SPINE Volume 26, Number 7, pp 724–730
©2001, Lippincott Williams & Wilkins, Inc.
Impaired Postural Control of the Lumbar Spine Is
Associated With Delayed Muscle Response Times in
Patients With Chronic Idiopathic Low Back Pain
Andrea Radebold, MD, Jacek Cholewicki, PhD, Gert K. Polzhofer, BA, and Hunter S. Greene, MD
Study Design. Balance performance in unstable sitting
and trunk muscle response to quick force release were
measured in 16 patients with chronic low back pain and
14 matched healthy control subjects.
Objectives. To determine whether patients with low
back pain will exhibit poorer postural control, which will
be associated with longer average muscle response
times.
Summary of Background Data. Larger postural sway
during standing and delayed trunk muscle response
times for patients with low back pain have been reported
in several independent studies.
Methods. Unstable sitting test was accomplished by
attaching different sized hemispheres to the bottom of a
seat. Subjects performed trials with eyes open and closed
while the displacements of the center of pressure were
measured with a force plate underneath the seat. Re-
sponse to a quick force release was recorded from 12
major trunk muscles with surface electromyography.
Subjects performed isometric trunk exertions in a semi-
seated position when the resisted force was suddenly
released with an electromagnet. Average muscle re-
sponse times and balance performance were correlated
using a linear regression analysis.
Results. Patients with low back pain demonstrated
poorer balance performance than healthy control volun-
teers, especially at the most difficult levels. Patients also
had delayed muscle response times to quick force re-
lease. Average muscle onset times together with age and
weight correlated significantly with balance performance
with closed eyes (R2 ⫽ 0.46), but not with eyes opened (R2
⫽ 0.18).
Conclusions. Patients with chronic low back pain dem-
onstrated poorer postural control of the lumbar spine and
longer trunk muscle response times than healthy control
volunteers. Correlation between these two phenomena
suggests a common underlying pathology in the lumbar
spine. [Key words: low back pain, postural control, pos-
turography, random walk, sudden loading] Spine 2001;
26:724 –730
The maintenance of whole-body balance is a complex
task, involving the interaction of three major sensory
input systems (visual, vestibular, and somatosensory)
and the precisely coordinated motor output at many
From the Biomechanics Research Laboratory, Department of Ortho-
paedics and Rehabilitation, Yale University School of Medicine, New
Haven, Connecticut.
Supported by the Biomedical Engineering Research Grant from the
Whitaker Foundation.
Acknowledgment date: January 17, 2000.
First revision date: May 4, 2000.
Second revision date: August 23, 2000.
Acceptance date: August 28, 2000.
Device status category: 1.
Conflict of interest category: 14.
724
joints.17 When the balance is impaired or lost tempo-
rarily, the recovery strategy must encompass the mainte-
nance of lumbar spine stability to avoid injuries to that
region.3,4,20 Low back injuries among industrial workers
often occur because of losses of balance like slips and
trips while handling loads.1,9,19,28
Several studies have identified impairment in postural
control in patients with low back pain (LBP) using pos-
turography.2,14,18,27 The increased body sway in patients
with LBP was hypothesized to stem from, among other
things, injury and/or damage to proprioceptive tissues in
the lumbar spine. The disparity in postural control was
more pronounced in patients with LBP in trials with
closed eyes.2,18 Thus, the lack of visual feedback seems to
overburden an impaired proprioception when perform-
ing postural control tasks.
Poor proprioception also was hypothesized to cause
delayed muscle response to sudden trunk loading in pa-
tients with LBP.11,12,16,22,30 This hypothesis is further
supported by observations that patients with LBP have
poorer lumbar spine positional sense.10,21,25 An overall
longer psychomotor speed in patients with LBP also was
documented by several studies.14,15,26,29
In general, three levels of motor control (spinal reflex,
brain stem balance, and cognitive programming) com-
bine to produce appropriate muscle response.13 The spi-
nal reflex pathway uses proprioceptive input from mus-
cle spindles and Golgi tendon organs. The brain stem
pathway coordinates vestibular and visual input using
proprioception from joint receptors. Cognitive program-
ming is based on repeated and stored central commands,
which lead to voluntary adjustments. If a deficiency in
proprioception is the main underlying cause of delayed
muscle response to sudden loading and poor balance
performance in individuals with LBP, then a correlation
between the measures of these two phenomena should
exist, especially in the absence of visual input.
