INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pulmonary diseases due to multidrug-resistant mycobacteria are responsible for therapeutic failure in about half the patients (1) with mortality rates as high as 22% in infections with multidrug-resistant tuberculosis (MDR-TB) (1) and 30% in infections with Mycobacterium avium complex (MAC) (3). The extension of the disease, particularly in cases associated with cavities, as well as the number of drugs to which the strains are resistant, are strongly associated with medical treatment failure (1).

Therefore, several investigators have proposed surgical resection of the infected parts of the lungs to improve the outcome of these patients (4). Resection of the diseased areas of the lungs had been shown to improve the outcome of patients with tuberculosis before the introduction of modern antimycobacterial chemotherapy (10). However, we are not aware of controlled trials of resectional surgery for pulmonary diseases caused by multidrug-resistant mycobacteria. Several retrospective reports suggest that selected patients may benefit from this kind of surgery (4, 5, 7). Unfortunately, Iseman and coworkers (9) and Moran and coworkers (5) have reported that two-thirds of the patients, essentially those at the highest risk of medical treatment failure, were not eligible for excisional surgery because of the extension of the disease or because of poor respiratory function (1). In the remaining patients, postoperative complications were frequent, with rates ranging from 30 to 50% (4).

Collapse therapy with plombage has been proposed in the early 1950s in patients with persistent cavitary disease despite specific antimycobacterial drug therapy (10). This technique consisted of denuding the ribs overlying the diseased area, collapsing selectively the parenchymal cavitiesthe major sources of bacillary proliferationand maintaining the collapse by filling the subcostal extraperiostal space with a "plomb," mainly lucite spheres. Compared with staged extrapleural thoracoplasty from which it is descended, the major advantages of this procedure were as follows: (1) collapse was obtained with only one operation; (2) it did not cause any physical deformity; (3) it was applicable to poor risk patients; (4) it preserved lung function; (5) it was applicable to bilateral disease; (6) hospital stay was short; (7) it markedly reduced postoperative complications (11). When practiced, this procedure cured up to three-fourths of the patients (10) but was abandoned after a few years, as were other surgical procedures, when the improvements of antimycobacterial chemotherapy made it sufficient to cure almost all of the patients.

However, the emergence of diseases caused by multidrug-resistant mycobacteria led us to propose collapse therapy with plombage as a complementary therapy. We report here a series of patients with pulmonary diseases caused by multidrug-resistant mycobacteria, considered at high risk of treatment failure and unsuitable for pulmonary resection, who underwent collapse therapy with plombage.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Seven patients with multidrug-resistant mycobacterial pulmonary infection underwent collapse therapy with plombage between September 1992 and March 1996 at the Val d'Or hospital in St. Cloud, France (Table 1). None of them had been considered a candidate for resection because of extension of the disease or of poor clinical and functional status. There were six men and one woman. Mean age was 46 yr (range, 26 to 68 yr).

Four patients had MDR-TB resistant to isoniazid (1 mg/ml) and rifampin (5 mg/ml) on the average to 5.5 antituberculous drugs. All patients had received multiple courses of chemotherapy at intervals for 2 to 16 yr before referral (mean 7.5 yr).

In two patients, who had advanced chronic obstructive disease (COPD), infection was due to Mycobaterium xenopi, in one to MAC. All three patients met the diagnostic criteria recommended by the American Thoracic Society in 1997 (12) and had treatment failure despite antimycobacterial regimen containing clarithromycin administered for at least 12 mo.

All patients had extensive bilateral and cavitary disease. In four patients the cavities were bilateral. Six patients had active disease at time of surgery. One patient infected with M. tuberculosis had bacteriological conversion 1 mo before collapse therapy but was considered at high risk of relapse because of extensive bilateral cavitary disease, heavy pretreatment, and poor drug susceptibility pattern of his infectious strain. HIV status was negative in all patients.

Pulmonary function tests were not available in three patients because of contamination risks. In the other four, FEV₁ was respectively 30, 31, 36, and 48% of predicted values (mean 36%). In three patients partial pressure of oxygen while breathing room air was below 60 mm Hg.

