HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2383041825v1 239/1/131    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Onishi, H. Articles by Nakamura, H. Search for Related Content PubMed PubMed Citation Articles by Onishi, H. Articles by Nakamura, H.

Hepatic Metastases: Detection with Multi–Detector Row CT, SPIO-enhanced MR Imaging, and Both Techniques Combined¹

¹ From the Department of Radiology, Osaka University Graduate School of Medicine, Osaka, Japan (H.O., T.M., T.K., M.H., M.K., H.A., S.N., K.O., K.T., H.N.); and Department of Radiological Sciences, University of Rome-La Sapienza, Rome, Italy (R.I., R.P.). Received October 25, 2004; revision requested December 29; revision received February 24, 2005; accepted March 15; final version accepted, May 5. Address correspondence to T.M., Department of Diagnostic and Interventional Radiology, Kinki University School of Medicine, 377-2, Ohno-higashi, Osakasayama, Osaka 589-8511, Japan (e-mail: murakami{at}med.kindai.ac.jp).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
Purpose: To retrospectively compare the accuracy in detection of hepatic metastases among contrast material–enhanced multi–detector row computed tomography (CT) alone, superparamagnetic iron oxide (SPIO)-enhanced magnetic resonance (MR) imaging alone, and a combination of contrast-enhanced CT and SPIO-enhanced MR imaging.

Materials and Methods: The ethics committee did not require its approval or informed consent for this retrospective study, which was compliant with Declaration of Helsinki principles. Data in 38 patients (22 men, 16 women; mean age, 64.5 years; range, 35–78 years) suspected of having hepatic metastases who underwent both contrast-enhanced CT and SPIO-enhanced MR imaging were retrospectively analyzed. Twenty-one of the 38 patients had 61 metastases. Seventeen of the 61 metastases were confirmed histologically; the remaining 44 metastases were defined with imaging follow-up. At MR imaging, SPIO-enhanced heavily T1-weighted images, T2*-weighted gradient echo images, and T2-weighted fast spin-echo images were evaluated. Contrast-enhanced multi–detector row CT images obtained in the portal phase were evaluated. Four blinded observers independently reviewed CT images, MR images, and the combination of CT and MR images. Diagnostic accuracy was evaluated by using the alternative free-response receiver operating characteristic (AFROC) method. Sensitivities and positive predictive values were also analyzed with the Fisher protected least significant difference test and generalized estimating equations.

Results: The mean area under the AFROC curve for the combined approach (0.70) was significantly higher than that for SPIO-enhanced MR imaging alone (0.58, P < .05, Fisher protected least significant difference test), and there was no significant difference between each of them and that for contrast-enhanced CT alone (0.66). For all lesions, the mean sensitivity of combined imaging (0.59) was significantly higher than that of CT (0.48) or MR imaging (0.43) alone (P < .05, Fisher protected least significant difference test and generalized estimating equations). For all lesions, the mean positive predictive values were 0.82, 0.89, and 0.81, for combined MR and CT, CT alone, and MR alone, respectively.

Conclusion: The addition of SPIO-enhanced MR imaging to contrast-enhanced multi–detector row CT (ie, combined analysis of SPIO-enhanced MR images and contrast-enhanced CT images) can improve sensitivity in the detection of hepatic metastases, although this improvement in sensitivity was not significant at AFROC analysis.

© RSNA, 2006

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
Superparamagnetic iron oxide (SPIO) is a liver-specific particulate magnetic resonance (MR) imaging contrast agent that is taken up by the reticuloendothelial system of the liver and improves the focal hepatic lesion-to-liver contrast-to-noise ratio and hepatic tumor detection. Seneterre et al (1) have demonstrated that, for the detection of hepatic metastases, SPIO-enhanced MR imaging is more accurate than or at least as accurate as computed tomography (CT) during arterial portography (which was considered to be the most sensitive modality). Some researchers (2–7) have compared the capability of SPIO-enhanced MR imaging with that of contrast material–enhanced helical CT for detecting malignant hepatic lesions and have found SPIO-enhanced MR imaging to be superior. In those studies, however, contrast-enhanced helical CT was performed with a single–detector row scanner. To our knowledge, no study has yet been conducted in which SPIO-enhanced MR imaging was compared with contrast-enhanced helical CT performed with a multi–detector row scanner for detection of hepatic metastases.

Multi–detector row CT scanners enable better spatial resolution in the direction of the body axis and greater anatomic coverage during a single breath hold (8). Some authors have reported that use of a thinner section thickness (ie, 2.5 or 5 mm) at CT improves the detection of small hepatic lesions (9,10). With the rapid introduction of multi–detector row CT scanners to the clinical environment, the use of a thinner section thickness at contrast-enhanced CT for the detection of hepatic metastases has become a routine practice. The expected improvement in detection of hepatic metastases with contrast-enhanced multi–detector row CT could potentially reduce the superiority of SPIO-enhanced MR imaging.

