Introduction
Knee osteoarthritis (KOA) is a chronic disorder characterized by joint pain, increasing
physical disability, and progressive cartilage degeneration, which can lead to total
knee arthroplasty (TKA). Despite extensive research, the complex pathophysiology of
KOA remains incompletely understood. As a result, no established disease‐modifying
treatment exists today, and the management of KOA still relies on a combination of
nonpharmacologic and pharmacologic modalities 1 primarily intended to relieve the
symptoms of pain and loss of knee function. In clinical practice, acetaminophen and
nonsteroidal antiinflammatory drugs (NSAIDs) are generally recommended to relieve
pain and improve joint function 2, 3, 4, 5. However, due to their vascular and gastrointestinal
toxicity 6, 7, the risk‐benefit ratio of these rapid‐acting drugs might not be favorable,
particularly for long‐course therapies and within the aging population in which KOA
is most prevalent 8. Intraarticular (IA) injections of corticosteroids such as triamcinolone
hexacetonide and methylprednisolone acetate are also commonly prescribed. As a systemic
absorption occurs following IA corticosteroid injection, systemic adverse events (AEs)
can be expected and precautions should be observed in patients with concomitant diseases
such as hypertension or diabetes mellitus 9, 10, 11, 12, 13.
Box 1
Significance & Innovations
Besides efficacy and safety, this review also addresses less frequently evaluated
aspects of the intraarticular use of hyaluronic acid (HA) in the management of knee
osteoarthritis (KOA) such as effectiveness and efficiency.
This is the first time that a group of experts recommends systematic repeated intraarticular
HA treatment in KOA patients who had a beneficial response to a previous cycle of
treatment.
IA hyaluronic acid (HA) is a local treatment modality devoid of the systemic AEs observed
after IA corticosteroid injection or oral administration of analgesics and NSAIDs.
Thus, IAHA represents an alternative to analgesics and NSAIDs in patients with comorbidities,
as well as a secondary option in case of inadequate response to first‐line pharmacologic
KOA treatments 14. Registered as medical devices in the US, IAHA preparations are
currently approved for the symptomatic treatment of KOA, one of the most commonly
affected joints. Despite being widely employed in the daily management of KOA for
almost 20 years in the US (about 30 years in Europe and Japan), controversies persist
regarding their efficacy and safety, as highlighted by discrepancies in the guidelines
related to the use of IAHA in clinical practice 2, 3, 4, 5, 15, which largely reflect
the divergent conclusions drawn by meta‐analyses on the topic 16, 17, 18, 19, 20,
21, 22, 23, 24, 25, 26, 27, 28, 29, 30.
For this purpose, the European Society for Clinical and Economic Aspects of Osteoporosis
and Osteoarthritis (ESCEO) convened an international task force of experts in OA and
clinical research methodology. Seven members of this working group (CC, FR, PR, OB,
GH‐B, AM, and DU) were entrusted with the task of preparing a review on diverse aspects
of the use of IAHA in the management of KOA. Current knowledge on the mode of action,
efficacy, effectiveness, safety, and cost‐effectiveness of this treatment modality
was presented and discussed at a 1‐day meeting in January 2016 in Geneva (Switzerland).
The objective of this review is to provide specialists and practicing physicians with
clear, concise, and reasoned answers to questions they might have on the use of IAHA
in the management of KOA.
Efficacy
Efficacy of IAHA on pain and joint function has been evaluated in numerous randomized
controlled trials (RCTs). Consequently, 15 meta‐analyses/systematic reviews assessing
the symptomatic effects of IAHA on KOA have been published to date 16, 17, 18, 19,
20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30. Six of them concluded on a clear efficacy
of IAHA 17, 20, 24, 27, 29, 30, 4 considered it as a marginally efficacious treatment
modality 16, 19, 25, 26, and 2 meta‐analyses found no significant difference in efficacy
between IAHA and IA placebo 18, 21 (Table 1).
