Comparison Studies of CT and MRI in Patients With Hepatic Metastases

June 1, 2000

The increasing use of systemic and directed liver therapy for patients with hepatic metastases has created a demand for improved accuracy of noninvasive imaging techniques. Computed tomography (CT) and magnetic

ABSTRACT: The increasing use of systemic and directed liver therapy for patients with hepatic metastases has created a demand for improved accuracy of noninvasive imaging techniques. Computed tomography (CT) and magnetic resonance imaging (MRI) are the most common studies used for imaging the liver in oncology patients. Both modalities have undergone substantial technical improvement in the past decade, and it is often unclear which technique is better suited to specific clinical circumstances. This article will review the major recent developments in each modality, including dual-phase spiral CT, CT angiography, ultrafast MRI, and MRI enhanced with liver-specific contrast agents. Studies that directly compared state-of-the-art CT and MRI will be emphasized. This review will give the reader a better understanding of the capabilities and limitations of these techniques and will clarify which is best used for specific clinical situations. [ONCOLOGY 14(Suppl 3): 21-28, 2000]

Computed tomography (CT) and magnetic resonance imaging (MRI) are commonly used for hepatic imaging in oncology patients. Both CT and MRI have advanced rapidly over the past 10 years as a result of improvements in computer hardware, software, and other electronic components. These changes have increased the accuracy of each technique in detecting and characterizing focal hepatic neoplasms. However, the improvements in both modalities have led to some confusion over their relative advantages.

Many studies have directly compared the diagnostic efficacy of MRI and CT in patients with focal hepatic malignancies.[1-18] These reports determined that, although the spatial resolution of CT remains superior, MRI has better contrast resolution and has repeatedly been proven to detect and characterize focal liver lesions with greater accuracy than CT.

Despite this obvious advantage, CT remains the primary tool for routine diagnostic evaluation of patients with known or suspected liver malignancies in most centers. The relative underutilization of MRI may be due, at least in part, to a lack of understanding of the advantages and disadvantages of each technique.

This article will review the current literature on state-of-the-art techniques for imaging the liver using both CT and MRI. Direct comparison studies of CT and MRI will be reviewed, and the relative advantages of each technique will be discussed. With this knowledge, the reader will be better able to choose an imaging study for evaluating oncologic patients with known or suspected liver disease.

CT or MRI?

Both MRI and CT of the liver have various relative advantages and disadvantages (Table 1). Careful consideration of these advantages and disadvantages is warranted before selecting the study that is most appropriate for an individual patient. In many cases, the choice may not be clear, and either study would be considered the standard of care. However, a closer examination will show that each has some fairly specific benefits and drawbacks.

In general, CT is less costly than MRI because it involves lower capital equipment costs and a shorter examination time. Computed tomography is also widely available and often does not have the same “backlog” of patients as does MRI. Most physicians are relatively comfortable with evaluating CT images themselves, although nearly all still rely on the interpretation of a diagnostic radiologist for a final diagnosis.

The major limitation associated with the use of hepatic CT studies is the need for iodinated contrast material. Both the conventional and the newer nonionic contrast agents are nephrotoxic, and, as a result, their use is restricted in patients with renal insufficiency.[19,20] Iodinated contrast is also associated with a relatively high rate of adverse reactions.[21] In addition, CT uses ionizing radiation, the potential harm of which is poorly understood. Most significantly, studies have shown that CT is less sensitive and specific than MRI in detecting and characterizing focal hepatic disease—a fact that will be discussed in greater detail below.

Magnetic resonance imaging of the liver has several advantages over CT of the liver. Magnetic resonance imaging provides excellent contrast that can reveal subtle variations in tissues of differing histology. No ionizing radiation or nephrotoxic contrast media is used, and the most commonly administered contrast agents, extracellular gadolinium chelates, have very favorable safety profiles.[19] Magnetic resonance images can be acquired in any orientation and, unlike CT scans, are not limited to the axial plane. Together, these advantages of MRI lead to better detection and characterization capabilities than are available with CT.

