| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Division of Abdominal Imaging and Intervention,
Massachusetts General Hospital, Ellison 234-E, 55 Fruit St., Boston, MA
02114.
2 Decision Analysis and Technology Assessment Group, Massachusetts General
Hospital, Boston, MA 02114.
3 Department of Surgical Oncology, Massachusetts General Hospital, Boston, MA
02114.
Received July 25, 2001;
accepted after revision January 22, 2002.
Presented at the annual meeting of the Radiological Society of North
America, Chicago, November 2000.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Forty-two consecutive subjects with malignant liver tumors scheduled for resection were studied with multiphase MDCT. The first 22 subjects underwent both multiphase MDCT angiography and catheter angiography before surgery. The subsequent 20 subjects underwent only preoperative CT angiography. Postprocessing was performed, and the images were analyzed for the depiction of arterial, portal vein, and hepatic vein anatomy and for the identification of important vascular variants. The postprocessing findings were compared and correlated with the findings from catheter angiography (22/42) or intraoperative sonography (42/42) and surgery (42/42).
RESULTS. Arterial anomalies were detected on the images of 17 of 42 patients, including a replaced right hepatic artery in five, replaced left hepatic artery in six, accessory right and left hepatic arteries in two, common trunk for the celiac and superior mesenteric arteries in one, and early bifurcation of the celiac artery in one. In 22 patients in whom catheter angiography confirmation was available, the number of arteries and almost all the significant anomalies were correctly identified on CT angiography (accuracy, 97%; sensitivity, 94%; specificity, 100%). In the subset of 20 patients who underwent MDCT angiography without catheter angiography confirmation, all clinically relevant information was provided by CT angiography. The portal and hepatic vein anatomy and the relationships of the liver tumors to the neighboring venous structures were shown on CT.
CONCLUSION. Multidetector CT provides valuable preoperative information about hepatic vascular architecture and can be used as a noninvasive alternative to catheter angiography before oncologic liver surgery.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In this article, we describe the technique of CT angiography using an MDCT scanner and discuss the comparative performance of CT angiography and catheter angiography.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study was conducted in two phases. In the first 22 patients, the vascular anatomy was confirmed on catheter angiography; in the subsequent 20 patients, only CT angiography was performed preoperatively. In all 42 patients, confirmation was obtained at surgery and on intraoperative sonography.
CT Technique
Dual-phase CT was performed in all patients on an MDCT scanner (LightSpeed
OX/i; General Electric Medical Systems, Milwaukee, WI) after mechanical
injection of 150 mL of nonionic iodinated contrast medium with a concentration
of 300 mg I/mL (Oxilan 300 [ioxilan]; Cook, Boston, MA). The contrast material
was injected at a rate of 5 mL/sec through an 18-gauge IV cannula. Scanning
was performed using a pitch of 6:1 (HiSpeed mode), a 0.8-sec scanning time per
rotation, and a detector configuration of 4.0 x 2.5 mm. These parameters
resulted in a table speed of 15 mm per rotation or 18.7 mm/sec. Other scanning
parameters included 140 kVP and 220-260 mA. For the hepatic arterial phase
imaging, the scanning delay was 25 sec after the initiation of a contrast
bolus; and for the portal venous phase imaging, the scanning delay was 65 sec
after the initiation of a contrast bolus.
Images in each phase were acquired in a single breath-hold, and the approximate time of acquisition was 10-15 sec. Images with a slice thickness of 2.5 mm were reconstructed at every 1.25 mm.
Image Processing
The axial raw data images were processed on a dedicated image processing
workstation (Advantage Windows 3.1; General Electric Medical Systems) for
multiplanar reformations, maximum intensity projection, curved planar
reformations, and volume-rendered technique reconstructions. A trained
technologist and a radiologist performed the image postprocessing to create
three-dimensional (3D) reconstructions of the hepatic arteries, portal vein,
and hepatic veins. The average time spent on the image processing for each
study was 30 min.
Catheter Angiography
Digital subtraction angiography was performed in 16 patients, and cut-film
angiography was performed in six patients. All angiograms were obtained by a
vascular radiologist, and the patients were studied via a transfemoral artery
approach with a 5- to 6.5-French catheter. Selective celiac and superior
mesenteric artery injections were performed with nonionic iodinated contrast
media injected at a rate of 5 mL/sec. Evaluation of the celiac and superior
mesenteric arteries was extended to the portal vein in at least one
injection.
