T thoracic cavity volume, either to prevent or to control

Transcription

1 Thoracoplasty in the Context of Current Surgical Practice George Peppas, D, Thomas. olnar, D, Kumarasingham Jeyasingham, RCS, and Alan B. Kirk, RCS Department of Thoracic Surgery, renchay Hospital, Bristol, England Although widely employed for well over a century as a procedure for reducing the capacity of the thoracic cavity, thoracoplasty in current practice has become a rarity. A retrospective analysis of 37 patients (29 men, 8 women) who underwent the procedure under the care of one thoracic surgeon in a 16-year period provides the basis for this presentation. Ages ranged from 23 to 82 years with a mean age of 58 f The mean follow-up was 8.5 years. Nineteen patients underwent the procedure for complications after lung resection for lung cancer. There were four perioperative deaths in this group (21.1%) and 6 long-term survivors (31.6%). Eighteen patients without lung cancer underwent thoracoplasty as a planned treat- ment or for complications. There were no perioperative deaths, two late deaths, and 16 long-term survivors (88.9%) in the group. In the entire series, the overall perioperative mortality rate was 10.8% with no major long-term morbidity. Although proper timing and proper patient selection are essential in the use of thoracoplasty as a procedure to cope with the septic complications of lung cancer resection, it is overall a safe and successful procedure that has a relatively low mortality and that leads to considerable improvement in quality of life. ( 1993;56:903-9) horacoplasty is a remodeling procedure to reduce the T thoracic cavity volume, either to prevent or to control infection already established in the pleural cavity. Hippocrates is credited with the first rib resection for drainage of an infected pleural space [l]. In more modern times, Simon in 1869 and Estlander [2] in 1879 were the earliest surgeons to employ thoracoplasty to control empyema thoracis. In 1885, de Cerenville [3] described the removal of the second and third ribs with their periosteum to diminish the volume of the apex of the chest. Schede [4] in 1890 described the first total thoracoplasty, removing the ribs with their periosteum and the intercostal muscles, thereby leaving only the pectoral muscles and skin as a cover for the underlying thoracic viscera. Apart from minor modifications to these techniques by riedrich [5], Wilms [6], and Sauerbruch [7], the next important step came with Semb s [8] technique of apicolysis to lower the cupola of the chest by releasing the apical fascia, thereby encouraging the lung to drop from the apex. Alexander [9] finally popularized the operation of thoracoplasty as we know it today. Until the early 1950s, the main indications for thoracoplasty were to treat pulmonary tuberculosis and control empyema. In recent years, the main indications for thoracoplasty have been restricted to postpneumonic and postresectional empyema with or without a bronchopleural fistula [lo]. Taking into consideration the complexity of the problems and the wide variety of clinical presentations, we Accepted for publication Dec 4, Address reprint requests to r Jeyasingham, Department of Thoracic Surgery, renchay Hospital, Bristol BS16 ILE, England. decided a prospective study of thoracoplasty and its applications was not feasible. However, we thought that a retrospective analysis of the use of thoracoplasty in one surgeon s practice in the past 17 years would provide some useful information. aterial and ethods Patient Population During the period 1975 through 1991, 37 patients underwent thoracoplasty under the care of one surgeon. Data relating to these patients were collected retrospectively. There were 29 men and 8 women with a mean age of years. Nineteen patients had the operation to control complications of resection for lung cancer, and 18 patients had the operation during the course of management of disease not related to lung cancer. In these 18 patients, the male to female ratio was 2:l as opposed to 18:l in the lung cancer group. Ten variables were examined: type of thoracoplasty; use of myoplasty; closure of bronchopleural fistula; obliteration of pleural space; control of space infection; healing of chest wound and sinuses; chronic morbidity related to thoracoplasty; perioperative mortality; long-term survival; and in patients who underwent operation for lung cancer, TN staging of the cancer. Operative Technique Thoracoplasty was undertaken only when adequate drainage of the infected pleural space had been achieved with an intercostal drain positioned in a dependent site. General anesthesia was maintained with a double-lumen endotracheal tube to isolate each half of the bronchial tree by The Society of Thoracic Surgeons /93/$6.00

