| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gastrointestinal Imaging |
1 From the Department of Diagnostic Radiology, Research Institute of Radiological Science, Severance Hospital (M.J.K., J.H.K., J.J.C., J.S.L., Y.T.O.), Brain Korea 21 Project for Medical Science (M.J.K.), and Department of Diagnostic Radiology, Yong Dong Severance Hospital (M.S.P.), Yonsei University College of Medicine, Seodaemun-ku, Shinchon-dong 134, Seoul 120-752, Republic of Korea; and Department of Diagnostic Radiology, NHIC Ilsan Hospital, Gyonggi-do, Korea (J.J.C.). Received June 18, 2002; revision requested August 20; final revision received December 29; accepted January 20, 2003. Address correspondence to M.J.K. (e-mail: kimnex@yumc.yonsei.ac.kr).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Unenhanced, dynamic gadolinium-enhanced, and SPIO-enhanced hepatic MR images were obtained in 134 patients. SPIO-enhanced MR imaging was performed immediately after gadolinium-enhanced dynamic MR imaging in 50 patients, 1 day after gadolinium-enhanced dynamic MR imaging in 40 patients, and before gadolinium-enhanced dynamic MR imaging in 44 patients. Two radiologists independently reviewed the gadolinium image set (unenhanced and gadolinium-enhanced dynamic MR images) and the SPIO image set (unenhanced and SPIO-enhanced MR images) in random order. Lesion detection sensitivity and lesion characterization accuracy were compared by analyzing the area under the receiver operating characteristic curve (Az).
RESULTS: Overall lesion detection accuracy for pooled data was significantly higher with the SPIO set (Az = 0.903) than with the gadolinium set (Az = 0.857) (P < .05). When hypovascular lesions were excluded, the detection rate was similar with the two sets. When hepatocellular carcinomas were excluded, the detection rate was significantly higher with the SPIO set (P < .01). Readers were more accurate in differentiating benign from malignant lesions with the gadolinium set (Az = 0.915) than with the SPIO set (Az = 0.847) (P < .01). Detection accuracy tended to be better with the images obtained after the second contrast agent was used.
CONCLUSION: Hypovascular lesion detection was better with SPIO-enhanced MR images than with gadolinium-enhanced MR images. Detection and characterization of hypervascular lesions were improved with gadolinium-enhanced MR images.
© RSNA, 2003
Index terms: Contrast media, comparative studies • Gadolinium • Iron • Liver neoplasms, diagnosis, 761.3121, 761.3199, 761.323, 761.33 • Liver neoplasms, MR, 761.12143 • Magnetic resonance (MR), contrast media
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Several authors have compared gadolinium chelate– and SPIO-enhanced MR images for the detection and characterization of focal hepatic lesions (10–14). Blakeborough et al (10) reported that SPIO-enhanced MR images had better sensitivity than gadolinium-enhanced dynamic MR images in the detection of focal hepatic lesions. However, most of the subjects in the study of Blakeborough et al (10) had metastatic lesions rather than primary hepatic tumors. Vogl et al (11) reported that significantly more hepatocellular carcinoma (HCC) lesions were detected on SPIO-enhanced MR images than on gadolinium-enhanced MR images. However, Tang et al (12) reported that gadolinium-enhanced MR imaging revealed more HCC lesions than SPIO-enhanced MR imaging. This difference may be attributed to tumor size differences in the two studies, because the greater accuracy of gadolinium-enhanced MR imaging was noted mainly in the detection of small (<1.5–2 cm) HCCs (12,14). Moreover, the better performance of gadolinium-enhanced dynamic MR imaging versus SPIO-enhanced MR imaging was found to be more prominent in patients with cirrhosis (13).
