Asian Cardiovasc Thorac Ann 2007;15:134-138
© 2007 Asia Publishing EXchange Ltd
Active Pulmonary Tuberculosis: Experience with Resection in 106 Cases
Rishendran Naidoo, FC Cardio (SA)
Department of Cardiothoracic Surgery, Inkosi Albert Luthuli Central Hospital, Mayville, South Africa
For reprint information contact: Rishendran Naidoo, FC Cardio (SA) Tel: 27 31 240 2114 Fax: 27 31 240 2113 Email: rishendran{at}mweb.co.za, Inkosi Albert Luthuli Central Hospital, Private Bag X03, Mayville, 4058, South Africa.
 |
ABSTRACT
|
|---|
The surgical management of pulmonary tuberculosis has evolved since collapse therapy was the mainstay of treatment. Despite this, resection for active tuberculosis is viewed with circumspection. Details of 106 patients with pathologically proven active pulmonary tuberculosis, who were operated on from January 1997 to January 2005, were reviewed retrospectively. Demographic data, radiographic profiles, indications for surgery, sputum status, and preoperative drug therapy were analyzed in relation to outcomes. The indications for surgery included multidrug-resistant tuberculosis in 27 patients, hemoptysis in 44, bronchiectasis in 27, and diagnostic dilemmas where a tumor could not be excluded in 8. All patients were operated on while receiving antituberculous therapy, and 17 were sputum positive at the time of surgery. Two (1.9%) patients died postoperatively. Morbidity was 16.9%, including 6 cases of postpneumonectomy empyema and one of bronchopleural fistula. Surgery for active tuberculosis may be undertaken with acceptable morbidity and mortality.
|
INTRODUCTION
|
|---|
The surgical management of pulmonary tuberculosis (TB) has evolved considerably since the days of collapse therapy. The current indications for surgery include: multidrug-resistant (MDR) TB, defined as resistance to isoniazid and rifampicin; hemoptysis; disease sequelae; and complications of treatment.1,2 Surgery in the face of active disease has largely been avoided in view of the higher complication rate in these patients.3–5 Lung resection is usually undertaken after a preoperative course of appropriate medical therapy has either been commenced or completed. However, where emergency resection for hemoptysis is undertaken, preoperative antituberculous therapy is not always possible. It is also possible that active disease may be discovered on postoperative pathological assessment, despite a belief of preoperative cure.
|
PATIENTS AND METHODS
|
|---|
A retrospective review was undertaken of data from the Thoracic Surgery Unit in Durban on patients treated between January 1997 and January 2005. The inclusion criteria were: anatomical pulmonary resection, with the minimum being a lobectomy; histologically active tuberculosis, defined as the presence of acid-fast bacilli and necrotizing granulomatous inflammation; surgery indicated by MDR-TB, hemoptysis, bronchiectasis with or without evidence of a destroyed lung, and diagnostic dilemmas. Sputum status was not used to define disease activity as it was not available in all patients, especially those presenting with hemoptysis. Routine preoperative evaluation included suitability for resection based on clinical grounds, pulmonary function testing, and radiographic evaluation. This entailed chest radiographs and high resolution computed tomography. Pulmonary function testing could not be performed in patients presenting with massive hemoptysis. In this instance, room air arterial blood gases were measured. HIV testing was routine, with consent; CD4 estimates were made if positive. Elective surgery was deferred in patients with CD4 counts below 400 cells·mm–3 as there was no recourse to antiretroviral medication at that stage. In the case of life-threatening hemoptysis, a low CD4 count did not preclude surgery.
Patients were grouped according to the indications for surgery. Group 1 had MDR-TB and had been treated medically, based on culture sensitivities, with at least 3 months of appropriate therapy prior to surgery.6 The indications for surgery included localized disease with failure to sputum-convert after appropriate therapy for 4–6 months and recurrent disease. Postoperative medical therapy was continued for 18 months.7 Group 2 had hemoptysis; our protocol for this condition was determined by the stability of these patients. Patients were treated with intravenous antibiotics and antituberculous therapy, and sedated with parenteral morphine. Unstable patients with active hemoptysis and resectable disease on radiographic grounds were resuscitated and taken to the operating room. Stable resectable patients were assessed radiographically, temporarily stabilized by bronchial artery embolization if active disease was suspected, and operated on electively, usually during the same admission. Non-resectable patients were embolized and managed conservatively with antibiotics and antituberculous therapy. Group 3 patients were deemed on radiographic grounds to have burnt-out disease with features of bronchiectasis. A course of antituberculous therapy was standard in this group, and surgery was indicated for recurrent chest infections. This group also included patients with radiographic evidence of a destroyed lung. Group 4 comprised patients who posed a diagnostic dilemma. Malignancy is notoriously difficult to distinguish from tuberculosis in some cases. Where radiographic assessment, sputum cytology and microbiology, bronchial lavage, and even transthoracic fine-needle aspiration were unhelpful, exploratory thoracotomy was undertaken. In the early part of this series, the frozen section facilities were geographically separate from the thoracic unit, thus lobectomy or pneumonectomy was performed with the justification that the greater evil was being treated. Currently, this is not the case, and frozen section analysis now avoids extensive pulmonary resection.
