Search tips
Search criteria 


Logo of westjmedLink to Publisher's site
West J Med. 2000 February; 172(2): 91–94.
PMCID: PMC1070762

Randomized, controlled trial of glucosamine for treating osteoarthritis of the knee


Objective To determine the effectiveness of glucosamine in reducing pain from osteoarthritis of the knee. Design Randomized, double-blind parallel trial of glucosamine 500 mg three times daily or a placebo for 2 months. Setting Veterans Affairs Medical Center, Prescott, AZ. Participants Ninety-eight patients aged 34 to 81 being treated for osteoarthritis of the knee. Main outcome measures Pain intensity both at rest and while walking as assessed by a visual analog scale at baseline and after 30 and 60 days of treatment. Results Forty-nine patients were randomly allocated to each group. There was no statistical difference between the two groups in scores on the visual analog scale at 30 days for resting (mean [SD] score placebo group 3.5 [2.7] vs 3.3 [2.4] glucosamine group, P=0.66) or walking (5.1 [2.6] vs 5.3 [2.4], P=0.69); there was also no difference at 60 days for resting (3.4 [2.5] vs 3.2 [2.5], P=0.81) or walking (4.9 [2.2] vs 4.9 [2.8], P=0.90). There was also no statistical difference between groups in the mean change from baseline in scores on the visual analog scale (mean [SD] change for walking at 60 days placebo group -1.5 [2.5] vs glucosamine group -1.4 [3.0], P=0.77). Two participants taking glucosamine and 4 taking placebo withdrew from the study due to adverse side effects (P=0.67). Conclusion Glucosamine was no better than placebo in reducing pain from osteoarthritis of the knee in this group of patients.

Glucosamine sulfate is widely purported in the media and on the Internet to be an effective treatment for osteoarthritis. Its popularity can be traced, in part, to the success of glucosamine in veterinary medicine and to the anecdotal reports of patients with arthritis.1 Some controlled clinical trials have examined the efficacy of glucosamine in treating osteoarthritis.2,3,4,5,6These trials have identified some benefit in using oral or intra-articular glucosamine when compared with nonsteroidal anti-inflammatory drugs or placebo. Despite these data, glucosamine is not recommended as a treatment by the American College of Rheumatology.7 Nevertheless, it may offer a comparatively safe alternative to nonsteroidal anti-inflammatory drugs for treating osteoarthritis, and further study is warranted, especially in veterans, who tend to be older and have more comorbidity than the general population.

Summary points

  • Glucosamine has been reported to be effective in reducing pain in osteoarthritis
  • Controlled clinical trials have found that glucosamine may be effective in treating osteoarthritis when compared with nonsteroidal anti-inflammatory drugs or placebo
  • Glucosamine was no more effective than placebo after 2 months of treatment for osteoarthritis of the knee in this study of older patients


Selection of participants

Patients were eligible to participate in the study if they had a history of osteoarthritis of the knee and had radiographic findings consistent with the disease. Patients were recruited by referral from primary care providers in the outpatient clinics of the medical center. Knee radiographs were graded by a radiologist who was blinded to the purpose of the study. Grading was based on criteria described by Kellgren and Lawrence.8 Grade 0 indicated no arthropathy; grade 4, severe arthropathy. Patients who had been treated earlier with either glucosamine or chondroitin, or both, who were not ambulatory, or who had radiographic findings classed as less severe than grade 1 were excluded from the study. A human studies committee approved this study, and informed consent was obtained from all participants.

Allocation and treatment

Participants were randomly allocated in a double-blind design to treatment with either 500 mg glucosamine (Applehart Laboratories, Bedford, NH) three times daily or a placebo. Randomization was performed using a computer generated list of random numbers. Treatment lasted 2 months. Glucosamine was given with or without food. Patients who were taking other analgesics were instructed to continue them for the duration of the study.

