Content uploaded by Peter Michaelson
Author content
All content in this area was uploaded by Peter Michaelson on Apr 23, 2015
Content may be subject to copyright.
Correspondence: Peter Michaelson, Lule å University of Technology, Department of Health Science, SE 971 87 Lule å , Sweden. E-mail: Peter.michaelson@ltu.se
(Rece ived 18 October 2011 ; accept ed 8 March 2012 )
the same time, it is well known that LBP patients
often have multidimensional problems, where patho-
anatomical, neurophysiological and psychosocial fac-
tors interact (1). This contributes to the complexity
of the disorder and complicates both the clinical
examination and diagnosis. For a successful out-
come, it may be of great importance to classify LBP
patients into uniform groups and offer them tailored
treatment (6).
A pathophysiological explanation to how persis-
tent LBP can emerge is that acute LBP generates pain,
which then causes a modifi ed neuromuscular activa-
tion in the stabilizing muscles of the lower back (7,8).
This tends to overload the structures and aggravate
the injury (9). In some cases, the condition becomes
prolonged and causes hypotrophy in the strong type
2 muscle fi bers of the back extensors (10).
For persistent LBP, there is evidence of a positive
treatment effect through a variety of training regi-
mens, which concludes that physical activity can
ORIGINAL ARTICLE
Treating persistent low back pain with deadlift training – A single
subject experimental design with a 15-month follow-up
DAVID HOLMBERG
1 , HENRIK CRANTZ
2 & PETER MICHAELSON
3
1 Luleå University of Technology; Department of Health Science, Luleå Sweden, Cederkliniken Primary Health Care Centre,
Pite å , Sweden,
2 Luleå University of Technology; Department of Health Science, Luleå Sweden, Fristaden Primary Health Care
Centre, Eskilstuna, Sweden, and
3 Lule å University of Technology; Department of Health Science, Division of Health and
Rehabilitation, Lule å , Sweden
Abstract
Low back pain (LBP) is a common disorder in the western world. Persistent LBP can be caused by pathological changes in
the discs and disturbed neuromuscular activation, which can cause hypotrophy of the strong type 2 muscle fi bers of the exten-
sors. Deadlift (DL) is an exercise that may address all these pathological transformations. The aim of the study was to inves-
tigate the effect of DL training on patients with persistent LBP. A single subject experimental design with an AB-design and
multiple baselines was applied in this pilot study. Two patients with discogenic LBP and one with arthrogenic LBP were
treated with DL training over 8 – 10 weeks. A follow-up was performed at 15 months. The subjects with discogenic LBP showed
positive response to DL training regarding pain intensity and functional status, but the exercise did not affect their mental
health. The subject with arthrogenic LBP did not seem to have positive effect from DL training. As a result of this pilot study,
the authors hypothesize that DL training may be a successful treatment for subjects with LBP of discogenic origin.
Key words: Arthrogenic , discogenic , exercise , neuromuscular activation , training
Introduction
Low back pain (LBP) is a common problem in the
western world (1). van Tulder (2) defi nes LBP as
distinct pain, ache or discomfort, located between
the lower ribs and gluteal folds. The condition is
often categorized as acute , subacute or persistent LBP
(2). Although the categorization is based on time of
inconvenience, it is generally accepted, since the
cause of LBP seldom is known (1). After experien-
cing acute LBP, there is an increased risk of deve-
loping subacute or persistent symptoms, where the
prognosis of recovery is poor (3,4).
Regarding persistent LBP, Manchikanti et al. (5)
states that the pain originates from the discs in 40%
of the cases, from the facet joints in 30% of the cases
and from the sacroiliac joints in 30% of the cases.
However, pain sensations can also originate from
ligaments, fascia, muscles and nerve roots (1). The
absence of valid diagnostic criteria makes it diffi cult
to detect which structures generate the pain (1). At
Advances in Physiotherapy, 2012; 14: 61–70
ISSN 1403 -8196 print/ISSN 1651-1948 online © 2012 Infor ma Hea lthca re
DOI : 10. 3109 /140 38196 .2 012.6 74973
Adv Physiother Downloaded from informahealthcare.com by Lulea University Of Technology on 10/15/13
For personal use only.
62 D. Holmberg et al.
Participants
The inclusion criteria were patients between the age
of 18 and 55 diagnosed with LBP of discogenic or
arthrogenic origin, where the symptoms had lasted
for at least 12 weeks or had occurred at least twice
in the last 2 years. Patients with acute ritzhopathy,
disc protrusion (diagnosed less than a year ago),
chronic diseases (rheumatoid arthritis, Mb Bech-
terew etc.), neurological diseases or generalized pain
syndrome were excluded. Five subjects met the cri-
teria and were invited to participate in the study,
where one declined participation. Therefore four
subjects were included, three with discogenic and
one with arthrogenic LBP. One of the discogenic
subjects withdrew after only 1 week because of
new employment, whereas three subjects completed
the study.
