International Journal of Surgery 39 (2017) 95e103

Contents lists available at ScienceDirect

International Journal of Surgery

journal homepage: www.journal-surgery.net

Review

Efficacy and safety of intraarticular hyaluronic acid and corticosteroid

for knee osteoarthritis: A meta-analysis
Wei-wei He a, b, 1, Ming-jie Kuang a, b, 1, Jie Zhao a, b, Lei Sun a, Bin Lu a, Ying Wang a,
Jian-xiong Ma a, **, Xin-long Ma a, c, *
a
Biomechanics Labs of Orthopaedics Institute, Tianjin Hospital, Tianjin, 300050, People's Republic of China
b
Tianjin Medical University, Tianjin, 300070, People's Republic of China
c
Department of Orthopedics, Tianjin Hospital, Tianjin, 300211, People's Republic of China

h i g h l i g h t s

Intraarticular corticosteroid is more effective on pain relief than intraarticular hyaluronic acid in short term (up to 1 month), while it reverses up to 6
months.

Intraarticular corticosteroid and hyaluronic acid benefit similarly for knee function improvement.

Both two methods are relatively safe.

a r t i c l e i n f o a b s t r a c t

Article history: Objective: A meta analysis to compare efficacy and safety of intraarticular hyaluronic acid (HA) and
Received 17 October 2016 intraarticular corticosteroids (CS) in patients with knee osteoarthritis.
Received in revised form Method: Potential studies were searched from the electronic databases included PubMed, Embase, web
20 January 2017
of science and the Cochrane Library up to August 2016. High quality randomized controlled trials (RCTs)
Accepted 23 January 2017
Available online 27 January 2017
were selected based on inclusion criteria. RevMan 5.3 were used for the meta-analysis.
Results: 12 RCTs containing 1794 patients meet the inclusion criteria. Visual analog scale (VAS) score in
CS group decrease more than HA group up to 1 month (p ¼ 0.03) and it shows equal efficacy at 3 months
Keywords:
Intraarticular hyaluronic acid
(p ¼ 0.29); HA is more effective than CS at 6 months (p ¼ 0.006). To Western Ontario and McMaster
Corticosteroids Universities Osteoarthritis Index (WOMAC) score, there is no significant difference for two groups at 3
Knee osteoarthritis months (p ¼ 0.29); HA shows greater relative effect than CS at 6 months (p ¼ 0.005). No significant
Meta-analysis difference is found on proportion of rescue medication use after initiation of treatment (p ¼ 0.58) and
proportion of withdrawal for knee pain (p ¼ 0.54). HA and CS exhibit equal efficacy on improvement of
active range of knee flexion at 3 months (p ¼ 0.73) and 6 months (p ¼ 0.43). More topical adverse effects
occurred in intraarticular HA group when compared with intraarticular CS group.
Conclusion: Intraarticular CS is more effective on pain relief than intraarticular HA in short term (up to 1
month), while HA is more effective in long term (up to 6 months). Two therapies benefit similarly for
knee function improvement. Both two methods are relatively safe, but intraarticular HA causes more
topical adverse effects compared with intraarticular CS.
© 2017 IJS Publishing Group Ltd. Published by Elsevier Ltd. All rights reserved.

