Professional Documents
Culture Documents
PAUL W. HODGES, PT, PhD, DSc, MedDr, BPhty (Hons)1 • LIEVEN DANNEELS, PT, PhD2
B
ack muscle function is a prerequisite for optimal control ments: different mechanisms,
of spinal stiffness and movement. Muscle structure affects time dependencies, and relation-
muscle function, function affects structure, and pain/ ships to structure, function, and
outcomes. Results are variable
Copyright © 2019 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
cle fiber change, fatty infiltration, reduced and back muscle changes has many ele- with progressively longer/superficial fibers
crossing up to 5 segments79 (often referred
UUSYNOPSIS: Spinal health depends on optimal to subacute/recurrent to chronic states of the to as the superficial multifidus [SM]75,86)
back muscle performance, and this is determined condition. Further, these adaptations have been (FIGURE 1). Lateral to the multifidus are the
by muscle structure and function. There has been shown to be explained by different time-dependent longissimus and iliocostalis, which include
substantial research evaluating the differences in mechanisms. This has substantial impact on the
lumbar78 and thoracolumbar78 portions
structure and function of many back muscles, in- rationale for rehabilitation approaches. The aim
cluding the multifidus and erector spinae, but with (collectively referred to as the erector spi-
of this commentary was to review and consolidate
considerable variation in results. Many studies the breadth of research investigating adaptation nae [ES]). Functionally, the DM primarily
have shown atrophy, fat infiltration, and connective in back muscle structure and function, to consider provides compression, with a limited ex-
tissue accumulation in back muscles, particularly explanations for some of the variation between tension moment,9 with relevance for seg-
deep fibers of the multifidus, but the results are studies, and to propose how this model can be mental control,75,86 whereas the SM and ES
not uniform. In terms of function, results are also used to guide rehabilitation in a manner that is
somewhat inconsistent, often reporting lower mul- generate spine extension, and a lesser con-
tailored to individual patients and to underly-
tifidus activation and augmented recruitment of tribution to lateral flexion and rotation,9
ing mechanisms. J Orthop Sports Phys Ther
more superficial components of the multifidus and
2019;49(6):464-476. doi:10.2519/jospt.2019.8827
to move the spine or increased stiffness
erector spinae, but, again, with variation between when cocontracting with antagonist mus-
studies. A major recent observation has been the UUKEY WORDS: acute back pain, chronic back
cles.14,75,86 Further, recent work has shown
identification of time-dependent differences in pain, electromyography, imaging, multifidus
features of back muscle adaptation, from acute muscle, rehabilitation that multifidus muscle fibers are shorter
and arranged in tightly packed bundles,
1
Clinical Centre for Research Excellence in Spinal Pain, Injury and Health, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia.
2
Department of Physical Therapy and Rehabilitation, Ghent University, Ghent, Belgium. Dr Hodges receives book royalties from Elsevier. Professional and scientific bodies have
reimbursed him for travel costs related to presentation of research on pain, motor control, and exercise therapy at scientific conferences/symposia. He has received fees for
teaching practical courses on motor control training. He is also supported by a Senior Principal Research Fellowship from the National Health and Medical Research Council of
Australia (APP1102905). The authors certify that they have no affiliations with or financial involvement in any organization or entity with a direct financial interest in the subject
matter or materials discussed in the article. Address correspondence to Dr Paul Hodges, School of Health and Rehabilitation Sciences, The University of Queensland, Level 3,
Therapies Annex (84A), St Lucia, QLD 4072 Australia. E-mail: p.hodges@uq.edu.au t Copyright ©2019 Journal of Orthopaedic & Sports Physical Therapy®
and decreased.87 During walking, Arendt- is inconsistent with the stereotypical pain were observed at multiple levels45 but
Nielsen et al3 showed increased ES EMG adaptation proposed by Lund et al.73 In- was consistent with localized immediate
during swing (the phase of low ES acti- stead, changes appear to take advantage of reduction in excitability of spinal neural
vation) and decreased ES EMG during the versatility of the complex trunk mus- pathways, assessed using stimulation of
stance (the phase of high ES activation). cle system to enhance spine protection the spinal cord.49 This parallels reduced
Copyright © 2019 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
Together, these data show task-specific in response to acute noxious input using multifidus reflex responses to electrical
changes in the multifidus and ES. One person-specific strategies, as predicted IVD stimulation in pigs after infusion
interpretation is that when challenged by contemporary models of pain adapta- of physiologic saline into a facet joint.59
by acute noxious input, multifidus and ES tion.54 High-load tasks have been found to Both observations could be explained by
activation is decreased when the muscles have unchanged muscle activation (Dan- reflex inhibition, similar to that observed
produce spine extension (prone arm el- neels et al, unpublished data), which may after knee injury100 (FIGURE 3). In contrast,
evation, forward weight shift, stance indicate that no option for modification response of the multifidus to motor cor-
phase), but increased when the activation of the recruitment strategy is available if tex stimulation increased after IVD le-
prevents spine motion (rapid arm flexion, output is to be maintained, or that subtle sion.49 Whether multifidus activation is
Journal of Orthopaedic & Sports Physical Therapy®
swing phase). individual variation was induced but not facilitated or inhibited depends on the
Experimental pain during dynamic observed in group analysis. balance between increased excitability
trunk movements reveals similar task- The conclusion from studies of nox- of descending input from the brain and
specific observations. Reduced trunk flex- ious back input is that back muscle acti- decreased spinal cord excitability. As this
ion velocity and range are accompanied vation is modified in acute LBP, but this may differ between tasks and individu-
by absence of expected ES relaxation at varies between tasks and individuals. Al- als, it is reasonable to speculate that this
terminal flexion, but reduced ES EMG though multifidus and ES activation has could explain diversity in response to ex-
when it extends the spine to upright.118 been reported to increase or decrease, perimental pain in humans; differences
When contraction history is estimated when apparently inconsistent data are in the relative contribution of spinal and
from T2 shifts in muscle function mag- taken together, the data can be reconciled descending inputs to back muscle activa-
netic resonance imaging (MRI), prone to imply a general goal to protect the tion between tasks may shift the balance
trunk extension (between 45° and 0° of spine. This phenomenon is characterized from inhibition to facilitation.
