DOI: 10.1148/rg.276075034
RadioGraphics 2007;27:1653-1666
© RSNA, 2007
Comprehensive Preoperative Assessment of Pancreatic Adenocarcinoma with 64-Section Volumetric CT1
Darren D. D. Brennan, MB, BCh, BAO, MSc(Rad Sci), MRCPI, FFR(RCSI), FRCR,
Giulia A. Zamboni, MD,
Vassilios D. Raptopoulos, MD, and
Jonathan B. Kruskal, MD, PhD
1 From the Department of Radiology, Beth Israel Deaconess Medical Center, 330 Brookline Ave, CC-377, Boston, MA 02215. Presented as an education exhibit at the 2005 RSNA Annual Meeting. Received February 28, 2007; revision requested March 16 and received May 31; accepted June 11. G.A.Z. received a research grant from Toshiba; V.D.R. received research grants from Toshiba and E-Z-EM; remaining authors have no financial relationships to disclose.
Address correspondence to D.D.D.B. (e-mail: dbrennan{at}bidmc.harvard.edu).
 |
Abstract
|
|---|
Pancreatic adenocarcinoma is a common gastrointestinal malignancy that has a poor prognosis and for which successful surgical resection is the only method of cure. Preoperative staging and assessment can be performed with a number of modalities. 
Multidetector (64-section) volumetric computed tomography (CT) allows rapid anatomic coverage coupled with excellent spatial resolution. Understanding the technical parameters necessary for successful pancreatic CT angiography is crucial. Carefully timed scan acquisition maximizes the difference in attenuation between the neoplasm and the pancreatic parenchyma and allows accurate local and distant staging as well as assessment of local resectability. In addition, angiographic data sets can be rendered to create displays of the local venous and arterial anatomy that are familiar to surgeons. Advanced rendering can also be used to create pancreaticographic type images. The TNM system of staging for pancreatic adenocarcinoma is not frequently included in radiology reporting but is important for deciding on optimal therapy and neoadjuvant therapy.
© RSNA, 2007
|
Introduction
|
|---|
Pancreatic carcinoma is the fourth leading cause of death from cancer in the United States (1,2). The disease has a dismal prognosis, with a mortality rate similar to its incidence: The overall 5-year survival rate is less than 5% (3). Early diagnosis and resection remain the only potential cure, but only a minority (5%–30%) of tumors are detected when they are still resectable (4,5). Screening has not proved to be effective in the general population and is not recommended (6). In patients with pancreatic cancer, the detection of disease at an early and therefore likely resectable stage is very important.
In this article, we describe our experience with 64-section volumetric computed tomography (CT) in the preoperative assessment of pancreatic adenocarcinoma in terms of technical factors, scanning protocol, and postprocessing techniques that can maximize the information obtainable with 64-section CT. In addition, we review the relevant arterial and venous anatomy, an understanding of which is crucial when determining the advisability of en bloc resection, and describe relevant vascular variants and disease entities that should be sought out as part of a comprehensive preoperative assessment.
|
Role of Imaging
|
|---|
The role of imaging in pancreatic adenocarcinoma has evolved over the past few years as technology has evolved. Imaging is reserved for the evaluation of patients who are suspected of having pancreatic cancer on the basis of clinical, endoscopic, or laboratory findings. As mentioned earlier, screening in the general population is not recommended (6), and even in patients with familial pancreatic carcinoma, the role of radiology in screening is not well defined. Ultrasonography (US), endoscopic US, contrast material–enhanced US, magnetic resonance imaging, and multidetector CT provide complementary, sometimes competing, information, and their use worldwide reflects local expertise and availability. At our institution, we routinely use multidetector CT for preoperative evaluation, since it couples good spatial resolution with wide anatomic coverage and thus allows comprehensive local and distant disease assessment in a single session. In addition, the latest 64-section multidetector CT scanners, coupled with the newest generations of power injectors, allow accurate timing of the contrast material bolus, thus in turn allowing reproducible imaging that potentially maximizes the difference in attenuation between the tumor and the background pancreatic parenchyma. Finally, multidetector CT data sets can subsequently be rendered in a number of ways, which improves diagnostic yield and utility. Whichever imaging modality is used, the aim of preoperative assessment should be to (a) localize the pancreatic adenocarcinoma, (b) stage the tumor and determine if it is locally resectable, and (c) advise the surgeon preoperatively of relevant anatomic vascular variants.
|
Multidetector Volumetric CT
|
|---|
The advent of slip-ring technology, which heralded the appearance of spiral CT, has proved to be one of the most significant advances in the history of medical technology. Single-section technology gave way to multisection technology, and the number of detector rows has increased almost exponentially, to the point that 64-section CT platforms are now available worldwide and are produced by a number of manufacturers. A section thickness of approximately 0.5–0.625 mm, coupled with gantry rotation speeds of 0.33–0.5 seconds, allows vast anatomic coverage during comfortably short breath holds. The rapid acquisition eliminates artifact that might arise from peristalsis or breathing, and the advent of isotropic voxels makes multiplanar imaging a clinical reality and allows flawless volumetric reconstruction.
|
Technical Factors
|
|---|
Timing of the contrast material bolus should be optimized to maximize sensitivity for the detection of primary and metastatic tumors. With 64-section CT, multiple discrete phases of vascular and parenchymal enhancement can easily be achieved, but the radiologist must be cognizant of the heavy radiation exposure involved and strive to obtain the maximum amount of information while minimizing radiation exposure.
