The development of major surgery was retarded for centuries by a lack of knowledge and technology. Significantly, the general anesthetics ether and chloroform were not developed until the middle of the nineteenth century. These agents made major surgical operations possible, which created an interest in repairing wounds to the heart, leading some investigators in Europe to conduct studies in the animal laboratory on the repair of heart wounds. The first simple operations in humans for heart wounds soon were reported in the medical literature.

On July 10, 1893, Dr. Daniel Hale Williams (Fig. 1-1), a surgeon from Chicago, successfully operated on a 24-year-old man who had been stabbed in the heart during a fight. The stab wound was slightly to the left of the sternum and dead center over the heart. Initially, the wound was thought to be superficial, but during the night the patient experienced persistent bleeding, pain, and pronounced symptoms of shock. Williams opened the patient’s chest and tied off an artery and vein that had been injured inside the chest wall, likely causing the blood loss. Then he noticed a tear in the pericardium and a puncture wound to the heart “about one-tenth of an inch in length.”¹

The wound in the right ventricle was not bleeding, so Williams did not place a stitch through the heart wound. He did, however, stitch closed the hole in the pericardium. Williams reported this case 4 years later.¹ This operation, which is referred to frequently, is probably the first successful surgery involving a documented stab wound to the heart. At the time Williams’ surgery was considered bold and daring, and although he did not actually place a stitch through the wound in the heart, his treatment seems to have been appropriate. Under the circumstances, he most likely saved the patient’s life.

A few years after Williams’ case, a couple of other surgeons actually sutured heart wounds, but the patients did not survive. Dr. Ludwig Rehn (Fig. 1-2), a surgeon in Frankfurt, Germany, performed what many consider the first successful heart operation.² On September 7, 1896, a 22-year-old man was stabbed in the heart and collapsed. The police found him pale, covered with cold sweat, and extremely short of breath. His pulse was irregular and his clothes were soaked with blood. By September 9, his condition was worsening, as shown in Dr. Rehn’s case notes:

Ten years after Rehn’s initial repair, he had accumulated a series of 124 cases with a mortality of only 60%, quite a feat at that time.³

Dr. Luther Hill was the first American to report the successful repair of a cardiac wound, in a 13-year-old boy who was a victim of multiple stab wounds.⁴ When the first doctor arrived, the boy was in profound shock. The doctor remembered that Dr. Luther Hill had spoken on the subject of repair of cardiac wounds at a local medical society meeting in Montgomery, Alabama. With the consent of the boy’s parents, Dr. Hill was summoned. He arrived sometime after midnight with six other physicians. One was his brother. The surgery took place on the patient’s kitchen table in a rundown shack. Lighting was provided by two kerosene lamps borrowed from neighbors. One physician administered chloroform anesthesia. The boy was suffering from cardiac tamponade as a result of a stab wound to the left ventricle. The stab wound to the ventricle was repaired with two catgut sutures. Although the early postoperative course was stormy, the boy made a complete recovery. That patient, Henry Myrick, eventually moved to Chicago, where, in 1942, at the age of 53, he got into a heated argument and was stabbed in the heart again, very close to the original stab wound. This time, Henry was not as lucky and died from the wound.

Another milestone in cardiac surgery for trauma occurred during World War II when Dwight Harken, then a U.S. Army surgeon, removed 134 missiles from the mediastinum, including 55 from the pericardium and 13 from cardiac chambers, without a death.⁵ It is hard to imagine this type of elective (and semielective) surgery taking place without sophisticated indwelling pulmonary artery catheters, blood banks, and electronic monitoring equipment. Rapid blood infusion consisted of pumping air into glass bottles of blood.

Martin Kirschner reported the first patient who recovered fully after undergoing pulmonary embolectomy in 1924.⁶ In 1937, John Gibbon estimated that nine of 142 patients who had undergone the procedure worldwide left the hospital alive.⁷ These dismal results were a stimulus for Gibbon to start work on a pump oxygenator that could maintain the circulation during pulmonary embolectomy. Sharp was the first to perform pulmonary embolectomy using cardiopulmonary bypass, in 1962.⁸

Pericardial resection was introduced independently by Rehn⁹ and Sauerbruch.¹⁰ Since Rehn’s report, there have been few advances in the surgical treatment of constrictive pericarditis. Some operations are now performed with the aid of cardiopulmonary bypass. In certain situations, radical pericardiectomy that removes most of the pericardium posterior to the phrenic nerves is done.

