Thoracic Anesthesia


The Principles of Thoracic Anesthesia


1 History and Scope of Anesthesia for Thoracic Surgery

2 Practice Improvement and Patient Safety in Thoracic Anesthesia: A Human Factors Perspective

3 Physiology of One-Lung Ventilation

4 Perioperative Management of the Patient with Pulmonary Hypertension

5 Lung Separation Techniques

6 Mechanisms of Pain in Thoracic Surgery

7 The Biology of Lung and Esophageal Cancer

8 Anatomy, Imaging and Practical Management of Selected Thoracic Surgical Procedures

History and Scope of Anesthesia for Thoracic Surgery

Marcelle Blessing
Edmond Cohen

The history of anesthesia for thoracic surgery encompasses much of the history of anesthesia because the modern practice of thoracic anesthesia relies on major advances in preoperative evaluation, airway management, intraoperative monitoring, pharmacological agents, and improvements in postoperative pain management and intensive care management. These advances equip the modern thoracic anesthesiologist with the tools and techniques to care for even the frailest patients undergoing complex surgical operations. Many patients who would have been deemed inoperable in the past are operative candidates today, because of improvements in both anesthetic and surgical techniques that have augmented safety for all patients undergoing thoracic surgery. The current practice of anesthesia for thoracic surgery represents a culmination of 100 years of advances in anesthesia techniques, and these techniques continue to evolve.


The safe delivery of anesthesia for thoracic surgery is a relatively late development in the history of anesthesia because of the ingenuity needed to overcome the unique challenges of safely performing surgery in the thorax. It is easy to forget today that prior to advances in general anesthesia techniques, specifically positive pressure ventilation and controlled respiration with endotracheal intubation, surgery that trespassed the chest wall posed grave risks to patients. Although inhalational anesthesia was introduced in the 1840s, it took another 100 years before anesthesiologists made significant headway in providing safe care for patients undergoing operations in the chest. Improvements in anesthetic practice permitted thoracic surgery to flourish as a specialty; the growth of no other surgical subspecialty depended so heavily on the progress of anesthesia. Although intrathoracic procedures have become routine, thoracic surgeons and anesthesiologists retain a unique relationship; few areas of surgery require as much communication and cooperation between surgeon and anesthesiologist.

As the scope of thoracic surgery has increased greatly, so has the scope of anesthetic practice for it. Today, knowledge of anesthesia techniques for thoracic surgery has become more important than ever. Greater numbers of and types of procedures for lung, esophageal, mediastinal, anterior and posterior spinal, thoracic aortic and cardiac surgery rely on thoracic approaches that require use of one-lung ventilation (OLV). Also, more intrathoracic procedures are being performed with minimally invasive approaches that rely on OLV for adequate surgical exposure. To safely provide the OLV that is universally favored by thoracic surgeons, anesthesiologists must be knowledgeable about the physiology of OLV, be familiar with the range of tools available for providing it, and be aware of techniques for preventing hypoxemia. A variety of double lumen endotracheal tubes and modern endobronchial blockers are now available to provide safe and reliable OLV for most patients.

The Pneumothorax Problem

The inherent danger of performing surgery within the thorax has been known since antiquity. Almost 2000 years ago, the Roman encyclopedia author Celsus, in De Medicina, described the problem succinctly by noting that when a knife penetrates the chest death ensues at once, even though the belly can be opened safely with the patient breathing spontaneously.1 Celsus was describing the so-called “pneumothorax problem”: When the chest is opened and the lung is exposed to atmospheric pressure, the operative lung suddenly collapses because of the loss of the normally negative intrapleural pressure. The collapsed lung would paradoxically expand during expiration and collapse again during inspiration, as air was transferred from healthy to collapsed lung by the patient struggling to breathe. The transfer of air between the two lungs became known as “pendelluft.” A surgeon brave enough to open the chest wall of a spontaneously breathing patient would also face vigorous side-to-side movement of the mediastinum with respiration known as “mediastinal flapping” that would cause compression of the contralateral lung. The patient would quickly become tachypneic and cyanotic while struggling to breathe spontaneously. Only the briefest intrathoracic procedures could be performed under these circumstances, and for this reason thoracic surgery was mostly limited to procedures of the extrathoracic chest wall in the first third of the 20th century. John W. Strieder, an eminent thoracic surgeon, gave a colorful description of operating in the “good old days” where “the period of operation was, with dismaying frequency, a race between the surgeon and the impending asphyxia of the patient.”2

After the advent of inhalational anesthesia in the 1840s, the delivery of general anesthesia became routine and permitted the rapid growth of most areas of surgery. However, the techniques favored for delivering inhalational anesthesia until the 1930s remained mask or open drop administration of ether or chloroform, with or without nitrous oxide. Muscle relaxants did not yet exist, and endotracheal intubation was considered an invasive procedure that was used only by a few experts. Typically, patients would breathe spontaneously, and thus could control the depth of anesthesia with their depth and rate of respirations. Also, long before the use of cuffed endotracheal tubes, an intact cough reflex was valued during general anesthesia for protecting the lungs from gastric aspiration, so relatively light planes of anesthesia were the norm. Prior to the introduction of antibiotics in the 1930s, patients frequently presented with empyema, pulmonary abscess or tuberculosis, and often had copious secretions and formidable coughs. Clearly, operating conditions were poor for the early 20th century thoracic surgeon facing a lightly-anesthetized patient asphyxiating and coughing with an unprotected airway. With this in mind, it is no surprise that thoracic surgery remained in its infancy well into the 20th century, until anesthetic techniques progressed sufficiently to provide improved operating conditions.

Prior to the development of antibiotics, the main indication for thoracic surgery was infection. Opening the pleural cavity did not necessarily pose a significant danger to these patients because long-standing infections often resulted in adhesions that formed between the lung and chest wall, preventing the formation of a significant open pneumothorax or mediastinal shift. In fact, repeated aspirations were often attempted to promote the formation of adhesions so that subsequently the pleural cavity could be opened safely, or substances such as air or water were injected into the pleural space as irritants, also to encourage adhesions, in preparation for surgery.3,4 Another brutal technique to cope with the open pneumothorax, “Muller’s handgrip,” was also woefully inadequate: The surgeon would grab the lung while the chest was open and pull it into the wound to plug the thoracotomy incision.5 Pulmonary resection was usually performed using a snare or tourniquet technique, and reoperation would be necessary to remove necrotic tissue. Not surprisingly, such staged procedures could result in sepsis from remaining necrotic tissue. Complications were common and mortality from thoracic surgery was astonishingly high. In a review from 1922 by Howard Lillienthal, the mortality rate for lobectomy performed for chronic pus formation was 42%, and in the 10 cases where more than one lobe was involved, mortality was 70%.6

Differential Pressure Breathing

Better anesthetic techniques were sought to meet the demands of thoracic surgeries. The first promising solution to prevent the development of an open pneumothorax was developed in Germany by the surgeon Ernst Ferdinand Sauerbruch.7 In 1893, Sauerbruch’s mentor Johann von Mikulicz-Radecki asked him to tackle the “pneumothorax problem,” and his solution, differential pressure breathing, became the prevailing method for anesthetic management in thoracic surgery until World War II. Sauerbruch performed thoracotomies on dogs and found that spontaneous ventilations were sustained without lung collapse if the exposed lung was kept at a pressure 10 cm H2O below atmospheric pressure, and later used this technique during thoracotomies on humans (Figure 1–1). To provide the negative pressure, the patient and surgical team had to work within a negative pressure chamber constructed of steel, and the patient’s head extended outside the cabinet and was exposed to ambient pressure. Throughout the surgery, the patient would breathe spontaneously and the lungs would remain inflated from the negative pressure within the chamber. The surgeon within the chamber was separated from the anesthetist by a thick wall so that communication was only possible by telephone, and then only with difficulty due to the loud noise in the chamber created by the pump that created the negative pressure. Although impractical, Sauerbruch’s method was considered a triumph and differential pressure breathing was adapted throughout Europe and America; Sauerbruch became a very influential figure in the history of thoracic anesthesia.


