Manual Therapy 16 (2011) 573e577

Contents lists available at ScienceDirect

Manual Therapy
journal homepage: www.elsevier.com/math

Original article

The relationship of transversus abdominis and lumbar multifidus clinical muscle

tests in patients with chronic low back pain
Julie Hides a, b, *, Warren Stanton a, M. Dilani Mendis a, Margot Sexton a
a
School of Physiotherapy, Australian Catholic University, McAuley at Banyo, Queensland 4014, Australia
b
Mater Back Stability Clinic, Mater Health Services, South Brisbane, Queensland 4101, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Introduction: Previous research of transversus abdominis (TrA) and multifidus muscle function in the
Received 2 June 2010 presence of chronic low back pain (LBP) has investigated these muscles in isolation. In clinical practice, it
Received in revised form is assumed that a relationship exists between these muscles and so they are often assessed and reha-
26 April 2011
bilitated together. However, no studies have tested or documented this association. This study aimed to
Accepted 9 May 2011
examine the relationships between clinical muscle testing and other measures taken in the course of
a clinical assessment at a back clinic.
Keywords:
Methods: This retrospective chart audit examined the files of 82 patients (40 Males, 42 Females) for
Low back pain
Transversus abdominis muscle
results of clinical tests of TrA and multifidus muscle contraction, multifidus muscle size measurements
Multifidus muscle and other clinical measures such as distribution of pain and pain on manual examination.
Clinical muscle tests Results: The ability to contract multifidus was related to the ability to contract TrA with the odds of
a good contraction of multifidus being 4.5 times higher for patients who had a good contraction of TrA. A
poor ability to contract multifidus was related to poor TrA contraction. Patients with unilateral LBP had
more multifidus muscle asymmetry (11.6%) than those with bilateral/central pain (0.01%) and had a poor
multifidus contraction on the affected side (p < 0.01). No other significant relationships were found.
Discussion & conclusion: Current clinical practice of assessment and rehabilitation of both TrA and
multifidus muscles in patients with chronic LBP is supported by the findings of this study. Future studies
may investigate if a neurophysiological relationship exists between these muscles.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Changes in motor control of abdominal muscles have been

reported in subjects with LBP. There is evidence of delayed activa-
There has been considerable debate about the role of trunk tion of the TrA muscle in clinical and experimental studies of LBP
muscles such as the transversus abdominis (TrA) and lumbar (Hodges and Richardson, 1996, 1998; Hodges et al., 2003b). Imaging
multifidus muscles in lumbo-pelvic stability and low back pain studies using magnetic resonance imaging (MRI) and ultrasound
(LBP). There is considerable evidence that these muscles provide an imaging have documented alterations in motor control of the
important contribution. For example, there is evidence that the TrA abdominal muscles (Hides et al., 2008b, 2010). It has been proposed
muscle is controlled independently of the other abdominal muscles that the documented motor control changes, such as dysfunction of
in a range of tasks (Hodges and Richardson, 1996, 1998). It has been the TrA muscle, are associated with higher long-term incidence of
proposed that the TrA muscle may contribute to stability of the LBP (Moseley, 2004).
lumbo-pelvic region via its effects on intra-abdominal pressure and There is also considerable evidence for the role of the lumbar
by affecting fascial tension (Hodges et al., 2003a, 2005; Barker et al., multifidus muscle in stabilization of the lumbar spine. Biome-
2006). In addition, biomechanical models have proposed that the chanical studies have highlighted the role of the multifidus muscle
TrA muscle may play a role in support of the lumbo-pelvic region in provision of segmental stiffness (Panjabi, 1992a,b; Wilke et al.,
for weight-bearing (Snijders et al., 1995). 1995), control of the spinal segment’s neutral zone (Panjabi et al.,
1989; Panjabi, 1992b), and its capacity to stabilize the spine when
spinal stability is challenged (Keifer et al., 1997, 1998; Moseley,
* Corresponding author. School of Physiotherapy, Australian Catholic University,
McAuley at Banyo, Queensland 4014, Australia. Tel.: þ61 7 36237530; fax: þ61 7
2004).