1. Introduction
* Corresponding author. No. 155, Munan Road, Heping District, Tianjin, 300050,
People's Republic of China. Knee osteoarthritis (OA) is a clinical syndrome of joint pain
** Corresponding author. No. 155, Munan Road, Heping District, Tianjin, 300050,
accompanied by varying degrees of functional limitation and
People's Republic of China.
E-mail addresses: 2010021204@tmu.edu.cn (W.-w. He), doctorkmj@tmu.edu.cn decreased quality of life, characterised by loss of cartilage, remod-
(M.-j. Kuang), zhaoj_91@163.com (J. Zhao), sunleigys@163.com (L. Sun), eling of adjacent bone and associated with inflammation [1]. A
578794146@qq.com (B. Lu), 337533607@qq.com (Y. Wang), mbiomechanics@126. large proportion of US adults aged 60 and older have radiographic
com (J.-x. Ma), maxinlong432@sina.com (X.-l. Ma).
1
knee OA (37.4%) and symptomatic radiographic knee OA (12.1%); or
These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.ijsu.2017.01.087
1743-9191/© 2017 IJS Publishing Group Ltd. Published by Elsevier Ltd. All rights reserved.
96 W.-w. He et al. / International Journal of Surgery 39 (2017) 95e103
an estimated 13.3 million persons have radiographic knee OA, 4.3
million persons suffer symptomatic radiographic knee OA [2]. Knee
OA is a progressive disease, knee pain and loss of function are main
symptoms, which can lead to disability, joint replacement and low
quality of life. In advance of effective disease-modifying medical
interventions for knee OA, treatments are mainly symptomatic
essentially, mainly including intraarticular HA and intraarticular CS
[3].
HA is a natural glycosaminoglycan and a component of synovial
fluid which act as a lubricant and elastic shock absorber during
joint movements. Both concentration and molecular weight of HA
decreases in osteoarthritic joints [4]. Intraarticular HA can restore
the effect of synovial fluid, protect against cartilage erosion, reduce
synovial inflammation [4,5]. Moreover, HA have direct and indirect
analgesic effect on the joints [5]. Although numerous clinical
studies [6e9] and systematic review [10] have shown benefits (e.g.,
improvements in knee pain and function, longer time to knee
arthroplasty) on knee OA, several studies [11e13] reported
increased risk of local or serious adverse events after viscosupple-
mentation of the knee. The safety of intraarticular HA remains
controversial.
Intraarticular CS is a long-standing treatment for OA, and the
first clinical trial of intraarticular CS in knee OA was performed in
1958 by Miller and colleagues [14]. Corticosteroids have marked
anti-inflammatory and immunosuppressive effect, besides, CS can
increase both relative viscosity and concentration of HA [15] in
arthritic knee. While there is a debate on the effective time of
intraarticular CS. Some studies [14,16,17] suggest a short-term (up
to 12 weeks) effect for knee OA, whereas there are also papers
[18,19] report that a significant improvement can be sustained up to
24 weeks.
In spite of the efficacy and cost-effectiveness of treatment
modalities for knee OA are frequently debated and guidelines
have changed over time, intraarticular CS and HA remain com-
mon treatment for knee OA [3]. Among the efficacy and safety
between intraarticular HA and intraarticular CS, there still haven't
reached a consencus. In a recent network meta-analysis, Trojian
et al. [20] propose that intraarticular HA is superior to intra-
articular CS. Conflicts also exist in guidelines. Osteoarthritis
Research Society International [21] recommend both intra-
articular HA and intraarticular CS for patients with knee OA, while
American College of Rheumatology [22] and National Institute for
Health and Care [1] just recommend the use of intraarticular CS.
Thus, we conduct a meta analysis to compare efficacy and safety
of intraarticular HA and intraarticular CS in patients with knee
OA.
2. Materials and methods
2.1. Search strategy
Two reviewers performed an electronic literature search for
randomized controlled trials (RCTs) comparing the efficacy or
safety of intraarticular hyaluronic acid injections with intraarticular
corticosteroid injections in the management of knee osteoarthritis.
The electronic databases include PubMed, Embase, web of science
and the Cochrane Library up to August 2016. The following terms
were used as key words:knee osteoarthritis, gonarthrosis, hyal-
uronic acid (and trade names for hyaluronic acid), viscosupple-
mentation and corticosteroid (and the trade name for
corticosteroid).
In addition, further articles were obtained by reviewing refer-
ences of the selected articles. The detail retrieval process is shown
in Fig. 1.
2.2. Inclusion criteria
We included published RCTs that used human beings and
compared the efficacy or safety of intraarticular HA with intra-
articular CS to treat knee OA. Each RCT wad required to contain at
least one outcome, including the visual analog scale (VAS), Western
Ontario and McMaster Universities Osteoarthritis Index (WOMAC),
proportion of rescue medication use after initiation of treatment,
proportion of withdrawal for knee pain, range of motion of the
knee, and adverse events. Eligible studies were assessed indepen-
dently by two authors. In case of disagreement, a consensus was
reached through discussion between two authors.
2.3. Data extraction
Two reviewers independently retrieve the relevant data from
articles using a standard data extraction form. The publication date,
author, study design, number and demographics of participants,
HA/CS dose, regimen and frequency, withdrawal rate, follow up