Figure 1–1. Sauerbruch’s experimental negative pressure box for performing thoracotomies on dogs. The dog’s chest is enclosed in the box in which the pressure is –10 mm Hg (1904). (From: Mushin WW, Rendell-Baker L, eds. The Principles of Thoracic Anesthesia. Springfield, IL: Charles C Thomas; 1953, with permission. Copyright Wiley-Blackwell.)

Even though Sauerbruch’s methods were highly commended, his negative pressure technique did not gain many followers because it required investment in an expensive and cumbersome negative pressure chamber. However, another option became popular that was based on essentially the same principle. Ludolph Brauer, a colleague of Sauerbruch’s, conducted his own experiments and developed a different solution simultaneously. In fact, his description of the use of a positive pressure chamber was initially presented in the same issue of the same journal as Sauerbruch’s initial presentation of his research. Instead of applying negative pressure to the open chest, Brauer increased the intrapulmonary pressure by placing his subject’s head in a positive pressure chamber. His original apparatus was simply a large box into which the head of the patient was placed after induction of anesthesia. Anesthesia was maintained using oxygen and chloroform, and the patient would breathe spontaneously without assistance. Prior to the opening of the chest, the pressure inside the box would be raised by adding compressed air. Brauer found that if the pressure were raised 10 mm Hg above the atmospheric pressure applied to the open lung, no pneumothorax developed. An obvious limitation and challenge of Brauer’s apparatus was that the anesthetist had no access to the patient’s head during surgery.7 Brauer’s device bears a striking resemblance to the helmets used for delivering noninvasive continuous positive airway pressure (CPAP), a tool for treating acute respiratory failure outside of the operating room (OR).8

Brauer’s positive pressure technique was the favored method for preventing pneumothorax because the equipment involved was simpler, less bulky, and inexpensive compared with negative pressure chambers. At the same time, even more complicated versions of Sauerbruch’s negative pressure chamber were developed. The American surgeon Willy Meyer modified the Sauerbruch device to create his “universal differential pressure chamber” that included both a positive and negative pressure chamber so that the patient’s head, an anesthetist and assistant could be enclosed in a positive pressure chamber.9He constructed the only negative pressure chamber for such a use in the United States. By using both chambers, the pressure gradient could be maintained by applying positive pressure to the head, negative pressure to the open chest, or both. The chamber was used not only to support respiration during surgery, but also to improve wound drainage and lung expansion postoperatively.10

Both the positive-pressure and negative-pressure methods relied on maintaining a pressure gradient between the air inside and outside the lungs, otherwise known as differential pressure anesthesia. Both methods were successful at preventing the formerly inevitable open pneumothorax after thoracotomy, but both were ultimately doomed to become historical relics because they provided dangerously inadequate ventilation. Hypoventilation, hypercarbia, hypoxemia, and impaired venous return were significant problems during prolonged cases. Clinical deterioration was not unusual after long cases using these techniques, even though the lungs never collapsed. Meyer recognized that carbon dioxide accumulation was probably the culprit in such cases of unexplained shock, and recommended periodically deflating the lungs by applying rhythmic variations in pressure coinciding with spontaneous respirations to assist ventilation.10


While the debate continued regarding the merits of positive versus negative pressure application, an alternative anesthetic method for preventing the development of the open pneumothorax evolved from earlier discoveries in tracheal intubation and mechanical ventilation and became popular in America. This new method, called tracheal insufflation anesthesia, was the clear precursor to the endotracheal anesthesia that is universally used for thoracic surgery today.

Endotracheal Intubation

Tracheal intubation and mechanical ventilation were by no means new discoveries. Because of widespread reluctance to accept tracheal intubation for common use, the course of its development did not follow a smooth path. Andreas Vesalius described tracheal intubation and positive pressure ventilation of a pig in 1543. He performed a tracheotomy and passed a reed into the trachea of a pig and blew into the tube to provide artificial ventilation during a thoracotomy and was able to prevent a potentially lethal pneumothorax. However, his findings went unnoticed and had to be rediscovered. In the 18th century, interest in artificial ventilation for resuscitation grew. In 1788, Charles Kite resuscitated victims of drowning from the Thames using curved metal cannulas that he placed blindly in the trachea. Application of these resuscitation techniques to anesthesia delivery began soon after the discovery of inhalational anesthetics in the 1840s. In 1869, Friedrich Trendelenburg used a tracheostomy tube with an inflatable cuff to administer chloroform during head and neck procedures. William MacEwen, a Scottish surgeon, is credited with the first use of oral tracheal intubation for an anesthetic. On July 5, 1878, MacEwen placed a flexible metal tube in the larynx of an awake patient who was to have an oral tumor removed at the Glasgow Royal Infirmary.11 Unaware of earlier uses of intubation, Joseph O’Dwyer, a pediatrician, performed blind oral tracheal intubations on children suffering from diphtheria in 1885.12 O’Dwyer later developed a rigid tube with a conical tip that occluded the larynx sufficiently to facilitate positive pressure ventilation. In 1893, George Fell attached O’Dwyer’s metal tube to a bellows and T-piece, creating the Fell-O’Dwyer apparatus. Fell used the apparatus to provide ventilatory support for opioid-induced respiratory depression (Figure 1–2).


Figure 1–2. The Fell-O’Dwyer apparatus (c. 1888). O’Dwyer’s laryngeal tube has a curved right angle and uses fitted, interchangeable, conical heads of different sizes designed to fit securely into the larynx. Rings were provided for the operator’s fingers and the operator’s thumb was placed over the expiratory orifice during inflation. (From: Mushin WW, Rendell-Baker L, eds. The Principles of Thoracic Anesthesia. Springfield, IL: Charles C Thomas; 1953, with permission. Copyright Wiley-Blackwell.)

By the 1890s, there was interest in the application of endotracheal anesthesia to thoracic surgery as a possible solution to the pneumothorax problem. Two French surgeons, Tuffier and Hallion, reported on their use of tracheal intubation and artificial ventilation for thoracotomies on animals in 1896.11 Their device also incorporated a bellows that was used for rhythmic inflation of the lungs, and a water valve that could control the degree of resistance to expiration, a precursor of the modern use of positive end-expiratory pressure (PEEP). Rudolph Matas, among his many pioneering contributions to anesthesiology, made modifications to the Fell-O’Dwyer apparatus so it could be used for surgical purposes. He cited the research of Tuffier and Hallion as his inspiration, and was convinced that these methods were ideal for thoracic cases. His modifications to the Fell-O’Dwyer apparatus included a graduated cylinder for delivery of precise volumes of gases and a mercurial manometer for direct measurement of intrapulmonary pressures. He also added an intralaryngeal cannula connected by a stopcock to a rubber tube and funnel that could be used for administration of chloroform.13

Tracheal Insufflation

Until 1907, endotracheal techniques always involved an endotracheal tube that was similar in width to the trachea, through which inspiration and exhalation occurred. However, in 1907, a different method was introduced by Barthélemy and Dufour that became known as “tracheal insufflation”.11 Insufflation anesthesia consisted of placing a thin tube in the trachea and then continuously blowing gases under positive pressure into the lower portion of the trachea. Expired gases escaped between the tracheal tube and the tracheal wall. Meltzer and Auer, American physiologists, used this technique extensively in animal studies. They showed that curarized dogs could be anesthetized by blowing air and ether continuously into a tube inserted into the trachea, and that gas exchange would occur “without any normal or artificial rhythmical respiratory movements whatever” because expired gases could escape around the tracheal tube.14 Because gas was insufflated continuously, and no rhythmic applications of pressure were used, the insufflation method was similar to Brauer’s positive pressure method; however, efficiency of gas exchange was improved by the decrease in dead space achieved by placing the cannula in the trachea.