Magnetic resonance imaging also has some important drawbacks, however. In most regions, it is a more expensive examination than CT. There are fewer MRI scanners than CT scanners, which can result in scheduling difficulties. In general, the MRI examination takes longer and has more contraindications than a CT scan. Finally, MRI has more contraindications than does CT. These include, but are not limited to, the presence of a pacemaker or implanted device (such as a defibrillator or insulin pump), some aneurysm clips and heart valves, and recently placed vascular stents.

Technical Developments

Two types of CT technology are currently in clinical use: axial (or conventional) and spiral (or helical) CT.

Axial vs Spiral CT

Axial CT has been available for more than 20 years, although it has been improved continuously. In this technique, the patient is moved a short distance incrementally and a single image is formed, followed by another incremental move and another image acquisition, and so on.

Spiral CT uses a constant linear movement of the patient through the scanner as the x-ray tube revolves continuously 360°, forming a “spiral” around the patient. This allows for more rapid scanning than can be achieved with axial CT, and in general, overcomes many of the constraints of conventional CT of the liver.[22] Faster scanning is important because it makes possible the acquisition of more images during a single period, reducing respiratory and other motion artifacts. Multiple image sets can then be obtained following infusion of a contrast agent.

Dual-Phase Contrast-Enhanced CT

Dual-phase contrast-enhanced CT is a recently developed improvement in liver imaging. In this technique, a complete set of images is obtained during the hepatic-arterial dominant phase of liver perfusion, followed by a second set obtained during the portal-venous phase. Some lesions, such as hepatocellular carcinoma (HCC) and certain metastases, have blood supplied mostly from the hepatic artery and will have higher attenuation values than the liver parenchyma during the arterial phase (Figure 1). Most metastases have a portal-venous blood supply and are hypoattenuating compared to the rest of the liver during the portal phase of perfusion.

The dual-phase technique offers advantages for both lesion detection and characterization over the single-phase studies, and many dedicated CT liver studies are now performed using this technique.[23-25] Dual-phase examinations require higher injection rates of contrast agent than are routinely used; for this reason, nonionic contrast is customarily administered to reduce patient discomfort and the incidence of adverse reactions.

CT Angiography

Angiographically assisted CT is an invasive technique used to increase the sensitivity of CT.[15-18,26-30] In this study, a catheter is placed either in the celiac or hepatic artery (CT hepatic arteriography [CTHA]) or in the superior mesenteric or splenic artery (CT during arterial portography [CTAP]). A rapid bolus of contrast is delivered, which results in a large contrast load reaching the liver either directly from the hepatic artery (CTHA) or indirectly via the portal or splenic vein (CTAP). Although these angiographically assisted techniques are more sensitive than noninvasive CT, they are not routinely used because of their invasive nature.

Computed tomography during arterial portography has been recognized as the most effective imaging technique for the preoperative determination of hepatic tumor resectability, but this distinction was challenged recently by MRI studies using liver-specific contrast agents.[15-18] One significant drawback of CTAP, aside from its invasive nature, is its lowered specificity due to the false-positive “perfusion defects” seen in the liver.[29] The perfusion defects are caused by altered hepatic vasculature and can appear identical to focal hepatic metastases. Delayed images can help reduce the number of false-positive focal lesions, but CT angiography remains a sensitive, but less specific, technique.

Ultrafast MRI

Magnetic resonance studies use several varied pulse sequences to uncover subtle histologic differences between normal and abnormal tissues in the liver. At present, MRI can acquire some types of images more than 100 times faster than was possible 10 years ago. This has enabled a significantly shorter study time, and has allowed the acquisition of many different sequences, both before and after administration of intravenous contrast.

Today, a basic liver MRI protocol involves three to four pulse sequences that can generally be acquired within 20 minutes. Hepatic imaging protocols usually include T2-weighted, inversion recovery, and T1-weighted sequences. Specific sequences, such as inversion recovery, are highly sensitive for the detection of hepatic neoplasms.[30] Once depicted on MRI, focal lesions can often be accurately characterized as malignant or benign, cyst or solid tumor, and so forth, based on their appearance and relative signal intensity (Figure 2).

