| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Departments of Radiology (M.J.G., J.B.K., J.S., S.N.G., V.R.) and Transplantation Surgery (D.W.H.), Beth Israel Deaconess Medical Center/Harvard Medical School, One Deaconess Rd, Boston, MA 02215. From the 2001 RSNA scientific assembly. Received November 7, 2002; revision requested January 13, 2003; revision received January 28; accepted March 11. Address correspondence to J.B.K. (e-mail: jkruskal@bidmc.harvard.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: In potential liver donors, 100 consecutive hepatic CT angiograms were obtained after intravenous bolus administration of 150–180 mL of nonionic contrast material. Arterial phase images (1.25-mm collimation, 7.5 mm/ 0.8-second table speed) were acquired after test dose injection. Portal phase images were acquired at 60 seconds (2.5-mm collimation, 15 mm/0.8-second table speed). Postprocessing depicted arterial, portal, and hepatic vein anatomy traversing the anticipated surgical hepatectomy plane to the right of the middle hepatic vein (MHV) and separating the right and left lobes of the liver. Two radiologists interpreted the images, and data were agreed on by consensus. Data collected included intrahepatic anatomy and origin of the artery and vein supplying segment IV; the venous drainage from segments V and VIII; and the presence, size, and distance from the right hepatic vein (RHV) confluence of accessory hepatic veins in the surgical plane.
RESULTS: Thirty-one donors had conventional hepatic vascular anatomy. Vessels that traversed the hepatectomy plane included the artery supplying segment IV in seven (7%) patients, dominant portal vein supply to segment IV from the right portal vein in two (2%) patients or from both right and left portal vein branches in three (3%) patients, segment VIII draining into the MHV in 67 (67%) patients or both the MHV and RHV in 18 (18%) patients (the major draining vein was >7 mm in diameter in 23%), segment V draining into the MHV in 10 (10%) patients, or both the MHV and RHV in 19 (19%) patients (the major draining vein from segment V was 7–10 mm in diameter in 70 patients, and larger than 10 mm in five). Forty-four accessory hepatic veins were identified in 40 patients; seven drained segment V, while the majority drained segments VI and VII. The mean diameter was 5.3 mm and 45% were larger than 6 mm. The average distance to the RHV–inferior vena cava confluence was 28.7 mm. Of 70 patients with drainage from segment V into RHV, 22 (31%) had an accessory RHV. However, atypical drainage into the MHV was noted in seven (70%) of 10 patients and into the MHV and RHV in 11 (58%) of 19 patients.
CONCLUSION: In the majority of potential donors, CT angiography depicted a wide range of vascular anatomic variations that traverse the hepatectomy plane.
© RSNA, 2003
Index terms: Computed tomography (CT), angiography, 761.12116, 952.12916 • Computed tomography (CT), multi–detector row, 761.12116 • Hepatic arteries, CT, 952.12916 • Hepatic veins, CT, 952.12916 • Liver, transplantation, 761.451
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Given the potential risks to the donor, very careful work-up is mandated. Computed tomography (CT) and magnetic resonance (MR) imaging play a primary role in the workup of potential adult donors prior to transplantation (3–7). Depending on available segmental volumes, hepatectomy is performed along a plane that separates the liver into left and right lobes. This plane runs along a line to the right of the middle hepatic vein (MHV)—which connects the gallbladder fossa and the inferior vena cava—where it emerges from the liver surface, which is known as the Cantlie line. The accurate definition of structures traversing this relatively avascular plane that separates segment IV (which, along with segments II and III, remain in the donor) from segments V and VIII is necessary to prevent vascular injury, which would compromise essential hepatic metabolic function in the healthy donor.
In our institution, we perform multiphase multi–detector row CT angiography as part of the donor workup prior to harvesting the right lobe of the liver. The improved spatial, contrast, and temporal resolution provided by multi–detector row CT allows accurate depiction of small intrahepatic vessels. Indeed, Kamel et al (7) have shown that multi–detector row CT angiography is accurate in the depiction of third-order intrahepatic arteries. Anatomic variations of the extrahepatic vasculature are common, well recognized, and easily depicted with CT angiography; however, intrahepatic vascular variants, specifically as they relate to the anticipated surgical plane for split-liver transplantation, have not been well documented. Additionally, since certain anomalies may result in modification of the planned surgical procedure, it is important to document variations in arterial and venous anatomy that may occur in donors, particularly those variations that traverse the anticipated hepatectomy plane. Inadvertent damage to a major vessel may cause ischemic injury to the graft or to the donor liver.