Table 1
Meta‐analyses evaluating intraarticular hyaluronic acid in the symptomatic treatment
of knee osteoarthritisa
Reference
Comparator
Main outcome
ES (95% CI)
General outcome
Lo et al, 2003 16
IA placebo
Pain change from BL at M1–4
SMD = 0.32 (0.17, 0.47)
Intermediate
Wang et al, 2004 17
IA placebo
Pain on movement
MPD = 7.9 (4.1, 11.7)
Positive
Arrich et al, 2005 18
IA placebo
Pain on movement at W2–6
WMD = 3.8 (−1.4, 9.1) I2 = 81%
Negative (no clinical difference)
Pain on movement at W10–14
WMD = 4.3 (0.9, 7.6) I2 = 0%
Pain on movement at W22–30
WMD = 7.1 (2.4, 11.8) I2 = 0%
Modawal et al, 2005 19
IA placebo
Pain change at W5–7
MD = 17.6 (7.5, 28.0)
Intermediate
Pain change at W8–12
MD = 18.1 (6.3, 29.9)
Pain change at W15–22
MD = 4.4 (−15.3, 24.1)
Bellamy et al, 2006 20
IA placebo
Pain/function at W5–13
WMD/SMD variable for different HA products
Positive
Medina et al, 2006 21
IA placebo
Pain
MPD = 2.4 (−0.6, 5.5)
Negative (no statistical difference)
Function
MPD = 2.0 (−0.8, 4.9)
Stiffness
MPD = 5.5 (2.2, 8.8)
Reichenbach et al, 2007 22
IA hylan
Pain at end of followup
SMD = −0.27 (−0.55, 0.01) I2 = 88%
HA > hylan
Bannuru et al, 2009 23
IA CS
Pain change from BL at W2
SMDg = −0.39 (−0.65, −0.12) I2 = 47%
Positive (HA > CS from W8 onward)
Pain change from BL at W4
SMDg = −0.01 (−0.23, 0.21) I2 = 37%
Pain change from BL at W8
SMDg = 0.22 (−0.05, 0.49) I2 = 47%
Pain change from BL at W12
SMDg = 0.35 (0.03, 0.66) I2 = 49%
Pain change from BL at W26
SMDg = 0.39 (0.18, 0.59) I2 = 0%
Bannuru et al, 2011 24
IA placebo
Pain change from BL at W4
SMDg = 0.31 (0.17, 0.45) I2 = 75%
Positive
Pain change from BL at W8
SMDg = 0.46 (0.28, 0.65) I2 = 75%
Pain change from BL at W12
SMDg = 0.25 (0.15, 0.36) I2 = 60%
Pain change from BL at W16
SMDg = 0.20 (0.11, 0.30) I2 = 7%
Pain change from BL at W24
SMDg = 0.21 (0.10, 0.31) I2 = 32%
Colen et al, 2012 25
IA placebo
Pain change from BL at M3
vs. IA placebo: WMD = 10.2 (4.4, 16.0) I2 = 92%
Intermediate
vs hylan SMD = 0.07 (−0.10, 0.24) I2 = 72%
Rutjes et al, 2012 26
Sham or no intervention
Pain at end of followup
SMD: 0.37 (0.28, 0.46) τ2 = 0.09
Intermediate
Miller and Block, 2013 27
IA placebo
Pain at W4–13
SMD = 0.43 (0.26, 0.60) I2 = 73%
Positive
Pain at W14–26
SMD = 0.38 (0.21, 0.55) I2 = 75%
Function at W4–13
SMD = 0.34 (0.16, 0.51) I2 = 54%
Function at W14–26
SMD = 0.32 (0.18, 0.45) I2 = 69%
Bannuru et al, 2014 28
NSAIDs
Pain change from BL at W4
SMDg = −0.01 (−0.18, 0.15) I2 = 0%
Positive (HA = NSAIDs)
Pain change from BL at W12
SMDg = 0.05 (−0.17, 0.28) I2 = 30%
Pain change from BL at end of followup
SMDg = 0.07 (−0.10, 0.24) I2 = 16%
Bannuru et al, 2015 29
Other options
Pain change from BL at M3
vs IA placebo: SMDg = 0.34 (0.26, 0.42)b
Positive
vs oral placebo: SMDg = 0.63 (0.39, 0.88)b
Function change from BL at M3
vs IA placebo: SMDg = 0.30 (0.20, 0.40)b
vs oral placebo: SMDg = 0.45 (0.08, 0.84)b
Richette et al, 2015 30
IA placebo
Pain at W12
SMD = 0.20 (0.12, 0.29) I2 = 32%
Positive
Function at W12
SMD = 0.12 (0.02, 0.22) I2 = 0%
a
Positive effect sizes (ES) favor intraarticular hyaluronic acid, negative ES favor
comparator. 95% CI = 95% confidence interval; IA = intraarticular; BL = baseline;
M = month; SMD = standardized mean difference; MPD = mean percent difference; W = week;
WMD = weighted mean difference; MD = mean difference; HA = hyaluronic acid; hylan = hylan
G‐F 20; CS = corticosteroid; SMDg = standardized mean difference adjusted for small
sample size (Hedges' g); NSAIDs = nonsteroidal antiinflammatory drugs.