The development of fast MRI sequences has reduced acquisition times to the point that the entire liver can be imaged in 15 to 30 seconds (the length of time most people can hold their breath). Using this technique, multiple sets of gadolinium-enhanced, T1-weighted images can be acquired in a manner similar to dual-phase hepatic CT; ie, one set can be obtained during the hepatic arterial phase, one set during the portal phase, and then one or more sets during delayed-equilibrium phases. The pattern with which focal hepatic lesions enhance on these multiple sets of images is used to characterize the lesions. Specific lesions have distinctive enhancement patterns; hepatic hemangiomas can be characterized in this manner with a specificity approaching 100% (Figure 3).[31]

Liver-Specific MRI Contrast Agents

Until recently, gadolinium chelates were the only clinically available contrast agents for MRI. Two new agents, ferumoxides and mangafodipir trisodium injection (Teslascan) recently gained Food and Drug Administration (FDA) approval and are presently in routine clinical use in the United States.

Many other contrast agents are in various stages of clinical development. These new contrast agents have been designed specifically for hepatic imaging, although they do have other indications as well.[32] They selectively target, accumulate, and remain within the liver, allowing much longer periods during which imaging can be performed after administration. These agents have increased the ability to both detect and characterize focal liver lesions, and will be discussed in greater detail below.

CT-MRI Comparison Studies

Initial Studies

Many of the initial CT/MRI comparison studies were performed in the late 1980s and early ’90s using technology that would currently be considered out of date; thus, the results of those studies are not directly applicable today. Since spiral CT technology was not available at that time, only axial CT imaging was used. Not only was axial CT slow, but also it created studies that used thicker imaging slices (10 mm) than are generally employed with spiral technology (5 mm).

Magnetic resonance technology was also substantially slower in the late 1980s and early ’90s than it is today. This, again, resulted in studies that (1) used thicker imaging slices, reducing detection of smaller lesions; and (2) were acquired too slowly to perform breath-hold imaging, resulting in greater respiratory motion and image blurring. In addition, none of the currently used contrast agents was available, limiting both the sensitivity and specificity of the examinations.

Because of these limitations, the early studies comparing CT with MRI found few significant differences in the accuracy rates for detecting focal hepatic disease. One of the most widely referenced studies is that of the Radiology Diagnostic Oncology Group II.[1] From 1989 to 1993, this group conducted a multi-institutional trial that evaluated CT and MRI in 478 patients with colorectal carcinoma. The authors found the accuracy of MRI and CT in detecting liver metastases to be identical (85%). Computed tomography was less sensitive than MRI (62% and 70%, respectively) but had a slightly greater specificity (97% and 94%, respectively).

Rummeny et al compared conventional (axial) CT with unenhanced MRI in 39 patients.[2] Their direct comparison of the best MRI technique (a T2-weighted spin-echo sequence that is no longer used routinely) and the best CT technique (incremental dynamic bolus CT) showed a strong trend in favor of T2-weighted MRI over incremental dynamic bolus CT. However, the differences did not reach statistical significance.

Current Studies

Most of the studies discussed here were performed over the past several years. The MRI techniques varied substantially from study to study, while the CT techniques varied somewhat less. The MRI studies used different contrast agents and disparate pulse sequences. The CT studies used variable protocols that altered the contrast volumes and injection rates, the speed of the table motion, and the “pitches” for the spiral studies. Although these individual variables may have differed from study to study, they still represent, for the most part, the “cutting edge” of technology.

Most of the studies used the best current techniques or protocols available in order to compare optimal CT with optimal MRI. Specific technique protocols are detailed extensively in each report, and radiologists are encouraged to review the reference articles if they wish to reproduce the techniques at their own institutions. Since the CT studies vary less technically than the MRI studies, the reports will be grouped according to the type of contrast agent used, with the exception of CT angiographic studies, which will be reviewed separately.

Gadolinium-Enhanced MRI

Gadolinium has a great number of MRI applications throughout the body and has been used routinely worldwide for over 10 years. It is the most commonly used contrast agent for MRI, and several formulations are currently available. As an intravenous contrast agent, it acts in a manner similar to that of the iodinated contrast agents used for CT.

Gadolinium chelates are very useful for characterizing focal hepatic lesions because specific lesions have distinctive enhancement patterns on dynamic gadolinium-enhanced MRI.[31] However, its usefulness in increasing the detection of focal hepatic lesions has been disputed.[3,33]

In a recent study, Hamm et al questioned the effectiveness of gadolinium-enhanced MRI vs unenhanced MRI in increasing the detection of hepatic neoplasms.[34] They found no statistically significant difference in the detection rate of hepatic metastases between the use of unenhanced and gadolinium-enhanced MRI.

