I geons first applied thoracoplastic procedures to the

Transcription

1 ORIGINAL ARTICLES Thoracoplasty: Current Application to the Infected Pleural Space Terrence P. Horrigan, D, and Norman J. Snow, D Divisions of Cardiothoracic Surgery, Case-Western Reserve University School of edicine and Cleveland etropolitan General -. Hospital, Cleveland, Ohio Thoracoplasty, once commonly used in the management of cavitary pulmonary disease, continues to find application in the obliteration of infected pleural spaces. This study reports a series of 13 patients receiving thoracoplasty between 1976 and ive patients had chronic apical empyema spaces without prior resection of lung tissue. Two of the empyemas were due to tuberculosis, two were due to atypical mycobacteria, and one was due to postpneumonic empyema. All patients had extensive destruction of upper lobe tissue. Eight patients had undergone prior pulmonary resection; 3 had persistent infected spaces in the early postoperative period, 3 had development of empyemas and bronchopleural fis- tulas late (5 to 19 years) after pulmonary resection, and 2 had postpneumonectomy empyema. All patients had rigid cavity walls preventing space obliteration by rib removal alone and required concomitant resection of the thickened pleura and intercostal muscle tissues. Bronchopleural fistulas were present in 11 patients and were closed with adjacent nonintercostal muscle. All patients survived and had successful obliteration of the infected spaces with acceptable physiological and cosmetic results. We conclude that thoracoplasty remains a useful procedure in the management of the infected pleural space in select patients. (Ann Thoruc Surg 1990;50:695-9) t is slightly more than 100 years since European sur- I geons first applied thoracoplastic procedures to the chronically infected pleural space [l]. Numerous individual techniques evolved, and in the 1930s and 1940s, an entire generation of thoracic surgeons became well versed in the application of thoracoplasty, predominantly for collapse therapy of cavitary pulmonary tuberculosis. Over the past several decades, the use of thoracoplasty has been progressively declining because of improvements in antimicrobial therapy and increasing safety of pulmonary resection. Despite these advances, there remains a small or editorial comment, see page 689. group of patients with chronically infected pleural spaces in whom the lung cannot be expanded because of intrinsic disease or prior resection, either partial or total. Although these patients can be managed by the more conservative procedures of rib resection and open tube drainage [2] or Eloesser flap construction [3], these methods have obvious disadvantages in the otherwise healthy patient. This report details our experience with the application of thoracoplasty to this group of patients. aterial and ethods Patient Population Between 1976 and 1989, 13 patients underwent thoracoplasty procedures at Cleveland etropolitan General Presented at the Thirty-sixth Annual eeting of the Southern Thoracic Surgical Association, Scottsdale, AZ, Nov 9-1 1, Address reprint requests to Dr Horrigan, Department of Surgery, St. Paul-Ramsey edical Center, 640 N Jackson St, St. Paul, N Hospital. This group represents approximately one fifth of all patients with pleural space infection complicated by unexpandable or absent lung tissue. Patients who did not receive thoracoplasty were either medically unsuitable (advanced age, uncontrolled cancer, debility) or refused the procedure. These patients were managed by open tube drainage or Eloesser flap construction. The 13 patients having thoracoplasty are grouped according to whether or not they had undergone prior pulmonary resection. APICAL EPYEAS WITHOUT PRIOR RESECTION. This group of 5 patients was seen with extensive destruction of pulmonary tissue from the disease process, often with destruction of the entire upper lobe (Table 1; ig 1). The size of the apical space varied considerably, depending on the degree to which the upper lobe had contracted before the development of the subsequent empyema. All patients in this group had bronchopleural fistulas, and the required thoracoplasty size ranged between four and seven ribs. All these patients had chronic fever and weight loss, and 3 had recurrent clinically significant hemoptysis. POSTRESECTION SPACES. There were 8 patients in this group (Table 2). Six had undergone prior lobectomy, bilobectomy, or resection of giant apical bullae. One patient who had had bilobectomy also had undergone resection of the superior segment of the lower lobe (for coccidioidomycosis). Three patients had persistent fever in the early postoperative period (<3 months), and their spaces failed to resolve with intercostal tube drainage. Apical spaces developed late (5 to 19 years) after partial by The Society of Thoracic Surgeons /90/$3.50

