Most pulmonary resection today is for bronchogenic carcinoma. This disease for all practical purposes did not exist at the turn of the century! Although the first case was reported by Agricola in 1521, and the first case in the U.S. was reported in 1851, a mere 211 cases had been reported in the surgical literature by the year 1900, and only 374 cases by 1912. Many of these cases were metastases or sarcomas. By 1927 the condition was considered epidemic in the western world; second only to gastric cancer in numbers. Mystified physicians tried to explain the marked increase in cases to syphyllis, tuberculosis, influenza, cobalt, tar, automobile exhaust gases etc. but Rigdon credits Soemmerling with being the first to suspect the true culprit, cigarette smoking. We can now credit American and British tobacco companies with the causation of most of the the many millions of deaths caused by this disease in the past century.

Thus, in the early experience, bronchogenic carcinoma was a rare disease, and one which was rarely discovered at an early enough stage to allow resection. Heidenhain resected a small lung cancer incidentally in a case of bronchiectasis in 1901 and Davies performed a lobectomy for lung cancer in 1912, but the patient died post-op. Chest roentgenogram of Davies 1912 case

Tuberculosis was the most common pulmonary disease, which killed approximately 90,000 Americans in 1930, but it was usually bilateral and thus not suitable for surgical ablation. Bronchiectasis, an unusual disease outside of the third world nations today, was then a common and debilitating disease. The fact that it was often unilateral and limited to one or two lobes, made it the logical target of most early pulmonary resections. Furthermore, it usually involved the lower lobes which are technically easier to resect.

In 1932 Evarts Graham performed a pneumonectomy upon Dr. Gilmore, who had a large bronchogenic carcinoma, and the patient survived . Most textbooks and articles today refer to Graham's pneumonectomy as a spectacular technical feat and an epochal turning point in thoracic surgery. This is certainly not accurate. In reality, pulmonary resection techniques developed painfully slowly over the fifty year period after 1883, in hundreds of hospitals and surgical clinics throughout the world. In the 1880s animal experimentation was carried out by Biondi, Gluck, Block and Schmid. During the 1890s sporadic attempts at resection in humans were carried out by Tuffier, MacEwen and Gluck.44 During the period 1904 to 1929, surgeons began to specialize and perform series of pulmonary resections. By 1929, thoracic surgery had become an established specialty, and surgeons throughout the world were involved in a collaborative effort that culminated in safe one-stage pulmonary resections. General scientific principles proved to be of less importance than development and accretion of simple and reliable techniques. Success was finally achieved through the combined efforts of a number of surgeons working in the U.S., Canada and England who communicated frankly with one other in journals and in surgical societies, and cooperated so that other surgeons would have the benefit of the lessons learned by them in a painful process of trial and error. Surgical morbidity and mortality were horrifying at first, and only the bravest patients and the most resilient surgeons chose to continue to work in the field.

Before the advent of successful resection, most surgical treatment was limited to techniques designed to collapse the lung. This followed upon observations by clinicians that the course of tuberculosis seemed to be improved if the diseased lung could be put at rest by collapsing it.15 This collapse was either temporary via induced pneumothorax (Forlanini and Murphy)19 20 or permanent by phrenicectomy , thoracoplasty or plombage. It is difficult to be sure today whether the gruesome deformities inflicted on patients subjected to extensive thoracoplasties had any beneficial effects on tuberculosis, but it was clear to the physicians of the time that collapse therapy was useless in the bronchiectatic patient. This patient was condemned to a life of misery from recurrent infections, and rendered unemployable and a social outcast because of his cough and malodorous sputum. Relief could only be obtained by surgical removal of the chronically infected lung tissue. It was on these unfortunate patients that most of the first lobectomies were performed.

