Journal reference: Wandel S, Jüni P, Tendal B, et al. Effects of glucosamine, chondroitin, or placebo in patients with osteoarthritis of hip or knee: network meta-analysis. BMJ 2010; 341: c4675
Published: 17 September 2010
|Glucosamine does not appear to be effective for knee or hip osteoarthritis pain.
The evidence is robust. Even using the most favourable interpretation of the estimate of effect, glucosamine does not have a clinically significant effect on knee or hip osteoarthritis pain.
Doris, a 58-year old lady suffers from osteoarthritis. She experiences this condition mostly as chronic knee pain. One day she asks me, “I’ve heard a lot about glucosamine. What do you think?”
Glucosamine is certainly popular and recommended enthusiastically by many. However, what does the research evidence say?
Does oral glucosamine reduce the pain from (knee) osteoarthritis?
What does the research evidence say?
Step 1: The Cochrane Library
The Cochrane Library has a systematic review and meta-analysis on glucosamine therapy for osteoarthritis by Towheed et al. (2009) that is up-to-date to November 2008. 1 Although this Cochrane review is commonly cited as evidence for the efficacy of glucosamine, a benefit is only found when low quality studies are included in the analysis. When the analysis is restricted to high quality studies, glucosamine does not appear different to placebo.
Step 2: PubMed
Are there any newer systematic reviews and/or meta-analyses? I used the search strategy:
glucosamine and osteoarthritis; limited to systematic review OR meta-analysis, English, humans and published since November 2008 only
A newer meta-analysis by Wandel et al. (2010) that was published in the BMJ was identified. Let’s have a closer look at this article. 2
This article is similar to a systematic review so I will use the appraisal sheet for systematic reviews available from the Centre for Evidence Based Medicine. 3
What PICO question does the systematic review ask?
In people with knee or hip osteoarthritis (Participants); what is the effect of glucosamine sulphate, glucosamine hydrochloride, chondroitin sulphate, or the combination (Intervention); compared to placebo (Comparator); on pain reported in any of nine time windows (from up to 3 months to 22 months or more) (Outcome).
Is it clearly stated?
Is it unlikely that important studies were missed?
Yes. The authors searched four large electronic databases. They also manually searched conference proceedings, textbooks, and reference lists of all obtained papers. The authors contacted content experts.
Were the criteria used to select articles for inclusion appropriate?
Probably yes. The authors included randomised controlled trials (RCTs) that compared any of the three aforementioned substances, or their combination, with either placebo or head-to-head. They excluded trials that used sub-therapeutic doses (<1500 mg/d of glucosamine and <800 mg/d of chondroitin). Notably, the authors only included larger RCTs that with an average of at least 100 patients per study arm – which would have excluded many smaller trials. It is not entirely clear why the authors used this criterion.
Were the included studies sufficiently valid for the question asked?
Probably yes. Two of the authors independently assessed the quality of the included studies. Of the included studies most described adequate concealment of allocation and adequate blinding of patients.
Were the results similar between studies?
Yes. There was little heterogeneity between trials for pain.
What were the results?
The authors identified 10 randomised trials. Glucosamine resulted in slightly less pain than placebo, with an average improvement of -4 mm (95% confidence interval -7 mm to -1 mm) on a 100 mm visual analogue scale (VAS) compared to placebo. The result is effectively the same when the analysis was limited only to glucosamine sulphate.
Although there is a statistically significant effect, this difference is not even close to being clinically significant. The authors suggest that the minimum clinically significant change is -9 mm on the 100 mm VAS for osteoarthritis. This is not reached with glucosamine, even with the most favourable interpretation of the confidence interval. 2 Curiously, when the analysis is restricted to trials that had funding that was independent of industry, the effect vanishes – glucosamine as compared to placebo resulted in a non-statistically significant and miniscule -1 mm difference on a 100 mm VAS.
I advised Doris that the best evidence so far suggests that glucosamine won’t help her knee pain, though it does appear to be very safe. 1 I encouraged her to keep up with her exercise regimen and to use paracetamol regularly.
The confidence interval gives the range within which the true effect is likely to lie. It can be seen as the plausible upper and lower bounds on the size of any true effect. Moreover, the width of the confidence interval provides information on the precision of the estimate of the true effect. 4
- Towheed T, Maxwell L, Anastassiades TP, et al. Glucosamine therapy for treating osteoarthritis. Cochrane Database of Systematic Reviews 2005, Issue 2. Art. No.: CD002946.
- Wandel S, Jüni P, Tendal B, et al. Effects of glucosamine, chondroitin, or placebo in patients with osteoarthritis of hip or knee: network meta-analysis. BMJ 2010; 341: c4675
- Systematic Review: Are the results of the review valid? Centre for Evidence Based Medicine, University of Oxford. http://www.cebm.net/index.aspx?o=1157 Retrieved: 2012 October 23
- Davies HT, Crombie IK. What are confidence intervals and p-values? Hayward Medical Communications: 2009 Aug. http://www.medicine.ox.ac.uk/bandolier/painres/download/whatis/what_are_conf_inter.pdf Retrieved: 2012 October 23