European Journal of Radiology 57 (2006) 182186

A new look at the fetus: Thick-slab T2-weighted

sequences in fetal MRI
Peter C. Brugger a, , Christoph Mittermayer b , Daniela Prayer c
a
c

Center of Anatomy and Cell Biology, Integrative Morphology Group, Medical University of Vienna, Vienna, Austria
b Department of Neonatology and Intensive Care, University Hospital of Vienna, Austria
Department of Neuroradiology, University Clinics of Radiodiagnostics, Medical University of Vienna, Vienna, Austria
Received 11 November 2005; received in revised form 14 November 2005; accepted 16 November 2005

Abstract
Although magnetic resonance imaging (MRI) of the fetus is considered an established adjunct to fetal ultrasound, stacks of images alone
cannot provide an overall impression of the fetus. The present study evaluates the use of thick-slab T2-weighted MR images to obtain a
three-dimensional impression of the fetus using MRI. A thick-slab T2-weighted sequence was added to the routine protocol in 100 fetal MRIs
obtained for various indications (19th to 37th gestational weeks) on a 1.5 T magnet using a five-element phased-array surface coil. Slice
thickness adapted to fetal size and uterine geometry varied between 25 and 50 mm, as did the field of view (250350 mm). Acquisition of one
image took less than 1 s. The pictorial essay shows that these images visualize fetal anatomy in a more comprehensive way than is possible
with a series of 34 mm thick slices. These thick-slab images facilitate the assessment of the whole fetus, fetal proportions, surface structures,
and extremities. Fetal pathology may be captured in one image. Thick-slab T2-weighted images provide additional information that cannot
be gathered from a series of images and are considered a valuable adjunct to conventional 2D MR images.
2005 Elsevier Ireland Ltd. All rights reserved.
Keywords: MR imaging; Fetus; Fetal MRI; Technique

1. Introduction
In the past several years, the use of magnetic resonance
imaging (MRI) of the fetus has been increasingly reported as
an adjunct to ultrasound in the evaluation of fetal pathologies
[14]. High soft-tissue contrast, high resolution, the multiplanar imaging capabilities, and the capability to visualize the
whole fetus, even in late stages of pregnancy, have been cited
as advantages of this method compared to ultrasound [5].
Yet, an overall impression of the fetus is hardly provided by
stacks of images alone and great expertise is required to compile a large number of sections into a 3D mental model [6].
Thus, attempts were also made to perform three-dimensional
reconstructions based on in vivo fetal MRI [69]. However,
fetal motion and maternal breathing movements can limit
the acquisition of a continuous series of slices. Moreover,

Corresponding author. Tel.: +43 1 4277 61163; fax: +43 1 4277 61142.
E-mail address: peter.brugger@meduniwien.ac.at (P.C. Brugger).

0720-048X/$ see front matter 2005 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.ejrad.2005.11.018

the relatively large slice thickness (35 mm), as well as large

intersection gaps in relation to fetal size, introduce additional
difficulties such as steps in the reconstructed images [6,10],
which, therefore, lack a smooth contour. In particular, the
evaluation of the fetal extremities may be complicated, as
extremity movements are more frequent than body movements, and thus, may not be well depicted on 34 mm thick
slices in continuity [11]. The present study evaluates the use
of thick-slab T2-weighted images in fetal MRI as an alternative to three-dimensional reconstructions.

2. Materials and methods

Fetal MRI examinations were performed on a 1.5 superconducting unit (Philips Medical Systems, Best, The Netherlands) using a five-element phased-array surface coil. MR
examinations were performed for clinical purposes (CNS
and extra-CNS indications), and written, informed consent

P.C. Brugger et al. / European Journal of Radiology 57 (2006) 182186

183

was obtained. No sedation of either mother or fetus was

used. In 100 fetal MRIs (fetuses from 19th to 37th gestational weeks (GW) of age) made between July 2003 and
August 2004, a thick-slab T2-weighted, single-shot turbo
spin echo sequence (repetition time (TR)/echo time (TE):
8000/400800 ms; one signal acquired with a 256 256
matrix; field of view (FOV): 250350 mm; flip angle 90 ;
echo train length: 205) was added to the routine protocol [12].
Slice thickness was adapted to fetal size or the body part in
question and varied between 25 and 50 mm. The slab was
preferably planned in the sagittal plane, and care was taken
not to include the placenta or the uterine wall. The acquisition
of one image took less than 1 s.

