Justin Gibson, BS, CSCS
Introduction
Low back
pain is a common and often debilitating condition affecting an average of 12%
of all persons at any given time and a lifetime prevalence upwards of 60% of
the population across sex, socioeconomic, occupational, and national boundaries
(Hoy, et al., 2012) . Researchers the world over have searched
for underlying causes and contributing factors to low back pain, and the
methods and results they have obtained vary substantially from one study to the
next.
One such
potential factor is posture. The position and alignment of the lumbar vertebrae
is a logical point of interest when studying pain in the region, and as such
has been the focal point of many researchers. However, the literature
surrounding posture and low back pain is complicated, with a great deal of
variety in both defining “posture” and “pain”, and a consensus among
researchers is often illusive. In this paper, I will attempt to provide an
overview of major studies in the realm of static standing posture in the
sagittal plane and its link -or lack thereof- to pain.
Posture and Low Back Pain: The Theory
All studies
begin with a question based on a thorough analysis of the research that came
before them. To ask the question, “Does one’s posture affect their risk of
developing lower back pain?”, there must be a reasonable theory behind why that
might be the case. And, indeed, many studies and textbooks have put forth
potential structural and physiological factors linking the two.
In their
study, “Effect of low back posture on the morphology of the spinal canal”, Chung, Lee, Kim, Chung, & Ahn (2000)
observed structural changes in the spine when subjects were placed in differing
positions during MRI imaging of 20 healthy young volunteers. Using foam
bolsters and stabilization belts, the researchers looked at L3-L4 and L4-L5
relationships when subjects were put into neutral, flexed, extended, and
rotated positions. They found that the gap between the posterior disc and the
anterior margin of the facet joint increased with flexion and decreased with
extension or rotation. This narrowing of the spinal canal during extension and
rotation, the researchers argued, could potentially explain the
posture-dependent symptom of spinal stenosis. However, the authors of this
study acknowledge the limitations of measuring the behavior of the spine in a
cramped MRI machine, including a restriction of range of motion and the
inability to axially load the spine. Still, changes in the spinal canal were
observed, and give some credence to the link between certain spinal ailments
and certain postures.
Another
study that looked at potential risk factors for postural aberrations was performed
by Burnett et al on female rowers. A
non-invasive electromagnetic device was attatched at L3 and S2 to measure axial
rotation range of motion in different circumstances. For both standing and
sitting, subjects achieved the highest degree of rotation in a “neutral zone”
determined by the experience physiotherapists guiding the study to be the
midpoint between end-range flexion and extension. When the subjects were guided
into end-range extension, axial rotation was significantly reduced; the
researchers theorized that this was due to the bony stiffness provided by the
facet joints of the vertebra. In end-range flexion, axial rotation was even
more significantly reduced, this time possibly due to passive stiffness from
the posterior fibers of the annulus and the posterior ligaments. The results of
the study suggest that in end-ranges of motion, more force may be absorbed by
the comparitively less resilient passive structures of the spine rather than
the active, muscular structures, and could potentially lead to injury.
While these
studies are useful in determining potential mechanisms for injury during
significant flexion or extension, they say little about “natural” standing
posture, as most people do not stand with their spines approaching end ranges
of either flexion or extension. In fact, “natural” standing postures can be
frustrating to study due to the immense variance between individuals. Such was
the goal of Claeys, Brumagne, Deklerck,
Vanderhaeghen, & Dankaerts (2016) when they attempted to look at
correlations between all spinal regions in 99 healthy first year physiotherapy
students. These researchers looked at lumbar curve, thoracic inclination, trunk
angle, pelvic tilt, lumbar angle, head angle, neck angle, and cervicothoracic
angle to any and all correlations between them. What they found with respect to
standing posture was that only trunk angle was significantly predictive of
other angles, and even then, the correlations were mostly medium to small. In
discussing their findings, the authors conclude that optimal posture probably
does not exist in a way that can be applied to any large population, and the
“optimal” posture for any individual will be heavily determined by genetic and
environmental factors unique to them.
