This study compared two T2-weighted FLAIR sequences for imaging the posterior fossa in pediatric patients - a conventional sequence with rectilinear k-space coverage and a BLADE sequence with rotating blade-like k-space trajectories. Ratings of 26 patients found significantly fewer motion and pulsation artifacts with BLADE, allowing better differentiation of lesions from artifacts. BLADE improved delineation of small lesions and increased interobserver agreement.

1. MAGNETOM Flash
The Magazine of MR
Issue Number 4/2010
Special Edition for the USA
Clinical
Reduction of Motion
Artifacts in the
Posterior Fossa
Page 6
3D Knee Imaging
using syngo SPACE
Page 12
Musculoskeletal MRI
in Sports Medicine
Page 20
Traumatic Lesion of
the Brachial Plexus
Page 32
How I do it
Visualizing the
Distal Biceps Tendon
Page 36
USA

2. Editorial
Matthias Lichy, M.D.
Dear MAGNETOM user,
At the recent annual meeting of the ISMRM, MAGNETOM Flash magazine and our
the radiological community heard about internet platform, MAGNETOM World
the latest steps in MR technology and how (www.siemens.com/magnetom-world),
Siemens is a leader in contributing to the also support this sharing of knowledge.
innovations of these developments. A central Those who attended these meetings
theme was how new technologies can help received their own copies of the two latest
us improve our day-to-day patient care and MAGNETOM Flash issues including clinical
productivity. Future steps in MR technology and research highlights from world-wide
were also a focal point, including ultra high known and honored researchers and clini-
field imaging, parallel transmission and com- cians from the United States of America.
pressed MR image acquisition. For those who could not attend either
meeting, the editorial team of MAGNETOM
The Siemens user community was also able Flash has selected articles from the two
to participate in another remarkable event: latest international versions. We hope they
the MAGNETOM World Summit, held in will provide you with new ideas and under-
Shenzhen, China. More than 500 participants standing.
from all corners of the globe exchanged
ideas on how to use innovative imaging We hope you will enjoy reading this special
techniques such as syngo TimCT or syngo U.S. issue of MAGNETOM Flash!
Tissue4D in daily routine. This exchange
of knowledge lies at the heart of what
every MAGNETOM World Summit organized
by Siemens Healthcare MR is all about. Matthias Lichy, M.D.
2 MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world

3. Editorial
The Editorial Team
We appreciate your comments.
Please contact us at magnetomworld.med@siemens.com
Antje Hellwich Okan Ekinci, M.D. Peter Kreisler, Ph.D. Heike Weh,
Associate Editor Center of Clinical Competence – Collaborations & Applications, Clinical Data Manager,
Cardiology, Erlangen, Germany Erlangen, Germany Erlangen, Germany
Bernhard Baden, Ignacio Vallines, Ph.D., Wellesley Were Milind Dhamankar, M.D.
Clinical Data Manager, Applications Manager, MR Business Development Sr. Director, MR Product
Erlangen, Germany Erlangen, Germany Manager Marketing, Malvern, USA
Australia and New Zealand
Michelle Kessler, US Gary R. McNeal, MS (BME) Dr. Sunil Kumar S.L.
Installed Base Manager, Advanced Application Specialist, Senior Manager Applications,
Malvern, PA, USA Cardiovascular MR Imaging Canada
Hoffman Estates, USA
MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world 3

4. Content Content
Content 15
syngo NATIVE TrueFISP
20
MRI in Sports Medicine
40
syngo ASL
68
Klippel-Trénaunay-Weber Syndrome
Further clinical information Clinical Clinical Clinical Clinical
Visit the MAGNETOM
Pediatric Imaging k How I do it Whole-Body k How I do it
World Internet pages at 6 Diagnostic Relevant Reduction 36 FABS View of the Elbow 57 Case Reports: Tumor Detection by 76 Non Contrast-Enhanced MR
www.siemens.com/ of Motion-Artifacts in the Posterior for Visualization of Distal Diffusion-Weighted MRI and Angiography (syngo NATIVE)
magnetom-world Fossa by syngo BLADE Imaging Biceps Tendon ADC-Mapping with Correlation to Manuela Rick, et al.
for further clinical T. von Kalle, et al. Peter Cazares, et al. PET/CT Results
information and talks Matthias Philipp Lichy, et al.
by international 38 Tips for T2-weighted TSE Shoulder
experts. Clinical Imaging with Spectral Fat Saturation
Orthopedic imaging Zhang, Wei Jun, Clinical
Hans-Peter Hollenbachl.
12 3D High Resolution MRI of the Women’s Health
Knee at 3T Using a Moderately T2- 62 Whole-Body MRI for Accurate
weighted 3D-TSE-fs (syngo SPACE) Clinical Assessment of Tumor Load of
sequence – Useful or Not? Neurology Bone Metastases Originate from
A. Horng, et al. Mamma Carcinoma
40 Cerebral Blood Flow Imaging Philip Aschoff, et al.
20 Case Reports: Musculoskeletal MRI in with 3D GRASE ASL Sequence
Sports Medicine Increases SNR and Shortens 68 Case Report: Klippel-Trénaunay-
Heinz-Peter Schlemmer, et al. Acquisition Time Weber Syndrome
David A. Feinberg, Chawla Tanya, et al.
