump was used. Then the roller pumps, because of their ease of use and reliability, gained popularity. In more recent years, the centrifugal pump described by Rafferty et al. in 1968 (102) has become commercially available as the BioMedicus Biopump (BioMedicus, Inc., division of Medtronic, Inc., Minneapolis, MN) (Fig. 1.10). Some of the advantages of this pump are reliability, ease of use over a wide range of flows, less likelihood to pump air, absence of spallation, and low hemolysis. This pump was originally developed and used for perfusions lasting hours or days. However, surgeons in growing numbers have been impressed by the centrifugal pump's performance and advantages and have begun in increasing numbers to use it for routine ECC.

FIG 1.10. The BioMedicus disposable centrifugal flow blood pump, with ease of use over a wide range of flows and a number of other advantages (see text), has gained steadily increasing acceptance for routine open heart operations and also for longer time circulatory support. Three sizes are available.

As CPB became predictably reliable for a wide variety of congenital malformations, beginning in 1956 open cardiotomy was successfully applied to revolutionize the treatment of patients with acquired valvular heart disease (103,104) and subsequently to the treatment of an even larger group afflicted with coronary arteriosclerosis (105,106). By 1967, the ultimate landmark of successful human heart transplantation was reached by two surgeons, Drs. Barnard and Shumway, who had trained together in the late 1950s in my cardiac program at the University of Minnesota.

Today, the primary challenge is the need to further widen the benefits of heart replacement by both increasing the availability of donors and by an effective permanent intracorporeal mechanical heart. This latter seems a likelihood in the foreseeable future. Because of the shortage of donors, only 10% to 12% of potential recipients are being served. Short of a breakthrough with xenotransplantation (which is quite possible), the gap inevitably will have to be filled by a reliable, practical, fully implantable, total artificial heart.


Back to Quick Links

CPB by ECC for open heart surgery and even replacement of the heart itself were just dreams only 50 years ago. Today, after millions of total body perfusions, CPB has become a standard, widely used, low-risk procedure with immense benefits to humankind. We got where we are today by a worldwide catalytic combination of research, heterodoxy, and serendipity.


Back to Quick Links

  • Dr. F. John Lewis performed the first successful open heart operation (closure of ASD) using general hypothermia and inflow occlusion on September 2, 1952.

  • Other than one successful case by Dr. John H. Gibbon, Jr. in 1953, early clinical experience with CPB was discouraging and had unacceptably high morality rates.

  • The "azygos flow concept" led to the first clinical use of controlled cross-circulation for closure of VSD on March 26, 1954 by Dr. C. Walton Lillehei.

  • Dr. John W. Kirklin and coworkers at the Mayo Clinic modified the IBM-Gibbon screen film oxygenator and used it in a large series of patients in the mid to late 1950s.

  • Three major problems associated with early open heart surgery were finding an acceptable method for emptying the heart for reasonable lengths of time; unfamiliar pathology, inaccurate diagnoses, and new surgical techniques for effective repair; and lack of sophisticated postoperative care.

  • Despite concerns over the potential for air embolism, the DeWall-Lillehei helix bubble oxygenator was used clinically beginning in May 1955 in a large series of patients and became the method of choice worldwide for open heart operations.

  • Reasons for the success of the bubble oxygenator were its simplicity, efficiency, and low cost.

  • Commercially available, prepackaged, sterile bubble oxygenators of the unitized sheet and later hard-shell design with integral heat exchanger further popularized its use.

  • The rotating disc oxygenator, developed by Drs. Frederick Cross and Earl Kay, also was used widely for early open heart surgery in the United States.

  • Membrane oxygenators were developed and used clinically in the 1950s through the 1970s, but lack of demonstrable benefit for short CPB times contributed to their infrequent use by most groups until microporous designs became predominant in the mid-1980s.

  • Hemodilution with nonblood solutions was a major technologic advance in CPB that improved tissue perfusion, reduced hemolysis, and avoided donor blood exposure.


