FLEXION MOBILIZATIONS WITH MOVEMENT TECHNIQUES:

THE IMMEDIATE EFFECTS ON RANGE OF MOVEMENT AND

PAIN IN SUBJECTS WITH LOW BACK PAIN
Kika Konstantinou, PhD, MSc,a Nadine Foster, DPhil, PGCE,b Alison Rushton, EdD, MSc,c
David Baxter, DPhil,d Christine Wright, C. Math,e and Alan Breen, PhD, DC f

ABSTRACT
Objective: This study investigates the immediate effects of flexion mobilizations with movement techniques (MWMs)

on spinal range of movement in individuals with low back pain and also their impact on pain. A preliminary attempt has
been made to describe the clinical profiles of subjects who were thought to benefit from MWMs.
Method: A small-scale explanatory study was conducted using a crossover design, placebo-controlled, with subjects and
assessors blinded. After assessment by physiotherapists, 26 subjects with low back pain with pain on lumbar flexion,
thought to be appropriate for treatment with MWMs, participated. Subjects received an MWM intervention and a placebo
intervention in a randomized order. Lumbar spinal flexion and extension and pain during flexion were recorded
immediately before and after each intervention, using double inclinometry and visual analogue scales.
Results: Mean spinal range of movement increased significantly with the MWM intervention, as compared with the
placebo (true flexion: MWMs 49.28 [SD 16.4], placebo 45.38 [SD 14.1], P = .005; total flexion: MWMs 76.78 [SD 22.4],
placebo 69.78 [SD 21.5], P = .005). Mean pain scores did not change.
Conclusions: The MWMs produced statistically significant, but small, immediate spinal mobility increases but no pain
reduction when compared with placebo. By introducing clinical judgment into the subject selection process for the trial, 19
(73%) of 26 subjects benefited from MWMs techniques in terms of range of movement and/or pain intensity, whereas
9 (35%) subjects showed such changes with the placebo intervention. (J Manipulative Physiol Ther 2007;30:178-185)
Key Indexing Terms: Physical Therapy; Range of Motion

t has been estimated that 60% to 80% of all adults

experience low back pain (LBP) at some point in their
lives, although not all seek medical care.1-3 A number of

Spinal Physiotherapy Specialist/Research Physiotherapist,

Haywood Hospital/ Primary Care Sciences Research Centre, Keele
University, UK.
b
Senior Lecturer in Therapies (Pain Management), Primary
Care Sciences Research Centre, Keele University, UK.
c
Lecturer in Physiotherapy, School of Health Sciences, University of Birmingham, UK.
d
Dean of Physiotherapy, Centre for Physiotherapy Research,
University of Otago, New Zealand.
e
Senior Lecturer in Research Methods, School of Health
Sciences, University of Birmingham, UK.
f
Director, Institute for Musculoskeletal Research and Clinical
Implementation, Anglo-European College of Chiropractic, Bournemouth, UK.
Submit requests for reprints to: Kika Konstantinou, PhD, MSc,
MCSP, mMACP, Primary Care Sciences Research Centre, Keele
University, Keele, Staffordshire, ST5 5BG, UK
(e-mail: k.konstantinou@cphc.keele.ac.uk).
Paper submitted October 13, 2006; in revised form November
11, 2006; accepted December 6, 2006.
0161-4754/$32.00
Copyright D 2007 by by National University of Health Sciences.
doi:10.1016/j.jmpt.2007.01.015

178

studies suggest that between 10% and 50% of patients with

LBP receive physiotherapy,4-6 with a more recent study
suggesting that physiotherapists treat approximately 9% of all
the subjects with LBP in Britain annually.7 Manual (or
manipulative) therapy is part of osteopathic and chiropractic interventions and is an area of specialization within
physiotherapy that is most commonly used in the management of LBP in Britain.8,9 Manual therapy is a broad term
that incorporates many different concepts and approaches.
However, there is limited research evidence for the efficacy
and effectiveness of specific techniques as used by physiotherapists for patients with LBP. In addition, it is still unclear
which patients with LBP benefit more from specific manual
therapy approaches.
Mulligan10,11 pioneered a relatively new concept in
manual therapy; these techniques are known as mobilizations
with movement (MWMs), or as SNAGs, an acronym for
bsustained natural apophyseal glides.Q These have been
extensively described elsewhere.11 In the case of lumbar
spine MWMs, the technique involves the application of an
accessory glide along the plane of the zygapophyseal (facet)
joint in a weight-bearing position during active movement.
Mulligan11,p.44 proposes that these spinal techniques improve
signs and symptoms by directly facilitating the restricted

