PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of bmjclinevidLink to Publisher's site
 
BMJ Clin Evid. 2007; 2007: 1124.
Published online 2007 August 1.
PMCID: PMC2943775

Rheumatoid arthritis

Karen Walker-Bone, Senior Lecturer in Rheumatology

Abstract

Introduction

Rheumatoid arthritis usually starts as a symmetrical polyarthritis, and its course is marked by flares and remissions. The aims of treatment are to relieve pain and swelling, and to improve function. In addition, disease-modifying antirheumatic drugs (DMARDs) may reduce disease progression.

Methods and outcomes

We conducted a systematic review and aimed to answer the following clinical questions: What are the effects of drug treatments in people with rheumatoid arthritis who have not previously received any disease-modifying antirheumatic drug treatment? How do different drug treatments compare in people with rheumatoid arthritis who have either not responded to or are intolerant of first-line disease-modifying antirheumatic drugs? We searched: Medline, Embase, The Cochrane Library and other important databases up to June 2005 (BMJ Clinical Evidence reviews are updated periodically, please check our website for the most up-to-date version of this review). We included harms alerts from relevant organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA).

Results

We found 62 systematic reviews, RCTs, or observational studies that met our inclusion criteria. We performed a GRADE evaluation of the quality of evidence for interventions.

Conclusions

In this systematic review we present information relating to the effectiveness and safety of the following interventions: adalimumab, anakinra, antimalarial drugs, azathioprine, ciclosporin, corticosteroids, cyclophosphamide, etanercept, infliximab plus methotrexate, leflunomide, methotrexate (alone; or plus sulfasalazine plus hydroxychloroquine), oral gold, parenteral gold, penicillamine, sulfasalazine.

Key Points

Rheumatoid arthritis is a chronic inflammatory disorder that mainly affects the peripheral joints and surrounding tissue.

  • It usually starts as a symmetrical polyarthritis, and its course is marked by flares and remissions.
  • The aims of treatment are to relieve pain and swelling, and to improve function. In addition, disease-modifying antirheumatic drugs (DMARDs) may reduce disease progression.

The DMARD methotrexate is widely used as first-line treatment in people with rheumatoid arthritis because of consensus about its effectiveness in practice.

  • Sulfasalazine and combined treatment with methotrexate and sulfasalazine are as effective as methotrexate in improving pain, joint swelling, and function in people with early rheumatoid arthritis who have not previously received DMARDs.
  • Antimalarials may improve symptoms and function in DMARD-naïve people, and are reasonably well tolerated, but radiological evidence of erosion is more marked with antimalarials than with sulfasalazine.

There is a variety of DMARDs available for second-line treatment of rheumatoid arthritis, and we found no clear evidence that one is superior.

  • Methotrexate, sulfasalazine, penicillamine , and leflunomide cause similar improvements in symptoms and function when given to people as second-line DMARD treatment, although methotrexate causes fewer adverse effects.
  • The combination of methotrexate plus sulfasalazine plus hydroxychloroquine is more effective in reducing measures of disease activity in people receiving second-line treatment than any of the drugs used alone. Adding the cytokine inhibitors infliximab or etanercept to methotrexate is more effective than using methotrexate alone.
  • Although antimalarials and oral gold seem to improve clinical disease activity when given as second-line treatment, they are not as effective as methotrexate or sulfasalazine. Although parenteral gold is more effective than oral gold, it leads to higher levels of toxicity than most of the other commonly used DMARDs.
  • Ciclosporin offers short-term control of rheumatoid arthritis when used as second-line treatment, but is associated with nephrotoxicity.
  • We don′t know whether cyclophosphamide is as effective as other DMARDs for second-line treatment.
  • Cytokine inhibitors may offer an alternative to traditional DMARDs for second line treatment of rheumatoid arthritis, but more research is needed.
  • Etanercept may be as effective as methotrexate in improving symptoms, function, and radiological evidence of progression, but more evidence for its effect is needed
  • Azathioprine is less effective and is less well tolerated than methotrexate.
  • We don't know whether anakinra or adalimumab are as effective as other DMARDs for second-line treatment.
  • Although widely used for the initial short-term relief of clinical disease activity in rheumatoid arthritis, we don't know how corticosteroids compare with other drugs for first or second-line treatment.

About this condition

Definition

Rheumatoid arthritis is a chronic inflammatory disorder. It is characterised by chronic pain and swelling that primarily affects the peripheral joints and related periarticular tissues. It usually starts as an insidious symmetrical polyarthritis, often with non-specific symptoms such as malaise and fatigue.

Incidence/ Prevalence

Studies from the USA have suggested age-adjusted incidence rates of between 0.7 and 0.4 per 1000 person years at risk, but data from European studies suggest a slightly lower incidence rate (0.25/1000 person years). With the exception of some Native American populations where incidence is higher, there is marked consistency in the prevalence of rheumatoid arthritis worldwide. All studies suggest a female incidence rate between two and three times higher than the male rate, and that incidence rates increase progressively with age.

Aetiology/ Risk factors

The cause of rheumatoid arthritis is, as yet, unknown. Genetic factors, hormonal influences, obesity, diet, and cigarette smoking have all been implicated as risk factors. The most widely accepted cause of rheumatoid arthritis is an infection with a micro-organism in a genetically susceptible host.

Prognosis

Rheumatoid arthritis is a chronic condition. In most cases, it follows a course of relapses and remissions (polycyclic pattern). Relapses ("flares") are associated with generalised pain, swelling, and stiffness, which may affect most joints simultaneously. People with rheumatoid arthritis have reduced life expectancy compared with healthy controls, as shown by a longitudinal cohort study undertaken in the UK including 1010 people with rheumatoid arthritis (standardised all-cause mortality among men: 1.45, 95% CI 1.22 to 1.71; standardised all-cause mortality among women: 1.84, 95% CI 1.64 to 2.05). People with rheumatoid arthritis also have excess cardiovascular disease mortality (standardised cardiovascular mortality among men: 1.36, 95% CI 1.04 to 1.75; standardised cardiovascular mortality among women: 1.93, 95% CI 1.65 to 2.26).

Aims of intervention

The aims of rheumatoid arthritis management are not only symptom control during active disease flares, but also suppression of disease activity in order to prevent permanent joint damage, and reducing long-term disability, with minimal adverse effects of treatment.

Outcomes

Patient global assessment; physician global assessment (as assessed by Likert scale etc); pain scores (as measured by visual analogue scale, Likert scale); early morning stiffness; tender joint count, swollen joint count, or both as assessed by Ritchie articular index; disability (as measured by walking distance, dressing, getting in/out of bath); overall function as measured by validated scales such as the American College of Rheumatology core response criteria; Disease Activity Score; European League Against Rheumatism response criteria (based on the Disease Activity Score); radiological erosions (as measured by Sharp or Larsen scores) at 1 year; quality of life; mortality; cardiovascular mortality; adverse effects including neutropenia, anaemia, thrombocytopenia, abnormal liver function, proteinuria, skin rashes, shortness of breath, pulmonary reactions.

Methods

BMJ Clinical Evidence search and appraisal June 2005. The following databases were used to identify studies for this review: Medline 1966 to June 2005, Embase 1980 to June 2005, and The Cochrane Library and Cochrane Central Register of Controlled Clinical Trials Issue 2, 2005. Additional searches were carried out using these websites: NHS Centre for Reviews and Dissemination (CRD) — for Database of Abstracts of Reviews of Effects (DARE) and Health Technology Assessment (HTA), Turning Research into Practice (TRIP), and National Institute for Health and Clinical Excellence (NICE). Abstracts of the studies retrieved from the initial search were assessed by an information specialist. Selected studies were then sent to the contributors for additional assessment, using predetermined criteria to identify relevant studies. Study design criteria for inclusion in this review were: published systematic reviews and RCTs in any language including at least 20 people. Studies are at least single blind. We excluded all studies described as "open", "open label", or not blinded unless blinding was impossible. There was no minimum length of follow-up required to include studies and we did not exclude on the basis of size of follow-up. In addition we use a regular surveillance protocol to capture harms alerts from organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA), which are added to the reviews as required. We have performed a GRADE evaluation of the quality of evidence for interventions included in this review (see table ).

Table
GRADE evaluation of interventions for rheumatoid arthritis

Glossary

American College of Rheumatology (ACR) criteria
for assessing response includes seven measures in its core data set: swollen joint count, tender joint count, patient assessment of global status, an acute phase reactant [erythrocyte sedimentation rate (ESR) or C-reactive protein (CRP)], health professional assessment of global status, physical function, and pain. Improvement criteria for the are based on improvement of at least 20% in both tender and swollen joint counts, and three of the five additional measures (ACR 20), and corresponding ACR 50, 50% improvement, and ACR 70, 70% improvement.
Disease Activity Score (DAS)
is a clinical index of disease activity that combines information from swollen joints, tender joints, erythrocyte sedimentation rate (ESR), and general health or global disease activity measured on a visual analogue scale.
European League Against Rheumatism (EULAR) response criteria
classifies trial participants as “good”, “moderate”, or “non-responders” using individual change from baseline in Disease Activity Score.
Health Assessment Questionnaire (HAQ)
assesses eight functional categories: dressing and grooming, arising, eating, walking, hygiene, reach, grip, and common daily activities. For each of these domains, patients report the amount of difficulty they have had in performing two to three specific activities in the previous week, assessing each activity on a scale from 0 (without any difficulty) to 3 (unable to do). By convention, the HAQ Disability Index (HAQ-DI) is expressed on a scale from 0 to 3 units, representing the mean of the eight domain scores. A HAQ-DI of 0 indicates no functional disability, while a HAQ-DI of 3 indicates severe functional disability.
High-quality evidence
Further research is very unlikely to change our confidence in the estimate of effect.
Larsen score
Assesses radiological damage by scoring joints from 0 (normal) to 5; possible score range 0–250.
Likert scale
When responding to a Likert questionnaire item, respondents specify their level of agreement to a statement. Traditionally a 5 point scale is used: strongly disagree, disagree, neither agree nor disagree, agree, strongly agree.
Low-quality evidence
Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Moderate-quality evidence
Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Ritchie articular index
A graded assessment of 26 joint regions to assess tenderness plus a 44 joint count to assess swelling.
Sharp score
Assesses radiological damage by measuring erosions and joint space narrowing in 44 different joints and reporting an aggregated score ranging from 0 to 448.
Very low-quality evidence
Any estimate of effect is very uncertain.