b
95% credible interval.
The reported effect sizes (ES) for pain favored IAHA over IA placebo and ranged from
0.20 (30) to 0.46 (24). Heterogeneity of outcomes between trials was relatively high
in several meta‐analyses 16, 20, 24, 27. As shown by Bannuru et al, the benefits of
IAHA vary over time, with a maximal effect on pain at 8 weeks (ES 0.46 [95% confidence
interval (95% CI) 0.28, 0.65]) that is still observable at 24 weeks (ES 0.21 [95%
CI 0.10, 0.31]) 24. This time dependency may partly contribute to the differences
in pain estimates that have been observed between meta‐analyses. Another possible
explanation is the inclusion of clinical trials that widely differ in methodological
quality. Thus, when the analysis was restricted to high‐quality trials, an ES of 0.34
(95% CI 0.02, 0.67) and of 0.20 (95% CI 0.03, 0.37) in favor of IAHA was obtained
at 8 and 24 weeks, respectively 24. These results were confirmed by Richette et al,
who examined clinical data obtained from IA placebo controlled trials with a low risk
of bias only (n = 8) in order to reach the highest level of evidence. Based on an
ES of 0.20 (95% CI 0.12, 0.29) for pain at 12 weeks, the authors concluded that IAHA
provided a moderate but real effect on pain in patients with KOA 30.
Since the estimates for pain mentioned above were based on the difference between
the IA placebo and the IAHA effects, it is worth noting that evidence for a clinically
significant response of the IA delivery method itself has recently been demonstrated
in a meta‐analysis, with an ES of 0.29 (95% credible interval [95% CrI] 0.04, 0.54)
in favor of IA placebo compared with oral placebo 29. In the same study, different
treatment options of KOA were compared based on their ES versus oral placebo at 3
months. IAHA was found to be the most efficacious intervention with an ES of 0.63
(95% CrI 0.39, 0.88), followed by IA corticosteroids (ES 0.61 [95% CrI 0.32, 0.89]),
diclofenac (ES 0.52 [95% CrI 0.34, 0.69]), ibuprofen (ES 0.44 [95% CrI 0.25, 0.63]),
naproxen (ES 0.38 [95% CrI 0.27, 0.49]), celecoxib (ES 0.33 [95% CrI 0.25, 0.42]),
and IA placebo. With an ES of 0.18 (95% CrI 0.04, 0.33), acetaminophen was considered
as not superior to oral placebo in relieving pain 29. In direct comparison, IAHA was
shown to be not significantly different from continuous oral NSAID treatment at 4
and 12 weeks regarding pain, function, and stiffness 28, and superior to IA corticosteroids
from 8 to 26 weeks regarding pain 23.
Effectiveness
The extent to which IAHA achieves its intended effect in the “real‐life” clinical
setting cannot be measured in RCTs, since patients in such studies are not representative
of those seen in usual practice. Therefore, observational studies of real practice
are better suited in evaluating effectiveness of IAHA.
Petrella and Wakeford retrospectively assessed the effectiveness of IA crosslinked
HA (hylan) using the Southwestern Ontario database, a Canadian real‐world cohort 31.
For this purpose, they identified 1,263 patients with OA in 1 or both knees that received
2 consecutive series of IA hylan injections and no other prescribed OA medications.
