610039

research-article2015
CRE0010.1177/0269215515610039Clinical RehabilitationLi et al.

CLINICAL
Article REHABILITATION

Clinical Rehabilitation

The effects of resistance exercise 2016, Vol. 30(10) 947­–959

© The Author(s) 2015
Reprints and permissions:
in patients with knee sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0269215515610039

osteoarthritis: a systematic cre.sagepub.com

review and meta-analysis

Yanan Li, Youxin Su, Shaoqing Chen, Yingjie Zhang,

Ziyi Zhang, Changyan Liu, Meili Lu, Feiwen Liu, Shuzhen Li,
Zhen He, Yiru Wang, Lu Sheng, Wenting Wang,
Zhengxuan Zhan, Xu Wang and Naixi Zheng

Abstract
Objective: To analyze the effectiveness of resistance exercise in the treatment of knee osteoarthritis on
pain, stiffness, and physical function.
Design: Systematic review and meta-analysis of randomized controlled trials.
Data sources: PubMed, Embase, Cochrane Central Register of Controlled Trials, the Web of Science,
and Chinese Biomedical Literature Database were searched from the date of inception to August 2015.
Methods: Trials comparing effects of resistance exercise intervention with either non-intervention or
psycho-educational intervention were selected by two reviewers independently. The risk of bias was
assessed and studies with similar outcomes were pooled using a fixed or random effects model.
Results: Data from 17 randomized clinical trials including 1705 patients were integrated. The main
source of methodological bias in the selected studies was lack of double blinding. The meta-analysis
results suggested that resistance exercise training relieved pain (standard mean difference [SMD]: -0.43;
95% confidence interval [CI]: -0.57 to -0.29; P < 0.001), alleviated stiffness (SMD: -0.31; 95%: CI -0.56 to
-0.05; P = 0.02), and improved physical function (SMD -0.53; 95% CI: -0.70 to -0.37; P < 0.001).
Conclusion: Resistance exercise is beneficial in terms of reducing pain, alleviating stiffness, and improving
physical function in patients with knee osteoarthritis.

Keywords
Resistance exercise, knee osteoarthritis, pain, meta-analysis

Received: 6 May 2015; accepted: 5 September 2015

Department of Rehabilitation Medicine College, Fujian

Introduction University of Traditional Chinese Medicine, Fujian, China

Osteoarthritis is the most common age-related rheu- Corresponding author:

Youxin Su, Department of Rehabilitation Medicine College,
matic disease and a major cause of chronic pain and Fujian University of Traditional Chinese Medicine, 1 Qiuyang
disability worldwide,1 affecting approximately 52.5 Street, Shangjie, Fuzhou, Fujian, China.
million adults, or 22.7% of the population.2 Due to Email: suyouxin777@hotmail.com

Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
948 Clinical Rehabilitation 30(10)
its predilection for the knee,3 patients with sympto-
matic knee osteoarthritis often complain of intermit-
tent pain, limitation of joint movement, and difficulty
in daily activities. In light of the aging of the world’s
population, there is a large increase in the demand
for treatments of knee osteoarthritis.4
Therapeutic exercise is recommended in numerous
international guidelines as a nonpharmacologic treat-
ment for knee osteoarthritis.5–7 Resistance exercise
— defined as use the external resistance load (e.g.
body weight, resistance bands, machines) to force
skeletal muscles contract — may be successful in
normalizing muscle firing patterns and joint biome-
chanics for knee osteoarthritis.8 Resistance exercise
can be performed in a variety of ways, and numerous
studies have shown that it can reduce pain.9–11
Several previous systematic reviews have out-
lined and shown the effectiveness of resistance exer-
cise for knee osteoarthritis.12–15 One of these reviews
included other interventions, such as aerobic exer-
cise, performance exercise, or mixed exercise.14 A
review reported in 2005 focused on home based
quadriceps strengthening exercise, excluded other
types of resistance exercise.12 In addition, another
review included a variety of the surgical patients and
included studies comparing with active control
group(e.g. paracetamol, short-wave diathermy treat-
ment, stretching exercise),15 variation in this review
may effects resistance exercise outcomes. A review
reported in 2013 meta analysed the effect of weight-
bearing strengthening exercise and non-weight-
bearing strengthening exercise separately, and this
study looked at pain only.13 Furthermore, none of
them found evidence that resistance exercise ther-
apy had a significant benefit on stiffness alleviating,
the evidence is still inconclusive. Because of the
expansion of research in the area of resistance exer-
cise in the past years, an updated systematic review
and meta-analysis of research data in this area would
be useful in informing clinical practice.
The aims of this systematic review were to
answer the following questions. Should resistance
exercise be used to:
•• reduce pain in patients with knee osteoarthritis?
•• alleviate stiffness in patients with knee
osteoarthritis?
Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
•• improve physical function in patients with knee
osteoarthritis?
Methods
Inclusion and exclusion criteria for studies were
based on study design, participants, interventions,
comparisons, and outcomes as follows. Available
human randomized controlled trials (RCTs) were
included. Observational studies were excluded.
Participants who had been diagnosed with unilat-
eral or bilateral knee osteoarthritis were consid-
ered. Age, sex, disease duration and severity were
unrestricted. The diagnosis had to be made based
on the classification criteria of the American
College of Rheumatology.16, 17 Participants who
had undergone a knee surgery or had history of
rheumatoid arthritis were excluded. Studies per-
forming resistance exercise were included. Studies
with multiple interventions (for example, resist-
ance exercise plus aerobic exercise, neuromuscular
electrical stimulation, or ultrasound) were
excluded. Studies comparing with no treatment or
educational interventions were included. The out-
come measures pain, stiffness, or physical function
were included.
Original articles were searched in five elec-
tronic databases: PubMed, Embase, Cochrane
Central Register of Controlled Trials (CENTRAL),
the Web of Science, and Chinese Biomedical
Literature Database (CBM) from their inception
to August 2015. The search strategies can be
found in the appendix, available online. Reference
lists of retrieved RCTs and previous systematic
reviews were also searched.
Two reviewers (ZYZ and CYL) independently
screened titles and abstracts of retrieved studies to
exclude obviously irrelevant studies. Full texts of
the potential studies were reviewed to determine
eligible trials. Articles were considered relevant
based on the inclusion and exclusion criteria.
Disagreements between reviewers were resolved
by discussion.
Data from all eligible trials were extracted by two
reviewer (MLL and FWL) independently. A special
standardised forms, developed for this review, was
used to record authors’ names, publication year,
Li et al. 949
demographics of participants, content of interven-
tion, outcome measures and other information. If
the reported data were incomplete or ambiguous,
we requested additional information from corre-
sponding authors. Conflicts were resolved through
discussion.
When a given study reported more than one
pain, stiffness, or physical function measure, we
gave preference to the Western Ontario and
McMaster Universities Osteoarthritis Index
(WOMAC) scale, which has been widely used
worldwide and employed in most trials.18
Two reviewers (MLL and FWL) independently
assessed the quality of studies, and the Cochrane
Collaboration’s tool was used for assessing the risk
of bias.19 Recommended domains were considered
as following: selection bias, performance bias,
detection bias, attrition bias, reporting bias, and
other bias. For the “other bias” item on the risk of
bias scale, we assessed the compliance acceptable
in all groups. We requested additional information
from corresponding authors if it is necessary, and
disagreements were resolved through discussion.
For the continuous data, if studies used different
scales to assess the same outcome, the standardized
mean difference (SMD) with 95% confidence
intervals (CI) were calculated. If studies used same
scale to assess the same outcome, the mean differ-
ences (MD) with 95% CI were used.
Statistical heterogeneity among studies was
assessed using the Chi-square test or by calculating
Higgins I2 values.20 The results were pooled using
a fixed effect model if the I2 value was < 40%.
Otherwise, a random effect model was applied.
However, if the I2 value was > 75%, the heteroge-
neity was regarded as substantial, and overall
meta-analysis was considered inappropriate. In
such cases, subgroup analysis was considered to
measure the pooled effect.
The Review Manager Software (version 5.3.5,
updated 2014 Cochrane Collaboration) was used
for data analysis, and all P values were two-sided.
Evaluation of overall effect size was based on
Cohen’s categories: effect sizes of 0.2–0.5 was
considered small, effect sizes of 0.5–0.8 consid-
ered moderate, and effect sizes > 0.8 considered
large effects.21
Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
If data within any study were insufficient for
pooling even after requesting details from the
authors, we excluded this study from meta-analyses
and described the result narratively.
Results
Study selection and characteristics
Seventeen RCTs were included in this review,9–11, 22–35
and the flowchart of study selection is shown in
Figure 1. In total, 1705 patients with the mean age
of 63.5 years were included, and characteristics of
the included studies are shown in Table 1. Studies
were conducted in the United States,9, 24, 25, 27, 30, 34, 35
England,26, 32 Australia,10, 23, 31, 33 Turkey,22 and
China.11, 28, 29
The included studies varied substantially in deliv-
ery mode, resistance load, duration, repetitions and
frequency in resistance exercise program. Resistance
exercise was delivered individually,22, 26, 27, 31, 32, 34
in a class-based format,35 performed under supervi-
sion in facility,11, 27–29 or at home.24, 26, 30–32 Resistance
loads varied from low resistance (10% 1RM) to high
resistance (80% 1RM), and there is a periodic
increase. Intervention duration ranged from eight
weeks to 24 months, and the most commonly utilized
program was three days per week, with three sets at
8–12 repetitions per set.
Majority of the included studies used the self-
report WOMAC scale to measure pain, stiffness,
and physical function, while the Visual Analog
Scale (VAS),22 Osteoarthritis Screening Index
(OASI),27 Knee Pain Scale (KPS)35 were also used
to measure outcomes.
Risk of bias in included studies
Methodological quality scores are shown in
Supplementary Figure 1 and 2. According to the
Cochrane Collaboration, none of the RCTs received
the maximum quality score. In all of the included
RCTs, random sequences were generated by using a
computer or random table, except one study used
pseudo-randomly.9 The most common source of
likely methodological bias was lack of double blind-
ing. Only two of the 17 included studies reported
950 Clinical Rehabilitation 30(10)