Notes

Disclaimer

The information contained in this publication is intended for medical professionals. Categories presented in Clinical Evidence indicate a judgement about the strength of the evidence available to our contributors prior to publication and the relevant importance of benefit and harms. We rely on our contributors to confirm the accuracy of the information presented and to adhere to describe accepted practices. Readers should be aware that professionals in the field may have different opinions. Because of this and regular advances in medical research we strongly recommend that readers' independently verify specified treatments and drugs including manufacturers' guidance. Also, the categories do not indicate whether a particular treatment is generally appropriate or whether it is suitable for a particular individual. Ultimately it is the readers' responsibility to make their own professional judgements, so to appropriately advise and treat their patients.To the fullest extent permitted by law, BMJ Publishing Group Limited and its editors are not responsible for any losses, injury or damage caused to any person or property (including under contract, by negligence, products liability or otherwise) whether they be direct or indirect, special, incidental or consequential, resulting from the application of the information in this publication.

References

1. Silman AJ. Rheumatoid arthritis. In: Silman AJ, Hochberg MC, eds. Epidemiology of the rheumatic diseases, 2nd ed. Oxford, Oxford Press, 2004: chapter 2, 31–71.
2. Masi AT. Articular patterns in the early course of rheumatoid arthritis. Am J Med 1983;75(suppl6A):16–26. [PubMed]
3. Goodson N, Marks J, Lunt M, et al. Cardiovascular admissions and mortality in an inception cohort of patients with rheumatoid arthritis with onset in the 1980s and 1990s. Ann Rheum Dis 2005;64:1595–1601. [PMC free article] [PubMed]
4. Sharp JT, Lidsky MD, Collins LC, et al. Methods of scoring the progression of radiological changes in rheumatoid arthritis. Correlation of radiologic, clinical and laboratory abnormalities. Arthritis Rheum 1971;14:706–720. [PubMed]
5. Larsen A, Dale K, Eek M. Radiographic evaluation of rheumatoid arthritis and related conditions by standard reference films. Acta Radiol Diagn (Stockh) 1977;18:481–491. [PubMed]
6. Dougados M, Combe B, Cantagrel A, et al. Combination therapy in early rheumatoid arthritis: a randomised, controlled, double blind 52 week clinical trial of sulphasalazine and methotrexate compared with the single components. Ann Rheum Dis 1999;58:220–225. [PMC free article] [PubMed]
7. Haagsma CJ, van Riel PL, de Jong AJ, et al. Combination of sulphasalazine and methotrexate versus the single components in early rheumatoid arthritis: a randomized, controlled, double-blind, 52 week clinical trial. Br J Rheumatol 1997;36:1082–1088. [PubMed]
8. The Australian Multicentre Clinical Trial Group. Sulfasalazine in early rheumatoid arthritis. J Rheumatol 1992;19:1672–1677. [PubMed]
9. Hannonen P, Mottonen T, Hakola M, et al. Sulfasalazine in early rheumatoid arthritis. A 48-week double-blind, prospective, placebo-controlled study. Arthritis Rheum 1993;36:1501–1509. [PubMed]
10. Williams HJ, Ward JR, Dahl SL, et al. A controlled trial comparing sulfasalazine, gold sodium thiomalate, and placebo in rheumatoid arthritis. Arthritis Rheum 1988;31:702–713. [PubMed]
11. Nuver-Zwart IH, van Riel PL, van de Putte LB, et al. A double blind comparative study of sulfasalazine and hydroxychloroquine in rheumatoid arthritis: evidence of an earlier effect of sulfasalazine. Ann Rheum Dis 1989;48:389–395. [PMC free article] [PubMed]
12. Van der Heijde DM, van Riel PL, Nuver-Zwart IH, et al. Effects of hydroxychloroquine and sulphasalazine on progression of joint damage in rheumatoid arthritis. Lancet 1989;1:1036–1038. [PubMed]
13. Clark P, Casas E, Tugwell P, et al. Hydroxychloroquine compared with placebo in rheumatoid arthritis. A randomized controlled trial. Ann Intern Med 1993;119:1067–1071. [PubMed]
14. Anon. A randomized trial of hydroxychloroquine in early rheumatoid arthritis: the HERA Study. Am J Med 1995;98:156–168. [PubMed]
15. Pavelka K Jr, Sen KP, Peliskova Z, et al. Hydroxychloroquine sulphate in the treatment of rheumatoid arthritis: a double blind comparison of two dose regimens. Ann Rheum Dis 1989;48:542–546. [PMC free article] [PubMed]
16. Tett SE, Cutler DJ, Beck C, et al. Concentration-effect relationship of hydroxychloroquine in patients with rheumatoid arthritis – a prospective, dose ranging study. J Rheumatol 2000;27:1656–1660. [PubMed]
17. Fielder A, Graham E, Jones S, et al. Royal College of Ophthalmologists guidelines: ocular toxicity and hydroxychloroquine. Eye 1998;12:907–909. [PubMed]
18. Criswell LA, Robinson VA, Saag KKS, et al. Moderate-term, low-dose corticosteroids for rheumatoid arthritis. In: The Cochrane Library, Issue 2, 2005. Chichester, UK: John Wiley & Sons, Ltd. Search date 1998, primary sources Medline; hand searches of Arthritis and Rheumatism, Scandinavian Journal of Rheumatology, Arthritis and Rheumatism abstracts, and references of all identified studies; and contact with authors of unpublished manuscripts.
19. Gotzsche PC, Johansen HK. Short-term low-dose corticosteroids vs placebo and nonsteroidal antiinflammatory drugs in rheumatoid arthritis. In: The Cochrane Library, Issue 2, 2005. Chichester, UK: John Wiley & Sons, Ltd. Search date 2004, primary sources Pubmed, Cochrane Central Register of Controlled Trials (CENTRAL), and hand searches of reference lists.
20. Blumenauer B, Judd M, Cranney A, et al. Infliximab for the treatment of rheumatoid arthritis. In: The Cochrane Library, Issue 2, 2005. Chichester, UK: John Wiley & Sons, Ltd. Search date 2002; primary sources Biological Abstracts, Cinahl, Current Contents, Dissertation Abstracts, EBM Reviews, Healthstar, and Medline. [PubMed]
21. Maini RN, Breedveld FC, Kalden JR, et al. Therapeutic efficacy of multiple intravenous infusions of anti-tumour necrosis factor alpha monoclonal antibody combined with low-dose weekly methotrexate in rheumatoid arthritis. Arthritis Rheum 1998;41:1552–1563. [PubMed]
22. Maini R, St Clair EW, Breedveld F, et al. Infliximab (chimeric anti-tumour necrosis factor monoclonal antibody) versus placebo in rheumatoid arthritis patients receiving concomitant methotrexate: a randomised phase III trial. ATTRACT Study Group. Lancet 1999;354:1932–1939. [PubMed]
23. Osiri M, Robinson VA, Shea BJ, et al. Leflunomide for treating rheumatoid arthritis. In: The Cochrane Library, Issue 2, 2005. Chichester, UK: John Wiley & Sons, Ltd. Search date 2002, primary sources Medline, Embase, Healthstar, Current Contents, Cochrane Controlled Trials Register, Cochrane Musculoskeletal Specialized Register, hand searches of reference lists and abstracts from the American College of Rheumatology annual scientific meetings, and contact with experts. [PubMed]
24. Bao C, Chen S, Gu Y, et al. Leflunomide, a new disease-modifying drug for treating active rheumatoid arthritis in methotrexate-controlled phase II clinical trial. Chin Med J (Engl) 2003;116:1228–1234. [PubMed]
25. Felson DT, Anderson JJ, Meenan RF. The comparative efficacy and toxicity of second line drugs in rheumatoid arthritis. Results of two metaanalyses. Arthritis Rheum 1990;33:1449–1461. Search date 1989; primary sources Medline, hand searches of reference lists of retrieved articles and of Arthritis Rheum and Scand J Rheumatol. [PubMed]
26. Felson DT, Anderson JJ, Meenan RF. Use of short-term efficacy/toxicity tradeoffs to select second line drugs in rheumatoid arthritis. A metaanalysis of published clinical trials. Arthritis Rheum 1992;35:1117–1125. Search date 1990; primary sources Medline, hand searches of reference lists of retrieved articles and of Arthritis Rheum and Scand J Rheumatol. [PubMed]
27. Weinblatt ME, Polisson R, Blotner SD, et al. The effects of drug therapy on radiographic progression of rheumatoid arthritis. Results of a 36-week randomized trial comparing methotrexate and auranofin. Arthritis Rheum 1993;36:613–619. [Erratum in: Arthritis Rheum 1993;36:1028] [PubMed]
28. Rau R, Herborn G, Karger T, et al. A double blind randomized parallel trial of intramuscular methotrexate and gold sodium thiomalate in early erosive rheumatoid arthritis. J Rheumatol 1991;18:328–333. [PubMed]
29. Rau R, Herborn G, Karger T, et al. A double-blind comparison of parenteral methotrexate and parenteral gold in the treatment of early erosive rheumatoid arthritis: an interim report on 102 patients after 12 months. Semin Arthritis Rheum 1991;21:13–20. [PubMed]
30. Alarcon GS, Lopez-Mendez A, Walter J, et al. Radiographic evidence of disease progression in methotrexate treated and nonmethotrexate disease modifying antirheumatic drug treated rheumatoid arthritis patients: a meta-analysis. J Rheumatol 1992;19:1868–1873. Search date 1991; primary sources ′the English language literature on the treatment of rheumatoid arthritis with methotrexate and hand searches of abstracts from national meeting of the American College of Rheumatology 1988-1991. [PubMed]
31. Jeurissen ME, Boerbooms AM, van de Putte LB, et al. Influence of methotrexate and azathioprine on radiologic progression in rheumatoid arthritis. A randomized, double-blind study. Ann Intern Med 1991;114:999–1004. [PubMed]
32. Jeurissen ME, Boerbooms AM, van de Putte LB, et al. Methotrexate versus azathioprine in the treatment of rheumatoid arthritis. A forty-eight-week randomized, double-blind trial. Arthritis Rheum 1991;34:961–972. [PubMed]
33. Willkens RF, Urowitz MB, Stablein DM, et al. Comparison of azathioprine, methotrexate, and the combination of both in the treatment of rheumatoid arthritis. A controlled clinical trial. Arthritis Rheum 1992;35:849–856. [PubMed]