Charles Elsberg, a thoracic surgeon in New York City, was inspired by Meltzer and Auer’s physiology research, and he utilized and modified their insufflation technique for application to thoracic surgery. His modifications included replacing the bellows of Meltzer and Auer’s apparatus with an electric motor. He also favored placement of the tracheal cannula under direct vision using either a Killian bronchoscope or Chevalier Jackson’s laryngoscope, after topicalization with cocaine.15 Elsberg first used insufflation to resuscitate a myasthenic patient who had become cyanotic and pulseless. After initiation of insufflation, her color improved and pulse returned, but resuscitation was discontinued after 5 hours because she did not regain consciousness. Bolstered by his success, Elsberg presided over the first use of tracheal insufflation anesthesia for thoracotomy.16 In February 1910, a 55-year-old butcher was admitted to the Mt. Sinai Hospital with a 13-month history of productive cough. A diagnosis of lung abscess was made, and the thoracic surgeon Howard Lilienthal sought a definitive operative cure, so he enlisted Elsberg for his expertise with tracheal insufflation. When the pleura was opened, 15 mm Hg pressure was applied, and the lung was described as “two-thirds of its capacity, mottled, and rosy pink in color.” Different pressures were applied and the lung readily collapsed and distended. He recommended the periodic interruption in the stream of insufflation every 2 or 3 minutes to allow the lungs to collapse and improve carbon dioxide elimination, bringing this method closer to modern positive pressure ventilation. The anesthetic was considered a great success, and Elsberg promoted tracheal insufflation for all types of surgery requiring general anesthesia. Just 1 year later, he published on his experiences anesthetizing over 200 patients with this technique.17 Elsberg’s method of tracheal insufflation strongly resembles the modern practice of oxygen insufflation used during rigid bronchoscopy that was introduced by Sanders in 1968.18

In the 1920s and 30s, tracheal insufflation anesthesia was the most popular anesthetic method for thoracic surgery in the United States. Insufflation anesthesia became popular in Europe for head and neck surgery because it gave the surgeon better access than mask or hand drop techniques. However, differential pressure anesthesia remained the preferred anesthetic technique for thoracic procedures in Europe for years to come. Reasons for this include the dominance of Sauerbruch and his refusal to endorse any other method. In 1916, Sauerbruch’s own assistant, Giertz, conducted experiments on animals that showed that rhythmic inflation of the lungs was superior to either continuous negative or positive pressure anesthesia. Giertz’s experiments demonstrated that differential pressure anesthesia resulted in inadequate ventilation, carbon dioxide retention, and impaired venous return causing circulatory depression.7 Tracheal insufflation was far from perfect. Carbon dioxide accumulation frequently occurred if gas flow went uninterrupted, so modifications to Elsberg’s apparatus were developed that periodically stopped airflow to permit lung collapse. Also, dangerously high-intrapulmonary pressure could occur when the return of gas was impeded. Cases of alveolar rupture and surgical emphysema referred to as “wind-tumor” occurred, probably as a result of vocal cord spasm around thin insufflation catheters interfering with the exit of expired gases.11

Advances in Laryngoscopy

Even though instruments for direct laryngoscopy existed by the 1920s, they were infrequently used. Blind placement of endotracheal tubes required considerable skill and could be a traumatic procedure. Prior to 1895, direct visual examination of the larynx was assumed to be impossible. Alfred Kirstein, a physician in Berlin, is credited with inventing the first direct laryngoscope in 1895.19 Kirstein’s “autoscope” was not used by anesthetists, but it was the prototype for most laryngoscopes to follow. In 1913, Chevalier Jackson developed his own laryngoscope and published a landmark paper on proper positioning and technique for laryngoscopy.20 In the 1940s, there was a renewed interest in laryngoscope blade design. Robert Miller created the familiar Miller blade in 1941, but the origins of its design are evident in Kirstein’s and Jackson’s laryngoscopes. Only 2 years later, Sir Robert Macintosh released his familiar curved blade that would go on to become the most popular blade in the world.

Endotracheal Tubes

Alongside the development of direct laryngoscopy, came the development of improved endotracheal tubes. World War I produced many casualties with head and neck injuries requiring reconstructive plastic surgery. In 1919, the British anesthetists Ivan Magill and Stanley Rowbotham were assigned to work with the British army plastics unit, and they were forced to adapt endotracheal anesthesia to safely care for these patients. They became experts in blind nasal tracheal intubations to provide unhindered access to the face and airway. They were dissatisfied with the thin endotracheal catheters that were in use for insufflation anesthesia, so they progressed to using wide-bore tubes that more closely resemble what are in use today. By using larger tubes, they returned to the older inhalation method where respiration occurred in both directions through one tube. By doing this, they rejected the popular insufflation technique. Magill’s wide-bore red rubber tubes resisted kinking and were better suited to the contours of the upper airway. “Magill tubes” remained the standard endotracheal tubes until plastic tubes were introduced.

In 1928, Arthur Guedel and Ralph Waters introduced an endotracheal tube with a detachable inflatable cuff (Figure 1–3).21 Prior to Guedel and Waters, there were sporadic proponents of cuffed tubes. As early as 1871, Trendelenburg fitted a cuff on a tracheotomy cannula, followed by Eisenmenger in 1893, and Dorrance in 1910.22 However, none of these early attempts to apply cuffs to endotracheal tubes attracted much interest. Guedel demonstrated the effectiveness of the cuff’s seal with his colorful “dunked dog” demonstrations. He submerged his intubated and anesthetized dog, Airway, in an aquarium, from which he emerged unharmed.23 Guedel’s cuffed endotracheal tube could prevent the aspiration of gastric contents, so it was no longer necessary to keep the patient “light” so the cough reflex would remain intact. Also, since deeper planes of anesthesia could be used, suctioning the trachea without coughing was possible. Aside from aspiration prevention, cuffed endotracheal tubes enabled the most important advancement in anesthetic management for thoracic surgery: the use of controlled positive pressure ventilation. By increasing the depth of anesthesia and delivering controlled breaths using a cuffed endotracheal tube, hyperventilation was now possible to suppress respiratory efforts.


Figure 1–3. Guedel and Waters “new intratracheal catheter” (1928). The catheter is shown deflated and then inflated. The tube was 14 inches long and made of rubber. (From: Mushin WW, Rendell-Baker L, eds. The Principles of Thoracic Anesthesia. Springfield, IL: Charles C Thomas; 1953, with permission. Copyright Wiley-Blackwell.)

