The Spine Journal 1 (2001) 95–101

Effects of aerobic exercise on low back pain patients in treatment

Arthur Daniel Sculco, MSa, Donald C. Paup, PhDa,*, Bo Fernhall, PhDa, Mario J. Sculco, MDb
a
The School of Public Health and Health Services, The George Washington University Medical Center, 817 23 rd St., NW, Washington, DC 20052, USA
b
330 Washington St, Norwich, CT 06360, USA
Received 31 August 2000; revised 22 November 2000; second revision 20 February 2001; accepted in revised form 5 March, 2001

Abstract Background context: Aerobic exercise (AE) has been prescribed to improve fitness and well-being
in apparently healthy individuals and cardiac, orthopedic, and other patient populations. AE has not
previously been studied as a sole treatment for low back pain patients (LBPP).
Purpose: This study evaluated the effects of low to moderate aerobic exercise as an adjunct treat-
ment for LBPP, 30 to 60 years of age, in a neurosurgical practice during a 2.5-year follow-up to an
initial 10-week exercise program. The purpose of this study was to determine the effects of short-
and long-term AE on LBPP. The initial 10-week phase compared AE and nonexercising controls
on mood states and pain/symptoms.
Study design/setting: A matched stratified design was used to input LBPP with similar previous
clinical treatments as well as age and sex into AE or control groups.
Patient sample: After screening 68 LBPP from a New England private neurosurgical practice, 40
patients met qualification criteria, and 35 volunteered for this AE research study. The LBPP in
this study were 30 to 60 years old and had the following medical diagnoses: herniated nucleus pul-
posus at one or more lumbar levels, degenerative discopathy, lumbosacral strain, and spinal canal
and/or foraminal stenosis.
Outcome measures: The measure of mood states was the Profile of Mood States, and the measure
of pain was the Brief Pain Inventory. The 2.5-year follow-up phase compared AE and nonexercise
patients on the following treatment variables: medical office visits for pain/symptoms, physical
therapy referrals, epidural steroid injections for pain/symptoms, prescription of pain medications,
and work status.
Methods: Thirty-five LBPP were matched stratified into an AE or nonexercise control group for
a 10-week exercise program. After the 10-week exercise program, all subjects were given the op-
portunity to cross over to the opposite group. Those patients choosing to exercise were advised to
follow a low to moderate aerobic exercise prescription (walking or cycling, 60% age-predicted
maximal heart rate, 4 days per week for 45 minutes per day). None of the original AE group
crossed over to the nonexercise group because of symptoms relating to their previous exercise
participation. All participants were contacted at 6-month intervals, and the number of medical of-
fice visits for pain/symptoms, physical therapy referrals, number of epidural steroid injections, and
number of prescriptions for pain was charted for 2.5 years. Work status was evaluated by compar-
ing the change in number of patients not working, working part time, working full time, or number
changing from full time to part time or not working from randomization to the end of follow-up.
Patients following the exercise prescription at least 50% of the time were compared with those ex-
ercising less than 50% of the time during the 2.5-year follow-up. Significance was determined at
the .05 level using Fisher’s exact test or the Kruskal-Wallis test.

I hereby certify that, to the best of my knowledge, no aspect of my cur-

rent personal or professional circumstances places me in the position of
having a conflict of interest with any interest of NASS relating to the
manuscript. I further hereby certify that, to the best of my knowledge, nei-
ther I (including any member of my immediate family) nor any individual
or entity with whom or with which I have a significant working knowledge
have (has) received something of value from a commercial party related
directly or indirectly to the subject of this manuscript.
* Corresponding author. Tel.: 1-202-994-7112; fax: 1-202-994-1420.
E-mail address: DPAVP@GWU.EDU

1529-9430/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved.
PII: S1529-9430(01)00 0 2 6 - 2
96 A.D. Sculco et al. / The Spine Journal 1 (2001) 95–101

Results: The initial 10-week AE phase of the study indicated that low to moderate AE signifi-
cantly improved mood profile (AE X9.58; control X19.11; p.01) but did not alter pain lev-
els. AE patients in the 2.5-year follow-up phase received significantly fewer pain medication pre-
scriptions (AE X2.76; control X13.35; p.02) and were given fewer physical therapy referrals
(AE X0.17; control X1.64; p.002). There was no significant difference in the number of med-
ical office visits for pain or epidural blocks administered to either group. Work status was im-
proved only in exercising patients (AE X0.24; control X0.35; p.04).