Journal of Manipulative and Physiological Therapeutics

Volume 30, Number 3

mobility of the facet joints and simultaneously influencing

the mobility of the intervertebral joint. The clinical appropriateness and effectiveness of these techniques are based on
whether they can bring about immediate changes in pain
status and spinal mobility in a pain-free fashion.11,p.44,45
To date, there is a lack of research on MWMs for the
spine; and the available studies have been related to
peripheral joint MWMs for the ankle12,13 and especially
the elbow.14-17 Although this research provides a useful
basis, extrapolation of these results to the treatment of
subjects with LBP is not appropriate.
A recent survey in the UK specifically investigated the
use of MWM techniques in LBP management to describe
the practice of physiotherapists and explore the reported
effects of MWMs.18 The findings suggested that at least 1 in
3 therapists currently involved in LBP management uses
MWMs as part of his or her treatment approach and that
physiotherapists use their clinical decision making to select
subjects whom they feel might benefit from these techniques. Therapists reported that the most common effects seen
immediately after the use of MWMs were increases in
patients active spinal range of motion (ROM) (54.4% of
survey respondents, n = 253) and pain relief (27.5%, n =
128). This was also reflected in the outcomes used to
evaluate improvement, with 47.6% (n = 198) of all
therapists most commonly using active ROM and 37%
(n = 154) most commonly using pain relief as an outcome
measure. The choice of outcomes reported by this survey
suggested that therapists choose MWMs as a form of
treatment based on the assessment of immediate changes in
spinal ROM impairments and/or pain responses.
Based on the survey findings,18 an explanatory smallscale trial was designed in the UK National Health Service
(NHS) physiotherapy outpatient setting. The primary aim of
the study was to investigate the claimed immediate effects
of lumbar spine flexion MWMs on spinal ROM in a sample
of patients with LBP referred to physiotherapy and, secondly,
to investigate their impact on pain. In addition, a preliminary
attempt has been made to describe the group of patients
with LBP thought to be appropriate for MWM techniques,
in terms of clinical profile and response to treatment.
Flexion MWMs were investigated because they have been
reported to be the most often used in clinical practice.18

Konstantinou et al
Spinal Flexion MWMs in LBP

Fig 1. Summary of screening process and entry of patients in

the trial.
flexion MWMs on 4 dependent variables: true and total
lumbar spine flexion ROM, total lumbar spine extension
ROM, and pain response during spinal flexion.

Inclusion and Exclusion Criteria

Ethical approval for the study was obtained from the

Local Research Ethics Committees of Warwickshire and
Birmingham NHS Health Authorities, UK.

The study took place in 2 NHS physiotherapy outpatient

departments and included subjects referred for physiotherapy
by their general practitioner or consultant for LBP with or
without leg pain. The recruitment period lasted from April
2001 to March 2002. Subjects older than 18 years were
included in the study only if the clinical impression of the
treating clinician was that they had LBP appropriate for
physiotherapy treatment and that flexion MWMs were
appropriate and potentially beneficial. A relatively high
number of subjects (n = 96) were not included in the study
after assessment because they were not thought eligible for
flexion MWMs. The subjects selected had pain primarily
upon spinal flexion and ROM limitations in the same
direction. In addition, subjects had to understand written
and spoken English so they could provide informed consent.
Subjects were also excluded from participating if they had
known contraindications to manual therapy, had undergone
previous spinal surgery, or were pregnant. Figure 1 summarizes the progress of all subjects screened for entry to the trial.

Design

Sample Size

This small-scale, explanatory clinical study was a crossover (same subject), subject- and assessor-blind, placebocontrolled design investigating the immediate effects of

The required sample size was estimated using the

formula described by Howell19 (Fig 2 for details) for
matched-samples t tests, based on the change in lumbar

METHODS

179

180

Konstantinou et al
Spinal Flexion MWMs in LBP

Fig 2. Explanation of formula used for sample size calculations.

spine true flexion ROM. The true flexion ROM difference
of interest was set at 78, which is above the upper limit of
the 95% confidence interval (CI) of the standard error of
measurement found by our double inclinometry (DI)
reliability study conducted before the trial. In the absence
of information relating to what constitutes a true difference
or a clinically important difference in ROM, 78 was chosen
so that the anticipated difference in the flexion ROM
variable needed to be at least higher than the average
measurement error.20 Any such change can then be
considered with some degree of confidence as a real change
in subjects movement. The standard deviation (SD) was set
at 128, derived from the literature on normative ROM data
on asymptomatic and symptomatic populations.21-30 This
was later adjusted to 6.58 based on analysis of available data
from the ongoing study.
Twenty-six subjects (15 men, 11 women) participated in
the study. This sample size was chosen so that the study
would have 90% power at a 5% significance level to detect a
difference in true lumbar spinal flexion ROM of 78 or more.