They compared them to a cohort of 3,318 demographically matched KOA patients who were
never treated with IAHA. All patients were evaluated fully between 2006 and 2012.
Results showed that in the group of patients who received repeated treatments of IA
hylan, pain at rest and pain after a 6‐minute walk decreased by mean ± SD 3.7 ± 1.8
points and 5.6 ± 1.7 points on a 10‐point visual analog scale, respectively. In parallel,
the distance walked in a 6‐minute walk test increased on average by 115 meters in
this patient group. These improvements in pain and physical function were significantly
greater than those achieved in KOA‐matched patients treated with other prescribed
OA medications (intergroup comparison P < 0.012 for pain at rest, P < 0.001 for pain
after a 6‐minute walk, and P < 0.001 for distance walked in a 6‐minute walk test).
Previous observational studies already found that IAHA was effective in relieving
pain and improving knee function for up to 6 months 32, 33, 34. Such pragmatic trials
were also adapted in the assessment of longer‐term outcomes.
Three recent retrospective studies examined the time from diagnosis of KOA to TKA
and whether this time would be influenced by the use of IAHA 35, 36, 37. Large populations
of insured subjects were screened and about 250,000 TKAs were reviewed after exclusions.
In these studies, the median time to arthroplasty from the index time of KOA diagnosis
was 114, 86, and 326 days in those not receiving IAHA. The median time from diagnosis
to TKA in those receiving IAHA was 484, 585, and 908 days. Two studies including 65‐year‐old
subjects showed a 1‐year increase in time to surgery 35, 36. The study including only
those younger than age 65 years showed an increase in time to surgery of 1.6 years
37. All studies showed increased time to surgery with increased series of IAHA, and
2 studies suggested additional benefit from higher molecular weight (MW) HA products
36, 37. The consistency of results from the 3 studies suggests both clinical and economic
benefits to IAHA. However, they are insufficient to prove that the observed delay
in time to TKA results from the use of IAHA, as treatment was not allocated randomly.
Thus, between‐group differences might reflect a bias in patient selection.
As the evidence for efficacy of IAHA on knee pain and function is currently established,
the ESCEO task force suggests promoting pragmatic trials in order to focus clinical
research on the effectiveness of IAHA in real‐life conditions, considering the patient
situation in terms of individual and disease characteristics, medical history, and
comorbidities.
Safety
Seven meta‐analyses of RCTs comparing IAHA to IA placebo evaluated the safety of the
intervention 17, 18, 20, 21, 26, 27, 29. AEs occurred slightly more often among patients
who received an IAHA treatment (relative risk [RR] 1.08 [95% CI 1.01, 1.15]) 18. However,
they were usually mild transient local reactions such as pain at the injection site
and swelling, with an RR of 1.19 (95% CI 1.01, 1.41) according to Wang et al 17 or
RR 1.34 (95% CI 1.13, 1.60) according to Rutjes et al 26. The use of IAHA has also
been associated with increased risk of flares (RR 1.51 [95% CI 0.84, 2.72]) and effusion
at the injected knee (RR 1.15 [95% CI 0.38, 3.54]), which were not statistically significant
26. In contrast, the meta‐analysis of US‐approved HA products by Miller and Block
reported no statistically significant differences between IAHA and IA placebo for
any safety outcomes, including serious AEs (P = 0.12), treatment‐related serious AEs
(P = 1.0), study withdrawal (P = 1.0), and AE‐related study withdrawal (P = 0.46)
27. Only Rutjes et al raised concerns about the safety of IAHA. Their meta‐analysis
emphasized an increased risk of serious AEs (RR 1.41 [95% CI 1.02, 1.97]) and dropouts
due to AEs (RR 1.33 [95% CI 1.01, 1.74]). However, these findings have been criticized
regarding the methodological rigor with which the serious AEs were analyzed and the
biologic plausibility of the reported events 38, 39.
As shown in a review by Pagnano and Westrich 40, the safety of IAHA appears to remain
unchanged with multiple courses of treatment, while there might be an increase in
AEs after the first hylan injection(s). Leopold et al reported that patients receiving
multiple cycles of IA hylan exhibited more than an 8‐fold increase in the frequency
of acute local reactions compared with patients receiving only 1 course 41.
