683000

research-article2016
CRE0010.1177/0269215516683000Clinical RehabilitationWang et al.

CLINICAL
Original Article REHABILITATION

Clinical Rehabilitation

Effects of short-wave therapy in 1­–12

© The Author(s) 2016
Reprints and permissions:
patients with knee osteoarthritis: sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0269215516683000

A systematic review and cre.sagepub.com

meta-analysis

Haiming Wang1,2,*, Chi Zhang1,2,3,*, Chengfei Gao1,2,

Siyi Zhu1,2, Lijie Yang4, Quan Wei1,2 and Chengqi He1,2

Abstract
Objective: To evaluate the efficacy and safety of short-wave therapy with sham or no intervention for
the management of patients with knee osteoarthritis.
Methods: We searched the following databases from their inception up to 26 October 2016: MEDLINE,
CENTRAL, EMBASE, Physiotherapy Evidence Database, CINAHL and OpenGrey. Studies included
randomized controlled trials compared with a sham or no intervention in patients with knee osteoarthritis.
The results were calculated via standardized mean difference (SMD) and risk ratio for continuous variables
outcomes as well as dichotomous variables, respectively. Heterogeneity was explored by the I2 test and
inverse-variance random effects analysis was applied to all studies.
Results: Eight trials (542 patients) met the inclusion criteria. The effect of short-wave therapy on pain was
found positive (SMD, −0.53; 95% CI, −0.84 to −0.21). The pain subgroup showed that patients received
pulse modality achieved clinical improvement (SMD, –0.83; 95% CI, –1.14 to −0.52) and the pain scale in
female patients decreased (SMD, −0.53; 95% CI, −0.98 to −0.08). In terms of extensor strength, short-
wave therapy was superior to the control group (p < 0.05, I2 = 0%). There was no significant difference in
the physical function (SMD, −0.16; 95% CI, −0.36 to 0.05). For adverse effects, there was no significant
difference between the treatment and control group.
Conclusion: Short-wave therapy is beneficial for relieving pain caused by knee osteoarthritis (the pulse
modality seems superior to the continuous modality), and knee extensor muscle combining with isokinetic
strength. Function is not improved.

Keywords
Knee osteoarthritis, short-wave therapy, diathermy, systematic review, meta-analysis

Date received: 12 July 2016; accepted: 8 November 2016

1Department of Rehabilitation Medicine, West China Hospital, *Contributed equally

Sichuan University, Chengdu, Peoples’ Republic of China
2Rehabilitation Key Laboratory of Sichuan Province, Chengdu,
Corresponding author:
Peoples’ Republic of China Chengqi He, Rehabilitation Medicine Center, West China
3Department of Rehabilitation Medicine, Affiliated Hospital of
Hospital, Sichuan University, Guo Xue Road, 610041
Southwest Medical University, Luzhou, Peoples’ Republic of Chengdu, Sichuan, People’s Republic of China.
China Email: hxkfhcq2015@126.com
4Department of Stomatology, First Affiliated Hospital of