Subject one was a male in his early thirties (height
192 cm, weight 99 kg). He was used to exercising
fi ve or six times per week. His work as a woodsman
included physical strain such as walking or skiing in
rugged terrain. The fi rst episode of LBP occurred
in his late adolescence. Over time, his condition
evolved to recurrent LBP and the symptoms reached
a point where physical strain resulted in severe pain
that affected his daily life. His previous treatment
consisted of conventional physical therapy and
trunk stabilizing exercises, both without success. He
was diagnosed with discogenic LBP.
Subject two was a male in his late twenties (height
194 cm, weight 94 kg). He was used to weight training
three or four times per week. He was working as an
auto mechanic. The fi rst episode of LBP occurred in
his late adolescence because of a heavy lift and since
then he suffered from recurrent LBP. Prior to the
study, he had never received any treatment for his
symptoms. He was diagnosed with arthrogenic LBP.
Subject three was a male in his early thirties (height
176 cm, weight 91.5 kg) who previously had been
physically active in sports. He was working within pri-
mary healthcare and had a long history of LBP. An
acute worsening of his condition occurred in the spring
of 2007 because of a heavy lift. After that, he was forced
to withdraw from all types of physical activity. His pre-
vious treatment consisted of McKenzie therapy and
trunk stabilizing exercises without any signifi cant results.
He was diagnosed with discogenic LBP.
Procedure
The recruitment of subjects was carried out from
January 7, 2008 to March 7, 2008 in three healthcare
centers in a municipality in northern Sweden. A total
of nine physiotherapists were involved in the initial
recruitment process. After a fi rst screening, each
decrease pain and increase functional status (6).
However, there is no consensus regarding which
exercise is most effective for a specifi c disorder (11).
This might be explained by the fact that in most
articles the intervention consists of several different
exercises provided to heterogeneous groups (6). This
is a major problem that makes it impossible to deter-
mine which part of the intervention that may have
decreased the symptoms (11). To improve the treat-
ment, it is essential to investigate the effect of a sin-
gle exercise on subjects with a specifi c disorder.
Deadlift (DL) is a free weight exercise performed
with a barbell that may address all pathological trans-
formations of persistent LBP. The exercise creates
high activation of the stabilizing muscles, increases
the muscular strength and improves the patient ’ s lift-
ing technique, which prevents aggravation of the
injury. When performing DL, angular movement
emerges in several joints and therefore stresses a
majority of the muscles in the body, and especially
the back extensors and trunk stabilizing muscles. The
patient has to maintain the spine in a fi xed extended
position throughout the movement, which means
that the stabilizing muscles are activated (12). DL
actually activates the stabilizing muscles to a greater
extent than commonly used exercises performed on
a Swiss ball (12). Thereby, DL training can correct
neuromuscular activation and restore trunk stabiliza-
tion. DL also generates compression forces to the
spine, which increase nutrition in the discs through
diffusion of fl uid in the alternating compression –
decompression phase (13). Compressive mechanical
stimuli are also crucial in regenerating cartilage (14),
and may possibly affect LBP of facet joint origin.
High-intensity DL training will cause hypertrophy in
the type 2 muscle fi bers of the extensors (15) and
also contribute to a general strengthening of the
trunk muscles. The aim of this study was to investi-
gate the effect of DL training on patients with per-
sistent LBP of discogenic or arthrogenic origin.
Method
This pilot study was carried out as a single subject
experimental design (SSED) with an AB-design
and multiple baselines (16). SSED was considered
appropriate since there were no scientifi c reports
about DL training as treatment for persistent LBP
when the study started. Furthermore, SSED is a
method considered suitable for exploring new sci-
entifi c fi elds (16). The study design was in accor-
dance with the Helsinki declaration and was
approved by the Ethics group at Department of
Health Science, Lule å University of Technology
(dnr 50 - 2007).
Adv Physiother Downloaded from informahealthcare.com by Lulea University Of Technology on 10/15/13
For personal use only.
Treating persistent low back pain with deadlift training 63
subject that met the criteria and was interested in
participating was thoroughly examined by another
physiotherapist, specializing in orthopedic manual
therapy and with more than 20 years of professional
experience. The purpose of this examination was to
identify subjects based on clinical reasoning and
validated test methods (17). A total of fi ve subjects
fulfi lled the inclusion criteria and received both ver-
bal and written information about the pilot study.
One declined participation. The others signed a con-
sent form and were included. One subject withdraw
early from the study because of new employment.
The other three completed the study.
Outcome data (pain intensity, functional status
and mental health) was collected once a week at the
healthcare center during both baseline and the inter-
vention. A multiple baseline design was obtained
since the baseline lasted 5 weeks for subject one, 7
weeks for subject two and 4 weeks for subject three.
