Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Keogan, M. T.
Right arrow Articles by Edelman, R. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Keogan, M. T.
Right arrow Articles by Edelman, R. R.
(Radiology. 2001;220:310-320.)
© RSNA, 2001

State of the Art

Technologic Advances in Abdominal MR Imaging

Mary T. Keogan, MD and Robert R. Edelman, MD

1 From the Department of Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, One Deaconess Rd, Boston, MA 02215 (M.T.K.); and the Department of Radiology, Evanston Northwestern Healthcare, Ill (R.R.E.). Received October 4, 1999; revision requested November 16; final revision received August 8, 2000; accepted August 31. Address correspondence to M.T.K. (e-mail: mkeogan@caregroup.harvard.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 HARDWARE
PULSE SEQUENCES
 EVOLVING CLINICAL APPLICATIONS...
 CONCLUSION
 REFERENCES
 
Magnetic resonance (MR) imaging is finding an ever-growing role in the evaluation of a wide range of conditions in the abdomen. No longer confined to problem solving regarding abnormalities in solid organs, such as the liver and kidneys, MR imaging is increasingly being applied to the evaluation of the pancreatic and biliary ductal systems and even the bowel. Recent technical advances in hardware and software have allowed the acquisition of MR images that are largely free of artifact secondary to bowel peristalsis or respiratory motion; images providing excellent anatomic detail can now be obtained routinely. Faster sequences have reduced image acquisition time, thereby improving patient acceptance and allowing more efficient utilization of machine time. New three-dimensional sequences allow rapid image acquisition, reducing section misregistration and motion artifact while improving multiplanar reformations. The potential of MR imaging to provide functional and anatomic information is intriguing, and new techniques, including diffusion and perfusion imaging, are being evaluated. This review considers the advances in imaging hardware and pulse sequence design that underlie the increasing role of MR imaging in evaluation of the abdomen and discusses evolving clinical applications.

Index terms: Abdomen, MR, 70.12141, 70.12143, 95.12142 • Magnetic resonance (MR), technology, 70.12141, 70.12143, 95.12142 • State of the Art


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
HARDWARE
PULSE SEQUENCES
 EVOLVING CLINICAL APPLICATIONS...
 CONCLUSION
 REFERENCES
 
During the past few years, the use of cross-sectional imaging to assist in the diagnosis of abdominal disorders has grown enormously. This imaging has mostly involved computed tomography (CT), as a direct result of developments in helical scanning technology. Despite its initial promise, magnetic resonance (MR) imaging is less commonly applied in the abdomen than in other anatomic areas, such as the brain and spine, and its effect in abdominal imaging has been far less than that of CT.

Recent technologic advances, however, are propelling MR imaging to a new level of utility with abdominal applications. For instance, the optimization of sequences—such as single-shot rapid acquisition with relaxation enhancement (RARE) (GE Medical Systems, Milwaukee, Wis), half-Fourier RARE (half-acquisition single-shot turbo spin-echo; Siemens Medical Systems, Erlangen, Germany), and true fast imaging with steady-state free precession (FISP) sequences—has led to the implementation of high-quality MR cholangiopancreatography (MRCP), which is becoming an important noninvasive modality for assessment of the hepatobiliary system. Dynamic imaging with a combination of intravenously administered contrast agents (notably gadolinium chelates) and a large–flip angle, fast breath-hold, three-dimensional (3D), gradient-echo acquisition has allowed accurate diagnosis of disorders affecting the abdominal vasculature (Figs 1, 2). Breath-hold 3D acquisitions obtained by using a reduced–flip angle excitation may permit diagnosis of small hepatic tumors with unprecedented accuracy.



View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Coronal 3D contrast-enhanced MR angiogram (repetition time [TR] msec/echo time [TE] msec, 5/2; 40° flip angle; 40 mL gadopentetate dimeglumine) demonstrates a thrombus occluding the main portal vein (short straight arrow), with a patent superior mesenteric vein (long straight arrow) and multiple collateral vessels (curved arrow) arising from the coronary vein and communicating with the esophageal and gastric veins.

 


View larger version (147K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Coronal 3D contrast-enhanced MR angiogram (5/2, 40° flip angle, 40 mL gadopentetate dimeglumine) obtained in a patient with an arteriovenous fistula (presumed congenital) demonstrates the abdominal aorta (thick arrow), superimposed tortuous splenic artery and vein (thin arrows), and early appearance of signal intensity in the portal vein (arrowhead).

 
In this review, we consider recent hardware and software developments that are enabling new MR imaging applications in the abdomen. The clinical applications, advantages, and limitations of these methods are addressed, along with potential developments.


   HARDWARE
TOP
 ABSTRACT
 INTRODUCTION
 HARDWARE
PULSE SEQUENCES
 EVOLVING CLINICAL APPLICATIONS...
 CONCLUSION
 REFERENCES
 
Phased-Array Coils
Compared with a body coil, phased-array coils (containing four to six elements) provide a large (eg, two- to fourfold) improvement in the signal-to-noise ratio. This benefit decreases, but it is still substantial, as one images deeply situated organs such as the pancreas. The increased signal-to-noise ratio permits one to use faster imaging techniques. For instance, breath-hold imaging with gradient-echo or turbo spin-echo methods is feasible with a phased-array coil, but the images may be excessively noisy if only the body coil is used.

A promising new technology for fast imaging exists. Known as partially parallel imaging, the technology uses the intrinsic sensitivity variation of phased-array coils to speed the process of phase encoding. Several approaches are possible, including simultaneous acquisition of spatial harmonics, or SMASH, and sensitivity encoding, or SENSE (1,2). Aside from appropriate phased-array coils, no special hardware is required. The improved rate of data acquisition permits twofold or greater reductions in imaging time and corresponding improvements in spatial resolution along the phase-encoding direction. The effect of this technology in abdominal MR imaging awaits evaluation in clinical trials, but shorter imaging times are likely to be especially beneficial in patients with limited breath-holding ability.

Gradients
Magnetic field–gradient technology has greatly improved during the past few years. The peak gradient amplitude (measured in milliteslas per meter) and rise time to peak amplitude (measured in microseconds) determine the ultimate spatial resolution. State-of-the-art systems now offer peak gradients of 20–40 mT/m, with rise times of 150–300 µsec. Strong, fast gradients with high-duty cycles are needed to perform echo-planar imaging, as well as to obtain the short TRs and TEs used for MR angiography and true FISP imaging. Drawbacks of the newer gradient coil designs include, in some cases, restricted fields of view over which the gradient performance is linear (nonlinear gradients cause geometric image distortions) and high decibel levels that require adequate ear protection.

Magnet Design
Short-bore (reduced-length bore in which the patient is positioned for imaging) high-field magnets reduce claustrophobia while permitting improved access for interventional procedures. Open magnets provide the most direct access for interventions. While most are currently limited to lower field strengths (typically 0.2–0.3 T), superconductive open systems will offer fields as strong as 1 T. The signal-to-noise ratio decreases in proportion to the field strength; with low-field-strength open magnets, one generally must use thicker sections or larger fields of view to compensate, especially with abdominal imaging.