The purpose of the present study was twofold. First,
postural control of the lumbar spine was compared be-
tween healthy subjects and patients with LBP by measur-
ing their balance performance in unstable sitting. The
sitting task was chosen to verify that postural control
impairment could still be identified when the lumbar
spine was studied in isolation from the posture control of
lower body joints. Second, a relation was sought be-
tween trunk muscle response latencies to sudden loading,
representing the spinal reflex pathway, and the balance
performance in the unstable sitting, representing the
brain stem pathway. It was hypothesized that 1) patients
 Lumbar Postural Control in Low Back Pain • Radebold et al
Table 1. Anthropometric Subject Data
Patients Control Subjects
Number 1F, 15M 1F, 13M
Age, yrs 38.8 (10.1) 38.1 (9.6)
Weight, kg 81.9 (15.3) 80.4 (17.5)
Height, m 1.76 (0.09) 1.77 (0.09)
T9–L4/L5 distance, m 0.21 (0.03) 0.20 (0.04)
Average age, weight, height, and torso length defined as the distance from T9
to L4/L5 (standard deviations) for patients with low back pain and matched
healthy control subjects.
F ⫽ female; M ⫽ male.
will perform poorer in the unstable sitting test, 2) aver-
age trunk muscle response times will correlate signifi-
cantly with balance performance in an unstable sitting
test, and 3) average muscle response times, along with
age and body weight, will show a stronger association
with balance performance with closed eyes than with
open eyes.
Methods
Sixteen patients with chronic idiopathic LBP (Table 1) and 14
matched healthy control subjects volunteered for this study and
signed the consent form approved by Yale University Human
Investigation Committee. Low back pain was defined as a per-
sisting or periodic pain lasting longer than 6 months. Patients
with LBP included in this study had no neurologic deficits,
structural deformities, genetic spinal disorders, or previous spi-
nal surgery. Their radiographs showed only normal, age-
related changes. Patients had experienced LBP for periods
ranging from 6 months to 35 years, with pain intensity that
varied from mild to severe with some pain-free intervals. On a
10-cm visual analog scale,24 patients expressed their overall
LBP as 2.7 (SD ⫽ 2.0). The consumption of analgesics, mostly
nonsteroidal anti-inflammatories, varied from daily use to
medication as needed. The Roland disability questionnaire23
showed, on average, low scores (5.1 out of 24, SD ⫽ 4.2)
reflecting the ability of all patients to continue working with
some sick leave taken for days with intolerable pain only. All
patients were screened by an orthopedic surgeon before the
testing to assure that inclusion criteria were met.
Matched healthy control subjects (Table 1) were recruited
through advertisement. Control subjects had no history of any
neuromuscular or postural disorder and had never experienced
back pain lasting longer than 3 consecutive days. None of the
tested subjects had vestibular or visual disorders. All subjects
were tested in two separate experiments described below.
Balance Performance in an Unstable Sitting Posture. This
test was designed to assess the brain stem postural control
pathway. A sitting posture was chosen to isolate postural con-
trol of the lumbar spine from the control of lower body joints.5
Subjects were placed on a seat equipped with a foot support to
prevent any lower body movement (Figure 1). The seat was
designed to allow for attachment of polyester hemispheres of
varying diameters to the bottom, while preserving constant seat
height. Four levels of the seat instability (0, 1, 2, and 3) were
achieved by using smaller diameters of the hemispheres: infinity
(flat surface), 50, 44, and 22 cm. The seat was placed on a force
plate (model 9286AA; Kistler, Germany) at the edge of a table.
725
Figure 1. Unstable sitting test. Decreasing the diameter of the
hemispheres on the bottom of the seat allowed for increasing the
seat instability level. The leg and foot support prevented postural
adjustments through joints of the lower extremities. Displacements
of the center of pressure (CoP) underneath the seat were mea-
sured with a force plate.
