Hyaluronic Acid for Joints: Injections, Oral Forms and Osteoarthritis

Every movable joint in the body sits in a small pool of thick, slippery fluid, and much of that thickness comes from one molecule: hyaluronic acid, also called hyaluronan. The body builds it from two simple sugars — glucuronic acid and N-acetylglucosamine, a close relative of glucosamine — and strings them into chains that can reach millions of daltons. In a healthy joint those long chains give the fluid its unusual, gel-like flow. In osteoarthritis and rheumatoid arthritis the chains are fewer and shorter.

That observation led to one of the most studied, and most argued-over, treatments in joint medicine: injecting hyaluronic acid straight into the knee, a practice called viscosupplementation. This page follows the story from the biology of joint fluid, through four decades of trials whose results split in two directions, to the 2019 American College of Rheumatology guideline, the hip studies, the smaller body of research on hyaluronic acid taken by mouth, and the safety findings. Favourable and unfavourable analyses are reported side by side, each attributed to its authors.

Table of Contents

  1. 1. Hyaluronan in Healthy Synovial Fluid
  2. 2. What Changes in Osteoarthritis and Rheumatoid Joints
  3. 3. Viscosupplementation: The Injection Approach
  4. 4. Early Meta-Analyses and the Cochrane Review
  5. 5. Larger Trials and the 2022 BMJ Analysis
  6. 6. What Guidelines Say About Joint Injections
  7. 7. Hip Osteoarthritis and Molecular Weight
  8. 8. Oral Hyaluronan and Knee Comfort Studies
  9. 9. Safety Findings From Injection Trials
  10. Key Research Papers
  11. Connections

1. Hyaluronan in Healthy Synovial Fluid

Synovial fluid is the lubricating liquid inside the capsule of a joint such as the knee, hip or knuckle. Hyaluronan is described in a classic 1996 overview by Laurent, Laurent and Fraser as “a major component of synovial tissue and fluid.” A 2018 review of hyaluronic acid research gives its concentration in synovial fluid as about 3–4 mg per millilitre — some thirty to forty times the level in the vitreous jelly of the eye (about 0.1 mg/mL), and similar to the level in the umbilical cord’s Wharton’s jelly.

What makes hyaluronan special is not just how much of it there is but how long the chains are. Each molecule is a single unbranched strand of repeating sugar pairs. Laurent and colleagues explain that even at dilute concentrations — under 1 mg/mL — these strands tangle into networks, and that tangling gives hyaluronan solutions their unusual rheological (flow) properties. At the 3–4 mg/mL found in a joint, hyaluronan is several times above that tangling threshold.

Proposed roles in the joint

Researchers have long proposed several jobs for joint hyaluronan, including lubrication, holding water in the tissue, and scavenging free radicals and debris. Laurent, Laurent and Fraser were careful about this list: they described the joint roles as “more or less well documented hypotheses,” meaning some are better supported than others. The free-radical idea is examined in detail on the sibling page Hyaluronic Acid as an Antioxidant.

A molecule in constant turnover

Joint hyaluronan is not a permanent fixture. The body keeps making it and breaking it down. One review estimates its half-life in cartilage at about 1–3 weeks, compared with roughly a day in skin; across the whole body, Stern estimated that a 70-kg adult holds about 15 g of hyaluronan, of which about 5 g is turned over every day. The joint’s supply therefore depends on the lining cells of the joint continuing to make new chains — which is exactly what appears to falter in arthritis.

Back to Table of Contents

2. What Changes in Osteoarthritis and Rheumatoid Joints

In arthritis, synovial-fluid hyaluronan is found at a lower concentration and in smaller pieces than in healthy joints. Two separate lines of research help explain why: the joint lining makes less of it, and what is there gets broken apart.

Less made, more broken down

Hyaluronan is made by three enzymes called hyaluronan synthases (HAS1, HAS2 and HAS3) and broken down by enzymes called hyaluronidases. In 2004 a team led by Yoshida measured the genetic messages for these enzymes in knee synovium — the tissue lining the joint — from 17 people with osteoarthritis, 14 with rheumatoid arthritis and 20 healthy donors. In both diseases, messages for HAS1 and HAS2 were lower than in the healthy tissue, while the message for the breakdown enzyme HYAL2 was higher. The authors offered this pattern as a possible reason the joint fluid hyaluronan is lower in concentration and smaller in size in these conditions.