2 696 HORRIGAN AND SNOW 1990;50:695-9 Table 1. Apical Empyema Without Prior Resection Patient Age No. Sex (Y) Organism Prior Procedure Ribs Resected 1 54 tuberculosis + mixed Rib resection PYogens open drainage 2 63 ixed pyogens None tuberculosis + mixed Rib resection PYogens open drainage 4 64 avium intracellulare None avium intracellulare None A niger - A = Aspergillus; = ycobacteriurn lung resection in 3 patients; the residual lung had failed to expand completely in the early postoperative period, leaving a fluid cap in the apex of the pleural cavity which later became infected, with the development of bronchopleural fistulas in each (ig 2). Two patients with postpneumonectomy empyema underwent thoracoplasty: 1 patient had a bronchopleural fistula, and the other had a failed topical antibiotic sterilization of the infected pleural space, or Clagett procedure [4]. Operative Technique All procedures were performed in one stage. Subperiosteal rib resection was performed first, exposing the rigid, thickened underlying pleural membrane. All empyema cavities were found to be so rigid that rib resection alone failed to provide adequate collapse of the cavities. The empyema cavity was entered, and the intercostal muscles, neurovascular bundles, and thickened empyema cavity wall were resected, with liberal use of electrocautery. It is important to resect enough of the cavity wall so that a residual space is not permitted by a shelving edge or overhanging lip. The serratus posterior superior and rhomboid muscles were preserved and used for fistula closure when indicated. The first rib was resected in all patients, as were the transverse processes of the appropriate vertebrae. Once the empyema cavity wall had been resected, the exposed lung tissue or mediastinum was inspected for bronchopleural fistulas; they were present in 11 patients and were small and multiple in 9. The fistulas were occluded with small slips of viable muscle sutured into the openings with fine silk. Adjacent trapezius, latissimus, serratus, or rhomboid muscle was sutured to the raw apical lung tissue over the fistula repair to reinforce the fistula closure. The shoulder was mobilized to evaluate whether or not the scapula tended to lock on the uppermost residual rib; if it did, the lower third of the scapula was also resected. The extent of the thoracoplasties were as follows: four ribs in 4 patients, five ribs in 2 patients, 6 ribs in 4 patients, and seven ribs in 3 patients. Results All patients survived operation and were discharged from the hospital. Two patients required revision for inade- quate space obliteration based on postoperative chest roentgenograms. In 1 patient, the first rib had been intentionally left, and its removal provided satisfactory collapse of the cavity. The other patient was reoperated on to remove an additional lower rib. A third patient with extensive atypical tuberculous infection of the residual lung required tracheostomy and prolonged ventilatory support but was ultimately discharged. The thoracic drains were removed from 6 patients during the hospital admission; in the remainder, small air leaks persisted at the time of hospital discharge, and the large drain was exchanged for a small-caliber rubber tube after discharge; this was removed 2 to 6 months later. In the late postoperative period, all patients had permanent, satisfactory control of the infectious process with no late recurrences. One patient died 3 years postoperatively of a myocardial infarction, and 1 patient died at 2 years of carcinoma of the contralateral lung. One patient underwent successful myocardial revascularization 7 years postoperatively. Comment Thoracoplasty gained widespread acceptance for the management of cavitary pulmonary disease and chronic pleural space infections in an era noted for its poor antimicrobial therapy and lack of sophisticated ventilatory support. In that period, properly performed thoracoplasty gave results superior to other forms of collapse therapy [l]. The disadvantages of this procedure have been reported to include progressive scoliosis, chronic postoperative pain and hypesthesia of the chest wall, progressive pulmonary insufficiency, and a "mutilating" cosmetic appearance [5, 61. Although the number of patients presented here is small, we have been favorably impressed with the infrequency of these side effects. Lindskog [7] intimated that leaving the first rib in place may minimize scoliosis, and in a recent series, Hopkins and associates [8] at Duke left the first rib in 20% of their patients. Gregoire and co-workers [9], reporting a series of thoracoplasties for postpneumonic empyema, were able to achieve adequate collapse by leaving the first rib in all patients. Loynes [lo], however, reported that the degree of scoliosis was related to the total number of ribs resected as well as the removal of the transverse pro-

