Skip navigation

Glucosamine and Arthritis

Check out the updated version of this story (March 2001) here

In Bandolier 's Question Time ( Bandolier 42 ), Dr Proctor from Gainsborough asked what evidence there was that glucosamine was an effective treatment for arthritis. Given the amount of publicity it had been getting, we thought this was a good question. We had a number of responses, and decided to follow this up by trying to find papers and assess what evidence we could find.

Searching

This involved MEDLINE searching for articles on glucosamine, finding Internet pages which featured glucosamine, reference lists from retrieved articles, and references provided by Bandolier readers.

Results

Eight randomised trials [1-8] involving oral or intramuscular glucosamine were found and the details are in the large Table below. Articles on intra-articular injection, or where the material used was not clearly defined as glucosamine were excluded, as were non-randomised case series. All those included examined oral and/or intramuscular glucosamine in patients with arthritis over periods of up to eight weeks. Most had well-described methods and six had quality scores of 3 or more on a five-point scale [9]. Oral doses of glucosamine sulphate were 1.5 grams a day, and intramuscular doses were 400 mg twice or three times a week.
Reference Patients Design Drugs & Doses Outcomes Results Quality score
Mund-Hoym, 1980 80 patients with "vertebralsyndome". Mean age 58 years Randomised, parallel-group 3 injections of 400 mg glucosamine sulphate a week, plus oral glucosamine 250 mg 2/3 times a day on non-injection days (40 patients) Daily injections of 600 mg phenylbutazone (40 patients) All IM injections Number of days to clinical improvement. Global good/bad Glucosamine 32 days (± 1.3, range 21 - 48). Phenylbutazone 46 days (± 1.7, range 27 - 80). Good/bad outcome 34/6 for glucosamine, 29/11 for phenylbutazone R =1 DB=0 WD=0 Total=1
Pujalte et al, 1980 24 patients with osteoarthritis of the knee. Mean age 60 years Randomised, double-blind, parallel-group, 6-8 weeks 500 mg glucosamine sulphate, three times daily or identical placebo No other analgesics allowed Articular pain, swelling, movement using categorical scale (patient and doctor) Glucosamine significantly (p<0.01) better than placebo for pain, tenderness and swelling. Time for clinical improvement 14 days for glucosamine cf 40 days for placebo. No AE with glucosamine. R =2 DB=2 WD=1 Total=5
Crolle & D'Este, 1980 30 in-patients with chronic osteoarthrosis. Mean age 72 years Randomised, double-blind, parallel-group, 21 days 400 mg intramuscular glucosamine sulphate for 7 days, followed by 500 mg orally three times a day for 14 days. IM piperazine/chlorbutanol for 14 days, followed by oral placebo for 14 days. Pain at rest or movement, categorical scale. Walking time over 20 metres. Glucosamine significantly (p<0.01) better than placebo for pain and function restriction at 21 days. Overall symptom score better with glucosamine. No difference in walking times, though better. No AE. R =1 DB=2 WD=0 Total=3
D'Ambrosio et al, 1981 30 in-patients with chronic degenerative osteoarthrosis. Mean age 75 years. No steroids or NSAIDs for 2 weeks before trial. Randomised, open, parallel-group, 21 days 400 mg intramuscular glucosamine sulphate for 7 days, followed by 500 mg orally three times a day for 14 days. IM piperazine/chlorbutanol for 14 days, followed by oral placebo for 14 days. Pain at rest or movement, categorical scale. Glucosamine significantly (p<0.01) better than placebo for symptom score at 21 days. (pain at rest and movement appear significantly improved, though no stat test done). No AE. R =1 DB=0 WD=0 Total=1
Vaz, 1982 40 out-patients with unilateral osteoarthritis of the knee without complications. Mean age 58 years. Randomised, double-blind, parallel-group, 8 weeks 1.5 g/day of glucosamine sulphate or 1.2 g ibuprofen. Pain. categorical scale. Ibuprofen significantly better than glucosamine at 2 weeks, glucosamine significantly better than ibuprofen at 8 weeks. AE (mild) reported by 2 on glucosamine and 5 on ibuprofen. Overall efficacy (doctor) good 8/18 glucosamine and 3/20 ibuprofen. R =1 DB=1 WD=1 Total=3
Rovalti, 1992, Study 1 252 out-patients with gonarthrosis. Randomised, double-blind, parallel-group, 4 weeks 1.5 g/day of glucosamine sulphate or placebo. Lesquesne index, responders/non-responders Glucosamine significantly (p<0.05) better than placebo for symptom score at 4 weeks. Responders 66/126 glucosamine, 46/126 placebo. Minor AE 8/126 glucosamine, 13/126 placebo. R =1 DB=1 WD=1 Total=3
Reichelt et al, 1994 (duplicated in Rovalti, 1992, Study 2) 155 out-patients with gonarthrosis. Randomised, double-blind, parallel-group, 6 weeks Intramuscular glucosamine sulphate 400 mg or placebo twice a week. Lesquesne index, responders/non-responders Glucosamine significantly (p<0.04) better than placebo for symptom score at 4 weeks. Responders 40/79 glucosamine, 23/76 placebo. Minor AE 5/79 glucosamine, 3/76 placebo. R =1 DB=2 WD=1 Total=4
Müller-Fsassbender et al, 1994 (duplicated in Rovalti, 1992, Study 3) 199 in-patients with active gonarthrosis. Randomised, double-blind, parallel-group, 4 weeks 1.5 g/day of glucosamine sulphate or 1.2 g/day ibuprofen. Lesquesne index, responders/non-responders No significant difference at 4 weeks. Responders 48/100 glucosamine, 51/99 ibuprofen. Minor AE 7/100 glucosamine, 37/99 placebo. Drop outs 1/100 glucosamine, 7/99 ibuprofen. R =1 DB=2 WD=1 Total=4

