Complete the 2 tables using the article attached. See the example for guidance .
SPECIAL EDITORIAL REVIEW
Intra-articular platelet-rich plasma injections for knee osteoarthritis: An overview of systematic reviews and risk of bias considerations
Dan XING,1,2,* Bin WANG,3,* Wei ZHANG,1,2 Ziyi YANG,1,2 Yunfei HOU,1,2
Yaolong CHEN4,5 and Jianhao LIN1,2
1Arthritis Clinic & Research Center, Peking University People’s Hospital, 2Arthritis Institute, Peking University, Beijing, 3Department of Orthopaedics, The Second Hospital of Shanxi Medical University, Taiyuan, Shanxi, 4Evidence-Based Medicine Center, School of Basic Medical Sciences, Lanzhou University, Lanzhou, and 5Chinese GRADE Center, Gansu, China
Abstract Objectives: Numerous systematic reviews investigating the efficacy of platelet-rich plasma (PRP) in treating knee osteoarthritis (OA) have been recently published. The purpose of the present study was (1) to perform an over-
view of overlapping systematic reviews investigating PRP for knee OA via evaluating methodological quality and
risk of bias of systematic reviews and (2) to provide recommendations through the best evidence.
Methods: A systematic search of systematic reviews published through Feb 2017 was conducted using the MED- LINE, EMBASE and Cochrane Library. The methodological quality and risk of bias of included systematic
reviews were assessed by AMSTAR instrument and ROBIS tool respectively. Best evidence choice procedure was
conducted according to the Jadad decision algorithm. The systematic reviews with high quality of methodology
and low risk of bias were selected ultimately.
Results: Ten systematic reviews were eligible for inclusion. The Jadad decision making tool suggested that the reviews with highest AMSTAR score should be selected. According to the ROBIS tool, there were 4 systematic
reviews with low risk of bias and 6 with high risk of bias. As a result, two systematic reviews conducted by Dai et
al and Meheux et al with highest AMSTAR score and low risk of bias were selected as the best evidence.
Conclusions: The present overview demonstrates that PRP is an effective intervention in treating knee OA with- out increased risk of adverse events. Therefore, the present conclusions may help decision makers interpret and
choose PRP with more confidence.
Key words: intra-articular injection, knee osteoarthritis, overview, platelet-rich plasma, ROBIS.
INTRODUCTION
Knee osteoarthritis (OA) is a common pathological
condition with rising prevalence1–3 that plagues mainly
the elderly with pain, functional damage and
impairment of life quality.4 Knee OA might have also
increased the socioeconomic burden in terms of cost of
medical care for both government and individuals. At
present, the ultimate goal of treating knee OA is to
relieve symptoms and to improve joint function and
thus quality of life. Although total knee replacement
(TKA) may serve as an effective alternative for patients
with severe knee OA, the risk of complications of TKA
cannot be completely eliminated. Numerous conserva-
tive treatment options are currently available, including
analgesic medication, physical therapy, unloaded
Correspondence: Dr Jianhao Lin, Arthritis Clinic & Research Center, Peking University People’s Hospital, Peking Univer- sity, Beijing 100044, China. Email: linjianhao@pkuph.edu.cn *The first two authors contributed equally to this work. Registration: PROSPERO (http://www.crd.york.ac.uk/PROS PERO) [CRD42017058244].
© 2017 Asia Pacific League of Associations for Rheumatology and John Wiley & Sons Australia, Ltd
International Journal of Rheumatic Diseases 2017; 20: 1612–1630
bracing, intra-articular injections and so on, aiming to
relieve pain of the knee joints and delay the TKA inter-
vention.5
Platelet-rich plasma (PRP) is a portion of whole
blood with an elevated platelet concentration. Some
kinds of growth factors, stored in the alpha granules of
these increased concentrations of platelets, regulate
some biological processes in tissue repair.6 Preparation
of PRP involves centrifugation of autologous blood to
separate and extract the plasma and buffy-coat portion
of the blood, which result in high concentrations of
platelets. Thus, PRP contains a higher concentration of
platelet-derived growth factors (PDGFs) than in autolo-
gous platelets.7 Platelets mediate their effects on tissue
repair through degranulation, in which vascular
endothelial growth factor (VEGF), PDGF, epidermal
growth factor (EGF), basic fibroblastic growth factor
(bFGF), and transforming growth factor-b1 (TGF-b1) are released from alpha granules.8 Platelets also store
antibacterial and fungicidal proteins, metalloproteases,
coagulation factors and membrane glycoproteins,
which may influence inflammation by inducing the
synthesis of other integrins, interleukins and chemo-
kines.6 The preparation and injection of PRP can be
achieved or accomplished in the outpatient clinic.6,9 In
the early inflammatory phases of the healing process,
platelets are the first cell type to arrive at the site of tis-
sue injury and are particularly active in playing an
important role in treatment.
Among all the possible intra-articular therapeutic
options, PRP has been frequently applied in treating
knee OA.10 These aforementioned growth factors in pla-
telets with higher concentrations could act in chondro-
cytes to promote synthesis of extracellular matrix,
protein transcription and chondrocytes growth and
migration.11 Theoretically, the supraphysiological
release of PDGFs can directly act on the affected site. The
regenerative and anti-inflammatory effects of PRP may
play an important role in the cartilage-healing process of
a wide range of musculoskeletal disorders.12,13 Regard-
ing the treatment of knee OA, PRP may stimulate the
cascade of natural healing and mediate the anti-inflam-
matory response in knee OA.14,15 Currently, several PRP
products are commercially available in knee OA. The
techniques used to generate PRP products differ in the
amount of whole blood, the use of anticoagulant, the
time and speed of centrifugation, the final volume, and
the number of platelets in the PRP products. The benefit
of including leukocytes in the PRP product remains con-
troversial, and few studies have been conducted to eval-
uate the effects of the interaction between platelets and
leukocytes on the growth factor concentrations.16,17 Xu
et al.18 reported that the therapeutic effects of PRP on
cartilage regeneration could be enhanced by removing
leukocytes to avoid the activation of the nuclear factor
(NF)-jB pathway. Furthermore, the application of PRP can vary, based on the use of either bicarbonate, to buf-
fer the acidic nature of PRP derived with acid citrate dex-
trose, or thrombin, to initiate the clotting cascade
through degranulation of platelets.19
Several systematic reviews or meta-analyses concern-
ing this topic have been published. Although favorable
outcomes of PRP injections have been obtained,20–29
limitations such as less-optimized primary study design
and/or varied methodologies in conducting systematic
reviews could affect our recommendations in terms of
decision making. As stated in the American Academy of
Orthopedic Surgeons guideline for knee OA, it is unfea-
sible, with inconclusive strength of recommendation,
to make the stand for or against PRP use for patients
with symptomatic knee OA.30 Hence, positive conclu-
sions about using PRP treating knee OA cannot yet be
made with absolute confidence.