By 1930, all aspects of airway management necessary to conquer the “pneumothorax problem” were well-described. However, these methods did not gain immediate acceptance. Sauerbruch’s differential pressure method was still in wide use until World War II. Also, the advantage of cuffed tubes was not universally recognized. For example, in 1948, in a review of a series of 309 anesthetics for thoracic surgery, the authors still advocated placing the patient in steep Trendelenburg position to promote drainage of secretions through and around uncuffed tracheal tubes. Also, the authors of this review did not even recommend routine use of controlled ventilation.24


As anesthetic techniques improved, pulmonary surgery made slow progress in the 1920s, leading to significant advances in the 1930s. The two-stage snare or tourniquet technique for lung resection was replaced by the individual-structure ligation technique that reduced complications such as air leak, tension pneumothorax, hemorrhage, and infection from necrotic residual lung tissue. Harold Brunn was the first to use this technique extensively in 1929.25 Many landmark thoracic surgeries were preformed in the 1930s: Rudolph Nissen performed the first successful two-stage pneumonectomy in 1931 and Evarts Graham performed the first successful one-stage total pneumonectomy for a malignant tumor in 1933.26,27 Graham’s historic pneumonectomy anticipated the future of pulmonary surgery where cases of malignant disease would soon overshadow those of infection. Another critical development in thoracic surgery in the 1930s was the introduction of routine postoperative pleural drainage with closed-chest thoracostomy. Esophageal surgery also progressed in the 1930s. In 1933, the first successful transthoracic esophagectomy was performed in Japan.28 Not long before, these surgeries were too perilous to be attempted, but anesthetic techniques had improved sufficiently to accommodate the surgical advances of the 1930s. With the introduction of OLV and mechanical ventilation in the 1930s and 40s, anesthesiologists were able to further assist the development of thoracic surgery.


Before the 1940s, when general anesthesia using endotracheal intubation became routine for thoracic surgery, regional and spinal anesthesia for thoracic surgery were frequently used. Advocates of regional techniques believed that the cough reflex was preserved and that respiration was not detrimentally affected. According to Magill, spinal anesthesia was an excellent technique for thoracoplasty, lobectomy, and even pneumonectomy! He found that patients generally would breathe well and infrequently needed supplemental oxygen and that the cough reflex was well-maintained. He also found it helpful to have a cooperative patient who could participate in breath holding, an advantage at the time over general anesthesia that infrequently employed controlled ventilation.29 Others did not have similar success with spinal anesthesia. Nosworthy declared, “I like my anaesthetic technique to be such that I have the whole situation under control. I do not feel that I am in a position to cope with any emergency when chest surgery is performed under spinal anaesthesia.”30 Nosworthy found the cough reflex to be inadequate and dyspnea to be frequent during procedures on the open thorax under spinal anesthesia.


With the union of direct laryngoscopy, tracheal intubation, and controlled ventilation, interest in lung separation soon followed. Prior to the development of lung separation techniques, most thoracic surgeries were still performed for cases of infection, and spillage from the infected lung was a frequent problem. Patients frequently had copious secretions, requiring aggressive preoperative physiotherapy to reduce the risk of spillage in the OR. In 1931, Gale and Waters reported the first use of OLV for thoracic surgery.31 Their technique was simply to intubate the healthy bronchus blindly with a long endotracheal tube. They used a standard rubber Guedel-Waters cuffed endotracheal tube. The tube was softened with hot water and molded to have a lateral curve. It was placed in the trachea and then blindly advanced into either bronchus, stopping as soon as resistance was felt. The cuff provided a seal for the intubated bronchus, while also occluding the bronchus of the diseased lung. Their desire was to prevent the “pneumothorax syndrome” by isolating the lung exposed to ambient pressure. Also, they acknowledged the advantages of an immobile lung, quiet surgical field, and the prevention of secretions from entering the trachea. Prevention of contamination from infection was not of paramount importance for them. Although the simplicity of their technique was admirable, it was not very stable and did not become widely used.

In 1936, Rovenstine attempted to improve upon Gale and Waters’ technique by using a single-lumen endotracheal tube with two cuffs.32 By having two cuffs, either one lung or both lungs could be ventilated. Rovenstine’s endobronchial tube was made of woven silk and also would be molded in hot water to create a lateral curve and then be advanced blindly into either bronchus. The upper cuff was inflated first, above the carina. With the upper cuff alone inflated, both lungs could be ventilated. When the lower cuff was inflated, the trachea was occluded at the carina.

Bronchial Blockers

Bronchial blockade, another method of lung separation, was also first introduced into anesthetic practice in the 1930s. By placing an obstruction to ventilation in the bronchus to a lung or lobe, the unventilated lung distal to the obstruction will subsequently collapse. In 1935, Archibald gave the first description of the use of bronchial blockade. In order to control secretions during lobectomy, he used an inflatable balloon at the end of a rubber catheter to occlude the main bronchus of the diseased lung. He confirmed appropriate placement with x-ray films.33 He reported that this balloon was easy to place and that it prevented escape of pus from the diseased lobe. Although the use of x-ray made this technique cumbersome, the idea of using a balloon for bronchial blockade was promising and numerous refinements of this technique have been made and are still in current use.

In 1936, Magill improved on Archibald’s design by designing a similar bronchial blocker (BB) that could be placed under direct vision using a device of his own called a tracheoscope, eliminating the need for x-ray guidance. The BB was a long tube with a balloon at its distal end that was inserted alongside an endotracheal tube. It also had a suction catheter for the blocked lung, and Magill recommended the use of the blocker for the control of secretions; however, he did also acknowledge its ability to promote atelectasis and thus improve surgical exposure. Magill recommended its placement after topicalizing the larynx, but prior to induction of general anesthesia, so secretions could be suctioned during anesthetic induction. Magill also performed lung separation using endobronchial intubation using a technique similar to Gale and Waters’, but his endobronchial tube was made of rubber over fine metal tubing and was placed under direct vision with an endoscope through the lumen.29 All of these techniques required considerable skill and experience, limiting their popularity. Thompson’s BB, introduced in 1943, was modeled after Magill’s, and is the prototype for all BBs to follow. Thompson’s blocker consisted of two tubes fused together. One tube inflated a gauze-covered balloon and the other provided suction from the blocked bronchus. The blocker had a stylet and was placed through a rigid bronchoscope.34 Many other devices were employed as BBs prior to the development of the plastic BBs that are in use today. In 1938, Crafoord described his bronchial tamponage technique that used a ribbon gauze tampon for the control of secretions. The tampon was inserted through a rigid bronchoscope into the selected bronchus while the healthy lung would be ventilated by an endotracheal tube in the carina.35

In the 1950s, multiple single-lumen endotracheal tubes were developed with incorporated BBs. In 1953, Stuertzbecher introduced an endotracheal tube with an incorporated styletted BB and suction catheter. Vellacott introduced a similar tube in 1954, as did Macintosh and Leatherdale in 1955 and Green in 1958.36-38 These tubes are the clear predecessors of the Univent tube, the first modern endotracheal tube designed for bronchial blockade. First marketed in 1982, the Univent tube (Univent, Fuji Systems Corp., Tokyo, Japan) is a large endotracheal tube with a small internal lumen that contains a retractable, cuffed bronchial blocker.39 Once a procedure requiring bronchial blockade is over, the blocker can be retracted to its internal lumen and the tube functions as a conventional single-lumen tube. Although this is a convenient method for providing OLV in patients who may require postoperative mechanical ventilation, the Univent tube is bulky and has a larger diameter than standard single-lumen endotracheal tubes of the same ID number, causing potentially increased airflow resistance.40

Balloon-tipped catheters designed for other uses, such as Fogarty embolectomy catheters, Swan-Ganz catheters, and Foley catheters have been used as bronchial blockers. Although the Fogarty embolectomy catheter was designed as a tool for vascular surgery, there are numerous reports of its successful use as a bronchial blocker.41 Fogarty catheters have significant limitations because they were not designed for use as bronchial blockers. They have low-volume, high-pressure cuffs capable of damaging bronchial mucosa and there is no communicating channel for suction or oxygen insufflation. Positioning of Fogarty catheters is especially difficult in the left main bronchus because they were not designed with a mechanism to guide them. Recently, new balloon-tipped bronchial blockers have been designed that are specifically intended for bronchial blockade.42-44 All use balloons with low-pressure cuffs to decrease bronchial trauma, and all are intended for placement with guidance by flexible fiber-optic bronchoscopy.