Compared with high-field-strength imaging, low-field-strength imaging requires a longer TE to permit the use of lower-bandwidth sequences (which improves the signal-to-noise ratio). Despite the long TE, susceptibility and chemical shift artifacts are minimal on low-field-strength images. Unfortunately, the long TE reduces the number of sections that can be acquired per breath hold with a given TR. Out-of-phase gradient-echo imaging is feasible with suitable manipulation of the TE, although the required change in TE (and thus the difference in T2* weighting) is much larger than with low-field-strength imaging than at high -field-strength imaging (eg, TE change of {approx}17.2 msec at 0.2 T vs {approx}2.3 msec at 1.5 T).

Fat suppression may cause problems with chemical shift-dependent methods because of the small absolute frequency difference between fat and water at low field strengths. Fat suppression can be achieved by using a short tau inversion-recovery turbo spin-echo sequence, which is particularly useful for musculoskeletal applications. However, the intrinsically low signal-to-noise ratio with this type of imaging is a particular drawback for low-field-strength abdominal imaging. Although high–field strength imaging is preferred for abdominal applications, low-field-strength open systems can be useful for claustrophobic patients and for MR imaging–guided biopsy.


   PULSE SEQUENCES
TOP
 ABSTRACT
 INTRODUCTION
 HARDWARE
PULSE SEQUENCES
EVOLVING CLINICAL APPLICATIONS...
 CONCLUSION
 REFERENCES
 
Liver MR imaging may be helpful in problem solving when initial findings with ultrasonography (US) or CT are inconclusive. Another important indication is maximizing the detection of liver metastases (eg, in a candidate for partial hepatectomy). Other indications include characterization of a suspected cavernous hemangioma, focal nodular hyperplasia, or adenoma and detection of dysplastic nodules and hepatoma in the cirrhotic liver. Numerous approaches for abdominal MR imaging exist, and the optimal approach depends on the particular software and hardware. Depending on the operator preference and system capabilities, breath-hold imaging should replace non–breath-hold imaging with cooperative patients (3).

For nonenhanced breath-hold T1-weighted acquisitions, a spoiled gradient-echo sequence with a short TR and short TE is used. The TE should be chosen to be in phase (eg, 4.6 msec at 1.5 T) so that fat and water signals add rather than cancel. Out-of-phase imaging should also be performed (eg, TE of 2.3 msec at 1.5 T) to detect lipid within hepatocellular neoplasms, fatty infiltration, or fatty sparing (4,5). However, out-of-phase imaging alone is not sufficient because it may reduce the conspicuity of masses in a fatty liver (ie, a fatty liver will lose signal intensity, and a hypointense metastasis may become isointense relative to the surrounding liver parenchyma). For pancreatic imaging, fat suppression greatly improves the conspicuity of the pancreatic parenchyma and pancreatic masses. In patients with pancreatitis, fat suppression may reduce the visibility of stranding within the peripancreatic fat. However, it facilitates the depiction of hyperintense hemorrhagic peripancreatic collections against a background of hypointense peripancreatic fat (6). For non–breath-hold T1-weighted imaging, either a standard spin-echo sequence with a short TR (eg, 400–600-msec) or a short echo train (eg, three-echo) turbo spin-echo sequence with multiple signals acquired is generally used.

For breath-hold T2-weighted imaging, several techniques have been proposed, most of which are variations of the RARE sequence (eg, turbo spin-echo, fast spin-echo sequences). To reduce breath-hold times to an acceptable duration, a long echo train must be used (eg, >20 echoes). Short interecho spacing (eg, 4 msec) is suggested to minimize artifacts due to T2-dependent signal decay that occurs during a long echo train; these sequences test the limits of the gradients but are feasible with state-of-the-art gradient systems. Because such sequences are radio-frequency intensive, it is useful on high-field-strength systems to reduce the refocusing flip angle from the usual value of 180° (eg, to {approx}130°) to avoid specific absorption rate limits. Image quality is not adversely affected. Fat suppression is helpful, since the short interecho spacing reduces J-coupling effects so that fat appears bright. Inversion-recovery imaging provides uniform fat saturation, even with low field strengths. In our experience, a breath-hold inversion-recovery fast spin-echo sequence proved superior to conventional spin-echo sequences in the detection of hepatic tumors at 1.5 T (7).

An alternative non–breath-hold approach to fast T2-weighted imaging is fat-suppressed respiratory-triggered fast spin-echo imaging. Compared with both inversion-recovery and non–fat-suppressed conventional spin-echo sequences, this sequence has been shown (8) to have increased accuracy for liver lesion detection. In a recent study (9) in which four fat-suppressed T2-weighted sequences were compared in the detection of focal hepatic lesions, the fat-suppressed respiratory-triggered fast spin-echo sequence was more accurate than the conventional spin-echo, breath-hold fast spin-echo, and breath-hold multishot spin-echo echo-planar sequences. Depending on the particular system configuration, we suggest that either a respiratory-triggered fast spin-echo sequence or a breath-hold inversion-recovery fast spin-echo sequence is preferable to a conventional spin-echo sequence for T2-weighted hepatic imaging.

T2-weighted sequences have proven useful in the distinction of cysts and hemangiomas from malignant lesions (10). Without breath holding, they are sensitive to blurring and ghost artifacts due to respiration. Sequences sensitive to magnetic susceptibility (eg, T2*-weighted sequences such as gradient-echo sequences) are useful for the detection of hemochromatosis, for imaging of focal liver lesions following the injection of superparamagnetic particles, and for distinguishing siderotic nodules from hepatomas in patients with cirrhosis (11–13).

Gadolinium-enhanced T1-weighted; nonenhanced; and arterial-, portal-, and delayed-phase contrast material–enhanced imaging have become central to current hepatic MR imaging for lesion characterization. Although controversy exists as to whether enhanced images are more sensitive for the detection of liver metastasis (14–16), enhancement patterns on arterial phase images, in particular, are widely accepted as being important in the detection and characterization of hemangiomas, hepatomas, adenomas, and focal nodular hyperplasia (Fig 3) (15–17).



View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a. Transverse breath-hold T1-weighted MR images (131/4.1, 80° flip angle) obtained in a patient with focal nodular hyperplasia in the left lobe of the liver. (a) Before contrast enhancement. Note the almost isointense lesion (arrowheads) with a hypointense scar (arrow). (b) Arterial phase image (approximately 25 seconds after contrast material injection). Note the early marked enhancement of the lesion with a nonenhancing central scar (arrow). (c) Portal venous phase image (approximately 60 seconds after injection). Note that the lesion is almost isointense to the liver, and note the delayed enhancement of the central scar (arrow).

 


View larger version (133K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b. Transverse breath-hold T1-weighted MR images (131/4.1, 80° flip angle) obtained in a patient with focal nodular hyperplasia in the left lobe of the liver. (a) Before contrast enhancement. Note the almost isointense lesion (arrowheads) with a hypointense scar (arrow). (b) Arterial phase image (approximately 25 seconds after contrast material injection). Note the early marked enhancement of the lesion with a nonenhancing central scar (arrow). (c) Portal venous phase image (approximately 60 seconds after injection). Note that the lesion is almost isointense to the liver, and note the delayed enhancement of the central scar (arrow).