Reactive oxygen chops the chains

The second process is chemical. An inflamed joint produces reactive oxygen species, including the very reactive hydroxyl radical. In a 2003 laboratory study, Yamazaki and colleagues showed that hydroxyl radicals generated in a Fenton-reaction system depolymerised (cut up) hyaluronan, and related this to the smaller hyaluronan found in arthritic joints. A 2006 review led by Ladislav Šoltés described how reactive oxygen species degrade the high-molecular-weight hyaluronan of synovial fluid, and noted that the resulting loss of molecular weight and viscosity is used in laboratories as a marker of oxidative damage.

Why fragment size matters

Broken hyaluronan is not simply weaker hyaluronan. Stern and colleagues reviewed evidence that large chains are space-filling and tend to dampen inflammation and immune activity, while small fragments do the opposite — they are inflammatory, stimulate the immune system and can act as internal “danger signals.” Fragments created during injury, including by reactive oxygen, can therefore add to inflammation rather than calm it. This size-dependence later became the rationale for testing high-molecular-weight injection products, discussed in section 7.

Back to Table of Contents

3. Viscosupplementation: The Injection Approach

The idea behind viscosupplementation is simple to state: if the joint fluid has lost its thick, long-chain hyaluronan, put some back. A clinician injects a hyaluronic acid gel directly into the joint space, most often the knee, usually as a single injection or a short series.

From rooster combs to fermentation tanks

Medical hyaluronic acid began as an animal extract, part of the long tradition of medicines drawn from natural sources. According to the 2018 review by Fallacara and colleagues, hyaluronan was isolated during the 1930s to 1950s from human umbilical cord, rooster comb and streptococcal bacteria, and the first pharmaceutical-grade hyaluronic acid was produced in 1979 by Balazs from rooster combs and umbilical cords. The same review describes bacterial fermentation as the later industrial route. The molecule itself is identical across sources — the same glucuronic-acid and N-acetylglucosamine chain the body makes in its own joints.

Not one product but many

“Hyaluronic acid injection” covers a wide family of products. They differ in the source of the hyaluronan, in its molecular weight (low, medium or high), in concentration and volume, and in whether the chains are chemically cross-linked to make a firmer, longer-lasting gel. In the 2012 meta-analysis led by Rutjes, 22 of 89 trials used a cross-linked product. Authors of almost every review below flag this product variety as one reason pooled results are hard to interpret.

What the injections were expected to do

Supporters proposed that injected hyaluronan would restore the fluid’s viscosity and cushioning, and perhaps encourage the joint lining to make its own. Whether the benefit exists at all, and how large it is, is the question the trials tried to answer.

Back to Table of Contents

4. Early Meta-Analyses and the Cochrane Review

The first wave of pooled analyses was favourable to viscosupplementation.

Cochrane, 2006

In 2006 a Cochrane review led by Bellamy pooled 76 trials of viscosupplementation for knee osteoarthritis. The reviewers concluded that viscosupplementation was effective, with the clearest effects between 5 and 13 weeks after treatment, and that few adverse events were reported. They also warned that products differed from one another, so results from one could not simply be applied to another.

The 2015 network meta-analysis

In 2015 Bannuru and colleagues published a network meta-analysis in the Annals of Internal Medicine comparing the common drug treatments for knee osteoarthritis — 137 studies covering 33,243 people. Ranked against oral placebo at three months, intra-articular hyaluronic acid showed the largest effect on pain of any treatment examined (effect size 0.63). The authors added an important caution: intra-articular injections of any kind carry a sizeable placebo effect of their own, because an injected placebo (usually saline) also tends to improve pain. Part of the apparent advantage of hyaluronic acid over oral drugs, they noted, may reflect that injection effect rather than the hyaluronan.

Why these results looked strong

Both analyses pooled the eligible trials together, without setting the largest, best-blinded or unpublished trials apart. Later work, described in the next section, asked what happens when the analysis gives more weight to the biggest, best-blinded trials — and when unpublished trials are included.

Back to Table of Contents

5. Larger Trials and the 2022 BMJ Analysis

The second wave of analyses, by a group whose authors include Jüni and da Costa, reached a very different conclusion.