Intraoperative Sonography
An experienced gastrointestinal radiologist performed the intraoperative
sonography using a 7.5-MHz linear transducer (SSH-140 CE; Toshiba, Tokyo,
Japan). Intraoperative sonography was performed after adequate surgical
mobilization of the liver. The entire liver was studied for both focal lesion
detection and accurate segmental location on all lesions. The status of the
liver free of tumor was evaluated. Portal and hepatic veins and their
relationship to the tumor were also evaluated.
Standard of Reference
Whenever available, catheter angiography constituted the standard of
reference for the arterial anatomy, and visual inspection during surgery
constituted the alternate standard of reference for the arterial anatomy.
Catheter angiography or intraoperative sonography constituted the standard of
reference for the portal vein anatomy, and intraoperative sonography
constituted the standard of reference for hepatic vein anatomy.
Image Analysis
All images, including two-dimensional reformations and 3D reconstruction
images, were sent to a PACS (picture archiving and communication system)
workstation that permits interactive analysis.
Two radiologists with expertise in CT and catheter angiography who were unaware of the patients' clinical statuses, the laboratory data, and the results of other imaging modalities, independently evaluated each study. Catheter angiographic images were evaluated for patency and anatomy of the hepatic artery and portal vein systems. The radiologists made observations on CT in each patient concerning the origin of the celiac artery, superior mesenteric artery, right hepatic artery, and left hepatic artery and the presence of accessory arterial supply to the liver. Likewise, a description of hepatic veins, hepatic vein confluence, portal vein bifurcation, and the presence of accessory draining veins was made. Detectability of tumor on CT and tumor impingement (invasion, encasement, or obstruction) on hepatic vessels were recorded. The catheter angiographic images were considered to be the standard of reference for hepatic artery and portal vein anatomy in 22 of 42 patients. Intraoperative sonography served as an alternate standard of reference for the portal vein anatomy in 20 of 42 patients. In all patients, intraoperative sonography (42/42) was used as a standard of reference for hepatic vein anatomy. Surgical confirmation was available in 42 of 42 patients.
Statistical Analysis
Statistical analysis was performed using the McNemar test.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The conditions of three patients with colon cancer metastases that were considered operable on the preoperative CT were deemed inoperable because of additional metastases detected in the liver on intraoperative sonography in two of the three patients and a peritoneal implant seen in one of the three patients. These patients were treated with chemotherapy via the placement of an intraarterial chemotherapy pump.
Arterial Anomalies
Excellent arterial opacification was revealed on CT in all patients.
Tertiary order branches in the liver as small as 1 mm in diameter as well as
the artery to segment IV were routinely identified in all patients. Arterial
anomalies were observed in 17 of 42 patients. Replaced right hepatic artery
from the superior mesenteric artery was seen in five patients (Fig.
1A,1B).
Accessory supply to the right lobe of the liver from the superior mesenteric
artery was observed in three patients (Fig.
2A,2B,2C).
Replaced left hepatic artery from the left gastric artery was seen in six
patients (Fig.
3A,3B);
in one patient, the left gastric artery was a separate branch of the abdominal
aorta. Accessory supply to the left lobe of the liver from the left gastric
artery was observed in one patient. In another patient, the celiac and
superior mesenteric arteries had a common origin (Fig.
4A,4B,4C,4D).
One of the three patients with liver metastasis who underwent intraarterial
chemotherapy pump placement had an accessory left hepatic artery that was
ligated during surgery. However, a small accessory right hepatic artery
originating from the splenic artery and an accessory left hepatic artery from
the left gastric artery in two different patients were not prospectively
identified on CT angiography. The latter was obvious on retrospective review
and the former was clinically insignificant. No statistically significant
difference in the performance of CT angiography and catheter angiography was
observed (p = 0.5). The overall accuracy of MDCT angiography was 97%
(95% confidence interval [CI], 87-99.9%) with a sensitivity of 94% (95% CI,
71-99.9%) and a specificity of 100% (95% CI, 86-99.9%). The CT angiographic
findings were in agreement with the surgical findings in 95.8% of patients
(95% CI, 78-99.9%).
|
|
|
|
|
|
|
|
|
|
|
Portal Veins
Excellent opacification of the portal vein was seen on CT in 41 of 42
patients. CT could confidently assess the patency and caliber of the portal
vein and its main branches in all patients. Portal vein anatomy was best
displayed on the reconstructions in the coronal plane. Portal vein anomalies
were identified in seven patients. Trifurcation of the main portal vein was
seen in four patients. In two patients, the right posterior branch originated
from the main portal vein before the bifurcation
(Fig. 5). In one patient,
segment VIII received dual supply from the right and the left portal veins.
