|
Clinical question
Are intra-articular injections of hyaluronic acid effective for
osteoarthritis of the knee?
Bottom line
The evidence that intra-articular hyaluronic acid helps patients with knee
osteoarthritis is of poor quality. Improvements in pain at rest and pain
during exercise is seen in a minority of studies, and those studies were of
lower quality than those showing no benefit. There is no evidence of
functional improvement. Injections like this have a potentially powerful
placebo effect, so any benefit seen in unblinded studies without concealed
allocation is likely represent the placebo effect rather than any effect of
the drug.
Reference
Arrich J, Piribauer F, Mad P, Schmid D, Klaushofer K, Mullner M. Intra-articular
hyaluronic acid for the treatment of osteoarthritis of the knee: systematic
review and meta-analysis. CMAJ 2005; 172:1039-43.
Study design: Meta-analysis (randomized controlled trials)
Setting: Outpatient (specialty)
Synopsis
A previous meta-analysis (JAMA 2003; 290:3115-21) found that intra-articular
hyaluronic acid (Synvisc) injections for osteoarthritis of the knee were
minimally effective at best, while a recent Cochrane Review (whose authors
received funding from the manufacturer) found it much more effective. This
meta-analysis looked at the same question, but differed from the previous 2
analyses by examining separate outcomes (pain at rest, pain during exercise,
and function), rather than lumping them together. The current study was
sponsored by a national insurance program in Austria (we can imagine that
they would prefer a more skeptical analysis than the manufacturer). After a
careful literature search, 22 randomized controlled trials were included.
The study quality was generally poor: only 7 concealed allocation; 6
presented usable data from an intention-to-treat analysis; 16 had blinded
outcome assessment; and only 4 did all of this correctly. Eight studies with
468 patients reported 10 outcomes for pain at rest at 2 to 6 weeks. Six of
10 did not demonstrate any benefit; 1 poorly designed, industry-sponsored
study reported a large benefit for both 20 mg and 40 mg outcomes. There was
too much heterogeneity to combine studies and poorer quality trials tended
to find a greater benefit. Nine studies with 1141 patients reported 10
outcomes for pain during exercise. Pooled data at 10 weeks to 14 weeks and
20 weeks to 30 weeks found a statistically -- but probably not clinically --
significant benefit. Hyaluronic acid had no significant effect on function
at any time. The previous JAMA analysis found that 17 of 22 trials were
industry sponsored.
|