Conclusions: Low to moderate aerobic exercise appears to improve mood states and work status
and reduce the need for physical therapy referrals and pain medication prescriptions for LBPP in
the care of a neurosurgeon.
Keywords: Aerobic exercise; Mood states; Low back pain; Exercise therapy

Introduction self-esteem [13]. Martinsen [28] found patients with mild to

moderate depression to benefit from aerobic exercise and
Low back pain (LBP) with related symptoms is a com-
weight training. In young adults, running and aerobic dance
mon and pervasive problem in today’s society, affecting
were shown to be more effective in improving mood state
60% to 90% of all US adults during their lifetime [1,2]. The
than weight training as measured by the Profile of Mood
cost of medical care and lost work time have been estimated
States (POMS) [10]. In addition, normal 40-year-old male
to be in the billions of dollars annually [2–6].
and female adults showed improved self-concept and de-
Current nonsurgical and noninvasive treatment modali-
creased tension (POMS) resulting from a prescribed 8-week
ties for LBP include bed rest, medication (analgesic and/or
program of walking, jogging, or cycling [29].
anti-inflammatory), manipulation, patient education with
Because of the potential positive physical and psycho-
emphasis on biomechanics/ergonomics, physical therapy
logical benefits of AE and benefits from reduced medical
(PT), strength training, and aerobic exercise (AE). In previ-
services and return to work, the purpose of this study was to
ous research, AE has been used in combination with other
evaluate the effects of AE as a sole exercise intervention for
treatments [1,3,7–9] in the rehabilitation of LBP, but not as
patients with LBP. This included a 10-week exercise phase
a sole intervention. AE may contribute to the overall treat-
to determine the effects of a low to moderate program of AE
ment plan for patients with LBP because of the benefits for
on mood profile and pain severity in patients with LBP and
the general population, which include improvements in
a 30-month follow-up comparing use of medical services
mood state [10–13], strengthening the supporting low back
and changes in work status between AE and nonexercising
musculature [12,14], increasing nutrition for the disc [15],
patients with LBP.
prevention of further low back injuries [16,17], and preven-
tion of general disease states [18–20].
AE may be important in LBP rehabilitation and mainte-
nance of low back health, because it is a form of exercise Methods
that can minimize spinal loading while presenting low back
muscles with a sufficient load to lead to a training effect Study population
[21–24]. Low back tissues may need mild stressing (stimu-
Thirty-five patients (21 men and 14 women) with a mean
lation from walking or other forms of exercise) to enhance
age of 47.68 years (range, 30–60 years), sedentary but in
their health, but too much loading can be detrimental [22–
good general health, and free of any medical or psychologi-
24]. Some of the benefits of AE for patients with LBP may
cal contraindications for low- to moderate-intensity AE
include enhanced oxidative capacity of skeletal muscle and
were recruited for the study. Thirty-four of the above pa-
improved neuromotor control and coordination [25,26].
tients with LBP (one control patient died of an unrelated
Prevention of low back injury has been suggested by Cady
cause) participated in the 30-month follow-up program.
et al. [16,17], who found that firefighters who were less fit
had more episodes of low back injury and incurred more
cost for work-related injuries. Nutter [20] found that AE can Inclusion criteria
improve strength and endurance and prevent fatigue. This
effect of AE could prevent improper body mechanics result- The criteria for selection of subjects were men or women
ing from fatiguing muscles and thus reduce low back pain 25 to 65 years of age of any race or ethnic group, sedentary
and injury. AE offers potential psychological benefits to pa- (no current exercise routine) and in good general health,
tients with LBP by improving mood state and overall feel- free of any medical or psychological contraindications for
ing of well being [10,13,27–29]. low- to moderate-intensity aerobic exercise, and willing to
Exercise studies of 6- to –20-week durations are gener- adhere to the exercise intervention prescribed for the exper-
ally associated with decreased depression and an enhanced imental group.
 A.D. Sculco et al. / The Spine Journal 1 (2001) 95–101
Table 1
Classification of subjects by age, gender, and previous interventions
Age (yrs) Gender
Group Mean S.D. Male Female PLS EPI A EPI B NPI
AE 47.2 9.03 10 7 6 10 3 4
C 48.1 7.28 11 7 4 4 8 3 plete their pretest appointment and were dropped from the
AE  aerobic exercise; C  control; EPI A  epidural steriod blocks
within 6 months preceding the study; EPI B  epidural steroid blocks be-
fore the 6 months preceding the study; PLS  past lumbar surgery; NPI 
no previous interventions.