Double-Lumen Endobronchial Tubes

The origins of the double-lumen tube date back to 1889 when Head used a tube with two lumens to study respiratory physiology in dogs. In 1949, Bjork and Carlens designed the first double-lumen tube (DLT) used for thoracic surgery, although it was initially developed for differential bronchospirometry.45 The tube was designed for intubation of the left main bronchus. Carlens’ tube consisted of two cuffed tubes of unequal length fused together, a longer tube for intubating the left main bronchus and a shorter tube that ends in the trachea. Because the endobronchial intubation was performed blindly, a carinal hook was used to grip the carina and aid placement (Figure 1–4). In 1959, Bryce-Smith modified the Carlens tube by eliminating the carinal hook since it could cause damage to the trachea and it often hindered more than it helped in correct placement.46 Both of these tubes were suitable for all types of intrathoracic procedures except for left pneumonectomy, where the left main bronchus is cut close to the carina. For this reason, effective means of intubating the right main bronchus were needed. Because it is difficult to intubate the right main bronchus without occluding the opening of the right upper lobe bronchus, most endobronchial and DLTs were initially designed for use on the left. In 1960, Bryce-Smith and Salt described a right-sided DLT that included a slit in the endobronchial cuff for ventilation of the right upper lobe, and White designed a right-sided version of the Carlens’ tube that also used a slit in the endobronchial cuff.47,48


Figure 1–4. Bjork and Carlen’s double-lumen catheter (1949). This is the first double-lumen endobroncheal tube intended for intubation of the left mainstem bronchus. Note the presence of the carinal hook. (From: Mushin WW, Rendell-Baker L, eds. The Principles of Thoracic Anesthesia. Springfield, IL: Charles C Thomas; 1953, with permission. Copyright Wiley-Blackwell.)

These early DLTs were fraught with problems. Occlusion by kinking, trauma from carinal hooks, high-airway resistance during OLV and simply difficult placement were not uncommon. In 1962, Robertshaw introduced a new DLT that had no carinal hook and novel cross-sectional D-shaped lumens (Figure 1-5).49 The D-shaped lumens provided a larger cross-sectional area and reduced resistance to airflow compared with the older round lumens. Disposable plastic (polyvinyl chloride) DLTs have replaced these older red rubber tubes, but contemporary DLTs strongly resemble Robertshaw’s original design. Advantages of plastic tubes include large lumens that reduce resistance to gas flow, suctioning, or bronchoscopy. Also, plastic tubes use high-volume, low-pressure cuffs, as opposed to the older low-volume, high-pressure cuffs that could potentially cause more airway trauma. However, red rubber reusable tubes are still used in parts of the world where resources are scarce.


Figure 1–5. Robertshaw’s red rubber (left-sided) double-lumen tube (1962).

Prior to the introduction of fiberoptic bronchoscopy in the 1970s, endobronchial placement of DLTs was essentially blind and confirmation of placement relied on clinical examination. Small, flexible fiberoptic bronchoscopes permit precise evaluation of the positioning of DLTs, endobronchial tubes, and endobronchial blockers. In the 1980s, fiberoptic bronchoscopes were first used for positioning of DLTs in the OR, and this has now become a common practice. All catheter-guided BBs rely on fiberoptic bronchoscopy for positioning. Positioning can also easily be reconfirmed once patients are moved from supine to lateral decubitus position by using fiberoptic bronchoscopy, and positioning can also be easily reassessed mid-operation. In addition to its utility in precise placement of DLTs and bronchial blockers, fiberoptic bronchoscopy assists examination and identification of unusual airway anatomy and can guide tracheal toilet. Because of these advantages, many experts recommend its routine use for placement of DLTs.50,51


Although the “pneumothorax problem” was solved by the application of positive pressure ventilation to the lungs, the routine use of intermittent positive pressure ventilation was impractical before the development of muscle relaxants and mechanical ventilators. In 1909, Meltzer and Auer recommended the use of curare in their animal studies of tracheal insufflation; however, curare was only first used as part of a general anesthetic in 1942.14,52 In 1946, Harroun used curare, nitrous oxide, and morphine for thoracic surgery, an important new technique because it included no flammable agents and permitted the use of electrocautery during surgery.53 Muscle relaxation facilitated the use of controlled ventilation by suppressing spontaneous respiratory efforts with lighter planes of anesthesia and without hyperventilation. Safety problems with curare soon became evident, but numerous safer neuromuscular agents were introduced to replace curare, eventually making the administration of muscle relaxants a routine part of a general anesthetic.

Some early mechanical ventilators have already been described here, such as the Fell-O’Dwyer apparatus from 1892, and Matas’ modification of the Fell-O’Dwyer apparatus that incorporated manometry and delivery of inhaled anesthetics. Giertz’ experiments and writings advocating controlled ventilation had gone mostly ignored; however, Frenckner, a Swedish otolaryngologist, familiar with Giertz, developed the “Spiropulsator” in 1934 for rhythmic inflation of the lungs. In 1938, Crafoord, a colleague of Frenckner, modified the apparatus by including a reservoir bag from which the patient could take a breath, since spontaneous respirations were not uncommon because these early air-driven ventilators predated muscle relaxants.54 The reservoir bag was intended to keep the patient from fighting the ventilator. Crafoord and Frenckner routinely intubated patients under topical anesthesia and then used their ventilator during thoracic surgery. The “Spiropulsator” became popular in Scandinavia, but in the 1930s and 40s there was very limited interest in controlled ventilation elsewhere, since continuous positive pressure delivered by mask and tracheal insufflation were widely used (Figure 1–6).


Figure 1–6. The Frenckner Spiropulsator (1934). Note the cuffed endotracheal tube lying to the right. (From: Mushin WW, Rendell-Baker L, eds. The Principles of Thoracic Anesthesia. Springfield, IL: Charles C Thomas; 1953, with permission. Copyright Wiley-Blackwell.)

Routine use of ventilators intraoperatively only occurred after they were routinely used outside of the OR. In 1952, an epidemic of poliomyelitis in Copenhagen overwhelmed Blegdam’s hospital. Three thousand patients presented with polio, and one-third of these presented with paralysis. Faced with so many patients with respiratory insufficiency, the hospital turned to an anesthesiologist, Bjorn Ibsen, for help. Ibsen believed that performing tracheostomies and providing controlled ventilation for weak children would increase survival rates.55 Initially, since the hospital had few mechanical ventilators, medical students squeezed breathing bags in shift, but toward the end of the epidemic they were replaced with mechanical ventilators. Ibsen proved correct, survival rates increased dramatically and the modern intensive care unit (ICU) was born and the iron lung was left behind. In 1955, Björk and Engstrom used their ventilator for postoperative ventilatory support for the frailest of their thoracic surgery patients.56 After proving their safety and efficacy in the ICU, mechanical ventilators gained acceptance in OR in the 1960s and 70s.