flexion) induces lesser T2 shifts when by increased activation of the ES (and Summary In the acute phase, multifidus
the task is performed with than without some evidence of DM activation) when activation can be reduced or increased,
experimental pain.29 Although T2 rest the spine is challenged into flexion or as depending on the task. The mechanism
values differed between the DM and SM part of a cocontraction, but by reduced appears to be neural, through mecha-
(which might be explained by different multifidus (and some evidence of ES) nisms including spinal reflex inhibition
muscle fiber composition30), pain simi- activation when the task involves active and increased descending drive. The
larly affected both muscle regions.31 extension of the spine (FIGURE 2). Further, observation of atrophy of the multifidus
Other trunk movements reveal less ste- an understanding of the mechanisms for implies that, although multifidus activa-
reotypical outcomes. Although activation task-specific differences has been pro- tion may be inhibited or facilitated, the
of the DM was not recorded, net muscle vided by animal studies. net effect of inputs to the multifidus in
activation recorded with surface EMG Animal Studies of Tissue Injury Ani- the acute phase is likely to be inhibition.
electrodes over 12 abdominal and back mal studies enable investigation of the Although many studies focus on the mul-
with flexors
hypertrophy, this implies that early muscle
atrophy is not related to loss of muscle fi-
ber mass, and simple activation to over-
come inhibition is sufficient to restore
muscle health. This is likely to require Arm elevation in prone67 Activation to extend spine DM decreased
specific attention to the pattern of muscle to aid arm elevation
activation used during the exercise, and
Prone trunk extension29 Activation to extend spine DM decreased
simple extension of the spine is unlikely SM decreased
to be sufficient, as many different patterns
Journal of Orthopaedic & Sports Physical Therapy®
of activation are available to extend the Trunk flexion118 Lowering: relaxation at Lowering: no ES relax-
spine,48 and these might not involve the end flexion ation (ie, increased
Elevation: activation to activation)
multifidus. This would argue for an ap-
extend trunk Elevation: ES decreases
proach that specifically targets activation during elevation
of the multifidus, such as a motor control
approach for rehabilitation of back pain.55
Third, many current clinical practice Walking3 Stance: activation during Stance ES decreased
guidelines recommend that patients not stance Swing ES increased
seek care for an acute episode of uncom- Swing: relaxation during
plicated back pain and instead remain swing
active. However, the potential for multifi-
dus changes to recover with exercise but
to fail to recover with general functional
activity, and the potential for training to Slow trunk flexion and Cocontraction of flexor and ES variable increased
extension around extensor muscles to
reduce recurrence of LBP (all shown in a
neutral47 stabilize trunk around
small study of acute LBP and requiring neutral
reproduction),106 provides a foundation to
consider that benefit may be gained from
early intervention to restore multifidus
activation. This needs to be undertaken
with an emphasis on optimizing spine FIGURE 2. Changes in back muscle function in acute experimental back pain. Summary of tasks tested, function
health, with care not to instill a belief attributed to the trunk muscles in these tasks, and changes that have been observed in muscle activation. Data
in the patient that the spine is “at risk,” support the proposal that adaptation in acute pain depends on the function performed by the muscle in a specific
task. Abbreviations: DM, deep multifidus; ES, erector spinae; SM, superficial multifidus.
which may promote unhealthy attitudes
comparison with other approaches. controls, with greater reduction on the the task. Extensor muscle EMG was also
previously painful side.74 Superficial higher when acting as an antagonist dur-
Changes in Back Muscle Structure multifidus EMG increased earlier than ing flexion and when approaching the
and Function in Subacute LBP on the previously painful side or in pain- return to upright. Deep multifidus EMG
and During Remission free controls.74 was higher in the period before move-
Copyright © 2019 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
Subacute and Recurrent Clinical LBP Of This change in function and differ- ment, but during flexion, participants with
interest is the potential role for persis- entiation between regions of the mul- a history of unilateral LBP demonstrated
Injury
ACUTE PHASE
Reflex inhibition
Initial lesion
66,67
Journal of Orthopaedic & Sports Physical Therapy®
Pain
spine control Loss of slow fibers (subacute); increased fast fibers (chronic)
3 mo
51
Sensitization of Increased fatigability
Nociceptor nociceptors of muscle83
stimulation
61 M2-to-M1 transition of muscle
Fear of pain; and adipose macrophages61
catastrophization
Pain
6 mo
Muscle unloading
FIGURE 3. Proposed model of the timeline and mechanisms underlying the structural and inflammatory changes in the multifidus muscle after intervertebral disc lesion. Three
phases are shown, with different mechanisms and different changes in structure and function: acute (top), subacute-early chronic (middle), and chronic (bottom). Citations
from the text that provide evidence for the proposal’s features and the causal links are provided. Abbreviations: IL, interleukin; TNF, tumor necrosis factor.