32 Case Report: Traumatic Lesion of the Matthias Günther
Left Brachial Plexus
Markus Lentschig 48 Case Report: Non-Contrast-
Enhanced Evaluation of Perfusion
Clinical
Deficiencies of the Brain Cardiovascular
David J. Panasci,
Steven L. Mendelsohn 73 syngo NATIVE TrueFISP in the
Assessment of the Transplanted
52 Multimodal MRI of the Brain for Kidney
Improved Diagnosis and Therapy Peter Weale, et al.
Planning in a Case of Glioblastoma
Multiforme
Elna-Marie Larsson
4 MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world 5

5. Clinical Pediatric Imaging Pediatric Imaging Clinical
Diagnostic Relevant Reduction of the same time reduce the visibility
of small brain lesions. We therefore
2
compared the image quality of two
Motion Artifacts in the Posterior Fossa by T2-weighted fluid attenuated inversion
recovery (T2w FLAIR) sequences with
syngo BLADE Imaging different k-space trajectories (conven-
tional Cartesian and BLADE) with respect
to artifacts and depiction of small hyper-
T. von Kalle1; B. Blank2; C. Fabig-Moritz1; P. Müller-Abt1; M. Zieger1; K. Wohlfarth3; P. Winkler1 intense brain lesions [5].
1
Department of Pediatric Radiology, Olgahospital Klinikum Stuttgart, Germany Imaging techniques
2
Department of Pediatric Oncology, Hematology and Immunology, Olgahospital Klinikum Stuttgart, Germany We used a 1.5T scanner (MAGNETOM
3
Siemens Healthcare, Erlangen, Germany Avanto, Siemens Healthcare, Erlangen,
Germany) and the Siemens 12-channel
head matrix coil.
For each patient we compared two
sequences with transverse 4 mm sec-
1 Although movement and pulsation arti- tions (gap 10%) that were applied in
facts are a frequent problem in daily rou- identical slice positions:
tine [1–4] especially in the diagnostics 1. The T2w FLAIR standard sequence, a
of pediatric patients, only few articles on spin-echo sequence with conventional
this topic can be found in the literature. rectilinear Cartesian k-space trajectories
According to our experience mainly (image parameters: TI 2500 ms, TR 9000
MR images of the posterior fossa, the ms, TE 100 ms, BW 150 Hz/pixel, turbo
cerebellum and the brain stem, may be factor 19, TA160 s, FOV 230 x 230 mm,
significantly impaired by artifacts from matrix 256 x 256).
pulsatile flow of blood or cerebrospinal 2. The BLADE FLAIR sequence with rotat- 2 Movement artifacts caused by head movements (arrowheads) and pulsation artifacts
fluid even without patient head move- ing blade-like k-space trajectories (image caused by pulsatile flow (arrows). T2w FLAIR sequence with conventional rectilinear
ment [5, 6]. Sedation or general anes- parameters: TI 2500 ms, TR 9000 ms, k-space trajectories.
L thesia rarely influence these pulsation or
flow artifacts. However, accurate assess-
TE 107 ms, BW 250 Hz/pixel, turbo factor
29, TA 260 s, BLADE-Coverage 130 %,
ment of small brain lesions is essential FOV 230 x 230 mm, matrix 256 x 256).
in many pediatric patients, especially in During the acquisition of both
those with malignant brain tumors. sequences children were encouraged process the direction of this “blade” is 2 years 7 months to 17 years (median
MR imaging with “rotating blade-like not to move their head. They were rotated around the k-space centre such age 8 years 5 months) were included in
k-space covering” (BLADE) and “Perio- offered video films or audio programs that the complete series covers the the study.
dically Rotated Overlapping Parallel during the examination. whole k-space (Fig. 1). Images can be
Lines with Enhanced Reconstruction” displayed with or without an additional Image evaluation
(PROPELLER) have been shown to effec- BLADE technique motion correction algorithm. Four experienced pediatric radiologists
tively reduce artifacts in healthy volun- syngo BLADE is the product name of independently assessed unlabeled
teers and adult patients [7, 8, 9], as the motion insensitive Siemens turbo- Patients images of both FLAIR sequences of each
well as in pediatric patients [4, 10] and spin-echo sequence which utilizes the The typical hyperintense white matter patient. Structures of the posterior fossa,
therefore have the potential to reduce PROPELLER k-space trajectory [11]. abnormalities in the cerebellum of cerebellum and brain stem, were evalu-
the frequency of anesthesia in children. The technique is approved for diagnostic patients with neurofibromatosis type 1 ated according to the presence of move-
1 k-space trajectory in BLADE imaging. The k-space is covered by a series of blades each of As these MR techniques reduce motion MR examinations of patients. It consists (NF 1) [12] served the purpose of our ment artifacts (caused by head move-
which consists of the lowest phase encoding lines. The centre of the k-space (red circle) with artifacts by fast segmental image acqui- of blade-like rotating k-space coverage. study to assess and compare the visibil- ments) or pulsation artifacts (caused by
diameter L is resampled for every blade. Data are then combined to a high resolution image. sition combined with mathematical During each echo train of a BLADE ity of small and low contrast lesions. pulsatile flow of blood and/or cerebro-
algorithms, we assumed that it might at sequence the lowest phase encoding We re-evaluated images of children with spinal fluid) (Fig. 2), their differentiation
lines of a conventional rectilinear NF 1, who had been routinely scanned from lesions, and their delineation from
k-space are acquired. The number of for optic pathway gliomas and who the surrounding tissue by contrast (dif-
lines, which depends on the length of had been examined with T2w FLAIR ference in signal intensities) and edge
the echo train, determines the resolu- sequences of both techniques. 26 definition (clearly or poorly defined mar-
tion of the image. During the acquisition patients, 10 girls and 16 boys from gins) as has been described elsewhere
6 MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world 7

6. Clinical Pediatric Imaging Pediatric Imaging Clinical
3A 3B
Table 1: Scores for movement and pulsation artifacts, and the differentiation of artifacts
and lesions in the posterior fossa.
