Asian Cardiovasc Thorac Ann 2007;15:290-296
© 2007 Asia Publishing EXchange Ltd
Surgical Results in Bronchiectasis: Analysis of 149 Patients
Thomas Stephen, MCh,
Roy Thankachen, MCh,
Andrew P Madhu, MCh,
Nithya Neelakantan, MSc1,
Vinayak Shukla, MCh,
Roy J Korula, MCh
Cardiothoracic Surgery Unit II
1 Department of Biostatistics, Christian Medical College, Vellore, India
For reprint information contact: Roy J Korula, MCh Tel: 91 416 228 2186 Fax: 91 416 223 2035 Email: roykorula{at}hotmail.com, Department of Cardiothoracic Surgery, Christan Medical College, Vellore 632 004, India.
 |
ABSTRACT
|
|---|
Bronchiectasis remains a serious problem in developing countries. We reviewed the morbidity, mortality, and functional outcome of surgical treatment for bronchiectasis in our institution. Between 1992 and 2003, 149 patients (105 males, 44 females) underwent pulmonary resection for bronchiectasis. Their mean age was 33.7 years (range, 5–66 years). The indications for surgery were failure of conservative treatment in 59 (40%) patients, recurrent hemoptysis in 53 (36%), bronchial obstruction by a tumor in 9 (6%), and destroyed lung in 28 (19%). Bilateral disease was seen in 24 (16%) patients. Surgical treatment included pneumonectomy in 55 (37%) patients, lobectomy in 55 (37%), bilobectomy in 37 (25%), and lobectomy and/or segmentectomy in 2 (1%). There was one operative death (mortality, 0.67%) and morbidity occurred in 22 (14.8%) patients. Follow-up was complete in 94 patients, for a mean of 4.8 years (range, 3 months to 12 years). After surgery, 51 (34%) patients were asymptomatic. Surgical treatment for bronchiectasis can achieve good results with acceptable morbidity and mortality, not only in localized disease but also in extensive disease, if complete resection can be achieved.
|
INTRODUCTION
|
|---|
Bronchiectasis is permanent dilation of one or more bronchi.1 The prevalence of bronchiectasis has declined in developed countries with improved healthcare and the availability of suitable antibiotics. It continues to be an important problem in developing countries because of the increased prevalence of tuberculosis, pneumonia, and childhood infections such as rubella and pertussis.2 Patients with early disease can be treated successfully by conservative measures. In those with disease involving multiple lobes of the lung or poor compliance with long-term medical treatment, resection has a definitive role in curing as well as improving the quality of life. We retrospectively reviewed our 12-year experience of surgical resection for bronchiectasis.
|
PATIENTS AND METHODS
|
|---|
Between 1992 and 2003, 149 patients with bronchiectasis were operated on in the Department of Thoracic Surgery at Christian Medical College, Vellore. All patients were evaluated by taking a detailed history and by a physical examination. Preoperative assessment included chest radiography and pulmonary function tests. Bronchography was carried out in the initial 41 patients before 1995, and high-resolution computed tomography in the subsequent 108 patients after 1995. Pulmonary function tests included measurements of ventilatory capacity, static lung volume, and response to bronchodilators. Gas exchange studies were not routinely undertaken. Fiberoptic bronchoscopy was performed in all patients to rule out any intraluminal pathology. The parameters evaluated were patient demographics, symptoms, indications for surgery, choice of surgical procedure, morbidity, mortality, and functional outcome.
Most patients were referred by a chest physician after long-term medical treatment. Failure of conservative treatment was the reason for referral to surgery in 59 (40%) patients. This included those who were disabled by persistent sputum production and recurrent infective exacerbations despite all reasonable medical measures having been taken over a period of at least 12 months. Other indications were recurrent hemoptysis in 53 (36%) patients, destroyed lung in 28 (19%), and bronchial obstruction by a tumor in 9 (6%). Most patients underwent intensive chest physiotherapy and a course of antibiotics and bronchodilators to ensure that the sputum volume was less than 50 mL per day prior to surgery. All except those with recurrent hemoptysis had postural drainage. Three patients (2%) underwent emergency surgery for massive hemoptysis. Patients with bilateral disease were accepted for surgery only if the less-involved lung had minimal disease and could support ventilatory function postoperatively; a minimum of 8 normally functioning bronchopulmonary segments were considered necessary. Simultaneous bilateral resections were not practiced. Patients found to have residual disease in the remaining lung after resection were managed conservatively. Re-operation was undertaken for symptomatic residual disease following a lobectomy or bilobectomy.