Rutjes and colleagues, 2012

Rutjes and colleagues pooled 89 randomised trials in the Annals of Internal Medicine. Across all trials, viscosupplementation reduced pain with an effect size of −0.37. But the effect shrank sharply in the most reliable trials: in large trials with blinded outcome assessment it was −0.11, which the authors described as clinically irrelevant, and in unpublished trials it was −0.03 — essentially zero. The pattern suggested that small and published trials had overstated the benefit.

Pereira and colleagues, BMJ 2022

Ten years later, Pereira and colleagues updated the work in the BMJ with 169 trials and 21,163 participants. In the large, placebo-controlled trials the pooled difference in pain was a standardised mean difference of −0.08, which they translated as about 2 mm on a 100 mm pain scale — below the smallest difference patients notice. Using a cumulative analysis, they reported “conclusive evidence of clinical equivalence” between viscosupplementation and placebo since 2009, and concluded that their “findings do not support broad use” of viscosupplementation for knee osteoarthritis.

Putting the two waves together

The two waves drew on overlapping sets of trials, and much of the difference lies in what was weighted. The favourable analyses pooled trials broadly; the unfavourable ones showed that the benefit shrinks as trials get larger, better blinded and more complete, including unpublished ones. The Bannuru analysis had already raised a related point: an injected placebo also improves knee pain, so the comparison that matters is hyaluronic acid against a saline injection.

Back to Table of Contents

6. What Guidelines Say About Joint Injections

Clinical guidelines translate trial evidence into formal recommendations. The most widely cited in North America is the 2019 guideline of the American College of Rheumatology and the Arthritis Foundation for osteoarthritis of the hand, hip and knee, led by Kolasinski and published in Arthritis & Rheumatology in 2020.

The 2019 ACR/Arthritis Foundation recommendation

The guideline’s recommendations on intra-articular hyaluronic acid, as stated in its full text, are:

The guideline explains that in trials with a low risk of bias, the effect of hyaluronic acid injections compared with saline injections “approaches zero.”

How this fits the evidence above

The guideline panel’s position lines up with the Rutjes and Pereira analyses, which emphasise the large, low-bias trials. It does not deny that individual earlier trials and the 2006 Cochrane review found benefit; it reflects a judgement that, once bias is accounted for, the added effect over a saline injection is very small.

Back to Table of Contents

7. Hip Osteoarthritis and Molecular Weight

Most viscosupplementation research concerns the knee; the hip has been studied in far fewer trials.

The 2025 hip review

In 2025 Migliorini, Maffulli and colleagues published a systematic review of randomised trials of hyaluronic acid injections for hip osteoarthritis, covering 982 patients. They compared high-molecular-weight hyaluronic acid with medium-molecular-weight hyaluronic acid, placebo and control groups. At 3–4 months there were no significant differences between the groups. At 4–6 months, the high-molecular-weight group reported lower pain scores on a visual analogue scale than each of the other groups; on the WOMAC osteoarthritis index, both high- and medium-molecular-weight hyaluronic acid did better than control, with no difference between the two.

One detail in the paper is worth noting for accuracy: the concluding sentence of its published abstract refers to “knee OA symptoms,” although the review itself concerns the hip. This appears to be a slip in the abstract; the results reported on this page are the hip results.

Why molecular weight is tested at all

The interest in high-molecular-weight products follows from the biology in section 2: long hyaluronan chains are associated with the joint’s normal viscosity and with damping inflammation, while short fragments are associated with inflammation. A product closer to the size of natural joint hyaluronan was therefore expected to perform better. The hip review’s result is consistent with that expectation, though it rests on a small number of trials.

Reading it beside the guideline

The Migliorini finding sits alongside the 2019 ACR/Arthritis Foundation guideline, which strongly recommends against intra-articular hyaluronic acid for hip osteoarthritis. The two are not directly comparable — the review compared product types and included non-saline control groups, and it was published after the guideline — but readers will find both positions cited in discussions of hip injections.

Back to Table of Contents

8. Oral Hyaluronan and Knee Comfort Studies

Hyaluronic acid is also sold as a capsule or drink. Here the questions are different: not only whether it helps, but whether a large sugar chain swallowed by mouth reaches the joints at all.

