- Email: [email protected]
Bronchoplastic procedures for tuberculous bronchial stenosis
Bronchoplastic procedures for tuberculous bronchial stenosis Thirty-six patients underwent tracheobronchoplastic procedures for treatment of tuberculous tracheobronchial stenosis. The modes of operations were left upper sleeve lobectomy in 13 patients, sleeve resection of the left main bronchus in 12 patients (two underwent concomitant left upper lobectomy), right upper sleeve lobectomy in five patients, sleeve resection of the right intermediate bronchus in two patients, right sleeve superior segmentectomy of the lower lobe in one patient, sleeve resection of the trachea with concomitant left pneumonectomy in one patient, carinal resection with right upper sleeve lobectomy and middle lobectomy in one patient, and dilatation of the left main bronchus with a free skin graft reinforced with a steel wire in one patient. One patient died of pulmonary edema of unknown cause on the first postoperative day. Anastomotic stenosis occurred in seven patients. One of these patients underwent reoperation and six underwent endoscopic dilatation. One patient died in the hospital of massive bleeding during endoscopic dilatation 4 months after operation. Slight to moderate stenosis resulted in the remaining patients. Although there are some complications, we believe bronchoplastic operation is worthwhile for restoring pulmonary function in patients with tuberculous tracheobronchial stenosis. (J THoRAe CARDIOVASC SURG 1993;106:1118-21)
Ryoichi Kato, MD, Tooru Kakizaki, MD, Nanae Hangai, MD, Makoto Sawafuji, MD, Tatsuya Yamamoto, MD, Teruhisa Kobayashi, MD, Masazumi Watanabe, MD, Mitsuo Nakayama, MD, Masafumi Kawamura, MD, Koji Kikuchi, MD, Koichi Kobayashi, MD, and Tsuneo Ishihara, MD, Tokyo, Japan
h e incidence of pulmonary tuberculosis has decreased in most developed countries, and most patients can be successfully treated with antituberculosis drug therapy. However, there are still a few patients who have bronchial stenosis from cicatricial healing of endobronchial tuberculosis.' We have recently seen some patients who underwent bronchoplastic procedure for treatment of tuberculous tracheobronchial stenosis. We reviewed the medical records of the patients who underwent bronchoplastic procedures for tuberculosis to renew our knowledge of this now relatively rare lesion. In this article, we describe the experience in these patients and discuss the indications, perioperative care, and operative mode of From the Department of Surgery, Schoolof Medicine, KeioUniversi-
ty, Tokyo, Japan. Received for publicationDec. 11, 1992. Accepted for publicationMarch 1, 1993. Address for reprints: Ryoichi Kato, MD, Department of Surgery, Schoolof Medicine, Keio University, 35 Shinanomachi,Shinjukuku, Tokyo 160, Japan. Copyright @ 1993 by Mosby-Year Book, Inc. 0022-5223/93 $1.00 +.10 12/1/46876
1 1 18
bronchoplastic procedure. The role of surgery in relation to other modalities of therapy is also discussed. Patients and methods At Keio University Hospital, the first experience in bronchoplastic procedure was a patient with tuberculous bronchial stenosis who underwent dilatation of the left main bronchus with a free skin graft in 1955. Thirty-five more tracheobronchoplastic procedures had been performed for tuberculous lesions by the end of 1991. Thirty patients were women and six were men. Age varied from 20 to 68 years and the average age was 35 years. Fifteen patients were in their 20s, 11 were in their 30s, six were in their 40s, three were in their 50s, and one patient was 68 years old. Complete medical records were available for 24 patients. Of these 24 patients, 20 had symptoms. Eleven patients (55% of those with symptoms) reported cough, eight (40%) reported excessive sputum, five (25%) reported wheezing, five (25%) reported dyspnea, two (10%) reported fever, and one (5%) had episodes of pneumonia. The remaining four patients had no symptoms but had abnormal findings in their routine check up chest radiographs. The abnormal findings were peripheral infiltrative shadow in two patients who had active pulmonary tuberculosis, obscured left main bronchus in one patient, and volume loss ofthe right middle and lower lobe in one pa tient. The tracheal or bronchial stenoses were initially diagnosed by bron-
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 6
