Comparison of three-dimensional virtual endoscopy with bronchoscopy in patients with oesophageal carcinoma infiltrating the tracheobronchial tree

Citation
U. Rapp-bernhardt et al., Comparison of three-dimensional virtual endoscopy with bronchoscopy in patients with oesophageal carcinoma infiltrating the tracheobronchial tree, BR J RADIOL, 71(852), 1998, pp. 1271-1278
Citations number
23
Categorie Soggetti
Radiology ,Nuclear Medicine & Imaging","Medical Research Diagnosis & Treatment
Journal title
BRITISH JOURNAL OF RADIOLOGY
ISSN journal
00071285 → ACNP
Volume
71
Issue
852
Year of publication
1998
Pages
1271 - 1278
Database
ISI
SICI code
Abstract
Virtual endoscopy (VE) is a technique for performing simulated bronchoscopy using helical CT data of the tracheobronchial tree. In order to evaluate a virtual three-dimensional (3D) endoluminal procedure for the tracheobronch ial tree, comparison was made between bronchoscopy, axial CT images and min imal intensity projections (MIP). 21 patients were referred for helical CT because of oesophageal carcinoma shown by bronchoscopy to infiltrate into t he trachea or bronchi. Axial CT images obtained on a helical scanner were t ransferred to a Sparc20 workstation. VE was compared with the axial CT imag es and the MIP concerning additional information on the location and degree of stenosis gained after 3D reconstruction of the inner surface of the tra cheobronchial tree. The accuracy of this VE system was compared with bronch oscopy. Follow-up was performed in two patients to evaluate the tracheobron chial system after stent implantation. All stenoses were identified by VE w ith no statistically significant difference in detection of location or gra ding of the stenosis to real time bronchoscopy. Passage of subtotal stenosi s was only possible with VE. VE is suitable for following up stent implanta tion. Submucosal lesions of the tracheobronchial tree could not be detected by VE. There was no statistically significant difference regarding the loc ation of the stenoses between VE, axial CT slices, MIP and bronchoscopy. Th e VE showed only a statistically significant difference with regard to the degree of stenosis which was underrated on axial CT slices and MIPs. Pitfal ls including mucus plugs and wall defects due to the wrong threshold value were a limitation of VE. VE is presently too time-consuming to use in every patient with an infiltrating tumour into the tracheobronchial tree. In con clusion, while VE cannot replace endoscopy of the tracheobronchial tree or the oesophagus, it is an accurate and non-invasive method for identifying e ndoluminal tumours, grading stenoses and visualizing the tracheobronchial t ree beyond stenoses in a small number of patients who are not amenable to e ndoscopy.