In addition to the AEs mentioned above, postmarketing device surveillance highlighted
the occurrence of rare cases of localized inflammatory reactions such as pseudosepsis/severe
acute inflammatory reactions, predominantly reported after avian high‐MW crosslinked
HA injections (hylan) 42.
In comparison with other pharmacologic interventions for the treatment of KOA, IAHA
therapy was shown to exhibit fewer systemic AEs than acetaminophen or oral NSAIDs,
but more local reactions 43. These events were reported to be similar between different
IA therapies, i.e., HA and corticosteroids. Withdrawals due to AEs were more common
among patients receiving acetaminophen or NSAIDs than IA therapies 29. As a result,
after almost 30 years of use, IAHA is usually recognized as a safe treatment modality
for KOA.
Current guidelines
As shown in the previous sections dealing with the efficacy, effectiveness, and safety
of IAHA, there is increasing evidence that the risk‐benefit balance is favorable to
the use of IAHA in the management of KOA 24, 26, 27, 29, 30, 44. However, while IAHA
was initially clearly recommended by national and international professional societies,
such as the Osteoarthritis Research Society International (OARSI) 45, the European
League Against Rheumatism (EULAR) 2, and the American College of Rheumatology (ACR)
46, some clinicians, researchers, and decision makers feel that current guidelines
are conflicting and less favorable than 10 years ago 47, 48. Nevertheless, there is
relative general agreement on the place of IAHA in the management of KOA across organizations
(Table 2). Apart from the American Academy of Orthopedic Surgeons, which refutes the
use of IAHA based on insufficient clinical efficacy compared with IA saline 15, and
the National Institute for Health and Care Excellence, which does not assess the modality
at all 5, the use of IAHA is recommended as second‐line treatment of KOA (Table 2).
Table 2
Recommendations for the use of intraarticular hyaluronic acid in the management of
knee osteoarthritisa
Organization
(issue date)
Recommendation
AAOS (2013)
Recommends not using it (recommendation 9 based on lack of evidence for efficacy,
and not on potential harm) 15.
ACR (2012)
Does not recommend it in the initial management of the disease. Conditionally recommends
it in patients with no satisfactory response to prior recommended treatments 4.
ESCEO (2014)
Recommends it in patients who are severely symptomatic or still symptomatic despite
the use of NSAIDs, or in case of contraindications to NSAIDs (step 2) 14.
EULAR (2003)
Recommends it for pain reduction and functional improvement 2.
NICE (2014)
Recommends not offering it 5.
OARSI (2014)
Recommends it only after physician‐patient interaction for determining whether it
can have merit in the context of their individual characteristics, comorbidities,
and preferences (based on an uncertain appropriateness) 3.
a
AAOS = American Academy of Orthopedic Surgeons; ACR = American College of Rheumatology;
NSAIDs = nonsteroidal antiinflammatory drugs; ESCEO = European Society for Clinical
and Economic Aspects of Osteoporosis and Osteoarthritis; EULAR = European League Against
Rheumatism; NICE = National Institute for Health and Care Excellence; OARSI = Osteoarthritis
Research Society International.
Based on the available evidence and guidelines, the ESCEO recommends using IAHA in
patients remaining symptomatic despite continuous or intermittent treatment with conventional
pharmacologic treatment modalities, i.e., acetaminophen, symptomatic slow‐acting drugs,
and NSAIDs, as well as in patients with comorbidities precluding the use of NSAIDs.
This positioning is clearly highlighted in the ESCEO treatment algorithm for the management
of KOA and may help the prescribing physician prioritize interventions 14, 49. The
ACR guidelines move in the same direction and present IAHA as an alternative to unsatisfactory
initial pharmacologic therapy 46. Furthermore, they emphasize its particular interest
for patients ages >75 years, who are not recommended to take oral NSAIDs 4, 14. EULAR
considers the use of IAHA for pain relief and knee functional improvement based on
level 1B evidence 2.