In adults, a double-lumen endotracheal tube was used to provide isolated ventilation to each lung. A posterolateral thoracotomy was performed in all patients. Adhesions were released by meticulous dissection with electrocautery. All resections were carried out using standard techniques. Hilar vascular structures were isolated individually and ligated. Large vessel ligatures were reinforced with 4/0 Prolene sutures. Bronchial closure was achieved by figure-of-eight suturing with 2/0 nonabsorbable Ethibond suture. Stump coverage with autogenous tissue was not routinely undertaken. The majority of patients were extubated in the operating room. Those with extensive dissection, significant blood loss, or lungs not fully expanded were ventilated until the next morning.
All patients were monitored in the thoracic intensive care unit for at least 48 hours. A single 32F tube was placed for pleural drainage after pneumonectomy, and securely clamped until the following morning. Careful watch was kept for signs of mediastinal shift, a drop in hematocrit, or worsening in the hemodynamic status, which would indicate excessive bleeding requiring release of the chest tube clamp. Otherwise, the clamp was released the following morning for a few minutes, and the tube removed if there was no excessive drainage. In lobectomy and bilobectomy cases, two 32F intercostal drainage tubes were placed; one towards the base and the other towards the apex. They were connected to an underwater seal, and continuous low suction (–10 to –15 cm H2O) was applied. Depending on the drainage and expansion of the lung, the tubes were removed successively. Patients with prolonged air leak (> 1 week) were managed by delayed chest tube removal, which was effective in most cases. Postoperative empyema or residual space problems following lobectomy or bilobectomy were also managed by delayed chest tube removal. After 3 weeks of continuous low suction, the chest tubes were opened to the atmosphere to make sure the lung was stuck to the chest wall. On radiographic confirmation of this, the chest tubes were cut short close to the chest wall, secured with large safety pins to prevent slippage into the chest cavity, and gradually removed in the outpatient department over 2 to 3 weeks. Postoperative empyema with collapsed lung, even after 3 weeks of intercostal drainage and continuous suction, was managed by open thoracostomy and an Eloesser flap. Post-pneumonectomy empyema, with or without bronchopleural fistula (confirmed by bronchoscopy), was initially treated by intercostal drainage and appropriate antibiotics. After 10 days, when the mediastinum had stabilized, open thoracostomy with an Eloesser flap was carried out. Muscle-flap closure of the space and the associated fistula was performed at a later date once the infection had settled. Thoracoplasty procedures were not undertaken in any of these patients. Postoperative management also included intensive chest physiotherapy and administration of antibiotics.
Complete resection was defined as anatomic resection of all affected segments that had been determined preoperatively by either computed tomography or bronchography. Incomplete resection was accepted if there was bilateral disease. All resected specimens were subjected to histopathologic examination to confirm the diagnosis. Operative mortality was defined as death within 30 days after thoracotomy or during the same hospitalization. Follow-up information was obtained by history taking and physical examination during periodic outpatient clinic visits. The functional outcome of surgery was assessed by asking the patients to describe their current status with respect to their preoperative status. It was a purely subjective evaluation. All of the patients were able to come up with a definite answer. This was classified as: excellent, complete absence of the preoperative symptoms that led to surgery; improved, a marked reduction in preoperative symptoms; or unchanged, no reduction in preoperative symptoms. No criteria for objective evaluation were used for functional outcome; however, patients who continued to have hemoptysis after surgery (any amount) were categorized in the unchanged group. Symptomatic patients were evaluated by chest radiography, and if symptoms were severe, by high-resolution computed tomography.
Data for continuous variables were analyzed by Student’s t test and expressed as mean ± standard deviation. Significant differences between proportions were determined by chi-squared analysis (Pearson’s or Fisher’s exact test). The patient survival curve was generated by the Kaplan-Meier method. All analyses were carried out using Statistical Package for Social Sciences version 11.5 (SPSS, Inc., Chicago, IL, USA). Probability values < 0.05 were considered significant.
|
RESULTS
|
|---|
The sex distribution was found to be 70% (105/149) male and 30% (44/149) female. Ages ranged from 5 to 66 years (mean, 33.7 years). The majority of patients were in the 30–40-year age group (Figure 1
). Symptoms included cough with purulent sputum alone in 35 (24%) patients, hemoptysis alone in 32 (21%), and both in 82 (55%). The duration of symptoms was 6 months to 25 years (mean, 5.61 years). Preoperative medical treatment included antibiotics and bronchodilators in all patients, postural drainage in 72 (48%), and antituberculosis treatment in 97 (65%). The majority of patients (52%) had undergone medical treatment for 2 years or more prior to surgery.
