SPINE Volume 31, Number 6, pp E175–E178
©2006, Lippincott Williams & Wilkins, Inc.
An MRI Investigation Into the Function of the
Transversus Abdominis Muscle During “Drawing-In”
of the Abdominal Wall
Julie Hides, PhD,* Stephen Wilson, PhD,† Warren Stanton, PhD,* Shaun McMahon, PhD,‡
Heidi Keto, BPhty,* Katie McMahon, PhD,§ Martina Bryant, B App Sc,§
and Carolyn Richardson, PhD*
Study Design. An operator blinded dual modality trial
of measurement of the abdominal muscles during “draw-
ing-in” of the abdominal wall.
Objectives. 1) To investigate, using magnetic reso-
nance imaging (MRI), the function of the transversus
abdominis muscle bilaterally during a drawing-in of the
abdominal wall. 2) To validate the use of real-time ultra-
sound imaging as a measure of the deep abdominal mus-
cle during a drawing-in of the abdominal wall.
Summary of Background Data. Previous research has
implicated the deep abdominal muscle, transversus ab-
dominis, in the support and protection of the spine and
provided evidence that training this muscle is important
in the rehabilitation of low back pain. One of the most
important actions of the transversus abdominis is to
“draw-in” the abdominal wall, and this action has been
shown to stiffen the sacroiliac joints. It is hypothesized
that in response to a draw in, the transversus abdominis
muscle forms a deep musculofascial “corset” and that
MRI could be used to view this corset and verify its mech-
anism of action on the lumbopelvic region.
Methods. Thirteen healthy asymptomatic male elite
cricket players aged 21.3 ⫾ 2.1 years were imaged using
MRI and ultrasound imaging as they drew in their abdomi-
nal walls. Measurements of the thickness of the transversus
abdominis and internal oblique muscles and the slide of the
anterior abdominal fascia were measured using both MRI
and ultrasound. Measurement of the whole abdominal
cross-sectional area (CSA) was conducted using MRI.
Results. Results of the MRI demonstrated that, as a
result of draw-in, there was a significant increase in thick-
ness of the transversus abdominis (P ⬍ 0.001) and the
internal oblique muscles (P ⬍ 0.001). There was a signif-
icant decrease in the CSA of the trunk (P ⬍ 0.001). The
mean slide (⫾SD) of the anterior abdominal fascia was
From the *School of Health and Rehabilitation Sciences (Division of
Physiotherapy), †School of IT and Electrical Engineering, and §Centre
for Magnetic Resonance Imaging, University of Queensland, Brisbane,
Australia; and ‡Commonwealth Bank Cricket Academy, Brisbane,
Australia.
Acknowledgment date: February 11, 2005. First revision date: May 25,
2005. Second revision date: July 28, 2005. Acceptance date: August 2,
2005.
Supported by Cricket Australia Sports Science Medicine Research Pro-
gram.
The manuscript submitted does not contain information about medical
device(s)/drug(s).
Other funds were received in support of this work. No benefits in any
form have been or will be received from a commercial party related
directly or indirectly to the subject of this manuscript.
Address correspondence and reprint requests to Julie Hides, PhD, Di-
vision of Physiotherapy, School of Health and Rehabilitation Sciences,
University of Queensland, Brisbane, Australia, 4072. E-mail: j.hides@
shrs.uq.edu.au
1.54 ⫾ 0.38 cm for the left side and 1.48 ⫾ 0.35 cm for the
right side. Ultrasound measurements of muscle thickness
of both transversus abdominis and the internal oblique,
as well as fascial slide, correlated with measures obtained
using MRI (interclass correlations from 0.78 to 0.95).
Conclusions. The MRI results demonstrated that during
a drawing-in action, the transversus abdominis contracts
bilaterally to form a musculofascial band that appears to
tighten (like a corset) and most likely improves the stabili-
zation of the lumbopelvic region. Real-time ultrasound im-
aging can also be used to measure changes in the transver-
sus abdominis during the draw-in maneuver.
Key words: transversus abdominis, abdominal mus-
cles, lumbopelvic stabilization, magnetic resonance imag-
ing, real-time ultrasound imaging. Spine 2006;31:E175–E178
Over the last 10 years, low back pain researchers have
shown an increasing interest in the function of the ab-
dominal muscle synergists in the protection of the lum-
bopelvic region. Recently, researchers have focused on
the role of the transversus abdominis muscle as it is the
only muscle that consistently attaches into the thoraco-
lumbar fascia, a structure considered to assist in the sup-
port of the lumbar spine.1 In 1997, Hodges and Richard-
son2 studied the role of transversus abdominis in the
stabilization of the lumbar spine from a motor control
perspective and demonstrated an independence of action
of the transversus abdominis muscle in normal subjects
relative to the other abdominal muscles and differences
in this control between subjects with and without a his-
tory of low back pain.
Clinical muscle testing of the transversus abdominis
muscle has been based on its anatomic structure and
horizontal fiber arrangement. Two muscle tests include
observation of the abdominal wall during either a cogni-
tive “drawing-in” of the abdominal wall,3 or by using
automatic responses of the muscle to expiration.4 The
action of the bilateral muscle bellies of transversus abdo-
minis can be viewed using real-time ultrasound.5 Each
muscle belly thickens as well as shortens in length during
this action, to give the appearance of a deep muscle “cor-
set.” A recent randomized controlled trial showed that
using ultrasound imaging to provide feedback of trans-
versus abdominis muscle activation was superior to clin-
ical instruction alone in normal subjects.6 Clinical trials
that have focused on re-educating transversus abdominis
muscle function through its “draw-in” action have been
successful in decreasing lumbopelvic pain.7–9
E175
E176 Spine • Volume 31 • Number 6 • 2006