Lastly, even
though perfect posture may be undefinable for the population as a whole, we can
at least identify certain patterns in standing postures that make defining
different postures possible. Researchers appear to have no set consensus on
what categories of sagittal plane posture should exist, and the studies
presented in this paper use a mixture of quantitative and categorical variables
in attempting to do so. One such category of standing posture is sway, which is
seen in O'Sullivan, et al. (2002) and in Smith,
O'Sullivan, & Straker (2008) and is defined, loosely, as an anterior
shifting of the pelvis resulting in a posterior trunk lean. In the former
study, muscular activity was measured in sway standing as compared to neutral
standing (neutral, here, was defined as having markers on the acromion, greater
trochanter, and lateral malleolus stacked in a straight line on top of one
another). Sway posture was of interest to the study because it was deemed a
“passive” posture that one might adopt to save energy by relying on passive
structures of the body rather than active ones, akin to a slumped sitting
posture (which was also measured in this study). The researchers found that
muscle activity changed dramatically between neutral and sway postures: the
lumbar multifidus, internal oblique, and thoracic erector spinae all showed a
significantly decreased activations when
the subject shifted into the sway posture, replaced by a significant increase
in rectus abdominus activity. When discussing the clinical relavence of these
findings, the researchers speculate that individuals who habitually adopt
passive postures deactivate and potentially decondition the stabilizing muscles
of the lumbopelvic region, which may leave it vulnerable to strain,
instability, or injury.
To summarize
the theory behind looking at the relationship between posture and low back
pain, we see in the research that as the spine moves further away from
“neutral” postures (individualistic though they may be), it relies more on
passive structures to absorb force rather than the more resilient active
structures. These deviations may also decrease the space in the spinal canal,
which can potentially increase the risk of certain spinal ailments such as
stenosis. However, most individuals do not approach end-ranges of motion in
their normal standing posture, and as such the happenings-on at these ranges
may be of limited use.
Correlations in Observational Studies
Though
flawed, there is enough evidence linking posture to potential mechanisms of low
back pain to have warranted many studies on the subject. The most common way to
study most correlations is through observational studies, which attempt to find
commonalties between or across groups at a set point in time, without
establishing any cause an effect relationship. Though not experiments, they can
be useful in identifying potential links between two variables.
An author of
one of our previous studies about sway postures, O’Sullivan, who also took part
in a similar study linking that posture and others to low back pain in
adolescents. Specifically, the authors divided subjects into sway, flat,
hyperlordotic, and neutral categories. What they found was interesting: subjects
with non-neutral postures demonstrated higher odds for all measures of back
pain, but no one posture was consistently associated with pain in isolation (Smith, O'Sullivan, & Straker, 2008) . For instance, males
with sway posture were more likely to have back pain lasting three months or
more, but females with this same posture were no more likely to experience pain
than those with neutral posture. Likewise, females with a flat posture were
more likely to have pain aggravated by sport, but males were not.
Indeed, the
issue seems to be more complicated than a one-to-one relationship between
static standing posture and low back pain. A different study on adolescents
showed no correlations between postural
displacements and back pain using the PosturePrint internet-based computer
system (Ozyurek, Genc,
Karaali, & Algun, 2017) . One study finds an association with a
loss of lumbar lordosis (Jackson & McManus,
1994) ,
and another study finds an association with an increase in lumbar lordosis (Christie, Kumar, & Warren, 1995) .
One study,
refreshingly titled “Low Back Pain and Posture” (Tuzun, Yorulmaz, Cindas, & Vatan, 1999) grouped subjects
into acute pain, chronic pain, and a pain-free control groups of 50 each, then
measured their degrees of thoracic kyphosis, lumbar lordosis, and sacral
inclination in a relaxed standing posture with no shoes. Some correlations were
found between age and thoracic kyphosis, and between BMI and lumbar lordosis.
However, when comparing these postural angles between the subject groups, the
researchers detected no statistically significant differences. In their conclusion,
the researchers state that it is not reasonable to make a generalization
regarding these angles and low back pain.
This is not
to say, however, that the two are unrelated. Three possible explanations were
offered by Christie, Kumar, and Warren, who were mentioned above. They found
that those in acute pain were more likely to have greater thoracic kyphosis and
forward head positions, and that chronic back pain sufferers showed
significantly greater lumbar lordosis. The postural parameters they studied only
had moderate value in the prediction of pain, and they would not state whether
posture lead to pain or precipitation of pain causes postural aberrations.