Measurement of pain intensity

Participants were evaluated at the beginning of the study and at 30 and 60 days after starting treatment. Pain intensity was assessed with a visual analog scale. Numerous trials have validated the visual analog scale as a means of evaluating pain intensity, especially for subjective knee complaints.9, 10 Participants were instructed to make a mark on a horizontal line that was 10 cm long to reflect the average rating of the discomfort they had experienced during the previous week. The mark on the line was measured from the left using a metric ruler; a mark at 0 was classed as “no discomfort” and at 10 cm as “severe discomfort.” Participants completed two visual analog assessments at each visit, one representing pain intensity while at rest and the other representing pain while walking. Side effects were assessed at each visit by one of us (JPR) asking the patient if they had experienced any changes in their physical symptoms since the previous visit. Side effects were noted on the same form used for data collection

Data analysis

Demographic data were compared between groups using an independent Student t test for means (age, baseline visual analog scores, duration of arthritis) and a z test for proportional data (for example, the percentage taking nonsteroidal anti-inflammatory drugs) using SigmaStat software (Jandel, San Rafael, CA). Scores on the visual analog scale at rest and while walking were averaged, and means between treatment groups were compared at each visit using a one-way analysis of variance. Changes in scores from baseline were also computed, and the means were compared using a one-way analysis of variance. Statistical significance was set at α = 0.05. We calculated that an absolute difference of 1.1 on the visual analog scale should be detectable if we evaluated 100 participants, assumed a standard deviation of 2, a power of 80%, and an α of 0.05.11 The dropout rate from side effects was compared between treatment groups using a z test.


Altogether, 114 participants were enrolled in the study. They were well matched with respect to demographic data, duration of arthritis, baseline score on the visual analog scale, the concomitant use of analgesics, and radiographic stage (table 1). The mean (SD) age of participants in the placebo group was 64 (11) years and in the glucosamine group was 63 (12). The mean (SD) duration of arthritis in the placebo group was 14 (13) years and in the glucosamine group 12 (10). All participants met the criteria for knee osteoarthritis as described by the American College of Rheumatology.7 Five patients in each group were lost to follow up and were not included in the analysis. Data were analyzed on 98 patients, 49 in each group.

Table 1
Demographic and baseline characteristics of 98 patients with osteoarthritis of the knee who were randomly allocated to treatment with either placebo or 500 mg glucosamine three times daily

No statistical difference was noted between the glucosamine group and the placebo group in mean scores for resting and walking at the 30-day and 60-day assessment (mean [SD] score for resting at 60 days: placebo group 3.4 [2.5] vs 3.2 [2.5] glucosamine group, P=0.81; score for walking: 4.9 [2.2] vs 4.9 [2.8], P=0.90) (table 2). There was also no statistical difference between the groups when the mean change in scores from baseline was calculated and compared (mean [SD] change for walking at 60 days: placebo group -1.5 [2.5] vs -1.4 [3.0] glucosamine group, P=0.77) (table 3). There was a similar distribution of change in scores at the 60-day assessment for both resting and walking (figures (figures11 and and22).

Figure 1
Distribution of changes in scores of pain intensity at rest (60-day visit), as measured by a visual analog scale (VAS) among patients with osteoarthritis of the knee. A score of 0 indicates 11no discomfort” and 10 indicates “severe discomfort.” ...
Figure 2
Distribution of changes in scores of pain intensity while walking (60-day visit), as measured by a visual analog scale (VAS) among patients with osteoarthritis of the knee. A score of 0 indicates “no discomfort” and 10 indicates “severe ...
Table 2
Scores of pain intensity as measured with a visual analog scale*
Table 3
Mean change from baseline in score of pain intensity as measured with a visual analog scale*

Seventeen patients (34%) taking glucosamine experienced side effects compared with 11 (23%) taking placebo. Most side effects in both groups were mild and self-limiting; side effects included loose stools, nausea, heartburn, and headache. Two participants taking glucosamine withdrew from the study because of side effects. One patient had significant diarrhea and the other, dizziness. Side effects subsided after treatment was stopped. Four participants in the placebo group also withdrew because of side effects which included rash, sedation, diarrhea, and constipation. The rate of withdrawal as a result of side effects was not statistically different between the groups (P=0.67) (data not shown).


