Eur J Cardiothorac Surg 1999;16:187-193
© 1999 Elsevier Science NL
Surgery increased the chance of cure in multi-drug resistant pulmonary tuberculosis
Sook-Whan Sunga,
Chang Hyun Kanga,
Young Tae Kima,
Sung Koo Hanb,
Young-Soo Shimb,
Joo Hyun Kima
a Department of Thoracic and Cardiovascular Surgery, Seoul National University Hospital, 28 Yongon, Chongno, Seoul, 110-744, South Korea
b Department of Internal Medicine, Seoul National University Hospital, Seoul, South Korea
Corresponding author. Tel.: +82-2-7603637; fax: +82-2-7643664
e-mail: swsung{at}snu.ac.kr,
 |
Abstract
|
|---|
Background: Medical treatment of multiple drug resistant (MDR) pulmonary tuberculosis is generally quite unsuccessful. Recently, surgical management is increasing and shows promise. We analyzed our experience to identify the benefits and complications of pulmonary resection in MDR pulmonary tuberculosis. Methods: A retrospective review was performed in 27 patients undergoing pulmonary resection for MDR pulmonary tuberculosis between January 1994 and March 1998. Their average ages were 40 years and were diagnosed a median of 15 months before surgery. All patients had resistance to an average of 4.4 drugs including isoniazid and rifampin, and had received second line drugs selected according to drug sensitivity test preoperatively. Most patients (92.6%) had cavitary lesions. Bilateral lesions were also identified in 19 patients (70.4%), but the main focus was recognized in one side of the lung. Most patients were indicated to operation for those who could not achieve negative sputum despite adequate medical treatment (n=16, 59.3%); or for negative patients who had significant pulmonary parenchymal lesion (n=11, 40.7%) which would have had a high probability of recurrence. Pneumonectomy was done in nine patients, lobectomy in 16 and segmentectomy in two. Results: There was no operative mortality. Morbidity occurred in seven patients (25.9%); prolonged air leakage in three patients, reoperation due to bleeding in two, bronchopleural fistula in one, and reversible blindness in one. The median follow up period was 15 months (range 3–45). Sputum negative conversion was achieved in 22 patients (81.5%) initially. However, continued postoperative chemotherapy could convert to negative in another four patients (14.8%). Only one pneumonectomy patient (3.7%) failed because of considerable contralateral cavity. Conclusion: For patients with MDR pulmonary tuberculosis which is localized, and with adequate pulmonary reserve function, surgical pulmonary resection combined with appropriate pre and postoperative anti-tuberculosis chemotherapy can achieve high success rate with acceptable morbidity.
Key Words: Multi-drug resistant • Tuberculosis • Lung • Surgery
|
1. Introduction
|
|---|
The treatment of pulmonary tuberculosis has been changed dramatically since the development of anti-tuberculosis drugs. After 1950 resectional surgery could be performed more safely due to introduction of streptomycin (SM), para-aminosalicylic acid (PAS), and isoniazid (INH). However, the establishment of multi-drug therapy including rifampin (RFP) has made the surgical role in the pulmonary tuberculosis useful only in a limited field. In the past, most patients could be managed by medical treatment only. But recently, drug resistance is becoming an important problem in the management of pulmonary tuberculosis. Some changes of treatment strategy are now being considered.
The incidence of multi-drug resistant (MDR) pulmonary tuberculosis, which means resistant to both INH and RFP is decreasing in Korea. Percentage incidence was 10.5% in 1985, 7.1% in 1990, and 5.2% in 1997 [1]. Nevertheless, that portion is still a big threat to tuberculosis patients. The failure rate of medical therapy is over 40%, and a poor long-term survival rate is reported [2]. Since the introduction of surgery in MDR pulmonary tuberculosis [3], several centers reported a high rate of sputum conversion and improved long-term survival [3–8].