Instead, they offer three possible explanations for the correlations they
found:
1.With the onset of pain, all aspects
of the spinal curve initially respond to the pain with increased forward head
posture to decrease lumbar pain
2. Individuals with acute low back
pain may have had a preexisting FHP that resulted in flexion of the T spine and
signs and symptoms in the lumbar spine
3. Chronic LBP sufferers have adapted
to the pain with a localization of the postural changes the lumbar spine that
balances the upper spine back towards normal.
The lack of
consensus in these studies are probably due to the fact that methodologies for
measuring posture, pain, or both varied significantly between one study and the
next. For example, our two studies on adolescents came to different conclusions:
one took a relatively simple measurement of posture, administered a simple
yes/no questionnaire on lifetime back pain, and got no correlation (Ozyurek, Genc, Karaali, & Algun, 2017) ; the other took a
more detailed questionnaire inquiring into different types of pain and correlated
that with specific postures, separated further by male and female, and got the
more nuanced results discussed earlier (Smith, O'Sullivan, & Straker, 2008) . Perhaps both “posture”
and “low back pain” are simply not specific enough terms on their own to ensure
consistent definitions and methodologies across a broad range of studies. As
Christie et al put it, “Ideal posture has not been universally agreed to and
several different definitions have been advanced.” However, classifications of ‘neutral’,
‘hyperlordotic’, ‘flat’, and ‘sway’ have been used by several authors in
several studies: O'Sullivan, et al., 2002, Smith, O'Sullivan, & Straker, 2008, and Christie, Kumar, & Warren, 1995, so perhaps
there is still enough consistency to make useful comparisons.
Longitudinal Studies on Posture
and Low Back Pain
Longitudinal
experiments are widely considered to be the gold standard at establishing cause
and effect in a population. However, the time and monetary commitment they
require make them more prohibitive than observational studies. The following
two studies took a baseline measurement of a subject pool, then followed those
same subjects over a period of time to see the long-term impacts, if any, of a
multitude of factors, including posture.
Widhe, 2001 measured the thoracic kyphosis, lumbar
lordosis, spinal mobility, and incidences of back pain in a group of 90
children at ages 5 and 6, and then again at ages 15 and 16. He found that
posture did change significantly during the study period: both thoracic
kyphosis and lumbar lordosis increased by roughly 6°, and gender differences
that were not present at ages 5-6 presented themselves at age 15-16. He also
found that mobility significantly decreased during the 10 years between
measurements, particularly in the thoracic spine. Finally, he reported that
about one-third of the children at 15-16 stated that they had occasional low
back, which was similar to other reports using comparable methods. However,
these incidences of low back pain were not related to posture, spinal mobility,
or even physical activity in the sample population he took.
Widhe offers
some insights into his findings. He suggests that the sharp rise in the prevalence
of low back pain between pre-pubescence and puberty could reflect the
transition from “living in the present” to “bodily awareness”; the pain may not
be a pathological condition of the spine itself, but a new dimension of the
mind. This, he states, may explain why it has been hard to find any “physical
parameters” predicting future low back pain. He also contrasts LBP in
youngsters and adults, as the definition of back pain in the latter group is
often defined by its consequences, such as work incapacity, as opposed to
simply pain and discomfort.
For a
perspective on adults, Adams et al., 1999 followed 403 healthcare workers aged
18-40 for three years, with initial testing that included functional
assessments of anthropometry, muscle strength and endurance, mobility, and
posture, as well as psychological assessment via the MSPQ+ZUNG. The subjects
also answered a questionnaire every 6 months to check for low back pain. The
study found the following to be consistent predictors of low back pain: reduced
range of lumbar lateral bending, a long back, reduced lumbar lordosis,
increased psychological distress (measured by the MSPQ+ZUNG), and previous
nonserious low back pain. Only the latter three were consistent predictors of
“any” back pain.
However,
even in a longitudinal study, correlation still does not equal causation. The
results of the study state that these factors were much better predictors of
low back pain than chance alone, but even when combined (posture was only one
of many factors considered predictive) the researches only obtained an R² % value of 11.6. This means that less than
12% of the volunteers studied could attribute their back pain to any of the psychological
or physical risk factors considered in the study.
