| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Diagnostic Radiology, Chonbuk National University Hospital,
Conju, Korea.
2 Department of Diagnostic Radiology, Wonkwang University School of Medicine,
Iksan, Korea.
3 Department of Diagnostic Radiology, Korean Keyryong Army Hospital, Daejeon,
Korea.
4 Department of Radiology, Seoul National University Hospital and College of
Medicine, 28, Yongon-Dong, Chongno-Gu, Seoul 110-744, Korea.
Received August 27, 2003;
accepted after revision November 11, 2003.
Address correspondence to J. M. Lee.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Twenty-nine consecutive patients with 35 hepatocellular carcinomas underwent gadobenate dimeglumine–enhanced MRI (unenhanced, arterial, portal, and equilibrium phases) using 3D fat-saturated volumetric interpolated imaging and SPIO-enhanced MRI on a 1.5-T unit. SPIO-enhanced T2-weighted turbo spin-echo and T2*-weighted gradient-echo sequences were performed 48 hr after completion of the dynamic study. Three observers independently interpreted the images in random order, separately, and without patient identifiers. Diagnostic accuracy was evaluated using the alternative free response receiver operating characteristic method. Sensitivity and positive predictive value were also evaluated.
RESULTS. The mean sensitivity and positive predictive value of SPIO-enhanced imaging were 81.0% and 85.0%, respectively, and those of gadobenate dimeglumine–enhanced MRI were 91.4% and 88.1%, respectively. A significant difference was seen in the sensitivity of the two MRI examinations (p < 0.05). The mean value of the area under the ROC curve (Az) for gadobenate dimeglumine–enhanced imaging (Az = 0.97 ± 0.01) was significantly higher than that for SPIO-enhanced imaging (Az = 0.90 ± 0.02) (p = 0.004).
CONCLUSION. Gadobenate dimeglumine–enhanced 3D dynamic imaging showed better diagnostic performance than SPIO-enhanced imaging for the detection of hepatocellular carcinomas.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Gadobenate dimeglumine (MultiHance, Bracco) is a gadolinium-based paramagnetic contrast agent that combines the properties of a conventional nonspecific gadolinium-based agent with those of a liver-targeted agent [15, 16]. Unlike conventional gadolinium chelates, gadobenate dimeglumine has almost twofold greater T1 relaxivity, and 3–5% of the injected dose is taken up by functioning hepatocytes and excreted in the bile [17]. Detection of hepatocellular carcinomas with extracellular fluid contrast agents on dynamic arterial phase CT or MRI is of great importance because of the increased arterial supply in most hepatocellular carcinoma tumors [18]. Yoshimitsu et al. [19] showed the usefulness of dynamic MRI with double-dose gadolinium for the detection of hypervascular hepatocellular carcinomas. The twofold greater T1-relaxivity of gadobenate dimeglumine over that of conventional gadolinium chelates led us to propose that gadobenate dimeglumine–enhanced MRI may be superior to SPIO-enhanced MRI for the detection of hepatocellular carcinomas.
To our knowledge, no comparative study has been performed of SPIO-enhanced versus gadobenate dimeglumine–enhanced MRI for the detection of hepatocellular carcinomas. The purpose of our study is to compare the diagnostic performance of SPIO-enhanced MRI with that of gadobenate dimeglumine–enhanced MRI for the detection of hepatocellular carcinomas using alternative free response receiver operating characteristic (ROC) analysis [20] with multiple observers.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In all patients, liver cirrhosis was determined by clinical findings, blood chemistry tests (aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, bilirubin, albumin, and globulin), and sonographically guided biopsy (four patients). The severity of the hepatic disease was evaluated according to the Child-Pugh classification [21]. Twelve patients were classified as Child-Pugh classification A (mild liver cirrhosis), 13 as classification B (moderate liver cirrhosis), and four as classification C (severe liver cirrhosis).