The purpose of the present study was: (i) to conduct
an overview of overlapping systematic reviews that
compared the efficacy and safety of standalone PRP
injections in treating knee OA with placebo or other
injection agents; (ii) to evaluate the quality of method-
ology and risk of bias of relevant systematic reviews;
(iii) to determine which systematic reviews offer the
best evidence with lower risk of bias and provide rec-
ommendations for using PRP in treating knee OA.
MATERIALS AND METHODS
The present study was performed according to the
guideline of Preferred Reporting Items for Systematic
reviews and Meta-analysis (PRISMA) checklist.31 Based
on the (Evidence-based Practice Centers) EPC guidance,
existing systematic reviews have been integrated into a
new review.32
Search strategy All meta-analyses or systematic reviews that met the fol-
lowing inclusion criteria were searched in databases,
including MEDLINE, EMBASE and the Cochrane library
from database inception to 28 February, 2017. The fol-
lowing MeSH items or free words were used: knee,
osteoarthritis, platelet rich plasma, PRP, corticosteroid,
placebo, viscosupplementation, hyaluronic acid, meta-
analysis and systematic review. In addition to electronic
literature search, the references of searched articles were
International Journal of Rheumatic Diseases 2017; 20: 1612–1630 1613
Platelet-rich plasma for knee osteoarthritis
also reviewed to identify other meta-analyses or system-
atic reviews. The publication language was restricted to
English.
Inclusive and exclusive criteria Meta-analyses or systematic reviews of PRP in treating
knee OA patients that meet the following inclusive/ex-
clusive criteria were be eligible for inclusion:
(1) type of studies: meta-analysis or systematic review (2) participants: knee OA patients who were treated
with intra-articular injection
(3) interventions: the included systematic reviews had compared all types of PRP with the following
agents: placebo, corticosteroids, hyaluronic acid
(HA) and other pharmacological agents.
Exclusion criteria included the following items:
(1) PRP used in hand, hip, ankle and other joints with OA
(2) PRP used in or after arthroplasty (3) basic science review and systematic reviews based
on in vitro, in vivo preclinical studies
(4) abstract, commentary, methodological study, over- view, narrative review, guidelines.
Studies selection Two authors independently assessed the titles and
abstracts of the systematic reviews of choice for the eli-
gibility criteria. They were uninformed of the journals/
authors’ information and affiliations. Subsequently, sys-
tematic reviews which were regarded as potentially rele-
vant by any of the reviewers were obtained in full text
for further review independently by the same reviewers.
Any disagreement was resolved by reaching a consensus
through discussion.
Date extraction Two reviewers independently extracted the data from
each included systematic review using a standard data
extraction form. The following items were extracted:
title, authors, original study design, total number of pri-
mary studies, the pooled results, methodological vari-
ables, classification of PRP and PRP preparation
methods. Any disagreement concerning the extracted
data was resolved by discussion.
Methodological quality assessment for systematic reviews The Assessment of Multiple Systematic Reviews
(AMSTAR) tool33 was used to evaluate the quality of
methodology of included systematic reviews. The
assessment process was independently performed by
two reviewers. Any disagreement was resolved by dis-
cussion or consulting a third reviewer. The AMSTAR is a
measurement tool with 11 items for assessing method-
ological quality of published meta-analyses or system-
atic reviews.34
Heterogeneity within systematic reviews Heterogeneity of each variable was summarized for each
included systematic review with pooling results. The fol-
lowing two questions were also assessed: whether sensi-
tivity analysis was performed in meta-analysis and
whether the included systematic reviews had evaluated
potential sources of heterogeneity across primary stud-
ies. According to the Cochrane Handbook, heterogene-
ity of each outcome between 0% and 40% is considered
as not important, between 30% and 60% as moderate,
between 50% and 90% as substantial, and between
75% and 100% as considerable. I2 value was adopted to
demonstrate the heterogeneity among primary studies.
Choice of best evidence Best evidence choice procedure was conducted based
on the Jadad decision algorithm,35 which aims to help
select reliable ones among all the systematical reviews.
The assessment criteria of Jadad decision algorithm
included: clinical question, study selection and inclu-
sion, data extraction, assessment of study quality,
assessment of ability to combine studies, and statistical
methods for data synthesis.35 The procedure was con-
ducted by two independent reviewers. The consensus
was reached through agreeing on which of the included
systematic reviews could provide best evidence based
on current information.
Risk of bias assessment for systematic reviews The risk of bias assessment of included systematic
reviews was performed independently by two authors
with the help of the ROBIS tool.36 Disagreements were
resolved by discussion. According to the ROBIS tool,
risk of bias was evaluated by assessing the following
four domains: study eligibility criteria, identification
and selection of studies, data collection and study
appraisal, and synthesis and findings. These four
domains covered the main review processes.
Each domain was evaluated for information used to
support the judgments, signaling questions, and judg-
ment of concern about risk of bias. The answers for the
signaling questions included ‘Yes’, ‘Probably Yes’, ‘No’,
‘Probably No’ and ‘No Information’. The answer only
1614 International Journal of Rheumatic Diseases 2017; 20: 1612–1630
D. Xing et al.
with ‘Yes’ indicates low concerns. Thus, each domain
for risk of bias was classified as ‘Low’, ‘High’ or
‘Unclear’. If one domain was categorized as low level of
concern, all signaling questions for the domain were
Yes or Probably Yes. Concern about bias was elevated if
any signaling questions were reported as ‘No’ or ‘Proba-
bly No’.36
RESULTS Literature search A total of 145 titles and abstracts were preliminarily
reviewed, and 10 published systematic reviews20–29 ulti-
mately met the eligibility criteria (Fig. 1). Two system-
atic reviews with primary studies of other joints
included were excluded. Eight narrative reviews and
two commentaries were excluded. One literature was
not included because it conducted overview based on
the published systematic reviews. Two were excluded
because they aimed to review the basic science or pre-
clinical studies of PRP in treating OA. Two studies were
omitted because of PRP used in or after TKA. Two stud-
ies were excluded because they included studies that
compared PRP with other drugs. One was excluded due
to comparing leukocyte-poor PRP with leukocyte-rich
PRP.
Characteristics of included systematic reviews The characteristics of included systematic reviews are
presented in Table 1. The systematic reviews were pub-
lished from 201329 to 2017.20 The number of primary
studies included in systematic reviews varied from five
in the study published in 201427 to 16 in the one pub-
lished in 201428 (Table 2). Two systematic reviews24,25
conducted qualitative synthesis without pooled data.