2 904 I EPI AS ET AL ODERN THORACOI LASTY 1993;56:90>9 Appropriate perioperative antibiotics and prophylactic low-dose heparin sodium therapy were routinely employed. A posterolateral thoracotomy incision that extended vertically upward from its posterior limit provided adequate exposure of the upper ribs. Commencing with the second rib, a variable number of ribs down to the eighth rib were resected. In the majority of patients, four to five ribs were removed, with care taken that the inferior tip of the scapula did not impinge on the remaining ribs. The periosteum over the first rib was incised the length of the rib and separated from it, while further blunt dissection was continued upward into the apex to free the apical thoracic fascia. Then the apex of the lung and the soft tissues were retracted downward to obliterate the underlying space and approximate the soft tissues to the mediastinum (apicolysis). Excision of the ribs from two downwards was done subperiosteally and fashioned in such a manner that an increasing length of the anterior ribs was retained as the rib resections were carried inferiorly. Special attention was paid to removal of the back ends of the ribs completely. The transverse processes of the vertebrae were always preserved on the left side and were removed on the right side only if the paravertebral space was not already obliterated by granulation tissue. In patients with a bronchopleural fistula, direct suture was attempted, and where appropriate, a myoplastic flap was created from the intercostal muscles to support the repair of the fistula. Drainage of the pleural space was continued until such time as complete obliteration of the space was achieved. The thoracoplasty incision was repaired in the standard fashion by approximating the divided pectoral muscles and approximating the skin edges. When the standard procedure was carried out, we designated the operation a standard thoracoplasty. When the second to the fifth ribs were removed, the procedure was called a standard limited thoracoplasty. When the extent of rib resection extended to the sixth rib or further, we designated it a standard full thoracoplasty. In a small number of patients with unresolved localized empyema over a contracted lung, a tailored thoracoplasty was used to collapse the overlying chest wall. When ribs were removed as part of a procedure for correction of spinal deformities, we designated the procedure a cosmetic thoracoplasty. Postoperative anagement The postoperative management of patients included the application of compression pads to the chest wall to maintain the obliteration of the pleural space and to minimize the accumulation of blood in the subscapular space (Semb s space). An axillary and an anterior pectoral cotton pad of sufficient bulk were retained in position by torso bandages. This last step of compression is possibly the most important single step in the entire operation of thoracoplasty. Compression is maintained for a minimum of 2 to 3 weeks. Physical therapy to ensure good chest expansion and shoulder mobility is essential from the moment the patient recovers from anesthesia. Results Lung Cancer Group TYPES O THORACOPLASTY. Among the 1 9 patients who had thoracoplasty because of complications of resection for lung cancer, the indication for thoracoplasty was an already established empyema (13 patients), an uncontrolled bronchopleural fistula (16 patients), or both (Table 1). In 1 patient, thoracoplasty was done to control a malignant septic pleural effusion over a contracted, cancerous lung without previous surgical intervention. In 7 patients, the bronchopleural fiistula was repaired