| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Report |
1 Department of Diagnostic Radiology, University Hospital Essen, Hufelandstr.
55, D-45122 Essen, Germany.
2 Department of Internal Medicine and Cancer Research, University Hospital
Essen, D-45122 Essen, Germany.
Received October 17, 2001;
accepted after revision February 11, 2002.
Address correspondence J. Barkhausen.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. The 3D volumetric interpolated breath-hold whole-body MR imaging examination for metastases screening correlates well with CT and scintigraphy. The use of the rolling table platform permits rapid whole-body imaging in an average of 11 min. The preliminary results indicate that the described technique has the potential to emerge as an all-encompassing alternative to conventional multimodality tumor staging strategies.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
On the basis of the potential to overcome some of the outlined limitations, whole-body MR imaging has been evaluated for the detection of bone metastases [4, 5] and has been proposed for generalized tumor staging [6, 7]. Limitations of this compelling whole-body MR imaging concept have included long examination times exceeding 40 min [4] in combination with limited spatial resolution and the presence of imaging artifacts [6].
Recently, the volumetric interpolated breath-hold examination, a fat-saturated three-dimensional (3D) gradient-echo sequence with nearly isotropic resolution, has become available [8]. Although preserving adequate anatomic coverage and uniform fat saturation, the volumetric interpolated breath-hold 3D data sets are collected within the confines of a single breath-hold. In the abdomen, the volumetric interpolated breath-hold examination has been shown to provide image quality comparable with that of conventional fat-saturated two-dimensional gradient-echo images [9]. Inherently low signal-to-noise ratio requires the use of surface coils for signal reception. Extending the volumetric interpolated breath-hold high-resolution coverage from single to multiple anatomic regions toward a whole-body examination mandates rapid patient movement in conjunction with surface coil—based data reception. These requirements are fulfilled by a rolling table platform with integrated surface coils developed for whole-body MR angiography (AngioSURF system for unlimited rolling field of view; MR Innovation, Essen, Germany) [10]. The technique was adapted for the purpose of whole-body MR imaging.
The purpose of this study was to evaluate the feasibility and accuracy of a volumetric interpolated breath-hold examination in conjunction with the AngioSURF platform for whole-body MR imaging using CT and nuclear medicine techniques as the standard of reference in patients with metastases.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MR Imaging
MR examinations were performed on a 1.5-T system (Magnetom Sonata; Siemens
Medical Systems, Erlangen, Germany) equipped with high-performance gradient
systems characterized by a maximal gradient amplitude of 40 mT/m and a slew
rate of 200 mT/m/msec. The volumetric interpolated breath-hold examination is
based on a 3D spoiled gradient-echo acquisition. The sequence parameters
include a TR/TE of 3.1/1.2 msec, a flip angle of 12°, and a bandwidth of
490 Hz/pixel. A slab thickness of 312 mm was used for all measurements.
Sixty-five partitions were obtained that were subsequently interpolated to 104
partitions resulting in a final slice thickness of 3 mm. A rectangular (5/8)
field of view of 350 x 219 mm2 and a data acquisition matrix
of 120 x 256 data points resulted in an in-plane resolution of 1.8
x 1.4 mm2, which was interpolated by zero filling to a matrix
of 240 x 512 for a final pixel size of 0.9 x 0.7 mm2.
As a result of partial Fourier in both the ky (7/8) and kz (6/8)
phase-encoding directions, the data-acquisition time of a 3D volume was 22
sec. The 3D sequence incorporates a frequency-selective fat-saturation pulse
before each partition loop, which is centrically recorded to maximize fat
saturation.
Patients were placed in the supine position on the rolling table platform (Fig. 1), which is mounted on the original patient table of any MR imaging system (Symphony or Vision; Siemens Medical Systems). For easy movement in the z direction, the rolling table platform (length, 270 cm; width, 33-50 cm) is placed on seven pairs of roller bearings that are easily installed on the patient table. Signal reception is accomplished using two elements of the spine coil, which are integrated into the patient table, and the body phased array coil, which remains stationary by being attached to the original patient table. The rolling table platform glides the patient over the stationary patient table through the isocenter of the magnet between the posteriorly located spine coils and the anteriorly located body phased array coil.
|
A coronal localizer sequence of the upper abdomen that encompassed the entire liver was acquired. From this basis, further landmarks were defined in 26-cm steps in the cranial direction for imaging the thorax and skull and in the caudal direction for imaging the pelvis, femur, and knee region. All 3D data sets were collected in the axial plane, resulting in a craniocaudal coverage of 31.2 cm. To compensate for possible wrap around artifacts at the border of each 3D data set, we overlapped the acquisitions by 2.6 cm.
