2013 a Systematic Review and Meta-Analysis of Tai Chi for Osteoarthritis of the Knee

8/9/2019 2013 a Systematic Review and Meta-Analysis of Tai Chi for Osteoarthritis of the Knee

1/11

Complementary Therapies in Medicine (2013) 21, 396—406

Available online at www.sciencedirect.com

journal homepage: www.elsevierheal th.com/ journals/ct im

A systematic review and meta-analysis of

Tai Chi for osteoarthritis of the knee

R. Lauche∗, J. Langhorst, G. Dobos, H. Cramer

Department of Internal and Integrative Medicine, Kliniken Essen-Mitte, Faculty of Medicine, University of

Duisburg-Essen, Germany

Available online 1 July 2013

KEYWORDS

Osteoarthritis of theknee;Complementarytherapies;Tai Chi;Meta-analysis;Systematic review

Summary

Objectives: This paper aimed to systematically review and meta-analyze the effectiveness of

Tai Chi for osteoarthritis of the knee.

Methods:MEDLINE, the Cochrane Library, EMBASE, Scopus, PsycInfo and CAMBASEwere screened

through April 2013. Randomized controlled trials (RCTs) comparing Tai Chi to control conditions

were included. Two authors independently assessed risk of bias using the risk of bias tool rec-

ommended by the Cochrane Back Review Group. Outcome measures included pain, physical

functional, joint stiffness, quality of life, and safety. For each outcome, standardized mean

differences and 95% confidence intervals were calculated.

Results: 5 RCTs with a total of 252 patients were included. Four studies had a low risk of

bias. Analysis showed moderate overall evidence for short-term effectiveness for pain, physicalfunction, and stiffness. Strong evidence was found for short-term improvement of the physical

component of quality of life. No long-term effects were observed. Tai Chi therapy was not

associated with serious adverse events.

Conclusion: This systematic review found moderate evidence for short-term improvement of

pain, physical function and stiffness in patients with osteoarthritis of the knee practicing Tai

Chi. Assuming that Tai Chi is at least short-term effective and safe it might be preliminarily

recommended as an adjuvant treatment for patients with osteoarthritis of the knee. More high

quality RCTs are urgently needed to confirm these results.

© 2013 Elsevier Ltd. All rights reserved.

Contents

Introduction.......................................... ........................... .......................... ............... 397

Methods.................................................................................................................. 397

Protocol and registration ............................................................................................ 397

Eligibility criteria....................... .......................... ........................... ........................ 398

∗ Corresponding author at: Knappschafts-Krankenhaus, Am Deimelsberg 34a, 45276 Essen, Germany. Tel.: +49 201 174 25054;fax: +49 201 174 25000.

E-mail address: [email protected](R. Lauche).

0965-2299/$ — see front matter © 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.ctim.2013.06.001

http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.ctim.2013.06.001http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.ctim.2013.06.001http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.ctim.2013.06.001http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.ctim.2013.06.001http://www.sciencedirect.com/science/journal/09652299http://www.elsevierhealth.com/journals/ctimmailto:[email protected]://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.ctim.2013.06.001http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.ctim.2013.06.001mailto:[email protected]://crossmark.dyndns.org/dialog/?doi=10.1016/j.ctim.2013.06.001&domain=pdfhttp://www.elsevierhealth.com/journals/ctimhttp://www.sciencedirect.com/science/journal/09652299http://localhost/var/www/apps/conversion/tmp/scratch_5/dx.doi.org/10.1016/j.ctim.2013.06.001


2/11

Tai Chi for osteoarthritis of the knee 397

Literature search.................................................................................................... 398

Study selection............... .......................... ........................... .......................... ........ 398

Data collection......................... .......................... .......................... ......................... 398

Outcomes..................................................................................................... 398

Risk of bias in individual studies .............................................................................. 398

Data analysis ........................................................................................................ 398

Assessment of overall effect size ............................................................................. 398

Assessment of heterogeneity ................................................................................. 400Subgroup and sensitivity analyses............................................................................. 400

Risk of bias across studies .................................................................................... 400

Results................................................................................................................... 400

Study selection............... .......................... ........................... .......................... ........ 400

Study characteristics ................................................................................................ 401

Setting and participant characteristics ....................................................................... 401

Intervention characteristics .................................................................................. 401

Outcome measures ........................................................................................... 401

Risk of bias in individual studies .............................................................................. 404

Analyses of overall effects.................................................................................... 404

Sensitivity analyses of overall effects......................................................................... 404

Risk of bias across studies .................................................................................... 404

Safety..... .......................... .......................... ........................... .................... 404

Discussion................................................................................................................ 405Summary of main results ............................................................................................ 405

Applicability of evidence ............................................................................................ 405

Quality of evidence ................................................................................................. 405

Agreements and disagreements with other systematic reviews ...................................................... 405

Strengths and weaknesses ........................................................................................... 405

Conclusion... .......................... .......................... ........................... .......................... ... 405

Source of funding ........................................................................................................ 405

Conflicts of interest ...................................................................................................... 405

References............................................................................................................... 405

Introduction

Osteoarthritis of the knee is one of the most common chronicdiseases among older adults with high impact on physi-cal function1; about one fourth of people over 55 yearswill report a significant episode of pain in the knee in thelast year.2 Osteoarthritis of the knee is a condition whichis associated with articular cartilage destruction in addi-tion to underlying bony changes at the joint margins.3 Maincomplaints include pain and functional impairment duringeveryday activities which severely affects quality of life inthese patients.4

Symptomatic therapeutic approaches mainly consist of

physiotherapy, pharmacological therapy or, if therapies fail,joint replacement therapy.5,6 Patients are also encouragedto use some kind of joint-friendly strengthening and aerobicexercises,6,7 as it may reduce pain, increase function andreduce the progression of the osteoarthritis.