Lesion Confirmation
Thirty-five tumors (size range, 0.5–2 cm; mean, 1.5 ± 0.4 cm)
in 29 patients were included in this study. Twenty-five patients had solitary
lesions, two patients had two lesions, and two patients had three lesions. The
final diagnosis of hepatocellular carcinoma was proven by surgical specimens
in seven patients and by core needle biopsy in 10 patients. The remaining 12
patients underwent transarterial chemoembolization and had no histologic
confirmation; their diagnoses of hepatocellular carcinoma depended on a
combination of clinical findings such as liver cirrhosis and viral hepatitis
B, characteristic angiographic findings, elevated serum 
-fetoprotein
levels (> 400 ng/mL), and iodized oil (Lipiodol, Guerbet) uptake on CT
[22].
For iodized oil CT, hepatic angiography was performed by one interventional radiologist using a digital angiographic unit (Angiostar, Siemens). An emulsion of 5–10 mL of iodized oil (Lipiodol) and anticancer drugs—20–50 mg of doxorubicin hydrochloride (Adriamycin, Pharmacia & Upjohn) and 4–10 mg of mitomycin (Mitomycin-C, Kyowa Hakko Kogyo)—followed by gelatin sponge particles (Gelfoam, Upjohn) was injected through a catheter, the tip of which was placed superselectively into the segmental or subsegmental artery feeding the tumor. At follow-up CT of the liver 1 month after the procedure, nodular areas of retained Lipiodol were diagnosed as hepatocellular carcinoma by consensus of three radiologists.
Determination of the total number of hepatocellular carcinomas in seven patients who underwent hepatic surgery was based on pathologic analysis of the surgical specimens and intraoperative sonography. If additional hepatic nodules suspected of being hepatocellular carcinoma were found on intraoperative sonography, immediate frozen section analysis was done by a hepatobiliary pathologist. For frozen section analysis, a surgical specimen was quickly frozen at –20°C and stained with H and E; in this way, one small hepatocellular carcinoma was found that had not been detected on MRI. In 20 patients with 25 hepatocellular carcinomas who underwent transcatheter arterial chemoembolization, the standard of reference for the presence of hepatocellular carcinoma was the combined results of iodized oil CT after transcatheter arterial chemoembolization and CT angiography. Diagnoses of the remaining two patients with two hepatocellular carcinomas, both of whom underwent radiofrequency thermal ablation, were based on the results of imaging-guided biopsy and the findings of 13- and 15-month follow-up CT and MRI, respectively. At 13- and 15-month follow-up imaging, no new liver mass was found, so the total number of hepatocellular carcinomas in these two patients was regarded as one lesion. Follow-up contrast-enhanced three-phase helical CT or MRI was performed for a minimum of 6 months (range, 6–22 months) in all 29 patients.
MRI
All MRI was performed on a 1.5-T superconducting imager (Magnetom Symphony,
Siemens) with phased array multicoils for signal reception. The liver was
imaged in the axial plane in all the following sequences: baseline MRI,
including a respiratory-triggered T2-weighted turbo spin-echo sequence; a
breath-hold T2*-weighted fast imaging with steady-state free precession (FISP)
sequence; and a breath-hold T1-weighted fast low-angle shot (FLASH) sequence.
Respiratory-triggered T2-weighted turbo spin-echo imaging was performed using
parameters of TR range/TE, 3,300–5,500/85; echo-train length, 5; matrix,
120 x 256; and a signal average of 2. Breath-hold T2*-weighted FISP
imaging was performed using parameters of TR/TE, 180/12; flip angle, 30°;
matrix, 96 x 256; and a signal average of 1. Breath-hold T1-weighted
FLASH imaging was performed with parameters of 120/4; flip angle, 70°;
matrix, 120 x 256; and a signal average of 1. For all sequences, a 7-mm
slice thickness was used with a 10% intersection gap and a field of view of
35–40 cm, depending on the size of the liver.