A safety railing surrounded the force plate, providing support
when balance was lost. Subjects were instructed to maintain
balance while sitting upright with arms crossed. Data collec-
tion was initiated only after subjects had reached a steady state
of balance control. Postural sway was evaluated using force
plate measurements of the center of pressure (CoP) displace-
ments. Center of pressure coordinates were recorded at 1600
Hz and low-pass filtered at 10 Hz using a forth order Butter-
worth digital filter to eliminate noise.
Subjects performed five 7-second trials with eyes opened
and closed at each seat instability level. A 30-second rest was
given between trials. One minute of practice was allowed be-
fore collecting data at levels 1–3 with eyes open and eyes
closed. Subjects were asked to hold on to the safety railing at all
times between the trials to prevent additional learning. The
sequence of trials was constant, beginning with eyes open, fol-
lowed by trials with eyes closed, at each instability level from
0 –3. Thus, any increase in sway because of the increased seat
instability level or lack of visual feedback would be observed
despite additional learning.
Center of pressure trajectories were quantified with both
summary statistics and random walk analysis. Summary statis-
tics included maximum (MAX) and root mean square (RMS)
displacements of the CoP in the anterior/posterior (subscript x)
and lateral (subscript y) directions and a total CoP path length/s
(PATH). Descriptive statistics were averaged across five trials
in each experimental condition.
Random walk analysis quantified the effective stochastic
activity of the CoP motion in the form of short-term and long-
term diffusion coefficients (DS, DL) and scaling exponents (HS,
HL) extracted from stabilograms.5–7 A stabilogram was created
by plotting averaged squared distances traveled by CoP against
the corresponding time intervals. All five trials were concate-
nated to produce one stabilogram in each experimental condi-
tion. The critical point (CP) identified a transition between two
distinct slopes on the stabilograms and divided them into short-
726 Spine • Volume 26 • Number 7 • 2001
Figure 2. Quick force release test. Subjects exerted isometric
trunk flexion, extension, and lateral bending to the left and right.
Resisted force was suddenly released with an electromagnet, and
the response of 12 major trunk muscles was measured with
surface electromyogram.
term and long-term regions. Slopes of the linear approxima-
tions to these regions formed the diffusion coefficients (D), and
the exponential approximations formed the scaling exponents
(H). Diffusion coefficients (D) reflect the level of stochastic ac-
tivity and/or energy of the CoP motion. Scaling exponents (H)
quantify the correlation between the step increments making
up a stabilogram series. For H ⫽ 0.5, the increments in CoP
displacement are statistically independent. For H ⬎ 0.5, if the
CoP was moving in a particular direction, it will be likely to
continue in this direction (persistence). For H ⬍ 0.5, if the CoP
was moving in a particular direction for time t0, it would be
likely to move in an opposite direction for t ⬎ t0 (antipersis-
tence).6,7 Collins at al6,7 interpreted the short-term and long-
term regions as reflective of open-loop and closed-loop motor
control strategies.
Differences in balance performance between patients with LBP
and control volunteers were tested with three-factor repeated
measures ANOVA and Tukey’s post hoc pairwise comparison
test (P ⬍ 0.05). Level 3 was excluded from ANOVA because only
13% of patients finished this level with closed eyes.
Muscle Response Times in a Quick-Release Test. This
test was designed to test the spinal reflex motor control path-
way. A previously established quick-release protocol was used
to study the muscle response times.22 Subjects were placed in a
semi-seated position in an apparatus (Figure 2) that prevented
motion in the lower extremities. Subjects exerted isometric
trunk flexion, extension, and lateral bending. The resisted force
was suddenly released while surface electromyography was re-
corded from 12 major trunk muscles. The muscle onset and the
offset times in response to the force release were identified ac-
cording to a previously established protocol.22 Latencies longer
than 300 milliseconds were not considered because they could
have represented voluntary activities. Differences in response
times between patients with LBP and control volunteers were
tested with two-factor, repeated measures ANOVA (P ⬍ 0.05).
