INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Malignant pleural effusions are a common complication in advanced malignancy. Metastatic lung and breast cancer account for 75% of cases, with the remaining 25% occurring in a wide range of neoplastic diseases. Prognosis in these advanced tumors is generally very limited (1, 2). Most effusions do not respond to systemic therapy. Dyspnea, chest pain, and cough impair quality of life. Symptomatic relief can be achieved by drainage of pleural fluid. However, rapid recurrence of pleural effusion requires repeated thoracocentesis, thereby increasing the risk of complications. Therefore, for most patients, a safe, simple, and reliable method for pleurodesis is warranted after successful initial drainage. This can be attempted through instillation of a sclerosing agent via tube thoracostomy, insufflation of talc powder by thoracoscopy, or surgical pleurectomy.

Because tube drainage alone is not sufficiently effective for long-term effusion control, bedside instillation of a chemical agent via thoracostomy has become a popular method for pleurodesis (3). It is easy to perform, but its long-term success rate, like that of tube drainage, is limited. Currently, bleomycin instillation is one of the commonly used methods for effusion control. It has an 81% success rate at 30 d at best, which drops below 50% after 6 mo (6). In contrast, intracavitary application of talc, either through poudrage via thoracoscopy or via instillation of a slurry, has lasting success in 88 to 100% of all cases (7). Surgical pleurectomy for malignant effusion can be highly effective, but carries the morbidity of an extensive operation and a high cost. It is rarely indicated as a palliative procedure, and is generally reserved for patients with unsuccessful pleurodesis (10). Thoracoscopic talc poudrage is an established technique for the application of talc. Although general anesthesia is preferred during its use in some medical centers, local anesthesia makes the method a simple, inexpensive, and minimally invasive nonsurgical procedure, often referred to as "medical thoracoscopy." Both methods are widely used throughout Europe, but have not been compared prospectively.

We designed the prospective randomized trial reported here to compare the short- and long-term efficacy of thoracoscopic talc poudrage under local anesthesia with that of bleomycin instillation via a small-bore tube. We prospectively assessed side effects, factors influencing success rates, objective and subjective parameters of patient comfort, and determinants of cost of the two procedures.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

All patients with symptomatic pleural effusions and underlying malignancy were potential candidates for the study, and underwent thoracostomy with a small-bore tube (O.D. = 2.7 mm; Plastimed, France) under sonographic guidance. Inclusion in the study required documentation of malignant pleural disease, complete lung expansion on chest radiography at 12 to 24 h after drainage, improvement of symptoms after drainage, and expected survival of more than 1 mo. Patients with loculated effusions, those with previous drainages, and those with previous attempts at pleurodesis were not enrolled; nor were patients with known adverse reactions to the study medications or those with a severe coagulation disorder. In most cases, the pleurodesis procedure was scheduled on the next working day after thoracostomy. There were no explicit exclusion criteria based on performance status or medical condition, but patients had to be judged as capable of undergoing medical thoracoscopy in order to be included. The subjects gave written informed consent according to the study protocol approved by the local ethics board. Between August 1996 and October 1998, 36 patients were enrolled in the study and randomized to undergo talc poudrage or bleomycin instillation.

Bleomycin Pleurodesis

Bleomycin pleurodesis was done at bedside by a trained pulmonologist, using the small-bore tube inserted previously. At 15 min after instillation of 20 ml of lidocaine 1%, 60 IU of bleomycin (ASTA, Wangen, Switzerland) in 100 ml of normal saline were instilled and the tube was clamped. After 2 h, the tube was opened and was kept under suction (25 cm H₂O) until removal at least 48 h later.

