The influence of simulated transversus abdominis muscle force on sacroiliac
joint flexibility during asymmetric moment application to the pelvis
Rafael Gnat, Kees Spoor, Annelies Pool-Goudzwaard
PII:
DOI:
Reference:

S0268-0033(15)00170-9
doi: 10.1016/j.clinbiomech.2015.06.006
JCLB 3986

To appear in:

Clinical Biomechanics

Received date:
Accepted date:

19 February 2015
8 June 2015

Please cite this article as: Gnat, Rafael, Spoor, Kees, Pool-Goudzwaard, Annelies, The
inuence of simulated transversus abdominis muscle force on sacroiliac joint exibility
during asymmetric moment application to the pelvis, Clinical Biomechanics (2015), doi:
10.1016/j.clinbiomech.2015.06.006

This is a PDF le of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its nal form. Please note that during the production process
errors may be discovered which could aect the content, and all legal disclaimers that
apply to the journal pertain.

ACCEPTED MANUSCRIPT
The influence of simulated transversus abdominis muscle force on sacroiliac

IP

a,b,c

, Kees Spoor, PhD , Annelies Pool-Goudzwaard, PhD

a,d

SC
R

Rafael Gnat, PhD

joint flexibility during asymmetric moment application to the pelvis

Department of Neuroscience, Erasmus MC, University Medical Centre Rotterdam, P.O. Box

NU

2040, 3000 CA Rotterdam, The Netherlands.

Motion Analysis Laboratory, Faculty of Physiotherapy, University of Physical Education, ul.

MA

Mikolowska 72, 40-065 Katowice, Poland.

Faculty of Physiotherapy, Academy of Business, ul. Cieplaka 1c, 41-300 DbrowaGrnicza,

Poland.

Research Institute MOVE, Faculty of Human Movement Sciences, VU University Amsterdam,

Rafael Gnat

CE
P

Corresponding author:

TE

Van der Boechorststraat, 9, 1081 BT Amsterdam, The Netherlands

University of Physical Education, Faculty of Physiotherapy

AC

ul. Mikolowska 72
40-065 Katowice
Poland

tel: +48 515 957 646; e-mail: rafal.gnat@interial.pl

Word count abstract: 262

Word count text: 3088

ACCEPTED MANUSCRIPT
ABSTRACT
Background

The role of so-called local muscle system in motor control of the lower back and

IP

pelvis is a subject of ongoing debate. Prevailing beliefs in stabilizing function of

SC
R

this system were recently challenged. This study investigated the impact of in
vitro simulated force of transversely oriented fibres of the transversus abdominis

NU

muscle (a part of the local system) on flexibility of the sacroiliac joint during
asymmetric moment application to the pelvis.

MA

Methods

In 8 embalmed specimens an incremental moment was applied in the sagittal

plane to one innominate with respect to the fixed contralateral innominate.

TE

Ranges of motion of the sacroiliac joint were recorded using the Vicon Motion

CE
P

Capture System. Load-deormation curves were ploted and flexibility of the

sacroiliac joint was calculated separately for anterior and posterior rotations of
the innominate, with and without simulated muscle force.

AC

Findings

Flexibility of the sacroiliac joint was significantly bigger during anterior rotation
of the innominate, as comapred to posterior rotation (Anova P<0.05). After
application of simulated force of transversus abdominis, flexibility of the joint did
not change both during anterior and posterior rotations of the innominate.
Interpretation
A lack of a stiffening effect of simulated transversus abdominis force on the
sacroiliac joint was demonstrated. Earlier hypotheses suggesting a stiffening
influence of this muscle on the pelvis cannot be confirmed. Consistent with

ACCEPTED MANUSCRIPT
previous findings smaller flexibility of the joint recorded during posterior rotation
of the innominate may be of clinical importance for physio- and manual

therapists. However, major limitations of the study should be acknowledged: in

SC
R

IP

vitro conditions and simulation of only solitary muscle force.

AC

CE
P

TE

MA

NU

Keywords: pelvis, sacroiliac joint, transversus abdominis, flexibility

ACCEPTED MANUSCRIPT
1. INTRODUCTION
The role of so-called local muscle system (Bergmark 1989) in motor

control of the lower back and pelvis is a subject of ongoing debate.

IP

Prevailing beliefs in an exclusive stabilizing function of this system as well as

SC
R

studies leading to the formulation of such ideas (e.g. Hodges and

Richardson 1997ab, 1998, Hodges and Gandevia 2000, Sapsford et al.