Under the section of their article titled “What Caused Most of the Back Pain?”,
the researchers wrote: “Despite these possibilities, it is becoming apparent that
there are no simple explanations for LBP, only a large number of interacting
causes that must be identified and pieced together like a jig-saw puzzle. As
far as relatively trivial LBP is concerned, the effort to pursue these causes
may simply not be worthwhile, because the failure of all the physical factors
to predict “any” LBP in the current study surely indicates that mechanical influences
in the reporting of such pain are negligible.”
Conclusion
Posture is a complicated topic that,
despite having a general consensus around “neutral” being better, researchers
have been yet to quantify in exact terms. Low back pain is an enormously
complex subject that not only has a plethora of physical factors to consider,
but psychological, social, and environmental factors as well. Without a
concrete understanding of the complexities of either, it is very difficult to
make accurate predictions about the relationship between low back pain and
posture to a large population, and harder still to apply these broad concepts
in a clinical setting where possibly the only firmly established truth about
either is that they are both highly individualized. It is my opinion that
static standing posture has not been proven to be a reliable predictor of low
back pain, and that dynamic postures and loading probably play a more
significant role.
References
Adams, M., Mannion, A., & Dolan, P. (1999).
Personal Risk Factors for First-Time Low Back Pain. Spine, 2497-2505.
Burnett, A., O'Sullivan, P., Ankarberg, L., Gooding,
M., Nelis, R., Offermann, F., & Persson, J. (2008). Lower lumbar spine
axial rotation is reduced in end-range sagittal postures when compared to a
neutral spine posture. Manual Therapy, 300-306.
Cadar, I., & Pop, L. (2015). Correlations
between cervical spine posture and low back pain. Human & Veterinary
Medicine, 178-181.
Christie, H., Kumar, S., & Warren, S. (1995).
Postural Aberrations in Low Back Pain. Archives of Physical Medicine and
Rehabilitation, 218-224.
Chung, S., Lee, C., Kim, S., Chung, M., & Ahn,
J. (2000). Effect of low back posure on the morphology of the spinal canal. Skeletal
Radiology, 217-223.
Claeys, K., Brumagne, S., Deklerck, J.,
Vanderhaeghen, J., & Dankaerts, W. (2016). Sagittal evaluation of usual
standing and sitting spinal posture. Journal of Bodywork and Movement
Therapies, 326-333.
Hoy, D., Bain, C., Williams, G., March, L., Brooks,
P. B., Woolf, A., . . . Buchbinder, R. (2012). A Systematic Review of the
Global Prevalence of Low Back Pain. Arthritis & Rheumatism,
2028-2037.
Jackson, R., & McManus, A. (1994). Radiographic
Analysis of Sagittal Plance Alignment and Balance in Standing Volunteers and
Patients with Low Back Pain Matched for Age, Sex, and Size. Spine,
1611-1618.
O'Sullivan, P., Grahamslaw, K., Kendell, M.,
Lapenskie, S., Moller, N., & Richards, K. (2002). The Effect of Different
Standing and Sitting Postures on Trunk Muscle Activity in a Pain-Free
Population. Spine, 1238-1244.
Ozyurek, S., Genc, A., Karaali, H., & Algun, Z.
(2017). Three-dimensional evaluation of pelvic posture in adolescents with
and without a history of low back pain. Turkish Journal of Medical
Sciences, 1885-1893.
Pillastrini, P., Mugnai, R., Bertozzi, L., Costi,
S., Curti, S., Guccione, A., . . . Violante, F. (2010). Effectiveness of an
ergonomic intervention on work-related posture and low back pain in video
display terminal operators: A 3 year cross-over trial. Applied Ergonomics,
136-443.
Smith, A., O'Sullivan, P., & Straker, L. (2008).
Classifcaton of Sagittal Thoraco-Lumbo-Pelvic Alignment of the Adolescent
Spine in Standing and Its Relationship to Low Back Pain. Spine,
2101-2107.
Tuzun, C., Yorulmaz, L., Cindas, A., & Vatan, S.
(1999). Low Back Pain and Posture. Clinical Rheumatology, 308-312.
Widhe, T. (2001). Spine: Posture, mobility and pain.
A longitudinal study from childhood to adolescence. European Spine Journal,
188-123.