The outcome questionnaires were mailed to the sub-
jects at the 15-month follow-up.
Intervention
A conventional DL technique with a barbell and
Olympic weight plates was used in this pilot study.
The subject was standing upright with his feet
shoulder-width apart. In this position, with the
lumbar spine in neutral to extended lordosis, the
trunk and lumbar muscles were contracted to
achieve maximum stability of the spine (Figure 1a).
The subject leaned forward, by fl exing the hips and
knees with the trunk maintained in a fi xed posi-
tion, and grasped the barbell (Figure 1b). The sub-
ject lifted the barbell in a controlled straight
vertical movement, with the hips and knees
extended at the same time (Figure 1c and d). After
the concentric phase, the subject lowered the bar-
bell and returned to an upright position before the
next repetition.
DL was used both as warm-up and intervention
to ensure that a potential treatment effect derived
from this specifi c exercise. To ensure a perfect lifting
technique, the intervention was lead by a physio-
therapist with extensive experience of powerlifting.
In addition, sessions 1, 8 and 16 were monitored by
another physiotherapist (DH) who is a qualifi ed
instructor for the Swedish Powerlifting Federation.
In the beginning of the intervention, the subjects
had to learn a correct lifting technique. This was
achieved by training with low intensity, where the
rest between the sets was 3 min. Since the learning
time varied, the intensity progression was individu-
ally adjusted. The intensity was two to fi ve repeti-
tions in fi ve sets, as close to maximum effort as
possible. The rest between the sets was approxi-
mately 2 min. Progression was ensured by increas-
ing the intensity or adding repetitions or a
combination of both. The intervention stretched
over a period of 8 weeks with two training sessions
per week. Since subject one was absent during
weeks 6 – 8 of the intervention (four sessions),
because of sickness and work, the intervention was
extended for 2 weeks.
Outcome measures
The selected outcome measures were pain intensity,
functional status and health-related quality of life as
recommended by the World Health Organization
(18). All outcome measures were evaluated once
a week.
Pain intensity . The average pain intensity was evalu-
ated using the 100-mm visual analogue scale (VAS).
The validity, reliability and responsiveness of the VAS
for pain rating are documented (19,20).
Functional status . Profi le Fitness Mapping (PFM) is
a self-rated questionnaire used to assess functional
status for subjects with LBP (21). It consists of two
parts where the fi rst part evaluates symptoms (as
stiffness, weakness, tiredness) and the second part
evaluates the self-estimated degree of functional
Figure 1. (a – d) Performing the deadlift exercise.
Adv Physiother Downloaded from informahealthcare.com by Lulea University Of Technology on 10/15/13
For personal use only.
64 D. Holmberg et al.
limitation (of walking, standing, bending, sitting).
The results are expressed as a percentage of the total
bodily function, where 100% represents no disability.
The questionnaire holds good validity and reliability
regarding subjects with persistent LBP (21).
Mental health . Health-related quality of life was
assessed using the acute version of Short Form-36
(SF-36) with the subscales psychic and mental health.
This study only used quality of life, i.e. “ mental com-
ponent score ” (MCS) that summarizes vitality, social
functioning, role emotional and mental health. The
value of MCS varies from 0 to 100, where 100 cor-
respond to best possible quality of life (22). The
Swedish version holds good validity and reliability
(23). Normally the SF-36 also consists of a physical
component score (PCS).
Data analysis
The outcome measures were visually analyzed
by using the baseline mean, the two standard devi-
ation band (SD-band) of the baseline mean, the
intervention mean and the follow-up value. The
visual analysis included trend, level (mean) and
variability between baseline and the intervention
(24). In order to detect statistical signifi cance, the
two SD-band method was used to confi rm visual
fi ndings (16). A statistically signifi cant difference
( p ⬍ 0.05) has been detected when two successive
points of intervention values fall outside of the two
SD-bands (16). It has been suggested that a decrease
of the VAS score by 30% represent a decreased pain
intensity that the individual perceives as clinically
relevant (25). The data analyses therefore included
comparing individual intervention measurements to
the baseline mean. No similar recommendations
exist for PFM and MCS. All data analysis was done
using Microsoft Excel 2003.
Results
Subject one, discogenic LBP
Subject one started the DL training at a load of
30 kg for fi ve repetitions. At the end of week 5 the
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
B1 B2 B3 B4 B5 T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 15m
Perr cent i mproved
Pain intensity
Subject one
Measurement
Baseline
mean
Baseline+2sd
Baseline - 2sd
Interventi on
mean
Percent (%)
improved
Figure 2. The level, variability and percentage of improvement compared with baseline mean of the pain intensity. B1–5 represents the
baseline, T1–10 the inter vention and 15m the follow-up 15 months after the intervention start.