Search methodology The comprehensive search methodology which was
adopted by included systematic reviews is presented in
Table 3. Medline, Embase and Cochrane database are
the most frequently used searching databases of the
included studies.
Methodological quality of systematic reviews Table 4 presents the methodological characteristics of
included systematic reviews. Half of the included sys-
tematic reviews20–24 only included randomized clinical
trials (RCTs), while others25–29 included RCTs and non-
RCTs. The evidence degree of each systematic review
was Level II. REVMAN and STATA software were used
in systematic reviews with pooling data. Sensitivity and
subgroup analysis were conducted in five21–23,28,29 and
four21,23,26,28 included studies, respectively. Two sys-
tematic reviews21,26 assessed quality of evidence in their
study.
The total score by the AMSTAR instrument with each
question of included systematic reviews are presented
in Table 5. The average AMSTAR score of included liter-
atures was 7.3, ranging from 425 to 9.20,23,24 All the
included systematic reviews declared that there was no
conflict of interest in making the investigation. Three
systematic reviews conducted by Dai et al.,23 Shen
et al.20 and Meheux et al.24 had the highest quality.
Heterogeneity assessment The data heterogeneity of each variable in each system-
atic review are presented in Table 6. The I2 value was
shown to present the heterogeneity among primary
studies. The outcomes of almost all the included sys-
tematic reviews had moderate or high heterogeneity.
Jadad decision algorithm All pooled results reported in systematic reviews are
demonstrated in Figure 2. According to the following
Jadad decision algorithm, the eligible systematic reviews
were elected on account of the methodological quality
(Fig. 3). Therefore, the above-mentioned three stud-
ies20,23,24 with highest AMSTAR score were selected.
Risk of bias of systematic reviews The risk of bias of included systematic reviews by
ROBIS tool is presented in Table 7, as are the assess-
ment results of each item in phase 2 of ROBIS tool. The
third phase demonstrated whether the systematic
reviews as a whole was at risk of bias. There were four
systematic reviews23,24,26,28 with low risk of bias, and
six20–22,25,27,29 with high risk of bias. Judgments regard-
ing each ROBIS item are presented as percentages across
all the included systematic reviews in Figure 4. Based
on the AMSTAR instrument and ROBIS tool, two sys-
tematic reviews23,24 with higher methodological quality
and lower risk of bias were selected to provide best
evidence.
DISCUSSION
Meta-analysis and systematic review are generally con-
sidered to be the best way to obtain evidence for health-
care decision making, and thus can be used to resolve a
wide range of clinical problems. Decision makers in
medical institutions look forward to consistent, stable
and unbiased recommendations based on systematic
International Journal of Rheumatic Diseases 2017; 20: 1612–1630 1615
Platelet-rich plasma for knee osteoarthritis
Figure 1 The systematic reviews selection and inclusion process.
Table 1 Characteristics of included systematic reviews
Authors Journal Date of last
literature search
Date of
publication
No. of
included
studies
No. of included
RCTs
No. of grey
literature
Shen et al. (2017)20 Journal of Orthopaedic
Surgery and Research
December,
2016
January,
2017
14 14 0
Dai et al. (2016)23 Arthroscopy April, 2016 December,
2016
10 10 0
Sadabad et al.
(2016)22 Electronic Physician August, 2015 Mar, 2016 6 6 0
Meheux et al.
(2016)24 Arthroscopy February, 2015 Mar, 2016 6 6 0
Kanchanatawan et al.
(2016)21 Knee Surgery, Sports
Traumatology,
Arthroscopy
August, 2015 May, 2016 9 9 0
Lai et al. (2015)25 PM&R October, 2013 Jun, 2015 8 2 0
Laudy et al. (2015)26 British Journal of Sports
Medicine
June, 2014 May, 2015 10 6 0
Anitua et al. (2014)27 Arthroscopy October, 2013 August, 2014 5 2 0
Chang et al. (2014)28 Archives of Physical
Medicine and
Rehabilitation
September, 2013 Mar, 2014 16 5 0
Khoshbin et al.
(2013)29 Arthroscopy February, 2013 December. 2013 6 4 0
RCTs, randomized clinical trials.
1616 International Journal of Rheumatic Diseases 2017; 20: 1612–1630
D. Xing et al.
T ab
le 2
P ri m a ry
st u d ie s in cl u d e d in
sy st e m a ti c re v ie w s
P ri m a ry
st u d y
S tu d y d e si g n
S h e n et
a l.
(2 0 1 7 )
D a i et
a l.
(2 0 1 6 )
S a d a b a d
et a l.
(2 0 1 6 )
M e h e u x
et a l.
(2 0 1 6 )
K a n ch
a n a ta w a n
et a l. (2 0 1 6 )
L a i et
a l.
(2 0 1 5 )
L a u d y
et a l.
(2 0 1 5 )
A n it u a
et a l.
(2 0 1 4 )
C h a n g
et a l.
(2 0 1 4 )
K h o sh b in
et a l.
(2 0 1 3 )
C e rz a et
a l.
(2 0 1 2 )
R C T
+ +
+ +
+ �
+ �
+ +
D u y m u s et
a l.
(2 0 1 6 )
R C T
+ +
� �
� �
� �
� �
F il a rd o et
a l.
(2 0 1 5 )
R C T
+ +
� +
+ �
� �
� �
F o ro g h et
a l.
(2 0 1 6 )
R C T
+ �
� �
� �
� �
� �
G € o rm
e li et
a l.
(2 0 1 5 )
R C T
+ +
� �
+ �
� �
� �
L i et
a l. (2 0 1 1 )
R C T
+ �
� �
� �
+ �
+ +
P a te l et
a l.
(2 0 1 3 )
R C T
+ +
� +
+ +
+ �
+ +
P a te rs o n et
a l.
(2 0 1 6 )
R C T
+ +
� �
� �
� �
� �
R a e is sa d a t et
a l.
(2 0 1 5 )
R C T
+ +
+ +
+ �
� �
� �
S m it h et
a l.
(2 0 1 6 )
R C T
+ +
� �
� �
� �
� �
S p a k o v a et
a l.
(2 0 1 2 )
R C T
+ �
+ �
� +
+ �
+ +
S a n ch
e z et
a l.
(2 0 1 2 )
R C T
+ +
+ +
+ �
+ +
+ �
V a q u e ri z o et
a l.
(2 0 1 3 )
R C T
+ +
+ +
+ �
+ +
� �
S a n ch
e z et
a l.
(2 0 0 8 )
R e tr o sp e ct iv e
co h o rt st u d y
� �
+ �
� �
� +
� �
F il a rd o et
a l.
(2 0 1 2 )
R C T
� �
� �
+ +
+ �
+ +
R a y e g a n i et
a l.