Hence we reviewed retrospectively the medical records of 27 patients who had undergone resectional surgery due to MDR pulmonary tuberculosis, and tried to clarify the surgical role in the management MDR pulmonary tuberculosis.
|
2. Patients and methods
|
|---|
All 27 consecutive patients had undergone resectional surgery for multi-drug resistant (MDR) pulmonary tuberculosis from January 1994 to March 1998 at Seoul National University Hospital. MDR pulmonary tuberculosis was defined as the tuberculosis strain that is resistant to at least two fundamental drugs, INH and RFP. Eleven drugs were evaluated for drug sensitivity; these were INH, RFP, SM, PAS, ethambutol (EMB), kanamycin (KM), enviomycin (EVM), prothionamide (PTH), cycloserine (CS), ofloxacin (OFL), and pyrazinamide (PZA).
We defined the duration of pulmonary tuberculosis from the first diagnosed time irrespective of primary treatment or surgical retreatment. Negative sputum conversion was considered when sputum smear and/or culture was documented as negative for tuberculosis at least three serial tests monthly. If tubercle bacilli were identified after postoperative negative conversion or progression of pulmonary lesion was recognized on chest X-films, we defined those as relapse.
In 27 patients, 18 were male and nine were female. The mean age was 40.1±12.4 years. The duration of pulmonary tuberculosis was a median of 7.0 years before surgery (range 1–29), and an operation time lag after diagnosis of MDR tuberculosis was a median of 15 months (range 1–168). Most patients were referred from small community clinics, and at our hospital a median of 6 months were required to be operated (range 0–135). Eighteen patients (66.7%) had a history of relapse of pulmonary tuberculosis and were under a re-treatment protocol.
Drug sensitivity testing was performed in all patients, and the resistance ratio of each drug is shown in Fig. 1. INH and RFP resistance was documented in all patients. The resistance to the primary drug was higher than the secondary drug. The average number of resistant drugs was 4.4 (range 2–10). The medication history was different in each patient. Modification of drugs was necessary in most of the patients who were referred from small clinics. The selection principles of the drugs were as following: (1) no resistance was documented by drug sensitivity testing, (2) no history of previous usage, and (3) use of bactericidal drugs as long as possible. The patients had taken an average of 5.0 drugs (range 4–7 drugs) before operation and most of the drugs were secondary anti-tuberculosis drugs. The duration of drug treatment was scheduled more than three months before surgery.
View larger version (47K):
[in this window]
[in a new window]
|
Fig. 1. Preoperative resistance ratio of each drug. The resistance of primary drugs was higher than secondary drugs.
|
|
Routine preoperative evaluations were done in every patient, such as simple chest X-ray, computed tomography (CT), and bronchoscopic examination, sputum smear and or culture for tuberculosis. To assess resectability, pulmonary function testing and lung perfusion scans were also performed. According to the CT findings, every patient showed various kinds of radiologic lesions and mixtures. The most common radiologic findings were cavities and nodules (Table 1). Cavitary lesions could be found in 25 patients (92.6%). Bilateral lesions were also identified in 19 patients (70.4%), but the main severe lesions were confined at one side of the lung. Contralateral nodules, cavities and fibrosis were very small in bilateral patients; and these small lesions were seemed to be managed fairly well with drugs postoperatively. Every patient had to have reasonably localized disease and adequate pulmonary reservoir to be considered as surgical candidates (Fig. 2). Considerable bilateral disease was not indicated and not included in our series.
View larger version (40K):
[in this window]
[in a new window]
|
Fig. 2. Algorithm of the surgery for the management of multi-drug resistant pulmonary tuberculosis in our hospital (with permission, from Koh et al. [15]).
|
|
The patients were indicated to operation who did not achieve negative sputum despite adequate medical treatment (n=16, 59.3%), or those negative patients who had significant pulmonary parenchymal lesions (n=11, 40.7%); such as totally destroyed lung and/or large cavities, which would have a high probability of recurrence. One of the AFB positive patients was due to tuberculous empyema combined with bronchopleural fistula (Table 2).
The operation was aimed to remove major diseased areas and preservation of relatively healthy parenchyma as much as possible. Lobectomy was performed in 16 patients (59.3%) and pneumonectomy was performed in nine (33.3%), and segmentectomy in two (7.4%). In six patients among lobectomy group, wedge resection was added concomitantly for other lobar localized small lesions. There were no cases of bilateral resection. Bronchial stump was reinforced with adjacent viable tissue in 21 patients (77.8%), except six. A pericardial fat pad was used in 20 patients (74.1%) and pleural flap in one patient (3.7%). In the lobectomy group, when the remaining lungs were large enough to fill the ipsilateral thoracic cavity, we omitted bronchial wrapping in four patients. We also omitted wrapping in two segmentectomy patients. Bronchial stump with viable tissue was reinforced in all pneumonectomy groups.