Thoracoscopic Talc Poudrage

Thoracoscopy was done by a trained pulmonologist in an endoscopy suite, assisted by another physician and a trained nurse. Pneumothorax was induced under fluoroscopic control via the drainage tube with the patient in the left or right lateral decubitus position. About 20 to 30 ml of lidocaine 1% were used for local anesthesia, and sedatives (midazolam) and analgetics (pethidine) were given intravenously as needed. Oxygen saturation (Sa_O2), pulse, and blood pressure were assessed regularly. After complete evacuation of all remaining effusion fluid, incidental adhesions were disrupted with the blunt, closed biopsy forceps. Under visual control, 5 g of asbestos-free talc (Luzenac, Italy) were sprayed into the pleural cavity, using the Wolf equipment (Treier, Beromünster, Switzerland). The thoracoscope was then withdrawn, a thoracostomy tube (O.D. = 6.7 mm) was inserted posteriorly, and the pneumothorax was evacuated immediately. The drainage tube was kept under suction (25 cm H₂O) for at least 48 h.

Postinterventional Treatment

All patients were seen by respiratory therapists and encouraged to use incentive spirometry. For pain control, ibuprofen or a synthetic morphine analogue (pethidine, tramadol) were prescribed according to pain intensity. The chest tube was removed at least 48 h after the procedure and as soon as less than 200 ml of fluid were drained per day.

Data Collection

Tumor type, effusion-fluid pH, sonographic evidence of bulky pleural disease, and the duration of drainage were assessed. Appropriate lung expansion was determined by chest radiography before randomization, after treatment, and after removal of the thoracostomy tube. The use of medications and the volume of fluid drained were monitored daily. Subjective parameters were followed through daily interviews before and after treatment for as long as the drainage tube was in place. Patients were asked to express their level of dyspnea, pain, cough, and general well-being on a visual analogue scale (VAS) with a range of 0 to 100. If the patient remained in the hospital because of diagnostic or other therapeutic procedures unrelated to pleurodesis, discharge was scored at the day on which the patient could theoretically have left the hospital.

Follow-Up

Patients were followed individually by personal contact and regular telephone calls to primary care physicians or oncologists, and were called in for examination if indicated. No patient was lost to follow-up. Follow-up continued after the analysis period of 180 d. The primary endpoint of the study was a recurrence of pleural effusion, which was defined as a newly detected effusion needing drainage or occupying more than 33% of the pleural space on chest radiography as compared with the first radiograph after chest-tube removal, or death from any cause. Relapse was scored at the day of its radiologic detection or at the day of the first intervention for its treatment.

Study Design

The study was designed as a randomized, controlled trial with stratification according to tumor type and planned chemotherapy. Assumed success rates were 95% for thoracoscopic talc pleurodesis and 70% for bleomycin pleurodesis. With a statistical power of 80%, at a significance level of 95%, a significant difference was expected to be seen after analysis of 100 treatments. Because therapeutic success beyond 30 d after treatment turned out to be well below the expected rate of 70% for bleomycin pleurodesis, randomization was stopped after enrollment of 36 patients.

Statistical Analysis

Statistical analysis was done with StatView software for Macintosh computers (Apple, Inc., Cupertino, CA). Survival was calculated with Kaplan-Meier estimates. Unpaired nonparametric data (e.g., death, recurrence) were compared through the chi-square test and Fisher's exact test. Age, side of treatment, days of pain medication, tube drainage, follow-up, and unpaired comparisons of VAS data were evaluated with the Mann-Whitney U test. Paired comparisons of VAS data were done with Wilcoxon's signed ranks test. Calculations are based on data available at Day 180 after enrollment of the last patient.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient and Tumor Characteristics

Of 36 randomized patients, one refused the allocated treatment and one patient died unexpectedly before the treatment was applied. Three patients died of their malignancies at 7, 9, and 10 d, respectively, after bleomycin instillation, and none of these deaths was related to the procedure. There was no significant difference in mortality in the talc poudrage and bleomycin groups. The groups were also not different in terms of age or sex. Thirty cases (83%) were carcinomas of breast and lung (Table 1). Mortality was moderately linked to cancer type, with earlier deaths in lung versus breast cancer patients (p = 0.06).