36237650. Among subjects with LBP, impairments of the multifidus
E-mail address: [email protected] (J. Hides). muscle have been documented using imaging techniques. There is

1356-689X/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.math.2011.05.007
574 J. Hides et al. / Manual Therapy 16 (2011) 573e577

evidence that the cross-sectional area (CSA) of the multifidus is 2. Methods

selectively decreased compared with other lumbo-pelvic muscles
in patients with chronic LBP (Danneels et al., 2000). Multifidus 2.1. Subjects
muscle atrophy has been successfully quantified using MRI and
Computerized Tomography (CT) scanning in terms of both Data for this study was based on chart audits from the files of 82
decreased muscle CSAs (Barker et al., 2004) and presence of patients (42 females, 40 males) presenting to a hospital based back
alterations in muscle consistency (due to fatty deposits or fibrous/ pain clinic between 1998 and 2005 for assessment and manage-
connective tissue infiltration) and atrophy (Kader et al., 2000). ment of chronic LBP. The mean age of the patients was 43.21  13.09
Researchers have used real-time ultrasound imaging to demon- years. Patients who attended the clinic during these years and were
strate segmental decrease in the CSA of the multifidus, ipsilateral included in this study had a history of LBP in excess of three
to painful symptoms, in patients with acute unilateral LBP (Hides months, with LBP defined as pain localized between the T12
et al., 1994, 1996). A similar localized (rather than generalized) vertebral level and the gluteal fold. Individual cases were excluded
pattern of muscle atrophy of the multifidus muscle has been from review if they met any of the following criteria: previous
demonstrated in subjects with chronic LBP with unilateral pain lumbar surgery, pregnancy, systemic disease plus any reflex and or
presentations (Hides et al., 2008a; Wallwork et al., 2009). This motor signs of nerve root or cauda equina compression. Written
study also provided evidence of a corresponding reduced ability to informed consent was obtained from patients during their initial
voluntarily contract the atrophied muscle (Wallwork et al., 2009). visit to the clinic and the project met ethical requirements of the
In the clinical setting, possible methods of testing TrA muscle hospital.
function involve palpation of the abdominal wall (Hides et al.,
2000) and the use of a pressure cuff (pressure biofeedback unit
Chattanooga, USA) placed under the abdomen with the patient in 2.2. Procedure
a prone lying position (Hides et al., 2004). This test represents an
inner range concentric contraction of the TrA muscle to lift the Chart audits were conducted on the archived files of LBP
abdominal contents and wall and thereby decrease the pressure in patients from two physiotherapists at the clinic, who were the
the pressure biofeedback unit. The multifidus muscle can be most experienced in muscle measurement by ultrasound and for
assessed by the palpation of muscle bulk and by the quality of whom consistency has been verified by inter-rater and intra-
voluntary contraction at each lumbar vertebral level (Hides et al., rater reliability (Wallwork et al., 2007). The researcher con-
2004). Real-time ultrasound imaging is another method that is ducting measurements of the muscle system was blinded to the
used in physiotherapy clinical practice both for assessment of TrA results of patient questionnaires, which were administered,
and multifidus muscle function and size as well as for retraining collected and scored by a research assistant. De-identified audit
purposes (Bunce et al., 2004; Teyhen et al., 2005; Koppenhaver information from the patients’ initial clinical muscle testing and
et al., 2009; Wallwork et al., 2009; Hides et al., 2010). assessment was then entered onto a spreadsheet, including the
It would seem from previous research that changes in the TrA measures of multifidus muscle contraction, CSA of the multifidus
and multifidus muscles are common, and may even possibly muscle, TrA muscle contraction, and other clinical measures
represent a marker of chronic LBP. However, studies of impairments described below.
of the TrA and multifidus muscles have been performed in isolation.
