Management of the patient with a symptomatic malignant pleural

Transcription

1 1992 A Randomized Comparison of Indwelling Pleural Catheter and Doxycycline Pleurodesis in the Management of Malignant Pleural Effusions Joe B. Putnam, Jr., M.D. 1 Richard W. Light, M.D. 2 R. Michael Rodriguez, M.D. 2 Ronald Ponn, M.D. 3 Jemi Olak, M.D. 4 Jeffrey S. Pollak, M.D. 5 Robert B. Lee, M.D. 6 D. Keith Payne, M.D. 7 Geoff Graeber, M.D. 8 Kevin L. Kovitz, M.D. 9 1 Department of Thoracic and Cardiovascular Surgery, the University of Texas M. D. Anderson Cancer Center, Houston, Texas. 2 Department of Medicine, Saint Thomas Hospital and Vanderbilt University, Nashville, Tennessee. 3 Department of Surgery, Hospital of Saint Raphael and Yale University, New Haven, Connecticut. 4 Department of Surgery, University of Chicago, Chicago, Illinois. 5 Department of Radiology, Yale University School of Medicine, New Haven, Connecticut. 6 Department of Surgery, Emory University, Atlanta, Georgia. 7 Department of Medicine, Louisiana State University Medical Center, Shreveport, Louisiana. 8 Department of Surgery, West Virginia University, Morgantown, West Virginia. 9 Department of Medicine, Tulane University, New Orleans, Louisiana. Presented in part at the American College of Chest Physicians Annual Meeting, New Orleans, Louisiana, October Supported by Denver Biomaterials, Denver, Colorado. Drs. Richard Light, R. Michael Rodriguez, and Joseph Putnam own shares of stock in Surgimedics, which is a parent company of Denver Biomaterials. Address for reprints: Richard W. Light, M.D., Director of Pulmonary Disease Program, Saint Thomas Hospital, P.O. Box 380, 4220 Harding Road, Nashville, TN Received June 7, 1999; accepted July 2, BACKGROUND. The purpose of this study was to compare the effectiveness and safety of a chronic indwelling pleural catheter with doxycycline pleurodesis via tube thoracostomy in the treatment of patients with recurrent symptomatic malignant pleural effusions (MPE). METHODS. In this multi-institutional study conducted between March 1994 and February 1997, 144 patients (61 men and 83 women) were randomized in a 2:1 distribution to either an indwelling pleural catheter or doxycycline pleurodesis. Patients receiving the indwelling catheter drained their effusions via vacuum bottles every other day or as needed for relief of dyspnea. RESULTS. The median hospitalization time was 1.0 day for the catheter group and 6.5 days for the doxycycline group. The degree of symptomatic improvement in dyspnea and the quality of life was comparable in each group. Six of 28 patients who received doxycycline (21%) had a late recurrence of pleural effusion, whereas 12 of 91 patients who had an indwelling catheter (13%) had a late recurrence of their effusions or a blockage of their catheter after the initially successful treatment (P 0.446). Of the 91 patients sent home with the pleural catheter, 42 (46%) achieved spontaneous pleurodesis at a median of 26.5 days. CONCLUSIONS. A chronic indwelling pleural catheter is an effective treatment for the management of patients with symptomatic, recurrent, malignant pleural effusions. When compared with doxycycline pleurodesis via tube thoracostomy, the pleural catheter requires a shorter hospitalization and can be placed and managed on an outpatient basis. Cancer 1999;86: American Cancer Society. KEYWORDS: malignant pleural effusion, pleurodesis, indwelling catheter. Management of the patient with a symptomatic malignant pleural effusion (MPE) remains problematic. The current management of MPE entails the production of a pleurodesis (fusion of the visceral and parietal pleura). This can be done by instilling a sclerosing agent through a chest tube or via talc insufflation at thoracoscopy. 1 These treatments require the insertion of a chest tube and several days of hospitalization. An alternative treatment is intermittent or continuous drainage of the pleural fluid with a chronic indwelling pleural catheter. There have been several other reports with small numbers of patients that demonstrated the utility of an indwelling drainage catheter in patients with MPE. 2 7 Based on these preliminary reports, a pleural catheter was developed that could be chronically implanted in the pleural space for the treatment of MPE. The purpose of this randomized study was to compare the effec American Cancer Society

