See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/328747960

Non-pharmacological and non-surgical interventions to manage patients with knee

osteoarthritis: An umbrella review
Article in Acta reumatologica portuguesa · July 2018

CITATIONS READS
15 696

3 authors:

Ricardo Ferreira José Alberto Duarte

Polytechnic Institute of Maia University of Porto

7 PUBLICATIONS 20 CITATIONS 415 PUBLICATIONS 8,783 CITATIONS

SEE PROFILE SEE PROFILE

Rui Soles Gonçalves

Polytechnic Institute of Coimbra - Coimbra Health School
56 PUBLICATIONS 308 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

"Proteomic study of salivary disorders induced by diabetes"- PTDC/QUI/72683/ 2006 View project

Implications of amphetamine psycostimulants abuse in brain aging View project

All content following this page was uploaded by Ricardo Ferreira on 09 November 2018.

The user has requested enhancement of the downloaded file.

 ARTIGO DE REVISÃO

Non-pharmacological and non-surgical interventions to

manage patients with knee osteoarthritis: an umbrella
review
Ferreira RM1,2, Duarte JA2, Gonçalves RS3,4

ACTA REUMATOL PORT. 2018;43:182-200

AbstrAct cacy from the systematic reviews to support their use.

Conclusion: Comparing to the last known umbrella
Objective: Update the last known umbrella review review, similar results were achieved on Acupuncture
and summarize the available high-quality evidence and Exercise interventions to improve the patients’ pain,
from systematic reviews on the effectiveness of non- stiffness, function and quality of life, but different results
phar-macological and non-surgical interventions for were found regarding the utilization of Trans-cutaneous
patients with knee osteoarthritis (OA). Electrical Nerve Stimulation and Low-Le vel Laser
Methods: The systematic reviews were identified Therapy as they do not improved the patients’ pain and
throught electronic databases, such as: MEDLINE, Em- physical function.
base, Physiotherapy Evidence Database (PEDro), The
Cochrane Library, SciELo, Science Direct, Google Keywords: Non-pharmacologic; Non-surgical;
Scholar, Research Gate and B-ON. The studies’ selec- Knee; Osteoarthritis
tion respected the following terms to guide the search
strategy using the P (humans with knee osteoarthritis) I
(non-pharmacological and non-surgical treatments) C IntroductIon
(pharmacological, surgical, placebo, no intervention, or
other non-pharmacological/ non-surgical conserva-tive Osteoarthritis (OA) is the most common form of arthri-tis
treatments) O (pain, functional status, stiffness, in- that can affect all the movable joint tissues and is a
flammation, quality of life and patient global assess- major contributor to functional and social impairment,
ment) model. disability, reduced independence and poorer quality of
Results: Following the PRISMA statement, 41 systema- life in older adults1-9. From all the joints that can be af-
tic reviews were found on the electronic databases that fected by OA the knee is the most prevalent (especial-ly
could be included in the umbrella review. After me- in elderly women), where a third of older adults in the
thodical analysis (R-AMSTAR), only 35 had sufficient general population show radiological evidence of knee
quality to be included. There is gold evidence that Stan- OA10-16. Moreover, there is an increasing need for
dard Exercise programs can reduce pain and improve attention to this disease due to the societal trends such
physical function in patients with knee OA. Additio - as ageing, obesity prevalence and joint injury, which are
nally, there is silver evidence for Acupuncture, Aqua tic estimated to increase the number of people affect-ed
Exercise, Electroacupuncture, Interferential Current, Ki- with OA by 50% over the next 20 years7,15,17,18.
nesio Taping, Manual Therapy, Moxibustion, Pulsed Knee OA evolution is highly variable, with the di sea-
Electromagnetic Fields, Tai Chi, Ultrasound, Yoga, and se improving in some patients, remaining stable in
Whole-Body Vibration. For other interventions, the others and gradually worsening in others19, 20. Treat-
quality of evidence is low or did not show sufficient effi - ment strategies for OA include pharmacological, non--
phar macological, surgical and non-surgical interven-
1. Physical Education and Sports Department, N2i, Institute
tions7,12,21-28. However, as the majority of the non--
Polytechnic of Maia; pharmacological and non-surgical interventions are
2. Faculty of Sport, CIAFEL, University of Porto; safe, low cost, low tech, incorporate self-management
3. Coimbra Health School, Physical Therapy Department,
Polytechnic Institute of Coimbra performed at home or in the community and have a
4. Centre for Health Studies and Research, University of Coimbra substantial public health impact, they are nowadays the

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

182
 Ferreira rM et al

first step in the knee OA management and play and B-ON. Papers were accepted only in English
a criti-cal role in its treatment 7,12,22, 25-29. and excluded if duplicated. The search begun in
In the last few years, evidence-based practice has May 2017 and finished in August 2017.
become increasingly popular. Evidence-based practice The studies’ selection respected the following terms
uses the available literature to guide clinical decision to guide the search strategy using the Population (P)
making and assess the strength of clinical recommen- (humans with knee osteoarthritis), Intervention (I) (non-
dations30,31. When diagnosing and treating patients, pharmacological and non-surgical treatments), Control
practitioners employ evidence, frequently from sys- (C) (pharmacological, surgical, placebo, no
tematic reviews of randomized controlled trials (RCTs), intervention, or other non-pharmacologi cal/non--
to advocate for or against an intervention30. Although surgical conservative treatments), Outcome (O) (pain,
systematic reviews summarize the effects of a specific functional status, stiffness, inflammation, quality of life
intervention for a specific condition, an umbrella re- and patient global assessment) model.
view typically assesses the quality, collate the results The keywords used in the search were:
and summarizes the evidence providing a wider pictu- “Knee”; “Os-teoarthr*”; “Gonarthr*”. These
re on the research topic32-39. In 2008 Jamtvedt et al.32, keywords were identi-fied after preliminary
published an umbrella review about knee OA, with very literature searches and by cross-checking them
useful results and conclusions that increase the against previous relevant systematic reviews.
knowledge and evidence-based practice, establishing An example of an online search strategy draft used
as well a platform for future investigation in this to pic. in MEDLINE database is presented: (Systematic Re-
However, it was shown that at least 10% of all sys- view[ptyp]) AND (“2007/01/01” [Pdat] : “2017/05/
tematic reviews need updating at the publication time /29”[Pdat]) AND (“humans”[MeSH Terms]) AND (En-
because of the length of time taken in preparing a sys- glish[lang]) AND (((“Knee”[All Fields]) AND (“Os-
tematic review and the accelerated pace of new evi- teoarthr*”[All Fields])) OR (“Gonarthr*”[All Fields]))
dence scientific production40. There is, to our know- Additional publications that were not found
ledge, no available updated umbrella review on the during the original database search were
effectiveness of non-pharmacological and non-surgical identified through manual searches in the
interventions for knee OA. Therefore, the aim of this related articles and reviews refe-rence lists.
umbrella review is to summarize and update the avai-
lable high-quality evidence from systematic reviews on study selectIon
the effectiveness of non-pharmacological and non-sur- In this paper, the two reviewers independently
gical interventions for patients with knee OA. screened the titles and abstracts yielded by the
search against the inclusion and exclusion
criteria and se - lected the potential studies. The
Methods inclusion and exclu-sion criteria applied to this
review are described in Table I.
There is, to our knowledge, no widely accepted The full versions of the systematic reviews that
guide-line to conduct an overview39. Therefore, in appea red to meet the inclusion criteria or where there
an attempt to ensure a high-quality study, this was any uncertainty were obtained. As the last known
overview was con-ducted following the PRISMA umbrella review was the Jamtvedt et al. 32 study, it was
(Preferred Reporting Items for Systematic reviews chosen to start the search for systematic reviews of
and Meta-Analyses) guidelines41. non-pharmacological and non-surgical treatments for
knee OA published in the electronic databases after
dAtA sources And seArches January 2007. Furthermore, due to biomechanical and
The literature search aimed to identify systematic re-views disease relationship, other systematic reviews explo-
that evaluated the effect of non-pharmacologi-cal and non- ring hip and knee OA were included, only if the re-sults
surgical treatments for knee OA. Systema-tic and from patients with knee OA could be extracted
comprehensive searches were conducted in electronic separately. The reviewers read the full text versions
databases: MEDLINE, Embase, Physiothera - py Evidence and decided whether they actually meet the inclusion
Database (PEDro), The Cochrane Library, SciELo, Science cri-teria. When insufficient data was presented, the au-
Direct, Google Scholar, Research Gate thors were contacted by email in order to request fur-