within 48 hours after its occurrence, and in these patients, repair of the fistula was combined with thoracoplasty as a concomitant procedure to support the repair. In 2 patients, both of whom had an acute bronchopleural fistula, a pedicled myoplasty was used as an additional step to lend support to the bronchial stump. In 8 patients, the bronchopleural fiistula had been established for some time, with gross sepsis of the chest cavity. In these 8, although an atteinpt was made to surgically close the stump, success of stump closure was dependent more on space obliteration and infection control than on the stump suture. In 1 patient with a long-established epithelialized sinus track from the bronchus to the skin after a lower lobectomy, a pedicled myoplasty was combined with thoracoplasty to achieve successful closure of the fistula. ourteen patients required a standard full thoracoplasty, 2 patients had a standard limited procedure, and 3 patients underwent a tailored thoracoplasty, with the second rib left in place. SUCCESS O THORACOPLASTY. The success of thoracoplasty as an appropriate procedure for control of bronchopleural fistula in the 16 patients was judged on the basis of closure of the fistula, obliteration of the pleural space, and control of infection. our patients died in the perioperative period of infection and respiratory failure complicating sepsis of the contralateral lung. Two patients survived 6 and 12 weeks without achieving these criteria, the cause of death being attributed to tumor disease. our patients were alive for 4 to 12 months with successful control of space infection. Six patients are alive 5 to 16 years after thoracoplasty and have met the three criteria of success. In this group of patients, none of the long-term survivors has any chronic morbidity related to the operation, whereas in the short-term survivors, it was difficult to distinguish morbidity related to sepsis from that related to the thoracoplasty itself. The perioperative mortality was 4 of 19 (21.1%). Six patients (31.6%) survived 5 years or more after thoracoplasty, with a mean follow-up of 8.5 years (range, 5 to 16 years). Group Without Lung Cancer TYPES O THORACOPLASTY. Of the 18 patients who underwent thoracoplasty for disease not related to lung cancer, 9 required a standard limited operation, 4 had a standard full thoracoplasty, and 3 had a tailored procedure (Table

3 1993;56:90>9 PEPPAS ET AL 905 Table 1. Summary of Data on Patients Who Had Thoracoplasty After Resection for Lung Cancer Lung Patient Age Cancer Indication Type of Previous Current No. (y) Sex Staging for Thoracoplasty Thoracoplasty Ribs Date Operation Status 1 56 T2 N1, BP L Pneumo 1975 Alive, 16 y 2 57 T2 N2, BP L Pneumo 1975 Dead, periop, 1 d 3 43 T2 NO, BP Std Itd R Pneumo 1976 Alive, 14 y 4 60 T2 N1, BP Tailoredb L Pneumo 1977 Dead, 17 mo 5 63 T2 NO, BP L Pneumo 1979 Dead, periop, 5 wk 6 57 T2 N1, BP R Pneumo 1979 Dead, periop, 4 d 7 67 T2 N1, BP R Pneumo 1979 Dead, periop, 3 wk 8 71 T3 Nx Neoplastic effusion Tailored Dead, 6 wk 9 60 T1 NO, BP R Pneumo 1981 Dead, 9 mo T2 NO, BP b RUL 1984 Dead, 1 y T1 NO, BP L Pneumo 1985 Dead, 4 mo T3 NO Tailored R Pneumo 1985 Alive, 6 y T2 NO, BP L Pneumo 1985 Alive, 6 y T3 N2, BP R Pneumo 1986 Dead, 9 mo T2 NO, BP R Pneumo 1986 Alive, 5 y T1 NO R Pneumo 1984 Dead, 4 mo T2 NO, BP L Pneumo 1986 Dead, 10 mo T2 NO, BP b RLL 1986 Alive, 5 y T2 NO, BP R Pneumo 1987 Dead, 6 mo a Simultaneous OG and lung resection were performed. Concomitant myoplasty was done. Simultaneous gastric bypass and lung resection were done. BP = bronchopleural fistula; ltd = limited; Pneumo = pneumonectomy; RLL = right lower lobectomy; RUL = right upper lobectomy; Std = standard. 