Subject one
50.0
60.0
70.0
80.0
90.0
100.0
B1 B2 B3 B4 B5 T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 15m
Disability (%)
Measurement
Baseline mean
Baseline+2sd
Baseline-2sd
Intervention mean
Figure 3. The level and variability of functional status. B1 – 5 represents the baseline, T1 – 10 the intervention and 15m the follow-up
15 months after the intervention start.
Adv Physiother Downloaded from informahealthcare.com by Lulea University Of Technology on 10/15/13
For personal use only.
Treating persistent low back pain with deadlift training 65
load was 57.5 kg for fi ve repetitions and at the end
of week 10 it was 110 kg for two repetitions.
Pain intensity . The level of pain intensity was higher
during baseline than during the intervention, yet no
statistical signifi cance was achieved. Regarding variabil-
ity, the pain intensity fl uctuated during both baseline
and the fi rst weeks of intervention. From week 5 of the
intervention, the pain intensity stabilized at low values,
which was maintained at the follow-up, with an average
decrease of perceived pain intensity of just over 40%,
which was considered clinically relevant (Figure 2).
However, subject one suffered from a new disc injury
that had occurred a few weeks after the intervention
was fi nalized, and at the follow-up he reported that the
pain intensity was fl uctuating (Figure 2).
Functional status . The functional status increased
during intervention and the intervention mean almost
reached the two SD-band. A statistical signifi cance
was achieved since four points were outside the
two SD-band. The variability decreased after weeks
4 – 5 of the intervention and the functional status
stabilized at a higher level. At the follow-up, the func-
tional status had decreased, probably because of the
new LBP symptoms (Figure 3).
Mental health . The mean MCS rating was slightly higher
during the intervention than during baseline. The MCS
rating showed an increasing trend during baseline, and
early in the intervention a statistical signifi cance was
obtained. The MCS rating then decreased to the base-
line mean and was almost steady until the follow-up.
At the same time, the variability decreased (Figure 4).
Subject two, arthrogenic LBP
Subject two started the DL training at a load of 30
kg for fi ve repetitions. At the end of week 4 the load
was 140 kg for fi ve repetitions and at end of week 8
it was 165 kg for three repetitions.
Pain intensity . The pain intensity showed an increa sing
trend during baseline and a decreasing trend during
the intervention. The variability was especially high in
the transition between baseline and intervention. In
Subject one
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
B1 B2 B3 B4 B5 T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 15m
Mental compent score
Measurement
Baseline mean
Baseline+2sd
Baseline-2sd
Intervention mean
Figure 4. The level and variability of the mental component score (MCS). B1–5 represents the baseline, T1–10 the intervention and 15m
the follow-up 15 months after the intervention start.
Subject two
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
B1 B2 B3 B4 B5 B 6 B7 T1 T2 T3 T4 T5 T6 T7 T8 15m
Pain intensity
0
10
20
30
40
50
60
70
Measurement
Baseline mean
Baseline+2sd
Baseline-2sd
Intervention
mean
Percent (%)
improved
Percent improved
Figure 5 . The level, variability and percentage of improvement compared with baseline mean of the pain intensity. B1–7 represents the
baseline, T1–8 the inter vention and 15m the follow-up 15 months after the intervention start.
Adv Physiother Downloaded from informahealthcare.com by Lulea University Of Technology on 10/15/13
For personal use only.
66 D. Holmberg et al.
the latter part of the intervention, the pain intensity
seemed to stabilize at lower values, which was almost
maintained at the follow-up. No statistical signifi -
cance and no difference regarding changes of clinical
relevance (in%) was achieved (Figure 5).
Functional status . The functional status mean was slightly
higher during the intervention than during baseline, but
no statistical signifi cance was achieved. The variability
was slightly higher during baseline. The functional sta-
tus increased in the end of the intervention, but had
decreased slightly at the follow-up (Figure 6).
Mental health . The intervention mean was slightly
lower than the baseline mean and no statistical sig-
nifi cance was achieved. Regarding the MCS rating,
no specifi c trend was detected and the variability was
relatively low throughout the study (Figure 7).
Subject three, discogenic LBP
Subject three started the DL training at a load of
30 kg for fi ve repetitions. At the end of week 4 the
load was 50 kg for fi ve repetitions and at the end of
week 8 it was 75 kg for three repetitions.
Pain intensity . The pain intensity mean was signifi -
cantly lower during the invention than during base-
line. Statistical signifi cance was achieved since six out
of eight measurements were below the two SD-band.
Even though the pain intensity varied, there was a
decreasing trend during the baseline that continued
through the initial part of the intervention. The pain
intensity seemed to stabilize at a low level from week
5. The decrease was considered clinically relevant
since six out of eight measures showed a 30% decrease
compared with the baseline mean. The pain intensity
had increased at the follow-up (Figure 8).