Zhengzhou University, Zhengzhou, Peoples’ Republic of China

2 Clinical Rehabilitation 
Introduction
Knee osteoarthritis, characterized by loss of joint
cartilage that leads to pain and loss of function pri-
marily,1 is an age-related condition that affects a lot
more women (18%) than men (9.6%) aged
>60 years.1 In the absence of a curative approach,
knee osteoarthritis treatment aims at pain reduction
and function improvement as well.2 Despite that
advanced knee osteoarthritis patients could benefit
from the pain reduction effect of total knee replace-
ment, nearly all of them will be in need of a long-
term pain control.3,4
Short-wave therapy is a non-invasive treatment
modality that typically employs electromagnetic
radiation at 27.12 MHz delivered either in a continu-
ous or a pulse mode.5 Clinically, as one of the most
frequently prescribed electrotherapy modalities by
physicians,6 short-wave therapy has been applied to
treat an acute and chronic musculoskeletal pain for
many decades.7 Short-wave therapy has the possibil-
ity of becoming an attractive alternative or adjunct
therapy to pharmacological-based pain therapies,
even becoming a home-use pain therapy.7
However, the Osteoarthritis Research Society
International guideline for the non-surgical man-
agement of knee osteoarthritis in 2014 did not men-
tion short-wave therapy.8 One of the most crucial
reasons may be the conflicting results produced by
the small sample sizes of knee osteoarthritis in rel-
evant studies.6,9–11 A previous systematic review of
short-wave therapy found small, significant effects
on pain and muscle performance only when this
therapy evoked a local thermal sensation.12 The
author also stated that definitive conclusions were
difficult to draw. Meanwhile, certain debates were
raised regarding the methodologies of such system-
atic review.13,14 Moreover, there has not been a sys-
tematic review since 2012 and an increased number
of new clinical trials should be included in an
updated systematic review. Thus, we performed this
updated systematic review and meta-analysis to
explore whether short-wave therapy is an effective
treatment for knee osteoarthritis.
Methods
Databases that were searched to identify articles
included MEDLINE (via pubmed), CENTRAL,
EMBASE, Physiotherapy Evidence Database,
CINAHL (via EBSCOhost) and OpenGrey, all
from the implementation to 26 October 2016. The
detailed example of the full electronic search strat-
egy for MEDLINE is provided in Appendix 1,
available online.
Study selection
Two authors independently selected and evaluated
articles, and solved disagreements through consen-
sus. If necessary, the third reviewer author was
involved. Only randomized controlled trials that
compared short-wave therapy with sham (placebo)
or no intervention being assessed for pain and physi-
cal function in patients with knee osteoarthritis were
included. There were no language limits. The review-
ers screened titles and abstracts to select relevant tri-
als first. Subsequently, the full texts of the selected
and the remaining articles were read in details to
identify their eligibility. Criteria details of consider-
ing studies for this review were as follows.
1. Types of studies: All prospective randomized
controlled trials (both parallel and cross-over
trial designs were acceptable) were included,
whereas trials of no-randomized, clinical
observations, case reports, letters, abstracts
and reviews were eliminated.
2. Types of participants: The diagnosis was based
on the American College of Rheumatology diag-
nostic criteria for knee osteoarthritis,15 without
restrictions on age, gender, ethnicity and setting.
3. Types of intervention: The type, dose, fre-
quency and mode of short-wave therapy were
not limited. Trials were included subject to the
condition that short-wave therapy was supple-
mented by an additional modality or additional
modalities for intervention groups. Furthermore,
the study also contained a comparison group
applying the additional modality or the addi-
tional modalities alone.
4. Types of outcome measures: Primary outcomes
were pain intensity and physical function. Pain
including visual analogue scale16,17 and numer-
ical rating scales.18 Function measurement
including Western Ontario and McMaster
Universities (WOMAC) physical function
Wang et al. 3
scores,19 Lequesne indexes,20 arthritis impact
measurement scale21 and other functional
scales. Secondary outcomes were stiffness of
knee, muscle strength and adverse events
caused by short-wave therapy.
Data extraction and
management
Pairs of review authors extracted data indepen-
dently. We resolved any disagreements by consen-
sus or by consulting a third review author. A test
data form was applied to extract the following data:
Study participants (first author, year published,
clinical setting, samples size, gender and age),
short-wave therapy interventions used (frequency,
wave-mode, intensity, dose and device), combined
intervention, outcome measures (characteristics of
pain, physical function, safety outcomes, other
measures mentioned and follow-up). For any effect
sizes that were not reported or could not be calcu-
lated in the study, the authors would be contacted.
The extracted data were entered into a database
(Table 1, available online).
Assessment of risk of bias
included studies
The risk of bias was assessed in accordance with
methods of the Cochrane Collaboration’s ‘Risk of
bias’. Figures generated by the RevMan 5.3 (http://
ims.cochrane.org/revman) were presented to pro-
vide summary assessments of the risk of bias.
Moreover, the methodological quality of the
included studies was also assessed by the
Physiotherapy Evidence Database (PEDro)22 tool,
providing a score between 0 and 10 points (http://
www.pedro.fhs.usyd.edu.au/scale_item.html).
Scores of trials were either provided by the PEDro
website, or scored by our research group. Two
authors independently assessed the risk of bias in
included trials. We resolved any disagreement by
consensus or by consulting a third review author.
Statistical analyses
Where available and appropriate, with respect
to continuous data, standardized mean difference
(SMD) was employed to estimate the treatment
effect. All outcomes were compared applying the
random-effect model with 95% confidence inter-
vals (CI). The high and low dose short-wave ther-
apy subgroups in one trial were combined into a
single group, and the subgroups containing both
placebo short-wave therapy as well as control sub-
group without therapy in the same trial were also
combined into one single group.10 This was the for-
mula for calculation following the Cochrane
Handbook for systematic reviews (version 5.1.0).
The posttreatment means and deviations of trials
were pooled on the basis that the trials were suffi-
cient. Also, subgroup analysis was conducted
based on short-wave therapy mode and partici-
pants’ gender. We planned to test the heterogeneity
between comparable trials applying a standard I2
test considered statistically significant with a P
value <0.05. The I2 statistic value thresholds23 were
interpreted as follows: 0% to 40%, might not be