 


View larger version (138K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3c. Transverse breath-hold T1-weighted MR images (131/4.1, 80° flip angle) obtained in a patient with focal nodular hyperplasia in the left lobe of the liver. (a) Before contrast enhancement. Note the almost isointense lesion (arrowheads) with a hypointense scar (arrow). (b) Arterial phase image (approximately 25 seconds after contrast material injection). Note the early marked enhancement of the lesion with a nonenhancing central scar (arrow). (c) Portal venous phase image (approximately 60 seconds after injection). Note that the lesion is almost isointense to the liver, and note the delayed enhancement of the central scar (arrow).

 
Timing of image acquisition is critically important for arterial phase imaging and may be inaccurate if fixed image delays that do not allow for patient variables, such as cardiac output, are used. There are several approaches for timing the start of data acquisition with respect to the administration of contrast material. These include the administration of a timing bolus of 1–2 mL of gadopentetate dimeglumine (18), automatic triggering based on the detection of vessel enhancement with a tracker region (Smartprep, GE Medical Systems; Carebolus, Siemens Medical Systems) (19), and fluoroscopic imaging with manual triggering of the 3D acquisition. Of these, the timing bolus technique is most useful in hepatic imaging.

A 3D T1-weighted spoiled gradient-echo acquisition with gadolinium enhancement (also called volume-interpolated breath-hold examination, or VIBE) minimizes vascular pulsation artifacts and permits the acquisition of thinner (eg, 1–3-mm) sections. With such thin sections, one can better detect and characterize small lesions. Moreover, the data may be reconstructed in various planes to more clearly depict lesion and vascular relationships (Figs 4, 5) (20). Compared with a typical contrast-enhanced 3D MR angiographic sequence, the anatomic sequence requires modifications: (a) A reduced flip angle (eg, 10° vs 30°–60° for MR angiography) causes background tissues, such as liver, to appear hyperintense, and (b) the echo is centered (MR angiography sequences commonly involve an echo that is asymmetrically centered within the sampling window to minimize the TE). Additionally, while fat saturation may not always be necessary for 3D MR angiography (larger flip angles with MR angiography may reduce the signal from fat), fat suppression is always applied with the anatomic sequence.



View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Three-dimensional T1-weighted spoiled gradient-echo MR images (4/1.6, 12° flip angle) obtained in the arterial phase after gadolinium enhancement. Top: Transverse fat-saturated breath-hold image (1-mm section thickness) demonstrates the sharp edge definition of the normal pancreas (P). Note the excellent demonstration of the celiac axis (short arrow) and common hepatic artery (long arrow). Bottom: Curved multiplanar reconstruction provides a different perspective of the pancreas and clearly shows its relationship to the normal superior mesenteric artery (arrow).

 


View larger version (172K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a. (a-c) Transverse 3D T1-weighted spoiled gradient-echo MR images (4/1.6, 12° flip angle) obtained in a patient with hepatocellular carcinoma. (a) Nonenhanced fat-saturated image shows excellent contrast between tumor (arrowheads) and normal liver. Note the high signal intensity (arrow) in the left-sided collecting system, which appeared after a timing bolus was administered. (b) Gadolinium-enhanced arterial phase image shows a heterogeneously enhancing tumor (arrow). (c) Portal-phase gadolinium-enhanced image shows diffuse tumor enhancement (arrow). (d) Transverse 90-second-delayed gradient-echo image (130/4.6, no fat saturation) shows delayed enhancement of the tumor capsule (arrows). (e) Maximum intensity projection reconstructed in an oblique transverse plane (1.5-cm slab) shows the arterial supply to the tumor (arrow) from the right hepatic artery (arrowheads).

 


View larger version (154K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b. (a-c) Transverse 3D T1-weighted spoiled gradient-echo MR images (4/1.6, 12° flip angle) obtained in a patient with hepatocellular carcinoma. (a) Nonenhanced fat-saturated image shows excellent contrast between tumor (arrowheads) and normal liver. Note the high signal intensity (arrow) in the left-sided collecting system, which appeared after a timing bolus was administered. (b) Gadolinium-enhanced arterial phase image shows a heterogeneously enhancing tumor (arrow). (c) Portal-phase gadolinium-enhanced image shows diffuse tumor enhancement (arrow). (d) Transverse 90-second-delayed gradient-echo image (130/4.6, no fat saturation) shows delayed enhancement of the tumor capsule (arrows). (e) Maximum intensity projection reconstructed in an oblique transverse plane (1.5-cm slab) shows the arterial supply to the tumor (arrow) from the right hepatic artery (arrowheads).

 


View larger version (154K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5c. (a-c) Transverse 3D T1-weighted spoiled gradient-echo MR images (4/1.6, 12° flip angle) obtained in a patient with hepatocellular carcinoma. (a) Nonenhanced fat-saturated image shows excellent contrast between tumor (arrowheads) and normal liver. Note the high signal intensity (arrow) in the left-sided collecting system, which appeared after a timing bolus was administered. (b) Gadolinium-enhanced arterial phase image shows a heterogeneously enhancing tumor (arrow). (c) Portal-phase gadolinium-enhanced image shows diffuse tumor enhancement (arrow). (d) Transverse 90-second-delayed gradient-echo image (130/4.6, no fat saturation) shows delayed enhancement of the tumor capsule (arrows). (e) Maximum intensity projection reconstructed in an oblique transverse plane (1.5-cm slab) shows the arterial supply to the tumor (arrow) from the right hepatic artery (arrowheads).

 


View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5d. (a-c) Transverse 3D T1-weighted spoiled gradient-echo MR images (4/1.6, 12° flip angle) obtained in a patient with hepatocellular carcinoma. (a) Nonenhanced fat-saturated image shows excellent contrast between tumor (arrowheads) and normal liver. Note the high signal intensity (arrow) in the left-sided collecting system, which appeared after a timing bolus was administered. (b) Gadolinium-enhanced arterial phase image shows a heterogeneously enhancing tumor (arrow). (c) Portal-phase gadolinium-enhanced image shows diffuse tumor enhancement (arrow). (d) Transverse 90-second-delayed gradient-echo image (130/4.6, no fat saturation) shows delayed enhancement of the tumor capsule (arrows). (e) Maximum intensity projection reconstructed in an oblique transverse plane (1.5-cm slab) shows the arterial supply to the tumor (arrow) from the right hepatic artery (arrowheads).

 


View larger version (160K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5e. (a-c) Transverse 3D T1-weighted spoiled gradient-echo MR images (4/1.6, 12° flip angle) obtained in a patient with hepatocellular carcinoma. (a) Nonenhanced fat-saturated image shows excellent contrast between tumor (arrowheads) and normal liver. Note the high signal intensity (arrow) in the left-sided collecting system, which appeared after a timing bolus was administered. (b) Gadolinium-enhanced arterial phase image shows a heterogeneously enhancing tumor (arrow). (c) Portal-phase gadolinium-enhanced image shows diffuse tumor enhancement (arrow). (d) Transverse 90-second-delayed gradient-echo image (130/4.6, no fat saturation) shows delayed enhancement of the tumor capsule (arrows). (e) Maximum intensity projection reconstructed in an oblique transverse plane (1.5-cm slab) shows the arterial supply to the tumor (arrow) from the right hepatic artery (arrowheads).