time, and outcome measures were extracted for each trial. Where
necessary, means and standard deviation were approximated from
the figures in the studies. Besides, we calculated missing standard
deviations from other available data such as standard errors, or the
formula in Cochrane Handbook for Systematic Reviews of In-
terventions [23]. The data were extracted independently by two
reviewers, and any disagreement was discussed until a consensus
was reached.
2.4. Quality assessment
Two reviewers independently assessed the quality of the RCTs
by using modified Jadad scale [24,25]. Maximum score is 7 points,
studies with a total score of 3 points were considered as low
quality studies, while a total score of 4 points were considered
high-quality. It includes generation of allocation sequence, alloca-
tion concealment, blind method, and description of withdrawals
and drop-outs of the RCTs.
2.5. Statistical and subgroup analysis
We used Review Manager Software for windows (Version 5.3.
Copenhagen:The Nordic Cochrane Centre, The Cochrane Collabo-
ration, 2014) to perform the meta-analysis. For continuous variable
outcomes, mean difference (MD) and 95% confidence interval (CI)
were used to assess it. For dichotomous outcomes, relative risks
(RR) with a 95% CI were presented. Heterogeneity between studies
was assessed by I2 and c2 test. While I2<50% and P>0.1, we used a
fixed-effects model to evaluate, otherwise, a random-effects was
used. Besides, subgroup analysis was performed to explore the
source of heterogeneity when heterogeneity existed.
3. Results
3.1. Search results
From the databases and other sources (e.g.references), we
identified a total of 105 studies, of which 77 were excluded by title
and abstract. Among the rest of 28 studies, 14 studies [26e39] meet
the inclusion criteria after reading the full text in details. Never-
theless, two [38,39] of the 14 studies didn't report sufficient in-
formation for data extraction and analysis. Therefore, the meta-
analysis was performed on the basis of 12 studies (Fig. 1). The
included 12 studies were all RCTs and published between 1995 and
2016. A total of 1794 participants (673 males, 1121 females) were
included. Overall, 971 participants were randomly allocated to HA
 W.-w. He et al. / International Journal of Surgery 39 (2017) 95e103
group and 823 participants were in CS group. The characteristic of
the participants and the 12 studies are shown in Tables 1a and 1b.
Of the 12 RCTs included in analysis, one compared Hyalgan (Fidia
SPA,20 mg, 5-weekly) with triamcinolone hexacetonide (20 mg,
single injection) [26], one compared Hyalgan (Fidia, Padua, Italy,
20 mg, 5-weekly) with methylprednisolone acetate (Depo-Medrol,
Pharmacia & Upjohn, Milan, Italy,40 mg, 3-weekly) [27], one
compared sodium hyaluronate (Orthovisc, Anika Therapeutics, Inc,
Woburn, MA,30 mg, 3-wekly) with 6-methylprednisolone acetate
(Depo-Medrol, Eczacibaşi, Istanbul, Turkey,40 mg, 3-weekly) [28],
one compared hylan G-F 20(Synvisc®,16 mg, 3-weekly) with
triamcinolone hexacetonide (Aristospan®,40 mg, single injection)
[29], one compared sodium hyaluronate (Ostenil®,20 mg, 5-
weekly) with triamcinolone acetonide (Volon A10®,10 mg, 5-
weekly) [30], one compared non-animal stabilized hyaluronic
acid (NASHA)(Durolane®,60 mg, single injection) with triamcino-
lone acetonide (VolonA10®,10 mg, single injection) [31], one
compared sodium hyaluronate (Artzdispo; Kaken Pharmaceutical,
Tokyo, Japan,25 mg, 5-weekly) with decadron (Banyu Pharmaceu-
tical, Tokyo, Japan, 4 mg, single injection) [32], one compared
hylastan SGL-80 (Jonexa™, Genzyme Biosurgery, Cambridge, MA,
USA,4 or 8 ml,1 or 2-weekly) with methylprednisolone acetate
(40 mg, single injection) [33], one compared NASHA (Q-Med AB,
Uppsala, Sweden, 60 mg, single injection) with methylprednisolone
acetate (40 mg, single injection) [34], one compared hyaluronic
Acid (Fidia Farmaceutici S. p.A, Italy, 2 ml, single injection) with
corticosteroid (40 mg, single injection) [35], one compared HYADD
Fig. 1. The search results and selection procedure.
97
4(Fidia Farmaceutici Spa., Abano Terme, Italy, 2-weekly) with 6-
Methylprednisolone acetate (Depo-Medrol, Pfizer, New York, USA,
40 mg, 2-weekly) [36], and the final study compared hylan G-F
20(Synvisc; Genzyme Biosurgery, 6 ml, single injection) with
triamcinolone acetonide (40 mg, single injection) [37]. Details of
the studies are shown in Tables 1a and 1b. As for quality of the 12
studies, all trials were randomized design. Six RCTs reported allo-
cation concealment [30,33e37], while others were unclear. Seven
RCTs adopted double-blinded method [26,30,31,33e35,37], two
RCTs adopted single-blinded method [29,36]. while one RCT chose
an open design [28], and the blind method of the rest two RCTs
[27,32] were not clear. The points of included 12 RCTs assessed by
modified Jadad scale were also shown in Tables 1a and 1b.
3.2. Meta-analysis
3.2.1. VAS score
Data from six studies [26e28,32,35,37] including 484 patients
reported VAS score of target knee at 1 month after treatment.
Heterogeneity exists between the six studies (Chi2 ¼ 14.87, df ¼ 5,
P ¼ 0.01, I2 ¼ 66%; Fig. 2). So, we adopt a random-effects model. It
shows that intraarticular CS reduced VAS score more than intra-
articular HA, and the difference is statistically significant
(MD ¼ 0.67,95%CI:0.07 to 1.27, p ¼ 0.03, Fig. 2). A subgroup analysis
is performed for VAS score at 1month (Table 2).
Eight studies [26,28e31,35e37] enrolling 800 patients reported
VAS score of target knee at 3 months after treatment. There is
98 W.-w. He et al. / International Journal of Surgery 39 (2017) 95e103