(2 0 1 4 )
R C T
� �
� �
+ �
� �
� �
K o n et
a l.
(2 0 1 0 )
P ro sp e ct iv e
co h o rt st u d y
� �
� �
� +
� �
+ �
International Journal of Rheumatic Diseases 2017; 20: 1612–1630 1617
Platelet-rich plasma for knee osteoarthritis
T ab
le 2
(c o n ti n u e d )
P ri m a ry
st u d y
S tu d y d e si g n
S h e n et
a l.
(2 0 1 7 )
D a i et
a l.
(2 0 1 6 )
S a d a b a d
et a l.
(2 0 1 6 )
M e h e u x
et a l.
(2 0 1 6 )
K a n ch
a n a ta w a n
et a l. (2 0 1 6 )
L a i et
a l.
(2 0 1 5 )
L a u d y
et a l.
(2 0 1 5 )
A n it u a
et a l.
(2 0 1 4 )
C h a n g
et a l.
(2 0 1 4 )
K h o sh b in
et a l.
(2 0 1 3 )
S a m p so n et
a l.
(2 0 1 0 )
P ro sp e ct iv e
co h o rt st u d y
� �
� �
� +
� �
+ �
K o n et
a l.
(2 0 1 1 )
P ro sp e ct iv e
co h o rt st u d y
� �
� �
� +
+ �
+ +
F il a rd o et
a l.
(2 0 1 1 )
P ro sp e ct iv e
co h o rt st u d y
� �
� �
� +
� �
+ �
G o b b i et
a l.
(2 0 1 2 )
P ro sp e ct iv e
co h o rt st u d y
� �
� �
� +
� �
+ �
S a y et
a l.
(2 0 1 3 )
P ro sp e ct iv e
co h o rt st u d y
� �
� �
� �
+ +
� �
F il a rd o et
a l.
(2 0 1 2 )
P ro sp e ct iv e
co h o rt st u d y
� �
� �
� �
+ +
+ �
W a n g -S a e g u sa
et a l. (2 0 1 1 )
S in g le
a rm
st u d y
� �
� �
� �
� �
+ �
N a p o li ta n o
et a l. (2 0 1 2 )
S in g le
a rm
st u d y
� �
� �
� �
� �
+ �
H a lp e rn
et a l.
(2 0 1 3 )
S in g le
a rm
st u d y
� �
� �
� �
� �
+ �
Ja n g et
a l.
(2 0 1 3 )
S in g le
a rm
st u d y
� �
� �
� �
� �
+ �
R C T , ra n d o m iz e d cl in ic a l tr ia ls .
1618 International Journal of Rheumatic Diseases 2017; 20: 1612–1630
D. Xing et al.
reviews.37 However, it is not uncommon to have several
systematic reviews under the same topic published eval-
uating the same interventions, yet without consistent
conclusions. This also occurred in the study of PRP
injection in treating knee OA. Although several system-
atic reviews had supported PRP injections, the current
guideline is unable to recommend for or against PRP
for knee OA.30
The methodological quality and risk of bias among
individual systematic reviews may account for the dis-
crepancy in outcomes of systematic reviews. The follow-
ing types of biases can be induced when systematic
review is applied at all steps of the review process,
including study eligibility criteria, study selection, data
collection and evidence synthesis. Thus, the decision
makers should put methodological quality and risk of
bias in systematic reviews into consideration when
pooled conclusions are used.33,36 Therefore, the
AMSTAR tool33 was used in the present study to evalu-
ate the methodological quality of systematic reviews
about PRP in treating knee OA, and Jadad decision
algorithm35 was utilized to select the best evidence. Fur-
thermore, in collecting the systematic reviews and
assessing the risk of bias, a newly developed tool,
Table 3 Databases mentioned by included systematic reviews during literature searches
Authors Search database
Medline Embase Cochrane BIOSIS EBSCO Google Scholar Others
Shen et al. (2017)20 + + + � � � + Dai et al. (2016)23 + + + � � � + Sadabad et al. (2016)22 + � + � � � + Meheux et al. (2016)24 + � + � � � + Kanchanatawan et al. (2016)21 + � � � � � + Lai et al. (2015)25 + + � � � � + Laudy et al. (2015)26 + + + � � � + Anitua et al. (2014)27 + + + � � � + Chang et al. (2014)28 + � � � � � + Khoshbin et al. (2013)29 + + + � � � �
Table 4 Methodological characteristics of included systematic reviews
Authors Primary study
design
Level of
evidence
Software Sensitivity
analysis
Subgroup
analysis
Evidence body
assessment
Shen et al.
(2017)20 RCT Level II REVMAN No No No
Dai et al. (2016)23 RCT Level II STATA and REVMAN Yes Yes No
Sadabad et al.
(2016)22 RCT Level II STATA and REVMAN Yes No No
Meheux et al.
(2016)24 RCT Level II N/A N/A N/A N/A
Kanchanatawan
et al. (2016)21 RCT Level II STATA Yes Yes Yes
Lai et al. (2015)25 RCT and non-RCT Level II N/A N/A N/A N/A
Laudy et al.
(2015)26 RCT and non-RCT Level II REVMAN No Yes Yes
Anitua et al.
(2014)27 RCT and non-RCT Level II REVMAN No No No
Chang et al.
(2014)28 RCT and non-RCT Level II STATA Yes Yes No
Khoshbin et al.
(2013)29 RCT and non-RCT Level II REVMAN Yes No No
RCT, randomized clinical trials.
International Journal of Rheumatic Diseases 2017; 20: 1612–1630 1619
Platelet-rich plasma for knee osteoarthritis
T ab
le 5
A M S T A R cr it e ri a fo r in cl u d e d sy st e m a ti c re v ie w s
It e m s
S h e n et
a l.
(2 0 1 7 )
D a i et
a l.
(2 0 1 6 )
S a d a b a d
et a l. (2 0 1 6 )
M e h e u x
et a l. (2 0 1 6 )
K a n ch
a n a ta w a n
et a l. (2 0 1 6 )
L a i et
a l.
(2 0 1 5 )
L a u d y
et a l.
(2 0 1 5 )
A n it u a
et a l.
(2 0 1 4 )
C h a n g
et a l.
(2 0 1 4 )
K h o sh b in
et a l.
(2 0 1 3 )
W a s a p ri o r d e si g n
p ro v id e d ?
1 0
0 1
0 0
0 0
0 0
W a s th e re
d u p li ca te
se le ct io n a n d d a ta
e x tr a ct io n ?
1 1
1 1
0 0
1 1
1 1
W a s a co m p re h e n si v e
li te ra tu re
se a rc h
p re fo rm
e d ?
1 1
1 1
1 1
1 1
0 1
W a s th e st a tu s o f
p u b li ca ti o n u se d a s a n
in cl u si o n cr it e ri o n ?