Although cardiac catheterization is not considered heart surgery, it is an invasive procedure, and some catheter procedures have replaced heart operations. Werner Forssmann is credited with the first heart catheterization. He performed the procedure on himself and reported it in Klrinische Wochenschrift.¹¹ In 1956 Forssmann shared the Nobel Prize in Physiology or Medicine with Andre F. Cournand and Dickenson W. Richards, Jr. His 1929 paper states, “One often hesitates to use intercardiac injections promptly, and often, time is wasted with other measures. This is why I kept looking for a different, safer access to the cardiac chambers: the catheterization of the right heart via the venous system.”

In this report by Forssmann, a photograph of the x-ray taken of Forssmann with the catheter in his own heart is presented. Forssmann, in that same report, goes on to present the first clinical application of the central venous catheter for a patient in shock with generalized peritonitis. Forssmann concludes his paper by stating, “I also want to mention that this method allows new options for metabolic studies and studies about cardiac physiology.”

In a 1951 lecture Forssmann discussed the tremendous resistance he faced during his initial experiments.¹² “Such methods are good for a circus, but not for a respected hospital” was the answer to his request to pursue physiologic studies using cardiac catheterization. His progressive ideas pushed him into the position of an outsider with ideas too crazy to give him a clinical position. Klein applied cardiac catheterization for cardiac output determinations using the Fick method a half year after Forssmann’s first report.¹³ In 1930, Forssmann described his experiments with catheter cardiac angiography.¹⁴ Further use of this new methodology had to wait until Cournand’s work in the 1940s.

The first clinical attempt to open a stenotic valve was carried out by Theodore Tuffier on July 13, 1912.¹⁵ Tuffier used his finger to reach the stenotic aortic valve. He was able to dilate the valve supposedly by pushing the invaginated aortic wall through the stenotic valve. The patient recovered, but one must be skeptical as to what was accomplished. Russell Brock attempted to dilate calcified aortic valves in humans in the late 1940s by passing an instrument through the valve from the innominate or another artery.¹⁶ His results were poor, and he abandoned the approach. During the next several years, Brock¹⁷ and Bailey and colleagues¹⁸ used different dilators and various approaches to dilate stenotic aortic valves in patients. Mortality for these procedures, which was often done in conjunction with mitral commissurotomy, was high.

Elliott Cutler worked for 2 years on a mitral valvulotomy procedure in the laboratory. His first patient underwent successful valvulotomy on May 20, 1923, using a tetrasomy knife.¹⁹ Unfortunately, most of Cutler’s subsequent patients died because he created too much regurgitation with his valvulotome, and he soon gave up the operation.

In Charles Bailey’s 1949 paper entitled, “The Surgical Treatment of Mitral Stenosis,” he states, “After 1929 no more surgical attempts [on mitral stenosis] were made until 1945. Dr. Dwight Harken, Dr. Horace Smithy, and the author recently made operative attempts to improve mitral stenosis. Our clinical experience with the surgery of the mitral valves has been five cases to date.” He then describes his five patients, four of whom died and only one of whom lived a long life.^20,21

A few days after Bailey’s success, on June 16 in Boston, Dr. Dwight Harken successfully performed his first valvulotomy for mitral stenosis.²²

The first successful pulmonary valvulotomy was performed by Thomas Holmes Sellers on December 4, 1947.²³

Charles Hufnagel reported a series of 23 patients starting September 1952 who had operation for aortic insufficiency.²⁴ There were four deaths among the first 10 patients and two deaths among the next 13. Hufnagel’s caged-ball valve, which used multiple-point fixation rings to secure the apparatus to the descending aorta, was the only surgical treatment for aortic valvular incompetence until the advent of cardiopulmonary bypass and the development of heart valves that could be sewn into the aortic annulus position.