important; 30% to 60%, may represent moderate
heterogeneity; 50% to 90%, may represent sub-
stantial heterogeneity; and 75% to 100%, consider-
able heterogeneity. If the heterogeneity was
considered to have serious effects on the validity of
results, we would give up combining the data.
Also, any evidence of heterogeneity would be
identified so that it could determine whether there
were significant differences between trials that
might trigger heterogeneity. If the I2 test was not
significant (p > 0.05), sensitivity analysis would be
conducted to investigate the potential sources of
heterogeneity. Such analysis was performed
through omitting one study and identifying its
impact on the overall pooled estimate. P < 0.05 was
regarded as statistically significant. All statistical
analyses were performed by RevMan 5.3.
Results
Study selection
Eight studies (542 patients) were included in this
research. Figure 1 shows the flow details of the
included trials selected in different phases. In the
phase of full text articles assessed for eligibility,
two trials10,24 that were conducted in the same hos-
pital by the same group at different time (2008 and
2011) met the inclusion criteria, with the subject
4 Clinical Rehabilitation 
Figure 1.  Flow of information through the different phases of our systematic review.
collections of the former covered by the latter.
Unfortunately, we were not able to get in touch
with the authors. In order to draw a more conserva-
tive result, we included the latest trial, in accord-
ance with the previous systematic review.12
Preferred Reporting Items for Systematic reviews
and Meta-Analyses (PRISMA) guidance25 were
confirmed during this research flow figure.
Description of included studies
Eight parallel groups randomized placebo con-
trolled studies published between 2005 and 2012
were included. Three trials were conducted in
Turkey,6,9,26 one in each of United Kingdom,27
Italy,28 Brazil,10 Israel29 and Thailand.11 In total, the
studies included 542 participants who were
enrolled from a clinical setting. The ages of patients
ranged from 42 to 85 years old. Four trials9–11,26
included only females, while the other four tri-
als6,27–29 contained both females and males, with a
greater proportion of females compared with
males. The duration of intervention varied between
two to eight weeks, and the sessions of which
ranged from six to 24 treatments. The frequency of
short-wave therapy was 27.12 HZ, with exception
of one trial27 (27 HZ). Furthermore, the frequency
applied in another trial28 was not mentioned. The
output intensity of all trials ranged from 1.8 W to
20 W, while three studies9,26,28 did not present the
specific intensity. For short-wave mode, four tri-
als6,9,11,26 applied the continuous short-wave ther-
apy (we could not make contact with the authors of
one trial,26 but the previous systematic review’s
author12 did, and hence we employed their opin-
ion), while four others10,27–29 applied the pulse
Wang et al. 5
(one28 of which was inferred through the author’s
objective). Two trials10,29 included two treatment
groups that contained both high- and low-dose
short-wave therapy, while the other six trials
involved only one treatment group. When it came
to the control groups, no short-wave therapy was
given to three trials.9,26,28 Four trials6,11,27,29 included
a placebo or sham short-wave intervention and the
device of all of them was operationalized without
current output or power, appearing similar in
appearance to the treatment group device. One
trial10 involved two control groups, receiving no
short-wave therapy treatment and placebo short-
wave therapy without output. The description and
characteristics of the included trials are summa-
rized in Table 1, available online.
Risk of bias of included studies
Two risk of bias assessment tools were employed
in this systematic review. Both can be found in the
online figures and tables, with Figure 2 showing
the Cochrane library assessment tool, and Table 2
displaying the PEDro methodological quality
assessment results.
In order to promise randomization and conceal-
ment, issue and sealed envelopes containing the
names or codes of groups were used in two tri-
als.10,11 Moreover, the person who opened the
envelopes was also described, namely one11 by an
independent physical therapist, and the other10 by
an individual not involved in this study. The rand-
omization and concealment of one6 of the trials
was performed by the principal centre using adap-
tive assignment. One9 trial only reported with the
use of concealed envelopes, in the absence of any
details to describe how to generate the randomiza-
tion numbers and to perform the concealment. Four
trials26–29 provided insufficient details on the rand-
omization and concealment.
When it came to blinding, sham or placebo, a
short-wave therapy device was utilized in three tri-
als6,11,29 without any output delivery. The use of a
sham short-wave therapy was also reported in one
trial27 in the control group. Yet, there was not suf-
ficient information to ensure the sham device. One
trial10 conducted both a control group without
treatment and a placebo short-wave therapy with-
out output. The other three trials9,26,28 involved the
additional modality or modalities alone without
blinding the patients.
Four trials9,11,28,29 conducted a less than three-
month follow-up, while one trial6 conducted both a
three-month and six-month follow-up. Only one
trial10 was reported with a 12-month follow-up.
Three trials6,10,11 performed intention-to-treat anal-
ysis in outcomes analysis. As for financing, only
one trial27 received the funding support supplied by
their hospital trust fund. The other seven tri-
als6,9–11,26,28,29 did not mention information whether
they were supported by any commercial organiza-
tions or other funds.
Synthesis of results
Table 3, available online, demonstrates an overall
summary of the effects of short-wave therapy on
all synthesized results of interest.
Pain. Five trials assessed pain utilizing a visual
analogue scale. Three of these five trials were
measured in centimetres, whereas the other two
were measured in millimetres. One trial10 used the
numerical rating scales. Since both visual analogue
scale and numerical rating scales are high reliabil-
ity tools for assessing the generic unidimensional
pain questionnaires17 on knee osteoarthritis, the
pain results of six trials were pooled. The aggre-
gated results suggested that short-wave therapy
could reduce pain in knee osteoarthritis patients.
Subgroup analysis was performed based on gender
and different modes (pulse mode and continuous
mode). For trials that contained only females, the
pooled results of three trials suggested that short-
wave therapy was capable of reducing pain. As for
trials that contained both females and males, they
did not favour short-wave therapy treatment. In
addition, the aggregated results of three pulse mode
trials suggested that short-wave therapy would
reduce pain. By contrast, the pooled results of three
continuous mode short-wave therapy trials did not
support pain reduction (Figure 3).
6 Clinical Rehabilitation 