choscopy in 21 patients (889'0), by bronchography in two patients (89'0), and by chest tomography in one patient (49'0). The patients whose airway stenosis was initially diagnosed by bronchography or tomography underwent bronchoscopy for close preoperative observation. Eleven of the 24 patients were found to be infected with tubercle bacilli by sputum culture or stain when the bronchial stenosis was diagnosed but were free of infection at operation. They had had 4 to 34 months (average 14 months) of antituberculous drug therapy and had been infection free for 3 to 32 months before operation. Eight patients had a history of pulmonary tuberculosis but had been infection free for more than 6 months by operation. Antituberculous drugs were administered to seven of these eight patients I to 21 months (average 10 months) before the operation. Two patients were infected with tubercle bacilli at operation. One of these patients was operated on under diagnosis of stenotic left main bronchus ofunknown cause; the result of preoperative sputum culture was found to be positive after the operation. The other patient had been treated for pulmonary tuberculosis with 17 months of intermittent antituberculous drugs over 5 years. The tubercle bacilliwere thought to be from coexisting cavitary lesion, which wasresected concomitantly. Three patients had neither positive resultsof examination for tubercle bacilli nor a history of tuberculosis. In these patients, the stenosis was surmised to have been caused by endobronchial tuberculosis after pathologic examination of the resected specimen, though there was no specific finding of tuberculosis. All operations were performed electively with patients under generalanesthesia and intratracheally intubated. A patient who underwent sleeve resection of the trachea and left pneumonectomy was operated on through combined median sternotomy and anterior thoracotomy. All the other patients were operated on through standard thoracotomy. The mode of operations were left upper sleeve lobectomy in 13 patients, sleeve resection of the left main bronchus in 12 patients (two underwent concomitant left upper lobectomy), right upper sleeve lobectomy in five patients, sleeve resection of the right intermediate bronchus in two patients (the main stem bronchuswas also resected in wedge shape in one patient), right sleeve superior segmentectomy of the lower lobe in one patient, sleeve resection of the trachea with concomitant left pneumonectomy in one patient, carinal resection with right upper sleeve lobectomy and middle lobectomy in one patient, and dilatation of the left main bronchus with a free skin graft reinforced with a steel wire? in one patient (Fig. I). Silk suture was used in the first patient who underwent dilatationof the left main bronchus with a skin graft. Catgut suture (2-D) was used in the next 15 patients, monofilament nylon suture O-O} was used in the next eight patients, monofilament polypropylene suture O-O} was used in the next four patients, and braided polyglycolic acid suture (3-0) was used in the last 9 patients. All types of suture materials were carried on atraumatic needles. The patients were usually followed up for several years after being discharged. Antituberculous drugs were administered for at least 3 months (usually 6 to 12 months) after operation.
Results There was one operative death. This patient died from pulmonaryedema of unknown cause on the first postop-
Kata et al.
A
B
c:::>
c:::>
13 cases
lOcases
C
0
c:::>
c:::>
2 cases
5 cases
I I I9
~ ~~ ~ ~ ~f i
J4:f 1case
Fig. 1. Modes of operation in 36 patients.
erative day. There was one case of in-hospital death, which occurred 4 months after operation. Thecauseofthe death was massive bleeding during the endoscopic resection ofthe granulation tissue at the anastomosis. No other in-hospital deaths were seen, and there were no late deaths to our knowledge. Anastomotic stenosis occurred in seven patients. The operations performed in these seven patients were sleeve resection of the left main bronchus in four patients (one underwent concomitant left upper lobectomy), right upper sleeve lobectomy in two patients, and left upper sleeve lobectomy in one patient. One patient underwent reoperation for stenosis, after which restenosis eventually developed. Five patients underwent dilatation with a dilatation bougie and resection of granulation tissues with biopsy forceps through a rigid bronchoscope. One patient underwent dilatation through flexible bronchoscope with a balloon-tipped catheter. One patient who underwent dilatation through rigid bronchoscope died of massive bleeding when the granulation tissue was resected with a
1 120
The Journal of Thoracic and Cardiovascular Surgery
Kata et al.