Congenital cardiac surgery began when John Strieder at Massachusetts General Hospital first successfully interrupted a ductus on March 6, 1937. The patient was septic and died on the fourth postoperative day. At autopsy, vegetations filled the pulmonary artery down to the valve.²⁵ On August 16, 1938, Robert Gross, at Boston Children’s Hospital, operated on a 7-year-old girl with dyspnea after moderate exercise.²⁶ The ductus was ligated and the patient made an uneventful recovery.

Modifications of the ductus operation soon followed. In 1944, Dr. Gross reported a technique for dividing the ductus successfully. The next major congenital lesion to be overcome was coarctation of the aorta. Dr. Clarence Crafoord, in Stockholm, Sweden, successfully resected a coarctation of the aorta in a 12-year-old boy on October 19, 1944.²⁷ Twelve days later he successfully resected the coarctation of a 27-year-old patient. Dr. Gross first operated on a 5-year-old boy with this condition on June 28, 1945.²⁸ After he excised the coarctation and rejoined the aorta, the patient’s heart stopped suddenly. The patient died in the operating room. One week later, however, Dr. Gross operated on a second patient, a 12-year-old girl. This patient’s operation was successful. Dr. Gross had been unaware of Dr. Crafoord’s successful surgery several months previously, probably because of World War II.

In 1945, Dr. Gross reported the first successful case of surgical relief for tracheal obstruction from a vascular ring.²⁹ In the 5 years that followed Gross’s first successful operation, he reported 40 more cases.

The famous Blalock-Taussig operation also was first reported in 1945. The first patient was a 15-month-old girl with a clinical diagnosis of tetralogy of Fallot with a severe pulmonary stenosis.³⁰ At age 8 months, the baby had her first cyanotic spell, which occurred after eating. Dr. Helen Taussig, the cardiologist, followed the child for 3 months, and during that time, cyanosis increased, and the child failed to gain weight. The operation was performed by Dr. Alfred Blalock at Johns Hopkins University on November 29, 1944. The left subclavian artery was anastomosed to the left pulmonary artery in an end-to-side fashion. The postoperative course was described as stormy; the patient was discharged 2 months postoperatively. Two additional successful cases were done within 3 months of that first patient.

Thus, within a 7-year period, three congenital cardiovascular defects, patent ductus arteriosus, coarctation of the aorta, and vascular ring, were attacked surgically and treated successfully. However, the introduction of the Blalock-Taussig shunt probably was the most powerful stimulus to the development of cardiac surgery because this operation palliated a complex intracardiac lesion and focused attention on the pathophysiology of cardiac disease.

Anomalous coronary artery in which the left coronary artery communicates with the pulmonary artery was the next surgical conquest. The surgery was performed on July 22, 1946, and was reported by Gunnar Biorck and Clarence Crafoord.³¹ The anomalous coronary artery was identified and doubly ligated. The patient made an uneventful recovery.

Muller³² reported successful surgical treatment of transposition of the pulmonary veins in 1951, but the operation addressed a partial form of the anomaly. Later in the 1950s, Gott, Varco, Lillehei, and Cooley reported successful operative variations for anomalous pulmonary veins.

Another of Gross’s pioneering surgical procedures was surgical closure of an aortopulmonary window on May 22, 1948.³³ Cooley and colleagues³⁴ were the first to report on the use of cardiopulmonary bypass to repair this defect and converted a difficult and hazardous procedure into a relatively straightforward one.

Glenn³⁵ reported the first successful clinical application of the cavopulmonary anastomosis in the United States in 1958 for what has been termed the Glenn shunt. Similar work was done in Russia during the 1950s by several investigators. On January 3, 1957, Galankin,³⁶ a Russian surgeon, performed a cavopulmonary anastomosis in a 16-year-old patient with tetralogy of Fallot. The patient made a good recovery with significant improvement in exercise tolerance and cyanosis.

The development of the heart-lung machine made repair of intracardiac lesions possible. To bypass the heart, one needs a basic understanding of the physiology of the circulation, a method of preventing the blood from clotting, a mechanism to pump blood, and finally, a method to ventilate the blood.