While it may be assumed that these impairments are related, no 2.2.1. Multifidus muscle testing
studies have tested and documented this association. In clinical The ability of the patient to perform an isometric contraction of
practice, these muscles are often rehabilitated together (Hides the multifidus muscle at the L5 vertebral level was assessed by
et al., 2004), based on the assumption that a relationship exists manual palpation (Hides et al., 2004), and documented as ‘unable’,
between these muscles. This is an important point as there is ‘poor’ or ‘good’ for the nominated symptomatic or affected side
evidence to support the efficacy of this approach. Macedo et al. (19.5%, 59.8%, 20.7% of the sample respectively) and least painful or
(2009) examined motor control training for persistent LBP unaffected side (19.5%, 52.4%, 28.0% respectively). Multifidus
(subacute, chronic and recurrent). Of the 14 randomized controlled muscle cross-sectional area was also measured at the L5 vertebral
trials included, the results of 7 trials showed that motor control level using real-time ultrasound imaging as described by Hides
exercise (alone or as supplement to another intervention), was et al. (2008a).
better than a minimal intervention in reducing pain at short-term,
intermediate and long-term follow-up, and in reducing disability at 2.2.2. Transversus abdominis muscle testing
long-term follow-up. In addition to the lack of current information The ability to contract the TrA muscle was assessed using the
of the relationship between the two muscles, it is unknown if there clinical muscle test (prone test with the pressure biofeedback unit
is a relationship between the results of muscle tests of the TrA and or PBU) (Hides et al., 2004). The results of three aspects of the
multifidus muscles and other clinical measures. The relationship clinical test were recorded in the patient files (either an increase or
between the distribution of reported symptoms, findings of manual decrease in mmHg measured on the PBU, presence or absence of
joint examination and clinical assessment of TrA and multifidus spinal movement and presence or absence of bulging of the
muscle function has not been established in a clinical population abdominal wall). This information was used to code the patients
with chronic LBP. into four groups; ‘poor with spinal movement’ (33.3%), ‘poor with
Therefore, this study was designed to investigate the relation- increased PBU pressure’ (42.3%), ‘fair with decreased PBU pressure
ships between clinical muscle testing and other measures taken in by 0e2 mmHg’ (16.7%) or ‘good with decreased PBU pressure by
the course of a clinical assessment at a back clinic. The aims of the 3e6 mmHg’ (7.7%). Subsequently, due to insufficient subject
study were to test for concordance between (i) clinical assessments numbers for analysis of four groups, the TrA muscle contraction
of TrA muscle contraction and multifidus muscle contraction, and groups ‘poor with spinal movement’ and ‘poor with increased PBU
(ii) multifidus muscle contraction and multifidus muscle size and pressure’ were combined to represent ‘poor’ contraction (75.6%),
asymmetry, and (iii) the association of these measures and other and the groups ‘fair with decreased PBU pressure by 0e2 mmHg’
clinical outcome measures such as pain on manual examination and ‘good with decreased PBU pressure by 3e6 mmHg’ were
and pain distribution. combined as ‘good’ contraction (24.4%).
 J. Hides et al. / Manual Therapy 16 (2011) 573e577 575

2.2.3. Clinical measures & self-reported factors TrA muscle (coded as ‘poor’ or ‘good’), pain on manual examination
The manual examination of each vertebral level involved central (coded as ‘pain’ or ‘no pain’), and pain distribution (coded as
lumbar postero-anterior intervertebral movements (PAIVMs) as ‘unilateral’ or ‘bilateral/central’). Due to the large number of inde-
described by Maitland (1986). Symptomatic levels of the lumbar pendent variables in this analysis, type of pain and gender were not
spine were determined based on pain provocation elicited upon included as they did not contribute to the model. The variables of
manual examination. Patients also reported on the following age, VAS, RMQ, HAQ and symptom duration were entered as
factors on the day of assessment: general level of pain based on the covariates in the analysis.
visual analogue scale (VAS; rated 0e10) (Huskisson, 1983), level of
disability based on the Roland Morris Disability Questionnaire 3. Results
(RMQ; rated 0e24) (Roland and Morris, 1983), amount of general
physical activity based on the Habitual Activity Questionnaire The mean (SD) of the VAS pain scores was 4.41  2.47, with
(HAQ; rated 0e10) (Baecke et al., 1982), and symptom duration (in individual pain ratings ranging between zero and maximum (10/
months). 10) at the time of examination. HAQ scores ranged between 3.75
Other data entered and subsequently used in analyses included: and 10.35 (out of a possible 15) with a mean score of 7.22  1.45.