 
In our experience, a combination of the following sequences has proven effective at 1.5 T, in terms of throughput and accuracy: transverse breath-hold inversion-recovery turbo spin-echo, transverse in- and out-of-phase precontrast T1-weighted gradient-echo, coronal half-Fourier RARE, and transverse postcontrast breath-hold fat-suppressed 3D MR sequences. The entire study can be completed within 30 minutes, including less than 5 minutes of imaging time (Table).


View this table:
[in this window]
[in a new window]

  		Sample Protocol for Hepatic MR Imaging with a Breath-hold Technique at 1.5 T

 
Subsecond Imaging
Subsecond imaging with a T1-weighted inversion-recovery rapid gradient-echo sequence (snapshot fast low-angle shot or turbo fast low-angle shot sequence; Siemens Medical Systems) is notable for lack of pulsation artifact and motion artifact (21,22). Because spatial resolution is inferior, this sequence should not replace the more standard breath-hold T1-weighted sequences (fast low-angle shot or spoiled multiplanar gradient-recalled acquisition in the steady state; GE Medical Systems), except when the patient’s condition renders conventional imaging impractical.

Although multishot RARE imaging (ie, that involving multiple radio-frequency excitations) as described previously permits high-definition imaging of the abdomen, it is sensitive to motion artifacts. Single-shot echo-planar imaging permits a T2-weighted image to be acquired in as short as one-tenth of a second. Although echo-planar imaging of the abdomen is feasible and may be useful for hepatic imaging and experimental applications such as diffusion imaging, the presence of distortions due to magnetic susceptibility variations and imperfect fat suppression limits its routine use. By reducing the echo train length with a multishot echo-planar sequence, diagnostically useful abdominal images can be obtained with much less chemical shift and fewer susceptibility artifacts (23).

Chemical shift and susceptibility artifacts are negligible with RARE acquisitions. With a single-shot RARE acquisition, a long TE is chosen (on the order of 0.5–1.0 second) so that only fluids are depicted. Since there is little background signal, one can acquire a single image of a thick section (eg, 20–100 mm) that demonstrates a large proportion of the hepatobiliary system or urinary tract.

Because data can be acquired in less than a second (or, in some implementations, in less than 0.5 second), the partial-Fourier subsecond RARE sequence (half-Fourier RARE or single-shot fast spin-echo sequence) is helpful in patients who are unable to hold their breath or when a rapid survey is needed. Compared with RARE, half-Fourier RARE can have a shorter TE (typically on the order of 20–80 msec) because the center of the k space is acquired near the start of the echo train rather than in the middle (24–27). Compared with single-shot RARE, drawbacks include blurring in the phase-encoding direction caused by the partial Fourier reconstruction and the inability to use thick sections due to partial volume averaging. The sequence is also not as sensitive as standard breath-hold T2-weighted sequences for small low-contrast lesions. Nonetheless, half-Fourier RARE permits a time-efficient survey of the entire abdomen in multiple section planes. In an uncooperative patient, this can be enormously helpful (Fig 6). It also can be used to display tissue relationships that are difficult to see on a single-shot RARE image because of the lack of signal from background tissue.



View larger version (161K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Coronal half-Fourier RARE MR image (4.4/64) obtained in an immunocompromised patient with right-sided colitis (secondary to cytomegalovirus infection) demonstrates mural thickening and intramural high signal intensity (arrow) in the ascending colon; this finding is consistent with diffuse edema. High-signal-intensity pericolonic and perihepatic fluid secondary to ascites (arrowheads) is noted.

 
With various implementations of the RARE sequence, MRCP has been studied in a wide range of benign and malignant diseases of the biliary tree and has been shown to have accuracy comparable to that of endoscopic retrograde cholangiopancreatography (ERCP) (24,28). The most well established role for MRCP is in the detection of biliary calculi. Findings of several studies (24,28) have shown that the sensitivity of MRCP for the diagnosis of choledocholithiasis is close to 98%, which is comparable with that of ERCP, although both techniques may fail to depict or cause misinterpretation of small calculi. Because MRCP does not offer the option of stone extraction, the major role for this technique has been as a screening test when the clinical suspicion of calculi is low or when contraindications to ERCP exist. Other applications include use with suspicion of common duct calculi when the ducts are not dilated and US findings are negative, in the preoperative detection of aberrant ductal anatomy (29), in the evaluation of biliary ductal obstruction (30), in patients in whom ERCP has failed or in whom the postsurgical bowel anatomy precludes endoscopic access (31,32), and in the surveillance of patients with sclerosing cholangitis (33,34).

True FISP
Ultrafast imaging of the vascular and hepatobiliary systems can be performed by using the true FISP sequence (35,36). The true FISP sequence is a gradient-echo sequence with a balanced structure that compensates first-order phase shifts produced by flow or other kinds of motion. The TR and TE are kept as short as possible (eg, 3 and 1 msec, respectively) to minimize motion and susceptibility artifacts; this sequence requires the maximum capabilities of the gradient system. The image contrast is related to the T2*-T1 ratio. Tissues with a high ratio, such as bile, blood, and fat, appear bright. With the short TRs now possible, images are acquired in less than 1 second. In a single breath hold, one can survey the entire abdominal portal and systemic venous systems (Fig 7). True FISP is much more efficient than two-dimensional time-of-flight MR angiography (which permits the acquisition of only one or, at most, a few sections with each breath hold). Unlike with 3D MR angiography, contrast agent administration is not needed and, in fact, has only a minimal effect on tissue contrast. The bile and pancreatic ducts are also well shown (Fig 7). Drawbacks of the method include bandlike artifacts due to off-resonance effects and flow artifacts, which may be seen within the aorta and other large arteries (35). Both kinds of artifacts are reduced as the TR is shortened. In our experience, this sequence is of considerable value in patients who have difficulty with breath holding and when a motion-insensitive method is needed. True FISP may also be of value for both needle localization and therapy imaging assessment during interventional procedures performed with low field strengths (36).



View larger version (199K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7a. (a, b) Coronal true FISP MR images (6/3, 70° flip angle) obtained in a patient with right-sided renal carcinoma. (a) Note the large tumor (thick straight solid arrow) with mixed signal intensity, area of abnormally low signal intensity that represents a thrombus in the vena cava (thin straight solid arrow), and peritumoral collateral vessels (curved arrow). Note also the normal high signal intensity in the portal vein (open arrows) and normal jejunal loops (arrowheads). (b) Anterior image demonstrates the pancreatic body (short straight arrow), high-signal-intensity normal pancreatic duct (long straight arrow), superior mesenteric vein (arrowhead), and multiple low-signal-intensity gallstones (curved arrow). (c) Transverse contrast-enhanced T1-weighted MR image (131/4.1, 80° flip angle) demonstrates the large tumor (thick arrow) and low-signal-intensity thrombus within the inferior vena cava (thin arrow).