One of the key requirements of the heart-lung machine was anticoagulation. Heparin was discovered in 1915 by a medical student, Jay McLean, working in the laboratory of Dr. William Howell, a physiologist at Johns Hopkins.³⁷

John Gibbon contributed more to the success of the development of the heart-lung machine than anyone else.

Gibbon’s work on the heart-lung machine took place over 20 years in laboratories at Massachusetts General Hospital, the University of Pennsylvania, and Thomas Jefferson University. In 1937, Gibbon reported the first successful demonstration that life could be maintained by an artificial heart and lung and that the native heart and lungs could resume function. Unfortunately, only three animals recovered adequate cardiorespiratory function after total pulmonary artery occlusion and bypass, and even they died a few hours later.³⁸ Gibbon’s work was interrupted by World War II; afterward, he resumed his work at Thomas Jefferson Medical College in Philadelphia (Table 1-1).

Forest Dodrill and colleagues used the mechanical blood pump they developed with General Motors on a 41-year-old man⁴³ (Fig. 1-3). The machine was used to substitute for the left ventricle for 50 minutes while a surgical procedure was carried out to repair the mitral valve; the patient’s own lungs were used to oxygenate the blood. This, the first clinically successful total left-sided heart bypass in a human, was performed on July 3, 1952, and followed from Dodrill’s experimental work with a mechanical pump for univentricular, biventricular, or cardiopulmonary bypass. Although Dodrill and colleagues had used their pump with an oxygenator for total heart bypass in animals,⁵⁴ they felt that left-sided heart bypass was the most practical method for their first clinical case.

Later, on October 21, 1952, Dodrill and colleagues used their machine in a 16-year-old boy with congenital pulmonary stenosis to perform a pulmonary valvuloplasty under direct vision; this was the first successful right-sided heart bypass.⁴⁴ Between July 1952 and December 1954, Dodrill performed approximately 13 clinical operations on the heart and thoracic aorta using the Dodrill—General Motors machine, with at least five hospital survivors.⁵⁵ Although he used this machine with an oxygenator in the animal laboratory, he did not start using an oxygenator with the Dodrill—General Motors mechanical heart clinically until early 1955.

Hypothermia was another method to stop the heart and allow it to be opened.⁴⁴

John Lewis closed an atrial septal defect (ASD) in a 5-year-old girl on September 2, 1952 using a hypothermic technique.⁴⁴

The use of systemic hypothermia for open intracardiac surgery was relatively short-lived; after the heart-lung machine was introduced clinically, it appeared that deep hypothermia was obsolete. However, during the 1960s it became apparent that operative results in infants under 1 year of age using cardiopulmonary bypass were poor. In 1967, Hikasa and colleagues,⁵⁶ from Kyoto, Japan, published an article that reintroduced profound hypothermia for cardiac surgery in infants and used the heart-lung machine for rewarming. Their technique involved surface cooling to 20°C, cardiac surgery during circulatory arrest for 15 to 75 minutes, and rewarming with cardiopulmonary bypass. At the same time, other groups reported using profound hypothermia with circulatory arrest in infants with the heart-lung machine for cooling and rewarming. Results were much improved, and subsequently the technique also was applied for resection of aortic arch aneurysms.

After World War II, John Gibbon resumed his research. He eventually met Thomas Watson, chairman of the board of the International Business Machines (IBM) Corporation. Watson was fascinated by Gibbon’s research and promised help. Soon afterward, six IBM engineers arrived and built a machine that was similar to Gibbon’s earlier machine, which contained a rotating vertical cylinder oxygenator and a modified DeBakey rotary pump. Gibbon operated on a 15-month-old girl with severe congestive heart failure (CHF). The preoperative diagnosis was ASD, but at operation, none was found. She died, and a huge patent ductus was found at autopsy. The next patient was an 18-year-old girl with CHF owing to an ASD. This defect was closed successfully on May 6, 1953, with the Gibbon-IBM heart-lung machine. The patient recovered, and several months later the defect was confirmed closed at cardiac catheterization.⁵⁷ Unfortunately, Gibbon’s next two patients did not survive intracardiac procedures when the heart-lung machine was used. These failures distressed Dr. Gibbon, who declared a 1-year moratorium for the heart-lung machine until more work could be done to solve the problems causing the deaths.