OARSI guidelines for the nonsurgical management of KOA probably raise the most considerable
amount of misunderstanding. However, as clearly stated in the main part of the article,
an “uncertain” classification was “…not intended to be a negative recommendation or
preclude use of that therapy. Rather it indicates a role for physician‐patient interaction
in determining whether this treatment may have merit in the context of its risk: benefit
profile and the individual characteristics, co‐morbidities, and preferences of the
patient” 3. In other words, OARSI privileges the use of IAHA for specific clinical
phenotypes, which should be defined by the prescribing physician.
Finally, as recently highlighted by Altman et al 50, clinical practice guidelines,
and a fortiori recommendations for the use of IAHA in the management of KOA, are not
intended to create uniformity and suppress treating physicians' self‐analysis of the
patient situation; they are designed to provide the best evidence‐based information
available to help physicians in making a treatment decision.
Mode of action
In the US, from a purely regulatory point of view, the symptomatic effects of most
of the IAHA preparations are considered to primarily result from the unique properties
of HA in solution; when injected intraarticularly, exogenous HA is able to compensate
for the drop in HA concentration and chain length that has been observed during the
progression of OA, thus restoring the elastic and viscous properties of the synovial
fluid, which are responsible for its resistance to compression and its lubricating
effect, respectively 51, 52. However, these direct mechanical effects of exogenous
HA cannot account for its long‐term benefits observed in clinical trials 24, as it
is cleared from the joint within a few days, depending on the IAHA preparation. Over
the past 30 years, several possible pharmacologic mechanisms of action explaining
how the clinical effects of IAHA could persist for several months have been proposed.
A recent review of the preclinical basic science literature performed by Altman et
al 53 highlighted the major role of HA binding to cluster of differentiation 44 receptors
in this complex mechanism, as numerous of its mediated effects (e.g., inhibition of
interleukin [IL]–1β, IL‐6, and matrix metalloproteinase [MMP] expression, and reduction
in prostaglandin E2 synthesis) contribute to the chondroprotection, proteoglycan/glycosaminoglycan
synthesis, antiinflammatory, and subchondral effects as observed in vitro. Alternative
pathways involving HA binding to intercellular adhesion molecule 1 and modulation
of transient receptor potential vanilloid channel 1 activity 54 have also been described.
In addition, the toll‐like receptor (TLR) signaling pathway has been suggested to
contribute to the antiinflammatory and anticatabolic (inhibition of inflammation‐induced
activation of MMPs) effects of IAHA in joint tissues 55, 56. Indeed, there is evidence
that HA fragments that are highly expressed in the OA joint can transduce inflammatory
signals through TLR2, TLR4, or both. When exogenous HA is injected in the joint, the
proportion of these fragments is reduced, thus modulating TLR‐mediated innate immune
responses.
The ESCEO group is not convinced that one mode of action is sufficient to explain
the therapeutic trajectory of IAHA and privilege the hypothesis that several mechanisms
overlap and interact to relieve OA pain. Furthermore, in vitro results obtained on
chondrocytes may not be the same as what happens clinically (i.e., in an anaerobic
medium).
Systematically repeated treatment
Is it useful and acceptable to propose to re‐inject patients with a low level of pain?
This very relevant issue in daily clinical practice has been addressed in the AMELIA
(Osteoarthritis Modifying Effects of Long‐Term Intra‐Articular Adant) study 44, a
long‐term, randomized, placebo‐controlled trial carried out in 306 patients with symptomatic
KOA who received 4 consecutive cycles (5 weekly IA injections) of either noncrosslinked
HA of biofermentative origin or placebo. Followup visits were conducted at 6 months
after the first and second cycles and at 12 months after the third and fourth cycles,
resulting in a total study duration of 40 months. At the end of the study, the responder
rate according to the Outcome Measures in Rheumatology/OARSI response criteria 57
was significantly higher in the IAHA group than in the control group (80.5% versus
65.8%; RR 1.22 [95% CI 1.07, 1.41], P = 0.004). More interestingly, the number of
responders to IAHA progressively increased after each treatment cycle, while response
to IA placebo remained fairly stable, with a statistically significant between‐group
difference from 1 year onward (P < 0.05). No increase in AEs occurred with repeated
cycles of IAHA.