Outcome Measures and Equipment

The main outcomes of interest were true and total lumbar
spine flexion ROM. Secondary outcomes of interest were
pain during lumbar spine flexion and total lumbar spine
extension ROM.
Active ROM for true and total lumbar spinal flexion and
total lumbar spine extension was measured with DI with the
subject standing. True lumbar flexion is the ROM measured
by placing one inclinometer on T12 and the other on S2 and
having the patient bend forward; the difference between
their readings (T12 S2) is thought to reflect the mobility
of the lumbar spine alone.21 Total lumbar flexion is the
reading from the inclinometer only on T12; this is thought to
represent lumbar spine mobility, pelvic mobility, and soft
tissue extensibility such as hamstrings, etc.21
The inclinometers used were spirit level clinical goniometers with 2 points of contact and a reported accuracy, by

Journal of Manipulative and Physiological Therapeutics

March/April 2007

the manufacturer, of F18 (MIE Medical Research Ltd,

Leeds, UK). An intratester reliability study conducted
before the clinical trial yielded intraclass correlation
coefficients of 0.81, 0.81, and 0.91 for true lumbar flexion,
total lumbar flexion, and total lumbar extension, respectively. Fifteen subjects and 5 testers participated in the
reliability study. The subjects were measured on 3 occasions, 5 minutes apart, by each of the 5 testers. The starting
position of the subject for the performance and measurement
of spinal movements was standardized in the same way for
both reliability study and clinical trial.
Pain intensity during lumbar spine flexion was measured with a visual analogue scale (VAS).31-33 The VAS
consisted of a 10-cm horizontal line with 2 word
descriptors at either limit of the scale: bno painQ at the
left-hand side and bmaximum painQ at the right-hand side.
The subject was asked to draw a line across the scale
representing the severity of pain felt during active lumbar
spine flexion in standing position. The distance of the
mark from the lower limit of the scale was measured with
a ruler to the nearest 0.1 cm.
All measures were taken initially, then after the first
intervention, and again after the second intervention. The
same assessor, blinded to the order of intervention, took all
ROM measures for each subject.

Descriptive Baseline and Treatment Information

To permit description of those subjects with LBP
thought to benefit from MWMs, the following were also
collected: subjects sex and age, their functional scores on
the self-completed Roland-Morris Disability Questionnaire
(RMDQ),34 duration of the current LBP episode (in months),
history of LBP, area of pain on body chart, easing and
aggravating factors, therapists clinical impression of the
subjects condition (diagnosis), subjects psychological
profile using the self-completed Distress and Risk Assessment Method (DRAM),35 spinal level(s) treated, number of
sets, and repetitions and type of MWM techniques used by
the treating physiotherapists.

Procedure
Consenting subjects were told that the study was
investigating 2 different treatments and that they would
receive both in the same treatment session. Subjects
completed the RMDQ, and the history of the subjects back
problem was obtained. The physical examination included
assessment of active lumbar spine ROM, passive manual
examination of the lumbar spine, and any other tests the
treating clinicians felt necessary to carry out. Based on the
history and physical examination, a clinical impression was
formed by the physiotherapist and a decision was made
about whether the subject might benefit from the use of
flexion MWMs. Subjects who were considered by therapists
to be eligible for the use of flexion MWMs were randomly

Journal of Manipulative and Physiological Therapeutics

Volume 30, Number 3

allocated to 1 of 2 groups by coin toss as they presented for

treatment. This was done by the treating clinician. The
assessor, blinded to intervention order, performed the baseline measurements of spinal ROM and asked the subject to
rate their pain during flexion. After the baseline measurements, the assessor left the room and the treating clinician
applied the intervention as dictated by the randomization.
Upon completion of the intervention, the assessor returned
and repeated the same measurement procedures of pain and
spinal ROM as for baseline. The assessor again left the room,
the second intervention was applied, and the assessor
returned and repeated the same measurements as above. At
the end of the session, the subject completed the DRAM.

Konstantinou et al
Spinal Flexion MWMs in LBP

Table 1. Summary of demographic and descriptive clinical

information
Characteristics
Variable

Mean

SD

Range

Age (y)
Duration of symptoms (mo)
Pain intensity of current episode
at its worst (VAS, 0-10 scale)
Pain intensity during flexion at
assessment (VAS, 0-10 scale)
RMDQa
Sex
Previous LBP history
Pain pattern (body chart)

38.3
26.8
7

11.7
47.9
2

18-61
0.1-240
3-10

5.2

1.9

1.7-9.8

Interventions
Six senior physiotherapists from the 2 physiotherapy
departments participated in the study. They had a mean of
9 years postgraduate experience, and all but 1 had been on
accredited MWM courses. In addition, before data collection, training sessions were held regarding the use and
application of MWMs and the standardization of procedures.
The flexion MWMs were applied either centrally or
unilaterally, at the discretion of the treating therapist, to the
symptomatic spinal level(s). The treatment consisted of
mobilizing between 1 and 3 levels, using 2 to 3 sets of 4 to
6 repetitions on each spinal level. Level of symptom
irritability guided the physiotherapist in the application of
the technique, reflecting normal clinical practice (eg,
subjects with nonirritable symptoms are thought to require
more repetitions or sets, whereas those with irritable, painful
symptoms require less, as described by Maitland36,p.107 and
Mulligan.11,p.46
The placebo intervention consisted of each subject lying
on the treatment couch in a comfortable position. This could
be either side lying, prone, crook lying, or supine. Subjects
were told that this position would enable the muscles around
the spine to relax and therefore reduce pain. The subject
remained in this position for the same amount of time
required to apply the flexion MWM technique: approximately 3 minutes. This allowed the effects of time and
repeated testing to be taken into account and provided some
assessment of the bnonspecificQ effects of treatment.37,38
The placebo intervention described above was chosen
because it was considered to be nonspecific.