During this period, C. Walton Lillehei and colleagues at the University of Minnesota studied a technique called controlled cross-circulation.⁵⁸ With this technique, the circulation of one dog was used temporarily to support that of a second dog while the second dog’s heart was stopped temporarily and opened. After a simulated repair in the second dog, the animals were disconnected and allowed to recover.

Lillehei and colleagues⁵⁸ used their technique at the University of Minnesota to correct a ventricular septal defect (VSD) in a 12-month-old infant on March 26, 1954 (Fig. 1-4). Either a parent or a close relative with the same blood type was connected to the child’s circulation. In Lillehei’s first clinical case, the patient made an uneventful recovery until death on the eleventh postoperative day from a rapidly progressing tracheal bronchitis. At autopsy, the VSD was closed, and the respiratory infection was confirmed as the cause of death. Two weeks later, the second and third patients had VSDs closed by the same technique 3 days apart. Both remained long-term survivors with normal hemodynamics confirmed by cardiac catheterization.

In 1955, Lillehei and colleagues⁵⁹ published a report of 32 patients that included repairs of VSDs, tetralogy of Fallot, and atrioventricularis communis defects. By May of 1955, the blood pump used for systemic cross-circulation by Lillehei and colleagues was coupled with a bubble oxygenator developed by Drs. DeWall and Lillehei, and cross-circulation was soon abandoned after use in 45 patients during 1954 and 1955. Although its clinical use was short-lived, cross-circulation was an important steppingstone in the development of cardiac surgery.

Meanwhile, at the Mayo Clinic only 90 miles away, John W. Kirklin and colleagues launched their open-heart program on March 5, 1955.⁵⁰ They used a heart-lung machine based on the Gibbon-IBM machine but with their own modifications. Kirklin wrote:⁶¹

By the end of 1956, many university groups around the world had launched into open-heart programs. Currently, it is estimated that more than 1 million cardiac operations are performed each year worldwide with use of the heart-lung machine. In most cases, the operative mortality is quite low, approaching 1% for some operations. Little thought is given to the courageous pioneers in the 1950s whose monumental contributions made all this possible.

Extracorporeal Life Support

Extracorporeal life support (ECLS) is an extension of cardiopulmonary bypass. Cardiopulmonary bypass was limited initially to no more than 6 hours. The development of membrane oxygenators in the 1960s permitted longer support. Donald Hill and colleagues in 1972 treated a 24-year-old man who developed shock lung after blunt trauma.⁶² The patient was supported for 75 hours using a heart-lung machine with a membrane oxygenator, cannulated via the femoral vein and artery. The patient was weaned and recovered. Hill’s second patient was supported for 5 days and recovered. This led to a randomized trial supported by the National Institutes of Health to determine the efficacy of this therapy for adults with respiratory failure. The study was conducted from 1972 to 1975 and showed no significant difference in survival between patients managed by ECLS (9.5%) and those who received conventional ventilatory therapy (8.3%).⁶³ Because of these results, most U.S. centers abandoned efforts to support adult patients using ECLS, also known as extracorporeal membrane oxygenation (ECMO).

One participant in the adult trial decided to study neonates. The usual causes of neonatal respiratory failure have in common abnormal postnatal blood shunts known as persistent fetal circulation (PFC). This is a temporary, reversible phenomenon. In 1976, Bartlett and colleagues at the University of Michigan were the first to treat a neonate successfully using ECLS. More than 8000 neonatal patients have been treated using ECLS worldwide, with a survival rate of 82% (ELSO registry data).

Melrose and colleagues⁶⁴ in 1955 presented the first experimental study describing induced arrest by potassium-based cardioplegia. Blood cardioplegia was used “to preserve myocardial energy stores at the onset of cardiac ischemia.” Unfortunately, the Melrose solution proved to be toxic to the myocardium, and as a result cardioplegia was not used widely for several years.