Tai Chi, developed as martial art in China, has been prac-ticed for centuries. After introduction in Europe and Americathe perception of Tai Chi shifted and it is nowadays regardeda form of exercise or gymnastics. Tai Chi typically includesa series of dance-like movements that combine to posturesor forms. The forms are executed using slow and smoothmovements that flow into each other. Tai Chi not only is amovement therapy, but it also includes meditative aspects.8

Due to its meditative character it may also reduce stress andincrease psychological well-being.Patients with osteoarthritis of the knee might benefit

fromTai Chi by increasing lower extremitymuscular strengthand joint stability,9 f or example Tai Chi has been found toreduce falls in older adults.10 The meditative aspect mightfurther decrease stress and improve well-being and sinceTai Chi can be offered as group therapy this might fostercontact and social support. Therefore Tai Chi might be a sup-plementary therapeutic option for patients suffering fromosteoarthritis of the knee.To our knowledge, a recent review has investigated the

effectiveness of Tai Chi for osteoarthritis in general11 how-

ever no long-term effects were evaluated. Therefore, theaim of this review was to systematically assess and meta-analyze the short- and long-term effectiveness of Tai Chitherapy for osteoarthritis of the knee.

Methods

Protocol and registration

The review was planned and conducted in accordance withthe PRISMA guidelines for systematic reviews and meta-analyses12 and the recommendations of the Cochrane Back


3/11

398 R. Lauche et al.

Review Group.13,14 The protocol had not been registeredprior to conducting the review.

Eligibility criteria

The following conditions had to be met in order to includea study into this review:

1) Types of studies. Only randomized controlled trials(RCTs) were considered eligible.

2) Types of participants. Studies of adults (18 years of ageand older) with osteoarthritis of the knee were eligi-ble. Diagnosis had to be based on valid instruments, suchas the Classification Criteria of the American College of Rheumatology,3,15 radiographic or laboratory evidence,or medical records. No further restriction regarding dis-ease duration and intensity were applied.

3) Types of interventions. Studies that compared Tai Chiwith no treatment, usual care, placebo or any activetreatment were eligible. No restrictions were maderegarding details of Tai Chi protocol. Co-interventions

were allowed.4) Types of outcome measures. Studies were eligible if

they assessed at least one patient-centered outcome,namely 1) pain, 2) physical functional, or 3) stiffness.If available, data on quality of life, and safety served assecondary outcomes measures.

5) Length of follow-up. Studies were eligible if theyassessed outcomes at least once, either directly at theend of the program (short-term effects), or closest to 6months (long-term effects) following treatment.

Studies were only eligible if they were published as fullpaper. All types of languages were eligible. No restriction

regarding publication date was applied.

Literature search

The following electronic databases were searched fromtheir inception through 04/15/2013: Pubmed/Medline,Embase, PsycInfo, Scopus, the Cochrane Library and CAM-BASE. The literature search was constructed around searchterms for Tai Chi and search terms for osteoarthritis of the knee and adapted for each database as necessary.For Pubmed/Medline, the search strategy was the fol-lowing: ‘‘(tai ji [MeSH Terms] OR tai ji [Title/Abstract]OR tai chi [Title/Abstract] OR t’ai chi [Title/Abstract])

AND (osteoarthritis [MeSH Terms] OR arthritis [MeSHTerms] OR arthralgia [MeSH Terms] OR joint disease [MeSHTerms] OR osteoarthritis [Title/Abstract] OR osteoarthro-sis [Title/Abstract] OR gonarthrosis [Title/Abstract]OR degenerative arthritis [Title/Abstract] OR arthritis[Title/Abstract])’’.

The reference lists of identified original or review articleswere searched manually for further articles.

Study selection

After duplicate removal the abstracts of all identi-fied records were each screened by two reviewers and

hardcopies of those studies that were found potentially eligi-ble were obtained. The full articles were then read in detailand those eligible were included in the systematic reviewand meta-analysis.

Data collection

Two reviewers independently extracted data on study

characteristics such as participants, interventions, controlconditions, co-interventions, outcomemeasures and results.Disagreements were rechecked with a third reviewer andresolved by discussion.

Outcomes

The outcomes were defined as follows.1) Pain intensity had to be measured on a visual ana-

log scale, on a numerical rating scale or on the pain scaleof the Western Ontario and McMaster Universities Arthri-tis Index (WOMAC). 2) Physical function was included whenmeasured on a visual analog scale, a numerical rating scale,the WOMAC physical function scale or the Knee injury and

Osteoarthritis Outcome Score — Physical Function ShortForm (KOOS-PS). 3) Stiffness was included when measuredon a visual analog scale, a numerical rating scale or theWOMAC stiffness scale. 4) Quality of life was included whenassessed using a validated generic measure, e.g. the SF-36,or a validated specificmeasure. 5) Safety was included whenmeasured as the number of adverse events during the studyor the number of drop-outs due to health problems.