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

183
 NoN-pharMacological aNd NoN-surgical iNterveNtioNs to MaNage patieNts with kNee osteoarthritis: aN uMbrella review

tAble I. InclusIon And exclusIon crIterIA

Inclusion Exclusion

The systematic reviews must include: The systematic reviews cannot include:
at least one of the keywords; papers with experimental or control group composed by
any kind of animal;
papers with an intervention group that has primary papers with participants that do not have a knee OA
knee OA, either clinical or radiological criteria (healthy subjects) or who have secondary knee OA
(or a combination); (traumatic or post-surgical);
with or without meta-analysis exclusively from with or without meta-analysis of randomized
randomized controlled trials after January 2007; controlled trials prior to January 2007;
papers with non-pharmacological and non-surgical papers with exclusively pharmacological or surgical
interventions interventions;
peer-reviewed scientific literature journals; books, non-randomized controlled trials, case reports,
expert opinions, conference papers or academic thesis;
papers that evaluate pain or other knee-related symptoms, papers with subjects with other illnesses namely cancer,
functional status or quality of life; heart diseases, kidney diseases, neurological diseases,
respiratory diseases, rheumatoid arthritis, gout
arthritis, septic arthritis or Paget’s disease;
detailed description of the non-pharmacological papers with subjects exclusively with osteoarthritis in
and non-surgical intervention; the hip, foot, shoulder, elbow, wrist and/or fingers.
their full version, in English.

ther data. In case of study selection disparities, studies, design of primary studies, consistency,
the re-viewers reached an agreement through and di-rectness. An overall assessment of the
verbal discus-sion or arbitration. quality of evi-dence was based on a summary
of these 4 criteria, as presented in Table II.
dAtA extrActIon And QuAlIty AssessMent
For this review, the authors independently scored the dAtA syntheses And AnAlysIs
bias of the studies by using the R-AMSTAR (Revised A The data that was extracted from the selected publica-
MeaSurement Tool to Assess systematic Reviews) 11-- tions to assess the effectiveness of non-pharmacologi-
item questionnaire. In R-AMSTAR each domain’s score cal and non-surgical interventions included: title, au-
ranges between 1 (minimum) and 4 (maximum), and the thors’ name, year of publication, knee OA conditions,
total score has a range of 11 (minimum) to 44 (maxi- participants’ sample size and their characteristics, ob-
mum) that, posteriorly based on the overall score, can jectives, description of the interventions, description of
translate in A (high quality: 44-33 score), B (moderate the control groups, studies’ outcomes, assessment
quality: 32-23 score), C (low quality: 22-13 score) and D times, studies’ results and studies’ conclusions. Also,
(very low quality: 12-11 score) quality grade 42. Con- studies were combined using qualitative best evidence
sidering the recommendations that only total scores of synthesis. Considering the broad scope of clinical con-
23/44 are considered to have at least moderate metho - ditions, it was decided to restrict the work to pain,
dological quality, it was established as the cutting-point functional status, stiffness, inflammation, quality of life
for include a systematic review in this overview42. and patient global assessment45.
Furthermore, principles from GRADE (Grading of
Recommendations Assessment, Development, and
Evaluation) were used for an overall assessment and results
integration of the strength of the evidence for each in-
tervention43, 44. The GRADE concept is based on an as- selectIon of the studIes
sessment of the following criteria: quality of primary A set of 2188 records were identified through database

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

184
 Ferreira rM et al

tAble II. GrAdInG QuAlIty of evIdence

Level Criteria

High-quality evidence (A) One or more updated, high-quality systematic review

(Further research is very unlikely to change our that are based on at least 2 high-quality primary studies
confidence in the estimate of effect) with consistent results
Moderate-quality evidence (B) One or more updated systematic reviews of high or
(Further research is likely to have an important impact moderate quality
on our confidence in the estimate of effect and may • Based on at least 1 high-quality primary study
change the estimate) • Based on at least 2 primary studies of moderate quality
with consistent results
Low-quality evidence (C) One or more systematic reviews of variable quality
(Further research is very likely to have an important • Based on primary studies of moderate quality
impact on our confidence in the estimate of effect and • Based on inconsistent results in the reviews
is likely to change the estimate) • Based on inconsistent results in primary studies
Very low-quality evidence (D) No high-quality systematic review was identified on
(Any estimate of effect is very uncertain) this topic

searching. After the application of the inclusion and atic reviews, with an average of 1,490.06±
exclusion criteria, 41 articles have emerged 46-86. The 1,797.11 (maximum=8,21876; minimum=16583 per
se-lection process is summarized Figure 1. systematic review and 82.08±47.2 per RCT. In the
studies, sever-al non-pharmacological and non-
MethodoloGIcAl QuAlIty surgical treatments were found (Figure 2).
After the selection of the studies, the reviewers inde- Supplementary file Table IV provides a summary
pendently applied the R-AMSTAR to evaluate the of the included systematic reviews characteristics.
methodological quality of the 41 selected papers 46-86. After
this process, they reached an agreement through verbal
discussion or arbitration. The percentage of agreement for dIscussIon
individual items ranged from 36.4% to 100%. The
methodological quality assessment using the R-AMSTAR The discussion will be presented according to the
revealed a mean score of 32.7 (range 18 in-terventions of the selected systematic reviews.
– 40). At the end, 6 of the systematic reviews 46,54,57,
66,71,81
were excluded because they did not reach PhysIcAl ActIvIty
23/44, rai - sing the mean score to 34.9. The The physical component of 4 different activities
classifications obtai-ned are described in Table III. were investigated, such as Aquatic Exercise,
Standard Exer-cise, Tai Chi and Yoga.
study chArActerIstIcs Aquatic Exercise was only investigated in one
Overall, the 35 included systematic reviews47-53,55,56,58- study47. According to the authors47 this type of
65,67-70,72-80,82-86 were published from 200748 to 201759,62,80 exercise decreases pain and disability, and
and were conducted in America (Canada64), Asia (Chi- increases quality of life. However, this was only
found in short-term fol-low-up (12 weeks), but it could
na49,53,56,60-63,72,77,79,80,84-86, Japan73-75, Saudi Arabia83 and
South Korea70,78), Europe (Denmark47,55, England65,76, not be confirmed in long-term. Additionally, this data
France50,67, Germany58, Ireland52, Norway48 and was gathered in knee and hip OA and only
Switzerland68,69) and Oceania (Australia51,59,82).
moderate-evidence can support these statements.
The total RCTs included in the systematic reviews
Regarding land type Standard Exercise, it was ex-
were 571, with an average of 16.3±14.41 studies (ma -
plored in 10 systematic reviews50,51,55,63,67,73-76,82. High--
ximum=6076; minimum=452,60) per systema tic review.
evidence shows that Standard Exercise programs are
Overall, 52,152 subjects were enrolled in the system-
effective in pain and stiffness reduction and in increa -

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

185
NoN-pharMacological aNd NoN-surgical iNterveNtioNs to MaNage patieNts with kNee osteoarthritis: aN uMbrella review

Additional records identified Records identified through

throuhg bibliography databases electronic databases searching
(n=78) (n=2126)

Identification

Records after duplicates removed

(n=2188)
Screening

Records excluded, because:

• Without access to full-text or without
Eligibility
all the information needed (n=2033)
Records screened • Were not a systematic review of
(n=2188) randomized controlled trials (n=76)
• Not published in peer-review
Included journals (n=2)

Full-text articles excluded, because:

• Did not use a non-pharmacological
Full-text articles or non-surgical intervention (n=25)
assessed for eligibility • Include subjects with no or
(n=77) secondary knee osteoarthritis (n=9)
• Analyzed mix non-pharmacological
or non-surgical interventions (n=2)

Studies included in
(n=41)

fIGure 1. Results of the inclusion and exclusion criteria (flow diagram applied in this umbrella review)

sing function and quality of life51,55,63,74-76,82. Also, pain There was only one study58 that explored the re-
and disability reduction were found in moderate-quali-ty peated practice of Tai Chi as a method to treat patients
evidence50,73. All types of Standard Exercise programs with knee OA. Lauche et al. 58 perceived, from modera-
(aerobic, strengthening or resistance) were found to be te-evidence, that Tai Chi practice could decrease pain
beneficial in patients with knee OA. However, there are and stiffness, and increase function in these patients,
still some doubts between the low and high-intensity with the best results reached in patients that were inter-
training. Li et al.63 concluded from high-evidence that vened at least 2 times/week, 30 or more min/session.
high-intensity training program was more effective than Similarly, Kan et al.56 analyzed the effectiveness of
the low-intensity training, Regnaux et al.67 found no Yoga regular practice to treat these patients and saw
differences between low and high-intensity training positive effects on pain and mobility, for the most
programs from low to very low-evidence and Zacharias common Yoga protocols (40-90 min/session, every day
et al.82 have shown that both were beneficial to this type during at least 8 weeks).
of population, however there was high-evidence to Overall, Activities can be recommended to patients
support the low-intensity training program and low to with knee OA, especially aerobic, resistance, strengthe-
moderate-evidence to support the high-intensity trai - ning or combined programs to improve pain, stiffness,
ning program. function and quality of life, regardless of the patient’s

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

186
 Ferreira rM et al

tAble III. MethodoloGIcAl QuAlIty of elIGIble studIes (n = 41)