Other researchers have found gadolinium chelates to be useful in the detection of metastases. Semelka et al recently reported the results of a study that compared single-phase spiral CT and MRI images, obtained prior to and following gadolinium enhancement, with respect to liver lesion detection and characterization.[3] In this study, MRI detected and characterized significantly more focal lesions than did CT.

The effect on patient management was determined by combined chart review and interview of the patients’ physicians and by retrospective clinical assessments performed by a surgical oncologist and medical oncologist separately. Magnetic resonance imaging had a greater effect on patient management than did single-phase spiral CT in more than 61% of patients.

MRI With RES-Specific Agents

Several new contrast agents have been designed that specifically target the cells of the recticuloendothelial system (RES). In most patients, the bulk of an injected dose will accumulate and stay in the liver. The RES agents are composed of iron oxide particles. Each of the iron oxide agents consists of an iron oxide core and a biodegradable coating. They produce a dose-dependent decrease in signal intensity on both T1- and T2-weighted images and significantly increase detection of focal hepatic lesions on MRI.[4-6,34-41]

Among the numerous formulations of the RES agents developed, three have been studied extensively for hepatic imaging. These include the ferumoxides (Feridex I.V.), SHU-555A (Resovist Injection), and AMI-227 (Combidex).

Following intravenous administration, the RES agents circulate in the blood before they are extracted by RES phagocytosis. In a normally functioning liver, about 80% of a ferumoxides dose is taken up by the Kupffer cells, 6% by the spleen, and a small amount by the bone marrow.[42]

There is differential uptake of ferumoxides by RES-containing normal liver parenchyma, as compared with liver that has been replaced by other tissue. Normal liver tissue appears much darker on T2-weighted images while liver metastases do not show any signal change because they lack substantial RES activity. This difference in enhancement makes focal hepatic lesions more conspicuous, since they are easier to identify as bright masses against a dark liver background and, thus, increases the sensitivity of the examination for lesion detection (Figure 4).

Many well-controlled studies, some using surgical pathology or intra-operative ultrasound as gold standards, have supported the efficacy of ferumoxides-enhanced MRI vs CT and unenhanced MRI.[4-6,32-41]

Studies of Ferumoxides-Enhanced MRI

One of the initial comparative studies was performed by Hagspiel et al. They imaged patients with liver metastases who were candidates for curative surgery.[5] The patients underwent surgery and intraoperative ultrasound after submitting to preoperative ultrasound, dynamic axial CT, and pre– and post–ferumoxides-enhanced MRI. The standard of reference was the total number of metastases identified at intraoperative ultrasound and pathologic examination.

In this study, ferumoxides-enhanced MRI was significantly more sensitive in detecting lesions than ultrasound, axial CT, or unenhanced MRI. The enhanced MRI identified 99% of the focal lesions detected by all of the other noninvasive modalities (unenhanced MRI, ultrasound, and axial CT) combined but only 56% of the lesions detected by intraoperative ultrasound and pathologic examination.

In the United States, phase III clinical trials again showed that ferumoxides-enhanced MRI examinations were statistically superior to other noninvasive imaging techniques.[6] The enhanced MRI studies depicted additional lesions in 27% of patients, as compared with unenhanced MRI, and more lesions in 40% of patients, as compared with contrast-enhanced axial CT. It is important to note, however, that this study was performed in the early and mid-1990s and did not use spiral CT technology.

More recent studies using spiral CT have confirmed the superiority of ferumoxides-enhanced MRI over other techniques. Muller et al compared the efficacy of ferumoxides-enhanced MRI and dual-phase spiral CT in detecting liver metastases and hepatocellular carcinoma.[7] They studied 38 patients with a total of 144 malignant hepatic lesions who underwent CT and MRI. MRI following administration of ferumoxides was associated with the highest rate of detection and was significantly superior to unenhanced MRI and double spiral CT.