Risk of bias in individual studies

Risk of bias was assessed independently by two reviewersusing the 2009 Updated Method Guidelines for SystematicReviews in the Cochrane Back Review Group.14 These guide-lines recommend12 quality criteria, each of which was ratedwith YES, NO or UNCLEAR. These criteria constitute the fol-lowing risk of bias domains: selection bias, performancebias, attrition bias, reporting bias, and detection bias. If study data were inconclusive trial authors were contactedfor further study details. Studies that scored positive on 6of the 12 criteria at the minimum and had no serious flawwere rated as having low risk of bias. Studies that met fewerthan 6 criteria and/or showed a serious flaw were rated ashaving high risk of bias.14 This classification as low or highrisk study was used for sensitivity analyses.

Data analysis

Studies were analyzed separately for short-term and long-term effects. Short-term follow-up was defined as measurestaken directly at the end of the intervention, and long-termfollow-up as measures obtained closest to 6 months afterrandomization.16

Assessment of overall effect size

If at least two studies were available on an outcome,meta-analyses were calculated utilizing Review Manager5 software (Version 5.1, The Nordic Cochrane Centre,Copenhagen). Standardized mean differences (SMD) with95% confidence intervals (CI) were calculated as the meangroup difference divided by the pooled standard deviation.17


4/11


5/11


270 records identified

through database searching- 83 Pubmed/Medline

- 1 Cochrane

- 0 Cambase

- 9 PsycInfo

- 126 Embase

- 51 Scopus

154 records excluded

after abstract screening

5 of studies includedin quantitative synthesis

(meta-analysis) [20-24]

0 full-text articles excluded

170 records after

duplicates removed

16 full-text articles

assessed for eligibility [20-35]

5 of studies includedin qualitative synthesis [20-24]

2 of additional records

identified through other sources

11 full-text articles excluded- 3 mixed patient samples [28,31,35]

- 5 no RCT [29,30,33,34,36]

- 1 no clinical outcomes [27]

- 1 withdrawn study [32]

- 1 study reported twice [25]

Figure 1 Flowchart of the results of the literature search.

Where no standard deviations were available, standard

errors, confidence intervals or t values were used to cal-culate them.

The magnitude of the overall effect size was judgedaccording to Cohen’s categories: small effect size:SMD= 0.2—0.5: moderate effect size: SMD= 0.5—0.8 andlarge effect size SMD> 0.8: large effect sizes.16

A negative standardized mean difference was defined toindicate beneficial effects of Tai Chi therapy compared tothe control intervention for pain and physical function (i.e.decreases complaints), for quality of life on the other handa positive SMD would correspond with enhanced well-being.If necessary, scores were inverted and the mean score wassubtracted from the instruments maximum score.13

Levels of evidence were judged using the criteria fromthe Cochrane Back Review Group with 1) Strong evidence:consistent findings amongmultiple RCTs with low risk of bias;2) Moderate evidence: consistent findings among multiplehigh-risk RCTs and/or one low-risk RCT; 3) Limited evidence:one RCTwith high risk of bias; 4) Conflicting evidence: incon-sistent findings among multiple RCTs; 5) No evidence: noRCTs.17

Assessment of heterogeneity

Statistical heterogeneity between studies was quantifiedby determination of I2. I2 > 30%, I2 > 50% and I2 > 75% weredefined to indicate moderate, substantial and considerable

heterogeneity, respectively.13 A p value ≤0.10 from the Chi2

test was defined to indicate significant heterogeneity.13

Subgroup and sensitivity analyses

If there were at least 2 studies in each subgroup, subgroupanalyses were conducted for type of control treatment (i.e.no treatment, usual care, placebo or any active treatment).Sensitivity analyses to test the robustness of significant

resultswere conducted by removing studies based on the fol-lowing methodological quality criterion: quality score< 6.18

If statistical heterogeneity was present in the respectivemeta-analysis, the subgroup and sensitivity analyses werealso used to explore the nature of this heterogeneity.

Risk of bias across studiesIf at minimum 10 studies were included in a meta-analysis,risk of publication bias was assessed by visual analysis of fun-nel plots generated by Review Manager 5.1 software. Nearlysymmetrical funnel plots indicate low risk while asymmetri-cal funnel plots indicate high risk of publication bias.19

Results

Study selection

Literature search retrieved 272 records, 102 of themwere duplicates (Fig. 1). Sixteen full-text articles were


6/11


T a b l e

2

R i s k o f b i a s a s s e s s m e n t o

f t h e i n c l u d e d s t u d i e s u s i n g t h e C o c h r a

n e r i s k o f b i a s t o o l .

A u t h o r , y e a r

B i a s

T o t a l r i s k :

( m a x .