Back to Quick Links

    1. Lewis FJ, Taufic M. Closure of atrial septal defects with the aid of hypothermia; experimental accomplishments and the report of one successful case. Surgery 1953;33:52–59.

    2. Bigelow WG, Callaghan JC, Hopps JAGeneral hypothermia for experimental intracardiac surgery. Ann Surg 1950;132:531–539.

    3. Swan H, Zeavin I, Blount SG Jr., et al. Surgery by direct vision in the open heart during hypothermia. JAMA 1953;153:1081–1085.

    4. Lewis FJ In discussion: Bigelow WG, Mustard WT, Evans JGSome physiologic concepts of hypothermia and their applications to cardiac surgery. J Thorac Surg 1954;28:463–480.

    5. Lewis FJ, Varco RL, Taufic MRepair of atrial septal defects in man under direct vision with the aid of hypothermia. Surgery 1954;36:538–556.

    6. Lewis FJ In discussion: Watkins E, Gross REExperiences with surgical repair of atrial septal defects. J Thorac Surg 1955;30:469–491.

    7. Dennis C, Spreng DS Jr. , Nelson GE, et al. Development of a pump-oxygenator to replace the heart and lungs: an apparatus applicable to human patients and application to one case. Ann Surg 1951;134:709–721.

    8. Dennis CPerspective in review: one group's struggle with development of a pump-oxygenator. Trans Am Soc Artif Intern Organs 1985;31:1–11.

    9. Edwards JE Personal communication, 1951.

    10. Gibbon JH Jr. Artificial maintenance of circulation during experimental occlusion of pulmonary artery. Arch Surg 1937;34:1105–1131.

    11. Gibbon JH Jr. , Miller BJ, Dobell AR, et al. The closure of interventricular septal defects in dogs during open cardiotomy with the maintenance of the cardiorespiratory functions by a pump oxygenator. J Thorac Surg 1954;28:235–240.

    12. Gibbon JH Jr. Application of a mechanical heart and lung apparatus to cardiac surgery. Minn Med 1954;37:171–185.

    13. Historic Operation. Time News Magazine 1953 May:70.

    14. Kirklin JW. Open heart surgery at the Mayo Clinic—the 25th Anniversary. Mayo Clin Proc 1980;50:339–341.

    15. Swan H In discussion: Lewis FJ, Varco RL, Taufic M. Repair of atrial septal defects in men under direct vision with the aid of hypothermia. Surgery 1954;36:538–556.

    16. Helmsworth JA, Clark LC Jr. , Kaplan S, et al. An oxygenator pump for use in total bypass of heart and lungs. J Thorac Surg 1953;26:617–631.

    17. Dodrill FD, Hill E, Gerisch RA, et al. Pulmonary valvuloplasty under direct vision using the mechanical heart for a complete bypass of the right heart in a patient with congenital pulmonary stenosis. J Thorac Surg 1953;26:584–597.

    18. Mustard WT, Thomson JA. Clinical experience with the artificial heart lung preparation. J Can Med Assoc 1957;76:265–269.

    19. Clowes GHA Jr. , Neville WE, Hopkins A, et al. Factors contributing to the success of failure in the use of a pump oxygenator for complete bypass of the heart and lung, experimental and clinical. Surgery 1954;36:557–579.

    20. Cohen M, Hammerstrom RW, Spellman MW, et al. The tolerance of the canine heart to temporary complete vena caval occlusion. Surg Forum 1952;3:172–177.

    21. Cohen M, Lillehei CW. A quantitative study of the "azygos factor" during vena caval occlusion in the dog. Surg Gynecol Obstet 1954;98:225–232.

    22. Andreason AT, Watson F. Experimental cardiovascular surgery. Br J Surg 1952;39:548–551.

    23. Gibbon JH Jr. . The maintenance of life during experimental occlusion of the pulmonary artery followed by survival. Surg Gynecol Obstet 1939;69:602–614.