SPIO-enhanced MR imaging for the evaluation of hepatic metastatic tumors is usually performed as an adjunct to contrast-enhanced CT. However, to our knowledge, although many studies have been performed to compare SPIO-enhanced MR imaging alone with CT alone, no study has as yet involved comparison of the combination of CT and SPIO-enhanced MR imaging with CT alone or with SPIO-enhanced MR imaging alone.

Thus, the purpose of our study was to retrospectively compare the accuracy in detection of hepatic metastases of contrast-enhanced multi–detector row CT alone with that of SPIO-enhanced MR imaging alone and that of a combination of contrast-enhanced CT and SPIO-enhanced MR imaging.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
Patients and Reference Standards
Seventy-three patients underwent SPIO-enhanced MR imaging for the evaluation of liver metastasis at the Department of Radiology, Osaka University Graduate School of Medicine, Osaka, Japan, from April 1999 to June 2003. For 56 of these patients, both contrast-enhanced CT with a multi–detector row scanner and SPIO-enhanced MR imaging were performed within 30 days. This study followed Declaration of Helsinki principles (11). The ethics committee at our institution does not require its approval or informed consent for retrospective studies such as this one. Informed consent for the imaging examinations was obtained from all patients. Eighteen of the 56 patients were excluded from our study because they were undergoing or had undergone chemotherapy (n = 15) or because there were too many nodules (ie, more than 10 in one patient) to be analyzed (n = 3). The remaining 38 patients (22 men and 16 women) were included in this retrospective study. The mean age was 64.5 years (range, 35–78 years).

Seventeen of the 38 patients had no hepatic metastasis, as confirmed with follow-up imaging studies performed over the course of at least 1 year (range, 12–54 months). The remaining 21 patients had hepatic metastases. Fourteen patients had metastases from colorectal cancer; five patients, metastases from pancreatic cancer; one patient, metastases from cecal cancer; and one patient, metastases from bile duct cancer. Eight of the 21 patients underwent definitive surgery with intraoperative ultrasonography (US) within 40 days after MR imaging and CT. Seventeen metastases were confirmed histologically in these eight patients. The 44 visible hepatic tumors in 13 patients for which histopathologic confirmation of disease was not available were considered to be hepatic metastases on the basis of tumor growth observed at follow-up examinations (performed 2–24 months after MR imaging and CT) or on the basis of the presence of increased serum levels of tumor marker when extrahepatic disease was not detected at CT or other examinations. There were no patients with hepatic cirrhosis in this study.

The presence or absence of hepatic metastases was decided in consensus by two radiologists (T.M. and H.O., with 18 and 7 years of experience, respectively, in gastrointestinal and hepatobiliary imaging) who were not among the four blinded readers who performed image analysis (see below). The two radiologists determined the presence or absence of metastases on the basis of findings obtained with contrast-enhanced CT, CT during arterial portography, CT hepatic arteriography, US, and gadolinium-enhanced MR imaging; findings obtained at follow-up US, CT, and MR imaging; and findings obtained at definitive surgery that involved intraoperative US, biopsy, and serologic examination. As a result, confirmation of 61 hepatic metastases (mean diameter, 14.6 mm; range, 3–130 mm) was obtained for 21 of the 38 patients (Table 1).

MR Imaging
MR imaging was performed with a superconducting magnet operating at 1.5 T (Magnetom Vision, Siemens Medical Systems, Erlangen, Germany, for 11 patients; Signa Horizon, GE Medical Systems, Milwaukee, Wis, for 27 patients) and a body phased-array coil. In our analysis, SPIO-enhanced heavily T1-weighted gradient-echo images, T2*-weighted gradient-echo images, and T2-weighted fast spin-echo images were assessed.

For 31 of the 38 patients, ferumoxides (Feridex; Eiken and Tanabe Pharmaceutical, Osaka, Japan) was administered. A dose of 0.56 mg (10 µmol) of iron per kilogram of body weight was diluted with 100 mL of 5% dextrose solution and infused intravenously at a rate of 3 mL/min. Approximately 60 minutes after the initiation of intravenous drip infusion of the SPIO for 30 minutes, contrast-enhanced images were acquired. For the remaining seven patients, ferucarbotran (Resovist; Nihon Schering, Osaka, Japan) was administered. A dose of 0.56 mg (10 µmol) of iron per kilogram (injection volume, 0.9–1.4 mL) was preloaded into a connecting intravenous tube and manually injected through a filter with a 5-µm pore size by flushing the connecting catheter with 20 mL of saline solution for 10 seconds or less (injection rate, approximately 2 mL/sec). Approximately 15 minutes after the initiation of the contrast medium injection, contrast-enhanced images were acquired.