1 2 )

S e l e c t i o n b i a s :

P e r f o r m a n c e b i a s :

A t t r i t i o n b i a s :

R e p o r t i n g

b i a s :

D e t e c t i o n b i a s :

A d e q u a t e

r a n d o m

s e q u e n c e

g e n e r a t i o n

A d e q u a t e

a l l o c a t i o

n

c o n c e a l m

e n t

S i m i l a r

b a s e l i n e c h a r -

a c t e r i s t i c s

A d e q u a t e

p a r t i c i p a n t

b l i n d i n g

A d e q u a t e

p r o v i d e r

b l i n d i n g

S i m i l a r o r n o

c o - i n t e r v e n t i o n s

A c c e p t a b l e

c o m p l i a n c e

A c c e p t a b l e

a n d d e s c r i b e d

d r o p - o u t r a t e

I n c l u s i o n o f a n

i n t e n t i o n - t o -

t r e a t

a n a l y s i s

N o s e l e c t i v e

o u t c o m e

r e p o r t i n g

A d e q u a t e

o u t c o m e

a s s e s s o r

b l i n d i n g

S i m i l a r t i m i n g

o f o u t c o m e

a s s e s s m e n t

B r i s m e e e t a l . ,

2 0 0 7 2 0

Y e s

U n c l e a r

Y e s

N o

N o

Y e s

Y e s

N o

N o

Y e s

Y e s

Y e s

7

L e e e t a l . ,

2 0 0 9 2 3

Y e s

Y e s

Y e s

N o

N o

Y e s

U n c l e a r

Y e s

Y e s

Y e s

Y e s

Y e s

9

S o n g e t a l . ,

2 0 0 3 2 2

Y e s

Y e s

Y e s

N o

N o

U n c l e a r

U n c l e a r

N o

N o

N o

U n c l e a r

Y e s

4

T s a i e t a l . ,

2 0 1 2 2 4

Y e s

Y e s

Y e s

N o

N o

Y e s

U n c l e a r

Y e s

Y e s

Y e s

Y e s

Y e s

9

W a n g e t a l . ,

2 0 0 9 2 1

Y e s

Y e s

Y e s

N o

N o

Y e s

Y e s

Y e s

Y e s

Y e s

Y e s

Y e s

1 0

assessed for eligibility20—36 (Fig. 1). Eleven full text arti-cles were excluded because they investigated mixed patientsamples,28,31,35 were not randomized,29,30,33,34,36 did notmeasure relevant outcomes27 or were withdrawn.32 Anotherstudywas published twice,23,25 they were reported as 1 studyonly.23 Five studies, involving a total of 252 patients, wereincluded in qualitative and quantitative analysis.20—24

Study characteristics

Characteristics of the samples, interventions, outcomemeasures and results are shown in Table 1.

Setting and participant characteristics

The trials originated from the US20,21,24 or Korea.22,23

Patients were recruited in an outpatient clinic,22 in seniorcenters/senior residential complexes24 or externally bynewspaper announcements.20,21,23

Subjects were aged 50 years,20,23 55 years21,22 or 60years and older24 and were diagnosed with osteoarthritisaccording to the clinical criteria of the American Col-

lege of Rheumatology,15,20,21 with clinical and radiographicfindings35 of at least mild to moderate OA.21—23 In onestudy OA of the knee was confirmed via medical records.24

Patients were excluded if they had undergone bilateralarthroplasty,20 surgery within the past 6 months21,23,24 orif they already participated in Tai Chi or an exerciseprogram.21—23

On average patients were in their 60 s and 80 s and female(86.5%, range from 72.7 to 100.0%). Data on ethnicity wereavailable only for one study with the majority being Cau-casian.

Intervention characteristics

Tai Chi programs lasted 8 weeks,23 12 weeks21,22 or 18weeks20 or 20 weeks20 with courses offered two times aweek,21,23 three times a week24 or at changing frequenciesafter expansion of home practice.20,22 Tai Chi was taught asthe Yang-Style,20,21 the Sun-Style22,24 or as Tai Chi Qigong,23

all including the typical slow, smooth and flowing move-ments of Tai Chi.Control interventions included a wait-list and an atten-

tion control. The wait-list received no treatment23 orcontinued standard care,22 the attention control receivedhealth lectures,20 wellness education, stretching and dietaryadvice21 or health education and culture-related or socialactivities.24

Patients received no co-interventions or received them

to the same extent in 4 studies at least.20,21,23,24

Outcome measures

Pain was assessed in all 5 studies. One of them assessed painintensity using a visual analog scale20 and 4 used the WOMACpain scale.21—24 Physical function was measured with theWOMAC physical function scale and stiffness was measuredby means of the WOMAC stiffness scale in all studies.20—24

Two studies measured quality of life using the SF-36,21,23 andsafety was reported in 3 studies.20,22,24

Short-term effectiveness was evaluated in all studies;long-term effects (6 months) on the other hand were onlyreported in 2 studies.20,21


7/11


Figure 2 Forrest plot for short-term effects of Tai Chi on pain, function, stiffness and quality of life.


8/11


9/11


Risk of bias in individual studies

Four of 5 studies had low risk of bias,20,21,23,24 see Table 2.All studies had low risk of selection bias with only Brismeeet al.20 not reporting allocation concealment. Risk of perfor-mance bias mainly was high, mostly due to lack of blinding of patients and care providers. Risk of attrition bias wasmixed;only 3 studies had acceptable dropout rates and includedintention-to-treat analyses.21,23,24 Risks of reporting bias and

detection bias were low in general.