Conventional syngo BLADE
yes no yes no
Movement 29 75 4 100
Pulsation 99 5 25 79
Differentiation 53 30 81 2
lesion/artefacts
Yes = artifacts present, lesions and artifacts distinguishable. No = artifacts not present, lesions and artifact not distinguishable. Maximum score for each
sequence = 104. Last line. Patients with neither artifacts nor lesions were excluded. Results of statistical tests cf Table 2.
3C 3D
[5] The transverse diameters of the alignment similar to a trellis [13] have In our study, all observers found more
largest and smallest lesions were mea- been shown to effectively reduce arti- pulsation artifacts than movement arti-
sured in both sequences of each patient. facts in T1- and T2-weighted images. facts in images of both the conventional
Signal intensities of a representative Studies on pediatric [4] and adult and the syngo BLADE sequence (Table
lesion and the adjacent normal brain patients [7] found a comparable detect- 1). Artifacts were reported significantly
tissue were measured. [5] ability of lesions in contrast-enhanced less often in images acquired with
T1-weighted images of FLAIR BLADE BLADE technique than in images with
Discussion sequences and conventional spin-echo rectilinear k-space trajectory (Table 1
Techniques with rotating blade-like sequences. However, image quality and and 2). These results confirm that
k-space covering (BLADE) [9 ], with delineation of small or poorly delineated artifacts caused by pulsation, flow and
periodically rotated overlapping parallel lesions in MR images in T2-weighted motion are significantly reduced by the
lines with enhanced reconstruction FLAIR acquired with these techniques BLADE technique in comparison to the
(PROPELLER) [8,10], and with k-space have not been systematically studied. standard sequence with conventional
Table 2: Posterior fossa. Comparison of conventional vs. BLADE images. 3 T2w FLAIR images of the posterior fossa of two different patients. Rectilinear k-space coverage (left), BLADE (right). Lesions
McNemar’s test. p-values for each observer. Cohen’s kappa for interobserver agreement. typical of neurofibromatosis type 1: low contrast confluent (arrowhead), high contrast round (arrow). Artifacts and lesions not
reliably distinguishable in conventional images (left).
Presence of Presence of Differentiation of Delineation of
movement artifacts pulsation artifacts lesions and artifacts lesions
Edge Contrast
definition rectilinear k-space covering. As the images led to an only fair interobserver tional T2w FLAIR images they could not
children were encouraged not to move agreement (Cohen’s kappa < 0.4, Table differentiate lesions from artifacts in
Observer 1 0.03 0.02 0.04 > 0.05 > 0.05 during the examination, there were only 2). These artifacts did not disturb the the posterior fossa, rendering nearly a
moderate to minor movement artifacts depiction of lesions. fourth of the examinations with conven-
Observer 2 0.02 0.00002 0.002 0.0027 0.0009
even in conventional T2w FLAIR images. There was good observer agreement tional k-space trajectory inadequate for
Observer 3 0.041 0.000007 0.001 > 0.05 > 0.05 Pulsation artifacts were more frequent (Cohen’s kappa 0.74, Table 2) that arti- a reliable diagnosis. In none of the BLADE
and more severe in conventional images facts compromised the assessment of sequences more than one observer
Observer 4 0.013 0.00002 0.013 > 0.05 > 0.05 and sometimes also present in BLADE lesions more often and more severely in considered artifacts and lesions to be
images. Ratings of subtle movement conventional T2w FLAIR images than in indistinguishable (Fig. 3).
Cohen’s kappa 0.34 0.37 0.72 0.11 0.01 artifacts in conventional images and BLADE images. In 6 of all 26 patients
subtle pulsation artifacts in BLADE (23%) observers agreed that in conven-
8 MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world 9

7. Clinical Pediatric Imaging Pediatric Imaging Clinical
4A As Gill et al. reported for their sample Observers of our study had the subjec- Conclusion
[12], most hyperintense lesions in the tive impression that the appearance of
thalami, brain stem and cerebellum FLAIR BLADE images differed slightly BLADE technique reduces movement 7 Forbes KP, Pipe JG, Karis JP, Heiserman JE (2002)
Improved image quality and detection of acute
were confluent or diffuse with poorly from that of conventional images which and pulsation artifacts in T2w FLAIR
cerebral infarction with PROPELLER diffusion-
defined edges. A smaller number of did not impair image quality and lesion images without relevant loss of image weighted MR imaging. Radiology 225:551-555.