NU

2001ab, Ferreira et al. 2006, 2007, Tsao and Hodges 2007) were recently
challenged (Cleland et al. 2002, Koumantakis et al. 2005, Mills et al. 2005,

MA

Allison and Morris 2008, Mannion et al. 2008, Lederman 2010, Burns et al.
2011, Reeves et al. 2011, Hodges et al. 2013). Indeed, earlier evidence

suggested that local (deep) muscles of this region, inserting on or deriving

TE

from the lower lumbar spine and pelvis (e.g. transversus abdominis (TrA),

CE
P

lumbar multifidus, pelvic floor), have the potential to decrease flexibility and
thus stabilize the articular junctions. In case of the pelvis, Richardson et al.
(2002) reported that after voluntary contraction of the TrA, laxity of the

AC

sacroiliac joint decreases. Pool-Goudzwaard et al. (2004) provided partial

confirmation of this finding demonstrating decreased flexibility of the
sacroiliac joint during in vitro simulation of pelvic floor activity. However,
such an effect was limited to female specimens and occurred after
mimicking multiple pelvic floor muscles. In line with these studies, Pel et al.
(2008) created a biomechanical simulation model showing an increase in
compression of the sacroiliac joint linked to activation of transversely
oriented TrA fibres as suggested earlier by Snijders et al. (1995). In contrast
with these studies, recently Gnat et al. (2013) demonstrated lack of a

ACCEPTED MANUSCRIPT
stiffening effect of simulated TrA muscle force on the pubic symphysis in
vitro. In line with their results, also the claim of a stiffening effect of TrA on

the sacroiliac joint (SIJ) seems questionable, although the anatomy and

IP

hence biomechanical action of this muscle (Askar 1977, Rizk 1980)

SC
R

suggests a compressive effect on the SIJ, as partly demonstrated by

previous studies (Snijders et al. 1995, Richardson et al. 2002, Pool-

NU

Goudzwaard et al. 2004, Pel et al. 2008).To our knowledge no in vitro study
has proven the stiffening effect of simulated TrA activity on the SIJ.

MA

The present study investigated the impact of in vitro simulated force

of transversely oriented fibres of the TrA on flexibility of the SIJ during

asymmetric loading of the pelvis. Flexion (linked to SIJ counternutation) and

TE

extension (linked to SIJ nutation) rotations in the sagittal plane of one

CE
P

innominate with respect to the other were used to mimic the natural
behaviour of the pelvis during e.g. locomotion. Our secondary objective was
to compare SIJ flexibility for flexion and extension rotations of the

AC

innominate. Such objective was formulated since previous studies using

similar methodology (e.g. Jacob and Kissling 1995, Smidt et al. 1995,
Sturesson et al. 2000 ab, Agarwal et al. 2014) did not address this issue.
Moreover, for quite a long time extension of the innominate (SIJ nutation)
has been regarded as the position of stability (or close-packed position)
protected by larger ligamentous guard, as opposed to flexion of the
innominate (SIJ counternutation) limited only by the thin anterior
ligamentous complex and long dorsal sacroiliac ligament (Vleeming et al.
1989ab, 1996).

ACCEPTED MANUSCRIPT
Findings of this study can be of importance in the debate whether
local muscles indeed have a stabilizing function in contrast to other muscles

of the trunk. Outcome of this debate can interfere with current treatment

IP

modalities employed in e.g. management of patients with postpartum pelvic

SC
R

girdle pain (Stuge et al. 2004) or demonstrating difficulties in transferring

load across the pelvic ring (Mens et al. 2001, Beales et al. 2010). Analysis of

NU

SIJ flexibility for flexion and extension of the innominate relative to the

MA

sacrum may also add to our understanding of SIJ function in human gait.

2. METHODS

2.1. Material

TE

Eight embalmed specimens (6 females, mean age at time of death

CE
P

70.3 (5) years, embalming time 3-6 months) consisting of the pelvis and L5
vertebrae with all ligaments and capsules intact were available.

AC

2.2. Experimental set-up

In our specimens, an incremental torque was applied to one

innominate in the sagittal plane while the other innominate was fixed to a
custom-made frame (Fig. 1). To enable both torque application to one
innominate and fixation of the opposite innominate, a rigid metal plate was
screwed to each innominate through the compact bone at the height of the
iliac crest, above the acetabulum and through the ischial tuberosity. One
plate was attached to the frame to fix the bone, while the second was
connected to a steel bar and axle to allow application of an incremental force

ACCEPTED MANUSCRIPT
resulting in a moment in the sagittal plane. To allow three-dimensional
movements of the innominate on the non-fixed side and the sacrum, the axle

was equipped with two universal joints and one prismatic sliding joint (Fig. 1).

IP

Torque was exerted on the axle by a custom-made pneumatic traction

SC
R

system installed on the frame. The force was transmitted by non-elastic cord
to a steel disk (diameter 300 mm) and then to the axle and innominate. To

NU

calculate the real moment applied to the bone a torque transducer was
placed between the steel disk and the bar with two universal joints (Fig.1).