One average onset and offset response time was calculated for
each subject and used for the regression test. Both genders were
included in the same statistical tests because earlier studies ex-
amining the latency of trunk muscles and the postural sway did
not show any gender-based differences.16,18,22,30
The best subset of variables that maximized adjusted R2 in a
linear regression analysis (Minitab Inc., State College, PA, re-
lease 12.23) was used to correlate the balance performance
with trunk muscle response times. The CoP PATH was selected
as a dependent variable, because it was found to be the most
reproducible.5 Averaged onset and offset muscle response times
were the dependent variables. Age, body weight, height, and
T9 –L4/L5 distance were included in the regression analysis
because they were shown to be correlated with CoP move-
ment.5 An additional four subjects (two patients and two con-
trol volunteers) who could not be anthropometrically matched
were included in the regression analysis for a total of 34
observations.
Results
All subjects completed both tests without experiencing
any pain or discomfort.
Balance Performance Test
All of the control subjects finished the most difficult seat
instability level3 with open eyes, and 71% finished it
with closed eyes. Conversely, 69% of patients with LBP
finished level 3 with open eyes, and only 13% finished it
with closed eyes, indicating their stronger dependence on
visual feedback. The two patients who finished level 3
with closed eyes were 22 and 31 years old, which was
younger than the average age of 39 years (Table 1).
All summary statistics of CoP motion (RMS, MAX,
PATH) were greater for patients with LBP (Figures 3, 4,
5) and increased significantly with increasing seat insta-
bility level and lack of visual feedback. Interactions be-
tween the subject’s status (patient/control) and seat in-
stability level and between seat instability levels and
visual feedback were also significant. Patients performed
significantly poorer than control volunteers in the ante-
rior/posterior (a/p) direction (subscript x) on the seat
instability level 1 and 2 (Figure 3A, 4A). In the lateral
direction (subscript y), significant differences were
present only for instability level 2 (Figure 3B, 4B). Bal-
ance performance degraded in both groups more rapidly
with eyes closed than with eyes open as seat instability
increased.
The random walk analysis of CoP stabilograms re-
vealed short-term and long-term regions separated by a
critical point (CP). The CP for the patients occurred later
than for control volunteers, but this was not statistically
significant (Table 2). The average distance traveled by
CoP in the CP time interval, however, was significantly
longer for patients than for control volunteers on the seat
instability level 2. Both short-term and long-term diffu-
sion coefficients (DS and DL) were larger for patients
(Table 2). The significant interaction between the status
of subjects and seat instability level indicated that differ-
 Lumbar Postural Control in Low Back Pain • Radebold et al 727

Figure 3. A, B, Root mean square (RMS) of the CoP movement
measured in the anterior/posterior (subscript x) and lateral (sub-
script y) directions. Increasing seat instability levels were
achieved with decreasing diameters of hemispheres underneath
the seat. * indicates statistically significant difference from healthy
control subjects (P ⬍ 0.05).
Figure 4. A, B, Maximum CoP deflection (MAX) measured in the
anterior/posterior (subscript x) and lateral (subscript y) directions.
Increasing seat instability levels were achieved with decreasing
diameters of hemispheres underneath the seat. * indicates statis-
tically significant difference from healthy control subjects (P ⬍
0.05).

ences in diffusion coefficients (DS and DL) between pa-

tients and control volunteers were greater at higher seat
instability levels. The significant interaction between seat
instability level and visual feedback condition indicated
their multiplicative effect on the DS and DL for both sub-
ject groups.
Short-term scaling parameters (HS) were smaller for
patients than for control volunteers (Table 2). For both
groups, however, HS was greater than 0.5, indicating
that CoP moved persistently away from the relative equi-
librium point. Neither the seat instability level, visual
feedback, nor the subject status affected the long-term
scaling parameters (HL). All long-term parameters were
less than 0.5, indicating that the CoP was likely to return
to its relative equilibrium point.
Quick Release Test
Trunk muscle response times for onsets and offsets were Figure 5. Total CoP path length traveled per second measured at
greater for patients with LBP than for control volunteers increasing seat instability levels. * indicates statistically signifi-
in all directions (Table 3), as previously observed.22 cant difference from healthy control subjects (P ⬍ 0.05).