Resection of a bronchiectatic lobe was first done in 1901 by Heidenhain. Subsequent to Sauerbruch's paper there was increased activity in the field of thoracic surgery as surgeons in different countries began to try to carry out pulmonary resections. At first these were isolated, often desperate operations, but eventually a few men began to accumulate experience and perform organized series of cases - Sauerbruck in Germany, Davies , Roberts and Edwards in England, Meyer and Lilienthal in New York , Graham in St. Louis, Robinson in Minnesota, Alexander in Michigan and Shenstone and Archibald in Canada. Looking back it is hard to understand the rationale for some of the methods used. They often seem absurd and completely illogical. The techniques were, however, completely consistent with the limits of the technology existing at the time.

Most important were the deficiencies of the anaesthetic techniques. Working without adequate control of the airway, ventilatory support was difficult and the patient quickly deteriorated once the chest was open. This meant that the operation had to be completed quickly. This ruled out careful dissection and individual vessel ligations. Anatomical knowledge was meager for this purpose anyway. For similar reasons, conditions were not conducive to attempts at secure bronchial closure. These conditions mandated a clamp, mass ligature or tourniquet type of resection, and this method was adopted almost universally. A clamp, mass ligature or tourniquet was first placed on the hilum of the lobe to be removed. Sauerbruck at first merely ligated the lobar pulmonary artery and closed. The result was infarction and necrosis of the lung and pleural empyema which could be drained at a second stage. Robinson, at the Mayo Clinic, who had studied with of Sauerbruch, advocated a multiple stage approach with a preliminary partial thoracoplasty and pleurodesis followed by a second stage in which the hilum of the lower lobe was clamped and the lobe amputated. The individual vessels and bronchus were individually ligated, and then a mass ligature was tied behind the clamp.The pleural space was then left open and packed until the inevitable empyema and bronchial fistula had healed. Lilienthal did a thoracotomy without thoracoplasty and a pleurodesis with tincture of iodine or iodoformized gauze, and one week later returned to ligate the lobar hilum with heavy silk "chain" ligatures followed by resection, treatment of the bronchial stump with phenol and open packing of the pleural cavity. Mortality in these two series was 43 and 45%! Friederich's two and Sauerbruck's three attempts at one stage lobectomy with closure of the wound all resulted in death by tension pneumothorax.

If the surgeon chose to resect the lobe and oversew the tissues at the hilum, including pulmonary artery, vein and bronchus, the process was speeded up, but this was a crude and dangerously uncertain method. A slipped ligature meant sudden exsanguinating hemorrhage. Bronchopleural fistula, empyema and death were almost inevitable consequences of breakdown of the inadequate bronchial closure. Closed drainage was seldom used and was generally felt to be deleterious. If the chest was closed, blood and serous secretions would fill the hemithorax, collapsing the lung, and air leak, almost inevitable due to the crudity of technique, represented a grave risk of death by tension pneumothorax.

One alternative was to leave the infarcted lung in place and to pack the wound open. The lung would then necrose and slough leaving the surgeon and patient to deal with an almost inevitable bronchopleural fistula and empyema, a condition scarcely better than the preoperative illness. Another problem was that if the pleural cavity was left open, the remaining lobes would, of course, collapse in the presence of an open thoracostomy. What to do? Some surgeons did a preliminary rib resection to allow partial collapse of the chest wall so that the lung could be exteriorized. This also had the supposed advantage of stabilizing traction on the mediastinum. The method was not effective and was soon abandoned. The solution adopted by most surgeons was preliminary pleurodesis. In most cases of inflammatory disease of the lung, there are strands of scar tissue, adhesions, that afix the lung to the chest wall. It was early noted that, if there were such adhesions, the remaining lobes would not collapse, and there would be less respiratory distress during the operation. Also, the pleural cavity would be partially obliterated and thus less subject to contamination by empyema. Finally, the size of the post-operative empyema cavity would be smaller and healing of the bronchopleural fistula could be more rapid. These considerations led to a number of different methods of staged resection. During the first stage the surgeon made the thoracotomy incision to the level of the parietal pleura. If he could see the lung moving, through the transparent parietal pleura, he knew that there were no adhesions. Adhesions could then be produced by any of the twenty-one different methods of pleurodesis described by Bethune, and sometimes a partial thoracoplasty, was performed at the same time. The chest was then reclosed and a period of time allowed to pass for development of adhesions. If there were adhesions seen at the time of the first stage operation, the surgeon could proceed directly with the resection.