    24. Cohen M, Lillehei CW. Autogenous lung oxygenator with total cardiac bypass for intracardiac surgery. Surg Forum 1953;4:34–40.

    25. Dodrill FD, Hill E, Gerish RA. Some physiologic aspects of the artificial heart problem. J Thorac Surg 1952;24:134–150.

    26. Dodrill ARC, Hill E, Gerish A. Temporary mechanical substitute for the left ventricle in man. JAMA 1952;150:642–644.

    27. Warden HE, Cohen M, DeWall RA, et al. Experimental closure of intraventricular septal defects and further physiologic studies on controlled cross circulation. Surg Forum 1954;5:22–28.

    28. Warden HE, Cohen M, Read RC, et al. Controlled cross circulation for open intracardiac surgery. J Thorac Surg 1954;28:331.

    29. Lillehei CW, Cohen M, Warden HE, et al. Direct vision intracardiac surgery: by means of controlled cross circulation or continuous arterial reservoir perfusion for correction of ventricular septal defects, atrio-ventricularis communis, isolated infundibular pulmonic stenosis, and tetralogy of Fallot. In Proceedings of Henry Ford Hospital Symposium . Philadelphia: W.B. Saunders, 1955:371–392.

    30. Lillehei CW. Controlled cross circulation for direct vision intracardiac surgery correction of ventricular septal defects, atrioventricularis communis, and tetralogy of Fallot. Postgrad Med 1955;17:388–396.

    31. Lillehei CW, Cohen M, Warden HEE, et al. The direct vision intracardiac correction of congenital anomalies by controlled cross circulation; results in 32 patients with ventricular septal defects, tetralogy of Fallot, and atrioventricular communis defects. Surgery 1955;38:11–29.

    32. Lillehei CW, Cohen M, Warden HE, et al. Direct vision intracardiac surgical correction of the tetralogy of Fallot, pentalogy of Fallot, and pulmonary atresia defects: report of first ten cases. Ann Surg 1955;142:418–455.

    33. Lillehei CW, Cohen M, Warden HE, et al. The results of direct vision closure of ventricular septal defects in eight patients by means of controlled cross circulation. Surg Gynecol Obstet 1955;101:446–466.

    34. Lillehei CW, Cohen M, Warden HE, et al. Complete anatomical correction of the tetralogy of Fallot defects: report of a successful surgical case. Arch Surg 1956;73:526–531.

    35. Lillehei CW. A personalized history of extracorporeal circulation. ASAIO J 1982;28:5–16.

    36. Lillehei CW, Varco RL, Cohen M, et al. The first open heart repairs of ventricular septal defect, atrioventricular communis, and tetralogy of Fallot using extracorporeal circulation by cross circulation: a 30-year follow-up. Ann Thorac Surg 1986;41:4–21.

    37. Lillehei CW, Varco RL, Cohen M, et al. The first open heart corrections of tetralogy of Fallot. A 26–31 year follow-up of 106 patients. Ann Surg 1986;204:490–502.

    38. Gott VL, CWalton . Lillehei and total correction of tetralogy of Fallot. Ann Thorac Surg 1990;49:328–332.

    39. Campbell GS, Crisp NW, Brown EB. Total cardiac bypass in humans utilizing a pump and heterologous lung oxygenator (dog lungs). Surgery 1956;40:364–371.

    40. Campbell GS, Vernier R, Varco RL, et al. Traumatic ventricular septal defect. Report of two cases. J Thorac Surg 1959;37:496–501.

    41. Mustard WT, Chute AL, Keith JD, et al. A surgical approach to transposition of the great vessels with extracorporeal circuit. Surgery 1954;36:39–51.

    42. Warden HE, DeWall RA, Read RC, et al. Total cardiac bypass utilizing continuous perfusion from a reservoir of oxygenated blood. Proc Soc Exp Biol Med 1955;90:246–250.

    43. Warden HE, Read RC, DeWall RA, et al. Direct vision intracardiac surgery by means of a reservoir of "arterialized venous" blood. J Thorac Surg 1955;30:649–657.

