| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Science to Practice |
1 Department of Diagnostic Radiology, Yale University School of Medicine, 789 Howard Ave, TE2, New Haven, CT 06520-8042 james.brink@yale.edu
|
Although polyps larger than 1 cm in diameter are generally considered to be of importance in most colon screening programs, current screening strategies in the United States and the United Kingdom triage patients into various screening algorithms partly on the basis of the presence of multiple (three or more) adenomas of any size (1,2). In this issue of Radiology, Taylor and colleagues (3) report on the in vitro evaluation of the technical parameters that govern polyp detection at four-channel multi–detector row computed tomographic (CT) colonography.
The Science
In recent years, investigators have used in vitro models to evaluate the effects of CT scanning parameters and colonic orientation on polyp depiction (4–6). Whereas some investigators (5) have created artificial polyps in excised animal colon specimens, others (4,6) have embedded spherical objects into tubular structures to simulate polyps within the colon. In the study under discussion here, Taylor et al closely replicated the in vivo condition by using a freshly excised total colectomy specimen from a patient with familial polyposis. Within a single specimen, 38 polyps measuring 2–15 mm in diameter provided a robust data set with which to test the image acquisition parameters of interest.
|
The orientation of the colon with respect to the longitudinal axis of the gantry affects the diagnostic performance for polyp detection. Taylor and colleagues showed substantial degradation in the detection of small polyps in colonic segments oriented orthogonally to the direction of patient travel (ie, the transverse colon). The detection of polyps in the transverse colon relies on the longitudinal resolution of the scan, whereas the detection of polyps in the ascending and descending colon relies on the in-plane resolution of the scan. Since longitudinal resolution is typically worse than in-plane resolution, the detection of polyps in the transverse colon was notably reduced.
The Practice
Clinical use.—The diagnostic performance of multi–detector row CT colonography for the detection of polyps is predicated on the interaction between the CT acquisition and the distended colon being interrogated. Although the size threshold for clinical concern regarding adenomatous polyps is generally held at 1 cm, the presence of multiple adenomas of any size affects patient care in many screening algorithms (1,2). As such, the use of high-resolution imaging is appropriate, as long as the radiation dose is carefully controlled. Such a strategy works well in imaging applications that make use of high-contrast interfaces such as the air–soft-tissue interface in CT colonography. In these applications, one may accept a higher level of image noise than one might otherwise prefer in such low-contrast imaging applications as the detection of metastatic lesions in the liver. Although Taylor and colleagues have shown that a low-pitch strategy helped improve detection of small polyps in the 2–4 mm range, breathing-related artifacts that may be expected with a low-pitch technique could overwhelm the improved detection of small polyps under these conditions. Thus, the protocol favored by Taylor and colleagues (collimation, 1.25 mm; pitch, 1.5; tube current, 50 mA) represents a reasonable compromise between radiation dose and polyp detection.
Future opportunities and challenges.—Recent improvements in CT scanners have outstripped the pace of scientific validation, and additional experimentation will be necessary to determine the appropriate technical parameters for each generation of this technology. Owing to the widespread and rapid deployment of eight- and 16-channel multi–detector row CT systems, the results of this four-channel study are already in need of reevaluation. Further advances in CT are expected beyond 16-channel systems. Volumetric CT, in which a flat-panel detector system is mounted on a CT gantry, will likely emerge in the near future with even further improvements in resolving capacity.
Summary
Using a freshly excised human colectomy specimen from a patient with familial polyposis, Taylor and colleagues have shown that four-channel multi–detector row CT colonography was best performed with 1.25-mm collimation. Although polyp detection was somewhat better with a beam pitch of 0.75, the reduction of respiratory artifacts associated with a pitch of 1.5 was thought to outweigh the slight improvement in the detection of polyps in the 2–4-mm range that would have been achieved with a pitch of 0.75. Reduction of tube current to 50 mA did not degrade polyp detection in this in vitro model. However, care must be taken to evaluate the tube current used in each patient, respectful of each individual’s unique body habitus.
FOOTNOTES
See also the article by Taylor et al in this issue.
REFERENCES
Related Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