728 Spine • Volume 26 • Number 7 • 2001

Table 2. Random Walk Analysis Parameters Comparing Balance Performance at Increasing Seat Instability Levels
Between Patients with LBP and Healthy Control Subjects
Level 0 Level 1 Level 2

Seat Instability Patients Controls Patients Controls Patients Controls

CP time, s 0.42 (0.16) 0.36 (0.12) 0.78 (0.37) 0.70 (0.36) 0.93 (0.62) 0.81 (0.34)
CP, mm2 1.3 (1.1) 1.7 (3.4) 66.9 (98.9) 34.5 (49.7) 172.5 (180.0)* 85.5 (78.7)
D short, mm2/s 1.5 (1.2) 2.4 (4.0) 40.3 (48.0) 24.2 (21.0) 99.8 (89.2)* 59.3 (54.1)
D long, mm2/s 0.3 (0.5) 0.2 (0.3) 14.9 (25.7) 5.8 (9.0) 33.7 (35.3)* 14.8 (15.5)
H short 0.56 (0.18) 0.59 (0.07) 0.78 (0.15) 0.82 (0.06) 0.77 (0.14)* 0.83 (0.06)
H long 0.20 (0.12) 0.20 (0.13) 0.23 (0.13) 0.23 (0.15) 0.24 (0.11) 0.22 (0.12)
CP ⫽ critical point; D ⫽ diffusion coefficient, H ⫽ scaling exponent.
Eyes open and closed trials were averaged.
* Statistically significant difference from control subjects (P ⬍ 0.05).

Linear regression was used to correlate balance per-
formance quantified with CoP path length (PATH) at the
seat instability level 2. At this level, balance performance
discriminated the best between patients and healthy con-
trol subjects. Age, body weight, and the average muscle
onset time were the only independent variables that were
significantly correlated (P ⬍ 0.05) with balance perfor-
mance in unstable sitting with eyes closed (adjusted
R2⫽0.46):
PATH 关mm/s兴 ⫽ ⫺ 20.6 ⫹ 0.361 * Age 关years兴
⫹ 0.192 * Weight关kg兴 ⫹ 0.2 * Onset Time 关ms兴
The only significantly correlated independent variable
with balance performance with eyes open, however, was
bodyweight. The above regression model (Equation1)
explained only 18% of variance in the PATH measured
during the tests with eyes open(R2⫽0.18).
Discussion
This study demonstrated that patients with chronic, id-
iopathic LBP have poorer postural control of the lumbar
spine than matched healthy control volunteers. These
differences became greater as the task difficulty level in-
creased. In the absence of visual feedback, poor postural
control performance correlated significantly with longer
trunk muscle response times to sudden force release. Al-
though similar findings have been reported in the past,
several important differences characterize the present
study.2,16,18,30 First, very strict inclusion criteria for pa-
tients with LBP were observed to exclude the possibility
of poor motor performance because of known factors
such as structural spine deformities, fractures, spinal ste-
nosis, disc herniation, or neurologic deficits. Second,
seated positions were chosen for the tests in this study to
(1) eliminate any adjustments and possible impairments in
joints of the lower extremities. All past studies examined
standing postures, and the conclusions drawn about the
postural control deficits in patients with LBP could not
be fully attributed to the lumbar region. Finally, an as-
sociation was found between delayed trunk muscle re-
sponse to sudden force release and poor postural control
of the lumbar spine in the absence of visual feedback that
suggests the existence of a common pathology underly-
ing both phenomena.
Several hypotheses have been presented in the litera-
ture explaining the impairment in postural control and
the delayed trunk muscle response to sudden loading
observed in individuals with LBP. A deficit in proprio-

Table 3. Average Trunk Muscle Response Times on Force Release

OFF [ms] Agonists ON [ms] Antagonists

Patients Controls Patients Controls

Extension: 63 (27)* 48 (16) 74 (15)* 69 (18)

Agonists ⫽ extensors
Antagonists ⫽ flexors
Flexion: 68 (40)* 53 (37) 80 (20)* 63 (9)
Agonists ⫽ flexors
Antagonists ⫽ extensors
Lateral bending: 57 (21) 53 (20) 80 (16)* 70 (13)
Agonists ⫽ ipsilateral muscles
Antagonists ⫽ contralateral muscles

Average Response Time: 63 (24) 51 (12) 78 (11)* 67 (9)

Average (standard deviation) reaction time (ms) of all agonistic muscles that shut off and all antagonistic muscles that switched on in response to the sudden force
release.