Study R-AMSTAR Items R-AMSTAR GRADE

(A to Z; Year) 1 2 3 4 5 6 7 8 9 10 11 Score (11 – 44) (A–D)
Bjordal et al. (48) 4 1 3 3 3 4 1 2 4 4 3 32 C
Lange et al. (57) 2 1 3 2 1 1 1 1 4 1 1 18
Rutjes et al. (69) 4 4 4 2 3 4 4 4 4 4 3 40 A
Rutjes et al. (68) 4 4 4 2 3 4 4 4 4 4 3 40 A
French et al. (52) 3 1 4 1 3 4 3 3 1 1 1 25 B
Cao et al. (49) 2 4 4 3 3 4 4 4 4 4 1 37 B
Silva et al. (71) 2 2 2 3 1 4 4 1 1 1 1 22
Lauche et al. (58) 4 4 4 2 4 4 4 4 2 2 3 37 B
Negm et al. (64) 4 2 4 2 3 4 2 2 4 4 3 34 D
Parkes et al. (65) 2 2 3 2 3 4 4 3 4 4 3 34 B
Tanaka et al. (73) 4 4 1 1 1 4 4 4 4 4 3 34 A
Uthman et al. (76) 3 4 4 3 3 1 3 1 4 4 1 31 B
We et al. (78) 1 4 4 2 3 4 4 4 4 4 1 35 A
Juhl et al. (55) 3 1 4 2 3 4 4 4 4 4 1 34 A
Quintrec et al. (66) 1 1 3 2 1 4 1 1 1 1 2 18
Tanaka et al. (74) 2 4 4 3 1 4 3 2 4 4 1 32 A
Ye et al. (81) 3 2 4 2 1 4 2 1 1 1 1 22
Zacharias et al. (82) 3 4 3 2 4 4 4 4 4 4 2 38 A
Zeng et al. (84) 2 4 4 2 3 4 2 2 4 4 2 33 B
Anwer et al. (46) 2 2 3 1 1 4 1 1 1 1 3 20
Fransen et al. (51) 4 4 4 2 3 4 4 4 4 4 3 40 A
Huang et al. (53) 3 2 3 2 3 4 3 2 4 4 2 32 B
Li et al. (61) 3 4 4 3 3 4 4 4 4 4 3 40 A
Regnaux et al. (67) 4 4 4 2 3 4 4 4 4 4 3 40 D
Tanaka et al. (75) 3 2 4 2 3 4 4 1 4 4 1 32 B
Wang et al. (77) 3 2 4 2 3 4 4 2 4 4 3 35 B
Zafar et al. (83) 2 4 2 2 4 4 4 3 4 4 1 34 B
Zeng et al. (85) 3 4 4 2 2 4 4 3 4 4 3 37 B
Zhang et al. (86) 3 4 4 3 3 4 3 3 4 4 3 38 B
Bartels et al. (47) 4 4 4 2 3 4 4 4 4 4 3 40 B
Coudeyre et al. (50) 3 2 2 2 3 4 3 2 4 4 3 32 B
Kan et al. (56) 3 4 2 2 3 4 3 1 1 1 2 26 B
Li et al. (60) 2 4 2 2 3 4 4 2 4 4 2 33 B
Li et al. (63) 3 4 2 2 3 4 3 3 4 4 3 35 A
Shim et al. (70) 4 4 3 3 3 4 4 4 4 4 1 38 B
Song et al. (72) 4 4 4 4 3 4 4 2 4 4 1 38 B
Xiang et al. (79) 3 3 3 2 3 3 3 1 4 4 1 30 A
Jorge et al. (54) 2 1 2 3 3 4 1 2 1 1 1 21
Lee et al. (59) 4 1 3 4 3 4 4 2 4 4 3 36 C
Li et al. (62) 4 4 3 1 3 4 4 4 4 4 1 36 B
Xu et al. (80) 2 1 4 4 3 4 3 2 4 4 1 32 B
Average 3 3 3.3 2.3 2.7 3.8 3.2 2.6 3.4 3.4 2 32.7 B

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

187
NoN-pharMacological aNd NoN-surgical iNterveNtioNs to MaNage patieNts with kNee osteoarthritis: aN uMbrella review

there is the need to have more evidence

Physical Activity (especially of high-quality).
Aquatic Exercise (47)
Standard Exercise (50, 51, 55, 63, 67, 73-76, 82) AcuPuncture
Tai Chi (58) From the Cao et al.49 systematic review and based on
Yoga (56) its high-quality evidence, pain and function can im-prove
Acupuncture (49) with needle Acupuncture in patients with knee OA.
Clinical Devices However, his recommendation cannot be fully achieved
Cupping Therapy (59) because the RCTs analyzed in the systematic review
Electrotherapy (85) used different acupuncture points and different
Electroacupuncture (70) protocols for treating these patients. Yet, generally the
Pulsed Electromagnet Fields (64, 78) best results were achieved following acupuncture points
Transcutaneous Electrical Nerve Stimulation (69) based on the Traditional Chinese Medicine meridian
Insoles (65) theory to treat the knee joint, known as the “Bi”
syndrome. These points consisted of 4 local points
Low-Level Laser Therapy (53)
(Yanglinquan [gall bladder 34], Yinlinquan [spleen 9],
Mudpack Therapy (79)
Zhusanli [stomach 36], Dubi [stomach 35]) and 4 dis-tal
Ultrasound (68, 84, 86)
points (Kunlun [urinary– bladder 60], Xuanzhong [gall
Whole-Body Vibration (61, 77, 83)
bladder 39], Sanyinjiao [spleen 6], and Taixi [kid-ney 3]),
Manual Therapy (52, 80)
done at least 2 times/week, 2 h/session. Yet, apart from
Moxibustion (60, 72)
being an invasive treatment and the lack of
Multimodal (48) standardization showed, it is imprudent to recommend
Kinesio Taping (62) its use based on just one systematic review. Conse-
quently, it is necessary to develop further high quality
fIGure 2. The non-pharmacological and non-surgical systematic reviews that assess this intervention.
interventions (n=35)

clInIcAl devIces
Cupping Therapy, Electrotherapy, Insoles, Low-Level
age, sex, BMI, radiographic status or baseline. This Laser Therapy, Mudpack Therapy, Ultrasound and
could be explained by the initial neuromuscular res- Whole-Body Vibration were the different interventions
ponse in an attempt to adapt to that specific exercise, approached in the included systematic reviews.
usually followed by muscular hypertrophy. Also, the The Cupping Therapy was only investigated in one
general feeling of well-being, reduced pain and greater study59. Although improvements in pain and physical
ability to perform tasks could be due to the gait con-trol function on patients with knee OA were found, name-ly
mechanisms or to the central release of endorphins. using a protocol of 10-20 min/session, 3-5 times/ /week,
Definitive conclusion on the best exercise program could this was only supported by weak-quality evi-dence.
not be achieved, because of the lack of differen - ces Thus, to recommend its use, further research (of high-
among several exercise interventions and the small quality) is needed to cease any uncertainty that this
number of included studies. Nevertheless, the studies’ intervention raised.
evidence pointed out more frequently that, among all, Electrotherapy was explored in 5 systematic re-
low-intensity isokinetic (concentric-eccentric) muscu-lar views64,69,70,78,85. Shim et al.70 studied the effects of Elec-
strengthening exercises (with special focus in the knee troacupuncture in patients with knee OA and saw im-
extensor muscles), done 3 or more times/week, with at provements in pain and quality of life from mo derate-
least 12 supervised sessions, could lead to faster and quality evidence. However, the Elec-troacupuncture
long lasting results. Aquatic Exercise, Tai Chi and Yoga, protocols used in the RCTs varied and comparing with
despite showing some impact on the knee OA patients’ sham Electroacupuncture the results were less
life, cannot be fully recommended because the evidence significant. Hence, despite the good results in patients
gathered was of moderate-quality data and from one with knee OA, its recommendation has to be carefully
systematic review from each method of treat-ment. taken, because they were reached mainly in moderate-
Therefore, for the treatments mentioned before quality RCTs, and they were less significant