In this study glucosamine had little effect on the intensity of pain in patients with osteoarthritis of the knee. This is the first study to report negative findings in the use of oral glucosamine to treat osteoarthritis, and thus our findings contradict those reported in other prospective trials of this treatment. One trial of 200 patients showed that glucosamine 1.5 g/day was as effective as ibuprofen 1.2 g/day in bringing about a reduction in pain scores on the Lequesne index.2 A trial of 40 patients with osteoarthritis of the knee found that glucosamine 1.5 g/day was statistically superior to ibuprofen 1.2 g/day in reducing pain scores at 8 weeks.3 Glucosamine was less effective than ibuprofen when patients were assessed at 4 weeks. The largest study included 252 patients and found that glucosamine 1.5 g/day was superior to placebo in reducing pain scores on the Lequesne index.4

It is not clear why there is disagreement between the results of our study and those of previous investigations. In this study, patients tended to be older, heavier, and had had arthritis longer than participants in other trials, suggesting that our patients had more pronounced arthropathy, which can be seen when radiographic findings from this study are compared to the largest placebo-controlled study.4 Although methods of grading knee radiographs differed slightly between studies, more patients in our study had more severe disease than patients in other studies, suggesting more pronounced joint pathology. It is possible that patients with more severe disease may not respond as readily to glucosamine as patients with less severe arthropathy. This would make theoretical sense in that glucosamine is believed to be a precursor of proteoglycans. Proteoglycans are thought to be instrumental in helping cartilage retain water and in promoting formation of an elastic layer, factors which may improve the functional characteristics of cartilage.12 Older patients with a longer history of arthritis may have more damage to their cartilage and the cartilage could thus be less responsive to the effects of glucosamine. Additional studies are required to examine the effects of glucosamine treatment in patients with osteoarthritis of a longer duration.

Another reason for our negative results could be that 2 months of treatment was insufficient to effect clinical improvement in our patients. Other studies involving apparently healthier patients have found clinical improvement within 4-8 weeks of initiating treatment. The largest study of glucosamine identified significant benefit when it was compared with placebo at 4 weeks of treatment but not in the preceding weeks.4 One trial found superior pain relief after 8 weeks of treatment with glucosamine when compared with ibuprofen.2 The efficacy of glucosamine when given for longer than 8 weeks is uncertain.

Any trial that has a negative finding raises the prospect of a type II error. With 100 patients, we estimated that we would have an 80% chance of detecting a difference between the two groups of 1.1 points in scores on the visual analog scale. This would have represented a 31% reduction in pain scores at rest and a 21% reduction while walking when compared with placebo. It is difficult to assess what change in score is clinically significant when measuring the intensity of pain. One large study in which patients with knee osteoarthritis were treated with piroxicam used a 30% difference in scores as an indicator of efficacy13; piroxicam achieved this efficacy measure more often than placebo. Another large study using oxaprozin and nabumetone found a statistically significant 35% to 48% difference from baseline in scores on the visual analog scale.14 We observed practically no difference between the groups when absolute visual analog scores were compared. There was a slight difference in favor of glucosamine when the change in score was compared with the score at rest, however, this did not approach statistical significance. We believe that a much larger cohort of patients would be necessary to achieve significance in a trial of this sort in a similar group of patients.

It is unclear whether the addition of chondroitin sulfate to glucosamine would have influenced the outcome of this study. This combination is available in health food stores and is advertised extensively. Chondroitin has been shown to stimulate the production of proteoglycans and hyaluronic acid and to inhibit the proteolytic enzymes which may damage cartilage.15 Chondroitin has been shown to be more effective than placebo in reducing pain in patients with osteoarthritis of the knee.16,17 Only one controlled study has compared this combination with placebo in patients with osteoarthritis.18 In this double-blind crossover study of 21 servicemen glucosamine with chondroitin was superior to placebo in reducing scores of pain intensity on the visual scale after 8 weeks of treatment. It is not clear if chondroitin contributed to the reduction in pain found in this study since other trials have suggested that 3-6 months of treatment are needed to derive any benefit from this compound. The National Institutes of Health are planning a study of 1000 patients that will compare the efficacy of glucosamine, glucosamine and chondroitin, and placebo in the treatment of osteoarthritis to clarify the role of combination treatment. In our patients with osteoarthritis of the knee glucosamine was no better than placebo in reducing the intensity of pain.