Use of complex intraoperative patient monitors is commonplace today; however, prior to the 1960s, intraoperative monitoring consisted of merely observation, palpation and auscultation with an anesthesiologist relying on a blood pressure cuff, electrocardiogram, and esophageal stethoscope. Hypoxemia was only detected by the presence of peripheral cyanosis, frequently a late and unreliable sign. The development of accurate invasive monitoring of peripheral arterial, central venous, and pulmonary arterial pressures improved monitoring, but noninvasive monitors of oxygenation and ventilation, specifically pulse oximetry and end-tidal capnography, have become crucial elements of safe anesthetics for all types of surgery, especially during OLV. Severinghaus declared, “Pulse oximetry is arguably the most important technological advance ever made in monitoring the well-being and safety of patients during anesthesia, recovery, and critical care.”57 In 1942, Millikan developed the first oximeter for the ear intended for use by pilots in World War II to warn them of hypoxia from an oxygen supply failure. Takuo Aoyagi, a Japanese engineer, refined the oximeter to measure pulse in addition to oxygen saturation, creating the first pulse oximeter.58 Pulse oximetry was not used in the OR routinely until the 1980s, making it a relatively recent addition to the routine monitors available to anesthesiologists. The history of capnography also followed a similar course. The initial application of infrared absorption to measure expired carbon dioxide occurred in 1943; however, capnography did not appear routinely in ORs until the 1980s.59 Pulse oximetry and capnography have decreased the need for direct measurement of arterial blood gases, but have not replaced it entirely. Both provide rapid and continuous guides to gas exchange, and serve as a guide when direct blood gas measurements are appropriate.


In 1956, Halothane was introduced in England, and it quickly replaced ether and cyclopropane because of its favorable safety profile, high potency, less noxious odor, nonflammability, and kinetic properties that provided a more rapid induction and emergence.60 Halothane’s potency made nitrous oxide unnecessary during OLV, and it could be used safely with electrocautery because it is nonflammable. Although halothane has largely been replaced by isoflurane, sevoflurane, and desflurane because of its association with liver toxicity and cardiac arrhythmias, the practice of using potent inhalational agents without nitrous oxide for maintenance of anesthesia remains standard during OLV.

Even though potent inhaled anesthetics made delivery of 100% oxygen possible during OLV, hypoxemia was still frequently encountered due to blood shunted through the nonventilated lung. CPAP and PEEP are both ventilatory maneuvers that were developed for respiratory support outside of the OR, but have found their respective roles for improving oxygenation during OLV. In 1971, CPAP was first described for use in infants with idiopathic respiratory distress syndrome.61 Since the 1980s, CPAP applied to the nonventilated lung has been used as an effective treatment for hypoxemia during OLV.62However, CPAP often cannot be used during thoracoscopic procedures because it may interfere with surgical exposure. PEEP applied to the ventilated lung is also frequently included to improveoxygenation during OLV.63 High-frequency jet ventilation (HFJV) with oxygen to the nondependent lung has also been used during OLV to improve oxygenation.64 HFJV uses a jet of fresh gas delivered from a high-pressure source directly into the airway through a small catheter at rates of approximately 100 to 400 breaths per minute. Because of the small tidal volumes delivered, the collapsed lung field remains quiet for the surgeon. HFJV has advantages in many situations, including ventilating patients with bronchopleural fistulas and in interrupted airways such as in patients with tracheal stenosis or those undergoing reconstructive surgery of the airway.


Advances in pain management have improved care for patients undergoing thoracic surgery. Very severe pain results from thoracotomy incisions, and post-thoracotomy pain has a profound impact on recovery after surgery by interfering with return of pulmonary function. Also, inadequate treatment of acute pain following thoracic surgery can contribute to the development of disabling chronic pain. Awareness by anesthesiologists and thoracic surgeons of the impact of inadequately managed acute pain on morbidity has sparked the development of multiple modalities of pain management. Prior to the 1980s, the only option for patients was systemic opioids, frequently administered intramuscularly. Today, options include systemic opioids, regional local anesthesia, epidural local anesthesia, and epidural opioids; and all can be delivered using patient-controlled analgesia (PCA). Between the variety of pharmacologic agents available and the possibility of combining modes of analgesia, the options for patients are numerous and analgesic regimens can be tailored on an individual basis.

The introduction of neuraxial opioids to the analgesic armamentarium has probably been the most significant improvement. Thoracic epidural analgesia had been attempted in the past for post-thoracotomy pain, but when limited to local anesthetics, hypotension was frequently encountered, so this method was not considered viable for routine use.65 The first advocate for the use of neuraxial opioids was Rudolf Matas himself, who, in 1900, combined morphine with cocaine for spinal anesthesia, to reduce the excitatory effect on the central nervous system caused by cocaine.66 Interest in neuraxial opioid use remained dormant until the 1970s. In 1979, Behar et al, first described the use of epidural morphine for the treatment of pain, and noted its long duration of action.67 Numerous studies have demonstrated advantages of epidural over intravenous opioid analgesia, and epidural analgesia, usually combining dilute local anesthetics with lipophilic opioids, has become routine practice; some advocate it should be used for all post-thoracotomy patients.68 The routine placement of epidural catheters for management of postoperative pain has contributed to the formation of acute pain services, and thus expanded the role of anesthesiologists.69 Paravertebral blockade has also been advocated as an alternative to thoracic epidural analgesia. Many studies have demonstrated analgesic equivalence between the two techniques, while paravertebral blocks consistently have fewer side effects.70 Whether paravertebral blockade will replace thoracic epidural analgesia as the gold standard technique remains to be determined. Intercostal blocks can also be performed before or after thoracic procedures for postoperative analgesia, and they can be performed by the surgeon from within the thorax.


Thoracic surgical procedures have increased in both numbers and complexity, and the increased quality and diversity of anesthetic methods for caring for these patients has contributed to this development. Lung cancer continues to be a major public health problem, with 219,440 estimated new cases of lung cancer in the United States in 2009.71 Since the development of antibiotics, malignancy has been the most common indication for pulmonary surgery. However, important procedures for nonmalignant disease, such as lung transplantation and lung volume reduction surgery (LVRS), are now performed routinely at academic centers, thus making the frailest patients surgical candidates. Lung transplantation has increased from 33 transplants performed in the United States in 1988 to 1478 in 2008.72 The most common indications for transplantation are severe chronic obstructive pulmonary disease (COPD), followed by idiopathic pulmonary fibrosis, cystic fibrosis, alpha1-antitrypsin deficiency and primary pulmonary hypertension. LVRS is an option for patients with severe upper lobe emphysema to improve quality of life; however, the surgery remains controversial because of the very high cost of the surgery and rehabilitation. Also, postoperative mortality is high and careful preoperative selection of patients is crucial for identifying those who will benefit. Alternative, nonsurgical, approaches to lung volume reduction surgery, specifically endobronchial insertion of bronchial valves or injection of tissue fibrin glue, are under investigation and being introduced into clinical practice.

Progress in surgically treating patients with such compromised pulmonary function has increased the need for anesthesiologists to be involved as perioperative physicians, in addition to their role intraoperatively. Careful preoperative evaluation of patients for thoracic surgery is crucial so that anesthetic management can be tailored appropriately, including making appropriate plans for postoperative management. Anesthesiologists are also involved in perioperative pain management as well as management of those patients requiring ICU care postoperatively. Because of the variety of roles anesthesiologists fill when caring for patients undergoing thoracic surgery, care for these patients exemplifies the expanded role of anesthesiologists as perioperative physicians.