Data Analysis
All data were analyzed using the Statistical Package for
the Social Sciences version 13 (SPSS Inc, Chicago, Ill).
Statistical analyses on the 2 primary outcomes (true and total
lumbar spine flexion) and the 2 secondary outcomes (pain
and total lumbar spine extension) were conducted using the
2-stage approach described by Everitt.39 The first stage
tested for a carryover effect by using an independent-samples
t test on the data collected at baseline, postintervention, and

Clinical impression/diagnosis

DRAM (2 missing values)

11.4
4.7
4-22
11 female, 15 male
12 yes, 14 no
20 LBP, 3 LBP
and leg pain, 3 leg
pain only
8 arthrogenic,
3 discogenic,
12 description of
signs (neural,
flexion stiffness),
3 others
8 normal, 11 at risk,
4 distressed/somatic,
1 distressed/depressive

VAS, visual analogue scale; RMDQ, Roland Morris disability questionnaire (scores 0-24, higher scores = higher disability); DRAM, distress and
risk assessment method; SD, standard deviation.
a
Scores, 0-24; higher scores = higher disability.

postplacebo. This use of baseline measurements provided a

more powerful test for a carryover effect by essentially
eliminating the variation between different subjects.39
Carryover or residual effect in crossover designs is the
possibility that part of the effect of the intervention
administered in the first period may remain during the
second period and therefore contaminate the effect of the
intervention administered in the second period.39,40 In this
study, the statistical method described by Everitt39 was used
to assess whether the effect of the first intervention
remained during the application of the second intervention.
Other possible effects are that there is a change over time
irrespective of treatment or the time of administration of the
intervention has an effect on outcome. These last 2
possibilities do not apply to this study.
As described by Everitt,39 if the first stage of analysis had
shown a significant carryover effect, it would have been
impossible to continue the study as a crossover design and it
would have been necessary to change to an independent
design using only 1 measurement from each subject.
However, based on the findings from this first stage (ie,
nonsignificant carryover effect, P = .202), the second stage
used a paired-samples t test on the data collected postMWM
intervention and postplacebo to test for an MWM effect.
Levels of significance were set at 5% for primary
outcomes41,42 (true and total lumbar flexion) and at 1%

181

182

Konstantinou et al
Spinal Flexion MWMs in LBP

Journal of Manipulative and Physiological Therapeutics

March/April 2007

Table 2. Summaries for values at baseline, postplacebo, and

post-MWMs, and P values from the tests on the effects of MWMs
compared with placebo

Table 3. Number (and percentage) of subjects with improvements

according to cutoff scores for ROM and pain (z78 in ROM, z2
points on VAS) based on raw data

Intervention

True
lumbar
flexion

Total
lumbar
flexion

Total
lumbar
extension

Pain

MWMs
Placebo

11 (38.5%)
6 (23%)

14 (53.8%)
5 (19.2%)

8 (30.8%)
3 (11.5%)

11 (42.3%)
6 (23%)

Outcome measure

Summary

True
flexiona
ROM (8)

Total
flexiona
ROM (8)

Baseline measurement
Mean (SD)
44.4 (14.7)
69.5 (19.0)
CI
38.5-50.3
61.8-77.1
Postplacebo intervention
Mean (SD)
45.3 (14.1)
69.7 (21.5)
CI
39.6-51.0
61.0-78.4
PostMWM intervention
Mean (SD)
49.2 (16.4)
76.7 (22.4)
CI
42.6-55.8
67.7-85.8
Difference (MWM placebo)
Mean (SD)
3.9 (6.5)
7.0 (9.3)
CI
1.3-6.5
3.3-10.8
P value
.005
.005
(2 tailed)
Mean difference = MWM intervention
a
95% CI.
b
99% CI.

Total
extensionb
ROM (8)

Pain
scoresb
(cm)

21.9 (10.2)
16.3-27.5

5.2 (1.9)
4.2-6.2

21.2 (11.1)
15.1-27.3

4.3 (2.2)
3.1-5.5

24.0 (11.0)
18.0-30.0

4.2 (2.5)
2.8-5.6

2.8 (6.8)
0.9-6.5
.045

-0.1 (0.3)
0.8-0.6
.800

placebo intervention.

for secondary outcomes41,42 (lumbar spine extension, pain).

Summaries of the data were provided through mean values,
SDs, and CIs for the mean (95% for the primary outcomes,
99% for the 2 secondary outcomes).
Cutoff scores were defined to categorize subjects into
those who seemed to have a real ROM change (ie, beyond
that which could be explained by measurement error) and
those who seemed to experience a clinically meaningful
reduction in pain intensity. In this study, a change N78 was
considered as representing a real change in the subjects
flexion-extension ROM. The cutoff score for significant
changes in pain intensity was based on Farrar,43 who
suggested a change of 2 points (cm) or more on the VAS to
be clinically important.