34. Willkens RF, Sharp JT, Stablein D, et al. Comparison of azathioprine, methotrexate, and the combination of the two in the treatment of rheumatoid arthritis. A forty-eight-week controlled clinical trial with radiologic outcome assessment. Arthritis Rheum 1995;38:1799–1806. [PubMed]
35. Willkens RF, Stablein D. Combination treatment of rheumatoid arthritis using azathioprine and methotrexate: a 48 week controlled clinical trial. J Rheumatol Suppl 1996;44:64–68. [PubMed]
36. Whiting-O'Keefe QE, Fye KH, Sack KD. Methotrexate and histologic hepatic abnormalities: a meta-analysis. Am J Med 1991;90:711–716. Search date not reported; primary sources ′computerized literature search for all English language studies reporting liver biopsy data on patients with psoriasis, psoriatic arthritis, or rheumatoid arthritis receiving low-dose methotrexate.′ [PubMed]
37. Barrera P, Laan RFJM, Van Riel PLCM, et al. Methotrexate-related pulmonary complications in rheumatoid arthritis. Ann Rheum Dis 1994;53:434–439. [PMC free article] [PubMed]
38. Patatanian E, Thompson DF. A review of methotrexate-induced accelerated nodulosis. Pharmacotherapy 2002;22:1157–1162. Search date 2001; primary sources Medline, Toxline, and Embase and hand searches of reference lists of key articles. [PubMed]
39. Ince A, Yazici Y, Hamuryudan V, et al. The frequency and clinical characteristics of methotrexate (MTX) oral toxicity in rheumatoid arthritis (RA): a masked and controlled study. Clin Rheumatol 1996;15:491–494. [PubMed]
40. O'Dell JR, Haire CE, Erikson N, et al. Treatment of rheumatoid arthritis with methotrexate alone, sulfasalazine and hydroxychloroquine, or a combination of all three medications. New Engl J Med 1996;334:1287–1291. [PubMed]
41. O'Dell JR, Leff R, Paulsen G, et al. Treatment of rheumatoid arthritis with methotrexate and hydroxychloroquine, methotrexate and sulfasalazine, or a combination of the three medications: results of a two-year, randomized, double-blind, placebo-controlled trial. Arthritis Rheum 2002;46:1164–1170. [PubMed]
42. Paulus HE, Williams HJ, Ward JR, et al. Azathioprine versus D-penicillamine in rheumatoid arthritis patients who have been treated unsuccessfully with gold. Arthritis Rheum 1984;27:721–727. [PubMed]
43. van Rijthoven AW, Dijkmans BA, Goie The HS, et al. Comparison of cyclosporine and D-penicillamine for rheumatoid arthritis: a randomized, double blind, multicenter study. J Rheumatol 1991;18:815–820. [PubMed]
44. Weinblatt ME, Reda D, Henderson W, et al. Sulfasalazine treatment for rheumatoid arthritis: a metaanalysis of 15 randomized trials. J Rheumatol 1999;26:2123–2130. Search date 1998; primary sources Medline, Excerpta Medica, and Derwent Drug File. [PubMed]
45. Jones G, Halbert J, Crotty M, et al. The effect of treatment on radiological progression in rheumatoid arthritis: a systematic review of randomized placebo-controlled trials. Rheumatology 2003;42:6–13. Search date 2000; primary sources Cochrane Controlled Trials Register; Medline; Embase; hand searches of bibliographies of reviews, papers, and conference proceedings of the American College of Rheumatology; and contact with pharmaceutical companies and authors. [PubMed]
46. Bird HA, Le Gallez P, Dixon JS, et al. A single-blind comparative study of auranofin and hydroxychloroquine in patients with rheumatoid arthritis. Clin Rheumatol 1984;3:57–66. [PubMed]
47. Ahern MJ, Harrison W, Hollingsworth P, et al. A randomised double-blind trial of cyclosporin and azathioprine in refractory rheumatoid arthritis. Aust N Z J Med 1991;21:844–849. [PubMed]
48. Kruger K, Schattenkirchner M. Comparison of cyclosporin A and azathioprine in the treatment of rheumatoid arthritis – results of a double-blind multicentre study. Clin Rheumatol 1994;13:248–255. [PubMed]
49. Boers M, Dijkmans BA, van Rijthoven AW, et al. Reversible nephrotoxicity of cyclosporine in rheumatoid arthritis. J Rheumatol 1990;17:38–42. [PubMed]
50. Yocum DE, Klippel JH, Wilder RL, et al. Cyclosporin A in severe, treatment-refractory rheumatoid arthritis. A randomized study. Ann Intern Med 1988;109:863–869. [PubMed]
51. Berg KJ, Forre O, Djoseland O, et al. Renal side effects of high and low cyclosporin A doses in patients with rheumatoid arthritis. Clin Nephrol 1989;31:232–238. [PubMed]
52. Boers M, Dijkmans BA, van Rijthoven AW, et al. Reversible nephrotoxicity of ciclosporin in rheumatoid arthritis. J Rheumatol 1990;17:38–42. [PubMed]
53. Blumenauer B, Judd M, Cranney A, et al. Etanercept for the treatment of rheumatoid arthritis. In: The Cochrane Library, Issue 2, 2005. Chichester, UK: John Wiley & Sons, Ltd. Search date 2003; primary sources Biological Abstracts, Current Contents, Dissertation Abstracts, EBM Reviews, and all Cochrane electronic databases, hand searches of proceedings of major rheumatology conferences and of reference lists from standard rheumatology textbooks, comprehensive reviews, and identified clinical trials and contact with content experts and pharmaceutical companies that manufacture etanercept. [PubMed]
54. Klareskog L, van der Heijde D, de Jager JP, et al. Therapeutic effect of the combination of etanercept and methotrexate compared with each treatment alone in patients with rheumatoid arthritis: double-blind randomised controlled trial. Lancet 2004;363:675–681. [PubMed]
55. Currey HL, Harris J, Mason RM, et al. Comparison of azathioprine, cyclophosphamide, and gold in treatment of rheumatoid arthritis. BMJ 1974;3:763–766. [PMC free article] [PubMed]
56. den Broeder A, van de Putte L, Rau R, et al. A single dose, placebo controlled study of the fully human anti-tumor necrosis factor-alpha antibody adalimumab (D2E7) in patients with rheumatoid arthritis. J Rheumatol 2002;29:2288–2298. [PubMed]
57. van de Putte LB, Rau R, Breedveld FC, et al. Efficacy and safety of the fully human anti-tumour necrosis factor alpha monoclonal antibody adalimumab (D2E7) in DMARD refractory patients with rheumatoid arthritis: a 12 week, phase II study. Ann Rheum Dis 2003;62:1168–1177. [PMC free article] [PubMed]
58. van de Putte LB, Atkins C, Malaise M, et al. Efficacy and safety of adalimumab as monotherapy in patients with rheumatoid arthritis for whom previous disease modifying antirheumatic drug treatment has failed. Ann Rheum Dis 2004;63:508–516. [PMC free article] [PubMed]
59. Furst DE, Schiff MH, Fleischmann RM, et al. Adalimumab, a fully human anti tumor necrosis factor-alpha monoclonal antibody, and concomitant standard antirheumatic therapy for the treatment of rheumatoid arthritis: results of STAR (Safety Trial of Adalimumab in Rheumatoid Arthritis). J Rheumatol 2003;30:2563–2571. [PubMed]
60. Keystone EC, Kavanaugh AF, Sharp JT, et al. Radiographic, clinical, and functional outcomes of treatment with adalimumab (a human anti-tumor necrosis factor monoclonal antibody) in patients with active rheumatoid arthritis receiving concomitant methotrexate therapy: a randomized, placebo-controlled, 52-week trial. Arthritis Rheum 2004;50:1400–1411. [PubMed]
61. Bresnihan B, Alvaro-Gracia JM, Cobby M, et al. Treatment of rheumatoid arthritis with recombinant human interleukin-1 receptor antagonist. Arthritis Rheum 1998;41:2196–2204. [PubMed]
62. Bresnihan B, Newmark R, Robbins S, et al. Effects of anakinra monotherapy on joint damage in patients with rheumatoid arthritis. Extension of a 24-week randomized, placebo-controlled trial. J Rheumatol 2004;31:1103–1111. [PubMed]
63. Nuki G, Bresnihan B, Bear MB, et al. Long-term safety and maintenance of clinical improvement following treatment with anakinra (recombinant human interleukin-1 receptor antagonist) in patients with rheumatoid arthritis: extension phase of a randomized, double-blind, placebo-controlled trial. Arthritis Rheum 2002;46:2838–2846. [PubMed]
64. Cohen SB, Woolley JM, Chan W, et al. Interleukin 1 receptor antagonist anakinra improves functional status in patients with rheumatoid arthritis. J Rheumatol 2003;30:225–231. [PubMed]
65. Cohen SB, Moreland LW, Cush JJ, et al. A multicentre, double blind, randomised, placebo controlled trial of anakinra (Kineret), a recombinant interleukin 1 receptor antagonist, in patients with rheumatoid arthritis treated with background methotrexate. Ann Rheum Dis 2004;63:1062–1068. [PMC free article] [PubMed]
66. Fleischmann RM, Schechtman J, Bennett R, et al. Anakinra, a recombinant human interleukin-1 receptor antagonist (r-metHuIL-1ra), in patients with rheumatoid arthritis: a large, international, multicenter, placebo-controlled trial. Arthritis Rheum 2003;48:927–934. [PubMed]
67. Tesser J, Fleischmann R, Dore R, et al. Concomitant medication use in a large, international, multicenter, placebo controlled trial of anakinra, a recombinant interleukin 1 receptor antagonist, in patients with rheumatoid arthritis. J Rheumatol 2004;31:649–654. [PubMed]
68. Schiff MH, DiVittorio G, Tesser J, et al. The safety of anakinra in high-risk patients with active rheumatoid arthritis: six-month observations of patients with comorbid conditions. Arthritis Rheum 2004;50:1752–1760. [PubMed]
69. Saag KG, Criswell LA, Sems KM, et al. Low-dose corticosteroids in rheumatoid arthritis. A meta-analysis of their moderate-term effectiveness. Arthritis Rheum 1996;39:1818–1825. Search date 1994; primary sources Medline, hand searches of reference lists of retrieved articles and of Arthritis Rheum and Scand J Rheumatol. [PubMed]
70. Vischer TL, Sinniger M, Ott H, et al. A randomized, double-blind trial comparing a pulse of 1000 with 250 mg methylprednisolone in rheumatoid arthritis. Clin Rheumatol 1986;5:325–326. [PubMed]
71. Iglehart IW III, Sutton JD, Bender JC, et al. Intravenous pulsed steroids in rheumatoid arthritis: a comparative dose study. J Rheumatol 1990;17:159–162. [PubMed]
72. Shipley ME, Bacon PA, Berry H, et al. Pulsed methylprednisolone in active early rheumatoid disease: a dose-ranging study. Br J Rheumatol 1988;27:211–214. [PubMed]
2007; 2007: 1124.
Published online 2007 August 1.