MA

Throughout the whole measurement its signals were registered with a

sampling frequency of 10 Hz.

Prior to flexibility analysis, it was necessary to record ranges of

TE

motion of the SIJ in the sagittal plane during the specimen loading. To

CE
P

achieve this the Vicon MX Motion Capture System (Vicon Motion Systems,
Oxford, UK) was used. A total of 11 reflective markers (diameter 9.5 mm)
were screwed to the specimen (4 per each innominate, 3 per sacrum). To

AC

minimize interference by bone deformations (especially in the lower part of

the innominate), in the current analysis only the 9 markers located close to
the SIJs (Fig.2) were used and the remaining two markers in the pubic
symphysis area (not shown in the Fig.2) were omitted. Markers were
illuminated by an infrared light source mounted on each of four video
cameras equipped with a 20 mm lens (Sigma, Tokyo, Japan). A sampling
frequency of 100 Hz was used which was fairly enough for the employed,
semi-static mode of load application. The precision of the measurement

ACCEPTED MANUSCRIPT
estimated on a mechanical model was equal to 0.1 deg for angular and 0.1
mm for linear measurements.

During each test an incremental torque was first applied to either the

IP

left or right innominate bone (randomly chosen) in a semi-static, step-wise

SC
R

manner. The opposite innominate was properly fixed and remained

immovable throughout the trial. During consecutive stages of the procedure

NU

(see below) the torque increased/decreased in load-steps ranging from 3 to

7 Nm applied with 20 s intervals. Starting from the unloaded state, we used

MA

larger load-steps (7 Nm) and approaching maximal torques they were

gradually becoming smaller (minimally 3 Nm). This prevented specimen

destruction when maximal loads were in use. The torque during each load-

TE

step increased/decreased at a rate of 2 Nm/s. After each load-step was

CE
P

completed, the position of the markers was captured in the Vicon system
with a 10 s latency.

Using this mode of torque application, after application of the maximal

AC

flexion torque in the sagittal plane (FT, anterior rotation about Y-axis of the
reference system (Fig.2)), the load was step-wise released until the moving
innominate returned to its neutral position. This was directly followed by a
step-wise increasing of extension torque (ET, posterior rotation about Y-axis
of the reference system). Subsequently, the ET load was released until the
neutral position. Finally, FT was step-wise applied again until its maximum.
After each step positions of the markers were captured. The number of loadsteps during application of FT and ET was not fixed. To measure the range
of motion during the procedure an electronic goniometer was mounted on

ACCEPTED MANUSCRIPT
the steel disk (Fig.1). The load increments were ceased when during two
consecutive, final load-steps the range of motion (ROM) recorded by the

goniometer increased by less than 0.1 deg.

IP

The same procedure was repeated with simulation of the force

SC
R

exerted by transversely oriented fibres of the TrA on the pelvis. To achieve

this, a rubber band was transversely screwed to the innominates through

NU

holes at the height of the anterior superior iliac spines (Fig. 2), pulling them
towards each other with a force of 120 5 N (Mens et al. 2006, Pel et al.

MA

2008).

The whole procedure was repeated with the torque applied to the

other innominate. Therefore, in total 4 series of measurement were

TE

performed: 1) torque applied to side 1 of the pelvis (L or R depending on

CE
P

randomization); 2) torque applied to side 1 with simulated TrA force; 3)

torque applied to side 2 of the pelvis; 4) torque applied to side 2 with

AC

simulated TrA force.

2.3. Data analysis

The first stage of each series of measurement for each specimen (i.e.
unloaded state to maximal FT) was considered a preconditioning stage and
this part of the data was ignored. In this way the tissues could adjust to the
load and artefacts associated with this process were avoided. Therefore,
only the following two trajectories of a single loading cycle were studied: 1)
t1: maximal FT unloaded state maximal ET; and 2) t2: maximal ET

ACCEPTED MANUSCRIPT
unloaded state maximal FT (Fig. 3 shows qualitatively an example of a
load deformation curve).

The parameters of the rotation axis of relative motion between bones

IP

were calculated from the positions of the sacral and innominate markers

SC
R

after each loading step (Spoor and Veldpaus 1980, Sderkvist and Wedin
1993).The sagittal plane SIJ ROM is represented by the rotations of the

NU

innominate in relation to the sacrum about the Y axis of the reference

system (Fig.2). These rotations are defined as follows: flexion is anterior

MA

rotation about the Y-axis; extension is posterior rotation about the Y-axis
The coordinate system was attached to the fixed innominate bone.