Exclusion criteria
Subjects were excluded from the study if they exhibited
cardiovascular contraindications to AE, had acute severe
low back symptoms (i.e., severe back pain and/or radiating
pain/symptoms down either or both legs), had low back
symptoms that prevented them from participating in AE ac-
tivities, or had undergone any surgical or PT interventions
for LBP during the past 6 months.
Recruitment
Patients were recruited from a neurosurgical practice in a
community serving a population of about 85,000. Potential
subjects were identified by review of neurosurgical appoint-
ment logs from the prior 2 years. Patients meeting the inclu-
sion criteria were identified by chart review and recruited to
participate in the study by office staff.
Subject diagnoses
Thirty-five of 40 subjects with subacute (at least 4
months after injury) or chronic (pain duration of 6 months
or more) LBP fulfilled all inclusion and exclusion criteria
and volunteered to participate. Their diagnoses included
herniated nucleus pulposus at one or more lumbar levels
with varying degrees of radicular compression based on
magnetic resonance imaging (MRI), computed tomographic
(CT) scan, and clinical evaluation; degenerative discopathy
with and without bulging at all lumbar levels; lumbosacral
strain without significant MRI, CT, or clinical focal abnor-
malities; and spinal canal and/or foraminal stenosis. Before
participation, each subject signed an informed consent indi-
cating their willingness to participate in the study. In addi-
tion, each patient obtained written permission from his/her
treating physician to participate in this study. The study was
approved by The George Washington University Medical
Center institutional review board.
Experimental and control matching technique
A matched stratified design was used to control for the
possible confounding effect of previous clinical treatments
(epidural steroid blocks within the 6 months preceding the
study, epidural steroid blocks undergone before the 6
months preceding the study, and past lumbar surgery) as
97
well as gender and decade of life. For each patient, a coin
was tossed and the assignment to either intervention or con-
trol group was made. When two patients in a row were as-
signed to one group, the next patient was assigned to the
other group. Five of the 40 recruited patients failed to com-
study, thus leaving 35 patients for the study. Table 1 shows
the AE (n17) and control (n18) patient demographic af-
ter matched stratification.
Compliance to study safety guidelines
During the initial assessment for the 10-week AE study,
all subjects were encouraged not to lift loads greater than 20
pounds, lift awkward loads (a load in which the weight was
not evenly distributed), or perform body maneuvers in
which the individual does not move the trunk and pelvis
along the same axis, for example, forceful wiggling or
twisting.
Assessment tools
Assessment tools included questionnaires on mood pro-
file and pain intensity along with chart review and tele-
phone interviews for outcomes regarding use of medical
services and work status. The questionnaire used to evaluate
the changes in mood profile hypothesized to occur as a re-
sult of the AE was the POMS short form. This form con-
tains 65 adjective rating items that are rated on a scale of 0
to 4 (0not at all, 1a little, 2moderately, 3quite a bit,
4extremely). The adjectives are used to assess the follow-
ing mood states: confusion/bewilderment, tension/anxiety,
depression/dejection, anger/hostility, vigor, and fatigue. In
addition, a composite total mood disturbance score was
used to evaluate mood state [30]. For each POMS test ad-
ministered, the patients were instructed to rate how they
felt, on average, during the past 24 hours. The Brief Pain In-
ventory (BPI) was used to evaluate pain as related to the av-
erage, least, and worst pain the patient had during the previ-
ous week. Pain was scored on a 0–10 scale in these three
areas [31].
Experimental protocol
Each subject completed a POMS and BPI baseline as-
sessment at time zero, a second assessment at the end of the
fifth week, and a third and final assessment at the end of the
tenth week. All subjects in the intervention (AE) group met
with an exercise physiologist and were instructed on how to
take their pulse (carotid and radial artery) and how best to
follow the exercise prescription. Baseline assessments were
completed under supervision, and the second and third as-
sessments were mailed to each subject for completion and
return to the study investigators. Completion of these as-
sessments was 100%.