Methotrexate (first-line treatment)

Summary

DISEASE ACTIVITY Compared with sulfasalazine: Methotrexate may be as effective as sulfasalazine at reducing disease activity over 12 months as first-line treatment in people with rheumatoid arthritis ( low-quality evidence ). Compared with sulfasalazine plus methotrexate: Methotrexate alone may be as effective as sulfasalazine plus methotrexate at reducing disease activity over 12 months as first-line treatment in people with rheumatoid arthritis ( low-quality evidence ). ADVERSE EFFECTS The risk of adverse effects seems to be similar with methotrexate and with sulfasalazine, including headache, vertigo, gastrointestinal upsets, abnormal liver function tests, stomatitis, and leukopenia. Adverse effects may increase when the drugs are combined. NOTE We found no direct information about whether methotrexate is better than no active treatment in people who had not previously received disease-modifying antirheumatic drugs. We found no clinically important results about the effects of methotrexate compared with antimalarial drugs in people with early rheumatoid arthritis.

Benefits

Methotrexate versus placebo:

We found no systematic review or RCTs comparing methotrexate versus placebo in people naïve to disease-modifying antirheumatic drugs (DMARDs).

Methotrexate versus sulfasalazine or versus methotrexate plus sulfasalazine:

We found two RCTs including 310 adults with early (< 12 months since diagnosis) active (Disease Activity Score > 3.0) rheumatoid arthritis who had not previously taken DMARDs. Participants in both RCTs were included only if they had not used corticosteroids before entry to the trial. Corticosteroids were used by 12 people during the second RCT (3 taking methotrexate, 4 sulfasalazine, and 4 combination treatment). In both RCTs, participants received methotrexate alone, sulfasalazine alone, or a combination of methotrexate plus sulfasalazine. A step-up approach to dosing was used (up to a maximum of methotrexate 15 mg/week and sulfasalazine 3 g/day). Analysis was by intention to treat in both RCTs. Both RCTs found similar improvements in disease activity over 12 months with methotrexate, sulfasalazine, and methotrexate plus sulfasalazine as assessed by patient rated Global Assessment score, improvement in Disease Activity Score, proportion achieving American College of Rheumatology response criteria, proportion achieving European League Against Rheumatism response criteria, Health Assessment Questionnaire, swollen joint count, tender joint count, pain scores, and grip strength (see table 1 ). In the first RCT, global assessment was evaluated using a Likert scale where 0 = asymptomatic and 5 = very severe, joint tenderness was assessed using the Ritchie articular index, and function was assessed using the Health Assessment Questionnaire. In the second RCT, global assessment was evaluated as the proportion of people judging disease to be “moderately/much improved” and this RCT found that methotrexate alone or in combination with sulfasazine significantly increased global satisfaction rates compared with sulfasalazine alone. The first RCT also evaluated radiological changes according to the modified Sharp score. It found that radiological progression over 12 months was modest in both groups compared with baseline, with no significant difference in erosion scores among groups (see table 1 ). There were, however, lower scores in people taking combination treatment, suggesting less radiological progression.

Table 1
Methotrexate, sulfasalazine, or the combination in disease-modifying antirheumatic drug (DMARD) naïve people with early rheumatoid arthritis.

Methotrexate versus antimalarial drugs:

We found no systematic review or RCTs in people naive to DMARDs.

Harms

Methotrexate versus placebo:

We found no RCTs.

Methotrexate versus sulfasalazine or versus methotrexate plus sulfasalazine:

In both RCTs, all interventions were associated with headache, vertigo, gastrointestinal upsets, abnormal liver function tests, stomatitis, and leukopenia. In the first RCT, a significantly higher proportion of people taking combined treatment had one or more adverse effects (91% with methotrexate plus sulfasalazine v 75% with either methotrexate or sulfasalazine alone; P = 0.025 for combined v either methotrexate or sulfasalazine alone; absolute numbers not reported). There was no significant difference among groups in the proportion of people who had one or more adverse effects in the second RCT (27/35 [77%] with methotrexate alone v 30/34 [88%] with sulfasalazine alone v 32/36 [89%] with methotrexate plus sulfasalazine; reported as not significant, P value not reported).

Methotrexate versus antimalarial drugs:

We found no RCTs.

Comment

Clinical guide:

Although methotrexate is widely used as a first-line treatment in people with active rheumatoid arthritis, few controlled trials have been performed among DMARD-naïve patients, and placebo-controlled studies of methotrexate in early rheumatoid arthritis are now widely considered unethical, because of its observed effectiveness in practice. Methotrexate has also been widely used as the “gold standard” in RCTs assessing newer DMARDs and biological agents. As such, it has probably become the most widely used first-line drug in people naïve to DMARDs. In the absence of poor prognostic signs (high titre of rheumatoid factor at baseline, erosions on baseline x rays, involvement of large numbers of joints at baseline, high disability scores, high levels of inflammatory markers), there are no data to indicate relative benefit of methotrexate over sulfasalazine, and many clinicians will be guided by patient choice, having explained the risks and benefits of each drug. Methotrexate is contraindicated in pregnancy and among men planning to father a child, and patients are advised to use contraceptives for 3 months after stopping treatment. The British Society of Rheumatology recommends that all patients commencing methotrexate are screened with a baseline chest x ray. Monitoring for toxicity is usually carried out every 2 weeks for the first 6 weeks and monthly thereafter for 6 months, checking full blood count and liver function tests. Renal function is assessed every 3 months. Abstinence or restriction of alcohol is recommended during methotrexate treatment.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Sulfasalazine (first-line treatment)

Summary

DISEASE SEVERITY Compared with placebo: Sulfasalazine as first-line therapy may not improve overall disease severity compared with placebo in people with rheumatoid arthritis who had not previously received disease-modifying antirheumatic drugs ( very low-quality evidence ). Compared with methotrexate: Sulfasalazine as first-line treatment may be as effective as methotrexate at reducing disease activity over 12 months in people with rheumatoid arthritis ( low-quality evidence ). Compared with sulfasalazine plus methotrexate: Sulfasalazine alone as first-line treatment in people with rheumatoid arthritis may be as effective as sulfasalazine plus methotrexate at reducing disease activity over 12 months (low-quality evidence). Compared with hydroxychloroquine: Sulfasalazine is as effective as chydroxychloroquine at improving symptoms and function in people with rheumatoid arthritis ( moderate-quality evidence ). JOINT PAIN AND TENDERNESS Compared with placebo: Sulfasalazine as first-line therapy may reduce joint pain and tenderness compared with placebo after 6–12 months ( low-quality evidence ). ADVERSE EFFECTS The risk of adverse effects seems to be similar with sulfasalazine and methotrexate , including headache, vertigo, gastrointestinal upsets, abnormal liver function tests, stomatitis, and leukopenia. Adverse effects may increase when the drugs are combined.