    44. Jones RE, Donald DE, Swan HJC, et al. Apparatus of the Gibbon type for mechanical bypass of the heart and lungs. Proc Mayo Clin 1955;30:105–113.

    45. Donald DE, Harshbarger HG, Hetzel PS, et al. Experiences with a heart lung bypass (Gibbon type) in the experimental laboratory. Proc Mayo Clin 1955;30:113–115.

    46. Kirklin JW, DuShane JW, Patrick RT, et al. Intracardiac surgery with the aid of a mechanical pump oxygenator system (Gibbon type): report of eight cases. Proc Mayo Clin 1955;30:201–206.

    47. Kirklin JW, McGoon DC, Patrick RT, et al. What is adequate perfusion? In: Allen JG , ed. Extracorporeal circulation . Springfield, IL: Charles C Thomas, 1958:125–138.

    48. Lillehei CW, DeWall RA, Read RC, et al. Direct vision intracardiac surgery in man using a simple, disposable artificial oxygenator. Dis Chest 1956;29:1–8.

    49. DeWall RA, Warden HE, Read RC, et al. A simple, expendable, artificial oxygenator for open heart surgery. Surg Clin North Am 1956:1025–1034.

    50. DeWall RA, Warden HE, Gott VL, et al. Total body perfusion for open cardiotomy utilizing the bubble oxygenator. J Thorac Surg 1956;32:591–603.

    51. DeWall RA, Warden HE, Varco RL, et al. The helix reservoir pump-oxygenator. Surg Gynecol Obstet 1957;104:699–710.

    52. Hodges PC, Sellers RD, Story JL, et al. The effects of total cardiopulmonary bypass procedures upon cerebral function evaluated by the electroencephalogram and a blood brain barrier test. In: Allen JG , ed. Extracorporeal circulation . Springfield, IL: Charles C Thomas, 1958:279–294.

    53. Gott VL, DeWall RA, Paneth M, et al. A self-contained, disposable oxygenator of plastic sheet for intracardiac surgery. Thorax 1957;12:1–9.

    54. Gott VL, Sellers RD, DeWall RA, et al. A disposable unitized plastic sheet oxygenator for open heart surgery. Dis Chest 1957;32:615–625.

    55. Lillehei CW, Varco RL, Ferlic RM, et al. Results in the first 2,500 patients undergoing open heart surgery at the University of Minnesota Medical Center. Surgery 1967;62:819–832.

    56. Naef AP. The story of thoracic surgery . Toronto: Hografe and Huber, 1990:113–119.

    57. Cooley DA. Recollections of early development and later trends in cardiac surgery. J Thorac Cardiovasc Surg 1989;98:817–822.

    58. Cross FS, Berne RM, Hirose Y, et al. Description and evaluation of a rotating disc type reservoir oxygenator. Surg Forum 1956;7:274–278.

    59. Kay EB, Zimmerman HA, Berne RM, et al. Certain clinical aspects in the use of the pump oxygenator. JAMA 1956;162:639–641.

    60. Bjork VO. Brain perfusions in dogs with artificially oxygenated blood. Acta Chir Scand 1948;96[Suppl]:137.

    61. Harken DE. Personal communication, May 1992.

    62. Gross RE. Unpublished data, 1992.

    63. Gross RE. Shattuck lecture. Open heart surgery for repair of congenital defects. N Engl J Med 1959;260:1047–1057.

    64. Barnhorst DE, Moffitt EA, McGoon DC. Clinical use of the Bentley-Temptrol oxygenating system. In: Ionescu MI , Wooler GH , eds. Current techniques in extracorporeal circulation . London: Butterworths, 1976:91–116.

    65. Bartlett RH, Harken DE. Instrumentation for cardiopulmonary bypass-past, present, and future. Med Instrum 1976;10:119–124.