2 Indwelling Catheter for Malignant Effusions/Putnam et al FIGURE 1. The Pleurx indwelling pleural catheter is shown. The valve on the distal end of the pleural catheter is closed except when the access tip of the drainage line is inserted. tiveness of the indwelling pleural catheter with tube thoracostomy and doxycycline pleurodesis in the management of patients with recurrent, symptomatic MPE. We hypothesized that use of the newly developed pleural catheter compared with doxycycline pleurodesis would decrease hospital stay and result in a comparable improvement in quality of life without affecting overall mortality. MATERIALS AND METHODS A prospective, randomized, multicenter study was conducted comparing the treatment of symptomatic, recurrent MPE with an indwelling pleural silicone catheter (Pleurx; Surgimedics, Denver Biomaterials, Denver, CO) to doxycycline pleurodesis via tube thoracostomy. Description of Indwelling Pleural Catheter Apparatus The apparatus for the indwelling pleural catheter consists of the catheter and a drainage line with an access tip matched to the catheter (Fig. 1). The pleural catheter is a 15.5 Fr. silicone rubber catheter, 66 cm in length, with fenestrations along the proximal 24 cm. On the distal end is a valve designed to enhance the safety of the product. The valve prevents fluid or air from passing in either direction through the catheter unless the catheter is accessed with the matched drainage line. The pleural fluid is drained by inserting the access tip of the drainage line into the valve of the catheter and then draining the fluid via an external tube into vacuum bottles. Radiologic Evaluation Pleural effusions were semiquantitated as follows: 1) no effusion, 2) effusions only detected on decubitus films, 3) blunting of the costophrenic angles, 4) moderate effusion fluid between the costophrenic angle and hilar level, 5) large effusion fluid above hilar level but not completely opacified, and 6) complete opacification. Patient Selection Patients were required to have a malignancy with at least a moderate sized pleural effusion and dyspnea relieved after therapeutic thoracentesis. Exclusion criteria included chylothorax, previous lobectomy or pneumonectomy on the affected side, previous attempts at pleurodesis, autoimmunodeficiency syndrome, Karnofsky performance status score 50, 8 bilateral moderate or larger pleural effusions, multiple loculations, mediastinal shift toward the side of the effusion, pleural infection, or abnormal coagulation profile. Study participants were not allowed to receive concurrent intrapleural chemotherapy or radiation therapy to the ipsilateral chest. Patients were allowed to receive systemic chemotherapy or mediastinal radiation. Study Design After a written informed consent was obtained, patients were randomized via consecutively numbered envelopes to receive the indwelling catheter or the doxycycline pleurodesis. A 2:1 distribution was used with the greater number of patients receiving the indwelling catheter. Each center was provided a set of consecutively numbered envelopes containing the treatment assignment that was generated from a table of random numbers. The patient s disability was assessed by using the

3 1994 CANCER November 15, 1999 / Volume 86 / Number 10 modified Borg scale score for dyspnea (scale, 0 10) at rest and after walking 100 feet on the level. 9 A qualityof-life questionnaire, the dyspnea component of Guyatt Chronic Respiratory Questionnaire (CRQ), also was administered. 