Glucosamine and Osteoarthritis
|
| Researchers (year) | Place | No. sub-jects | Length of study | Recruitment | Type glucosamine & dosage | Outcome measure | Findings (pain, movement, joint space) |
| 1. Hughes and Carr (2002) | UK | 80 | 6 mths | Clinic outpatients, all severities | Oral g sulphate 1500mg | Global assessment of pain + measured knee flexion | No significant difference in pain but significant difference in knee flexion |
| 2. Reginster et al. (2001) | Belgium | 212 | 3 years | Patients? | 1500 mg gl sul, oral | Joint space + WOMAC score | No joint space loss with glucosamine, slight improvement of WOMAC score |
| 3. Reichelt et al
(1994) (from McAlindon 2000) |
Germany | 155 | 6 weeks | Lequesne's criteria, radiol. stage I-III | i.m. gl sul 400 mg 2x week | Red of at least 3 points in Lesquesne index + subjective judgement | Significant decrease in index with glucosamine |
| 4. Houpt, McMillan & Wein (1999) | Canada | 101 | 8 weeks | Examination | glucosamine HCl?dose? | Sign. diff in WOMAC pain score + daily diary + knee examination | No sign. diff in pain score but dd and knee exam favorable? |
By far the majority of glucosamine studies have been favourable (2,3,4, from Table 1). Many of the earlier glucosamine studies had methodological flaws such as small numbers of subjects, drop out rate not clarified, no control group, or too short a study period. Many researchers are sceptical because most of the studies on glucosamine to date have been sponsored by glucosamine manufacturers.
It is difficult to compare the effects of glucosamine between studies because differing outcome measures have been used. In most, reduction of pain is the primary outcome but different scales are used. While reduction of pain is a benefit to the OA sufferer, degree of pain may not necessarily reflect the extent or progression of OA. In other studies the degree of knee flexion or changes in joint space are measured. A narrowing of joint space is taken to indicate progression of OA. There is an argument, however, that this is not the case (Chard & Dieppe 2001).
Most of the earlier studies did not account for dropout rate. People who drop out of a study may have experienced adverse effects or may not feel they are getting any benefit. This can bias results towards a positive finding.
Two independent mega analyses (McAlindon et al. 2000, Towheed et al. 2001) assessed only studies with sound methodological standards. Both analyses conclude that glucosamine is beneficial in the treatment of OA.
From all glucosamine studies from 1966 to 1999 McAlindon et al. (2000) identified only six studies that achieved the methodological standards they were seeking, i.e. double blind, placebo controlled, trials of more than 4 weeks duration.
The studies were further evaluated using a 'quality assessment instrument' (Chalmers et al. 1981, Rochon et al. 1994, from McAlindon et al.). The pooled results showed 'moderate to large treatment effects on symptoms of OA' but the authors believed that the shortcomings of the methodology in most of the trials meant that the actual efficacy would be 'more modest'. Nevertheless they concluded that because glucosamine was safe it had 'considerable utility in OA treatment'. Interestingly, the analysis found a larger effect with longer trials. This is supported by other research and is consistent with the proposed action of glucosamine.
With less stringent selection criteria Towheed et al. (2001) identified 16 appropriate randomised controlled trials of glucosamine. These were quality assessed by different means (Jalad 1995, from Towheed et al.). Outcome measures were pooled, in similar fashion to McAlindon et al. Glucosamine was found to be both effective and safe. In 12 out of 13 studies glucosamine was shown to be more effective than a placebo, and in 2 out of 4 studies glucosamine was more effective than an NSAID.
Both mega-analyses concluded that further research was necessary to confirm long term efficacy and safety of glucosamine.
A third mega-analysis was conducted in from 1998 to 2000 by a special EULAR (European League Against Rheumatism) task force (Pendleton et al. 2000). After searching and analysing all literature on treatments for OA (more than 2,800 publications in total), the group recommended the use of glucosamine for treatment of OA.
Since the release of these mega-analyses three more studies have been published (Reginster et al. 2001, Hughes & Carr 2002, Pavelka et al. 2001). These studies present conflicting results.