Analyses of overall effects

Meta-analyses (Fig. 2, Table 3) revealed moderate evidencefor short-term effects on pain (SMD=−0.72; 95% CI−1.00 to−0.44; P < 0.00001), physical function (SMD=−0.72; 95% CI−1.01 to−0.44; P < 0.00001) and stiffness (SMD=−0.59; 95%CI −0.99 to −0.19; P = 0.004). Strong evidence was foundfor physical component of quality of life (SMD= 0.88; 95%CI 0.42 to 1.34; P < 0.0001), but no evidence for the mentalcomponent. No evidence was found for long-term effects.

Based on Cohen’s categories, the short-term effects wereof moderate size, and the effect on the physical component

of quality of life was large.

Sensitivity analyses of overall effects

After exclusion of the high risk RCT22 the results for pain,physical function and stiffness remained significant, theeffect size was approximately the same.Subgroup analyses. Subgroup analyses were conducted forthe comparisons Tai Chi vs. wait-list and Tai Chi vs. attentioncontrol, see Table 4.Tai Chi vs. wait-list. Limited evidence was found for short-term effects on pain, physical function and stiffness of TaiChi compared to wait-list. Effect sizes were moderate, seeTable 4. No long-term effects were evaluated.Sensitivity analyses of Tai Chi vs. wait-list. No sensitivity

analysis could be conducted for the comparison Tai Chi vs.Wait-list because only 2 RCTs were available for the com-parison, 1 low risk and 1 high risk RCT. Results of the lowrisk RCT however showed no significant effect in favor of TaiChi.Tai Chi vs. attention control:. Moderate evidence wasfound for short-term effects of pain and physical func-tion with large and moderate effect size respectively, seeTable 4.Sensitivity analyses of Tai Chi vs. attention control. Nosensitivity analysis for the comparison Tai Chi vs. attentioncontrol was necessary, because all trials had low risk of bias.

Risk of bias across studiesSince less than 10 studies were included in each meta-analysis, funnel plots were not analyzed.

Safety

Adverse events were reported in 3 studies.20,21,24 Brismeeet al.20 f ound sporadic complaints of minor muscle sorenessand foot and knee pain at the beginning of the interven-tion, Wang et al.21 f ound temporarily increased knee pain.She also reported two serious adverse events, namely newlydiagnosed breast and colon cancer, which obviously werenot related to Tai Chi. These adverse events occurred in theintervention and the control group. Tsai et al.24 stated that

T a b l e

4

E f f e c t s i z e s o f T a i C h i v e r s u s c o n t r o l s .

C o m p a r i s o n

O u t c o m e

a

N o . o f s t u d i e s

N o . o f p a t i e n t s

( T a

i C h i )

N o . o f p a t i e n t s

( c o n t r o l )

S M

D ( 9 5 % C I )

P

H e

t e r o g e n e i t y I 2 ; C h i 2 ; P

T a i C h i v s . A t t e n t i o n

c o n t r o l

S h o r t - t e r m : P a i n

3

6 8

6 0

−

0 . 7

9 [ − 1 . 1

6 , − 0 . 4

3 ]

< 0 . 0

0 0 1

6 % ; 2 . 1

3 ; 0 . 3

4

L o n g - t e r m : P a i n

2

4 0

3 3

−

0 . 2

9 [ − 1 . 0

6 ,

0 . 4

8 ]

0 . 4

6

6 2 % ,

2 . 6

4 ; 0 . 1

0

S h o r t - t e r m : F u n c t i o n

3

6 8

6 0

−

0 . 7

6 [ − 1 . 1

2 , − 0 . 4

0 ]

< 0 . 0

0 0 1

0 % ; 1 . 8

4 ; 0 . 4

0

L o n g - t e r m : F u n c t i o n

2

3 8

3 3

−

0 . 3

3 [ − 0 . 9

5 ,

0 . 2

8 ]

0 . 2

9

4 0 % ; 1 . 6

7 ; 0 . 2

0

S h o r t - t e r m : S t i f f n e s s

3

6 8

6 0

−

0 . 5

9 [ − 1 . 3

1 ,

0 . 1

3 ]

0 . 1

1

7 4 % ; 7 . 7

6 ; 0 . 0

2

L o n g - t e r m : S t i f f n e s s

2

3 8

3 3

0 .

0 6 [ − 0 . 7

2 ,

0 . 8

3 ]

0 . 8

9

6 2 % ; 2 . 6

4 ; 0 . 1

0

T a i C h i v s . W a i t - l i s t

S h o r t - t e r m : P a i n b

2

5 1

3 6

−

0 . 6

1 [ − 1 . 0

6 , − 0 . 1

7 ]

0 . 0

0 7

0 % ; 0 . 0

5 ; 0 . 8

2

S h o r t - t e r m : F u n c t i o n b

2

5 1

3 6

−

0 . 6

7 [ − 1 . 1

2 , − 0 . 2

3 ]

0 . 0

0 3

0 % ; 0 . 6

0 ; 0 . 4

4

S h o r t - t e r m : S t i f f n e s s b

2

5 1

3 6

−

0 . 5

6 [ − 1 . 0

1 , − 0 . 1

2 ]

0 . 0

1

0 % ; 0 . 1

8 ; 0 . 6

7

A b b r e v i a t i o n s : C I , C o n fi d e n c e i n t e r v a

l ; S D ,

S t a n d a r d d e v i a t i o n .

a

O u t c o m e s a r e o n l y s h o w n i f s u f fi c i e n t d a t a f o r m e t a - a n a l y s i s w e r e a v a i l a b l e

.

b

N o s e n s i t i v i t y a n a l y s i s p o s s i b l e d u e t o t h e l a c k o f l o w r i s k s t u d i e s .