significant heterogeneity between the eight studies (Chi2 ¼ 46.04,
df ¼ 7, P < 0.00001, I2 ¼ 85%; Fig. 2). Therefore, a random-effects
model is used. The effect size (MD: 0.46, 95%CI: 1.31 to 0.39,
p ¼ 0.29; Fig. 2) suggests equal efficacy.
Seven studies [26e29,32,36,37] including 646 patients reported
VAS score of target knee at 6 months after injection. Heterogeneity
exists between the seven studies (Chi2 ¼ 13.67, df ¼ 6, P ¼ 0.03,
I2 ¼ 56%; Fig. 2). Thus, a random-effects model is performed. The
effect size (MD: 0.73,95%CI: 1.25 to 0.21, p ¼ 0.006; Fig. 2) favors
the intraarticular HA.
3.2.2. WOMAC score
WOMAC score at 3 months was shown in five studies
[29,34e37] enrolling 1002 patients. There is significant heteroge-
neity (Chi2 ¼ 17.31, df ¼ 4, P ¼ 0.002, I2 ¼ 77%; Fig. 3), so we conduct
a random-effects model. Intraarticular HA is found to reduce
WOMAC score more than intraarticular CS, while it's not statisti-
cally significant (MD: 2.3,95%CI: 6.53 to 1.93, p ¼ 0.29; Fig. 3).
WOMAC score at 6 months was shown in four studies
[29,34,36,37] including 848 patients. Statistical heterogeneity ex-
ists in the four studies (Chi2 ¼ 10.01, df ¼ 3, P ¼ 0.02, I2 ¼ 70%;
Fig. 3).
Thus, a random-effects model is used. The effect size (MD:
5.51,95%CI: 8.77 to 1.54, p ¼ 0.005; Fig. 3) favors the intraarticular
HA.
3.2.3. Proportion of rescue medication use
Data from three studies [27e29] enrolling 369 patiens
mentioned proportion of rescue medication use. No statistical dif-
ference is observed between the two groups (RR:1.04,95%CI:0.9 to
1.2, p ¼ 0.58; Fig. 4).
3.2.4. Proportion of withdrawal for knee pain
Five studies [26,28,30e32] including 286 patients mentioned
proportion of withdrawal for knee pain. No significant statistical
difference is observed between the two groups (RR:1.29,95%CI:0.57
to 2.92, p ¼ 0.54; Fig. 5).
3.2.5. Active range of knee flexion
Active range of knee flexion at 3 months was shown in two
studies [28,37] enrolling 154 patients. The pooled data
(MD:0.49,95%CI: 2.3 to 3.29, p ¼ 0.73; Fig. 6) suggests equal efficacy.
Active range of knee flexion at 6 months was shown in two
studies [28,37] enrolling 154 patients. There is significant hetero-
geneity (Chi2 ¼ 3.87, df ¼ 1, P ¼ 0.05, I2 ¼ 74%; Fig. 6). Thus, a
random-effects is performed. The effect size (MD:1.77,95%CI: 4.09
to 7.63, p ¼ 0.55, Fig. 7) suggests equal efficacy.
3.2.6. Treatment-related adverse effects
Six studies [28,29,33,34,36,37] enrolling 1352 patiens reported
treatment-related adverse effects. Significant difference is found
between the two groups regarding to treatment-related adverse
effects (RR:1.66,95%CI:1.34 to 2.06, p < 0.00001; Fig. 8).
4. Discussion
This meta-analysis is performed to compare intraarticular hy-
aluronic acid injection with intraarticular corticosteroid injection in
the treatment of knee osteoarthritis. Both intraarticular HA and
intraarticular CS are first-line therapy for knee OA patients, while
attitudes of guidelines towards the two therapys diverse from each
other [21,22,40].
A number of studies have proven analgesic effect of intra-
articular HA and CS [6,10,16,18]. Our meta-analysis suggests that
analgesic effect of the two therapys varies over time. The VAS score
at 1 month in intraarticular CS group is found to be significant
lower than in intraarticular HA group, CS shows a greater pain relief
efficacy than HA in a short term; but by 3 months, no significant
difference in VAS score is found between the two groups, two
treatments exhibit equal efficacy; up to 6 months, patients in HA
group show significant lower VAS score than CS group (Fig. 2), thus
HA exhibits a greater analgesic effect than CS in the long term.
Whereas it’s important to note that this a meta-analysis of intra-
articular HA and CS, which does not compare their efficacy with
placebo. There are some studies claim that the effect of intra-
articular CS is largely absent by 6 months [14], if so, absolute effect
of intraarticular HA may be modest. WOMAC index is a compre-
hensive instrument evaluating the condition of patients with
osteoarthritis of the knee and hip. It can be self-administered and
was developed at Western Ontario and McMaster Universities in
1982. The Index is self-administered and assesses the three di-
mensions of pain, disability and joint stiffness in knee and hip
osteoarthritis using a battery of 24 questions. There is no significant
difference on WOMAC score in the two groups at 3 months, while