Meanwhile, some authors have reported the feasibility and utility of the combined use of a gadolinium chelate and SPIO in a single MR imaging session for evaluating focal hepatic lesions (15–18). In a study by Ward et al (18), additional gadolinium-enhanced dynamic MR imaging was performed immediately after SPIO-enhanced MR imaging and was found to result in significant improvement in HCC detection accuracy as compared with either unenhanced or SPIO-enhanced MR imaging alone. Kondo et al (17) reported that use of a combination of SPIO-enhanced MR imaging and gadolinium-enhanced MR imaging (with SPIO-enhanced MR imaging performed 1 day after gadolinium-enhanced MR imaging), may also improve detection accuracy for malignant hepatic tumors. However, routine use of dual contrast material–enhanced MR imaging is not warranted in clinical practice, and the indications for and recommended sequence of dual contrast-enhanced MR imaging have not been established.
We undertook this study to compare the performances of radiologists in detecting and characterizing focal hepatic lesions at gadolinium- and SPIO-enhanced MR imaging with different contrast agent administration sequences.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Two hundred sixty-four focal hepatic lesions were confirmed either histologically or clinically in these 134 patients. The lesions included 103 HCCs in 55 patients, 72 metastases in 33 patients, 37 hemangiomas in 23 patients, 35 cysts in 28 patients, seven focal inflammatory lesions in two patients, five cholangiocarcinomas in three patients, four focal nodular hyperplasias (FNHs) in four patients, and one bile duct adenoma in one patient. No pathologic lesion was found in 11 patients. Twenty-two patients had two or more types of lesion (two patients had HCC, hemangiomas, and cysts; three patients had HCC and hemangioma; four patients had HCC and cysts; one patient had HCC and bile duct adenoma; seven patients had metastases and cysts; two patients had metastases and hemangiomas; three patients had hemangiomas and cysts; one patient had hemangioma and FNH; and one patient had an inflammatory nodule with cyst). Liver metastases arose from the following primary tumors: colorectal carcinoma (53 lesions in 26 patients), gastric carcinoma (11 lesions in four patients), pancreatic carcinoma (seven lesions in two patients), and renal cell carcinoma (one lesion in one patient). The diameters of all lesions ranged from 0.3 to 12.0 cm (mean, 1.8 cm). Specifically, the diameters of cysts ranged from 0.3 to 4.7 cm (mean, 1.2 cm); the diameters of HCCs, from 0.4 to 12.0 cm (mean, 2.5 cm); the diameters of hemangiomas, from 0.4 to 4.1 cm (mean, 1.4 cm); the diameters of metastases, from 0.6 to 10.1 cm (mean, 1.6 cm); the diameters of FNHs, from 1.2 to 3.4 cm (mean, 2.1 cm); and the diameters of the remaining lesions, from 0.7 to 8.3 cm (mean, 1.8 cm).
Diagnoses were established as follows: A diagnosis of HCC was based on findings at percutaneous biopsy (n = 15 patients), on findings in a surgical specimen (n = 16), or on typical clinical and laboratory findings in combination with the progression of the disease as depicted on follow-up computed tomographic (CT) or single- or dual-contrast–enhanced MR images (n = 24). In patients with metastasis to the liver, diagnosis was based on histologic findings of the primary tumor and rapid disease progression as depicted on serial follow-up images (12 lesions in seven patients), on findings in a percutaneous biopsy specimen (six lesions in three patients), on results of intraoperative ultrasonography (US) and palpation (20 lesions in six patients), or on results of gross examination of excised liver lesions (34 lesions in 17 patients). Hemangiomas were diagnosed on the basis of their typical appearance at either US or CT and on the basis of an absence of growth during a follow-up period of at least 6 months. Cysts were confirmed by typical imaging findings on precontrast T1- and T2-weighted MR images correlated with either US or CT findings. Seven inflammatory lesions in two patients were confirmed on the basis of biopsy results and appropriate clinical findings. Intrahepatic cholangiocarcinomas were diagnosed by means of either surgery (n = 2) or biopsy (n = 1), and FNHs were confirmed at either surgery (n = 2) or biopsy (n = 2). Finally, one bile duct adenoma was demonstrated at surgery.