* Significant difference from healthy control subjects (P ⬍ 0.05).
 Lumbar Postural Control in Low Back Pain • Radebold et al
ception in the lumbar spine appears a likely scenario.
Although proprioception was not measured in this study
and a definite conclusion cannot be drawn, a propriocep-
tive deficit hypothesis fits well with the results of this and
earlier studies. Direct measurements of trunk position
and movement sense indicated that patients with LBP
have poorer proprioception than healthy control sub-
jects.10,21,25 Further, the presence of bilateral asymme-
tries that correlated with injury in the study of Parkhurst
and Burnett21 suggests that injury causes localized pro-
prioceptive deficits. Trunk muscles and ligaments are the
main dynamic stabilizers of the lumbar spine and may
contain damaged proprioceptors in patients with
LBP.12,20 –22,25
In the present study, patients with LBP performed
worse with their eyes closed than did healthy control
volunteers. At the most difficult seat instability level,
with closed eyes only 13% of patients (69% of controls)
completed the trials. Patients who completed the trials
were the youngest of their group at ages 22 and 31 years.
Further, only when the eyes were closed did the average
muscle response time show a significant association with
balance performance. Therefore, it appears that in the
absence of visual feedback, the remaining sensory input
systems were more challenged, which resulted in a more
pronounced deficiency in postural control. These find-
ings are also in agreement with previous studies.2,18
The results of the random walk analysis, interpreted
according to Collins and De Luca’s hypothesis6 and a
previous study,5 provide more insight into open-loop
and closed-loop motor control mechanisms. Open-loop
mechanisms typically work without feedback and are
quantified with the short-term parameters in the time
interval before the CP. Patients had larger short-term
diffusion coefficients than control volunteers, demon-
strating that their CoP moved more. Although both
groups showed a persistent movement in short-term in-
tervals (HS⬎0.5), patients had less persistence than con-
trol volunteers because their HS exponents were smaller.
Perhaps patients generated larger CoP movement in the
short-term intervals to overcome larger sensory thresh-
olds (i.e., dead zones)6 caused by low back injury. The
larger movement then could lead to an enhanced feed-
back and consequently to an improved postural control
of the spine.8 This hypothesis is supported by the time
interval results when the CP was reached, indicating the
switch from the open-loop to the closed-loop postural
control mechanism. The CP was only slightly longer for
patients than for control volunteers, but the CoP moved
significantly more within this time interval.
Closed-loop mechanisms involve feedback and were
quantified with the long-term parameters. Patients had
significantly greater long-term diffusion coefficients DL
than control volunteers, indicating greater CoP move-
ment. The long-term scaling parameters (HL) for patients
and control volunteers were similar, however, for all seat
instability levels as well as the trials lacking visual feed-
back. Both groups showed an anti-persistent movement
729
(HL ⬍ 0.5), indicating that the CoP returned to the point
of equilibrium. The long-term scaling parameters, how-
ever, were found to be the least reproducible in a previ-
ous study.5
Summary statistics of the CoP movement revealed
that patients had poorer balance control than healthy
control volunteers under all testing conditions. In addi-
tion, control subjects performed better in the anterior/
posterior (a/p) than in the lateral direction, which is con-
sistent with previous results.5 Patients with LBP, in
contrast, showed no difference between the anterior/
posterior and lateral directions (Figures 3A , 3B, 4A, 4B).
Perhaps postural adjustments are more easily executed in
the anterior/posterior direction, because all joints re-
sponsible for postural adjustments move either exclu-
sively (i.e., knee), or have a much greater range of motion
(i.e., ankles, hips, inteverterbral joints) in that direction.
Fine postural adjustments thus might be easier in the
anterior/posterior direction than in the lateral direction.