1 1
1 1
1 1
1 1
1 1
W a s a li st o f in cl u d e d /
e x cl u d e d st u d ie s
p ro v id e d ?
0 0
0 0
0 0
0 0
0 0
W e re
th e p ro fi le s o f th e
in cl u d e d st u d ie s
p ro v id e d ?
1 1
0 1
1 1
1 1
0 1
W a s th e m e th o d o lo g ic a l
q u a li ty
o f th e in cl u d e d
st u d ie s e v a lu a te d a n d
d o cu m e n te d ?
1 1
0 1
1 0
1 1
1 1
W a s th e sc ie n ti fi c q u a li ty
o f th e in cl u d e d st u d ie s
u se d a p p ro p ri a te ly
in
fo rm
u la ti n g co n cl u si o n s?
1 1
1 1
0 0
1 0
1 1
W e re
th e m e th o d s u se d to
co m b in e th e fi n d in g s o f
st u d ie s a p p ro p ri a te ?
1 1
1 1
1 1
1 1
1 0
W a s th e p u b li ca ti o n b ia s
e v a lu a te d ?
0 1
0 0
1 0
0 0
1 0
W e re
th e co n fl ic ts o f
in te re st st a te d ?
1 1
1 1
1 0
1 1
1 1
T o ta l sc o re
9 9
6 9
7 4
8 7
7 7
1620 International Journal of Rheumatic Diseases 2017; 20: 1612–1630
D. Xing et al.
T ab
le 6
H e te ro g e n e it y o f e a ch
o u tc o m e in
in cl u d e d sy st e m a ti c re v ie w s
O u tc o m e s
S h e n et
a l.
(2 0 1 7 )
D a i et
a l.
(2 0 1 6 )
S a d a b a d
et a l.
(2 0 1 6 )
M e h e u x
et a l. (2 0 1 6 )
K a n ch
a n a ta w a n
et a l. (2 0 1 6 )
L a i et
a l.
(2 0 1 5 )
L a u d y
et a l.
(2 0 1 5 )
A n it u a
et a l.
(2 0 1 4 )
C h a n g
et a l.
(2 0 1 4 )
K h o sh b in
et a l.
(2 0 1 3 )
W O M A C p a in
a t
3 m o n th s (P R P vs .
co n tr o l)
9 4 %
W O M A C p a in
a t
6 m o n th s (P R P vs .
co n tr o l)
9 6 %
W O M A C p a in
a t
1 2 m o n th s (P R P vs .
co n tr o l)
8 6 %
W O M A C p a in
a t
6 m o n th s (P R P vs . H A )
9 6 %
N A
W O M A C p a in
a t
1 2 m o n th s (P R P vs . H A )
7 9 %
W O M A C p a in
a t 6 w e e k s
(P R P vs . sa li n e )
N A
W O M A C p a in
a t
3 m o n th s (P R P vs . S a li n e )
N A
W O M A C p a in
a t
6 m o n th s (P R P v s sa li n e )
N A
N A
W O M A C p a in
a t
1 2 m o n th s (P R P vs .
sa li n e )
N A
W O M A C p a in
(< 1 y e a r)
(P R P vs . H A )
9 0 .5 %
W O M A C p a in
(< 1 y e a r)
(P R P vs . sa li n e )
8 5 .5 %
W O M A C fu n ct io n a t
3 m o n th s (P R P v s
co n tr o l)
9 1 %
W O M A C fu n ct io n a t
6 m o n th s (P R P vs .
co n tr o l)
9 7 %
W O M A C fu n ct io n a t
1 2 m o n th s (P R P vs .
co n tr o l)
8 4 %
W O M A C fu n ct io n a t
3 m o n th s (P R P vs . H A )
8 6 %
International Journal of Rheumatic Diseases 2017; 20: 1612–1630 1621
Platelet-rich plasma for knee osteoarthritis
T ab
le 6
(c o n ti n u e d )
O u tc o m e s
S h e n et
a l.
(2 0 1 7 )
D a i et
a l.
(2 0 1 6 )
S a d a b a d
et a l.
(2 0 1 6 )
M e h e u x
et a l. (2 0 1 6 )
K a n ch
a n a ta w a n
et a l. (2 0 1 6 )
L a i et
a l.
(2 0 1 5 )
L a u d y
et a l.
(2 0 1 5 )
A n it u a
et a l.
(2 0 1 4 )
C h a n g
et a l.
(2 0 1 4 )
K h o sh b in
et a l.
(2 0 1 3 )
W O M A C fu n ct io n a t
6 m o n th s (P R P vs . H A )
8 7 %
9 4 %
N A
W O M A C fu n ct io n a t
1 2 m o n th s (P R P vs . H A )
3 1 %
W O M A C fu n ct io n a t
6 w e e k s (P R P vs . sa li n e )
N A
W O M A C fu n ct io n a t
3 m o n th s (P R P vs . sa li n e )
N A
W O M A C fu n ct io n a t
6 m o n th s (P R P vs . sa li n e )
+ N A
W O M A C fu n ct io n a t
1 2 m o n th s (P R P vs .
sa li n e )
+
W O M A C fu n ct io n
(< 1 y e a r)
(P R P v s H A )
9 5 .8 %
W O M A C fu n ct io n
(< 1 y e a r)
(P R P vs .
sa li n e )
9 4 .2 %
W O M A C st if fn e ss
a t
6 m o n th s (P R P v s H A )
N A
W O M A C st if fn e ss
(< 1 y e a r)
(P R P vs . H A )
9 2 .9 %
W O M A C st if fn e ss
(< 1 y e a r)
(P R P vs .
sa li n e )
�
T o ta l W O M A C a t
3 m o n th s (P R P vs .
co n tr o l)
9 0 %
T o ta l W O M A C a t
6 m o n th s (P R P vs .
co n tr o l)
9 7 %
8 9 %
T o ta l W O M A C a t
1 2 m o n th s (P R P vs .
co n tr o l)
8 5 %
T o ta l W O M A C (<
1 y e a r)
(P R P vs . H A )
9 3 %
9 6 .6 %
T o ta l W O M A C (<
1 y e a r)
(P R P vs . sa li n e )
9 3 .6 %
1622 International Journal of Rheumatic Diseases 2017; 20: 1612–1630
D. Xing et al.
T ab
le 6
(c o n ti n u e d )
O u tc o m e s
S h e n et
a l.
(2 0 1 7 )
D a i et
a l.
(2 0 1 6 )
S a d a b a d
et a l.
(2 0 1 6 )
M e h e u x
et a l. (2 0 1 6 )
K a n ch
a n a ta w a n
et a l. (2 0 1 6 )
L a i et
a l.
