| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorial |
1 From the Department of Radiology, Mayo Clinic E-2, 200 First St SW, Rochester, MN 55905. Received April 20, 2001; accepted April 21. Address correspondence to the author.
Index terms: Computed tomography (CT), contrast enhancement, 77.12112, 77.12114, 77.12115 • Computed tomography (CT), helical, 77.12115 • Editorials • Pancreas, CT, 77.12112, 77.12114, 77.12115 • Pancreas, neoplasms, 77.321
Today’s multi–detector row (ie, four-row) computed tomographic (CT) scanner provides radiologists with unprecedented speed for CT image acquisition. This newly acquired speed allows radiologists to select among several data acquisition options, such as larger volumes of data during a similar acquisition time, smaller volumes of data at more precise enhancement intervals (ie, shorter times), thinner collimated sections, or, in some cases, all three advantages combined. Equipped with this new technology, the radiologist is often confused with the smorgasbord of added capabilities and is tempted to exploit it in all manners possible to benefit the patient. In patients suspected of having pancreatic carcinoma, it is common to scan the pancreas up to four times: prior to the intravenous injection of contrast material and during the arterial, pancreatic parenchymal, and portal venous phases of contrast enhancement. All of this can be done without a substantial time penalty or x-ray tube heating problems and with the reward of acquiring exquisite cross-sectional images.
Depending on the multisection CT acquisition mode used for data collection, it is possible to retrospectively (ie, without rescanning the patient) reconstruct the raw CT data into images that are thinner than the original set of reconstructed images. This provides a second set of images that have improved anatomic detail although they were acquired during the same optimal phase of contrast enhancement as the first set of reconstructed images. These narrower images can be reformatted into off-axis planes to produce angiographic images and unique anatomic displays that were not previously possible.
Before long, the radiologist is confronted with a staggering stack of images to review and potentially high film and/or data storage costs. The constant dilemma is in determining which of these sets of images to acquire, which scans to view to solve specific imaging tasks, and which scans to exclude without compromising patient care.
In this issue of Radiology, McNulty et al (1) provide insight into an intelligent protocol for pancreatic CT performed by using a currently available multi–detector row (ie, four-row) helical scanner. This technology, if used properly, as McNulty and coauthors (1) showed, can translate into images with optimal contrast differences between the normal pancreatic parenchyma and the pancreatic adenocarcinoma, as well as optimal arterial and venous opacification. Optimal anatomic definition should translate into improved pancreatic cancer staging and better triage of patients to either surgery or palliative therapy.
The need for optimized protocols for the evaluation of patients suspected of having pancreatic carcinoma should not be understated. Pancreatic cancer remains a major health problem today: It is the fourth leading cause of cancer-related deaths in men and women (2). Despite major advances in the multiple imaging techniques used to directly visualize the pancreas, the 5-year survival with this disease has not been altered (2). Most patients die within a year or two after diagnosis and often suffer profoundly. One of the most common reasons for our failure to accurately stage patients with pancreatic cancer is that the tumor often has spread beyond the pancreas at diagnosis. The high recurrence rate in patients following "curative resection" implies that metastatic disease is usually present but invisible at operation, despite heroic efforts to remove the tumor and all the surrounding tissue and organs.
Adequate preoperative tumor detection and staging are the cornerstones of treating patients with this disease. Patients with clearly nonresectable disease (most commonly considered to be vascular encasement of the superior mesenteric artery or celiac artery, occlusion of the superior mesenteric vein or portal vein, or metastases to the regional lymph nodes or other distant organs) can benefit from being spared an extensive operation associated with substantial morbidity and mortality, cost, pain, and personal disruption. Patients with tumors that are considered to be localized and amenable to surgical removal have the option of operation and a hope for cure. Accurate triage of patients into each management group is usually the responsibility of the radiologist.
Of central importance to tumor staging is the initial detection of the tumor and the assessment of its size and location in reference to key vital peripancreatic structures. In this regard, the article by McNulty et al (1) is especially timely. By using the techniques recommended, maximum tumor-pancreas contrast can be achieved, and the mass can be seen to the best advantage. The authors of this article point out that maximum tumor conspicuity can be achieved during either the pancreatic parenchymal or portal venous phase of contrast enhancement. These two phases also are associated with maximum opacification of the celiac and superior mesenteric arteries (pancreatic phase) and of the superior mesenteric and portal veins (portal venous phase). Because of the proclivity of these tumors to invade these local structures and preclude complete resection, accurate preoperative assessment is critical. Imaging of the upper abdomen during both of these phases provides a robust technique for the detection and staging of this important neoplasm. Arterial phase image acquisition is unnecessary, unless CT angiographic images are desired. Precontrast images need to be obtained only for mapping a volume of tissue to be scanned later during contrast enhancement.