MR Imaging
MR imaging was performed with 1.5-T MR imaging units (Signa Horizon; GE Medical Systems, Milwaukee, Wis). All images were obtained in the transverse plane by using a phased-array multicoil. A rectangular field of view of 22–24 x 29–32 cm (adjusted for each patient) was held constant for all sequences.
Unenhanced MR imaging was performed in all patients. However, the order of use of the two contrast agents was varied by means of random group assignment. In group I patients (n = 50), SPIO-enhanced imaging was performed immediately after gadolinium-enhanced MR imaging. In group II patients (n = 40), SPIO-enhanced MR imaging was performed 1 day after gadolinium-enhanced MR imaging. In group III patients (n = 44), gadolinium-enhanced dynamic MR imaging was performed immediately after SPIO-enhanced MR imaging.
Unenhanced MR imaging was performed with the following sequences: (a) a respiratory-triggered T2-weighted fast spin-echo (FSE) sequence with an effective repetition time msec/effective echo time msec of 3,500–10,900/96–105, an echo train length of 16, two signals acquired, a matrix of 256 x 256, superior and inferior spatial presaturation and chemically selective fat saturation, and an 8-mm section thickness with a 2-mm gap; (b) a breath-hold T1-weighted fast multiplanar spoiled gradient-recalled-echo (GRE) in-phase sequence (150–200/4.2–4.4); (c) an out-of-phase sequence (120–180/1.5–2.2) with a flip angle of 90°, one signal acquired, a matrix of 256 x 128, a 10-mm section thickness, and zero gap, interleaved; and (d) a breath-hold T2-weighted single-shot half-Fourier sequence with an effective echo time of 180, a matrix of 256 x 160, and an 8-mm section thickness with a 2-mm gap.
Ferumoxides (Feridex I.V.; Advanced Magnetics, Cambridge, Mass) was administered at 15 µmol of iron per kilogram of body weight. The ferumoxides suspension was diluted in 100 mL of a 5% glucose solution and administered intravenously for 30 minutes. Patients underwent postcontrast MR imaging from 30 minutes to 1 hour after the administration of the contrast agent. SPIO-enhanced MR imaging was performed with the following sequences: (a) a respiratory-triggered T2-weighted FSE sequence and (b) a breath-hold T2*-weighted GRE sequence with 120–130/10, a 30° flip angle, a section thickness of 10 mm with zero gap, a matrix of 256 or 512 x 192, one signal acquired, and two or three data acquisitions. The respiratory-triggered T2-weighted FSE MR images were obtained by using the same parameters used to obtain the unenhanced images.
For gadolinium-enhanced dynamic MR imaging, either gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) or gadodiamide (Omniscan; Nycomed Amersham, Oslo, Norway) was injected as a rapid bolus by hand injection at a dose of 0.1 mmol/kg and immediately followed by a 10–20-mL saline flush. Chemically selective fat-suppressed spoiled GRE MR images were obtained in the transverse plane during suspended respiration immediately after intravenous injection of the gadolinium chelate, and additional images were obtained at 30–35 seconds, 65–70 seconds, and 5 minutes. The imaging parameters included 180–200/1.5–2.2, a 90° flip angle, a 256 x 128 matrix with a three-quarter rectangular field of view, one signal acquired, and 8–10-mm-thick sections with zero intersection gap.
Image Analysis
Two experienced gastrointestinal radiologists (J.J.C., M.S.P.) working at other hospitals (NHIC Ilsan Hospital and Yong Dong Severance Hospital, respectively) were invited to retrospectively analyze the images. The radiologists knew only that the patients had been referred for the evaluation of known or suspected focal hepatic malignancies and were unaware of all other information regarding patient history, laboratory results, findings of other imaging modalities, and final diagnosis. They independently reviewed two sets of MR images in random order: (a) unenhanced and gadolinium-enhanced dynamic MR images (the gadolinium set) and (b) unenhanced and SPIO-enhanced MR images (the SPIO set). To minimize learning bias, patients’ names, ages, identification numbers, and imaging parameters were hidden during the review. Intervals between the reviews of the two sets of images were at least 1 month.