Placebo-controlled trials

Five trials compared glucosamine with placebo. All showed statistical superiority of glucosamine. Four of these had dichotomous outcomes for calculating NNTs, which ranged from 1.7 to 6.3 in individual trials. Overall the NNT was 5.0 (3.5 to 8.9). This means that one of every five patients with arthritis who are treated with glucosamine, one would have short term benefits in reduced pain and tenderness who would not have had if they had been given a placebo.
Outcomes for oral or intramuscular glucosamine in arthritis
Trial Administration Outcome Glucosamine response Placebo response Relative benefit (95% CI) NNT (95% CI)
Pujalte et al, 1980 Oral Improved pain and tenderness at 6-8 weeks 8/10 2/10 4.0 (1.1 to 14) 1.7 (1.1 to 4.0)
Crolle & D'Este, 1980 Oral Symptom free at 3 weeks 4/15 0/15 100) 3.8 (2.0 to 27)
Rovalti, 1992 Oral Responder at 4 weeks 66/126 46/126 1.4 (1.1 to 1.9) 6.3 (3.6 to 26)
Reichelt et al, 1994 Intramuscular Responder at 6 weeks 40/79 23/76 1.7 (1.1 to 2.5) 4.9 (2.8 to 19)
Combined 118/230 71/227 1.6 (1.3 to 2.1) 5.0 (3.5 to 8.9)

Active-controlled trials

Three trials compared glucosamine with NSAID (phenylbutazone or ibuprofen). There was no difference between ibuprofen (1.2 g/day) and oral glucosamine (1.5 g/day).

Adverse effects

Few adverse effects or study withdrawals were reported for glucosamine. They tended to occur less frequently with glucosamine than with NSAID. A large open study of 1208 arthritis patients taking oral glucosamine 1.5 g/day for 13 to 99 days [10] had 28 patients who stopped taking glucosamine because of adverse effects. Those adverse effects reported in more than 1% of patients were epigastric pain/tenderness, heartburn, diarrhoea and nausea.

Comment

Bandolier was surprised to find as many as eight randomised trials. While it is possible to criticise all of the trials to some extent, as a group they are no worse than others used to support commonly-used therapies. There is a consistent thrust of efficacy over placebo, and an inability to distinguish glucosamine from NSAID. But all trials were relatively short-term, and longer-term observations for adverse effects would be welcome.

The bottom line is that there is a body of evidence supporting the efficacy of oral and intramuscular glucosamine in arthritis.

References:

  1. W-D Mund-Hoyn. Konservative behandlung von Wirbelsäulenarthrosen mit glucosaminsulphat un phenylbutazone. Therapiewoche 1980 30: 5922-8.
  2. JM Pujalte, EP Llavore, FR Ylescupidez. Double-blind clinical evaluation of oral glucosamine sulphate in the basic treatment of osteoarthrosis. Current Medical Research and Opinion 1980 7: 110-4.
  3. G Crolle, E D'Este. Glucosamine sulphate for the management of arthrosis: a controlled clinical investigation. Current Medical Research and Opinion 1980 7: 114-9.
  4. E D'Ambrosio, B Casa, R Bompani, G Scali, M Scali. Glucosamine sulphate: a controlled clinical investigation in arthrosis. Pharmacotherapeutica 1981 2: 504-8.
  5. AL Vaz. Double-blind clinical evaluation of the relative efficacy of ibuprofen and glucosamine sulphate in the management of osteoarthrosis of the knee in out-patients. Current Medical Research and Opinion 1982 8: 145-9.
  6. LC Rovati. Clinical Research in osteoarthritis: design and results of short-term and long-term trials with disease-modifying drugs. International Journal of Tissue Reaction 1992 XIV: 243-51.
  7. H Müller-Fassbender, GL Bach, W Haase, LC Rovati, I Setnikar. Glucosamine sulphate compared to ibuprofen in osteoarthritis of the knee. Osteoarthritis and Cartilage 1994 2: 61-9.
  8. A Reichelt, KK Fuorster, M Fischer, LC Rovati, I Setnikar. Efficacy and safety of intramuscular glucosamine sulphate in osteoarthritis of the knee. Arzneim-Forsch/Drug Research 1994 44: 75-80.
  9. AR Jadad, RA Moore, D Carroll, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Controlled Clinical Trials 1996 17: 1-12.
  10. MJ Tapadinhas, IC Rivera, AA Bignamini. Oral glucosamine sulphate in the management of arthrosis: report on a multi-centre open investigation in Portugal. Pharmacotherapeutica 1982 3: 157-68.


previous or next story in this issue