Another major advance in thoracic surgery has been the development of minimally invasive techniques. The success of laparoscopy for minimally invasive abdominal surgery in the 1980s, as well as improvements in endoscopic video systems and instruments spurred thoracic surgeons to develop minimally invasive techniques of their own. Video-assisted thoracoscopic surgery (VATS) has been widely performed since the early 1990s, and is increasingly replacing traditional open approaches for more and more complex procedures. VATS requires lung separation with OLV for adequate surgical exposure because retraction of the operative lung by the surgeon is not possible. The benefits of VATS over open techniques include less postoperative pain and shorter hospital stays with faster recovery of preoperative function.73 Patients are demanding minimally invasive surgery and forcing surgeons to become more agile with these techniques. There have been concerns, particularly in Europe, that VATS may be inferior treatment for early-stage malignancy; however, available data confirms that VATS lobectomy for early-stage lung cancer is equivalent to thoracotomy in terms of survival.74 Minimally invasive esophagectomies and mediastinal procedures are also now frequently performed. Robotic-assisted techniques for thoracic procedures have also attracted interest, but the benefits and utility of the robot in the thorax still needs to be defined.

Modern thoracic anesthesia with the prevalence of VATS has increased the need for anesthesiologists to master OLV, and has spurred the development of new techniques for lung separation, especially the proliferation of bronchial blockers in recent years. The Arndt blocker (Cook Critical Care, Bloomington, IN), introduced in 1994, is wire-enabled and requires coaxial placement for fiber-opticbronchoscopic guided placement. In 2004, the Cohen tip deflecting endobronchial blocker (Cook Critical Care) was introduced. It possesses a rotating wheel for flexing the tip of the blocker and can be placed under either coaxial or parallel bronchoscopic guidance. Fuji Systems now also manufactures a bronchial blocker, the Uni-blocker that is essentially the bronchial blocker from the Univent tube sold separately. The newest bronchial blocker, the EZ blocker (EZ blocker BV, Rotterdam, The Netherlands) has a novel design featuring a bifurcated distal end.

Dual lumen endotracheal tubes (DLTs) also continue to be refined. At least five different manufacturers now produce DLTs for either the right or the left bronchus in a variety of sizes. A relatively new DLT, the Silbroncho, is a left-sided DLT made of silicone rubber with a wire-reinforced tip. Proposed advantages of the Silbroncho include a smaller cuff to prevent left upper lobe occlusion, and the flexible, reinforced tip is intended to prevent bronchial lumen compression.75

With the array of bronchial blockers and DLTs now available, OLV is easier, safer, and more versatile than ever. Today, single-lumen endotracheal tubes are only rarely used for OLV for adults because of the availability of DLTs and bronchial blockers that are better suited for lung separation. However, they are still used frequently for children because the relatively small airways of infants and small children cannot accommodate DLTs and placement of bronchial blockers can also be very challenging. The proliferation of tools and techniques for OLV has also been spurred by increased use of thoracic and minimally invasive approaches to spinal, cardiac, esophageal, and vascular procedures. Such a wide range of procedures requiring OLV has made facility with these techniques a necessity for most anesthesiologists because these surgical techniques must be aborted without adequate lung separation.

Anesthesiologists are also frequently involved in other types of thoracic procedures. Tumors of the bronchi and trachea are frequently treated with stents and or laser therapy. Airway stenting to palliate patients with severe airway obstructions, usually due to malignant causes, has become increasingly common. These procedures may require special ventilatory techniques, such as high-frequency jet ventilation or the Sanders injection system. Also, stents are now frequently placed by interventional pulmonologists outside of the OR, posing unique challenges to the anesthesiologist.

The variety and complexity of procedures now routinely performed by thoracic surgeons would not be possible without the improvements in anesthetic techniques described here. Anesthesiologists have refined methods of securing the airway, lung isolation, physiologic monitoring, and ventilatory techniques to the point where anesthetizing frail patients for complex procedures appears deceptively easy.


1. Celsus A. De Medicina. Loeb Classical Library, 1938. Accessed September 12, 2009.

2. Strieder JW. Anesthesia from the viewpoint of the thoracic surgeon. Anesthesiology. 1950;11(1):60-64.

3. Souttar HS. A British Medical Association lecture on recent advances in the surgery of the chest. Br Med J. 1926;1(3413):933-936.

4. Matas R. On the management of acute traumatic pneumothorax. Ann Surg. 1899;29(4):409-434.

5. Barry JE, Adams AP. The development of thoracic anesthesia. In: Hurt R. The History of Cardiothoracic Surgery from Early Times. New York: Parthenon; 1996:131-152.

6. Lilienthal H. Resection of the lung for suppurative infections with a report based on 31 operative cases in which resection was done or intended. Ann Surg. 1922;75(3):257-320.

7. Mushin WM, Rendell-Baker L. The Principles of Thoracic Anaesthesia: Past and Present. Springfield, IL: Charles C Thomas; 1953:48-66.

8. Bellani G, Patroniti N, Greco M, Foti G, Pesenti A. The use of helmets to deliver non-invasive continuous positive airway pressure in hypoxemic respiratory failure. Minerva Anestesiol. 2008;74(11):651-656.

9. Meyer W. Pneumectomy with the aid of differential air pressure: an experimental study: the new type of apparatus used. JAMA. 1909;53(24):1978-1987.

10. Meyer W. Some observations regarding thoracic surgery on human beings. Ann Surg. 1910;52(1):34-58.

11. Gillespie NA. The evolution of endotracheal anaesthesia. J Hist Med Allied Sci. 1946;1(4):583-594.

12. O’Dwyer J. Fifty cases of croup in private practice treated by intubation of the larynx. Med Rec. 1887;(32):557-561.

13. Matas R. Intralaryngeal Insufflation. JAMA. 1900;34(23):1468-1473.

14. Meltzer SJ, Auer J. Continuous respiration without respiratory movements. J Exp Med. 1909;11(4):622-625.

15. Elsberg CA. Clinical experiences with intratracheal insufflation (Meltzer), with remarks upon the value of the method for thoracic surgery. Ann Surg. 1910; 52(1):23-29.

16. Lilienthal H. The first case of thoracotomy in a human being under anaesthesia by intratracheal insufflation. Ann Surg. 1910;52(1):58-66.

17. Elsberg CA. Experiences in thoracic surgery under anaesthesia by the intratracheal insufflation of air and ether. Ann Surg. 1911;54(6):749-757.

18. Mette PJ, Sanders RD. Ventilation bronchoscopy: a new technic. Anaesthesist. 1968;17(10):316-321.

19. Kirstein A. Autoskopie des Larynx und der Trachea. Berl Klin Woch. 1895;(32): 476-478.

20. Jackson C. The technique of insertion of intratracheal insufflation tubes. Surg Gyn Obst. 1913;(17):507-509.

21. Guedel AE, Waters RM. A new intratracheal catheter. Current Researches in Anesth Analg. 1928;(7):238-239.

22. White GM. Evolution of endotracheal and endobronchial intubation. Brit J Anaesth. 1960;32(5):235-246.

23. Waters RM. Arthur E Guedel. Brit J Anaesth. 1952;24(4):292-299.

24. Fisher K, Lee LD, Stegeman DE, et al. Anesthetic management for surgery within the thorax: a report of 309 cases. Anesthesiology. 1948;9(6):623-636.

25. Brunn H. Surgical priniciples underlying one-stage lobectomy. Arch Surg. 1929;18(1):490-515.

26. Herbsman H. Early history of pulmonary surgery. J Hist Med Allied Sci. 1958;13(3):329-348.

27. Graham AE, Singer JJ. Successful removal of an entire lung for carcinoma of the bronchus. JAMA. 1933;101(18):1371-1374.