age, gender (51.2% female), pain on manual examination (58.5% Baseline RMQ disability scores ranged between 0 and 21 (out of
yes), distribution of painful symptoms based on completion of a possible 24) with a mean score of 8.61  5.81. The mean duration
a body chart, and type of pain based on symptom duration and of symptoms was 51.21  91.72 months, with a range of 3
variability. Distribution of pain was coded as ‘bilateral/central’ monthse40 years (480 months), reflecting the prolonged chro-
(61%) or ‘unilateral’ (39%) in accord with the procedure of Hides nicity of the LBP population presenting to the clinic.
et al. (2008a). The type of pain was coded as either ‘chronic
recurring’ (30.9%) defined as recurring episodes of LBP with reso- 3.1. Contraction of the multifidus muscle
lution between episodes spanning 3 months or more; ‘continual’
(38.2%) defined as no resolution of pain for more than 3 months; or Results of the analysis of LBP patients (shown in Table 1) indi-
‘continual low/recurrent/exacerbated’ (30.9%) defined as low level cated that the ability to contract the multifidus muscle was related
of pain persisting with periods of exacerbated pain. to the ability to contract the TrA muscle. The odds of a good
contraction of the multifidus muscle were 4.45 times higher for
2.3. Statistical analysis patients who had good contraction of the TrA muscle compared to
those who had a poor ability. Conversely, a poor ability to contract
The Statistical Package for Social Sciences (SPSS) was used for multifidus was related to poor TrA contraction. Notably, none of the
data analysis. Data from subjects reporting unilateral LBP was re- other clinical measures of pain, disability or risk factors were
categorized as ‘affected’ or ‘unaffected’ based on the reported side related to multifidus contraction (p > 0.05) and there were no
of symptoms. The affected side for those with bilateral or central gender differences (p > 0.05).
pain was taken as the smallest side. This procedure provided a more
conservative test of the difference in asymmetry due to pain 3.2. Multifidus muscle size and asymmetry
distribution, as any significant effect reported for the unilateral pain
group would be greater than the absolute amount of asymmetry Results of the analysis for the CSA of the multifidus at the L5
among the bilateral group. vertebral level showed significant muscle asymmetry (smaller
Patients who were documented as unable to activate the mul- affected side; F ¼ 11.12, p ¼ 0.002), which interacted with pain
tifidus muscle (n ¼ 20) were excluded, leaving 62 cases for analysis. distribution (unilateral or bilateral/central) (F ¼ 9.92, p ¼ 0.003).
This group who were unable to contract the multifidus at all were The means presented in Table 2 show that patients with unilateral
excluded as it was possible they simply were unable to follow the LBP had more muscle asymmetry (11.6%) than the bilateral/central
instruction to contract the multifidus muscle, rather than having pain group (0.01%). In addition, there was a significant interaction
muscle dysfunction. This procedure also ensured that multifidus
muscle contraction was coded the same way as TrA muscle
contraction (poor and good), and was a linear uni-dimensional Table 1
measure as required for regression analysis. Logistic regression results of variables associated with ability to contract the mul-
tifidus muscle.