December 1993
Fig. 2. A, Preoperativebronchoscopy showsbilateral main bronchi in a patient who underwent sleeve resection of the left main bronchus. The left main bronchus was markedly stenotic and the bronchi distal to the stenosis could not be observed. B, Preoperativebronchogramreveals marked stenosis of the left main bronchusand normal appearance of the bronchidistal to the stenosis. C, Postoperative bronchoscopy showsbilateral main bronchi.Good patency of the left main bronchus was obtained. biopsy forceps. Autopsy revealed no bronchopulmonary arterial fistula, and the bleeding was thought to have been from the bronchial artery. Slight to moderate anastomotic stenoses resulted in other patients who underwent endoscopic dilatation. The results of the preoperative and postoperative pulmonary function tests among those whose pulmonary parenchyma was not resected were available in seven patients (six patients underwent sleeve resection of the left main bronchus [Fig. 2] and one patient underwent sleeve resection of the right intermediate bronchus with partial resection of the main bronchus). Postoperative pulmonary function tests were performed 1 to 3 months after the operation. Preoperative and postoperative vital capacity, vital capacity percentage of predicted, forced expiratory volume in 1 second, and percent of forced vital capacity expired in 1 second were 2.54 ± 0.68 Land 2.37 ± 0.72 L,85 ± 17% and 78 ± 18%,1.55 ± 0.36 Land 1.92 ±
0.60 L, 66 ± 11% and 82 ± 9%, respectively. There were significant increases in forced expiratory volume in 1 second (p < 0.05) and percent of forced vital capacity expired in 1 second (p < 0.01), but decreases in vital capacity and vital capacity percentage of predicted were not significant (paired t test).
Discussion Because tuberculous tracheobronchial stenosis IS a benign disease, although it has considerable influence on quality oflife, indications for surgical procedures must be strictly determined. The patients with severe symptoms such as dyspnea on exertion, frequent respiratory tract infections, or persistent cough but with good risk may be the best candidates for operation. The deaths in our series of 36 patients were two early cases (5.6%), including a late hospital death. We believe that risk is low, and in younger patients with good risks operation is indicated whenev-
The Journal of Thoracic and Cardiovascular Surgery
Kato et al.
Volume 106, Number 6
er feasible because the lesion may damage quality of life inthe future when pulmonary function deteriorates further. The results of the preoperative and postoperative pulmonary function tests in seven of the patients without pulmonary parenchymal resection showed significant increases in forced expiratory volume in I second and percent of forced vital capacity expired in I second without significant decrease in vital capacity and vital capacity percentage of predicted. This improvement also seems tojustify the indication of the operation, beyond the relief ofsymptoms. Inolderpatients and patients with poor risks, alternative methods such as balloon dilatation.': 4 use of expandable metallic stents,5 and vaporization by laser" should be considered, Although these techniques are less invasive than surgery, the effect may only be temporary, or a risk exists for erosion into the pulmonary vasculature. We therefore believe that surgery should be the therapy of choice in patients who can tolerate the operation. Because the disease is benign and seldom life-threatening, the operation should be elective. There should be no bronchoscopic findings of active endotracheobronchial tuberculosis at operation. We generally administer antituberculous drugs at least 6 months before operation when the bronchial stenosis is thought to be caused by tuberculosis, to ensure that there is no latent tuberculous inflammation at operation. The mode of operation should be determined according tolocation, extent, and degree of stenosis. Resection of pulmonary parenchyma should be minimal, limited to the portion with irreversible deterioration, because the purpose ofthe operation is to restore pulmonary function to the hypoventilated lung. In determining the extent of bronchial resection, several facts should be taken into consideration. Complete resection of the affected bronchus is seldom possible because theextentof the disease is usually wide;complete resection isnotalwaysnecessary because a slight remnant ofstenosis would affect ventilatory function only slightly. It should be noted that, unlike with the bronchoplastic procedures for tumorous lesions, remnant disease at the sur~cal margin is permissible in operation for tuberculous tracheobronchial stenosis. Transection of the bronchusshould bestarted close to or at the stenotic segment. Additional resection to obtain appropriate bronchial