Scan timing can be determined with three methods: empiric timing, use of a test bolus, or bolus tracking. With the test bolus technique, multiple low-dose scans are acquired at a single level, typically around the level of the celiac axis, following the administration of 20 mL of iodinated contrast material. A temporal graph of the contrast enhancement in the aorta is obtained by positioning a region of interest over the aorta, and the time of peak enhancement can be used to calculate the optimum timing for each phase of acquisition.
With the bolus tracking technique, low-dose continuous monitoring scans are obtained after the administration of 150–200 mL of contrast material (concentration, 350 mg of iodine per milliliter) with automatic power injectors at a rate of 4–6 mL/sec through an 18- or 20-gauge intravenous catheter placed in an antecubital vein. A region of interest of 40–50 mm2 and covering about 70% of the cross-sectional area of the aorta is positioned on the abdominal aorta, again typically at about the level of the celiac axis, and scanning is then triggered when a predefined attenuation level (in Hounsfield units) is reached.
With 16–64-section scanners, the use of a fixed empiric delay is unlikely to allow the capture of peak parenchymal enhancement in all patients, since acquisition times in the pancreas are only about 3–4 seconds. Consequently, accurate bolus timing becomes critical for proper pancreatic imaging, and we do not favor the use of an empiric approach.
Many factors influence maximal pancreatic parenchymal enhancement, including technical factors such as the generation of CT scanner; contrast material–related factors like volume and concentration of iodine injected and the rate of injection; and patient factors such as age, weight, and cardiac output. Hence, there is wide intraindividual variability as to when peak parenchymal enhancement is achieved, although the use of multiphase imaging should allow differentiation between tumor and normal parenchyma in a greater number of patients. Investigators are divided as to the optimal timing for imaging of the pancreatic parenchyma, since heterogeneous techniques have been applied in the past. Table 1 summarizes the literature in this regard. The technique used at our institution (64-section volumetric CT) is described below.
|
Scanning Protocol at Our Institution
|
|---|
Our protocol makes use of rapid bolus administration as described in the previous section, with automatic power injectors from Envision (Medrad, Indianola, Pa). The injected vein is initially tested with 30 mL of saline solution to ensure proper function. Patients drink 500 mL of whole milk, 4% fat milk, or water as an oral contrast agent prior to the study, which reduces artifact during postprocessing. We routinely use bolus tracking, positioning a region of interest on the abdominal aorta at the level of the celiac axis and using an enhancement threshold of 150 HU.
The triple phase acquisition includes a nonenhanced phase, a late arterial phase (10-second delay from the time of peak aortic enhancement), and a portal venous phase (35-second delay). Contrast-enhanced acquisitions are performed craniocaudally with thin collimation (0.5 or 0.625 mm), 120 kVp, and automatic modulation of the milliamperes. Five-millimeter axial images, as well as 5 x 5-mm coronal and sagittal reformatted images, are sent to the picture archiving and communication system, and 0.5-mm-thick sections are sent to workstations for tailored postprocessing.
Late Arterial (Pancreatic) Phase
The late arterial phase of imaging closely corresponds to the so-called pancreatic phase of parenchymal enhancement, during which there is maximal conspicuity of tumors, which are hypovascular, with maximization of attenuation differences between the tumor and the pancreas. This phase also allows adequate arterial and mesenteric venous enhancement for the detection of vascular invasion (12,13). In our experience, which is similar to that of others (7,8), this phase of imaging also provides excellent enhancement of the arterial system, so that we do not routinely perform a pure arterial phase study. However, on the rare occasion when advanced rendering of the arteries is specifically required preoperatively, we will perform an additional early arterial phase study for CT arteriography.
Portal Venous Phase
The portal venous phase of imaging is the ideal phase for detecting liver metastases, as well as for creating reconstructed images of the venous structures, which might be essential for surgical planning. It also provides a second look at the pancreas, which can occasionally be useful.
|
Image Postprocessing
|
|---|
State-of-the-art multidetector CT scanners, with their thin collimation, which makes isotropic voxels a reality, have increased the importance of reformatted images in the preoperative evaluation of pancreatic tumors. A variety of postprocessing techniques might be used for different purposes, including multiplanar reformation (MPR), maximum intensity projection (MIP), minimum intensity projection (mIP), volume rendering (VR), and curved planar reformation.
Recent reports have shown that the use of high-quality MPR images provides more accurate information regarding vascular invasion and resectability of pancreatic carcinoma (14,15). Other reports have shown the usefulness of curved planar reformatted images (16,17).
At our institution, for the preoperative evaluation of pancreatic tumors, we routinely evaluate axial images, coronal and sagittal MPR images, vascular MIP or VR reformatted images from data obtained during both the pancreatic parenchymal and portal venous phases, and curved MPR images obtained along the planes of the great vessels. We also reconstruct curved MPR mIP images of the hepatobiliary system. Vascular reformatted images and thick-slab oblique images are routinely prepared by technologists in the three-dimensional imaging laboratory and viewed on the picture archiving and communication system by the radiologists. When additional image processing is required, it is usually performed by technologists after a discussion with radiologists. Advanced processing including surgical simulation can also be performed, allowing the prediction of which vessels cross the anticipated surgical field.
|
Diagnosis of Pancreatic Adenocarcinoma
|
|---|
With the protocols described earlier, the normal pancreas shows vigorous enhancement during the pancreatic parenchymal phase (Fig 1), and imaging during this phase maximizes attenuation differences between the hypovascular tumor and the surrounding parenchyma (Fig 2). Usually, the tumor can be clearly seen against the enhanced background pancreatic parenchyma, and secondary signs of malignancy such as pancreatic ductal or biliary dilatation or vascular occlusion can also be used to aid in tumor localization. About 10% of pancreatic adenocarcinomas are isoattenuating relative to the background pancreatic parenchyma, making diagnosis more difficult, and in such cases these secondary signs can be extremely useful (Fig 3) (18). The "double duct sign," in which a tumor obstructs both the pancreatic and contiguous parts of the intrahepatic common bile duct, is a reliable indicator of an obstructing lesion, although it is not specific for pancreatic adenocarcinoma (19).