It was now time to resect the lung. At the second operation, the adhesions surrounding the lobe to be resected were taken down, but the adhesions to the other lobes were left undissected. The adhesions would now be a distinct disadvantage. Freeing up the lobe to be resected would require difficult dissection, resulting in bleeding and air leakage from the delicate tissue of the lung. Since hilar dissection would be very difficult under these conditions, a mass ligature technique was applied.

Shenstone did a thoracoplasty and six weeks later a lobectomy "attended by considerable haemorrhage and shock". Later he omitted the thoracoplasty and proceeded directly to thoracotomy, using snare devices to "lasso" the hilum of the lobe before proceeding with amputation, suture and closure with suction drainage.52 Others used a clamp or a rubber tourniquet for the same purpose. The lung tissue distal to the clamp or tourniqet could be amputated, and the pedicle oversewn, with the suture material closing pulmonary artery, vein and bronchus in a mass ligature. Some surgeons decided to avoid the immediate risk of massive hemorrhage due to slipped ligature and left the lung in place. The wound was left open and packed. The lobe could either be removed at a third stage or allowed to necrose and slough out . Graham destroyed the ligated lung with a cautery. Operative mortalities for these proceedures were as high as 50% in the best hands. It is not surprising that Dr. Samuel Robinson could find little cause for optimism in his presidential address of 1925 to the Association of Thoracic Surgeons. He found that most surgeons became quickly discouraged and gave up doing thoracic surgery. Here Robinson discribes a typical thoracic surgical operation of his day.

"The patient is placed on the operating table . The posture is uncomfortable. There may be cyanosis. It induces coughing. The anaesthetist is greeted by an evacuation of a large amount of pungent , purulent sputum, incident to the posture on the table . The whole bronchial tree may be filled with this material as the anesthetist begins. If regional blocking and paravertebral procainization is carried out, the pleura is no sooner opened and traction on the diaphragm commenced than the need of general anesthesia is obvious . As the secretions well towards the trachea, the cyanosis increases. The lower lobe obstinately resists being delivered; the pleural adhesions are strong and widespread; the attachments to the diaphragm are ropelike and tenacious, Finger dissection is inadequate. Work with the knife and scissors is blind. Cleavages are sought in vain. The pericardium is dangerously involved in the adhesions. Meanwhile, the patient's condition may become distressing and perhaps alarming. If open pneumothorax is adding insult to injury, the lung cannot be used to plug the thoracic gap, because the lobe is not deliverable. If differential pressure is being employed , that too may be acting badly. The mucopurulent secretion may interfere with proper intratracheal air insufflation. It interferes with the respiration under positive pressure. There may be cyanosis even with the head outside a negative pressure cabinet. And then the difficulties multiply. The complete liberating at one sitting may have to be abandoned. There is bleeding and infectious leakage from the lung, and bleeding from the diaphragm. Tight closure of the chest without drainage seems inadvisable under such conditions, and yet necessary to avoid the ills of postoperative pneumothorax. Suddenly, it is obviously time to return the patient to his bed. Not much has been accomplished..... And, after all, the greatest dangers occur after the operation."

Despite this discouraging outlook, modern pulmonary resection emerged over the following ten years. During this era, surgeons began to specialize in thoracic surgery. Such specialists began to experience decreasing mortality as their experience and skill improved. The beginning of the modern era began with Harold Brunn in San Francisco who reported on a series of one stage pulmonary resections in 1929. The major difference in his technique was that he drained all chests with a chest tube which was aspirated by hand syringe every two hours . Shenstone, in Canada used basically the same technique but added siphon drainage.

A very important aspect of these early experiences was the collegiality and honesty of the surgeons involved. Robinson, Lilienthal and Meyer met and frankly discussed their morbidity and mortality results and described the drawbacks of their techniques. So also did Shenstone, who credited Brunn with providing information that resulted in improvements in his own results.

continue

Back to the Home Page