2). Two patients underwent a cosmetic thoracoplasty as part of a corrective procedure for kyphoscoliosis. In 11 patients, the indication for thoracoplasty was an uncontrolled empyema that had not resolved with drainage and decortication. Three of these patients had postpneumonic empyemas. Three empyemas were associated with a fistula; two were bronchopleural fistulas, and one was a tracheoesophagopleural communication resulting from recurrent Hodgkin s disease previously treated with chemotherapy and radiotherapy. In 1 patient an empyema developed after a left upper lobectomy for an unresolved lung abscess with a bronchopleural communication. In 5 patients, thoracoplasty was done for Aspergillusrelated problems. our of these patients had a mycetoma. Three of the mycetomas were resected, and a concomitant limited thoracoplasty was performed. One patient had a cavernostomy and limited thoracoplasty. In the fifth patient, a bronchopleural fistula was sealed with a myoplasty, and a standard full thoracoplasty was completed in two stages. Nine patients had a history of previous tuberculosis that had been treated with antituberculosis drugs. In 3 of these patients, artificial pneumothorax had been induced as part of the therapy 23 to 42 years previously. One patient with a mycetoma had undergone a right upper lobectomy for a lung cyst 28 years previously, and another had undergone a left upper lobectomy for carcinoma 7 years previously, although the subsequent empyema was associated with an active infection with ycobacterium kansasii. One patient had undergone previous pleurectomy for recurrent right pneumothoraces, after which she was treated for active tuberculosis, the final indication for operation being a mycetoma in the cavitating right upper lobe. Three of the patients, in whom aspergillomas developed were in respiratory failure caused by chronic obstructive airway disease. SUCCESS O THORACOPLASTY. Success of the operation as judged by control of space infection, space obliteration, and closure of the bronchopleural fistula was achieved in 15 of 16 patients. Only 1 patient had a persistent discharging sinus despite successful closure of an underlying bronchopleural fistula with a combined myoplasty and thoracoplasty. The two cosmetic operations were deemed successful on the grounds of good skin healing and cosmetic result. Thus the objective was achieved in 17 (94.4%) of 18 patients. Only 1 patient had long-term morbidity related to the thoracoplasty. This patient was in chronic respiratory failure at the time of operation and had to be artificially ventilated for several weeks. A frozen shoulder with fixation of the scapula developed, and continued to be a cause of chronic pain. With the exception of the cosmetic operations for kyphoscoliosis, none of these patients has postoperative deformities (igs 1, 2). There was no perioperative mortality in this group. Two patients died 7 months and 4 years after thoraco-

4 906 PEPPAS ET AL 1993;56:903-9 Table 2. Summary of Data on Patients Without Lung Cancer Who Underwent Thoracoplasty Patient No. Age (Y) Indication for Type of Other Current Sex Diagnosis Thoracoplasty Thoracoplasty Ribs Date Procedures Status Pneumonia TB Pneumonia Pneumonia Hodgkin s Aspergillosis Aspergillosis Aspergillosis Scoliosis Scoliosis ycetoma, BP Empyerna, BP Em p y e m a Em p y e m a ycetoma ycetoma TEP ycetoma ycetoma ycetoma Hump Hump Tailored Tailored Tailored Std Itd Std Itd Cosmetic Cosmetic S yoplasty Decortication, 1980 RUL, 1981 LUL, 1980 Decortication, 1985 Pleurectomy, 1978; RUL, 1987 Radiotherapy, chemotherapy RUL, 1960 Cavernostomy LUL, 1988 Scoliosis repair Scoliosis repair RUL, 1991 Alive, 15 y Dead, 4 y Alive, 13 y Alive, 13 y Alive, 11 y Alive, 10 y Alive, 7 y Alive, 6 y Alive, 5 y Alive, 5 y Alive, 4 y Alive, 4 y Alive, 3 y Dead, 7 mo Alive, 2 y Alive, 2 y Alive, 6 mo Alive, 6 mo a This was a concomitant procedure APN = artificial pneumothorax; BP = bronchopleural fistula; Itd = limited; LUL = left upper lobectomy; RUL = right upper lobectomy; Std = standard; TB = tuberculosis; TEP = tracheoesophagopleural fistula. plasty. In both, chronic respiratory failure contributed to death. Sixteen patients are alive 6 months to 15 years after thoracoplasty with a mean follow-up of 8.5 years. ig 1. (Patient 15, Table 2.) Postoperative roentgenogram made after a left-sided standard limited thoracoplasty performed after left upper lobectomy for a mycetoma. A frozen shoulder developed as the result of prolonged artificial ventilation in the immediate postoperative period. It has become more mobile, but the roentgenogram reveals that the scapula has not dropped. Therefore the patient retains a satisfactoy appearance, both from the front and the