10/11


no adverse events occurred. No other serious adverse eventswere reported.

Discussion

Summary of main results

This meta-analysis found moderate overall evidence forshort-term improvement of pain, physical function and stiff-ness, and strong evidence for short-term improvement of thephysical component of quality of life. Compared to attentioncontrol, Tai Chi showed moderate evidence for short-termimprovement of pain and physical function, and compared towait-list there was limited evidence for short-term improve-ment of pain, physical function and stiffness, however dueto the high risk of bias the results of the latter might be dueto bias alone. Available safety data suggest that Tai Chi wasnot associated with serious adverse events.

Applicability of evidence

Trials were conducted with outpatients or externallyrecruited patients in the US or Korea. The vast majority of participants were female with an average age between 60and 80 years. The studies included patients with at leastmildto moderate osteoarthritis according to radiographic evi-dence. The results of the studies might or might not apply tothe majority of osteoarthritis patients; there are not enoughstudies for conclusive judgment.

Quality of evidence

There was some variability of the methodological quality of studies; the effects on pain, physical function and stiffness

were robust against potential methodological biases. Theresults of the comparisons Tai Chi vs. wait-list could not beseparated from bias.

Agreements and disagreements with other

systematic reviews

Systematic reviews are available for Tai Chi in the treatmentof osteoarthritis in general.11,37 One review also includeda meta-analysis,11 which found encouraging effects forosteoarthritis of the knee regarding pain and function. How-ever this meta-analysis did not cover the long-term effectsand further included a study31 that was later withdrawn

because there was considerable doubt on the authenticityof the data.38,39 After exclusion of the study from Ni et al.,31

inclusion of the recent study results by Tsai et al.24 and underconsideration of the long-term effects the results are moredifferentiated than reported there.11 Altogether short-termeffects were only found when compared to attention con-trol. Compared to wail-list controls the effects could not bedistinguished from bias.

The modes of action are not known in detail, but itcan be assumed that Tai Chi acts like other kinds of physi-cal exercises.11 Physical movement can increase muscularstrength and flexibility, and stabilize the knee joint.9 Itmay also help to improve balance and reduce the risk of

falling.10 The meditative aspect of Tai Chi can reduce stressand increase psychosocial well-being.

Strengths and weaknesses

This review and meta-analysis included only randomizedcontrolled trials and evaluated long-term effects of Tai Chi.Patient-centered outcomes were used for meta-analysesas recommended by the task force of the OsteoarthritisResearch Society.40

The primary limitation of this review is the total numberof eligible trials. Therefore, sensitivity analyses for subgroupcomparisons were partly impossible. More studies that com-pare Tai Chi therapy to guideline endorsed therapies; otherforms of exercises or placebo are urgently needed. The num-ber of trials also limits judgment on side effect of Tai Chi.Although no adverse events were associated with Tai Chidefinite conclusions are not possible here. It only can beassumed that Tai Chi is a treatment with low risk of injury.Overall risk of bias was low but 1 out of 5 studies had a

high risk of bias. Statistical heterogeneity was not present

in the meta-analyses. Another limiting factor is the shortfollow-up time frame. Only 2 studies evaluated long-termfollow-up and the longest follow-up was conducted at 24weeks. Recommendations of the Osteoarthritis ResearchSociety International include longer time frames40,41 whichshould be considered in following studies.

Conclusion

This systematic review found moderate evidence for short-term improvement of pain, physical function and stiffnessin patients with osteoarthritis of the knee practicing TaiChi. Assuming that Tai Chi is at least short-term effective

and safe it might be preliminarily recommended as an adju-vant treatment for patients with osteoarthritis of the knee.More high quality RCTs are urgently needed to confirm theseresults.

Source of funding

This review was supported by a grant from the Rut- andKlaus-Bahlsen-Foundation. The funding source had no rolein the study design, in the collection, analysis and interpre-tation of data; in the writing of the manuscript; and in thedecision to submit the manuscript for publication.

Conflicts of interest

The authors declare that no financial or non-financial con-flict of interest exists.

References

1. Guccione AA, Felson DT, Anderson JJ, et al. The effects of spe-cific medical conditions on the functional limitations of eldersin the Framingham study. American Journal of Public Health1993;84:351—8.

2. Peat G, McCarney R, Croft R. Knee pain and osteoarthritisin older adults: a review of community burden and current

http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0005


11/11


use of primary health care. Annals of the Rheumatic Diseases2001;60:91—7.

3. AltmanR,AschE, Bloch D,et al. Development of criteria for theclassification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Diagnostic and Therapeutic CriteriaCommittee of the American Rheumatism Association. Arthritis& Rheumatism 1986;29:1039—49.

4. Reginster JY. The prevalence and burden of arthritis. Rheuma-tology (Oxford) 2002;41(Suppl. 1):3—6.

5. Pendleton A, Arden N, Dougados M, et al. EULAR recommen-dations for the management of knee osteoarthritis: report of a task force of the standing committee for international clini-cal studies including therapeutic trials (ESCISIT). Annals of theRheumatic Diseases 2000;59:936—44.