Paramagnetic contrast material (gadopentetate dimeglumine, Magnevist; Schering, Berlin, Germany) was applied IV into the antecubital vein using an automatic injector (Spectris; Medrad, Germany). Injection parameters included a dose of 0.2 mmol/kg gadopentetate dimeglumine and a flow rate of 3 mL/sec followed by rapid injection of 20 mL of normal saline at the same flow rate. Depending on body height, patients were examined in seven (n = 5) or eight steps (n = 3). After a delay of 20 and 60 sec after contrast administration, the first two 3D data sets of the abdomen, encompassing the entire liver, were performed in the axial plane with patients breath-holding. Subsequently, the rolling table platform was moved manually to the chest region, where a coronal 3D data set was collected. The 8 sec required to move the platform was used to provide breathing instructions. Subsequently, the platform was moved to permit the acquisition of axial 3D data sets of the pelvis, the femur or knee, and the skull. Finally, the platform was moved back to the abdomen, where a third abdominal data set was collected. The entire imaging time was 3 min 52 sec for the seven-step protocol and 4 min 22 sec for the eight-step protocol.
Reference Examinations
All patients underwent thoracic, abdominal, and pelvic helical CT. In three
patients, contrast-enhanced helical CT was performed to exclude brain
metastases. Furthermore, five patients underwent skeletal scintigraphy. All
examinations were performed 3-13 days (average, 7.6 days) before the
whole-body MR examination.
Data Analysis
The 3D MR imaging data sets of each whole-body examination were analyzed
interactively by two radiologists on a workstation (Virtuoso; Siemens Medical
Systems) in the multiplanar reformation mode. Discrepancies were resolved by
consensus. Reviewers did not have information about the results of other
imaging modalities. Information about patient age, sex, and primary tumor were
provided. Images were evaluated for the presence of metastatic disease.
Metastases were numerically quantified for lung; liver; bones; and other
anatomic regions including the cerebrum, retroperitoneum, and skin. The
diameter of the smallest metastasis detected on whole-body MR imaging for each
anatomic region was measured. Whole-body MR findings were directly compared
with the reference examinations to evaluate agreement.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Whole-body MR imaging detected hepatic metastases in six patients (Fig. 2A,2B). In three patients, more than five hepatic metastases were identified. The smallest hepatic metastasis measured 6 mm in diameter. No false-negative findings were revealed on MR imaging. Whole-body MR imaging detected pulmonary metastases in five patients. In one patient, a lung metastasis measuring 5 mm in diameter on CT was missed on MR imaging. The smallest pulmonary metastasis measured 8 mm in diameter (Fig. 3A,3B). In addition, MR imaging revealed retroperitoneal metastases in one patient with testicular cancer and a metastases in the dorsal abdominal wall in another patient. Both findings were confirmed on CT.
|
|
|
|
Bone metastases evidenced by increased signal intensity on contrast-enhanced volumetric interpolated breath-hold images were detected on MR imaging in four patients. Locations included the spine (n = 3), pelvis (n = 3), and ribs (n = 2). Furthermore, in two patients, osseous metastases were seen in the scapula and the femur. Bone scintigraphy, used as the standard of reference, revealed good correlation with MR imaging (Fig. 4A,4B,4C) except in one patient. In this patient, MR imaging had revealed osseous lesions in the lumbar spine, whereas skeletal scintigraphy did not. Eventually, the presence of bone metastasis was confirmed at biopsy. All other MR imaging findings regarding osseous metastases were confirmed on skeletal scintigraphy.
|
|
|
Cerebral metastases ranging between 8 and 15 mm in diameter were detected on MR imaging in two patients. All metastases, and their absence in a third patient, were confirmed on CT (Fig. 5A,5B).
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The concept of whole-body MR imaging is not new [4,5,6,7]. Long scanning times and poor image quality characterized by extensive artifacts have limited its clinical impact [4, 6, 7]. In conjunction with a fat-saturated ultrafast 3D gradient-echo sequence, the application of the rolling table platform overcomes some of these limitations. After being placed on the platform, the patient is moved through the MR scanner and through an integrated coil sandwich without a new landmark being acquired for each step of the whole-body examination. Hence, high image quality can be obtained throughout all stations, and even small lesions are readily detected and distinguished from surrounding tissues.
