Skip Navigation


Rheumatology Advance Access originally published online on May 23, 2007
Rheumatology 2007 46(8):1223-1233; doi:10.1093/rheumatology/kem116
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
46/8/1223    most recent
kem116v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (11)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Canter, P. H.
Right arrow Articles by Ernst, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Canter, P. H.
Right arrow Articles by Ernst, E.
Related Collections
Right arrow Osteoarthritis and Cartilage
Right arrow Rheumatoid Arthritis
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

REVIEWS

The antioxidant vitamins A, C, E and selenium in the treatment of arthritis: a systematic review of randomized clinical trials

P. H. Canter, B. Wider and E. Ernst

Department of Complementary Medicine, Peninsula Medical School, Universities of Exeter and Plymouth, 25 Victoria Park Road, Exeter EX2 4NT, UK.

Correspondence to: P. H. Canter, Department of Complementary Medicine, Peninsula Medical School, Universities of Exeter and Plymouth, 25 Victoria Park Road, Exeter EX2 4NT, UK. E-mail: peter.canter{at}pms.ac.uk


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Conclusion
 References
 
Objective. To systematically review the evidence from randomized clinical trials (RCTs) for the effectiveness of the antioxidant vitamins A, C, E or selenium or their combination in the treatment of arthritis.

Methods. A systematic search of computerized databases from inception to September 2006 for relevant RCTs, application of pre-defined inclusion/exclusion criteria and independent data extraction by two authors. Methodological quality was assessed using the Jadad scale.

Results. The searches identified 20 unique RCTs meeting the inclusion criteria: 11 in inflammatory arthritis and 9 in osteoarthritis (OA). The studies included are generally of poor quality. They fall into three main clusters: selenium for rheumatoid arthritis (n = 5); vitamin E for inflammatory arthritis (n = 5) and vitamin E for OA (n = 7). One RCT suggests superiority of vitamin E over placebo and three RCTs suggest equivalence between vitamin E and diclofenac in the treatment of inflammatory arthritis. In OA, four RCTs compared vitamin E with placebo. Two shorter-term studies were positive and two longer-term studies were negative. Two further RCTs suggest equivalence between vitamin E and diclofenac in the treatment of OA. Findings for selenium, vitamin A and a combination product in inflammatory arthritis and for vitamin A, and a combination product in OA were negative. An isolated positive result for vitamin C in OA is of doubtful clinical significance.

Conclusions. Clinical trials testing the efficacy of vitamin E in the treatment of OA and inflammatory arthritis have been methodologically weak and have produced contradictory findings. There is presently no convincing evidence that selenium, vitamin A, vitamin C or the combination product selenium ACE is effective in the treatment of any type of arthritis.

KEY WORDS: Vitamins, Selenium, Arthritis, Inflammatory arthritis, Rheumatoid arthritis, Osteoarthritis, Systematic review


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Conclusion
 References
 
Antioxidant supplements and diets have long been advocated for the treatment of rheumatoid arthritis (RA), osteoarthritis (OA) and other inflammatory arthritis. However, the value of antioxidants in the prevention and treatment of a wide range of serious diseases including stroke, cancer, diabetes, cataracts, Parkinson's disease, Alzheimer's disease and arthritis has been questioned in the light of more recent negative research findings and studies suggesting that antioxidant properties may be absent or reduced in vivo, may only be important in those with a deficiency or may even be harmful [1]. It therefore seems timely to assess the clinical evidence supporting the use of antioxidants specifically in arthritis. If shown to be safe and effective, antioxidants may be an alternative to treatment with non-steroidal anti-inflammatory drugs (NSAIDs) or other drugs that are associated with adverse effects during long-term use.

In RA, reactive oxygen species and other free radicals are associated with the inflammation process via numerous pathways (reviewed in [2, 3]). These include the role of nitrous oxide in regulating vascular tone, superoxide in fibroblast proliferation and hydrogen peroxide in the transcription of cytokines IL-2 and TNF-{alpha}. During inflammation, oxidation modifies low-density lipoproteins, inactivates {alpha}-1-protease inhibitor, damages DNA and causes lipid peroxidation [2]. Reactive oxygen species also damage cartilage and the extracellular matrix and inhibit collagen and proteoglycan synthesis [3]. Evidence that increased oxidative stress or deficient antioxidant status are important in the pathogenesis of RA comes from several studies (reviewed in [2–4]). Epidemiological studies have shown that low intake of dietary antioxidants is associated with the incidence of RA [2]. Furthermore, animal studies have demonstrated an anti-inflammatory role for some antioxidants including superoxide dismutase (SOD) and vitamin E in experimentally induced arthritis [1, 2].