10 Protocol for Catheter Patients The Seldinger technique 11 was used to insert the wire into the pleural effusion at approximately the anterior axillary line. A 1 2 cm incision was made over the wire. A chest wall tunnel (5 8 cm in length) was created with a counter incision. The catheter was pulled through the tunnel and out next to the wire. After dilation of the wire tract with a Teflon peel-away sheath, the indwelling catheter was inserted into the chest. The counter incision was closed primarily, and the catheter was secured to the skin medially with a suture. After catheter insertion, up to 1500 ml of pleural fluid initially were drained. If the effusion was not drained completely, as evaluated with the chest radiographs, then another 1000 ml of fluid were drained every 8 hours until drainage was complete. If the patient was found to have a trapped lung, then the catheter was left in place. The patient was considered for discharge after a chest radiograph demonstrated the absence of fluid. Prior to discharge, the patient s dyspnea was rerated using the Borg scale. The patient assessed the most severe and average levels of discomfort during the 24 hours after catheter insertion using a visual analog scale. Prior to discharge, the patients and/or their care givers were provided with detailed oral and written instruction for draining the pleural fluid. Patients were instructed to drain the pleural fluid completely every other day. Protocol for Doxycycline Pleurodesis Patients After tube thoracostomy (chest tube size was at the discretion of the investigator) was performed, the pleural fluid was drained. If the underlying lung failed to expand by 72 hours, then the patient was assumed to have a trapped lung, and chemical pleurodesis was not attempted. If the lung expanded and the drainage was 150 ml/24 hours, then doxycycline (500 mg in a total volume of 50 ml) was injected into the pleural space. If the chest tube had been in place for 4 days and the drainage was 300 ml over the previous 24 hours, then doxycycline was injected. If the drainage exceeded 300 ml during Day 4, then chemical pleurodesis was not attempted. After doxycycline injection, the chest tube was clamped for 4 hours, and the patient was rotated. Suction was then reapplied. The following day, the patient assessed the most severe and average levels of pain during the previous 24 hours using a visual analog scale. The chest tube was removed when the 24-hour drainage fell below 100 ml. If the 24-hour drainage volume did not fall below 100 ml within 4 days, then doxycycline was readministered. If the drainage remained above 100 ml/24 hours for 4 days after the second injection, then the patient was considered to have experienced treatment failure. All patients who were thought to be initial treatment successes had chest radiographs within 8 hours of chest tube removal and had their dyspnea reassessed. Follow-Up Procedures Follow-up clinic visits were scheduled at 4 weeks, 8 weeks, and 12 weeks. For the first 12 weeks, patients were called on a weekly interval if a clinic visit was not scheduled. After the initial 12 weeks, patients were called on every other week. Chest radiographs, interval history, and physical examination were obtained at each clinic visit. The patient s dyspnea was quantitated with a Borg score, and their quality of life was assessed by the Guyatt CRQ. Pleural catheter patients were followed until either death or catheter removal. Pleural catheter patients were advised to return for a follow-up visit sooner if there was no pleural fluid drainage on 3 consecutive occasions. If pleurodesis had occurred, then the catheter was removed. After discharge, the patients were said to have failed treatment if the effusion was large (Grade 5) or if the effusion was moderate (Grade 4) and the patient was symptomatic. Statistical Analysis The data are presented as the mean standard deviation when the data are distributed normally and as the median with the range when the data are not distributed normally. The characteristics of the patients in the two groups (pleural catheter or doxycycline pleurodesis) were compared by using unpaired t tests. If the data failed the normality test, then the results were compared by using the nonparametric Mann Whitney rank-sum test (SigmaStat; Jandel Scientific, San Rafael, CA). The chi-square test was used for statistical analysis when proportions in the treatment groups were compared. Differences in the treatment results were considered significant when P When the treatment results were analyzed, only the patients who received the appropriate treatment without protocol violation were included.

