Pleural disease, especially malignant pleural effusion, is an increasing clinical problem throughout the world. Both the incidence of lung cancer in developed countries and the incidence of other cancers worldwide are rising. The annual incidence of malignant pleural effusion in developed countries has been predicted to be more than 500,000 new cases.1–4 Of these cases, nearly 60% will be diagnosed by cytologic investigation of pleural aspirate.5 The remaining 200,000 cases may require more invasive diagnostic procedures. Furthermore, the incidence of mesothelioma is rising.6
Medical thoracoscopy (MT) for cases of exudative pleural effusion without a diagnosis by either clinical, radiologic, laboratory, or cytologic investigation is performed as an established diagnostic method for the diagnosis of pleural diseases. In addition, MT can be performed for therapeutic as well as diagnostic purposes.7–9 As a result, the use of MT for pleural effusions is likely to rise in the coming years.1,9
Efficiency, complications, and contraindications of MT should be discussed. In particular, the reliability, safety, and costs for normal-risk and high-risk groups of patients should be established.
The aims of the present study were to establish the safety of MT in a group of patients at high-risk for complications and to determine the factors that affect the development of complications.
PATIENTS AND METHODS
This study was conducted in the Chest Diseases Department of the Medical Faculty of Eskisehir Osmangazi University from January 2002 to January 2009, which is a high-volume center for MT. The study was approved by the Ethical Committee of Eskisehir Osmangazi University Medical Faculty (2006-06-07 and 2006/387; PR-10-09-23-11; 2010/246). A trial registration was also developed (Clinicaltrials.gov Identifier: NCT01193439).
In the year of 2002, a database for the pleural diseases was created, which would be prospectively filled in the department. For cases with pleural diseases, details regarding the epidemiological characteristics, physical, laboratory, and radiologic findings, diagnostic methods, diagnosis, medications, outcome features, and characteristics of follow-up were recorded in this database.
Patients and Follow-up
Consecutive patients with the following criteria were admitted to participate in the study: evidence of exudative pleural effusion for which a specific diagnosis could not be determined by cytologic, microbiological, or clinical examinations and willingness to participate in the study and undergo MT. Exclusion criteria were as follows: patients under the age of 18 years, pleural thickening or a pleural-based mass without pleural effusion as determined by radiologic investigation, and the presence of clinically or microbiologically diagnosed parapneumonic effusion that did not need bronchopleural fistula treatment or drainage. Patients were thoroughly informed before thoracoscopy, and all patients provided written informed consent.
Before MT, patients were grouped into 2 subgroups: patients at high risk for complications of MT (high-risk patients) and patients with no risk for complications (normal-risk patients). After MT, as mentioned above, thoracoscopic findings, histopathologic results, microbiological results, early and late complications of MT, other diagnostic or therapeutic procedures, and the treatment outcome of each patient were recorded in the database. All patients were followed up for at least 24 months to track the late complications of MT.
Complications established in this study were grouped as early and late according to their time of onset after the procedure. If a complication was observed within 7 days after the procedure, then it was accepted as an early complication.10 Early complications were classified as transient attack of hypertension or hypotension, air leaking for >1 day or a prolonged air leak of over 5 days, subcutaneous emphysema or mediastinal emphysema, pain, cutaneous infection localized at the entry site, pleural infection arising <7 days, arrhythmias, hemorrhage of >20 mL, an uncomfortable cough, fever, and transient attack of hypoxemia, which improves with oxygen inhalation during procedure.10,11 Late complications was identified as a local tumoral invasion through the entry site and empyema.10 It is well known that pleurodesis with talc increases the risks of complications by itself. To discuss this specific point, we investigated the complications considering whether to perform talc pleurodesis or not.
Risk factors were determined by considering the clinical states that are classified as relative contraindications or factors that adversely affect the outcome of thoracoscopy in the previous publications.1,9,12 Accordingly, risk factors contributing to the high risk of developing complications were identified as advanced age (>80 y), cardiac valvular disease or cardiomyopathy/heart failure (ejection fraction<45%), bilateral pleural effusion and/or pericardial effusion, hypoxemic (PaO2: 50 to 60 mm Hg) and/or hypercapnic respiratory failure (PCO2: 45 to 60 mm Hg), diabetes mellitus regulated with insulin, low Karnofsky Performance Status (KPS<80), the presence of deep-vein thrombosis, recent (within the previous week) pulmonary embolism (PE), and renal failure (creatinine>2 g/dL). In the cases who have bilateral pleural effusion, MT was performed to the side having large pleural effusion.