All images were acquired in the transverse plane with a section thickness of 6–8 mm, an intersection gap of 1–2 mm, and a field of view of 28–32 cm. Presaturation pulses were applied above and below the imaging volume for artifact reduction. After the injection of SPIO, heavily T1-weighted gradient-echo images (repetition time msec/echo time msec, 150/1.3–2.2; flip angle, 90°; number of signals acquired, one; matrix, 256 x 128) and T2*-weighted gradient-echo images with long echo times (150/10–12; flip angle, 60°, number of signals acquired, one; matrix, 256 x 128) were obtained. With one imaging unit (Signa Horizon; GE Medical Systems), heavily T1-weighted and T2*-weighted images were obtained by using a fast spoiled gradient-recalled acquisition in the steady state (SPGR) sequence, and with the other (Magnetom Vision; Siemens Medical Systems), they were obtained by using a fast low-angle shot sequence. For 11 of 38 patients, conventional T2-weighted spin-echo images (1800–2000/70–90; number of signals acquired, two; matrix, 256 x 160) were obtained. For the remaining 27 patients, respiratory-triggered T2-weighted fast spin-echo images were obtained (3750–8000/65–80 [effective]; echo train length, eight; number of signals acquired, three; matrix, 512 x 160). In our analysis, SPIO-enhanced heavily T1-weighted gradient-echo images, T2*-weighted gradient-echo images, and T2-weighted fast spin-echo images were assessed.

CT Examination
For 22 of the 38 patients, CT was performed with a four–detector row scanner (LightSpeed QX/i; GE Medical Systems) with a tube voltage of 120 kV, a tube current of 300 mA, a rotation period of 0.5 second, a detector collimation of 4 x 2.5 mm (four detector rows with 2.5-mm section thickness), and a table increment of 15 mm per rotation. For the remaining 16 patients, CT was performed with an eight–detector row scanner (LightSpeed Ultra; GE Medical Systems) with a tube voltage of 120 kV, a tube current of 300 mA, a rotation period of 0.5 second, a detector collimation of 8 x 2.5 mm, and a table increment of 27 mm per rotation. Images with an effective section thickness of 5 mm were reconstructed every 5 mm to provide contiguous sections.

In our analysis, contrast-enhanced images obtained in the portal venous phase were assessed. Nonionic contrast medium, either iopamidol (Iopamiron 300; Nihon Schering, Osaka, Japan) or iohexol (Omnipaque 300; Daiichi Pharmaceutical, Tokyo, Japan), with 300 mg of iodine per milliliter, was administered intravenously as a bolus of 2.0 mL/kg at a rate of 2–5 mL/sec with a power injector (Auto Enhance A-50; Nemoto Kyorindo, Tokyo, Japan). Image acquisition in the portal venous phase began 60–70 seconds after initiation of the contrast medium injection in all patients.

Image Assessment
Images from the combined MR imaging sequences (including heavily T1-weighted, T2*-weighted, and T2-weighted sequences performed after SPIO enhancement) and contrast-enhanced CT images were interpreted independently by four readers (T.K., M.H., M.K., and H.A., with 15, 11, 8, and 7 years of experience, respectively, in gastrointestinal and hepatobiliary imaging). These four readers were different from the two radiologists who had determined the presence or absence of tumors on the basis of radiologic and pathologic findings. The readers had worked mainly as gastrointestinal and hepatobiliary radiologists for 7–15 years and had interpreted CT and MR images of the liver as part of their daily clinical and research practice. They knew that the patients were at risk for hepatic metastases but were blinded to all other information about the patients' history, including the site of the primary tumor. All images were interpreted in soft copy at a commercially available Microsoft Windows–based workstation (Virtual Place 21; Aze, Tokyo, Japan).

At first, each reader interpreted MR images and recorded the presence, location, and size of each hepatic lesion on a form on which there were schematic drawings of transverse sections of the whole liver. Then, they assigned a confidence rating to each lesion by using a four-point scale, on which a score of 1 indicated that a lesion was probably not present; a score of 2, that lesion presence was equivocal; a score of 3, that a lesion was probably present; and a score of 4, that a lesion was definitely present.

In determining whether a tumor was malignant or benign at SPIO-enhanced MR imaging, we used the criteria of Kumano et al (16), according to which a lesion that appeared hyperintense on SPIO-enhanced T1-weighted images was considered a hemangioma, while a lesion that was hypointense on T1-weighted images, of intermediate intensity on T2*-weighted images, and strongly hyperintense on T2-weighted images was considered a cyst, and a lesion that was hypointense on T1-weighted images and hyperintense on T2*-weighted and T2-weighted images was considered a malignant tumor. More than 2 weeks (2–3 weeks) after the first session, each reader viewed the CT images and recorded the findings in a similar fashion. After reading the CT images for a certain patient, the readers then looked at the MR images for the same patient and additionally recorded the decisions they made on the basis of both MR and CT images. At the time of scoring, the readers were aware that sensitivity calculations were performed with inclusion of only those lesions awarded a confidence rating of 3 or 4. The CT images were presented to each reader in a different order than were the MR images in an attempt to avoid learning bias.