 


View larger version (201K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7b. (a, b) Coronal true FISP MR images (6/3, 70° flip angle) obtained in a patient with right-sided renal carcinoma. (a) Note the large tumor (thick straight solid arrow) with mixed signal intensity, area of abnormally low signal intensity that represents a thrombus in the vena cava (thin straight solid arrow), and peritumoral collateral vessels (curved arrow). Note also the normal high signal intensity in the portal vein (open arrows) and normal jejunal loops (arrowheads). (b) Anterior image demonstrates the pancreatic body (short straight arrow), high-signal-intensity normal pancreatic duct (long straight arrow), superior mesenteric vein (arrowhead), and multiple low-signal-intensity gallstones (curved arrow). (c) Transverse contrast-enhanced T1-weighted MR image (131/4.1, 80° flip angle) demonstrates the large tumor (thick arrow) and low-signal-intensity thrombus within the inferior vena cava (thin arrow).

 


View larger version (144K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7c. (a, b) Coronal true FISP MR images (6/3, 70° flip angle) obtained in a patient with right-sided renal carcinoma. (a) Note the large tumor (thick straight solid arrow) with mixed signal intensity, area of abnormally low signal intensity that represents a thrombus in the vena cava (thin straight solid arrow), and peritumoral collateral vessels (curved arrow). Note also the normal high signal intensity in the portal vein (open arrows) and normal jejunal loops (arrowheads). (b) Anterior image demonstrates the pancreatic body (short straight arrow), high-signal-intensity normal pancreatic duct (long straight arrow), superior mesenteric vein (arrowhead), and multiple low-signal-intensity gallstones (curved arrow). (c) Transverse contrast-enhanced T1-weighted MR image (131/4.1, 80° flip angle) demonstrates the large tumor (thick arrow) and low-signal-intensity thrombus within the inferior vena cava (thin arrow).

 

   EVOLVING CLINICAL APPLICATIONS AND TECHNIQUES
TOP
 ABSTRACT
 INTRODUCTION
 HARDWARE
PULSE SEQUENCES
 EVOLVING CLINICAL APPLICATIONS...
CONCLUSION
 REFERENCES
 
Cholecystitis
MR imaging is rarely used for the evaluation of patients presenting with right upper quadrant pain, because these patients are traditionally evaluated with US. However, the MR imaging findings for acute cholecystitis have recently been described in a small patient group (37). With the half-Fourier RARE sequence, a sensitivity of 91% and a specificity of 79% for acute cholecystitis was noted, based on the detection of pericholecystic high-signal-intensity fluid, with good interobserver agreement. In addition, the accuracy of half-Fourier RARE imaging was greater than that of US for the detection of common duct stones, although MR imaging has the disadvantage that the Murphy sign cannot be assessed. To our knowledge, the potential role of MR imaging versus CT and radionuclide scanning has not been studied to date.

Sclerosing Cholangitis
An evolving indication for MRCP is the evaluation of patients with sclerosing cholangitis (38–40). These patients are at increased risk for cholangiocarcinoma and biliary calculi, but they have an increased rate of serious complications (including ascending cholangitis) following ERCP. As resolution of smaller ducts has improved with sequences such as single-shot RARE, MRCP now offers an alternative to ERCP. A relative drawback of MRCP is that it cannot be used to distend the intrahepatic ducts, as ERCP can; therefore, subtle ductal changes may not be detected. However, MRCP may have in a role as a follow-up or surveillance study once the diagnosis has been made. Its use in this way may reduce the need for multiple ERCP studies.

Hepatic parenchymal changes have been reported with the use of conventional (non-MRCP) sequences in patients with sclerosing cholangitis. Peripheral wedge-shaped areas of high signal intensity (on both T1- and T2-weighted images) that may be related to underlying perfusion changes and dilated bile ducts (likely related to bile duct inflammation) may be seen on nonenhanced images (41,42). Contrast-enhanced MR imaging may be useful in patients with sclerosing cholangitis to both demonstrate peripheral parenchymal enhancement and, with the addition of delayed imaging, improve detection of cholangiocarcinoma (42).

Pancreatitis
In patients suspected of having early chronic pancreatitis, demonstration of side-branch changes may be important and may necessitate ERCP. Techniques to increase the caliber of branches and thereby improve their visualization at MR imaging have been evaluated. Matos et al (43) have reported the usefulness of dynamic MR with a single-shot turbo spin-echo sequence following intravenous administration of secretin (to increase exocrine pancreatic secretion) in the evaluation of side branches or papillary stenosis (Fig 8). This group subsequently described progressive enhancement of the parenchyma following secretin stimulation as an insensitive but specific sign of early chronic pancreatitis (44). A disadvantage of this approach is the limited supply and cost of secretin, at least in the United States.



View larger version (161K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8a. Coronal RARE MR images (2,800/1,100, 20-mm section thickness). (a) Image obtained before the administration of secretin shows high signal intensity within the gallbladder (GB), stomach (S), and duodenum (D). The common bile duct (short arrow) is seen, but the main pancreatic duct is seen only in the pancreatic head (long arrow) on this thick section. The accessory duct (arrowhead) is barely seen. (b) Two minutes after the intravenous injection of secretin, the pancreatic duct (short arrows) is distended and well depicted in the pancreatic body; it inserts into the major papilla (long arrow). The accessory duct is now seen in the pancreatic head (arrowhead). Note increased fluid in the jejunum (J).

 


View larger version (167K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8b. Coronal RARE MR images (2,800/1,100, 20-mm section thickness). (a) Image obtained before the administration of secretin shows high signal intensity within the gallbladder (GB), stomach (S), and duodenum (D). The common bile duct (short arrow) is seen, but the main pancreatic duct is seen only in the pancreatic head (long arrow) on this thick section. The accessory duct (arrowhead) is barely seen. (b) Two minutes after the intravenous injection of secretin, the pancreatic duct (short arrows) is distended and well depicted in the pancreatic body; it inserts into the major papilla (long arrow). The accessory duct is now seen in the pancreatic head (arrowhead). Note increased fluid in the jejunum (J).

 
Bowel Imaging
Recently, interest has focused on the use of heavily T2-weighted sequences, such as half-Fourier RARE, that can be used to display high-signal-intensity fluid within dilated bowel loops without the use of ingested contrast agents. This technique may be particularly useful for the evaluation of bowel obstruction. Standard imaging techniques for the evaluation of potential small-bowel obstruction include plain radiography; contrast-enhanced studies; and, increasingly, CT (45). MR imaging has been much less frequently used in the past because, in addition to limitations due to expense and availability, the spin-echo sequences typically used were limited by substantial motion artifacts resulting from breathing and peristalsis. The few studies of small-bowel obstruction performed with these sequences also involved the use of air insufflation, muscle relaxants, and intravenous administration of contrast material, which substantially add to the complexity and time required for the study (46).

The half-Fourier RARE sequence provides good image quality with minimal motion artifact. Moreover, one can image repetitively with a half-Fourier RARE–type sequence to view bowel peristalsis. Both normal and abnormal bowel loops and mucosal folds are well demonstrated (Fig 5). For the diagnosis of small-bowel obstruction, half-Fourier RARE has a sensitivity of 90% and specificity of 85%, and it has allowed correct diagnosis of the level of obstruction in 73% of cases (47). The cause of obstruction is more difficult to define, but as with a contrast-enhanced study, adhesive obstruction is more likely in the setting of sharply angulated bowel loops, while conditions such as Crohn disease may be suggested by typical distribution of areas of wall thickening. With the intravenous administration of a gadolinium chelate, areas of active inflammation can be identified, and it has been suggested that transmural inflammation due to Crohn disease may be differentiated from limited submucosal involvement in ulcerative colitis (48).