The surgical technique involved a standard posterolateral thoracotomy with bronchial isolation achieved by means of double-lumen endotracheal tubes in adults and bronchus blockers in children. The prone position was also used in selected pediatric resections. Pre- and postoperative bronchoscopy was performed. The bronchial stump was sutured with interrupted absorbable Vicryl or nonabsorbable Ethibond suture, depending on surgeon preference. Muscle flap reinforcement of the bronchial stump was not routine, and stapling devices were not used in view of the cost.
Early ambulation was encouraged. Follow-up is difficult in our environment due to the poor socioeconomic circumstances of most of our patients, although we do have a self-referral system in place. Postoperative antituberculous therapy was continued for a minimum of 6 months in patients with active disease, and the MDR group received 18 months of therapy.
Statistical analyses were undertaken with SPSS software (SPSS Inc., Chicago, IL, USA) using Fischer’s exact test. A p value less than 0.05 was considered to be statistically significant.
|
RESULTS
|
|---|
There were 106 patients operated on over the 8-year period. Their mean age was 33.6 years (range, 2–69 years), and 66 (62%) were male. Preoperative antituberculous therapy had been administered for a mean duration of 82 days (range, 1–515 days) at the time of resection. Sputum studies were positive in 17 patients prior to surgery.
Surgery was undertaken in the MDR group for localized cavitary disease in 12 patients, persistent sputum positivity in 11, and recurrent disease in 4. The most common resection performed in this group was left pneumonectomy (Table 1
). The overall morbidity in this group was 25.9%, with post-pneumonectomy empyema occurring in 2 sputum-positive patients (Table 2). Two patients remained sputum positive following surgery, resulting in a cure rate of 92.6%.
Hemoptysis was the most frequent indication (44/106, 42%) for surgical intervention. Bronchial artery embolization had been undertaken in 7.7% of group 2 patients prior to surgical intervention; they were all stable. The decision to offer stable patients bronchial artery embolization was based on history of a recent bout of massive hemoptysis. It is felt that intraoperative bleeding is reduced if the collaterals are embolized. In 4 patients presenting as emergencies, surgery was carried out without considering embolization, to avoid delay. One patient in group 2 developed atelectasis of the residual lung after lobectomy, with no infective component, which required intensive physiotherapy. There were 4 sputum-positive patients in this group; one developed postoperative pneumonia not requiring ventilation. The overall morbidity in this group was 11.4%.
Three of the 27 patients operated on for bronchiectasis developed post-resection empyema requiring prolonged drainage and sterilization of the space. The morbidity in group 3 was 18.5%. Three patients were sputum positive prior to surgery. They had received a course of anti-TB treatment and did not have drug-resistant disease on cultures. It was unfortunate that the patient who underwent pneumonectomy during an exploratory thoracotomy developed post-pneumonectomy empyema. Endobronchial tuberculosis was not demonstrated in the specimen. There was no associated bronchopleural fistula (BPF), and drainage was all that was required.
The most frequent resection was left pneumonectomy (31/106, 29%), in agreement with the report of Ashour and colleagues8 supporting the left bronchus syndrome hypothesis. The overall morbidity was 16.9%, and the major morbidity rate was 8.5% (Table 2
). There was no statistical correlation between the indications for surgery and the complications ( p = 0.719). Sputum status, when available, did not contribute to morbidity ( p = 0.118).
There were 2 deaths (mortality, 1.9%) during the study period, both occurring in patients with hemoptysis. One patient died suddenly on the 3rd postoperative day, with no obvious cause being found; a postmortem was not performed. The second patient was retroviral positive and was found dead on day 7 from a suspected pulmonary embolus.