Subjects in the intervention (AE) group were prescribed
a 10-week home-based aerobic training program consisting
of walking or cycling. Training was performed four times
98 A.D. Sculco et al. / The Spine Journal 1 (2001) 95–101

per week at an intensity of 60% of age-predicted maximal Statistics

heart rate: training heart rate.6(220age). The patients
The Statistical Package for Social Sciences was used for
exercised for 20 minutes during week 1, 30 minutes during
data analysis. Statistical significance was set at the .05 level
week 2, and for 45 minutes during week 3 and the remain-
(p.05). Descriptive statistics were calculated for the
der of the 10 weeks. This exercise intensity, duration, and
POMS and BPI tests at 0, 5, and 10 weeks of the study. To
frequency meets the American College of Sports Medicine
adjust for unequal baseline differences, a one-way analysis
guidelines for increasing aerobic fitness [32]. The subjects
of covariance was performed where the covariate was the
self-monitored radial or carotid pulse rate at various inter-
baseline value of a subject.
vals throughout the session. Date, time, duration, exercise
The outcomes obtained from chart review and patient
heart rate, and mode of training were recorded by the sub-
telephone interview were number of doctor’s office visits
ject at each exercise session and mailed weekly to the study
for pain/symptoms, number of physical therapy referrals,
investigator. Subjects in the control group were instructed to
number of prescriptions for pain medications, and number
continue their normal routine and not participate in any for-
of epidural steroid injections. These outcomes were com-
mal exercise program for the duration of the 10-week por-
pared across the two groups of patients who exercised more
tion of the study.
or less than 50% of the 2.5-year follow-up. Work status was
Subjects in both groups (AE and control) completed the
documented from telephone interviews and patient charts
weekly safety compliance log, which was mailed to the
and evaluated by the change in number of subjects not
study investigator each week in order to keep regular up-
working, working part time, working full time, or the num-
dates on the subjects’ compliance to the study safety guide-
ber of patients changing from full time to part time or not
lines, document any injury that a subject may have incurred
working. One point, positive or –negative, was assigned for
during ordinary living or AE activity, and give the controls
each change in work status with more work being positive
a sense of active participation in the study. Patients in the
and less work being negative. Significance was determined
intervention group mailed a weekly log of their aerobic ex-
at the .05 (p.05) level using Fisher’s exact test or the
ercise program, including length of exercise time, intensity
Kruskal-Wallis test.
of exercise (heart rate), and mode of exercise (walking or
cycling). Because the exercise intervention was home based
and unsupervised, both groups had equal personal contact Results
with the study investigator.
At the completion of the 10-week phase of the study, all As can be seen in Table 2, the patients with LBP in this
subjects (AE and control groups) were advised to exercise study adhered to the exercise prescription during the 10-
following the guidelines given to the intervention group. All week exercise program. The average prescribed target heart
subjects were contacted every 6 months by telephone to re- rate was 103.41 beats per minute, whereas the average
port their log of aerobic exercise participation. At the end of achieved heart rate by the AE group was 103.58 beats per
the 2.5-year follow-up, all subjects were divided into two minute for an average exercise frequency of 3.62 days per
groups: those who exercised according to the intervention week out of a 4 day per week prescribed maximum. The ex-
guidelines for the majority of the 30-month follow-up pe- ercise duration was 38.10 minutes per session, whereas 41
riod and those who did not. Exercisers were defined as those Table 3
who exercised at least 16 times per month for at least 16 of The effect of aerobic exercise on mood states as measured by the Profile of
the 30 follow-up months. Those exercising less than 16 of Mood states
the 30 follow-up months were classified as nonexercisers. Week 0 Week 5 Week 10
Semiannual telephone interviews and medical record re- Group Mean S.D. Mean S.D. Mean S.D.
views were employed to assess the following outcomes for
C depression 6.16 8.35 5.88 10.17 9.44 12.31
all subjects: office visits for pain/symptoms, medication E depression 3.64 4.06 2.351 4.12 3.641 5.74
prescriptions for pain/symptoms, use of epidural steroid in- C fatigue 7.72 6.51 8.33 6.31 8.05 7.33
jections, physical therapy referrals, and work status. These E fatigue 4.58 4.43 4.41 4.89 3.88 4.24
variables were compared between the exercise and nonexer- C anger 4.11 5.49 5.77 7.10 8.16 11.27
E anger 2.35 2.99 0.941 2.19 1.821 3.57
cise groups at the end of the 30-month follow-up period. C confusion 1.88 4.43 2.00 3.27 3.38 4.91
E confusion 1.52 5.02 0.88 2.71 0.58 2.59
C tension 4.16 6.80 4.50 6.64 4.27 7.02
Table 2 E tension 2.05 3.63 0.121 4.13 0.94 4.94
Exercise prescription and compliance during the 10-week aerobic exercise C vigor 15.61 5.61 15.00 7.00 13.72 6.96
program E vigor 15.00 6.08 16.52 8.46 17.88 7.32
C TMD 11.22 32.15 11.94 32.94 19.11 43.72
Heart rate
E TMD 2.11 17.31 8.00 21.69 9.581 19.89
Group Mean S.D. Days/week Minutes/session
1
p  .05.