Presence of tumor thrombosis in the left branch portal vein was diagnosed in
one patient with primary liver cancer (Fig.
6A,6B).
In three patients with liver metastases, the tumor was seen surrounding the
right portal vein in two and the left portal vein in one.
|
|
|
MDCT angiograms were in agreement with the catheter angiograms obtained in the venous phase of celiac angiography in 22 of 22 patients and on intraoperative sonography in 42 of 42 patients (accuracy, 100%; 95% CI, 84-99.9%).
Hepatic Veins
Hepatic vein trunks were confidently identified in all the patients.
Excellent opacification of the hepatic veins was seen on CT in 40 of 42
patients. In two patients, the hepatic vein opacification was suboptimal;
however, the vein anatomy could still be seen. Hepatic veins were displayed in
the axial and coronal planes. The axial plane best revealed the early
branching and early bifurcation of the middle hepatic vein, which may alter
the plane of liver resection.
In 29 patients, the middle and left hepatic veins joined in a common trunk that subsequently drained into the inferior vena cava. Hepatic vein anomalies were identified in 11 of 42 patients. In eight patients, an accessory right hepatic vein was draining posteriorly and inferiorly into the inferior vena cava. In three patients, a large tributary (>5 mm) was draining segment VIII, which then emptied into the middle hepatic vein (early branching). This caused confusion in segmental localization of the tumor in one patient; however, coronal reformatted images clarified this finding. In addition, in 10 patients, the hepatic veins were involved by the tumor; two of these patients had tumor compressing the inferior vena cava. In another, the tumor in the right lobe was seen extending to the confluence of the middle and left hepatic veins (Fig. 7). Multiplanar two-dimensional and 3D reconstructions were helpful in showing the relationship of the tumors to the hepatic veins and the inferior vena cava. There was 100% agreement between CT and intraoperative sonography in the evaluation of hepatic veins (95% CI, 91-99.9%). CT accurately predicted tumor involvement of the hepatic veins, which correlated well with the intraoperative sonographic and surgical (10/10) findings.
|
The image quality on MDCT and CT angiography was excellent, without any significant motion or stairstep artifact. In one patient who had a prior cholecystectomy and in another patient who had a prior right colon resection, a few beam-hardening artifacts emanated from metallic surgical clips but did not interfere in the evaluation of the vascular anatomy on the axial raw data images. However, stairstep artifacts were present on the 3D reconstruction images in these two patients.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
At times, the presence of a vascular anomaly may indicate modification to the surgical technique. For example, some patients have their right anterior portal vein originating from the left portal vein. In these patients, left portal vein resection proximal to the origin of the anterior portal vein would compromise portal perfusion to segments IV, V, and VIII, resulting in segmental ischemia and subsequent atrophy [13,14,15,16, 18]. Likewise, preoperative mapping of the hepatic vein system is indispensable because the transection plane is determined by the anatomic distribution of the hepatic veins. In right lobe resection, the middle hepatic vein and the left hepatic vein should be preserved to prevent parenchymal damage to the remnant liver. At times, a large (>5 mm) tributary vein may drain segment VIII into the middle hepatic vein, and resection of the middle hepatic vein during a left hepatectomy may compromise venous drainage of segment VIII, resulting in congestive ischemia and atrophy. Likewise, inferior accessory hepatic veins, which typically drain segments V and VI directly into the inferior vena cava, can also be detected and characterized on CT. These aberrant vessels are also important in certain surgical settings. Understanding the relationship of the tumor to the adjacent hepatic vein is equally significant for planning a resection to include an adequate tumor margin [6, 14,15,16]. Hepatic veins are typically not depicted on catheter angiography unless direct hepatic venography is conducted. However, hepatic veins can adequately be depicted on CT angiography or MR angiography and, possibly, on sonography.