The purpose of this study was to evaluate the relevant arterial and venous anatomy of the hepatectomy plane lateral to segment IV by using multi–detector row CT with respect to adult living related transplantation of the right lobe of the liver.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Image Acquisition and Processing
Either 1,000 mL of 4% milk (n = 56) or 1,000 mL of water (n = 44) was used for bowel opacification (7). A four-channel multi–detector row CT scanner (Lightspeed QXi; GE Medical Systems, Milwaukee, Wis) was used to perform multiphase image acquisition after intravenous bolus administration of 150–180 mL of nonionic contrast material (ioversol, Optiray 300; Mallinckrodt Medical, St Louis, Mo) at a rate of 5–6 mL/sec. The bolus phase delay was determined with a test injection of 15–18 mL of the nonionic contrast material, also at a rate of 5–6 mL/sec. A collimation of 1.25 mm with table speed of 7.5 mm/0.5 second or a gantry rotation of 0.8 was used for the arterial phase. Venous or nonequilibrium phase imaging started 45 seconds after commencement of the arterial phase and used a collimation of 2.5 mm with a table speed of 15 mm per gantry rotation (7). For image processing, 50% overlap was used, created retrospectively, and multiplanar maximum intensity projection and three-dimensional volume renderings were performed on both the arterial and venous phase acquisitions.
Image Interpretation and Data Collection
Images that incorporated postprocessing were available at a picture archiving and communication system that facilitated data analysis and interactive scrolling. Images were reconstructed at a commercial workstation (Advantage, Windows 4.0; GE Medical Systems). Liver segments were assigned according to the classification described by Couinaud, as modified by Bismuth (8). All images were jointly reviewed by two abdominal radiologists (J.B.K., V.R.), each with more than 10 years of experience in reading abdominal CT scans. When consensus could not be reached, a third abdominal radiologist (S.N.G.) with similar experience interpreted the image.
Images were considered adequate if the relevant hepatic vasculature in the appropriate phase of contrast material enhancement was depicted sufficiently, so that the radiologist could confidently predict the segmental vascular anatomy. Specific image data collected included adequacy of depiction of all relevant arteries and veins. A virtual hepatectomy plane was defined as a plane that occurred 1 cm lateral to the MHV and extended from the level of the suprahepatic inferior vena cava down to the gallbladder fossa (9) (Fig 1). Once determined, this became the central focus for subsequent data analysis.
|
|
Anatomy was considered normal if segment IV was supplied by the left hepatic artery and not by the right hepatic artery, centrally, in a codominant manner, or directly from the common hepatic artery; if segment IV portal venous supply arose from the left portal vein and not the right portal vein, which bifurcates (not trifurcates); if drainage from segment V passed into the RHV and not the MHV, accessory hepatic vein, or dual hepatic vein drainage; and if drainage from segment VIII passed into the MHV and not into the RHV, both the MHV and RHV, or the inferior vena cava. Since 100 patients were evaluated, raw data and percentages are equivalent. Diameter was measured with electronic calipers.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Hepatic Arteries
When arising from the right, arteries to segment IV may cross the main lobar fissure. In seven of 100 donors, this artery traversed the anticipated surgical plane. The artery arose solely from the right hepatic artery in six of the seven donors, and separate arteries arose from both the right and left hepatic arteries in one. Inadvertent ligation of this artery during harvesting may produce ischemia to segment IV in the remaining donor liver. With a small left lateral lobe (segments II and III), there may be insufficient parenchyma remaining to sustain metabolic function. With dual origin of these arteries to segment IV, many surgeons sacrifice the artery arising from the right hepatic artery when removing the right lobe. Anecdotal evidence suggests that segment IV functions well with arterial supply from the remaining artery that arises from the left side. Experienced surgeons will dissect the right hepatic artery down to the origin of the branch to segment IV and divide the right hepatic artery distal to this (16). To facilitate this, we measured the distance from the bifurcation of the common hepatic artery to the origin of the branch to segment IV. Five arteries that supplied segment IV arose from the bifurcation or proper hepatic artery, and one arose from the common hepatic artery. The surgeon may also dissect the right hepatic artery to the right of the common duct after the right hepatic artery passes posterior (or occasionally anterior) to it, which is distal to the takeoff of the segment IV artery in almost all cases. Hepatic artery anatomy has been thoroughly studied (17), with an emphasis on whole organ transplantation (18,19) but not split-liver transplantation. An anatomic study by Onishi et al (20) showed a middle hepatic artery to segment IV arose from the hepatic artery proper in 5.5% of patients, and a dual supply was identified in 5.5%. These variants will not alter surgery, unless the artery is more than 2 mm in diameter (21,22). There is considerable variation in the reported anomalous origin of the RHA ranging from 6%–11% (23,24) to 40%–50% (25–27).
