Ann Thorac Surg 1996;62:817
© 1996 The Society of Thoracic Surgeons
Invited Commentary
Invited Commentary
David P. Naidich, MD
Department of Radiology, NYU Medical Center/Bellevue Hospital, 27th St and 1st Ave, New York, NY 10016
See also page 811.
It is with appreciation that I read this article. Although I agree with LoCicero and associates' basic premise that CT is of value in assessing the central airways, it is notable that they fail to emphasize the considerable body of data already available documenting the accuracy of routine axial images for assessing airway pathology, especially when acquired helically. A number of issues, both technical and methodologic, concerning this work need to be addressed.
Concerning LoCicero and associates' technique for acquiring spiral CT scans, it is unclear why they feel it necessary to first perform a "standard thoracic survey" using 8-mm sections before obtaining a second data set using 3-mm sections for multiplanar reconstructions. This involves considerable wasted time and, more important, unnecessary radiation exposure. Most current scanners allow adequate coverage of the central airways from the outset using either 3- or 5-mm collimation, eliminating the need for additional scan acquisitions. Given their statement that these studies were obtained in patients with known or suspected tracheobronchial pathology, it is also unclear why they felt it necessary to obtain 3-mm sections through the entire thorax. There seems little purpose in scanning through the lung bases using 3-mm collimation if the indication for CT evaluation is the central airways.
Methodologically, LoCicero and associates assert that both multiplanar reconstructions and 3-D reconstructions are "additive" to routine transaxial images. Did LoCicero and associates evaluate these images independently or in association with corresponding axial images? Their statement that all scans were reviewed prospectively does not clarify this issue. Although I have little doubt that multiplanar and 3-D reconstructions add important data concerning the true length of tracheal and main bronchial lesions, it is most unlikely that any truly unique information was gained compared with the initial transaxial source images. (Their illustrations of a widely patent tracheobronchial fistula or a tracheal hamartoma are good examples.) What were the relative contributions of multiplanar reconstructions compared with 3-D surface images (or, for that matter, the minimum intensity projection images referred to but not discussed in the text)? This is not a trivial issue as the time it takes to generate multiplanar reconstructions is a fraction of that necessary to generate acceptable 3-D images which are truly labor intensive. In my experience, the use of 3-D reconstructions, unlike multiplanar reconstructions, hardly justifies the additional time or expense.
Finally, LoCicero and associates might have pointed out that in addition to multiplanar and 3-D reconstructions there are other means available for using spiral scan data to assess the central airways. The ability to acquire images first in deep inspiration and then in expiration, for example, allows "dynamic" assessment of central airway mechanics-a point obscured by LoCicero and associates' suggestion that the dynamic component of airway obstruction is "better" assessed by bronchoscopy. In addition, they fail to mention the technique of projectional imaging, or so-called virtual bronchoscopy. The ability to reconstruct axial images to visualize the endoluminal appearance of the airways represents another extremely promising application of spiral scanning deserving mention.
Related Article
-
Spiral CT With Multiplanar and Three-Dimensional Reconstructions Accurately Predicts Tracheobronchial Pathology
- Joseph LoCicero, III, Phillip Costello, Christian T. Campos, Nicola Francalancia, Kevin M. Dushay, Ronald C. Silvestri, and Joseph D. Zibrak
Ann. Thorac. Surg. 1996 62: 811-817.
[Abstract]
[Full Text]