activation has also been observed, but that inferred from lower resting T2 MRI mea- that multifidus muscle CSA recovers after
tends to occur on the side that was not the sures of the multifidus (but not the ES the acute period, but that fat deposits in-
previously painful side. and psoas) during LBP remission on both crease in those with ongoing or recurrent
Back muscle structure changes have sides at lower lumbar levels.26 As resting symptoms.
been evaluated in cross-sectional stud- T2 measures reflect the resting metabolic Human Experimental Pain Studies Dur-
Copyright © 2019 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
ies using T1-weighted MRI during re- state of the multifidus, lower values in ing LBP Remission When pain is induced
mission of unilateral LBP.27 Although no the LBP group would be consistent with experimentally during LBP remission,
A
Acute Subacute-early chronic Chronic
Acute inhibitory/regeneration phase Proinflammatory cytokine phase Disuse/deconditioning phase
• Muscle inhibition • Slow-to-fast muscle fiber transformation • Muscle/muscle fiber atrophy
• Acute size reduction (vascular/other mechanism) • Fibrosis • Fibrosis
• Acute adipose activation • Fatty infiltration • Fatty infiltration
• Regeneration pathway activation
Journal of Orthopaedic & Sports Physical Therapy®
1 wk 3 mo 6 mo 9 mo 12 mo
B
Motor control training Initial motor control training to restore recruitment pattern
to overcome
Rehabilitation Strategy
inhibition
Strength and endurance
training to restore muscle
size and reduce fat and
connective tissue changes
Exercise to promote anti-inflammatory effect
and fast-to-slow muscle fiber transition
FIGURE 4. Model of the interaction and overlap between the multiple putative mechanisms for structural remodeling of the multifidus after intervertebral disc injury (A), along
with (B) proposed interventions. Each mechanism has a different time course, physiological basis, and consequences for multifidus structure. Although correction of muscle
recruitment patterns is likely to be relevant at each phase, the relative importance of muscle loading to train strength, endurance, and hypertrophy will increase over time.
Adapted with permission from Hodges et al.50
change in muscle structure in detail. At tial stimulus for muscle adaptation is is to address the problem with exercise.
3 and 6 months after IVD lesion, there is reduced muscle loading. This was origi- Exercise can promote M1-to-M2 (anti-in-
no atrophy of lean muscle or individual nally proposed on the basis of differen- flammatory) macrophage polarization,70
muscle fibers.50,51 There is no upregula- tial impact of muscle unloading on “slow” reduce inflammatory cytokine expression
tion of molecular pathways for atrophy and “fast” muscles in rats39; atrophy was in the multifidus (in a rat model of IVD
Copyright © 2019 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
or downregulation of hypertrophy path- greater in the soleus than in the extensor degeneration),61 and promote transition
ways.50 Adipose CSA increased along with digitorum. The hypothesis that reduction from fast to slow muscle fibers.93
connective tissue (fibrosis, which cannot of muscle loading due to gravity would Which type of exercise can promote
be differentiated in human MRI),50,51 preferentially impact the multifidus was an anti-inflammatory effect and modify
and the proportion of slow type I mus- investigated by exposure to 3 months of muscle structure? Animal studies show
cle fibers reduced.51 Although changes bed rest. The CSA of the multifidus, but that regular general physical activity
were localized to the injured level at 3 not those of the ES and quadratus lum- (mice using a running wheel) can pro-
months, they were more generalized at 6 borum, decreased,7 and psoas CSA in- mote polarization to the anti-inflam-
months.51 In parallel studies of IVD lesion creased. Thus, inactivity affects the back matory M2 macrophages and reduce
Journal of Orthopaedic & Sports Physical Therapy®
in rabbits, passive mechanical properties muscles nonuniformly, with a similar dis- connective tissue accumulation, but gen-
(stiffness) increased and multifidus fiber tribution (biased toward the multifidus) eral exercise did not prevent changes in
density decreased at 12 weeks, but not 4 to that found in LBP. some components of the extracellular
weeks. Changes were not related to fiber Summary Back muscle structure chang- matrix in the multifidus.60 These data
type or protein (titin) changes, but were es persist beyond acute LBP resolution. have 2 important implications. First,
probably related to collagen reorganiza- Although lean muscle atrophy tends to the data highlight that general physical
tion.10 These data imply an important recover, there is evidence from carefully activity can prevent inflammatory and
role for adaptation of noncontractile tis- controlled animal studies suggesting that fibrotic changes in muscles, but whether
sue in the multifidus after injury. structural changes in the multifidus de- the changes, once developed, can be re-
Animal studies have enabled detailed velop to include fibrosis, fat infiltration, versed by general physical activity has not
analysis of mechanism. On the basis of and slow-to-fast muscle fiber transition. been established. This provides a basis
evidence of involvement of proinflam- Thus, although neural mechanisms can to consider potential anti-inflammatory
matory cytokines in persistent LBP113 explain changes in the acute context, sub- effects of early introduction of physical
and muscle remodeling,92 involvement acute changes appear to be more likely to activity. Second, general activity was not
of inflammatory cytokines in multifidus be explained by a muscle inflammatory sufficient to rectify all muscle changes,
structural changes has been examined response (FIGURE 3). In terms of function, which implies that it may be necessary
in the subacute period.50,51 These stud- DM activation is generally decreased, to introduce more specific exercise to
ies have identified elevated messenger and most consistently on the painful address fibrotic changes in the DM mus-
ribonucleic acid expression of proinflam- side. Even if DM activation is increased, cle. There is preliminary evidence that
matory cytokines (tumor necrosis factor, as is suggested in some studies, compro- loaded exercise reduces fat proportion.90
interleukin-1β) within the multifidus by mised muscle structure suggests that the In terms of muscle fiber transformation,
6 months after IVD injury,51 despite the muscle output would be less. Frequently strength training appears to be neces-
absence of injury to the muscle. More observed augmented activation of more sary to increase the proportion and size
recent work highlights that M1 (proin- superficial back muscles (including the of slow-type muscle fibers.