2.3.1. Contraction of the multifidus muscle Variablesa Chi-square Odds 95% Confidence
Logistic regression analysis was used to investigate factors ratio interval

associated with contraction of the multifidus muscle. Ability to Age (older) 0.01 0.99 (0.94, 1.06)
Pain VAS (higher) 0.19 0.91 (0.60, 1.38)
contract the multifidus muscle (measured by manual palpation and
HAQ (higher) 0.01 1.01 (0.59, 1.71)
coded ‘unable’, ‘poor’ or ‘good’) was used as the dependent variable. RMQ (higher) 0.37 0.95 (0.80, 1.12)
The independent variables were entered in the following blocks; (a) Symptom duration (longer) 0.98 1.01 (0.99, 1.01)
age, VAS, RMQ, HAQ and symptom duration, (b) contraction of the TrA muscle 4.45* 2.59 (1.07, 6.29)
TrA muscle (coded as ‘poor’ or ‘good’), pain on manual examination contraction (poor, good)
Pain on manual 0.12 0.78 (0.18, 3.30)
(coded as ‘pain’ or ‘no pain’), pain distribution (coded as ‘unilateral’ examination (no, yes)
or ‘bilateral/central’), type of pain (coded as ‘chronic recurring’, Pain distribution 1.68 0.58 (0.26, 1.32)
‘continual’, or ‘continual low/recurrent/exacerbated’) and gender. (bilateral/central, unilateral)
Pain type
(continual low/recurrent
2.3.2. Multifidus muscle size and asymmetry
/exacerbated)
For this repeated measures analysis of covariance, the cross- (chronic recurring) 0.26 1.58 (0.27, 9.14)
sectional area (size) of the multifidus muscle was used as the (continual) 0.07 0.79 (0.14, 4.65)
dependent variable. The repeated measures factor in the analysis Gender (female, male) 0.33 0.62 (0.12, 3.24)
was muscle asymmetry (coded as affected or unaffected side). The *p < 0.05.
a
between subjects factors in the analysis were: contraction of the For each variable the odds ratio refers to the category in italics.
576 J. Hides et al. / Manual Therapy 16 (2011) 573e577

Table 2 same vertebral level, and this result was not present at other
Factors associated with multifidus muscle cross-sectional area at the L5 vertebral vertebral levels. However, in the study by Wallwork et al. (2009),
level.
the results for the multifidus contraction were averaged across
Variable Pain distributiona sides. The results of the current study can go one step further by
Bilateral/central Unilateral showing there was a significant interaction between pain distri-
Vertebral side bution (unilateral or bilateral/central) and ability to contract the
Affected 357.0 (25.4) 355.0 (29.1) multifidus muscle on the affected side. Patients with unilateral LBP
Unaffected 360.2 (30.3) 401.4 (34.6) had a greater mean difference in multifidus CSA and were smaller
Multifidus contraction on the side that they had a poor multifidus contraction. These
Poor 385.8 (28.2) 323.5 (39.3) results suggest that the alterations of motor control seen in chronic
Good 338.2 (41.4) 432.9 (48.0)
LBP patients are very specific and may require equally localized
a
Values are mean (standard error) in mm2. rehabilitation strategies.
While there are inherent similarities between the results of the
between pain distribution (unilateral or bilateral/central) and current investigation and previous studies (Richardson et al.,
ability to contract the multifidus muscle on the affected side 2004a; Hides et al., 2008b, 2010; Wallwork et al., 2009), there are
(F ¼ 8.18, p ¼ 0.007). Compared to the bilateral/central pain group, methodological differences. The previous studies described, all
the unilateral pain group had a greater mean difference in multi- used imaging techniques (ultrasound imaging and MRI) to docu-
fidus CSA and was smaller on the side that they had a poor mul- ment the results of muscle tests. MRI and ultrasound imaging
tifidus contraction (Table 2). Furthermore, duration of symptoms testing of the TrA muscle was performed in a supine position. The
was significantly related to multifidus muscle size (F ¼ 6.34, prior studies also used comparison groups and blinded assessors.
p ¼ 0.016). The direction of effect, indicated by a negative correla- As the results of these studies are similar, the results suggest that
tion, showed that the longer duration of symptoms was related to well trained clinicians can adequately perform these muscle tests in
smaller multifidus muscle size. There was no significant effect for the field without necessarily requiring sophisticated equipment.