Ward et al evaluated 51 hepatic resection candidates with known colorectal metastases.[8] These patients were examined using pre– and post–ferumoxides-enhanced MRI and state-of-the-art dual-phase spiral CT. The authors used an alternative free-response receiver-operator characteristic method to analyze the results. The gold standard for the presence of a lesion consisted of findings from surgery, intraoperative ultrasound scans, and histopathologic studies in 31 patients and consensus review of all other imaging studies and clinical follow-up in 20 patients.

These investigators found that the mean sensitivity of MRI was significantly higher than that of dual-phase spiral CT. They determined the sensitivity to be 79.8% for MRI and 75.3% for CT for all lesions, and 80.6% for MRI and 73.5% for CT for malignant lesions.

Instead of using MRI as the primary imaging modality, many centers rely on MRI as a problem-solving method that is employed when CT findings are either inconclusive or conflict with other clinical data. Using this approach, Helmberger et al evaluated ferumoxides-enhanced MRI in 46 patients in whom a primary or secondary hepatic malignancy was suspected, but in whom dual-phase spiral contrast-enhanced CT was inconclusive.[9] Spiral CT was performed followed by pre– and post–ferumoxides-enhanced MRI. This study used a mixed gold standard (histologic proof in 30 of 36 cases, long-term follow-up in 16 of 46 cases).

Ferumoxides-enhanced MRI revealed significantly more lesions than CT. MRI had both greater sensitivity (97%) and specificity (88%). The differences between the modalities were even more pronounced in the detection of lesions smaller than 10 mm; again, ferumoxides-enhanced MRI was the most sensitive method.

Economic Considerations

Economic factors also play an important role in determining which imaging modality will be used. Schultz et al performed a retrospective chart review comparing ferumoxides-enhanced MRI with contrast-enhanced CT in the preoperative imaging of hepatic neoplasms.[10] They also evaluated the clinical impact and overall health care costs associated with ferumoxides-

enhanced MRI. A comparison of CT and MRI findings with findings from intraoperative ultrasound and pathologic specimens showed a significant difference in sensitivity (ferumoxides-enhanced MRI, 86%; contrast-enhanced CT, 58%; P < .001). These researchers also determined that MRI was an accurate predictor of eventual surgical and intraoperative ultrasound (IOUS) findings.

In this study, in 95% of the patients imaged preoperatively with ferumoxides-enhanced MRI, there were no additional lesions detected using IOUS. Compared to contrast-enhanced CT, ferumoxides-enhanced MRI altered the clinical management in 67% of patients. These management changes corresponded to an overall net cost savings of $108,368 ($1,901 per patient). The authors concluded that ferumoxides-enhanced MRI is an economically feasible imaging method that will alter the clinical management in a substantial number of patients as compared with contrast-enhanced CT.

MRI Using Hepatobiliary Agents

Several MRI agents have been developed that are actively taken up by hepatocytes.[11-14,31] Teslascan was approved in the United States in late 1997. Gadolinium (Gd)-BOPTA (MultiHance) and Gd-EOB-DTPA (Eovist Injection) are two new hepatobiliary agents that are currently in clinical trials. These new agents are actively taken up by hepatocytes and produce enhancement of the liver. Unlike gadolinium chelates, they each remain within the hepatocytes for a relatively prolonged period.

The hepatobiliary agents produce an increase in liver-to-lesion contrast and increase lesion distinction following administration. Most metastases do not enhance, but hepatocellular tumors and some endocrine metastases do enhance, making manganese-DPDP useful for lesion characterization.

Manganese-DPDP–Enhanced MRI

The results of the US phase III studies of manganese-DPDP have not yet been reported, but results of European trials have been published.[11-13] Torres et al reported results from two independent trials in Europe in which a total of 624 patients were evaluated.[12] They compared the results of manganese-DPDP–enhanced MRI with contrast-enhanced CT in a subgroup of 137 patients. An analysis of this subset of patients determined that the manganese-DPDP–enhanced images were significantly superior to contrast-enhanced CT images in the detection of lesions.

Our group recently reported the results of a study of 27 consecutive patients with colorectal carcinoma referred for hepatic resection who underwent dual-phase spiral CT, unenhanced MRI, and enhanced MRI.[14] All 27 of the patients had undergone surgery within 10 days of imaging, and the findings were correlated with the intraoperative sonography and pathologic analysis of the resected hepatic specimens. A total of 62 metastases were confirmed by pathologic analysis and intraoperative sonography. The analysis showed that spiral CT depicted 93% of the lesions; unenhanced MRI, 84%; and Mn-DPDP–enhanced MRI, 97%.

