| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Department of Radiology, New York University Medical Center, 530 First Ave, HCC Basement-MRI, New York, NY 10016 (V.S.L., M.T.L., N.M.R., G.A.K., J.C.W.); and Siemens Medical Systems, Erlangen, Germany (G.L., D.M.T.). Received April 12, 1999; revision requested June 10; final revision received September 3; accepted September 15. Address correspondence to V.S.L. (e-mail: lee@mri.med.nyu.edu).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Fifty patients underwent 3D spoiled gradient-echo imaging (4.2/1.8 [repetition time msec/echo time msec]; flip angle, 12°; interpolation in three directions; intermittent fat saturation; pixel size 
2.5 mm in all dimensions) before and dynamically after administration of gadopentetate dimeglumine, with the first enhanced acquisition timed for hepatic arterial dominance by using a test bolus of contrast material. Qualitative and quantitative measures of image quality were determined. Patterns of arterial and venous anatomy were assessed. Ten patients (20%) underwent repeat imaging within 6 months, and reproducibility was evaluated.
RESULTS: Hepatic contrast-to-noise ratios for nonenhanced and arterial, portal venous, and equilibrium phase studies averaged 13.0 ± 12.6 (SD), 17.4 ± 11.8, 30.4 ± 16.2, and 28.6 ± 21.1, respectively. During arterial phase, the liver enhanced a mean of 29% of the maximal enhancement as measured during portal venous phase. Hepatic vascular anatomic variants were comparable in distribution to those cited in published articles. Repeat studies were not significantly different in image quality when compared with original studies.
CONCLUSION: High-quality arterial phase 3D volumetric interpolated breath-hold images can be obtained reliably and reproducibly when timed by using a test dose of contrast material.
Index terms: Liver, blood supply, 761.92 • Liver, MR, 761.121412, 761.121415, 761.121419, 761.12144 • Magnetic resonance (MR), pulse sequences, 761.121412, 761.121415, 761.121419, 761.12144 • Magnetic resonance (MR), technology, 761.121412, 761.121415, 761.121419, 761.12144 • Magnetic resonance (MR), three-dimensional, 761.121419 • Magnetic resonance (MR), treatment planning, 761.121412, 761.121415, 761.121419, 761.12144 • Magnetic resonance (MR), vascular studies, 761.121412, 761.121415, 761.121419, 761.12144
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Second, high spatial resolution helps improve evaluation of small hepatic lesions. Historically, the greatest obstacles to this goal have been the use of relatively large section thicknesses and gaps between two-dimensional (2D) imaging sections, which contribute to partial volume artifacts.
Third, accurate three-dimensional (3D) rendering of segmental and vascular anatomy by means of contrast-enhanced studies can be helpful for surgical planning, including assessment of the feasibility of hepatic resection (10) and hepatic transplantation (11,12). Although these features would be achieved best with isotropic 3D imaging, until recently, dynamic T1-weighted magnetic resonance (MR) imaging of the liver has been limited predominantly to 2D imaging.
In an earlier work (13), we described an MR approach that incorporates interpolation schemes and intermittent fat-saturation pulses to achieve fat-saturated 3D images of the entire liver with nearly isotropic pixel size (approximately 2 mm in all three dimensions) in less than 25 seconds. We refer to this approach as the volumetric interpolated breath-hold examination. On the basis of qualitative and quantitative measures, image quality for this 3D method was shown to be comparable to or improved over conventional 2D gradient-echo imaging when used for nonenhanced or delayed contrast-enhanced imaging of the abdomen. As a breath-hold technique, volumetric interpolated breath-hold examinations can be performed repeatedly following injection of 0.1 mmol of gadopentetate dimeglumine per kilogram of body weight to obtain dynamic 3D images of the liver. We hypothesize that the nearly isotropic resolution can facilitate reconstruction methods such as multiplanar reconstruction and angiographic postprocessing for definition of vascular and segmental anatomy.