Statistical Analysis
Alternative free-response receiver operating characteristic (AFROC) analysis was performed on a tumor-by-tumor basis. Although the conventional receiver operating characteristic method allows for only one response per image, the AFROC method allows for multiple responses per image (17). An AFROC curve was fitted to each reader's confidence rating by using a maximum-likelihood estimation program (ROCKIT 0.9B; C. E. Metz, University of Chicago, Chicago, Ill, 1998). The diagnostic performance of both methods and readers was estimated by calculating the area under the AFROC curve (A_z). The sensitivities and positive predictive values for each reader and each method were also determined by using only those lesions allocated a confidence rating of 3 or 4. In addition to analyses that included all lesions, we also performed analyses for a subgroup that included lesions 1 cm in diameter or smaller and a subgroup that included lesions larger than 1 cm in diameter.

Logistic regression analysis with generalized estimating equations (18) was also used to account for the clustering effects of multiple tumor nodules in the same patient by using a statistical software package (SPSS 11.0 for Windows; SPSS Japan, Tokyo, Japan) and a macro program (GEE95, version 1.01; courtesy of Methodology and Statistics Group, Oregon Research Institute, Eugene, Ore). To consider the association between tumor size and detection rate, tumor size was included as one of the factors in the logistic regression analysis.

For A_z, sensitivity, and positive predictive values, the statistical significance of any difference among CT alone, MR imaging alone, and the combination of CT and MR imaging was assessed by using the Fisher protected least significant difference test. A P value of less than .05 was considered to indicate a statistically significant difference.

So that we could assess interreader variability in image interpretation, we used the unweighted statistic with binary data to measure the extent of agreement among the four readers. The binary values of 0 (not a lesion) and 1 (a lesion) were assigned to lesions with a confidence rating of 2 or less and to lesions with a confidence rating of 3 or greater, respectively. The extent of disagreement was not factored into the calculation. Values of up to 0.40 were considered to indicate positive but poor agreement; values of 0.41–0.75, good agreement; and values greater than 0.75, excellent agreement (19).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
AFROC Analysis
When all lesions were considered (Table 2), the mean A_z value for the combination of CT and SPIO-enhanced MR imaging was significantly greater than that for SPIO-enhanced MR imaging alone (P = .03) and greater (but not significantly greater) than that for CT alone (P = .45). For lesions 1 cm in diameter or smaller, the mean A_z value for the combination of CT and SPIO-enhanced MR imaging was greater than that for CT alone and that for SPIO-enhanced MR imaging alone, although there were no statistically significant differences (P = .56 and P = .13, respectively). The mean A_z value for CT alone was greater than that for SPIO-enhanced MR imaging alone, although there were no statistically significant differences (P = .56 for all lesions and P = .10 for lesions 1 cm in diameter or smaller). In several instances, the A_z value for lesions larger than 1 cm was degenerated because it was based on a small data set or a data set with many tied values.

False-Positive Findings and Positive Predictive Value
There was no significant difference between mean positive predictive values for CT alone, SPIO-enhanced MR imaging alone, and combined CT and SPIO-enhanced MR imaging (Table 4). False-positive lesions were recorded in 1 or 2 (mean, 1.5) of the 17 patients without liver metastases at CT alone, between 3 and 7 patients (mean, 4.0) at SPIO-enhanced MR imaging alone, and between 1 and 4 patients (mean, 2.5) with combined CT and SPIO-enhanced MR imaging.

View larger version (133K): [in this window] [in a new window] [Download PPT slide]   Figure 1a: Surgically proved liver metastasis from rectal cancer in 71-year-old man. (a) Transverse contrast-enhanced CT scan shows a small and slightly hypoattenuating nodule (arrow). (b) On transverse SPIO-enhanced respiratory-triggered T2-weighted fast spin-echo MR image (4615/64 [effective]; echo train length, eight), the lesion is seen as a high-signal-intensity nodule (arrow). Although on (c) a transverse SPIO-enhanced fast SPGR MR image (150/10; flip angle, 60°), the lesion is seen as a high-signal-intensity nodule (arrow) and on (d) a transverse SPIO-enhanced T1-weighted fast SPGR MR image (150/1.3; flip angle, 90°), the lesion is seen as a low-signal-intensity nodule (arrow), it is difficult to distinguish the true lesion from vessels owing to the small size of the lesion. With CT images alone, two readers assigned this lesion a confidence score of 0, another reader assigned a score of 1, and the remaining reader assigned a score of 2. With MR images alone, three readers assigned this lesion a confidence score of 0 and the remaining reader assigned a score of 1. With both CT and MR images, one reader assigned this lesion a confidence score of 0, another reader assigned a score of 1, and the remaining two readers assigned a score of 3. The combination of contrast-enhanced CT and SPIO-enhanced MR imaging increased the confidence level for this lesion.