Portal Veins
The dominant portal vein supplying segment IV arose from the left portal vein in 95 of the donors and from the RPV in two. In three patients, a codominant portal venous blood supply arose from both the left and the right portal veins. The hepatic artery to segment IV arose from the left side in all patients with a portal vein origin to segment IV that did not arise solely from the left portal vein (eg, arose in part from the right portal vein). We did not identify any patients with left portal origin of the right liver segments, which is a recognized variant and relative contraindication to right lobe harvesting. Most portal variations occur in the right hemiliver and range from absence of a right portal trunk to accessory portal branches (28). Portal vein trifurcations were identified with multi–detector row CT in 16 (16%) of the donors in our study, while portal vein trifurcations were identified with US in 10.8% of patients in a study by Atri et al (29) and with surgery in 10% of patients in a study by Marcos et al (11). This variant is not a contraindication to surgery, but isolation of portal vein branches is not possible until the portal vein has been fully exposed by transecting parenchyma and unroofing the vein; hence, preoperative knowledge of this is important to surgical care. Marcos et al also found a higher percentage (32.5%) of right portal veins that supply segment IV than we found. Other important variants include four different intrahepatic branching patterns that result from absence of the right portal vein (28). Another variant that is surgically relevant is an early posterior right portal branch. When two separate right allograft portal inflows are present and are too short to insert into a single orifice, a Y-graft can be fashioned by joining the portal vein of the recipient to the segments of the donor (30), or the two portal vein branches can be fashioned and an anastomosis can be created side-to-side to form a common single orifice.
Hepatic Veins
As surgical expertise improves with right lobe grafting, the anatomy of hepatic veins draining segments V and VIII has become more important (31). Venous drainage from segments V and VIII may traverse the hepatectomy plane, altering the number of venovenous anastomoses that are required. Complex hepatic venous anatomy, such as duplicate RHVs or MHVs, can increase the duration of surgery and the associated morbidity, requiring venoplasty to create a single outflow orifice for each main hepatic vein (31).
The current technique for right lobe harvesting preserves the donor MHV to ensure the integrity of segment IV, with transection of tributaries that drain the anterior segments (V and VIII) of the liver. The RHV that drains segments V and VIII is harvested. In the optimal surgical setting, all venous outflow from segments V and VIII occurs via the RHV; however, dominant drainage from segment VIII occurred via the MHV in the majority (67%) of our donors.
If the diameter of a hepatic vein indirectly reflects the volume of the corresponding drainage area, our study indicates that a single venovenous anastomosis is required in most donors. The major vein that drains segment VIII and crosses the proposed hepatectomy plane was larger than 7 mm in diameter in 23 donors, and none were larger than 10 mm in diameter. Dominant drainage of segment VIII was via the RHV in only 12 of our donors, although there were invariably smaller RHV branches that contributed to outflow in the remainder of donors. Segment V, however, was drained via the RHV in 70 of the donors. The major draining vein from segment V had a diameter of 7–10 mm in 70 donors, and a diameter larger than 10 mm in five. In a series of 17 patients who underwent surgery, Cheng et al (32) ligated (without creating an anastomosis) all venous structures smaller than 5 mm with no substantial graft congestion or venous complications.