Dynamic imaging was performed after the administration of gadobenate dimeglumine at a dose of 0.1 mmol/kg followed by a 20-mL saline flush. Determination of the scanning delay for image acquisition was achieved by a test bolus technique in which 1 mL of gadobenate dimeglumine was injected with a saline flush; the vessel of interest (abdominal aorta) was then scanned approximately once per second. T1-weighted imaging in the arterial (20–35 sec), portal (45–60 sec), and equilibrium (3 min) phases was performed using 3D Fourier transform gradient-echo imaging (volumetric interpolated breath-hold examination, Siemens) using the following parameters: 3.4/1.5; flip angle, 12°; bandwidth, 490 Hz/pixel; matrix, 256 (readout direction) x 120 (phase) x 64–72 (partition); effective slice thickness, 2.3 mm; and field of view, 32–35 cm [23]. All images were acquired during breath-hold, and sampling was done by 70% in the z direction and by 70% in the direction of phase using volumetric interpolation. Image reconstruction with 6-mm thickness was performed with source images at an MRI workstation.
After completion of the dynamic MRI, SPIO-enhanced MRI was performed after an interval of 48 hr. SPIO-enhanced imaging comprised a respiratory-triggered T2-weighted turbo spin-echo sequence, a breath-hold T2*-weighted FISP sequence, and a breath-hold T1-weighted FLASH sequence, with the same parameters as those used for the baseline MRI. The SPIO agent ([ferumoxides] Feridex, Advanced Magnetics), at a dose of 15 µmol of iron per kilogram of body weight, was diluted in 100 mL of a 5% dextrose solution and injected IV through a specific 5-µm filter for 30 min. Imaging began approximately 70 min (range, 50–90 min) after the IV infusion of SPIO.
Image Analysis
All MR images were evaluated independently and separately by three
gastrointestinal radiologists who had at least 10 years' experience in
interpreting MR images of the liver. The three radiologists were unaware of
the design of this study. These radiologists knew that the patients had liver
cirrhosis and were at risk for hepatocellular carcinoma but were unaware of
the results of all the other imaging findings or the final diagnosis. Two sets
of images were analyzed: 3D dynamic gadobenate dimeglumine– enhanced
images (arterial, portal, and equilibrium phases) and two sequences of
SPIO-enhanced T2-weighted images (T2*-weighted and respiratory-triggered
T2-weighted turbo spin-echo images). To minimize any learning bias, a 3-week
interval was inserted between the two interpretations. Hard-copy images were
reviewed for both gadobenate dimeglumine–enhanced and SPIO-enhanced
images.
Each observer recorded the presence and the segmental location of the lesions, assigning each a confidence level on a 5-point scale: 1, definitely or almost definitely absent; 2, probably absent; 3, possibly present; 4, probably present; and 5, definitely or almost definitely present. To avoid a mismatch between the findings of the scored lesions and those of the gold standard for determining the total number of lesions, each observer recorded the individual image number, the segmental location of all lesions, and the size of each lesion. In the three patients who had two lesions in one segment, the observers added further description of the size and location of the mass in each segment to avoid confusion in the data analysis.
On the basis of the reviews of the three observers, alternative free response ROC curve analysis was performed on a tumor-by-tumor basis [20]. For each imaging set, an alternative free response ROC curve was fitted to each observer's confidence rating data using a maximum likelihood estimation program (ROCKIT 0.9B, Metz CE). The diagnostic accuracy of each imaging set and observer was determined by calculating the area under the ROC curve (Az). The differences between imaging sets in terms of the mean area under the ROC curves were statistically analyzed using the two-tailed Student's t test for paired data. The sensitivity and positive predictive values for 3D dynamic gadobenate dimeglumine–enhanced images and SPIO-enhanced images were then calculated. The sensitivity of each observer and each set of images was determined by the number of lesions assigned a confidence level of 4 or 5 from among the 35 hepatocellular carcinomas. The sensitivity and positive predictive value were compared using the Student's t test. A two-tailed p value of less than 0.05 was considered significant. Agreement between blinded observers is reported in terms of kappa values, with values greater than 0 indicating positive correlation. Values up to 0.4 indicate positive but poor correlation, and those of 0.41–0.75 indicate good correlation.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
The sensitivity and positive predictive value for each observer and for each technique were calculated, and the mean values were also determined (Table 2). Overall, a trend was seen toward increased sensitivity for gadobenate dimeglumine–enhanced images compared with SPIO-enhanced images, and the mean sensitivities of gadobenate dimeglumine–enhanced images were significantly greater than those of SPIO-enhanced images (91.4% vs 81.0%). Among 35 lesions, one was not detected by any observer on either SPIO-enhanced images or gadobenate dimeglumine–enhanced images. This lesion, which could not be seen on retrospective viewing, was about 4 mm in diameter and was confirmed on surgical resection. Three small (0.5-, 0.6-, 0.8-cm) lesions were not detected by any observer on SPIO-enhanced images but were detected on gadobenate dimeglumine–enhanced dynamic arterial phase images (Fig. 1A, 1B, 1C, 1D). However, two (0.4- and 2.0-cm) lesions were not detected by any observer on gadobenate dimeglumine–enhanced images but were detected on SPIO-enhanced images (Fig. 2A, 2B, 2C). Retrospective analysis of the missed lesions by all observers showed that two of the three missed lesions on SPIO-enhanced images showed decreased signal intensity by SPIO uptake and the other lesion was misdiagnosed as a vessel. On the gadobenate dimeglumine–enhanced images, the two lesions missed by all observers showed equal intensity to surrounding liver parenchyma but were confirmed at percutaneous biopsy.