Benefits

Sulfasalazine versus placebo:

We found no systematic review but found three RCTs in adults with early active rheumatoid arthritis (< 12 months since diagnosis) who had not previously received disease-modifying antirheumatic drugs (DMARDs). The first RCT (105 people aged 22–78 years with early non-erosive rheumatoid arthritis) compared sulfasalazine 2 g daily versus placebo over 6 months. Corticosteroid use was not allowed during the trial. In total, 65 people (62%) completed the trial; analysis was by intention to treat for all outcomes except radiological progression. The RCT found that sulfasalazine significantly improved joint tenderness measured by the Ritchie articular index and the number of swollen and tender joints compared with placebo (see table 2 ). It found no significant difference between sulfasalazine and placebo in pain or morning stiffness, although results were better in people taking sulfasalazine. It also found no significant difference between groups in quality of life as measured by the Gill Quality of Life Instrument (reported as not significant, no further data reported). At 12 months, 23/105 (22%) people had paired films available for assessment of radiological progression: two people taking placebo had erosions in the hands and five people (2 taking sulfasalazine, 3 taking placebo) had erosions in the feet, with no significant difference between groups (reported as not significant). The second RCT (80 adults with early active rheumatoid arthritis, with or without radiographic erosions at baseline) compared sulfasalazine 2 g daily versus placebo for 48 weeks. Five people receiving sulfasalazine and three receiving placebo also took corticosteroids (5 mg/day). Seventy-eight people (98%) completed the study but only 29 (36%) were taking the original allocation; significantly more people taking placebo than sulfasalazine changed treatment to intramuscular gold because of lack of efficacy (14/38 [37%] with sulfasalazine v 28/40 [70%] with placebo; P = 0.003). The RCT found that sulfasalazine significantly reduced joint tenderness (measured by Ritchie articular index), the number of swollen joints, and patient's global assessment at 48 weeks compared with placebo (absolute results presented graphically; P < 0.05 for all outcomes; analysis by intention to treat; last observation carried forward). The improvement in swollen joint count with sulfasalazine was evident after only 4 weeks. There was no significant difference in early morning stiffness or physician's global assessment (reported as not significant, P value not reported; absolute results presented graphically).The RCT found that more people taking sulfasalazine than placebo had no evidence of radiographic progression after 12 months, but did not assess the significance of the difference between groups (people with no evidence of progression: 12/36 [33%] with sulfasalazine v 8/37 [22%] with placebo). The third RCT (186 adults with active rheumatoid arthritis of more than 6 months' duration) compared three interventions: sulfasalazine 2 g daily, placebo, or intramuscular gold injections 50 mg a week for 37 weeks. Gold is no longer given as first-line treatment for people with rheumatoid arthritis, so we did not report results for gold in this section. Participants could take corticosteroids (up to 10 mg prednisone daily) for 1 month before and during the trial; corticosteroid dose was required to remain stable. In total, 109 people completed the study taking their allocated treatment (68% of those taking sulfasalazine, 45% intramuscular gold, and 62% placebo). In the intention-to-treat analysis, the RCT found that sulfasalazine significantly increased the proportion of people with clinically important improvements in tender joint counts compared with placebo (see table 3 ). The RCT found greater improvements in physician global assessment, patient global assessment, and swollen joint count in people taking sulfasalazine compared with placebo, but did not assess the significance of the difference between these two groups. An analysis of difference among people in all three groups (sulfasalazine, gold, or placebo) for these outcomes was not significant. However, in this study, high response rates were seen in people taking placebo for all assessed measures. Patient and physician global assessment were assessed on a scale from 1 to 5 (1 = asymptomatic and 5 = very severe disease activity).

Table 2
Sulfasalazine versus placebo in DMARD naïve people with early rheumatoid arthritis.
Table 3
Sulfasalazine versus placebo or im gold in DMARD-naïve people with early rheumatoid arthritis.

Sulfasalazine versus methotrexate or versus methotrexate plus sulfasalazine:

See benefits of methotrexate.

Sulfasalazine versus hydroxychloroquine:

We found two RCTs comparing sulfasalazine versus hydroxychloroquine. Corticosteroid use was not allowed during either trial. The first RCT (60 people aged 18–75 years with active rheumatoid arthritis who had not previously received DMARDs) compared sulfasalazine versus hydroxychloroquine over 18 months. It found that, after 48 weeks, mean patient global assessment, Ritchie articular index, pain scores, morning stiffness, tender and swollen joint counts, and grip strength were improved in all participants, with no significant difference between groups (see table 4 ). Patient global assessment was assessed on a 10 cm visual analogue scale (0 cm = very well, 10 cm = very bad). Improvement in clinical measures was generally slower in people taking hydroxychloroquine compared with sulfasalazine. The second RCT (60 people aged 22–75 years with active rheumatoid arthritis not adequately controlled by non-steroidal anti-inflammatory agents who had not previously received DMARDs) compared radiological changes in people taking sulfasalazine versus those in people taking hydroxychloroquine. It found that, after 48 weeks, more people taking sulfasalazine than hydroxychloroquine had no erosions (see table 4 ). Both the increases in the number of new erosions and the total number of erosions were significantly lower with sulfasalazine than with hydroxychloroquine over 48 weeks in an analysis of 57/60 [95%] participants.

Table 4
Sulfasalazine versus hydroxychloroquine in DMARD-naïve people with early rheumatoid arthritis.

Harms

Sulfasalazine versus placebo:

In the first RCT, rashes, liver function test abnormalities, and gastrointestinal upsets were more frequent in people taking sulfasalazine compared with placebo, and withdrawals were higher among people taking sulfasalazine (14/53 [26%] with sulfasalazine v 4/52 [8%] with placebo; significance not assessed). In the second RCT, adverse effects leading to withdrawal were few but, again, higher among people taking sulfasalazine compared with placebo (5/38 [13%] with sulfasalazine v 1/40 [3%] with placebo; significance not assessed), and included rash, headache, and gastrointestinal upsets. The third RCT found that more people taking sulfasalazine than placebo withdrew because of adverse effects (16% with sulfasalazine v 10% with placebo; significance assessment not reported). Adverse effects included rash, pruritis, gastrointestinal upset, abnormal liver function tests, and pneumonitis.

Sulfasalazine versus methotrexate or versus methotrexate plus sulfasalazine:

See harms of methotrexate.

Sulfasalazine versus hydroxychloroquine:

In the first RCT, 1/30 (3%) people taking sulfasalazine withdrew because of agranulocytosis, and 3/30 (10%) because of gastrointestinal upset. One person taking hydroxychloroquine withdrew because of borborygmi. Adverse effects not resulting in withdrawal in both groups included gastrointestinal upset, abnormal liver function tests, and abnormal taste. The second RCT did not assess adverse effects.

Comment

Clinical guide:

Although sulfasalazine is used widely as initial DMARD treatment for rheumatoid arthritis, there are no systematic reviews and only small numbers of RCTs among DMARD-naïve patients. However, the available RCTs suggest that sulfasalazine is superior to placebo, hydroxychloroquine, and intramuscular gold. Sulfasalazine is often preferred over methotrexate as first-line treatment as, despite the lack of evidence from RCTs, it is generally regarded in clinical practice as a less toxic drug, which is safe in pregnancy. It can be taken by men intending to father a child, although it can be associated with oligospermia. Compared with methotrexate, sulfasalazine requires less rigorous monitoring (full blood count and liver function tests every 2 weeks for 6 weeks and then every 3 months thereafter). Alcohol consumption is not prohibited.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Antimalarial drugs (first-line treatment)

Summary

DISEASE SEVERITY Hydroxychloroquine compared with placebo: Hydroxychloroquine may improve disease severity compared with placebo in people with rheumatoid arthritis who have not previously received disease-modifying antirheumatic drugs ( very low-quality evidence ). Hydroxychloroquine compared with sulfasalazine: Hydroxychloroquine is as effective as sulfasalazine at improving symptoms and function in people with rheumatoid arthritis ( moderate-quality evidence ). JOINT PAIN AND TENDERNESS Hydroxychloroquine compared with placebo: Hydroxychloroquine may improve joint pain and tenderness compared with placebo in people with rheumatoid arthritis who have not previously received disease-modifying antirheumatic drugs ( low-quality evidence ). NOTE We found no direct information about whether chloroquine is better than no active treatment, or how chloroquine compares with methotrexate.

Benefits

Chloroquine versus placebo:

We found no systematic review or RCTs comparing chloroquine versus placebo in people naïve to disease-modifying antirheumatic drugs (DMARDs).

Hydroxychloroquine versus placebo:

We found no systematic review but found two RCTs. The first RCT (126 people with early rheumatoid arthritis, < 5 years since diagnosis) compared hydroxychloroquine 400 mg daily versus placebo. Corticosteroid use was not allowed during the trial. It found that hydroxychloroquine significantly improved patient and physician global assessment, tender and swollen joint count, pain, and grip strength over 6 months (analysis of 121/126 [96%] participants, last observation carried forward). See table 5 for full results. Patient and physician global assessment were assessed on a scale from 1 to 5 (1 = asymptomatic and 5 = very severe disease activity). The second RCT (19 people with early active rheumatoid arthritis, < 2 years since diagnosis) found that, compared with placebo, hydroxychloroquine at a maximum dose of 7 mg/kg daily (maximum 400 mg/day) significantly improved swollen and tender joint count, and composite pain scores, but did not improve disability as assessed by the Health Assessment Questionnaire or the Arthritis Impact Measurement Scales or psychological wellbeing over 12 months. Intramuscular corticosteroid use was allowed from weeks 2 to 24 of the trial and the RCT found no significant difference between hydroxychloroquine and placebo in the dose of corticosteroids required, although doses were lower in people taking hydroxychloroquine. See table 5 for full results: absolute results are standardised scores covering three domains of pain, physical functioning, and psychological functioning, where score is zero at outset of treatment and positive values denote improvement. Although this standardised scoring system was developed for this RCT, it was devised from pooling of validated measures, for example, the “joint index” combined tender joint count, swollen joint count, grip strength, and duration of early morning stiffness.

Table 5
Hydroxychloroquine versus placebo in DMARD naïve people with early rheumatoid arthritis.

