Cross-sectional imaging of the abdomen in oncology patients presents unique challenges and opportunities. A close working relationship between the oncologist and radiologist is essential for the exchange of the clinical and
ABSTRACT: Cross-sectional imaging of the abdomen in oncology patients presents unique challenges and opportunities. A close working relationship between the oncologist and radiologist is essential for the exchange of the clinical and imaging information necessary for optimizing patient diagnosis and management. Compared to helical computed tomography (CT), magnetic resonance imaging (MRI) of the abdomen and pelvis offers important advantages, including superior soft-tissue contrast. The multiplanar capabilities of MRI allow for direct coronal or sagittal imaging, providing a truer anatomic presentation of abdominal and pelvic masses. Recent advances in MRI, including the use of intravenous (IV) and oral contrast agents, the development of high-performance imagers, and improved surface coil designs, facilitate more rapid abdominal imaging with superior image quality. All of these features combine to produce a versatile imaging examination with exquisite sensitivity for depicting abdominal and pelvic tumors. In this article, we will review the clinical applications for hepatic and extrahepatic abdominal MRI in the oncology patient. The MRI techniques and protocols described can be applied to most commercially available high-field magnetic resonance imagers. [ONCOLOGY 14(Suppl 3):5-14, 2000]
The imaging evaluation of oncology patients requires accurate depiction and characterization of all hepatic and extrahepatic tumors. While helical computed tomography (CT) has been the workhorse of most radiology departments, recent advances in abdominal magnetic resonance imaging (MRI) have moved it to the forefront of oncologic imaging at our institution, the Sharp Memorial Hospital.[1,2]
Compared to helical CT, MRI of the abdomen offers the potential advantages of superior soft-tissue contrast and multiplanar imaging. The extent to which we are able to distinguish tumor from normal abdominal soft tissues is central to our ability to accurately depict the dimensions of a tumor. Magnetic resonance images display a much wider range of soft-tissue contrast, making tumor masses easier to distinguish from adjacent soft tissues (Figure 1). The addition of contrast agents provides differential enhancement of tumor and normal soft tissues, thus further improving the delineation of hepatic and extrahepatic tumor (Figure 2).
Among the recent advances in MRI are faster pulse sequences, breath-hold imaging, and use of intravenous contrast agents and surface coils, all of which have improved image quality and shortened examination times. The versatility of abdominal MRI is unmatched by any other imaging examination. For these reasons, at our institution, MRI has evolved from a problem-solving study to become the primary imaging examination in many patients with malignancy. In this review, we will discuss the spectrum of oncologic applications for abdominal MRI and will highlight areas in which newer MRI techniques offer significant advantages over helical CT.
The depiction of a focal liver lesion requires a difference in signal intensity between the lesion and the adjacent liver parenchyma. The lesion may be either more or less intense than the surrounding liver. Most liver masses are easily depicted on unenhanced T1-weighted or T2-weighted MRIs (Figure 3). However, some tumors produce minimal changes in T1- and T2-relaxation and, therefore, show limited contrast with the surrounding liver.
Contrast agents are used to accentuate the inherent differences in liver-lesion signal intensity. These contrast agents facilitate differential enhancement of liver parenchyma and masses, as well as promote lesion depiction and, to some degree, lesion characterization (Figure 4).[3,4]
Gadolinium chelates were the first intravenous contrast agents to be approved. These nonspecific extracellular agents rapidly equilibrate from the intravascular space into the extracellular space after injection. The use of gadolinium chelates has become an integral part of MRI of the liver and extrahepatic abdomen. Although liver-specific contrast agents are now available, gadolinium chelates continue to offer significant advantages in abdominal MRI.
Gadolinium chelates uniquely provide important information about tumor perfusion, which is key in the assessment of liver masses. They assist with the detection and characterization of liver lesions and in the establishment of the volume of viable perfused tumor.