4 Indwelling Catheter for Malignant Effusions/Putnam et al TABLE 1 Comparison of Demographics of the Patients in the Two Treatment Groups Characteristic Pleural catheter Doxycycline pleurodesis No. of patients Female gender (%) Age (yrs) Karnofsky performance status No. of primary tumors (%) Lung 39 (39) 19 (42) Breast 25 (25) 14 (31) Adenocarcinoma (primary unknown) 5 (5) 0 Renal cell carcinoma 4 (4) 2 (4) Melanoma 6 (6) 0 Lymphoma leukemia (from other) 4 (4) 0 Other 16 (16) 10 (22) Initial size of effusion a Right-sided effusions (%) 60 of 99 (61) a 17 of 45 (39) Pleural fluid glucose (mg/dl) Pleural fluid protein (gm/dl) Pleural fluid LDH (IU/L) (median) 397 ( ) 390 ( ) Positive pleural fluid cytology (%) 68 of 84 (81) 36 of 39 (90) a P RESULTS One hundred forty-four patients were enrolled in the study at 11 institutions. Forty-five patients (31.2%) were randomized to the doxycycline arm, and 99 patients (68.8%) were randomized to the indwelling pleural catheter arm. The demographics of the patients assigned to the two treatment groups were similar (Table 1). The most common primary malignancy in each group was lung, and the second most common was breast (Table 1). The mean initial size of the effusions in the indwelling catheter group was significantly larger (P 0.031) than in the doxycycline pleurodesis group. The mean pleural fluid glucose and LDH levels and the percentage with positive cytology were similar in both groups. Doxycycline Pleurodesis Patients Two of the 45 patients who were randomized to the doxycycline pleurodesis group did not receive chest tubes. One patient was withdrawn from the study when his primary care physician elected to treat with radiotherapy rather than a chest tube. A second patient withdrew from the study after he learned that he was randomized to receive a chest tube rather than an indwelling catheter. Of the remaining 43 patients, 2 were withdrawn due to protocol violations. One patient was mistakenly given talc rather than doxycycline intrapleurally, whereas a second patient received a second administration of doxycycline before it was TABLE 2 Treatment Results from the Two Treatment Groups Characteristic Pleural catheter Doxycycline pleurodesis No. of patients Hospitalization days (median) a Fluid drained first 24 hours (ml) Recurrence of effusion postdischarge (%) 12 of 91 (13) 6 of 28 (21) Complications (%) Pleural infection 1 of 91 (1) 0 Cellulitis 6 of 91 (7) 0 Obstruction of catheter 2 of 91 (3) 0 Tumor seeding of catheter tract 3 of 91 (3) 0 a P compared with pleural catheter group. justified in the protocol. The initial treatment was unsuccessful in 13 of the remaining 41 patients (32%) for the following reasons: 300 ml of fluid/24 hours after 4 days of tube drainage (n 7 patients), incomplete drainage (n 3 patients), trapped lung (n 1 patient), recurrent large effusion immediately after chest tube removal (n 1 patient), and death 3 days after chest tube insertion (n 1 patient). The median chest tube size was 28 Fr (range, Fr). Indwelling Pleural Catheter Patients Three of the 99 patients who were randomized to the indwelling catheter group did not have the catheter inserted. In 2 patients, the radiographs were misinterpreted. The third patient had a loculated effusion. Two of 96 patients who received the indwelling catheter were withdrawn because of protocol violations: One patient had a concomitant chest tube, and the other had a chylothorax. In the remaining 94 patients, the chest radiograph revealed at most a small effusion in 91 patients (97%). A multiloculated effusion, a hemothorax, and a pleural infection each occurred in 1 patient. Hospitalization Times The median hospitalization time necessary for the treatment of pleural effusions was significantly less in the pleural catheter group (median, 1.0 day) than in the doxycycline pleurodesis group (median, 6.5 days; P 0.001) (Table 2). This time was the interval from randomization until the patients were eligible for discharge based on their response to treatment and its complications. The study was designed to have all indwelling catheter patients discharged after being observed for hours in the hospital after insertion of the catheter. Forty of the indwelling catheter patients were discharged within 24 hours. Hospitaliza-