If, however, proprioception in patients with LBP is dis-
turbed, then those subtle differences may vanish. There-
fore, the results of the present study suggest that balance
performance in unstable sitting, measured in the anteri-
or/posterior direction, discriminates better between pa-
tients with LBP and healthy control volunteers. In con-
trast, Mientjes and Frank18 found greater differences in
postural sway between patients and control subjects in
the lateral direction. Their study, however, was per-
formed in a standing posture.
The present study addressed two of three different
motor control pathways: the spinal reflex and the brain
stem pathway.13 Both of these pathways are dependent
on proprioception, other sensory inputs, information
processing, and the appropriate motor output. A deficit
in either of these components will result in poor postural
control and delayed muscle response to sudden loading.
Luoto et al15 suggested that a long psychomotor reaction
time observed in individuals with LBP might stem from
deficits in central information processing. The reaction
times measured in that study represented mainly the de-
cision time (central information processing), however,
and they cannot be compared with the results of the
present study.
Other suggested mechanisms underlying impaired
motor control in patients with LBP were decreased nerve
conduction velocity and reflex inhibition.11,12 A reflex
inhibition because of pain was unlikely in the present
study because neither patients nor control subjects expe-
rienced pain during the testing.
This study demonstrated that postural control of the
lumbar spine could be assessed by measuring balance
performance in unstable sitting. The correlation between
average trunk muscle response time and balance perfor-
mance indicates that those delayed latencies in patients
with LBP also may contribute to impairment in postural
control, as was hypothesized in other studies.12,21,22,25 If
postural control is impaired, the dynamic stability of the
lumbar spine may be compromised, making a person
730 Spine • Volume 26 • Number 7 • 2001
vulnerable to sustaining a low back injury or aggravating
an existing back problem. The causal effect , however, 13.
has yet to be determined. The tests presented here could
be used to screen for deficits in postural control of the
14.
lumbar spine to help to detect or prevent low back prob-
lems or to rehabilitate patients with LBP.
15.
Key Points 16.
● Balance performance in unstable sitting and
trunk muscle response to quick force release were 17.
measured to study postural control of the lumbar
spine in 16 patients with chronic low back pain
(LBP) and 14 matched control subjects. 18.
● Patients with LBP performed poorer in the bal-
ance performance test and had delayed trunk mus- 19.
cle response times.
● Average trunk muscle onset time, together with 20.
age and body weight, were significantly correlated 21.
with balance performance with closed eyes, but not
with open eyes. 22.
● Correlation between impaired postural control
and delayed muscle response times for patients 23.
with LBP suggests a common underlying pathology
in the lumbar spine. 24.
25.
References
26.
1. Bigos SJ, Spengler DM, Martin NA, et al. Back injuries in industry: A retro-
spective study. II. Injury factors. Spine 1986;11:246 –51.
2. Byl NN, Sinnott PL. Variations in balance and body sway in middle-aged 27.
adults. Subjects with healthy backs compared with subjects with low-back
dysfunction. Spine 1991;16:325–30.
3. Cholewicki J, McGill SM. Mechanical stability of the in vivo lumbar spine: 28.
Implications for injury and chronic low back pain. Clin Biomech 1996;11:
1–15. 29.
4. Cholewicki J, Panjabi MM, Khachatryan A. Stabilizing function of trunk
flexor– extensor muscles around a neutral spine posture. Spine 1997;22:
2207–12. 30.
5. Cholewicki J, Polzhofer GK, Radebold A. Postural control of trunk during
unstable sitting. J Biomech 2000;33:1733–7.
6. Collins JJ, De Luca CJ. Open-loop and closed-loop control of posture: A
random-walk analysis of center-of-pressure trajectories. Exp Brain Res
1993;95:308 –18.
7. Collins JJ, De Luca CJ, Burrows A, et al. Age-related changes in open-loop
and closed-loop postural control mechanisms. Exp Brain Res 1995;104:
480 –92.
8. Cordo P, Inglis JT, Verschueren S, et al. Noise in human muscle spindles
[letter]. Nature 1996;383:769 –70.