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

188
 Ferreira rM et al

when comparing with sham Electroacupuncture. Also high-quality RCTs, that this therapy did not improve
this intervention is invasive, leading to need to redou- pain or function in patients with knee OA. Therefore,
ble care. So, further investigation on this intervention is following this B (R-AMSTAR – 32) quality conducted
needed to increase the quality of RCTs and try to build systematic review, the use of this therapy cannot be
a standard protocol of treatment. Another paper 69 re-commended. Simillary, only Xiang et al.79 studied
studied the effectiveness of Transcutaneous Nerve the impact of Mudpack Therapy in patients with knee
Stimu ationl in knee OA individuals and it could not be OA and did not find statistical significant differences
confirmed that this intervention is beneficial in pain be-tween the several experimental and control
reduction. This conclusion has been reached because groups in the functional and pain outcomes. So, it
only small trials of questionable-quality were included. does not exist sufficient support to recommend the
As this was an A (R-AMSTAR – 40) conducted-quality use of Mudpack Therapy in this population.
systematic review, we agree on the authors’ The effectiveness of Ultrasound in patients with knee
conclusions and cannot confirm their fully OA was evaluated in 3 systematic reviews68,84,86. Appa-
recommendations. Zeng et al.85 not only reached the rently, the continuous and, even more, the pulsed Ul-
same conclusion regarding the use of Transcutaneous trasound modes (especially the 1MHz, 2.5 W/cm2, 15
Nerve Stimulation, but also found similar results in the min/session, 3 session/week, during 8 weeks protocol)
use of Neuromuscular Elec-trical Stimulation, can be effective in the patients’ pain and physical func-
Noninvasive Interactive Neurosti-mulation and Pulsed tion. The raised hypothesis for this positive results is the
Electromagnetic Fields. The only Electrotherapy that chondrocyte proliferation and matrix production in
this systematic review group fully recommended, due to human articular cartilage. However, these findings are
the high-quality RCTs support, is the Interferential only supported by moderate to low quality RCTs. So,
Current, where the greatest signi-ficant differences due to the quality of the RCTs, some prudence is still
comparing with the control group were found. From all necessary to recommend its utilization in patients with
Interferential Current protocols, the most promising knee OA. The performance of new high-quality RCTs is
were those that used 100Hz du-ring 20 min, for 3-5 also waranted, in order to support the use of ul-trasound
sessions/week, through 4 weeks. However, Negm et in these patients.
al.64 and We et al.78 did not agree with the previous Finally, the Whole-Body Vibration therapy was ex-
author’s results regarding the utiliza-tion of Pulsed plored by Li et al.61, Wang et al.77 and Zafar et al.83, that
Electomagnetic Fields. Negm et al.64 found, from low to obtained very disperse results. Whole-body vibration
very-low quality RCTs, positive re-sults on knee OA exercises are a strength type exercises that uses vibra-
patient’s physical function but not on pain. Alternatively, tions generated by a vibrating plataform, in order to
We et al.78 confirms from high-qua-lity RCTs, not only stimulate muscles and tendons, by the contac of the
the increase in the patients’ physi-cal function but also human body with the vibrating surfaces 61,83. Li et al.61
in pain. Therefore, the use of Pulsed Electromagnetic saw that Whole-Body Vibration was not different com-
Fields in this population is still uncertain though, paring with other forms of exercises in pain, strength
tendentially, this intervention has shown to be a useful and self-reported status. Curiously, when added to
therapy (especially using a fre-quency of 5-12Hz for 30 squat exercise (namely 20 min session consisting in 6
min, during at least 3 weeks) based on the most recent to 9 reps per session of non-weight bearing squat, 3 sec
high quality systematic re-views and larger RCTs. of isometric at 60° knee flexion and 3 sec of isometric
Regarding the use of Insoles, it has only been stu flexion at 10°; plus the vibration plate with frequency of
died by Parkes et al.65. Although the overview of all 35 to 40Hz, 20 to 70 s, amplitude of 4 mm, and ac-
studies showed a statistically significant difference celeration that ranged from 2.78 to 3.26G – 3 times per
between the use of Lateral Wedges and decreased pain week) it was more efficacious when compared with
in medial knee OA, when comparing these insoles with squat exercises alone, implying that this machine/exer-
neutral insoles no statistical significant or clinically cise can be a good complement to more usual and well
important association were established. Thus, with these established exercises. Although Zafar et al.83 also
B (R-AMSTAR – 34) quality findings, the use of Late ral reached the conclusion that this exercise can be bene-
Wedges for these patients cannot be supported. Simi - ficial to the patients’ pain and function, Wang et al.77
larly, Low-Level Laser Therapy was only studied by one only observed positive effects in function, but not in
group53 wich reached the conclusion, by analyzing pain, raising more questions than answers regarding

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

189
NoN-pharMacological aNd NoN-surgical iNterveNtioNs to MaNage patieNts with kNee osteoarthritis: aN uMbrella review

this recent and not well known topic. Thus, besides tion in quality of life, pain and physical function. The
being safe (only requires the person to be over the mechanisms that can explain this positive effects might
vi-brating plate) and a good supplement to exercise be the generated heat and the stimulation of some spe-
(es-pecially, in increasing function), there is still cific acupoints that: 1) acts through the local system
some un-certainty about the true influence of this neural network and releases some neurotransmission,
machine on the patients with knee OA. such as opioidergics, b-endorphins, and adenosine
Overall, as explored above, there is some uncertain-ty triphosphate; 2) modulates the inflammatory reactions
about the effect of clinical devices. It seems that the best through the degranulation of local mastocytes and acti-
clinical devices used to manage knee OA patients are vation of thermoreceptors; 3) this enhanced activation of
Electroacupuncture, Interferential Current, Pulsed Elec- the thermoreceptors could also lead to a decrease of
tromagnetic Fields, Ultrasound and Whole-Body Vibra-tion. the nociceptive painful transmission. The acupoints that
On the other hand, Transcutaneous Nerve Sti-mulation, were more commonly used and that obtai ned posi-tive
Neuromuscular Electrical Stimulation, Nonin vasive results more often were similar to those used in the
Interactive Neurostimulation, Insoles, Low-Level Laser acupuncture interventions, such as the gall bladder 34,
Therapy and Mudpack Therapy either did not show spleen 9 and 10, stomach 34, 35 and 36, and the 2, 4
sufficient efficacy, or did not gather enough high-quality and 5 extra lower extremity points, done at least 3
RCTs support, or even did not reach ho-mogeneous times/week, 15-20 min/session. With the moderate to
results and cannot be recommended. high-quality evidence gathered in these systematic re-
views, it seems that Moxibustion can be a relatively safe
MAnuAl therAPy intervention (only skin flushing was observed, howe-ver
From the available systematic reviews52,80 based on it disappeared within 3 days) and a viable alterna-tive to
moderate-quality evidence, Manual Therapy can be usual care on the knee OA patients’ health mana-
recommended to treat patients with knee OA, because gement.
it can improve the pain, stiffness and physical function.
But, once again, the protocols used and the interven- KInesIo tAPInG
tions varied a lot: Swedish Massage, usual Mobiliza- A recent systematic review, conducted by Li et al.62 eva -
tion, Maitland, Acupressure, Tui Na, Shi Manipulation, luated the effects of the elastic therapeutic bandages in
usual Manipulation and Myofascial Mobilization. This patients with knee OA. In this systematic review, based
could be due to the experience of the treatment provider in moderate-quality data, it was found improvements in
and the individual clinical presentation, where usually pain, flexibility, proprioception and knee-related health
the intervener tries to adapt his treatment (type of status. However, identical results were achived in
intervention, dosage, force, amplitude, rate, repeti-tion strength between the experimental and the control group
and duration) according to the patient’s case and were found. The included RCTs had different ex-
severity. Nevertheless, the intervention that has shown perimental protocols, such as the type of cut (I-strip, Y-
more consistent positive results was Massage (one of strip or the combination of several strips), direction
the most widely used intervention in this type of po- (center extremities or extremities center), patient pre
pulation). Yet, since most measure instruments are paration (some placed the knee in full flexion, others the
self--reported and subjective, sometimes a single touch knee in semi-flexion; some paid attention to the
on the patient may induce him or her to report better re- presence of sweat and hair removing, others did not)
sults. Therefore, a placebo effect cannot be ruled out. and tension (from 10 to 75%) of the Kinesio Tape.
Consequently, although Manual Therapy is a safe and However, the results obtained were, in general, similar.
economic way to treat patients with knee OA, more This may indicate that it is not the application that
high-quality RCTs and a better exploration of their counts, but rather the effect that the Kinesio Tape can
methods are needed, in an attempt to bring up more induce in patients. Since Kinesio Tape fibers are ma-
standard protocols and improve its recommendation. nufactured with a wave-like pattern, convolutions will
appear after the recoil, creating more space between the
MoxIbustIon tissues and providing a tactile input through the skin,
Moxibustion treatments were analyzed in two systema- stimulating type 2 cutaneous mechanoreceptors. This
tic reviews60, 72. Both showed that the Moxibustion in- may result in an improved fluid exchange, mus-cle
tervention was superior to control and sham Moxibus- function re-education and pain reduction. Although