View larger version (114K): [in this window] [in a new window] [Download PPT slide]   Figure 1b: Surgically proved liver metastasis from rectal cancer in 71-year-old man. (a) Transverse contrast-enhanced CT scan shows a small and slightly hypoattenuating nodule (arrow). (b) On transverse SPIO-enhanced respiratory-triggered T2-weighted fast spin-echo MR image (4615/64 [effective]; echo train length, eight), the lesion is seen as a high-signal-intensity nodule (arrow). Although on (c) a transverse SPIO-enhanced fast SPGR MR image (150/10; flip angle, 60°), the lesion is seen as a high-signal-intensity nodule (arrow) and on (d) a transverse SPIO-enhanced T1-weighted fast SPGR MR image (150/1.3; flip angle, 90°), the lesion is seen as a low-signal-intensity nodule (arrow), it is difficult to distinguish the true lesion from vessels owing to the small size of the lesion. With CT images alone, two readers assigned this lesion a confidence score of 0, another reader assigned a score of 1, and the remaining reader assigned a score of 2. With MR images alone, three readers assigned this lesion a confidence score of 0 and the remaining reader assigned a score of 1. With both CT and MR images, one reader assigned this lesion a confidence score of 0, another reader assigned a score of 1, and the remaining two readers assigned a score of 3. The combination of contrast-enhanced CT and SPIO-enhanced MR imaging increased the confidence level for this lesion.

View larger version (126K): [in this window] [in a new window] [Download PPT slide]   Figure 1c: Surgically proved liver metastasis from rectal cancer in 71-year-old man. (a) Transverse contrast-enhanced CT scan shows a small and slightly hypoattenuating nodule (arrow). (b) On transverse SPIO-enhanced respiratory-triggered T2-weighted fast spin-echo MR image (4615/64 [effective]; echo train length, eight), the lesion is seen as a high-signal-intensity nodule (arrow). Although on (c) a transverse SPIO-enhanced fast SPGR MR image (150/10; flip angle, 60°), the lesion is seen as a high-signal-intensity nodule (arrow) and on (d) a transverse SPIO-enhanced T1-weighted fast SPGR MR image (150/1.3; flip angle, 90°), the lesion is seen as a low-signal-intensity nodule (arrow), it is difficult to distinguish the true lesion from vessels owing to the small size of the lesion. With CT images alone, two readers assigned this lesion a confidence score of 0, another reader assigned a score of 1, and the remaining reader assigned a score of 2. With MR images alone, three readers assigned this lesion a confidence score of 0 and the remaining reader assigned a score of 1. With both CT and MR images, one reader assigned this lesion a confidence score of 0, another reader assigned a score of 1, and the remaining two readers assigned a score of 3. The combination of contrast-enhanced CT and SPIO-enhanced MR imaging increased the confidence level for this lesion.

View larger version (140K): [in this window] [in a new window] [Download PPT slide]   Figure 1d: Surgically proved liver metastasis from rectal cancer in 71-year-old man. (a) Transverse contrast-enhanced CT scan shows a small and slightly hypoattenuating nodule (arrow). (b) On transverse SPIO-enhanced respiratory-triggered T2-weighted fast spin-echo MR image (4615/64 [effective]; echo train length, eight), the lesion is seen as a high-signal-intensity nodule (arrow). Although on (c) a transverse SPIO-enhanced fast SPGR MR image (150/10; flip angle, 60°), the lesion is seen as a high-signal-intensity nodule (arrow) and on (d) a transverse SPIO-enhanced T1-weighted fast SPGR MR image (150/1.3; flip angle, 90°), the lesion is seen as a low-signal-intensity nodule (arrow), it is difficult to distinguish the true lesion from vessels owing to the small size of the lesion. With CT images alone, two readers assigned this lesion a confidence score of 0, another reader assigned a score of 1, and the remaining reader assigned a score of 2. With MR images alone, three readers assigned this lesion a confidence score of 0 and the remaining reader assigned a score of 1. With both CT and MR images, one reader assigned this lesion a confidence score of 0, another reader assigned a score of 1, and the remaining two readers assigned a score of 3. The combination of contrast-enhanced CT and SPIO-enhanced MR imaging increased the confidence level for this lesion.