contraction of the TrA muscle, pain on manual examination or the It is also important to note that in the current study, we cannot
other clinical measures assessed (p > 0.05). determine cause and effect with respect to the presence of LBP. We
are not able to determine whether the LBP caused the deficit in
4. Discussion motor control of the TrA and multifidus muscles, or if a deficit in the
motor control of the deep abdominal muscle predisposed the
The main result of the current study was that the ability to patients to LBP. However, in support of the argument for LBP
contract the multifidus muscle (at the L5 vertebral level) was preceding changes in motor control, results of laboratory studies
related to the ability to contract the TrA muscle, i.e. the odds of using induced experimental LBP have demonstrated alterations in
a good contraction of the multifidus muscle were 4.45 times higher motor control of the TrA and multifidus muscles (Hodges et al.,
for patients who had a good contraction of the TrA muscle 2003b; Kiesel et al., 2008). We can however confirm results from
compared with those who had a poor ability. other studies that these changes in motor control are positively
The muscle test for the TrA muscle was performed formally in related to an increased duration of symptoms. Previous research
prone lying, which is an anti-gravity position for this muscle (Barker et al., 2004) has also shown a relationship between
(Hides et al., 2004). To be in the ‘good’ category for this muscle test decreased multifidus muscle size and duration of symptoms, indi-
(0e2 and 3e6 mmHg decrease in pressure) is quite difficult, as this cating that multifidus muscle atrophy is associated with longer
represents a concentric, anti-gravity, inner range contraction of symptom duration. In the current study, other measures such as the
the TrA muscle. Approximately 24% percent of this study’s subjects HAQ, RMQ and VAS were unrelated to multifidus size, asymmetry
presenting to the back clinic could perform this test, indicating and ability to contract the muscle.
that while dysfunction of the TrA (as assessed by the formal The result showing the ability to contract the multifidus muscle
muscle test) is common in patients with chronic LBP, it is not was related to the ability to contract the TrA muscle, is not likely to
present in all subjects in this population. Other studies have also be surprising to clinicians in the field, who commonly assess and
examined the ability to draw in the abdominal wall. Using MRI, treat these muscles together in clinical practice. However, this
differences in the ability to draw in the abdominal wall have relationship has not been previously tested and documented in
similarly been documented in subjects with LBP (Richardson et al., a quantitative research study, though there is evidence to support
2004b; Hides et al., 2008b, 2010). In these studies, subjects with this approach to rehabilitation (Macedo et al., 2009). It is important
LBP were shown to be less able to concentrically shorten the TrA to note that the current investigation is of a clinical nature. This is
muscle to decrease the CSA of their trunk (Richardson et al., different from laboratory studies which have shown that the acti-
2004b; Hides et al., 2008b). However, a limitation of these vation of the TrA is neurophysiologically linked to the activation of
studies is that only muscle testing of the TrA muscle was reported. the pelvic floor muscles (Sapsford et al., 2001). Future laboratory
One of these studies (Hides et al., 2008b) reported a concurrent studies would be required to verify if a neurophysiological rela-
decreased CSA of the multifidus (plus lumbar erector spinae) tionship exists between the TrA and multifidus muscles.
muscles in those with LBP, but the results of muscle testing of the
multifidus were not reported. 4.1. Limitations and future directions
For the multifidus muscle, the results of the current study are
consistent with those of Wallwork et al. (2009). Wallwork et al. The data for this investigation were obtained by a retrospective
(2009) compared the CSA and the ability to voluntarily perform chart audit (clinical data) rather than from a rigorously designed
an isometric contraction of the multifidus muscle at four vertebral scientific investigation. Limitations of this clinical data include
levels in subjects with and without chronic LBP. Results showed interpretation of clinical notes and lack of blinding of assessors to
a significantly smaller CSA of the multifidus muscle for the subjects patient presentation, as this was not possible in the clinical situa-
in the chronic LBP group, compared with subjects from the healthy tion. However, the results of the study can be used to guide clinical
group at the L5 vertebral level. Results of the muscle test showed decisions and formulate appropriate questions for future research
a corresponding smaller contraction for subjects with LBP at the studies.
 J. Hides et al. / Manual Therapy 16 (2011) 573e577 577

Acknowledgements Hodges PW, Richardson CA. Inefficient muscular stabilisation of the lumbar spine
associated with low back pain: a motor control evaluation of transversus
abdominis. Spine 1996;21:2640e50.
The authors would like to acknowledge the subjects studied, Ms Hodges PW, Richardson CA. Delayed postural contraction of transversus abdominis
Linda Blackwell (Director of Physiotherapy, Mater Health Services), in low back pain associated with movement of the lower limb. Journal of Spinal
Skye Shaw (Receptionist, Mater/ACU Back Stability Clinic), Craig Disorders 1998;11(1):46e56.