Gadolinium chelates are equally important for MRI of the extrahepatic abdomen. The interstitial accumulation of these agents within peritoneal, omental, and gastrointestinal tumor produces marked enhancement and is crucial to accurate tumor staging. Depiction of lesions within solid visceral organs, such as the pancreas, kidneys, and spleen, also improves following gadolinium injection.
For effective hepatic MRI, we combine unenhanced axial T1-weighted spoiled gradient recalled echo (SPGR), and fat suppressed T2-weighted MRI with rapid, serial dynamic gadolinium-enhanced SPGR MRI. Three sets of dynamic gadolinium-enhanced liver images are obtained during the arterial, portal-venous, and equilibrium phases of liver enhancement. On high-field strength systems, rapid SPGR T1-weighted sequences are obtained after a 0.1-mmol/kg bolus injection of gadolinium chelate. By imaging the entire liver volume during a short period of suspended respiration, motion artifact is eliminated, while the intravascular gadolinium provides important information on tumor perfusion.
Compared to CT with iodinated contrast material, MRI has proven to be superior in evaluating metastatic liver disease.[5-9] Semelka et al compared single-phase helical CT and MRI with fat suppressed T2-weighted and gadolinium-enhanced SPGR in 89 patients with focal liver masses. In 49% of patients, MRI detected more lesions than did helical CT, with MRI depicting 519 true-positive liver masses vs 295 lesions depicted with helical CT.
Delineation of lesion borders is often easier on MRI, allowing for more consistent measurement of hepatic metastasis (Figure 4).
Hypervascular Hepatic Metastases
Hypervascular metastases from renal cell carcinoma, pancreatic islet cell tumors, breast carcinoma, thyroid carcinoma, sarcomas, and carcinoid tumors are supplied by the hepatic artery and enhance rapidly following the injection of gadolinium chelates. Since 75% to 80% of the blood supply to the liver derives from the portal vein, there is only minimal liver enhancement on these early images. Thus, on arterial phase images, hypervascular metastases will show marked enhancement against a background of minimally enhancing liver parenchyma (Figure 5).
Hypovascular Hepatic Metastases
Hypovascular metastases arise from colon carcinoma, pancreatic carcinoma, transitional cell carcinoma, and lung cancer and are best depicted on portal-venous phase gadolinium-enhanced images. These metastases receive a minimal supply of blood from the hepatic artery or portal vein. During the portal-venous phase, the liver parenchyma demonstrates marked enhancement while the hypovascular metastases show minimal enhancement, producing the greatest difference in liver-lesion signal intensity (Figure 6).[6,7,9] Gradual peripheral enhancement and heterogenous central enhancement of these lesions occurs on later images.
On arterial phase images, hypovascular metastases are poorly visualized, appearing only vaguely distinct from the adjacent liver. Since neither the tumor nor the liver is enhancing at this point, these hypovascular lesions may not be depicted. However, due to the high-contrast resolution of MRI, it is not uncommon to see thin peripheral enhancement of hypovascular metastases on gadolinium-enhanced capillary phase images. On equilibrium phase gadolinium-enhanced images, hypovascular metastases will become indistinct due to nonspecific interstitial accumulation of the contrast material. The pattern of peripheral washout of the contrast may also be noted.
Hepatocellular carcinomas (HCC) manifest a highly variable appearance on unenhanced T1-weighted and T2-weighted images. On T1-weighted imaging, HCC are most often hypointense compared with normal liver, although hyperintense lesions or areas of hyperintensity within a hypointense HCC can be seen. These hyperintense regions within the HCC reflect the presence of fat, copper, or protein. On T2-weighted imaging, HCC are generally hyperintense, although well-differentiated HCC may be isointense compared with liver parenchyma, thus limiting their detection.
Due to the variable appearance of HCC on T1-weighted and T2-weighted MRIs, dynamic gadolinium-enhanced imaging can play an important role in the diagnosis of this primary liver tumor.[10,11] Compared to helical CT with iodinated contrast material, the superior contrast resolution of MRI facilitates detection of small enhancing HCC on arterial phase gadolinium-enhanced SPGR images. In the setting of cirrhosis, any abnormal foci on arterial phase images should be considered highly suggestive of a developing HCC. In a recent study, Yamashita et al found that arterial phase gadolinium-enhanced MRI is superior to helical CT for detecting hepatocellular carcinoma.