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Coronal reformatted pancreatic parenchymal phase image shows intense enhancement of the normal pancreas. Note the excellent enhancement of the common hepatic artery (arrow) and the superior mesenteric artery (SMA) (arrowhead).
|
|
View larger version (133K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Contrast-enhanced CT scan shows a large, locally unresectable adenocarcinoma of the pancreatic head (arrow). Note the difference in attenuation between the tumor and the avidly enhancing normal pancreas.
|
|
View larger version (98K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Curved MPR mIP image shows abrupt cut-off of the pancreatic duct (arrow) in the region of the neck of the pancreas. No mass was visualized at CT. Exploratory surgery showed the finding to represent an isoattenuating pancreatic adenocarcinoma. Analysis of images to look for secondary signs of pancreatic carcinoma such as duct or vessel cut-off is important, since not all tumors will be directly visualized.
|
|
In practice, patients with pancreatic adenocarcinoma frequently undergo endoscopic retrograde cholangiopancreatography prior to presenting for CT, and indwelling stents as well as any postprocedural inflammation can significantly limit the evaluation of pancreatic malignancy.
Although radiologists do not frequently use the TNM staging system when issuing reports, we believe that there is merit in using this system as a basis for reporting.
|
Staging of Pancreatic Cancers
|
|---|
Pancreatic cancers are staged according to the TNM system, which is endorsed by the American Joint Committee on Cancer (AJCC) and the Union Internationale Contre le Cancer (UICC) (Tables 2, 3).
Some centers still use the Japanese Pancreatic Society system for lymph node staging, which numbers the lymph nodes, whereas the AJCC-UICC system groups lymph nodes according to their anatomic location. In current practice, radiologic staging of pancreatic cancer usually focuses on identifying disease that is locally resectable while excluding distant metastases. Most patients in whom resection appears to be viable at radiologic assessment will undergo laparoscopy prior to surgical exploration to rule out small peritoneal implants or liver disease, the presence of which precludes curative resection.
T Staging
Approximately 90% of pancreatic adenocarcinomas manifest as a focal mass, with the remainder manifesting as more diffuse involvement. Radiologic imaging is highly sensitive for assessment of the T stage. T1 and T2 tumors are distinguished on the basis of size, and this assessment can usually be accurately made, particularly when multi-planar images are used (Fig 4). Pancreatic adenocarcinoma typically manifests as a lesion that is somewhat hypoattenuating relative to the normally enhancing pancreatic parenchyma. This decreased tumor enhancement is caused by fibroblastic proliferation and decreased vascularity. T3 disease is defined as extension into the peripancreatic soft tissues, without invasion into the stomach, colon, celiac axis, or SMA, which invasion characterizes a T4 tumor. Again, reference to MPR images is useful when evaluating for local disease extension (Fig 5), which can usually be detected easily at CT. Lymphatic channels within the pancreas drain into lymph nodes around the celiac axis and the superior mesenteric vein (SMV). As these lymphatic vessels become engorged with tumor cells, they produce a perivascular cuff of soft tissue that represents extrapancreatic T4 disease (Fig 6). Occasionally, this perivascular cuff is the only radiologic sign of a pancreatic carcinoma. Such vascular invasion, with involvement of the celiac axis or SMA (T4 tumor), renders the disease unresectable.
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4. Coronal reformatted pancreatic parenchymal phase image shows a focal hypoattenuating tumor (arrow). Whipple resection revealed that the tumor was entirely confined to the pancreas and represented a T2 tumor. The vigorous enhancement seen in the pancreas during the pancreatic parenchymal phase facilitates depiction of the tumor.
|
|
View larger version (131K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. (a) Pancreatic CT angiogram shows a large exophytic mass (arrows) arising from the neck of the pancreas and invading the stomach (arrowhead), a finding that represents a radiologic T3 tumor (ie, extension into the peripancreatic tissues that does not involve either the celiac axis or the SMA). The patient did not undergo surgery because metastatic disease was present elsewhere. (b) Coronal reformatted pancreatic parenchymal phase image obtained in a different patient with T3 pancreatic adenocarcinoma delineates the inferior extrapancreatic extent of the tumor (arrows).
|
|
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. (a) Pancreatic CT angiogram shows a large exophytic mass (arrows) arising from the neck of the pancreas and invading the stomach (arrowhead), a finding that represents a radiologic T3 tumor (ie, extension into the peripancreatic tissues that does not involve either the celiac axis or the SMA). The patient did not undergo surgery because metastatic disease was present elsewhere. (b) Coronal reformatted pancreatic parenchymal phase image obtained in a different patient with T3 pancreatic adenocarcinoma delineates the inferior extrapancreatic extent of the tumor (arrows).
|
|
View larger version (133K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6. Contrast-enhanced CT scan obtained in a patient with pancreatic adenocarcinoma shows a cuff of soft tissue (arrows) surrounding the celiac axis. Such a cuff is thought to represent lymphovascular invasion and defines a tumor as a T4 lesion. Involvement of either the celiac axis or the SMA renders a tumor unresectable.
|
|
T1–T3 tumors are all potentially resectable, although the chance of effecting a surgical cure lessens with more advanced disease. As described earlier, pancreatic adenocarcinoma infiltrates lymphatic vessels early, and local infiltrative disease can manifest as subtle infiltration of peripancreatic tissue. This local invasion can cause underestimation of the true extent and stage of the tumor and is a cause of aborted surgical resection if not identified preoperatively.
N Staging
Previous work has shown that CT is not accurate in the assessment of nodal involvement in pancreatic ductal adenocarcinoma; therefore, in patients with an otherwise resectable tumor, the CT depiction of enlarged peripancreatic nodes should not preclude resection with curative intent (20,21). Nevertheless, prognosis is directly related to lymph node involvement (22), and extensive, meticulous, and comprehensive surgical lymph node staging is an integral part of the assessment of disease burden. The classification of lymph node distribution according to a name-based (AJCC-UICC) or number-based (Japanese Pancreatic Society) system has been shown to allow more accurate pathologic dissection of lymph node groups and to improve subsequent lymph node yield, thereby allowing more accurate staging (23). For pancreatic adenocarcinoma, the UICC recommendation is for a minimum yield of 10 lymph nodes. Hence, although CT cannot accurately help predict whether one or more lymph nodes are involved by tumor, preoperative assessment of tumor should include a description of the distribution of visualized nodes. This information can be used as an intraoperative guide by the surgeon, since different surgical techniques can be used to improve lymph node yield (24). More important, it can be used by the pathologist to facilitate optimal pathologic dissection and maximize lymph node yield.
At our institution, we use the name-based AJCC-UICC system for describing lymph node distribution. This system divides lymph nodes into four categories, depending on whether they are (a) superior to the body and head of the pancreas; (b) anterior, including anterior pancreaticoduodenal, pyloric, and proximal mesenteric; (c) inferior to the body and head of the pancreas; or (d) posterior, including posterior pancreaticoduodenal, common bile duct, and proximal mesenteric (Fig 7).
The Japanese Pancreatic Society further subdivides these lymph nodes and names 18 individual sites, but this approach has been shown by at least one retrospective review not to improve yield (23). The use of MPR images is invaluable in assigning individual lymph nodes to a likely group (Fig 8).
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8a. (a) Contrast-enhanced CT scan shows a large pancreatic adenocarcinoma (arrowhead) as well as multiple surrounding peripancreatic lymph nodes (arrows). (b) Sagittal reformatted image clearly delineates the inferior extent of the retroperitoneal lymphadenopathy (arrow). Subtle strands of tumor extend into the gastrocolic ligament (arrowheads). By referring to MPR images, the radiologist can get a clear picture of lymph node distribution. MPR images are also invaluable in assigning a T stage to a tumor.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8b. (a) Contrast-enhanced CT scan shows a large pancreatic adenocarcinoma (arrowhead) as well as multiple surrounding peripancreatic lymph nodes (arrows). (b) Sagittal reformatted image clearly delineates the inferior extent of the retroperitoneal lymphadenopathy (arrow). Subtle strands of tumor extend into the gastrocolic ligament (arrowheads). By referring to MPR images, the radiologist can get a clear picture of lymph node distribution. MPR images are also invaluable in assigning a T stage to a tumor.
|
|
The hepatic artery node, which is important in the assessment of pancreatic adenocarcinoma, should be considered part of the celiac drainage basin, and malignancy in this node cannot be determined on the basis of size at radiologic imaging (25). However, when this node is positive for malignancy, the survival rate is as poor as for liver metastases or peritoneal carcinomatosis (25). This node is consistently located at the site of take-off of the gastroduodenal artery, and any obvious nodal enlargement at the site should raise some suspicion.
M Staging
As mentioned earlier, CT allows global disease staging, since it affords extensive anatomic coverage in a single session. Nevertheless, CT remains limited in its capacity to help detect small liver metastases (Fig 9) and peritoneal deposits (Fig 10), which can cause upstaging of the patient’s condition to stage IV disease. It is recommended that particular attention be paid to these areas at staging CT. Laparoscopy is often performed prior to formal laparotomy to evaluate the peritoneum and can also be combined with laparoscopic US to evaluate the liver for focal metastases. Distant metastases preclude curative resection, but patients may undergo surgical bypass procedures. Screening CT of the lung is not routinely recommended for the preoperative evaluation of patients with pancreatic carcinoma because lung metastases are rare in patients with no other contraindications for resection (26).
View larger version (138K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9. Contrast-enhanced portal venous phase CT scan demonstrates a focal lesion in segment V of the liver (arrow). Note also the biliary stent. Subsequent biopsy showed the lesion to represent a metastasis.
|
|
View larger version (128K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10. Contrast-enhanced portal venous phase image shows an approximately 1-cm nodule on the omentum (arrow), a finding that proved to be a metastatic deposit at subsequent laparoscopy. Arrowhead indicates the primary (pancreatic) tumor.
|
|
Although in our experience CT is well suited for both tumor detection and assessment of resectability, endoscopic US would appear to play a complementary role in tumor assessment and is probably the optimal method for lymph node staging, since it allows sampling of any suspect lymph nodes that are present. Comparative studies suggest that when preoperative staging is performed with both multidetector CT and endoscopic US, with one of the modalities being used for initial screening and the other in potentially resectable cases, the two modalities play complementary roles (27,28). One recent study has suggested that endoscopic US may be even better than CT at helping predict local unresectability (29). This study used only single-section CT, and the reported accuracy for correctly helping predict resectability (60%) was significantly less than that reported elsewhere in the literature.
|
Assessment of Vascular Resectability
|
|---|
In addition to diagnosing and staging a pancreatic adenocarcinoma, the radiologist can and should define the relationship of the tumor to critical arterial and venous structures, since their involvement can preclude resection. Multidetector CT has been used extensively to predict resectability, and various grading systems based on CT findings have been devised (30,31). However, tumor resectability does not equate to patient cure, and an oncologically unsound (R1 or R2) resection in which either microscopic or gross disease is left behind provides little or no survival benefit compared with palliative bypass surgery.
Venous Invasion
Before the advent of CT, potential resectability was decided on the basis of the venous phase of SMA angiographic findings, with bilateral waist-like narrowing on axial images indicating unresectability. This extension to the left side of the SMV implied SMA extension and, hence, an unresectable T4 tumor. With cross-sectional imaging, the relationship between the tumor and the superior mesenteric vessels can be assessed more accurately. Combining axial source data with VR and curved MPR images has been suggested as the optimal method for evaluating for unresectability (15,16). Specific signs of unresectability include a circumferentially narrowed or occluded SMV or portal vein (PV) (Figs 11–13). Other CT criteria that have been shown to suggest tumor involvement include increasing degrees of circumferential venous involvement by tumor; the "teardrop" mesenteric vein sign (Fig 14); and the presence of small, dilated peripancreatic veins (32). The teardrop mesenteric vein sign was described by Hough et al (31) as a specific sign of tumor involvement of the SMV. It refers to a focally tethered, teardropshaped SMV, which was originally reported to be a reliable sign for predicting the unresectability of adenocarcinoma of the pancreatic head and can contribute significantly to preoperative planning.
View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11. Contrast-enhanced pancreatic parenchymal phase image shows a low-attenuation cuff (small arrows) surrounding a narrowed SMV. A tumor (large arrow) is seen extending into the peripancreatic transverse mesocolon anteriorly. Laparotomy showed the tumor to be unresectable.
|
|
View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12. Coronal MIP image shows a long narrowed segment (arrow) of the SMV extending into the superior mesentericoportovenous complex and the proximal PV. SV = splenic vein. Use of MIP and VR images allows accurate assessment of the extent of venous involvement and is particularly favored by surgeons when a venous bypass procedure is being considered.
|
|
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13a. (a) Contrast-enhanced CT scan shows a pancreatic mass (arrowhead) causing severe flattening of the SMV (large arrow). Note also the cuff of soft tissue (small arrows) surrounding the celiac axis. (b) Coronal reformatted image clearly depicts the degree of venous deformity and the low-attenuation mass (arrow), which was found to represent an unresectable tumor.
|
|
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13b. (a) Contrast-enhanced CT scan shows a pancreatic mass (arrowhead) causing severe flattening of the SMV (large arrow). Note also the cuff of soft tissue (small arrows) surrounding the celiac axis. (b) Coronal reformatted image clearly depicts the degree of venous deformity and the low-attenuation mass (arrow), which was found to represent an unresectable tumor.
|
|
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14. Contrast-enhanced thin-section pancreatic parenchymal phase CT scan shows eccentric deformity of the SMV (arrow) caused by tumor fronds extending from a tumor in the pancreatic head. This eccentric pattern of deformity has been likened to a teardrop and is frequently seen in cases (such as this one) of locally unresectable tumor.
|
|
When assessing axial images, we use the grading system proposed by Raptopoulos et al (31): grade 0: normal, with a fat plane or normal pancreas between the tumor and vessel; grade 1: loss of the fat plane between the tumor and vessel, with or without smooth displacement of the vessel; grade 2: flattening or slight irregularity of one side (ie, < 180° of the perimeter) of the vessel, a finding that is associated with questionable venous involvement; grade 3: encased vessel with tumor extending around at least two sides (ie, > 180° of the perimeter), altering vessel contour and producing concentric or eccentric narrowing of the lumen (associated with definite tumor invasion but does not necessarily exclude resection because en bloc venous resection may still be possible); and grade 4: at least one occluded major vessel (mostly unresectable, although a short-segment focal venous occlusion with superior and inferior patency may allow local resection).
Individual assessment of the PV, splenic vein, and SMV as well as of the major trunks of the latter vessel (Fig 15) should be systematically made when assessing the mesentericoportal veins, using a combination of axial and reformatted images as described earlier.
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15. Pancreatic parenchymal phase CT angiogram shows good enhancement of the SMV. A hypoattenuating tumor is seen extending along the right side of the SMV that anteriorly cannot be separated from the gastrocolic trunk (arrow). Surgery revealed that the tumor could not be resected due to vascular invasion.
|
|
Over the past few years, significant attention has been focused on expanding the scope of surgery by performing en bloc venous resection of the SMV-PV confluence when this entity is the only potential barrier to resection (33). Although a few reports of arterial resection exist, SMA or celiac axis involvement remains a relative contra-indication for resection because the mesenteric neural plexus is also frequently involved and the ensuing resection would not only be technically difficult but probably oncologically unsound. It should be stressed that en bloc vascular resection is performed, not to increase the local soft-tissue resection volume or to allow extended lymphadenectomy, but in an effort to achieve a negative resection margin in a patient population in whom an R0 resection would otherwise be impossible. To this end, there is evidence in the literature that segmental en bloc SMV-PV resection can be beneficial in selected patients (34). Studies have shown that CT can be used to predict the need for venous resection (35), and if venous resection is anticipated, it is important to discuss each individual case in detail with the surgeon. Multiple strategies for revascularization exist, and the surgeon should be apprised of the position of critical venous structures such as the gastrocolic trunk, the inferior mesenteric vein, and the first jejunal branch, as well as their respective distances from each other, the tumor, and the mesentericoportal confluence. In this way, likely "take-off" and "landing" zones for bypass grafts can be selected. Rendering of volumetric data is invaluable in these cases. Although each case must be systematically evaluated individually, it is important to pay particular attention to the relationship of the tumor to the SMV-PV confluence. The intact confluence is not readily mobile, so that end-to-end anastomosis between it and the SMV cannot be performed if more than 2 cm of the SMV is removed. In these cases, it is usually better to conserve the SMV-PV anastomosis and insert an interposition graft than to excise the confluence and the splenic vein, since the latter maneuver can lead to sinistral portal hypertension with upper gastrointestinal variceal hemorrhage. If the SMV-PV confluence is involved by tumor, however, extensive resection with ligation of the splenic vein may be required.
Arteries
Assessment of unresectability based on arterial involvement is also made using a combination of axial source images and postprocessed rendered images. As was previously discussed, involvement or encasement of the celiac trunk, hepatic artery, proximal gastroduodenal artery, or SMA makes a tumor unresectable. With axial images, a periarterial cuff of soft tissue can be directly visualized.
MIP and VR images cannot depict a perivascular cuff of soft tissue; hence, the diagnosis is based on changes in caliber. Postprocessed images do, however, allow better assessment of the extent of involvement than do axial images (Fig 16). Optimal results are achieved when axial source images and postprocessed images are combined (31). Again, grading systems such as those proposed earlier (30,31) can be used to quantify circumferential arterial encasement by tumor and to predict the likelihood of successful resection.
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16a. (a) Contrast-enhanced CT scan shows a cuff of soft tissue (arrows) surrounding the origin of the common hepatic artery from the celiac axis. (b) VR image demonstrates circumferential narrowing of the origins of the common hepatic artery and splenic artery (arrows).
|
|
View larger version (108K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16b. (a) Contrast-enhanced CT scan shows a cuff of soft tissue (arrows) surrounding the origin of the common hepatic artery from the celiac axis. (b) VR image demonstrates circumferential narrowing of the origins of the common hepatic artery and splenic artery (arrows).
|
|
The specific circumstances under which curative resection will be attempted vary by both institution and country, with more extensive surgeries favored in Japan and Europe than in the United States. Table 4 shows the criteria that are used at the M.D. Anderson Cancer Center in Houston, Texas, which are similar to those in place at our institution.
Performance of CT
With advances in CT technology, the capacity of CT to help predict resectability has increased. In one single-center retrospective study, the rates of aborted Whipple procedure were studied during three discrete time periods that coincided with the CT technology then available (1993–1996, spiral CT; 1996–2001, spiral CT angiography; 2001–2004, multidetector CT angiography). The study showed that the chances that CT will correctly help predict resectability are 7.47 times higher with multidetector CT angiography than with spiral CT (36). False-negative results most often occur because of unsuspected liver surface metastases, peritoneal deposits, or unsuspected vascular invasion.
The positive predictive values of helical CT for unresectability range between 89% and 100%, with accuracies of 85%–95% (30,37). However, the positive predictive values for resectability are much lower (45%–79%) (37–41).
Relative to helical CT, helical CT angiography has been shown to provide additional information, thereby improving staging for resectability of pancreatic tumors. In a retrospective study, the predictive value for vascular resectability was 96% for axial helical CT with CT angiography and only 70% for axial helical CT alone (P = .021) (31). Multidetector CT is being used with increasing frequency, especially in the preoperative evaluation of oncology patients. Four-section CT has been reported to have excellent negative predictive value (100%) for vascular invasion and good negative predictive value (87%) for overall tumor resectability in patients with pancreatic adenocarcinoma (14,16).
More recent research suggests that there is no difference in the diagnostic accuracy for preoperative staging of pancreatic adenocarcinoma across the different generations of multidetector CT scanners (42).
Preoperative Planning
Pancreatic CT angiography can be used, not only for the detection and characterization of tumors and for local and distant tumor staging, but also as a fundamental tool for surgical planning. Preoperative knowledge of the normal vascular anatomy and of the presence of variants relevant to the Whipple procedure is crucial: This information can be obtained with CT angiography, and advanced rendering techniques can be used to create displays that are familiar to the surgeon (Fig 17).
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 17. VR image from multidetector CT data shows an aberrant origin of the right hepatic artery from the SMA (arrow). This common variant should be sought out because the aberrant artery will pass posterior to the PV and could be inadvertently ligated during cystic duct exploration.
|
|
|
Conclusions
|
|---|
Multidetector (64-section) volumetric CT allows comprehensive preoperative assessment of pancreatic adenocarcinoma. Carefully timed scan acquisition maximizes the difference in attenuation between the neoplasm and the pancreatic parenchyma and allows accurate local and distant staging as well as assessment of local resectability. In addition, angiographic data sets can be rendered to create displays of the local venous and arterial anatomy that are familiar to surgeons. Advanced rendering can also be used to create pancreaticographic type images.
|
Footnotes
|
|---|
Abbreviations: AJCC = American Joint Committee on Cancer, MIP = maximum intensity projection, mIP = minimum intensity projection, MPR = multiplanar reformation, PV = portal vein, SMA = superior mesenteric artery, SMV = superior mesenteric vein, UICC = Union Internationale Contre le Cancer, VR = volume-rendered
|
References
|
|---|
- Sahmoun AE, D’Agostino RA Jr, Bell RA, Schwenke DC. International variation in pancreatic cancer mortality for the period 1955–1998. Eur J Epidemiol 2003;18:801–816.[CrossRef][Medline]
- Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2006. CA Cancer J Clin 2006;56:106–130.[Abstract/Free Full Text]
- American Cancer Society. Cancer facts and figures 2004. Atlanta, Ga: American Cancer Society, 2004.
- Cooperman AM, Kini S, Snady H, Bruckner H, Chamberlain RS. Current surgical therapy for carcinoma of the pancreas. J Clin Gastroenterol 2000; 31:107–113.[CrossRef][Medline]
- Wray CJ, Ahmad SA, Matthews JB, Lowy AM. Surgery for pancreatic cancer: recent controversies and current practice. Gastroenterology 2005;128: 1626–1641.[CrossRef]
- U.S. Preventive Services Task Force. Guide to clinical preventive services, 2006. Rockville, Md: Agency for Healthcare Research and Quality, 2006.
- Fletcher JG, Wiersema MJ, Farrell MA, et al. Pancreatic malignancy: value of arterial, pancreatic, and hepatic phase imaging with multi–detector row CT. Radiology 2003;229:81–90.[Abstract/Free Full Text]
- McNulty NJ, Francis IR, Platt JF, Cohan RH, Korobkin M, Gebremariam A. Multi–detector row helical CT of the pancreas: effect of contrast-enhanced multiphasic imaging on enhancement of the pancreas, peripancreatic vasculature, and pancreatic adenocarcinoma. Radiology 2001;220:97–102.[Abstract/Free Full Text]
- Graf O, Boland GW, Warshaw AL, Fernandez-del-Castillo C, Hahn PF, Mueller PR. Arterial versus portal venous helical CT for revealing pancreatic adenocarcinoma: conspicuity of tumor and critical vascular anatomy. AJR Am J Roentgenol 1997;169:119–123.[Abstract/Free Full Text]
- Hollett MD, Jorgensen MJ, Jeffrey RB Jr. Quantitative evaluation of pancreatic enhancement during dual-phase helical CT. Radiology 1995;195: 359–361.[Abstract/Free Full Text]
- Keogan MT, McDermott VG, Paulson EK, Sheafor DH, Frederick MG, de Long DM, Nelson RC. Pancreatic malignancy: effect of dual-phase helical CT in tumor detection and vascular opacification. Radiology 1997;205:513–518.[Abstract/Free Full Text]
- Lu DS, Vedantham S, Krasny RM, Kadell B, Berger WL, Reber HA. Two-phase helical CT for pancreatic tumors: pancreatic versus hepatic phase enhancement of tumor, pancreas, and vascular structures. Radiology 1996;199:697–701.[Abstract/Free Full Text]
- Boland GW, O’Malley ME, Saez M, Fernandez-del-Castillo C, Warshaw AL, Mueller PR. Pancreatic-phase versus portal vein-phase helical CT of the pancreas: optimal temporal window for evaluation of pancreatic adenocarcinoma. AJR Am J Roentgenol 1999;172:605–606.[Abstract/Free Full Text]
- Catalano C, Laghi A, Fraioli F, et al. Pancreatic carcinoma: the role of high-resolution multislice spiral CT in the diagnosis and assessment of resectability. Eur Radiol 2003;13:149–156.[Medline]
- Brugel M, Link TM, Rummeny EJ, Lange P, Theisen J, Dobritz M. Assessment of vascular invasion in pancreatic head cancer with multislice spiral CT: value of multiplanar reconstructions. Eur Radiol 2004;14:1188–1195.[Medline]
- Vargas R, Nino-Murcia M, Trueblood W, Jeffrey RB Jr. MDCT in pancreatic adenocarcinoma: prediction of vascular invasion and resectability using a multiphasic technique with curved planar reformations. AJR Am J Roentgenol 2004;182:419–425.[Abstract/Free Full Text]
- Fukushima H, Itoh S, Takada A, et al. Diagnostic value of curved multiplanar reformatted images in multislice CT for the detection of resectable pancreatic ductal adenocarcinoma. Eur Radiol 2006; 16:1709–1718.[CrossRef][Medline]
- Prokesch RW, Chow LC, Beaulieu CF, Bammer R, Jeffrey RB Jr. Isoattenuating pancreatic adenocarcinoma at multi–detector row CT: secondary signs. Radiology 2002;224:764–768.[Abstract/Free Full Text]
- Plumley TF, Rohrmann CA, Freeny PC, Silverstein FE, Ball TJ. Double duct sign: reassessed significance in ERCP. AJR Am J Roentgenol 1982;138:31–35.[Abstract/Free Full Text]
- Roche CJ, Hughes ML, Garvey CJ, et al. CT and pathologic assessment of prospective nodal staging in patients with ductal adenocarcinoma of the head of the pancreas. AJR Am J Roentgenol 2003; 180:475–480.[Abstract/Free Full Text]
- Tamm EP, Silverman PM, Charnsangavej C, Evans DB. Diagnosis, staging, and surveillance of pancreatic cancer. AJR Am J Roentgenol 2003; 180:1311–1323.[Free Full Text]
- Geer RJ, Brennan MF. Prognostic indicators for survival after resection of pancreatic adenocarcinoma. Am J Surg 1993;165:68–73.[CrossRef][Medline]
- Chaudhry IH, Campbell F. An audit of pathology lymph node dissection techniques in pylorus preserving Kausch-Whipple pancreatoduodenectomy specimens. J Clin Pathol 2001;54:758–761.[Abstract/Free Full Text]
- Jones L, Russell C, Mosca F, et al. Standard Kausch-Whipple pancreatoduodenectomy. Dig Surg 1999;16:297–304.[CrossRef][Medline]
- Maithel SK, Khalili K, Dixon E, et al. Impact of regional lymph node evaluation in staging patients with periampullary tumors. Ann Surg Oncol 2007; 14:202–210.[Abstract/Free Full Text]
- Nordback I, Saaristo R, Piironen A, Sand J. Chest computed tomography in the staging of pancreatic and periampullary carcinoma. Scand J Gastroenterol 2004;39:81–86.[CrossRef][Medline]
- Soriano A, Castells A, Ayuso C, et al. Preoperative staging and tumor resectability assessment of pancreatic cancer: prospective study comparing endoscopic ultrasonography, helical computed tomography, magnetic resonance imaging, and angiography. Am J Gastroenterol 2004;99:492–501.[CrossRef][Medline]
- Yovino S, Darwin P, Daly B, Garofalo M, Moesinger R. Predicting unresectability in pancreatic cancer patients: the additive effects of CT and endoscopic ultrasound. J Gastrointest Surg 2007;11: 36–42.[CrossRef][Medline]
- Kala Z, Valek V, Hlavsa J, Hana K, Vanova A. The role of CT and endoscopic ultrasound in pre-operative staging of pancreatic cancer. Eur J Radiol 2007;62:166–169.[CrossRef][Medline]
- Lu DS, Reber HA, Krasny RM, Kadell BM, Sayre J. Local staging of pancreatic cancer: criteria for unresectability of major vessels as revealed by pancreaticphase, thin-section helical CT. AJR Am J Roentgenol 1997;168:1439–1443.[Abstract/Free Full Text]
- Raptopoulos V, Steer ML, Sheiman RG, Vrachliotis TG, Gougoutas CA, Movson JS. The use of helical CT and CT angiography to predict vascular involvement from pancreatic cancer: correlation with findings at surgery. AJR Am J Roentgenol 1997;168:971–977.[Abstract/Free Full Text]
- Hough TJ, Raptopoulos V, Siewert B, Matthews JB. Teardrop superior mesenteric vein: CT sign for unresectable carcinoma of the pancreas. AJR Am J Roentgenol 1999;173:1509–1512.[Abstract]
- Fuhrman GM, Leach SD, Staley CA, et al. Rationale for en bloc vein resection in the treatment of pancreatic adenocarcinoma adherent to the superior mesenteric-portal vein confluence. Pancreatic Tumor Study Group. Ann Surg 1996;223:154–162.[CrossRef][Medline]
- Bachellier P, Nakano H, Oussoultzoglou PD, et al. Is pancreaticoduodenectomy with mesentericoportal venous resection safe and worthwhile? Am J Surg 2001;182:120–129.[CrossRef][Medline]
- Bold RJ, Charnsangavej C, Cleary KR, et al. Major vascular resection as part of pancreaticoduodenectomy for cancer: radiologic, intraoperative, and pathologic analysis. J Gastrointest Surg 1999;3: 233–243.[CrossRef][Medline]
- Raptopoulos V, Kruskal JB, Gkogkas C, Katsikani A, Baptista J, Siewert B. Multidetector CT angiography in pre-operative planning for pancreatic adenocarcinoma. Presented at the 15th annual meeting and postgraduate course of ESGAR [European Society of Gastrointestinal and Abdominal Radiology], Geneva, Switzerland, June 15–18, 2004.
- Gouma DJ, van Geenen RC, van Gulik TM, et al. Rates of complications and death after pancreaticoduodenectomy: risk factors and the impact of hospital volume. Ann Surg 2000;232:786–795.[CrossRef][Medline]
- Freeny PC, Traverso LW, Ryan JA. Diagnosis and staging of pancreatic adenocarcinoma with dynamic computed tomography. Am J Surg 1993; 165:600–606.[CrossRef][Medline]
- Bluemke DA, Cameron JL, Hruban RH, et al. Potentially resectable pancreatic adenocarcinoma: spiral CT assessment with surgical and pathologic correlation. Radiology 1995;197:381–385.[Abstract/Free Full Text]
- Tabuchi T, Itoh K, Ohshio G, et al. Tumor staging of pancreatic adenocarcinoma using early and late-phase helical CT. AJR Am J Roentgenol 1999;173:375–380.[Abstract/Free Full Text]
- Warshaw AL, Fernandez-del Castillo C. Pancreatic carcinoma. N Engl J Med 1992;326:455–465.[Medline]
- Zamboni G, Kruskal JB, Vollmer CM, Baptista J, Callery MP, Raptopoulos V. Value of MDCT angiography in the preoperative evaluation of pancreatic adenocarcinoma. Radiology 2007;245 (in press).
Top
Abstract
Introduction