MT was conducted with a rigid thoracoscope (Karl Storz, Tuttlingen, Germany) under mild sedation and local anesthesia. The outer size of trocar was 11 mm. Patients received atropine and morphine sulfate intramuscularly 45 minutes before the procedure. Midazolam (1 mg) was administered to the patients at the beginning of the procedure. During the procedure, midazolam was administered intermittently to a total dose of 5 mg in proportion to the reactions of the patient. During the entire procedure, cardiac rhythm, arterial blood pressure, and oxygen saturation were monitored. The monitorization time was extended up to 6 hours after the procedure at bedside.
Biopsy samples were taken for histopathologic investigation and, if necessary, for microbiological investigation as previously defined.13
For cases in which MT indicated pleurodesis, the pleural cavity was examined after the thoracoscope was inserted. Fluid was completely drained, and talc pleurodesis was conducted with the talc insufflation method.14 When MT was used for diagnostic purposes, pleurodesis was also conducted with talc insufflation during the procedure if appearances suggested underlying malignancy. In uncertain cases, a chest tube was left in situ, and the procedure was completed. In cases in which histopathologic evaluation resulted in a malignant diagnosis, talc pleurodesis was carried out through the chest tube with a talc slurry instillation.15 A sterile 5-g dry talc powder (Talc; Laborsan, Eskisehir, Turkey) was utilized in both the methods. During the study period, the talc had mixed particle size.
Suction was not applied routinely after thoracoscopy. The chest tube was directly connected to an underwater drainage bottle and allowed to self-drain.
We used thoracoscopy in the patients with mild hypoxemia (PaO2: 50 to 60 mm Hg) and/or hypercapnia (PaCO2: 45 to 60 mm Hg). The procedure was conducted as quickly as possible, and a chest tube and the underwater sealed drainage bottle were available for use if the procedure was terminated at any time. Controlled oxygen was given to hypoxemic patients and during the procedure, patients were laid partially on their backs with a chest-head angle of 30 degrees from the horizontal.
SPSS version 12.0 software (SPSS Inc., Chicago, IL) was used to perform the statistical analysis. Statistical significance for intergroup differences was assessed by the χ2 test (the Yates adjusted or the Fisher exact test) for categorical variables and the Student t test for continuous variables (eg, age). Complications were considered as primary outcome variables. To evaluate the relationships between variables (age, sex, indication to MT, coexisting disease, application of talc pleurodesis, and the presence of risk factors) and complications, multivariable-adjusted logistic regression models were applied, in which the contribution of each individual component of some of the variables was also analyzed. Finally, separate exploratory analyses were performed to evaluate the influence of complications. For all regression models, adjustment was made based on age, sex, existence of malignancy, presence of risk factors, and application of talc pleurodesis. A P value of <0.05 was considered as significant.
The total number of patients included in the study was 355. There were 209 (58.9%) men and 146 (41.1%) women. Of the 355 patients, 86 (24.2%) had 1 risk factor, 18 (5.1%) had 2 risk factors, and 2 (0.6%) had 3 risk factors. These 106 patients (29.9%) had one or more risk factors that placed them in the group of high-risk patients for complications of MT, whereas the remaining 249 (70.1%) had no risk factors, placing them in the group of the normal-risk patients (Table 1).
The distribution of age, sex, indications for MT, and diagnoses of patients according to groups are shown in Table 2. A difference was not observed between the 2 groups in terms of indications. Mesothelioma was more common in normal-risk patients in our population, and metastatic pleural diseases were prevalent in high-risk patients. However, when malignant patients were examined together, the ratios were similar in both the groups.