Figure 3.  Forest plot of trials comparing the effects of short-wave therapy and control changes on knee
osteoarthritis pain. (a) The aggregated results of pain changes (six trials). (b) The aggregated results of pain changes
in groups contained female only (three trials). (c) The aggregated results of pain changes in groups contained
both female and male (three trials). (d) The aggregated results of pain changes in groups applied pulse short-wave
therapy (three trials). (e) The aggregated results of pain changes in groups applied continuous short-wave therapy
(three trials).

Physical function. All of the eight included trials
assessed physical function. The data supplied in the
articles of Rattanachaiyanont (2008)11 and Cantar-
ini (2007)28 were insufficient. Although efforts
were made to make attempts to contact both authors,
we failed to obtain physical function results from
Cantarini. However, Rattanachaiyanont did provide
us with specific data. One study10 employed the
Knee Injury and Osteoarthritis Outcome Score to
assess physical function, high scores showing high
level function opposite to the other tools (WOMAC,
Lequesne indexes or arthritis impact measurement
scale). Therefore, this study was excluded. As a
result, six trials involving 377 patients were aggre-
gated for meta-analysis, indicating no beneficial
effects. We also conducted a subgroup analysis
based on gender and different modes (pulse mode
and continuous mode). Both gender and different
Wang et al. 7

Figure 4.  Forest plot of trials comparing the effects of short-wave therapy and control on physical function
changes. (a) The aggregated results of physical function changes (six trials). (b) The aggregated results of physical
function changes in groups only contained female people (three trials). (c) The aggregated results of physical
function changes in groups contained both female and male people (three trials). (d) The aggregated results of
pain changes in groups applied pulse short-wave therapy (two trials). (e) The aggregated results of pain changes in
groups applied continuous short-wave therapy (four trials).