Table 1a
Characteristic of included studies and participants.

Author, year (ref.) Participants Gender (M/F) Mean Age Interventions Jadad score (points)
(years)

HA CS HA CS HA CS

Jones et al., 1995 [26] 32 31 39/24 71.4 69.5 Hyalgan Triamcinolone hexacetonide 5
20 mg 20 mg
5 weekly injections Single injection
Frizziero and Pasquali 52 47 46/53 49 50 Hyalgan Methylprednisolone acetate 4
Ronchetti, 2002 [27] 2 ml (20 mg) 1 ml (40 mg)
5 weekly injections 3 weekly injections
€
Tascioglu and Oner 2003 30 30 0/60 57.4 60.1 Orthovisc 6-Methylprednisolone acetate 3
[28] 2 ml (30 mg) 1 ml (40 mg)
3 weekly injections 3 weekly injections
Carbon et al., 2004 [29] 113 103 93/123 62.5 63.7 Synvisc Triamcinolone hexacetonide 5
2 ml (16 mg) 2 ml (40 mg)
3 weekly injections Single injection
Skwara et al., 2009 [30] 21 21 17/25 60.8 61.3 Ostenil Triamcinolone acetonide 7
2 ml (20 mg) 1 ml (10 mg)
5 weekly injections 5 weekly injections
Skwara et al., 2009 [31] 24 26 27/23 60.92 61.81 NASHA Triamcinolone acetonide 6
3 ml (60 mg) 1 ml (10 mg)
Single injection Single injection
 W.-w. He et al. / International Journal of Surgery 39 (2017) 95e103 99

Table 1b
Characteristic of included studies and participants.