    66. Rygg IH, Kyvsgaard E. A disposable polyethylene oxygenator system applied in a heart-lung machine. Acta Chir Scand 1956;112:433–437.

    67. Clark LC Jr. , Gollan F, Gupta VB. The oxygenation of blood by gas dispersion. Science 1950;111:85–87.

    68. Crafoord CA, Senning A. Utvecklingen av extra-corporeal cirkulation med hjart-lungmaskin. Nordisk Med 1956;56:1263.

    69. Clowes GHA Jr. , Neville WE. The membrane oxygenator. In: Allen JG , ed. Extracorporeal circulation . Springfield, IL: Charles C Thomas, 1958:81–100.

    70. Melrose DM. A mechanical heart-lung for use in man. Br Med J 1953;2:57–66.

    71. Gerbode F, Osborn JJ, Bramson ML. Experiences in the development of a membrane heart-lung machine. Am J Surg 1967;114:16–23.

    72. DeWall RA, Grage TB, McFee AS, et al. Theme and variations on blood oxygenators. Surgery 1961;50:931–940.

    73. DeWall RA, Grage TB, McFee AS, et al. Theme and variations on blood oxygenators. II. Film oxygenators. Surgery 1962;51:251–257.

    74. Kolff WJ, Effler DB, Groves LJ, et al. Disposable membrane oxygenator (heart-lung machine) and its use in experimental surgery. Cleve Clin Q 1956;23:69–97.

    75. Edmunds LH Jr. , Ellison N, Colman RW, et al. Platelet function during cardiac operations. Comparison of membrane and bubble oxygenators. J Thorac Cardiovasc Surg 1982;83:805–812.

    76. Sade RM, Bartles DM, Dearing JP, et al. A prospective randomized study of membrane and bubble oxygenators in children. Ann Thorac Surg 1980;29:502–511.

    77. Trumbell HR, Howe J, Mottl K, et al. A comparison of the effects of membrane and bubble oxygenators on platelet counts and platelet size in elective cardiac operations. Ann Thorac Surg 1980;30:52–57.

    78. van den Dungen JAM, Karlicek GF, Brenken U, et al. Clinical study of blood trauma during perfusion with membrane and bubble oxygenators. J Thorac Cardiovasc Surg 1982;83:108–116.

    79. van Oeveren W, Kazatchkine MD, Descamps-Latscha B, et al. Deleterious effects of cardiopulmonary bypass. A prospective study of bubble versus membrane oxygenation. J Thorac Cardiovasc Surg 1985;89:888–899.

    80. Boers M, van den Dungen JJAM, Karlicek GF, et al. Two membrane oxygenators and a bubbler. A clinical comparison. Ann Thorac Surg 1983;35:455–462.

    81. Lande AJ, Dos SJ, Carlson RG, et al. A new membrane oxygenator-dialyzer. Surg Clin North Am 1967;47:1461–1470.

    82. DeWall RA, Bentley DJ, Hirose M, et al. A temperature controlling (omnithermic) disposable bubble oxygenator for total body perfusion. Dis Chest 1966;49:207–211.

    83. DeWall RA, Najafi H, Roden T. A hard shell temperature controlling disposable blood oxygenator. JAMA 1966;197:1065–1068.

    84. Kalke BR, Castaneda A, Lillehei CW. A clinical evaluation of the new Temptrol (Bentley) disposable blood oxygenator. J Thorac Cardiovasc Surg 1969;57:679–687.

    85. Aberg T, Kihlgren M, Jonsson L, et al. Improved cerebral protection during open-heart surgery. A psychometric investigation on 339 patients. In: Becker R , Katz J , Polonius M-J , Speidel H , eds. Psychopathological and neurological dysfunctions following open-heart surgery . Berlin: Springer-Verlag, 1982:343–351.

    86. Aberg T, Ahlund P, Kihlgren M. Intellectual function late after open heart operation. Ann Thorac Surg 1983;36:680–683.