modes subgroups analysis suggested that short-
wave therapy was not able to make any improve-
ment in physical function without heterogeneity
(Figure 4).
Stiffness.  Three trials9,11,29 conducted the WOMAC
stiffness subscale. We contacted one author11 to
obtain the specific data as the original article did not
manifest sufficient information about the results.
The pooled results of these trials were not in favour
of short-wave therapy (Figure 5).
Muscle strength. Three trials assessed muscle
strength with a focus on isokinetic knee motion.
Two of these trials conducted strength at three
speeds (60, 120 and 180°/s), while the other one
only assessed the speed at 90°/s. Thus, we aggre-
gated the two trials in three different speeds
8 Clinical Rehabilitation 
Figure 5.  Forest plot of trials comparing the effects of short-wave therapy and control on WOMAC stiffness
subscale changes (three trials).
involving both extensor and flexor. The extensor
strength in all three speeds showed significant
improvement in the treatment group combining with
isokinetic exercise, without heterogeneity (p < 0.05,
I2 = 0%). Additionally, no significant difference was
found in the flexor strength of both groups in all
three speeds with very low heterogeneity (p > 0.05,
I2 < 28%) (Appendix 2, available online).
Follow-up.  Six trials carried out follow-up: One6 of
which performed both three-month and six-month
follow-up after the beginning, another trial10 only
did the 12 months one, and the other four trials9,11,28,29
conducted the follow-up at no more than three
months. The pooled results of pain follow-up at no
more than three months (three trials) suggested that
there was no significant difference between the
treatment group and the comparison group in
accordance with the indication of pain follow-up for
more than three months. The aggregated results of
physical function of no more than three months fol-
low-up manifested that there was no significant dif-
ference between the short-wave therapy treatment
group and the comparison group. Nevertheless, data
were not sufficient to be pooled for a long-term fol-
low-up regarding physical function. In terms of stiff-
ness with follow-up of no more than three months,
the pooled results of three trials did not favour short-
wave therapy treatment. Also, the long-term follow-
up results could not be aggregated for only one trial6
data (Figure 6).
Adverse events. Only two trials11,28 recorded the
treatment adverse events. One28 of these two trials
did not encounter side-effects in the short-wave
therapy group, while the other11 was observed with
light events in four patients similar with the pla-
cebo group, showing no statistical differences. No
adverse event was reported in the other six trials.
Safety results manifested that short-wave therapy
was a safe treatment, and the risk ratio could not be
calculated.
Discussion
The main finding of this systematic review and
meta-analysis indicates that short-wave therapy
treatment is an effective approach of pain relief but
does not improve physical function in knee osteo-
arthritis patients. In terms of subgroup analysis on
pain, it shows that the pulse modality seems supe-
rior to the continuous modality (without heteroge-
neity) and effective outcomes may be found in
females (moderate heterogeneity). By combining
short-wave therapy with isokinetic exercise, knee
extensor strength can be improved. Yet, such ther-
apy modality is not beneficial for stiffness relief.
The low number of adverse responses during short-
wave therapy shows it could be a safe treatment for
knee osteoarthritis cure.
As the main result of our study supports pain
relief, it is inconsistent with the earlier review.12 In
contrast to the previous systematic review, we only
included trials applying generic unidimensional
pain questionnaires, and excluded trials using
generic multidimensional pain questionnaires, pri-
marily WOMAC pain subscale. WOMAC, which
is the most commonly used knee osteoarthritis
measure, has shown high correlations between the
pain and physical function subscale. Thus, the
Wang et al. 9

Figure 6.  Forest plot of trials comparing the follow-up effects of short-wave therapy and control. (a) The
aggregated no more than three months pain follow-up results of pain changes (three trials). (b) The aggregated
more than three months pain follow-up results (two trials). (c) The aggregated no more than three months physical
function follow-up results (four trials). (d) The aggregated no more than three months WOMAC stiffness subscale
follow-up results (three trials).