Author, year (ref.) Participants Gender (M/F) Mean Age Interventions Jadad score (points)
(years)

HA CS HA CS HA CS

Shimizu et al., 2010 [32] 26 25 13/38 75.9 75.3 Sodium hyaluronate Decadron 4
25 mg 4 mg
5 weekly injections Single injection
Housman et al., 2014 [33] 259 132 130/261 61.3 60.1 Hylastan SGL-80 Methylprednisolone acetate 7
4 ml 1 ml (40 mg)
1 or 2 weekly injections Single injection
Leighton et al., 2014 [34] 218 215 220/213 61.9 61.5 NASHA Methylprednisolone acetate 7
3 ml (60 mg) 1 ml (40 mg)
Single injection Single injection
Askari et al., 2016 [35] 71 69 21/119 58.5 57.0 Hyaluronic Acid Corticosteroid 7
2 ml 40 mg
Single injection Single injection
Bisicchia et al., 2016 [36] 75 75 47/103 71.5 68.6 HYADD 4 6-Methylprednisolone acetate 7
Not mentioned 1 ml (40 mg)
2 weekly injections 2 weekly injections
Tammachote et al., 2016 [37] 50 49 20/79 62.6 61 Synvisc Triamcinolone acetonide 7
6 ml 1 ml (40 mg)
Single injection Single injection

Fig. 2. A forest plot diagram showing the VAS score.

100 W.-w. He et al. / International Journal of Surgery 39 (2017) 95e103
Table 2
Subgroup analysis of VAS score at 1 month.
Subgroup Studies
Allocation concealment 2 [35,37]
Analgesic or NSAIDs 3 [27,28,35]
Dose of CS  40 mg 4 [26,32,35,37]
VAS, visual analog scale; MD, mean difference; CI, confidence interval.
up to 6 months, WOMAC score in intraarticular HA group is fewer
than intraarticular CS group which favors the intraarticular HA
(Fig. 3). During follow-up visit of the 12 RTCs, some patients ask for
withdraw for unbearable knee pain or need rescue medicine for
analgesia. Proportion of rescue medication use after initiation of
treatment and proportion of withdrawal for knee pain can be in-
direct evidence for analgesic effect. However, both two indexes
exhibit no significant difference between the two therapys (Figs. 4
and 5). Deterioration of range of motion is one of the major
outcome of osteoarthritis, thus active range of knee flexion after
injections can reflect the efficacy of the two treatments. Both of
intraarticular HA and CS are beneficial to improving range of mo-
tion for affected knee [41,42]. Our meta-analysis indicates an equal
efficacy on active range of knee flexion at 3 and 6 months (Figs. 6
and 7). As for safety, few studies report serious adverse events
associated with intraarticular HA or intraarticular CS [6,18]. Most of
treatment-related adverse effects are topical adverse effects
(e.g.knee pain, joint swelling, joint stiffness). Our meta-analysis
shows that intraarticular HA group exhibits greater incidence of
treatment-related adverse effects than intraarticular CS group,
some of these may can be attributed to higher injection frequency
in intraarticular HA group. Three studies [29,33,34] report serious
systemic adverse events in either intraarticular HA or intraarticular
CS group, while none of them is related to study treatment or
procedure, thus, both therapys are relatively safe. For patients of
the 12 RCTs included in the analysis, their OA stages are all between
Kellgren-Lawrence grade IeIII. Patients with Kellgren-Lawrence
grade Ⅳ are excluded. For these patients, severe sclerosis and
definite deformity of bone contour exsit, surgery treatments e.g.
Fig. 3. A forest plot diagram showing the WOMAC score.
Effect estimate
c2 MD 95% CI I2 (%) P
0.7 0.72 [0.21,1.24] 0 0.006
8.35 0.98 [0.02,1.94] 76 0.05
2.35 0.52 [0.11,0.92] 0 0.01
total knee arthroplasty may be a better choice [43].
Wang F et al. also conducted a meta-analysis to compare the
efficacy and safety of intraarticular HA and CS in the treatment of
knee OA. They found that intraarticular HA and intraarticular CS
have equal efficacy for pain in short term (1 month) and intra-
articular HA is more effective than CS after 3 months; besides, no
difference was observed between HA and CS for adverse effects.
Nevertheless, the number and quality of the included studies in
their meta-analysis are restricted, and they mistook the data of the
maximum knee flexion during gait analysis for active range of knee
flexion. Our meta-analysis includes more high quality RCTs, be-
sides, data extraction is more strict.
Twelve RCTs meet the inclusion criteria of meta-analysis. All
RCTs have high quality of which Jadad score 4 except one [28]
(Tables 1a and 1b), and a subset analysis confined to the other 11
RCTs yield results that are consistent with the overall analysis. All
RCTs report randomization, but method of three [26,28,29] is un-
clear. Six RCTs [30,33e37] mention allocation concealment, while
others are nor clear. All inclusive studies show comparable baseline
data, but only four studies [27,29,33,34] take intend-to-treat anal-
ysis. This methodological weakness are considered when analyzing
outcomes. Seven RCTs adopted double-blinded method
[26,30,31,33e35,37], two RCTs adopted single-blinded method
[29,36], while one RCT chose an open design [28], and the blind
method of the rest two RCTs [27,32] were not clear. Thus, perfor-
mance and detection bias should be taken into account. Moreover,
subgroup analysis is performed to find out the source when sig-
nificant heterogeneity exists (Table 2).
Current meta-analysis has following limitations [1]: Only twelve
 W.-w. He et al. / International Journal of Surgery 39 (2017) 95e103 101

Fig. 4. A forest plot diagram showing the proportion of rescue medication use.