Each observer recorded the presence, size (maximum diameter), and site (Couinaud segment) of one or more of the lesions with a four-point confidence rating scale in which a score of 1 indicated that a lesion was probably not present; 2, a lesion was possibly present; 3, a lesion was probably present; and 4, a lesion was definitely present. The ability of an observer to detect a lesion was assessed by using lesions assigned a score of 3 or 4 (18). To prevent lesion misallocation (mismatch), we used a standardized template form for each examination on which the interpreter indicated the segmental location of each lesion. Each reader recorded the individual image number, the segmental location, and the size of each lesion. If multiple lesions with the same identification parameters were scored, the observers added further comments to distinguish the lesions.
Each observer also indicated the possibility of malignancy by using the following scores: 1, definitely benign; 2, probably benign; 3, possibly malignant; 4, probably malignant; and 5, definitely malignant. Observers were aware that for statistical analysis purposes, only scores of 3–5 would be considered as indicating malignant lesions. Each observer also tried to specify the tumor type according to established criteria. For unenhanced MR images, strong hyperintensity on double-echo T2-weighted images, isointensity on T1-weighted images, and hypointensity on T2-weighted images were regarded as signs of benignity. The shape, border, and internal texture of the lesions were also used as references. On gadolinium-enhanced MR images, absence of enhancement (typical of cysts), peripheral globular enhancement with a gradual filling-in pattern (typical of hemangiomas), and homogeneously increased enhancement on arterial phase images with isointensity on delayed phase and both T1- and T2-weighted images (typical of FNH) were considered to indicate benignity. Early enhancement with rapid washout (typical of HCC) and irregular peripheral enhancement with a peripheral washout sign on delayed phase images (typical of cholangiocarcinoma or metastases) were used as indicators of malignancy. At SPIO-enhanced MR imaging, lesions that showed no enhancement were considered to be malignant or cystic (especially when they showed high signal intensity on nonenhanced T2-weighted turbo spin-echo MR images), and lesions that showed enhancement on T2-weighted double-echo turbo spin-echo images were considered to be benign (11,19,20). All images were reviewed by using a local picture archiving and communication system (PACS) monitor and Digital Imaging and Communications in Medicine (DICOM) image viewing software (
-view version 4.XX; Mediface, Seoul, Korea).
Statistical Analysis
An alternative-free response receiver operating characteristic curve was fitted to each reader’s confidence ratings by using a maximum-likelihood estimation (ROCKIT 0.9B; C. E. Metz, University of Chicago, Ill, 1998). Each observer’s performance in detecting and characterizing focal hepatic lesions with each imaging technique and imaging protocol was assessed by using the area under the receiver operating characteristic curve (Az) (21). Differences between the receiver operating characteristic curves were determined by using a univariate z score test. A two-tailed P value of less than .05 was considered to indicate a statistically significant difference (22). So that we could determine whether diagnostic accuracy varied according to the type of lesion, we also plotted receiver operating characteristic curves for the composite data of both readers after hypovascular lesions (ie, metastases, cysts, cholangiocarcinomas, and inflammatory lesions) were excluded. The same procedure was also performed after nonsolid lesions (hemangiomas and cysts) were excluded and after HCCs were excluded.