Prescribed 103.41 5.53 4.00 41.00 2
C  control group; E  aerobic exercise group; TMD  total mood
Achieved 103.58 5.47 3.62 38.10 disturbance.
 A.D. Sculco et al. / The Spine Journal 1 (2001) 95–101 99

Table 4 Table 5
The effect of aerobic exercise on pain as assessed by the Brief Pain Number of office visits for pain/symptoms for all subjects during the 30-
Inventory during the 10-week aerobic exercise program month follow-up
Week 0 Week 5 Week 10 Exercise Nonexercise
BPI Group Mean S.D. Mean S.D. Mean S.D. Treatment Group N Mean S.D. N Mean S.D.
C average 4.27 2.21 4.00 2.24 4.05 2.33 Aerobic exercise 10 1.18 1.51 7 1.85 2.14
E average 3.70 1.31 3.05 2.19 3.23 1.67 Nonexercise 10 1.57 3.61 7 1.73 4.23
C worst 5.55 3.05 5.94 2.18 5.11 2.80
E worst 4.76 2.19 3.29 2.56 3.47 2.00
C least 2.94 2.33 2.88 1.81 3.16 2.40
E least 1.82 1.42 1.52 1.32 2.05 1.81
the majority of the 30-month follow-up period showed sig-
BPI  Brief Pain Inventory; C  control group; E  aerobic exercise nificantly fewer prescriptions for pain (AE X2.76,
group.
S.D.4.42; control X13.35, S.D.16.61; p.02) and
were given fewer physical therapy referrals (AE X0.17,
S.D.0.39; control X1.64; S.D.1.54; p.002). There
minutes was prescribed for the 10-week phase of the study.
was no difference in number of office visits for pain/symp-
These exercise intensity, duration, and frequency results are
toms (Table 5) or in number of epidural blocks administered
within the American College of Sports Medicine guidelines
to either group (Fig. 1). In addition, work status was signifi-
for inducing an aerobic training effect [32]. There were no
cantly improved (AE X0.24, S.D.0.56; control X3.53,
reports of injury by patients adhering to the prescribed AE
S.D.0.702; p.04) for the subjects who exercised for the
training program during the 10-week phase or the 30-month
majority of the 30-month follow-up period when compared
follow-up period. In addition, all AE and control patients re-
with those who did not (Fig. 2).
ported adherence to the study safety guidelines and were
free of other injury during the 10-week phase of the study.
There were no statistically significant differences in the Discussion
mean scores between the AE and control groups on any of
The present investigation used a matched stratified de-
the baseline POMS parameters. After 5 weeks of interven-
sign to assess the effect of a low to moderate program of
tion, the AE group showed a significant decrease in depres-
aerobic exercise, isolated as a sole exercise intervention, for
sion (p.012), anger (p.002), and tension (p.020). The
patients with chronic or subacute LBP in treatment in a neu-
control group did not change in any of the above parame-
rosurgical practice. Ten weeks of AE was associated with
ters. After 10 weeks of the intervention, the AE group
positive psychological outcomes, as indicated by the POMS
showed significantly less depression (p.019), anger
questionnaire. These outcomes include reduced depression,
(p.013), and total mood disturbance (p.009) than the
anger, and total mood disturbance for the AE patients com-
control group (Table 3). The difference between the AE and
pared with control patients. Other studies using AE in con-
control group for depression appears to be the result of
junction with other LBP treatments have also shown de-
the control group becoming more depressed while the AE
creases in depression, anxiety, and total mood disturbance
group remained at pretest levels. The AE group showed im-
[33–35]. In comparison to other studies using AE to treat
proved mood change for anger and total mood disturbance,
LBP, this research was unique in that AE was isolated as the
whereas the control group showed more disturbance. There
sole exercise intervention and the exercise prescription was
were no significant changes in pain for the AE or control
carried out individually through a home-based program.
groups after the 10-week phase of the study as assessed by
Thus, the positive psychological changes occurred without
the BPI (Table 4).