Three recent studies have found real-time ultrasound

imaging to be reliable for assessment of changes in thick-
ness of the abdominal muscles when they contract.10 –12
Other studies have related these thickness measures to
levels of muscle activation using fine wire EMG assess-
ments.13,14 Hodges et al13 showed that, for low level
isometric contactions, muscle thickness correlated with
EMG activity for transversus abdominis and internal
oblique but not for the external oblique muscle.
While ultrasound imaging has been successfully used
to measure the abdominal muscles, magnetic resonance
imaging (MRI) is considered a “gold standard” for view-
ing these muscles in cross section, both at rest and during
the draw-in maneuver. Ultrasound imaging, despite a
limited field of view, may be more practical and just as
accurate. It is much less expensive; transversus abdomi-
nis muscle thickness is well visualized and correlates well
with fine-wire EMG.14
The aims of this study were to observe and assess the
pattern of activation of the muscles of the anterolateral
abdominal wall during the draw-in action, through the
use of MRI, and to validate real-time ultrasound mea-
sures obtained from the same subjects repeating the
maneuver.

Materials and Methods

This study involved MRI and ultrasound measures of abdom-
inal muscle function conducted on an eligible sample of 26 elite
cricketers attending a cricket academy. Thirteen subjects were
recruited into the study and 13 subjects were excluded due to a
history of low back pain. The mean age of the participants was
21.3 ⫾ 2.1 years, mean height was 174 ⫾ 5.8 cm, and weight
was 75.8 ⫾ 6.3 kg. This study was approved by the Human
Research Ethics Committee at the University of Queensland,
Brisbane, Australia. Subjects were instructed on the muscle test
(drawing-in action) of the transversus abdominis muscle before
undergoing MRI and ultrasound imaging. MRI imaging and
ultrasound imaging were performed in a hospital setting, on the
same day. Subjects were assessed first using MRI and then with
ultrasound imaging.
Procedures
MRI Assessment. Subjects were first screened for contrain-
dications to MRI by a medical practitioner. Subjects were po-
sitioned in supine lying with their hips and knees resting on a
foam wedge. While lying in the scanner, subjects were in-
structed on the muscle contraction required. They were asked
to draw in the abdominal wall without moving the spine. The
cross-sectional area (CSA) of the deep musculofascial system
was measured first at rest and then in the contracted state. A 1.5
Tesla Siemens Sonata MR system was used. A True FISP se-
quence using 14 ⫻ 7 mm contiguous slices centered on L3–L4
disc was used for static images. A cine sequence of one 7 mm
slice was taken during a contraction at a frame rate of 500
milliseconds for 7 seconds.
Rest images were performed in a breath-hold at midex-
piration (Figure 1a). Contraction images were commenced af-
ter vocal initiation of the contraction by the operator (Figure
Figure 1. a, MRI of the deep musculofascial corset at rest, showing
thickness measurements of the transversus abdominis (䡺) and inter-
nal oblique muscles (䡺 䡺), and the cross-sectional area of the trunk
(outlined). b, MRI of the deep musculofascial corset taken during the
abdominal draw-in action, showing the thickness measurements of
the transversus abdominis (䡺) and internal oblique muscles (䡺 䡺)
and the cross-sectional area of the trunk (outlined).
1b). The cine series was commenced before the initiation of a
contraction to ensure complete temporal coverage of the
maneuver.
Ultrasound Assessment. A Toshiba Aplio SSA770 (Nasu,
Japan) apparatus was used. A linear transducer with a fre-
quency of 7.5 MHz was used. Subjects were positioned in su-
pine lying as conducted for the MRI examination. A transverse
image of the anterolateral abdominal wall was obtained just
inferior to the level of the umbilicus for left and right sides.
Subjects were asked to perform the drawing-in action of the
abdominal wall, and ultrasound images were obtained at rest
and on contraction, one side at a time (Figure 2).
Measurements. Images from MRI and ultrasound assess-
ments were saved onto CD for later analysis using a measure-
ment software package on a laptop computer. Measurements
on MRI included (Figure 1):
Thickness of transversus abdominis and internal oblique at
rest (left and right)
Thickness of transversus abdominis and internal oblique on
contraction (left and right)
 “Drawing-In” of the Abdominal Wall • Hides et al E177