I I 2I
lumen should be carried out after observation of the bronchial lumen from inside. Resection of the affected bronchus should be sufficient that the bronchial wall at the anastomosis not be malacic. The bronchial wall at the anastomotic site may be thickened or hard, but this seldom affects the healing of the anastomosis if there is no active inflammation present. It should be kept in mind that excessivetension on anastomosis caused by extensive resection would do more harm than would slight remaining stenosis. Postoperative bronchoscopy 2 or 3 weeks after operation is needed to confirm good healing of the bronchial anastomosis. We administer antituberculous drug for 6 to 12 months after operation to prevent the recurrence of pulmonary tuberculosis. The necessity of this therapy has not been proved, but because there are few side effects from antituberculosis drugs, we believe it is justified. In conclusion, although there were some complications, we believe that bronchoplastic procedure for tuberculous tracheobronchial stenosis is worthwhile.
Addendum Wehave recently performed leftlower sleeve lobectomy with total sleeve resection of the left main-stem bronchus in a 23-year-old woman for tuberculous bronchial stenosis, with a good result. REFERENCES 1. Ip MSM, So SY, Lam WK, Mok CK. Endobronchial
tuberculosis revisited. Chest 1986;89:727-30. 2. Gebauer PW. Reconstructive surgery of the trachea and bronchi: late results withdermal grafts. J THORAC CARDlOVASC SURG 1951 ;22:568-84. 3. Cohen MD, Weber TR, Rao Cc. Balloon dilatation of tracheal and bronchial stenosis. AJR Am J Roentgenol 1984; 142:477-8. 4. Nakamura K, Terada N, Ohi M, Matsushita T, Kato N, Nakagawa T. Tuberculous bronchial stenosis: treatment with balloon bronchoplasty. AJR Am J Roentgenol 1991; 157:1187-8. 5. Wallace MJ, Charnsangavej C, Ogawa K, et aL Tracheobronchial tree: expandable metallic stents used in experimental and clinical applications. Radiology 1986; 158:30912. 6. Dumon JF, Rebound E, Garbe L. Aucomte F, Meric B. Treatmentof tracheobronchial lesions by laser photoresection. Chest 1982;81 :278-84.
Ryoichi Kato, MD, Tooru Kakizaki, MD, Nanae Hangai, MD, Makoto Sawafuji, MD, Tatsuya Yamamoto, MD, Teruhisa Kobayashi, MD, Masazumi Watanabe, MD, Mitsuo Nakayama, MD, Masafumi Kawamura, MD, Koji Kikuchi, MD, Koichi Kobayashi, MD, and Tsuneo Ishihara, MD, Tokyo, Japan
h e incidence of pulmonary tuberculosis has decreased in most developed countries, and most patients can be successfully treated with antituberculosis drug therapy. However, there are still a few patients who have bronchial stenosis from cicatricial healing of endobronchial tuberculosis.' We have recently seen some patients who underwent bronchoplastic procedure for treatment of tuberculous tracheobronchial stenosis. We reviewed the medical records of the patients who underwent bronchoplastic procedures for tuberculosis to renew our knowledge of this now relatively rare lesion. In this article, we describe the experience in these patients and discuss the indications, perioperative care, and operative mode of From the Department of Surgery, Schoolof Medicine, KeioUniversi-
ty, Tokyo, Japan. Received for publicationDec. 11, 1992. Accepted for publicationMarch 1, 1993. Address for reprints: Ryoichi Kato, MD, Department of Surgery, Schoolof Medicine, Keio University, 35 Shinanomachi,Shinjukuku, Tokyo 160, Japan. Copyright @ 1993 by Mosby-Year Book, Inc. 0022-5223/93 $1.00 +.10 12/1/46876
1 1 18