The 
test was used to assess interobserver variability in terms of lesion detection and differentiation of benign from malignant focal hepatic lesions (23). Degrees of agreement were categorized as follows: values of 0.00–0.20 were considered to indicate poor agreement; values of 0.21–0.40, fair agreement; values of 0.41–0.60, moderate agreement; values of 0.61–0.80, good agreement; and values of 0.81–1.00, excellent agreement.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
The sensitivities of the imaging techniques with respect to the two observers are shown in Table 4. In terms of detecting HCCs, sensitivity was higher with the gadolinium set than with the SPIO set for both observers. Also, when the data for the non-HCC lesions were pooled, sensitivity with the SPIO set was significantly higher than that with the gadolinium set for observer 1 but not for observer 2.
|
= 0.62) and SPIO ( = 0.65) sets. However, for the differentiation of benign from malignant lesions, interobserver agreement was higher with the gadolinium set ( = 0.60) than with the SPIO set ( = 0.41). |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
In terms of focal hepatic lesion detection, when pooled data from the two observers were assessed, the radiologists’ performance was found to be significantly better with unenhanced and SPIO-enhanced MR images (the SPIO set) than with unenhanced and gadolinium-enhanced MR images (the gadolinium set). However, when hypovascular or nonsolid lesions such as metastases, cysts, and hemangiomas were excluded from the analysis, the radiologists’ performance in detecting other focal lesions was not significantly different. In contrast, when HCCs were excluded from the analysis, the difference between the radiologists’ performance with the two imaging methods was further increased. In terms of the sensitivity of each imaging technique, the gadolinium set had higher sensitivity than the SPIO set for detecting HCCs, while the SPIO set had higher sensitivity than the gadolinium set for detecting non-HCC lesions. These results suggest that SPIO-enhanced MR images are better for detecting hypovascular lesions, including metastases, while gadolinium-enhanced dynamic MR images are better for detecting hypervascular lesions such as HCCs.
Recent reports comparing gadolinium-enhanced and SPIO-enhanced MR imaging also support the superiority of gadolinium-enhanced MR imaging for detecting HCCs, particularly small (<1.5–2 cm) lesions and lesions in patients with cirrhosis (12–14). For example, some small or well-differentiated HCCs may appear isointense on SPIO-enhanced MR images because they may retain some Kupffer cell activity, but they may appear hyperintense on dynamic gadolinium-enhanced arterial phase MR images because they are mainly supplied by hepatic arteries. On the other hand, in a cirrhotic liver, reduced Kupffer cell activity may decrease the contrast enhancement effect of SPIO, while the hyperintense area of hepatic fibrosis may obscure HCC lesions.
Results are controversial with regard to the detection of metastatic lesions (10,13). Blakeborough et al (10) reported that SPIO-enhanced MR imaging was significantly better at depicting malignant lesions, which were mostly metastatic, than gadolinium-enhanced MR imaging. In contrast, Matsuo et al (13) recently found that gadolinium-enhanced MR imaging was significantly better at depicting both HCCs and metastases than SPIO-enhanced MR imaging; although their study included a relatively larger number of patients with cirrhosis, the superiority of gadolinium-enhanced MR images versus SPIO-enhanced MR images was also noted in a group of patients without cirrhosis. Our study results indicate that SPIO-enhanced MR images may be better than gadolinium-enhanced dynamic MR images at depicting hepatic metastases. The reason for the different results is not clear but may be the different methods of image analysis. Matsuo et al (13) included unenhanced images at the review of gadolinium-enhanced MR images. However, they did not do so at the review of SPIO-enhanced images. In the present study, unenhanced MR images were included at the review of both gadolinium-enhanced and SPIO-enhanced MR images.
Results of recent studies have indicated that SPIO-enhanced MR imaging is highly accurate in depicting hepatic metastases (5,20,27,28), with results that parallel or even surpass those of CT during arterial portography (6,24). On the other hand, controversy exists concerning the utility of gadolinium-enhanced MR images for improving the detection of liver metastases (25,26,29). In some studies, the detection of non-HCC tumors (including metastases) did not increase when dynamic gadolinium-enhanced MR images were used (25,26). However, in a study by Semelka et al (29), unenhanced and gadolinium-enhanced dynamic MR imaging and spiral CT during arterial portography were found to be equivalent in terms of lesion depiction in patients who were evaluated preoperatively for resection of liver metastases.