Based on these results, the ESCEO task force encourages the use of repeated cycles
of IAHA in patients who responded to the first injection, starting a new treatment
cycle as soon as the first symptoms appear.
Molecular weight
There are more than 80 marketed IAHA preparations worldwide. They differ in many characteristics,
including origin (animal versus biofermentation), mean MW (500–6,000 kDa) and MW distribution,
molecular structure (linear, crosslinked, and a mix of both), method of crosslinking,
concentration (0.8–30 mg/ml), volume of injection (0.5–6.0 ml), and posology. Some
of the preparations include different concentrations of additives, such as mannitol,
sorbitol, or chondroitin sulfate. While each of these parameters may theoretically
have an impact on the effect of the IAHA treatment, research has mostly focused on
the potential differences resulting from the size of HA. For this purpose, the exogenous
HA available for IA injections are divided into 3 MW categories: low (500–730 kDa),
intermediate (800–2,000 kDa), and high (2,000–6,000 kDa), including crosslinked formulations
of HA 58. The rationale for an influence of the MW results from research on the mode
of action of IAHA. Indeed, as summarized in the review by Altman et al 53, basic preclinical
science showed that higher MW HAs may provide superior chondroprotective, proteoglycan/glycosaminoglycan
synthesis, antiinflammatory, mechanical, and analgesic effects. Whether these differences
observed in vitro translate into clinical evidence is not clearly established yet.
Two meta‐analyses comparing different IAHA preparations have been published to date
(Table 1). The first one compared hylan with lower‐MW IAHA and found no clinically
relevant benefits in terms of efficacy of either type of preparation. However, the
risks for local AEs (RR 1.91 [95% CI 1.04, 3.49]; I2 = 28%) and postinjection flares
(RR 2.04 [95% CI 1.18, 3.53]; I2 = 0%) were observed to be twice as high with hylan
than with low‐ or intermediate‐MW HA 22. In addition, a higher number of pseudosepsis
cases were reported with hylan than with other IAHA preparations, whose risk might
increase with subsequent courses 40. In the second meta‐analysis, comparing different
commercially available IAHA products, the authors were not able to conclude that one
brand had a better efficacy than another due to the heterogeneity of the studies and
outcomes 25.
In addition, most head‐to‐head RCTs performed to date have found noninferiority with
respect to symptomatic efficacy between the various HA preparations evaluated 59,
60, 61, 62, 63, 64. To our knowledge, only 1 RCT was able to demonstrate a statistically
significant difference between 2 IAHA preparations varying in MW regarding symptomatic
efficacy. Indeed, this study conducted in 400 patients with KOA showed that an intermediate‐MW
HA provided statistically superior pain relief at 6 months than a low‐MW HA. Secondary
end points confirmed the primary finding 58.
Based on the above, the ESCEO task force considers that there is currently no clinical
evidence supporting an advantage in efficacy of one product over another. Furthermore,
if certain intrinsic properties (e.g., MW) of particular IAHA preparations provide
beneficial results in comparison with other IAHA products, characteristics related
to the patient are likely to prevail.
Efficiency
Only 9 pharmacoeconomic studies evaluating the cost‐effectiveness of IAHA have been
published to date (Table 3) 65, 66, 67, 68, 69, 70, 71, 72, 73. Most of them were
conducted in patients with KOA who received IA injections of hylan 65, 66, 67, 71.