bronchoplastic procedure. The role of surgery in relation to other modalities of therapy is also discussed. Patients and methods At Keio University Hospital, the first experience in bronchoplastic procedure was a patient with tuberculous bronchial stenosis who underwent dilatation of the left main bronchus with a free skin graft in 1955. Thirty-five more tracheobronchoplastic procedures had been performed for tuberculous lesions by the end of 1991. Thirty patients were women and six were men. Age varied from 20 to 68 years and the average age was 35 years. Fifteen patients were in their 20s, 11 were in their 30s, six were in their 40s, three were in their 50s, and one patient was 68 years old. Complete medical records were available for 24 patients. Of these 24 patients, 20 had symptoms. Eleven patients (55% of those with symptoms) reported cough, eight (40%) reported excessive sputum, five (25%) reported wheezing, five (25%) reported dyspnea, two (10%) reported fever, and one (5%) had episodes of pneumonia. The remaining four patients had no symptoms but had abnormal findings in their routine check up chest radiographs. The abnormal findings were peripheral infiltrative shadow in two patients who had active pulmonary tuberculosis, obscured left main bronchus in one patient, and volume loss ofthe right middle and lower lobe in one pa tient. The tracheal or bronchial stenoses were initially diagnosed by bron-
The Journal of Thoracic and Cardiovascular Surgery Volume 106, Number 6
choscopy in 21 patients (889'0), by bronchography in two patients (89'0), and by chest tomography in one patient (49'0). The patients whose airway stenosis was initially diagnosed by bronchography or tomography underwent bronchoscopy for close preoperative observation. Eleven of the 24 patients were found to be infected with tubercle bacilli by sputum culture or stain when the bronchial stenosis was diagnosed but were free of infection at operation. They had had 4 to 34 months (average 14 months) of antituberculous drug therapy and had been infection free for 3 to 32 months before operation. Eight patients had a history of pulmonary tuberculosis but had been infection free for more than 6 months by operation. Antituberculous drugs were administered to seven of these eight patients I to 21 months (average 10 months) before the operation. Two patients were infected with tubercle bacilli at operation. One of these patients was operated on under diagnosis of stenotic left main bronchus ofunknown cause; the result of preoperative sputum culture was found to be positive after the operation. The other patient had been treated for pulmonary tuberculosis with 17 months of intermittent antituberculous drugs over 5 years. The tubercle bacilliwere thought to be from coexisting cavitary lesion, which wasresected concomitantly. Three patients had neither positive resultsof examination for tubercle bacilli nor a history of tuberculosis. In these patients, the stenosis was surmised to have been caused by endobronchial tuberculosis after pathologic examination of the resected specimen, though there was no specific finding of tuberculosis. All operations were performed electively with patients under generalanesthesia and intratracheally intubated. A patient who underwent sleeve resection of the trachea and left pneumonectomy was operated on through combined median sternotomy and anterior thoracotomy. All the other patients were operated on through standard thoracotomy. The mode of operations were left upper sleeve lobectomy in 13 patients, sleeve resection of the left main bronchus in 12 patients (two underwent concomitant left upper lobectomy), right upper sleeve lobectomy in five patients, sleeve resection of the right intermediate bronchus in two patients (the main stem bronchuswas also resected in wedge shape in one patient), right sleeve superior segmentectomy of the lower lobe in one patient, sleeve resection of the trachea with concomitant left pneumonectomy in one patient, carinal resection with right upper sleeve lobectomy and middle lobectomy in one patient, and dilatation of the left main bronchus with a free skin graft reinforced with a steel wire? in one patient (Fig. I). Silk suture was used in the first patient who underwent dilatationof the left main bronchus with a skin graft. Catgut suture (2-D) was used in the next 15 patients, monofilament nylon suture O-O} was used in the next eight patients, monofilament polypropylene suture O-O} was used in the next four patients, and braided polyglycolic acid suture (3-0) was used in the last 9 patients. All types of suture materials were carried on atraumatic needles. The patients were usually followed up for several years after being discharged. Antituberculous drugs were administered for at least 3 months (usually 6 to 12 months) after operation.