using specialized equipment22 or training Findings for muscle fiber type pro- variable results (see van Dieën et al109).
programs (eg, leg-loading tasks requiring portions in chronic LBP are variable.11,76 This is exemplified by studies of dynamic
control of the trunk).97 As an example, a matched case-control trunk movements that show no differ-
muscle biopsy study revealed lower pro- ence2,68,72,89 or decreased1,12,99,116 ES EMG
Changes in Lumbar Muscle Structure portions of type I fibers and higher pro- during lumbar extension and no differ-
Copyright © 2019 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
and Function in Chronic/Persistent LBP portions of type II and intermediate type ence,68,89 decreased,1,2,116 or increased99 ES
Chronic/Persistent Clinical LBP When IIc fibers (suggesting ongoing fiber tran- EMG during lumbar flexion. When the
LBP persists, back muscle structure sition) in LBP.83 There was no difference multifidus has been studied specifically,
changes become more extensive. Most in the CSA of individual fibers,83 consis- Danneels et al23 showed lower multifidus
studies identify smaller multifidus CSA tent with animal studies,50 suggesting a EMG during gentle lordosis coordination
that is bilateral (unlike changes that smaller area of muscle occupied with type exercises, but lower activity of all back
are specific to a painful side in acute I fibers and lower fatigue resistance.83 muscles during high-load exercises. This
LBP).8,25,37,112 Results differ between stud- Further evidence is provided by a neg- was corroborated by higher T2 shifts in
ies for other muscles. Parkkola et al91 re- ative correlation between proportion of the multifidus and ES after trunk exten-
Journal of Orthopaedic & Sports Physical Therapy®
ported atrophy of a combined measure type I fibers and duration of pain, but a sion, consistent with reduced back mus-
of the ES and multifidus, whereas sev- positive correlation with type II fibers.82 cle endurance.40
eral studies have reported atrophy of the Comparison of T2 resting values shows We speculate that variation is likely
multifidus but not of the ES in chronic a tendency, although nonsignificant, to- to have several possible explanations.
LBP.8,25 Kamaz et al64 reported more ward a lower mean value for the multifi- First, different adaptations have been
generalized, smaller multifidus, psoas, dus and ES in LBP, suggesting a higher observed in different patient groups (eg,
and quadratus lumborum CSAs. Smaller type II fiber proportion.40 In contrast, LBP subgroups20). For instance, studies
multifidus (and psoas) CSA was observed Crossman et al18 found no differences in have shown opposite changes in mul-
with longer-duration LBP.17 Comparisons fiber size, type I fiber proportion, or area tifidus activation (recorded with surface
between continuous versus intermittent occupied by type I fibers in mild disabling EMG electrodes) depending on whether
LBP found no difference in the multifidus LBP, despite earlier failure during a back the individual with LBP typically adopts
or ES.40,57 extensor muscle fatigue test. Some varia- a flexed or extended posture in sitting.20
Analysis of individuals with asym- tion may be explained by different biopsy Second, individual differences in mo-
metrical pain provides some variable locations, symptom severity, and control tor adaptation have been observed, but
data. Some report smaller CSA of the samples harvested from cadavers with with a similar goal.54 For superficial
multifidus,58,62 psoas,19 or both the mul- unclear LBP history.76 An important con- back muscles, including the ES, esti-
tifidus and psoas5 on the painful side, but sideration is that human studies are all mates from mathematical modeling
extending over multiple segments.5 In cross-sectional, and no longitudinal data of the net outcome of muscle adapta-
horses, facet degeneration of unknown are available. Animal studies that test tion (regardless of individual pattern)
duration involves localized reduction in separate animals at different time points show increased “stability” or protection
multifidus CSA.102 but in very carefully controlled conditions of the spine.35,107 Although this could
Fat depositions that are either lo- provide sensitive evidence of muscle fiber be reasonable to protect the spine from
calized to the multifidus or distributed differences.50 Longitudinal human studies further pain/injury in the short term, it
more generally have been reported using would help resolve this issue. has the long-term costs54 of increased
controlled, can lead to variation. muscles for spine protection, and some general exercise.
