In Vivo Intravoxel Incoherent Motion Measurements in
the Human Placenta Using Echo-Planar Imaging at 0.5 T
R.J. Moore,
1
B. Issa,
1
P. Tokarczuk,
1
K.R. Duncan,
2
P. Boulby,
1
P.N. Baker,
2
R.W. Bowtell,
1
B.S. Worthington,
1
I.R. Johnson,
2
and P.A. Gowland
1
*
This paper presents the first in vivo measurements of intravoxel
incoherent motion in the human placenta, obtained using the
pulsed gradient spin echo (PGSE) sequence. The aims of this
study were two-fold. The first was to provide an initial estimate
of the values of the IVIM parameters in this organ, which are
currently unknown. The second aim was then to use these
results to optimize the sequence timings for future studies. The
moving blood fraction (f), diffusion coefficient (D), and pseudo-
diffusion coefficient (D*) were measured. The average value of
f was 26 6 % (mean SD), D was 1.7 0.5 10
3
mm
2
/sec,
and D* was 57 41 10
3
mm
2
/sec. For the optimized values
of b, the expected percentage uncertainty in the fitted values of
f, D, and D* for the placenta were f/f 14.9%, D/D 14.3%,
D*/D* 44.9%, for an image signal-to-noise of 20:1, and a
total imaging time of 800 sec. Magn Reson Med 43:295–302,
2000. © 2000 Wiley-Liss, Inc.
Key words: IVIM; human placenta; EPI; optimization; PGSE
This work aims to demonstrate the ability of magnetic
resonance imaging to make noninvasive in vivo measure-
ments of intravoxel incoherent motion (IVIM) in the pla-
centa (1). Pilot data has been acquired and has then been
used to optimize the sequence for use in any future stud-
ies. This work forms part of a project to determine the
potential of echo-planar magnetic resonance imaging (EPI)
in assessing the compromised fetus. Intrauterine growth
restriction (IUGR) is a major cause of perinatal mortality
and morbidity. Pregnancies complicated by IUGR are
known to be associated with defective trophoblastic inva-
sion during placental development, which impedes utero-
placental blood flow. It is expected that blood movement
measurements made using this technique will eventually
provide a method for probing placental structure and func-
tion, with the aim of assisting in the prediction of IUGR.
There has been much deliberation over the interpreta-
tion of IVIM measurements, in particular over the link
between IVIM and classical perfusion (2,3,4). Classical
perfusion is a measure of the blood delivered to and used
by a specified mass of tissue, and is measured in units of
ml/min/100 g. It is often measured using spin labelling
techniques in MRI (5). In contrast, IVIM measures quasi
random blood movement within a single imaging voxel
and results in a bi-exponential signal attenuation in a
standard pulsed gradient spin echo (PGSE) experiment.
The pseudo-diffusion coefficient (D*) is associated with
perfusion and is measured in units of mm
2
/sec. Signal
attenuation due to D* is observed at lower values of b as it
relates to larger scale movement. The affects of diffusion,
quantified by the diffusion coefficient (D) and also mea-
sured in units of mm
2
/sec, are observed at higher values of
b and are related to smaller scale movement of water (6).
The value of f measures the total volume of blood moving
in the voxel compared to the total voxel volume and is
quoted as a percentage.
The PGSE sequence (7) has previously been used to
measure f, D, and D*, predominantly in the brain where it
has been criticized for its low sensitivity (8). However, the
expected volume of moving blood in the placenta is high,
reducing image signal-to-noise requirements and improv-
ing sensitivity to f and D*, suggesting that the placenta
would be an ideal site in which to use this sequence.
The development of obstetric MRI has been hampered
by the presence of motion artifacts in images of the fetus.
The unpredictable and nonperiodic nature of fetal move-
ment makes it impossible to use simple techniques such as
gating or retrospective gating to reduce these artifacts.
However, it is possible to produce good quality images of
the fetus using high-speed imaging techniques such as EPI
(9), FLASH (10), or HASTE (11) to freeze physiological
motion. A further advantage of high-speed imaging is that
it provides a method of obtaining quantitative results in a
reasonable imaging time, thereby minimising patient dis-
comfort or stress, which is particularly important when
scanning pregnant subjects. EPI takes 130 msec to form an
image as implemented here (approximate echo train
length), with an echo time to the origin of k-space of 35
msec. It is ideally suited to imaging the fetus because it
produces minimal RF power deposition. Furthermore, de-
spite the sensitivity of EPI to variations in magnetic sus-
ceptibility, there are no significant image artifacts in fetal
imaging because the susceptibility of the fetus is well
matched to the uterine environment.
In order to produce meaningful quantitative results in
vivo using MRI, the measurement parameters must always
be chosen to maximize the reproducibility of the measured
values. This is particularly important in this study because
pregnant subjects cannot lie comfortably in the scanner for
long periods and the work is conducted at low field
strength. Therefore, it is important to select the optimum
combination of diffusion sensitivity or b values to maxi-
mize the reproducibility in f, D, and D*. The measure-
1
Magnetic Resonance Centre, School of Physics and Astronomy, University
of Nottingham, UK.
2
Obstetrics and Gynaecology, City Hospital, University of Nottingham, UK.
Dr. B. Issa’s present address is Department of Physics, P.O. Box 17551, UAE
University, Al-Ain, United Arab Emirates.
Dr. P. Tokarczuk’s present address is Imaging Science and Biomedical En-
gineering, University of Manchester, Manchester, UK.
Dr. P. Boulby’s present address is MR Centre, National Society for Epilepsy,
Chalfont St. Peter, Buckinghamshire SL9 ORJ, UK.
This work was presented at the 5th Meeting of the ISMRMB in Vancouver,
April 1997.
*Correspondence to: Dr. P. A. Gowland, Magnetic Resonance Centre, Uni-
versity of Nottingham, Nottingham NG7 2RD, UK.
Received 17 July 1998; revised 18 October 1999; accepted 19 October 1999.
Magnetic Resonance in Medicine 43:295–302 (2000)
© 2000 Wiley-Liss, Inc. 295