Although an area much less studied, free radicals may also play a role in the pathogenesis of OA [5–7] and in particular, via the effects upon lipids and cartilage [8, 9]. Observational and epidemiological studies suggest that diets deficient in antioxidants may be associated with an increased incidence of OA or faster disease progression [6, 10].

Numerous clinical studies testing the effectiveness of specific antioxidants or particular antioxidant diets in the treatment of arthritis have been published during the last 30 yrs. Here, we systematically review the evidence from randomized clinical trials (RCTs) for the effectiveness of the antioxidant vitamins A, C and E and for selenium in the treatment of any type of arthritis. Selenium is included because although it is not itself an antioxidant, it is an essential component of the endogenous antioxidant enzyme glutathione peroxidase.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Conclusion
 References
 
The electronic databases Pubmed, AMED, EMBASE, Cochrane Library and Health Technology Assessments (HTA) were searched from their respective inception to September 2006 for RCTs of vitamin A, vitamin C, vitamin E or selenium for arthritis. Search terms were arthritis and antioxidant*; ascorbic acid; vitamin; vitamin E; vitamin A; vitamin C; carot*, tocopherol; ascorb*; selenium. These terms were used in MeSH and text searches in Pubmed and as text searches in the other databases. Articles so identified, were downloaded to Reference Manager and duplicates were deleted. The first author (P.C.) read titles and/or abstracts for all articles and those that might describe RCTs meeting the inclusion criteria were retrieved and read in full. Reference lists of retrieved articles were searched for further potentially relevant RCTs. There were no language restrictions on articles for inclusion.

Criteria for inclusion in this systematic review were:

  1. Human participants diagnosed with any type of arthritis.
  2. Testing vitamin E, vitamin C, selenium, vitamin A, or a combination of these.
  3. Testing these supplements in combination with other interventions if the design allowed the separate evaluation of one of these supplements or the separate evaluation of a combination product of these supplements.
  4. Testing the effectiveness of the intervention in the treatment of arthritis.
  5. Using clinically relevant outcome measures, i.e. validated scales for rating symptoms or patient- or doctor-rated clinical assessment of disease symptoms.
  6. RCT.

Criteria for exclusion were:

  1. Clinical trials testing the effectiveness of the intervention only in the prevention of arthritis.
  2. Studies employing only surrogate endpoints without clinically relevant outcomes relating to disease symptoms.

Assessment of articles for inclusion/exclusion was carried out independently by two authors (P.C. and B.W.) and disagreements were resolved by discussion between all three authors. Quality of reporting was assessed using the Jadad scale [11]. This is a validated measure of quality of reporting in which points are awarded if the study is described as randomized (+1); the means of carrying out randomization is described and appropriate (+1); the study is described as double-blind (+1); the means of double-blinding is described and appropriate (+1); and there is a description of withdrawals giving number and reason in both groups (+1). Points are deducted if the method to generate the sequence of randomization is described and is inappropriate (–1); or if the method of double blinding is described and is inappropriate (–1).

Data were extracted from articles in English by P.C. and from foreign language articles by B.W. according to a pre-defined data extraction sheet. Independent data extraction was carried out for all articles by the third author (E.E.). Where articles failed to report between-group differences in clinical outcomes and there were sufficient data (mean and S.D.), a weighted mean difference was calculated using Review Manager [12]. For equivalence trials, effect sizes with S.D. are reported for each group wherever possible. Where changes from baseline are not reported in the original article and sufficient data are available, these were calculated. When necessary, S.D. of the changes were calculated using a coefficient of 0.4.

Best evidence synthesis was used to combine the results of studies of the same intervention in similar classes of arthritis. In order to reach a positive conclusion we were looking for replicated and consistent positive findings in methodologically strong studies. We did not expect data to be sufficiently homogeneous clinically or statistically for a metanalysis to be performed.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Conclusion
 References
 
The searches identified 156 references. After reading titles and abstracts, 39 of these articles were identified as potentially meeting the inclusion criteria. Full-length articles were obtained and read for these. Of these 39 articles, 22 [13–34] described 20 unique RCTs meeting our inclusion criteria. Two RCTs were each reported in two articles [18, 19, 21, 22] and two RCTs [17, 25] were described in abstracts only. A flow diagram describing the results of our search and selection procedure with reasons for exclusion is given in Fig. 1.