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

190
 Ferreira rM et al

to recommend its use, some precautions are needed -quality RCTs and systematic reviews to satisfy
because this conclusion is based on just one study with the raised uncertainties.
medium-quality RCTs. Therefore, more information The study had the limitation of only accepted
needs to be gathered for recommending its use. RCTs in English. If the search had be done in other
languages, more studies (possibility of high quality)
MultIModAl and more in-terventions probability could have been
Usually, physiotherapy interventions are multimodal, included, hel-ping in overview of the study aim.
where a great variety of interventions, especially non--
surgical and non-pharmacological are used. Bjordal et corresPondence to
48 Ricardo Luis de Almeida Maia
al. analyzed the effectiveness of different treatments
Ferreira Avenida Carlos de Oliveira
in knee OA patients. In this systematic review, Bjordal Campos, Castêlo da Maia, Porto E-
and his group reached the conclusion that the best mail: rferreira@ipmaia.pt
treat-ments to these patients are the Transcutaneous
Electri-cal Nerve Stimulation, Electroacupuncture and references
Low-Level Laser Therapy. On the other hand, Manual 1. Neogi T, Zhang Y. Epidemiology of osteoarthritis. Rheumatic
Disease Clinics of North America. 2013;39(1):1-19.
Acupuncture, Pulsed Electromagnetic Field and Static
2. Baum T, Joseph G, Karampinos D, Jungmann P, Link T, Bauer
Magnets did not show significant statistical differences J. Cartilage and meniscal T2 relaxation time as non-invasive
compared to the control groups. Conversely, these con- biomarker for knee osteoarthritis and cartilage repair proce-
clusions were only supported by this umbrella review in dures. Osteoarthritis and Cartilage. 2013;21(10):1474-1484.
the Electroacunpuncture intervention. Regarding the 3. Zhang S. Recent changes in evidence-based, non-
pharmaco-logical treatment recommendations for
Transcutaneous Nerve Stimulation, Low-Level Laser
acupuncture and Tai Chi for knee osteoarthritis. Journal
Therapy and Pulsed Electromagnetic Field, the results of Sport and Health Science. 2013;2(3):158-159.
we found were exactly the opposite, maybe due to the 4. Stevenson JD, Roach R. The benefits and barriers to
higher among of information gathered with this type of physical activity and lifestyle interventions for
study. osteoarthritis affecting the adult knee. Journal of
Orthopaedic Surgery and Research. 2012;7(15):1-7.
5. Finan PH, Buenaver LF, Bounds SC, Hussain S, Park RJ,
Haque UJ, et al. Discordance between pain and
conclusIon radiographic severity in knee osteoarthritis: Findings from
quantitative sensory test-ing of central sensitization.
Arthritis & Rheumatology. 2013;65(2):363-372.
In conclusion, based on the included systematic re-
6. Felson DT, Lawrence RC, Dieppe PA, Hirsch R,
views, there is good evidence that Standard Exercise Helmick CG, Jordan JM, et al. Osteoarthritis: New
programs can reduce pain and improve physical func- insights. Part 1: The dis-ease and its risk factors.
tion in patients with knee OA. Additionally, there is Annals of Internal Medicine. 2000;133(8):635-646.
moderate evidence that Acupuncture, Aquatic Exer-cise, 7. Sakalauskiene G, Jauniskiene D. Osteoarthritis:
etiology, epi-demiology, impact on the individual and
Electroacupuncture, Interferential Current, Kine-sio
society and the main principles of management.
Taping, Manual Therapy, Moxibustion, Pulsed Elec- Medicina (Kaunas). 2010;46(11): 790-797.
tromagnetic Fields, Tai Chi, Ultrasound, Yoga and 8. Martel-Pelletier J, Pelletier J-P. Is osteoarthritis a
Whole-Body Vibration (more as a complement of exer - disease invol-ving only cartilage or other articular
tissues? Eklem Hastalik Cerrahisi. 2010;21(1):2-14.
cise than a single intervention) are effective in the res -
9. Felson DT, Hodgson R. Identifying and treating
pective evaluated outcomes. For other interventions, the preclinical and early osteoarthritis. Rheumatic Disease
quality of evidence is low or does not show suffi-cient Clinics of North Ameri-ca. 2014;40(4):699-710.
efficacy from the systematic reviews to support its use. 10. Peat G, McCarney R, Croft P. Knee pain and
osteoarthritis in older adults: A review of community
Furthermore, comparing to Jamtvedt et al. 32 umbrella
burden and current use of primary health care. Annals
review, we confirmed the worthy use of Acupuncture of the Rheumatic Diseases. 2001;60(2):91-97.
and Exercise but, according to our se-lected systematic 11. McAlindon T, Snow S, Cooper C, Dieppe P. Radiographic
reviews, we found different results re-garding the use pat-terns of osteoarthritis of the knee joint in the
Transcutaneous Electrical Nerve Stim-ulation and Low- community: The importance of the patellofemoral joint.
Annals of the Rheuma-tic Diseases. 1992;51(7):844-849.
Level Laser Therapy to reduce the patients’ pain and
12. Michael J, Schlüter-Brust KU, Eysel P. The epidemiology,
improve physical function. Addi-tionally, there is an etiolo-gy, diagnosis, and treatment of osteoarthritis of the knee.
urgent need to develop new high- Deutsches Ärzteblatt International. 2010;107(9):152-162.

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

191
NoN-pharMacological aNd NoN-surgical iNterveNtioNs to MaNage patieNts with kNee osteoarthritis: aN uMbrella review

13. Pereira D, Peleteiro B, Araujo J, Branco J, Santos R, 31. Ernst E, Pittler MH. Assessment of therapeutic safety in
Ramos E. The effect of osteoarthritis definition on sys-tematic reviews: Literature review. British Medical
prevalence and incidence es-timates: A systematic review. Journal. 2001;323(7312):546-547.
Osteoarthritis and Cartilage. 2011;19(11):1270-1285. 32. Jamtvedt G, Dahm KT, Christie A, Moe RH, Haavardsholm
14. Hannan MT, Anderson JJ, Zhang Y, Levy D, Felson E, Holm I, et al. Physical therapy interventions for patients
DT. Bone mineral density and knee osteoarthritis in with osteoarthritis of the knee: An overview of systematic
elderly men and women. The Framingham Study. reviews. Physical Therapy. 2008;88(1):123-136.
Arthritis & Rheumatology. 1993;36(12):1671-1680. 33. Ioannidis JP. Integration of evidence from multiple meta-ana-
15. Alnahdi AH, Zeni JA, Snyder-Mackler L. Muscle lyses: A primer on umbrella reviews, treatment networks and
impairments in patients with knee osteoarthritis. Sports multiple treatments meta-analyses. Canadian Medical Associa-
Health. 2012;4(4):284--292. tion Journal. 2009;181(8):488-493.
16. Thomas E, Peat G, Croft P. Defining and mapping the 34. Higgins JP, Green S. Cochrane handbook for systematic
person with osteoarthritis for population studies and reviews of interventions: John Wiley & Sons; 2011.
public health. Rheumatology. 2013(53):338-345. 35. Aromataris E, Fernandez R, Godfrey CM, Holly C, Khalil H,
17. Anderson AS, Loeser RF. Why is osteoarthritis an age- Tungpunkom P. Summarizing systematic reviews:
related disease? Best Practice & Research Clinical Methodolog-ical development, conduct and reporting of an
Rheumatology. 2010;24(1):15-26. umbrella review approach. International Journal of
18. Golightly YM, Allen KD, Caine DJ. A comprehensive Evidence-based Healthcare. 2015;13(3):132-140.
review of the effectiveness of different exercise 36. Pollock M, Fernandes RM, Hartling L. Evaluation of AMSTAR
programs for patients with osteoarthritis. The Physician to assess the methodological quality of systematic reviews in
and Sportsmedicine. 2012;40(4): 52-65. overviews of reviews of healthcare interventions. British Medi-
19. Felson DT. Osteoarthritis of the knee. New England cal Council Medical Research Methodology. 2017;17(48):1-13.
Journal of Medicine. 2006;354(8):841-848. 37. McKenzie JE, Brennan SE. Overviews of systematic
20. Conditions NCCfC. Osteoarthritis: national clinical guidelines reviews: great promise, greater challenge. Systematic
for care and management in adults: Royal College of Reviews. 2017;6(2017):185-189.
Physicians; 2008. 38. Caird J, Sutcliffe K, Kwan I, Dickson K, Thomas J. Mediating
21. Felson DT, Lawrence RC, Hochberg MC, McAlindon T, policy-relevant evidence at speed: Are systematic reviews of
Dieppe PA, Minor MA, et al. Osteoarthritis: New sys-tematic reviews a useful approach? Evidence & Policy: A
insights. Part 2: Treat-ment approaches. Annals of Jour-nal of Research, Debate and Practice. 2015;11(1):81-97.
Internal Medicine. 2000;133(9): 726-737. 39. Hartling L, Chisholm A, Thomson D, Dryden DM. A descrip-tive
22. Nelson AE, Allen KD, Golightly YM, Goode AP, Jordan JM, ed- analysis of overviews of reviews published between 2000 and
itors. A systematic review of recommendations and guidelines 2011. Public Library of Science One. 2012;7(11):e49667.
for the management of osteoarthritis: The chronic osteoarthri- 40. Shojania KG, Sampson M, Ansari MT, Ji J, Doucette S, Moher D.
tis management Initiative of the US bone and joint initiative. How quickly do systematic reviews go out of date? A survival
Seminars in Arthritis and Rheumatism; 2014: Elsevier. analysis. Annals of Internal Medicine. 2007;147(4):224-233.
23. Ringdahl E, Pandit S. Treatment of knee osteoarthritis. 41. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting
Ameri-can Family Physician. 2011;83(11):1286-1292. items for systematic reviews and meta-analyses: The PRISMA
24. Scott D, Kowalczyk A. Osteoarthritis of the knee. Britsh statement. Annals of Internal Medicine. 2009;151(4):264-269.
Medi-cal Journal Clinical Evidence. 2007;2007(9):1-28. 42. Kung J, Chiappelli F, Cajulis OO, Avezova R, Kossan G, Chew L, et
25. Scott D, Shipley M, Dawson A, Edwards S, Symmons D, Woolf al. From systematic reviews to clinical recommendations for
A. The clinical management of rheumatoid arthritis and os- evidence-based health care: validation of revised assessment of
teoarthritis: Strategies for improving clinical effectiveness. multiple systematic reviews (R-AMSTAR) for grading of clini-cal
British Journal of Rheumatology. 1998;37(5):546-554. relevance. The open dentistry journal. 2010;4:84-91.
26. Sinkov V, Cymet T. Osteoarthritis: Understanding the 43. Brozek J, Akl E, Alonso-Coello P, Lang D, Jaeschke R, Williams J,
patho-physiology, genetics, and treatments. Journal of et al. Grading quality of evidence and strength of recommen-
the National Medical Association. 2003;95(6):475-482. dations in clinical practice guidelines. Allergy. 2009;64(5):669.
27. Sinusas K. Osteoarthritis: Diagnosis and treatment. 44. Guyatt GH, Oxman AD, Kunz R, Falck-Ytter Y, Vist GE,
American Family Physician. 2012;85(1):49-56. Liberati A, et al. Rating quality of evidence and strength of
28. Bruyère O, Cooper C, Pelletier J-P, Branco J, Brandi ML, recommen-dations: Going from evidence to recommendations.
Guillemin F, et al., editors. An algorithm recommendation BMJ: British Medical Journal. 2008;336(7652):1049.
for the management of knee osteoarthritis in Europe and 45. Bellamy N, Kirwan J, Boers M, Brooks P, Strand V, Tugwell P,
interna-tionally: A report from a task force of the European et al. Recommendations for a core set of outcome measures
Society for Clinical and Economic Aspects of Osteoporosis for fu-ture phase III clinical trials in knee, hip, and hand
and Os-teoarthritis (ESCEO). Seminars in Arthritis and osteoarthri-tis. Consensus development at OMERACT III. The
Rheumatism; 2014: Elsevier. Journal of Rheumatology. 1997;24(4):799-802.
29. Dziedzic KS, Hill JC, Porcheret M, Croft PR. New 46. Anwer S, Alghadir A, Zafar H, Al-Eisa E. Effect of whole
models for pri-mary care are needed for osteoarthritis. body vibration training on quadriceps muscle strength in
Physical Therapy. 2009;89(12):1371-1378. indivi duals with knee osteoarthritis: A systematic review
30. Harris JD, Quatman CE, Manring M, Siston RA, Flanigan and meta-ana ysisl. Physiotherapy. 2015;102(2):145-151.
DC. How to write a systematic review. The American 47. Bartels EM, Lund H, Hagen KB, Dagfinrud H, Christensen R,
Journal of Sports Medicine. 2014;42(11):2761-2768. Danneskiold-Samsøe B. Aquatic exercise for the treatment of

