From 1995 to 2004, 28 consecutive patients underwent surgical treatment, including mainstem bronchoplasty with sparing of pulmonary parenchyma, for benign bronchial stenosis at the Samsung Medical Center. Data were collected by reviewing the hospital and office records. Twenty-one patients were female and 7 were male. The mean patient age was 34.4 yr, and their ages ranged from 2 to 69 yr.

Most of the patients (25 out of 28) had respiratory symptoms at the time of diagnosis such as cough, dyspnea, sputum, wheezing, hemoptysis with or without fever. In three asymptomatic cases, the diagnostic workup was started from the unresolved atelectasis on chest radiograph which was incidentally found for other reasons. All of the enrolled patients underwent fiberoptic bronchoscopy, a standard chest radiograph and thoracic CT. On the standard radiograph, lobar or segmental atelectasis (n=18), pneumonia (n=2), pulmonary fibrosis, bronchiectasis, a mass shadow in the main bronchus were identified. Two patients did not show any abnormality on standard chest radiograph. Preoperative pulmonary function tests were available for 20 patients, and the tests showed that the average forced expiratory volume for 1 sec (FEV₁) was 1.76 L (64% of the normal predicted value), with a range from 0.89 L to 3.57 L (29% to 114% of the normal predicted value).

All the patients underwent main bronchus reconstruction for stenosis caused by benign diseases including tuberculosis (n=21), trauma (n=4), or benign endobonchial mass (1 endobronchial leiomyoma, 1 inflammatory pseudotumor and 1 lipoma). In all of the cases of traumatic main bronchial stenosis, diagnosis was not suspected until the atelectasis was not resolved by more than 2-weeks of pulmonary toileting. Endobronchial masses were pathologically confirmed before the operation with bronchoscopic biopsy. Twenty of all tuberculous main bronchial stenosis patients had histories of anti-tuberculosis medication. None of the patients arbitrarily stopped their medication and the duration of the medication ranged from 6 months to 12 months. One of the tuberculous stenosis patient started anti-tuberculosis medication after initial bronchoscopy due to actively caseating and stenotic endobronchial tuberculosis and then surgically treated after 3 months of medication. The average duration from the completion of medication to the time of operation was 116 months, with a range from 6 months to 468 months.

For the patients showing pneumonia or atelectasis on their preoperative radiograph, aggressive toileting of sputum (including therapeutic bronchoscopy) and two weeks antibiotic coverage before operation were applied.

Nine patients who had fibrous stricture due to tuberculosis (n=8) or trauma (n=1) underwent intervention treatment before operation; balloon dilatation under the rigid bronchoscope was performed in three patients; removal of granulation tissue under rigid bronchoscopy was done in one patient and application of stent was done in five patients. All of them showed no improvement with the non-surgical intervention.

For the patients showing main bronchial stenosis we initially performed a thorough examination with flexible bronchoscopy and a CT scan. If anatomical correction was necessary, we then tried rigid bronchoscopic evaluation and we tried non-surgical interventions such as laser, balloon dilatation, or stent insertion, as well as controlling the underlying infectious disease. However, if surgical treatment was determined to be helpful to the patients or failure of non-surgical intervention was determined, we then performed bronchoplastic procedures without any unnecessary delay. When the patients were less than 40 yr, they had adequate pulmonary function tolerating general anesthesia and thoracotomy, and lung parenchymal destruction accompanied by bronchial stenosis was none or minimal, we enthusiastically recommended bronchoplastic operation.

Single or double-lumen endotracheal tubes were used for the safe conduct of anesthesia. Posterolateral thoracotomy of the involved bronchial side was usually used, however, when carinal reconstruction was intended, right posterolateral thoracotomy was used. Dissection and mobilization of the distal trachea and the bilateral main bronchus was performed in most of the cases. In every case, end-to-end anastomosis was done using 4-0 or 5-0 Vicryl (Ethicon, Somerville, NJ, U.S.A.). Interrupted sutures were always used and placed starting at both ends of the resection; they were evenly spaced and were added one by one from each side to reach the side opposite to the small part of the bronchus and were tied after all the sutures had been placed. Simple resection and end-to-end anastomosis between the distal trachea and distal main bronchus was sufficient in these cases. However, in 19 patients, concomitant procedures such as lobectomy, wedge resection of lung, or carina reconstruction were necessary. The indications and methods of the concomitant procedures are listed in Table 1. In order to release the tension around the anastomosis, mobilization of pulmonary ligament was sufficient in 14 cases. But, additional mobilization techniques were needed in 14 patients: pretracheal release for 4, hilar release for 1, and cervical flexion for 1. Especially for left side main bronchoplasty, mobilization of the aortic arch by detaching the ligament of Botalli and Marshall's fold (n=8) was needed. The anastomosis was wrapped with pedicled pericardial fat (n=1) or parietal pleura (n=3). At the end of the operation, before extubation, the bronchial tree was inspected using flexible bronchoscope to evaluate the suture line, the patency, and the alignment of the bronchial anastomosis and to aspirate secretions and blood.

Follow-up bronchoscopy after the operations was done on postoperative day seven and at approximately 2 months. Chest CT with volume rendering reconstruction of the airway was also done at postoperative 2-month. If the bronchoscopy results were good, only the chest CT was checked at 6 month, and then it was checked every year.

The institutional approval of this study was obtained and that each patient within the study gave informed consent for serving as a subject on January 2005.