3 HORRIGAN AND SNOW 1990;50:6959 cesses, rather than to the presence or absence of the first rib. We have been disappointed in the adequacy of collapse when the first rib has been left and advocate its 697 Table 2. Empyema After Pulmonaw Resection, No, Patient Sex Age (v) Prior Resection Early Space (<3 onths) 6 43 RUL, RL, SS (coccidioidomycosis) 20 RUL (trauma) RUL (cancer) Late Space (5 to 19 Years) 9 44 Bullectomy 64 RUL (tuberculosis) RUL (cancer) 11 Postpneumonectomy Empyema Pneumonectomy (cancer) Pneumonectomy (lung abscess) Ribs Removed All of these patients had rib resection and open drainage before thoracoplasty. SS = RL = right middle lobectomy; RUL = right upper lobectomy; superior segment, lower lobe. A B ig 1. (Patient 5. ) ( A ) Preoperative chest roentgenogram. The right upper lobe has been destroyed by atypical mycobacterial infection, with resultant empyema and bronchopleural fistulas. ( B ) Chest roentgenogram showing resolution of infected space 8 months after four-rib thoracoplasty. removal. One of our patients required reoperation to resect the first rib to achieve adequate cavity obliteration. All of our patients have scoliosis to some degree. In none, however, has it been severe, and none have had symptoms referable to the curvature. The cosmetic result has not been unacceptable (ig 3). With respect to chronic chest wall pain, only 1 of our patients (patient 10) had long-term incapacitating pain. He was managed by neurosurgical referral and dorsal rhizotomy and achieved acceptable pain relief. The most troublesome complaint experienced by our patients has been restriction of shoulder motion on the involved side. The scapula becomes adherent to the chest wall, especially if a portion of that bone has been resected, and the appearance is that of a frozen shoulder. Abduction and elevation are accomplished by tilting the torso away from the involved side, and this disability has been reasonably well accepted. We did not see symptoms of progressive pulmonary failure in our patients, as described by Gaensler and Strieder [ 111and by Huang and Lyons [ 121. These observers noted this finding only in postpneumonectomy patients, and our series has only 2 patients in this category, probably too small a number to be meaningful. ore than half of our patients were discharged with the thoracic drains still in place because of persistent small air leaks. We exchanged the large-bore tubes placed at operation for smaller rubber catheters after several weeks. In the first years of this series, these bronchocutaneous were by periodic sinogramsj and the tubes were removed Only when there was no evidence of a bronchial communication. ore recently, believing that a new space would not develop in the midst of a firmly healed wound, we have simply waited several months

4 698 HORRIGAN AND SNOW A 1990;50:695-9 of skeletal muscle is highly effective in reducing postoperative air leaks; this also contributed to earlier tube removal in the patients operated on more recently. As all of our patients demonstrated empyema cavities with thick, unyielding walls, we thought it necessary to resect the cavity walls to effectively obliterate the space. This technique of rib resection combined with excision of the wall of the empyema cavity is generally referred to as a Schede thoracoplasty [2,5, 61. The procedure as originally described entailed leaving the wound open, using packing to fill the space, rather than primary wound closure, as we have practiced. It also involved a much more extensive resection of ribs, both in number and extent, than we have done. By confining the rib removal below the third rib to the more posterior aspect of the chest, the severe deformity that characterizes most descriptions of the Schede thoracoplasty can be avoided. Both Wangensteen [13] and Kergin [ 5 ] attempted to improve the cosmetic result by leaving the intercostal muscle bundles in place. In our patients, we found the intercostal tissues to have an unreliable vascular supply after rib resection and think that their removal did not detract from the final appearance or function. any of the patients in our series had the infected pleural space drained before thoracoplasty by rib resection and open drainage tubes; this group has had the B ig 2. (Patient 9.) (A) Preoperative chest roentgenogram. The right lung had incompletely expanded after resection of large apical bullae; 5 years later the apical fluid became infected, and a bronchopleural fistula developed. The space was initially treated by rib resection and open drainage. (B)Chest roentgenogram made 2 year after six-rib thoracoplasty and scapular resection. until the thoracoplasty wound was securely healed and then removed the catheters. None of the patients displayed a residual space. In contrast to Gregoire and co-authors [9], we believe that meticulous closure of every identifiable pulmonary fistula by plugging it with a slip ig 3. (Patient 13.) Cosmetic result 12 years after seven-rib thoracoplasty and scapular resection for postpneumonectomy emwema.