Recurrence

The five patients mentioned previously who were not treated or did not survive for more than 14 d were excluded from further analysis. Among the remaining 31 patients with 32 treatments, follow-up was complete. One patient with breast carcinoma was randomized and treated twice for effusions on both sides on occasions separated by several weeks. A similar proportion of patients in both groups received chemotherapy within 180 d after treatment (Table 2). We observed a trend toward more frequent relapse at 30 d after bleomycin instillation which reached significance after 90 d and achieved greater significance at 180 d (Table 3). The incidence of recurrences after bleomycin treatment was continuous over time. In contrast, the success of pleurodesis with talc poudrage was constant after the first month (Figure 1). Recurrences led to secondary treatments within 180 d after bleomycin in eight cases and after talc pleurodesis in two cases (p = 0.04). Follow-up beyond 180 d resulted in one further recurrence at Day 214 after bleomycin treatment. In the patients analyzed for recurrence, there was no significant difference between groups in mortality at 30, 90, and 180 d (Table 4).

View larger version (12K): [in this window] [in a new window]   Figure 1.   Kaplan-Meier survival plot with p = 0.01 for number of patients without recurrences until 300 d after pleurodesis. Recurrences after thoracoscopic talc poudrage (in two of 15 patients) were observed only in the very early postprocedural period, with a stable success rate of 87% after 30 d. Bleomycin pleurodesis was also followed by early and late recurrences. Index cases = recurrence; censored cases = death.

Treatments and Subjective Parameters

After bleomycin pleurodesis, more fluid was drained than with talc poudrage, and the duration of chest tube drainage was longer by 1 d. Although not significant, this difference accounts for the shorter hospital stay of patients who had thoracoscopic talc poudrage (6.7 d, versus 7.7 d for bleomycin pleurodesis, p = NS). At baseline, no significant differences were found between the study groups for dyspnea, cough, pain, and overall well-being (Figure 2). Dyspnea and cough were significantly reduced by drainage. In the first 3 d after treatment, scores for dyspnea, cough, pain, and overall well-being were not significantly different in the tale poudrage and bleomycin groups. Pain intensity, however, was scored higher by bleomycin-treated patients. In contrast, a nonsignificant trend toward more intensive use of pain medication was observed for the talc-treated patients.

View larger version (20K): [in this window] [in a new window]   Figure 2.   Dyspnea, cough, pain, and overall well-being as assessed with visual analogue scales (VAS; range 0 to 100) at baseline and at Days 1, 2, and 3 after pleurodesis. (A and B) Significant improvement of dyspnea and cough occurred from baseline to postprocedural Day 1 (p = 0.01). (C ) Neither pain nor overall well-being (D) changed after treatment. At no time could a significant difference between treatments be found. Values shown are mean ± SEM. Circles = bleomycin; squares = talc.

Side Effects

Side effects of both treatments were minor. Temporary low-grade fever  38° C occurred in two patients in each treatment group. One of the patients in the talc-treated group turned out to have pneumonia on the side contralateral to pleurodesis. One patient complained about transient nausea after bleomycin treatment and another patient complained about shoulder pain, possibly related to the lateral decubitus position during thoracoscopy. All problems were successfully managed with routine measures. There was no major complication or death.

Prediction of Recurrence

There were two recurrences of effusion in the thoracoscopic talc pleurodesis group. One patient with ovarian carcinoma had an uncontrolled malignant peritoneal effusion, which was not drained separately. Because it was not possible at any point to remove the pleural drain from this patient we assumed that peritoneal fluid was drained transdiaphragmatically. The patient died 10 wk after the intervention, and autopsy was refused. In the analysis for factors discriminating between patients with successful therapy and those with recurrence, none of the three variables of cancer type, pleural fluid pH, or bulky pleural disease could be identified as predictive (Table 2).