Figure 1.  Flowchart of randomised controlled trials included in the systematic review.

blinding of both participants and outcomes asses-
sors.10, 33 However, in the item of blinding assess-
ment, eight studies presented a low risk.11, 23, 28, 29, 31,
32, 34, 35 For the item intent-to-treat four studies pre-
sented a high risk9, 10, 33, 34 and 13 studies a low
risk.11, 22-32, 35 The description of compliance was
generally insufficient in many studies.
Effect of intervention
The effect of resistance exercise is shown in
Figure 2. Pooling of the data for resistance exercise
using a random effect model revealed a small effect
size (SMD: -0.43; 95% CI: -0.57 to -0.29; P <
0.001) for pain relief (2030 subjects) and a moder-
ate effect size (SMD: -0.53; 95% CI: -0.70 to -0.37;
P < 0.001) for physical function improvement
(2077 subjects). Pooling the data using a fixed
effect model revealed a small effect size (SMD:
-0.31; 95% CI: -0.56 to -0.05; P = 0.02) for stiff-
ness alleviation (254 subjects). The study heteroge-
neity was calculated as I2 = 46%, 0%, and 58% for
pain, stiffness, and physical function, respectively.
The effect of high intensity and low intensity
resistance exercise on pain, stiffness, and physical
function is shown in Figure 3. Both categories

Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
 Table 1.  Characteristics of included studies.
Study Participants Intervention Frequency Outcome Outcome Outcome
Li et al.

(per week) (pain) (stiffness) (function)

Size I/C Age(y) Sex (F%) I/C BMI(kg/m2)
±SD I/C ±SD I/C
Schilke, 199627 10 / 10 64.5 ± NR 85.0 (in all NR 8 weeks of muscle- 3 days OASI OASI subscale AIMS physical
68.4 ± NR subjects) strength training vs. subscale activity
control
Ettinger,199735 146 / 149 68 ± 6.0 73 / 69 NR 18 months of 3 days KPS _ Self-report of
69 ± 6.0 resistance exercise Physical Disability
vs. health education
program
Maurer,199934 57 / 56 66.3 ± 8.8 43.4 / 35.7 NR 8 weeks of exercise vs. 3 days WOMAC _ WOMAC function
64.5 ± 8.4 education pain subscale
subscale
O’Reilly,199926 113 / 78 61.9 ± 10.0 NR NR 6 months of exercise 7 days WOMAC _ WOMAC function
62.3 ± 9.7 vs. control pain subscale
subscale
Baker, 200130 23 / 23 69.0 ± 6.0 73.9 / 82.6 31.0 ± 4.0 4 months of home 3 days WOMAC _ WOMAC function
68.0 ± 6.0 32.0 ± 5.0 based progressive pain subscale
resistance training vs. subscale
a nutrition education
program
Gür, 200222 9/8/6 55 ± 12.0 NR 32.1 ± NR 8 weeks of concentric 3 days VAS _ _
56 ± 12.0 31.6 ± NR vs. concentric-
57 ± 9.0 32.2 ± NR eccentric vs. control
Topp, 200224 32 / 35 / 35 63.5 ± 1.9 65.6/71.4/80.0 NR 16 weeks of isometric 3 days WOMAC WOMAC WOMAC function

Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
65.6 ± 1.8 exercise vs. dynamic pain stiffness subscale
60.9 ± 1.8 exercise vs. control subscale subscale
Thomas, 200232 235 / 78 61.5 ± 9.58 64.0 / 65.0 28.0 ± 4.18 24 months of exercise 7 days WOMAC WOMAC WOMAC function
61.9 ± 9.39 28.1 ± 4.81 vs. control pain stiffness subscale
subscale subscale
Jan, 200811 34 / 34 / 34 63.3 ± 6.6 79.4 /79.4/83.3 24.1 ± NR 8 weeks of high- 3 days WOMAC _ WOMAC function
31.8 ± 7.1 24.0 ± NR strength training vs. pain subscale
62.8 ± 6.3 24.1 ± NR low-strength training subscale
vs.control
(Continued)
951
 Table 1. (Continued)
952

Study Participants Intervention Frequency Outcome Outcome Outcome

(per week) (pain) (stiffness) (function)
Size I/C Age(y) Sex (F%) I/C BMI(kg/m2)
±SD I/C ±SD I/C
Lim, 200823 VM VM VM VM 12 weeks of home- 5 days WOMAC _ WOMAC function
26 / 26 67.2 ± 6..7 50.0 28.2 ± 3.7 based quadriceps pain subscale
NA 66.6 ± 8.9 46.2 30.3 ± 5.3 strengthening program subscale
27 / 28 NA NA NA vs. control
64.1 ± 9.3 62.9 29.0 ± 5.2
60.8 ± 7.8 60.7 28.4 ± 5.0
Jan, 200928 35 / 35 63.2 ± 6.8 71.4 / 68.6 25.1 ± NR 8 weeks of nonweight- 3 days _ _ WOMAC function
62.2 ± 6.7 24.9 ± NR bearing exercise vs. subscale
control
Lin, 200929 36 / 36 61.6 ± 61.6 66.7 / 72.2 23.7 ± NR 8 weeks of strength 3 days WOMAC _ WOMAC function
62.2 ± 6.7 24.7 ± NR training vs. control pain subscale
subscale
Bennell, 201031 45 / 44 64.5 ± 9.1 51.1 / 45.5 27.5 ± 4.7 12 weeks of hip 5 days WOMAC _ WOMAC function
64.6 ± 7.6 28.4 ± 4.1 strengthening exercises pain subscale
vs. control subscale
Foroughi, 201133 26 / 28 66.0 ± 8.0 NR 31.4 ± 5.4 6 months of exercise 3 days WOMAC WOMAC WOMAC function
65.0 ± 7.0 32.7 ± 8.4 vs. sham-exercise pain stiffness subscale
subscale subscale
Foroughi, 201135 26 / 28 64.0 ± 7.0 NR 31.9 ± 5.2 6 months of 3 days WOMAC WOMAC WOMAC function
64.0 ± 8.0 33.2 ± 8.1 progressive resistance pain stiffness subscale
training vs. sham subscale subscale
exercise

Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
Wortley, 20139 15 / 9 69.5 ± 6.7 69.2 / 66.7 30.5 ± 6.0 10 weeks of resistance 2 days WOMAC WOMAC WOMAC function
70.5 ± 5.0 30.0 ± 6.2 training program vs. pain stiffness subscale
control subscale subscale
Aaboe, 201425 15 / 15 57.1 ± 60.0 27.1 ± 12 weeks of quadriceps 3 days WOMAC WOMAC WOMAC function
7.7 (in all (in all subjects) 4.0 (in all strengthening group pain stiffness subscale
subjects) subjects) vs.control subscale subscale

F: Female; BMI: body mass index; I/C: intervention /control group; NR: not reported; OASI: the Osteoarthritis Screening Index; KPS: knee pain scale; VAS: visual analog scale;
WOMAC: Western Ontario and McMaster Universities Osteoarthritis Index; VM: varus malalignment subjects; NA: neutral alignment subjects.
Clinical Rehabilitation 30(10)
Li et al. 953

Figure 2.  Forest plots for effects of resistance exercise (RE) on pain, stiffness, and physical function.

Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
954 Clinical Rehabilitation 30(10)

Figure 3.  Forest plots for effects of resistance exercise (RE) intensity on pain, stiffness, and physical function.

Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
Li et al. 955
achieved significant treatment benefits in terms of
pain and physical function. Pooling of the data for
high intensity resistance exercise revealed a large
effect size for pain relief and physical function
improvement. Meanwhile, pooling the data in low
intensity resistance exercise revealed a small effect
size. However, for both high intensity and low
intensity resistance exercise, there is no significant
treatment benefits in terms of stiffness.
The effect of resistance exercise duration on
pain, stiffness, and physical function is shown in
Figure 4 and Figure 5. Both categories achieved
significant treatment benefits in terms of pain and
physical function. Pooling of the data for 12 or less
weeks resistance exercise revealed a moderate
effect size on stiffness alleviation. However, there
is no benefits for more than 12 weeks resistance
exercise.
Discussion
According to the meta-analyses, outcomes showed
statistically significant results comparing resist-
ance exercise with control group, and the resist-
ance exercise is beneficial in terms of reducing
pain, alleviating stiffness, and improving physical
function.
Knee osteoarthritis has been linked to decreased
muscular strength, increased ligament stiffness,
and alterations in muscle activation patterns.36
Resistance exercise may normalize muscle firing
patterns and joint biomechanics leading to reduc-
tions in pain and cartilage degradation, and restore
muscle strength and joint mechanics while improv-
ing physical function.8
Our results support previous reviews which
reported improvements in pain and function with
resistance exercise,12, 15 and showed a moderate
effect size (SMD: -0.53; 95% CI: -0.70 to -0.37)
compared with Roddy’s review a small effect size
(SMD: -0.32; 95% CI: -0.23 to -0.41) for physical
function improvement. This may be due to Roddy’s
review included only home-based quadriceps
strengthening trails which ignored hamstrings and
hip muscles strengthening trails.
Previous studies involving hip abductor and
adductor strengthening and resistance exercises
Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
found improvement in muscle strength.10, 37 Only
one included study addressed the role of hip mus-
cles strengthening in the treatment of knee osteoar-
thritis,31 and majority of the studies highlight the
strengthening of the quadriceps and hamstrings
with a gradually increased load ignoring the fact
that individuals with knee osteoarthritis experience
weakness in the hip muscles.38
Previous reviews did not investigate the effect
of resistance training on stiffness. Our study dem-
onstrated that resistance training resulted in allevi-
ating stiffness comprising with control group.
However, the effect size was small (SMD: -0.31;
95% CI: -0.56 to -0.05), it may due to the small
sample size (254 subjects).
Components of a resistance exercise program
include intensity, delivery mode, duration, and fre-
quency. Exercise intensity is commonly reported as
the percentage of 1RM (repetition maximum). Our
study demonstrated that both high intensity (6 to 8
repetitions at 80% of 1RM) and low intensity (12
to 15 repetitions at 60% of 1RM) resistance exer-
cise are benefit to pain and physical function, and
high intensity resistance exercise program seems
more effective than low intensity. Resistance exer-
cise can be performed under supervision in facil-
ity,11, 27–29 or at home.24, 26, 30–32 And a study reported
that the differences between the various forms of
exercise delivery were not statistically significant
for knee osteoarthritis,14 but encouraging patients
to maintain exercise programmes beyond a super-
vised period of instruction is a major challenge.
There was considerable variation in the duration
of the exercise programme, length of intervention
ranged from eight weeks to 24 months. Although,
our results showed that there was no significant
treatment benefits between within 12 weeks and
more than 12 weeks duration for pain and physical
function, within 12 weeks duration is effective for
stiffness alleviation. For the frequency of the resist-
ance exercise program, three days per week pro-
gram was used in majority of the studies. It may
due to moderate weekly frequency could promote
marked gains on muscle hypertrophy and power
gains in elderly subjects.39
There are also some important caveats to note
about resistance exercise. First of all, it concerns
956 Clinical Rehabilitation 30(10)