Table 1. Measurements of Thickness of the Transversus

Abdominis (TrA), Internal Oblique (IO), Slide of the
Transversus Abdominis Fascia, and Trunk Cross-sectional
Area (CSA) Obtained by Ultrasound Imaging and MRI
Left Right

Relaxed Contracted Relaxed Contracted

TrA thickness (cm)

MRI 0.71 ⫾ 0.16 1.08 ⫾ 0.25 0.68 ⫾ 0.17 1.04 ⫾ 0.27
Ultrasound 0.70 ⫾ 0.15 1.05 ⫾ 0.17 0.68 ⫾ 0.17 1.00 ⫾ 0.25
IO thickness (cm)
MRI 1.71 ⫾ 0.24 2.04 ⫾ 0.36 1.57 ⫾ 0.20 1.94 ⫾ 0.35
Ultrasound 1.65 ⫾ 0.24 1.93 ⫾ 0.32 1.54 ⫾ 0.18 1.89 ⫾ 0.30
TrA slide (cm)
MRI 0 1.54 ⫾ 0.38 0 1.48 ⫾ 0.35
Ultrasound 0 1.55 ⫾ 0.40 0 1.51 ⫾ 0.44
Values are mean ⫾ SD.

and slide of the transversus abdominis fascia was determined

by the interclass correlation (ICCs, Model 3) of the scores for
each muscle (transversus abdominis and internal oblique) in
each condition (relaxed and contracted) and for each side (left
and right).