lesions were well circumscribed with detectability (Figs. 3 and 4). These quality. It therefore markedly improves 8 Wintersperger BJ, Runge VM, Biswas J, Nelson
edges that were distinct from the adja- subtle differences have already been depiction of small and low contrast CB, Stemmer A, Simonetta AB, Reiser MF,
cent normal tissue. Mean ratios of signal reported for PROPELLER technique [2]. brain lesions in children in the posterior Naul LG, Schoenberg SO (2006) Brain magnetic
resonance imaging at 3 Tesla using BLADE com-
intensities were only slightly lower in BLADE images might therefore be identi- fossa of pediatric patients. This can be
pared with standard rectilinear data sampling.
the posterior fossa than in cerebrum and fied by experienced radiologists despite crucial especially in patients after Invest Radio 41: 586-592.
midbrain. “blinding”. We have, however, not identi- surgery of malignant brain tumors. In 9 Forbes KP, Pipe JG, Karis JP, Farthing V,
If lesions were not obscured by artifacts, fied bias due to this fact in our study. the absence of major artifacts lesions Heiserman JE (2003) Brain imaging in the unse-
dated pediatric patient: Comparison of periodi-
visibility of lesions with both clearly and In our study, data acquisition time and of all sizes were depicted in comparable
cally rotated overlapping parallel lines with
poorly defined edges appeared to be image reconstruction time together quality by both techniques. enhanced reconstruction and single-shot fast
comparable in images of both tech- were approximately 2–3 minutes longer spin-echo sequences. AJNR 24: 794-798.
niques (Fig. 4). Observers’ comparisons for syngo BLADE sequences than for References 10 Pipe JG (1999) Motion correction with
of both imaging techniques according to conventional FLAIR sequences. The 1 Böck JC, Neumann K, Sander B, Schmidt D, PROPELLER MRI: Application to head motion and
Schörner W (1991) Prepontine artifacts due to free-breathing cardiac imaging. Magn Reson
contrast and edge definition did not advantage of artifact reduction by the
pulsation of cerebrospinal fluid in T2-weighted Med 42: 963-969.
reveal consistently significant differ- BLADE technique clearly outweighed the 11 Gill DS, Hyman SL, Steinberg A, North KN (2006):
coronal MRI. Clinical relevance, incidence and a
ences for lesions of the posterior fossa prolonged duration. technique for efficient artefact suppression. Age-related findings on MRI in neurofibromato-
(Table 2). There only was a tendency to There are limitations of our retrospective RoeFo 154: 202-205. sis type 1. Pediatr Radiol 36: 1048-1056.
better edge definition in BLADE images. study: We were not able to measure the 2 Forbes KP, Pipe JG, Bird CR, Heiserman JE (2001) 12 Maclaren JR, Bones PJ, Millane RP, Watts R (2008)
PROPELLER MRI: Clinical testing of a novel MRI with TRELLIS: a novel approach to motion
Visibility of lesions also was indepen- degree of patients’ movements. There-
technique for quantification and compensation correction. Magn Res Imaging 26: 474-483.
dent of size. Even the smallest lesions of fore, like other investigators [2, 4, 9], we 13 Iskandar BJ, Sansone JM, Medow J, Rowley HA
of head motion. J Magn Reson Imaging 14:
our sample (2–3 mm) were equally can only assume that these parameters 215-222. (2004) The use of quick-brain magnetic reso-
4B
depicted by both techniques. were similar in statistical mean through- 3 Penzkofer AK, Pfluger T, Pochmann Y, Meissner nance imaging in the evaluation of shunt-treated
O, Leisinger G (2002) MR imaging of the brain hydrocephalus. J Neurosurg (Pediatrics 2) 101:
BLADE technique is based on standard out the examination. As both sequences
in pediatric patients: Diagnostic value of HASTE 147-151.
image acquisition techniques and there- were performed successively, pulsation 14 Ba-Ssalamah A, Schick S, Heimberger K, Linnau
sequences. AJR 179: 509-514.
fore has the advantage of providing and flow artifacts were also assumed as 4 Alibek S, Adamietz B, Cavallaro A, Stemmer A, KF, Schibany N, Prokesch R, Trattnig S (2000)
image characteristics equal to standard unchanged for both acquisitions. For our Anders K, Kramer M, Bautz W, Staatz G (2008) Ultrafast magnetic resonance imaging of the
sequences. As an alternative to syngo retrospective analysis we had to accept Contrast-enhanced T1-weighted fluid-attenuat- brain. Magn Reson Imaging 18: 237-43.
BLADE imaging, pulsation artifacts may that in addition to the k-space trajectory ed inversion-recovery BLADE Magnetic Reso-
nance Imaging of the brain: An alternative
be identified by a second data acquisi- some of the parameters of both
to spin-echo technique for detection of brain
tion after changing the phase encoding sequences were not identical. As the lesions in the unsedated pediatric patient? Contact
direction [1] or the slice orientation, BLADE technique benefits from long Acad Radiol 15: 986-995. Thekla v. Kalle, M.D.