[6]. Benell KL,Hunter DH, Hinmann RS.Management of osteoarthri-tis of the knee. BMJ 2000;345:e4934.

[7]. Roddy E, Zhang W, Doherty M, et al. Evidence-based recom-mendations for the role of exercise in the management of osteoarthritis of the hip or knee—–theMOVEconsensus. Rheuma-tology (Oxford) 2005;44:67—73.

[8]. Lam P. New horizons. developing tai chi for health care. Aus-tralian Family Physician 1998;27:100—1.

[9]. Christou EA, Yang Y, Rosengren KS. Taiji training improves kneeextensor strength and force control in older adults. Journals of

Gerontology Series A: Biological Sciences and Medical Sciences2003;58:763—6.

[10].Li F, Harmer P, Fischer KJ, et al. Tai Chi and fall reductions inolder adults: a randomized controlled trial. Journal of Geron-tology 2005;60A:187—94.

[11].Kang JW, Lee MS, Posadzki P, et al. T’ai chi for the treatmentof osteoarthritis: a systematic review and meta-analysis. BMJOpen 2011;1:e000035.

[12].Moher D, Liberati A, Teztlaff J, et al. Preferred reporting itemsfor systematic reviews and meta-analyses: the PRISMA State-ment. Annals of Internal Medicine 2009;51:1—7.

[13].Higgins JPT, GreenS.Cochranehandbookforsystematic reviewsof interventions. Version 5.1.0. The Cochrane Collaboration;2011, http://www.cochrane-handbook.org(assessed09.02.12).

[14].FurlanAD,Pennick V, Bombardier C, et al.2009 updated method

guidelines for systematic reviews in the Cochrane Back ReviewGroup. Spine 2009;34:1929—41.

[15].Belo JN, Berger MY, Koes BW, et al. The prognostic value of the clinical ACR classification criteria of knee osteoarthritis forpersisting knee complaints and increase of disability in generalpractice. Osteoarthritis Cartilage 2009;17:1288—92.

[16].Cohen J. Statistical power analysis for the behavioral sciences.Hillsdale, NJ: Lawrence Erlbaum Associates; 1998.

[17].van Tulder M, Furlan A, Bombardier C, et al. Updated methodguidelines for systematic reviews in theCochraneCollaborationBack Review Group. Spine 2003;28:1290—9.

[18].CramerH, LaucheR, LanghorstJ, et al.Effectiveness of yoga formenopausal symptoms—–a systematic review and meta-analysisof randomized controlled trials. Evidence-Based Complemen-tary and Alternative Medicine 2012;2012:863905.

[19].Egger M, Davey Smith G, et al. Bias in meta-analysis detectedby a simple, graphical test. BMJ 1997;315:629—34.

[20].Brismée JM,Paige RL,Chyu MC,et al.Group andhome-based taichi in elderly subjects with knee osteoarthritis: a randomizedcontrolled trial. Clinical Rehabilitation 2007;21:99—111.

[21].Wang C, Schmid CH, Hibberd PL, et al. Tai Chi is effectivein treating knee osteoarthritis: a randomized controlled trial. Arthritis and Rheumatism 2009;61:1545—53.

[22].Lee HJ, Park HJ, Chae Y, et al. Tai Chi Qigong for the qualityof life of patients with knee osteoarthritis: a pilot, ran-domized, waiting list controlled trial. Clinical Rehabilitation2009;23:504—11.

[23].Song R, Lee EO, Lam P, et al. Effects of tai chi exercise on pain,balance, muscle strength, and perceived difficulties in physical

functioning in older women with osteoarthritis: a randomizedclinical trial. Journal of Rheumatology 2003;30:2039—44.

[24].Tsai PF, Chang JY, Beck C, Kuo YF, Keefe FJ. A pilot cluster-randomized trial of a 20-week tai chi program in elders withcognitive impairment and osteoarthritic knee: effects on painand other health outcomes. Journal of Pain and SymptomMan-agement 2012;45:660—9.

[25].Song R, Lee EO, Lam P, et al. Effects of a Sun-style Tai Chi exer-cise on arthritic symptoms,motivation and the performance of

health behaviors in women with osteoarthritis. Taehan KanhoHakhoe Chi 2007;37:249—56.

[26].Song R, Roberts BL, Lee EO, et al. A randomized study of the effects of t’ai chi on muscle strength, bone mineraldensity, and fear of falling in women with osteoarthritis. Journal of Alternative and Complementary Medicine 2010;16:227—33.

[27].Fransen M, Nairn L, Winstanley J, et al. Physical activity forosteoarthritis management: a randomized controlled clinicaltrial evaluating hydrotherapy or Tai Chi classes. Arthritis and Rheumatism 2007;57:407—14.

[28].Lee HY. Comparison of effects among Tai-Chi exercise, aquaticexercise, and a self-help program for patients with kneeosteoarthritis.Taehan KanhoHakhoe Chi 2006;36:571—80 [Arti-cle in Korean].

[29].Lee HY, Kim HI. Effects of sun style Tai Chi program on mus-cle strength and physical function in osteoarthritis patients. Journal of Korean Nurses Association for Complementary Alter -

native Therapy 2004;1:43—52.[30].Hartman CA, Manos TM, Winter C, et al. Effects of T’ai Chi

training on function and quality of life indicators in older adultswith osteoarthritis. Journal of the American Geriatrics Society 2010;48:1553—9.

