Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Hyaluronic acid injections for OA knee

Systematic reviews

Hyaluronic acids are large glucosminoglycans in synovial fluid. They have high but variable molecular weight and viscosity, and injecting them into the knee joint is aimed at providing lubrication and shock absorption. Usually three to five weekly injections comprise a course of treatment.

There has been a degree of controversy over efficacy of this treatment, and two new systematic reviews [1,2] are sceptical. There are a number of variables, including the molecular weight of the hyaluronic acid preparation used, over what period injections are given, when benefits might be seen, and what those benefits are.

Systematic reviews

The first [1] had a wide search strategy to February 2003 and included only randomised trials, of at least single blind, testing intra-articular hyaluronic acid administered at least weekly for three weeks, against intra-articular placebo, in patients with osteoarthritis (OA). Pain had to be measured and reported using standard pain measures in osteoarthritis. The second [2] used a similar strategy to October 2002, included case series, but excluded trials before 1995 in order to examine the most up-to-date literature.


The first review [1] included 22 trials, 19 published in full, with 2949 patients. Trial size was 24 to 408 participants. Effect size was calculated for each study, and pooled. Of the 22 trials, only three individually had a statistically significant effect size. Overall the effect size was 0.3 (95% CI 0.2 to 0.5), indicating a small effect. Omitting three trials with the largest molecular weight (6,000 kD), the effect size was even smaller at 0.2 (0.1 to 0.3).

Three trials (268 patients) used 6,000 kD hyaluronic acid, one of which was very small, with just 30 patients. The two larger studies differed in their conclusion, one with a very large effect size, and one no different from placebo.

The second review included 13 randomised trials and five case series. The randomised trials were included in the first review. Three of the five case series were prospective, were small, and lasted six months to two years. Three used 6,000 kD hyaluronic acid, but only one was prospective. All reported some degree of pain relief in some patients.

Adverse events reported included injection site pain and swelling in 2% to 23% of injections. Gastrointestinal adverse events and back pain were also reported.


The sort of effect size seen in these trials is small, about that for the effect of NSAIDs over paracetamol. The result itself is not robust. Trials were often small, there was clinical heterogeneity regarding type of hyaluronic acid, dose, outcome measured, and duration of the study (which could have been six weeks to a year). So even the small effect seen might be overstated. Most of the randomised trials were company sponsored.

The evidence for a big effect is underwhelming. The evidence for any effect carries limited weight. The evidence is that there will be harm to be balanced against any small benefit. Not entirely convincing, this.

The real disappointment comes from the reporting. Effect size is not intuitively helpful, though it is useful when trying to pool information from different outcomes. Bandolier looks for outcomes that are more meaningful, like patients improved, or changes in a scale, or, better still, some clinically useful but simply described outcome that we can understand. Then we have the chance of comparing interventions, and can check whether the patients in different trials are the same. Here we failed.


  1. GH Lo et al. Intra-articular hyaluronic acid in treatment of knee osteoarthritis: a meta-analysis. JAMA 2003 290: 3115-3121.
  2. A Aggarwal, IP Sempowski. Hyaluronic acid injections for knee osteoarthritis. Systematic review of the literature. Canadian Family Physician 2004 50: 249-256.

previous or next story