Hydroxychloroquine 200 mg versus 400 mg:

We found two RCTs (77 people aged 18–75 years with moderately active rheumatoid arthritis), which compared hydroxychloroquine 200 mg daily versus 400 mg daily. It found improvements in measures of disease activity as assessed by patient and physician global Assessments, swollen and tender joint counts, Ritchie articular index, pain, early morning stiffness, and grip strength with both doses of hydroxychloroquine, with no significant difference between groups (see table 6 ). In the first RCT, patient global assessment was assessed as “very satisfied”, “satisfied”, or “dissatisfied”. Physician global assessment was assessed as “very good”, “good”, “poor”, or “no effect”. In the second RCT, patient global assessment was measured on a 10 cm visual analogue scale (0 cm = unwell). In both RCTs, participants could take corticosteroids (up to 10 mg/day prednisone) for 1 month before and during the trial; dose was required to remain stable. In both RCTs, results were based on an analysis of people who completed the trial, which was 80–90% of participants.

Table 6
Hydroxychloroquine 200 mg versus 400 mg for second-line treatment.

Hydroxychloroquine versus sulfasalazine:

See benefits of sulfasalazine.

Antimalarial drugs versus methotrexate:

We found no systematic review or RCTs.

Harms

Chloroquine versus placebo:

We found no RCTs.

Hydroxychloroquine versus placebo:

In the first RCT, similar rates of people taking hydroxychloroquine or placebo had adverse effects, primarily nausea or gastric pain (28/63 [44%] with hydroxychloroquine v 28/58 [48%] with placebo; significance not assessed). No one in the trial had ocular toxicity. The second RCT also found similar rates of adverse effects in people taking hydroxychloroquine and placebo (39 adverse events with hydroxychloroquine v 38 adverse events with placebo; significance not assessed). Adverse effects included gastrointestinal upset, rash, and ocular disturbance (blurring, halo, flashing, poor focus).

Hydroxychloroquine 200 mg versus 400 mg:

Both RCTs found similar rates of withdrawal because of adverse effects, including rash, gastrointestinal upset, and blurred vision with both doses of hydroxychloroquine (significance of difference between groups not assessed in either RCT).

Hydroxychloroquine versus sulfasalazine:

See harms of sulfasalazine.

Antimalarial drugs versus methotrexate:

We found no RCTs.

Comment

Clinical guide:

There are currently few RCT data, but antimalarial drugs appear to offer modest clinical benefit in the treatment of people with rheumatoid arthritis who have not previously taken DMARDs. We found no evidence about whether antimalarial drugs delay disease progression, either radiologically or over the long term, as this was not assessed by the RCTs. In clinical practice, few specialists would favour antimalarial drugs over methotrexate or sulfasalazine as first-line treatment for active rheumatoid arthritis, because of the relatively modest clinical effects and lack of radiographic data. However, antimalarial drugs are used in people with comorbidities — for example, hepatic or renal impairment, or overlap syndromes such as systemic lupus erythematosus — and some people, anxious about the risks of toxicity of other DMARDs, will consent to the use of antimalarial drugs as a “safer” alternative. Chloroquine and other antimalarial drugs are associated with dose-related retinal toxicity, and pretreatment ophthalmological screening is recommended. After a review of published evidence, the Royal College of Ophthalmologists and Royal College of Physicians jointly recommended that pretreatment ophthalmological screening prior to the commencement of hydroxychloroquine was not indicated, but that eye symptoms, particularly changes in colour vision, should be monitored ophthalmologically.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Corticosteroids (first-line treatment)

Summary

JOINT PAIN AND TENDERNESS Compared with chloroquine: Oral corticosteroids are as effective as chloroquine at reducing joint pain and tenderness in people with early rheumatoid arthritis who have not previously received disease-modifying antirheumatic drugs ( moderate-quality evidence ). NOTE We found no direct information about whether oral corticosteroids are better than no active treatment.

Benefits

Oral corticosteroids versus placebo:

We found two systematic reviews (search date 1998, search date 2004), which identified no RCTs comparing corticosteroids versus placebo in people with early rheumatoid arthritis who had not previously received disease-modifying antirheumatic drugs.

Oral corticosteroids versus chloroquine:

We found one systematic review (search date 1998), which identified one RCT (56 people, mean age 69 years with early < 1 year rheumatoid arthritis). It found no significant difference between oral prednisone and chloroquine in tender joint count, swollen joint count, or functional status over 2 years (see table 7 ).

Table 7
Prednisone versus chloroquine for second-line treatment.

Harms

Oral corticosteroids versus placebo:

We found no RCTs.

Oral corticosteroids versus chloroquine:

The review gave no information on adverse effects. The RCT found that there was a higher rate of bone mass loss over 2 years in people taking prednisolone.

Comment

Clinical guide:

Corticosteroids (oral or parenteral) offer short-term relief of clinical disease activity in people with rheumatoid arthritis, but the longer-term safety and effectiveness of this strategy in people with rheumatoid arthritis has not been evaluated, particularly given the well recognised long-term risks of high-dose corticosteroids.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Infliximab plus methotrexate (second-line treatment)

Summary

DISEASE ACTIVITY Infliximab plus methotrexate compared with methotrexate alone: Infliximab plus methotrexate reduces disease activity after 6 months compared with methotrexate alone ( high-quality evidence ). ADVERSE EFFECTS Infliximab has been associated with an increased risk of infections including reactivation of tuberculosis. Its long-term safety is unknown. NOTE We found no clinically important results about the effects of infliximab compared with adalimumab, anakinra, antimalarial drugs, azathioprine, ciclosporin, cyclophosphamide, etanercept, leflunomide, penicillamine, or sulfasalazine.

Benefits

Infliximab plus methotrexate versus methotrexate alone:

We found one systematic review (search date 2002), which identified two RCTs (529 people with rheumatoid arthritis). The review combined results in these trials even though they had slightly different designs. The first RCT compared four interventions: infliximab 1 mg/kg (with and without methotrexate), 3 mg/kg (with and without methotrexate), 10 mg/kg (with and without methotrexate), or placebo infusion plus methotrexate. The second RCT compared three different doses of infliximab versus placebo in people who were all receiving a stable background dose of methotrexate. Infusions were given every 4 or 8 weeks. The review found that infliximab at any dose plus methotrexate significantly improved patient global assessment, physician global assessment, and American College of Rheumatology (ACR) 20, 50, and 70 response rates compared with methotrexate at 6 months (see table 8 ). It also found that infliximab plus methotrexate significantly reduced radiographic progression at 12 months compared with methotrexate.

Table 8
Infliximab plus methotrexate versus methotrexate alone for second-line treatment.

Infliximab versus adalimumab, anakinra, antimalarial drugs, azathioprine, ciclosporin, cyclophosphamide, etanercept, leflunomide, penicillamine, or sulfasalazine:

We found no systematic review or RCTs.

Harms

The review did not combine data for all doses of infliximab when assessing adverse effects. It found no significant difference between infliximab 10 mg plus methotrexate and methotrexate alone in the proportion of people who withdrew because of adverse effects (9/95 [9%] with infliximab plus methotrexate v 7/92 [8%] with methotrexate alone; RR 1.07, 95% CI 0.40 to 2.06). Adverse effects were primarily infections (type of infection not specified).

Infliximab versus adalimumab, anakinra, antimalarial drugs, azathioprine, ciclosporin, cyclophosphamide, etanercept, leflunomide, penicillamine, or sulfasalazine:

We found no RCTs.

Drug safety alert

FDA issues a drug safety alert on the risk of opportunistic fungal infections associated with TNF-alpha blockers (tumour necrosis factor alpha-blockers). Some of these invasive fungal infections could be fatal (4 September 2008). FDA issues drug safety alert on the increased risk of lymphoma and other malignancies in children and adolescents, and the risks of leukaemia and new onset psoriasis, associated with TNF blockers (04 August 2009).

The FDA has issued a drug safety alert on the risk of opportunistic fungal infections associated with TNF-alpha blockers (tumour necrosis factor alpha-blockers). Some of these invasive fungal infections could be fatal (http://www.fda.gov). A drug safety alert has been issued on the increased risk of lymphoma and other malignancies in children and adolescents, and the risks of leukaemia and new onset psoriasis, associated with TNF blockers (http://www.fda.gov).

Comment

Clinical guide:

Infliximab allows rapid disease control and reduces rheumatoid arthritis disease activity. It appears to have an acceptable safety profile in these trials. Available efficacy and toxicity data is relatively short term (6 and 12 months), and further long-term studies are needed to assess safety and sustainability of response. Antibodies to double-stranded DNA have been identified following treatment with infliximab. Reactivation of tuberculosis has been documented and people should be screened for previous tuberculosis before treatment.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Leflunomide (second-line treatment)

Summary

DISEASE ACTIVITY Compared with methotrexate: Leflunomide is as effective as methotrexate at reducing disease activity after 6–24 months ( high-quality evidence ). Compared with sulfasalazine: Leflunomide is as effective as sulfasalazine at reducing disease activity over 6–12 months. ADVERSE EFFECTS Leflunomide causes more gastrointestinal adverse effects and hypertension than methotrexate, but is less likely to cause elevated liver function tests. The long-term safety of leflunomide is unknown. NOTE We found no clinically important results about the effects of leflunomide compared with placebo, adalimumab, anakinra, antimalarial drugs, azathioprine, ciclosporin, corticosteroids, cyclophosphamide, etanercept, gold, infliximab or penicillamine.