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

192
 Ferreira rM et al

knee and hip osteoarthritis. Cochrane Database of 2016;30(10):947-959.

Systematic Reviews. 2016;4(3):1-67. 64. Negm A, Lorbergs A, Macintyre N. Efficacy of low frequency
48. Bjordal JM, Johnson MI, Lopes-Martins RA, Bogen B, Chow R, pulsed subsensory threshold electrical stimulation vs placebo
Ljunggren AE. Short-term efficacy of physical interventions in on pain and physical function in people with knee osteoarthritis:
osteoarthritic knee pain. A systematic review and meta-analy- Systematic review with meta-analysis. Osteoarthritis and Car-
sis of randomised placebo-controlled trials. BioMed Central tilage. 2013;21(9):1281-1289.
Musculoskeletal Disorders. 2007;8(1):51-65. 65. Parkes MJ, Maricar N, Lunt M, LaValley MP, Jones RK,
49. Cao L, Zhang X-L, Gao Y-S, Jiang Y. Needle acupuncture for Segal NA, et al. Lateral wedge insoles as a
os-teoarthritis of the knee. A systematic review and updated conservative treatment for pain in patients with medial
meta--analysis. Saudi Medical Journal. 2012;33(5):526-532. knee osteoarthritis: A meta-analy-sis. Journal of the
50. Coudeyre E, Jegu A, Giustanini M, Marrel J, Edouard American Medical Association. 2013;310(7):722-730.
P, Pereira B. Isokinetic muscle strengthening for knee 66. Quintrec J, Verlhac B, Cadet C, Bréville P, Vetel JM, Gauvain
osteoarthritis: A systematic review of randomized JB, et al. Physical exercise and weight loss for hip and knee
controlled trials with meta--analysis. Annals of Physical os-teoarthritis in very old patients: A systematic review of the
and Rehabilitation Medicine. 2016;59(3):207-215. lit-erature. Open Rheumatoly Journal. 2014;8:89-95.
51. Fransen M, McConnell S, Harmer AR, Van der Esch M, Simic 67. Regnaux J-P, Trinquart L, Boutron I, Nguyen C, Brosseau L,
M, Bennell KL. Exercise for osteoarthritis of the knee. Ravaud P. High-intensity versus low-intensity physical activity
Cochrane Database Systematic Reviews. 2015;4(1):1-144. or exercise in patients with hip or knee osteoarthritis. Cochrane
52. French H, Brennan A, White B, Cusack T. Manual therapy Database of Systematic Reviews. 2015(10):1-63.
for osteoarthritis of the hip or knee – A systematic review. 68. Rutjes AW, Nüesch E, Sterchi R, Jüni P. Therapeutic
Manual Therapy. 2011;16(2):109-117. ultrasound for osteoarthritis of the knee or hip. The
53. Huang Z, Chen J, Ma J, Shen B, Pei F, Kraus V. Effectiveness Cochrane Library. 2010(1):1-43.
of low-level laser therapy in patients with knee osteoarthritis: A 69. Rutjes AW, Nüesch E, Sterchi R, Kalichman L, Hendriks E,
systematic review and meta-analysis. Osteoarthritis and Carti- Osiri M, et al. Transcutaneous electrostimulation for
lage. 2015;23(9):1437-1444. osteoarthritis of the knee. The Cochrane Library. 2009(4):1-81.
54. Jorge MSG, Zanin C, Knob B, Wibelinger LM. Effects 70. Shim J-W, Jung J-Y, Kim S-S. Effects of
of deep heating to treat osteoarthritis pain: Systematic electroacupuncture for knee osteoarthritis: A systematic
review. Revista Dor. 2017;18(1):79-84. review and meta-analysis. Evi-dence-Based
55. Juhl C, Christensen R, Roos EM, Zhang W, Lund H. Impact of Complementary and Alternative Medicine. 2016;2016:1-18.
exercise type and dose pain and disability in knee osteoarthri- 71. Silva A, Serrão P, Driusso P, Mattiello S. The effects of
tis: A systematic review and meta regression analysis of ran- thera-peutic exercise on the balance of women with
domized controlled trials. Arthritis & Rheumatology. knee os-teoarthritis: A systematic review. Osteoarthritis
2014;66(3):622-636. and Cartilage. 2012;16(1):1-9.
56 Kan L, Zhang J, Yang Y, Wang P. The effects of yoga on pain, mo- 72. Song G-M, Tian X, Jin Y-H, Deng Y-H, Zhang H, Pang
bility, and quality of life in patients with knee osteoarthritis: A X-L, et al. Moxibustion is an alternative in treating knee
systematic review. Evidence-Based Complementary and Alter- osteoarthritis: The evidence from systematic review and
native Medicine. 2016;2016:1-10. meta-analysis. Medicine. 2016;95(6):1-11.
57. Lange AK, Vanwanseele B. Strength training for treatment 73. Tanaka R, Ozawa J, Kito N, Moriyama H. Efficacy of
of os-teoarthritis of the knee: A systematic review. Arthritis strength-ening or aerobic exercise on pain relief in
Care & Re-search. 2008;59(10):1488-1494. people with knee os-teoarthritis: A systematic review
58. Lauche R, Langhorst J, Dobos G, Cramer H. A systematic review and meta-analysis of rando-mized controlled trials.
and meta-analysis of tai chi for osteoarthritis of the knee. Com- Clinical Rehabilitation. 2013;27(12): 1059-1071.
plementary Therapies in Medicine. 2013;21(4):396-406. 74. Tanaka R, Ozawa J, Kito N, Moriyama H. Effect of the frequen-
cy and duration of land-based therapeutic exercise on pain re-
59. Lee EY, Kim J-E, Lee KK, Wang Y. Cupping therapy
lief for people with knee osteoarthritis: A systematic review and
for treating knee osteoarthritis: The evidence from
systematic review and meta-analysis. Complementary meta-analysis of randomized controlled trials. Journal of Phy -
Therapies in Clinical Practice. 2017;28(2017):152-160. sical Therapy Science. 2014;26(7):969-975.
60. Li A, Wei Z-J, Liu Y, Li B, Guo X, Feng S-Q. Moxibustion 75. Tanaka R, Ozawa J, Kito N, Moriyama H. Does exercise
treat-ment for knee osteoarthritis: A systematic review and thera-py improve the health-related quality of life of people
meta-analy sis. Medicine. 2016;95(14):1-9. with knee osteoarthritis? A systematic review and meta-
61. Li X, Wang X-Q, Chen B-L, Huang L-Y, Liu Y. Whole-body vi- analysis of ran-domized controlled trials. Journal of
bration exercise for knee osteoarthritis: A systematic review Physical Therapy Science. 2015;27(10):3309-3314.
and meta-analysis. Evidence-Based Complementary and 76. Uthman OA, van der Windt DA, Jordan JL, Dziedzic KS, Healey EL,
Alternative Medicine. 2015;2015:1-11. Peat GM, et al. Exercise for lower limb osteoarthritis: Sys-tematic
review incorporating trial sequential analysis and net-work meta-
62. Li X, Zhou X, Howe Liu NC, Liang J, Yang X, Zhao G,
analysis. Briths Medical Journal. 2013;347:1-13.
et al. Ef-fects of elastic therapeutic taping on knee
77. Wang P, Yang X, Yang Y, Yang L, Zhou Y, Liu C, et al. Effects of
osteoarthritis: A sys-tematic review and meta-analysis.
whole body vibration on pain, stiffness and physical functions in
Aging and Disease. 2017;8(6):1-13.
63. Li Y, Su Y, Chen S, Zhang Y, Zhang Z, Liu C, et al. The effects patients with knee osteoarthritis: A systematic review and meta-
analysis. Clinical Rehabilitation. 2015;29(10):939-951.
of resistance exercise in patients with knee osteoarthritis: A
sys-tematic review and meta-analysis. Clinical Rehabilitation. 78. We SR, Koog YH, Jeong K-I, Wi H. Effects of pulsed electro-