Radiography showed 28 (19%) patients had features suggestive of destroyed lung. In 9 cases, the bronchiectasis was due to a bronchial tumor: 8 adenomas (6 on the right side, 2 on the left) and 1 pseudoinflammatory tumor of the right main bronchus. Bronchiectasis was more common on the left side (96/149; 64%), with the left lower lobe alone involved in 32 (21%) patients; bilateral disease was seen in 24 (16%). Sputum culture and sensitivity showed no bacterial growth in 68 (46%) patients, and mixed growth of mainly gram-negative organisms in the others: Escherichia coli and Klebsiella in 35 (23%), Haemophilus influenzae in 24 (16%), Pseudomonas in 13 (9%), and Streptococcus pneumoniae in 8 (5%). Sputum acid-fast bacillus was positive in 6 (4%) patients preoperatively; all were given antituberculous treatment and underwent surgery once the sputum was negative. Pulmonary function tests were carried out in all patients: 75 (50%) had a restrictive defect, 70 (47%) had an obstructive defect, and 4 had normal pulmonary function.
The surgical approach was through a posterolateral thoracotomy in all 149 patients. Operative details are given in Table 1. Four patients had history of previous treatment for lung abscess, and 2 had undergone previous surgery; both of these patients had a lobectomy for bronchiectasis, and one of them had active empyema at the time of the second surgery. Completion pneumonectomy was performed in both cases. Three patients underwent emergency surgery for massive hemoptysis: a right pneumonectomy in 1, and a right upper lobectomy in 2. Complete resection was possible in 84% (125/149). The other 24 (16%) patients (including the 2 previously operated on) had bilateral disease and complete resection was not attempted. Eighteen (12%) patients had biopsy-proven tuberculosis. Table 2 shows the type of resection according to symptoms.
Operative morbidity was 14.8% (Table 3). Operative mortality (0.67%) was encountered in one patient who had extensive disease of the right lung with dense adhesions. He suffered significant blood loss intraoperatively and 5 units of blood were transfused. In the immediate postoperative period, he developed disseminated intravascular coagulation and died. A repeat thoracotomy was carried out in 4 patients; for postoperative bleeding in 2 (from bronchial vessels at the bronchial stump in 1, and for collapsed residual left upper lobe with blood clots in 1) and for early bronchopleural fistula in 2. Both patients with bronchopleural fistula had associated infection; thoracostomy with an Eloesser flap was performed in the early postoperative period. Bronchopleural fistula closure with a pectoralis major muscle flap and window closure was carried out in both patients 1 year later. Empyema developed in 12 patients, which was treated by intercostal drainage and prolonged antibiotic administration; 7 of these, including the patient who had a repeat thoracotomy and clot removal, needed an Eloesser flap for persistent infection in the residual space. One patient with pleural effusion was treated by intercostal aspiration. All minor air leaks subsided on prolonged controlled suction through the intercostal drainage tubes. Factors affecting postoperative empyema were evaluated by univariate analysis (Table 4); no significant risk factor was found.
The follow-up was completed by December 2003. Complete follow-up was achieved in 94 patients over a mean period of 4.8 years (range, 3 months to 12 years). The other patients could not be followed up in our outpatient clinic because they lived too far away. The outcomes in the patients followed up are listed in Table 5. There was one late death at 2 months postoperatively. This patient, who had a repeat thoracotomy and clot removal followed by an Eloesser flap for residual space infection, developed persistent fever with an air fluid level in the left upper zone 3 weeks after discharge from the hospital. Under ultrasound guidance, a Malecot catheter was introduced into this cavity, which drained pus. He was discharged with underwater seal drainage, and was advised gradual shortening of the Malecot catheter. About 2 weeks later, he was readmitted with torrential bleeding from the Malecot catheter, which was found to have eroded into the subclavian artery. Although hemostasis could be obtained in the operating room, he had another episode of bleeding 2 days later in the ward, and he could not be resuscitated. Residual bronchiectasis was seen in 34% (32/94) of the followed-up patients (Table 5); 13 of them had bilateral disease preoperatively and the other 19 had either recurrence of minimal disease that was missed preoperatively or newly developed bronchiectasis. Other causes of persistent postoperative symptoms included chronic bronchitis or chronic obstructive pulmonary disease, tumor recurrence, and arteriovenous malformation. Table 6 gives details of the symptomatic patients followed up, and Table 7 shows the relationship of postoperative symptoms to the type of resection. The cumulative symptom-free survival is shown in Figure 2.