Fifteen patients (14.4%) were HIV positive. CD4 counts were available in 12 patients, ranging from 350 to 874 cells·mm–3 (mean, 555 cells·mm–3). Three patients in this cohort developed complications: 1 post-pneumonectomy empyema, 1 BPF, and 1 minor wound sepsis, giving morbidity rates of 26.6% in the HIV-positive group and 17.6% in the HIV-negative group. This was not statistically significant ( p = 0.302). It was not possible to analyze the CD4 counts as they were not available for all patients.
Follow-up was complete in the MDR group over 18 months, at the MDR-TB clinic run by physicians. The overall follow-up period ranged from 1–18 months (mean, 2.79 months).
|
DISCUSSION
|
|---|
The indications for surgery in tuberculosis are well defined.1,2 Traditionally an adjunct to medical therapy, surgery is becoming popular as a means of biological and anatomical eradication in both the active form and in disease sequelae.1,9 Surgery in the face of active disease and in patients who have had no preoperative therapy is avoided in view of the higher complication rates.3
The most important complication is BPF with or without resultant space infections. Risk factors identified by both univariate and multivariate analysis include sputum positivity and a low preoperative forced expiratory volume in 1 second.4 In this series, sputum positivity was not a risk factor; this was possibly because of the limited availability of sputum results in our patients. Pomerantz6 noted a BPF rate of 30% in sputum-positive patients with MDR-TB, and advocated the routine use of muscle flap reinforcement of the bronchial stump. The BPF rate in our group with MDR-TB was only 3%. The overall morbidity of 25.9% is on the high side; however, the major morbidity in this group was only 18.5%, in agreement with the literature.6,10–14 Intercostal muscle flap reinforcement of the bronchial stump was used in 2 patients, neither of whom developed complications, but we had only 11 patients who were sputum positive. The role of routine muscle flap reinforcement in patients with MDR-TB is debatable, with reports demonstrating reasonable results without the use of a muscle flap.10–13 Further study is obviously indicated. Conlan and colleagues5 demonstrated a 5.6% BPF rate and a 15.3% incidence of post-pneumonectomy empyema. This was in the presence of benign disease with no active tuberculosis. Morbidity and mortality rates of 17.8% have also been noted, with BPF rates of 5.1% for disease sequelae.15 We had a fairly low BPF rate of 1% overall. This was despite the absence of muscle flap reinforcement of the bronchial stump and using interrupted absorbable sutures, which Deschamps and colleagues16 described as a risk factor for empyema and fistula formation. This report is limited by the inclusion of patients with malignant disease.
Post-pneumonectomy empyema is also a feared complication. Sputum positivity, low preoperative forced expiratory volume in 1 second, old age, right pneumonectomy, preoperative empyema, and the opening of a cavity intraoperatively are all risk factors for this complication.4 Post-pneumonectomy empyema occurs in 4.2%–32% of patients.3–6,10,17,18 The highest rate of this complication was 45.7%, as noted by Odell and Henderson18 reporting on their experience of pneumonectomy through an empyema space. Our rate of post-pneumonectomy empyema was 5.8%, despite the disease activity.
The number of pneumonectomies for bronchiectasis may seem high, but considering the pathology that we were faced with, it was necessary.19 The patients with destroyed lung were included in this group, and the resections were dictated by the extent of the disease. The impact of HIV requires further study; we now have access to antiretroviral medication and this will result in more elective surgery being undertaken. I hope to further evaluate the role of surgery in patients with HIV and TB. The fact that there was no correlation between morbidity and HIV status may possibly be explained by the fact that all of the patients we operated on were clinically well with CD4 counts over 200 cells·mm–3; the recognized level for AIDS.
It is evident that the morbidity and mortality in this review is in keeping with that of other reports. The presence of active disease did not appear to contribute significantly to the outcome, leading one to question the significance of ‘pathologically active disease’. The reasons for the equivalent morbidity are difficult to determine as we follow similar principles to those mentioned in the literature, with the exception of staple closure and muscle reinforcement of the bronchial stump. A factor to consider, however, is that there were only 17 sputum-positive patients preoperatively, despite the pathological evidence of active disease.