Figure 1
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 1. Flow diagram for inclusion and exclusion of articles.

 
An overview of the indications, interventions and control interventions for the studies included is provided in Table 1. Of the 20 RCTs included, 11 concerned inflammatory arthritis [13–25] including one RCT for ankylosing spondylitis [23] and one RCT for psoriatic arthritis [24]. Of these, five included comparisons of selenium with placebo [13–17] and five included comparisons of vitamin E with either placebo [17, 18] or diclofenac-sodium [20, 21, 23]. Two RCTs tested vitamin A against either placebo [17] or ibuprofen [24] and one RCT tested the combination product selenium ACE against placebo [25]. Nine RCTs for OA were found [26–34], including one RCT in spondylosis [32]. Of these, seven compared vitamin E with either placebo [26–29], diclofenac [30, 31] or vitamin A [32]. The remaining comparisons were between vitamin C and placebo [33], and selenium ACE and placebo [34].


View this table:
[in this window]
[in a new window]

  		TABLE 1. RCTs included in the systematic review

 
Quality of reporting was moderate with a mean Jadad score of 2.9 out of a possible 5 points (n = 20). Mean Jadad score was 2.7 for RCTs in inflammatory arthritis (n = 11) and 3.1 in RCTs for OA (n = 9). Only five of the studies adequately described the randomization procedure [23, 28, 29, 32, 33] and only four adequately described how double blinding was achieved [23, 27, 32, 33]. Scoring for individual items of the Jadad score is shown in Table 2. However, many of the studies had weaknesses in design, statistical analysis or other features of reporting, which are not reflected in the Jadad score. These weaknesses are highlighted either in the brief descriptions of each RCT that follow or in the tables. The main features and outcomes for each RCT are summarized in Table 3 for inflammatory arthritis and in Table 4 for OA.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Jadad scores of included studies

 

View this table:
[in this window]
[in a new window]

  		TABLE 3. Trials included in inflammatory arthritis

 

View this table:
[in this window]
[in a new window]

  		TABLE 4. OA Trials included in osteoarthritis

 
RCTs of selenium for inflammatory arthritis
Tarp [13] found that despite a marked increase in serum selenium concentrations (66 ± 3 µg/l to 223 ± 13 µg/l) during a 6-month trial, there were no statistically significant differences between selenium- and placebo-treated RA patients in any of the nine clinical outcome measures. However, treatment groups do not appear to have been very comparable. Duration and severity of disease at baseline was considerably greater in the selenium group than in the placebo group and there were marked differences between groups in the pattern of co-medications used.

Also in RA, Peretz [14], reports only selected clinical outcomes with significant within-group changes in the selenium group for a pain visual analogue scale (VAS) at 3 months and for the number of joints involved at 3 months and at 6 months. However, no inter-group statistical comparisons were conducted and insufficient data are presented for re-analysis. Again, there was a significant increase in plasma selenium concentration in the active treatment group. The validity of these findings cannot be assessed on the basis of the report available.

Peretz [15] observed no statistically significant between-group differences in any of the five clinical outcome measures at any test point in 90-day trial in RA comparing selenium and placebo. Two of 15 items on a quality of life questionnaire, those relating to arm movements and health perception, significantly favoured the selenium group but no correction appears to have been made in the statistical analysis for multiple comparisons.

Heinle [16] in a 3-month study in RA found a statistically significant increase in erythrocyte selenium concentration in the selenium group (90.2–108 µg/l) but no significant differences between groups in the three clinical outcome measures. The reduction in steroid and NSAID use reported in the selenium group is not accompanied by any statistical analysis.

Jäntti [17], reported in abstract form only, compared vitamin E, vitamin A, selenium, omega-3 fish oil and placebo in RA. Clinical status was monitored throughout the treatment period and at a 4-week follow-up, but no data are given for any clinical outcomes. The authors concluded that selenium, vitamin E or vitamin A have no clear effect in RA. The validity of these findings cannot be assessed on the basis of the report available.

RCTs of vitamin E for inflammatory arthritis
Serum vitamin E levels more than doubled (20.1–43.8 µmol/l) over 12 weeks in the treatment group in a trial in RA, but Edmonds [18, 19] found no significant differences between groups in three clinical outcomes regarded by the authors as indicators of inflammation. There were significant differences favouring vitamin E in patient recorded morning pain, evening pain and pain after a chosen activity, but these analyses used one-tailed tests. A re-analysis confirmed significant differences favouring the vitamin E group for morning and evening pain but not for pain after a chosen activity. It is not clear how daily patient data for the pain scales were gathered and which of these data were used in the analysis at 12 weeks. Differences between groups in patient global assessments failed to reach significance while investigator assessments significantly favoured vitamin E. Even so, according to investigator global assessments, 60% of patients showed no change or got worse. Baseline data indicate that the vitamin E group had more severe RA than controls. The authors concluded that vitamin E has analgesic but not anti-inflammatory effects. The latter conclusion was supported by haematological and biochemical data gathered during the study.

