Diagnostic value of medical thoracoscopy in malignant pleural effusion

The identification of malignant cells in a pleural lavage in patients without pleural effusion suggests micrometastatic disease, and our previous meta-analysis [1] showed that positive pleural lavage cytological findings are associated with a higher recurrence rate and significant poorer survival, with the overall hazard ratio for patients having malignant cells in pleural lavage was 5.61 (95% confidence interval 3.98–7.90). In non-small-cell lung cancer patients, the evidence of even a minimal pleural effusion at diagnosis is an independent prognostic factor for worse survival [2]. Malignant pleural effusion (MPE) is frequently observed in multiple malignancies, and lung cancer is the most common cause [3]. The existence of MPE in patients indicates systemic dissemination of cancer and declining in life expectancy and quality [4, 5].

The current guideline recommended that thoracentesis and/or closed pleural biopsy can be used as the first diagnostic steps in the diagnosis of MPE [6]. However, these procedures usually do not work when pleural effusion with thickness less than 10 mm on chest computed tomography (CT) scans. Instead, the more invasive approaches, such as medical thoracoscopy (MT), can be considered to identify whether pleural biopsy contains malignant cells [3, 7]. As a matter of fact, MT is a highly sensitive and safe method for diagnosing exudative pleural effusions [8‐10]. The recent developed semi-rigid MT is easy to use and can gain popularity among respiratory physicians who are accustomed to flexible bronchoscope [11, 12].

In the present retrospective study of patients with MPE having undergone at least one semi-rigid MT over a 9-year period in a Chinese 1600-bed general hospital, we analyzed the diagnostic efficiency and safety of MT in the diagnosis of MPE.

The study protocol and ethical approval was approved by the Institutional Review Board for human studies of Beijing Chao-Yang Hospital, China. Informed consents were not required as this was considered a review of clinical practice.

Information including medical history, clinical presentation, laboratory examination results, and image data of unexplained exudative pleural effusions patients who underwent MT in our hospital between July 2005 and June 2014 were gathered, and only MPE patients were finally included in the current study. Unexplained exudative pleural effusions were defined as the patients underwent the initial diagnostic approaches including thoracentesis and/or closed pleural biopsy, and their diseases remain undiagnosed. The characteristics of the study population are listed in Table 1.

MT procedures have been described in our previous publications [13, 14]. The diagnosis of MPE was established by the presence of the positive findings for malignancy in pleural biopsy.

Descriptive statistical methods were used in the data analysis (mean ± standard deviation [SD] or/and range).

Between July 2005 and June 2014, 833 patients with undiagnosed pleural effusions successfully underwent medical thoracoscopy [15]. Eventually, 342 patients with lymphocytic exudates were finally diagnosed with MPE; the mean age was 62.8 ± 9.7 years.

For 149 MPE patients, pleural fluid occurred only in the right side, for 133 only in the left, and for the rest 60 both sides were involved (Table 1). The size of a pleural effusion was clarified as small, moderate, or large based on CT imaging according to the methods described by Moy and colleagues [16]. In both unilateral and bilateral effusion, the proportions of small, moderate, and large size of pleural effusions were 16.7, 12.9, and 70.4%, respectively. The appearance of pleural effusion was blood-stained in 55.9% of patients, and in 44.1% was yellow.

In addition to pleural effusion, CT imaging revealed mediastinal and hilum lymphadenopathy, pleural thickening, pulmonary consolidation or infiltration, pulmonary mass or nodules, pulmonary atelectasis, and pleural nodularity (Table 1).

In all patients studied, we observed one or more abnormalities on the surface of parietal or/and visceral pleura under medical thoracoscopy. As shown in Table 2, pleural nodules, hyperemia, pleural adhesion, pleural plaques, ulcer, and the other pleural pathological changes were observed.

The most common etiological causes of MPE were metastatic carcinomas (n = 272), pleural malignant mesothelioma (n = 35), lymphoma (n = 10). It should be mentioned that we could not identify the original malignancies in 25 patients with MPE (Table 3). Among metastatic malignancies that resulted in MPE, the most common cancer included lung cancer, followed by breast cancer, ovarian cancer, pancreatic cancer (Table 4).

No serious adverse events were observed, and transient chest pain (43.9%) induced by the indwelling chest tube was the most frequent minor complication. Subcutaneous emphysema was found in 8.5% of patients who recovered after chest-tube drainage. Minor bleeding was recorded in 6.4% of patients. And 5.6% of patients appeared with transient self-limited fever (38 °C or more).

Because the prognosis for patients with MPE is poor, an efficacious procedure that can establish a definite diagnosis as early as possible with a minimum of risk and discomfort would be highly desirable. If a patient with undiagnosed pleural effusion is suspected as malignant, cytologic examination of pleural fluid is the first recommendation [6]. Although repeated thoracenteses can enhance the sensitivity of cytology, it is usually only 50 to 70% [17]. When cytology fails, closed percutaneous needle biopsy was traditionally performed blindly by an Abrams or Ramel needle [18‐20]. Nevertheless, its role in diagnosing MPE has been challenged, as the positive diagnostic rate of closed pleural biopsy was only about 50% [21, 22]. More recently, the real time image-guided pleural biopsy has been shown to be a promising procedure for sampling the pleura, since it can increase the sensitivity for diagnosing MPE to about 80% [21, 23‐25].