with the disadvantage of a longer echo trains, a turbo factor had been 5 Von Kalle T, Blank B, Fabig-Moritz C, Müller-Abt P, Department of Pediatric Radiology
Zieger M, Wohlfarth K, Winkler P (2009) Olgahospital Klinikum Stuttgart
examination time and the higher risk of chosen that was larger than that of our
reduced artefacts and improved assessment of Bismarckstr. 8
movement artifacts. Reduction of routine T2w FLAIR sequence. However, a 70176 Stuttgart
hyperintense brain lesions with BLADE MR
motion artifacts in non-sedated children change of these parameters would have imaging in patients with neurofibromatosis Germany
can also be achieved by rapid sequences had only minor influence on motion type 1. Pediatr Radiol 39: 1216-1222. Phone: +49 (0) 711/ 278-04
Kallmes DF, Hui FK, Mugler JP III (2001) Sup- t.vonkalle@klinikum-stuttgart.de
(e.g. single shot techniques) which, in artefacts, and presumably would not
pression of cerebrospinal fluid and blood flow www.klinikum-stuttgart.de
neuroimaging, have the disadvantages have changed the quality of results.
artifacts in FLAIR MR imaging with a single-slab
of a poorer differentiation of gray and Prospective studies of a larger patient three-dimensional pulse sequence: Initial
white matter [3, 10,14] and a lower group could avoid these shortcomings. experience. Radiographics 221: 251-255.
spatial resolution [15]. 6 Naganawa S, Satake H, Iwano S, Kawai H, Kubota
S, Komada T, Kawamura M, Sakurai Y, Fukatsu H
(2008) Contrast-enhanced MR imaging of
the brain using T1-weighted FLAIR with BLADE
compared with a conventional spin-echo
sequence. Eur Radiol 18: 337-342.
4 Absence of major artifacts and comparable visibility of lesions in images
of both techniques (T2w FLAIR: rectilinear k-space top, BLADE bottom), low
contrast confluent (arrowhead), high contrast round (arrow).
10 MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world 11

8. Clinical Orthopedic imaging Orthopedic imaging Clinical
3D High Resolution MRI of the Knee sion Recovery (SPAIR) technique. Parallel
imaging was performed with the k-space
based technique syngo GRAPPA with an
thickness for MPR of 1 mm (SPACE1mm) to
be optimal for the visualization of ana-
tomical structures (Fig. 1). This slice
despite slightly inferior CNR (joint fluid/
cartilage, joint fluid/menisci, fat/liga-
ments and bone marrow/subchondral
at 3T Using a Moderately T2-weighted acceleration factor R = 2. For signal
reception, a dedicated multichannel knee
thickness provides significantly higher
SNR for ligaments, subchondral bone
bone) as compared to 2D-TSE-fs.
coil with 8 independent RF-channels was and menisci and at least equal SNR for Clinical application
3D-TSE-fs (syngo SPACE) sequence – used. Reformation of the datasets was
performed on a syngo MultiModality
cartilage, bone marrow, muscle and fat
of syngo SPACE as compared to conven-
The reconstruction time for one syngo
SPACE dataset was below 30 s, the data
Useful or Not? Workstation (Leonardo, software version
VB15A, Siemens Healthcare, Erlangen,
tional 2D-TSE-fs. Though identification
of anatomical structures was compara-
acquisition time was 10 min 35 sec with
syngo SPACE versus 12 min 48 sec with
Germany). ble for syngo SPACE and 2D-TSE-fs, the 2D TSE in three planes (table 1). Thus
A. Horng1, M. Notohamiprodjo1, J. Raya1, J. Park1, W. Horger 2, A. Crispin1, M. F. Reiser1, C. Glaser1 Analysis of axial, sagittal and coronal SPACE1mm showed significantly better the overall acquisition time for syngo
reformations (MPR) of 0.5 mm, 1 mm visualization of menisci in axial sections SPACE was comparable to the acquisi-
1
Department of Clinical Radiology, University of Munich – Grosshadern Campus, Munich, Germany and 2 mm slice thickness suggest a slice and meniscal roots in coronal sections tion of the 2D-TSE-fs datasets in three
2
Siemens Healthcare, Erlangen, Germany
1
Background
Magnetic resonance imaging (MRI) of nervous system [4] and recently for the of the knee. Parallel imaging facilitates
the knee is justifiably one of the most body trunk [5, 6]. They enable data blockwise 3D-data acquisition with iso-
commonly performed MRI examinations, acquisition with high isotropic spatial tropic spatial resolution for evaluation of
as it offers excellent direct depiction of resolution and allow for an interactive the whole knee in a reasonable time
cartilage, ligaments, menisci and periar- 3-dimensional visualization. Such post- window. The acquisition time should be
ticular soft tissue. This can be achieved processing after an initial isotropic data either less or at least comparable to
by standard application of fat-saturated acquisition has been proven successful acquisition times of conventional 2D TSE
moderately T2-weighted 2D Turbo Spin in many other MR and CT-based applica- datasets in three planes. The advantage
Echo (TSE)-sequences in three orienta- tions. of an isotropic 3D-dataset is the possibil-
tions [1, 2]. However, conventional TSE- ity of 3-dimensional multiplanar refor-
sequences are not isotropic, hence struc-
Technical considerations matting (MPR), which may enhance the
tures and signal alterations / lesions with for syngo SPACE evaluation of small delicate or oblique
a size less than the usual slice thickness Recently a 3D-TSE-sequence with moder- structures like meniscal roots or the fas-
of 3 to 6 mm, i.e. meniscal roots, may not ate T2-weighting called “Sampling Perfec- cicles of the anterior cruciate ligament.