assessment of knee osteoarthritis pain employing
WOMAC is easily confounded by physical disabil-
ity.17 To decrease the heterogeneity and potential
bias, we adopted the unidimensional pain question-
naires to evaluate pain, and function measurement
to assess physical function, respectively. We also
contacted with the authors to obtain the outcomes
of two trials.11,28 One11 of the two trials was
included in the previous review without pooled
effect size. In addition, different dose short-wave
therapy subgroups were combined together to
make up the treatment group. In terms of compari-
son groups, both placebo short-wave therapy and
the control subgroup without therapy in the same
study were also combined into one single group,
while the previous systematic review chose the
strongest positive effect group to pool the results.
We excluded one trial30 of individuals with both
hip and knee osteoarthritis in our study, also differ-
ent from the previous review.14
10 Clinical Rehabilitation 
The positive effect on pain relief was in favour
of short-wave therapy, and the pooled pain esti-
mate was unchanged by the removal of each study,
showing a robust outcome without changing the
heterogeneity. In clinical practice, which type of
short-wave mode is much more effective in pain
relief for knee osteoarthritis patients is an impor-
tant problem the physicians want to know. Hence,
different modes of short-wave therapy were aggre-
gated in this study. The results suggested that pulse
short-wave therapy was of superiority compared
with the continuous mode when it comes to reliev-
ing pain. Such a result was robust without hetero-
geneity, exhibited by sensitivity analysis. Clinically,
whether the short-wave effects are different in dif-
ferent genders is another problem of physicians’
concern, as more women are epidemiologically
affected by knee osteoarthritis than men.1 Hence,
we pooled the trials containing only female people
and trials containing both female and male people.
The outcomes were indicative for the phenomenon,
which was that short-wave therapy seems effective
in women with knee osteoarthritis.
In the physical function assessment, six trials
were pooled for analysis results. The results
revealed that four trials used the WOMAC physical
function subscale, one trial applied Lequesne
indexes, and the other employed the arthritis impact
measurement scale. We performed sensitivity anal-
yses and subgroup analysis, and the aggregated
results did not favour short-wave therapy in the
improvement of physical function. Our study basi-
cally agreed with the previous results. The meas-
urement of the WOMAC stiffness subscale was
only reported in three trials. The aggregated result
did not support short-wave therapy with moderate
heterogeneity. As joint stiffness is one of the most
common symptoms in osteoarthritis,31 we sug-
gested clinical trials are needed to assess whether
short-wave therapy can improve the stiffness of
knee osteoarthritis people. We also pooled the fol-
low-up data for meta-analysis, which was missing
in the previous review. The aggregated results indi-
cated that short-wave therapy could act as an
immediate pain relief strategy after treatment.
In terms of muscle strength, both extensor and
flexor were conducted in all reported speeds,
whereas different from only the extensor of 60°/s
being performed in the previous systematic review.
Three trials included in our study assessed muscle
strength. Two of the three trials conducting strength
at the three speeds (60, 120 and 180°/s) could be
pooled for meta-analysis, while the other trial27
only reported the 90°/s speed. As suggested by the
pooled results, short-wave therapy brought bene-
fits to the knee extensor muscle combining with
isokinetic strength, showing no improvement of
the knee flexor muscle. Though the main effect of
short-wave therapy is not to improve muscle
strength, the effect of combining it with isokinetic
strength is valuable in clinical experiment, for the
important role of extensor in knee osteoarthritis.
Recent reviews32,33 regarded the impaired muscle
function as an important risk factor for develop-
ment of knee osteoarthritis, and the meta-analysis34
also demonstrated that knee extensor muscle weak-
ness was associated with an increased risk of devel-
oping knee osteoarthritis in both men and women.
The finding of our study was indicative of the
potential of the combination of short-wave therapy
and isokinetic exercise and they are becoming an
effective treatment for the improvement of exten-
sor strength of knee osteoarthritis patients.
When it comes to interpreting the results of our
study, some limitations should be cautiously con-