Table 3
Pharmacoeconomic studies evaluating the cost‐effectiveness of IAHA in the symptomatic
treatment of knee osteoarthritisa
Reference
Source of data
Country
Duration
Product vs. comparator
Cost‐effectiveness
ICER
Waddell et al, 2001 (65)
Cohortb
US
3 years
Synvisc + AC vs. AC
NA
Dominant
Torrance et al, 2002 (66)
RCT 74
Canada
12 months
Synvisc + AC vs. AC
NA
$10,000 Canadian per QALY
Kahan et al, 2003 (67)
RCT 67
France
9 months
Synvisc vs. CC
€829 for 32% improvement in Lequesne index
Dominant
Yen et al, 2004 (68)
Cohortb
Taiwan
6 months
Artz vs. naproxen
$1,538 US per QALY
$42,000 US per QALY
Mazières et al, 2007 (69)
Cohort 69
France
6 months
Suplasyn vs. CC (BL)
€528 for 27% improvement in Lequesne index
Dominant
Turajane et al, 2007 (70)
Retrospective 75
Thailand
2 years
Hyalgan + AC vs. AC
NA
NA
Chou et al, 2009 (71)
Cohort 71
Taiwan
6 months
Synvisc vs. CC (BL)
$299,456 NT per QALYc
NA
Artz vs. CC (BL)
$394,021 NT per QALY
NA
Miller and Block 2014 (72)
Cohort 76
US
2 years
Registered IAHA vs. CCb
$12,800 US per QALY
< $23,400 US per QALY
Hatoum et al, 2014 (73)
RCTs 77, 78
US
12 months
Euflexxa vs. CC (BL)
$21,281 US per QALY
Dominant
Hatoum et al, 2014 (73)
RCTs 74, 77, 78
US
12 months
Euflexxa vs. AC
$8,816 US per QALY
$38,741 US per QALY
a
ICER = incremental cost‐effectiveness ratio; AC = appropriate care (e.g., nonpharmacologic
therapy, nonsteroidal antiinflammatory drugs [NSAIDs], analgesics, corticosteroids,
total knee arthroplasty); NA = not available; RCT = randomized controlled trial; QALY = quality‐adjusted
life year; CC = conventional care (i.e., nonpharmacologic therapy, NSAIDs, analgesics);
BL = baseline.
b
Hypothetical.
c
$1 NT = $0.0315 US (2008 exchange).
Although pharmacoeconomic data are overall in favor of IAHA, only a few evaluations
included sensitivity analyses or were performed based on RCTs. Furthermore, some costs
were not included, such as those associated with potential AEs. Therefore, the ESCEO
task force recommends investigating more in the identification, measurement, and comparison
of the costs, risks, and benefits of IAHA in the management of KOA.
Conclusion
This review addresses major aspects related to the use of IAHA in the management of
KOA. It emphasizes its good safety profile and its moderate but real efficacy on symptoms,
which is in the same range as other pharmacologic treatment modalities (e.g., NSAIDs)
used in this indication. The effectiveness of IAHA has also been highlighted based
on real‐life clinical data and should therefore be helpful to clinicians when making
care decisions tailored to individual patient needs.
The ESCEO working group is aware that IAHA is not a panacea for treating a heterogeneous
disease such as KOA. However, KOA is known to mostly affect the elderly and individuals
with significant comorbidities, thereby constraining the use of several conventional
therapeutic options such as acetaminophen, NSAIDs, opioids, duloxetine, or TKA. Therefore,
the experts convened by the ESCEO feel that the medical community cannot afford to
neglect the use of IAHA, particularly now that systematic repetitive treatment cycles
have been shown to yield positive results in terms of both efficacy and safety. Indeed,
they consider that IAHA is a reasonable alternative in patients with KOA who have
not sufficiently responded to previous pharmacologic treatments, and a key weapon
in the therapeutic armamentarium for managing KOA in patients with contraindications
to conventional interventions.
The ESCEO task force calls for additional well‐conducted clinical trials (RCTs, cohort
studies) to define the predictive factors (i.e., patient phenotypes, treatment characteristics)
associated with an optimal risk‐benefit ratio. Such research may further help in determining
candidates that may take most advantage of IAHA.
AUTHOR CONTRIBUTIONS
All authors were involved in drafting the article or revising it critically for important
intellectual content, and all authors approved the final version to be submitted for
publication. Dr. Cooper had full access to all of the data in the study and takes
responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design
Cooper, Bruyère, Reginster.
Acquisition of data
Cooper, Rannou, Richette, Bruyère, Herrero‐Beaumont, Migliore, Uebelhart, Reginster.
Analysis and interpretation of data
Cooper, Rannou, Richette, Bruyère, Al‐Daghri, Altman, Brandi, Collaud Basset, Herrero‐Beaumont,
Migliore, Pavelka, Uebelhart, Reginster.
ADDITIONAL DISCLOSURE
Author Collaud Basset is an employee of TRB Chemedica International SA.