The second RCT comparing vitamin E with placebo is that of Jäntti [17] described earlier. The abstract available provides no assessable data but the authors concluded there was no clear effect of vitamin E on RA.

In a 3-week equivalence trial, Kolarz [20] compared vitamin E treatment with diclofenac in chronic polyarthritis. At 3 weeks, there were no statistically significant differences between groups in any of the seven clinical outcome measures. There were significant within-group changes in both groups in pain, morning stiffness, Ritchie Index and swelling as well as high responder rates (81% for vitamin E and 75% for diclofenac). Pain VAS scores were significantly correlated with vitamin E serum concentrations. No baseline demographic data are reported.

In a similar design, Wittenborg [21, 22] compared vitamin E and diclofenac treatment over 3 weeks in hospitalized chronic polyarthritis patients. Again, there were no significant between-group differences but the authors report significant within-group changes in both groups for all five clinical outcome measures, though levels of significance are not reported. Physicians rated therapy as successful in 54.8% of vitamin E patients and in 48.8% of diclofenac patients. Therapy was rated as successful by patients in 54.8% of the vitamin E group compared with 53.6% of patients in the diclofenac group. Alpha-tocopherol serum concentrations more than doubled (17.6–36.4 mg/l) in the supplemented group during the treatment period. The vitamin E group did appear to have more severe RA at baseline (median morning stiffness 90 min vs 68 min, median Ritchie score 56 vs 49).

Klein [23] compared a lower dose of vitamin E and diclofenac over 12 weeks in ankylosing spondylitis. There were no statistically significant between-group differences in any of the six clinical outcome measures. The authors report similar statistically significant within-group changes in most parameters for both groups at both 6 and 12 weeks. However, raw data are missing for several of the outcomes. The only difference between groups in within-group changes reported was that the lumbar Schobersche score had improved significantly by 12 weeks in the diclofenac group but not in the vitamin E group, but the relevant raw data is not reported. Baseline morning stiffness was considerably longer in the diclofenac group (47 ± 64 min vs 28 ± 18 min).

RCTs of vitamin A in inflammatory arthritis
Hopkins [24] compared the vitamin A derivative etretinate, an aromatic retinoid, with ibuprofen in psoriatic arthritis. There were high dropout rates in both groups and 7 of 20 patients withdrew from the etritanate group because of side effects including dyspepsia, mouth ulcers, cracked lips, flare-up of psoriasis, cracking finger nails, sore eyes, scaling skin, hair loss and impotence. With etretinate, there were no within-group improvements in any of the six clinical parameters at 24 weeks or between groups at any time point. There were large baseline differences in early morning stiffness (196 min in ibuprofen group vs 86 min with etretinate).

RCTs of combination products in inflammatory arthritis
Petersson [25], reported only in abstract form, compared the combination product selenium ACE with placebo in RA using a crossover RCT with 6-month treatment periods. No significant differences between active and placebo treatment were detected despite a significant increase in serum selenium concentration during treatment with selenium ACE. The validity of these findings cannot be assessed on the basis of the report available.

RCTs of vitamin E in OA
Machtey [26] compared vitamin E with placebo for 10 days in a single-blind, crossover RCT in patients with various OA diagnoses, mainly spondylosis. The only clinical outcome was a daily patient record of the condition categorized as ‘no pain’, ‘marked improvement’, ‘some improvement’, or ‘no change’. However, in the analysis, the data have been categorized as ‘significant improvement’, ‘mild improvement’ or ‘no improvement’ but how these categories relate to the four originally recorded by the patients is not explained. Nor is it explained how the daily patient records were converted to a single overall category for each of the 10-day treatment reports. The authors report that 15 (51.7%) of 29 patients experienced ‘significant improvement’ during treatment with vitamin E compared with 1 of 25 (4%) during treatment with placebo, a statistically significant difference (P < 0.05). Combining the categories, ‘significant’ and ‘mild improvement’ indicated that 22 (75.9%) patients benefited from vitamin E treatment and 13 (52.0%) from placebo treatment (P < 0.05). It should be noted that no washout period was incorporated into the crossover design and assessors were not blinded to treatment allocation.

