The squeaky joint gets the oil: lubricants for osteoarthritis

Hyaluronic acid (HA) is the lubricant that normally oils your joints. In osteoarthritis of the knee, the joints are stiff and painful, and there is less of this lubricant in the joint. So why not inject HA directly into the joints (viscosupplementation)?  Talk about biological plausibility. To many people, this must be very appealing. We know this because people around the world have paid hundreds of millions of dollars to have this stuff injected into their knees. And yet the latest review of the research concludes that HA injections do not provide any significant benefit and may increase the risk of harm. So how do they conclude that, when so many individual studies show that it works? Because the studies are biased.

There have been many studies comparing hyaluronic acid to a placebo or to other forms of treatment. There have also been many reviews, summarising those trials. Overall, when you pool all the individual studies, there is some evidence that hyaluronic acid may provide some short term benefit (even from the 2006 Cochrane review). However, this is often what you get when you provide an overall summary without teasing out the quality of the studies; separating the good from the bad.

Most regular readers know that better study methods lead to a reduction in bias, and that if bias is allowed to creep in, studies like this will favour of the product. So even if a product is not effective, it can seem effective when you add the results of less rigorous studies (with bias towards showing a benefit) to the results of the better studies (which are less likely to show any benefit); the average will come out somewhere in between ‘no effect’ (the truth) and ‘a strong effect’ (the biased results). At least one previous review has shown that the bias in HA studies has largely been responsible for the reported effectiveness, but other reviews (example) have simply added it up and concluded that HA works.

The presentstudy looked beyond this. Firstly, they included studies that were not published (necessary if you want to show all the evidence). But what’s more: they looked at how the results of different studies varied according to factors such as: funding source, blinding of assessors, concealment of randomisation, publication status, and use of a sham. These are the kinds of things that can bias a study. At this point, I got very interested in what they were going to find.

The findings

Pooling all the studies together showed that injecting HA reduced knee pain by the minimum amount deemed to be clinically important, so it just got over the line. No surprises there. The interesting stuff follows.

Studies that were not published showed that HA was not effective. Studies that used a blinded assessor (to reduce bias) showed that HA did not perform too well. Larger studies showed it to be ineffective. All of the factors they looked at showed that reducing the bias was associated with less effectiveness of the HA. For those who understand these things, the funnel plot was the nail in the coffin, showing statistically significant bias in favour of hyaluronic acid.

It’s complicated, but basically the more scientific the study, the less effect there was; the more accurately you looked for an effect, the harder it was to find. The trend was such that if we had the perfect (unbiased) study, it would show no effect whatsoever from HA injections.

Their conclusion:

“In patients with knee osteoarthritis, viscosupplementation is associated with a small and clinically irrelevant benefit and an increased risk for serious adverse events.”

Below is the information for patients, released in the same issue of the journal:

What is the problem and what is known about it so far?

Knee osteoarthritis is a common condition in which changes in the knee joints lead to pain. Non-steroidal anti-inflammatory drugs, such as aspirin, ibuprofen, and naproxen, are the most commonly prescribed agents for this condition, but they do not always work and can have adverse effects. An alternative treatment strategy called viscosupplementation uses hyaluronic acid injection to improve knee function. Hyaluronic acid occurs naturally in the joint fluid and acts as a lubricant and shock absorber, but people with osteoarthritis have less-than-normal amounts of it in their joints.

Why did the researchers do this particular study?

To summarize information about the efficacy and adverse effects of viscosupplementation for knee osteoarthritis.

Who was studied?

Adults with symptomatic knee osteoarthritis who participated in 89 trials of viscosupplementation.

How was the study done?

The authors reviewed clinical trials that randomly assigned adults with symptomatic knee osteoarthritis to receive viscosupplementation or a “control” intervention. The authors used a statistical method called meta-analysis to combine the results of the separate studies to estimate the effect of the therapy on knee pain intensity and physical function. They also used meta-analysis to estimate the safety of the therapy, including the occurrence of “flare-ups” (a hot, painful, swollen knee within 24 to 72 hours after injection) and effusions (excessive joint fluid inside the treated knee after an injection).

What did the researchers find?

The authors identified 177 reports describing 89 trials in 12,667 patients with knee osteoarthritis. When combined and analyzed, large studies of high quality suggested that viscosupplementation had a small, clinically irrelevant effect on pain and no effect on function. Conversely, the authors found an association between viscosupplementation and an increase in adverse events, including flare-ups and effusions, and also life-threatening events or events resulting in hospitalization or disability.

What were the limitations of the study?

The analysis was limited by the poor quality of many of the trials. In particular, many did not provide adequate data on adverse events.

What are the implications of the study?

Viscosupplementation may provide minimal or nonexistent pain relief or function improvement in patients with symptomatic knee osteoarthritis. It also appears to increase the risk for adverse events. Based on these findings, the authors discourage the use of viscosupplementation for knee osteoarthritis.

Addit 17 August 2012: Hyaluronic acid no better than saline in ankle arthritis (here)