be completely detected. A slice thickness tion with Application optimized Contrasts Disadvantages might be slightly
below 3 mm is rarely acquired because of using different flip angle Evolutions” decreased in-plane resolution as com- 2D-TSE-fs 3 mm SPACE 0.5 mm SPACE 1 mm SPACE 2 mm
its reduced signal-to-noise ratio (SNR) (syngo SPACE), was developed for 3T sys- pared to conventional 2D-TSE-fs-
1 Coronal syngo SPACE reconstructions in 0.5 mm, 1 mm and 2 mm show a good homogeneity throughout the image as
and contrast-to-noise ratio (CNR) and tems. A restore pulse and variable flip sequences and some additionally
compared to the T2w-2D-TSE-fs.
because of the prolonged acquisition angle distribution enable extremely large required time for the 3D reconstructions.
time for complete joint coverage. Fur- turbo factors. The variable flip angles Recently our research group evaluated
thermore post-processing options for provide a particular evolution of the sig- syngo SPACE for isotropic highly resolved
2D-sequences for the assessment of nal during the echo train resulting in a MRI of the knee at 3T (MAGNETOM Trio,
structures, which are captured in an “pseudo steady-state” with constant sig- Siemens Healthcare, Erlangen, Germany) Table 1: Sequence parameters for the syngo SPACE and the T2w-2D-TSE-fs-sequences.
oblique course through several slices, like nal level neglecting relaxation [7]. Addi- with consecutive 3-dimensional-MPR in
the anterior cruciate ligament or the fem- tionally, SAR is reduced by this acquisi- comparison to conventional 2D-TSE-fs-
oral trochlear cartilage [3] are limited. tion scheme. The usage of this sequences in three planes (coronal, sag- TR TE FA [°] Resolution FOV Matrix Parallel Takqis
In this setting the introduction of a technique on a high field 3T system ittal, axial) [10]. [ms] [ms] [mm3] [cm] Imaging
highly resolved 3D moderately allows integration of parallel imaging Sequence parameters for syngo SPACE
T2-weighted (3D-T2w-TSE) sequence with excellent SNR and CNR at reason- and for the moderately T2w-2D-TSE-fs- syngo SPACE 1200 30 120 0.5 16 320 x 320 GRAPPA r=2 10’35’’
may be useful. In the literature time effi- able acquisition times [8, 9]. sequence are given in table 1. Fat satura-
cient 3D-T2w-TSE sequences have The application of syngo SPACE at 3T tion in syngo SPACE was performed with T2w-2D-TSE-fs 3200 30 180 0.36 x 0.36 x 3 16 448 x 448 GRAPPA r=2 12’34’’
already been evaluated for the central might establish a new approach to MRI the SPectral selection Attenuated Inver-
12 MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world 13

9. Clinical Orthopedic imaging Orthopedic imaging Clinical
2 3
2D-TSE-fs 3 mm
2D-TSE-fs 3 mm
SPACE 1 mm
2 Coronal reconstructed syngo SPACE1mm and 2D-TSE-fs of a healthy volunteer. SPACE provides better visualization of the posterior medial
meniscal root as compared to T2w-2D-TSE-fs, where the insertion of the root is blurred because of larger partial volume effects.
planes suggesting that the technique is configuration is challenging to interpret ligament (Fig. 7) may aid in the evalua-
feasible for daily clinical use. on conventional angulated thick sagittal tion of primarily difficult anatomical sites
The advantage of syngo SPACE over or coronal sections (Fig. 5). The signal / or a complicated situation after injury.
2D-TSE-fs is the possibility of free multi- image characteristics of syngo SPACE
planar isotropic reconstructions at com- appear more similar to TSE image charac- Conclusion
parable SNR resulting in a slightly teristics than to GRE and therefore are Blockwise acquired syngo SPACE is a
improved detection and differentiation unlikely to require a big adjustment of new approach to MRI of the knee at 3T.
of relevant small ligamenteous (Fig. 2) the radiologist’s reading and interpreta- It allows highly-resolved isotropic true SPACE 1 mm
and meniscal structures (Figs. 3, 4). Clin- tion habits to the new sequence. 3-dimensional acquisition and subse-
ical relevance thus might be better visu- Usage of the free 3D-reformation accord- quent reconstruction. Overall acquisition 3 Axial sections of the medial meniscus of a healthy individual. syngo SPACE provides more detailed depiction of the meniscus throughout a
alization of small avulsive ligamenteous ing to the course of oblique anatomical time is shorter than that of three sepa- higher number of slices as compared to 2D-TSE-fs. Both the meniscal body and its attachments (meniscal roots) are clearly visualized in SPACE
lesions, e.g. of meniscal roots and of structures as seen for femoral trochlear rate 2-dimensional datasets and SNR for while in 2D-TSE-fs parts of those are masked by partial volume effects.