    87. Henriksen L. Evidence suggestive of diffuse brain damage following cardiac operations. Lancet 1984;1:816–820.

    88. Shaw PJ, Bates D, Carlidge NEF, et al. Early neurological complications of coronary artery bypass surgery. Br Med J 1985;291:1384–1387.

    89. Arom KV, Cohen DE, Strobl FT. Effect of intraoperative intervention on neurological outcome based on electroencephalographic monitoring during cardiopulmonary bypass. Ann Thorac Surg 1989;48:476–483.

    90. Zuhdi N, McCollough B, Carey J, et al. Hypothermic perfusion for open heart surgical procedures—report of the use of a heart lung machine primed with five percent dextrose in water inducing hemodilution. J Int Coll Surg 1961;35:319–326.

    91. Zuhdi N, McCollough B, Carey J, et al. Double helical reservoir heart lung machine designed for hypothermic perfusion primed with five percent glucose in water inducing hemodilution. Arch Surg 1961;82:320–325.

    92. Zuhdi N Discussion of paper by Yeh TJ, Ellison LT, Ellison RG. Hemodynamic and metabolic responses of the whole body and individual organs to cardiopulmonary bypass with profound hypothermia. J Thorac Cardiovasc Surg 1961;42:782–792; 827–828.

    93. DeWall R, Lillehei CW. Simplified total body perfusion-reduced flows, moderate hypothermia, and hemodilution. JAMA 1962;179:430–434.

    94. DeWall R, Lillehei R, Sellers R. Hemodilution perfusions for open heart surgery. N Engl J Med 1962;266:1078–1084.

    95. Lillehei CW. Hemodilution perfusions for open heart surgery. Use of low molecular dextran and 5 percent dextrose. Surgery 1962;52:30–31.

    96. Cuello-Mainardi L, Bhanganada K, Mack JD, et al. Hemodilution in extracorporeal circulation: comparative study of low molecular weight dextran and 5 percent dextrose. Surgery 1964;56:349–354.

    97. Long DM Jr. , Todd DB, Indeglia RA, et al. Clinical use of dextran-40 in extracorporeal circulation—a summary of 5 years' experience. Transfusion 1966;6:401–403.

    98. Todd DB, Indeglia RA, Simmons RL, et al. Comparative clinical study of hemolysis and renal function accompanying extracorporeal circulation utilizing hemodilution with different perfusates. Am J Cardiol 1965;15:149. (abst).

    99. Nakib A, Lillehei CW. Assessment of different priming solutions for oxygenators by renal blood flow and metabolism. Ann Thorac Surg 1966;2:814–822.

    100. Lillehei CW, Simmons RL, Miller ID, et al. Role of hemodilution and phenoxybenzamine (Dibenzyline) in prevention of renal ischemia during cardiopulmonary bypass. Circulation 1965;31Suppl 2:II-138. (abst).

    101. Todd DB Jr. , Indeglia RA, Lillehei RC, et al. An analysis of some factors influencing renal function following open heart surgery utilizing hemodilution and antiadrenergic drugs. Am J Cardiol 1967;19:154. (abst).

    102. Rafferty EH, Kletschka HD, Wynyard M, et al. Artificial heart: application of nonpulsatile force-vortex principle. Minn Med 1968;51:11–16.

    103. Lillehei CW, Gott VL, DeWall RA, et al. Surgical correction of pure mitral insufficiency by annuloplasty under direct vision. Lancet 1957;77:446–449.

    104. Lillehei CW, Gott VL, DeWall RA, et al. The surgical treatment of stenotic or regurgitant lesions of the mitral and aortic valves by direct vision utilizing a pump oxygenator. J Thorac Surg 1958;35:154–191.

    105. Favaloro RG. Saphenous vein graft in the surgical treatment of coronary artery disease, operative technique. J Thorac Cardiovasc Surg 1969;58:178–185.

    106. Johnson WD, Flemma RS, Lepley D Jr. Direct coronary surgery utilizing multiple-vein bypass grafts. Ann Thorac Surg 1970;9:436–444.