Ann Thorac Surg 2002;74:213-217
© 2002 The Society of Thoracic Surgeons
Original article: general thoracic
Early and late mortality after pleurodesis for malignant pleural effusion
Alain Bernard, MD*a,
Régis Bernard de Dompsure, MDa,
Olivier Hagry, MDa,
Jean Pierre Favre, MDa
a Chirurgie Thoracique, Hôpital Universitaire, Dijon, France
Accepted for publication March 18, 2002.
* Address reprint requests to Dr Bernard, Service de Chirurgie Thoracique, C.H.U. Hôpital du Bocage, Bd de Lattre de Tassigny, Cedex 21034 Dijon, France
e-mail: alain.bernard{at}chu-dijon.fr
 |
Abstract
|
|---|
Background. The purpose of this study is to analyze morbidity and mortality and to determine the relative contribution of each of these potential prognosis variables for predicting morbidity and mortality in patients after pleurodesis by thoracotomy or thoracoscopy.
Methods. Between March 1, 1996, and January 31, 2001, a total of 70 patients underwent pleurodesis for recurrent malignant pleural effusion. Thoracoscopy was performed in 54 patients (77%); pleurodesis was achieved by pleural abrasion (n = 15), pleurectomy (n = 5), and talc insufflation (n = 34). Thoracotomy was performed in 16 patients (23%) who also needed pleurectomy and decortication for a trapped lung.
Results. Postoperative complications occurred in 24 patients (34%). Factors adversely affecting morbidity with univariate analysis included: three or four metastatic sites (p = 0.003), and thoracotomy (p = 0.009). Factors adversely affecting morbidity with multivariate analysis included: thoracotomy (p = 0.0005) and number of metastatic sites (p = 0.007). Six patient deaths (8.6%) occurred during hospitalization. Factors adversely affecting in-hospital mortality with univariate analysis included: Eastern Cooperative Oncology Group Performance Status 2 to 3 (p = 0.001), lower preoperative serum hemoglobin (p = 0.001), lower preoperative serum albumin (p = 0.0001), and thoracotomy (p = 0.03). Factors adversely affecting in-hospital mortality with multivariate analysis included: preoperative serum albumin less than 60 g/L (p = 0.007) and ECOG Performance Status 2 to 3 (p = 0.008). Twelve patients (17%) died within 90 days after surgery. Factors adversely affecting 3-month mortality with univariate analysis included: ECOG Performance Status 2 to 3 (p = 0.001), lower preoperative serum hemoglobin (p = 0.03), higher preoperative white cells (p = 0.03), lower preoperative serum albumin (p = 0.03), and preoperative thoracentesis more than once per month (p = 0.03). Factors adversely affecting 3-month mortality with multivariate analysis included: ECOG Performance Status 2 to 3 (p = 0.01), preoperative thoracentesis more than once per month (p = 0.03), three or four metastatic sites (p = 0.02), and preoperative white blood cell count 
12,000/mm3 (p = 0.03).
Conclusions. Thoracotomy is not indicated in patients with a malignant effusion because of poor survival, a high frequency of complications, and prolonged hospital stay. Pleurodesis thoracoscopy is indicated in patients with good performance status coupled with good nutrition.
|
Introduction
|
|---|
In patients who develop progressive pleural effusions producing dyspnea and cough, quality of life is affected. Patients with symptomatic effusions may benefit from pleurodesis to relieve dyspnea and to prevent the reaccumulation of pleural fluid. These procedures are usually palliative and are performed depending on the patient’s symptoms, underlying medical conditions, extent disease, performance status, and prognosis. Dyspnea can be improved by pleural fluid drainage followed by a pleurodesis procedure, thereby preventing pleural fluid reaccumulation. Pleurodesis includes mechanical irritation of the pleural membranes or intrapleural instillation of chemical agents to induce pleural inflammation. Unfortunately, pleurodesis attempts fail in 10% to 40% of patients with recurrence of pleural fluid and dyspnea [1] according to type of procedure. Patients with malignant pleural effusion generally survive only a few months. The cause of death is related to the advanced nature of the patient’s underlying disease. Considering discomfort, inconvenience, and potential risk of pleurodesis, patients having a poor short-term survival can be treated with serial thoracentesis or other palliative measures to control dyspnea. For most patients, the method controlling symptomatic malignant pleural effusion that is associated with the least morbidity is chest tube drainage with the instillation of a sclerosing agent [1–3]. However, the literature reports the rate of failure ranging from 7% to 46% for these procedures [2, 3]. Thoracoscopy pleurodesis may be used to improve success. Thoracoscopy allows optimal preparation of the pleural surface and homogeneous distribution of the talc under visual control, maximizing the chances for complete pleurodesis. However, thoracoscopy is potentially associated with higher morbidity in debilitated patients with a short expected survival.