Most whole-body MR imaging concepts are based on echoplanar sequence designs [6, 11]. Johnson et al. [11] presented a technique on the basis of the acquisition of a stack of transverse T2-weighted echoplanar images, using the body coil for both signal transmission and reception. An automatic table moved the patient through the gantry. The results were encouraging because most abnormalities could be detected. Horvath et al. [6] reported similar results for whole-body echoplanar MR imaging of patients with breast cancer, which rendered correct findings in 95% of patients. Apparent limitations of the technique included a low signal-to-noise ratio resulting in reduced spatial resolution and magnetic susceptibility effects amplified by inconsistent shimming over the different body regions, causing considerable distortions in echoplanar image contrast and geometry. A recent study assessing true fast imaging with steady-state precession for whole-body MR imaging documented poor image contrast and low spatial resolution as significant limitations [7].
The whole-body MR examination overcomes these limitations. The three-dimensional volumetric interpolated breath-hold data sets provide high-resolution images with nearly isotropic voxels. In conjunction with zero-filling routines, the voxel dimensions could be reduced to 0.7 x 0.9 x 3.0 mm. The sequence is T1-weighted, thus documenting the enhancement of paramagnetic contrast material. In conjunction with fat saturation, areas characterized by increased contrast material uptake in the parenchymal organs and the skeletal system are easily detected. Fast imaging routines permit the collection of a first abdominal data set in the arterial phase followed by a second data set obtained in the portal venous phase. A delayed 3D data set completes the assessment of the liver. This type of dynamic 3D imaging has been documented to be most accurate in the detection and characterization of hepatic mass lesions [12]. Thus, all hepatic metastases documented on multiphase dynamic CT were also seen on whole-body 3D volumetric interpolated breath-hold MR imaging.
Outside the abdomen, the 3D data is collected in the equilibrium phase. The enhancement characteristics in this late phase are well suited for the detection of cerebral and osseous lesions. In this limited study, the MR 3D data sets proved sufficient for visualization of all cerebral and osseous metastases identified on CT and bone scintigraphy, respectively. In fact, whole-body MR imaging revealed one-bone metastases not detected on skeletal scintigraphy and later confirmed at biopsy. Limitations of skeletal scintigraphy are well known, as is the fact that MR imaging is more sensitive in the detection of metastases from some primary tumors [4]. Limitations inherent to the 3D volumetric interpolated breath-hold concept seemed evident in the lungs. Although only a single 5-mm lesion was missed, the underlying image quality compared poorly with CT scanning. Perhaps the incorporation of other sequences such as half-Fourier acquired single-shot turbo spin echo could have improved assessment of the lungs for metastatic disease.
The examination time is minimized by obviating patient and coil repositioning. The rolling table concept, which has been successfully applied to whole-body MR angiography [10], provides extended craniocaudal coverage in 1 min 12 sec. The short overall examination time of less than 15 min contributes to high patient acceptance. More important, from a diagnostic perspective, the time required to collect the 3D data sets after the contrast injection amounts to 4 min. During this time interval, sufficient paramagnetic contrast material will remain in the lesions to permit their ready identification.
Certainly, whole-body MR imaging will not replace dedicated MR examinations, which can exploit a host of different contrast mechanisms for the study of individual organ systems. The aim of the described strategy is focused solely on the detection and follow-up of metastatic lesions in patients with known primary tumors. If identified lesions cannot be reliably characterized as metastatic, additional examinations are required. Naturally, the clinical potential of whole-body MR imaging should be evaluated on the basis of larger patient cohorts—we are hopeful that this first report will motivate others to help in this regard.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Brauck, M. O. Zenge, F. M. Vogt, H. H. Quick, F. Stock, T. Trarbach, M. E. Ladd, and J. Barkhausen Feasibility of Whole-Body MR with T2- and T1-weighted Real-time Steady-State Free Precession Sequences during Continuous Table Movement to Depict Metastases Radiology, March 1, 2008; 246(3): 910 - 916. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kataoka, H. Ueda, T. Koyama, S. Umeoka, K. Togashi, R. Asato, S. Tanaka, and J. Ito Contrast-Enhanced Volumetric Interpolated Breath-Hold Examination Compared with Spin-Echo T1-Weighted Imaging of Head and Neck Tumors Am. J. Roentgenol., January 1, 2005; 184(1): 313 - 319. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. C. Lauenstein, S. C. Goehde, C. U. Herborn, M. Goyen, C. Oberhoff, J. F. Debatin, S. G. Ruehm, and J. Barkhausen Whole-Body MR Imaging: Evaluation of Patients for Metastases Radiology, October 1, 2004; 233(1): 139 - 148. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Antoch, F. M. Vogt, L. S. Freudenberg, F. Nazaradeh, S. C. Goehde, J. Barkhausen, G. Dahmen, A. Bockisch, J. F. Debatin, and S. G. Ruehm Whole-Body Dual-Modality PET/CT and Whole-Body MRI for Tumor Staging in Oncology JAMA, December 24, 2003; 290(24): 3199 - 3206. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