Some work has considered automatic evidence of compromised function of the There are several issues to consider.
control of spinal posture in chronic LBP. DM, with some differences explained by First, specific motor control training
Studies of trunk loading frequently re- patient subgroups. Notably, back muscle may not be more effective than general
ported delayed ES reaction in predictable function must be considered with respect exercise, despite physiological evidence
Copyright © 2019 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
and unpredictable perturbations,80 or to muscle structure. Following the sub- that implies that specific attention to the
only during predictable perturbations.69 acute inflammatory-related mechanisms structure and function of muscles such as
Back muscle function has also been in- for muscle changes, the dramatic muscle the DM may be required. Second, stud-
vestigated from the perspective of central structural changes in chronic LBP have ies that compare motor control training
nervous system function using transcra- been generally explained by disuse sec- to general exercise have generally ap-
nial magnetic stimulation. These studies ondary to changes in movement patterns plied motor control training to a group
show lower excitability of descending (shielding the DM from load15), pain/fear with nonspecific LBP, and in a manner
pathways to the ES101 and modified or- avoidance,65 or deconditioning.82 These that is not individualized to the patient.
ganization of the motor cortex repre- possible mechanisms require further ex- This contrasts with evidence summarized
Journal of Orthopaedic & Sports Physical Therapy®
sentation of the multifidus.105 This latter amination in longitudinal studies. above that highlights individual variation
feature was characterized by convergence This could include excessive protec- in changes in structure and function of
of the discrete brain representations of tion, often involving the more superficial the back muscles and implies that exer-
the DM and ES. This correlates with LBP ES, requiring strategies to reduce activa- cise must be tailored to the individual.
severity96 but was specific to individuals tion and a compromised DM and strate- Further, motor control intervention is un-
who presented with poor ability to differ- gies to enhance function and structure. likely to be appropriate for all individuals.
entiate lumbar from thoracolumbar mo- Identification of the features to address For instance, individuals with pain that is
tion.34 Changes in corticomotor function may be facilitated by assessment of move- primarily maintained by central sensi-
support the argument for compromised ments,33 posture,20 and pain characteris- tization may not be appropriate for this
multifidus function in LBP; however, tics. Of note, it is likely that restoration intervention (see Hodges46 for review).
further work is required to understand of fatty and fibrotic changes in muscle Although not a focus of this commentary,
the relationship between brain changes, structure would require resistance train- the biopsychosocial nature of LBP implies
motor function, and symptoms. ing once activation of the muscle is es- that psychosocial factors may be the pri-
Consideration of changes in mul- tablished. Danneels et al22 showed that ority targets for intervention for some in-
tifidus/ES muscle structure and muscle low-load motor control training was in- dividuals and may interact with biological
function together exposes an important sufficient to restore muscle CSA in this features, including back muscle structure
complication for interpretation. If lean case; controlled application of progres- and/or function. This has been confirmed
muscle is reduced in chronic LBP in as- sive overload after low-load training to in chronic (and acute) LBP65,87,115 and
sociation with fibrosis and fat infiltra- improve motor patterns was required to must be considered in comprehensive
tion, then force output would be lower produce hypertrophy in the multifidus LBP management. The relative weight-
despite identical or increased EMG. and reduce pain/disability.24 This is sup- ing of psychosocial and motor interven-
Notably, EMG may be greater relative to ported by results of a recent systematic tions will likely depend on the individual,
maximum voluntary contraction than in review.97 Such training might also reduce which has been confirmed in several ran-
controls but generate less force. Similar fat infiltration.85 Reduced type I muscle domized controlled trials.36,77,111
T
his commentary aimed to sum- the cross-sectional area of multifidus and psoas wasting in patients with chronic low back pain.
in patients with unilateral back pain: the relation- Br J Rheumatol. 1992;31:389-394. https://doi.
marize the state of knowledge with
ship to pain and disability. Spine (Phila Pa 1976). org/10.1093/rheumatology/31.6.389
respect to changes in back muscle 2004;29:E515-E519. 18. Crossman K, Mahon M, Watson PJ, Oldham JA,
structure and function in LBP, from acute 6. B attié MC, Niemelainen R, Gibbons LE, Dhillon Cooper RG. Chronic low back pain-associated
to chronic contexts. A major observation S. Is level- and side-specific multifidus asym- paraspinal muscle dysfunction is not the result
metry a marker for lumbar disc pathology? Spine of a constitutionally determined “adverse”
that explains much of the variation ob-
J. 2012;12:932-939. https://doi.org/10.1016/j. fiber-type composition. Spine (Phila Pa 1976).
served in the literature is the time-depen- spinee.2012.08.020 2004;29:628-634.
dent nature of changes and underlying 7. B elavý DL, Armbrecht G, Richardson CA, Felsen- 19. Dangaria TR, Naesh O. Changes in cross-sec-
mechanisms, and the need to consider berg D, Hides JA. Muscle atrophy and changes in tional area of psoas major muscle in unilateral
spinal morphology: is the lumbar spine vulner- sciatica caused by disc herniation. Spine (Phila
different approaches to managing LBP at
able after prolonged bed-rest? Spine (Phila Pa Pa 1976). 1998;23:928-931.
different times. Successful management 1976). 2011;36:137-145. https://doi.org/10.1097/ 20. D
ankaerts W, O’Sullivan P, Burnett A, Straker L.
will depend on individual examination BRS.0b013e3181cc93e8 Altered patterns of superficial trunk muscle activa-
Downloaded from www.jospt.org at on July 19, 2022. For personal use only. No other uses without permission.
of adaptation of back muscle structure 8. B eneck GJ, Kulig K. Multifidus atrophy is local- tion during sitting in nonspecific chronic low back
ized and bilateral in active persons with chronic pain patients: importance of subclassification.
and function and the relative impor-
unilateral low back pain. Arch Phys Med Rehabil. Spine (Phila Pa 1976). 2006;31:2017-2023. https://
tance of psychosocial features to develop 2012;93:300-306. https://doi.org/10.1016/j. doi.org/10.1097/01.brs.0000228728.11076.82
a treatment strategy with consideration apmr.2011.09.017 21. Danneels L, Cagnie B, D’Hooge R, et al. The ef-
of time-dependent mechanisms to tailor 9. B ogduk N, Macintosh JE, Pearcy MJ. A universal fect of experimental low back pain on lumbar
Copyright © 2019 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
intervention to an individual. t model of the lumbar back muscles in the upright muscle activity in people with a history of clinical
position. Spine (Phila Pa 1976). 1992;17:897-913. low back pain: a muscle functional MRI study.