Results There was one operative death. This patient died from pulmonaryedema of unknown cause on the first postop-
Kata et al.
A
B
c:::>
c:::>
13 cases
lOcases
C
0
c:::>
c:::>
2 cases
5 cases
I I I9
~ ~~ ~ ~ ~f i
J4:f 1case
Fig. 1. Modes of operation in 36 patients.
erative day. There was one case of in-hospital death, which occurred 4 months after operation. Thecauseofthe death was massive bleeding during the endoscopic resection ofthe granulation tissue at the anastomosis. No other in-hospital deaths were seen, and there were no late deaths to our knowledge. Anastomotic stenosis occurred in seven patients. The operations performed in these seven patients were sleeve resection of the left main bronchus in four patients (one underwent concomitant left upper lobectomy), right upper sleeve lobectomy in two patients, and left upper sleeve lobectomy in one patient. One patient underwent reoperation for stenosis, after which restenosis eventually developed. Five patients underwent dilatation with a dilatation bougie and resection of granulation tissues with biopsy forceps through a rigid bronchoscope. One patient underwent dilatation through flexible bronchoscope with a balloon-tipped catheter. One patient who underwent dilatation through rigid bronchoscope died of massive bleeding when the granulation tissue was resected with a
1 120
The Journal of Thoracic and Cardiovascular Surgery
Kata et al.
December 1993
Fig. 2. A, Preoperativebronchoscopy showsbilateral main bronchi in a patient who underwent sleeve resection of the left main bronchus. The left main bronchus was markedly stenotic and the bronchi distal to the stenosis could not be observed. B, Preoperativebronchogramreveals marked stenosis of the left main bronchusand normal appearance of the bronchidistal to the stenosis. C, Postoperative bronchoscopy showsbilateral main bronchi.Good patency of the left main bronchus was obtained. biopsy forceps. Autopsy revealed no bronchopulmonary arterial fistula, and the bleeding was thought to have been from the bronchial artery. Slight to moderate anastomotic stenoses resulted in other patients who underwent endoscopic dilatation. The results of the preoperative and postoperative pulmonary function tests among those whose pulmonary parenchyma was not resected were available in seven patients (six patients underwent sleeve resection of the left main bronchus [Fig. 2] and one patient underwent sleeve resection of the right intermediate bronchus with partial resection of the main bronchus). Postoperative pulmonary function tests were performed 1 to 3 months after the operation. Preoperative and postoperative vital capacity, vital capacity percentage of predicted, forced expiratory volume in 1 second, and percent of forced vital capacity expired in 1 second were 2.54 ± 0.68 Land 2.37 ± 0.72 L,85 ± 17% and 78 ± 18%,1.55 ± 0.36 Land 1.92 ±
0.60 L, 66 ± 11% and 82 ± 9%, respectively. There were significant increases in forced expiratory volume in 1 second (p < 0.05) and percent of forced vital capacity expired in 1 second (p < 0.01), but decreases in vital capacity and vital capacity percentage of predicted were not significant (paired t test).
Discussion Because tuberculous tracheobronchial stenosis IS a benign disease, although it has considerable influence on quality oflife, indications for surgical procedures must be strictly determined. The patients with severe symptoms such as dyspnea on exertion, frequent respiratory tract infections, or persistent cough but with good risk may be the best candidates for operation. The deaths in our series of 36 patients were two early cases (5.6%), including a late hospital death. We believe that risk is low, and in younger patients with good risks operation is indicated whenev-
The Journal of Thoracic and Cardiovascular Surgery
Kato et al.