5 1996 CANCER November 15, 1999 / Volume 86 / Number 10 TABLE 3 Comparison of Initial Borg and Guyatt Scores and Mean Improvements from the Initial Score in the Two Treatment Groups a Borg score rest Borg score exercise Guyatt CRQ Time group Catheter Doxycycline Catheter Doxycycline Catheter Doxycycline Initial (99) (44) (99) (45) (96) (44) Change after initial Rx (93) (30) (90) (28) Change at 30 days (62) (28) (60) b (26) (59) (27) Change at 60 days (49) (20) (46) (20) (44) (20) Change at 90 days (35) (21) (34) (20) (33) (21) CRQ: Chronic Respiratory Questionnaire. a The numbers in parentheses refer to the number of patients evaluated at each time period. All numbers are the mean standard deviation of differences from initial values. b P tions were prolonged in the other patients for therapy related to other conditions. Late Failures In the doxycycline group, late failure was defined as recurrence of the effusion after an initially successful pleurodesis, whereas, in the pleural catheter group, late failure was defined as the recurrence of the effusion after its initial successful control. The late failure rate was comparable in the doxycycline group (6 of 28 patients; 21%) and in the indwelling catheter group (12 of 91 patients; 13%) (chi-square 0.23; P 0.631). The 6 late failures in the doxycycline group were all documented by X-ray and by thoracentesis in four. All 6 recurrences occurred by 30 days. In the indwelling catheter group, the recurrence was due to loculations resulting in incomplete drainage of the pleural space in 7 patients. The catheter was repositioned in 1 of these patients, who subsequently developed a spontaneous pleurodesis. The recurrence occurred after a successful, spontaneous pleurodesis in 2 patients. In 2 patients, the catheter became occluded. It was replaced in 1 patient who subsequently developed a spontaneous pleurodesis. One patient had a documented recurrence at another medical center, and it was not clear whether the catheter was occluded. Spontaneous Pleurodesis Spontaneous pleurodesis occurred in 42 of the 91 patients (46%) who were treated successfully with the pleural catheter. The median time to pleurodesis was 29 days (range, days). The median amount of fluid drained in the first week as an outpatient was significantly less in the group that achieved pleurodesis than in the group that did not (460 ml vs ml; P 0.05). Overall, the median amount of fluid drained each week ranged from 500 ml to 1500 ml, with the maximum 4000 ml/week. Patients who achieved a pleurodesis tended to have a gradual diminution in the amount of fluid drained. Changes in Quality of Life The initial mean values for the resting and exercise Borg score and the dyspnea component of the Guyatt CRQ were similar among the two groups (Table 3). When the Borg scores were re-evaluated after the initial treatment was completed, the mean degree of improvement both at rest and after exercise were nearly identical among the two treatment groups. The Borg scores postexercise at 30 days, 60 days, and 90 days after the initial treatment showed a trend toward greater improvement in the indwelling catheter group, which was statistically significant at 30 days (P 0.05) (Table 3). The improvements in the Guyatt CRQ scores were similar 30 days, 60 days, and 90 days posttreatment in both groups. Concomitant Therapy At the time that the patients were enrolled in the study, 25 of the 135 evaluable patients were receiving chemotherapy. There was no significant relation between the results from the doxycycline group or the indwelling catheter group and whether or not the patients were receiving chemotherapy (Table 4). At 30 days postrandomization, an additional 3 patients in the doxycycline group and 13 patients in the indwelling catheter group were receiving chemotherapy. Again, there was no relation between chemotherapy and the results from the doxycycline group or the indwelling catheter group. There was no relation between the occurrence of a spontaneous pleurodesis and chemotherapy in the indwelling catheter group. Only 1 patient in the entire group received radiotherapy to the mediastinum or chest wall.