10. B
rown SH, Gregory DE, Carr JA, Ward SR, Masu- J Neurophysiol. 2016;115:851-857. https://doi.
ACKNOWLEDGMENTS: The forum on which this da K, Lieber RL. ISSLS Prize winner: adaptations org/10.1152/jn.00192.2015
body of research was based, “State-of-the-Art to the multifidus muscle in response to experi- 22. Danneels LA, Cools AM, Vanderstraeten
mentally induced intervertebral disc degenera- GG, et al. The effects of three different train-
in Motor Control and Low Back Pain: Inter-
tion. Spine (Phila Pa 1976). 2011;36:1728-1736. ing modalities on the cross-sectional area
national Clinical and Research Expert Fo- https://doi.org/10.1097/BRS.0b013e318212b44b of the paravertebral muscles. Scand J Med
rum,” was supported by the National Health 11. C
agnie B, Dhooge F, Schumacher C, et al. Fiber Sci Sports. 2001;11:335-341. https://doi.
and Medical Research Council of Australia, typing of the erector spinae and multifidus org/10.1034/j.1600-0838.2001.110604.x
muscles in healthy controls and back pain pa- 23. D
anneels LA, Coorevits PL, Cools AM, et al. Differ-
in collaboration with the North American
Journal of Orthopaedic & Sports Physical Therapy®
tients: a systematic literature review. J Manipula- ences in electromyographic activity in the multifi-
Spine Society. The forum was chaired by Dr tive Physiol Ther. 2015;38:653-663. https://doi. dus muscle and the iliocostalis lumborum between
Paul Hodges. org/10.1016/j.jmpt.2015.10.004 healthy subjects and patients with sub-acute and
12. C
assisi JE, Robinson ME, O’Conner P, MacMillan chronic low back pain. Eur Spine J. 2002;11:13-19.
M. Trunk strength and lumbar paraspinal muscle https://doi.org/10.1007/s005860100314
activity during isometric exercise in chronic low- 24. Danneels LA, Vanderstraeten GG, Cambier DC, et
REFERENCES
back pain patients and controls. Spine (Phila Pa al. Effects of three different training modalities on
1976). 1993;18:245-251. the cross sectional area of the lumbar multifidus
1. Ahern DK, Follick MJ, Council JR, Laser-
13. C
hae GN, Kwak SJ. NF-κB is involved in the muscle in patients with chronic low back pain.
Wolston N, Litchman H. Comparison of lumbar
TNF-α induced inhibition of the differentiation Br J Sports Med. 2001;35:186-191. https://doi.
paravertebral EMG patterns in chronic low
of 3T3-L1 cells by reducing PPARγ expression. org/10.1136/bjsm.35.3.186
back pain patients and non-patient con- Exp Mol Med. 2003;35:431-437. https://doi. 25. Danneels LA, Vanderstraeten GG, Cambier
trols. Pain. 1988;34:153-160. https://doi. org/10.1038/emm.2003.56 DC, Witvrouw EE, De Cuyper HJ. CT imaging
org/10.1016/0304-3959(88)90160-1 14. C
holewicki J, Panjabi MM, Khachatryan A. Stabi- of trunk muscles in chronic low back pain pa-
2. Arena JG, Sherman RA, Bruno GM, Young TR. lizing function of trunk flexor-extensor muscles tients and healthy control subjects. Eur Spine
Electromyographic recordings of 5 types of low around a neutral spine posture. Spine (Phila Pa J. 2000;9:266-272. https://doi.org/10.1007/
back pain subjects and non-pain controls in dif- 1976). 1997;22:2207-2212. s005860000190
ferent positions. Pain. 1989;37:57-65. https://doi. 15. C
laus AP, Hides JA, Moseley GL, Hodges PW. 26. D’Hooge R, Cagnie B, Crombez G, Vanderstraeten
org/10.1016/0304-3959(89)90153-X Different ways to balance the spine in sitting: G, Achten E, Danneels L. Lumbar muscle dys-
3. Arendt-Nielsen L, Graven-Nielsen T, Svarrer muscle activity in specific postures differs function during remission of unilateral recurrent
H, Svensson P. The influence of low back between individuals with and without a history nonspecific low-back pain: evaluation with mus-
pain on muscle activity and coordination of back pain in sitting. Clin Biomech (Bristol, cle functional MRI. Clin J Pain. 2013;29:187-194.
during gait: a clinical and experimental Avon). 2018;52:25-32. https://doi.org/10.1016/j. https://doi.org/10.1097/AJP.0b013e31824ed170
study. Pain. 1996;64:231-240. https://doi. clinbiomech.2018.01.003 27. D’Hooge R, Cagnie B, Crombez G, Vanderstraeten
org/10.1016/0304-3959(95)00115-8 16. C
laus AP, Hides JA, Moseley GL, Hodges PW. G, Dolphens M, Danneels L. Increased intramus-
4. Astfalck RG, O’Sullivan PB, Straker LM, et al. Different ways to balance the spine: subtle cular fatty infiltration without differences in lum-
Sitting postures and trunk muscle activity in changes in sagittal spinal curves affect re- bar muscle cross-sectional area during remission
adolescents with and without nonspecific chronic gional muscle activity. Spine (Phila Pa 1976). of unilateral recurrent low back pain. Man Ther.