The purpose of this study was to evaluate the feasibility and technical quality of an isotropic volumetric interpolated breath-hold examination as a dynamic 3D imaging method in 50 consecutive patients referred for MR examination of the liver, including patients with cirrhosis. To ensure accurate timing of the arterial phase acquisition, a timing examination was implemented by using a test dose of contrast material. To assess reproducibility of the imaging method, we compared baseline results with results from repeat studies performed within 6 months.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Within 6 months of the initial study, 10 patients (20%) were referred for repeat MR examination. These patients underwent imaging with the same imaging protocol, and reproducibility was evaluated.
MR Imaging Protocol
All patients underwent imaging with use of a 1.5-T MR system (Vision; Siemens, Erlangen, Germany) with high-performance gradients (25 mT/m maximum gradient strength and 600-µsec rise time) and a torso phased-array coil. Before the study, a 22-gauge intravenous catheter was placed in an arm vein and attached to an MR-compatible power injector (Spectris; Medrad, Pittsburgh, Pa). Before volumetric interpolated breath-hold examination, all patients underwent imaging with use of fast, short inversion time inversion-recovery and T1-weighted in-phase and out-of-phase gradient-echo sequences according to our routine protocol.
The volumetric interpolated breath-hold examination is a 3D spoiled gradient-echo acquisition with 4.2/1.8 (repetition time msec/echo time msec), a flip angle of 12°, and a bandwidth of 488 Hz/pixel. Sequence details have been reported previously (13) and are summarized briefly as follows: symmetric echo (160 points) in the read direction (kx, left-to-right) and 80% partial Fourier sampling (128 of 160 points) in the phase-encoding direction (ky, anterior-to-posterior), whereby each 160 x 160 partition is interpolated to a 256 x 256 matrix.
By using a field of view of 300–400 mm, in-plane spatial resolution of 2.5 mm or less is attained for the initial matrix of 160 x 160, whereas the imaging matrix of 256 x 256 yields a pixel size of 1.6 mm or less. In the section-select, or partition, direction (kz, superior-to-inferior), asymmetric echo sampling is performed to obtain 40–56 data points that are then interpolated by using zero filling (sinc interpolation) to produce 80–112 partitions. The 3D sequence incorporates a frequency-selected fat-saturation pulse prior to each partition (section) loop. Each partition loop is centric reordered to maximize fat saturation.
Section thicknesses ranged from 160 to 200 mm to ensure full coverage of the liver and yielded a partition thickness of 2.0–2.5 mm. For all studies, anatomic coverage was defined to include the entire liver. The mean acquisition time for volumetric interpolated breath-hold examination was 22.7 seconds (range, 18–28 seconds).
For all patients, a nonenhanced volumetric interpolated breath-hold examination was performed first and was followed by a timing examination and three dynamically acquired contrast-enhanced volumetric interpolated breath-hold acquisitions. The timing examination was performed according to the method described by Earls et al (14) by using a 1-mL test dose of gadopentetate dimeglumine (Magnevist; Berlex, Wayne, NJ). For contrast-enhanced studies, all patients received 19 mL of contrast material for a mean dose of 0.14 mmol/kg (range, 0.08–0.19 mmol/kg). The volumetric interpolated breath-hold acquisition was then repeated twice at 45-second intervals for portal vein and equilibrium phase acquisitions. All volumetric interpolated breath-hold studies were performed during breath holding at the end of expiration. The total table time for the dynamic volumetric interpolated breath-hold portion of the examination, including the timing examination, averaged 9 minutes ± 3 (SD).
Image Analysis
Qualitative assessment.—Overall image quality was evaluated subjectively by two investigators (M.T.L., V.S.L.) by means of consensus. Adequacy of spatial coverage to include the entire liver was assessed. The conspicuity of hepatic arteries for each contrast-enhanced data set was evaluated on a scale of 0–2: 0, no arteries identified; 1, arteries seen but demonstrating mild enhancement; 2, conspicuous arterial enhancement. The conspicuity of the portal vein and segmental branches was evaluated similarly for portal venous phase volumetric interpolated breath-hold acquisitions.
Region-of-interest analysis.—By using the commercial software of our system, region-of-interest analysis was performed by one investigator (M.T.L.). Signal intensity (SI) measurements of the hepatic artery, portal vein, hepatic vein, inferior vena cava, and aorta were recorded for each of the three volumetric interpolated breath-hold acquisitions for all patients. Vessels were identified on the enhanced images, and regions of interest were copied to identical sections on the nonenhanced studies. In addition, SI was measured in retroperitoneal fat, air, and liver, with care taken to exclude vessels from the regions of interest.