Functional status . The mean functional status was
considerably higher during the intervention than
during baseline. Statistical signifi cance was achieved
since fi ve measurements were outside the two SD-
band. The variability decreased and the functional
status seemed to stabilize at a high level from week
50.0
60.0
70.0
80.0
90.0
100.0
B1 B2 B3 B4 B5 B6 B7 T1 T2 T3 T4 T5 T6 T7 T8 15m
Disability (%)
Subject two
Measurement
Baseline mean
Baseline + 2sd
Baseline - 2sd
Intervention mean
Figure 6. The level and variability of functional status. B1–7 represents the baseline, T1–8 the intervention and 15m the follow-up 15
months after the intervention start .
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
B1 B2 B3 B4 B5 B 6 B7 T1 T2 T3 T4 T5 T6 T7 T8 15m
Mental compoent score
Subject two
Measurement
Baseline mean
Baseline + 2sd
Baseline - 2sd
Intervention mean
Figure 7. The level and variability of the mental component score (MCS). B1–7 represents the baseline, T1–8 the inter vention and 15m
the follow-up 15 months after the intervention start.
Adv Physiother Downloaded from informahealthcare.com by Lulea University Of Technology on 10/15/13
For personal use only.
Treating persistent low back pain with deadlift training 67
5 of the intervention, and this was maintained at the
follow-up (Figure 9).
Mental health . The MCS mean was slightly lower during
the intervention than during baseline but no statistical
signifi cance was achieved. The variability increased in
the second half of the intervention (Figure 10).
Discussion
Prior to this pilot study, DL training had not been
evaluated as treatment for subjects with persistent
LBP. Therefore, SSED was considered suitable, since
it enables systematical measurements of a dependent
variable throughout the intervention (26). One inher-
ited limitation with SSED is controlling the depen-
dent variable for confounding factors throughout the
intervention (16). However, SSED results can provide
information on whether a randomized controlled trial
(RCT) should be performed (26). To strengthen the
internal validity, a multiple baseline was used (26).
In this pilot study, a methodological weakness was
that not all baseline variables were stable when the
intervention started, and that all subjects at some point
showed a positive or negative baseline trend. A baseline
trend can indicate spontaneous recovery or a worsening
of the condition, but since all subjects had suffered from
LBP for many years, it is unlikely that they should all
recover during the short baseline phase. It is more likely
that they contacted the primary healthcare because of
a worsening of the LBP, and with that poor starting
point the condition could only improve (27), which can
be considered a regression towards the mean. In retro-
spect, it might have been better to extend the baseline
measurements to ensure stability before initiating inter-
vention. On the other hand, achieving stability may be
diffi cult, since self-rated variables such as pain intensity
and functional status normally vary over time for
patients with LBP (28).
When recruiting, fi ve subjects met the inclusion
criteria, where one declined participation without giv-
ing any reason and one dropped out because of new
0
10
20
30
40
50
60
70
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
B1 B2 B3 B4 T1 T2 T3 T4 T5 T6 T7 T8 15m
Pain intensity
Subject three
Measurement
Baseline mean
Baseline+2sd
Baseline-2sd
Intervention
mean
Percent (%)
improved
Percent improved
Figure 8. The level, variability and percent of improvement compared with baseline mean of the pain intensity. B1–4 represents the baseline,
T1–8 the intervention and 15m the follow-up 15 months after the intervention start .
50.0
60.0
70.0
80.0
90.0
100.0
B1 B2 B3 B4 T1 T2 T3 T4 T5 T6 T7 T8 15m
Disability (%)
Subject three
Measurement
Baseline mean
Baseline + 2sd
Baseline - 2sd
Intervention mean
Figure 9. The level and variability of functional status. B1–4 represents the baseline, T1–8 the intervention and 15m the follow-up 15
months after the intervention start.
Adv Physiother Downloaded from informahealthcare.com by Lulea University Of Technology on 10/15/13
For personal use only.
68 D. Holmberg et al.
employment. The three remaining subjects can be
considered a homogeneous group of male patients in
their thirties, physically active to a normal degree and
therefore not entirely representative of a general pop-
ulation of patients with persistent LBP. Yet, it should
be noted that these patients were not headhunted by
advertisement or personal contacts, but simply seek-
ing help from a physiotherapist for their LBP.