back. Comment The purpose of thoracopiasty is to establish contact between the chest wall and the mediastinum or residual lung to promote creation of granulation tissue and thus achieve space obliteration. An operation that was originally devised as a collapse therapy in the surgical management of pulmonary tuberculosis has, with the decreasing need for surgical intervention in this disease, come to be regarded as a procedure needing; an application. However, thoracoplasty still has an important role in the practice of modern thoracic surgery provided the basic principles of thoracoplasty are not violated, the criteria for its application are defined, and great care is exercised in patient selection. Otherwise, an excellent operative procedure developed by the pioneers of thoracic surgery will be misjudged by those not trained to perform it and will be relegated to the status of a has been. We believe that the four basic principles in performing a thoracoplasty are as follows: 1. Providing adequate drainage of space for the control of sepsis before operation. 2. Ensuring that the nutritional and cardiovascular status of the patient are optimum. 3. Removing a sufficient number of ribs for an adequate obliteration of space. The ribs should be disarticulated posteriorly at the costotransverse and costovertebral joints. When more than five ribs are to be removed, the procedure preferably is staged.

5 1993;56:903-9 PEPPAS ET AL 907 A B ig 2. (Patient 28, Table 2.) (A) rontal view and (B) posterior vim after a standard limited thoracoplasty for control of empyema after a right upper lobectomy for a chronic lung abscess. (C) Chest roentgenogram showing that the right scapula is maintained on a level with the left despite the limited thoracoplasty. c 4. Performing an extrafascial apicolysis whenever space to be obliterated is located in the apex of chest. the the The techniques of thoracoplasty in current use are numerous and varied in thei; details. A few examples incorporate the major principles in the current application of this procedure. Intrupleurul thorucoplusty as described by Schede [4] involves the excision of ribs, intercostal muscles, endothoracic fascia, periosteum, and parietal pleura, with only the muscles of the pectoral girdle covered by the skin and subcutaneous tissues left behind. This procedure results in considerable deformity of the chest and spine as well as loss of lung function [ll]. It is rarely performed nowadays in view of the grotesque deformity that follows but is applicable in patients whose chest wall is grossly calcified and in whom separate excision of the ribs with retention of the intercostal muscles is not feasible. Extrapleural thorucoplusty combined with Semb's apicolysis [8], popularized by Alexander [9], is a less radical and less deforming procedure that retains the periosteum of the ribs, the intercostal muscles, and the parietal pleura. When more than four ribs are excised, the procedure is often done in two or three stages to minimize operative blood loss and shock to the patient. The apicolysis ensures adequate collapse of the cupola of the lung and the mediastinum. Tuffier in 1891 introduced the technique of extrapleural pneumonolysis, which entails the separation of the lung and pleura in the extrapleural space over the parietal chest wall to achieve collapse of the apex of the lung [12]. This technique was practiced extensively around the turn of the century and was later modified by the addition of plombage material such as naphthalene balls and newer materials such as polyethylene in the form of balls [13] or packs [14], achieving the same objective. This technique is called plombage thorucoplusty. Iioka and associates [15] described a technique of separating the inner periosteum of the ribs with intercostal muscles and parietal pleura to create an extruperiosteul