Cost Estimation

All calculations were done with the assistance of hospital economists specialized in cost control. Because all patients in the study had the same diagnostic procedures and were hospitalized on the general ward, cost calculation was focused on procedure-related costs (Table 5). No difference in cost for managing side effects or complications had to be accounted for in the study. The most important cost-driving factors related to the procedures used in the study were personnel costs and intervention-room time for talc poudrage, and the cost for bleomycin itself. Direct procedural costs in our environment slightly favored bleomycin pleurodesis. The main cost-driving factor, however, was the time spent in the hospital (fixed costs per day). The trend toward longer drainage with bleomycin at least equalized the costs for the two treatments. Taking into account the significant excess number of recurrences that had to be treated after bleomycin pleurodesis, thoracoscopic talc poudrage is clearly more cost effective in the long term.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The remaining lifespan of patients with symptomatic malignant pleural effusions is very limited. The goal of treatment should be rapid and durable relief of symptoms. In all but the most seriously ill patients, palliative pleurodesis is the management of choice. The ideal procedure should be simple, safe, well tolerated, of lifelong success, and cost effective in combining a low initial expense and minimal recurrence rate. The present study is to our knowledge the first to directly compare thoracoscopic talc poudrage under local anesthesia with bleomycin instillation via a small-bore chest tube for pleurodesis. It demonstrates the superior efficacy and cost effectiveness of thoracoscopic talc pleurodesis.

Our study was a comparison of two standard procedures used in many European hospitals. It has to be taken into consideration that different substances and methods of application were used in our two study groups. Bleomycin pleurodesis is esteemed by many clinicians for its simplicity and rapid availability. In our study, bleomycin pleurodesis was safe and well tolerated. Although its short-term success rate was acceptable (59%), its long-term efficacy was disappointing, with almost half of the patients treated with it requiring repeated treatments within half a year. In contrast, we observed only two early failures of thoracoscopic talc pleurodesis under local anesthesia, both within the first month after therapy. One of these patients had uncontrolled malignant ascites, probably draining via the patient's chest tube. Long-term follow up after thoracoscopic talc poudrage under local anesthesia revealed a very good result (87% success rate) with no additional recurrence.

The patient characteristics, distribution of cancer types, and survival rates in our study appear to have been representative for malignant pleural effusion (1). A comparison of our results with those in other published reports is rather difficult because of divergent definitions of treatment failure, variable patient selection, methodologic incongruence, and limited published data. Thoracoscopic talc poudrage in various other case series and comparative studies (9, 11) had an 86% to 100% short- and long-term success rate, resembling our results. Talc poudrage clearly improves the 0% to 64% success rate of thoracoscopy and drainage alone (3). The outcome of our bleomycin-treated group at 30 days (59%) was slightly less satisfactory than expected from other studies that had reported early success rates of 62% to 81% with bleomycin (6, 8, 11, 15, 16). The long-term efficacy of the procedure has not been so well studied and seems less encouraging, at 34.6% at 6 mo in a previous trial (6) and 35% in our study.

Accepted predictors of failure of pleurodesis are a pleural fluid pH  7.2 and a high pleural tumor burden (14, 17). Both factors were less favorable in the talc-treated group in our study, and therefore cannot explain the difference in outcome. A small-bore tube for chemical pleurodesis, as used in our bleomycin-treated patients (I.D. = 1.5 mm), has been reported to improve patient comfort without impairing success rate (18). Because thoracoscopy was followed by insertion of a large-bore drain (I.D. = 4 mm), it could be argued that drainage after treatment was less effective in the bleomycin-treated patients. However, the effectiveness of small-bore catheters for chemical pleurodesis was demonstrated in a recent, prospective randomized trial with 108 patients. Thirty-day success rates of 72% with bleomycin and 79% with tetracycline were similar to those achieved with these agents in trials with large-bore catheters (16). In our study, all patients were successfully drained with small-bore catheters before treatment. Furthermore, the bleomycin-treated patients drained more fluid after treatment via small-bore drains than did the talc-treated patients with large-bore drainage tubes. We therefore do not consider drainage failure a relevant cause for the early recurrences of effusion after bleomycin treatment in our study. An influence of the catheter used on late recurrences seems unlikely.