Postoperatively every patient was scheduled to have intensive anti-tuberculosis chemotherapy regimens for at least 18 months. The regimens were not changed until resected specimen's culture was reported. When we noticed different results compared with preoperative sensitivity tests, modifications of drug regimens were followed as in seven patients (25.9%).
Data were expressed as median value or mean±SD. Comparisons between patients were made by unpaired Student's t-test for continuous variables and by chi square test for categorical variables. A P value of less than 0.05 was considered statistically significant.
|
3. Results
|
|---|
Among 27 patients, 11 patients (40.7%) showed negative sputum preoperatively, and 16 patients (59.3%) showed persistent positive smear and culture test. The duration of the secondary drug treatment was significantly longer in the sputum positive group compared to the negative converted group, 48.0±51.2 versus 11.6±13.4 months (P<0.05). There was no difference in the frequency of the applied drugs except cycloserine (Table 3). In negative converted patients group, cycloserine was also adapted more frequently than the other secondary drugs (P<0.05). The pathologic and radiologic findings between positive and negative converted group did not show any differences (Table 4).
View this table:
[in this window]
[in a new window]
|
Table 4. Comparison of the various clinical and radiological and surgical factors between the favorable group and unfavorable group
|
|
There was neither operative death nor late death. Complications after operation for MDR tuberculosis occurred in seven patients (25.9 %). Prolonged air-leakage more than one week was occurred in three patients, reoperation due to postoperative bleeding in two patients, transient blindness in one, and postoperative bronchopleural fistula in one. The cause of the transient blindness was not identified. Brain CT and fundus examination did not show any pathologic findings, and the patient recovered spontaneously without any neurologic sequelae in 2 weeks. Reoperation was performed in three patients. Two were due to postoperative bleeding and one was due to postoperative bronchopleural fistula, this patient received Eloesser procedure.
Follow-up was possible in all patients at our hospital. The median follow-up period was 29 months (range 3–53). In the early postoperative period, sputum conversion was achieved in 22 patients (81.5%), and the remaining five patients (18.5%) showed positive sputum (Fig. 3). Among negatively converted 22 patients, recurrence was identified in two patients. One patient, the first case of this series, showed a little increased size of the nodule 9 months after completing her postoperative medication. She resumed medication for only another four months because AFB was not demonstrated in smear and culture and the size of the nodule had not increased. The other patient had contralateral small cavity and showed positive AFB during 5-month follow-up period after surgery. However with continuous postoperative medical treatment, this patient could be converted successfully in 10 months after the relapse. The remaining five patients who had positive sputum after the operation, have undergone continued medical therapy, and negative conversion could be achieved in four patients after 1, 1, 2 and 9 postoperative months for each. Only one patient could not be converted to negative and he is the only patient who failed in this series. He is still under anti-tuberculosis medication for 3 years and 2 months after surgery, and the cavity size has been increasing slowly (Fig. 4).
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 3. Diagram for result of operation in 27 patients. MDR-TB, mutidrug resistant pulmonary tuberculosis; AFB, acid fast bacillus.
|
|
View larger version (137K):
[in this window]
[in a new window]
|
Fig. 4. Sequential chest films taken preoperatively, and 2, 9, 37 months postoperativly (film A,B,C,D, respectively). He received pneumonectomy due to totally destroyed left lung with large cavities. This is the only patient who failed to get negative conversion after operation. Contralateral small cavities have been increasing in size gradually postoperatively.
|
|
Until June 1988, 15 patients (55.6%) could cease medication based on the absence of recurrence after a median of 20 months of postoperative medication (range 17–26). The remaining 12 patients were under medication for a median of 11 months (range 3–45). The medication schedule was prolonged according to the activity of the residual lesion and sputum smear and/or culture test. Most of the residual lesions were nodular lesions, while cystic, streaky or atelectasis was identified in some patients postoperatively. Nevertheless, the kinds of the residual lesions did not affect the postoperative medication. The overall negative conversion rate after combined medical and surgical treatment was 96.3%.