The purpose of this study is to analyze morbidity and mortality and to determine the relative contribution of each of these potential prognosis variables for predicting morbidity and mortality in patients after undergoing pleurodesis by thoracotomy or thoracoscopy.
|
Material and methods
|
|---|
Between March 1, 1996 and January 31, 2001, a total of 70 patients underwent pleurodesis for recurrent malignant pleural effusion. The characteristics of patients are shown in Table 1.
Comorbidity diseases included coronary artery disease in 5, thrombosis in 7, diabetes mellitus in 6, other cancer in 7, embolism in 3, high blood pressure in 2, chronic obstructive pulmonary disease in 1, and alcoholism in 1. There was one metastatic site (pleura only) in 37 patients, two metastatic sites (pleura and one extrathoracic site) in 23 patients, and three or four metastatic sites (pleura and two or three extrathoracic sites) in 9 patients (Table 1).
Thoracoscopy was performed in 54 patients (77%). Pleurodesis was achieved by pleural abrasion (n = 15), pleurectomy (n = 5), and talc insufflation (n = 34).
Thoracotomy was indicated in 16 patients (23%) who had a collapsed unexpandable lung. These patients who had a trapped lung were not candidates for pleurodesis thoracoscopy. The cause of a chronic collapse of lung was related to a peel of visceral pleural tumor or a chronic effusion that created a limited fibrothorax. A trapped lung was readily recognized by a lack of expansion of the lung under thoracoscopy. In these patients, a thoracotomy was required to perform a decortication with pleurectomy.
Thoracoscopy was performed under general anesthesia in all patients. The patient was placed in a supine position on the operating table, and the arm was abducted 90 degrees. A 10-mm camera port was introduced at the fourth intercostal space in the mid axillary line. A second port was inserted at the fifth intercostal space in the anterior axillary line. Pleural effusions were drained, and a quantity of 3 to 6 g of talc powder was then insufflated intrapleurally. One chest tube was positioned. For pleural abrasion or pleurectomy under thoracoscopy, the patient was placed in a lateral position as in posterolateral thoracotomy. A 10-mm camera port was introduced at the fifth intercostal space behind the scapula. Two other ports were introduced to drain pleural effusion and then to perform pleural abrasion or pleurectomy. One chest tube was positioned.
Morbidity included all complications occuring during hospitalization only. In-hospital mortality included both those patients who died within first 30 days and those who died later but during the same hospitalization. Three-month mortality included those patients who died within 90 days after surgery. The failure of pleurodesis was defined by a need for repeat thoracentesis or tube thoracostomy to drain a recurrent pleural effusion during the 3 months after pleurodesis.
The primary endpoints of analysis were morbidity and mortality. The effects of risk factors on these endpoints were evaluated with both univariate and multivariate analysis. Risk factors represented by continuous variables were assessed using two-sample t tests [4] when the data were approximately normal and by using rank sum tests [5] when the data were found to be not sufficiently gaussian. The effects of categorical variable risk factors were evaluated using 
2 tests. Multiple logistic regression was used during multivariate analysis to simultaneously evaluate the effects of risk factors. Specifically, stepwise forward selection was used to identify significant predictors. All analyses were conducted using BMDP Statistical Software (University of California, CA) [6].
|
Results
|
|---|
Postoperative complications occurred in 24 patients (34%). These included prolonged air leak more than 7 days in 16 patients, persistent pleural space in 5 patients, acute respiratory failure in 4 patients (mechanical support > 24 hours), pneumonia in 3 patients, congestive heart failure in 2, dyspnea in 2, and wound infection in 1 patient. Some patients had more than one complication. Factors adversely affecting morbidity with univariate analysis (Table 2)
included three or four metastatic sites (p = 0.003), and thoracotomy (p = 0.009). Two variables were close to the threshold of significance: weight loss of 5 kg or more (p = 0.06) and delay between diagnosis of pleural effusion and pleurodesis (p = 0.07). Continuous variables not affecting morbidity with univariate analysis included age, serum hemoglobin, white blood cell count, serum albumin, and creatinine. Factors adversely affecting morbidity with multivariate analysis (Table 3) included thoracotomy (p = 0.0005) and number of metastatic sites (p = 0.007). A prolonged air leak more than 7 days for thoracotomy was 50% (n = 8) and for thoracoscopy was 15% (n = 8) (p < 0.003). A persistent pleural space for thoracotomy was 19% versus 4% for thoracoscopy (p < 0.04). The average duration of chest tube was 7 days (range 1 to 30 days). The average duration of postoperative hospitalization was 8.8 days (range 1 to 41 days). The average duration of postoperative hospitalization was 6 days (range 2 to 25 days) in the patients without postoperative complications versus 14 days (range 1 to 41 days) in the patients with postoperative complications (p = 0.00001).