Benefits

Leflunomide versus methotrexate:

We found one systematic review and one subsequent RCT comparing leflunomide 20 mg daily (with loading dose of 100 mg in 2 RCTs) versus methotrexate (maximum weekly 20 mg oral dose). Participants could take corticosteroids (up to 10 mg/day prednisone) during the trials; corticosteroid dose was required to remain stable. The review (search date 2002, 3 RCTs, 1680 people aged > 18 years) found no significant difference between leflunomide and methotrexate in American College of Rheumatology (ACR)20 response rates at 6, 12, or 24 months (2 RCTs at 12 months: 345/677 [51%] with leflunomide v 401/671 [60%] with methotrexate; RR 1.17, 95% CI 1.07 to 1.29). The subsequent RCT (566 people with active rheumatoid arthritis) compared leflunomide 20 mg daily (without loading dose) versus oral methotrexate 15 mg or greater, weekly for 24 weeks.Corticosteroid use was not allowed during the trial. Intention-to-treat analysis at 6 months found no significant difference in the proportion of people with ACR20 response rates between leflunomide and methotrexate (see table 9 ). Completer analysis of 504/566 (89%) people also found no significant difference in measures of disease activity at 6 months as measured by patient and physician global assessment, tender and swollen joint count, grip strength, resting pain, and duration of morning stiffness.

Table 9
Leflunomide versus methotrexate for second-line treatment.

Leflunomide versus sulfasalazine:

We found one systematic review (search date 2002) comparing leflunomide 20 mg daily (with and without loading regimen) versus sulfasalazine 2 g daily. Participants could take corticosteroids (up to 10 mg prednisone daily) during the trials; corticosteroid dose was required to remain stable. The review (search date 2002, 2 RCTs, 554 people aged > 18 years with active rheumatoid arthritis) found no significant difference between leflunomide and sulfasalazine in ACR20 response rates at 6 or 12 months (12 months, 1 RCT: 51/74 [69%] with leflunomide v 52/78 [67%] with sulfasalazine; RR 1.03, 95% CI 0.83 to 1.29). However, it found that leflunomide significantly increased ACR20 response rates at 24 months (1 RCT: 34/57 [60%] with leflunomide v 49/60 [82%] with sulfasalazine; RR 0.73, 95% CI 0.57 to 0.93).

Leflunomide versus adalimumab, anakinra, antimalarial drugs, azathioprine, ciclosporin, corticosteroids, cyclophosphamide, etanercept, gold, infliximab, or penicillamine:

We found no systematic review or RCTs.

Harms

Leflunomide versus methotrexate:

The review found that significantly more people taking leflunomide than methotrexate withdrew because of adverse effects over 12 months (1 RCT: 134/683 [20%] with leflunomide v 93/680 [14%] with methotrexate; RR 1.43, 95% CI 1.13 to 1.83). It found that, compared with methotrexate, leflunomide significantly increased gastrointestinal symptoms and hypertension (gastrointestinal symptoms: RR 1.25, 95% CI 1.11 to 1.41; hypertension: RR 2.29, 95% CI 1.42 to 3.69). However, significantly fewer people taking leflunomide than methotrexate had elevated liver function tests at 12 months (2 RCTs, 56/683 [8%] with leflunomide v 105/680 [15%] with methotrexate; OR 0.53, 95% CI 0.39 to 0.72). The subsequent RCT found that fewer people taking leflunomide than methotrexate withdrew because of adverse effects, mainly gastrointestinal upset (0.34% with leflunomide v 6.10% with methotrexate; P < 0.001; absolute results not reported).

Leflunomide versus sulfasalazine:

The review found that fewer people taking leflunomide than sulfasalazine withdrew because of adverse effects, but the difference between groups was not significant (1 RCT: 19/130 [15%] with leflunomide v 25/130 with sulfasalazine; RR 0.77, 95% CI 0.43 to 1.33). It also found no significant difference between leflunomide and sulfasalazine in the proportion of people who had elevated liver function tests, gastrointestinal symptoms, infections, or weight loss (elevated liver function tests: OR 0.60, 95% CI 0.15 to 2.44; gastrointestinal symptoms: OR 0.82, 95% CI 0.50 to 1.36; infections: OR 0.30, 95% CI 0.05 to 1.73; weight loss: OR 2.79, 95% CI 0.69 to 11.38).

Leflunomide versus adalimumab, anakinra, antimalarial drugs, azathioprine, ciclosporin, corticosteroids, cyclophosphamide, etanercept, gold, infliximab, or penicillamine:

We found no RCTs.

Comment

Clinical guide:

Leflunomide offers symptomatic relief of clinical rheumatoid arthritis over 6 months' treatment, with rates of adverse effects comparable to those of methotrexate. The use of the loading regimen seems associated with greater adverse effects. RCTs assessing the effects of leflunomide on radiological progression and the effects of long-term treatment with leflunomide are needed.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Methotrexate (second-line treatment)

Summary

DISEASE ACTIVITY Compared with sulfasalazine: Methotrexate may be as effective as sulfasalazine at reducing disease activity after 26 weeks ( low-quality evidence ). Compared with antimalarial drugs: Methotrexate may be more effective than hydroxychloroquine or chloroquine at reducing disease severity after 26 weeks (low-quality evidence). Compared with oral gold: Methotrexate may be more effective than oral gold at reducing disease activity after 26 weeks (low-quality evidence). Compared with parenteral gold: Methotrexate may be as effective as parenteral gold at reducing disease activity after 6–12 months (low-quality evidence). Compared with penicillamine: Methotrexate may be as effective as penicillamine at reducing disease activity after 26 weeks (low-quality evidence). Compared with leflunomide: Methotrexate is as effective as leflunomide at reducing disease activity after 6–24 months ( high-quality evidence ). Compared with azathioprine: Methotrexate may be as effective as azathioprine at reducing disease activity after 24–48 weeks ( very low-quality evidence ). Compared with etanercept: Methotrexate is as effective as etanercept at reducing disease activity after 12 months (high-quality evidence). Methotrexate plus sulfasalazine plus hydroxychloroquine compared with dual or monotherapy: Triple therapy with methotrexate plus sulfasalazine plus hydroxychloroquine is more effective at reducing disease activity after 2 years compared with methotrexate plus sulfasalazine or methotrexate plus hydroxychloroquine dual therapy or compared with methotrexate alone (high-quality evidence). Compared with methotrexate plus etanercept: Methotrexate alone is less effective at reducing disease activity compared with methotrexate plus etanercept after 52 weeks (moderate-quality evidence). Compared with methotrexate plus infliximab: Methotrexate alone is less effective at reducing disease activity compared with methotrexate plus infliximab after 6 months (high-quality evidence). ADVERSE EFFECTS Methotrexate is generally well tolerated but has been associated with hepatic abnormalities, pneumonitis, and nodulosis. NOTE We found no clinically important results about the effects of methotrexate compared with adalimumab, anakinra, ciclosporin, cyclophosphamide, or corticosteroids.

Benefits

Methotrexate versus sulfasalazine:

We found two systematic reviews (search date 1989, search date 1990). The second review was an update of the first and both had the same weaknesses in their methods (see comment below), so we present only limited data here. The reviews found that methotrexate (minimum 7.5 mg/week orally or parenterally) and sulfasalazine (minimum 2 g/day) were associated with comparable improvements in measures of disease activity over a mean 26 weeks as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (17 treatment groups, 507 people > 18 years with active rheumatoid arthritis).

Methotrexate versus antimalarial drugs:

We found two systematic reviews (see comment below). The reviews found that, compared with antimalarial drugs (hydroxychloroquine at a dose of at least 200 mg/day or chloroquine at a dose of at least 250 mg/day), methotrexate (at least 7.5 mg/week orally or parenterally) was associated with greater improvements in measures of disease activity over a mean 26 weeks as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (20 treatment arms, 588 people aged > 18 years with active rheumatoid arthritis).

Methotrexate versus oral gold:

We found two systematic reviews. One of the RCTs identified by the reviews assessed radiographic outcome data in a subsequent publication. The reviews found that, compared with oral gold (> 6 mg/day), methotrexate (> 7.5 mg orally or parenterally weekly) produced greater improvements in measures of disease activity over a mean 26 weeks as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (35 treatment arms, 1774 people aged > 18 years with active rheumatoid arthritis; see comment below). A subsequent publication reported the radiographic outcome data from one of the RCTs (281 people with active rheumatoid arthritis for > 20 years) identified by the reviews comparing methotrexate versus oral gold for 9 months. Radiographic data were available for 167/281 (59%) people. The RCT found that, after 9 months, there were significantly fewer new erosions in people taking methotrexate compared with oral gold (mean: 0.60 with methotrexate v 1.67 with oral gold; P = 0.007).

Methotrexate versus parenteral gold:

We found two systematic reviews and two subsequent RCTs. The reviews found that both methotrexate (minimum 7.5 mg/week orally or parenterally) and intramuscular gold (50 mg/week) were associated with similar improvements in measures of disease activity over a mean 26 weeks as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (37 treatment groups, 922 people > 18 years with active rheumatoid arthritis; see comment below). The first subsequent RCT (57 people aged 18–70 years with active rheumatoid arthritis but no radiographic erosions at baseline) compared parenteral methotrexate 15 mg weekly versus parenteral gold 50 mg weekly over 6 months. Participants could take corticosteroids (up to 10 mg prednisone daily) for 1 month before and during the trial; corticosteroid dose was required to remain stable. The RCT found that the Hannover Activities of Daily Living score, swollen and tender joint count, duration of morning stiffness, and grip strength improved significantly in both groups with no difference between groups (see table 10 for full results; analysis of 47/57 [82%] people who completed the trial). The second subsequent RCT (102 people aged 18–70 years with active rheumatoid arthritis but no radiographic erosions at baseline) compared parenteral methotrexate 15 mg weekly versus parenteral gold 50 mg weekly over 12 months. Participants could take corticosteroids (up to 5 mg prednisone daily) during the trial; corticosteroid dose was required to remain stable. The RCT found that measures of disease activity (tender joint count, swollen joint count, duration of morning stiffness, and Activities of Daily Living score) improved significantly from baseline in both groups (P < 0.001) with no significant difference between groups. Radiological progression assessed by Larsen score increased significantly from baseline in both groups (P < 0.001) but, again, with no significant difference between groups. See table 10 for full details.