6 Indwelling Catheter for Malignant Effusions/Putnam et al TABLE 4 Relation between Chemotherapy at the Time of Initial Treatment and Results of Treatment Result Doxycycline pleurodesis chemotherapy Indwelling catheter chemotherapy Yes No Yes No Primary failure Secondary failure Long term success Total Morbidity The degree of pain experienced in the two groups was similar. The maximum and average degree of pain experienced in the first 24 hours after the procedure were and , respectively, in the doxycycline pleurodesis group and and , respectively, in the pleural catheter group. Almost all patients received opiates for analgesia during the first day, and there was no significant difference in the amount of analgesic received. Early (in-hospital) morbidity occurred in 6 of 43 patients who received a chest tube: fever (n 2 patients), severe pain requiring PCA (n 2 patients), hydropneumothorax (n 1 patient), and occluded chest tube requiring tube replacement (n 1). Early (in-hospital) morbidity occurred in 10 of 96 patients with the indwelling catheter: fever (n 3 patients), pneumothorax (n 3 patients), misplacement of catheter (n 2 patients), re-expansion pulmonary edema (n 1 patient), and hypercapnic respiratory failure secondary to over-sedation (n 1 patient). No procedure-related deaths occurred. In the 90-day follow-up period, there were several complications in the pleural catheter group. Three patients developed tumor seeding of the catheter tract that did not require therapy. Six patients developed local cellulitis around the catheter tract that responded to oral antibiotics and did not necessitate catheter removal. Pain during fluid drainage was reported by an additional 7 patients. The only complication reported in the doxycycline pleurodesis group was pain at the chest tube site in 1 individual. Survival The median survival was poor but similar for both groups. For the chest tube patients (n 35 patients), the median survival was 90 days, whereas, for the pleural catheter group (n 87 patients), the median survival was 87 days. DISCUSSION The current study demonstrates that patients with symptomatic, recurrent MPEs can be managed effectively with a chronic indwelling pleural catheter. The primary advantage of this method is decreased hospitalization time compared with that necessary for chemical pleurodesis via tube thoracostomy. There were more complications during follow-up in the pleural catheter group, but these complications tended to be minor. Because the life expectancy of patients with MPEs is short, efforts should be made to minimize the duration of their hospitalization. The median survival of 90 days in the current study is similar to that reported in previous studies. 12,13 Therefore, a reduction in the median hospitalization time from 6.5 days to 1.0 day, which occurred in the current study with the indwelling catheter group, increases the time that the patients are outpatients by 5%. One possible criticism of the current study is the manner in which the efficacy of the two procedures was compared. Our primary measure of efficacy was the percentage of patients who had a late recurrence of the effusion and an initially successful procedure. The incidence rate of late recurrence was 6 of 28 patients (21%) in the doxycycline group and 12 of 91 patients (13%) in the indwelling catheter group. There are two different measures that can be used. The first is the initial success rate, which compares the percentage of initially successful pleurodesis in the doxycycline group with the percentage of initially complete drainages in the indwelling catheter group. By using this comparison, the indwelling catheter appears superior, because, with this measure, the treatment was successful in 91 of 94 patients (97%) in the indwelling catheter group but in only 28 of 41 patients (68%) in the doxycycline pleurodesis group. We chose not to use this analysis, because the design of the study obviously was biased in favor of the indwelling catheter group. Patients in the doxycycline group who had a trapped lung or large amounts of fluid after the chest tube was placed would be considered to have experienced failure in the doxycycline group but not in the indwelling catheter group. The second measure by which the two procedures can be compared is long term control of the effusion without the presence of the catheter. By using this comparison, doxycycline pleurodesis appears superior, because it was effective in 22 of 41 patients (54%), whereas the indwelling catheter was effective in only 30 of 91 patients (33%). This analysis is biased against the indwelling catheter group, because a pleurodesis would have to occur for the treatment to be classified