View this table:
[in this window]
[in a new window]
|
Table 3. Variables Remaining by Multivariate Analysis and Independently Associated With Postoperative Complications
|
|
Six patient deaths (8.6%) occurred during hospitalization. The cause of death was pneumonia in 2 patients, congestive heart failure in 2, and acute respiratory failure in 2 (mechanical support > 24 hours). Factors adversely affecting in-hospital mortality with univariate analysis (Tables 4 and 5)
included ECOG Performance Status 2 to 3 (p = 0.001), lower preoperative serum hemoglobin (p = 0.001), lower preoperative serum albumin (p = 0.0001), and thoracotomy (p = 0.03). Factors adversely affecting in-hospital mortality with multivariate analysis (Table 6) included preoperative serum albumin less than 60 g/L (p = 0.007) and ECOG Performance Status 2 to 3 (p = 0.008). Twelve patients (17%) died within 90 days after surgery. Factors adversely affecting 3-month mortality with univariate analysis (see Tables 4 and 5) included ECOG Performance Status 2 to 3 (p = 0.001), lower preoperative serum hemoglobin (p = 0.03), higher preoperative white cells (p = 0.03), lower preoperative serum albumin (p = 0.03), and preoperative thoracentesis more than once per month (p = 0.03). Factors adversely affecting 3-month mortality with multivariate analysis (Table 6) included ECOG Performance Status 2 to 3 (p = 0.01), preoperative thoracentesis more than once per month (p = 0.03), three or four metastatic sites (p = 0.02), and preoperative white blood cell count of 12,000/mm3 or more (p = 0.03).
View this table:
[in this window]
[in a new window]
|
Table 4. Categorical Variables Associated With In-Hospital Mortality and 3-Month Mortality in Univariate Analysis
|
|
View this table:
[in this window]
[in a new window]
|
Table 6. Variables Remaining by Multivariate Analysis and Independently Associated With In-Hospital Mortality and 3-Month Mortality
|
|
A total of 64 patients were followed for at least 3 months after pleurodesis. The failure rate within 3 months after pleurodesis was 6% (n = 4). All patients with a recurrent effusion underwent operation with thoracoscopy. The failure rate was 7% after pleural abrasion (1/14), 5.5% after pleurectomy (1/18), and 6% after talc pleurodesis (2/32).
|
Comment
|
|---|
Prolonged air leak was the most frequent postoperative complication after thoracotomy in this study. In these patients, a decortication was necessary in conjunction with the pleurectomy because the lung was trapped. To avoid this complication, which increased the duration of hospitalization in patients who had a short expected survival rate, our results suggest that thoracotomy is not indicated in patients with a malignant effusion. Some investigators have reported techniques using a pleuroperitoneal catheter or chronic indwelling pleural catheter, which are alternatives to thoracotomy [7, 8].
Our in-hospital mortality rate is higher than that reported for other series [7, 9, 10], rising from 0% to 2.8%. It is difficult to compare these data with the other series; the characteristics of the patients are different as well as the procedure of pleurodesis. In particular, we included the patients who underwent operation through thoracotomy for a trapped lung. As with other series [11], poor performance status and lower serum albumin increased in-hospital mortality. The mortality rate of 17% within the 3 months after the pleurodesis is keeping with the severity of disease responsible for the pleural effusion, as shown in the metaanalysis of Heffner and colleagues [12]. Repetitive thoracentesis before pleurodesis in our patients had an adverse effect on the 3-month mortality rate. It is a credible possibility that aggressive diseases are responsible for a rapid and plentiful recurrence of the pleural effusion. Sahn and Good [2] showed that this type of pleural effusion correlated with a pH of 7.28 or less or with a lower glucose concentration. These pleural fluid characteristics were not examined in our study. However, Heffner and colleagues [3] showed that the value of the PH has insufficient predictive accuracy for selecting patients for pleurodesis on the basis of estimated survival.
Performance status of the patient which is a prognosis factor, must be used to select the patients for pleurodesis. We used the ECOG Performance Status; other investigators [11] have applied the Karnofsky performance scale, which is the significant variable with multivariate analysis.
The patients with multiple metastatic sites have a high risk of dying within 3 months, dissuading a pleurodesis under general anesthesia. Heffner and colleagues [12] showed that the survival duration was influenced by the primary neoplasm. This observation was not made during our study.
Although we did demonstrate that a white blood cell count of 12,000/mm3 or more was associated with increased 3-month mortality, we have no explanation for this observation. Increased white blood cell count, however, could be due to the invasion of bone marrow by primary neoplasm.