Other reports (3–10) on the proper helical CT technique for patients with pancreatic carcinoma have been published. Unfortunately, with each generation of CT scanners, it may be necessary to restudy this issue. Previously, to acquire data throughout the pancreatic volume, single-section CT units often had to scan a single tissue volume during both the arterial and pancreatic parenchymal phases of contrast enhancement. These acquisitions may have been labeled arterial phase or pancreatic phase images, depending on the time the data acquisition began after the start of intravenous contrast material administration and the various author definitions. Multisection CT scanners, because of their rapid acquisition capability, can acquire the same volume of image data in less than half the time that single-section scanners can and provide separate data sets for both the arterial and pancreatic parenchymal phases of contrast enhancement. The tightly packed data acquisition at multisection CT provides the potential for more accurate assessment of specific enhancement characteristics of the pancreas and peripancreatic tissues, but with this capability comes the need for added precision in timing the start of each acquisition. Results of the McNulty et al study (1) demonstrate that multisection scanning that begins 35 and 60 seconds after the start of a 150 mL bolus of contrast material injected at 4 mL/sec can result in optimized images of pancreatic tumors and the regional vascular structures.
It is now well recognized that the contrast material injection rate and volume affect the timing of peak contrast enhancement. Higher doses and faster rates of contrast material administration provide higher levels of pancreatic enhancement (11,12). More study is needed to determine the optimal dose and optimal rate of contrast material that enable us to detect smaller lesions and triage only the curable patients to operation. Similarly, the thin-section collimation available with helical scanning has allowed radiologists to identify and routinely evaluate peripancreatic arteries and veins that were previously ignored. These thinner sections have also allowed us to acquire reformatted arteriographic and parenchymal images that previously were not obtainable. The optimal collimation has not yet been determined, but it will likely be thinner than those we routinely use today, and it may depend on the imaging task to be accomplished.
Although radiation dose is usually a secondary concern in patients suspected of having pancreatic cancer, many patients with questionable pancreatic disease who are routinely scanned are found to have normal pancreatic studies. McNulty et al (1) found 49 (64%) of 77 patients who were examined for known or suspected pancreatic carcinoma to have normal studies. Therefore, imaging protocols should limit the radiation dose whenever possible. The radiation dose can be reduced by at least one-third by eliminating the arterial phase acquisition, as recommended in the McNulty et al article.
For all CT systems, it is important to recognize that if a protocol is changed to acquire narrower section widths, the milliampere second (and therefore the radiation dose) must be increased to maintain the same level of image noise. For a decrease in section width by a factor of two, the milliampere second must be increased by a factor of two for the image noise to remain unchanged (13). It is also important to recognize that some CT systems will automatically increase the milliampere second (which is directly proportional to dose) in response to the selection of a decreased section width. That is, the dose can be increased for thinner scan widths without explicit action by the operator.
In addition, for multisection CT systems, narrower collimation settings (eg, 4 x 1-mm or 4.0 x 2.5-mm detector configurations with nominal radiation beam widths of 5 or 10 mm, respectively) are less dose efficient than are wider collimation settings (eg, 4 x 5-mm detector configuration with nominal radiation beam width of 20 mm). For example, with the same milliampere second and pitch, changing from a 5-mm to a 20-mm nominal radiation beam doubles the patient dose (14). This effect has been shown to be similar among multisection CT equipment manufacturers (15). Hence, in multisection CT, an increase in patient dose is associated with thinner scan widths owing to two cumulative effects: (a) the need to increase the milliampere second to maintain equivalent image noise and (b) the dose inefficiency of narrower beam collimations. Hence, the diagnostic benefit of thinner multisection CT acquisitions should reasonably outweigh concerns regarding the increased patient dose. Studies such as that of McNulty et al (1) are of considerable value in helping radiologists determine which imaging paradigms have the greatest diagnostic yield. Most experts today recommend 2.5–3.0-mm collimation of the pancreas (3–10). The optimal dose settings and fields of view have yet to be determined.
Technologic advances will continue to improve CT in the future. Just as we become comfortable with the newfound acquisition speed of four-row multisection scanners, manufacturers are ready to introduce eight-row scanners. In a march toward faster and thinner collimation acquisition machines, additional rows of detectors beyond eight are also being developed. With each exciting advance, we will need to restudy imaging protocols, just as McNulty and coauthors (1) did for pancreatic carcinoma. They have given us an efficient protocol to use with today’s CT machines and a template for rediscovery as new technology is being introduced.
FOOTNOTES
See also the article by McNulty et al (pp 97–102 ) in this issue.
REFERENCES
Related Article
This article has been cited by other articles:
|
|
 |
|
  K. Tawada, T. Ishihara, A. Kobayashi, T. Yamaguchi, T. Tsuyuguchi, M. Matsuyama, and O. Yokosuka Quantitative Analysis of Vascular Endothelial Growth Factor in Liver Metastases from Pancreatic Carcinoma as a Predictor of Chemotherapeutic Effect and Prognosis Clin. Cancer Res., November 15, 2008; 14(22): 7438 - 7443. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ichikawa, S. M. Erturk, H. Sou, H. Nakajima, T. Tsukamoto, U. Motosugi, and T. Araki MDCT of pancreatic adenocarcinoma: optimal imaging phases and multiplanar reformatted imaging. Am. J. Roentgenol., December 1, 2006; 187(6): 1513 - 1520. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Golding Multi-slice computed tomography (MSCT): the dose challenge of the new revolution Radiat Prot Dosimetry, May 17, 2005; 114(1-3): 303 - 307. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S Fenchel, T R Fleiter, A J Aschoff, R van Gessel, H-J Brambs, and E M Merkle Effect of iodine concentration of contrast media on contrast enhancement in multislice CT of the pancreas Br. J. Radiol., October 1, 2004; 77(922): 821 - 830. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