(2 0 1 5 )
L a u d y
et a l.
(2 0 1 5 )
A n it u a
et a l.
(2 0 1 4 )
C h a n g
et a l.
(2 0 1 4 )
K h o sh b in
et a l.
(2 0 1 3 )
L e q u e sn e sc o re
a t
3 m o n th s (P R P vs . H A )
N A
L e q u e sn e sc o re
a t
6 m o n th s (P R P vs . H A )
9 0 %
N A
L e q u e sn e sc o re
a t
1 2 m o n th s (P R P vs . H A )
N A
L e q u e sn e sc o re
(< 1 y e a r)
(P R P vs . H A )
9 7 %
V A S p a in
a t 6 m o n th s
(s in g le
P R P v s sa li n e )
N A
V A S p a in
a t 6 m o n th s (2
P R P vs . sa li n e )
N A
V A S p a in
a t 3 m o n th s
(P R P vs . H A )
2 1 %
N A
V A S p a in
a t 6 m o n th s
(P R P vs . H A )
0 %
V A S p a in
a t 6 m o n th s
(P R P vs . co n tr o l)
8 8 %
IK D C sc o re s (<
1 y e a r)
(P R P vs . H A )
9 0 .7 %
IK D C sc o re s a t 6 m o n th s
(P R P vs . H A )
N A
4 4 %
K O O S sc o re s a t 6 m o n th s
(P R P vs . H A )
N A
E Q -V A S sc o re
(< 1 y e a r)
(P R P vs . H A )
7 9 .9 %
T h e p o o le d W O M A C ,
IK D C , a n d L e q u e sn e
sc o re
a t 6 m o n th s (P R P
vs . H A )
9 5 %
T h e p o o le d W O M A C ,
IK D C , a n d L e q u e sn e
sc o re
a t 1 2 m o n th s (P R P
vs . H A )
9 4 %
E ff e ct si z e o f fu n ct io n a l
ch a n g e a t 2 m o n th s
9 7 .3 %
E ff e ct si z e o f fu n ct io n a l
ch a n g e a t 6 m o n th s
9 6 .3 %
International Journal of Rheumatic Diseases 2017; 20: 1612–1630 1623
Platelet-rich plasma for knee osteoarthritis
T ab
le 6
(c o n ti n u e d )
O u tc o m e s
S h e n et
a l.
(2 0 1 7 )
D a i et
a l.
(2 0 1 6 )
S a d a b a d
et a l.
(2 0 1 6 )
M e h e u x
et a l. (2 0 1 6 )
K a n ch
a n a ta w a n
et a l. (2 0 1 6 )
L a i et
a l.
(2 0 1 5 )
L a u d y
et a l.
(2 0 1 5 )
A n it u a
et a l.
(2 0 1 4 )
C h a n g
et a l.
(2 0 1 4 )
K h o sh b in
et a l.
(2 0 1 3 )
E ff e ct
si z e o f fu n ct io n a l
ch a n g e a t 1 2 m o n th s
9 8 .6 %
P a ti e n t sa ti sf a ct io n a t
6 m o n th s (s in g le
P R P vs .
sa li n e )
N A
P a ti e n t sa ti sf a ct io n a t
6 m o n th s (2
P R P vs .
sa li n e )
N A
P a ti e n t sa ti sf a ct io n a t
6 m o n th s (P R P vs .
co n tr o l)
9 0 %
A d v e rs e e v e n ts (P R P vs .
co n tr o l)
5 9 %
A d v e rs e e v e n ts (P R P vs .
H A )
6 6 %
� �
A d v e rs e e v e n ts (P R P vs .
sa li n e )
7 3 %
3 6 .1 %
A d v e rs e e v e n ts (s in g le
P R P vs . sa li n e )
N A
A d v e rs e e v e n ts (2
P R P vs .
sa li n e )
N A
W O M A C , W e st e rn
O n ta ri o a n d M cM
a st e r U n iv e rs it ie s O st e o a rt h ri ti s In d e x ; P R P , p la te le t- ri ch
p la sm
a ; H A , h y a lu ro n ic
a ci d ; V A S , v is u a l a n a lo g sc a le ; IK D C , In te rn a ti o n a l K n e e D o cu m e n ta -
ti o n C o m m it te e ; K O O S , K n e e In ju ry
a n d O st e o a rt h ri ti s O u tc o m e S co re ; E Q -V A S , E u ro Q o l – v is u a l a n a lo g sc a le ; N A , n o t a v a il a b le .
1624 International Journal of Rheumatic Diseases 2017; 20: 1612–1630
D. Xing et al.
Figure 2 Results of each included systematic review with quantitative synthesis. Red means favoring platelet-rich plasma, green means no difference, yellow means not reporting, and blue means favoring placebo. Arabic numerals mean the number of included primary studies.
International Journal of Rheumatic Diseases 2017; 20: 1612–1630 1625
Platelet-rich plasma for knee osteoarthritis
ROBIS (www.robis-tool.info) was used.36 Ultimately,
two systematic reviews23,24 with lower risk of bias and
offering the best evidence were selected in the present
study.
Dai et al.23 included 10 RCTs in the meta-analysis
and reported that PRP had similar effects like HA in
pain relief and function improvement at 6 months, but
significantly better pain relief and function improve-
ment at 12 months compared with HA. At 6 and
12 months, PRP was associated with significantly better
pain relief and function improvement compared with
saline. PRP did not increase the risk of adverse events
compared with HA and saline. In the aspect of risk of
bias, the following three domains in the ROBIS tool
had high risk of bias: study eligibility criteria, identifica-
tion and selection of studies, and synthesis and find-
ings. However, these three domains were appropriately
considered during results interpreting and conclusion
making. The results without significance were also
emphasized by the reviewers. Thus, this systematic
review was rated as low risk of bias.
Meheux et al.24 conducted a systematic review via
best-evidence synthesis instead of meta-analysis as a
result of high heterogeneity among outcomes in each
primary study. Clinical heterogeneity was induced, to a
certain degree, by preparation of PRP, concentration of
platelets, severity of OA, PRP spinning approach, dura-
tion of spin, and number or volume of injections. The
clinical heterogeneity cannot be absolutely resolved.
Among the primary studies, the different PRP systems
were classified according to the PAW classification sys-
tem38 that looks at platelet concentration, activation
method and white blood cell (WBC) count. Because of
the limited number of the primary studies, subgroup
analysis cannot be conducted according to the PAW
classification. Thus, the qualitative analysis was con-
ducted with five levels of evidence based on the quality
and results of the primary studies.39 Six RCTs were
Figure 3 Flow diagram of Jadad decision algorithm.