Subsequently, load-deformation curves were plotted (for the 4 series

TE

of the measurement and two directions of torque application: FT and ET) for

CE
P

trajectories t1 and t2. Slopes of the adjusted linear regression lines (ROM vs
torque) were calculated. As visible in Figure 3, two regression lines for each
direction of torque application (FT and ET) were drawn and two slopes were

AC

calculated (F1, F2 and E1, E2). The mean value of the two slopes for each
torque direction was considered a measure of SIJ flexibility. Defined in such
way, the slope is directly proportional to flexibility of the joint.

2.4. Statistical analysis

Since in small samples it is difficult to reliably test the data for normal
distribution, both parametric and non-parametric tests were applied. The
outcomes were, however, similar and therefore normal distribution of the
data was assumed. The results of the parametric tests were presented, i.e. a

10

ACCEPTED MANUSCRIPT
mixed model of Anova with one independent factor: torque direction (FT and
ET) as well as one repeated factor: series of measurement (side 1 of the

pelvis; side 1 with TrA; side 2 of the pelvis; side 2 with TrA). Significance

SC
R

IP

level was set at P<0.05.

3. RESULTS

NU

Maximal recorded FT and ET for all specimens were equal to 76.40

Nm and 72.53 Nm; the maximum values averaged over all specimens were

MA

51.75 (SD 11.14) Nm and 52.62 (SD 10.80) Nm, respectively. No significant
difference was found between them.

Mean values and standard deviations of the slopes recorded during

TE

application of the FT and ET are presented in the Table 1. Analysis of Anova

CE
P

results revealed significant main effect for the independent factor, i.e. torque
direction (F=4.826; P=0.037). The FT slope (marginal mean 0.022 deg/Nm)
was significantly bigger than ET slope (marginal mean 0.014 deg/Nm),

AC

which means that flexibility during application of FT was bigger (Fig.4A,

Tab.1). Main effect for the repeated factor (F=0.471; P=0.703) was not
significant (marginal means of the slopes for consecutive series of
measurement 0.017 deg/Nm, 0.019 deg/Nm, 0.017 deg/Nm, 0.018
deg/Nm)(Fig.4B). So was the interaction of the independent and repeated
factors (F=2.099; P=0.107)(Tab.1). This means that after application of
simulated TrA force flexibility of the SIJ did not change both during
application of FT and ET.

11

ACCEPTED MANUSCRIPT
4. DISCUSSION
In this study the SIJ flexibility was investigated in two experimental

conditions (without/with TrA force simulation). It is difficult to compare the

IP

recorded levels of flexibility with outcomes of other studies since only few

SC
R

investigated this same issue in a comparable way. Pool-Goudzwaard et al.

(2003) recorded slopes of regression lines (ROM vs torque) in the range of

NU

0.046-0.124 deg/Nm. In another study (Pool-Goudzwaard et al. 2004) slopes

of 0.06 deg/Nm in males and 0.13 deg/Nm in females were reported. In the

MA

current study lower levels of flexibility were registered (maximal mean slope
of 0.025 deg/Nm for FT with TrA force simulation in series 2, Tab.1). Such

observations may be explained by different specimen mounting or

TE

embalming time (the two studies mentioned above do not provide data on

CE
P

embalment duration).

The main objective of this study was to investigate whether in vitro

simulated action of the TrA exerts a stiffening effect on SIJ during

AC

asymmetric loading of the pelvis. Previous in vivo studies and biomechanical

models tended to support such an effect of TrA contraction (Snijders et al.
1995, Richardson et al. 2002, Pel et al. 2008). However, based on the
present data, this effect cannot be confirmed. No significant SIJ flexibility
decrease was recorded during simulation of the TrA force. This finding
should not be attributed to gradual tissue damage during 4 consecutive
series of measurement since flexibility shows systematic tendency to
increase after TrA force simulation (between series 1 and 2; 3 and 4) and to
decrease after removal of this force (between series 2 and 3)(Fig.4B). There

12

ACCEPTED MANUSCRIPT
are several possible explanations for such findings. First of all, exclusively
the transverse portion of the TrA fibres was mimicked in vitro. The rationale

for this was that only from these fibres a large compressive force on the

IP

pelvis could be expected (Snijders et al. 1995). In vivo, a more global

SC
R

activation of the TrA should be present, as well as activation of numerous

synergistic muscles of the local system, which were not included in the

NU

current setting. Such a simplified model was unable to reflect the whole
spectrum of physiologic behaviour of both the TrA and other local muscles. It

MA

is known that in vivo activation of the TrA takes place in synergy with pelvic
floor muscles (Sapsford et al. 2001ab) and the diaphragm (Allison et al.

1998, Hodges and Gandevia 2000). Simulation of these muscle forces might

TE

have more strongly influence the flexibility of the SIJ. Observations from an

CE
P

in vitro study by Pool-Goudzwaard et al. (2004) seem to support this notion.