radial or complex meniscal tears whose cartilage (Fig. 6) and the anterior cruciate 1 mm reconstructions is similar to con-
Continued on page 90
14 MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world 15

10. Clinical Orthopedic imaging Orthopedic imaging Clinical
4 6
4 Axial recon- 6 Axial recon-
structed syngo structed syngo
SPACE1mm and 2D- SPACE1mm and
TSE-fs of a patient 2D-TSE-fs of a
with a bucket patient with a
handle tear. trochlear cartilage
SPACE provides delamination. In
better visualiza- strictly axially
tion of the config- acquired 2D-TSE-
uration of the fs (Topo A/B and
bucket handle series A) the fem-
tear as compared oral trochlear car-
to T2w-2D-TSE-fs, tilage is partially
in which delinea- blurred because
tion is impaired of partial volume
due to partial vol- effects, whereas
2D-TSE-fs 3 mm SPACE 1 mm ume effects. depiction in
SPACE1mm (B/C) is
sharper. Axially
reconstructed
Topo A/B axial SPACE1mm (Topo
A/B and series B)
5 is able to cover
5 Coronal, sagit-
the trochlear car-
tal and axial
tilage on more
reconstructed
slices than
syngo SPACE1mm
2D-TSE-fs
images (row A)
enabling a more
show a good
detailed depic-
delineation of a
tion. MPR perpen-
horizontal tear
dicular to the
within the medial
trochlear cartilage
meniscus which
(Topo C and series
approach the
C) even allows a
quality of the
clearer depiction
T2w-2D-TSE-fs
of cartilage
sequence (row B)
height and the
and provide an
lesion.
even clearer
depiction of the
lesion’s borders Topo C
and its extent. perpendicular
to trochlea
A SPACE 1 mm
B 2D-TSE-fs 3 mm A B C
16 MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world 17

11. Clinical Orthopedic imaging
Continued from page 86
7A 7B 7C
7 Anterior
Musculoskeletal
cruciate liga-
ment of a
healthy volun-
teer (sequence
Advisory Board
A–F). syngo
SPACE1mm tilted
to the course of
the anterior
Provides Protocols
cruciate liga-
ment allows
excellent trace-
ability of the
for 1.5 and
two bundles.
3T MAGNETOM
7D 7E 7F
systems
We have launched the MSK Advisory
Board website, providing proven MSK
protocols (.edx files) for download.
To support Technologists there are
also coil positioning videos and
tips & tricks.
Board members are:
ventional 2D-TSE-fs. The identification 2 Protocols for MRI examinations of joints and versus that at 1.5 T: preliminary results in a ■ Christian Glaser, LMU Grosshadern, Germany
of anatomical structures at least equals the spine as recommended by the AG MSK of porcine model. Radiology, 2005. 236(1): ■ Jürg Hodler, Balgrist University Hospital, Switzerland
the Deutsche Röntgengesellschaft. Rofo, 2006. p. 140-50.
the conventional sequence and allows ■ Young-Jo Kim, Harvard Medical School, Children’s
178(12): p. 128-130. 9 Link, T.M., et al., 3.0 vs 1.5 T MRI in the detec-
superior discrimination of relevant small 3 Mosher, T.J., Musculoskeletal imaging at 3T: tion of focal cartilage pathology--ROC analysis
Hospital Boston, USA
ligamentous structures. current techniques and future applications. in an experimental model. Osteoarthritis Carti- ■ Tallal Charles Mamisch, Bern University, Switzerland
These data suggest that a protocol com- Magn Reson Imaging Clin N Am, 2006. 14(1): lage, 2006. 14(1): p. 63-70. ■ Michael Recht, New York University, USA
prising 1 mm syngo SPACE reconstruc- p. 63-76. 10 Notohamiprodjo, M., et al., A New Approach for ■ Siegfried Trattnig, AKH Wien, Austria
4 Mugler, J.P., 3rd, et al., Optimized single-slab High Resolution MRI of the Knee at 3T – Evalua-
tions in three orientations would be use- ■ Lawrence M. White, University of Toronto, Canada
three-dimensional spin-echo MR imaging of the tion of a Moderately T2-Weighted 3D-TSE-Fs
ful for clinical evaluation. The additional brain. Radiology, 2000. 216(3): p. 891-9. (SPACE) Sequence, Investigative Radiology,
possibility of free 3-dimensional recon- 5 Lichy, M.P., et al., Magnetic resonance imaging 2009 44(9): p.585-597.
struction depending on the specific of the body trunk using a single-slab, 3-dimen-
clinical need may become useful for the sional, T2-weighted turbo-spin-echo sequence Contact
with high sampling efficiency (SPACE) for high Annie Horng, M.D.
diagnosis of difficult anatomical situa-
spatial resolution imaging: initial clinical experi- Department of Clinical Radiology
tions and presurgical planning, i.e. for ences. Invest Radiol, 2005. 40(12): p. 754-60. University Hospitals Munich –
traumatic ligamenteous lesions or com- 6 Isoda, H., et al., MRCP imaging at 3.0 T vs. 1.5 T: Campus Grosshadern
plex meniscal tears. preliminary experience in healthy volunteers. J Marchioninistrasse 15
Magn Reson Imaging, 2007. 25(5): p. 1000-6. 81377 Munich Visit us at
7 Alsop, D.C., The sensitivity of low flip angle RARE Germany
References imaging. Magn Reson Med, 1997. 37(2):
1 Glaser, C., et al., [Meniscus and ligament p. 176–84.
Phone: +49 89 7095 3620
www.siemens.com/magnetom-world
annie.horng@med.uni-muenchen.de
injuries]. Radiologe, 2006. 46(1): p. 26-35. 8 Masi, J.N., et al., Cartilage MR imaging at 3.0
18 MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world 19

12. Clinical Orthopedic imaging Orthopedic imaging Clinical
Case Reports:
1A
1A Sagittal
PDw STIR.