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

193
NoN-pharMacological aNd NoN-surgical iNterveNtioNs to MaNage patieNts with kNee osteoarthritis: aN uMbrella review

magnetic field on knee osteoarthritis: A systematic review. 83. Zafar H, Alghadir A, Anwer S, Al-Eisa E. Therapeutic effects of
Rheumatology. 2013;52(5):815-824. whole-body vibration training in knee osteoarthritis: A syste-
79. Xiang J, Wu D, Li Ja. Clinical efficacy of mudpack therapy matic review and meta-analysis. Archives of Physical Medicine
in treating knee osteoarthritis: A meta-analysis of and Rehabilitation. 2015;96(8):1525-1532.
randomized con-trolled studies. American Journal of 84. Zeng C, Li H, Yang T, Deng Z-h, Yang Y, Zhang Y, et
Physical Medicine & Re-habilitation. 2016;95(2):121-131. al. Effec-tiveness of continuous and pulsed ultrasound
80. Xu Q, Chen B, Wang Y, Wang X, Han D, Ding D, et al. The ef- for the manage-ment of knee osteoarthritis: A
fectiveness of manual therapy for relieving pain, stiffness, and systematic review and network meta-analysis.
dysfunction in knee osteoarthritis: A systematic review and Osteoarthritis and Cartilage. 2014;22(8):1090--1099.
meta-analysis. Pain Physician. 2017;20(4):229-243. 85. Zeng C, Yang T, Deng Z-h, Yang Y, Zhang Y, Lei G-h.
81. Ye J, Cai S, Zhong W, Cai S, Zheng Q. Effects of tai chi for pa- Electrical stimulation for pain relief in knee osteoarthritis:
tients with knee osteoarthritis: A systematic review. Journal of Systematic re-view and network meta-analysis.
Physical Therapy Science. 2014;26(7):1133-1137. Osteoarthritis and Cartilage. 2015;23(2):189-202.
82. Zacharias A, Green RA, Semciw A, Kingsley M, Pizzari 86. Zhang C, Xie Y, Luo X, Ji Q, Lu C, He C, et al. Effects
T. Effi-cacy of rehabilitation programs for improving of thera-peutic ultrasound on pain, physical functions
muscle strength in people with hip or knee and safety out-comes in patients with knee
osteoarthritis: A systematic review with meta-analysis. osteoarthritis: A systematic review and meta-analysis.
Osteoarthritis and Cartilage. 2014;22(11): 1752-73. Clinical Rehabilitation. 2015;30(10):960--971.

ÓRGÃO OFICIAL DA SOCIEDADE PORTUGUESA DE REUMATOLOGIA

194
 suPPleMentAry fIle tAble Iv. systeMAtIc revIews suMMArIes (n=35)

Authors N° of inclued
Interventions (A to Z; year) Objectives RCTs (subjects) Results/Conclusions
Activities (physical)
Aquatic Bartels et al. To evaluate the effects of aquatic exercise 13 (n=1190) Based upon moderate quality evidence, aquatic exercise
Exercise (47) for people with knee or hip OA, or both, has beneficial effects on knee OA people. A small but
compared to no intervention. clinically relevant decrease in pain and disability, and
small but clinically relevant increase in QOL. There is a
small short-term effect on the knee OA people at the end
PORTUGUESASOCIEDADEDAOFICIALÓRGÃO REUMATOLOGIADE

of an aquatic training program. The long-term effect is

unclear due to the paucity of studies.
Standard Tanaka et al. To investigate the differences in the efficacies 8 (n=466) Moderate-evidence show that muscle strengthening
Exercise (73) between strengthening and aerobic exercises exercises with or without weight-bearing and aerobic
for pain relief in knee OA patients. exercises are effective for pain relief in knee OA people.
In particular, for pain relief by short-term exercise

Ferreira rM et al
intervention, the most effective exercise is
non-weight–bearing strengthening exercise.
Uthman et al. To determine if exercise interventions are 60 (n=8218) Significant benefits of exercise over no exercise in OA
195

(76) more effective than no exercise control and patients were showed. An approach combining exercises
to compare the effectiveness of different to increase strength, flexibility and aerobic capacity is
exercise interventions in relieving pain and likely to be most effective in the management of lower
improving function in lower limb OA limb OA.
patients.
Juhl et al. To identify the optimal exercise program, 48 (n=4028) All exercise (aerobic, resistance, performance and mixed)
(55) characterized by type and intensity of types are beneficial in reducing pain in knee OA patients.
exercise, length of program, duration of Therefore, optimal exercise programs for knee OA
individual supervised sessions, and number should focus on improving aerobic capacity, quadriceps
of sessions per week, for reducing pain and muscle strength and lower extremity performance. For
patient-reported disability in knee OA. best results, the program should be supervised and
carried out 3 x per week.
continues on the next page
 NoN-pharMacological aNd NoN-surgical iNterveNtioNs to MaNage patieNts with kNee osteoarthritis: aN uMbrella review
tAble Iv. contInuAtIon
Authors N° of inclued

Interventions (A to Z; year) Objectives RCTs (subjects) Results/Conclusions

Tanaka et al. To investigate the influence of land-based 17 (n=1816) Although the effect size did not increase over the time,
(74) exercise frequency and duration on pain continuous strengthening exercise or aerobic exercise
relief for knee OA people. intervention (especially ≥4x per week) had a significant
effect on knee pain of knee OA people after 9 weeks.
Zacharias et al. To analyze the effect of exercise-based 40 (n=3989) Both exercises (high and low-intensity) interventions at
(82) rehabilitation programs for improving lower short-term follow-up are beneficial for strength outcomes
PORTUGUESASOCIEDADEDAOFICIALÓRGÃO

limb muscle strength in hip or knee OA in knee OA patients in comparison to a control program.
individuals. However high-intensity resistance exercise showed low to
moderate levels of quality of evidence for greater and more
sustained benefits.
Fransen et al. To determine whether land-based 54 (n=6345) High-quality evidence indicates that land-based
(51) therapeutic exercise is beneficial for knee therapeutic exercise provides short-term benefit that is
OA people in terms of reduced joint pain or sustained for at least 2 to 6 months after cessation of
improved physical function and QOL. formal treatment in terms of reduced knee pain and
moderate-quality evidence shows improvement in
physical function and QOL among knee OA people.
196

Regnaux et al. To determine the benefits and harms 6 (n=656) It was found very low-quality to low-quality evidence for
(67) of high versus low-intensity physical activity no important clinical benefit of high-intensity compared
or exercise programs in hip or knee OA to low-intensity exercise programs in improving pain and
people. physical function in the short term.
REUMATOLOGIAD

Tanaka et al. To examine the effects of exercise therapy on 12 (n=1239) Regardless of its type, high to moderate-quality showed
(75) the health-related QOL of knee OA people. that exercise therapy can improve health-related QOL in
knee OA patients.
Coudeyre et al. To assess the rehabilitation framework of 9 (n=696) Moderate-evidence shows that isokinetic muscle
(50) isokinetic muscle strengthening for knee OA. strengthening is an effective way to propose dynamic
E

muscle strengthening for knee OA rehabilitation and has

a significant effect on pain and disability.
Li et al. (63) To analyze the effectiveness of resistance 17 (n=1705) High-quality evidence shows that both high intensity and
exercise in the treatment of knee OA on low intensity resistance exercise are beneficial in terms of
pain, stiffness, and physical function. reducing pain, alleviating stiffness and improving physical
function in knee OA patients. However, a high intensity
program showed to be more effective than a low intensity
program.
continues on the next page
 tAble Iv. contInuAtIon
Authors N° of inclued

Interventions (A to Z; year) Objectives RCTs (subjects) Results/Conclusions

Tai Chi Lauche et al. To analyze the effectiveness of Tai Chi 5 (n=252) Moderate-evidence for short-term improvement of
(58) for knee OA. pain, physical function and stiffness in knee OA
patients practicing Tai Chi.
Yoga Kan et al. (56) To assess the effects of yoga on pain, 9 (n=372) Moderate-evidence shows that yoga might have
mobility and QOL in knee OA patients. positive effects in relieving pain and mobility on knee OA
patients, but the effects on QOL are unclear.
PORTUGUESASOCIEDADEDAOFICIALÓRGÃO REUMATOLOGIADE