low back pain: an analysis based on subclassifi- 2009;34:E208-E214. https://doi.org/10.1097/ 2012;17:584-588. https://doi.org/10.1016/j.
cation. Spine (Phila Pa 1976). 2010;35:1387-1395. BRS.0b013e3181908ead math.2012.06.007
study. Man Ther. 2010;15:364-369. https://doi. wasting ipsilateral to symptoms in patients with 56. Hug F, Hodges PW, Tucker K. Muscle force can-
org/10.1016/j.math.2010.02.002 acute/subacute low back pain. Spine (Phila Pa not be directly inferred from muscle activation:
32. Ebenbichler GR, Oddsson LI, Kollmitzer J, 1976). 1994;19:165-172. illustrated by the proposed imbalance of force
Erim Z. Sensory-motor control of the lower 45. H
odges P, Holm AK, Hansson T, Holm S. Rapid between the vastus medialis and vastus lateralis
back: implications for rehabilitation. Med Sci atrophy of the lumbar multifidus follows ex- in people with patellofemoral pain. J Orthop
Sports Exerc. 2001;33:1889-1898. https://doi. perimental disc or nerve root injury. Spine Sports Phys Ther. 2015;45:360-365. https://doi.
org/10.1097/00005768-200111000-00014 (Phila Pa 1976). 2006;31:2926-2933. https://doi. org/10.2519/jospt.2015.5905
33. Elgueta-Cancino E, Schabrun S, Danneels L, org/10.1097/01.brs.0000248453.51165.0b 57. Hultman G, Nordin M, Saraste H, Ohlsèn H. Body
van den Hoorn W, Hodges P. Validation of a 46. H
odges PW. Hybrid approach to treatment tailor- composition, endurance, strength, cross-section-
clinical test of thoracolumbar dissociation in ing for low back pain: a proposed model of care. al area, and density of MM erector spinae in men
chronic low back pain. J Orthop Sports Phys J Orthop Sports Phys Ther. 2019;49:453-463. with and without low back pain. J Spinal Disord.
Journal of Orthopaedic & Sports Physical Therapy®
muscle activation during induced pain. Man org/10.1007/BF00307824 93. Putman CT, Xu X, Gillies E, MacLean IM, Bell
Ther. 2008;13:132-138. https://doi.org/10.1016/j. 81. M
annion AF, Dumas GA, Cooper RG, Espinosa FJ, GJ. Effects of strength, endurance and com-
math.2006.10.003 Faris MW, Stevenson JM. Muscle fibre size and bined training on myosin heavy chain content
68. Larivière C, Gagnon D, Loisel P. The comparison type distribution in thoracic and lumbar regions and fibre-type distribution in humans. Eur J
of trunk muscles EMG activation between sub- of erector spinae in healthy subjects without Appl Physiol. 2004;92:376-384. https://doi.
jects with and without chronic low back pain dur- low back pain: normal values and sex differ- org/10.1007/s00421-004-1104-7
Copyright © 2019 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
ing flexion–extension and lateral bending tasks. ences. J Anat. 1997;190:505-513. https://doi. 94. Reeves NP, Cholewicki J, van Dieën JH, Kaw-
J Electromyogr Kinesiol. 2000;10:79-91. https:// org/10.1046/j.1469-7580.1997.19040505.x chuk G, Hodges PW. Are stability and instabil-
doi.org/10.1016/S1050-6411(99)00027-9 82. M
annion AF, Käser L, Weber E, Rhyner A, Dvorak ity relevant concepts for back pain? J Orthop
69. Leinonen V, Kankaanpää M, Luukkonen M, J, Müntener M. Influence of age and duration of Sports Phys Ther. 2019;49:415-424. https://doi.
Hänninen O, Airaksinen O, Taimela S. Disc herniation- symptoms on fibre type distribution and size of org/10.2519/jospt.2019.8144
related back pain impairs feed-forward control the back muscles in chronic low back pain pa- 95. Saragiotto BT, Maher CG, Yamato TP, et al. Mo-
of paraspinal muscles. Spine (Phila Pa 1976). tients. Eur Spine J. 2000;9:273-281. https://doi. tor control exercise for nonspecific low back
2001;26:E367-E372. org/10.1007/s005860000189 pain: a Cochrane review. Spine (Phila Pa 1976).
70. Leung A, Gregory NS, Allen LA, Sluka KA. Regular 83. M
annion AF, Weber BR, Dvorak J, Grob D, Mün- 2016;41:1284-1295. https://doi.org/10.1097/
physical activity prevents chronic pain by altering resi- tener M. Fibre type characteristics of the lumbar BRS.0000000000001645
dent muscle macrophage phenotype and increasing paraspinal muscles in normal healthy subjects 96. Schabrun SM, Elgueta-Cancino EL, Hodges PW.
Journal of Orthopaedic & Sports Physical Therapy®
interleukin-10 in mice. Pain. 2016;157:70-79. https:// and in patients with low back pain. J Orthop Smudging of the motor cortex is related to the
doi.org/10.1097/j.pain.0000000000000312 Res. 1997;15:881-887. https://doi.org/10.1002/ severity of low back pain. Spine (Phila Pa 1976).