|
DISCUSSION
|
|---|
Bronchiectasis, a chronic necrotizing infection, is the endpoint of various infections of the lung and bronchi.1 In developing countries like India, the majority of cases present in the late stages of the disease, where a definite etiology is difficult to establish. Similar to other series, most of our patients had recurrent infections in their past, and this emphasizes the importance of adequate treatment of pulmonary infections in childhood.2–4 As it is an irreversible process, the only real cure results from resection. Many of our patients had prolonged disease with extensive involvement of one entire lung. The characteristic features of bronchiectasis; bronchial dilatation with thickening and lack of tapering of the bronchus, were found in these patients.5 Fibrosis with destruction of lung parenchyma (destroyed lung) was seen in 19% of patients, which was most often secondary to tuberculosis.6 Pulmonary function tests showed a substantial number of patients had a restrictive defect rather than the obstructive defect that normally predominates in bronchiectasis.7 This highlights the fact that functional impairment in bronchiectasis is related to the extent of lung damage, as determined by the number of bronchopulmonary segments involved.1 Although not statistically significant, many of these restrictive defects were associated with extensive disease or destroyed lung. Sputum culture and sensitivity showed that most patients had no growth or mixed growth, with gram-negative organisms predominating. This may be because they were treated with multiple courses of antibiotics over a prolonged period.
The initial treatment for symptomatic bronchiectasis is primarily medical, with the goals of reducing airway obstruction and eliminating bacteria from the lower respiratory tract.3 Most patients had prolonged medical treatment as they came from the lower socioeconomic classes and were unwilling to contemplate early surgery. After every acute infection, the surrounding normal pulmonary tissue is affected and bronchiectatic areas get larger, leading to destruction of more lung tissue. Patients with both cough with sputum and hemoptysis had the maximum number of resections (Table 5
). Thus the severity of symptoms correlates with the extent of resection. Lack of proper health education and primary healthcare as well as social deprivation prevent many cases from coming to light early.7 Many patients had extensive disease involving multiple lobes. Although earlier studies have shown that appropriate selection of patients is critical for optimal results, and early pulmonary resection while the disease is localized is preferred, diagnosis and treatment of early disease is still difficult in a third world country like ours.3 This explains the high number of pneumonectomies (38%) in our study.
Only one of our patients had history of active empyema and intercostal drainage. This young girl had bilateral bronchiectasis and had undergone a right lower lobectomy elsewhere. One year later, she presented with right-sided empyema and destroyed upper and middle lobes. Completion pneumonectomy was performed when the intercostal drainage had reduced to < 100 mL per day. The bronchial stump was reinforced with a mediastinal pleural flap. Her postoperative period was uneventful. The second patient who needed a completion pneumonectomy (previous right upper lobectomy) had residual bronchiectasis and recurrent hemoptysis 4 years after the initial surgery. Our experience of completion pneumonectomy and resection through an empyema is limited. Studies have shown that these procedures can be carried out with acceptable morbidity and mortality, and they have revealed many preoperative risk factors for postoperative empyema.8–10 We could not arrive at a definite conclusion, possibly due to inadequate sample size. The individual pulmonary function test results (e.g., forced vital capacity) were not entered into the risk analysis, as they were not available in most patient records.