Figure 4.  Forest plots for effects of resistance exercise (RE) duration on pain and stiffness.

the responsiveness of self-reported pain and physi- present. Secondly, it concerns the placebo effect
cal function. Exercise makes people feel better for knee osteoarthritis patients.
generally and in doing so may affect self-reported The most common source of methodological
outcome even if changes to function are not bias was lack of double blinding. Thus, the nature

Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
Li et al. 957

Figure 5.  Forest plots for effects of resistance exercise (RE) duration on physical function.

of the resistance exercise intervention, and the
awareness of being treated may lead to perfor-
mance bias. Consequently, we estimated the risk of
performance bias as high for all studies except two
studies that utilized a sham exercise control group,
whose patients trained on the same equipment as
the patients from the intervention group but did not
perform hip abduction and adduction.10, 33 However,
the patients could easily identify the group they
were assigned to, and the authors did not report if
their blinding approach was successful. Since the
sham exercise group was not a true sham control
group, these studies received an “unclear” score in
the performance bias domain.
There are several limitations in our study. First,
we may have missed some relevant studies pub-
lished in other languages other than English or
Chinese. Second, our analysis was not according
to ITT (Intention-To-Treat) due to some studies
did not reported ITT data, which may bias our
result. Third, since we excluded patients who had
undergone a knee surgery, many severe knee oste-
oarthritis participants were not included in this
review, which may exaggerate the outcomes.
Finally, the result of this study is lack of long-
term effect and overall beneficial effect of resist-
ance exercise.
Even though there are some limitations in this
review, there are also some implications for prac-
tice. For knee osteoarthritis, benefits can be
obtained with leg exercises, basic components of
the prescription should include seated leg presses,
leg extensions and leg curls (with ankle cuff
weights to provide resistance). Inclusion of hip
adduction and abduction can help with improving
and maintaining appropriate knee mechanics.
Although, both high intensity and low intensity
resistance exercise are benefit to pain and physical
function, high intensity program is more effective
than low intensity program. Moreover, within 12

Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
958 Clinical Rehabilitation 30(10)
weeks duration of resistance exercise program is
effective for stiffness alleviation.
The treatment effect for many of the studies was
only small and may have been influenced by the
low methodological rigor of the studies included in
the meta-analyses. Future randomized controlled
trials about knee osteoarthritis must follow the
rules of the CONSORT Statement,40 and optimal
resistance exercise prescription of randomized
clinical trials should be tested to provide greater
benefit and evidence. Initiate research to assess the
long-term effectiveness of exercise for people with
knee osteoarthritis.
Clinical messages
•• Resistance exercise is effective on pain
relief, stiffness alleviation, and physical
function improvement in knee
osteoarthritis.
•• High intensity resistance exercise had a
lager effect size for pain and physical
function than low intensity.
•• Within 12 weeks of resistance exercise
was effective for stiffness.
Conflict of interest
The authors declare that there is no conflict of interest.
Funding
The present study was supported by grants from the State
Administration of Traditional Chinese Medicine (No.
201307004).
References
1. McAlindon TE, Bannuru RR, Sullivan MC, et al.
OARSI guidelines for the non-surgical management of
knee osteoarthritis. Osteoarthritis and cartilage/OARS,
Osteoarthritis Research Society. 2014; 22: 363–388.
2. Ma VY, Chan L and Carruthers KJ. Incidence, prevalence,
costs, and impact on disability of common conditions
requiring rehabilitation in the United States: stroke, spi-
nal cord injury, traumatic brain injury, multiple sclerosis,
osteoarthritis, rheumatoid arthritis, limb loss, and back
pain. Archives of physical medicine and rehabilitation.
2014; 95: 986–995.
3. Johnson VL and Hunter DJ. The epidemiology of osteo-
arthritis. Best practice & research Clinical rheumatology.
2014; 28: 5–15.
Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
4. Cross M, Smith E, Hoy D, et al. The global burden of hip
and knee osteoarthritis: estimates from the global burden
of disease 2010 study. Ann Rheum Dis. 2014; 73: 1323–
1330.
5. National Collaborating Centre for Chronic C. National
Institute for Health and Clinical Excellence: Guidance.
Osteoarthritis: National clinical guideline for care
and management in adults. London: Royal College of
Physicians (UK) Royal College of Physicians of London,
2008.
6. McAlindon TE, Bannuru RR, Sullivan MC, et al. OARSI
guidelines for the non-surgical management of knee oste-
oarthritis. Osteoarthritis And Cartilage. 2014; 22: 363–
388.
7. Richmond J, Hunter D, Irrgang J, et al. American Academy
of Orthopaedic Surgeons clinical practice guideline on the
treatment of osteoarthritis (OA) of the knee. The Journal
of bone and joint surgery American volume. 2010; 92:
990–993.
8. Vincent KR and Vincent HK. Resistance exercise for knee
osteoarthritis. PM&R. 2012; 4: S45–S52.
9. Wortley M, Zhang SN, Paquette M, et al. Effects of resist-
ance and Tai Ji training on mobility and symptoms in
knee osteoarthritis patients. Journal Of Sport And Health
Science. 2013; 2: 209–214.
10. Foroughi N, Smith RM, Lange AK, et al. Progressive
resistance training and dynamic alignment in osteoarthri-
tis: A single-blind randomised controlled trial. Clinical
biomechanics (Bristol, Avon). 2011; 26: 71–77.
11. Jan MH, Lin JJ, Liau JJ, et al. Investigation of clinical
effects of high- and low-resistance training for patients
with knee osteoarthritis: A randomized control-led trial.
Physical therapy. 2008; 88: 427–436.
1 2. Roddy E, Zhang W and Doherty M. Aerobic walk-
ing or strengthening exercise for osteoarthritis of the
knee? A systematic review. Ann Rheum Dis. 2005; 64:
544–548.
13. Tanaka R, Ozawa J, Kito N, et al. Efficacy of strength-
ening or aerobic exercise on pain relief in people with
knee osteoarthritis: a systematic review and meta-analysis
of randomized controlled trials. Clinical rehabilitation.
2013; 27: 1059–1071.
14. Fransen M and McConnell S. Exercise for osteoarthritis of
the knee. The Cochrane database of systematic reviews.
2009: Cd004376.
15. Juhl C, Christensen R, Roos EM, et al. Impact of exercise
type and dose on pain and disability in knee osteoarthri-
tis: a systematic review and meta-regression analysis of
randomized controlled trials. Arthritis & rheumatology
(Hoboken, NJ). 2014; 66: 622–636.
16. Altman R, Asch E, Bloch D, et al. Development of cri-
teria for the classification and reporting of osteoarthritis.
Classification of osteoarthritis of the knee. Diagnostic
and Therapeutic Criteria Committee of the American
Rheumatism Association. Arthritis Rheum. 1986; 29:
1039–1049.
Li et al. 959
17. Belo JN, Berger MY, Koes BW, et al. The prognostic
value of the clinical ACR classification criteria of knee
osteoarthritis for persisting knee complaints and increase
of disability in general practice. Osteoarthritis and carti-
lage / OARS, Osteoarthritis Research Society. 2009; 17:
1288–1292.
18. Bellamy N. Validation study of WOMAC: a health status