Detailed anatomic and functional information about the small bowel has recently been obtained by using the technique of MR enteroclysis (49). In this study (which included patients with inflammatory bowel disease or small-bowel obstruction), the small bowel was distended by using a methyl cellulose-water solution. A single-shot fast spin-echo T2-weighted sequence was used for dynamic imaging, while anatomic information was obtained by using a combination of breath-hold T1- and T2-weighted sequences before and after the intravenous administration of a gadolinium-based contrast agent. The MR study was well tolerated, and thus resulted in good small-bowel distention, and the diagnostic information showed good correlation with findings at surgery and conventional enteroclysis.

Virtual endoscopy of the colon with the use of CT is becoming an established technique both in screening for polyps and in the evaluation of the bowel after incomplete colonoscopy (50,51). Initial reports suggest that MR imaging can also provide virtual colonoscopic images that may aid polyp detection (52). The technique described uses a breath-hold 3D spoiled gradient-recalled-echo sequence with the patient in the prone and supine positions after rectal administration of a highly dilute solution of gadopentetate dimeglumine (Fig 9). Additionally, to assess contrast enhancement characteristics, a two-dimensional multiplanar spoiled gradient-recalled-echo sequence was also performed before and after the intravenous administration of gadopentetate dimeglumine. Findings of a recent study (52) of MR colonography in 132 patients showed that while this technique is limited in the detection of small lesions (<=5 mm in diameter), 26 of 27 large (>10-mm) lesions were detected.



View larger version (167K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 9. Coronal T1-weighted 3D gradient-echo MR colonogram (5/2, 40° flip angle) obtained in a patient with cecal carcinoma. Abdominal image obtained after instillation of dilute gadolinium-based contrast material reveals a low-signal-intensity mass in the cecum (arrow); cecal adenocarcinoma was confirmed at resection. Note an out-of-phase effect secondary to the short TE, which resulted in reduced conspicuity of fat. (Image courtesy of Martina M. Morrin, MD, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Mass.)

 
Optimization of MR colonography depends not only on the sequence but also on the composition of enema. Enemas should be biocompatible and should be able to distend and homogeneously fill the colon. In addition, homogeneously increased and decreased signal intensity on T1- and T2-weighted images, respectively, is necessary. While gadolinium chelates have desirable signal characteristics for the imaging of bowel lesions, problems include low viscosity that causes reflux into the small bowel and the requirement for high volumes (as much as 40 mL of a 0.5 mol/L formulation), with the associated expense (53). An enema preparation consisting of iron glycerophosphate (with the addition of 0.8% wt/wt cellulose to increase viscosity) has been proposed as a cheaper alternative (49). Alternatively, one can perform imaging by instilling air into the colon, as with CT colonoscopy. This technique is advantageous because is has no cost and because rectal instillation of air is better tolerated than rectal instillation of fluid. One then performs imaging by using a half-Fourier RARE sequence, which makes a polyp appear bright in comparison with the dark air (54). Further validation is required before any method of MR colonoscopy can be considered for routine clinical use.

Functional information regarding bowel activity is a recent area of MR research activity. An MR imaging method has been developed to simultaneously measure gastric emptying and motility (55). After ingestion of a standardized test meal (labeled with gadoterate meglumine to provide positive contrast enhancement), a turbo spin-echo sequence was used to determine volume, and a coronal dynamic study (fast field-echo study) was used to determine motility. This technique allows tracking of a contraction throughout the stomach and allows determination of the frequency, propagation speed, and amplitude of each contraction. This technique offers a noninvasive and accurate method for the investigation of a range of disorders, including diabetic gastroparesis and pyloric stenosis.

Perfusion Imaging
Perfusion MR imaging offers a noninvasive measure of tissue function (56). Perfusion images are typically acquired at 1–2-second intervals following a bolus injection of contrast agent. Depending on the application, either a rapid T1- or T2*-weighted acquisition can be used. Perfusion imaging has been previously applied to the central nervous system (eg, in patients with cerebral ischemia) and heart (eg, for the detection of myocardial ischemia) (57–59). Recently, there has been interest in potential abdominal applications (60–62). In the liver, perfusion imaging has been used to characterize lesions (63). With a single-shot echo-planar technique (T2*-weighted imaging), gadolinium-based materials serve as negative contrast agents; the result is that enhancing lesions have lower signal intensity as the contrast agent accumulates. Characteristic perfusion patterns are seen in hepatocellular carcinoma, metastasis, and hemangioma. Furthermore, using a bolus injection of superparamagnetic iron oxide, investigators (64) have shown decreases in signal intensity that are substantially different for various histologic subtypes of hepatocellular carcinoma (poorly, moderately, and well differentiated). Perfusion imaging may be useful in the evaluation of tumor microvasculature in the context of new antiangiogenic agents that are being studied in clinical trials (65).

In the kidney, perfusion imaging (with a T1-weighted turbo fast low-angle shot sequence) has shown acute focal changes in renal hemodynamics in patients who have undergone extracorporeal shockwave lithotripsy. Investigators (66) have hypothesized that shunting of blood flow from the cortex to the medulla in the treated area may represent a protective mechanism against medullary ischemia. Prasad et al (67) have shown that perfusion imaging with the use of this T1-weighted sequence and captopril may be of value in assessing the functional importance of renal artery stenosis. Also, MR perfusion imaging has recently been described (68) as a novel technique that may be used to objectively demonstrate the preservation (or potential improvement) of myometrial perfusion in patients undergoing uterine artery embolization as a treatment for fibroids.

Diffusion Imaging
Diffusion imaging of the brain is routinely used for early detection and characterization of stroke (69). Because of its exquisite sensitivity to motion, diffusion imaging of the abdomen had, until recently, problems. However, good results have been obtained with single-shot echo-planar sequences (70–73). The various abdominal organs have unique diffusion characteristics, as measured with the apparent diffusion coefficient. For instance, the spleen has a small apparent diffusion coefficient, indicating reduced water mobility, whereas the renal cortex has a large apparent diffusion coefficient. Renal diffusion imaging is of potential interest because of the high blood flow and water transport functions of the kidney. Changes in the apparent diffusion coefficient have been documented in work with animals in conditions such as renal artery stenosis and renal obstruction (74).

The clinical utility of diffusion imaging in the abdomen is not yet clear, although potential applications in the characterization of tumors and infiltrative diseases exist. Differences in the mean apparent diffusion coefficients have been demonstrated for hepatocellular carcinoma, metastasis, and hemangiomas; these were all higher than that of normal liver (71). Single-shot echo-planar imaging has been evaluated with and without a small diffusion-sensitizing gradient, and it has been compared with breath-hold T2-weighted fast spin-echo sequences (72). This study demonstrated better lesion-to-liver signal intensity with echo-planar imaging than with fast spin-echo imaging; the use of the diffusion-sensitizing gradient caused suppression of high signal intensity from vessels and resulted in better conspicuity of small lesions. This sequence depicted more metastatic deposits, compared with T2-weighted fast spin-echo sequences.