Blankenhorn [27] compared a 6-week treatment with high-dose vitamin E with placebo. Physician-assessed global effectiveness significantly favoured vitamin E and three pain parameters indicated significantly earlier and greater pain reduction with vitamin E compared with placebo. Consumption of analgesics was reduced by 50% in the vitamin E group and 25% in the placebo group. Subgroup analyses of patients with arthritis of the knee and hip also indicated statistically significant superiority of vitamin E over placebo for pain during movement, pressure-induced pain, consumption of analgesics and physician-assessed global effectiveness. No baseline demographic data are reported.

Wluka [28] carried out a 2-yr RCT comparing vitamin E and placebo in OA of the knee. There were no statistically significant differences between treatment groups for loss of knee cartilage, a quality of life measure (SF-36) or in a disease specific, instrument (WOMAC pain, stiffness, function and total scores) completed by patients.

In a similar study lasting 6 months, Brand [29] compared vitamin E and placebo in OA of the knee. Increase in pain over the course of the 6-month study was significantly less in the placebo group but ANOVA indicated no significant difference between groups for change over time for any of the outcome measures. It should be noted that the placebo group had significantly more pain (P = 0.15) and higher body mass index (BMI) (P = 0.03) than the vitamin E group at baseline.

Scherak [30] compared short-term (3 weeks) treatment with vitamin E and diclofenac in active arthritis. No significant differences between groups were observed for any of the six outcomes relating to pain and function and improvements were similar in both groups. Analgesic effects were of earlier onset in the diclofenac group. Patients showing the largest increase in plasma-alpha-tocopherol concentration (>2 S.D.) appeared to experience a greater reduction of pain (8/11 patients vs 4/11 patients). At baseline the diclofenac group had significantly greater pressure-induced pain in (P < 0.018) and the vitamin E group had longer mean duration of arthritis (4.5 yrs vs 3.3 yrs).

Link [31] compared 3 weeks treatment with vitamin E and diclofenac in OA of the knee and/or hip. There were no significant differences between groups for pain on pressure and pain upon movement at any of the weekly testpoints but the relevant means are reported only for the second of these. There were no significant differences between groups in range of motion for the patients with OA of the knee (n = 21) or in hip-joint circumference (n = 8) but this data is presented only graphically. It appears that the comparator treatment, diclofenac had very little effect on these outcomes over the 3-week trial. There was a statistically significant difference favouring vitamin E at 3 weeks in change from baseline on the Keitel Function Test.

Mahmud [32] compared treatment with vitamin E, vitamin A and a combination of vitamin E and vitamin A in patients with radiologically confirmed spondylosis who had not responded to conventional treatment. After a preliminary single-blind trial, apparently non-randomized a subsequent double-blind RCT in 20 patients compared the same three treatments over 3 weeks. After 3 weeks, pain intensity was reported as ‘completely relieved’ in all patients treated with either vitamin E or vitamin E and A and as either the same or worse in those treated with vitamin A. No statistical comparison between groups for pain intensity is reported and the method for assessing pain intensity is not described. Results given for duration of symptoms combine data from the non-randomized and RCTs and are not, therefore, reported here. Absence of a placebo or standard comparator treatment make this study impossible to interpret.

RCTs of vitamin C for OA
Hertz Jensen [33] carried out a 2-week crossover RCT in radiographically verified OA of the knee and/or hip comparing daily treatment with calcium ascorbate and placebo. The placebo was described as identical in appearance and smell but patients were instructed to swallow the tablets whole in an attempt to avoid them detecting a difference in taste. It is therefore possible that they were unblinded. Pain, measured on a VAS, was reduced significantly more in the vitamin C group. Lequesne function and patient preference scores also showed superiority of calcium ascorbate over placebo. No baseline data for demographic or clinical parameters is provided.

RCTs of vitamin A for OA
A single RCT tested the efficacy of vitamin A for OA and is that of Mahmud [32] described earlier, comparing vitamin A, vitamin E and the combination of vitamin A and E. As noted above, this trial cannot be interpreted because it failed to use either a placebo control or standard comparator treatment, and because of inadequate statistical analysis and poor reporting.

RCTs of combination products in OA
A single trial conducted by Hill [34] compared 6 months treatment with selenium ACE and placebo in primary and secondary OA. There were no significant differences between or within groups at 3 or 6 months in VAS scores for pain, stiffness or change in general wellbeing or in blind-scored X-rays taken of the worst affected joint at 0 and 6 months. Patient assessments of efficacy as excellent, good, moderate or poor are not reported.