We thank Mary Poncel for help in interpreting knee radiographs.


Funding: None.

Competing interests: None declared


1. Tapadinhas MJ, Rivera IC, Bignamini AA. Oral glucosamine sulphate in the management of arthrosis: report on a mulit-centre open investigation in Portugal. Pharmatherapeutica 1982;3:157-168. [PubMed]
2. Muller-Fabbender H, Bach GL, Haase W, et al. Glucosamine sulfate compared to ibuprofen in osteoarthritis of the knee. Osteoarthritis Cartilage 1994;2:61-69. [PubMed]
3. Vaz AL. Double-blind clinical evaluation of the relative efficacy of ibuprofen and glucosamine sulphate in the management of osteoarthrosis of the knee in out-patients. Curr Med Res Opin 1982;8:145-149. [PubMed]
4. Noack W, Fisher M, Forster KK, et al. Glucosamine sulfate in osteoarthritis of the knee. Osteoarthritis Cartilage 1994;2:51-59. [PubMed]
5. Drovanti A, Bignamini AA, Rovati AL. Therapeutic activity of oral glucosamine sulfate in osteoarthrosis: a placebo-controlled double-blind investigation. Clin Ther 1980;3:260-272. [PubMed]
6. Vajaradul Y. Double-blind evaluation of intra-articular glucosamine in outpatients with gonarthrosis. Clin Ther 1980;3:260-272. [PubMed]
7. Hochberg MC, Altman RD, Brandt KD, et al. Guidelines for the medical management of osteoarthritis. Part II. Osteoarthritis of the knee. Arthritis Rheum 1995;38:1541-1546. [PubMed]
8. Kellgren JH, Lawrence JS. Radiologic assessment of osteoarthritis. Ann Rheum Dis 1957;16:494-502. [PMC free article] [PubMed]
9. Flaherty SA. Pain measurement tools for clinical practice and research. J Am Assoc Nurse Anesth 1996;64:133-139.
10. Flandry F, Hunt JP, Terry GC, et al. Analysis of subjective knee complaints using visual analog scales. Am J Sports Med 1991;19:112-118. [PubMed]
11. Du V Florey C. Sample size for beginners. BMJ 1993;306:1181-1184. [PMC free article] [PubMed]
12. Barclay TS, Tsourounis C, McCart GM. Glucosamine. Ann Pharmacother 1998;32:574-579. [PubMed]
13. Ravaud P, Auleley GR, Ayral X, et al. Piroxicam therapy: a double blind, randomized multicenter study comparing 2 versus 4 week treatment in patients with painful knee osteoarthritis with effusion. J Rheumatol 1998;25:2425-2431. [PubMed]
14. Makarowski W, Weaver A, Rubin B, et al. The efficacy, tolerability, and safety of 1200 mg/d of oxaprozin and 1500 mg/d of nabumetone in the treatment of patients with osteoarthritis of the knee. Clin Ther 1996;18:114-124. [PubMed]
15. Burkhardt D, Ghosh P. Laboratory evaluation of antiarthritic drugs as potential chondroprotective agents. Semin Arthritis Rheum 1987;17(suppl 1):3S-34S. [PubMed]
16. Uebelhart D, Thonar EJ, Delmas PD, et al. Effects of oral chondroitin sulfate on the progression of knee osteoarthritis: a pilot study. Osteoarthritis Cartilage 1998;6(suppl A):39-46. [PubMed]
17. Bucsi L, Poor G. Efficacy and tolerability of oral chondroitin sulfate as a symptomatic slow-acting drug for osteoarthritis (SYSADOA) in the treatment of knee osteoarthritis. Osteoarthritis Cartilage 1998;6(suppl A):31-36. [PubMed]
18. Leffler CT, Philippi AF, Leffler SG, et al. Glucosamine, chondroitin, and manganese ascorbate for degenerative joint disease of the knee or low back: a randomized, double-blind, placebo-controlled pilot study. Mil Med 1999;164:85-91. [PubMed]

Articles from The Western Journal of Medicine are provided here courtesy of BMJ Group