For subject one, the variability regarding pain
intensity was quite large from the beginning of base-
line through week 4 of the intervention, and a stable
baseline was not achieved. From week 5, the pain
intensity seemed to stabilize at a lower level. The 40%
decrease compared with the baseline mean can be
regarded as clinically detectable and relevant for the
subject (25). At the same time subject one ’ s lifting
capacity increased considerably (from 30 to 110 kg)
while maintaining the decreased pain intensity. For
subject three, the trend regarding pain intensity indi-
cated a spontaneous recovery during baseline that
with some variability continued for the fi rst week of
intervention. However, there was a signifi cant
decrease in pain intensity from week 5 of the inter-
vention, and the values stabilized at a level consider-
ably lower than the baseline mean. The 35 – 65%
decrease can be considered clinically detectable and
relevant for the subject (25). As for subject one, this
occurred while the lifting capacity increased (from
30 to 75 kg). This pattern was not repeated for sub-
ject two, where the intervention did not seem to
affect the pain intensity. However, for the last weeks
of the intervention, there was a slight decreasing
trend that was maintained at the 15-month follow-up,
but this was not considered clinically relevant (25).
Even though the pain intensity did not decrease for
subject two, the lifting capacity increased dramati-
cally (from 30 to 165 kg). Regarding this dramatic
increase, one should bear in mind that subject two
weight-trained frequently prior to the LBP.
Concerning functional status, the results were
similar to pain intensity, namely that a stable baseline
was not achieved and that the variability continued
through the fi rst weeks of intervention. For subjects
one and three, the PFM scores seemed to stabilize
at higher values from week 5 of the intervention, and
almost all values reached statistical signifi cance. For
subject two, the functional status was almost con-
stant during the intervention. A small increase
occurred after week 6, but this was not considered
relevant.
Subjects one and three, diagnosed with disco-
genic LBP, showed the same pattern regarding pain
intensity and functional status. The authors believe
that this is an effect of DL training. The results for
subject two were not as positive, which might indi-
cate that DL training is an effective treatment for
LBP of discogenic origin but has no effect on
arthrogenic LBP. All subjects increased their lifting
capacity. This can probably not be attributed solely
to increased muscular strength, but might also be a
result of overcoming their fear of putting strain on
their lower backs. The decreasing trend regarding
pain intensity late in the intervention suggests that
a longer intervention would have been motivated,
and should be considered in future studies. How-
ever, when designing this pilot study, the selected
treatment period of 8 – 10 weeks was in accordance
with most studies using training as intervention
(29). Since none of the subjects continued with the
exercise after the intervention, it was not surprising
that the functional status had decreased for all sub-
jects at the follow-up (30). It was surprising that
none of the subjects chose to continue with the
exercise, in spite of the positive effect on pain inten-
sity and functional status. However, one can only
speculate about their personal reasons since this
pilot study was designed to study the effect of DL
training to end of intervention.
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
B1 B2 B3 B4 T1 T2 T3 T4 T5 T6 T7 T8 15m
Mental co mpo nent s core
Subject three
Measurement
Baseline mean
Baseline + 2sd
Baseline - 2sd
Intervention mean
Figure 10. The level and variability of the mental component score (MCS). B1–4 represents the baseline, T1–8 the intervention and 15m
the follow-up 15 months after the intervention start .
Adv Physiother Downloaded from informahealthcare.com by Lulea University Of Technology on 10/15/13
For personal use only.
Treating persistent low back pain with deadlift training 69
DL training had no signifi cant effect regarding
mental health, although there was a slight improve-
ment at the follow-up for subjects two and three. For
subject one, statistical signifi cance was achieved dur-
ing weeks 2 – 3 of the intervention, but since the sub-
sequent measurements declined towards the baseline
mean, the authors interpreted this as a type I error.
For subject one, the intervention mean was slightly
higher than the baseline mean, where most of the
improvement occurred during baseline and then
slowly decreased throughout the intervention. In
summary, the results on mental health correspond to
other studies where exercise is compared with no or
passive treatment (31,32). Since PFM is specifi c for
disability related to LBP and there is an overlap
between the questions in PFM and PCS, the authors
found it non-essential to use PCS. However, the
results might have been different if the PCS values
had been evaluated.
No adverse effects from the DL training were
detected, which is probably related to a strict control
of the subjects ’ lifting technique and an individually
adjusted progression during the intervention. Sub-
ject one experienced new LBP problems that started
a few weeks after the intervention, but these new
symptoms were not linked to the intervention accord-
ing to the subject himself. During the initial 3 – 4
weeks of the intervention, all subjects experienced an
increased pain intensity, which could be considered
an adverse effect. Although exercise evidently is an
effective treatment for decreasing pain intensity and
increasing functional status (29), it involves putting
strain on affected structures, which can temporarily
increase the pain intensity. It may therefore be impos-
sible to strengthen the stabilizing muscles of the
spine without risking an initial increase of the pain
intensity.