Occasionally there is partial drainage of the right lobe via a small RHV with dominant outflow via the MHV. In 10 of our patients, the main outflow from segment V was via the MHV. If this anomaly is not appreciated prior to or during surgery, inadequate venous reconstruction may lead to graft congestion and dysfunction (33). Dominant drainage from the right lobe may occur into the MHV or via an inferior RHV. If these variations are not recognized and only the RHV is used for reconstruction, venous outflow obstruction with graft congestion and eventual graft failure will occur. Additionally, preservation of the MHV and its branch drainage of the left lobe are crucial to prevent outflow blockage from occurring in segment IV, which remains in the donor (33). When this vein traverses the hepatectomy plane, Cattral et al (34) advise including a segment of the MHV with its large venous tributaries, dividing it from the left hepatic vein, and restoring flow with a jump graft from the recipient left portal vein. It is interesting that these authors also found that triphasic CT was very useful for optimizing the depiction of venous outflow of the graft.
Of clinical concern is the potential for causing partial congestion of the harvested anterior segments of the liver by not including or fashioning sufficient venous outflow (35). Outflow problems that arise in the implanted graft may be devastating. A number of centers have modified their harvesting technique by extending the right lobe graft up to the MHV and forming a vein graft (36,37). Miller et al (38) have moved the hepatectomy plane from approximately 0.5 cm to the right of the MHV to immediately adjacent to the right border of the Cantlie line, allowing isolation and interposition graft reconstruction of important segment V and/or segment VIII tributaries. Of the variety of innovative procedures being performed, the recipient’s left portal vein may be used as an interposition graft to drain a dominant MHV in a right-lobe transplant (39). Given these recent surgical modifications, it is helpful to document venous drainage across the anticipated hepatectomy plane into the MHV, especially if this drainage encompasses a large area of the right lobe.
Accessory Hepatic Veins
We identified 44 surgically relevant, accessory inferior hepatic veins in 40 patients. The majority of these veins drained segments VI and VII, although they drained segment V in 8% of the patients. The mean diameter was 5.3 mm, and 45% were larger than 6 mm in diameter, which is the current cutoff at our institution. Marcos et al (39) preserves these veins if their diameter is larger than 5 mm. In our patients, the average coronal distance from caval insertion of the accessory inferior hepatic vein to the RHV-caval confluence was 28.7 mm. When this distance is more than 40 mm, it is technically difficult to implant the two veins in the recipient’s inferior vena cava concurrently with a single, partially occluding venous clamp. Of the 70 donors who had conventional drainage from segment V to the RHV, 45 (31%) also had an accessory inferior hepatic vein larger than 5 mm in diameter. When drainage occurred to the MHV or to both the RHV and MHV, the number of accessory inferior hepatic veins rose to 70% and 58%, respectively. Identification of an accessory vein should therefore heighten awareness of additional anomalous drainage that may alter surgery.
The presence of accessory veins increases surgical time. Only one of our patients had two accessory veins that were larger than 6 mm in diameter, while another had three veins, only two of which were larger than 6 mm.
Multi–detector row CT angiography can provide surgeons with most of the necessary preoperative vascular and volumetric data required for surgical planning. Previous studies have documented the vascular anatomy of specific liver segments, including descriptions of the portal vein anatomy of segment VIII (40) and the RHV and right portal vein anatomy of the entire right hemiliver (41). To our knowledge, no studies have focused on vascular structures that traverse the hepatectomy plane. MR imaging can provide similar multiplanar images of the liver without the use of iodinated contrast material or ionizing radiation and depict the intrahepatic biliary anatomy (6,42). It must still be shown, however, that present MR techniques can accurately depict the intrahepatic arterial anatomy, specifically the origin and location of the artery to segment IV, minor (<5 mm) hepatic vein, requiring back-table venoplasty, or small inferior hepatic veins requiring separate anastomoses with the recipient’s inferior vena cava (32). Anticipated advances in MR technology may permit these vascular structures to be well depicted. Early use of CT angiography is advocated by Pomfret et al (9) to depict anatomy that may result in donor exclusion, thus avoiding an unnecessary biochemical, hematologic, and extensive consultative workup. Conversely, Marcos et al (43,44) suggest that most variations in hepatic vascular anatomy can be accommodated without complications to the donor or complex reconstructions. Still, other authors (38) indicate that there are very few anatomic contraindications to surgery.