Because the number of cases was limited, the diagnosis distribution of 61 patients who were not included in this table was as follows: the final diagnosis for 2 patients was Meigs syndrome and 1 was chylothorax; 7 patients were diagnosed with rheumatoid arthritis–related pleural effusion; 5 were diagnosed with parapneumonic pleural effusion; 2 were diagnosed with hemothorax; 20 patients were diagnosed with various benign pleural diseases, including cardiac pleural effusion, renal failure, viral pleuritis, radiotherapy-related pleural effusion, hepatic pleural effusion, and PE; 5 patients were not definitively diagnosed because of spontaneous healing, these patients were classified as unidentified pleural effusion; 11 cases were followed up; and 8 patients could not be followed up, and we excluded them because of a lack of diagnosis. Talc pleurodesis was performed in 181 patients.
The distribution of complications based on the presence of risk factors for the development of complications is given in Table 3.
In subjects with early complications, the most common complication of the normal-risk patients was subcutaneous emphysema (10.3%), and the second most common was transient hypertension (6.4%). Pain (12.3%) and subcutaneous emphysema (7.7%) were the most frequent complications in the high-risk patients. Subcutaneous emphysema and pain were also the most frequent complications in all patients (9.4% and 6.8%). For both groups, only pain was observed more frequently in the high-risk patients compared with the normal-risk patients. For late complications, there was no significant difference between high-risk patients and normal-risk patients. Local tumoral invasion through the entry site was the most frequent complication for the patients with malignant pleural diseases (23/216;10.6%).
Talc pleurodesis was associated with prolonged air leak (8.3% vs. 2.3%; P=0.013) and empyema (5.0% vs. 0.6%; P=0.020).
Taking into consideration the patient characteristics, the presence of risk factors, and the pleurodesis procedure, the effects of these features on the development of early and late complications were examined. These findings are given in Table 4.
According to the data given in Table 4, the late complications of MT were more prevalent in patients who underwent MT for the purpose of pleurodesis and treatment of bronchopleural fistula and in patients with mesothelioma. In addition, talc pleurodesis also increases the rate of late complication. In our study, empyema was more commonly observed in patients who underwent talc pleurodesis. Local tumor invasion was more commonly observed in patients diagnosed with mesothelioma.
The influence of the variables mentioned in Table 4 that were found to be significant in univariate analysis was examined. To this end, 2 different sets of multivariable analyses were performed for both early and late complications.
In Table 5, multivariable models identified mesothelioma as independently associated with late complications. Unadjusted logistic regression detected an association between talc pleurodesis and the development of late complications, which was no longer significant after adjustment for age, sex, presence of malignancy, and existence of risk factors in multivariable adjustment.
For 355 patients who underwent MT, the rate of patients with no complication was 208 (58.6%). This rate accounted for 59.4% of the normal-risk patients (148/249) and 56.6% (60/106) of the high-risk patients. A significant difference was not observed (χ2=0.206; P=0.650). Of the 147 patients who experienced any complication, 19 (5.4%) had 1 complication, 114 (32.1%) had 2 complications, and 14 (3.9%) had 3 complications.
Among the normal-risk patients, 1 patient who experienced extensive subcutaneous and mediastinal emphysema died 8 days after thoracoscopy. Therefore, the mortality rate was 0.28% (1/355) for this patient series.
In cases with exudative pleural effusion without a diagnosis by clinical, radiologic, laboratory, cytologic investigation or closed needle biopsy, MT is the method of choice.7,9,14,16 Because of both the advantages of MT in terms of diagnosis and treatment with the possibility of lower side effects and the gradually increasing incidence of mesothelioma and malignant pleural disease, the use of MT has been increasing in pulmonary clinics in recent years. In a recent comprehensive review, the authors stated that, in 1999, 11 centers across the United Kingdom offered a local anesthetic thoracoscopy service, which increased to 17 centers in May 2004 and 37 centers in 2009.1 A similar trend can be observed throughout the world.17
Therefore, the spectrum of patients who will undergo MT in the coming years will expand. Most of these patients will have malignant pleural effusion.1,7,14 Because malignant pleural effusion will point to advanced disease, the use of MT will increase in patients with more deteriorated general conditions. Although many authors claim that MT is a minimally invasive procedure with a low complication rate,7,8,18 its safety in patients at high risk for complications has not been sufficiently examined.