View larger version (87K): [in this window] [in a new window] [Download PPT slide]   Figure 2a: Surgically proved liver metastasis from cecal cancer in 61-year-old woman. (a) Transverse contrast-enhanced CT scan shows the lesion as a faint hypoattenuating area (arrow). (b) Transverse SPIO-enhanced fast low-angle shot MR image (150/10; flip angle, 60°) shows the lesion as a hyperintense nodule (arrow). With CT images alone, all four readers assigned a confidence score of 0 to this lesion. With MR images alone, one reader assigned a confidence score of 2, two readers assigned a score of 3, and the remaining reader assigned a score of 4. With both CT and MR images, one reader assigned a score of 0, another reader assigned a score of 2, and the remaining two readers assigned a score of 3.

View larger version (116K): [in this window] [in a new window] [Download PPT slide]   Figure 2b: Surgically proved liver metastasis from cecal cancer in 61-year-old woman. (a) Transverse contrast-enhanced CT scan shows the lesion as a faint hypoattenuating area (arrow). (b) Transverse SPIO-enhanced fast low-angle shot MR image (150/10; flip angle, 60°) shows the lesion as a hyperintense nodule (arrow). With CT images alone, all four readers assigned a confidence score of 0 to this lesion. With MR images alone, one reader assigned a confidence score of 2, two readers assigned a score of 3, and the remaining reader assigned a score of 4. With both CT and MR images, one reader assigned a score of 0, another reader assigned a score of 2, and the remaining two readers assigned a score of 3.

View larger version (125K): [in this window] [in a new window] [Download PPT slide]   Figure 3a: Liver metastasis from rectal cancer in 59-year-old man. At follow-up CT 1 month later, the lesion showed growth and the tumor marker increased during the observation period. Hence, the lesion was considered to be a true metastatic tumor. (a) Transverse contrast-enhanced CT scan shows the lesion as a small hypoattenuating area (arrow). (b) Transverse SPIO-enhanced fast low-angle shot MR image (150/10; flip angle, 60°) shows the lesion as a triangular hyperintense nodule (arrow). With CT images alone, three readers assigned a confidence score of 2 to this lesion and the remaining reader assigned a score of 3. With MR images alone, three readers assigned a score of 0 and the remaining reader assigned a score of 1. With both CT and MR images, one reader assigned a score of 1, two readers assigned a score of 2, and the remaining reader assigned a score of 3. On MR images alone, it was difficult to distinguish the true lesion from vessels because the lesion appeared as a triangular rather than a round shape and mimicked the obliquely sectioned vessel.

View larger version (111K): [in this window] [in a new window] [Download PPT slide]   Figure 3b: Liver metastasis from rectal cancer in 59-year-old man. At follow-up CT 1 month later, the lesion showed growth and the tumor marker increased during the observation period. Hence, the lesion was considered to be a true metastatic tumor. (a) Transverse contrast-enhanced CT scan shows the lesion as a small hypoattenuating area (arrow). (b) Transverse SPIO-enhanced fast low-angle shot MR image (150/10; flip angle, 60°) shows the lesion as a triangular hyperintense nodule (arrow). With CT images alone, three readers assigned a confidence score of 2 to this lesion and the remaining reader assigned a score of 3. With MR images alone, three readers assigned a score of 0 and the remaining reader assigned a score of 1. With both CT and MR images, one reader assigned a score of 1, two readers assigned a score of 2, and the remaining reader assigned a score of 3. On MR images alone, it was difficult to distinguish the true lesion from vessels because the lesion appeared as a triangular rather than a round shape and mimicked the obliquely sectioned vessel.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 
In several clinical situations, SPIO-enhanced MR imaging is performed as an adjunct to contrast-enhanced CT for the detection of metastases because CT is superior in the detection of extrahepatic disease and the evaluation of the primary lesion in some malignancies. Therefore, we also evaluated the diagnostic capabilities of the combination of contrast-enhanced CT and SPIO-enhanced MR imaging for the detection of hepatic metastases. Our results show that the addition of SPIO-enhanced MR imaging to contrast-enhanced CT improved the sensitivity for the detection of hepatic metastases. The A_z value was also improved, although not statistically significantly so. The positive predictive value for evaluation of all lesions was decreased by adding SPIO-enhanced MR imaging to contrast-enhanced CT, although this decrease was not statistically significant. However, with regard to tumors 1 cm in diameter or smaller, an improvement in the positive predictive value was present, although again the improvement was not statistically significant. This discrepancy between the positive predictive value for all lesions and that for lesions 1 cm in diameter or smaller could be related to the fact that small cysts were difficult to distinguish from small metastatic tumors, with a consequent increase in the number of false-positive results at contrast-enhanced CT. By contrast, small cysts could be differentiated from small metastatic tumors by comparing T2-weighted images with long-echo-time gradient-echo images at SPIO-enhanced MR imaging.

Several authors (20–22) have described the difficulty of distinguishing small lesions from peripheral vessels and the interference of susceptibility effects with small lesions, which are more conspicuous after enhancement with SPIO contrast agents. Peripheral vessels are well recognized at contrast-enhanced CT. Thus, the combined evaluation of SPIO-enhanced MR and contrast-enhanced CT images may decrease these potential disadvantages of SPIO-enhanced MR imaging.