Authors demonstrated a tendency to decrease SIJ flexibility only after
simulation of multiple forces produced by pelvic floor muscles. Perhaps a

AC

more complex specimen than used in the present study (pelvis, lumbar and
lower thoracic spine with ribs), and reproduction of multiple vectors
generated by the TrA and other deep muscles, would help to more
realistically verify the hypothesis concerning their role in decreasing flexibility
of the pelvic girdle. These observations (Richardson et al. 2002, PoolGoudzwaard et al. 2004, Pel et al. 2008) allow to speculate that muscle
synergisms and coactivation within the local system have greater impact on
the flexibility of the pelvic region than isolated activation of single muscles.
This issue has already been addressed by others (e.g. Cholewicki et al.

13

ACCEPTED MANUSCRIPT
1997, Gardner-Morse and Stokes 1998). It might also be clinically relevant.
Within many therapeutic approaches (both older and modern) exercises are

implemented aiming to enhance the patients ability to activate single local

IP

muscles (Pilates and Miller 1945, Richardson and Jull 1995, Hides et al.

SC
R

1996, OSullivan et al. 1997, Comerford and Mottram 2001, Stuge et al.
2004). To date, no data are available demonstrating that restoration of such

NU

an ability is equivalent to restoration of the desired synergisms. Probably, in

lower back motor control training and testing the ability to coactivate

MA

numerous local muscles might deserve more attention. Further studies to

explain these issues are strongly needed.

Differences between flexibility recorded during application of FT and

TE

ET is also worth attention and conforms to previous studies. Extension of the

CE
P

innominate is linked to nutation of the SIJ, which is long regarded as the

more stable position of the joint than counternutation (linked to innominate
flexion)(Vleeming et al. 1989ab, 1996). It seems that flexibility of the SIJ

AC

might be subjected to modulation by different flexion/extension angles of the

innominate bone. This information may be of importance for therapists
dealing with pelvic dysfunctions and trying to find associations between
them and features of individual gait. For example, one may speculate that
pelvic pain occurring at heel strike is linked to excessive flexibility of the SIJ.
In the field of manual therapy this observation suggests that clinicians may
gain different manual sensations when assessing SIJ using movements of
anterior and posterior rotations of the innominate. Anterior rotation may
produce softer and less stiff feel as compared to stiffer posterior rotation. In

14

ACCEPTED MANUSCRIPT
therapy, when applying manual thrust techniques or mobilizations
incorporating anterior innominate rotation movement there may be a need to

IP

techniques incorporating posterior rotation.

cover larger ROM before reaching the motion barrier in comparison to

SC
R

Apart from the fact that only the transverse portion of the TrA fibres
was mimicked, other limitations of the present study need to be addressed.

NU

First, the limitations linked to in vitro studies are known and inherent.
Conclusions derived from cadaveric data cannot be directly applied to living

MA

organisms. Long-term embalming might decrease elasticity of the tissues,

and flexibility of the SIJ might differ from in vivo conditions. It is also possible

that loads applied in this study exceeded those associated with normal daily

TE

life activity. Moreover, the SIJ loads in vivo depend on muscle forces and

CE
P

are therefore different from the passive loads in cadaveric experiments, with
consequences for relative bone positions. The history of our specimens
remained unknown as well. It was only possible to exclude specimens with

AC

evident bony anomalies. The generalizability of our conclusions is therefore

limited. Additional studies on fresh cadaveric material, and also in vivo, are
needed to verify the results of this experiment.

5. CONCLUSIONS
1. A lack of a stiffening effect of simulated TrA muscle force on SIJ was
demonstrated. Earlier hypotheses suggesting a stiffening influence of this
muscle on the pelvis cannot be confirmed.

15

ACCEPTED MANUSCRIPT
2. Flexibility of the SIJ was smaller for extension of the innominate relative to

AC

CE
P

TE

MA

NU

SC
R

IP

the sacrum (SIJ nutation) than for flexion (SIJ counternutation).

16

ACCEPTED MANUSCRIPT
CONFLICT OF INTEREST
We inform that no sponsors were involved in the study. All expenses were covered by

former Department of Biomedical Physics and Technology, Erasmus University,

SC
R

IP

Rotterdam. Therefore, no conflict of interest is associated with the presented study.

ACKNOWLEDGEMENTS

NU

We would like to acknowledge the kind assistance of very talented Mrs. Karine

MA

Bollerot, PhD, who prepared for us figures 1 and 2.

REFERENCES

1. Agarwal Y, Doebele S, Windolf M, Shiozawa T, Gueorguiev B, Stuby FM.

TE

Two-leg alternate loading model a different approach to biomechanical

investigations of fixation methods of the injured pelvic ring with focus on the

CE
P

pubic symphysis. J Biomech, 2014, 47(2):380-6.