Huskisson EC. Visual analogue scales. In: Melzack R, editor. Pain measurement and
Gilmore (Physiotherapist), and Megan Bambeery for administering assessment. New York: Raven Press; 1983. p. 33e7.
and scoring the questionnaires. The contribution of Amanda Kader D, Wardlaw D, Smith F. Correlation bewteen the MRI changes in the lumbar
Bromley, Rebecca Davis, Angela Phillott and Rebecca Robinson in multifidus muscles and leg pain. Clinical Radiology 2000;55:145e9.
Keifer A, Shirazi-Adl A, Parnianpour M. Stability of the human spine in neutral
compiling information for this report is acknowledged. postures. European Spine Journal 1997;6:45e53.
Keifer A, Shirazi-Adl A, Parnianpour M. Synergy of the human spine in neutral
postures. European Spine Journal 1998;7:471e9.
Kiesel KB, Uhl T, Underwood FB, Nitz AJ. Rehabilitative ultrasound measurement of
References select trunk muscle activation during induced pain. Manual Therapy 2008;13:
132e8.
Baecke J, Burema J, Frejters J. A short questionnaire for the measurement of habitual Koppenhaver SL, Hebert JJ, Fritz JM, Parent EC, Teyhen DS, Magel JS. Reliability of
activity in epidemiological studies. The American Journal of Clinical Nutrition rehabilitative ultrasound imaging of the transversus abdominis and lumbar
1982;36:936e42. multifidus muscles. Archives of Physical Medicine and Rehabilitation 2009;
Barker KL, Shamley DR, Jackson D. Changes in the cross-sectional area of multifidus 90(1):87e94.
and psoas in patients with unilateral back pain: the relationship to pain and Macedo LG, Maher CG, Latimer J, McAuley JH. Motor control exercise for persistent,
disability. Spine 2004;29(22):E515e9. nonspecific low back pain: a systematic review. Physical Therapy 2009;89(1):
Barker PJ, Guggenheimer KT, Grkovic I, Briggs CA, Jones DC, Thomas CD. Effects of 9e25.
tensioning the lumbar fascia on segmental stiffness during flexion and exten- Maitland GD. Vertebral manipulation. 5th ed. London: Butterworth Heinemann;
sion. Spine 2006;31(4):397e405. 1986.
Bunce SM, Hough AD, Moore AP. Measurement of abdominal muscle thickness Moseley GL. Impaired trunk muscle function in sub-acute neck pain: etiologic in the
using M-mode ultrasound imaging during functional activities. Manual Therapy subsequent development of low back pain? Manual Therapy 2004;9(3):157e63.
2004;9(1):41e4. Panjabi M. The stabilizing system of the spine: part I: function, dysfunction,
Danneels LA, Vanderstraeten GG, Cambier DC, Witvrouw EE, De Cuyper HJ. CT adaptation and enhancement. Journal of Spinal Disorders 1992a;5:383e9.
imaging of trunk muscles in chronic low back pain patients and healthy control Panjabi M. The stabilizing system of the spine. part II: neutral zone and instability
subjects. European Spine Journal 2000;9(4):266e72. hypothesis. Journal of Spinal Disorders 1992b;5:390e7.
Hides JA, Boughen CL, Stanton WR, Strudwick MW, Wilson SJ. A magnetic resonance Panjabi M, Abumi K, Duranceau J, Oxland T. Spinal stability and intersegmental
imaging investigation of the transversus abdominis muscle during drawing-in muscle forces: a biomechanical model. Spine 1989;14:194e200.
of the abdominal wall in Elite Australian Football League Players with and Richardson C, Hides J, Hodges P. Principles of the ‘segmental stabilization’ exer-
without low back pain. Journal of Orthopaedic & Sports Physical Therapy 2010; cise model. In: Richardson CA, Hodges PW, Hides JA, editors. Therapeutic
40(1):4e10. exercise for lumbopelvic stabilization: a motor control approach for the
Hides J, Gilmore C, Stanton W, Bohlscheid E. Multifidus size and symmetry among treatment and prevention of low back pain. Edinburgh: Churchill Livingstone;
chronic LBP and healthy asymptomatic subjects. Manual Therapy 2008a;13(1): 2004a. p. 175e83.