6 908 PEPPAS ET AL 1993;56:903-9 thorucoplusty, the plombage material being the patient's own blood and serum, which accumulate in the space. The advantages of extraperiosteal thoracoplasty are said to be the lack of chest wall deformity, the preservation of lung function, and the fact that no foreign body is inserted in a potentially infectable space. The technique of standard thoracoplasty as practiced by us entails retaining the first rib, thereby ensuring that the deformity is minimized by preventing shoulder drop. The apicolysis, however, allows release of the apex of the pleura and collapse of the lung toward the hilum and the mediastinum. The number of ribs excised is dictated by the need of the individual patient and the size and location of the space obliteration required. Scoliosis is minimized by retaining the transverse processes. Like Gregoire and co-workers [16], we believe that preservation of the first rib has to be combined with apicolysis to achieve obliteration of the apical space. Omission of apicolysis may have accounted for some of the failures encountered by Hopkins and colleagues [17] and Horrigan and Snow [I. Contrary to the views expressed by these two groups, space obliteration can be achieved without removing the transverse processes, while avoiding the development of spinal deformities. The indications for thoracoplasty in the present series have ranged from management of bronchopleural fistula after lung resection for malignancy to prevention of bronchopleural fistula and space infection in patients undergoing operation for complications of mycetoma in Aspergillusinfiltrated lungs. In some patients with a bronchopleural fistula, thoracoplasty was undertaken within 24 to 48 hours after the occurrence of the fistula to cope with the excessive loss of tidal volume associated with the fistula once the pneumonectomy space had been controlled by evacuating all accumulated fluid by tube thoracostomy. The pneumonectomy space in these patients was already contaminated, but chronic sepsis had not yet been established. In 7 patients, thoracoplasty was delayed until such time as an already established empyema was drained adequately with tube thoracostomy, while ensuring that the loss of tidal volume was tolerated by the patient. Apart from the 16 patients in whom the empyema resulted from a bronchopleural fistula, 3 patients without previous lung resection had development of a fistula. Thoracoplasty was employed in these 19 patients in an effort to obliterate an empyema space in communication with the bronchus, although repair of the bronchial stump was attempted in every instance. It was thought that obliteration of the space and collapse therapy would aid the healing of the fistula, which otherwise would have been subjected to the negative intrapleural pressure. We believe that myoplasty has a role in the closure of bronchopleural fistulas in providing a viable vascular tissue support, but not as a means of space obliteration. The addition of thoracoplasty achieves that objective. yoplasty, however, should be performed in a relatively clean environment after drainage of sepsis. The presence of gross sepsis may have accounted for the multiplicity of operations (4.3 per patient) in the report by Pairolero and co-workers [19], who used myoplasty as a space-filling procedure. Our success rate for closure of bronchopleural fistulas was 68.4% overall and 86.6% with exclusion of the perioperative deaths. Our use of thoracoplasty during the course of management of primary empyema thoracis in this experience was restricted to patients in whom less radical procedures such as tube thoracostomy, fibrinoly sis, decortication, and fenestration had failed to achieve obliteration of the space. When empyema is associated with parenchymal disease requiring lung resection, space obliteration by limited thoracoplasty becomes mandatory. Resection procedures for mycetomas in patients with complicated aspergillosis and infiltration of the lung parenchyma by fibrotic tissue are fraught with problems associated with inadequate expansion of the residual lung, breakdown of the bronchial stump, and space infection [20]. Elective thoracoplasty was employed in these patients in an attempt to obliterate the space after resection, with the objective of preventing these disastrous complications. When cavernostomy was resorted to in patients with a mycetoma, the aim of'thoracoplasty was the obliteration of the residual space after adequate irrigation of the mycetoma cavity [21] with an antifungal agent. In the group of patients without lung