Following IV injection of gadolinium, the pattern and degree of enhancement of HCC is related to tumor differentiation and histologic subtype. Well-differentiated tumors often show minimal arterial phase enhancement that washes out on portal-venous phase images, leaving the tumor hypointense compared with the adjacent liver. The presence of a capsule will be indicated by a hypointense rim on early images that enhances on delayed images.
Moderately or poorly differentiated HCC are characterized by dilated sinusoidal spaces, which produce moderate or marked enhancement with gadolinium on arterial phase images. Enhancing tumor nodules are often best depicted on these arterial phase dynamic images. Portal-venous invasion by an HCC is most accurately identified on gadolinium-enhanced breath-hold SPGR images. Expansion of the portal vein and replacement of the intravascular gadolinium by tumor thrombus is evidence of portal-venous tumor extension.
Cholangiography is well established as the definitive imaging technique for evaluating the intraluminal component of biliary tumors. Unfortunately, cholangiography is insensitive in detecting the extrabiliary extent of primary or metastatic hepatobiliary tumors-an important factor in determining resectability.[12,13] Extrabiliary tumor is better shown with cross-sectional imaging techniques, such as CT, MRI, or sonography.
Although helical CT is an excellent imaging technique for detecting focal liver masses and generally provides high-quality images of malignant biliary obstruction, its sensitivity in detecting intrahepatic or hilar cholangiocarcinomas has been variable. Incomplete depiction of the extent of hilar and peripheral cholangiocarcinomas can lead to understaging.
In our experience, cholangiocarcinomas are best depicted on gadolinium-enhanced MRIs. Cholangiocarcinomas are typically hypovascular tumors. Their marked intensification on delayed gadolinium-enhanced MRI is due to the diffusion of water-soluble contrast material into the large interstitial space of these tumors (known to have an abundant fibrous connective tissue stroma). Compared to helical CT, MRI is more sensitive to subtle changes in contrast enhancement because of its superior contrast resolution and, thus, has an advantage in detecting bile duct tumors and defining their extrabiliary extent.
The comprehensive biliary magnetic resonance examination also includes a magnetic resonance cholangiopancreatogram (MRCP). By imaging the intrinsic high-signal intensity of bile, an MRCP noninvasively creates an image of the biliary tree similar to that obtained endoscopically with endoscopic retrograde cholangiopancreatography (ERCP). With more rapid MR imaging, an MRCP can be obtained in seconds, producing excellent pictures of the intrahepatic and extrahepatic bile ducts and the pancreatic duct.
In patients with malignant biliary obstruction, an MRCP can depict the level of biliary obstruction caused by primary biliary tumors or adjacent hepatic or nodal tumor. Combining an MRCP with multiplanar anatomic MRI provides a thorough evaluation of patients with primary or metastatic biliary obstruction.
Evaluation of pancreatic adenocarcinoma requires accurate depiction of local tumor extent, involvement of adjacent vascular structures, and distant metastases. At most institutions, helical CT has been the primary imaging study for pancreatic carcinoma. Use of thinner slices and early dynamic contrast-enhanced imaging of the pancreas has improved the depiction of smaller pancreatic masses. However, determining the resectability of a pancreatic tumor still poses a challenge. Understaging of pancreatic cancer may lead to unnecessary surgical intervention for an unresectable tumor.
The inherently superior contrast resolution of MRI facilitates depiction of small pancreatic carcinomas. Pancreatic masses appear as dark- or low-signal intensity masses on fat-suppressed T1-weighted images, on which the normal pancreatic parenchyma is intrinsically bright. Since pancreatic carcinomas are hypovascular, they are depicted on dynamic arterial phase gadolinium-enhanced MRIs as dark hypovascular masses (Figure 7). Use of high-resolution MRI in multiple planes will also likely further improve the depiction of pancreatic carcinoma.