Several parameters were calculated from the measured SI. For all tissues and vessels, contrast-to-noise ratio (CNR) was calculated by using retroperitoneal fat as the reference and was defined as (SIT - SIfat)/noise, where SIT is the SI of tissues and vessels and SIfat is the SI of fat. Noise was defined as the SD of SI measured in air outside the body; midline areas that could be affected by ghosting from vessels and respiratory artifact were avoided.
Relative enhancement was defined as (SIpost - SIpre)/SIpre, where SIpost is the SI after the administration of contrast material and SIpre is the SI before the administration of contrast material, and was measured for the liver and all vessels. The venous-to-arterial enhancement ratio for a given enhanced study was defined as (SIpostV - SIpreV)/(SIpostA - SIpreA), where V is vein, A is artery, and a ratio of 0 represents the ideal arterial phase image free of venous enhancement. This ratio was used as a measure of portal venous and hepatic venous enhancement relative to hepatic arterial enhancement during arterial phase volumetric interpolated breath-hold examination.
We also calculated the percentage of total hepatic parenchymal enhancement seen during arterial phase imaging as (SIapL - SIpreL)/(SIpvpL - SIpreL), where SIap is the SI during the arterial phase, SIpvp is the SI during the portal venous phase, and L is liver.
Hepatic anatomic assessment.—Patterns of hepatic vascular (arterial, portal venous, and hepatic venous) anatomy were assessed retrospectively by two investigators (V.S.L, M.T.L.) by means of consensus by using commercially available software (NUMARIS and VIRTUOSO; Siemens) to obtain multiplanar reconstruction images of the source data. Hepatic arterial anatomy was classified according to the Michels classification (15), and additional variants were specified. Briefly, the Michels classification is summarized as follows: I, proper hepatic artery divides into right and left hepatic arteries ("normal anatomy"); II, left hepatic artery replaced to left gastric artery; III, right hepatic artery replaced to superior mesenteric artery; IV, both right and left hepatic arteries replaced; V, accessory left hepatic artery replaced to left gastric artery; VI, accessory right hepatic artery replaced to superior mesenteric artery; VII, accessory right and left hepatic arteries; VIII, replaced right hepatic artery and accessory left hepatic artery or, conversely, replaced left hepatic artery with accessory right hepatic artery; IX, proper hepatic artery arising from the superior mesenteric artery; and X, proper hepatic artery arising from the left gastric artery.
Patterns of hepatic venous and portal venous anatomy were also classified by using source images (16–18). Variations in the confluence of the main hepatic veins were noted. The presence of any inferior hepatic veins was recorded, as was depiction of separate hepatic veins directly draining the caudate lobe. The pattern of intrahepatic portal venous branching was characterized, and the number of segmental portal branches seen was recorded. The presence of portal venous thrombosis and cavernous transformation was also assessed.
Statistical Analysis
The two-tailed Student t test was used to compare quantitative image parameters measured for all volumetric interpolated breath-hold acquisitions. Patient circulation time and image quality parameters were also compared by using the Student t test for patients with (n = 17) and those without (n = 34) cirrhosis of the liver as determined by means of MR imaging features (19). For studies of reproducibility, image quality parameters for baseline and repeat examinations were compared by using paired t test analysis.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
Image Quality
In two studies (4%), a small portion of the dome of the liver was excluded from the dynamic volumetric interpolated breath-hold study, presumably because of variations in patient breath-holding technique. Overall, conspicuity of the hepatic arteries on arterial phase volumetric interpolated breath-hold studies was high (score = 2) for all but one patient (98%); in the one patient with low arterial conspicuity (score = 0), arterial anatomy was defined by using portal venous phase images. Portal venous enhancement was graded as highly conspicuous on all but one study (98%).