Both gadolinium-BOPTA and Gd-EOB-DTPA are paramagnetic contrast agents that are selectively taken up and secreted into the bile by hepatocytes.[30] These two agents have not been approved for routine diagnostic use in the United States. Direct comparison studies of MRI enhanced with these agents and spiral CT should be available soon.

MRI vs CT Angiography

In the past, CTAP or CTHA had been considered the most sensitive examinations for the detection of focal liver lesions. Although these examinations are expensive and invasive procedures, many clinicians prefer to perform them prior to hepatic resection.

Several recent studies have challenged the view that CTAP should be the study of choice prior to hepatic intervention. Seneterre et al, using receiver-operator characteristic analysis, assessed the ability of unenhanced and ferumoxides-enhanced MRI and CTAP to detect metastatic involvement in 17 patients.[15] The accuracy rates of CTAP, unenhanced MRI, and ferumoxides-enhanced MRI were 92.5%, 90.8%, and 95.1%, respectively. These investigators concluded that ferumoxides-

enhanced MRI is at least as accurate as CTAP in detecting hepatic metastases and that CTAP could potentially be replaced by the noninvasive MRI examination. This study prompted an editorial by Soyer in Radiology that advocated the replacement of CTAP with ferumoxides-enhanced MRI.[28]

Several other studies have supported the contention that MRI could replace CTAP preoperatively. Semelka et al found that spiral CTAP and dynamic gadolinium-enhanced MRI were approximately equivalent in lesion detection.[17] Lencioni et al determined that the difference in sensitivity between ferumoxides-enhanced MRI and spiral CTAP was not statistically significant (P > .1).[18] In their study, spiral CTAP depicted nine false-positive lesions, while MRI had no false-positive results. The positive predictive value was 79% for spiral CTAP and 100% for combined pre- and postcontrast MRI.

Not every recent study has supported the superiority of MRI over spiral CTAP. Several authors have concluded that spiral CTAP is more sensitive than ferumoxides-enhanced MRI in patients who are to undergo liver resection.[16,43] Strotzer et al prospectively studied 35 patients using contrast-enhanced spiral CT, spiral CTAP, and ferumoxides-enhanced MRI.[16] They found that spiral CTAP had significantly higher sensitivity, based on a lesion-by-lesion analysis.


Liver CT and MRI have undergone rapid and substantial improvements in the past decade. Such developments as dual-phase spiral CT, CT angiography, ultrafast MRI, and MRI enhanced with liver-specific contrast agents have allowed for better detection and characterization of focal hepatic neoplasms.

Direct comparison studies have repeatedly confirmed that MRI has greater accuracy for detecting hepatic lesions, especially when the targeted contrast agents are employed. Angiographically assisted CT and MRI appear to have similar sensitivity for detecting hepatic neoplasms; however, MRI has greater specificity and the advantage of being noninvasive.

State-of-the-art spiral CT and MRI are each highly accurate in detecting focal hepatic neoplasms. Liver CT is most commonly employed as a routine screening tool because it is somewhat less costly and is more widely available. Magnetic resonance imaging has clear advantages when the highest possible accuracy is desired, such as when imaging patients prior to hepatic resection or intervention, or when the CT result is either inconclusive or conflicts with other clinical data.


1. Zerhouni EA, Rutter C, Hamilton SR, et al: CT and MR imaging in the staging of colorectal carcinoma: Report of the Radiology Diagnostic Oncology Group II. Radiology 200:443-451, 1996.

2. Rummeny EJ, Wernecke K, Saini S, et al: Comparison between high-field-strength MRI and CT for screening of hepatic metastases: A receiver operating characteristic analysis. Radiology 182:879-886, 1992.

3. Semelka RC, Worawattanakul S, Kelekis NL, et al: Liver lesion detection, characterization, and effect on patient management: Comparison of single-phase spiral CT and current MR techniques. J Magn Reson Imaging 7:1040-1047, 1997.