Acupuncture
Cao et al. (49) To evaluate the efficacy of treatment with 14 (n=3835) Moderate to high-quality evidence shows that
acupuncture for knee OA. acupuncture provides a significantly better relief from
knee OA pain and a larger improvement in function than
sham acupuncture, standard care treatment, or waiting for
further treatment.
Therapy RCTs of cupping therapy for treating knee that cupping therapy can effectively improve the

Clinical Devices
Cupping Lee et al. (59) To evaluate the available evidence from 7 (n=661) Only weak evidence can support the hypothesis
OA patients. treatment efficacy and physical function in knee OA
197

rM
patients.

et
Electrotherapy Zeng et al. (85) To investigate the efficacy of different 27 (n=1249) IFC seems to be the most promising pain relief

a
l
electrical stimulation therapies in pain relief treatment for the management of knee OA.
knee OA patients. Although the recommendation level of the other electrical
stimulations therapies is either uncertain (high-frequency
TENS) or not appropriate
(low- frequency TENS, NMES, PES and NIN)
for pain relief, it is likely that none of the
interventions is dangerous.
Electroacupuncture
Shim et al. (70) To verify the effects of electroacupuncture 31 (n=3187) Moderate to low-quality evidence shows that
treatment on knee OA electroacupuncture treatment can more significantly
improve pain and QOL of knee OA patients than
control interventions. However, comparing with sham
electroacupuncture, the difference in pain was less
significant.
continues on the next page
 NoN-pharMacological aNd NoN-surgical iNterveNtioNs to MaNage patieNts with kNee osteoarthritis: aN uMbrella review
tAble Iv. contInuAtIon
Authors N° of inclued

Interventions (A to Z; year) Objectives RCTs (subjects) Results/Conclusions

PEMF Negm et al. To determine if low frequency pulsed 7 (n=459) Current evidence of low and very low-quality suggests
(64) subsensory threshold electrical stimulation that low frequency (≤100 Hz) pulsed subsensory threshold
produced either through PEMF or PES vs electrical stimulation produced either through PEMF/PES
sham PEMF/PES intervention is effective in vs sham PEMF/PES is effective in improving physical
improving pain and physical function at function but not pain intensity at treatment completion in
treatment completion in knee OA adults adults with knee OA blinded to treatment.
PORTUGUESASOCIEDADEDAOFICIALÓRGÃO

blinded to treatment.
We et al. (78) To determine the efficacy of PEMF compared 14 (n=482) High-quality evidence supports PEMF efficacy in the
with a placebo in knee OA patients. management of knee OA pain and function.
TENS Rutjes et al. To compare TENS with sham or no specific 18 (n=813) It could not be confirmed that TENS is effective for pain
(69) intervention in terms of effects on pain and relief. The systematic review is inconclusive, hampered
withdrawals due to adverse events in knee by the inclusion of only small trials of questionable
OA patients. quality.
Insoles Parkes et al. To evaluate whether lateral wedge insoles 12 (n=885) Although meta-analytic pooling of all studies showed
(65) reduce pain in medial knee OA patients a statistically significant association between use of
198

compared with an appropriate control. lateral wedges and lower pain in medial knee OA,
restriction of studies to those using a neutral insole
comparator did not show a significant or clinically
important association. These findings do not support
the use of lateral wedges for this indication.
REUMATOLOGIAD

LLLT Huang et al. To investigate the efficacy of LLLT treatment 9 (n=518) The best available current evidence does not show neither
(53) of knee OA. early nor later benefits in reducing pain or improving
function, not supporting the use of LLLT as a therapy for
knee OA patients.
E

Mudpack Xiang et al. To evaluate the clinical efficacy of mudpack 10 (n=1010) Functional and pain improvements in knee OA patients
Therapy (79) therapy for the knee OA treatment and treated with mudpack therapy was not significantly
identify the likely factors associated with the different from the control subjects at the end of the
high heterogeneity of combined studies. 4- months follow-up.
US Rutjes et al. To compare therapeutic US with sham or 5 (n=341) Therapeutic US may be beneficial for patients with OA of
(68) no specific intervention in terms of effects the knee. Because of the low quality of the evidence, there
on pain and function safety outcomes in is uncertainty about the magnitude of the effects on pain
knee or hip OA patients. relief and function.
continues on the next page
 tAble Iv. contInuAtIon
Authors N° of inclued

Interventions (A to Z; year) Objectives RCTs (subjects) Results/Conclusions

Zeng et al. (84) To investigate the efficacy of continuous and 12 (n=814) Moderate-quality shows that pulsed US has a greater
pulsed US in the management of knee OA. probability of being the preferred mode, as it is more
effective in both pain relief and function improvement
when compared with the control group. However,
continuous US could only be considered as a pain relief
treatment in the management of knee OA.
PORTUGUESASOCIEDADEDAOFICIALÓRGÃO REUMATOLOGIADE

Zhang et al. To explore the effects of therapeutic US with 10 (n=645) Both continuous and pulsed therapeutic US modes can be
(86) sham or no intervention on pain, physical beneficial for reducing knee pain and improving physical
function and safety outcomes in knee OA functions in knee OA patients, however it has small
patients. significant differences.
WBV Li et al. (61) To assess the effects of WBV exercise on knee 5 (n=168) No differences were found in decreasing pain intensity or
OA patients. improving self-reported status, in addition to muscle

Ferre
ira
was more efficacious than squat exercise alone in
strength enhancement compared with other forms of
exercise. However, WBV combined with squat exercise
increasing the level of functional performance.
199

etrM
Wang et al. To assess the effects of WBV for pain, stiffness 5 (n=170) Eight-week and 12-week WBV are beneficial for
(77) and physical functions in knee OA patients. improving physical functions in knee OA patients, but not

al
in reducing pain.
Zafar et al. (83) To examine the current evidence regarding 5 (n=165) WBV has demonstrated limited but beneficial therapeutic
the effects of WBV in knee OA individuals. effects in knee OA individuals in pain and function.
Manual Therapy
French et al. To determine if manual therapy improves 4 (n=280) There is silver level evidence that manual therapy has a
(52) pain and/or physical function in hip or knee beneficial effect compared with exercise therapy, both in
OA people. the short and long-term for pain reduction and increased
physical function. But manual therapy, in the form of
massage therapy, is effective compared to no intervention
in knee OA.
Xu et al. (80) To evaluate the effectiveness and adverse 14 (n=841) Moderate-quality evidence shows that manual therapy
events of manual therapy compared to other might be an effective and safe treatment for improving
treatments for relieving pain, stiffness and pain, stiffness and physical function in knee OA patients.
physical dysfunction in knee OA patients.
continues on the next page
View publication stats

NoN-pharMacological aNd NoN-surgical iNterveNtioNs to MaNage patieNts with kNee osteoarthritis: aN uMbrella review
tAble Iv. contInuAtIon
Authors N° of inclued

Interventions (A to Z; year) Objectives RCTs (subjects) Results/Conclusions

Moxibustion
Li et al. (60) To determine whether the administration 4 (n=746) Limited evidence shows that moxibustion treatment has
of moxibustion is an effective treatment small significant differences comparing to control on
for knee OA. managing the symptoms and improving the QOL among
the selected knee OA patients.
Song et al. (72) To critically reassess the effects of 13 (n=1309) Moxibustion treatment is superior to usual care and sham
moxibustion on knee OA moxibustion in pain, QOL and physical function. But the
PORTUGUESASOCIEDADEDAOFICIALÓRGÃ REUMATOLOGIAD

effects of moxibustion on target population are nearly

equal to oral drug and intra-articular injection, however
with less secondary effects.
Multimodal
Bjordal et al. To access the efficacy of common 36 (n=2434) TENS, electroacupuncture and LLLT administered with
(48) non-pharmacological optimal doses in an intensive 2–4 weeks treatment
interventions for knee OA. regimen seem to offer clinically relevant short-term pain
relief for knee OA.
200

Manual acupuncture, PEMF, US and static magnets did

not show sufficient clinical significant evidence to
recommend their use.
Kinesio Taping
O

Li et al. (62) To assess the effects of elastic taping 11 (n=168) Significant improvements were found in self-reported pain
on knee OA patients. during activity, knee flexibility, knee-related health status
and proprioceptive sensibility compared with other forms
of treatments. However, no differences were found
between the Kinesio Taping group and control group for
knee muscle strength.
E

Abbreviations: IFC (InterFerential Current); LLLT (Low-Level Laser Therapy); NIN (Noninvasive Interactive Neurostimulation); NMES (NeuroMuscular Electrical Stimulation); OA
(Osteoarthritis); PEMF (Pulsed Electromagnetic Fields); PES (Pulsed Electrical Stimulation); QOL (Quality of Life); TENS (Transcutaneous Electrical Nerve Stimulation); US
(Ultrasound); WBV (Whole Body Vibration).