The value of narrow-band imaging bronchoscopy in diagnosing central lung cancer.

Aims: This research aimed to study the value of narrow-band imaging(NBI) in the diagnosis of central lung cancer. Materials and methods: This study included 916 patients with clinical suspected of central lung cancer or follow-up of patients after curative lung cancer surgery. All of the patients were examined by Olympus Evis Lucera electronic bronchoscope system, any sites that were abnormal when viewed by white-light bronchoscopy (WLB) or NBI were biopsied, four to six biopsies were taken at each site of the abnormal region visualized as lesions, we record the endoscopic features of NBI and compared with histopathology results, to evaluate the diagnostic value of NBI for central lung cancer and the relationship between vascular patterns of NBI and histological types of lung cancer, and try to establish a multinomial logistic regression model for predicting the histological types of lung cancer. The biopsy specimens were examined by CD34 antibody through immunohistochemistry (IHC) method, CD34 marked microvessel density(MVD), compared the number of microvessels between benign and malignant diseases and the number between different histological types of lung cancer, to verify the results of NBI. Results: NBI provided high sensitivity (91.7%), specificity (84.9%), positive predictive value (97.6%), negative predictive value (61.5%), and agreement rate (90.7%). The predominant vascular patterns in the well-defined histological types of lung cancer were dotted blood vessels (121 patients), tortuous blood vessels (248 patients), and abrupt-ending blood vessels (227 patients). Logistic regression analysis of the results showed that smoking status of the patient, combined with vascular patterns under NBI, and age partly affect the histological types of lung cancer. Conclusions: NBI is highly accurate for the diagnosis of central lung cancer.