Virtual bronchoscopy for evaluation of malignant tumors of the thorax

• Published in: Journal of thoracic and cardiovascular surgery Vol. 123; no. 5; pp. 967 - 972
• Main Authors: Finkelstein, Steven E., Summers, Ronald M., Nguyen, Dao M., Stewart, John H., Tretler, Jean A., Schrump, David S.
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Philadelphia, PA Elsevier Inc 01.05.2002; AATS/WTSA; Elsevier
• Subjects: Adult; Aged; Aged, 80 and over; Biological and medical sciences; Bronchoscopy - methods; Evaluation Studies as Topic; Female; Fiber Optic Technology; Humans; Image Processing, Computer-Assisted; Investigative techniques, diagnostic techniques (general aspects); Male; Medical sciences; Middle Aged; Prospective Studies; Radiodiagnosis. Nmr imagery. Nmr spectrometry; Respiratory system; Sensitivity and Specificity; Thoracic Neoplasms - pathology; User-Computer Interface; Bronchoscopy; Human; Fibroscopy; Radiodiagnosis; Pathophysiology; Respiratory disease; Virtual image; Helical computed tomography; Thorax; Malignant tumor; Respiratory system; Image reconstruction; Biomedical information; Medical imagery; Tridimensional image; Endoscopy; Diagnosis; Technique; Comparative study
• ISSN: 0022-5223; 1097-685X
• DOI: 10.1067/mtc.2002.121495

Objective: Virtual bronchoscopy is a novel technique making use of 3-dimensional reconstruction of 2-dimensional helical computed tomographic images for noninvasive evaluation of the tracheobronchial tree. This study was undertaken to evaluate the diagnostic potential of virtual bronchoscopy by comparing virtual bronchoscopic images with fiberoptic bronchoscopic findings in patients with thoracic malignant disease. Methods: Thirty-two consecutive patients with thoracic malignant tumors underwent virtual bronchoscopy for evaluation of suspected tracheobronchial lesions. For each virtual bronchoscopic examination, 200 to 300 contiguous 1.25-mm images of the thorax were obtained in only one or two 17-second breath holds by using a multislice computed tomographic scanner. Virtual bronchoscopy images were reconstructed and interpreted blind to the actual endoscopic findings. Results of virtual bronchoscopy were compared with fiberoptic bronchoscopic findings in 20 patients. Results: Anatomic computer simulation of the bronchial tree was successfully created in all patients. In 7 (35%) of 20 patients, results of fiberoptic bronchoscopy were found to be within normal limits. In all patients with normal anatomy, virtual bronchoscopy accurately correlated with the fiberoptic findings. Thirteen (65%) patients had a total of 22 abnormal findings on fiberoptic bronchoscopy. Virtual bronchoscopy detected 18 of 22 abnormal fiberoptic bronchoscopic findings: 13 of 13 obstructive lesions, 5 of 6 endoluminal lesions, and 0 of 3 mucosal lesions. The sensitivity of virtual bronchoscopy was 100% for obstructive lesions, 83% for endoluminal lesions, 0% for mucosal lesions, and 82% for all abnormalities; the specificity of virtual bronchoscopy was 100%. Conclusions: Preliminary evaluation indicates that virtual bronchoscopy may be a promising and noninvasive modality for identifying bronchial obstructions and endoluminal lesions, as well as for assessing the tracheobronchial tree beyond stenoses. However, at present, virtual bronchoscopy does not enable the detection of subtle mucosal lesions, and as such, this modality may not be appropriate for identifying premalignant lesions in the respiratory tract. Although fiberoptic bronchoscopy remains the standard modality for evaluating airway patency and mucosal lesions, virtual bronchoscopy may provide additional information that may be useful in the management of pulmonary malignant tumors. J Thorac Cardiovasc Surg 2002;123:967-72