2. Allison G, Kusuhara N, Yoshimura N, Tomita T, Easton PA. The role of the

AC

diaphragm during abdominal hollowing exercises. Aust J Physiother, 1998,

44:95-104.

3. Allison GT, Morris SL. Transversus abdominis and core stability has the
pendulum swung? Br J Sports Med, 2008, 42:930-1.
4. Askar OM. Surgical anatomy of the aponeurotic expansions of the anterior
abdominal wall. Annals of the Royal College of Surgeons of England, 1977,
59:313-21.
5. Beales DJ, OSullivan PB, Briffa NK. The effects of manual pelvic
compression on trunk motor control during an active straight leg raise in
chronic pelvic girdle pain subjects. Man Ther, 2010, 15(2):190-9.
17

ACCEPTED MANUSCRIPT
6. Bergmark A. Stability of the lumbar spine. A study in mechanical
engineering. Acta Orthop Scand Suppl, 1989, 230:1-54.

7. Burns SA, Foresman E, Kraycsir SJ, Egan W, Glynn P, Mintken PE,

IP

Cleland JA. A treatment-based classification approach to examination and

SC
R

intervention of lumbar disorders. Sports Health, 2011, 3(4):362-72.

8. Cholewicki J, Panjabi M, Khachatryan A. Stabilizing function of trunk flexor-

NU

extensor muscles around a neutral spine posture. Spine, 1997, 22: 2207-12.
9. Cleland J, Schulte C, Durall C. The role of therapeutic exercise in treating

MA

instability-related lumbar spine pain: a systematic review. J Back

Musculoskelet Rehabil, 2002, 16(2-3):105-15.

10. Comerford MJ, Mottram SL. Functional stability retraining: Principles and

TE

strategies for managing mechanical dysfunction. Man Ther, 2001, 6(1):3-14.

CE
P

11. Ferreira ML, Ferreira PH, Latimer J, Herbert RD, Hodges PW, Jennings MD,
Maher CG, Refshauge KM. Comparison of general exercise, motor control
exercise and spinal manipulative therapy for chronic low back pain: a

AC

randomized trial. Pain, 2007, 131:31-7.

12. Ferreira PH, Ferreira ML, Maher CG, Herbert RD, Refshauge KM. Specific
stabilization exercise for spinal and pelvic pain: a systematic review. Aust J
Physiother, 2006, 52:79-88.
13. Gardner-Morse MG, Stokes AF. The effects of abdominal muscle
coactivation on lumbar spine stability. Spine, 1998, 23:86-92.
14. Gnat R, Spoor K, Pool-Goudzwaard A. Simulated transversus abdominis
muscle force does not increase stiffness of the pubic symphysis and
innominate bone: an in vitro study. Clin Biomech, 2013, 28:262-7.

18

ACCEPTED MANUSCRIPT
15. Hides JA, Richardson CA, Jull GA. Multifidus muscle recovery is not
automatic after resolution of acute, first-episode low back pain. Spine, 1996,

21:2763-9.

IP

16. Hodges PW, Coppieters MW, MacDonald D, Cholewicki J. New insight into

SC
R

motor adaptation to pain revealed by a combination of modelling and

empirical approaches. Eur J Pain, 2013, 17(8):1138-46.

NU

17. Hodges PW, Gandevia SC. Changes in intra-abdominal pressure during

postural and respiratory activation of the human diaphragm. J Appl Physiol,

MA

2000, 3:967-76.

18. Hodges PW, Richardson CA. Contraction of the abdominal muscles

associated with movement of the lower limb. PhysTher, 1997a, 77:132-44.

TE

19. Hodges PW, Richardson CA. Delayed postural contraction of transversus

CE
P

abdominis associated with lower back pain. J Spinal Disord, 1998, 11:46-56.
20. Hodges PW, Richardson CA. Feedforward contraction of transversus
abdominis is not influenced by the direction of arm movement. Exp Brain

AC

Res, 1997b, 114:362-70.

21. Jacob HAC, Kissling RO. The mobility of the sacroiliac joints in healthy
volunteers between 20 and 50 years of age. Clin Biomech, 1995, 10(7):35261.
22. Koumantakis GA, Watson PJ, Oldham JA. Trunk muscle stabilization
training plus general exercise versus general exercise only: randomized
controlled trial of patients with recurrent low back pain. Phys Ther, 2005,
85(3):209-25.

19

ACCEPTED MANUSCRIPT
23. Lederman E. The myth of core stability. J Bodyw Mov Ther, 2010, 14(1):8498.

24. Mannion AF, Pulkovski N, Toma V, Sprott H. Abdominal muscle size and

SC
R

control subjects. J Anat, 2008, 213:173-82.