9. Frymoyer JW, Pope MH, Clements JH, et al. Risk factors in low-back pain.
An epidemiological survey. J Bone Joint Surg 1983;65A:213– 8.
10. Gill KP, Callaghan MJ. The measurement of lumbar proprioception in indi-
viduals with and without low back pain. Spine 1998;23:371–7.
11. Hodges PW, Richardson CA. Inefficient muscular stabilization of the lumbar
spine associated with low back pain. A motor control evaluation of trans-
versus abdominis. Spine 1996;21:2640 –50.
12. Hodges PW, Richardson CA. Delayed postural contraction of transversus
abdominis in low back pain associated with movement of the lower limb.
J Spinal Disord 1998;11:46 –56.
Lephart SM, Pincivero DM, Giraldo JL, et al. The role of proprioception in
the management and rehabilitation of athletic injuries. Am J Sports Med
1997;25:130 –7.
Luoto S, Taimela S, Hurri H, et al. Psychomotor speed and postural control
in chronic low back pain patients A controlled follow-up study. Spine 1996;
21:2621–7.
Luoto S, Taimela S, Hurri H, et al. Mechanisms explaining the association
between low back trouble and deficits in information processing. A con-
trolled study with follow-up. Spine 1999;24:255– 61.
Magnusson ML, Aleksiev A, Wilder DG, et al. Unexpected load and asym-
metric posture as etiologic factors in low back pain. Eur Spine J 1996;5:23–
35.
Mergner T, Rosemeier T. Interaction of vestibular, somatosensory, and vi-
sual signals for postural control and motion perception under terrestrial and
microgravity conditions: A conceptual model. Brain Res Rev 1998;28:118 –
35.
Mientjes MI, Frank JS. Balance in chronic back pain patients compared to
healthy people under various conditions in upright standing. Clin Biomech
1999;14:710 –716.
Omino K, Hayashi Y. Preparation of dynamic posture and occurrence of low
back pain. Ergonomics 1992;35:693–707.
Panjabi MM. The stabilizing system of the spine. Part I. Function, dysfunc-
tion, adaptation, and enhancement. J Spinal Disord 1992;5:383–9.
Parkhurst TM, Burnett CN. Injury and proprioception in the lower back.
J Orthop Sports Phys Ther 1994;19:282–95.
Radebold A, Cholewicki J, Panjabi MM, et al. Muscle response pattern to
sudden trunk loading in healthy individuals and in patients with chronic low
back pain. Spine 2000;25:947–54.
Roland M, Morris R. A study of the natural history of back pain. Part I:
Development of a reliable and sensitive measure of disability in low back
pain. Spine 1983;8:141– 4.
Scott J, Huskisson EC. Graphic representation of pain. Pain 1976;2:175– 84.
Taimela S, Kankaanpaa M, Luoto S. The effect of lumbar fatigue on the
ability to sense a change in lumbar position. A controlled study. Spine 1999;
24:1322–7.
Taimela S, Osterman K, Alaranta H, et al. Long psychomotor reaction time
in patients with chronic low back pain: Preliminary report. Arch Phys Med
Rehabil 1993;74:1161– 4.
Takala E-P, Korhonen I, Viikari-Juntura E. Postural sway and stepping re-
sponse among working population: Reproducibility, long-term stability, and
associations with symptoms of the low back. Clin Biomech 1997;12:7– 8.
Troup JD, Martin JW, Lloyd DC. Back pain in industry. A prospective
survey. Spine 1981;6:61–9.
Venna S, Hurri H, Alaranta H. Correlation between neurological leg deficits
and reaction time of upper limbs among low back pain patients. Scand J
Rehabil Med 1994;26:87–90.
Wilder DG, Aleksiev AR, Magnusson ML, et al. Muscular response to sud-
den load. A tool to evaluate fatigue and rehabilitation. Spine 1996;21:2628 –
39.
Address reprint requests to
Jacek Cholewicki, PhD
Biomechanics Laboratory
Department of Orthopaedics and Rehabilitation
Yale University School of Medicine
P.O. Box 208071
New Haven, CT 06520-8071
E-mail: jacek.cholewicki@yale.edu