The knee trials

In 2016 Oe and colleagues published a review in Nutrition Journal of randomised controlled trials of oral hyaluronan for knee pain conducted between 2008 and 2015. They reported that oral hyaluronan relieved knee pain and reduced synovial effusion (fluid swelling in the joint) in the trials reviewed. Several of the review’s authors were employees of a company that supplies hyaluronic acid. That connection does not by itself make the findings wrong, but no independent meta-analysis of oral hyaluronan for knee osteoarthritis was found for this page, and the oral trials are small compared with the injection trials.

Does swallowed hyaluronan get absorbed?

Animal studies disagree. In rats, Kimura and colleagues found that hyaluronan survived artificial stomach and small-intestine juices but was broken down by bacteria in the large intestine into small oligosaccharides, which could pass the gut wall and were later detected in skin. A team led by Lazníček, using radiolabelled hyaluronan in rats, found “no significant absorption” of the labelled molecule into the central circulation and only traces with a second labelling method. Kimura’s own paper summarises the situation plainly: the absorption of orally administered hyaluronan “remains controversial.” No human pharmacokinetic study was verified for this page. The full animal evidence is set out on the forms, absorption and safety page.

Food and the body’s own production

Hyaluronan is built by the joint lining from sugars derived from glucose, so the body does not depend on dietary hyaluronan to make it. Animal connective tissues such as rooster comb are traditional extraction sources, and bone broth contains hyaluronan together with chondroitin, but no verified source gives the hyaluronan content of any food or shows that a particular food raises joint hyaluronan in people. Related joint nutrients are covered on the site’s glucosamine, collagen and bone broth joint pages.

Back to Table of Contents

9. Safety Findings From Injection Trials

As with benefit, the safety findings for viscosupplementation depend on which trials are pooled and how adverse events are classified. Three analyses frame the discussion.

A safety meta-analysis: no excess serious events

In 2021 Miller and colleagues, publishing in Cartilage, pooled 35 randomised trials comparing intra-articular hyaluronic acid with saline in 8,078 patients with knee osteoarthritis. They found no difference between the groups in overall adverse events, serious adverse events or withdrawals because of adverse events. Non-serious local reactions were somewhat more common with hyaluronic acid (14.5% versus 11.7%) and typically resolved within days. The 2006 Cochrane review had likewise reported few adverse events.

Two analyses that found more serious events

The Rutjes 2012 analysis found a higher risk of serious adverse events with viscosupplementation than with control (relative risk 1.41, 95% confidence interval 1.02–1.97). The Pereira 2022 BMJ analysis, restricted to large placebo-controlled trials, found a similar signal (relative risk 1.49, 95% confidence interval 1.12–1.98). Both groups weighed this against a benefit they judged to be clinically negligible.

How the findings differ

The analyses differ in which trials they included, how they defined serious events, and which comparison groups they accepted. Each set of authors presents its own conclusion, and this page does not attempt to reconcile them beyond what they themselves say. The Miller analysis found the excess with hyaluronic acid to be in non-serious, short-lived local reactions; the Rutjes and Pereira analyses reported their excess in serious adverse events.