instrument for measuring clinically-important patient-rel-
evant outcomes following total hip or knee arthroplasty in
osteoarthritis. J Orthop Rheumatol. 1988; 1: 95–108.
19. Savovic J, Weeks L, Sterne JA, et al. Evaluation of the
Cochrane Collaboration’s tool for assessing the risk of
bias in randomized trials: focus groups, online survey,
proposed recommendations and their implementation.
Systematic reviews. 2014; 3: 37.
20. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring
inconsistency in meta-analyses. BMJ (Clinical research
ed). 2003; 327: 557–560.
21. Cohen J. Statistical power analysis for the behavioral sci-
ences. Academic press, 1988.
22. Gur H, Cakin N, Akova B, et al. Concentric versus com-
bined concentric-eccentric isokinetic training: effects on
functional capacity and symptoms in patients with osteo-
arthrosis of the knee. Archives of physical medicine and
rehabilitation. 2002; 83: 308–316.
23. Lim BW, Hinman RS, Wrigley TV, et al. Does knee
malalignment mediate the effects of quadriceps strength-
ening on knee adduction moment, pain, and function in
medial knee osteoarthritis? A randomized controlled trial.
Arthritis and rheumatism. 2008; 59: 943–951.
24. Topp R, Woolley S, Hornyak J, et al. The effect of
dynamic versus isometric resistance training on pain and
functioning among adults with osteoarthritis of the knee.
Archives of physical medicine and rehabilitation. 2002;
83: 1187–1195.
25. Aaboe J, Henriksen M, Bartholdy C, et al. The effect of
quadriceps-strengthening exercise on quadriceps and knee
biomichanics during walking in adults with knee osteoar-
thritis: A randomized controlled trial. Osteoarthritis and
Cartilage. 2014; 22: S80–S81.
26. O’Reilly SC, Muir KR and Doherty M. Effectiveness of
home exercise on pain and disability from osteoarthritis of
the knee: a randomised controlled trial. Ann Rheum Dis.
1999; 58: 15–19.
27. Schilke JM, Johnson GO, Housh TJ, et al. Effects of mus-
cle-strength training on the functional status of patients
with osteoarthritis of the knee joint. Nursing research.
1996: 45: 68–72.
28. Jan MH, Lin CH, Lin YF, et al. Effects of Weight-Bearing
Versus Nonweight-Bearing Exercise on Function, Walking
Speed, and Position Sense in Participants With Knee
Osteoarthritis: A Randomized Controlled Trial. Archives of
physical medicine and rehabilitation. 2009; 90: 897–904.
Downloaded from cre.sagepub.com at UNIV CALIFORNIA SAN DIEGO on October 31, 2016
29. Lin DH, Lin CH, Lin YF, et al. Efficacy of 2 non-weight-
bearing interventions, proprioception training versus
strength training, for patients with knee osteoarthritis: a
randomized clinical trial. The Journal of orthopaedic and
sports physical therapy. 2009; 39: 450–457.
30. Baker KR, Nelson ME, Felson DT, et al. The efficacy of
home based progressive strength training in older adults
with knee osteoarthritis: a randomized controlled trial.
The Journal of rheumatology. 2001; 28: 1655–1665.
31. Bennell KL, Hunt MA, Wrigley TV, et al. Hip strength-
ening reduces symptoms but not knee load in people
with medial knee osteoarthritis and varus malalignment:
a randomised controlled trial. Osteoarthritis and carti-
lage / OARS, Osteoarthritis Research Society. 2010; 18:
621–628.
32. Thomas KS, Muir KR, Doherty M, et al. Home based
exercise programme for knee pain and knee osteoarthritis:
randomised controlled trial. BMJ (Clinical research ed).
2002; 325: 752.
33. Foroughi N, Smith RM, Lange AK, et al. Lower limb
muscle strengthening does not change frontal plane
moments in women with knee osteoarthritis: A rand-
omized controlled trial. Clinical Biomechanics. 2011;
26: 167–174.
34. Maurer BT, Stern AG, Kinossian B, et al. Osteoarthritis
of the knee: isokinetic quadriceps exercise versus an edu-
cational intervention. Archives of physical medicine and
rehabilitation. 1999; 80: 1293–1299.
35. Ettinger WH Jr, Burns R, Messier SP, et al. A randomized
trial comparing aerobic exercise and resistance exercise
with a health education program in older adults with knee
osteoarthritis. The Fitness Arthritis and Seniors Trial
(FAST). Jama. 1997; 277: 25–31.
36. Bennell KL, Hunt MA, Wrigley TV, et al. Role of mus-
cle in the genesis and management of knee osteoarthritis.
Rheumatic diseases clinics of North America. 2008; 34:
731–754.
37. Sled EA, Khoja L, Deluzio KJ, et al. Effect of a home
program of hip abductor exercises on knee joint load-
ing, strength, function, and pain in people with knee
osteoarthritis: a clinical trial. Physical therapy. 2010; 90:
895–904.
38. Costa RA, Oliveira LM, Watanabe SH, et al. Isokinetic
assessment of the hip muscles in patients with osteoar-
thritis of the knee. Clinics (Sao Paulo, Brazil). 2010; 65:
1253–1259.
39. Cadore EL, Pinto RS, Bottaro M, et al. Strength and
endurance training prescription in healthy and frail
elderly. Aging and disease. 2014; 5: 183–195.
40. Schulz KF, Altman DG and Moher D. CONSORT 2010
statement: updated guidelines for reporting parallel
group randomised trials. International journal of surgery
(London, England). 2011; 9: 672–677.