Musculoskeletal MRI
in Sports Medicine
Heinz-Peter Schlemmer1,3; Tina Holder1; T. Nägele1,2; Claus D. Claussen1,2
1
Radiologie SpOrt Stuttgart, Germany
2
University Hospital of Tuebingen, Germany
3
German Cancer Research Center, Heidelberg, Germany
Background
and as a consequence insufficient train- with pain at training but without corre-
In contrast to its role in oncology, for ing and competitive performances, of sponding trauma are shown.
example, MR imaging in sports medicine athletes can have various causes and are All MRI exams shown in this article were
deals primarily with healthy and young one of the main indications for MRI. performed at 1.5 Tesla (MAGNETOM
individuals. This imaging technique is an Because of the importance of the results Avanto).
1B
invaluable tool in sports medicine, not of such MR exams to the athlete, the 1B Coronal
only because of its excellent soft tissue interdisciplinary approach is one of the Case 1 PDw STIR
contrast but also because of its non- key elements for optimal and responsi- 22-year-old male soccer player with
invasive and non-ionizing nature. While ble treatment and support. Conse- severe knee pain after traumatic knee
imaging in recreational sports is mainly quently, all cases shown in this article injury.
limited to an evaluation of the effects of were examined at and treated by an An extensive effusion is obvious. Already
severe traumatic events, such as a skiing association of different facilities includ- suspected by clinical signs, oedema and
accident, the role of musculoskeletal ing the Department of Sports Medicine tear of the anterior cruciate ligament
MRI in case of competitive sports is at Tuebingen University and the Olympic supports the diagnosis of a complete
much more extensive. For example, fol- Training Center Stuttgart. rupture of the ligament (arrowhead in
lowing an injury, MRI is also used to In this article, we present a selection of Fig. 1A). In addition, a horizontally
assist in the detailed evaluation of the cases. Whilst not being a representative shaped oedema within the dorsal medial
degree of performance impairment of selection, they do reflect very well the meniscus supports the suspicion of a
the athlete – with direct impact on treat- variety and range of musculoskeletal horizontal (smaller) meniscal tear (arrow
ment / training actions taken for effec- imaging in sports medicine: in cases 1 in Fig. 1B).
tive fast and full recovery. This implies and 2, MRI was used to support the clini- Images were acquired with the dedi-
that time-to-diagnosis and the easy cal diagnoses of ligamental tear / rup- cated CP extremity coil. Sequence
access and availability of MRI scan time ture. Cases 3 through 5 show typical parameters for the shown images were:
(also for follow-up exams) is important patterns of muscular tears and haema- ■ Sagittal PDw TSE with spectral fat
for these athletes. One should also take toma after trauma. In these patients, suppression: TR / TE = 3754 / 37 ms,
into account that the pattern and sever- MRI was used to evaluate the involve- SL 1.5 mm, FOV 140 x 140 mm,
ity of injuries can differ between recre- ment and extension of the different Matrix 269 x 384 px
ational and competitive sports. This has muscles to evaluate and quantify the ■ Coronal PDw TSE with spectral fat
a direct impact on the indication to MRI degree of performance impairment and suppression: TR / TE = 3754 / 37 ms,
and the required knowledge of technolo- to provide information for further reha- SL 1.5 mm, FOV 140 x 140 mm,
gists and radiologists. But also non-trau- bilitation training. And finally, in cases 6 Matrix 269 x 384 px
matic pain and limitation of mobility, and 7, the images of two young athletes
20 MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world 21

13. 2F
Clinical Orthopedic imaging
2A 2B 2C 2F–H 2nd
Case 2
trauma: trans-
23-year-old male soccer player after versal T2w
TSE.
traumatic ankle injury.
MRI demonstrates a complete rupture of
the anterior fibulotalar ligament (arrow
in Figs. 2A–C). In addition, a partial
rupture of the anterior syndesmosis liga-
ment is present and an extensive oedema
and haemorrhage of the surrounding
soft tissue can also be observed (Figs. 2D
and E). 7 months after the initial trau-
matic event, another trauma of similar
type occurred. Follow-up MRI showed a
thickened but now continuous fibulota-
lar ligament and ventral syndesmosis 2G
(arrowheads in Figs. 2D–F); no re-rup-
ture was found.
Images were acquired with the 4-chan-
2A–C 1st trauma: transversal T2w TSE. nel flex coil. Sequence parameters for
the shown images were:
■ Transversal T2w TSE: TR / TE = 5259 /
85 ms, SL 3 mm, FOV 140 x 140 mm,
2D 2E
Matrix 269 x 448 px
■ Coronal PDw STIR: TR / TE / TI = 6500 /
29 / 160 ms, SL 3 mm, FOV 160 x
160 mm, Matrix 169 x 256 px
■ Coronal T1w TSE: TR / TE = 537 / 9.5
ms, SL 3 mm, FOV 111 x 160 mm,
Matrix 250 x 448 px, TA
■ Follow-up MRI transversal T2w TSE:
TR / TE = 4110 / 105 ms, SL 3 mm, FOV
110 x 110 mm, Matrix 512 x 512 px
2H
2D 1st trauma: coronal PDw STIR. 2E 1st trauma: coronal T1w TSE.
22 MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world MAGNETOM Flash · 4/2010 · www.siemens.com/magnetom-world 23