4. Fretz CJ, Stark DD, Metz CE, et al: Detection of hepatic metastases: Comparison of contrast-enhanced CT, unenhanced MRI, and iron oxide-enhanced MR imaging. Am J Roentgenol 155:763-770, 1990.

5. Hagspiel KD, Neidl KFW, Eichenberger AC, et al: Detection of liver metastases: Comparison of superparamagnetic iron oxide-enhanced and unenhanced MRI at 1.5 T with dynamic CT, intraoperative US, and percutaneous US. Radiology 196(2):471-478, 1995.

6. Ros PR, Freeny PC, Harms SE, et al: Hepatic MR imaging with ferumoxides: A multi-center clinical trial of the safety and efficacy in the detection of focal hepatic lesions. Radiology 196:481-488, 1995.

7. Muller RD, Vogel K, Neumann K, et al: SPIO-MR imaging vs double-phase spiral CT in detecting malignant lesions of the liver. Acta Radiol 40:628-635, 1999.

8. Ward J, Naik KS, Guthrie JA, et al: 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 210:459-466, 1999.

9. Helmberger T, Gregor M, Holzknecht N, et al: Comparison of dual-phase helical CT with native and ferumoxides-enhanced magnetic resonance imaging in detection and characterization of focal liver lesions. Radiologe 39:678-684, 1999.

10. Schultz JF, Bell JD, Goldstein RM, et al: Hepatic tumor imaging using iron oxide MRI: Comparison with computed tomography, clinical impact, and cost analysis. Ann Surg Oncol 6:691-698, 1999.

11. Rummeny EJ, Torres CG, Kurdziel JC, et al: Manganese-DPDP for MR imaging of the liver. Results of an independent image evaluation of the European phase III studies. Acta Radiol 38:638-642, 1997.

12. Torres CG, Lundby B, Sterud AT, et al: Manganese-DPDP for MR imaging of the liver. Results from the European phase III studies. Acta Radiol 38:631-637, 1997.

13. Wang C, Ahistrom H, Ekholm S, et al: Diagnostic efficacy of Mn-DPDP in MR imaging of the liver. A phase III multicentre study. Acta Radiol 38:643-649, 1997.

14. Earls JP, Choti MA, Borman TL, et al: Detection of liver metastases prior to hepatic resection: Prospective comparison of dual-phase spiral CT, unenhanced and Mn-DPDP (Teslascan) enhanced MRI, intraoperative sonography, and surgical pathology. Paper presented at: Annual Meeting of the Radiological Society of North America (RSNA). Chicago, 1999.

15. Seneterre E, Taourel P, Bouvier Y, et al: Detection of hepatic metastases: Ferumoxides-enhanced MR imaging vs unenhanced MR imaging and CT during arterial portography. Radiology 200:785-792, 1996.

16. Strotzer M, Gmeinwieser J, Schmidt J, et al: Diagnosis of liver metastases from colorectal adenocarcinoma. Comparison of spiral-CTAP combined with intravenous contrast-enhanced spiral-CT and SPIO-enhanced MRI combined with plain MRI. Acta Radiol 38:986-992, 1997.

17. Semelka RC, Cance WG, Marcos HB, et al: Liver metastases: Comparison of current MR techniques and spiral CT during arterial portography for detection in 20 surgically staged cases. Radiology 213:86-91, 1999.

18. Lencioni R, Donati F, Cioni D, et al: Detection of colorectal liver metastases: Prospective comparison of unenhanced and ferumoxides-enhanced magnetic resonance imaging at 1.5 T, dual-phase spiral CT, and spiral CT during arterial portography. MAGMA 7:76-87, 1998.

19. Prince MR, Arnoldus C, Frisoli JK: Nephrotoxicity of high-dose gadolinium compared with iodinated contrast. J Magn Reson Imaging 6:162-166, 1996.

20. Deray G, Bellin MF, Boulechiar H, et al: Nephrotoxicity of contrast media in high-risk patients with renal insufficiency: Comparison of low- and high-osmolar contrast agents. Am J Nephrol 11:309-312, 1991.

21. Spring DB, Bettmann MA, Barkan HE: Deaths related to iodinated contrast media reported spontaneously to the United States Food and Drug Administration, 1978-1994: Effect of the availability of low-osmolality contrast media. Radiology 204:333-337, 1997.