1626 International Journal of Rheumatic Diseases 2017; 20: 1612–1630
D. Xing et al.
included in this systematic review and were summa-
rized by descriptive analysis. All primary studies except
Filardo et al.40 used leukocyte-poor PRP and showed
significant clinical and statistical improvements on
Western Ontario and McMaster Universities
Osteoarthritis Index (WOMAC) scores between HA and
PRP or HA and placebo groups. All primary studies
used different PRP preparations with three of six using
Table 7 Risk of bias assessment of systematic reviews using ROBIS tool
Figure 4 Risk of bias of the included systematic reviews with ROBIS tool. The ROBIS tool incorporates the assessment of study eli- gibility criteria, identification and selection of studies, data collection and study appraisal, and synthesis and findings. The overall risk of bias is determined based on the above four domains. Each risk of bias item is presented as the percentage across all the sys- tematic reviews, which indicates the proportion of different levels of risk of bias for each item.
International Journal of Rheumatic Diseases 2017; 20: 1612–1630 1627
Platelet-rich plasma for knee osteoarthritis
calcium chloride activator, one of six using leukocyte-
rich PRP, four of six using the single spin approach, and
two of six using the double spin approach. No primary
study directly compared leukocyte-poor PRP to leuko-
cyte-rich PRP. Therefore, the systematic review con-
cluded that PRP would lead to significant clinical
improvements up to 12 months after injection. Leuko-
cyte-poor PRP had better clinical outcomes and
WOMAC scores compared with HA at 3–12 months after intervention. However, the authors did not draw
the conclusion in leukocyte-rich versus leukocyte-poor
PRP due to limited evidence. Although two of the
ROBIS domains were at risk of bias, the systematic
review as a whole had low risk of bias due to full discus-
sion in the interpretation of review findings and consid-
ered limitations.
The strength of the present study is the combined
and simultaneous utilization of ROBIS tool, AMSTAR
instrument and Jadad decision algorithm for evaluating
the risk of bias and quality of methodology of the sys-
tematic reviews. The ultimate purpose is to help deci-
sion makers select the best evidence with low risk of
bias in terms of PRP injection for knee OA from discor-
dant systematic reviews. Thus, the conclusion of the
present study is consistent with the finding of the two
aforementioned systematic reviews. Hence, based on
the existing optimal evidence, we can take recommen-
dations that intra-articular PRP injection might be more
efficacious than HA in treating knee OA in terms of
pain relief and function improvement over HA within
12 months. However, due to the limitations of current
evidence and clinical heterogeneity among primary
studies, we cannot draw the conclusion that which PRP
systems are most effective for knee OA.
The limitations of the present study include the fol-
lowing. (i) English language systematic reviews were
included. Non-English language literatures could have
been omitted, leading to language bias. (ii) The quality
of methodology of the primary studies may radically
influence the results of included systematic reviews.
Although we assessed risk of bias and quality of the
included systematic reviews, the limitations of primary
studies, especially conflicts of interests, should be con-
sidered when the results are interpreted. (iii) The pri-
mary studies included in other systematic reviews
which were not regarded as the best evidence with low
risk of bias could be considered when making clinical
decisions.
In this overview of overlapping systematic reviews
investigating efficacy and safety of PRP for knee OA, the
best available evidence with low risk of bias suggested
that PRP is an effective intervention for knee OA with-
out increased risk of adverse events. Therefore, we have
relative confidence to recommend the use of PRP in
treating knee OA. To some extent, the OA guideline
developers can refer to the present overview when
developing clinical practice guidelines in terms of PRP.
AUTHOR CONTRIBUTIONS
Project conceptualization: D Xing and JH Lin. Study
design: D Xing, B Wang and JH Lin. Data collection/val-
idation: D Xing, B Wang, ZY Yang and JH Lin. Data
analysis: D Xing, B Wang and YF Hou. Result interpreta-
tion: D Xing, B Wang, Q Liu and YF Hou. Reporting &
editing: D Xing, YF Hou and W Zhang. Final approval
of the version to be submitted: D Xing, B Wang and JH
Lin. Project guarantor: YL Chen and JH Lin.
FUNDING
This study was supported by grants from the National
Natural Science Foundation of China (No. 81501919).
CONFLICT OF INTEREST
The authors declare that they have no conflicts of
interest.
REFERENCES
1 Sacks JJ, Luo YH, Helmick CG (2010) Prevalence of speci-
fic types of arthritis and other rheumatic conditions in the
ambulatory health care system in the United States, 2001– 2005. Arthritis Care Res (Hoboken) 62, 460–4.
2 Lawrence RC, Felson DT, Helmick CG et al. (2008) Esti-
mates of the prevalence of arthritis and other rheumatic
conditions in the United States. Part II. Arthritis Rheum 58 (1), 26–35.
3 Helmick CG, Felson DT, Lawrence RC et al. (2008) Esti-
mates of the prevalence of arthritis and other rheumatic
conditions in the United States. Part I. Arthritis Rheum 58 (1), 15–25.
4 Nyvang J, Hedstr€om M, Gleissman SA (2016) It’s not just
a knee, but a whole life: A qualitative descriptive study on
patients’ experiences of living with knee osteoarthritis and
their expectations for knee arthroplasty. Int J Qual Stud
Health Well-being 11, 30193. 5 Crawford DC, Miller LE, Block JE (2013) Conservative
management of symptomatic knee osteoarthritis: a flawed
strategy. Orthop Rev (Pavia) 5 (1), e2. 6 Cole BJ, Seroyer ST, Filardo G et al. (2010) Platelet-rich
plasma: where are we now and where are we going. Sports
Health 2, 203–10.
1628 International Journal of Rheumatic Diseases 2017; 20: 1612–1630
D. Xing et al.
7 Andia I, S�anchez M, Maffulli N (2012) Joint pathology
and platelet-rich plasma therapies. Expert Opin Biol Ther
12 (1), 7–22. 8 Eppley BL, Woodell JE, Higgins J (2004) Platelet quantifi-
cation and growth factor analysis from platelet-rich
plasma: implications for wound healing. Plast Reconstr
Surg 114, 1502–8. 9 Kon E, Filardo G, Di MA, Marcacci M (2011) Platelet-
rich plasma (PRP) to treat sports injuries: evidence to
support its use. Knee Surg Sports Traumatol Arthrosc 19, 516–27.
10 Cugat R, Cusc�o X, Seijas R et al. (2015) Biologic enhance-
ment of cartilage repair: the role of platelet-rich plasma
and other commercially available growth factors. Arthro-
scopy 31, 777–83. 11 Metcalf KB, Mandelbaum BR, McIlwraith CW (2013)
Application of platelet-rich plasma to disorders of the
knee joint. Cartilage 4, 295–312. 12 Padilla S, Orive G, Sanchez M, Anitua E, Hsu WK (2014)
Platelet-rich plasma in orthopaedic applications: evidence-
based recommendations for treatment. J Am Acad Orthop
Surg 22, 469–70. 13 Smyth NA, Murawski CD, Fortier LA, Cole BJ, Kennedy JG
(2013) Platelet-rich plasma in the pathologic processes of
cartilage: review of basic science evidence. Arthroscopy 29, 1399–409.