Diffusion-weighted echo-planar imaging may be helpful in the differentiation of mucin-producing pancreatic tumors from other entities (73). By using a diffusion-sensitizing gradient of 30 sec/mm2, the mean apparent diffusion coefficient of viscous fluid (in both the cystic cavity and main pancreatic duct) in mucin-producing tumors was substantially lower than that of serous tumors and cerebrospinal fluid; however, it was not lower than that of pseudocysts. Because pseudocysts do not have associated viscous material within the main duct, calculation of an apparent diffusion coefficient within the cyst and duct may be helpful in the differentiation of these cystic lesions. Unfortunately, the limited spatial resolution with echo-planar imaging precludes evaluation of small masses. For instance, small mucinous tumors confined to side branches may not be depicted.

Blood Oxygen Level–Dependent Imaging
Functional imaging of the kidney can be performed by using a blood oxygen level–dependent (BOLD) approach similar to that used for functional brain imaging. The BOLD technique uses a heavily T2*-weighted sequence to depict changes in blood oxygenation. Because the renal medulla normally functions in a hypoxic state, the BOLD signal is normally low. The signal is increased with pharmacologic agents, such as acetazolamide, or with water loading, which improves medullary oxygenation (74,75). The BOLD response is reduced in older persons and after the administration of nonsteroidal antiinflammatory drugs that inhibit prostaglandin synthesis. This method might have value in the prediction of the risk of acute tubular necrosis; however, further studies are required.

The BOLD signal response of the superior mesenteric vein after a meal may be of clinical value in the diagnosis of chronic mesenteric ischemia. In a healthy individual, the BOLD signal remains stable or increases after a meal because the greater metabolic demand is balanced by an increase in arterial blood flow, which provides sufficient oxygen to the tissues. However, with mesenteric ischemia, the blood flow response is blunted. The resultant ischemic bowel increases the quantity of deoxyhemoglobin (which has a reduced BOLD signal) in the superior mesenteric vein (76). One can also directly assess the patency of the mesenteric vessels by using gadolinium-enhanced 3D MR angiography (76).


   CONCLUSION
TOP
 ABSTRACT
 INTRODUCTION
 HARDWARE
PULSE SEQUENCES
 EVOLVING CLINICAL APPLICATIONS...
 CONCLUSION
REFERENCES
 
In conclusion, the technology available for MR imaging of the abdomen is improving at a dramatic rate. The combination of phased-array coils, faster magnetic-field gradients, contrast agents, and partially parallel imaging techniques may eventually provide spatial resolution that equals or exceeds that available with multi–detector row helical CT, with potentially superior detection and characterization of abdominal disease. The improvements in technology have enabled entirely new applications, such as MRCP. The technology for functional imaging of the abdomen is maturing, but its clinical validation is still limited. With ongoing improvements in technology and, hopefully, with more experience in large-scale multicenter clinical trials, the utility of MR imaging for imaging of the abdomen will be validated and will continue to improve.


   FOOTNOTES
 
Abbreviations: BOLD = blood oxygen level dependent, ERCP = endoscopic retrograde cholangiopancreatography, FISP = fast imaging with steady-state free precession, MRCP = MR cholangiopancreatography, RARE = rapid acquisition with relaxation enhancement, TE = echo time, TR = repetition time, 3D = three-dimensional


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 HARDWARE
PULSE SEQUENCES
 EVOLVING CLINICAL APPLICATIONS...
 CONCLUSION
 REFERENCES
 