The failure rate of the pleurodesis is comparable to the data of the literature [7, 9, 10, 13]. We believe that the quality of the pleurodesis is better when it is performed by thoracoscopy and under general anesthetia. The failure rate of other studies rise from 0% to 7% [7, 9, 10, 13] after thoracoscopy. When the authors perform an instillation of talc through thoracostomy tube, the failure rate goes from 11% to 24% [3]. Among the pleurodesis agents, the talc certainly gives the most effective results [2, 14]. The failures of pleurodesis are directly influenced by the biochemical characteristics of the pleural fluid [3]. A decision threshold for each test was a pH of 7.28 or less, a glucose level of 72 mg/dL or less, and serum lactate dehydrogenase of more than 146 [3]. Finally, a pleurodesis with talc that is performed by thoracoscopy under general anesthetic offers selected patients the best chance of success with the lowest morbidity and mortality.
In conclusion, this study has demonstrated that thoracotomy is not indicated in patients with a malignant effusion because of poor survival, a high frequency of complications, and prolonged hospital stay. Thoracoscopy pleurodesis was indicated in patients with good performance status coupled with good nutrition. Caution is advised in performing pleurodesis in patients with multiple metastatic sites.
|
References
|
|---|
- Sahn S.A. Malignant pleural effusions. Clin Chest Med 1985;6:113-117.[Medline]
- Sahn S.A., Good J.T. Pleural fluid pH in malignant effusion: diagnostic, prognostic and therapeutic implications. Ann Intern Med 1988;108:345-347.
- Heffner J.E., Niertert P.J., Barbieri C. Pleural fluid pH as predictor of pleurodesis failure. Analysis of primary data. Chest 2000;117:87-95.[Abstract/Free Full Text]
- Snedecor G.W., Cochran W.G. Statistical methods, 7th ed. Ames, Iowa: Iowa State University Press, 1980:75-78.
- Mehta C.R., Patel N.R. A network algorithm for performing Fisher’s exact test in rXc contingency tables. J Am Statist Assoc 1983;78:427-434.
- Dixon W.J. . BMDP Statistical Software. Los Angeles: University of California Press, 1990:739-807.
- Schulze M., Boehle A.S., Kurdow R., Dohrmann P., Henne-Bruns D. Effective treatment of malignant pleural effusion by minimal invasive thoracic surgery: thoracoscopic talc pleurodesis and pleuroperitoneal shunts in 101 patients. Ann Thorac Surg 2001;71:1809-1812.[Abstract/Free Full Text]
- Putnam J.B., Walsh G.L., Swisher S.G., et al. Outpatient management of malignant effusion by chronic indwelling pleural catheter. Ann Thorac Surg 2000;69:369-375.[Abstract/Free Full Text]
- Benko I., Molnar T.F., Horvath O.P. Palliative treatment of malignant pleural effusions by video-assisted thoracoscopic surgery. Acta Chir Hung 1999;38:131-133.[Medline]
- Furedi A., Keekes L., Gether P., Kiss B. Video-assisted thoracoscopic pleurodesis for malignant pleural effusions. Acta Chir Hung 1999;38:155-157.[Medline]
- Burrow C.M., Mathews C., Colt H.G. Predicting survival in patients with recurrent symptomatic malignant pleural effusions. An assessment of the prognostic values of physiologic, morphologic and quality of life measures of extent of disease. Chest 2000;117:73-78.[Abstract/Free Full Text]
- Heffner J.E., Nietert P.J., Barbieri C. Pleural fluid pH as a predictor of survival for patients with malignant effusions. Chest 2000;117:79-86.[Abstract/Free Full Text]
- Milanez de campos J.R., Vargas F.S., de Campos Werebe E., et al. Thoracoscopy talc poudrage. A 15-year experience. Chest 2001;119:801-806.[Abstract/Free Full Text]
- Diacon A.H., Wyser C., Bolliger C.T., et al. Prospective randomized comparison of thoracoscopic talc poudrage under local anesthesia versus bleomycin instillation for pleurodesis in malignant pleural effusions. Am J Respir Crit Care Med 2000;162:1445-1449.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
R. A Qureshi, S. L Collinson, R. J Powell, P. O Froeschle, and R. G Berrisford
Management of Malignant Pleural Effusion Associated with Trapped Lung Syndrome
Asian Cardiovasc Thorac Ann,
April 1, 2008;
16(2):
120 - 123.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
F. Rodriguez-Panadero, J. P. Janssen, and P. Astoul
Thoracoscopy: general overview and place in the diagnosis and management of pleural effusion.
Eur. Respir. J.,
August 1, 2006;
28(2):
409 - 422.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