To identify the risk factors for unfavorable outcome after surgery, we divided patients into two groups. The favorable group was those patients who showed negative sputum immediate postoperatively, and the unfavorable group was those who showed positive sputum postoperatively or relapsed after initial negative sputum conversion (Table 4). The risk factors of the duration of pulmonary tuberculosis and pneumonectomy showed differences among the two groups. This suggests that longer duration and advanced parenchymal lesion may affect the surgical outcome.
|
4. Discussion
|
|---|
The incidence of tuberculosis is under decreasing tendency in South Korea [1]. However, poor compliance of the patients and inadequate medical management by untrained physicians are the important problems in a small proportion of pulmonary tuberculosis patients until now. These problems can be the source of drug resistance in pulmonary tuberculosis. The co-resistance rates to INH and RFP has been decreasing in Korea, as evidence by 10.5% in 1985, 7.1% in 1995 and 5.2% in 1995 [1]. Nevertheless, drug resistance eventually raises treatment failure rate due to a limitation of the number of drugs available, low effectiveness and the high complication rate of the secondary drugs [9].
Medical treatment in MDR tuberculosis has been reported not to be satisfactory. Treatment failure rate has been reported to be 40–70% and the major cause of mortality being the progression of pulmonary tuberculosis [2,10,11]. As a result, surgical resections for MDR tuberculosis have been concerned and tried.
Iseman [3] reported the results of surgical treatment of MDR tuberculosis and suggested that the combination of medical and surgical treatment can achieve higher long-term sputum conversion rate. They proposed three surgical indications for MDR pulmonary tuberculosis. First, that drug resistance is so extensive that there is a high probability of failure or relapse. Second, the disease is sufficiently localized so that preponderance of radiographically visible disease can be resected, with the expectation of adequate cardiopulmonary capacity post-surgery. Third, that surgery is indicated when there are sufficient drugs to diminish the mycobacterial burden enough to facilitate probable healing of the bronchial stump. By our limited experience, we suggest that pulmonary resection can be done successfully even when there are not sufficient number of sensitive drugs. In addition, the time of operation is closely related to the indications. The earlier the operation is performed more salvage of lung parenchyma is possible. However, adequate medical management is demanded prior to the procedure so as to avoid unnecessary operation. Although no clear-cut guidelines are established, surgery is recommended in the early period during pharmacological therapy. Most of the authors performed surgery between 3 and 6 months after medication [4,5,9]. This concept was based on the report of Goble [2] that most of the negative conversion could be achieved between 1 and 8 months, and that the median value was 2 months.
Many patients in our series had a long history of pulmonary tuberculosis. Especially our early experience (1994–1995; n=13), the long duration was attributed to the fact that our physicians were very reluctant to advise an operation in MDR tuberculosis. Consequently, pneumonectomy represented major surgery in the treatment of MDR tuberculosis in our early patients. However, a recent trend has been to early operation and as much salvage of lung parenchyma as possible. Lobectomy has been the most frequently performed operation since 1996 (71.4%; n=14).
Bilateral lesions are frequently found in MDR tuberculosis patients [3,8]. In our study, bilateral lesions could be identified in 19 patients (70.4%). In most patients however, the prominent areas were sufficiently localized to one side of the lung. The contralateral lesion was minimal to mild in the extent of the severity of disease, and it was fibro-nodular lesion in nearly all cases. However, in three patients, small cavitary lesions were found preoperatively in the contralateral lung. In one patient, surgery failed to convert sputum to negative as described above (Fig. 4). In the second patient, initial negative conversion was achieved but later relapsed. Continued medical treatment was possible to convert to negative again. The third patient was successful. The indication of surgery for bilateral cavitary lesion is not conclusive well. The selection of the patient may lie in the condition of the patient and in the pulmonary residual capacity after bilateral resection. When we plan to operate bilateral cavitary disease, we have to be prepared to remove contralateral small cavitary lesions if sputum is not converted to negative after resection of one side major cavitary lesions.