Numerous tumor markers have been intensively examined for improving the diagnosis of MPE, however, seeking for a highly accurate pleural fluid tumor marker that reliably diagnoses MPE has been in vain so far [26]. Using one tumor marker alone for diagnosing MPE is not recommend according to the recent evidences, however, when combined two or more tumor markers together, the diagnostic sensitivity seems to be improved [27, 28]. The diagnostic performance of tumor markers for MPE seems to be similar with conventional tests including cytological examination–high specificity and low sensitivity. Tumor markers are less important in practice, since they do not complement the properties of conventional tests.

CT- or ultrasound-guided pleural biopsies are quite sensitive and safe, with the only reported complications being local hematoma and minor hemoptysis [21, 29]. The limitation of the image-guided pleural biopsy is the blindness of the procedure. MT overcomes this problem by allowing for the visualization of abnormal areas and for a direct biopsy, and thus improves the diagnostic accuracy of pleural effusions [8, 9]. Since June 2005, our institution started using MT as a routine method for patients with undiagnosed exudative pleural effusion in cases when either clinical, radiologic, laboratory, or cytologic investigation was failed. During a period of 9 years, 833 patients with unexplained pleural effusions underwent MT successfully, and among them, 342 were eventually diagnosed with MPE [15].

One or more abnormalities on the surface of parietal or/and visceral pleura were observed in the whole population in this study, including pleural nodules, hyperemia, pleural adhesion, pleural plaques, ulcer, and the other pleural pathological changes. Pathological examination revealed the presence of the positive findings for malignancy in pleural biopsy in 342 patients. The data in detail presented in the current study derived from our whole MT study population [15]. As reported in the previous publication [15], after a complete work-up including MT biopsies, the definite diagnoses of 92.6% (771/833) of patients with pleural effusions can be established definitely by MT followed by histopathological study.

It was noted that no etiological causes of pleural effusions can be identified in 7.4% (62/833) of patients even after MT [15]. All of these patients were followed up for at least 12 months, and did not complain about a new pleural effusion. No diagnosis other than benign pleural effusion was found in these 62 patients. Several studies suggest that among the patients with histological diagnosis of non-specific pleurisy made after MT, 8.3–18% of them were eventually diagnosed with MPE, usually pleural mesothelioma, during long term follow-up [30‐32]. Therefore, we cannot exclude the possibility that a few patients with MPE-negative MT results would be finally diagnosed with MPE if we prolong the follow-up.

It has been reported that carcinoma from any organ can metastasize to the pleura, but lung, and breast carcinomas and lymphomas are the most common causes, digestive and ovary carcinomas are less frequent [33]. Our recent unpublished data indicated that during the past 3 years, 23.7% (365/1541) of pleural effusion patients admitted to our hospital were diagnosed with MPE. In the present study, our data showed that the most common etiological causes of MPE confirmed by MT were metastatic carcinomas, and followed by pleural malignant mesothelioma, lymphoma, and the other malignancies. Among metastatic malignancies that resulted in MPE, the most common cancer was lung cancer, followed by breast cancer, ovarian cancer, and pancreatic cancer.

MT has an excellent safety profile when performed by a trained physician, and a mortality rate associated with MT is always ≤0.8% [34]. A recent meta-analysis further suggested that mortality associated with MT was not observed, and that the major complication rate of MT was 1.5% and the minor complication rate was 10.5% [10]. In our 9-year study, MT procedures were well tolerated with a low rate of complications without serious adverse events, and transient chest pain caused by the indwelling chest tube was the most frequent minor complication. Large volumes of pleural fluid could be safely aspirated, although some patients suffered from coughing and chest discomfort after lung re-expansion with a chest tube.

The strengthen of this study was that study population was in a large size, in which 342 MPE patients were included. At the same time, our study also had limitations. First, as a retrospective study, it’s impossible to collect and analyze the required data in a prospective way. Second, we could only retrospectively reviewed the data from patients with MPE, and no data from the other control group, such as the patients with tuberculous pleurisy, were available, it was therefore not possible to calculate the sensitivity and specificity of MT in diagnosing MPE. Third, blind needle biopsies or image-assisted biopsies were performed only in a few patients before undergoing MT in our series. This partially explained why there were so many pleural effusion patients (833 cases) receiving MT examination in our institution during 9 years.

In summary, MT is simple and safe with a high positive rate in the diagnosis of MPE. Due to its convenience and compatibility with existing bronchoscopy, MT appears to be a more widely performed procedure. Thus MT should be performed actively for proper patients with suspected MPE.