In our study, radiologists’ performance in terms of lesion characterization was better with the gadolinium-enhanced MR images than with the SPIO-enhanced MR images. Gadolinium-enhanced dynamic MR imaging is widely accepted as being important for the detection and characterization of hemangiomas, HCCs, adenomas, and FNHs (26,30). Some investigators have also reported the utility of SPIO-enhanced MR imaging in characterizing focal hepatic lesions (31,32). However, some difficulty has been experienced in distinguishing between small hemangiomas and metastases, because both kinds of lesion may appear similarly hyperintense on both T1- and T2-weighted SPIO-enhanced MR images (6,7). The lack of a dynamic imaging capability of ferumoxides, which was used in this study as a SPIO agent, can also be a disadvantage in the characterization of focal hepatic lesions.
The results of this study also indicate that the sequence of contrast agent administration may affect radiologists’ performance in both detecting and characterizing focal hepatic lesions. Gadolinium-enhanced MR imaging was found to be significantly better than SPIO-enhanced MR imaging for differentiating benign and malignant lesions when it was performed first but not when it was performed immediately after SPIO-enhanced MR imaging. We assumed that this was because some small lesions, such as hemangiomas, small HCCs, and metastases, appear hyperintense on SPIO-enhanced T1-weighted MR images, and this sometimes obscures their typical enhancement characteristics at subsequent gadolinium-enhanced dynamic MR imaging. Another reason might be that the radiologists in this study were less familiar with the effect of SPIO when it is used before gadolinium-enhanced MR imaging. However, use of a double-contrast MR imaging protocol in which SPIO is administered before a gadolinium chelate may result in an increase in the conspicuity of hypervascular lesions on gadolinium-enhanced dynamic MR images by decreasing the signal intensity of the background liver parenchyma (18).
For distinguishing between HCC and dysplastic nodules, SPIO-enhanced MR imaging may have both advantages and disadvantages. When a nodule in a cirrhotic liver exhibits SPIO uptake, this may mean that it is either a well-differentiated HCC or a dysplastic nodule. In this situation, whether the nodule shows increased arterial enhancement is an important indicator of HCC if other hypervascular lesions, such as FNH or adenoma, can be excluded. In contrast, when a nodule depicted on T1-weighted MR images shows hyperintensity, the early arterial enhancement that may indicate a HCC will be difficult to observe. In such a case, the absence of uptake or a reduced uptake of SPIO may indicate the presence of an HCC rather than a dysplastic nodule. Therefore, SPIO-enhanced MR images can be useful when findings at gadolinium-enhanced MR imaging are equivocal.
In terms of detection accuracy, the second-contrast-agent image set was better than the first-contrast-agent image set in all groups. Therefore, we expect that when either gadolinium- or SPIO-enhanced MR imaging is performed first, contrast enhancement with the second agent may improve detection accuracy compared with contrast enhancement with only a single agent. However, in the present study, a difference in detection accuracy with the two image sets was significant only in the group in which gadolinium-enhanced MR images were obtained initially and SPIO-enhanced MR images were obtained 1 day later. A possible explanation for the absence of a significant difference in detection accuracy with the two image sets in the other groups is that the small sample sizes obtained by grouping patients according to contrast agent administration sequence affected the statistical significance of the results. As noted above, the detection accuracies with the SPIO and gadolinium sets differed according to the pathologic nature of the lesions. Although a relatively large number of patients were enrolled in this study, the pathologic natures of the focal lesions included in each group were heterogeneous, and this may have affected the results of this study.
A limitation of this study is that many HCCs (n = 24) were not pathologically proved. Typical clinical and laboratory findings, in combination with evidence of disease progression such as that seen on follow-up images, were used as diagnostic criteria in these patients. Although these diagnostic criteria were applied as strictly as possible to exclude false-positive lesions, it is possible that a small number of HCCs were missed.