7 1998 CANCER November 15, 1999 / Volume 86 / Number 10 as successful. We maintain that a treatment is successful as long as there is no pleural fluid, whether or not a catheter is present. There are several methodological weaknesses to the current study. First, the study was not blinded for obvious reasons. Accordingly, the investigators could have been biased in assessing patient symptoms, radiographs, or capability of being discharged. Second, the study was designed so that pleural catheter patients could be discharged within hours, whereas the doxycycline pleurodesis patients had to remain in the hospital for a minimum of 3 days. If the sclerosing agent had been injected as soon as the lung re-expanded, and if the chest tube had been removed 24 hours after the injection of the sclerosing agent, as proposed previously, 14 then there may have been less of a difference in days of hospitalization. However, other studies have shown that the median duration of hospitalization for pleurodesis with tube thoracostomy is 6 days The results of the current study for doxycycline pleurodesis tend to be slightly worse than those reported previously with tetracycline pleurodesis. In a review of pleurodesis for MPEs, treatment was successful in 70% of patients who were treated with intrapleural tetracycline derivatives, 18 whereas, in the current study, the treatment was successful in only 54% of patients. However, in another study, treatment was successful in only 33% of patients at 30 days. 13 At the present time, talc slurry is one of the most popular sclerosing agents, and it may be more effective than the tetracycline derivatives. 18 However, acute respiratory failure has resulted from its administration, 19 and the duration of hospitalization required is comparable to that for patients who are receiving tetracycline derivatives Doxycycline rather than talc slurry was chosen for the current study because of the concern about the acute respiratory distress syndrome with talc slurry and the limited availability of sterile talc at the time the study was initiated. The indwelling pleural catheter required less hospitalization time; however, what were the drawbacks to its use? Table 2 shows that 12 of the 91 patients (13%) with the indwelling catheter experienced a complication after discharge from the hospital compared with no complications in the doxycycline pleurodesis patients. However, the patients with cellulitis were treated successfully with oral antibiotics, and the patients with tumor seeding of the catheter tract did not require treatment. Only the patient with the pleural infection required hospitalization. These additional complications in the indwelling catheter patients seem minor compared with the extended hospital stay associated with doxycycline pleurodesis. Patients with the indwelling catheter also had to drain their pleural fluid at home until they developed a spontaneous pleurodesis. After the patients and/or their care givers were provided detailed oral and written instruction for draining the pleural fluid, most patients had no difficulty in following the protocol. Although the study coordinators were available at all times for questions, they rarely were contacted except when the catheter stopped draining. It is noteworthy that spontaneous pleurodesis occurred in nearly 50% of the patients treated with the indwelling pleural catheter by unknown mechanisms. When patients are treated with tube thoracostomy for several days, pleurodesis will develop in a significant proportion of patients. 20,21 In contrast, it is very uncommon for patients who are treated with serial therapeutic thoracenteses to develop a spontaneous pleurodesis. 22 We believe that two conditions are necessary to induce a spontaneous pleurodesis, namely, the pleural space must be completely drained, and there must be inflammation in the pleural space. Placement of a chest tube results in pleural inflammation. 23 When a patient is seen with a symptomatic recurrent pleural effusion, available treatment options include serial thoracentesis, thoracoscopy with chemical pleurodesis, tube thoracostomy with chemical pleurodesis, insertion of a pleuroperitoneal shunt, or insertion of an indwelling pleural catheter. Factors that should be considered when making this choice include cost, hospitalization time, symptom relief, convenience, patient acceptance, complications, and success rate. We believe that the indwelling pleural catheter is a good option for most patients with MPE primarily because of the reduced hospitalization time and the potential for outpatient use. In this study, a hospitalization of less than 1 day was required for the majority of patients treated with the indwelling catheter. In contrast, median hospitalizations exceeding 5 days are reported commonly for patients who received pleurodesis with either tube thoracostomy or thoracoscopy. 