Hughes and Carr (2002) found mixed results for glucosamine in their randomised placebo-controlled trial of 80 patients with OA of the knee (6 months duration, 1,500 mg glucosamine daily). The primary outcome measure was patient's 'global assessment' of pain in the knee, but knee flexion was also measured. Although there was no significant difference between placebo and glucosamine groups for pain, knee flexion had significantly improved by 13% for the glucosamine group. Hughes and Carr claimed this was a small difference, and chose to interpret it as 'probably due to measurement error'. On the face of it the interpretation could have been positive.
Reginster et al. (2001), believing that long trials were necessary to allow for maximum effect of glucosamine on the progression of OA, followed subjects for 3 years in a randomised, double blind placebo-controlled trial of 212 patients with moderate OA of the knee. The team measured the space between the tibio-femoral joint and found that patients on the placebo had progressive narrowing of the joint space whereas this progressive degeneration did not occur with those patients taking glucosamine. As assessed by WOMAC scores, OA symptoms were slightly worse for those on the placebo.
Reginster et al. claimed that glucosamine was not only effective for allaying the symptoms of RA, but actually modified the degenerative structural changes associated with OA.
Of all the studies to date this one appears to be the most thorough. Unlike other studies Reginster took the 'dropout rate' into account when analysing his data, giving a 'worst scenario', where a poor outcome was assigned to those patients dropping out. The results were still significantly better for those patients taking glucosamine. It is fair to point out, however, that Reginster's research was sponsored by a major producer of glucosamine, Rotta Laboratories.
Reginster suggested that short-term efficacy of glucosamine on OA symptoms was due to anti inflammatory effects (Hua et al. 2002), while long-term efficacy was due to reported effects of glucosamine on cartilage metabolism synthesis of proteoglycans (Setnikar et al. 1999, Basleer et al. 1998, from Reginster).
The independent study by Pavelka (2001, from electronic replies to Chard & Dieppe) also lent support to the pro-glucosamine lobby.
The US National Institutes of Health glucosamine arthritis intervention trial (GAIT) is currently recruiting 1588 people over 40 years of age with OA of the knee to compare joint space width changes over two years after treatment. Treatment will be with either glucosamine, chondroitin, a combination of the two, an NSAID (celecoxib) or a placebo.
The team is well aware of the methodological mistakes and omissions of previous trials. The authors claim 'The study has been carefully constructed to definitely determine the efficacy' of glucosamine (and chondroitin) (ClinicalTrials.gov website 2002).
As outlined in this paper there seems to be general agreement from animal studies and in vitro work (see above) on the functions of glucosamine in vitro.
Questions have been raised over whether glucosamine is available in joint spaces from an oral dose, but several animal studies and in vitro studies have demonstrated fast uptake of glucosamine into joints after oral dose.
The major problems with general acceptance of glucosamine by the medical fraternity, however, lie with the purported inadequacies of the research. The solid methodology in the recent study by Reginster (2001) would have gone some way to countering these concerns, had Reginster not also been sponsored by Rotta Laboratories.
Many of the problems with accepting glucosamine as a serious medication have been expressed recently by Chard and Dieppe in an editorial, 'Glucosamine for OA: magic, hype, or confusion?' in the British Medical Journal, 2001. They argue that many people are using glucosamine because of media hype about a magic cure, but that 'there is no good evidence that it works', that there are questions whether it can be taken up orally, and that the most appropriate dosage is unknown. Chard and Dieppe also claim that there is no relationship between 'radiographic changes'. i.e. joint space differences and the severity and clinical expression (i.e. pain) of the disease.
On the question of dosage, it is true that no specific studies on dosage have been undertaken, but there has been general consensus on dosage in the countries that use glucosamine as a prescription drug. This dosage has been used in all trials to date, and although it may not be the most efficacious, this does not mean that glucosamine has not been proven to work. Chard and Dieppe contradict themselves about oral intake by later admitting that uptake of glucosamine is 'most likely proved'.
Joint space as a good indicator of progression of OA may also not be 'proved' in every individual case, but the statistical association with patient perception of symptoms is too strong to be dismissed, and the association makes sense in terms of structural progression of the disease. The GAIT trial is using joint space as an outcome measure, which is a strong vote for its relationship with severity and clinical expression of OA.
Chard and Dieppe think that company interference has been a major problem with glucosamine trials, but say that such interference also occurs with NSAID trials.