The post-operative mortality and morbidity rate for pulmonary tuberculosis is decreasing. This is largely owing to proper selection of the patient, development of the anesthesia technique, stapling devices, and the effective use of the anti-tuberculosis agent postoperatively [12]. The mortality rate of MDR tuberculosis after resectional surgery ranged from 0 to 3% [3–8,13]. We did not experience any mortality in our study. The complication rate is reported at 20–50%, but most complications were minor. Major complications are reported at around 15%, and included empyema, bronchopleural fistula, bronchogenic spread to contralateral lung field, postoperative bleeding and respiratory failure. Among them, bronchopleural fistula and empyema take an important position especially in those patients whose sputum was positive preoperatively. Some authors recommend using muscle flap to reinforce the bronchial stump and reported a low incidence of bronchopleural fistula [14]. On the other hand, others report that the important point is the peribronchial blood supply. Less dissection and avoidance of electrical diathermy could prevent bronchopleural fistula [6]. In our study, we did not use a muscle flap but most frequently used a pericardial fat or pleural flap, or either did not apply any reinforcement to bronchial stump. A bronchopleural fistula developed in only one patient after 4 months postoperatively, but he was the patient who had previous bronchopleural fistula and tuberculous empyema and received a pleuro-pneumonectomy. When the remaining lungs were big enough to fill the ipsilateral thoracic cavity, we omitted bronchial wrapping in four patients in lobectomy and two with segmentectomy. There were no bronchial healing complications in these patients. In our experience, bronchial reinforcement is not always necessary in lobectomy and segmentectomy.
The postoperative duration of anti-tuberculosis medication has not been defined clearly; however we could find two cases of relapse in initially negatively converted patients. Therefore adequate drug coverage is important in surgical operation. A short-term protocol in eight to twelve month period is reported [6]. However, the period of retreatment for relapsed pulmonary tuberculosis is usually over 18 months, so we recommend 18–24 months regimen postoperatively. In cases of relapse or positive sputum postoperatively the regimen schedule should be prolonged according to disease status. An algorithm for the general treatment strategy of MDR tuberculosis as practiced in our hospital is shown in Fig. 2 [15].
|
Footnotes
|
|---|
Presented at the 12th Annual Meeting of the European Association for Cardio-thoracic Surgery, Brussels, Belgium, September 20–23, 1998.
|
Appendix A
|
|---|
Conference discussion
Dr T.F. Molnar (Pecs, Hungary): In which circumstances, if ever, do you do a thoracoplasty in the case of TB? Do you carry on giving antituberculotic drug treatment after the surgery or not? And if yes, for how long after an operation?
Dr Sung: Regarding thoracoplasty, usually we do not do thoracoplasty anymore. If the lesion is resectable and the patient is tolerable to operation, it's better to do a resectional surgery than thoracoplasty. When we anticipate space problem after resectional surgery, we may add a lateral thoracoplasty to decrease postoperative morbidity. And postoperatively, we medicate the patient usually for 20 months. If the patient's sputum is not converted postoperatively and the patient has residual diseases, we have to prolong the medication. In these cases, the duration of medication is dependant on sputum status and responsiveness of residual diseases. Sometimes we have to medicate for 30 months or 40 months.
Mr J. Dussek (London, UK): You specifically state that you did not include HIV-positive patients in this series. Does this mean that you have also operated on HIV-positive patients with resistant TB? And if so, what was the outcome?
Dr Sung: In our country, we don't have so many HIV patients. But I think if the patient has HIV, the results would be quite different, and I don't have any experience on this, so I couldn't answer.
Mr A. Jilaihawi (Glasgow, UK): I have operated on a few. We still have remnants of TB in Scotland. The majority of these I did were the avium type. Have you classified your Mycobacterium, the human type, the avium type, or the resistant ones?
Dr Sung: All of these cases are not avium type. We don't have so many cases of avium type nationwide. Korea is still an endemic area of tuberculosis, and we rarely have non-tuberculous mycobacteriosis. The principles of surgical indications and postoperative drug management of avium type could be nearly the same as resistant tuberculosis.