The present study had some limitations. First, as is the case with some previous studies (1,4), a potential limitation to our study could be the lack of histologic proof for all nodules believed to be hepatic metastases. However, several kinds of confirmatory findings, including those at CT arterial portography and CT hepatic arteriography, intraoperative US, follow-up CT, and follow-up MR imaging, were available for all metastases. The presence of other proved metastatic tumors, the small size of many nodules, and/or the patient's clinical history frequently made biopsy of individual lesions unnecessary or impractical. Consequently, the exact number of metastatic tumors in each liver could not be determined, and, therefore, specificity and accuracy could not be calculated. In theory, it is possible that some false-positive or false-negative lesions may have been included inadvertently.

Second, in our study, CT was evaluated only in terms of contrast-enhanced images obtained in the portal venous phase. The use of a dual-phase technique with imaging during both the arterial and the portal venous phase may improve the depiction of liver tumors, especially hypervascular liver tumors (23,24). However, most hepatic metastases are hypovascular, and use of the dual-phase technique increases radiation dose. Therefore, only contrast-enhanced images in the portal venous phase are used for the detection of hepatic metastases in our daily practice.

Third, sensitivity and A_z values at AFROC analysis in our study were lower than those in a previous study by Ward et al (4). This difference could be related to the fact that the number of lesions 1 cm in diameter or smaller in our study was higher than that in the study by Ward et al.

Fourth, two different SPIO contrast agents (ie, ferumoxides and ferucarbotran) were used in our study. Thus, we cannot rule out the possibility that differences in enhancement between these two SPIO contrast agents might have influenced the results in our study. However, on the basis of our anecdotal experience, we believe that the lesion-to-liver contrast provided by these two SPIO contrast agents is virtually identical. Fifth, the number of patients in our study was small (n = 38), and this may have been a potential factor in the fact that we observed no significant difference in sensitivity between CT alone and SPIO-enhanced MR imaging alone. However, we believe that if there is a significant difference between CT alone and SPIO-enhanced MR imaging alone, it is small.

In summary, there was no significant difference in the sensitivity of detection of hepatic metastases between contrast-enhanced CT images obtained with a multi–detector row scanner alone and SPIO-enhanced MR images alone. However, the combined analysis of SPIO-enhanced MR images and contrast-enhanced multi–detector row CT images improves the detection of hepatic metastases in comparison with the detection achieved with independent analysis of these images.

	FOOTNOTES

 

Abbreviations: AFROC = alternative free-response receiver operating characteristic • A_z = area under AFROC curve • SPGR = spoiled gradient-recalled acquisition in the steady state • SPIO = superparamagnetic iron oxide

Author contributions: Guarantors of integrity of entire study, T.M., H.N.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, H.O., T.M., R.I.; clinical studies, H.O., T.M., T.K., M.H., M.K., H.A., S.N., K.O.; statistical analysis, H.O., M.H.; and manuscript editing, H.O., T.M., T.K., M.H., R.I., K.T., R.P., H.N.

Authors stated no financial relationship to disclose.

	References

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION References

 

