Comparison Trial of White Light, Autofluorescence and Narrow Band Imaging Bronchoscopy

Patients with known or suspected malignancies of the lung and with a medical indication for a bronchoscopy will be invited to participate in the study. All patients must be over 18 years old and must sign informed consent. The endoscopic examination of the bronchial tree will be performed using a commercially available flexible fluorescence bronchoscope (11004BI; Karl Storz; Culver City, CA) and the flexible white light video-bronchoscope (BF-T180; Olympus; Center Valley, PA).

Images obtained will be photographed using both video recordings and parallel images of white light and imaged bronchoscopy. These digital images will be electronically stored and evaluated at a later time. Biopsies will be obtained of all suspected abnormalities. These will be correlated with the images.

Images will be identified as normal, suspect or abnormal based upon each modality independently, paired with the spectral image and all three combined. The control will be an area considered normal by all modes of bronchoscopy. Images will be graded at the time of bronchoscopy. Later, all images will be graded in a blind fashion by two independent bronchoscopists. Standard statistical analysis using paired Student's t test will be applied. We anticipate that the combination of one or more imaged bronchoscopy techniques will increase accuracy by 25%. Power calculation to detect such a difference will require 140 abnormal specimens. If 50% of patients examined have an abnormality, the study population will need to be around 300 patients.

Histopathology reports will be obtained on all specimens. Abnormal results will be discussed with the individual patient and appropriate procedures will be performed based upon best medical practices. These procedures and patient follow-up are beyond the scope of the study.