Mr J. Thorpe (Leeds, UK): Just some technical points of view. I have been taught to do thoracoplasties in this sort of situation and I'm still very pleased to do them. In terms of your lobectomy/pneumonectomy, the lungs can be very thickened and infected. Do you do extrapleural dissection? And if so, what are your operative techniques?
Dr Sung: It depends on the disease extent. If the pleura is also involved in tuberculosis, we have to do a extrapleural dissection. If the pleura is not diseased, we can do just a simple lobectomy and pneumonectomy after adhesiolysis of pleura with electrocautery.
Mr Thorpe: And do you place any antibiotics in the pleural space?
Dr Sung: After pneumonectomy?
Mr Thorpe: At the time of operation.
Dr Sung: Not usually, we do not place antibiotics in the pleural space, even after pneumonectomy.
|
References
|
|---|
-
Hong Y.P. Tuberculosis in Korea: yesterday, today and tomorrow. Tuberc Respir Dis 1997;44:1-10.
-
Goble M., Iseman M.D., Madsen L.A., Waite D., Ackerson L., Horsburgh C.R. Treatment of 171 patients with pulmonary tuberculosis resistant to isoniazid and rifampin. N Engl J Med 1993;328:527-532.[Abstract/Free Full Text]
-
Iseman M.D., Madsen L., Goble M., Pomerantz M. Surgical intervention of pulmonary disease caused by drug-resistant Mycobacterium tuberculosis. Am Rev Respir Dis 1990;141:623-625.[Medline]
-
Pomerantz M., Madsen L., Goble M., Iseman M.D. Surgical management of resistant mycobacterial tuberculosis and other mycobacterial pulmonary infections. Ann Thorac Surg 1991;52:1108-1111.[Abstract]
-
Pomerantz M., Brown J. The surgical management of tuberculosis. Semin Thorac Cardiovasc Surg 1995;7:108-111.[Medline]
-
Leuven M.V., Groot M., Shean K.P., Oppell U.O. Pulmonary Resection as an adjunct in the treatment of multiple drug-resistant tuberculosis. Ann Thorac Surg 1997;63:1368-1373.[Abstract/Free Full Text]
-
Treasure R.L., Seaworth B.J. Current role of surgery in Mycobacterium tuberculosis. Ann Thorac Surg 1995;59:1405-1409.[Abstract/Free Full Text]
-
Jun H.J., Han D.K., Park S.K., Song S.D., Choi P.H., Woo J.S. Clinical evaluation of surgical resection on pulmonary tuberculosis-Multiple drug resistant pulmonary tuberculosis. Kor J Thorac Cardiovasc Surg 1997;30:786-792.
-
Iseman M.D. Treatment of multidrug-resistant tuberculosis. N Engl J Med 1993;329:784-791.[Free Full Text]
-
Mitchison D.A., Nunn A.J. Influence of initial drug resistance on the response to short-course chemotherapy of pulmonary tuberculosis. Am Rev Respir Dis 1986;133:423-430.[Medline]
-
Lee J.C., Yoo C.G., Cheong H.S. Clinical features and management of multidrug-resistant tuberculosis. Tuberc Respir Dis 1996;43:14-21.
-
Moran J.F. Surgical treatment of pulmonary tuberculosis. In: Sabiston D.C., Spencer F.C., eds. Surgery of the Chest, 6th edition. Philadelphia: Saunders, 1995:752-772.
-
Rizzi A., Rocco G., Robustelli M., Rossi G. Results of surgical management of tuberculosis: experience in 206 patients undergoing operation. Ann Thorac Surg 1995;59:896-900.[Abstract/Free Full Text]
-
Pomerantz M., Brown J.M. Surgery in the treatment of multidrug-resistant tuberculosis. Clin Chest Med 1997;18:123-130.[Medline]
-
Koh W.J., Lee J.H., et al. Yoo C.G. The adjunctive role of resectional surgery for the treatment of multidrug resistant pulmonary tuberculosis. Tuberc Respir Dis 1997;44:975-991.
Received September 21, 1998;
received in revised form April 27, 1999;
accepted April 27, 1999.
Top
Abstract
1. Introduction