In a comparison of helical CT and MRI of pancreatic carcinoma, Ichikawa et al reported that dynamic gadolinium-enhanced MRI detected 19 of 21 tumors, as compared with 16 tumors detected on helical CT; also, MRI was equal or superior to helical CT in determining local tumor extent and vascular involvement. Sheridan et al recently reported their experience in 33 surgical candidates with pancreatic carcinoma. In their study of the optimal helical CT and MRI technique, both imaging tests depicted 29 of 31 pancreatic tumors. However, MRI was significantly better in determining tumor resectability.
These findings agree with our experience at the Sharp and Children’s MRI Center. We found that MRI is superior to CT in depicting small pancreatic carcinomas and in detecting direct tumor extension, vascular involvement, and metastases.
The breath-hold, gadolinium-enhanced, fat-suppressed SPGR MRIs are obtained dynamically through the pancreas only at 0, 1, and 2 to 3 minutes to depict a low-intensity pancreatic carcinoma adjacent to the normally enhancing pancreatic parenchyma. We obtain these dynamic enhanced images of the pancreas in the axial plane. Alternatively, they may be obtained in the oblique coronal plane along the axis of the pancreas. Coronal gadolinium-enhanced images and delayed axial gadolinium-enhanced SPGR MRIs are then taken of the entire abdomen.
The multiplanar capabilities, excellent soft-tissue contrast, and versatility of MRI make it the examination of choice for evaluating renal masses. Using axial helical CT, it can be difficult to depict small solid renal masses arising from the upper or lower pole of the kidney. Coronal or sagittal MRIs provide a truer anatomic presentation of the kidneys, allowing for a better appreciation of the position of a renal mass and its relationship to surrounding structures. In our experience, the detection and characterization of renal cell carcinoma are superior on MRI.
Characterization of a renal mass as cystic or solid can be made with confidence on the gadolinium-enhanced images. Cysts show no evidence of enhancement on dynamic gadolinium-enhanced MRIs, whereas solid tumors that are vascular exhibit some degree of enhancement. Compared to helical CT, gadolinium-enhanced MRIs are much more sensitive to subtle enhancement of solid masses, allowing for more confident lesion characterization.
Transitional cell carcinomas can also be depicted on MRI. Angiomyolipomas contain fat that is easily depicted on T1-weighted MRIs, facilitating characterization of this benign renal tumor.
The staging and presurgical evaluation of renal cell carcinomas requires accurate depiction of local tumor extent, venous tumor invasion, and nodal metastases. Previous studies have shown that MRI is superior to helical CT in depicting venous tumor extension.
The versatility of MRI allows for the combination of multiplanar anatomic imaging with: (1) an MR venogram to assess venous tumor invasion, (2) an MR urogram to evaluate the collecting system, and (3) an MR arteriogram to assess the number and position of renal arteries. This single, comprehensive MR examination provides a complete presurgical imaging evaluation of a renal mass. By supplanting multiple imaging studies, MRI provides a cost-effective approach to renal imaging.
A set of flow-sensitive axial gradient-echo (GRE) images are obtained through the kidneys and inferior vena cava (IVC) to depict venous tumor extension.
Coronal gadolinium-enhanced, fat-suppressed SPGR MRIs are obtained dynamically through the kidneys only at 0, 1, and 3 minutes. We use as small a field of view as is possible by placing the patient’s hands over his or her head to avoid wrap artifact. Three quick passes are taken through the kidneys only following injection of gadolinium during successive breath holds. This is followed by a set of axial fat-suppressed SPGR MRIs through the entire abdomen. In specific cases, we may combine this anatomic MR examination with a MR angiogram to depict the renal arteries or an MR urogram to depict the renal collecting system and ureters.
Magnetic resonance imaging provides a simple, quick, and accurate means to characterize an adrenal mass as a benign adenoma or a malignant lesion. The depiction of lipid within an adrenal mass confirms the presence of a benign adrenal adenoma.