RESULTS
Twenty-six subjects participated in the study and
received both interventions. Baseline characteristics are
presented in Table 1. All data were normally distributed.
The first stage of analysis tested for carryover effect, and
this was found to be nonsignificant ( P N .20).
Table 2 presents the mean values, SDs, and appropriate
CIs for measurements at baseline, postplacebo, and post
MWM intervention. Differences in the mean values are
presented as an indication of MWM effects, and the
P values indicate significance or otherwise of these effects.
As can be seen, the mean difference between MWMs and

placebo intervention was 3.98 for true lumbar spine flexion,

78 for total lumbar spine flexion, and 2.88 for total lumbar
spine extension. For pain, the overall mean difference
between MWMs and placebo was 0.1 cm.
The second stage of analysis indicated statistically
significant mean differences in true ( P = .005) and total
lumbar flexion ( P = .005) but not in lumbar extension ( P =
.045) or pain intensity during lumbar flexion ( P = .800).
According to the cutoff scores defined, immediately after
the MWM application, 19 (73%) of the 26 subjects
presented with improvements in flexion-extension ROM
and/or pain scores (z78, z2 points on VAS); these subjects
were considered as having responded favorably to the
MWM intervention. With the placebo intervention, 9 of the
26 subjects (35%) showed immediate improvements in
ROM and/or pain outcomes. Table 3 presents the number
and percentages of subjects who improved for each
outcome measure.

,
Description of Subjects Profiles
The additional information collected on the clinical
profiles of subjects included in the study indicated that half
reported a first-time LBP episode and that most reported
pain confined to the low back with symptoms of more than
3 months duration. Most subjects presented with intermittent pain of moderate to high intensity when at its worst,
scored moderately on the functional disability scale, and
were classified as bnormalQ or bat riskQ for psychological
distress. Most of the subjects reported activities and/or
postures of flexion as aggravating factors and indicated rest
and cessation of the aggravating activity or posture as easing
factors (Table 4). During assessment, all subjects had pain
during lumbar flexion.

Nonrespondents to Treatment
Overall, 7 subjects did not change or were made worse
with the application of MWMs. Their data suggest that they
had longer symptom duration, had higher disability scores,
had pain that referred distally, were more often female, and
were reporting mostly a first-time LBP episode. Because of
the small number of subjects who did not benefit, statistical
comparison was not appropriate.

Journal of Manipulative and Physiological Therapeutics

Volume 30, Number 3

Konstantinou et al
Spinal Flexion MWMs in LBP

Table 4. Summary of characteristics of respondents and nonrespondents to treatment

Variable
Duration of
symptoms4 (mo)
RMDQa,4
Pain intensity of
current episode
at its worst
Pain intensity during
flexion at
assessment
Sex4
LBP history4
Aggravating factors
reported by patients4

Pain pattern4
(body chart)
Clinical impression/
diagnosis4

DRAM
(2 missing values)

Respondents
(n = 19)

Nonrespondents
(n = 7)

25.7 (mean)
(0.1 [3 d]-240)
10.6 (4-18)
7.0, SD 2.3 (3-10)

29.9 (2-60)
13.6 (8-22)
7.0, SD 0.8
(6-8)

5.1, SD 1.9
(1.7-9.8)

5.3, SD 1.8
(1.7-6.8)

6 female, 13 male
10 yes, 9 no
9 flexion,
5 extension,
3 flexion and
extension, 2 other
12 LBP, 7 leg pain

5 female, 2 male
2 yes, 5 no
2 flexion,
4 flexion and
extension, 1 other

5 arthrogenic,
1 discogenic,
11 description of
signs, 2 other
5 normal, 9 at risk,
3 distress/somatic

3 arthrogenic,
2 discogenic,
1 description
of signs, 1 other
3 normal, 2 at risk,
1 distress/somatic,
1 distress/
depressive

3 LBP, 4 leg pain

The asterisk denotes difference between characteristics; however, this

was not tested statistically.
RMDQ, Roland Morris disability questionnaire; DRAM, distress and risk
assessment method; LBP, low back pain; SD, standard deviation.
a
Scores, 0-24; higher scores = higher disability.

DISCUSSION
The study was designed to test the efficacy of flexion
MWMs in terms of those clinical effects attributed to these
techniques in clinical practice. The findings suggested that
flexion MWMs produced a statistically significant immediate improvement in ROM compared with the placebo
intervention for true and total lumbar spine flexion, but
not for total lumbar spine extension or pain scores.
There is little previous information to guide decision
making about what degree of change in lumbar spine ROM
may be considered clinically important, despite the fact that
in physiotherapy practice, ROM is regularly assessed and
considered to be an important goal.6,8,44,45 To define a
meaningful change for this trial, any score in lumbar ROM
above the average measurement error (78, as recorded in the
DI reliability study) was considered to represent a real
change in subject spinal mobility. Both true and total flexion
ROM measurements were well correlated at baseline as well
as after the MWMs intervention; this correlation is in
agreement with previous research.46
There are several potential reasons for the ROM increase
with the application of MWMs compared with the placebo.