Our study shows that in patients being examined for liver resection for neoplasms, MDCT with CT angiography of the liver permits excellent parenchymal and vascular preoperative evaluation. CT angiography depicted all the vascular anomalies (arterial, portal, and hepatic venous) except in two patients in whom a small accessory right hepatic artery and a left hepatic artery were not recognized prospectively. However, the latter could be identified on retrospective review, and the former had no significant impact on the surgical treatment. The surgeon considered 3D data sets to be helpful in planning liver resection, especially in tumors that were close to the inferior vena cava, the hepatic veins, and the portal veins. On the basis of the data provided to the surgeons, multiphase MDCT has become an essential part of preoperative evaluation of patients with liver neoplasms.
Many centers now prefer CT angiography to catheter angiography for preoperative planning because of its noninvasive nature, improved patient compliance, and reduced morbidity [7]. Our study shows that CT angiography using MDCT does provide vascular details equivalent to catheter angiography in detecting clinically relevant anatomy of hepatic arteries and the portal vein in the context of preoperative planning for tumor resection. In addition, the mapping of the hepatic vein and its relationship to the tumor could be shown on CT because selective visceral catheter angiography cannot provide information about the hepatic vein anatomy and its relationship to liver tumor unless direct venography is performed.
Most anomalies were diagnosed on axial images. Three-dimensional reformation was occasionally useful in studying complex vascular anatomy, as depicted in two patients with a replaced right hepatic artery from the superor mesenteric artery that was better visualized on 3D imaging. Curved plane reformations were helpful in the lengthwise displaying of the entire course of the artery (Fig. 2B). Likewise, coronal reconstructions of the venous phase images assisted in the accurate localization of the tumor and its relationship to the hepatic vein and the inferior vena cava.
The introduction of helical technology between 1993 and 1994 expanded the capability of CT to provide rapid vascular imaging in the optimal phase of vascular opacification. Since then, other investigators have reported numerous studies with single-detector CT and 3D imaging in the preoperative evaluation of the liver [17, 19,20,21,22,23,24]. Although pioneering work has been done with single-detector CT, several limitations have existed, including the need for thinner slice acquisition and z-axis coverage. Tube heating constraints were present, and the use of a wider pitch resulted in more 3D artifacts because of longitudinal sectional broadening. The overall breath-hold for the helical acquisition was longer, and to catch the optimal contrast bolus in the vessels, more contrast volume was needed. Visualization of the second and third order branches was often difficult.
The rapid scanning capability of MDCT allows optimal phase scanning in a short breath-hold, well within the time of dynamic administration of a single bolus of contrast material. Therefore, clinically useful information can be obtained even in patients having difficulty in suspending respiration. Other clinical advantages over single-detector CT include increased z-axis coverage and thinner slice acquisition (1.25 mm) during a shorter scanning time. These advantages have resulted in excellent spatial resolution and better depictions of fine vascular details. Retrospective reconstruction into thinner slices from the same raw data provides more robust 3D rendering with diminished helical artifacts. MDCT therefore permits increased scanning speed and improved scanning resolution compared with conventional helical scanners [8,9, 22, 23].
Chambers et al. [21] showed the ability of helical CT without reconstructions to correctly identify hepatic artery anatomy, with 96% sensitivity and 87% specificity for detecting aberrant hepatic artery anatomy. Our results corroborate the findings of previous studies [7, 21, 25]. However, two-dimensional and 3D images are aesthetically more pleasing. The projectional views are more acceptable to the surgeons as an alternative to catheter angiography. MDCT also provides more accurate and clearer anatomic relationships because it allows the lengthwise display of vessels (Fig. 2B). We consider this imaging to be of increasing importance for hepatic resection because the surgical technique and the placement of the chemotherapy pump are influenced significantly by the knowledge of hepatic vascular variants.
Our study also confirms previous reports by Johnson et al. [26] and Uchida et al. [20] that multiplanar and 3D reconstruction images, which were used in our study, depict the relationship of the tumor to the neighboring vessel. This principle was useful in one of the patients in our series in whom the tumor was contiguous to the proximal segment of the inferior vena cava. The 3D images correctly predicted that the most proximal segment of the inferior vena cava was spared.
CT angiography has been shown to be an alternative to catheter angiography for the evaluation of potential liver transplantation and liver resection candidates. This approach has been supported by Winter et al. [7], who showed the usefulness of shaded-surface display and maximum intensity projection in preoperative evaluation in comparison with the results of catheter angiography. This finding has a major impact on the cost of the total preoperative evaluation. As in other reports, the maximum-intensity-projection images were better in depicting the hepatic vein anatomy and the portal vein anatomy [8, 24].