   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Conclusion
 References
 
A systematic search of the literature identified 20 RCTs of the antioxidant vitamins A, C, E or selenium in the treatment of arthritis and there have been a similar number of such studies in inflammatory arthritis (n = 11) and in OA (n = 9). The quality of these studies is generally poor and in particular, descriptions of randomization and double-blinding are inadequate.

The studies included fall into three main clusters: five studies of selenium for RA [13–17]; five studies of vitamin E for inflammatory arthritis [17–23]; and seven studies of vitamin E for OA [26–32]. In addition to these three clusters, there were negative RCTs of vitamin A in inflammatory arthritis [17, 24], one inadequately reported [17] and the other, in psoriatic arthritis, had important methodological shortcomings and indicated a negative risk-benefit profile [24].

The remainder were all isolated RCTs and comprised studies of the combination product Selenium ACE in RA [25], or in OA [34], vitamin A in OA [32] or vitamin C in OA [33]. Of these, only the study of vitamin C reported any statistically significant differences favouring antioxidant treatment but effect sizes were approximately half that of conventional treatments [33].

Of the five RCTs testing selenium against placebo in RA, one failed to carry out any between-group statistical analyses [14] and another, reported only in abstract form [17], provides no data or statistical analyses that can be properly assessed for validity. Four of these studies showed that selenium concentrations could be significantly elevated by supplementation over 3 months or 6 months, but none reported any significant between-group differences in clinical outcome measures. A reported reduction in the use of co-medications in one trial [16] is not supported by any statistical analysis and significant differences favouring selenium in 2 of 15 subscales of a quality of life questionnaire in another [15] may be the result of uncorrected multiple hypothesis testing.

Five clinical trials tested vitamin E in inflammatory arthritis. Two of these used placebo controls [17, 18] and three compared vitamin E with diclofenac [20, 21, 23]. The placebo-controlled studies, both short-term and of relatively poor quality, reported contradictory findings. The first [17] is very poorly reported but found no clear effect of vitamin E on clinical status. The other [18] found significant positive effects of high-dose vitamin E treatment on pain but not on parameters related to inflammation. All three RCTs comparing vitamin E with diclofenac reported similar findings. There were no significant between-group differences but similar and statistically significant within-group changes in both groups for most outcome measures including pain measures. Authors generally regarded these findings as evidence of equivalence between the two treatments. However, it should be noted that conclusions about the effectiveness of vitamin E based on these equivalence studies depends critically upon the effectiveness attributed to diclofenac. Tests for statistical difference are designed to be conservative and are therefore unsuitable for establishing equivalence. Two of the equivalence studies [20, 21] had short-term treatment periods of only 3 weeks and apparently equivalent positive effects may be the result of placebo effect and/or regression to the mean in both groups. The third [23], which lasted 12 weeks provides raw data for only two of the six outcomes and neither are direct measures of pain. Equivalence can only be judged for finger-floor distance and morning stiffness and for the latter diclofenac was clearly superior. Taken together, these studies provide very little convincing evidence of a positive effect of vitamin E in inflammatory arthritis.

Of the seven clinical trials of vitamin E in OA, four compared the treatment with placebo [26–29], two against diclofenac [30, 31]; and one against vitamin A [32]. Of the placebo-controlled trials, two suggested effectiveness of vitamin E for pain but one of these [26] was methodologically very weak. The second, more robust trial [27] indicated greater effectiveness for both the whole patient sample and for a subgroup with OA of the knee and hip. This latter finding somewhat contradicts the results of the other two placebo-controlled trials [28, 29], which both specifically addressed OA of the knee and both of which had largely negative results. The latter negative studies were more recent, had larger sample sizes and relatively long treatment periods of 6 months [29] and 2 yrs [28] compared with the placebo-controlled trials [26, 27] and the two equivalence trials [30, 31] with more positive outcomes. The two equivalence trials comparing vitamin E treatment with diclofenac [30, 31] suggest similar effectiveness for the two treatments and one reported a statistically significant superiority of vitamin E over diclofenac on the Keitel Function Test [31]. Interpretation of the data from these two trials is subject to the same reservations as those previously expressed about equivalence trials against diclofenac in inflammatory arthritis. These data are consistent with either a short-term effect of vitamin E on subjective measures of pain in OA that is not maintained longer term or a spurious finding resulting from methodological weaknesses in the placebo-controlled trials and placebo and/or regression to the mean in the equivalence trials. The findings of the final trial [32] comparing vitamin E with vitamin A and the combination of the two vitamins is difficult to interpret because of the absence of either a placebo control or a standard comparator treatment and because of inadequate statistical analysis and poor reporting.