5 1990; HORRIGAN AND SNOW 699 opportunity to make a comparison of the quality of their lives between long-term drainage and thoracoplasty and has uniformly expressed preference for thoracoplasty as a long-term solution to the infected space. We believe that there remains a useful role for the application of this time-honored procedure in the management of pleural space infection in properly selected patients. We gratefully acknowledge the contributions to us on this subject by our surgical mentors, Drs Harvey J. endelsohn, Jay L. Ankenev, and lohn. Perry, Jr. References 1. Alexander J. In: Evolution of the surgical treatment of pulmonary tuberculosis. Steele JD, ed. The surgical management of pulmonary tuberculosis. Springfield: Charles C. Thomas, Samson PC. Empyema thoracis: essentials of present-day management. 1971;11:21G Eloesser L. An operation for tuberculous empyema. Surg Gynecol Obstet 1935;60: Clagett OT, Geraci JE. A procedure for the management of postpneumonectomy empyema. J Thorac Cardiovasc Surg 1963;45: Kergin G. An operation for chronic pleural empyema. J Thorac Surg 1953;26: Gaensler EA. The surgery for pulmonary tuberculosis. Am Rev Respir Dis 1982;125(Suppl): Lindskog GE. Treatment of pulmonary tuberculosisthoracoplasty and pneumonolysis. In: Glenn WWL, Liebow AA, Lindskog GE, eds. Thoracic and cardiovascular surgery with related pathology. 3rd ed. New York:Appleton- Century-Crofts, 1975: Hopkins RA, Ungerleider R, Staub EW, Young WG Jr. The modern use of thoracoplasty. 1985;40: Gregoire R, Deslauriers J, Beaulieu, Piraux. Thoracoplasty: its forgotten role in the management of nontuberculous postpneumonectorny empyema. Can J Surg 1987;30: 34s Loynes RD. Scoliosis after thoracoplasty. J Bone Joint Surg (Br) 1972;54: Gaensler EA, Strieder JW. Progressive changes in pulmonary function after pneumonectomy: the influence of thoracoplasty, pneumothorax, oleothorax, and plastic sponge plombage on the side of the pneumonectomy. J Thorac Surg 1951;22: Huang CT, Lyons HA. Cardiorespiratory failure in patients with pneumonectomy for tuberculosis. J Thorac Cardiovasc Surg 1977;74: Wangensteen OH. The pedicled muscle flap in the closure of persistent bronchopleural fistula. J Thorac Surg 1935;5: DISCUSSION DR JOHN R. HANKINS (Baltimore, D): I enjoyed this paper very much and have just one comment. Horrigan and Snow mentioned the transfer of small slips of muscle, but their ernphasis seemed to be more on the removal of ribs and much less on the use of chest wall muscles to obliterate the infected pleural space. In the late 1970s, our group at the University of aryland presented to this Association our experience with the use of chest wall muscle flaps to close bronchopleural fistulas. We found that if we used the larger muscles, such as the pectoralis major or the latissimus dorsi, they helped fill the infected space in addition to closing the fistulas, and we did not have to remove as many ribs. Would the authors please comment on this point? DR KAAL A. ANSOUR (Atlanta, GA): I challenge John Alexander s statement. We have now another operation that can close a space. I believe no presentation on management of the infected pleural space is complete without a discussion of muscle flaps. In our institution, my colleagues and I no longer do total thoracoplasties, whether for apical empyemas, postresection spaces, or postpneurnonectomy empyemas, nor do we resect the first rib or the scapula anymore. Use of the serratus anterior or latissimus dorsi muscle in connection with a very limited thoracoplasty has obviated the need for total thoracoplasty. DR THOAS B. ERGUSON (St. Louis, O): I have two comments. irst, I think the idea of trying to save the chest wall by filling the space with tissue is a good idea. Second, the problem with muscle flaps is that they are not always available, as many of these patients have had previous procedures. I have found that omentum works very well, particularly if the patient has a persistent bronchopleural fistula that you cannot close by other means. DR SNOW: I think both Dr Hankins and Dr ansour are absolutely right in terms of trying to fill these spaces with viable tissue if it is available. I was going to make the same comment that Dr erguson did, ie, that some of the patients have had a previous thoracotomy, which usually obviates the use of the latissimus, which is a primary large-bulk muscle for these procedures. I agree that to the extent that you can fill the space and save some of the chest wall, it is certainly the best thing to do, if possible.