Bile ducts may also traverse the surgical plane. CT angiography is not the optimal modality for use in their evaluation, but anatomic data may be obtained with either MR or CT cholangiopancreatography. This was not part of our study, however, and it may be a limitation of our results.
Vascular variants that are recognized preoperatively rarely result in donor exclusion. Knowledge of the anticipated hepatectomy plane and variant vascular anatomy, as depicted with CT angiography and in conjunction with the local surgical techniques employed, can however greatly assist the surgeon in preoperative planning and intraoperative management harvesting of the right lobe of the liver. The importance of vascular variants that traverse the hepatectomy plane for harvesting of the right lobe of the liver lies in their early recognition with appropriate individualization of each operation. We believe a close liaison between the radiologist and the operating surgeon is thus essential to tailor and optimize the surgical approach to improve technical success of this challenging procedure.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, J.B.K.; study concepts, M.J.G., J.B.K.; study design, J.B.K., S.N.G., V.R.; literature research, M.J.G., J.B.K., J.S.; clinical studies, M.J.G., J.B.K., S.N.G., V.R.; data acquisition, M.J.G., J.B.K., V.R.; data analysis/interpretation, M.J.G., J.B.K., J.S., S.N.G., V.R.; statistical analysis, M.J.G., J.B.K., J.S.; manuscript preparation, definition of intellectual content, editing, revision/review, and final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
scandal
: segment IV and liver transplantation. J Chir Paris 1993; 130:443-446.[Medline]
This article has been cited by other articles:
|
|
 |
|
  K. Shimada, H. Isoda, T. Okada, T. Kamae, S. Arizono, Y. Hirokawa, and K. Togashi Unenhanced MR Portography With a Half-Fourier Fast Spin-Echo Sequence and Time-Space Labeling Inversion Pulses: Preliminary Results Am. J. Roentgenol., July 1, 2009; 193(1): 106 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. A. Catalano, A. H. Singh, R. N. Uppot, P. F. Hahn, C. R. Ferrone, and D. V. Sahani Vascular and Biliary Variants in the Liver: Implications for Liver Surgery RadioGraphics, March 1, 2008; 28(2): 359 - 378. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ishigami, K. Yoshimitsu, H. Irie, T. Tajima, Y. Asayama, M. Hirakawa, and H. Honda Accessory Left Gastric Artery from Left Hepatic Artery Shown on MDCT and Conventional Angiography: Correlation with CT Hepatic Arteriography Am. J. Roentgenol., October 1, 2006; 187(4): 1002 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Atasoy and E. Ozyurek Prevalence and types of main and right portal vein branching variations on MDCT. Am. J. Roentgenol., September 1, 2006; 187(3): 676 - 681. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. T. Johnson and E. K. Fishman IV Contrast Selection for MDCT: Current Thoughts and Practice Am. J. Roentgenol., February 1, 2006; 186(2): 406 - 415. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Uchida, M. Ishibashi, N. Tomita, M. Shinagawa, N. Hayabuchi, and K. Okuda Hilar and Suprapancreatic Cholangiocarcinoma: Value of 3D Angiography and Multiphase Fusion Images Using MDCT Am. J. Roentgenol., May 1, 2005; 184(5): 1572 - 1577. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. J. Stemmler, E. K. Paulson, F. J. Thornton, S. R. Winters, R. C. Nelson, and B. M. Clary Dual-Phase 3D MDCT Angiography for Evaluation of the Liver Before Hepatic Resection Am. J. Roentgenol., December 1, 2004; 183(6): 1551 - 1557. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Onodera, T. Omatsu, J. Nakayama, T. Kamiyama, H. Furukawa, S. Todo, T. Nishioka, and K. Miyasaka Peripheral Anatomic Evaluation Using 3D CT Hepatic Venography in Donors: Significance of Peripheral Venous Visualization in Living-Donor Liver Transplantation Am. J. Roentgenol., October 1, 2004; 183(4): 1065 - 1070. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