[31].Ni GX, Song L, Yu B, et al. Tai chi improves physical functionin older Chinese women with knee osteoarthritis. Journal of Clinical Rheumatology 2010;16:64—7.

[32].Shen CL, James CR, Chyu MC, et al. Effects of Tai Chi ongait kinematics, physical function, and pain in elderly withknee osteoarthritis—–a pilot study. American Journal of ChineseMedicine 2008;36:219—32.

[33].Tsai PF, Beck C, Chang JY, et al. The effect of tai chi on kneeosteoarthritis pain in cognitively impaired elders: pilot study.Geriatric Nursing 2009;30:132—9.

[34].Adler PA. The effects of tai chi on pain and function in olderadults with osteoarthritis. Dissertation, Case Western ReserveUniversity, 2007.

[35].Lee HY, Lee KJ. Effects of Tai Chi exercise in elderly with kneeosteoarthritis. TaehanKanhoHakhoe Chi 2008;38:11—8 [Articlein Korean].

[36].Kellgren JH, Lawrence JS. Radiological assessmentof osteo-arthrosis. Annals of the Rheumatic Diseases1957;16(December):494—502.

[37].Lee MS, Pittler MH, Ernst E. Tai chi for osteoarthritis: a system-atic review. Clinical Rheumatology 2008;27:211—8.

[38].Schumacher HR.Tai Chi improves physical function in older Chi-nese women with knee osteoarthritis: retraction. Journal of Clinical Rheumatology 2010;16:357.

[39].WangC. Serious concerns related to thearticle entitled ‘‘Tai Chiimproves physical function in older Chinese women with kneeosteoarthritis’’. Journal of Clinical Rheumatology 2010;16:356.

[40].Altman R, Brandt K, Hochberg M, et al. Design and conductof clinical trials in patients with osteoarthritis: recommenda-tions from a task force of the Osteoarthritis Research Society,Results from a workshop. Osteoarthritis and Cartilage 1996;4:217—43.

[41].Hochberg MC, Altman RD, Brandt KD, et al. Design and conductof clinical trials in osteoarthritis: preliminary recommendationsfrom a task forceof theOsteoarthritis Research Society. Journalof Rheumatology 1997;24:792—4.
http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0010http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0015http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0015http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0015http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0015http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0015http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0015http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0015http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0015http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0015http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0015http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0020http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0020http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0020http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0020http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0020http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0020http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0020http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0025http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0025http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0025http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0025http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0025http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0025http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0025http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0025http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0025http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0025http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0030http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0030http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0030http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0030http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0030http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0030http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0035http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0035http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0035http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0035http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0035http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0035http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0035http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0035http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0035http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0035http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0040http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0040http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0040http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0040http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0040http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0040http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0040http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0045http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0045http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0045http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0045http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0045http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0045http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0045http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0045http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0050http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0050http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0050http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0050http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0050http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0050http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0050http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0050http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0055http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0055http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0055http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0055http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0055http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0055http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0055http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0055http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0060http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0060http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0060http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0060http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0060http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0060http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0060http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0060http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0065http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0065http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0065http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0065http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0065http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0070http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0070http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0070http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0070http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0070http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0070http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0070http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0070http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0075http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0075http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0075http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0075http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0075http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0075http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0075http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0075http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0075http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0080http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0080http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0080http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0080http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0085http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0085http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0085http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0085http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0085http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0085http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0085http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0085http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0090http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0090http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0090http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0090http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0090http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0090http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0090http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0090http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0095http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0095http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0095http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0095http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0095http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0095http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0095http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0100http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0100http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0100http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0100http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0100http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0100http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0100http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0100http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0105http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0105http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0105http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0105http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0105http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0105http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0105http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0110http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0110http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0110http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0110http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0110http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0110http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0110http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0110http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0115http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0115http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0115http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0115http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0115http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0115http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0115http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0115http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0115http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0120http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0120http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0120http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0120http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0120http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0120http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0120http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0120http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0120http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0120http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0125http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0125http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0125http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0125http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0125http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0125http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0125http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0125http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0125http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0125http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0130http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0130http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0130http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0130http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0130http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0130http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0130http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0130http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0130http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0135http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0135http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0135http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0135http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0135http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0135http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0135http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0135http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0135http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0140http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0140http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0140http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0140http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0140http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0140http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0140http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0140http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0140http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0145http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0145http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0145http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0145http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0145http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0145http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0145http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0145http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0150http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0150http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0150http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0150http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0150http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0150http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0150http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0150http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0155http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0155http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0155http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0155http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0155http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0155http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0155http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0155http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0160http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0160http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0160http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0160http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0160http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0160http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0160http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0160http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0160http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0165http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0165http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0165http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0165http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0165http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0165http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0165http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0175http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0175http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0175http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0175http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0175http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0175http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0175http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0175http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0180http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0180http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0180http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0180http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0180http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0180http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0180http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0205http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0200http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0195http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0190http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)00094-0/sbref0185http://refhub.elsevier.com/S0965-2299(13)

Date post:	01-Jun-2018
Category:	Documents
Upload:	ani-fran-solar
View:	217 times
Download:	0 times

2013 a Systematic Review and Meta-Analysis of Tai Chi for Osteoarthritis of the Knee

Documents