Figure 2. a, Ultrasound image of the muscles of the right antero-
lateral abdominal wall at rest showing transversus abdominis
(TrA), internal oblique (IO), and external oblique (EO) muscles. b,
Ultrasound image of the muscles of the right anterolateral abdom-
inal on drawing-in the abdomen, showing the thickness measure-
ments of the transversus abdominis (䡺) and internal oblique
muscles (䡺 䡺) and slide measurement (䡺 䡺 䡺). Note the lateral
pull of transversus abdominis (3).
Slide of the transversus abdominis fascia (left and right)
CSA of the trunk at rest
CSA of the trunk on contraction
The same measurements were performed on the ultrasound
images with the exception of the trunk CSA measures (Figure
2). Measurements conducted on the MRI and ultrasound im-
ages were performed by two independent operators who were
blinded to the others results. Repeatability and reliability of
linear and CSA measurements obtained using ultrasound and
MRI have been previously reported for the trunk muscles.15,16
Statistical Analysis. Analysis of variance was used to exam-
ine change in the muscle dimensions of thickness (transversus
abdominis and internal oblique), slide (transversus abdominis),
and trunk CSA. Separate analyses were conducted for the MRI
and the ultrasound imaging variables with repeated measures
of relaxed versus contracted and left versus right for the trans-
versus abdominis and internal oblique muscles. Validity of ul-
trasound measures against MRI measures for muscle thickness
Results
Analysis of the MRI data revealed that the draw-in ma-
neuver resulted in a significant decrease in the CSA of the
trunk (F ⫽ 221.58, P ⬍ 0.001). The mean CSA of the
trunk at rest was 393.90 ⫾ 29.10 cm2 and this decreased
to 362.61 ⫾ 31.92 cm2. There was a corresponding sig-
nificant increase in thickness of the transversus abdomi-
nis muscle due to the draw-in action as measured by both
MRI (F ⫽ 50.34, P ⬍ 0.001) and ultrasound imaging
(F ⫽ 115.66, P ⬍ 0.001). Similarly, the draw-in maneu-
ver resulted in a significant increase in thickness of the
internal oblique muscles for both MRI (F ⫽ 25.42, P ⬍
0.001) and ultrasound imaging (F ⫽ 31.29, P ⬍ 0.001).
Descriptive data for transversus abdominis and inter-
nal oblique muscle thickness, slide of transversus abdo-
minis, and trunk CSA is shown in Table 1. Analysis of
variance confirmed that there was not a statistically sig-
nificant effect due to left versus right side. The ICCs for
measures of muscle thickness and slide obtained by MRI
and ultrasound were in the high range from 0.78 to 0.95
(Table 2).
Discussion
This study on normal, pain-free subjects demonstrated
that the action of drawing-in the abdominal wall resulted
Table 2. Interclass Correlations for Transversus
Abdominis and Internal Oblique Muscle Thickness as
Measured by Ultrasound Imaging and MRI
Left Right
Relaxed Contracted Relaxed Contracted
TrA 0.84 0.85 0.93 0.94
IO 0.91 0.93 0.93 0.95
Slide — 0.91 — 0.78
E178 Spine • Volume 31 • Number 6 • 2006
in a concentric contraction of the transversus abdominis
muscle bilaterally, in conjunction with bilateral slide of
the anterior abdominal fascia, an increase in thickness
of the internal oblique muscles, and a decrease in CSA of
the trunk. The shortening of the transversus abdominis
muscle appears to tension the fascial attachments of the
muscle anteriorly (anterior abdominal fascia) and poste-
riorly (thoracolumbar fascia). Thickening of the internal
oblique muscle as a result of the draw-in maneuver is
consistent with the findings presented by Richardson et
al.17 In this study, contraction of the transversus abdo-
minis muscle was accompanied by an increase in EMG
activity of the oblique abdominal muscles. This would
suggest that the draw-in action does not activate the
transversus abdominis muscle in isolation.
MRI is considered a “gold standard” for viewing the
abdominal muscles in cross section, both at rest and dur-
ing the draw-in maneuver. Given the high ICCs and sim-
ilarity in the mean scores in this study, it could be pro-
posed that the variables measured in this investigation
could be adequately assessed using ultrasound imaging.
This would be useful where large numbers of subjects are
to be investigated or where portability is an issue. As
muscle contraction has a voluntary component, it was
anticipated that measures of contracted muscles may
show more variability than measures taken at rest.10,11
In this study, ICCs for muscle thickness measurements
were similar for both conditions for the muscles mea-
sured. This result is promising for future investigations
that may wish to use ultrasound imaging to measure
change in muscle thickness on contraction.
It has previously been reported that change in transver-
sus abdominis muscle thickness measured by ultrasound
imaging was able to discriminate subjects with a history of
low back pain from normal subjects.12 While this measure