   Conclusion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusion
References
 
Clinical trials testing the efficacy of vitamin E in the treatment of arthritis have been methodologically weak and have produced contradictory findings. Suggestions that there may be a positive effect on pain in some shorter-term studies of OA and inflammatory arthritis need to be replicated in RCTs using more methodologically robust protocols. In particular, there is a need for more clinical trials incorporating placebo control arms in order to avoid the difficulties inherent in interpreting the results in equivalence trials. There is presently no convincing evidence that selenium, vitamin A, vitamin C or the combination product selenium ACE are effective in the treatment of any type of arthritis.

Formula

The authors have declared no conflicts of interest.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Conclusion
 References
 

  1. Melton L. The antioxidant myth: a medical fairy tale. New Scientist (2006) August 5.
  2. Mahajan A, Tandon VR. Antioxidants and rheumatoid arthritis. J Indian Rheumatol Assoc (2004) 12:139–42.
  3. Hitchon CA, El-Gabalawy HS. Oxidation in rheumatoid arthritis. Arthritis Res Ther (2004) 6:265–78.[CrossRef][Web of Science][Medline]
  4. Hagfors L, Leanderson P, Skoldstam L, Andersson J, Johansson G. Antioxidant intake, plasma antioxidants and oxidative stress in a randomized, controlled, parallel, Mediterranean dietary intervention study on patients with rheumatoid arthritis. In: Nutrition J. (2003) 2:5. Available at: www.nutritionj.com/content/2/1/5.[CrossRef]
  5. Henrotin Y, Deby-Dupont G, Deby C, Debruin M, Lamy M, Franchimont P. Production of active oxygen species by isolated human chondrocytes. Br J Rheumatol (1993) 32:562–7.[Abstract/Free Full Text]
  6. McAlindon TE, Jaques P, Zhang Y, et al. Do antioxidant micronutrients protect against the development and progression of knee osteoarthritis? Arthr Rheum (1996) 39:648–56.[Web of Science][Medline]
  7. Kaiki G, Tsuji H, Yonezawa T, et al. Osteoarthrosis induced by intra-articular hydrogen peroxide injection and running load. J Orthop Res (1990) 8:731–40.[CrossRef][Web of Science][Medline]
  8. Tiku ML, Shah R, Allison GT. Evidence linking chondrocyte lipid peroxidation to cartilage matrix protein degradation. Possible role in cartilage aging and the pathogenesis of osteoarthritis. J Biol Chem (2000) 275:20069–76.[Abstract/Free Full Text]
  9. Black J, Shadle CA, Parsons JR, Brighton CT. Articular cartilage preservation and storage. Arthr Rheum (1979) 22:1102–8.[Web of Science][Medline]
  10. Kowsari B, Finnie SK, Carter RL, et al. Assessment of the diet of patients with rheumatoid arthritis and osteoarthritis. J Am Diet Assoc (1983) 82:657–9.[Web of Science][Medline]
  11. Jadad AR, Moore RA, Carrol D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Contr Clin Trials (1996) 17:1–12.[CrossRef]
  12. Review Manager (RevMan). In: Version 4.2 for Windows. (2003) Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration.
  13. Tarp U, Overvad K, Thorling EB, Graudal H, Hansen JC. Selenium treatment in rheumatoid arthritis. Scand J Rheumatol (1985) 14:364–8.[Web of Science][Medline]
  14. Peretz A, Neve J, Duchateau J, Famaey JP. Adjuvant treatment of recent onset rheumatoid arthritis by selenium supplementation: preliminary observations. Br J Rheumatol (1992) 31:281–2.[Free Full Text]
  15. Peretz A, Siderova V, Nève J. Selenium supplementation in rheumatoid arthritis investigated in a double blind, placebo-controlled trial. Scand J of Rheumatol (2001) 30:208–12.[CrossRef]
  16. Heinle K, Adam A, Gradl M, Wiseman M, Adam O. [Selenium concentration in erythrocytes of patients with rheumatoid arthritis. Clinical and laboratory chemistry infection markers during administration of selenium]. Medizinische Klinik (1997) 92(Suppl 3):29–31.