Breath-hold SPGR MRI can be obtained with a fat-sensitive technique using an opposed-phase echo time, in which the signal from fat and water protons cancels and produces signal loss. On such opposed-phase MRIs, a lipid-containing adrenal adenoma will appear dark (Figure 8). A second set of MRIs obtained with an in-phase echo time will show an increase in the signal of the adrenal mass. By comparing the appearance of the adrenal mass on the in-phase and opposed-phase breath-hold MRIs, we can confidently characterize an adrenal mass (Figure 8).[19,20]
Adrenal masses that do not show a loss of signal on lipid-sensitive opposed-phase MRIs should be evaluated further with tissue diagnosis to rule out adrenal metastases, primary adrenal carcinoma, or a pheochromocytoma. Approximately 10% of adrenal adenomas do not contain lipid and will be indistinguishable from an adrenal malignancy. However, the other 90% will be correctly characterized as benign on fat-sensitive MRIs, obviating further diagnostic evaluation. At our institution, this simple approach to adrenal mass characterization has simplified staging in patients with a primary malignancy and an adrenal mass.
Breath-hold SPGR MRI of the adrenal glands is performed with an in-phase echo time of 4.4 ms and an opposed-phase echo time of 2.2 ms. These two sets of breath-hold images quickly characterize an adrenal mass.
The depiction of small peritoneal implants and carcinomatosis is a challenge for cross-sectional imaging studies, including helical CT and unenhanced MRI. However, with gadolinium-enhanced MRI, small peritoneal tumors are routinely depicted with a level of clarity unmatched by other imaging studies.[21-23] Marked enhancement of small peritoneal implants with IV gadolinium on MRIs facilitates the detection of metastases to free peritoneal surfaces and bowel serosa.
Following the injection of gadolinium, peritoneal implants slowly accumulate the contrast material and, therefore, are most conspicuous on images obtained 5 to 10 minutes after an IV injection of the gadolinium chelate (Figure 9). The addition of fat suppression reduces the competing high signal of the adjacent fat and is an important element in this technique.
Several studies have shown that gadolinium-enhanced, fat-suppressed MRI is superior to helical CT in depicting peritoneal tumors (Figure 2).[21-23] The superior performance of enhanced MRI compared to CT scanning is most noticeable in the depiction of small (< 1 cm) tumor implants and carcinomatosis. In a recent study, gadolinium-enhanced MRI detected 75% to 80% of small tumor implants (< 1 cm) compared to 22% to 33% detected on helical CT (P < .0001).
Monitoring Treatment Response
Gadolinium-enhanced MRI is useful in women with ovarian cancer to monitor response to therapy by depicting residual peritoneal tumor and by detecting recurrence in patients with a rising serum CA-125 level. Detecting clinically occult tumor is critical in determining appropriate patient management.
In a recent 5-year longitudinal study, we compared the results of MRI to serum CA-125 level and physical examination in 69 women with treated ovarian cancer. The findings on MRI were compared with the clinical impression of tumor presence or absence, based on CA-125 values and physical examination. Our experience confirmed that
gadolinium-enhanced MR imaging can detect clinically occult tumors in women with treated ovarian cancer. Of the 69 patients, 39 were in clinical remission with a normal CA-125 level and physical examination. In our study, 23 (59%) of the 39 patients in clinical remission had residual subclinical tumor proven by laparotomy or clinical follow-up. Gadolinium-enhanced MRI correctly showed residual tumor in 20 of 23 patients.
For all 69 patients, MRI had a 91% sensitivity, 87% specificity, 90% accuracy, and 72% negative predictive value, and was superior to both serum CA-125 level (53% sensitivity, 94% specificity, 63% accuracy, and 38% negative predictive value; P < .0001) and physical examination (26%, 94%, 43%, and 29%; P < .0001) in detecting residual or recurrent tumor.
A second-look laparotomy was performed in 34 patients. There was no significant difference between second-look laparotomy and MRI, each of which had an 87% sensitivity, a 75% specificity, and an 85% accuracy. In this subset of patients, second-look laparotomy and MRI were superior to serum CA-125 level (60% sensitivity, 100% specificity, and 65% accuracy; P < .05).