|
|
|
|
|
|
|
|
Regarding the positive predictive values, no significant difference was seen between gadobenate dimeglumine–enhanced and SPIO-enhanced images (88.1% vs 85.0%). For all observers, 15 false-positive findings were found on SPIO-enhanced images and 13 false-positive findings on gadobenate dimeglumine–enhanced images. On SPIO-enhanced images, all false-positive lesions were smaller than 1 cm, and intrahepatic vessels (n = 8) or fibrosis (n = 7) were mistaken for tumor (Fig. 3A, 3B). On gadobenate dimeglumine–enhanced images, arterioportal shunts (n = 9) and vessels (n = 4) were mistaken for tumor (Fig. 4A, 4B, 4C).
|
|
|
|
|
For SPIO- and gadobenate dimeglumine–enhanced MR images, the kappa values for the three observers were 0.607–0.762, indicating good or excellent interobserver agreement for the presence of lesions (Table 3).
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our study results show that in alternative free response ROC analysis, all three observers achieved higher performance with 3D dynamic gadobenate dimeglumine–enhanced imaging (mean Az, 0.966) than with SPIO-enhanced imaging (mean Az, 0.902) (p = 0.004). In terms of sensitivity, gadobenate dimeglumine–enhanced MRI performed better than SPIO-enhanced MRI (91.4% vs 81.0%) (p < 0.05). Our results agree with those of Pauleit et al. [14], who showed the greater diagnostic capabilities of dynamic gadolinium-enhanced MRI compared with SPIO-enhanced MRI for detecting hepatocellular carcinoma. Furthermore, comparing the diagnostic performance of our study with that of Pauleit et al., the mean sensitivity and Az values of gadobenate dimeglumine–enhanced imaging and SPIO-enhanced imaging in our study are higher than in their study. Despite the fact that the hepatocellular carcinomas in our study were small, our diagnostic performance was comparable to or more accurate than that of previous studies composed of average-sized hepatocellular carcinomas greater than 2 cm in diameter [12–14, 24–26].
The superior diagnostic performance of gadobenate dimeglumine–enhanced dynamic MRI compared with SPIO-enhanced MRI in our study or the gadolinium-enhanced MRI of a previous study [14] may be explained by the use of a modified 3D gradient-echo technique and an almost twofold greater T1-relaxivity of gadobenate dimeglumine relative to a conventional gadolinium-based agent. In this study, we performed gadobenate dimeglumine–enhanced dynamic imaging using a modified 3D gradient-echo technique (i.e., volumetric interpolated breath-hold examination [22]) and mechanical injection using a test-bolus technique for optimization of contrast delivery to the hepatic arterial system and the tumor. Volumetric interpolated breath-hold examination, a modified 3D gradient technique, is optimized for short acquisition times and achieves high-spatial-resolution imaging for excellent anatomic detail through the use of asymmetric k-space sampling and interpolation [22]. In our study, the thinner slices of 3D dynamic imaging compared with those of SPIO-enhanced T2-weighted imaging might contribute to the greater diagnostic accuracy of gadobenate dimeglumine–enhanced imaging both for the detection of small hepatocellular carcinomas and for small lesion characterization. Gadobenate dimeglumine is the fifth gadolinium chelate with the potential to improve hepatocellular carcinoma detection because of its higher T1-relaxivity [15, 16]. We believe that the combination of a newly developed 3D dynamic MRI sequence and an MRI contrast agent with high T1 relaxivity can satisfy the twofold requirement of a liver imaging technique that is accurate in lesion detection and characterization, at least for the detection of hepatocellular carcinomas.