In summary, these results show that a radiologist’s performance in detecting and characterizing focal hepatic lesions at gadolinium-enhanced and SPIO-enhanced MR imaging may depend on the type of lesions. When both contrast agents are used, the sequence of contrast agent administration may also affect the radiologist’s performance. A combination of unenhanced and SPIO-enhanced MR imaging was found to be beneficial for detecting non-HCC lesions, while a combination of unenhanced and gadolinium-enhanced MR imaging was found to be useful for detecting HCC lesions and in overall lesion characterization. Both gadolinium chelates and SPIO may improve detection rates when used as second contrast agents. Therefore, we suggest that contrast agents be used as follows in hepatic MR imaging: Gadolinium enhancement should be used in routine hepatic MR imaging for the characterization and detection of focal lesions. SPIO-enhanced MR imaging should be performed first or in addition to gadolinium-enhanced MR imaging specifically for the preoperative determination of the number of hepatic metastases or HCCs when the detection of additional lesions may affect the treatment plan. Also, when results of a first contrast-enhanced MR imaging examination are equivocal, a second MR imaging examination may be useful for both detection and characterization.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, M.J.K.; study concepts and design, M.J.K., J.H.K.; literature research, M.J.K., J.H.K.; clinical studies, M.S.P., J.J.C., J.S.L., Y.T.O.; data acquisition, J.H.K., M.J.K., J.S.L., Y.T.O.; data analysis/interpretation, all authors; statistical analysis, M.J.K., J.H.K.; manuscript preparation, all authors; manuscript definition of intellectual content and editing, M.J.K., J.H.K.; manuscript revision/review, all authors; manuscript final version approval, M.J.K., J.H.K.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Guiu, R. Loffroy, and J.-P. Cercueil Double-Contrast MRI in Patients with Cirrhosis Am. J. Roentgenol., August 1, 2008; 191(2): W78 - W78. [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]
|
|
|
|
 |
|
 C. Charnsangavej, B. Clary, Y. Fong, A. Grothey, T. M. Pawlik, and M. A. Choti Selection of Patients for Resection of Hepatic Colorectal Metastases: Expert Consensus Statement Ann. Surg. Oncol., October 1, 2006; 13(10): 1261 - 1268. [Full Text] [PDF]
|
|
|
|
|
|
M.-J. Kim, J. H. Kim, J. Y. Choi, S. H. Park, J.-J. Chung, K. W. Kim, and D. G. Mitchell Optimal TE for SPIO-enhanced gradient-recalled echo MRI for the detection of focal hepatic lesions. Am. J. Roentgenol., September 1, 2006; 187(3): W255 - W266. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. S. Lee Science to Practice: Can MR Imaging Replace Liver Biopsy for the Diagnosis of Early Fibrosis? Radiology, May 1, 2006; 239(2): 309 - 310. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Lutz, J. K. Willmann, K. Goepfert, B. Marincek, and D. Weishaupt Hepatocellular Carcinoma in Cirrhosis: Enhancement Patterns at Dynamic Gadolinium- and Superparamagnetic Iron Oxide-enhanced T1-weighted MR Imaging Radiology, November 1, 2005; 237(2): 520 - 528. [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]
|
|
|
|
|
|
S. Bipat, M. S. van Leeuwen, E. F. I. Comans, M. E. J. Pijl, P. M. M. Bossuyt, A. H. Zwinderman, and J. Stoker Colorectal Liver Metastases: CT, MR Imaging, and PET for Diagnosis--Meta-analysis Radiology, October 1, 2005; 237(1): 123 - 131. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. H. Kim, D. Choi, S. H. Kim, J. H. Lim, W. J. Lee, M. J. Kim, H. K. Lim, and S. J. Lee Ferucarbotran-Enhanced MRI Versus Triple-Phase MDCT for the Preoperative Detection of Hepatocellular Carcinoma Am. J. Roentgenol., April 1, 2005; 184(4): 1069 - 1076. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