The improved sensitivity of gadolinium-enhanced MRI in depicting small-volume tumor compared to CA-125 levels alone, provides oncologists with information critical to patient management. It offers a more accurate means of monitoring response to adjuvant chemotherapy and detecting recurrence after initial response. In our experience, gadolinium-enhanced MRI often shows residual tumor in patients following adjuvant chemotherapy indicating a need for additional treatment (Figure 6).
The improved sensitivity of gadolinium-enhanced MRI to subtle peritoneal tumors is also useful in evaluating primary tumors of the pancreas and gastrointestinal tract that commonly spread by tumor cell exfoliation and peritoneal seeding. At our institution, the information from MRI is used to determine the appropriate course of therapy. For instance, in a patient with pancreatic carcinoma, depiction of peritoneal tumor spread will obviate unnecessary surgery for an unresectable tumor.
A fat-suppressed, breath-hold SPGR MRI following an IV gadolinium injection is a key imaging test for depicting peritoneal tumor. Images are obtained both immediately and 5 minutes following injection of 0.2 mmol/kg of IV gadolinium. The delayed images are most sensitive for depicting enhancing peritoneal tumors.
Bowel distention is critical for depicting serosal and mural tumor and is achieved with inexpensive dilute barium sulfate, typically used for CT scanning. We administer three 450-mL bottles starting 1 hour prior to the MR examination. Rectal water can also be used for distention of the rectosigmoid colon. At the time of gadolinium injection, 1 mg of IV glucagon is administered to eliminate bowel peristalsis on the gadolinium-enhanced SPGR images.
Primary and metastatic neoplasms involving the gastrointestinal tract are depicted on MRIs performed with adequate bowel distention.[24-27] Serosal tumor in patients with ovarian carcinoma is seen as an abnormal mural thickening and enhancement.[21,23] There is often adjacent enhancing peritoneal or mesenteric tumor. Similar metastatic involvement of the gastrointestinal tract may be depicted in patients with pancreatic carcinoma or disseminated tumor from any primary tumor.
Primary gastrointestinal malignancy is depicted as a dominant mass and/or mural thickening. The ability to detect subtle differences in soft-tissue contrast during MRI facilitates depiction of tumors of the stomach, small intestine, and colon. This is particularly true of gadolinium-enhanced MRIs, which easily detect enhancing intestinal tumors. Newer breathing-independent, single-shot RARE (single-shot fast spin echo [SSFSE] or half Fourier single-shot turbo spin echo [HASTE]) MRI provides very rapid T2-weighted images that are also ideally suited for evaluation of bowel disease.
On these images, intestinal fluid or orally administered water-soluble contrast appears bright, thus highlighting the adjacent bowel wall. In our practice, the combination of SSFSE T2-weighted images and breath-hold gadolinium-enhanced SPGR images with fat suppression is more sensitive than helical CT in depicting benign and malignant gastrointestinal tract disease. In a recent comparison of helical CT and MRI in patients with surgical evidence of disease, MRI depicted 24 (88%) of 29 sites of primary or metastatic intestinal tract tumor compared to 15 (74%) depicted on helical CT.
The MRI technique used for gastrointestinal disease is the same as that used for peritoneal tumors. Bowel distention is achieved with three bottles of 2% barium sulfate and rectal water. Glucagon (1 mg IV) is administered with the gadolinium to eliminate bowel peristalsis.
The most useful sequences are the single-shot RARE (SSFSE or HASTE) MRIs, which are breathing independent T2-weighted images, and the fat-suppressed, breath-hold SPGR MRIs following IV gadolinium injection. Images are obtained on the axial and coronal planes.
For depiction of osseous metastases, MRI offers a real and significant advantage over helical CT. The sensitivity of MRI for bone metastases or marrow infiltration in patients with lymphoma or leukemia is far superior to that of helical CT and is often better than a nuclear medicine bone scan.