The experience with MDR-TB has helped us realize that surgery may be undertaken in patients with active disease, with good results.6,7,10–14 The implications are that surgery may be undertaken earlier, prior to the development of bilateral disease and removing foci that may later house MDR disease. Whilst not advocating the routine role of surgery in active disease, this is a viable option if required. This would primarily be in the emergency setting where preoperative therapy is not possible. The results in this series demonstrate that surgery may be undertaken with acceptable morbidity and mortality.
|
ACKNOWLEDGMENTS
|
|---|
Thanks to Dr. J Odell for his constructive comments.
|
REFERENCES
|
|---|
- Treasure RL, Seaworth BJ. Current role of surgery in Mycobacterium tuberculosis. Ann Thorac Surg 1995;59:1405–9.[Abstract/Free Full Text]
- Freixinet J. Surgical indications for treatment of pulmonary tuberculosis. World J Surg 1997;21:475–9.[Medline]
- Massard G, Dabbagh A, Wihlm JM, Kessler R, Barsotti P, Roeslin N, et al. Pneumonectomy for chronic infection is a high-risk procedure. Ann Thorac Surg 1996;62:1033–8.[Abstract/Free Full Text]
- Kim YT, Kim HK, Sung SW, Kim JH. Long-term outcomes and risk factor analysis after pneumonectomy for active and sequela forms of pulmonary tuberculosis. Eur J Cardiothorac Surg 2003;23:833–9.[Abstract/Free Full Text]
- Conlan AA, Lukanich JM, Shutz J, Hurwitz SS. Elective pneumonectomy for benign lung disease: modern-day mortality and morbidity. J Thorac Cardiovasc Surg 1995;110(4 Pt 1):1118–24.[Abstract/Free Full Text]
- Pomerantz M. Surgery for the management of Mycobacterium tuberculosis and nontuberculous mycobacterial infections of the lung. In: Shields TW, Locicero J III, Ponn RB, editors. General thoracic surgery. 5th ed. Philadelphia: Lippincott, Williams & Wilkins, 2000:1066–75.
- Iseman MD, Madsen L, Goble M, Pomerantz M. Surgical intervention in the treatment of pulmonary disease caused by drug-resistant Mycobacterium tuberculosis. Am Rev Respir Dis 1990;141:623–5.[Medline]
- Ashour M, Pandya L, Mezraqji A, Qutashat W, Desouki M, al-Sharif N, et al. Unilateral post-tuberculous lung destruction: the left bronchus syndrome. Thorax 1990;45:210–2.[Abstract/Free Full Text]
- Souilamas R, Riquet M, Barthes FP, Chehab A, Capuani A, Faure E. Surgical treatment of active and sequelar forms of pulmonary tuberculosis. Ann Thorac Surg 2001;71:443–7.[Abstract/Free Full Text]
- van Leuven M, De Groot M, Shean KP, von Oppell UO, Willcox PA. Pulmonary resection as an adjunct in the treatment of multiple drug-resistant tuberculosis. Ann Thorac Surg 1997;63:1368–73.[Abstract/Free Full Text]
- Naidoo R, Reddi A. Lung resection for multidrug-resistant tuberculosis. Asian Cardiovasc Thorac Ann 2005;13:172–4.[Abstract/Free Full Text]
- Takeda T, Maeda H, Hayakawa M, Sawabata N, Maekura R. Current surgical intervention for pulmonary tuberculosis. Ann Thorac Surg 2005;79:959–63.[Abstract/Free Full Text]
- Sung SW, Kang CH, Kim YT, Han SK, Shim YS, Kim JH. Surgery increased the chance of cure in multi-drug resistant pulmonary tuberculosis. Eur J Cardiothorac Surg 1999;16:187–93.[Abstract/Free Full Text]
- Shiraishi Y, Nakajima Y, Katsuragi N, Kurai M, Takahashi N. Resectional surgery combined with chemotherapy remains the treatment of choice for multidrug-resistant tuberculosis. J Thorac Cardiovasc Surg 2004;128:523–8.[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]
- 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]
- Furak J, Trojan I, Szoke T, Tiszlavicz L, Morvay Z, Csada E, et al. Surgical intervention for pulmonary tuberculosis: analysis of indications and perioperative data relating to diagnostic and therapeutic resections. Eur J Cardiothorac Surg 2001;20:722–7.[Abstract/Free Full Text]
- Odell JA, Henderson BJ. Pneumonectomy through an empyema. J Thorac Cardiovasc Surg 1985;89:423–7.[Abstract]
- Blyth DF. Pneumonectomy for inflammatory lung disease. Eur J Cardiothorac Surg 2000;18:429–34.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