sidered. First, though we searched six main data-
bases and the references of the included trials, we
may miss some other grey literature and hand
search of relevant literature, which may cause
selection bias. Second, there are only a limited
number of trials added compared with the previ-
ous review. So, the evidence was limited owing to
the small number of included studies and hetero-
geneity. Further high-quality studies are required
to provide stronger evidence. However, the stabil-
ity of the results of the sensitivity analysis con-
firms that the main outcomes of this study are
robust. Third, there is variability in treatment
duration, number of sessions, intensity of output
and other parameters in the included trials, which
may cause heterogeneity and also limit the evi-
dence. Fourth, the included trials were only writ-
ten in English, though we searched for studies
without restriction of languages, which may also
produce a selection bias. Last, the follow-up data
were not sufficient, especially the long-term item,
Wang et al. 11
and the pooled results indicated moderate to sub-
stantial heterogeneity.
Clinical messages
•• Short-wave therapy is beneficial for
relieving pain of knee osteoarthritis.
•• Pulse mode short-wave therapy seems
superior to continuous modality.
•• Short-wave therapy increases knee exten-
sor strength.
Acknowledgements
The authors thank Vilai Kuptniratsaikul MD and Professor
Antonella Fioravanti for providing the initial data.
Contributors
HaiMing Wang and Chi Zhang were responsible for the
literature search and data extraction. ChengFei Gao and
SiYi Zhu performed the blinding literature search and
the data extraction. Quan Wei and Lijie Yang were in
charge of data interpretation and technical guidance.
ChengQi He made important intellectual contribution in
research design and manuscript revision.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest
with respect to the research, authorship, and/or publica-
tion of this article.
Funding
The author(s) disclosed receipt of the following financial
support for the research, authorship, and/or publication
of this article: The financial support for the research,
authorship and/or publication of this article was spon-
sored by National Natural Science Foundation of China
[Grant Number 81572236].
Reference
1. Woolf AD and Pfleger B. Burden of major musculo-
skeletal conditions. Bull World Health Org 2003; 81:
646–656.
2. Bennell KL, Hall M and Hinman RS. Osteoarthritis year
in review 2015: Rehabilitation and outcomes. Osteoarth
Cart 2016; 24: 58–70.
3. Losina E, Paltiel AD, Weinstein AM, et al. Lifetime medi-
cal costs of knee osteoarthritis management in the United
States: Impact of extending indications for total knee
arthroplasty. Arthritis Care Res. 2015; 67: 203–215.
4. Smith SR, Deshpande BR, Collins JE, Katz JN and Losina
E. Comparative pain reduction of oral non-steroidal anti-
inflammatory drugs and opioids for knee osteoarthritis:
Systematic analytic review. Osteoarth Cart 2016; 24:
962–972.
5. Shields N, Gormley J and O’Hare N. Short-wave dia-
thermy: Current clinical and safety practices. Physiother
Res Int 2002; 7: 191–202.
6. Atamaz FC, Durmaz B, Baydar M, et al. Comparison of
the efficacy of transcutaneous electrical nerve stimulation,
interferential currents, and shortwave diathermy in knee
osteoarthritis: A double-blind, randomized, controlled, mul-
ticenter study. Arch Phys Med Rehabil 2012; 93: 748–756.
7. Rawe IM. The case for over-the-counter shortwave ther-
apy: Safe and effective devices for pain management.
Pain Management 2014; 4: 37–43.
8. McAlindon TE, Bannuru RR, Sullivan MC, et al. OARSI
guidelines for the non-surgical management of knee oste-
oarthritis. Osteoarth Cart 2014; 22: 363–388.
9. Akyol Y, Durmus D, Alayli G, et al. Does short-wave
diathermy increase the effectiveness of isokinetic exercise
on pain, function, knee muscle strength, quality of life,
and depression in the patients with knee osteoarthritis?: A
randomized controlled clinical study. Eur J Phys Rehabil
Med 2010; 46: 325–336.
10. Fukuda TY, Alves da Cunha R, Fukuda VO, et al. Pulsed
shortwave treatment in women with knee osteoarthritis: A
multicenter, randomized, placebo-controlled clinical trial.
Phys Ther 2011; 91: 1009–1017.
11. Rattanachaiyanont M and Kuptniratsaikul V. No additional
benefit of shortwave diathermy over exercise program for
knee osteoarthritis in peri-/post-menopausal women: An
equivalence trial. Osteoarth Cart 2008; 16: 823–828.
12. Laufer Y and Dar G. Effectiveness of thermal and ather-
mal short-wave diathermy for the management of knee
osteoarthritis: A systematic review and meta-analysis.
Osteoarth Cart 2012; 20: 957–966.
13. Laufer Y and Dar G. Response to Letter to the Editor:
Comment on Laufer et al. entitled “Effectiveness of ther-
mal and athermal short-wave diathermy for the manage-
ment of knee osteoarthritis: A systematic review and
meta-analysis”. Osteoarth Cart 2014; 22: 607–608.
14. Zeng C, Gao SG and Lei GH. Comment on Laufer et al.
entitled “Effectiveness of thermal and athermal short-
wave diathermy for the management of knee osteoarthri-
tis: A systematic review and meta-analysis”. Osteoarth
Cart 2014; 22: 605–606.
15. 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.
16. Gentelle-Bonnassies S, Le Claire P, Mezieres M, Ayral
X and Dougados M. Comparison of the responsiveness
12 Clinical Rehabilitation 
of symptomatic outcome measures in knee osteoarthritis.
Arthritis Care Res 2000; 13: 280–285.
17. Hawker GA, Mian S, Kendzerska T and French M.
Measures of adult pain: Visual Analog Scale for Pain
(VAS Pain), Numeric Rating Scale for Pain (NRS Pain),
McGill Pain Questionnaire (MPQ), Short-Form McGill
Pain Questionnaire (SF-MPQ), Chronic Pain Grade Scale
(CPGS), Short Form-36 Bodily Pain Scale (SF-36 BPS),
and Measure of Intermittent and Constant Osteoarthritis
Pain (ICOAP). Arthritis Care Res 2011; 63 Suppl 11:
S240–252.
18. Ornetti P, Dougados M, Paternotte S, Logeart I and
Gossec L. Validation of a numerical rating scale to assess
functional impairment in hip and knee osteoarthritis:
Comparison with the WOMAC function scale. Annals
Rheumatic Dis 2011; 70: 740–746.
19. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J
and Stitt LW. Validation study of WOMAC: a health status
instrument for measuring clinically important patient rel-
evant outcomes to antirheumatic drug therapy in patients
with osteoarthritis of the hip or knee. J Rheumatology
1988; 15: 1833–1840.
20. Lequesne MG, Mery C, Samson M and Gerard P.
Indexes of severity for osteoarthritis of the hip and knee.
Validation—value in comparison with other assessment
tests. Scan J Rheumatology 1987; 65: 85–89.
21. Meenan RF, Gertman PM and Mason JH. Measuring
health status in arthritis. The arthritis impact measurement
scales. Arthritis Rheum 1980; 23: 146–152.
22. Maher CG, Sherrington C, Herbert RD, Moseley AM and
Elkins M. Reliability of the PEDro scale for rating qual-
ity of randomized controlled trials. Phys Ther 2003; 83:
713–721.
23. Higgins JPT GS. Cochrane Handbook for Systematic
Reviews of Interventions Version 5.1.0 [updated March
2011]. Available from www.cochrane-handbook.org2011.
24. Fukuda TY, Ovanessian V, Cunha RAD, et al. Pulsed
short wave effect in pain and function in patients with
knee osteoarthritis. J App Res 2008; 8: 189–198.
25. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA state-
ment for reporting systematic reviews and meta-analyses of
studies that evaluate health care interventions: Explanation
and elaboration. Ann Intern Med 2009; 151: W65–94.
26. Cetin N, Aytar A, Atalay A and Akman MN. Comparing
hot pack, short-wave diathermy, ultrasound, and TENS on
isokinetic strength, pain, and functional status of women
with osteoarthritic knees: A single-blind, randomized,
controlled trial. Am J Phys Med Rehabil 87: 443–451.
27. Callaghan MJ, Whittaker PE, Grimes S and Smith L. An
evaluation of pulsed shortwave on knee osteoarthritis
using radioleucoscintigraphy: A randomised, double blind,
controlled trial. Joint Bone Spine 2005; 72: 150–155.
28. Cantarini L, Leo G, Giannitti C, Cevenini G, Barberini P
and Fioravanti A. Therapeutic effect of spa therapy and
short wave therapy in knee osteoarthritis: A randomized,
single blind, controlled trial. Rheumatol Int 27: 523–529.
29. Laufer Y, Zilberman R, Porat R and Nahir AM. Effect
of pulsed short-wave diathermy on pain and function of
subjects with osteoarthritis of the knee: A placebo-con-
trolled double-blind clinical trial. Clin Rehabil 2005; 19:
255–263.
30. Moffett JA, Richardson PH, Frost H and Osborn A. A
placebo controlled double blind trial to evaluate the
effectiveness of pulsed short wave therapy for osteoar-
thritic hip and knee pain. Pain 67: 121–127.
31. Arya RK and Jain V. Osteoarthritis of the knee joint: An
overview. J Indian Academy Clin Med 2013; 14: 154–162.
32. Roos EM, Herzog W, Block JA and Bennell KL. Muscle
weakness, afferent sensory dysfunction and exercise in
knee osteoarthritis. Nature Rev Rheumatol 2011; 7: 57–63.
33. Bennell KL, Wrigley TV, Hunt MA, Lim B-W and
Hinman RS. Update on the role of muscle in the genesis
and management of knee osteoarthritis. Rheumatic Dis
Clinics N Am 2013; 39: 145–176.
34. Øiestad BE, Juhl CB, Eitzen I and Thorlund JB. Knee
extensor muscle weakness is a risk factor for development
of knee osteoarthritis. A systematic review and meta-
analysis. Osteoarth Cart 2015; 23: 171–177.