Where the joint evidence stands

Back to Table of Contents

Key Research Papers

  1. Laurent TC, Laurent UB, Fraser JR. The structure and function of hyaluronan: An overview. Immunol Cell Biol. 1996;74(2):A1-7. PubMed PMID: 8724014
  2. Fallacara A, Baldini E, Manfredini S, Vertuani S. Hyaluronic Acid in the Third Millennium. Polymers (Basel). 2018;10(7):701. PubMed PMID: 30960626
  3. Stern R. Hyaluronan catabolism: a new metabolic pathway. Eur J Cell Biol. 2004;83(7):317-25. PubMed PMID: 15503855
  4. Yoshida M, Sai S, Marumo K, Tanaka T, Itano N, Kimata K, Fujii K. Expression analysis of three isoforms of hyaluronan synthase and hyaluronidase in the synovium of knees in osteoarthritis and rheumatoid arthritis by quantitative real-time reverse transcriptase polymerase chain reaction. Arthritis Res Ther. 2004;6(6):R514-20. PubMed PMID: 15535829
  5. Yamazaki K, Fukuda K, Matsukawa M, Hara F, Yoshida K, Akagi M, Munakata H, Hamanishi C. Reactive oxygen species depolymerize hyaluronan: involvement of the hydroxyl radical. Pathophysiology. 2003;9(4):215-220. PubMed PMID: 14567924
  6. Soltés L, Mendichi R, Kogan G, Schiller J, Stankovska M, Arnhold J. Degradative action of reactive oxygen species on hyaluronan. Biomacromolecules. 2006;7(3):659-68. PubMed PMID: 16529395
  7. Stern R, Asari AA, Sugahara KN. Hyaluronan fragments: an information-rich system. Eur J Cell Biol. 2006;85(8):699-715. PubMed PMID: 16822580
  8. Bellamy N, Campbell J, Robinson V, Gee T, Bourne R, Wells G. Viscosupplementation for the treatment of osteoarthritis of the knee. Cochrane Database Syst Rev. 2006;2006(2):CD005321. PubMed PMID: 16625635
  9. Bannuru RR, Schmid CH, Kent DM, Vaysbrot EE, Wong JB, McAlindon TE. Comparative effectiveness of pharmacologic interventions for knee osteoarthritis: a systematic review and network meta-analysis. Ann Intern Med. 2015;162(1):46-54. PubMed PMID: 25560713
  10. Rutjes AW, Jüni P, da Costa BR, Trelle S, Nüesch E, Reichenbach S. Viscosupplementation for osteoarthritis of the knee: a systematic review and meta-analysis. Ann Intern Med. 2012;157(3):180-91. PubMed PMID: 22868835
  11. Pereira TV, Jüni P, Saadat P, Xing D, Yao L, Bobos P, Agarwal A, Hincapié CA, da Costa BR. Viscosupplementation for knee osteoarthritis: systematic review and meta-analysis. BMJ. 2022;378:e069722. PubMed PMID: 36333100
  12. Kolasinski SL, Neogi T, Hochberg MC, Oatis C, Guyatt G, Block J, Callahan L, Copenhaver C, Dodge C, Felson D, Gellar K, Harvey WF, Hawker G, Herzig E, Kwoh CK, Nelson AE, Samuels J, Scanzello C, White D, Wise B, Altman RD, DiRenzo D, Fontanarosa J, Giradi G, Ishimori M, Misra D, Shah AA, Shmagel AK, Thoma LM, Turgunbaev M, Turner AS, Reston J. 2019 American College of Rheumatology/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee. Arthritis Rheumatol. 2020;72(2):220-233. PubMed PMID: 31908163
  13. Migliorini F, Pilone M, Mazzoleni MG, Schäfer L, Katusic D, Maffulli N. Intra-articular hyaluronic acid injections for hip osteoarthritis: a level I systematic review. Eur J Orthop Surg Traumatol. 2025;35(1):180. PubMed PMID: 40343507
  14. Oe M, Tashiro T, Yoshida H, Nishiyama H, Masuda Y, Maruyama K, Koikeda T, Maruya R, Fukui N. Oral hyaluronan relieves knee pain: a review. Nutr J. 2016;15:11. PubMed PMID: 26818459
  15. Kimura M, Maeshima T, Kubota T, Kurihara H, Masuda Y, Nomura Y. Absorption of Orally Administered Hyaluronan. J Med Food. 2016;19(12):1172-1179. PubMed PMID: 27982756
  16. Laznicek M, Laznickova A, Cozikova D, Velebny V. Preclinical pharmacokinetics of radiolabelled hyaluronan. Pharmacol Rep. 2012;64(2):428-37. PubMed PMID: 22661195
  17. Miller LE, Bhattacharyya S, Parrish WR, Fredericson M, Bisson B, Altman RD. Safety of Intra-Articular Hyaluronic Acid for Knee Osteoarthritis: Systematic Review and Meta-Analysis of Randomized Trials Involving More than 8,000 Patients. Cartilage. 2021;13(1_suppl):351S-363S. PubMed PMID: 31735075

PubMed Topic Searches

  1. Viscosupplementation for knee osteoarthritis: meta-analyses
  2. Intra-articular hyaluronic acid for hip osteoarthritis
  3. Synovial fluid hyaluronan and molecular weight in osteoarthritis
  4. Oral hyaluronan and knee pain

Back to Table of Contents

Connections