IP

symmetry at rest and during abdominal hollowing exercises in healthy

25. Mens JMA, Vleeming A, Snijders CJ, Koes BW, Stam HJ. Reliability and

NU

validity of the Active Straight Leg Raise Test in posterior pelvic pain since
pregnancy. Spine, 2001, 26:1167-71.

MA

26. Mills JD, Taunton JE, Mills WA. The effect of a 10-week training regimen on
lumbo-pelvic stability and athletic performance in female athletes: a

randomized-controlled trial. Phys Ther Sport, 2005, 6(2):60-6.

TE

27. OSullivan PB, Twomey LT, Allison GT. Evaluation of specific stabilizing

CE
P

exercise in the treatment of chronic low back pain with radiologic diagnosis
of spondylolysis or spondylolisthesis. Spine, 1997, 15(24):2959-67.
28. Pel JJ, Spoor CW, Pool-Goudzwaard AL, Hoek van Dijke GA, Snijders CJ.

AC

Biomechanical analysis of reducing sacroiliac joint shear load by

optimization of pelvic muscle and ligament forces. Ann Biomed Eng, 2008,
36(3):415-24.
29. Pilates J, Miller WJ. Return to life through Contrology. Christopher
Publishing House, Boston, 1945.
30. Pool-Goudzwaard A, Hoek van Dijke G, Mulder P, Spoor C, Snijders C,
Stoeckart R. The iliolumbar ligament: its influence on stability of the
sacroiliac joint. Clin Biomech, 2003, 18:99-105.

20

ACCEPTED MANUSCRIPT
31. Pool-Goudzwaard A, Hoek van Dijke G, van Gurp M, Mulder P, Snijders C,
Stoeckart R. Contribution of pelvic floor muscles to stiffness of the pelvic

ring. Clin Biomech, 2004, 19:564-71.

IP

32. Reeves NP, NarendraKS, Cholewicki J. Spine stability: lessons from

SC
R

balancing a stick. Clin Biomech, 2011, 26(4):325-30.

33. Richardson CA, Jull GA. Muscle control-pain control. What exercises would

NU

you prescribe? Man Ther, 1995, 1:2-10.

34. Richardson CA, Snijders CJ, Hides JA, Damen L, Pas MS, Storm J. The

MA

relation between the transversus abdominis muscles, sacroiliac joint

mechanics, and low back pain. Spine, 2002, 27(4):399-405.

35. Rizk NN. A new description of the anterior abdominal wall in man and

TE

mammals. J Anat, 1980, 131:373-85.

CE
P

36. Sapsford RR, Hodges PW, Richardson CA, Cooper DH, Markwell SJ, Jull
GA. 2001a. Co-activation of the abdominal and pelvic floor muscles during
voluntary exercises.Neurourol Urodyn, 2001a, 1:31-42.

AC

37. Sapsford RR, Hodges PW. 2001b. Contraction of the pelvic floor muscles
during abdominal maneuvers. Arch Phys Med Rehabil, 2001b, 8:1081-8.
38. Smidt GL, McQuade K, Wei S-H, Barakatt E. Sacroiliac kinematics for
reciprocal straddle positions. Spine, 1995, 20:1047-54.
39. Snijders CJ, Vleeming A, Stoeckart R, Mens JMA, Kleinrensink GJ.
Biomechanical modelling of sacroiliac joint stability in different postures.
Spine: State Art Rev, 1995, 9:419-32.
40. Sderkvist I, Wedin PA. Determining the movements of the skeleton using
well-configured markers. J Biomech,1993, 26(12):1473-7.

21

ACCEPTED MANUSCRIPT
41. Spoor CW, Veldpaus FE. Rigid body motion calculated from spatial coordinates of markers. J Biomech, 1980, 13(4):391-3.

42. Stuge B, Lrum E, Kirkesola G, Vllestad N. The efficacy of a treatment

IP

program focusing on specific stabilizing exercises for pelvic girdle pain after

SC
R

pregnancy. A randomized controlled trial. Spine, 2004, 29(10):351-9.

43. Sturesson B, Uden A, Vleeming A. A radiostereometric analysis of

NU

movements of the sacroiliac joint during standing hip flexion test. Spine,
2000a, 25:364-8.

MA

44. Sturesson B, Uden A, Vleeming A. A radiostereometric analysis of the

movements of the sacroiliac joints in the reciprocal straddle position. Spine,

2000b, 25(2):214-17.

TE

45. Tsao H, Hodges PW. Immediate changes in feedforward postural

CE
P

adjustments following voluntary motor training. Exp Brain Res, 2007,

181:537-46.