22. Heiken JP, Brink JA, Vannier MW: Spiral (helical) CT. Radiology 189:647-656, 1993.

23. Baron RL, Oliver JH, Dodd Gd, et al: Hepatocellular carcinoma: Evaluation with biphasic, contrast-enhanced helical CT. Radiology 199:505-511, 1996.

24. Bonaldi VM, Bret PM, Reinhold C, et al: Helical CT of the liver: Value of an early hepatic arterial phase. Radiology 197:357-363, 1995.

25. Hollett MD, Jeffrey RB, Nino-Murcia M, et al: Dual-phase helical CT of the liver: Value of arterial phase scans in the detection of small (< 1.5 cm) malignant hepatic neoplasms. Am J Roentgenol 164:879-884, 1995.

26. Soyer P, Levesque M, Caudron C, et al: MRI of liver metastases from colorectal cancer vs CT during arterial portography. J Comput Assist Tomogr 17:67-74, 1993.

27. Nelson RC, Chezmar JL, Sugarbaker PH, et al: Preoperative localization of focal liver lesions to specific liver segments: Utility of CT during arterial portography. Radiology 176:89-94, 1990.

28. Soyer P: Will ferumoxides-enhanced MR imaging replace CT during arterial portography in the detection of hepatic metastases? Prologue to a promising future. Radiology 200:610-611, 1996.

29. Yamagami T, Arai Y, Matsueda K, et al: The cause of nontumorous defects of portal perfusion in the hepatic hilum revealed by CT during arterial portography. Am J Roentgenol 172:397-402, 1999.

30. Paulson EK, Baker ME, Paine SS, et al: Detection of focal hepatic masses: STIR MR vs CT during arterial portography. J Comput Assist Tomogr 18(4):581-587, 1994.

31. Quillin SF, Atilla S, Brown JJ, et al: Characterization of focal hepatic masses by dynamic contrast-enhanced MR imaging: Findings in 311 lesions. Magn Reson Imaging 15:275-285, 1997.

32. Earls JP, Bluemke DA: New MR imaging contrast agents. Magn Reson Imaging Clin N Am 7:255-273, 1999.

33. Hamm B, Mahfouz AE, Taupitz M, et al: Liver metastases: Improved detection with dynamic gadolinium-enhanced MR imaging? Radiology 202:677-682, 1997.

34. Majumdar S, Zoghbi S, Pope C, et al: Quantitation of MR relaxation effects of iron oxide particles in liver and spleen. Radiology 169:653-655, 1988.

35. Saii S, Stark DD, Hahn PF, et al: Ferrite particles: Superparamagnetic MR contrast agent for the reticuloendothelial system. Radiology 162:211-217, 1987.

36. Bellin MF, Zaim S, Auberton E, et al: Liver metastases: Safety and efficacy of detection with superparamagnetic iron oxide in MR imaging. Radiology 193:657-663, 1994.

37. Denys A, Arrive L, Servois V, et al: Hepatic tumors: Detection and characterization at 1-T MR imaging enhanced with AMI-25. Radiology 193:665-669, 1994.

38. Elizondo G, Weissleder R, Stark DD, et al: Hepatic cirrhosis and hepatitis: MR imaging enhanced with superparamagnetic iron oxide. Radiology 174:797-801, 1990.

39. Fretz CJ, Elizondo G, Weissleder R, et al: Superparamagnetic iron oxide enhanced MR imaging: Pulse sequence optimization for detection of liver cancer. Radiology 172:393-397, 1989.

40. Stark DD, Weissleder R, Elizondo G, et al: Superparamagnetic iron oxide: Clinical application as a contrast agent for MR imaging of the liver. Radiology 168:297-301, 1988.

41. Winter TC 3d, Freeny PC, Nghiem HV, et al: MR imaging with IV superparamagnetic iron oxide: Efficacy in the detection of focal hepatic lesions. Am J Roentgenol 161:1191-1198, 1993.

42. Weissleder R, Stark DD, Engeistad BL, et al: Superparamagnetic iron oxide: Pharmacokinetics and toxicity. Am J Roentgenol 152:167-173, 1989.