Osseous metastases may be visualized on unenhanced T1-weighted and T2-weighted MRIs. However, we find that the axial and coronal fat-suppressed, gadolinium-enhanced SPGR MRIs are exquisitely sensitive to enhancing tumor in the spine (Figure 5), bones of the pelvis, and femur. These images are routinely obtained during all abdominal MR examinations. An evaluation of the marrow for focal lesions or for diffuse tumor infiltration is a standard part of every abdominal MR examination.
Our oncologists have come to use abdominal MRI as a screening examination for hepatic and extrahepatic tumors. As the primary imaging examination used in an oncology patient, we depend on MRI for accurately detecting and characterizing all tumor in the abdomen and pelvis.
We recently compared the effectiveness of MRI and helical CT for depicting extrahepatic tumor in 57 oncology patients at two institutions that had surgical proof of extrahepatic tumor. Helical CT scans depicted 101 (66%) of 154 surgically confirmed extrahepatic tumors, compared to 139 tumors (90%; P < .0001) depicted on MRI. In addition, MRI detected more patients with tumor at 11 of the 17 anatomic sites, with MRI and helical CT uncovering an equivalent number at the remaining six sites. Anatomic sites at which MRI showed a significantly greater rate of detection of extrahepatic tumor included the peritoneum, bowel, and mesentery. In our continuing experience, patients with osseous metastases are also better imaged with MRI than with helical CT.
The MRI techniques described in this article are readily applicable to any 1.5-T imager. In fact, many of the MR examinations at one institution were performed with an 8-year old imaging machine. While newer high-performance imagers offer some added capabilities, they are not necessary to produce the images shown in this article. Excellent MRIs can be obtained with most imagers using commercially available pulse sequences and contrast agents.
Abdominal MRI is a powerful tool for evaluating the oncology patient. With its superior contrast resolution, multiplanar capabilities, and almost unlimited versatility, MRI can answer questions left unanswered by helical CT. It can accurately depict tumor extent and organ involvement, thereby positively affecting patient diagnosis and management.
1. Chezmar JL, Rumancik WM, Megibow AJ, et al: Liver and abdominal screening in patients with cancer: CT vs MR imaging. Radiology 168:43-47, 1988.
2. Low RN, Semelka RC, Worawattanakul S, et al: Extrahepatic abdominal imaging in patients with malignancy: Comparison of MR imaging and helical CT with subsequent surgical correlation. Radiology 210:625-632, 1999.
3. Low RN: Contrast agents for MR imaging of the liver. J Magn Reson Imaging 7:56-67, 1997.
4. Low RN: Current uses of gadolinium chelates for magnetic resonance imaging of the liver. Top Magn Reson Imaging 9:141-166, 1998.
5. Heiken JP, Weyman PJ, Lee JKT, et al: Detection of focal hepatic masses: Prospective evaluation with CT, delayed CT, CT during arterial portography, and MR imaging. Radiology 171:47-51, 1989.
6. Semelka RC, Shoenut JP, Kroeker MA, et al: Focal liver disease: Comparison of dynamic contrast-enhanced CT and T2-weighted fat-suppressed, FLASH, and dynamic gadolinium enhanced MR imaging at 1.5 T. Radiology 184:687-694, 1992.
7. 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.
8. Larson RE, Semelka RC, Baglely AS, et al: Hypervascular malignant liver lesions: Comparison of various MR imaging pulse sequences and dynamic CT. Radiology 192:393-399, 1994.
9. Semelka RC, Shoenut JP, Ascher SM, et al: Solitary hepatic metastasis: Comparison of dynamic contrast-enhanced CT and MR imaging with fat-suppressed T2-weighted, breath-hold T1-weighted FLASH, and dynamic gadolinium-enhanced FLASH sequences. J Magn Reson Imaging 4:319-323, 1994.
10. Yamashita Y, Mitsuzaki K, Yi T, et al: Small hepatocellular carcinoma in patients with chronic liver damage: Prospective comparison of detection with dynamic MR imaging and helical CT of the whole liver. Radiology 200:79-84, 1996.