Aside from the scientific points of difference, it does seem that with some sections of the medical fraternity a major underlying (but unstated) concern has been with the fact that the cause of glucosamine has been taken up by the popular press.
Attitudes to glucosamine seem to have polarised as a result, either towards glucosamine or towards NSAIDs. Even where research has been objective, some researchers appear to interpret their results depending on how they feel about glucosamine.
In the USA, Great Britain and Australia, NSAIDS still seem to have a monopoly on medicos' opinions. Yet some studies show that they appear no more effective than paracetamol, (Courtney & Doherty 2002) and they have been responsible for 16,500 deaths in the US every year through gastric complications, and can create problems with patients who are on multiple medications (Wolfe 1999).
Glucosamine, whether efficacious or not, has not been shown to have any serious side-effects.
Glucosamine may provide a safer and cheaper alternative than the NSAIDs commonly used to treat OA. They may also have a disease modifying effect, which NSAIDs do not have. It may be preferable to reserve final judgement until the results of the GAIT trial become known. But will it really end all argument?
In the meantime however, it is perhaps in the best interests of the OA sufferer to use glucosamine in their fight against pain and disability. Desperate people should not be discouraged from using safe remedies to get relief.
Further trials of the effects of glucosamine on the early stages of arthritis, i.e. initial cartilage damage, may show that early intervention is most important to slowing progression of this disease. Perhaps more work is also needed on the most efficacious dosages.
Bassleer C, Roviati L, Franchimont P 1998, Stimulation of proteoglycan production by glucosamine sulphate in chondrocytes isolated from human osteoarthritic cartilege in vitro, Osteoarthritis Cartilage, 6 (6)
Chard J & Dieppe P 2001, Glucosamine for osteoarthritis: magic, hype, or confusion? Editorial, BMJ, 322
Chard & Dieppe as above, electronic responses
ClinicalTrials.gov 2002, a service of the National Institutes of Health (http://clinicaltrials.gov/ct/gui/show/NCT00032890), Glucosamine/Chondroitin Arthritis Intervention Trial
Courtney P & Doherty M 2002, Key questions concerning paracetamol and NSAIDs for osteoarthritis, Annals of the Rheumatic Diseases 61: 767773
Giannoukakis N 2002 (in press), Immunosuppressive effects of glucosamine, JBC papers in press, The American Society for Biochemistry and Molecular Biology Inc.
Harvard Health Online 2000 (www.health.harvard.edu/medline/women), Supplements, Glucosamine for osteoarthritis
Houpt J, McMillan R, Wein C 1999, Effect of glucosamine hydrocloride in the treatment of pain of osteoarthritis of the knee, J Rheumatology, 26 (11)
Hua J, Sakamoto K, Nagaoka I 2002, Inhibitory actions of glucosamine, a therapeutic agent for osteoarthritis, on the functions of neutrophils, J of Leucocyte Biol, 71
Hughes R & Carr A 2002, A randomised, double blind, placebo-controlled trial of glucosamine suphate as an analgesic in OA of the knee, Rheumatology, 41
McAlindon T, LaValley M, Gulin J, Felson D 2000, Glucosamine and chondroitin for treatment of osteoarthritis, a systematic quality assessment and meta-analysis, JAMA, 283 (11)
Pendleton A, Arden N, Dougados M, Doherty M, Bannwarth B, Bijlsma JW et al. 2000, EULAR recommendations for the management of knee osteoarthritis: report of the Standing Committee for International Clinical Studies Including Therapeutic Trials (ESCISIT), Ann Rheum Dis 59 (12) 936944
Pavelka et al. 2001 cited in Chard & Dieppe electronic responses (see above)
Reginster J, Deroisy R, Rovati L, Lee R, Lejeune E, Bruyere O, Giacovelli G, Henrotin Y, Dacre J, Gosset C 2001, Long term effects of glucosamine sulphate on osteoarthritis progression: a randomised, placebo-controlled study, Lancet, 357 (9252)
Towheed T, Anastassiades T, Shea B, Houpt J 2001 , Glucosamine therapy for treating osteoarthritis, Cochrane Database Syst Rev, (1)
Wolfe M, Lichtenstein D, Singh G 1999, Gastrointestinal toxicity of Non-Steroidal Anti-Inflammatory Drugs, NE J Med, June