The placebo intervention consisted of a comfortable non

weight-bearing position adopted by the subject. The subjects
were led to believe that this resting position was a form of
treatment that might reduce their pain on subsequent
movement. It may therefore be suggested that the increases
in ROM with the application of MWMs were due to the
repeated flexion movement involved in the MWMs,
whereas the placebo did not involve movement. However,
designing a placebo that closely resembles an active
intervention without producing physiological effects is
exceptionally difficult in manual therapy.47,48 The placebo
in this study was designed to account for time and
measurement effects while considered nonspecific. The case
may be that repeating a spinal movement, although painful,
may lead to ROM increases simply by stretching effects.
However, not all subjects demonstrated increases in ROM,
although all of them performed repeated spinal flexion as
part of the treatment and measuring procedures, with a small
number of subjects getting worse. In addition, the statistical
analysis did not indicate any significant carryover treatment
effect, which may suggest that repeated movement was not
necessarily the explanation for ROM increases. Other
possible explanations could be the conditioning and expectancy related to the credibility of the intervention and the
triggering of central inhibitory systems.37 Neurophysiological muscular effects may play a role in ROM changes; total
flexion ROM (which combines pelvic mobility and hamstrings extensibility) was greater than the true flexion ROM.
This suggestion seems to have some support from limited
experimental research to date.49
The only published study to date exploring the effect
of posteroanterior mobilizations on spinal ROM in a
population with LBP did not report any significant ROM
improvements but demonstrated reduction in pain scores.50
In the current study, despite the overall immediate,
positive differences in spinal flexion ROM, no inferences
can be made about the clinical relevance of the small
differences observed.
One reason why differences in pain did not reach
statistical significance may have been the sample size in
this study, which was calculated using estimated differences
in ROM. To detect a difference of 2 points on the VAS for
the study population (power 90%, significance 5%), we
estimate that a sample of 46 would have been required.
The preliminary investigation to describe the clinical
profiles of subjects who were thought to benefit from
MWMs suggested that the treating physiotherapists considered subjects appropriate for treatment with flexion MWMs
if they had mainly local LBP symptoms that were
predominately intermittent, aggravated by flexion activities
and/or postures; pain produced on active lumbar flexion;
and restriction of spinal flexion. These subjects seemed to
have moderate functional restrictions but not high psychological distress as measured with the DRAM. Most could be
considered as having chronic LBP because they reported

183

184

Konstantinou et al
Spinal Flexion MWMs in LBP

more than 3 months duration of symptoms. More than half

were experiencing their first episode of LBP.
A point that merits consideration is the physiotherapists
who participated in this clinical trial, as they were responsible
for subject selection and treatment. All 6 therapists had
experience in LBP management, and all but 1 had attended a
formal course on the MWM concept and were using these
techniques in their normal practices. In addition, educational
sessions were held before data collection regarding the use
and application of MWMs. Subject selection in the trial was
designed to mirror clinical practice, that is, based on
individual clinical judgment from the assessment findings.
This, in turn, is based on training, experience, and, to some
extent, treatment preferences.51,52 It is not unreasonable to
suppose that different clinicians may have selected different
types of subjects for the study or may have differently
influenced the studys results. However, in the absence of
robust classification systems or well-described clinical
subgroups within the population of nonspecific LBP, it is
difficult to standardize the selection process except in general
terms. Two different NHS sites participated in the trial, and 6
senior physiotherapists were responsible for patient selection
and treatment; it is believed that this increases the external
validity of the results to some degree.
Limitations of the study include the relatively small
number of subjects (in terms of examining the effects upon
pain and description of subjects profile) and the lack of
investigation of the credibility of the placebo intervention as
a form of treatment for pain relief. It is fully acknowledged
that the placebo used in this study may not have been very
credible to patients; however, it was thought that it would be
appropriate as a totally nonspecific intervention because it
was not expected to produce any physiological effects such
as increases in spinal mobility or pain relief. The main
criticism of crossover designs is that the effect of the first
intervention may have remained during the application of
the second intervention because the 2 treatments were
applied within the same session. The researchers addressed
this through the use of baseline measurements for each
intervention and by conducting the recommended statistical
tests for carryover effect.

CONCLUSION
This study investigated the efficacy of spinal flexion
MWM techniques in terms of ROM and pain scores in
subjects with LBP selected for treatment with these
techniques. There were statistically significant differences
in spinal ROM but not in pain. Differences in ROM however
were quite small, and they may not be clinically important.
The study lends some support to the anecdotal evidence that
MWM techniques may produce immediate change in spinal
movement for some subjects. It also provides preliminary
information on the clinical characteristics of the population
with LBP thought to benefit from these techniques.

Journal of Manipulative and Physiological Therapeutics

March/April 2007

Practical Applications
! Flexion MWMs seemed to be more effective in
increasing active lumbar spine flexion when
compared with a placebo intervention in subjects
with LBP.
! The ROM changes however were small, and it is
unclear whether they have any clinical significance.
! Mobilizations with movement techniques may
result in immediate improvements in ROM and
pain levels for some subjects with LBP.