11. Earls JP, Theise ND, Weinreb JC, et al: Dysplastic nodules and hepatocellular carcinoma: Thin-section MR imaging of explanted cirrhotic livers with pathologic correlation. Radiology 201:207-214, 1996.
12. Low RN, Sigeti JS, Francis IR, et al: Evaluation of malignant biliary obstruction: Efficacy of fast multiplanar spoiled gradient-recalled MR imaging vs spin-echo MR, CT, and cholangiography. AJR Am J Roentgenol 162:315-323, 1994.
13. Semelka RC, Shoenut JP, Kroeker MA, et al: Bile duct disease: Prospective comparison of ERCP, CT, and fat-suppressed MRI. Gastrointest Radiol 17:347-352, 1992.
14. Soto JA, Barish MA, Yucel EK, et al: MR cholangiopancreatography: Findings on 3D fast spin-echo imaging. AJR Am J Roentgenol 165:1397-1401, 1995.
15. Megibow AJ, Zhou XH, Rotterdam H, et al: Pancreatic adenocarcinoma: CT vs MR imaging in the evaluation of resectability-report of the radiology diagnostic oncology group. Radiology 195:327-332, 1995.
16. Ichikawa T, Hardome H, Hachiya J, et al: Pancreatic ductal adenocarcinoma: Preoperative assessment with helical CT vs dynamic MR imaging. Radiology 202:655-662, 1997.
17. Sheridan MB, Ward J, Guthrie JA, et al: Dynamic contrast-enhanced MR imaging and dual-phase helical CT in the preoperative assessment of suspected pancreatic cancer: A comparative study with receiver operating characteristic analysis. AJR Am J Roentgenol 173:583-590, 1999.
18. Semelka RC, Shoenut JP, Magro CM, et al: Renal cancer staging: Comparison of contrast-enhanced CT and gadolinium-enhanced fat-suppressed spin-echo and gradient-echo MR imaging. J Magn Reson Imaging 3:597-602, 1993.
19. Dunnick NR, Korobkin M, Francis I: Adrenal Radiology: Distinguishing benign from malignant adrenal masses. AJR Am J Roentgenol 167:861-867, 1996.
20. Outwater EK, Siegelman ES, Tadecki PD, et al: Distinction between benign and malignant adrenal masses: Value of T1-weighted chemical-shift MR imaging. AJR Am J Roentgenol 165:579-583, 1995.
21. Low RN, Barone RM, Lacey C, et al: Peritoneal tumor: MR imaging with dilute oral barium and intravenous gadolinium-containing contrast agents compared with unenhanced MR imaging and CT. Radiology 204:513-520, 1997.
22. Semelka RC, Lawrence PH, Shoenut P, et al: Primary ovarian cancer: Prospective comparison of contrast-enhanced CT and pre- and postcontrast, fat-suppressed MR imaging, with histologic correlation. J Magn Reson Imaging 3:99-106, 1993.
23. Low RN, Saleh F, Song SYT, et al: Treated ovarian cancer: Comparison of MR imaging with serum CA-125 level and physical examination-a longitudinal study. Radiology 11:519-528, 1999.
24. Low RN, Francis IR: MR imaging of the gastrointestinal tract with IV gadolinium and diluted barium oral contrast media compared with unenhanced MR imaging and CT. AJR Am J Roentgenol 169:1051-1059, 1997.
25. Semelka RC, Shoenut JP, Silverman R, et al: Bowel disease: Prospective comparison of CT and 1.5-T pre- and postcontrast MR imaging with T1-weighted fat-suppressed and breath-hold FLASH sequences. J Magn Reson Imaging 1:625-632, 1991.
26. Semelka RC, Gesine J, Kelekis NL, et al: Small bowel neoplastic disease: Demonstration by MRI. J Magn Reson Imaging 6:855-860, 1996.
27. Matsushita M, Oi H, Murakami T, et al: Extraserosal invasion in advanced gastric cancer: Evaluation with MR imaging. Radiology 192:87-91, 1994.