CNRs for liver and vascular structures relative to retroperitoneal fat for nonenhanced, arterial, portal venous, and equilibrium phase studies are shown in Table 1.
|
Mean SI values of liver and major vessels for each of the volumetric interpolated breath-hold acquisitions are plotted in Figure 2. During arterial phase, aortic SI demonstrated an increase greater than 600% over nonenhanced images and was significantly greater during arterial phase than during portal venous or equilibrium phases (P < .05). Hepatic venous and inferior vena caval enhancement were not significantly different between portal venous and equilibrium phases (P > .4), whereas hepatic arterial enhancement decreased slightly but not significantly on portal venous phase studies (P > .2).
|
Hepatic Vascular Anatomy
The distribution of hepatic arterial anatomic variants is shown in Table 2, along with the original observations reported by Michels (15) (Fig 3). One of the six patients with an accessory left hepatic artery replaced to the left gastric artery (Michels type V) also had a variant of the right hepatic artery—the vessel arose off the proximal portion of the celiac artery. Of the three patients (6%) not classified according to one of the 10 original Michels types, one had a right hepatic artery arising directly from the aorta (Fig 4); two had right hepatic arteries arising from the proximal celiac artery and, in addition, in these two, one also had a left gastric artery arising directly from the aorta and the other had a middle hepatic artery arising directly from the aorta.
|
|
|
|
|
Patients with Cirrhosis versus Patients without Cirrhosis
Volumetric interpolated breath-hold image quality was compared in patients with (n = 17) and those without (n = 34) MR evidence of cirrhosis (Table 3). Parameters including CNR, relative enhancement ratios, and patient circulation times were all not significantly different for each contrast-enhanced volumetric interpolated breath-hold study. However, portal venous CNR, relative to retroperitoneal fat, was consistently lower in patients with cirrhosis compared with those without cirrhosis, although this difference did not achieve statistical significance (P = .06–.25) (Table 3). The venous-to-arterial enhancement ratio of the portal vein to the hepatic artery was significantly lower in patients with cirrhosis, averaging 0.22 ± 0.24 versus 0.47 ± 0.42 for patients without cirrhosis (P = .03).
|
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Arterial phase CT or MR imaging typically is performed following a fixed imaging delay, usually 15–30 seconds, after the start of intravenous injection of contrast material. However, imaging results vary as a result of differences in circulatory physiology among individuals (21). Using a test bolus of contrast material to time dynamic 2D gradient-echo MR acquisitions in a study of 28 patients, Earls et al (22) found that the transit time from arm vein to arterial enhancement (patient circulation time) varied between 8 and 32 seconds, whereas peak portal venous enhancement started between 20 and 55 seconds. In the same study, successful arterial phase examinations were achieved in 26 (93%) of 28 patients whose studies were timed to circulation time compared with only 17 (61%) of 28 of those who underwent imaging following a fixed 20-second delay after injection of contrast material.
We implemented a similar test bolus timing examination to obtain arterial phase 3D acquisitions and found the approach efficient and reliable. Overall, 49 (98%) of 50 studies were considered subjectively to have excellent conspicuity of arterial vessels during the arterial phase acquisition. These observations are supported by measurements of a nearly 40-fold increase in hepatic arterial CNR during arterial phase compared with that in nonenhanced images (Table 1). In a reproducibility study in 10 patients, a volumetric interpolated breath-hold examination in which a timing examination was used resulted in comparable MR imaging results, despite differences in circulation times of up to 8 seconds between studies.
These findings highlight the importance of customizing the timing of arterial phase acquisitions for each individual at the time of examination. Overall, the test bolus technique adds less than 4 minutes to table time and has the advantage of being implemented easily with any commercial system without special software or hardware (22,23). We emphasize, however, that volumetric interpolated breath-hold examination does not require a timing examination and can also be performed with conventional fixed imaging delays.
One potential limitation to the test bolus approach is the variable effect of breath holding on circulation time, because the test bolus examination is performed during free breathing, whereas volumetric interpolated breath-hold examination is performed during end-expiration breath holding. Krinsky et al (24) found that differences in circulation time with breath holding were as much as 6 seconds when measured at the level of the carotid arteries. One potential solution that remains to be explored is the use of breath holding during the initial portion of the 1-minute test bolus study.