Fig. 5. A forest plot diagram showing the proportion of withdrawal for knee pain.

Fig. 6. A forest plot diagram showing the active range of knee flexion at 3 months.

RCTs containing 1794 patients were included, the statistical efficacy
of our analysis would increase if more RCTs had been included [2].
Some indexes such as active range of knee flexion after injections
are only reported in two or three studies, it's a good assessing
indication for knee function. This may influence the result [3]. Only
English publications are included, thus, publication bias exists [4].
Potential source of heterogeneity is the use of different drugs in
varied doses and regimens. Therefore, we use subgroup analysis
and random-effects model to find and eliminate the influence [5].
For the 12 RCTs included in our analysis, the longest time of flow-up
visit is just 6 months, longer term effect of both HA and CS are not
included. Although several limitations exist, we strictly screened

Fig. 7. A forest plot diagram showing the active range of knee flexion at 6 months.
102 W.-w. He et al. / International Journal of Surgery 39 (2017) 95e103
Fig. 8. A forest plot diagram showing the treatment-related adverse effects.
available articles prior to inclusion to ensure high quality, besides,
Cochrane Handbook and PRISMA guidelines are used to evaluate
the quality of results contained in all included studies.
5. Conclusions
Our meta-analysis indicates that Intraarticular CS is more
effective on pain relief than intraarticular HA in short term (up to 1
month), while HA is more effective in long term (up to 6 months).
Two therapies benefit similarly for knee function improvement.
Both two methods are relatively safe, but intraarticular HA causes
more topical adverse effects compared with intraarticular CS.
Ethical approval
Have nothing to declare.
Sources of funding
National Natural Science Foundation of China (No. 81572154).
Tianjin Health Bureau Science and Technology Foundation (No.
15KG123).
Key project of Tianjin science and technology (No.
13ZCZDSY01700).
Author contribution
Wei-wei He and Ming-jie Kuang:Study conception and design,
write the paper.
Jie Zhao and Lei Sun:Collect and analysis data.
Bin Lu and Ying Wang:Revise draft and generate figures.
Jian-xiong Ma and Xin-long Ma:Design the study and proofread
the manuscript.
Conflict of interest statement
Conflict of interest:none.
Research registration unique identifying number (UIN)
Reviewregistry133.
Guarantor
Xin-long Ma.
References
[1] NICE, Osteoarthritis: Care and Management in Adults, 2014. Clinical guideline
CG177 Methods, evidence and recommendations.
[2] C.F. Dillon, E.K. Rasch, Q. Gu, R. Hirsch, Prevalence of knee osteoarthritis in the
United States: arthritis data from the third national Health and nutrition ex-
amination survey 1991e94, J. Rheumatol. 33 (11) (2006) 2271e2279.
[3] K.M. Koenig, K.L. Ong, E.C. Lau, T.P. Vail, D.J. Berry, H.E. Rubash, et al., The use
of hyaluronic acid and corticosteroid injections among medicare patients with
knee osteoarthritis, J. Arthroplasty 31 (2) (2016) 351e355.
[4] L.W. Moreland, Intra-articular hyaluronan (hyaluronic acid) and hylans for the
treatment of osteoarthritis: mechanisms of action, Arthritis Res. Ther. 5 (2)
(2003) 54e67.
[5] A. Gomis, A. Miralles, R.F. Schmidt, C. Belmonte, Intra-articular injections of
hyaluronan solutions of different elastoviscosity reduce nociceptive nerve
activity in a model of osteoarthritic knee joint of the Guinea pig. Osteoarthritis
and cartilage/OARS, Osteoarthr. Res. Soc. 17 (6) (2009) 798e804.
[6] F. Navarro-Sarabia, P. Coronel, E. Collantes, F.J. Navarro, A.R. de la Serna,
A. Naranjo, et al., A 40-month multicentre, randomised placebo-controlled
study to assess the efficacy and carry-over effect of repeated intra-articular
injections of hyaluronic acid in knee osteoarthritis: the AMELIA project,
Ann. Rheumatic Dis. 70 (11) (2011) 1957e1962.
[7] K.L. Ong, A.F. Anderson, F. Niazi, A.L. Fierlinger, S.M. Kurtz, R.D. Altman, Hy-
aluronic acid injections in medicare knee osteoarthritis patients are associated
with longer time to knee arthroplasty, J. Arthroplasty 31 (8) (2016)
1667e1673.
[8] J.E. DeCaria, M. Montero-Odasso, D. Wolfe, B.M. Chesworth, R.J. Petrella, The
effect of intra-articular hyaluronic acid treatment on gait velocity in older
knee osteoarthritis patients: a randomized, controlled study, Archiv. Geron-
tology Geriatrics 55 (2) (2012) 310e315.
[9] N.K. Arden, C. Akermark, M. Andersson, M.G. Todman, R.D. Altman,
A randomized saline-controlled trial of NASHA hyaluronic acid for knee
osteoarthritis, Curr. Med. Res. Opin. 30 (2) (2014) 279e286.
[10] K.A. Campbell, B.J. Erickson, B.M. Saltzman, R. Mascarenhas, B.R. Bach Jr.,
B.J. Cole, et al., Is local viscosupplementation injection clinically superior to
other therapies in the treatment of osteoarthritis of the knee: a systematic
review of overlapping meta-analyses, Arthrosc. J. Arthrosc. Relat. Surg. Off.
Publ. Arthrosc. Assoc. N. Am. Int. Arthrosc. Assoc. 31 (10) (2015) 2036e2045
e14.
[11] A.W. Rutjes, P. Juni, B.R. da Costa, S. Trelle, E. Nuesch, S. Reichenbach, Visco-
supplementation for osteoarthritis of the knee: a systematic review and meta-
analysis, Ann. Intern. Med. 157 (3) (2012) 180e191.
[12] L. Tahiri, K. Benbouazza, B. Amine, N. Hajjaj-Hassouni, Acute pseudoseptic
arthritis after viscosupplementation of the knee: a case report, Clin. Rheu-
matol. 26 (11) (2007) 1977e1979.
[13] W.B. Kim, R.O. Alhusayen, Skin necrosis from intra-articular hyaluronic acid
injection, J. Cutan. Med. Surg. 19 (2) (2015) 182e184.
[14] P. Juni, R. Hari, A.W. Rutjes, R. Fischer, M.G. Silletta, S. Reichenbach, et al.,
Intra-articular corticosteroid for knee osteoarthritis, Cochrane Database Syst.
Rev. (10) (2015) Cd005328.
[15] R.A. Jessar, M.A. Ganzell, C. Ragan, The action of hydrocortisone in synovial
inflammation, J. Clin. Invest. 32 (6) (1953) 480e482.
[16] J. Chao, C. Wu, B. Sun, M.K. Hose, A. Quan, T.H. Hughes, et al., Inflammatory
characteristics on ultrasound predict poorer longterm response to intra-
articular corticosteroid injections in knee osteoarthritis, J. Rheumatol. 37 (3)
(2010) 650e655.
[17] C.T. Hepper, J.J. Halvorson, S.T. Duncan, A.J. Gregory, W.R. Dunn, K.P. Spindler,
The efficacy and duration of intra-articular corticosteroid injection for knee
osteoarthritis: a systematic review of level I studies, J. Am. Acad. Orthop. Surg.