3. Results
On images obtained with this method, the whole fetus
(Figs. 18) could be recognized at once due to the large FOV
and large slice thickness. This eases the assessment of fetal
proportions (Figs. 13 and 5) and position (Fig. 6). Superb
contrast between amniotic fluid and the fetus and high resolution provide a smooth fetal surface and visualization of
anatomic detail. Small structures, such as myelomeningoceles (Fig. 1) and large external masses (Fig. 2), can be depicted.
The size and shape of fetal extremities (Figs. 13 and 5) are
displayed on a single image. In addition, fluid-filled cavities (stomach, urinary bladder, small intestines) and pathologies are recognizable, shining through the fetal surface
(Figs. 4 and 8). Using a lower TE enhances this effect (Fig. 8a
and b). Pathological fluid accumulations such as hydrops

Fig. 2. Fetus (20 + 2 GW) in breech presentation, 35 mm slice thickness, TE

800. The size of the multicystic sacrococcygeal teratoma in relation to the
fetus is easily appreciated. The urinary bladder is in a normal position and
not compressed.

Fig. 3. Fetus (35 + 1 GW) with thanatophoric dwarfism. Two radial 40 mmthick slices. (a) Note short and thickened extremities with short digits,
external genitals. (b) Frontal bossing is evident.

and pleural effusions (Fig. 4) or ovarial cysts (Fig. 8) could

be delineated. In addition, the wide fluid-filled subarachnoid
space allowed assessment of the gyral pattern, as well as the
size and content of the posterior fossa (Figs. 2, 3 and 57).
Furthermore, the configuration, shape, and course of the
umbilical cord were displayed (Figs. 1 and 4).

Fig. 1. Fetus at 20 + 1 GW with Chiari II malformation; 40 mm slice thickness. The small sacral myelomeningocele is visible. Both legs are well
depicted and normal in shape. Colpocephaly and the small posterior fossa
are evident (compare to normal cerebral anatomy in Fig. 2).

4. Discussion
Originally developed for magnetic resonance cholangiopancreaticography (MRCP) [13,14], this heavily T2-

184

P.C. Brugger et al. / European Journal of Radiology 57 (2006) 182186

Fig. 4. Fetus at 23 + 1 GW, 40 mm slice thickness. Fetal hydrops is most

pronounced in the trunk and neck. Pleural effusions are recognizable as
intrathoracic hyperintensities, while the hypoplastic lungs show low signal
intensity.

weighted sequence planned as a thick-slab allows a new look

at the fetus. However, while the surrounding tissues detail the
fluid-filled ducts in MRCP, this principle is reversed in fetal
imaging. The long T2 time of amniotic fluid, with little signal
from the fetal soft tissues, provides optimal contrast. Due to
summation of fetal tissues and amniotic fluid in the respective region of the volume, a spatial impression is generated:
structures surrounded by much amniotic fluid appear more
distant. This three-dimensional impression can be enhanced
by acquiring radial stacks, which can then be rotated when
viewed on a computer workstation. While a slice thickness
of at least 15 mm is necessary to generate a 3D impression,
maximum slice thickness is dictated by fetal size, the position of the fetus within the uterus, and the dimensions of the

Fig. 5. Fetus at 24 + 1 GW with severe intrauterine growth retardation

(30 mm slice thickness). The marked disproportion between head and body
is evident. Short umbilical cord.

Fig. 6. Stuck twin (19 + 3 GW) in a twin-to-twin transfusion syndrome,

50 mm slice thickness. The fetus shows an unnatural position with flexed
thighs. The stomach is filled with fluid but not the urinary bladder. Placental
vessels converging in the fetus are recognizable in the left margin. The wide
lateral ventricles are typical for this age. The head of the co-twin is also
visible in axial plane.

Fig. 7. (a) Paramedian slab through a 27 GW fetus, 30 mm slice thickness.

Cortical folding corresponds to gestational age: frontal, pre- and post-central
gyri are evident, and the insula is not yet covered by its opercula. (b)
32 + 1 GW fetus (25 mm slice thickness). A more complex gyral pattern
is present.

P.C. Brugger et al. / European Journal of Radiology 57 (2006) 182186

185

Fig. 8. Fetus at 32 + 2 GW with ovarian cyst. (a) TE 800 ms, (b) TE 400, same slice thickness (50 mm). Lower TE enhances transparency. Thoracic organs and
bowel are better recognized. (c) TE 400 ms with a slice thickness of 70 mm. Placenta and uterus are included and projected onto the fetus.

