Use of Intraarticular Hyaluronic Acid in the Management of Knee Osteoarthritis in Clinical Practice
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 2-5. 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 9-13. 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 2-5, 15, which largely reflect the divergent conclusions drawn by meta-analyses on the topic 16-30. 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 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 16-30. 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). 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. 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 32-34. 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 35-37. 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. 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. 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). Organization (issue date) 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. 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 the clinical effects of IAHA persist for several months have been recent review of the by Altman et al highlighted the role of HA to of in this complex as numerous of its effects of and and in contribute to the and effects as observed in HA to 1 and of transient 1 have also been In the has been suggested to contribute to the antiinflammatory and of of effects of IAHA in joint there is evidence that HA that are in the OA joint can inflammatory or exogenous HA is injected in the the of these is thus The ESCEO group is not that one mode of action is to the of IAHA and the that several mechanisms and to relieve OA Furthermore, in results obtained on may not be the same as clinically in an it and to to patients with a low level of This in daily clinical practice has been in the of study a in patients with symptomatic KOA who received 4 consecutive cycles IA of HA of or placebo. were at 6 months after the first and cycles and at 12 months after the and in a total study of months. the of the study, the according to the in response was significantly higher in the IAHA group than in the group versus RR [95% CI P = the of to IAHA increased after treatment while response to IA placebo with a statistically significant between-group difference from 1 (P < increase in AEs occurred with repeated cycles of IAHA. Based on these the ESCEO task force the use of repeated cycles of IAHA in patients who to the first injection, a treatment as as the first symptoms are more than IAHA preparations They differ in characteristics, including versus mean and molecular and a of method of concentration of injection and of the preparations different of such as or of these may have an on the effect of the IAHA treatment, research has on the differences from the of For this purpose, the exogenous HA available for IA injections are 3 low and high including crosslinked of HA The for an of the results from research on the mode of action of IAHA. as in the review by Altman et al showed that higher may provide superior and these differences observed in clinical evidence is not clearly established Two meta-analyses comparing different IAHA preparations have been published to date (Table 1). The first one compared hylan with IAHA and found no clinically benefits in terms of efficacy of of preparation. However, the for local AEs (RR [95% CI = and flares (RR [95% CI = were observed to be as high with hylan than with or HA In a higher of cases were reported with hylan than with other IAHA risk might increase with courses In the meta-analysis, comparing different available IAHA the authors were not able to that one a better efficacy than due to the of the studies and outcomes In most RCTs to date have found with to symptomatic efficacy between the HA preparations evaluated only 1 was able to a statistically significant difference between 2 IAHA preparations in regarding symptomatic this study in patients with KOA showed that an HA provided statistically superior pain relief at 6 months than a points confirmed the Based on the above, the ESCEO task force considers that there is currently no clinical evidence an in efficacy of one over Furthermore, properties of particular IAHA preparations provide results in comparison with other IAHA related to the patient are to Only studies evaluating the cost-effectiveness of IAHA have been published to date (Table of them were in patients with KOA who received IA injections of hylan data are in favor of IAHA, only a few or were based on Furthermore, some were not such as those associated with Therefore, the ESCEO task force recommends more in the and comparison of the and benefits of IAHA in the management of KOA. This review aspects related to the use of IAHA in the management of KOA. its safety profile and its moderate but real efficacy on which is in the same as other pharmacologic treatment modalities in this The effectiveness of IAHA has also been highlighted based on real-life clinical data and should be to when making to individual patient The ESCEO working group is that IAHA is not a for treating a disease such as KOA. However, KOA is to the and with significant comorbidities, the use of several conventional options such as acetaminophen, NSAIDs, or Therefore, the experts convened by the ESCEO feel that the medical cannot to the use of IAHA, particularly that treatment cycles have been shown to results in terms of both efficacy and they that IAHA is a alternative in patients with KOA who have not to previous pharmacologic and a in the for KOA in patients with to conventional The ESCEO task force for additional clinical trials cohort to the patient phenotypes, treatment associated with an risk-benefit Such research may help in determining that may take most of IAHA. All authors were in the or it for and all authors approved the to be for to all of the data in the study and for the of the data and the of the data is an of International