Furthermore, the substitution of CT angiography for catheter angiography as the method for determining vascular anatomy is much more convenient for the patient. CT angiography spares the patient the discomfort and the associated morbidity of catheter angiography. Analgesia, periprocedural nursing, and postprocedural observation are also not needed with CT angiography.
The limitations of our study include the small study sample with confirmation on catheter angiography. However, our results are compelling in support of MDCT as an efficient preoperative imaging tool for liver vascular evaluation. Although cone-beam artifacts are theoretically a potential problem of MDCT, they have not, to date, caused significant image degradation. Likewise, for optimal phase vascular enhancement, a timing acquisition may be required. This strategy is complex and requires additional time for two injections of contrast material. Our earlier experience with single-detector CT showed that the use of a timing acquisition did not offer advantages over the use of an empiric delay for CT angiography of the liver. In addition, preoperative biliary anatomy evaluation is important, specifically in candidates for liver transplantation. CT is limited in its ability to perform this task and to provide the pertinent information in a nondilated system.
Large data sets generated with CT angiographic studies must be considered in networking and image storage. Typically, more than 500 images are generated from each CT angiographic study, and it becomes impractical to view such a large data set on film. Handling the large volume of image data generated with MDCT mandates a PACS system or a workstation for image viewing, yet it creates a considerable burden for the PACS system to handle the data.
MDCT has improved the applications of CT angiography because of its speed and wider z-axis coverage. Our experience with MDCT suggests that it provides valuable preoperative information about hepatic vascular architecture and can be used as a noninvasive alternative to catheter angiography for all oncologic liver surgery workups.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. Minami, H. Chung, M. Kudo, S. Kitai, S. Takahashi, T. Inoue, K. Ueshima, and H. Shiozaki Radiofrequency Ablation of Hepatocellular Carcinoma: Value of Virtual CT Sonography with Magnetic Navigation Am. J. Roentgenol., June 1, 2008; 190(6): W335 - W341. [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]
|
|
|
|
|
|
C. B. Winston, N. A. Lee, W. R. Jarnagin, J. Teitcher, R. P. DeMatteo, Y. Fong, and L. H. Blumgart CT Angiography for Delineation of Celiac and Superior Mesenteric Artery Variants in Patients Undergoing Hepatobiliary and Pancreatic Surgery Am. J. Roentgenol., July 1, 2007; 189(1): W13 - W19. [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]
|
|
|
|
|
|
A. Shibuya, T. Nakazawa, K. Saigenji, K. Furuta, and K. Matsunaga Diaphragmatic hernia after radiofrequency ablation therapy for hepatocellular carcinoma. Am. J. Roentgenol., May 1, 2006; 186(5 Suppl): S241 - S243. [Full Text] [PDF]
|
|
|
|
|
|
N. Kalra, S. Suri, R. Gupta, S. K. Natarajan, N. Khandelwal, J. D. Wig, and K. Joshi MDCT in the Staging of Gallbladder Carcinoma. Am. J. Roentgenol., March 1, 2006; 186(3): 758 - 762. [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]
|
|
|
|
|
|
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]
|
|
|
|
 |
|
 V A Duddalwar Multislice CT angiography: a practical guide to CT angiography in vascular imaging and intervention Br. J. Radiol., December 1, 2004; 77(suppl_1): S27 - S38. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Saini Multi-Detector Row CT: Principles and Practice for Abdominal Applications Radiology, November 1, 2004; 233(2): 323 - 327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Sahani, A. Mehta, M. Blake, S. Prasad, G. Harris, and S. Saini Preoperative Hepatic Vascular Evaluation with CT and MR Angiography: Implications for Surgery RadioGraphics, September 1, 2004; 24(5): 1367 - 1380. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Romer, M. Chung, A. Chan, D. W. Townsend, F. Torok, B. McCook, M. P. Federle, and N. Avril Single-Detector Helical CT in PET-CT: Assessment of Image Quality Am. J. Roentgenol., June 1, 2004; 182(6): 1571 - 1577. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A M Connors and S P Blake Multiple-choice questionnaire: general Imaging, October 1, 2002; 14(5): 428 - 437. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