15 The initial success rate with the indwelling catheter exceeds 90%, and only 10% of the catheters fail when they have been successful initially. Patient acceptance is good, symptoms are relieved, and complications, for the most part, are minor. The major disadvantage of the indwelling catheter is the necessity for the repeated drainage of the pleural space at home should spontaneous pleurodesis not occur. Additional research is indicated to determine whether the injection of sclerosing agents through the indwelling catheter can be performed safely on an outpatient basis. Indwelling catheters have been used

8 Indwelling Catheter for Malignant Effusions/Putnam et al for repeated intrapleural injections of antineoplastic agents. 6,7 Based on this study we conclude that the indwelling pleural catheter is an alternative treatment to tube thoracostomy and chemical pleurodesis for the management of patients with symptomatic recurrent MPEs. The advantages of the indwelling pleural catheter are that it can be inserted on an outpatient basis, the initial success rate is high, and the associated complications, for the most part, are minor. REFERENCES 1. Light RW, Vargas FS. Pleural sclerosis for the treatment of pneumothorax and pleural effusion. Lung 1997;175: Zeldin DC, Rodriguez RM, Glassford DM, Cutrell G, Newman JH. Management of refractory MPEs with a chronic indwelling pleural catheter. Chest 1991;100:87S. 3. Leff RS, Eisenberg B, Baisen CE, Mosley KR, Messerschmidt GL. Drainage of recurrent pleural effusion via an implanted port and intrapleural catheter. Ann Intern Med 1986;104: Robinson RD, Fullerton DA, Albert JD, Sorensen J, Johnston MR. Use of pleural Tenckhoff catheter to palliate MPE. Ann Thorac Surg 1994;57: Grodzin CJ, Balk RA. Indwelling small pleural catheter needle thoracentesis in the management of large pleural effusions. Chest 1997;111: Driesen P, Boutin C, Viallat JR, Astoul PH, Vialette JP, Pasquier J. Implantable access system for prolonged intrapleural immunotherapy. Curr Respir J 1994;7: Patz EF Jr., McAdams HP, Goodman PC, Blackwell S, Crawford J. Ambulatory sclerotherapy for MPEs. Radiology 1996; 199: Karnofsky DA. Problems and pitfalls in the evaluation of anticancer drugs. Cancer 1965;18: Borg GAV. Psychophysical basis of perceived exertion. Med Sci Sports Exer 1982;14: Guyatt GH, Berman LB, Townsend M, Pugsley SO, Chambers LW. A measure of quality of life for clinical trials in chronic lung disease. Thorax 1987;42: Thal AP, Quick KL. A guided chest tube for safe thoracostomy. Surg Gynecol Obstet 1988;167: Chernow B, Sahn SA. Carcinomatous involvement of the pleura. An analysis of 96 patients. Am J Med 1977;63: Ruckdeschel JC, Moores D, Lee JY, Einhorn LH, Mandelbaum I, Koeller J, et al. Intrapleural therapy for MPEs. A randomized comparison of bleomycin and tetracycline. Chest 1991;100: Villanueva AG, Gray AW Jr., Shahian DM, Williamson WA, Beamis JF Jr. Efficacy of short term versus long term tube thoracostomy drainage before tetracycline pleurodesis in the treatment of MPEs. Thorax 1994;49: Heffner JE, Standerfer RJ, Torstveit J, Unruh L. Clinical efficacy of doxycycline for pleurodesis. Chest 1994;105: Yim AP, Chan AT, Lee TW, Wan IY, Ho JK. Thoracoscopic talc insufflation versus talc slurry for symptomatic MPE. Ann Thorac Surg 1996;62: Zimmer PW, Hill M, Casey K, Harvey E, Low DE. Prospective randomized trial of talc slurry vs. bleomycin in pleurodesis for symptomatic MPEs. Chest 1997;112: Walker Renard, Vaughan LM, Sahn SA. Chemical pleurodesis for MPEs. Ann Intern Med 1994;120: Kennedy L, Sahn SA. Talc pleurodesis for the treatment of pneumothorax and pleural effusion. Chest 1994;106: Groth G, Gatzemeier U, Haubingen K, Heckmayr M, Magnussen H, Neuhauss R, et al. Intrapleural palliative treatment of MPEs with mitoxantrone versus placebo (pleural tube alone). Ann Oncol 1991;2: Sorensen PG, Svendsen TL, Enk B. Treatment of MPE with drainage, with and without instillation of talc. Eur J Respir Dis 1984;65: Anderson CB, Philpott GW, Ferguson TB. The treatment of MPEs. Cancer 1974;33: Carvalho P, Kirk W, Butler J, Charan NB. Effects of tube thoracostomy on pleural fluid characteristics in sheep. J Appl Physiol 1993;106: