HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Brennan, D. D. Articles by Kruskal, J. B. Search for Related Content PubMed PubMed Citation Articles by Brennan, D. D. Articles by Kruskal, J. B. Social Bookmarking                      What's this? Hotlight (NEW!) What's Hotlight?

Virtual Whipple: Preoperative Surgical Planning with Volume-Rendered MDCT Images to Identify Arterial Variants Relevant to the Whipple Procedure

¹ Department of Radiology, Beth Israel Deaconess Medical Center, 1 Deaconess Rd., Boston, MA 02215.
² Department of Surgery, Beth Israel Deaconess Medical Center, Boston, MA.

Received July 24, 2006; accepted after revision November 8, 2006.

 
Address correspondence to D. D. Brennan.

G. Zamboni is the recipient of an educational grant from Toshiba America.

WEB This is a Web exclusive article.

Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
OBJECTIVE. The purposes of this study were to combine a thorough understanding of the technical aspects of the Whipple procedure with advanced rendering techniques by introducing a virtual Whipple procedure and to evaluate the utility of this new rendering technique in prediction of the arterial variants that cross the anticipated surgical resection plane.

CONCLUSION. The virtual Whipple is a novel technique that follows the complex surgical steps in a Whipple procedure. Three-dimensional reconstructed angiographic images are used to identify arterial variants for the surgeon as part of the preoperative radiologic assessment of pancreatic and ampullary tumors.

Keywords: CT angiography • CT arteriography • MDCT • pancreaticobiliary imaging

Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Pancreatic adenocarcinoma is the fourth leading cause of cancer death in the United States and the sixth in Europe. The American Cancer Society estimates a 29,000 cases per year incidence and a nearly 28,000 cases per year mortality [1]. Approximately 5-30% of patients with pancreatic cancer appear to have the tumor contained entirely within the pancreas at diagnosis. Surgical resection is recommended for this group of patients and provides the best option for long-term survival [2].

The Whipple procedure, or pancreatico-duodenectomy, is a complex surgical procedure used to resect tumors of the head of the pancreas, distal common bile duct, and duodenum. There is considerable anatomic variability in the arterial supply to this region, and preoperative knowledge of the variants is important. Although conventional axial imaging has been used in preoperative planning, the use of 3D reconstruction augments the ability to evaluate arterial anatomy. Raptopoulos et al. [3] found 3D reconstructions better than axial images for prediction of successful pancreatic resection. One of our aims was to combine thorough understanding of the technical aspects of the Whipple procedure with advanced rendering techniques by introducing a virtual Whipple procedure. The other aim was to evaluate the utility of this new rendering technique in predicting the arterial variants that cross the anticipated resection plane. We electronically simulated the steps of a Whipple operation and used the color-encoded volume-rendered angiographic images acquired to display the arterial vessels that cross the surgical plane.

Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
This study was performed in accordance with the requirements of a retrospective review. An institutional review board waiver was obtained, and no informed consent was required. We reviewed the preoperative CT angiographic (CTA) studies of all patients who underwent a Whipple procedure over a 26-month period (October 2003-December 2005) at a single academic institution.

CTA was performed with 8-, 16-, and 64-MDCT (LightSpeed Ultra, LightSpeed Pro 16, GE Healthcare; Aquilion 64, Toshiba America Medical Systems). Unenhanced, late arterial-pancreatic parenchymal phase, and early portal venous (20 seconds after the arterial phase) acquisitions were timed with a bolus-tracking technique. Numerous authors have advocated imaging in the pancreatic parenchymal phase, which is the late arterial phase of imaging that produces maximum contrast between the enhancing pancreatic parenchyma and the relatively hypoenhancing adenocarcinoma [4]. As part of a comprehensive pancreatic imaging strategy, we acquired images in this pancreatic phase and in the hepatic portal venous phase. Rate of injection, iodine concentration, and volume of injection influence peak pancreatic enhancement. The differences in acquisition times between successive classes of MDCT also influence peak pancreatic enhancement.

We used bolus-tracking techniques and identical contrast parameters to maximize reproducibility in our patient cohort over successive generations of MDCT scanners. Bolus tracking is a method of optimizing delivery of the contrast bolus so that scans are obtained during a desired phase of imaging. Generically, a region of interest is placed over an organ or a vessel of interest, and sequential low-dose scanning is performed after the initiation of contrast administration. At a given attenuation threshold, automated or manual scanning is begun, usually after a specified delay. We used an attenuation of 150 H and triggered from the aorta at the level of the celiac axis. Reports in the literature suggest that this method may be more reproducible than fixed-duration methods of contrast injection [5], although the topic continues to be debated. Vendors have various names for the technique (e.g., CareBolus, Siemens Medical Solutions; SmartPrep, GE Healthcare).

All patients were given 200 mL of contrast material with a concentration of 350 mg I/mL at a rate of 5 mL/s by automatic power injector (Envision, Medrad) through an 18- or 20-gauge IV catheter placed in an antecubital vein. Contrast-enhanced images were acquired in the craniocaudal axis according to the parameters in Table 1.

Whipple Procedure
A Whipple procedure, although complex, can be broken down into six essential maneuvers. The first requires dissection to the lesser sac and wide exposure of the pancreatic bed. Kocherization is used to physically lift the pancreas and duodenum out of the lesser sac to enable further dissection. The next step involves division of the common hepatic duct and performance of cholecystectomy. Blunt dissection is performed along the superior mesenteric vein before transection of the pancreatic head, and the head of the pancreas is removed. In a classic Whipple procedure, antrectomy is performed, and the pylorus and duodenum are removed. In a pylorus-preserving Whipple procedure, the pylorus and proximal duodenum are retained. The uncinate process is carefully dissected away from the superior mesenteric vein and removed. The final step of a Whipple procedure calls for mobilization of a loop of jejunum and construction of a pancreati-cojejunostomy, a choledochojejunostomy, and a duodenojejunostomy.

Virtual Whipple
The arterial phase of the CTA data sets was exported to an offline workstation (Vitrea 2, Vital Images) for rendering. A simple algorithm was used to identify and manually trace the margins of the gallbladder, which was isolated and electronically removed from the data set (Figs. 1A and 1B). The next few steps of the virtual Whipple procedure were performed simultaneously by manual tracing of the dissection margins of the duodenum and pancreas, including the uncinate process, and carefully avoiding the surrounding vascular structures (Fig. 1C, 1D, 1E, 1F). Careful reference to the anatomic features in three orthogonal planes allowed the user to avoid removal of nontargeted organs and vessels (Fig. 1G).

View larger version (137K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1A —44-year-old man with pancreatic adenocarcinoma. Images show sequential steps in performance of virtual Whipple procedure. Initially, margins of gallbladder are defined on sequential axial images (A) using freeform region of interest tool. Electronic sculpt function allows for subsequent electronic removal (B). Note that patient has had prior ERCP with stent placement. After electronically simulating cholecystectomy, pancreaticoduodenectomy is simulated using similar function. Reference to sagittal (C), coronal (D), and axial (E) multiplanar reformat images and volume-rendered projections (F) allow for real-time monitoring of progress and prevent inadvertent inclusion of vital structures such as mesenteric vessels. Pancreatic head, uncinate process, regional lymph nodes, and transverse mesocolon, if involved, are included and then removed electronically (G).

View larger version (143K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1B —44-year-old man with pancreatic adenocarcinoma. Images show sequential steps in performance of virtual Whipple procedure. Initially, margins of gallbladder are defined on sequential axial images (A) using freeform region of interest tool. Electronic sculpt function allows for subsequent electronic removal (B). Note that patient has had prior ERCP with stent placement. After electronically simulating cholecystectomy, pancreaticoduodenectomy is simulated using similar function. Reference to sagittal (C), coronal (D), and axial (E) multiplanar reformat images and volume-rendered projections (F) allow for real-time monitoring of progress and prevent inadvertent inclusion of vital structures such as mesenteric vessels. Pancreatic head, uncinate process, regional lymph nodes, and transverse mesocolon, if involved, are included and then removed electronically (G).

View larger version (103K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1C —44-year-old man with pancreatic adenocarcinoma. Images show sequential steps in performance of virtual Whipple procedure. Initially, margins of gallbladder are defined on sequential axial images (A) using freeform region of interest tool. Electronic sculpt function allows for subsequent electronic removal (B). Note that patient has had prior ERCP with stent placement. After electronically simulating cholecystectomy, pancreaticoduodenectomy is simulated using similar function. Reference to sagittal (C), coronal (D), and axial (E) multiplanar reformat images and volume-rendered projections (F) allow for real-time monitoring of progress and prevent inadvertent inclusion of vital structures such as mesenteric vessels. Pancreatic head, uncinate process, regional lymph nodes, and transverse mesocolon, if involved, are included and then removed electronically (G).

View larger version (98K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1D —44-year-old man with pancreatic adenocarcinoma. Images show sequential steps in performance of virtual Whipple procedure. Initially, margins of gallbladder are defined on sequential axial images (A) using freeform region of interest tool. Electronic sculpt function allows for subsequent electronic removal (B). Note that patient has had prior ERCP with stent placement. After electronically simulating cholecystectomy, pancreaticoduodenectomy is simulated using similar function. Reference to sagittal (C), coronal (D), and axial (E) multiplanar reformat images and volume-rendered projections (F) allow for real-time monitoring of progress and prevent inadvertent inclusion of vital structures such as mesenteric vessels. Pancreatic head, uncinate process, regional lymph nodes, and transverse mesocolon, if involved, are included and then removed electronically (G).

View larger version (81K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1E —44-year-old man with pancreatic adenocarcinoma. Images show sequential steps in performance of virtual Whipple procedure. Initially, margins of gallbladder are defined on sequential axial images (A) using freeform region of interest tool. Electronic sculpt function allows for subsequent electronic removal (B). Note that patient has had prior ERCP with stent placement. After electronically simulating cholecystectomy, pancreaticoduodenectomy is simulated using similar function. Reference to sagittal (C), coronal (D), and axial (E) multiplanar reformat images and volume-rendered projections (F) allow for real-time monitoring of progress and prevent inadvertent inclusion of vital structures such as mesenteric vessels. Pancreatic head, uncinate process, regional lymph nodes, and transverse mesocolon, if involved, are included and then removed electronically (G).

View larger version (106K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1F —44-year-old man with pancreatic adenocarcinoma. Images show sequential steps in performance of virtual Whipple procedure. Initially, margins of gallbladder are defined on sequential axial images (A) using freeform region of interest tool. Electronic sculpt function allows for subsequent electronic removal (B). Note that patient has had prior ERCP with stent placement. After electronically simulating cholecystectomy, pancreaticoduodenectomy is simulated using similar function. Reference to sagittal (C), coronal (D), and axial (E) multiplanar reformat images and volume-rendered projections (F) allow for real-time monitoring of progress and prevent inadvertent inclusion of vital structures such as mesenteric vessels. Pancreatic head, uncinate process, regional lymph nodes, and transverse mesocolon, if involved, are included and then removed electronically (G).

View larger version (125K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1G —44-year-old man with pancreatic adenocarcinoma. Images show sequential steps in performance of virtual Whipple procedure. Initially, margins of gallbladder are defined on sequential axial images (A) using freeform region of interest tool. Electronic sculpt function allows for subsequent electronic removal (B). Note that patient has had prior ERCP with stent placement. After electronically simulating cholecystectomy, pancreaticoduodenectomy is simulated using similar function. Reference to sagittal (C), coronal (D), and axial (E) multiplanar reformat images and volume-rendered projections (F) allow for real-time monitoring of progress and prevent inadvertent inclusion of vital structures such as mesenteric vessels. Pancreatic head, uncinate process, regional lymph nodes, and transverse mesocolon, if involved, are included and then removed electronically (G).

The scope of the Whipple procedure has altered over the years, and many surgeons now perform mesenteric-portal venous bypass on selected patients in whom venous involvement is the only factor precluding resection and in whom proximal and distal graft insertion sites can be identified. If this extended type of surgery is to be performed, the virtual Whipple field may have to be extended to include the involved venous segment and any adjacent arteries and variants.

The prepared data sets were interrogated in a number of ways. Altering the tissue thresholds and transparency settings allowed the user to produce CTA and soft-tissue renderings. The workstation automatically color coded vessels that were electronically removed during the virtual Whipple procedure. The resulting angiographic images highlighted the arteries that crossed the anticipated surgical field (Figs. 2A, 2B and 3A, 3B). Although the images in this study were acquired with a Vitrea workstation, similar images can be produced with any similar platform. Image segmentation was performed by a trained technologist, and the images were reviewed for accuracy by the interpreting physician.

View larger version (120K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2A —73-year-old woman with pancreatic adenocarcinoma. Volume renderings of arterial dominant phase of imaging show accessory hepatic artery (arrowhead) arising from superior mesenteric artery. Artery is not color encoded because it does not cross through surgical plane, although documenting its presence for surgeon is important. A superior pancreaticoduodenal artery has aberrant origin from this vessel, and its long course is in surgical plane (arrows). Because it crosses surgical plane, this aberrant superior pancreaticoduodenal artery is removed during workstation simulation of Whipple procedure and is color encoded in volume-rendered CT angiogram. Biliary stent (asterik, B) is in place and is also color encoded as it crosses surgical plane.

View larger version (111K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2B —73-year-old woman with pancreatic adenocarcinoma. Volume renderings of arterial dominant phase of imaging show accessory hepatic artery (arrowhead) arising from superior mesenteric artery. Artery is not color encoded because it does not cross through surgical plane, although documenting its presence for surgeon is important. A superior pancreaticoduodenal artery has aberrant origin from this vessel, and its long course is in surgical plane (arrows). Because it crosses surgical plane, this aberrant superior pancreaticoduodenal artery is removed during workstation simulation of Whipple procedure and is color encoded in volume-rendered CT angiogram. Biliary stent (asterik, B) is in place and is also color encoded as it crosses surgical plane.

View larger version (153K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3A —80-year-old man with small ampullary adenocarcinoma. Volume-rendered image shows large ileal artery from superior mesenteric artery feeding retrogradely to anastomose with gastroduodenal artery and in so doing crossing surgical plane (arrowhead). Biliary stent (asterisk) is in place. Also shown are two complete pancreaticoduodenal arcades.

View larger version (112K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3B —80-year-old man with small ampullary adenocarcinoma. MDCT scan shows presence of retrograde feeding artery (arrowhead) in oblique thick slab.

Top Abstract Introduction Materials and Methods Results Discussion References

Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Preoperative pancreatic imaging is essential in assessment for local resectability and distant metastasis. Although MRI [7], sonography [8], and endoscopic sonography [9] have been used to assess for local resectability, at our institution MDCT is the preferred imaging technique. In addition to assessment for local vascular invasion, tethering, and narrowing, which have been described as radiologic signs of unresectability [10-12], the MDCT data sets can be used to depict local arterial anatomy. Apart from determining resectability, documentation of arterial variants is relevant to the surgeon as an aid to operative planning. Knowledge of the presence of a replaced or accessory right hepatic artery is useful to a surgeon performing common bile duct transection. Knowledge of the relation of the gastroduodenal artery to the hepatic artery is important because these two arteries have to be teased apart to avoid accidental ligation of the hepatic artery. Because many tumors are situated in the uncinate process and this part of the pancreas has to be lifted off the superior mesenteric vessels, documentation of the presence or absence of aberrant vessels in this area is crucial. Finally, a healthy pancreatic anastomosis can be made only in the presence of a good arterial supply (mainly from the splenic and dorsal pancreatic arteries) [13]. Documentation of tortuosity, calcification, aneurysm formation, and vascular variants in this region also is crucial. Many working parties and groups have issued recommendations that preoperative angiography be performed to evaluate regional arterial vessels and to document these arterial variants to avoid unfavorable outcomes [14, 15]. In patients with celiac axis occlusion, the proper hepatic artery can be fed retrogradely through the gastroduodenal artery from the superior mesenteric artery. Therefore the patency of the celiac axis should be assessed.

Although the virtual Whipple images provide useful supplemental information to the surgeon, other rendered and source images are necessary to determine the key question of resectability and to present other key anatomic relations, particularly those of the tumor to the surrounding veins. Our technologists have a routine for pancreatic CTA data sets that consists of coronal and axial oblique volume-rendered images of both pancreatic parenchymal and hepatic venous data sets (Table 3). These images are judiciously supplemented by minimum intensity projections of the pancreatic and common bile duct systems. Virtual Whipple renderings are reserved for patients likely to undergo a Whipple procedure. Although we work closely with our hepatobiliary surgeons, we usually make this determination on the basis of the available imaging findings. We have developed a Web-based dash-board for our CT laboratory that enables interpreting radiologists to leave detailed instructions for the technologists, and this procedure streamlines workflow in our imaging laboratory. In difficult cases, surgeons, radiologists, and technologists frequently edit cases together.

Results of direct comparison tests have been used to document the equivalent sensitivity of CTA to catheter angiography in both the peripheral arterial system [16] and the abdomen. The use of sophisticated rendering methods as a tool in preoperative imaging is well established in the field of solid-organ transplantation imaging, particularly of the liver [17] and kidney [18], living-donor transplantation of which is becoming commonplace. The results of this study show the ease with which established rendering techniques can be applied to the Whipple surgical field and how the resultant images can be used for accurate identification of visceral arterial variants that cross the surgical planes of dissection.

The largest limitation to this study was that we did not have a reference standard against to which assess our CT studies. Although the surgical notes were reviewed, the surgeons did not comment on the presence or absence of small arterial variants, and we cannot confirm that small vessels predicted to cross the surgical plane did so. The surgical notes did confirm the presence of the largest vascular anomalies, and the presence of all accessory or replaced right hepatic arteries was confirmed. The technique described is of use only for assessing arterial variants that are relevant to the Whipple procedure. Other images are necessary for full assessment of the resectability and stage of a tumor. We are fortunate to have strong relationships with the pancreatic surgeons who helped us develop this technique. Before we developed the virtual Whipple procedure, we had to understand closely the steps involved in the actual Whipple procedure. It is important that any radiologists interested in this technique have their surgical colleagues teach them the exact extent of the surgery performed at their institutions.

In conclusion, this study introduced a novel technique that follows the complex surgical steps in a Whipple procedure. Three-dimensional reconstructed angiographic images were used to identify arterial variants relevant to the Whipple operation and to highlight these variants for the surgeon as part of the preoperative radiologic assessment of pancreatic and ampullary tumors.

References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. American Cancer Society. Cancer facts and figures 2005. Atlanta, GA: American Cancer Society, 2005. Available at: http://www.cancer.org/downloads/STT/CAFF2005f4PWSecured.pdf. Accessed August 6, 2006
2. Yeo CJ, Cameron JL, Sohn TA, et al. Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Ann Surg 1997;226 : 248-257[CrossRef][Medline]
3. 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 1997; 168:971 -977[Abstract/Free Full Text]
4. 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]
5. Kondo H, Kanematsu M, Goshima S, et al. MDCT of the pancreas: optimizing scanning delay with a bolus-tracking technique for pancreatic, peripancreatic vascular, and hepatic contrast enhancement. AJR2007; 188:751 -756[Abstract/Free Full Text]
6. Maithel SK, Khalili K, Dixon E. Impact of regional lymph node evaluation in staging patients with periampullary tumors. Ann Surg Oncol 2007; 14:202 -210[Abstract/Free Full Text]
7. Arslan A, Buanes T, Geitung JT. Pancreatic carcinoma: MR, MR angiography and dynamic helical CT in the evaluation of vascular invasion. Eur J Radiol 2001;38 : 151-159[CrossRef][Medline]
8. Minniti S, Bruno C, Biasiutti C, et al. Sonography versus helical CT in identification and staging of pancreatic ductal adenocarcinoma. J Clin Ultrasound 2003;31 : 175-182[CrossRef][Medline]
9. Yusoff IF, Mendelson RM, Edmunds SE, et al. Pre-operative assessment of pancreatic malignancy using endoscopic ultrasound. Abdom Imaging 2003;28 : 556-562[CrossRef][Medline]
10. Lu DS, Reber HA, Krasny RM, Kadell BM, Sayre J. Local staging of pancreatic cancer: criteria for un-resectability of major vessels as revealed by pancreatic-phase, thin-section helical CT. AJR1997; 168:1439 -1443[Abstract/Free Full Text]
11. Diehl SJ, Lehmann KJ, Sadick M, Lachmann R, Georgi M. Pancreatic cancer: value of dual-phase helical CT in assessing resectability. Radiology 1998;206 : 373-378[Abstract/Free Full Text]
12. Hough TJ, Raptopoulos V, Siewert B, Matthews JB. Teardrop superior mesenteric vein: CT sign for un-resectable carcinoma of the pancreas. AJR 1999; 173:1509 -1512[Abstract]
13. Callery MP, Strasberg SM, Doherty GM, Soper NJ, Norton JA. Staging laparoscopy with laparoscopic ultrasonography: optimizing resectability in hepatobiliary and pancreatic malignancy. J Am Coll Surg 1997; 185:33 -39[CrossRef][Medline]
14. Volpe CM, Peterson S, Hoover EL, Doerr RJ. Justification for visceral angiography prior to pancreaticoduodenectomy. Am Surg 1998; 64:758 -761[Medline]
15. Biehl TR, Traverso LW, Hauptmann E, Ryan JA Jr. Preoperative visceral angiography alters intraoperative strategy during the Whipple procedure. Am J Surg 1993;165 : 607-612[CrossRef][Medline]
16. Jakobs TF, Wintersperger BJ, Becker CR. MDCT-imaging of peripheral arterial disease. Semin Ultrasound CT MR2004; 25:145 -155[CrossRef][Medline]
17. Kamel IR, Kruskal JB, Keogan MT, Goldberg SN, Warmbrand G, Raptopoulos V. Multidetector CT of potential right-lobe liver donors. AJR 2001; 177:645 -651[Free Full Text]
18. Kawamoto S, Montgomery RA, Lawler LP, Horton KM, Fishman EK. Multidetector CT angiography for preoperative evaluation of living laparoscopic kidney donors. AJR 2003;180 : 1633-1638[Abstract/Free Full Text]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Brennan, D. D. Articles by Kruskal, J. B. Search for Related Content PubMed PubMed Citation Articles by Brennan, D. D. Articles by Kruskal, J. B. Social Bookmarking                      What's this? Hotlight (NEW!) What's Hotlight?