46. Vleeming A, Pool-Goudzwaard AL, Hammudoghlu D, Stoeckart R, Snijders

AC

CJ, Mens JMA. The function of the long dorsal sacroiliac ligament. Its
implication for understanding low back pain. Spine, 1996, 21(5):556-62.
47. Vleeming A, Stoeckart R, Snijders CJ. The sacrotuberous ligament: a
conceptual approach to its dynamic role in stabilizing the sacroiliac joint.
Clin Biomech, 1989a, 4:201-3.
48. Vleeming A, van Wingerden JP, Snijders CJ, Stoeckart R, Stijnen T. Load
application to the sacrotuberous ligament: Influences on sacro-iliac joint
mechanics. Clin Biomech, 1989b, 4:204-9.

22

ACCEPTED MANUSCRIPT
TABLES
Table 1. Mean values (standard deviations)(deg/Nm) of the slopes recorded for sagittal plane

rotations of the innominate in relation to the sacrum (about Y axis of the reference system)

IP

during application of flexion (FT) and extension (ET) torques and for four series of measurement:

SC
R

1) torque applied to side 1 of the pelvis; 2) torque applied to side 1 of the pelvis with simulated
transversus abdominis (TrA) force; 3) torque applied to side 2 of the pelvis; 4) torque applied to

series 1

series 2

series 3

series 2

0.022 (0.012)

0.025 (0.017)

0.020 (0.012)

0.020 (0.013)

0.13 (0.007)

0.12 (0.008)

0.14 (0.007)

0.15 (0.010)

AC

CE
P

TE

MA

Torque
FT
ET

NU

side 2 with TrA force.

23

ACCEPTED MANUSCRIPT
FIGURE CAPTIONS

Figure 1. Experimental set-up with metal plates (A) screwed to both

IP

innominates, one fixed to the frame, the other connected to a steel bar with

SC
R

two universal joints (B), a torque transducer (C), a pulley (D) and a prismatic
sliding joint (E). Electronic goniometer (F) was used to control range of

NU

motion during the procedure. To maintain clarity, pneumatic cylinders

MA

generating the necessary forces are not shown.

Figure 2. The coordinate system and position of the sacral (S 1,2,3) and

innominate (I 1(1), 2(2), 3(3)) markers. The spring represents an elastic

TE

rubber band imitating the force exerted by the transversus abdominis muscle

CE
P

on the two innominates.

Figure 3. A representative schematic load-deformation curve for the rotations of

AC

the innominate in relation to the sacrum about the Y axis of the reference
system (flexion(positive)/extension(negative)). Data from the first stage (i.e.
unloaded state maximal flexion torque (FT)) were ignored (not shown).
Only the two trajectories were used for the analysis: t1) maximal FT
unloaded state maximal extension torque (ET), and t2) maximal ET
unloaded state maximal FT. Adjusted linear regression lines for t1 and t2
are presented by dotted lines. Two regression lines were drawn for ET (E1
and E2) and two other for FT (F1 and F2). Slopes of these regression lines
were regarded as measures of flexibility. The mean value of slopes E1 and

24

ACCEPTED MANUSCRIPT
E2 and the mean value of slopes F1 and F2 were calculated and subjected
to further analysis. Defined this way, the slope is directly proportional to joint

IP

flexibility.

SC
R

Figure 4. Marginal mean values of the slopes of regression lines recorded for
the two factors of Anova design (main effects): independent (A) torque

NU

direction (flexion (FT) vs extension (ET)); and repeated (B) series of

measurement (series 1-4: 1) torque applied to side 1 of the pelvis; 2) torque

MA

applied to side 1 of the pelvis with simulated transversus abdominis (TrA)

force; 3) torque applied to side 2 of the pelvis; 4) torque applied to side 2

AC

CE
P

TE

with TrA force). Whiskers indicate 95% confidence intervals.

25

AC

CE
P

TE

MA

NU

SC
R

IP

ACCEPTED MANUSCRIPT

Figure 1

26

MA

NU

SC
R

IP

ACCEPTED MANUSCRIPT

AC

CE
P

TE

Figure 2

27

AC

Figure 3

CE
P

TE

MA

NU

SC
R

IP

ACCEPTED MANUSCRIPT

28

SC
R

IP

ACCEPTED MANUSCRIPT

AC

CE
P

TE

MA

NU

Figure 4

29

ACCEPTED MANUSCRIPT
HIGHLIGHTS
Flexibility of the sacroiliac joint was investigated in vitro.

The force of transversus abdominis muscle was simulated during the

IP

procedure.

Simulation o the muscle force did not decrease flexibility of the joint.

This effect ocurred both during anterior and posterior rotations of the

SC
R

Overall flexibility of the joint was smaller during innominate posterior

CE
P

TE

MA

rotation.

AC

NU

innominate.

30