Diagnostic algorithm of lung cancer employed in the study. BAL, bronchoalveolar lavage; ENB, endobronchial brushing; FTB, forceps tissue biopsy
Malignant neoplasm was diagnosed in 106 cases. A positive result (cancer cells) in bronchofiberscopy specimens was described in 45 subjects (42.0%). No severe complications of this procedure were observed. Mild complications including elevated body temperature and hemoptysis, were rare.
Cytology (BAL, ENB) was positive in 38 samples (35.8%) and histopathological examination (FTB) in 30 samples (28.3%). Based on BAL, 11 samples (10.3%) turned out positive, yielding nine non-small cell lung cancer (NSCLC), and one small cell (SCLC) and undifferentiated lung cancer each. Brushing was more effective than BAL yielding 27 positive samples (25.5%): twenty NSCLC, six SCLC, and one undifferentiated cancer. The FTB was performed in 33 patients (31.0%) providing the following diagnostic yield: twelve squamous cell carcinoma, nine SCLC, six adenocarcinomas, one lung carcinoid tumor, and three negative results; with the overall effectiveness of 90% (Fig. 2).
Cancer types. FTB, forceps tissue biopsy; BAL, bronchoalveolar lavage; ENB, endobronchial brushing; SCLC, small cell lung cancer; NSCLC, non-small cell lung cancer
The present study encompassed patients suspected for malignant tumors in the lungs based on the chest CT alone or in combination with symptoms suggestive of lung cancer. Further tests confirmed the presence lung cancer in 106 patients, which yields a true positive rate of 96.4%. This diagnostic result is comparable to that obtained for forceps tissue biopsy with a true positive rate of 90%, and much higher compared with the BAL and ENB results, which yielded a true positive rate 10.3% and 25.5%, respectively.
A high sensitivity of CT examination is due likely to the characteristics of the patients studied and the CT criteria for positive screening. Up-to-date, there is no consensus on the definition of a positive result which affects sensitivity of CT examination. In the present study, all patients referred to our institution had been suspected of lung cancer and presented symptoms of advanced stages of the disease. Oliveira and Saraiva (2010) have examined a cohort of 70 patients suspected of lung cancer who underwent chest CT and bronchoscopy. They confirmed cancer in 37 (52.9%) patients, which gives relatively low sensitivity. For low-dose CT screening tests, Toyoda et al. (2008) have shown sensitivity of 88.9% and specificity of 92.6%. Lung cancer was diagnosed when ground-glass opacity nodules were over 10 mm in diameter. Smaller lesions were observed but were not subjected to invasive treatment. In the NELSON study, scans with nodules of a diameter over 9.8 mm were regarded positive. Additionally, nodules between 5 and 9.8 mm were considered as suspected and observed. Such nodules were regarded positive in case of 25% growth in volume after 3 month (Horeweg et al. 2013). In the National Lung Screening Trial, non-calcified nodules or masses with diameter of at least 4 mm were considered a positive screening result (NLSTRT 2011).
Bronchofiberoscopy is generally regarded as a safe and effective way to diagnose and stage patients with lung cancer. BAL and ENB are often performed routinely in every patient suspected of lung cancer during bronchofiberoscopy. CT examination performed prior to bronchofiberoscopy helps obtain samples of lesions located beneath the mucosa or pathologically changed lymph nodes. BAL is mainly used for cytological evaluation. Additionally, measurement of potential biomarkers including hepatocyte growth factor, interleukins, high mobility group proteins, and chemokine ligands in the bronchoalveolar lavage fluid and serum help diagnose different types of lung cancer (Jakubowska et al. 2015; Naumnik et al. 2016; Pastor et al. 2016). A diagnostic value of BAL differs for cancers depending on the location of lesions and their characteristics. The BAL test is based on the evaluation of cells exfoliated from malignant tumor. Lower sensitivity of BAL may be rooted in the fact that some of the lung tumors do not exfoliate cells useful for the diagnostic process. Bezel et al. (2016) have examined patients with suspected peripheral lung cancer. They showed that cytologic examination was truly positive in 18% and truly negative in 34% of cases. It had a sensitivity of 29% and a negative predictive value of 44% for the diagnosis of malignancy, while other bronchoscopic techniques, such as transbronchial forceps biopsy (sensitivity of 41%), bronchial washing (sensitivity of 36%), and bronchial brush biopsy (sensitivity of 38%) had a better diagnostic performance. Labbe et al. (2015) have reported the overall sensitivity of BAL at a level of 11.6% and ENB of 16.5%. They also revealed a higher diagnostic yield in patients with larger lesions, central/intermediate distance from the hilum and in case of the presence of a bronchus sign. The results of the present study are generally in line with the findings of other researchers as we found a lower overall sensitivity of BAL and ENB than that of forceps tissue biopsy.
Bronchial forceps biopsy can be performed either in case of a visible bronchial lesion or more distal, endobronchially invisible lesions. In the present study, forceps tissue biopsy was carried out when macroscopic mucous membrane abnormalities were seen during bronchofiberoscopy. This type of biopsy has a lower complication rate than open biopsy has. However, peripheral lung lesions often remain inaccessible. Binesh et al. (2015) have compared a diagnostic value of transbronchial lung biopsy with that of BAL in patients with pulmonary lesions suggestive of lung cancer. They confirmed a low specificity of BAL for detecting lung cancer, along with its usefulness in case of peripheral lesions. Thus, a combination of BAL with forceps biopsy may increase a positive diagnostic rate.
Introduction of advanced modalities such as endobronchial ultrasound transbronchial aspiration (EBUS-TBNA), video-assisted thoracoscopic lung biopsy, and virtual bronchoscopy enables to further improve the diagnostics. Those techniques supplement conventional examinations, especially in the assessment of lymph nodes involvement, staging of lung cancer, or in difficult cases in which the diagnosis could not be made based on other methods (Fukusumi et al. 2016; Khan et al. 2016; Adamczyk et al. 2013; Szlubowski et al. 2009; Silvestri et al. 1996). Despite technological advances, patients still benefit from the use of conventional bronchoscopic techniques such as BAL, brush and forceps biopsy. Combining conventional bronchoscopic techniques with new diagnostic modalities reduces the length of time between the first abnormal CT finding and the diagnosis of lung cancer, and also helps limit the number of consecutive procedures, which hastens the commencement of curative treatment (Binesh et al. 2015; Verma et al. 2015; Salomaa et al. 2005).