has been shown to be repeatable and reliable, the combina-
tion with additional measurements used in this study, in-
cluding slide of the anterior abdominal fascia and change in
thickness of the internal oblique muscle, may provide addi-
tional information in future studies.
One obvious limitation of this study was that it was
conducted on a very specific group of subjects, i.e., elite
male cricket players of a similar age and activity level.
Future research could include examination of a represen-
tative population group as well as differences across sub-
populations. Future studies would therefore incorporate
larger subject numbers and investigate the differences in
these measures between normal subjects and those with
different classifications of low back pain. Other issues
that could be examined include effects of variables such
as age, gender, and laterality and assessment of different
functions of the abdominal muscles. These measures
could also be incorporated into biomechanical models.
Acknowledgments
The authors thank the subjects studied, Toshiba Austra-
lia and GE Australia for provision of ultrasound imaging
equipment, Tom Johnsen (Scientific Officer), and Dr.
Fiona Mack.
Key Points
● Transversus abdominis is important in the support
and protection of the spine, and training this muscle is
important in the rehabilitation of low back pain.
● The transversus abdominis muscle forms a deep
musculofascial “corset,” and MRI can be used to
view this corset.
● In asymptomatic subjects, the action of draw-
ing-in the abdominal wall resulted in thickening of
the transversus abdominis and internal oblique
muscles, slide of the anterior abdominal fascia, and
a decrease in cross-sectional area of the trunk.
● Measurements made using magnetic resonance
imaging correlated with measurements made using
real-time ultrasound imaging.
References
1. Tesh KM, Shaw Dunn J, Evans JH. The abdominal muscles and vertebral
stability. Spine 1987;12:501– 8.
2. Hodges PW, Richardson CA. Feedforward contraction of transversus abdo-
minis is not influenced by the direction of arm movement. Exp Brain Res
1997;114:362–70.
3. Richardson CA, Jull GA. An historical perspective on the development of
clinical techniques to evaluate and treat the active stabilizing system of the
lumbar spine. Aust J Physiother Monogr 1995;1:5–13.
4. Lacôte M, Clevalier AM, Miranda A, et al. Clinical Evaluation of Muscle
Function. Edinburgh: Churchill Livingstone 1987:290 –3.
5. Hides JA, Richardson CA, Jull GA. Use of real-time ultrasound imaging for
feedback in rehabilitation. Man Ther 1998;3:125–31.
6. Henry SM, Westervelt KC. The use of real-time ultrasound feedback in
teaching abdominal hollowing exercises to healthy subjects. J Orthop Sports
Phys Ther 2005;35:338 – 45.
7. Golby L, Moore A, Doust J, et al. A RCT investigating the efficacy of manual
therapy, exercises to rehabilitate spinal stabilisation and an education booklet in
the conservative treatment of chronic low back pain. In: Proceedings of the 1st
International Conference on Movement Dysfunction. Edinburgh: 2001, p19.
8. Stuge B, Veierød MB, Laerum E, et al. The efficacy of a treatment program
focusing on specific stabilizing exercises for pelvic girdle pain after preg-
nancy: a two year follow up of a randomized clinical trial. Spine 2004;29:
E197–203.
9. O’Sullivan PB, Twomey LT, Allison GT. Evaluation of specific stabilizing
exercise in the treatment of chronic low back pain with radiologic diagnosis
of spondylolysis or spondylolisthesis. Spine 1997;22:2959 – 67.
10. Misuri G, Colagrande S, Gorini M, et al. In vivo assessment of respiratory
function of abdominal muscles in normal subjects. Eur Respir J 1997;10:
2861–7.
11. Bunce SM, Moore AP, Hough AD. M-mode ultrasound: a reliable measure
of transversus abdominis thickness? Clin Biomech 2002;17:315–7.
12. Ferreira P, Ferreira M, Maher C, et al. Clinical ultrasound test for transver-
sus abdominis thickness: investigation of reliability. Proceedings of the Mus-
culoskeletal Physiotherapy Australia 13th Biennial Conference. Sydney,
2003:70.
13. Hodges PW, Pengel LHM, Herbert RD, et al. Measurement of muscle con-
traction with ultrasound imaging. Muscle Nerve 2003;27:682–92.
14. McMeekan JM, Beith ID, Newham DJ, et al. The relationship between EMG
and change in thickness of transversus abdominis. Clin Biomech 2004;19:
337– 42.
15. Hides JA, Cooper DH, Stokes MJ. Diagnostic ultrasound imaging for mea-
surement of the lumbar multifidus muscle in normal young adults. Phys-
iother Theory Pract 1992;8:19 –26.
16. Hides JA, Richardson CA, Jull GA. Magnetic resonance imaging and ultra-
sonography of the lumbar multifidus muscle. Spine 1995;20:54 – 8.
17. Richardson CA, Snijders CJ, Hides JA, et al. The relationship between the
transversus abdominis muscles, sacroiliac joint mechanics and low back
pain. Spine 2002;27:399 – 405.