The most common complications of MT cited in the literature are subcutaneous emphysema (0.6% to 4.9%), prolonged air leakage (0.5% to 8.1%), air embolism (0.2%), PE (2.7%), acute pain (10.5%), oversedation (1.4%), empyema (0.5% to 2.7%), wound infection (3%), hemorrhage (significant; 0.3% to 0.4%), hypotension requiring treatment (0.2%), cardiac arrhythmias (2%), chest wall seeding by malignancy (0.5% to 4%), reexpansion pulmonary edema (2.2%), respiratory failure (1.3%), acute respiratory distress (0.2%), and postoperative fever (2% to 6%).8,9,19
In our study, the most common complication for the normal-risk patients was subcutaneous emphysema (10.3%), and the second most common was transient hypertension (6.4%). Pain (12.3%) and subcutaneous emphysema (7.7%) were the most frequent complications in the high-risk patients. Pain possibly resulted from avoiding using opiate in high doses in the group with high risk because the number of patients with borderline respiratory failure was higher in this group. Suction is not applied routinely after MT in our clinic. That may be the reason for the high subcutaneous emphysema rate. From the results, we can easily say that these findings confirmed that MT is a safe procedure with a low complication rate.
In comprehensive studies,7,9,11,13,14 authors pointed out that because MT is performed under conscious sedation and local anesthesia in a spontaneously breathing patient with partial or near-total lung collapse, the patients must not have severe hypoxemia unrelated to pleural effusion or require ventilatory support. We used MT in the patients with mild hypoxemia (PaO2: 50 to 60 mm Hg) and/or hypercapnia (PaCO2: 45 to 60 mm Hg), and we did not observe any serious side effect in this group of patients. None of the patients required early termination of the procedure.
When there is a contralateral pleural involvement, moderate-size or massive-size pleural effusion, or moderate-size pericardial effusion, MT is not advisable. However, after draining of the contralateral pleural or pericardial space, MT could be undertaken safely. We think that a moderate decrease in ejection fraction is not a contraindication for MT.
In addition, previous researchers found refractory cough to be a relative contraindication for MT and recommended that treatment be considered before a procedure is scheduled. However, prilocaine inhalation administered by a nebulizer during the procedure decreased the coughing, and the procedure was continued comfortably after coughing was stopped.
After the assessment of the distribution of complications in high-risk and normal-risk patient groups, we wanted to determine the factors that might affect complication development in patients who underwent MT with univariate and multivariate analysis (Tables 4, 5). In the univariate analysis, development of late complications was significantly high in patients who underwent MT for pleurodesis indication, for the purpose of bronchopleural fistula treatment, in patients diagnosed with mesothelioma, and patients who underwent talc pleurodesis (Table 4). No effect of any of the investigated variables on the development of early complications in the multivariate analysis was found, and only mesothelioma diagnosis was found to have an effect on the development of late complications. Mesothelioma accounted for the majority of our malignant patients; tumor spread to the chest wall is a characteristic of mesothelioma tumor behavior,20,21 and it can be caused by needle biopsy, thoracoscopy, or thoracotomy on sites where invasive procedure has been performed over the chest wall.20 Chest wall seeding in mesothelioma patients constitutes a problem not only for MT but also for other invasive procedures, and it should not be considered a complication that would prevent MT from being performed or that would increase the risk.
Combined data from 47 studies indicated that death occurred in 16/4736 cases (0.34%, 95% confidence interval, 0.19%-0.54%).1 Mortality rates for MT are low and have been reported <0.8% in published studies.16,19 In our study, the mortality rate was 0.28%.
In summary, a clear conclusion of this work is that MT under local anesthesia and mild sedation is a safe and effective method for the diagnosis of patients with pleural effusion, even for patients at high risk for complications when performed by experienced physicians in the high-volume centers. However, in the low-volume centers, physicians need to be careful to perform MT in high-risk patients if they do not have good experience on MT. The rate of any complication in our series was 41.4%. Although most of the complications do not constitute a significant problem in terms of mortality and morbidity, they can affect the total procedure cost because they may necessitate further treatment. That is to say, even if the complications are minor, the number of patients with complications seems significant as a cost-increasing factor. Therefore, this factor should also be taken into consideration in future studies, and comparative studies should be conducted accordingly.