71. Li YP, Schwartz RJ. TNF-α regulates early dif- jor.1100150614 2017;42:1172-1178. https://doi.org/10.1097/
ferentiation of C2C12 myoblasts in an autocrine 84. M
ok NW, Brauer SG, Hodges PW. Failure to BRS.0000000000000938
fashion. FASEB J. 2001;15:1413-1415. https://doi. use movement in postural strategies leads to 97. Shahtahmassebi B, Hebert JJ, Stomski NJ,
org/10.1096/fj.00-0632fje increased spinal displacement in low back pain. Hecimovich M, Fairchild TJ. The effect of exercise
72. Lisiński P. Surface EMG in chronic low back Spine (Phila Pa 1976). 2007;32:E537-E543. training on lower trunk muscle morphology.
pain. Eur Spine J. 2000;9:559-562. https://doi. https://doi.org/10.1097/BRS.0b013e31814541a2 Sports Med. 2014;44:1439-1458. https://doi.
org/10.1007/s005860000131 85. M
ooney V, Gulick J, Perlman M, et al. Relation- org/10.1007/s40279-014-0213-7
73. Lund JP, Donga R, Widmer CG, Stohler CS. The ships between myoelectric activity, strength, and 98. Sihvonen T, Lindgren KA, Airaksinen O, Manninen
pain-adaptation model: a discussion of the rela- MRI of lumbar extensor muscles in back pain H. Movement disturbances of the lumbar spine
tionship between chronic musculoskeletal pain patients and normal subjects. J Spinal Disord. and abnormal back muscle electromyographic
and motor activity. Can J Physiol Pharmacol. 1997;10:348-356. findings in recurrent low back pain. Spine (Phila
1991;69:683-694. https://doi.org/10.1139/y91-102 86. M
oseley GL, Hodges PW, Gandevia SC. Deep Pa 1976). 1997;22:289-295.
74. MacDonald D, Moseley GL, Hodges PW. People and superficial fibers of the lumbar multifidus 99. Sihvonen T, Partanen J, Hänninen O, Soimakallio
with recurrent low back pain respond differently muscle are differentially active during volun- S. Electric behavior of low back muscles during
to trunk loading despite remission from symp- tary arm movements. Spine (Phila Pa 1976). lumbar pelvic rhythm in low back pain patients
toms. Spine (Phila Pa 1976). 2010;35:818-824. 2002;27:E29-E36. and healthy controls. Arch Phys Med Rehabil.
https://doi.org/10.1097/BRS.0b013e3181bc98f1 87. M
oseley GL, Nicholas MK, Hodges PW. Does 1991;72:1080-1087.
75. MacDonald D, Moseley GL, Hodges PW. Why anticipation of back pain predispose to back 100. Stokes M, Young A. The contribution of reflex
do some patients keep hurting their back? trouble? Brain. 2004;127:2339-2347. https://doi. inhibition to arthrogenous muscle weakness.
Evidence of ongoing back muscle dysfunction org/10.1093/brain/awh248 Clin Sci (Lond). 1984;67:7-14. https://doi.
during remission from recurrent back pain. Pain. 88. M
oseley GL, Nicholas MK, Hodges PW. Pain org/10.1042/cs0670007
2009;142:183-188. https://doi.org/10.1016/j. differs from non-painful attention-demanding 101. Strutton PH, Theodorou S, Catley M, McGregor
pain.2008.12.002 or stressful tasks in its effect on postural AH, Davey NJ. Corticospinal excitability in
76. MacDonald DA, Moseley GL, Hodges PW. The control patterns of trunk muscles. Exp Brain patients with chronic low back pain. J Spinal
lumbar multifidus: does the evidence support Res. 2004;156:64-71. https://doi.org/10.1007/ Disord Tech. 2005;18:420-424. https://doi.
clinical beliefs? Man Ther. 2006;11:254-263. s00221-003-1766-0 org/10.1097/01.bsd.0000169063.84628.fe
https://doi.org/10.1016/j.math.2006.02.004 89. N
ouwen A, Van Akkerveeken PF, Versloot JM. 102. Stubbs NC, Riggs CM, Hodges PW, et al. Os-
77. Macedo LG, Maher CG, Hancock MJ, et al. Pre- Patterns of muscular activity during movement in seous spinal pathology and epaxial muscle
(Phila Pa 1976). 2011;36:1721-1727. https://doi. chronic low back pain: a randomized controlled Atrophy of the multifidus muscle in patients
org/10.1097/BRS.0b013e31821c4267 trial. Eur J Pain. 2013;17:916-928. https://doi. with lumbar disk herniation: histochemical
106. Tsao H, Druitt TR, Schollum TM, Hodges PW. Mo- org/10.1002/j.1532-2149.2012.00252.x and electromyographic study. Orthopedics.
tor training of the lumbar paraspinal muscles in- 112. W allwork TL, Stanton WR, Freke M, Hides JA. 2003;26:493-495.
duces immediate changes in motor coordination The effect of chronic low back pain on size and 118. Zedka M, Prochazka A, Knight B, Gillard D,
in patients with recurrent low back pain. J Pain. contraction of the lumbar multifidus muscle.
Copyright © 2019 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
@ MORE INFORMATION
spine. Spine (Phila Pa 1976). 2003;28:834-841. J Pain. 2008;24:273-278. https://doi.org/10.1097/
108. van Dieën JH, Flor H, Hodges PW. Low-back pain AJP.0b013e31816111d3
patients learn to adapt motor behavior with ad- 114. W ard SR, Kim CW, Eng CM, et al. Architectural WWW.JOSPT.ORG
Journal of Orthopaedic & Sports Physical Therapy®