amniotic cavity. The use of very thick slices (80 mm) has been
suggested [15], but very thick slices alone have included the
placenta or uterus and have led to additional artifacts (Fig. 8c).
Therefore, uterine geometry must be considered in planning
this sequence. This may be achieved by using a survey scan
in three orthogonal section planes. When using more than
one slice, these may either be planned in a parallel fashion
or radially. This is advisable in larger fetuses to avoid summation of anatomic detail. The fast acquisition of one image
(less than 1 s) makes degradation by fetal motion unlikely.
The only limitation for visualization of the fetal surface is
the presence of enough amniotic fluid, which is essential for
image contrast.
Compared to 3D ultrasound, thick-slab T2-weighted
imaging is not limited by the FOV. Therefore, larger fetuses
can be imaged with this method. Moreover, while 3D ultrasound displays only fetal surface anatomy or bones [16],
the graded spectrum of shades of gray provides a shinethrough effect for the inner fetal organs as well. The transparency shows both normal (e.g., stomach, urinary bladder)
and abnormal fluid collections (cysts, pleural and pericardial
effusions) within the fetal body, and aids in the assessment
of cortical development and gyration [1,12].
The visualization of extremities on typical stacks of
images may be impaired by motion and complicated because
of the missing continuity of the limb [11]. Thick-slab images
can show the limb as a whole, combined with information
about form and position. This may be particularly helpful in
situations in which the limbs are deformed (dwarfism, conditions that restrict movement either because of a neurological
disease or CNS malformation, or external reasons, such as
membranes).
The thick-slab imaging technique extends the repertoire
of sequences available for fetal MRI. The large FOV offers
a global overview of fetal pathology or anatomy. The entire
fetus can be visualized, even in the third trimester, fetal proportions are more readily recognized, and pathology can be
captured in one image. The use of thick-slab T2-weighted
images can overcome the difficulties currently associated
with 3D reconstructions of in vivo fetal MRI data, espe-

cially the time-consuming segmentation [6,8,10]. Thus, the

thick-slab method is considered an alternative to 3D reconstructions, and, in addition, supplies information that, to date,
is unavailable with 3D reconstructions.
This imaging method is especially useful in those fetuses
in which an overall morphological impression is benefit in the
perception of fetal anomalies (e.g., external masses, fetal proportions, extremities, arthrogryphosis not related to oligohydramnios, pathological fluid accumulations within the fetus).
The combination of overall appearance and translucency provides a new look at the fetus. Finally, the images obtained
with this method can communicate information to an audience that may not be familiar with sectional anatomy and thus
can contribute to prenatal counseling.
Acknowledgments
The authors wish to thank J.R. Lindner and Fritz Stuhr for
technical assistance in sequence development.
References
[1] Garel C, Brisse H, Sebag G, et al. Magnetic resonance imaging of
the fetus. Pediatr Radiol 1998;28:20111.
[2] Hubbard AM, States LJ. Fetal magnetic resonance imaging. Top
Magn Reson Imag 2001;12:93103.
[3] Breysem L, Bosmans H, Dymarkowski S, et al. The value of fast
MR imaging as an adjunct to ultrasound in prenatal diagnosis. Eur
Radiol 2003;13:153848.
[4] Coakley FV, Glenn OA, Qayyum A, et al. Fetal MRI: a developing technique for the developing patient. AJR Am J Roentgenol
2004;182:24352.
[5] Levine D. Magnetic resonance imaging in prenatal diagnosis. Curr
Opin Pediatr 2001;13:5728.
[6] Kubik-Huch RA, Wildermuth S, Cettuzzi L, et al. Fetus and uteroplacental unit: fast MR imaging with three-dimensional reconstruction
and volumetry feasibility study. Radiology 2001;219:56773.
[7] Schierlitz L, Dumanli H, Robinson JN, et al. Three-dimensional magnetic resonance imaging of fetal brains. Lancet 2001;357:11778.
[8] Hata N, Wada T, Chiba T, et al. Three-dimensional volume rendering of fetal MR images for the diagnosis of congenital cystic
adenomatoid malformation. Acad Radiol 2003;10:30912.

186

P.C. Brugger et al. / European Journal of Radiology 57 (2006) 182186

[9] Keller TM, Rake A, Michel SC, et al. MR assessment of fetal lung
development using lung volumes and signal intensities. Eur Radiol
2004;14:9849.
[10] Levine D. Three-dimensional fetal MR imaging: will it fulfill its
promise? Radiology 2001;219:3135.
[11] Trop I, Levine D. Normal fetal anatomy as visualized with fast
magnetic resonance imaging. Top Magn Reson Imag 2001;12:317.
[12] Prayer D, Brugger PC, Prayer L. Fetal MRI: techniques and protocols. Pediatr Radiol 2004;34:68593.
[13] Laing AD, Gibson RN. Magnetic resonance cholangiopancreatography. Australas Radiol 1999;43:28493.

[14] Vitellas KM, Keogan MT, Spritzer CE, et al. MR cholangiopancreatography of bile and pancreatic duct abnormalities with emphasis on the single-shot fast spin-echo technique. Radiographics
2000;20:93957.
[15] Martin C, Darnell A, Duran C, et al. Magnetic resonance imaging of the intrauterine fetal genitourinary tract. Abdom Imag
2004;29:286302.
[16] Lee W. 3D fetal ultrasonography. Clin Obstet Gynecol
2003;46:85067.