  1. Sodickson DK, Griswold MA, Jakob PM. SMASH imaging. Magn Reson Imaging Clin N Am 1999; 7:237-254.[Medline]
  2. Pruessman KP, Weiger M, Scheidegger Boesiger P. SENCE: sensitivity encoding for fast MRI. Magn Reson Med 1999; 42:952-962.[CrossRef][Medline]
  3. Siegelman ES, Outwater EK. MR imaging techniques of the liver. Radiol Clin North Am 1998; 36:263-284.[CrossRef][Medline]
  4. Kreft BP, Tanimoto A, Baba Y, et al. Diagnosis of fatty liver with MR imaging. J Magn Reson Imaging 1992; 2:463-471.[Medline]
  5. Levenson H, Greensite F, Hoefs J, et al. Fatty infiltration of the liver: quantification with phase-contrast MR imaging at 1.5 T vs biopsy. AJR Am J Roentgenol 1991; 156:307-312.[Abstract/Free Full Text]
  6. Lecesne R, Taourel P, Bret PM, Atri M, Reinhold C. Acute pancreatitis: interobserver agreement and correlation of CT and MR cholangiopancreatography with outcome. Radiology 1999; 211:727-735.[Abstract/Free Full Text]
  7. Gaa J, Hatabu H, Jenkins RL, Finn JP, Edelman RR. Liver masses: replacement of conventional T2-weighted spin-echo MR imaging with breath-hold MR imaging. Radiology 1996; 200:459-464.[Abstract/Free Full Text]
  8. Keogan MT, Spritzer CE, Paulson EK, et al. Liver MR imaging: comparison of respiratory triggered fast spin echo with T2-weighted spin-echo and inversion recovery. Abdom Imaging 1996; 21:433-439.[CrossRef][Medline]
  9. Kanematsu M, Hoshi H, Itoh K, et al. Focal hepatic lesion detection: comparison of four fat-suppressed T2-weighted MR imaging pulse sequences. Radiology 1999; 211:363-371.[Abstract/Free Full Text]
  10. Ito K, Mitchell DG, Outwater EK, Szkaruk J, Sadek AG. Hepatic lesions: discrimination of nonsolid, benign lesions from solid, malignant lesions with heavily T2-weighted fast spin-echo MR imaging. Radiology 1997; 204:729-737.[Abstract/Free Full Text]
  11. Bellin MF, Zaim S, Auberton E, et al. Liver metastases: safety and efficacy of detection with superparamagnetic iron oxide in MR imaging. Radiology 1994; 193:657-663.[Abstract/Free Full Text]
  12. Mitchell DG. Hepatobiliary contrast material: a magic bullet for sensitivity and specificity?. Radiology 1993; 188:21-22.[Free Full Text]
  13. Thomsen C, Wiggers P, Ring-Larsen H, et al. Identification of patients with hereditary haemochromotoses by magnetic resonance imaging and spectroscopic relaxation time measurements. Magn Reson Imaging 1992; 10:867-879.[CrossRef][Medline]
  14. Blakeborough A, Ward J, Wilson D, et al. Hepatic lesion detection at MR imaging: a comparative study with four sequences. Radiology 1997; 203:759-765.[Abstract/Free Full Text]
  15. Peterson MS, Baron RL, Murakami T, et al. Hepatic malignancies: usefulness of acquisition of multiple arterial and portal venous phase images at dynamic gadolinium-enhanced MR imaging. Radiology 1996; 201:337-345.[Abstract/Free Full Text]
  16. Hamm B, Mahfouz AE, Taupitz M, et al. Liver metastases: improved detection with dynamic gadolinium-enhanced MR imaging. Radiology 1997; 202:677-682.[Abstract/Free Full Text]
  17. Hamm B, Thoeni RF, Gould RG, et al. Focal liver lesions: characterization with nonenhanced and dynamic contrast material–enhanced MR imaging. Radiology 1994; 190:417-423.[Abstract/Free Full Text]
  18. Earls JP, Rofsky NM, DeCorato DR, Krinsky GA, Weinreb JC. Hepatic arterial-phase dynamic gadolinium-enhanced MR imaging: optimization with a test examination and a power injector. Radiology 1997; 202:268-273.[Abstract/Free Full Text]
  19. Foo TKF, Saranathan M, Prince MR, Chenevert TL. Automated detection of bolus arrival and initiation of data acquisition in fast, three-dimensional, gadolinium-enhanced MR angiography. Radiology 1997; 203:275-280.[Abstract/Free Full Text]
  20. Rofsky NM, Lee VS, Laub G, et al. Abdominal MR imaging with a volumetric interpolated breath-hold examination. Radiology 1999; 212:876-884.[Abstract/Free Full Text]
  21. Matthaei D, Haase A, Henrich D, Duhmke E. Fast inversion recovery T1 contrast and chemical shift contrast in high-resolution snapshot FLASH MR images. Magn Reson Imaging 1992; 10:1-6.[CrossRef][Medline]
  22. Ward J, Baudouin CJ, Ridgway JP, Robinson PJ. Magnetic resonance imaging in the detection of focal liver lesions: comparison of dynamic contrast-enhanced turboFLASH and T2 weighted spin echo images. Br J Radiol 1995; 68:463-470.[Abstract]
  23. Butts K, Riederer SJ, Ehman RL, Felmlee JP, Grimm RC. Echo-planar imaging of the liver with a standard MR imaging system. Radiology 1993; 189:259-64.[Abstract/Free Full Text]
  24. Fulcher AS, Turner MA, Capps GW, et al. Half-Fourier rare MR cholangiopancreatography: experience in 300 subjects. Radiology 1998; 207:21-32.[Abstract/Free Full Text]
  25. Tang Y, Yamashita Y, Takahashi M. Ultrafast T2-weighted imaging of the abdomen and pelvis: use of single shot fast spin-echo imaging. J Magn Reson Imaging 1998; 8:384-390.[Medline]
  26. vanHoe L, Bosmans H, Aerts , et al. Focal liver lesions: fast T2-weighted MR imaging with half-Fourier rapid acquisition with relaxation enhancement. Radiology 1996; 201:817-823.[Abstract/Free Full Text]
  27. Regan F, Bohlman ME, Khazan R, Rodriguez R, Schultze-Haakh H. MR urography using half-Fourier acquisition single shot turbo spin-echo (HASTE) in the assessment of ureteric obstruction. AJR Am J Roentgenol 1996; 167:1115-1120.[Abstract/Free Full Text]
  28. Reinhold C, Taourel P, Bret PM, et al. Choledocholithiasis: evaluation of MR cholangiography for diagnosis. Radiology 1998; 209:435-442.[Abstract/Free Full Text]
  29. Taourel P, Bret PM, Reinhold C, Barkun AN, Atri M. Anatomic variants of the biliary tree: diagnosis with MR cholangiopancreatography. Radiology 1996; 199:521-527.[Abstract/Free Full Text]
  30. Lee M, Lee H, Kim M, et al. Extrahepatic biliary diseases: 3D MR cholangiopancreatography compared to endoscopic retrograde cholangiopancreatography. Radiology 1997; 202:663-669.[Abstract/Free Full Text]
  31. Soto JA, Yucel EK, Barish MA, Chuttani R, Ferrucci JT. MR cholangiopancreatography after unsuccessful or incomplete ERCP. Radiology 1996; 199:91-98.[Abstract/Free Full Text]
  32. Pavone P, Laghi A, Catalano C, et al. MR cholangiography in the examination of patients with biliary enteric anastomosis. AJR Am J Roentgenol 1997; 169:807-811.[Abstract/Free Full Text]
  33. Ernst O, Asselah T, Sergent G, et al. MR cholangiography in primary sclerosing cholangitis. AJR Am J Roentgenol 1998; 171:1027-1030.[Abstract/Free Full Text]
  34. Campbell WL, Ferris JV, Holbert BL, Thaete FL, Baron RL. Biliary tract carcinoma complicating primary sclerosing cholangitis: evaluation with CT, cholangiography, US, and MR imaging. Radiology 1998; 207:41-50.[Abstract/Free Full Text]
  35. Unger EC, Cohen MS, Gatenby RA, et al. Single breath-holding scans of the abdomen using FISP and FLASH at 1.5 T. J Comput Assist Tomogr 1998; 12:575-583.
  36. Duerk JL, Lewin JS, Wendt M, Petersilge C. Remember true FISP? A high SNR, near 1-second imaging method for T2-like contrast in interventional MRI at 0.2 T. J Magn Reson Imaging 1998; 8:203-208.[Medline]
  37. Reagan F, Schaefer DC, Smith DP, Petronis JD, Bohlman ME, Magnuson TH. The diagnostic utility of HASTE MRI in the evaluation of acute cholecystitis. J Comput Assist Tomogr 1998; 22:638-642.[CrossRef][Medline]
  38. Ernst O, Asselah T, Sergent G, et al. MR cholangiography in primary sclerosing cholangitis. AJR Am J Roentgenol 1998; 171:1027-1030.
  39. Campbell WL, Ferris JV, Holbert BL, Thaete FL, Baron RL. Biliary tract carcinoma complicating primary sclerosing cholangitis: evaluation with CT, cholangiography, US, and MR imaging. Radiology 1998; 207:41-50.
  40. Fulcher AS, Turner MA, Franklin KJ, et al. Primary sclerosing cholangitis: evaluation with MR cholangiography—a case-control study. Radiology 2000; 215:71-80.[Abstract/Free Full Text]
  41. Revelon G, Rashid A, Kawamoto S, Bluemke DA. Primary sclerosing cholangitis: MR imaging findings with pathologic correlation. AJR Am J Roentgenol 1999; 173:1037-1042.[Abstract/Free Full Text]
  42. Ito K, Mitchell DG, Outwater EK, Blasbalg R. Primary sclerosing cholangitis: MR imaging features. AJR Am J Roentgenol 1999; 172:1527-1533.[Abstract/Free Full Text]
  43. Matos C, Metens T, Deviere J, et al. Pancreatic duct: morphologic and functional evaluation with dynamic MR pancreatography after secretin stimulation. Radiology 1997; 203:435-441.[Abstract/Free Full Text]
  44. Matos C, Deviere J, Cremer M, Nicaise N, Struyven J, Metens T. Acinar filling during secretin-stimulated MR pancreatography. AJR Am J Roentgenol 1998; 171:165-169.[Abstract/Free Full Text]
  45. Maglinte DDT, Gage SV, Harmon BH, et al. Obstruction of the small intestine: accuracy and role of CT in diagnosis. Radiology 1993; 188:61-64.[Abstract/Free Full Text]
  46. Chou CK, Liu GC, Chen LT, et al. The use of MRI in bowel obstruction. Abdom Imaging 1993; 18:131-135.