DL is an exercise that might target several dys-
functions connected to LBP. When performing DL
correctly, the exercise enhances a correct posture of
lower back curvature and requires control of the lum-
bar neutral zone, which has previously been found to
decrease pain intensity and increase functional status
(33). However, the transfer effect from the exercise
into everyday life has not been included in this pilot
study. Today there is evidence that physical exercise
is an effective intervention for LBP, but there is no
conclusive answer to the question of which form of
exercise is most effective (34). DL training activates
stabilizing muscles (12) and enhances core stabiliza-
tion and motor control (7,9), which might offer a
possible pathophysiological explanation of why DL
may be an effective treatment for LBP. However, this
cannot be concluded from this pilot study since no
measures of pattern of muscular activation were
performed. DL also repetitively generates both high
spinal axial load and compression load, which can be
benefi cial for the disc regeneration as diffusion might
increase (13). Still, this process might take time,
which could potentially explain why the positive
effect occurred around week 5 of the intervention for
the subjects with discogenic LBP, yet this has to be
verifi ed before any conclusions can be drawn.
Reduced physical strength has been observed among
LBP patients (35) and DL training appears to
increase the lifting capacity. This is probably an effect
of increased neuromuscular activation and hypertro-
phy of type 2 muscle fi bers, combined with a psy-
chological effect of trusting your back again.
Based on this pilot study, it appears that DL
training can be considered a treatment for men with
LBP of discogenic origin. The effect should not be
expected until 5 weeks after the intervention start.
However, the mechanism that explains why DL
might work is still unknown. A future RCT study is
recommended in order to explore whether the results
apply for a larger population.
Conclusion
With respect to the limitations of this pilot study, the
authors hypothesize that DL training may be used as
a treatment for LBP of discogenic origin, and that a
positive response may be achieved regarding pain
intensity and functional status while the muscular
strength increases. However, the exercise does not
seem to affect mental health or LBP of arthrogenic
origin.
Declaration of interest: The authors report no con-
fl icts of interest. The authors alone are responsible for
the content and writing of the paper.
References
O ’ Sullivan P. Diagnosis and classifi cation of chronic low back 1.
pain disorders: Maladaptive movement and motor control
impairments as underlying mechanism. Man Ther. 2005;10:
242 – 55.
van Tulder M, Becker A, Bekkering T, Breen A, del Real MT, 2.
Hutchinson A, et al. Chapter 3 European guidelines for the
management of acute nonspecifi c low back pain in primary
care. Eur Spine J. 2006;15 Suppl 2:169 – 91.
Croft PR, Macfarlane GJ, Papageorgiou AC, Thomas E, 3.
Silman AJ. Outcome of low back pain in general practice:
A prospective study. BMJ. 1998;316:1356 – 9.
Joines JD. Chronic low back pain: Progress in therapy. Curr 4.
Pain Headache Rep. 2006;10:421 – 5.
Manchikanti L, Singh V, Pampati V, Damron KS, Barnhill 5.
RC, Beyer C, et al. Evaluation of the relative contributions of
various structures in chronic low back pain. Pain Physician.
2001;4:308 – 16.
Hayden JA, van Tulder MW, Malmivaara A, Koes BW. 6.
Exercise therapy for treatment of non-specifi c low back
Adv Physiother Downloaded from informahealthcare.com by Lulea University Of Technology on 10/15/13
For personal use only.
70 D. Holmberg et al.
pain. Cochrane Database of Syst Rev. 2005;20:
CD000335.
Hodges PW, Richardson CA. Altered trunk muscle recruit-7.
ment in people with low back pain with upper limb move-
ment at different speeds. Arch Phys Med Rehabil. 1999;80:
1005 – 12.
Leinonen V, Kankaanp ä ä M, Luukkonen M, Hanninen O, 8.
Airaksinen O, Taimela S. Disc herniation-related back pain
impairs feed-forward control of paraspinal muscles. Spine.
2001;26:367 – 72.
Cholewicki J, McGill SM. Mechanical stability of the in vivo 9.
lumbar spine: Implications for injury and chronic low back
pain. Clin Biomech. 1996;11:1 – 15.
Demoulin C, Crielaard JM, Vanderthommen M. Spinal muscle 10.
evaluation in healthy individuals and low-back-pain patients: A
literature review. Joint Bone Spine. 2007;74:9 – 13.
van Middelkoop M, Rubinstein SM, Verhagen AP, Ostelo 11.
RW, Koes BW, van Tulder MW. Exercise therapy for chronic
nonspecifi c low-back pain. Best Pract Res Clin Rheumatol.
2010;24:193 – 204.
Hamlyn N, Behm DG, Young WB. Trunk muscle activation 12.
during dynamic weight-training exercises and isometric insta-
bility activities. J Strength Cond Res. 2007;21:1108 – 12.
Adams MA, Hutton WC. The effect of posture on diffusion 13.
into lumbar intervertebral discs. J Anat. 1986;147:121 – 134.
Jung Y, Park MS, Lee JW, Kim YH, Kim SH, Kim SH. Car-14.
tilage regeneration with highly-elastic three-dimensional
scaffolds prepared from biodegradable poly (l-lactide-co-
varepsilon-caprolactone). Biomaterials. 2008;29:4630 – 6.