28. Ohsawa T. Surgery of the esophagus. Arch Jpn Chir. 1933;(10):605-608.

29. Magill IW. Anaesthesia in thoracic surgery with special reference to lobectomy. Proc Roy Soc Med. 1936;29(6):643-653.

30. Nosworthy MD. Anesthesia in chest surgery. Proc Roy Soc Med. 1941;34(8):479-506.

31. Gale JW, Waters RM. Closed endobronchial anesthesia in thoracic surgery. J Thorac Surg. 1931;(1):432-437.

32. Rovenstine EA. Anaesthesia for intrathoracic surgery: the endotracheal and endobronchial techniques. Surg Gyn Obst. 1936;(63):325-330.

33. Archibald E. A consideration of the dangers of lobectomy. J Thorac Surg. 1935;(4):335-351.

34. Rusby NL, Thompson VC. Carcinoma of the lung: diagnosis and surgical treatment. Postgrad Med J. 1943;19(207):44-53.

35. Crafoord C. On the technique of pneumonectomy in man. Acta Chir Scand. 1938;(81):5-142.

36. Vellacott WN. A new endobronchial tube for broncho-pleural fistula repair. Brit J Anaesth. 1954;26(6):442-444.

37. Macintosh R, Leatherdale RA. Bronchus tube and bronchus blocker. Brit J Anaesth. 1955;27(11):556-557.

38. Gordon W, Green R. Right lung anaesthesia: anaesthesia for left lung surgery using a new right endobronchial tube. Anaesthesia. 1957;12(1):86-93.

39. Kamaya H, Krishna PR. New endotracheal tube (Univent tube) for selective blockade of one lung. Anesthesiology. 1985;63(3):342-343.

40. Slinger PD, Lesiuk L. Flow resistances of disposable double-lumen, single-lumen, and Univent tubes. J Cardiothorac Vasc Anesth. 1998;12(2):142-144.

41. Ginsberg RJ. New technique for one-lung anesthesia using an endobronchial blocker. J Thorac Cardiovasc Surg. 1981;82(4):542-546.

42. Arndt GA, Buchika S, Kranner PW, DeLessio ST. Wire-guided endobronchial blockade in a patient with a limited mouth opening. Can J Anaesth. 1999;46(1):87-89.

43. Cohen E. The Cohen flexitip endobronchial blocker: an alternative to a double lumen tube. Anesth Analg. 2005;101(6):1877-1879.

44. Mungroop HE, Mijzen L, Woltersom B, et al. Feasibility of a new Y shaped bronchial blocking device and comparison with a DLT: a Manikin Study. 2009 ASA meeting. Abstract A541.

45. Bjork VO, Carlens E. The prevention of spread during pulmonary resection by the use of a double lumen catheter. J Thorac Surg. 1950;20(1):151-157.

46. Bryce-Smith R. A double-lumen endobronchial tube. Brit J Anaesth. 1959;(31):274-275.

47. Bryce-Smith R, Salt R. A right-sided double lumen tube. Brit J Anaesth. 1960;(32):230-231.

48. White GM. A new double lumen tube. Brit J Anaesth. 1960;(32):232-234.

49. Robertshaw FL. Low resistance double-lumen endobronchial tubes. Brit J Anaesth. 1962;(34):576-579.

50. Ehrenwerth J. Pro: proper positioning of a double-lumen endobronchial tube can only be accomplished with endoscopy. J Cardiothoracic Anesth. 1988;2(1):101-104.

51. Campos JH. Current techniques for perioperative lung isolation in adults. Anesthesiology. 2002;97(5):1295-1301.

52. Griffith HR and Johnson GE. The use of curare in general anesthesia. Anesthesiology. 1942;3(4):418-420.

53. Harroun P, Beckert FE, Hathaway HR. Curare and nitrous oxide anesthesia for lengthy operations. Anesthesiology. 1946;7(1):24-28.

54. Anderson E, Crafoord C, Frenckner P. A new and practical method of producing rhythmic ventilation during positive pressure anaesthesia with description of apparatus. Acta Oto-laryngologica. 1940;28(1):95-102.

55. Ibsen B. The anaesthetist’s viewpoint on the treatment of respiratory complications in poliomyelitis during the epidemic in Copenhagen, 1952. Proc Roy Soc Med. 1954;47(1):72-74.

56. Bjork VO, Engstrom CG. The treatment of ventilatory insufficiency after pulmonary resection with tracheotomy and prolonged artificial ventilation. J Thorac Surg. 1955;30(3):356-367.

57. Severinghaus JW, Honda Y. Pulse oximetry. Int Anesthesiol Clin. 1987;25(4):205-214.

58. Severinghaus JW. Takuo Aoyagi: discovery of pulse oximetry. Anesth Analg. 2007;105(supp 6):S1-4.

59. Luft K. Method der registrieren gas analyse mit hilfe der absorption ultraroten Strahlen ohne spectrale Zerlegung. Z Tech Phys. 1943;(24)97.

60. Raventos J. The action of fluothane-a new volatile anaesthetic. Br J Pharmacol. 1956;11(4):394-410.

61. Gregory GA, Kitterman JA, Phibbs RH, et al. Treatment of the idiopathic respiratory distress syndrome with continuous positive airway pressure. NEJM. 1971;284(24):1333-1340.

62. Capan LM, Turndorf H, Patel C, Ramanathan S, Acinapura A, Chalon J. Optimization of arterial oxygenation during one-lung anesthesia. Anesth Analg. 1980;59(11):847-851.

63. Cohen E, Eisenkraft JB. Positive end-expiratory pressure during one-lung ventilation improves oxygenation in patients with low arterial oxygen tensions. J Cardiothorac Vasc Anesth. 1996;10(5):578-582.

64. Godet G, Bertrand M, Ben Ammeur M, et al. Comparison of HFJV vs CPAP for separate lung ventilation in patients surgically treated aneurysm of the thoracic aorta. Ann Fr Anesth Reanim. 1989; (supp8): R266.

65. Shuman RL, Peters RM. Epidural anesthesia following thoracotomy in patients with chronic obstructive airway disease. J Thorac Cardiovasc Surg. 1976;71(1)82-88.

66. Matas R. Local and regional anesthesia with cocaine and other analgesic drugs, including the subarachnoid method, as applied in general surgical practice. Philadelph Med J. 1900;(6):820-843.

67. Behar M, Magora F, Olshwang D, Davidson JT: Epidural morphine in treatment of pain. Lancet. 1979;1(8115):527-529.

68. Slinger PD. Pro: every postthoracotomy patient deserves thoracic epidural analgesia. J Cardiothorac Vasc Anesth. 1999;13(3):350-354.

69. Ready LB, Oden R, Chadwick HS, et al. Development of an anesthesiology-based postoperative pain management service. Anesthesiology. 1988;68(1):100-106.

70. Joshi GP, Bonnet F, Shah R, et al. A systematic review of randomized trials evaluating regional techniques for postthoracotomy analgesia. Anesth Analg. 2008;107(3):1026-1040.

71. National Cancer Institute. Surveillance epidemiology and end results. Accessed October 12, 2009.

72. Organ procurement and transplant network data reports. Accessed October 12 2009.

73. Demmy TL and Nwogu C. Is video-assisted thoracic surgery lobectomy better? Quality of life considerations. Ann Thorac Surg. 2008;85(2):S719-728.

74. Flores RM and Alam N. Video-assisted thoracic surgery lobectomy (VATS), open thoracotomy, and the robot for lung cancer. Ann Thorac Surg. 2008;85(2):S710-715.

75. Lohser J, Brodsky JB Silbroncho double-lumen tube J Cardiothorac Vasc Anesth. 2006;20(1):129-131.