At the end of the 10-week pilot phase of the study, all
subjects were advised to exercise following the intervention
exercise prescription. During the 30-month follow-up, our
findings indicated that several control subjects began to ex-
ercise and some AE subjects discontinued their exercise.
After the 30-month follow-up period, all subjects (control
and AE) were categorized as those who followed the exer-
cise prescription (exercised at least 16 months of the 30-
month follow-up period versus those who did not) for the
majority of this period. The average monthly exercise par-
ticipation by the subjects (out of 30 possible months) in
these two groups was 26.29 and 5.35 months, respectively.
None of the original AE patients crossed over to the nonex-
ercise group because of any problems relating to their previ- Fig. 1. Relationship between aerobic exercise activity and medical treat-
ous exercise participation. The subjects who exercised for ments for low back pain.
100 A.D. Sculco et al. / The Spine Journal 1 (2001) 95–101
Fig. 2. Relationship between aerobic exercise activity and work status.
the influence of group camaraderie (which often develops in
group rehabilitation programs) or the influence of contact
with the study investigator.
In contrast to other LBP studies using AE in conjunction
with other treatments, 10 weeks of AE as a sole treatment in
patients with LBP was not effective in reducing pain/symp-
toms [19,36–38]. The baseline levels of pain in the present
study were approximately 2 to 5 points for both groups on a
0-point (no pain) to 10-point (most pain) visual analogue
scale. If AE were to reduce pain/symptoms, it might be eas-
ier to detect pain decreases in patients with higher initial
pain levels. However, another viewpoint might suggest that
if initial pain levels were high, patients might drop out (stop
prescribed exercise routine) and the positive psychological
changes might not have occurred. Should AE reduce symp-
toms of pain in patients with LBP, a more vigorous exercise
program than that prescribed in this study might be needed
to induce these changes. The positive changes in mood
states associated with AE might also have been caused by
an indirect influence of AE, which could give patients a re-
duced level of fear for daily activities and exercise. Other
psychological tests would need to be measured to evaluate
this possibility. Although pain intensity was not reduced in
this study, it is possible that pain affect (pain tolerance, in-
fluence of pain on daily activities, etc.) might have been
changed by AE. It was noted that no exercise sessions were
missed by the AE group because of elevated pain symptoms
or back injuries. Thus, it is concluded that a conservative
home-based low to moderate AE program is safe and im-
proves psychological mood states but does not alter symp-
toms of pain in patients with LBP.
The 30-month follow-up portion of the study comparing
exercisers with nonexercisers on their use of medical ser-
vices and their work status indicated that there were signifi-
cantly fewer prescriptions for pain medications, fewer refer-
rals for physical therapy, and improved work status for
those who adhered to the exercise program. An outcome
measure very relevant to medical costs in our society is the
work status of patients with chronic LBP. Return to work
has been used as an important outcome measure in trials of
various interventions [8,39–42] for chronic LBP. Our find-
ings using AE as a sole intervention for patients with LBP
were consistent with previous studies using AE in conjunc-
tion with other LBP treatments in showing significantly im-
proved work status [12,14,34,35,39,43].
After the 10-week AE versus control study, all patients
were encouraged to exercise and were given the original ex-
ercise prescription and guidelines. Within 6 months, one
half of both the AE and control groups had crossed over to
the other group and remained self-selected to AE or nonex-
ercise for the remaining 24 months. This grouping has the
potential flaw that the patients were not originally random-
ized into these groups. Because the original matched strati-
fied pretests showed no differences in pretested variables,
we can probably assume the crossover groups were roughly
equal at the start of the study and thus self-selection would
probably not explain the results. There were also no differ-
ences in pain between the AE and control groups after the
10-week study. Another limitation of this study is that the
generalizability of the results to other populations of pa-
tients with LBP remains unclear. The patients in this study
were from a single neurosurgical practice, were subacute
and chronic patients most of whom underwent several med-
ical and/or surgical interventions before this exercise inter-
vention. Thus, this was a fairly select group of patients. Fur-
thermore, because this was not a true randomized design,
generalizability of results can be questioned. However, the
matched stratified design we employed is probably a better
choice when a relatively small number of patients are en-
rolled, because our design ensures roughly equal numbers in
both the control and experimental groups.