ACKNOWLEDGMENT
This article was developed as part of a PhD program at
Coventry University. The authors would like to thank the
physiotherapy managers, the participating physiotherapists,
and the subjects who took part. The authors would also like
to acknowledge the State Scholarships Foundation (IKY),
Republic of Greece, and the Chartered Society of Physiotherapy Charitable Trust Fund for financial assistance.

REFERENCES
1. Andersson GBJ. Epidemiology of low back pain. Acta Orthop
Scand 1998;S281:28-31.
2. Waxman R, Tennant A, Helliwell P. A prospective followup study of low back pain in the community. Spine 2000;25:
2085-90.
3. COST B13 Working Group on Guidelines for the Management
of Acute Low Back Pain in Primary Care (2004) 8.
4. Croft P. Low back pain in the community and in hospitals. A
report to the Clinical Standards Advisory Group of the
Department of Health. Prepared by the Arthritis and Rheumatism Council, Epidemiology Research Unit, Manchester; 1994.
5. Faas A, van Eijk JTM, Chavannes AW, Gubbels JW.
A randomised trial of exercise therapy in patients with acute
low back pain: efficacy on sickness absence. Spine 1995;20:
941-7.
6. van der Valk RWA, Dekker J, van Baar ME. Physical therapy
for patients with back pain. Physiotherapy 1995;81:345-51.
7. Maniadakis N, Gray A. The economic burden of back pain in
the UK. Pain 2000;84:95-103.
8. Foster NE, Thompson KA, Baxter GD, Allen JA. Management
of nonspecific low back pain by physiotherapists in Britain and
Ireland. Spine 1999;24:1332-42.
9. Gracey JH, McDonough SM, Baxter GD. Physiotherapy
management of low back pain. A survey of current practice
in Northern Ireland. Spine 2002;27:406-11.
10. Mulligan BR. Manual therapy. bNags,Q bSnags,Q bMWMSQ etc.
3rd ed. New Zealand7 Plane View Services Ltd; 1995.
11. Mulligan BR. Manual therapy. bNags,Q bSnags,Q bMWMSQ etc.
4th ed. New Zealand7 Plane View Services Ltd; 1999.
12. OBrien T, Vicenzino B. A study of the effects of Mulligans
mobilisation with movement treatment of lateral ankle pain
using a case study design. Man Ther 1998;3:78-84.
13. Collins N, Teys P, Vicenzino B. The initial effects of a
Mulligans mobilisation with movement technique on dorsi-

Journal of Manipulative and Physiological Therapeutics

Volume 30, Number 3

14.
15.

16.
17.

18.

19.
20.

21.

22.

23.
24.
25.

26.
27.
28.

29.
30.

flexion and pain in subacute ankle sprains. Man Ther 2004;

9:77-82.
Vicenzino B, Wright A. Effects of a novel manipulative
physiotherapy technique on tennis elbow: a single case study.
Man Ther 1995;1:30-5.
Vicenzino B, Paungmali A, Buratowski S, Wright A. Specific
manipulative therapy treatment for chronic lateral epicondylalgia produces uniquely characteristic hypoalgesia. Man Ther
2001;6:205-12.
Abbott JH, Patla CE, Jensen RH. The initial effects of an elbow
mobilisation with movement technique on grip strength in
subjects with lateral epicondylalgia. Man Ther 2001;6:163-9.
Bisset L, Beller E, Jull G, Brooks P, Darnell R, Vicenzino B.
Mobilisation with movement and exercise, corticosteroid
injection, or wait and see for tennis elbow: randomised trial.
BMJ 2006;333:939.
Konstantinou K, Foster NE, Rushton A, Baxter GD. The
use and reported effects of mobilisation with movement
techniques in low back pain management: a cross-sectional
descriptive survey of physiotherapists in Britain. Man Ther
2002;7:206-14.
Howell DC. Statistical methods for psychology. 4th ed. USA7
Duxbury Press; 1997. p. 450-508 [ch 14].
Petersen CM, Johnson RD, Schuit D, Hayes KW. Intraobserver
and interobserver reliability of asymptomatic subjects thoracolumbar range of motion using the OSI CA 6000 spine
motion analyzer. J Orthop Sports Phys Ther 1994;20:207-12.
Keeley J, Mayer T, Cox R, Gatchel R, Smith J, Mooney V.
Quantification of lumbar function. Part 5: reliability of range of
motion measures in the sagittal plane and an in vivo torso
rotation measurement technique. Spine 1986;11:31-5.
Gomez T, Beach G, Cooke C, Hrudey W, Goyert P. Normative
database for trunk range of motion, strength, velocity, and
endurance with the Isostation B-200 Lumbar Dynamometer.
Spine 1991;16:15-21.
Russell P, Pearcy MJ, Unsworth A. Measurement of the range
and coupled movements observed in the lumbar spine. Br J
Rheumatol 1993;32:490-7.
Alaranta H, Hurri H, Heliovaara M, Soukka A, Harju R.
Flexibility of the spine: normative values of goniometric and
tape measurements. Scand J Rehabil Med 1994;26:147-54.
Dopf CA, Mandel SS, Geiger DF, Mayer PJ. Analysis of spine
motion variability using a computerised goniometer compared
to physical examination. A prospective clinical study. Spine
1994;19:586-95.
McGregor AH, McCarthy ID, Hughes SP. Motion characteristics of the lumbar spine in the normal population. Spine
1995;20:2421-8.
Vachalathiti R, Crosbie J, Smith R. Effects of age, gender and
speed on 3-dimensional lumbar spine kinematics. Aust J
Physiother 1995;41:245-53.
Esola MA, McClure PW, Fitzgerald GK, Siegler S. Analysis of
lumbar spine and hip motion during forward bending in
subjects with and without a history of low back pain. Spine
1996;21:71-8.
Hasten DL, Lea RD, Johnston FA. Lumbar range of motion in
male heavy laborers on the applied rehabilitation concepts
(ARCON) system. Spine 1996;21:2230-4.
McGregor AH, McCarthy ID, Dore CJ, Hughes SP. Quantitative assessment of the motion of the lumbar spine in the low
back pain population and the effect of different spinal
pathologies on this motion. Eur Spine J 1997;6:308-15.