43e9. Richardson CA, Hides JA, Wilson S, Stanton W, Snijders CJ. Lumbo-pelvic joint
Hides JA, Richardson CA, Hodges PW. Local segmental control. In: Richardson CA, protection against antigravity forces: motor control and segmental stiffness
Hodges PW, Hides JA, editors. Therapeutic exercise for lumbo-pelvic stabiliza- assessed with magnetic resonance imaging. Journal of Gravitational Physiology
tion: a motor control approach for the treatment and prevention of low back 2004b;11(2):119e22.
pain. Edinburgh: Churchill Livingstone; 2004. p. 185e219. Roland MA, Morris RM. A study of the natural history of back pain. Part I: devel-
Hides JA, Richardson CA, Jull GA. Multifidus muscle recovery is not automatic opment of a reliable and sensitive measure of disability in low back pain. Spine
following resolution of acute first episode low back pain. Spine 1996;21(23): 1983;8(2):141e4.
2763e9. Sapsford RR, Hodges PW, Richardson CA, Cooper DH, Markwell SJ, Jull GA. Co-
Hides JA, Scott Q, Jull G, Richardson CA. A clinical palpation test to check the activation of the abdominal and pelvic floor muscles during voluntary exercises.
activation of the deep stabilising muscles of the lumbar spine. International Neurourology and Urodynamics 2001;20(1):31e42.
SportMed Journal 2000;1(4):1e4. Snijders CJ, Vleeming A, Stoekart R, Mens JMA, Kleinrensink GJ. Biomechanical
Hides JA, Stanton W, Freke M, Wilson S, McMahon S, Richardson C. MRI study of the size, modeling of sacroiliac joint stability in different postures. Spine: State of the Art
symmetry and function of the trunk muscles among elite cricketers with and Reviews 1995;9:419e32.
without low back pain. British Journal of Sports Medicine 2008b;42(10):809e13. Teyhen DS, Miltenberger CE, Deiters HM, Del Toro YM, Pulliam JN, Childs JD, et al.
Hides JA, Stokes MJ, Saide M, Jull GA, Cooper DH. Evidence of lumbar multifidus The use of ultrasound imaging of the abdominal drawing-in maneuver in
muscle wasting ipsilateral to symptoms in patients with acute/subacute low subjects with low back pain. Journal of Orthopaedic & Sports Physical Therapy
back pain. Spine 1994;19:165e72. 2005;35(6):346e55.
Hodges PW, Eriksson AE, Shirley D, Gandevia SC. Intra-abdominal pressure Wallwork TL, Hides JA, Stanton WR. Intrarater and interrater reliability of assess-
increases stiffness of the lumbar spine. Journal of Biomechanics 2005;38: ment of lumbar multifidus muscle thickness using rehabilitative ultrasound
1873e80. imaging. Journal of Orthopaedic & Sports Physical Therapy 2007;37(10):
Hodges PW, Holm AK, Holm S, Ekstrom L, Cresswell A, Hansson T, et al. Interver- 608e12.
tebral stiffness of the spine is increased by evoked contraction of transversus Wallwork TL, Stanton WR, Freke M, Hides JA. The effect of chronic low back pain on
abdominis and the diaphragm: in vivo porcine studies. Spine 2003a;28(23): size and contraction of the lumbar multifidus muscle. Manual Therapy 2009;
2594e601. 14(5):496e500.
Hodges PW, Moseley GL, Gabrielsson A, Gandevia SC. Experimental muscle pain Wilke HJ, Wolf S, Claes LE, Arand M, Wiesend A. Stability increase of the lumbar
changes feedforward postural responses of the trunk muscles. Experimental spine with different muscle groups: a biomechanical in vitro study. Spine 1995;
Brain Research 2003b;151(2):262e71. 20:192e8.