1. Seneterre E, Taourel P, Bouvier Y, et al. Detection of hepatic metastases: ferumoxides-enhanced MR imaging versus unenhanced MR imaging and CT during arterial portography. Radiology 1996;200:785–792.[Abstract/Free Full Text]
2. Fretz CJ, Stark DD, Metz CE, et al. Detection of hepatic metastases: comparison of contrast-enhanced CT, unenhanced MR imaging, and iron oxide-enhanced MR imaging. AJR Am J Roentgenol 1990;155:763–770.[Abstract/Free Full Text]
3. Hagspiel KD, Neidl KF, Eichenberger AC, Weder W, Marincek B. Detection of liver metastases: comparison of superparamagnetic iron oxide-enhanced and unenhanced MR imaging at 1.5 T with dynamic CT, intraoperative US, and percutaneous US. Radiology 1995;196:471–478.[Abstract/Free Full Text]
4. Ward J, Naik KS, Guthrie JA, Wilson D, Robinson P. Hepatic lesion detection: comparison of MR imaging after the administration of superparamagnetic iron oxide with dual-phase CT by using alternative-free response receiver operating characteristic analysis. Radiology 1999;210:459–466.[Abstract/Free Full Text]
5. Muller RD, Vogel K, Neumann K, et al. SPIO-MR imaging versus double-phase spiral CT in detecting malignant lesions of the liver. Acta Radiol 1999;40:628–635.[Medline]
6. Bluemke DA, Paulson EK, Choti MA, et al. Detection of hepatic lesions in candidates for surgery: comparison of ferumoxides-enhanced MR imaging and dual-phase helical CT. AJR Am J Roentgenol 2000;175:1653–1658.[Abstract/Free Full Text]
7. Reimer P, Jahnke N, Fiebich M, et al. Hepatic lesion detection and characterization: value of nonenhanced MR imaging, superparamagnetic iron oxide-enhanced MR imaging, and spiral CT-ROC analysis. Radiology 2000;217:152158.[Abstract/Free Full Text]
8. Hu H. Multi-slice helical CT: scan and reconstruction. Med Phys 1999;26:5–18.[CrossRef][Medline]
9. Weg N, Scheer MR, Gabor MP. Liver lesions: improved detection with dual-detector-array CT and routine 2.5-mm thin collimation. Radiology 1998;209:417–426.[Abstract/Free Full Text]
10. Kawata S, Murakami T, Kim T, et al. Multidetector CT: diagnostic impact of slice thickness on detection of hypervascular hepatocellular carcinoma. AJR Am J Roentgenol 2002;179:61–66.[Abstract/Free Full Text]
11. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA 2000;284:3043–3045.[Free Full Text]
12. Mathieu D, Vilgrain V, Mahfouz A, et al. Benign liver tumors. Magn Reson Imaging Clin N Am 1997;5:255–288.[Medline]
13. Mortele KJ, Ros PR. Cystic focal liver lesions in the adult: differential CT and MR imaging features. RadioGraphics 2001;21:895–910.[Abstract/Free Full Text]
14. Horton KM, Bluemke DA, Hruban RH, Soyer P, Fishman EK. CT and MR imaging of benign hepatic and biliary tumors. RadioGraphics 1999;19:431–451.[Abstract/Free Full Text]
15. Kim T, Federle MP, Baron RL, Peterson MS, Kawamori Y. Discrimination of small hepatic hemangiomas from hypervascular malignant tumors smaller than 3 cm with three-phase helical CT. Radiology 2001;219:699–706.[Abstract/Free Full Text]
16. Kumano S, Murakami T, Kim T, et al. Using superparamagnetic iron oxide–enhanced MRI to differentiate metastatic hepatic tumors and nonsolid benign lesions. AJR Am J Roentgenol 2003;181:1335–1339.[Abstract/Free Full Text]
17. Chakraborty DP, Winter LH. Free-response methodology: alternate analysis and a new observer-performance experiment. Radiology 1990;174:873–881.[Abstract/Free Full Text]
18. Liang K, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika 1986;73:13–22.[Abstract/Free Full Text]
19. Fleiss JL. The measurement of interrater agreement. In: Statistical methods for rates and proportions. 2nd ed. New York, NY: Wiley, 1981; 212–236.
20. Stark DD, Weissleder R, Elizondo G, et al. Superparamagnetic iron oxide: clinical application as a contrast agent for MR imaging of the liver. Radiology 1988;168:297–301.[Abstract/Free Full Text]
21. Winter TC 3rd, Freeny PC, Nghiem HV, et al. MR imaging with i.v. superparamagnetic iron oxide: efficacy in the detection of focal hepatic lesions. AJR Am J Roentgenol 1993;161:1191–1198.
22. Ros PR, Freeny PC, Harms SE, et al. Hepatic MR imaging with ferumoxides: a multicenter clinical trial of the safety and efficacy in the detection of focal hepatic lesions. Radiology 1995;196:481–488.[Abstract/Free Full Text]
23. Hollett MD, Jeffrey RB Jr, Nino-Murcia M, Jorgensen MJ, Harris DP. Dual-phase CT of the liver: value of arterial phase scans in the detection of small (< or = 1.5 cm) malignant hepatic neoplasms. AJR Am J Roentgenol 1995;164:879–884.[Abstract/Free Full Text]
24. Bonaldi VM, Bret PM, Reinhold C, Atri M. Helical CT of the liver: value of an early hepatic arterial phase. Radiology 1995;197:357–363.[Abstract/Free Full Text]

This article has been cited by other articles:

				Y. S. Maetani, H. Isoda, A. Nomura, S. Arizono, Y. Hirokawa, T. Shibata, and T. Kaori CT Laparoscopy for Detecting Small Superficial Metastatic Lesions of the Liver Surface: Initial Experience Am. J. Roentgenol., May 1, 2008; 190(5): 1314 - 1317. [Abstract] [Full Text] [PDF]

				H. Furukawa, K. Uesaka, and N. Boku Treatment Decision Making in Pancreatic Adenocarcinoma: Multidisciplinary Team Discussion With Multidetector-Row Computed Tomography Arch Surg, March 1, 2008; 143(3): 275 - 280. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2383041825v1 239/1/131    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Onishi, H. Articles by Nakamura, H. Search for Related Content PubMed PubMed Citation Articles by Onishi, H. Articles by Nakamura, H.