14 Mascarenhas R, Saltzman BM, Fortier LA, Cole BJ (2015)
Role of platelet-rich plasma in articular cartilage injury
and disease. J Knee Surg 28 (1), 3–10. 15 Xie X, Zhang C, Tuan RS (2014) Biology of platelet-rich
plasma and its clinical application in cartilage repair.
Arthritis Res Ther 16 (1), 204. 16 Kobayashi Y, Saita Y, Nishio H et al. (2016) Leukocyte
concentration and composition in platelet-rich plasma
(PRP) influences the growth factor and protease concen-
trations. J Orthop Sci 21, 683–9. 17 Pavlovic V, Ciric M, Jovanovic V et al. (2016) Platelet rich
plasma: a short overview of certain bioactive components.
Open Med (Wars) 11, 242–7. 18 Xu Z, Yin W, Zhang Y et al. (2017) Comparative evalua-
tion of leukocyte- and platelet-rich plasma and pure plate-
let-rich plasma for cartilage regeneration. Sci Rep 7, 43301. 19 Marx RE (2001) Platelet-rich plasma (PRP): what is PRP
and what is not PRP. Implant Dent 10, 225–8. 20 Shen L, Yuan T, Chen S, Xie X, Zhang C (2017) The tem-
poral effect of platelet-rich plasma on pain and physical
function in the treatment of knee osteoarthritis: systematic
review and meta-analysis of randomized controlled trials.
J Orthop Surg Res 12 (1), 16. 21 Kanchanatawan W, Arirachakaran A, Chaijenkij K et al.
(2016) Short-term outcomes of platelet-rich plasma injec-
tion for treatment of osteoarthritis of the knee. Knee Surg
Sports Traumatol Arthrosc 24, 1665–77. 22 Sadabad HN, Behzadifar M, Arasteh F, Behzadifar M, Deh-
ghan HR (2016) Efficacy of platelet-rich plasma versus
hyaluronic acid for treatment of knee osteoarthritis: a sys-
tematic review and meta-analysis. Electron Physician 8, 2115–22.
23 Dai WL, Zhou AG, Zhang H, Zhang J (2017) Efficacy of
platelet-rich plasma in the treatment of knee osteoarthri-
tis: a meta-analysis of randomized controlled trials. Arthro-
scopy 33, 659–70. 24 Meheux CJ, McCulloch PC, Lintner DM, Varner KE, Harris
JD (2016) Efficacy of intra-articular platelet-rich plasma
injections in knee osteoarthritis: a systematic review.
Arthroscopy 32, 495–505. 25 Lai LP, Stitik TP, Foye PM, Georgy JS, Patibanda V, Chen B
(2015) Use of platelet-rich plasma in intra-articular knee
injections for osteoarthritis: a systematic review. PMR 7, 637–48.
26 Laudy AB, Bakker EW, Rekers M, Moen MH (2015) Effi-
cacy of platelet-rich plasma injections in osteoarthritis of
the knee: a systematic review and meta-analysis. Br J Sports
Med 49, 657–72. 27 Anitua E, Sanchez M, Aguirre JJ, Prado R, Padilla S, Orive
G (2014) Efficacy and safety of plasma rich in growth fac-
tors intra-articular infiltrations in the treatment of knee
osteoarthritis. Arthroscopy 30, 1006–17. 28 Chang KV, Hung CY, Aliwarga F, Wang TG, Han DS, Chen
WS (2014) Comparative effectiveness of platelet-rich
plasma injections for treating knee joint cartilage degener-
ative pathology: a systematic review and meta-analysis.
Arch Phys Med Rehabil 95, 562–75. 29 Khoshbin A, Leroux T, Wasserstein D et al. (2013) The
efficacy of platelet-rich plasma in the treatment of symp-
tomatic knee osteoarthritis: a systematic review with quan-
titative synthesis. Arthroscopy 29, 2037–48. 30 Brown GA (2013) AAOS clinical practice guideline: treat-
ment of osteoarthritis of the knee: evidence-based guide-
line, 2nd edition. J Am Acad Orthop Surg 21, 577–9. 31 Liberati A, Altman DG, Tetzlaff J et al. (2009) The PRISMA
statement for reporting systematic reviews and meta-ana-
lyses of studies that evaluate healthcare interventions:
explanation and elaboration. BMJ 339, b2700. 32 Robinson KA, Chou R, Berkman ND et al. (2016) Twelve
recommendations for integrating existing systematic
reviews into new reviews: EPC guidance. J Clin Epidemiol
70, 38–44. 33 Shea BJ, Grimshaw JM, Wells GA et al. (2007) Develop-
ment of AMSTAR: a measurement tool to assess the
methodological quality of systematic reviews. BMC Med
Res Methodol 7, 10. 34 Shea BJ, Bouter LM, Peterson J et al. (2007) External vali-
dation of a measurement tool to assess systematic reviews
(AMSTAR). PLoS One 2, e1350. 35 Jadad AR, Cook DJ, Browman GP (1997) A guide to inter-
preting discordant systematic reviews. CMAJ 156, 1411–6. 36 Whiting P, Savovi�c J, Higgins JP et al. (2016) ROBIS: a
new tool to assess risk of bias in systematic reviews was
developed. J Clin Epidemiol 69, 225–34.
International Journal of Rheumatic Diseases 2017; 20: 1612–1630 1629
Platelet-rich plasma for knee osteoarthritis
37 Young D (2005) Policymakers, experts review evidence-
based medicine. Am J Health Syst Pharm 62, 342–3. 38 DeLong JM, Russell RP, Mazzocca AD (2012) Platelet-rich
plasma: the PAW classification system. Arthroscopy 28, 998–1009.
39 Owens DK, Lohr KN, Atkins D et al. (2010) AHRQ series
paper 5: grading the strength of a body of evidence when
comparing medical interventions – agency for healthcare research and quality and the effective health-care program.
J Clin Epidemiol 63, 513–23. 40 Filardo G, Kon E, Di MA et al. (2012) Platelet-rich plasma
vs hyaluronic acid to treat knee degenerative pathology:
study design and preliminary results of a randomized con-
trolled trial. BMC Musculoskelet Disord 13, 229.
1630 International Journal of Rheumatic Diseases 2017; 20: 1612–1630
D. Xing et al.
Copyright of International Journal of Rheumatic Diseases is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.