At follow-up, 31 (33%) patients were found to have no change or worse symptoms than preoperatively, of whom 77% had residual bronchiectasis; however, 73% had residual bronchiectasis in the group with improved symptoms. Previous studies have shown various causes for persistent symptoms after resection.11 Of the 32 (34%) patients with residual bronchiectasis, 19 had either recurrence of minimal disease not detected preoperatively or newly developed bronchiectasis. Most of these symptomatic patients had undergone a lobectomy or bilobectomy. Resection of the most involved part of the lung parenchyma alone may reduce the severity of post-resection symptoms, but low-grade symptoms may persist. Repeated bronchoscopy and drainage in such patients could help by clearing the airways and obtaining sputum for microbiological studies to allow specific antibiotic therapy. This shows the necessity for critical preoperative evaluation. Chest radiographs are nonspecific and the extent and severity of the disease may be underestimated. Previous studies have shown that computed tomography of the thorax may be unreliable in cases of cylindrical and varicose patterns of bronchiectasis.12 Pitfalls in bronchography include a lack of opacification of some bronchi because of ventilation failure or mucoid impaction, failure to reach the upper division of the left upper lobe, or subsegmental lesions.12
Patients with complete resection clearly had a better prognosis, in agreement with other series.2–5,11 Survival analysis showed a median symptom-free survival of 24 months (Figure 2). This short asymptomatic period is characteristic of the population operated upon, many of whom had advanced disease and incomplete resection, with persistent infection leading to new-onset bronchiectasis. It was concluded from this experience that surgical treatment for bronchiectasis in third world countries is a different problem to that in developed countries. Treatment of patients with advanced disease and diffuse destruction of lung parenchyma is difficult, regardless of the extent of resection. Good results with acceptable morbidity and minimal mortality can be achieved if complete resection is ensured. Critical preoperative evaluation is important to minimize residual disease. Patients should be educated to report early symptoms. Close follow-up, continued treatment with specific antibiotics, and improvements in living conditions have important roles in reducing the recurrence rate of this disease.
|
REFERENCES
|
|---|
- Seaton D. Bronchiectasis. In: Seaton A, Seaton D, editors. Crofton and Douglas’s respiratory diseases. Volume I. 5th ed. New Delhi: Oxford University Press, 2000:794–808.
- Kutlay H, Cangir AK, Enon S, Sahin E, Akal M, Gungor A, et al. Surgical treatment in bronchiectasis: analysis of 166 patients. Eur J Cardiothorac Surg 2002;21:634–7.[Abstract/Free Full Text]
- Agasthian T, Deschamps C, Trastek VF, Allen MS, Pairolero PC. Surgical management of bronchiectasis. Ann Thorac Surg 1996:62:976–80.[Abstract/Free Full Text]
- Annest LS, Kratz JM, Crawford FA Jr. Current results of treatment of bronchiectasis. J Thorac Cardiovasc Surg 1982:83:546–50.[Medline]
- Fujimoto T, Hillejan L, Stamatis G. Current strategy for surgical management of bronchiectasis. Ann Thorac Surg 2001:72:1711–5.[Abstract/Free Full Text]
- Halezeroglu S, Keles M, Uysal A, Celik M, Senol C, Haciibrahimoglu G, et al. Factors affecting postoperative morbidity and mortality in destroyed lung. Ann Thorac Surg 1997;64:1635–8.[Abstract/Free Full Text]
- Ellis DA, Thornley PE, Wightman AJ, Walker M, Chalmers J, Crofton JW. Present outlook in bronchiectasis: clinical and social study and review of factors influencing prognosis. Thorax 1981;36:659–64.[Abstract]
- Fujimoto T, Zaboura G, Fechner S, Hillejan L, Schroder T, Marra A, et al. Completion pneumonectomy: current indications, complications, and results. J Thorac Cardiovasc Surg 2001;121:484–90.[Abstract/Free Full Text]
- Odell JA, Henderson BJ. Pneumonectomy through an empyema. J Thorac Cardiovasc Surg 1985;89:423–7.[Abstract]
- Deschamps C, Bernard A, Nichols FC 3rd, Allen MS, Miller DL, Trastek VF, et al. Empyema and bronchopleural fistula after pneumonectomy: factors affecting incidence. Ann Thorac Surg 2001;72:243–8.[Abstract/Free Full Text]
- Grenier P, Maurice F, Musset D, Menu Y, Nahum H. Bronchiectasis: assessment by thin-section CT. Radiology 1986;161:95–9.[Abstract]
- Sanderson JM, Kennedy MC, Johnson MF, Manley DC. Bronchiectasis: results of surgical and conservative management. A review of 393 cases. Thorax 1974;29:407–16.[Medline]
TOP
ABSTRACT
INTRODUCTION