  17. Jäntti J, Vapaatalo H, Seppálá E, Ruutsalo HM, Isomäki H. Treatment of rheumatoid arthritis with fish oil, selenium, vitamins A and E, and placebo. Scand J Rheumatol (1991) 20:225.
  18. Edmonds SE, Winyard PG, Guo R, et al. Putative analgesic activity of repeated oral doses of vitamin E in the treatment of rheumatoid arthritis. Results of a prospective placebo controlled double blind trial. Ann Rheum Dis (1997) 56:649–55.[Abstract/Free Full Text]
  19. Springer D. Rheumatoide Arthritis: Hochdosiertes Vitamin E zeigt analgetischen Effekt. Natura Med (1998) 13:30–2.
  20. Kolarz G, Scherak O, Shohoumi MEL, Blankenthorn G. Hochdosiertes Vitamin E bei chronischer Polyarthritis. Aktuelle Rheumatologie (1990) 15:233–7.[Web of Science]
  21. Wittenborg A, Petersen G, Lorkowski G, Brabant T. Effectiveness of vitamin E in comparison with diclofenac sodium in treatment of patients with chronic polyarthritis. Z Rheumatol (1998) 57:215–21.[CrossRef][Web of Science][Medline]
  22. Brabant T, Wittenborg A. Anti-phlogistical and analgetical effectivity of vitamin E (d-alpha-tocopherol acetate, spondyvite) in comparison to Diclofenac-sodium in the treatment of patients with chronic arthritis. Zeitschrift fur Rheumatologie (1993) 52:356.
  23. Klein KG, Blankenhorn G. Vergleich der klinischeen Wirksamkeit von Vitamin E und Diclofenac-Natrium bei Spondylitis ankylosans. Vitamine, Mineralstoffe, Spurenelemente (1987) 2:137–42.
  24. Hopkins R, Bird HA, Jones H, et al. A double-blind controlled trial of etretinate (Tigason) and ibuprofen in psoriatic arthritis. Ann Rheum Dis (1985) 44:189–93.[Abstract/Free Full Text]
  25. Petersson I, Majberger E, Palm S, Larsen A. Treatment of rheumatoid arthritis with selenium and vitamin E. Scand J Rheumatol (1991) 20:218.
  26. Machtey I, Ouaknine L. Tocopherol in osteoarthritis: a controlled pilot study. J Am Geriatr Soc (1978) 26:328–30.[Web of Science][Medline]
  27. Blankenhorn G. Clinical effectiveness of Spondyvit (vitamin E) in activated arthroses. A multicenter placebo-controlled double-blind study. Z Orthop (1986) 124:340–3.[Medline]
  28. Wluka AE, Stuckey S, Brand C, Cicuttini FM. Supplementary vitamin E does not affect the loss of cartilage volume in knee osteoarthritis: a 2 year double blind randomized placebo controlled study. J Rheumatol (2002) 29:2585–91.[Abstract/Free Full Text]
  29. Brand C, Snaddon J, Bailey M, Cicuttini F. Vitamin E is ineffective for symptomatic relief of knee osteoarthritis: a six month double blind, randomised, placebo controlled study. Ann Rheum Dis (2001) 60:946–9.[Abstract/Free Full Text]
  30. Scherak O, Kolarz G, Schodl C, Blankenhorn G. [High dosage vitamin E therapy in patients with activated arthrosis]. Z Rheumatol (1990) 49:369–73.[Web of Science][Medline]
  31. Link P, Dreher R. D-alpha-tocopherolacetat versus Diclofena-Na in der Therapie der aktivierten Arthrose. Der Kassenarzt (1990) 22:48–52.
  32. Mahmud Z, Ali SM. Role of vitamin A and E in spondylosis. Bangladesh Med Res Counc Bull (1992) 18:47–59.[Medline]
  33. Hertz Jensen N. [Reduced pain from osteoarthritis in hip joint or knee joint during treatment with calcium ascorbate. A randomized, placebo-controlled cross-over trial in general practice]. Ugeskr Laeger (2003) 165:2563–6.[Medline]
  34. Hill J, Bird HA. Failure of selenium-ACE to improve osteoarthritis. Br J Rheumatol (1990) 29:211–3.[Abstract/Free Full Text]
Submitted 16 January 2007; revised version accepted 26 March 2007.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
46/8/1223    most recent
kem116v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (11)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Canter, P. H.
Right arrow Articles by Ernst, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Canter, P. H.
Right arrow Articles by Ernst, E.
Related Collections
Right arrow Osteoarthritis and Cartilage
Right arrow Rheumatoid Arthritis
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?