17 (10) (2009) 638e646.
[18] A.B. Lomonte, M.G. de Morais, L.O. de Carvalho, C.A. Zerbini, Efficacy of
triamcinolone hexacetonide versus methylprednisolone acetate intraarticular
injections in knee osteoarthritis: a randomized, double-blinded, 24-week
study, J. Rheumatol. 42 (9) (2015) 1677e1684.
[19] B. Arroll, F. Goodyear-Smith, Corticosteroid injections for osteoarthritis of the
knee: meta-analysis, BMJ (Clinical Research ed) 328 (7444) (2004) 869.
[20] T.H. Trojian, A.L. Concoff, S.M. Joy, J.R. Hatzenbuehler, W.J. Saulsberry,
C.I. Coleman, AMSSM scientific statement concerning viscosupplementation
injections for knee osteoarthritis: importance for individual patient outcomes,
 W.-w. He et al. / International Journal of Surgery 39 (2017) 95e103
Br. J. Sports Med. 50 (2) (2016) 84e92.
[21] T.E. McAlindon, R.R. Bannuru, M.C. Sullivan, N.K. Arden, F. Berenbaum,
S.M. Bierma-Zeinstra, et al., OARSI guidelines for the non-surgical manage-
ment of knee osteoarthritis. Osteoarthritis and cartilage/OARS, Osteoarthr.
Res. Soc. 22 (3) (2014) 363e388.
[22] M.C. Hochberg, R.D. Altman, K.T. April, M. Benkhalti, G. Guyatt, J. McGowan, et
al., American College of Rheumatology 2012 recommendations for the use of
nonpharmacologic and pharmacologic therapies in osteoarthritis of the hand,
hip, and knee, Arthritis Care Res. 64 (4) (2012) 465e474.
[23] H. JPT, S G. Cochrane Handbook for Systematic Reviews of Interventions, 2012.
Version 5,1.0,updated Mar 2011, http://wwwcochrane-handbookorg/.
[24] R. Banares, A. Albillos, D. Rincon, S. Alonso, M. Gonzalez, L. Ruiz-del-Arbol, et
al., Endoscopic treatment versus endoscopic plus pharmacologic treatment for
acute variceal bleeding: a meta-analysis, Hepatol. (Baltimore, Md) 35 (3)
(2002) 609e615.
[25] A.R. Jadad, R.A. Moore, D. Carroll, C. Jenkinson, D.J. Reynolds, D.J. Gavaghan, et
al., Assessing the quality of reports of randomized clinical trials: is blinding
necessary? Control. Clin. Trials 17 (1) (1996) 1e12.
[26] A.C. Jones, M. Pattrick, S. Doherty, M. Doherty, Intra-articular hyaluronic acid
compared to intra-articular triamcinolone hexacetonide in inflammatory knee
osteoarthritis. Osteoarthritis and cartilage/OARS, Osteoarthr. Res. Soc. 3 (4)
(1995) 269e273.
[27] L. Frizziero, I. Pasquali Ronchetti, Intra-articular treatment of osteoarthritis of
the knee: an arthroscopic and clinical comparison between sodium hyaluro-
nate (500e730 kDa) and methylprednisolone acetate, J. Orthop. Traumatol. 3
(2) (2002) 89e96.
€
[28] F. Tascioglu, C. Oner, Efficacy of intra-articular sodium hyaluronate in the
treatment of knee osteoarthritis, Clin. Rheumatol. 22 (2) (2003) 112e117.
[29] D. Caborn, J. Rush, W. Lanzer, D. Parenti, C. Murray, A randomized, single-
blind comparison of the efficacy and tolerability of hylan G-F 20 and triam-
cinolone hexacetonide in patients with osteoarthritis of the knee,
J. Rheumatol. 31 (2) (2004) 333e343.
[30] A. Skwara, C.D. Peterlein, C.O. Tibesku, D. Rosenbaum, S. Fuchs-Winkelmann,
Changes of gait patterns and muscle activity after intraarticular treatment of
patients with osteoarthritis of the knee: a prospective, randomised, double-
blind study, Knee 16 (6) (2009) 466e472.
[31] A. Skwara, R. Ponelis, C.O. Tibesku, D. Rosenbaum, S. Fuchs-Winkelmann, Gait
patterns after intraarticular treatment of patients with osteoarthritis of the
kneeehyaluronan versus triamcinolone: a prospective, randomized, double-
blind, monocentric study, Eur. J. Med. Res. 14 (4) (2009) 157e164.
[32] M. Shimizu, H. Higuchi, K. Takagishi, T. Shinozaki, T. Kobayashi, Clinical and
biochemical characteristics after intra-articular injection for the treatment of
103
osteoarthritis of the knee: prospective randomized study of sodium hyalur-
onate and corticosteroid, J. Orthop. Sci. Off. J. Jpn. Orthop. Assoc. 15 (1) (2010)
51e56.
[33] L. Housman, N. Arden, T.J. Schnitzer, C. Birbara, T. Conrozier, N. Skrepnik, et al.,
Intra-articular hylastan versus steroid for knee osteoarthritis. Knee surgery,
sports traumatology, arthroscopy, Off. J. ESSKA 22 (7) (2014) 1684e1692.
[34] R. Leighton, C. Akermark, R. Therrien, J.B. Richardson, M. Andersson,
M.G. Todman, et al., NASHA hyaluronic acid vs. methylprednisolone for knee
osteoarthritis: a prospective, multi-centre, randomized, non-inferiority trial.
Osteoarthritis and cartilage/OARS, Osteoarthr. Res. Soc. 22 (1) (2014) 17e25.
[35] A. Askari, T. Gholami, M.M. NaghiZadeh, M. Farjam, S.A. Kouhpayeh,
Z. Shahabfard, Hyaluronic Acid Compared with Corticosteroid Injections for
the Treatment of Osteoarthritis of the Knee: a Randomized Control Trail vol. 5,
SpringerPlus, 2016, p. 442.
[36] S. Bisicchia, G. Bernardi, C. Tudisco, HYADD 4 versus methylprednisolone
acetate in symptomatic knee osteoarthritis: a single-centre single blind pro-
spective randomised controlled clinical study with 1-year follow-up, Clin. Exp.
Rheumatol. 34 (5) (2016) 857e863.
[37] N. Tammachote, S. Kanitnate, T. Yakumpor, P. Panichkul, Intra-Articular,
single-shot hylan G-F 20 hyaluronic acid injection compared with cortico-
steroid in knee osteoarthritis: a double-blind, randomized controlled trial,
J. Bone Jt. Surg. Am. 98 (11) (2016) 885e892.
[38] S.S. Leopold, B.B. Redd, W.J. Warme, P.A. Wehrle, P.D. Pettis, S. Shott, Corti-
costeroid compared with hyaluronic acid injections for the treatment of
osteoarthritis of the knee. A prospective, randomized trial, J. Bone Jt. Surg. Am.
85-a (7) (2003) 1197e1203.
[39] O.A. Kaplunov, S.N. Biriukov, V. Bersanov, Use of prolonged corticosteroids
and hyaluronic acid salts in treatment of gonarthrosis, Khirurgiia (1) (2015)
58e62.
[40] A. Chinese Orthopaedic, Diagnosis and treatment of osteoarthritis, Orthop.
Surg. 2 (1) (2010) 1e6.
[41] A.C. Tang, S.F. Tang, W.H. Hong, H.C. Chen, Kinetics features changes before
and after intra-articular hyaluronic acid injections in patients with knee
osteoarthritis, Clin. Neurol. Neurosurg. 129 (Suppl 1) (2015) S21eS26.
[42] S. Mehta, B.L. Shay, T. Szturm, H.S. El-Gabalawy, Kinematic analysis of gait
following intra-articular corticosteroid injection into the knee joint with an
acute exacerbation of arthritis, Physiother. Can. Physiother. Can. 63 (4) (2011)
395e404.
[43] M.J. Kuang, C. Han, J.X. Ma, F. Li, J. Zhao, L. Fu, et al., The efficacy of intra-
operative autologous platelet gel in total knee arthroplasty: a meta-analysis,
Int. J. Surg. (London, England) 36 (Pt A) (2016) 56e65.