Since periods of hepatic arterial and portal venous enhancement overlap, for acquisition periods that last 20–25 seconds, a mild degree of portal venous enhancement during arterial phase acquisitions is not unexpected. Hepatic parenchymal enhancement during the arterial phase averaged 29% of maximal enhancement, which is comparable to the expected 25%–30% were hepatic arterial supply the only contributor to enhancement (21). During arterial phase, portal venous enhancement, though not insubstantial, remained less than 40% of arterial enhancement, and similarly, hepatic venous enhancement averaged 15% of arterial enhancement. These results are consistent with our observation that angiographic images, including maximum intensity projection images, can be produced from arterial phase data without substantially overlapping venous contamination (Figs 1, 3, 4).
Although we did not observe significant differences in enhancement patterns or patient circulation times in patients with cirrhosis compared with those without, we did observe slightly less enhancement of the portal vein on images obtained during both portal venous and equilibrium enhancement phases in patients with cirrhosis, although differences did not achieve statistical significance. Our results do not support the findings of Soyer et al (25), who observed that patients with cirrhosis have delayed parenchymal enhancement; however, a limitation of our study was the reliance on MR imaging, which may be less sensitive than histopathologic sampling, for diagnosis of cirrhosis. Better correlation between delayed or impaired portal venous enhancement with severity of portal hypertension may be possible with the advent of fast phase-contrast measurements of portal venous flow patterns (26).
With nearly isotropic 3D imaging, the pitfalls of intersection gaps and partial volume artifacts associated with 2D imaging can be avoided (27,28). Other advantages include the opportunity to reformat images in any plane without loss of image resolution. The 3D approach obviates separate repeat 2D acquisitions in additional imaging planes, which are inevitably obtained in the delayed venous phase, that may be required when a 2D strategy is used. We hypothesize that by allowing a 3D assessment of patterns of contrast enhancement and washout, dynamic volumetric interpolated breath-hold examination may improve diagnostic accuracy and confidence in evaluation of parenchymal disease, in particular small hepatic lesions such as hypervascular masses and hemangiomas (Fig 6).
|
|
|
In addition to lesion depiction and characterization, arterial phase volumetric interpolated breath-hold acquisitions can be used to obtain MR angiograms and venograms without additional imaging time or contrast material (Figs 1, 3–6). Accurate knowledge of hepatic arterial anatomy is helpful in several clinical settings, including surgical planning for hepatic transplantation (11) or resection or in planning catheter-related interventional procedures such as chemoembolization. This capability may become even more critical as transplantation with hepatic tissue resected from living related donors becomes more popular (29). Moreover, the ability to detect both vascular and extravascular disease by using the volumetric interpolated breath-hold approach lends itself to evaluation of cirrhosis or hepatoma and to the postoperative evaluation of hepatic transplants (30).
One limitation of this study is that conventional angiographic confirmation was not available. However, the distribution of anatomic variants in our study is comparable to that in Michels' original findings (Table 2). Using a 36-second MR sequence with similar effective pixel size and 0.15 mmol/kg gadopentetate dimeglumine, Kopka et al (31) recently demonstrated agreement between MR angiographic assessment and conventional angiographic assessment of hepatic arterial anatomy in 57 (95%) of 60 patients.
The identification of hepatic portal venous variants can be critical in surgical planning (32–34). For example, an accessory inferior hepatic vein can allow preservation of the posteroinferior area of the right lobe despite transection of the right hepatic vein in partial hepatectomy (33). We identified more inferior hepatic veins (seen in 22 of 50 patients) than previously reported by using CT (18) and ultrasonographic (35) methods, and we attribute the difference to greater spatial resolution and CNR with our technique and to the ability to perform detailed retrospective analysis by using multiplanar reformations. Variations in portal venous anatomy, such as the left portal vein arising from a right portal venous branch, may alter segmental definition and also substantially affect surgical approach (34). Moreover, multiplanar reformatting of venous phase 3D image sets potentially can improve definition of segmental anatomy (36) and increase the confidence with which hepatic tumors can be localized (37–39).