Konstantinou et al
Spinal Flexion MWMs in LBP

31. Huskisson EC. Visual analogue scales. In: Melzack R,

editor. Pain measurement and assessment. NY7 Raven Press;
1983. p. 33-7.
32. Melzack R, Katz J. Pain measurement in persons in pain. In:
Wall PD, Melzack R, editors. The textbook of pain. 4th ed.
Hong Kong7 Harcourt Publishers Limited; 1999. p. 409-26.
33. Scrimshaw SV, Maher C. Responsiveness of visual analogue
and McGill pain scale measures. J Manipulative Physiol Ther
2001;24:501-4.
34. Stratford PW, Binkley J, Solomon P, Finch E, Gill C, Moreland
J. Defining the minimum level of detectable change for the
Roland-Morris questionnaire. Phys Ther 1996;76:359-65.
35. Main CJ, Wood PLR, Hollis S, Spanswick CC, Waddell G. The
distress and risk assessment method. A simple patient
classification to identify distress and evaluate the risk of poor
outcome. Spine 1992;17:42-52.
36. Maitland GD. Vertebral manipulation. 5th ed. UK7 Butterworth
Heinemann; 1986.
37. Wall PD. The placebo and the placebo response. In: Wall PD,
Melzack R, editors. The textbook of pain. 4th ed. Hong Kong7
Churchill Livingstone; 1999. p. 1419-30 [ch 62].
38. Mengshoel AM. Physiotherapy and the placebo effect. Phys
Ther Rev 2000;5:161-5.
39. Everitt BS. Statistical methods for medical investigations. 2nd
ed. NY7 Edward Arnold; 1994. p. 77-104 [ch 6].
40. Kenward MG, Jones B. The analysis of data from 2  2
cross-over trials with baseline measurements. Stat Med 1987;
6:911-26.
41. Prentice RL. On the role of analysis of secondary outcomes in
clinical trials. Control Clin Trials 1997;18:561-7.
42. Davis CE. Secondary end points can be validly analysed, even
if the primary end point does not provide clear statistical
significance. Control Clin Trials 1997;18:557-60.
43. Farrar JT. What is clinically meaningful: outcome measures in
pain clinical trials. Clin J Pain 2000;16:106-12.
44. Battie MC, Cherkin DC, Dunn R, Ciol MA, Wheeler KJ.
Managing low back pain: attitudes and treatment preferences
of physical therapists. Phys Ther 1994;74:219-26.
45. Jette AM, Smith K, Haley SM, Davis KD. Physical therapy
episodes of care for patients with low back pain. Phys Ther
1994;74:101-15.
46. Rainville J, Sobel JB, Hartigan C. Comparison of total
lumbosacral flexion and true lumbar flexion measured by a
dual inclinometer technique. Spine 1994;19:2698-701.
47. Hawk C, Azad A, Phongphua C, Long CR. Preliminary study
of the effects of a placebo chiropractic treatment with sham
adjustments. J Manipulative Physiol Ther 1999;22:436-43.
48. Pichar JG, Wheeler JD. Response of muscle proprioceptors to
spinal manipulative-like loads in the anesthetised cat.
J Manipulative Physiol Ther 2001;24:2-11.
49. Indahl A, Kaigle AM, Reikeras O, Holm SH. Interaction
between the porcine lumbar intervertebral disc, zygapophysial
joints, and paraspinal muscles. Spine 1997;22:2834-40.
50. Goodsell M, Lee M, Latimer J. Short-term effects of lumbar
posteroanterior mobilisation in individuals with low-back pain.
J Manipulative Physiol Ther 2000;23:332-42.
51. Turner PA, Whitfield TWA. Physiotherapists use of evidence
based practice: a cross-national survey. Physiother Res Int
1997;2:17-29.
52. Turner PA, Whitfield TWA. Physiotherapists reasons for
selection of treatment techniques: a cross-national survey.
Physiother Theory Pract 1999;15:235-46.

185