Volume 106, Number 6
er feasible because the lesion may damage quality of life inthe future when pulmonary function deteriorates further. The results of the preoperative and postoperative pulmonary function tests in seven of the patients without pulmonary parenchymal resection showed significant increases in forced expiratory volume in I second and percent of forced vital capacity expired in I second without significant decrease in vital capacity and vital capacity percentage of predicted. This improvement also seems tojustify the indication of the operation, beyond the relief ofsymptoms. Inolderpatients and patients with poor risks, alternative methods such as balloon dilatation.': 4 use of expandable metallic stents,5 and vaporization by laser" should be considered, Although these techniques are less invasive than surgery, the effect may only be temporary, or a risk exists for erosion into the pulmonary vasculature. We therefore believe that surgery should be the therapy of choice in patients who can tolerate the operation. Because the disease is benign and seldom life-threatening, the operation should be elective. There should be no bronchoscopic findings of active endotracheobronchial tuberculosis at operation. We generally administer antituberculous drugs at least 6 months before operation when the bronchial stenosis is thought to be caused by tuberculosis, to ensure that there is no latent tuberculous inflammation at operation. The mode of operation should be determined according tolocation, extent, and degree of stenosis. Resection of pulmonary parenchyma should be minimal, limited to the portion with irreversible deterioration, because the purpose ofthe operation is to restore pulmonary function to the hypoventilated lung. In determining the extent of bronchial resection, several facts should be taken into consideration. Complete resection of the affected bronchus is seldom possible because theextentof the disease is usually wide;complete resection isnotalwaysnecessary because a slight remnant ofstenosis would affect ventilatory function only slightly. It should be noted that, unlike with the bronchoplastic procedures for tumorous lesions, remnant disease at the sur~cal margin is permissible in operation for tuberculous tracheobronchial stenosis. Transection of the bronchusshould bestarted close to or at the stenotic segment. Additional resection to obtain appropriate bronchial
I I 2I
lumen should be carried out after observation of the bronchial lumen from inside. Resection of the affected bronchus should be sufficient that the bronchial wall at the anastomosis not be malacic. The bronchial wall at the anastomotic site may be thickened or hard, but this seldom affects the healing of the anastomosis if there is no active inflammation present. It should be kept in mind that excessivetension on anastomosis caused by extensive resection would do more harm than would slight remaining stenosis. Postoperative bronchoscopy 2 or 3 weeks after operation is needed to confirm good healing of the bronchial anastomosis. We administer antituberculous drug for 6 to 12 months after operation to prevent the recurrence of pulmonary tuberculosis. The necessity of this therapy has not been proved, but because there are few side effects from antituberculosis drugs, we believe it is justified. In conclusion, although there were some complications, we believe that bronchoplastic procedure for tuberculous tracheobronchial stenosis is worthwhile.
Addendum Wehave recently performed leftlower sleeve lobectomy with total sleeve resection of the left main-stem bronchus in a 23-year-old woman for tuberculous bronchial stenosis, with a good result. REFERENCES 1. Ip MSM, So SY, Lam WK, Mok CK. Endobronchial
tuberculosis revisited. Chest 1986;89:727-30. 2. Gebauer PW. Reconstructive surgery of the trachea and bronchi: late results withdermal grafts. J THORAC CARDlOVASC SURG 1951 ;22:568-84. 3. Cohen MD, Weber TR, Rao Cc. Balloon dilatation of tracheal and bronchial stenosis. AJR Am J Roentgenol 1984; 142:477-8. 4. Nakamura K, Terada N, Ohi M, Matsushita T, Kato N, Nakagawa T. Tuberculous bronchial stenosis: treatment with balloon bronchoplasty. AJR Am J Roentgenol 1991; 157:1187-8. 5. Wallace MJ, Charnsangavej C, Ogawa K, et aL Tracheobronchial tree: expandable metallic stents used in experimental and clinical applications. Radiology 1986; 158:30912. 6. Dumon JF, Rebound E, Garbe L. Aucomte F, Meric B. Treatmentof tracheobronchial lesions by laser photoresection. Chest 1982;81 :278-84.