Conclusions
Ten weeks of AE intervention at a low to moderate exer-
cise intensity is an effective treatment for improving psy-
chological status but may not be an effective therapy for re-
ducing perceptions of pain in patients with chronic LBP. In
addition, this study suggests that long-term AE is a safe
therapy for lumbar spinal patients with chronic low back
pain and causes no new pain or exacerbation of pain/symp-
toms. Based on the above results, it appears that regularity
in performing AE at a low to moderate intensity over a long
period of time might provide a continued benefit in overall
mood state and a possible improvement in functional status
because of the general benefits of AE. In addition, AE ap-
pears to be a useful therapy in reducing the use of pain med-
ications and reducing physical therapy referrals from a phy-
sician as well as aiding in the improvement in work status.
Therefore, we conclude that AE should be considered for
inclusion in the treatment of patients with chronic LBP.
References
[1] Casazza B, Young J, Herring S. The role of exercise in the prevention
and management of acute low back pain. Occup Med State Art Rev
1998;13:51–60.
[2] Deyo R, Andersson G, Bombardier C, et al. Outcome measures for
 A.D. Sculco et al. / The Spine Journal 1 (2001) 95–101
studying patients with low back pain. Spine 1994;19(18 Suppl):
2032S–6S.
[3] Deyo R, Diehl A. Measuring physical and psychosocial function in
patients with low-back pain. Spine 1983;8:635.
[4] Berwick DM, Budman S, Feldstein M. No clinical effect of back
schools in an HMO. Spine 1989:14(3):338–44.
[5] Pollock M, Leggett S, Graves J, et al. Effect of resistance training on
lumbar extension strength. Am J Sport Med 1989;17:624–9.
[6] Van Tulder MW, Koes BW, Bouter LM. Conservative treatment of
acute and chronic nonspecific low back pain: a systematic review of
randomized controlled trials of the most common interventions.
Spine 1997;22(18):2128–56.
[7] Faas A, Chavannes C, van Eijk J, et al. A randomized, placebo-con-
trolled trial of exercise therapy in patients with acute low back pain.
Spine 1993;18:1388–95.
[8] Keane G, Saal J. The sports medicine approach to occupational low
back pain. West J Med 1981;154:525–7.
[9] Mitchell RI, Carmen GM. Results of a multicenter trial using an in-
tensive active exercise program for the treatment of acute soft tissue
and back injuries. Spine 1990;15(6):514–21
[10] Dyer J, Crouch J. Effects of running and other activities on moods.
Percep Mot Skill 1988;67:43–50.
[11] Gaskin M, Greene A, Robinson M, et al. Negative affect and the ex-
perience of chronic pain. J Psychosom Res 1992;36:707–13.
[12] Mayer T. Discussions: exercise, fitness, and back pain. In: Bouchard
C, Shephard R, Stephens T, et al., editors. Exercise, fitness, and
health. Champaign (IL): Human Kinetics, 1990:541.
[13] Morgan W. Affective beneficence of vigorous physical activity. Med
Sci Sports Exerc 1984;17:94–100.
[14] Mayer D, Watkins L. Role of endorphins in endogenous pain control
systems. In: Enrich H, editor. The role of endorphins in neuropsychi-
atry: modern problems of pharmo-psychiatry. Basel, Switzerland:
Karger, 1981:68–96.
[15] Vanharanta H. The intervertebral disc: a biologically active tissue
challenging therapy. Ann Med 1994;26:395–9.
[16] Cady L, Bischoff D, O’Connell E, et al. Strength and fitness and sub-
sequent back injury in fire fighters. J Occup Med 1979;21:269–72.
[17] Cady L, Thomas P, Karawsky R. Program for increasing health and
physical fitness of fire fighters. J Occup Med 1985;27:110–4.
[18] Burnham JM. Exercise is medicine: health benefits of regular physi-
cal activity. J La State Med Soc 1998 ;150:319–23.
[19] Mannion A, Muentener M, Taimela S. A randomized clinical trial of
three active therapies for chronic low back pain. Spine 1999;24:
2435–48.
[20] Nutter P. Aerobic exercise in the treatment and prevention of low
back pain. Occup Med State Art Rev 1988;3:137–45.