The false-positive findings of both gadobenate dimeglumine–enhanced MRI and SPIO-enhanced MRI were similar for both types of imaging (Table 2). Although it has been reported that SPIO-enhanced imaging has limited specificity for detecting small hepatocellular carcinomas because the vascular structures and fibrosis can be mistaken for tumors [27, 28], the false-positive findings and positive predictive value of SPIO-enhanced imaging in our study were comparable to those of gadobenate dimeglumine–enhanced imaging. The difference in false-positive findings between our study and previous studies may be explained by the small number of patients corresponding to Child-Pugh classification C in our study and the two kinds of SPIO-enhanced images in our study (i.e., respiratory-triggered T2-weighted turbo spin-echo sequences and breath-hold T2*-weighted FISP sequences), both of which may contribute to fewer false-positive findings.
A limitation of our study is that not all lesions were histologically confirmed. However, acquiring histologic confirmation of all tumors would have been difficult because only a small percentage of the patients with hepatocellular carcinomas and liver cirrhosis were candidates for surgery. We assume that the most precise determination of the total number of lesions is possible only through liver transplantation and matched-pair analysis performed to verify that the lesions detected on MRI correspond to lesions in the resected liver specimen. However, in our study, we included only patients with firm evidence of the presence of hepatocellular carcinomas on studies that included intraoperative sonography and biopsy or the combined assessment of hepatic angiography, CT during arterioportography, and iodized oil CT, with at least 6-month follow-up of all studies.
In conclusion, for detecting hepatocellular carcinomas, gadobenate dimeglumine–enhanced dynamic imaging using 3D T1-weighted gradient-echo sequences showed better diagnostic performance than SPIO-enhanced imaging. Therefore, in cases in which only one MRI contrast agent should be used for evaluating patients with hepatocellular carcinomas, we recommend gadobenate dimeglumine–enhanced dynamic imaging using a 3D T1-weighted gradient-echo sequence because this technique offers a greater rate of detection than does SPIO-enhanced MRI, and at an acceptable specificity.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1.5cm) malignant hepatic neoplasms.
AJR 1995;164:879
–884
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  D. T. Boll and E. M. Merkle Diffuse Liver Disease: Strategies for Hepatic CT and MR Imaging RadioGraphics, October 1, 2009; 29(6): 1591 - 1614. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y K KIM, C S KIM, and Y M HAN Detection of small hepatocellular carcinoma: comparison of conventional gadolinium-enhanced MRI with gadolinium-enhanced MRI after the administration of ferucarbotran Br. J. Radiol., June 1, 2009; 82(978): 468 - 484. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
|
|
R. F. Hanna, D. A. Aguirre, N. Kased, S. C. Emery, M. R. Peterson, and C. B. Sirlin Cirrhosis-associated Hepatocellular Nodules: Correlation of Histopathologic and MR Imaging Features RadioGraphics, May 1, 2008; 28(3): 747 - 769. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. F. Hanna, N. Kased, S. W. Kwan, A. C. Gamst, A. C. Santosa, T. Hassanein, and C. B. Sirlin Double-Contrast MRI for Accurate Staging of Hepatocellular Carcinoma in Patients with Cirrhosis Am. J. Roentgenol., January 1, 2008; 190(1): 47 - 57. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. C. Lauenstein, K. Salman, R. Morreira, T. Heffron, J. R. Spivey, E. Martinez, P. Sharma, and D. R. Martin Gadolinium-Enhanced MRI for Tumor Surveillance Before Liver Transplantation: Center-Based Experience Am. J. Roentgenol., September 1, 2007; 189(3): 663 - 670. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