In conclusion, volumetric interpolated breath-hold examination provides high-quality nearly isotropic 3D images for evaluation of hepatic disease and can be performed dynamically with acquisition times of less than 25 seconds. When timed by using a test dose of contrast material, arterial phase 3D volumetric MR images can be obtained reliably and reproducibly for potentially improved depiction and characterization of hepatic lesions, with the added advantage of multiplanar and angiographic reconstruction images that can facilitate surgical or catheter-related interventional planning.
|
Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, V.S.L.; study concepts and design, V.S.L., N.M.R.; definition of intellectual content, V.S.L., N.M.R., G.L., D.M.T.; literature research, V.S.L., N.M.R., G.A.K.; clinical studies, V.S.L., N.M.R., G.A.K.; data acquisition and analysis, M.T.L., V.S.L.; statistical analysis, V.S.L.; manuscript preparation, V.S.L.; manuscript editing and review, V.S.L., N.M.R., G.A.K., D.M.T., G.L., J.C.W.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
1.5 cm) malignant hepatic neoplasms. AJR Am J Roentgenol 1995; 164:879-884.This article has been cited by other articles:
|
|
 |
|
  S. H. Choi, J. M. Lee, N. C. Yu, K.-S. Suh, J.-J. Jang, S. H. Kim, and B. I. Choi Hepatocellular Carcinoma in Liver Transplantation Candidates: Detection with Gadobenate Dimeglumine-Enhanced MRI Am. J. Roentgenol., August 1, 2008; 191(2): 529 - 536. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Willatt, H. K. Hussain, S. Adusumilli, and J. A. Marrero MR Imaging of Hepatocellular Carcinoma in the Cirrhotic Liver: Challenges and Controversies Radiology, May 1, 2008; 247(2): 311 - 330. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. S. Park, J. M. Lee, J.-Y. Choi, M. W. Lee, H. J. Kim, J. K. Han, and B. I. Choi Preoperative Evaluation of Bile Duct Cancer: MRI Combined with MR Cholangiopancreatography Versus MDCT with Direct Cholangiography Am. J. Roentgenol., February 1, 2008; 190(2): 396 - 405. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Heilmaier, R. Sutter, A. M. Lutz, B. Seifert, D. Weishaupt, B. Marincek, and J. K. Willmann Mapping of Hepatic Vascular Anatomy: Dynamic Contrast-enhanced Parallel MR Imaging Compared with 64 Detector Row CT Radiology, December 1, 2007; 245(3): 872 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M J White, R L O'Gorman, E M Charles-Edwards, P A Kane, J B Karani, M O Leach, and J J Totman Parametric mapping of the hepatic perfusion index with gadolinium-enhanced volumetric MRI Br. J. Radiol., February 1, 2007; 80(950): 113 - 120. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Hecht, A. E. Holland, G. M. Israel, W. Y. Hahn, D. C. Kim, A. B. West, J. S. Babb, B. Taouli, V. S. Lee, and G. A. Krinsky Hepatocellular Carcinoma in the Cirrhotic Liver: Gadolinium-enhanced 3D T1-weighted MR Imaging as a Stand-alone Sequence for Diagnosis. Radiology, May 1, 2006; 239(2): 438 - 447. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Ward, P. J. Robinson, J. A. Guthrie, S. Downing, D. Wilson, J. P. A. Lodge, K. R. Prasad, G. J. Toogood, and J. I. Wyatt Liver Metastases in Candidates for Hepatic Resection: Comparison of Helical CT and Gadolinium- and SPIO-enhanced MR Imaging Radiology, October 1, 2005; 237(1): 170 - 180. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J J Totman, R L O'Gorman, P A Kane, and J B Karani Comparison of the hepatic perfusion index measured with gadolinium-enhanced volumetric MRI in controls and in patients with colorectal cancer Br. J. Radiol., February 1, 2005; 78(926): 105 - 109. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kataoka, H. Ueda, T. Koyama, S. Umeoka, K. Togashi, R. Asato, S. Tanaka, and J. Ito Contrast-Enhanced Volumetric Interpolated Breath-Hold Examination Compared with Spin-Echo T1-Weighted Imaging of Head and Neck Tumors Am. J. Roentgenol., January 1, 2005; 184(1): 313 - 319. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
= aorta, 
= liver, • = portal vein, 
= hepatic vein.