cancer, all but 1 had a successful outcome. On these grounds, thoracoplasty can be safely recommended in the management of postpneumonic and tuberculous empyemas and complicated mycetomas. In the lung cancer group, the high perioperative mortality (21.1%) was the result of urgent intervention to close the bronchopleural fistula in an attempt to salvage the patients with an intolerable loss of tidal volume. All these patients had initial drainage of the pneumonectomy space. Closure of the fistula was necessitated by the excessive loss of tidal volume. The thoracoplasty was a concomitant procedure to achieve space obliteration. These patients would otherwise have perished. With such a policy, the crude long-term survival of patients in the lung cancer group who had thoracoplasty has not been any worse than the expected long-term survival of patients undergoing operation for lung cancer. Had we refrained from performing thoracoplasity in patients with nodal disease in whom complications developed necessitating the procedure, our long-term results would have been considerably better. Our policy for control of pleural space infection continues to be (1) drainage by tube thoracostomy, fenestration, or a Clagett procedure; (2) control of infection; and (3) if a space persists, an obliteration procedure such as myoplasty or thoracoplasty. When space complications after resection are anticipated in view of the underlying pathology, for example, persistence of the space or infection, we advocate prophylactic limited thoracoplasty. We do not advocate thoracoplasty as the first procedure of choice. With this policy, our objectives were achieved in 78.1% of patients who underwent thoracoplasty. In recent years, myoplasty has increasingly been employed by others to

7 1993;5690>9 PEPPAS ET AL 909 achieve the same objectives, but we believe that when the basic principles in performing thoracoplasty are strictly followed and the indications for undertaking it are critically appraised, a successful outcome can be expected. References 1. eade RH. A history of thoracic surgery. Springfield, IL: Charles C. Thomas, Estlander JA. Resection des c6tes dans I'empyema chronique. Rev ed Chir (Paris) 1879;3: De Cerenville EB. De l'intervention dans les maladies du poumon. Rev ed Suisse Romande 1885;5: Schede. Die Behandlung der Empyeme. Verh Cong Innere ed Wiesbaden 1890;9: riedrich PL. Die operative Beeinflussung einseitiger Lungphthise durch totale Brustwandobilisierung. Arch Klin Chir 1908; Wilms. Die Pfeilerresektion der Rippen zur Verengerung des Thorax bei Lungentuberculose. Ther Ggw 1913;54: Sauerbruch. Die Chirurgie der Brustorgane. vol 11. Berlin: Springer, 1925: Semb C. Thoracoplasty with apicolysis. Nationaltrykkeriet, Oslo, Alexander J. The collapse therapy of pulmonary tuberculosis. Springfield, IL: Charles C. Thomas, Pairolero PC, Trastek V. Surgical management of chronic empyema: the role of thoracoplasty [Editorial]. Ann Thorac Surg 1990;50: Jaretzki A 111. Role of thoracoplasty in the treatment of chronic empyema [Letter]. 1991; Tuffier T. Etat actuel de la chirurgie intrathoracique. Paris: arson & Cie, 1914:90-105, Wilson DA. Extrapleural pneumonolysis with Lucite plombage. J Thorac Surg 1948;17: orriston-davies H, Temple LJ, Stathatos C. The polythene pack operation for pulmonary tuberculosis. Thorax 1951;6: Iioka S, Sawamura K, ori T, et al. Surgical treatment of chronic empyema. A new one-stage operation. J Thorac Cardiovasc Surg 1985;90: Gregoire R, Deslauriers J, Beaulieu, Pireaux. Thoracoplasty: its forgotten role in the management of nontuberculous postpneumonectomy empyema. Can J Surg 1987;30: Hopkins RA, Ungerleider Rh4, Staub EW, Young WG Jr. The modern use of thoracoplasty. 1985;40: Horrigan TP, Snow NJ. Thoracoplasty: current application to the infected pleural space. 1990;50: Pairolero PC, Arnold PG, Piehler J, cgoon DC. Intrathoracic transposition of extrathoracic skeletal muscle. J Thorac Cardiovasc Surg 1983;86: Al-Zeerah, Jeyasingham K. Limited thoracoplasty in the management of complicated pulmonary aspergillomas. Thorax 1989;44: Daly RC, Pairolero PC, Piehler J, Trastek V, Payne WS, Bernatz PE. Pulmonary aspergilloma. J Thorac Cardiovasc Surg 1986;92:981-8.