Folland JP, Williams AG. The adaptations to strength train-15.
ing: Morphological and neurological contributions to
increased strength. Sports Med. 2007;37:145 – 68.
Domholdt E. Rehabilitation research: Principles and applica-16.
tions, 3rd ed. St. Louis, MO: Elsevier Saunders; 2005.
Cook C, Hegedus E. Orthopedic physical examination tests: 17.
An evidence-based approach. Englewood Cliffs, NJ: Prentice
Hall; 2007.
World Health Organisation. International Classifi cation of 18.
Functioning, Disability and Health. 2001 [cited May 15,
2011][about 1 p.]; Available from: http://www.who.int/
classifi cations/icf/en/
Beurskens AJ, de Vet HC, K ö ke AJ, Lindeman E, van der 19.
Heijden GJ, Regtop W, et al. A patient-specifi c approach for
measuring functional status in low back pain. J Manipulative
Physiol Ther. 1999;22:144 – 8.
Bolton JE. Accuracy of recall of usual pain intensity in back 20.
pain patients. Pain. 1999;83:533 – 9.
Bj ö rklund M, Hamberg J, Heiden M, Barnekow-Bergkvist 21.
M. The assessment of symptoms and functional limitations
in low back pain patients: Validity and reliability of a new
questionnaire. Eur Spine J. 2007;16:1799 – 811.
Sullivan M, Karlsson J, Ware JE. The Swedish SF-36 health 22.
Survey. Evaluation of data quality, scaling assumptions,
reliability and construct validity across general populations
in Sweden. Soc Sci Med. 1995;41:1349 – 58.
Sullivan M, Karlsson J. SF-36 Health questionnaire, Swedish 23.
manual and interpretation guide. 2nd ed. G ö teborg: Sahlg-
renska University Hospital; 2002 (English abstract).
Wolery M, Harris SR. Interpreting results of single-subject 24.
research designs. Phys Ther. 1982;62:445 – 52.
Ostelo RW, Deyo RA, Stratford P, Waddell G, Croft P, 25.
Von Korff M, et al. Interpreting change scores for pain and
functional status in low back pain: Towards international
consensus regarding minimal important change. Spine. 2008;
33:90 – 4.
Backman CL, Harris SR, Chisholm JA, Monette AD. 26.
Single-subject research in rehabilitation: A review of stud-
ies using AB, withdrawal, multiple baseline, and alternat-
ing treatments designs. Arch Phys Med Rehabil. 1997;78:
1145 – 53.
Kongsted A, Leboeuf-Yde C. The Nordic back pain sub-27.
population program: Course patterns established through
weekly follow-ups in patients treated for low back pain.
Chiropr Osteopat. 2010;15:18:2.
Wilkey A, Gregory M, Byfi eld D, McCarthy PW. A com-28.
parison between chiropractic management and pain clinic
management for chronic low-back pain in a national health
service outpatient clinic. J Altern Complement Med. 2008;14:
465 – 73.
van Tulder M, Goossens M, Waddell G, Nashemson A. 29.
Conservative treatment of low back pain (Kroniska l ä ndry-
ggsbesv ä r – konservativ behandling). Swedish Council on
Technology Assessment in Health Care (SBU). 2000;2:
65 – 114. In Swedish.
Waling K, J ä rvholm B, Sundelin G. Effects of training on 30.
female trapezius myalgia: An intervention study with a 3-year
follow-up period. Spine. 2002;27:789 – 96.
Cairns MC, Foster NE, Wright C. Randomized controlled 31.
trial of specifi c spinal stabilization exercises and conventional
physiotherapy for recurrent low back pain. Spine. 2006;31:
670 – 81.
Maul I, L ä ubli T, Oliveri M, Krueger H. Long-term effects 32.
of supervised physical training in secondary prevention of
low back pain. Eur Spine J. 2005;14:599 – 611.
Suni J, Rinne M, Natri A, Statistisian MP, Parkkari J, 33.
Alaranta H. Control of the lumbar neutral zone decreases
low back pain and improves self-evaluated work ability:
A 12-month randomized controlled study. Spine. 2006;31:
611 – 20.
Macedo LG, Maher CG, Latimer J, McAuley JH. Motor 34.
control exercise for persistent, nonspecifi c low back pain: A
systematic review. Phys Ther. 2009;89:9 – 25.
Hultman G, Nordin M, Saraste H, Ohlsen H. Body compo-35.
sition, endurance, strength, cross-sectional area, and density
of MM erector spinae in men with and without low back
pain. J Spinal Disord. 1993;6:114 – 23.
Adv Physiother Downloaded from informahealthcare.com by Lulea University Of Technology on 10/15/13
For personal use only.