[21] Callaghan J, Paila A, McGill S. 3D analysis of spine loading during
gait. Presented at the 1996 meeting of the American Society for Bio-
mechanics, Atlanta, GA, October 17–19, 1996.
[22] McGill S. Distribution of tissue loads in the low back during a variety
of daily and rehabilitation tasks. J Rehab Res Dev 1997;34(6):448–58.
[23] McGill S. Low back exercises: evidence for improving exercise re-
gimes. Phys Ther 1998;7:754–65.
[24] McGill S. Low back exercises: prescription for the healthy back and
when recovering from injury. In: Roitman J, ed. ACSM’s resource
manual for guidelines for exercise testing and prescription. 3rd ed.
Baltimore: Williams and Wilkins, 1998:116–26.
101
[25] Klaussen K, Andersen L, Pelle I. Adaptive changes in work capacity,
skeletal muscle capillarization and enzyme levels during training and
detraining. Acta Physiol Scand 1981;113:9–16.
[26] Nachemson A. Exercise, fitness, and back pain. In: Bouchard C,
Shephard R, Stephens T, et al., editors. Exercise, fitness, and health.
Champaign (IL): Human Kinetics, 1990:533.
[27] King A, Taylor C, Haskell W, et al. Influence of regular aerobic exer-
cise on psychological health: a randomized, controlled trial of healthy
middle-aged adults. Health Psych 1989;8:305–24.
[28] Martinsen E. Benefits of exercise for the treatment of depression.
Sport Med 1990;9:380–9.
[29] Wilfley D, Kunce J. Differential physical and psychological effects
of exercise. J Coun Psych 1986;33:337–41.
[30] McNair DM, Lorr LE, Dropplerman LF. EITS manual for the Profile
of Mood States. San Diego: Educational and Individual Testing Ser-
vice, 1971/1981:1–11.
[31] Serlin RC, Mendoza TR, Nakamura Y, et al. When is cancer pain
mild, moderate, or severe? Grading pain severity by its interference
with function. Pain 1995;61:277–84.
[32] American College of Sports Medicine. ACSM’s guidelines for exer-
cise testing and prescription. 6th ed. Baltimore: Lippincott Williams
and Wilkins, 2000:135–64.
[33] Davies JE, Gibson T, Tester L. The value of exercises in the treat-
ment of low back pain. Rheum Rehab 1979;18:243–7.
[34] Turner J, Clancy S, McQuade K, et al. Effectiveness of behavioral
therapy for chronic low back pain: a component analysis. J Cons Clin
Psych 1990;58:573–9.
[35] Turner J, Romano J. Evaluating psychologic interventions for chronic
pain: issues and recent developments. Adv Pain Res Ther 1984;7:
257.
[36] Carpenter DM, Nelson BW. Low back strengthening for the preven-
tion and treatment of low back pain. Med Sci Sports Exerc 1999;
31(1):18–24.
[37] Frost H, Lamb SE, Klaber-Moffett JA, Fairbank JC, Moser JS. A fit-
ness programme for patients with chronic low back pain: 2-year fol-
low-up of a randomised controlled trial. Pain 1998;75(2–3):273–9.
[38] Split W, Kotwica S, Woszczak M. Role of exercise therapy in the
prevention of lumbosacral backache in textile industry workers. Med
Pr 1984;35(6):465–9.
[39] Kellett K, Keller D, Nordholm L. Effects of an exercise program on
sick leave due to low back pain: a clinical trial. J Sports Ther 1995;
22:52–9.
[40] Lindstrom I, Ohlund C, Eek, C. et al. The effect of graded activity of
patients with subacute low back pain: a randomized prospective clini-
cal study with an operant-conditioning behavioral approach. Phys
Ther 1992;72(4):279–93.
[41] Malmivaara A, Kakkinen U, Aro T, et al. The treatment of acute low
back pain—bed rest, exercise, or ordinary activity? N Engl J Med
1995;332:351–5.
[42] Teasell RW, Harth M. Functional restoration: returning patients with
chronic low back pain to work—revolution or fad? Spine 1996 21(7):
844–7.
[43] Torstensen T, Ljundgren A, Meen H, et al. Efficiency and costs of
medical exercise therapy, conventional physiotherapy, and self-exer-
cise in patients with chronic low back pain. A pragmatic, randomized,
single-blinded, controlled trial with 1-year follow-up. Spine 1998;23:
2616–24.