Talc poudrage in patients with pneumothorax and a normal parietal pleura causes considerable pain. It might therefore be expected that thoracoscopic talc poudrage under local anesthesia would cause more pain than instillation of bleomycin. Furthermore, bleomycin was given through a small-bore catheter, which causes almost no pain, whereas patients who had talc poudrage were drained with large-bore chest tubes. As expected, patients with small-bore tubes had sightly lower pain scores before treatment. However, in our assessment of subjective side effects after the bleomycin and talc poudrage procedures, no difference was found for pain or overall well-being. In contrast, despite a trend toward a higher pain score after bleomycin, patients who had talc treatment subsequently received more pain medication. Probably, more medication was given routinely but without intent to patients after the more invasive treatment. Dyspnea and cough were reduced after drainage in both groups, and were not significantly different in the two groups in the first 3 d after treatment.

In general, local pain and low-grade fever as common side effects of pleurodesis were self limiting and could easily be controlled with routine medications. No major side effects of either method occurred, and have only occasionally been reported. Bleomycin is cleared renally and causes systemic toxicity when given for pleurodesis in patients with renal failure (19, 20). No serious complications were observed after talc treatment in our patients even though they included patients in poor medical condition. Reported side effects of thoracoscopic talc pleurodesis include acute respiratory distress syndrome (ARDS), pneumonitis, empyema, and local wound infections. Although serious respiratory complications traditionally were believed to occur very rarely with talc pleurodesis, and only with larger talc doses ( 10 g), recent retrospective studies have found a higher incidence with talc doses as low as 2 g and with a range of different techniques of application (7, 21, 22). Systemic distribution of talc is discussed as a possible mechanism for ARDS (21, 23). In the future, the safety of talc for pleurodesis will need close attention.

The superior cost efficiency of thoracoscopic talc poudrage in our study is founded on the excess recurrences of effusion that had to be treated after bleomycin pleurodesis; the direct costs for pleurodesis with the two procedures are not substantially different. However, this is true for our environment, and may not be applicable to other countries, where the balance between cost-driving factors is different. The cost efficiency of pleurodesis has recently been discussed in the literature (24). On the basis of published data, the investigators calculated that costs per symptom-free day slightly favored bleomycin pleurodesis over thoracoscopic talc poudrage. However, in that study, the procedural cost for thoracoscopic talc pleurodesis included general anesthesia, surgical staff costs, and costs of postoperative monitoring. Our data indicate that general anesthesia as a major cost-driving factor can be safely eliminated without impairing success rate of talc pleurodesis. Furthermore, medical thoracoscopy is performed by pulmonologists in an endoscopy suite, and can therefore be scheduled independently. This contributes to a shorter hospital stay and may make thoracoscopic talc pleurodesis under local anesthesia the more favorable method.

Recent reports show encouraging success rates for instillation of talc slurry via thoracostomy tube (8, 25). If these results are confirmed, instillation of talc slurry could be the method of choice for many patients with malignant pleural effusions. However, thoracoscopic talc pleurodesis would remain the method of choice in situations requiring visualization of the pleura for diagnostic or therapeutic purposes.

Conclusion

Thoracoscopic talc pleurodesis under local anesthesia is highly effective and clearly superior to bleomycin instillation, especially in the long term. Both methods are safe, provide symptomatic relief, and have similar subjective tolerances. The initial expense for pleurodesis is outweighed by the shorter drainage time and reduced need for additional treatments of recurrences with this technique. Thoracoscopic talc pleurodesis under local anesthesia is cost effective.