Collapse therapy consisted of denuding the ribs overlying the diseased area (extraperiostal apicolysis), collapse of the cavitary portion of the lungs, and filling the space with five to 18 polystyrene spheres (mean 10) of 4 cm diameter (volume 25 cm³). Collapse therapy was unilateral in all but one patient. In patients with bilateral cavities, the lung harboring the most voluminous cavities was collapsed. In Patient 2, because of voluminous cavities in both lungs, collapse therapy was bilateral. In Patient 1, removal of a homolateral aspergilloma preceded collapse. When possible, plombage was removed by thoracotomy as soon as the periosteum had regenerated in the collapsed position, making plombage superfluous.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Data relating outcome of surgery are displayed in Table 2. There were no immediate postoperative deaths and no patient developed significant complications. Hospital stay ranged from 8 to 22 d (mean 12 d). In all patients surgery was followed by complete collapse of parenchymal cavities. Figure 1 shows the pre- and postoperative chest radiograph and computed tomographic (CT) lung scan of Patient 1. Bacteriological conversion was obtained in five of the six patients with active disease at time of surgery. In the patient who converted 1 mo before surgery, long-standing conversion was maintained. Medical treatment was continued for 8 to 14 mo after operation (mean 11.5 mo). These six patients remained sputum-culture-negative for mycobacteria after 8 to 58 mo (mean 34 mo). One COPD patient infected with M. xenopi died 4 mo after surgery of an acute bout of respiratory failure. His sputum was still culture-positive until death, although this was not considered to be directly responsible for the fatal issue. Plombage could be removed in four patients 3, 8, 11, and 25 mo after initial surgery and no local complication was observed (Figure 2). One patient refused removal of the plombage and one patient originating from Algeria could not come back to France for withdrawal of the spheres. Patient 5, infected with MAC, died suddenly at home 8 mo after collapse therapy, 5 mo after removal of the plombage. The cause of death could not be established; no autopsy was performed.

View larger version (146K): [in this window] [in a new window]   View larger version (105K): [in this window] [in a new window]   View larger version (110K): [in this window] [in a new window]   View larger version (151K): [in this window] [in a new window]   Figure 1.   Preoperative (A) and postoperative (B) chest radiographs and CT lung scans at the level of the carina of Patient 1. The previous cavity of the right upper lung is no longer demonstrable after collapse therapy.

View larger version (167K): [in this window] [in a new window]   View larger version (108K): [in this window] [in a new window]   Figure 2.   Chest radiograph (top) and lung CT scan (bottom) of Patient 1 after withdrawal of the plombage. Arrow shows the calcifications of the periosteum which has regenerated in the collapsed position. The extrapleural space has filled spontaneously.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Prolonged sputum conversion was observed after collapse therapy in five of six patients and in the patient who had inactive disease at time of surgery long-standing sputum conversion was maintained by collapse therapy. All these patients were considered at high risk of treatment failure because of extensive, bilateral and cavitary disease, heavy pretreatment, and susceptibility pattern of the infectious strains. In one patient infected with M. xenopi, who died of acute failure of his COPD 4 mo after surgery, follow-up was too short to assess the effectiveness of collapse therapy.

When compared with excisional surgery, hospital stay was short in our patients and no immediate postoperative death or complication was observed although all patients were considered at high risk because of respiratory impairment and active disease at the time of operation for six of them (4). Because this procedure is strictly extrapleural, it does not expose the patients to bronchopleural fistula, prolonged air leak, or hemorrhage, the most common complications of resection in these patients. Moreover, no postoperative respiratory failure was observed. Indeed, by selectively collapsing the diseased area, this procedure does not alter significantly respiratory function (11) as illustrated in Figure 3.

View larger version (83K): [in this window] [in a new window]   View larger version (83K): [in this window] [in a new window]   Figure 3.   Radionucleotide perfusion lung scans of Patient 1 before (A) and after (B) collapse therapy with anterior (1), posterior (2), right (3), and left posterior oblique (4) views. Selective collapse of the diseased lung did not alter the distribution of pulmonary perfusion.

Complications after collapse therapy with plombage reported in the 1950s included basically local infection and mediastinal compression (11). The latter should be avoided by using a number of plombs just necessary to collapse the underlying cavities. More recent reports have related late local complications resulting from plombage inserted up to 40 yr earlier, including acute infection and hemorrhage, migration of the plombage, erosion of surrounding structures, or development of neoplasms (13). To prevent such deleterious complications we chose to remove the foreign bodies as soon as the plombage became superfluous to maintain the collapse. This occurred after an average of 3 to 5 mo. Plombage could be removed in four of our patients with no need for complementary thoracoplasty. The remaining space filled spontaneously and no reexpansion of the underlying cavities was observed. An alternate strategy could be to use extraperiosteal plombage with round silastic expander as the plomb as suggested in a case report (16).

In patients eligible for pulmonary resection, most surgeons will operate after bacteriological conversion, since an active disease at time of operation correlates directly with a high rate of postoperative complications (4, 7). In our patients active disease at time of collapse therapy did not interfere with postoperative outcome. Moreover, collapse therapy hastened bacteriological conversion presumably by dramatically reducing the bacillary population and may have prevented further emergence of drug resistance in these poor compliant patients.

In our experience, collapse therapy with plombage was a safe alternative therapy in patients with cavitary disease caused by resistant mycobacteria at high risk of medical treatment failure when resectional surgery is contraindicated.