Table 10
Methotrexate versus parenteral gold for second-line treatment.

Methotrexate versus penicillamine:

We found two systematic reviews (see comment below). The reviews found that methotrexate orally or parenterally (> 7.5 mg/week) and penicillamine at 500 mg daily produced comparable improvements in measures of disease activity over a mean 26 weeks as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (31 treatment groups, 857 people aged > 18 years with active rheumatoid arthritis).

Methotrexate versus leflunomide:

See benefits of leflunomide.

Methotrexate versus azathioprine:

We found two systematic reviews and two subsequent RCTs (reported in 5 publications) comparing methotrexate versus azathioprine. The first review found that, compared with azathioprine at a minimum dose of 1 mg/kg daily, methotrexate 7.5 mg weekly, orally or parenterally, was associated with greater improvements in measures of disease activity over a mean 26 weeks as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (24 treatment groups, 552 people aged > 18 years with active rheumatoid arthritis; see comment below). The second review (search date 1991, 2 RCTs, 100 people aged > 18 years with active rheumatoid arthritis) considered RCTs which evaluated radiographic progression. It found that people taking methotrexate had significantly slower rates of radiological progression over 24–48 weeks as defined by development of new radiographic erosions compared with azathioprine (weighted rate of progression assessed by Sharp score: 0.009 with methotrexate v 0.011 with azathioprine; P = 0.049). The review did not report whether participants in the RCTs were taking corticosteroids. RCTs were only included if all participants were accounted for at the end of the trial, but it is unclear whether the RCTs performed intention-to-treat analyses. The first subsequent RCT (64 people with active rheumatoid arthritis) compared weekly oral methotrexate at a maximum dose of 15 mg versus daily azathioprine at a maximum dose of 150 mg. Participants could take corticosteroids (up to 10 mg prednisone daily) during the trial; dose was required to remain stable. The RCT found that the rate of development of new erosions was significantly slower with methotrexate compared with azathioprine at 48 weeks (see table 11 ). The RCT assessed overall clinical response by assessing four variables: pain scores assessed on a 100 mm visual analogue scale, Ritchie articular index, and duration of morning stiffness. “Good or partial overall improvement” was defined as improvement of 30–50% on at least two variables with no worsening on the others. The RCT found that more people taking methotrexate than azathioprine had “good or partial overall improvement” at 48 weeks, but did not perform an intention-to-treat analysis assessing the significance of the difference between groups (see table 11 ). A second report of the same RCT reported no significant difference between groups in “clinical variables” in the intention-to-treat analysis at 48 weeks, but reported no results. This analysis is difficult to apply in clinical practice because, at 48 weeks, only 12/33 (36%) of people continued to take azathioprine, whereas 25/31 (81%) continued to take methotrexate. Therefore, despite the intention-to-treat analysis, it is unclear how effective azathioprine may be. From a clinical perspective, methotrexate seems to shows greater benefit in reducing rates of radiological progression over 48 weeks but, providing that patients are able to tolerate azathioprine, there do not appear to be significant differences in clinical responses between the two groups. The second subsequent RCT (209 people with active rheumatoid arthritis) compared three interventions: methotrexate (maximum 15 mg/week), azathioprine (maximum 150 mg/day), and the combination of both drugs (at the same maximum doses). Participants could take corticosteroids (up to 10 mg prednisone daily) during the trial; corticosteroid dose was required to remain stable. Response was defined as at least 30% improvement in at least three of four clinical measures (patient global assessment, physician global assessment, swollen joint count, and tender joint count). The RCT found no significant difference among groups in the proportion of people who responded at 24 or 48 weeks (see table 11 ; intention-to-treat analysis); although improvement rates were higher with both methotrexate and combination therapy than with azathioprine alone. Fewer people taking methotrexate alone than azathioprine alone or combined treatment had radiographic progression over 48 weeks, but there was no significant difference among groups (reported as not significant, P value and absolute numbers not reported).

Table 11
Methotrexate versus azathioprine or the combination versus either drug alone for second-line treatment.

Methotrexate in triple combination with sulfasalazine and hydroxychloroquine:

See benefits of methotrexate plus sulfasalazine plus hydroxychloroquine.

Methotrexate plus etanercept versus either drug alone:

See benefits of etanercept.

Methotrexate plus infliximab versus methotrexate alone:

See benefits of infliximab plus methotrexate.

Methotrexate versus adalimumab, or anakinra, ciclosporin, cyclophosphamide, corticosteroids:

We found no systematic review or RCTs.

Harms

Methotrexate versus sulfasalazine:

The reviews found that, compared with sulfasalazine, methotrexate was associated with lower rates of withdrawal because of toxicity, and both drugs were associated with a comparably low rate of severe toxicity (17 treatment groups, 507 people aged > 18 years with active rheumatoid arthritis; see comment below).

Methotrexate versus antimalarial drugs:

The reviews found that methotrexate was associated with a slightly higher risk of withdrawal because of toxicity and a slightly higher rate of severe toxicity than antimalarial drugs (20 treatment arms, 588 people aged > 18 years with active rheumatoid arthritis; see comment below).

Methotrexate versus oral gold:

The reviews found that methotrexate was associated with a slightly higher risk of withdrawal because of toxicity, and a slightly higher rate of severe toxicity than oral gold (35 treatment arms, 1774 people aged > 18 years with active rheumatoid arthritis; see comment below).

Methotrexate versus parenteral gold:

The reviews found that, compared with parenteral gold, methotrexate had a lower risk of withdrawal because of toxicity and a lower rate of severe toxicity (37 treatment groups, 922 people > 18 years with active rheumatoid arthritis; see comment below). The first subsequent RCT found that fewer people taking methotrexate than parenteral gold withdrew because of adverse effects, including nausea with methotrexate and stomatitis, proteinuria, and rash with gold (2/29 [7%] with methotrexate v 5/28 [18%] with gold; significance not assessed). The second subsequent RCT found similar results; fewer people taking methotrexate than parenteral gold withdrew because of adverse effects, and adverse effects causing withdrawal were also similar to those in the first RCT (7/52 [13%] with methotrexate v 16/50 [32%] with gold; significance not assessed).

Methotrexate versus penicillamine:

The reviews found that, compared with penicillamine, methotrexate was less likely to result in withdrawals because of toxicity and less likely to cause severe drug toxicity (31 treatment groups, 857 people aged > 18 years with active rheumatoid arthritis).

Methotrexate versus azathioprine:

The review found that, compared with azathioprine, methotrexate was associated with lower rates of withdrawal because of toxicity and lower rates of severe toxicity (24 treatment groups, 552 people aged > 18 years with active rheumatoid arthritis; see comment below). The second review did not assess adverse effects. The first subsequent RCT found that the number of withdrawals was significantly higher in the azathioprine group: after 48 weeks, 12/33 (36%) people treated with azathioprine and 25/31 (81%) people treated with methotrexate were still taking their initial drug. Withdrawals because of adverse effects were higher in people taking azathioprine after 48 weeks (2/31 [6%] with methotrexate v 13/33 [39%] with azathioprine; P < 0.03). Methotrexate withdrawals were primarily caused by elevated liver enzymes, and azathioprine withdrawals by nausea and abdominal pain. The second subsequent RCT found lower rates of withdrawals because of adverse effects with methotrexate than with azathioprine or combination treatment at 24 weeks, but did not assess the significance of the difference between groups (3/67 [4%] with methotrexate v 22/73 [30%] with azathioprine v 16/69 [23%] with methotrexate plus azathioprine). Withdrawals were primarily caused by gastrointestinal upset and elevated liver enzymes. The RCT did not assess withdrawals caused by adverse effects at 48 weeks.

Methotrexate in triple combination with sulfasalazine and hydroxychloroquine:

See harms of methotrexate plus sulfasalazine plus hydroxychloroquine.

Methotrexate plus etanercept versus either drug alone:

See harms of etanercept.

Methotrexate plus infliximab versus either drug alone:

See harms of infliximab plus methotrexate.

Methotrexate versus adalimumab, anakinra, ciclosporin, cyclophosphamide, or corticosteroids:

We found no RCTs.

Hepatic abnormalities:

We found one systematic review of histological hepatic abnormalities associated with methotrexate that included a meta-analysis (search date not reported, 15 observational studies, 636 people with either rheumatoid arthritis or psoriatic arthritis treated with methotrexate for a mean duration of 18 years). The review assessed histological grade using a scale from 1–4 similar to the Roenigk scale. It found that 28% (95% CI 24.5% to 31.6%) of people with rheumatoid arthritis taking methotrexate progressed by at least one histological grade, with risk of progression dependent upon average cumulative dose of methotrexate (P = 0.01) but independent of average duration of treatment, average dose per week, and maximum dose per week. Advanced histological changes were seen in 5% (95% CI 3.5% to 7.0%) of all people with rheumatoid arthritis treated with methotrexate. The risk of histological liver abnormality was greater in heavy (at least 100 g of alcohol/week) than light alcohol drinkers (18% for heavy drinkers v 4% for light drinkers; P = 0.0003).

Pulmonary complications:

We found one non-systematic review (6 prospective, 10 retrospective cohort studies, 1947 people with rheumatoid arthritis taking methotrexate), which estimated pneumonitis prevalence of 0.3–11.6% over 2–8 years of observation of people taking methotrexate.

Nodulosis (accelerated subcutaneous rheumatoid nodules):

One systematic review (search date 2001) investigated nodulosis associated with methotrexate, and found 27 case reports and two case series (one with 10 people and the other with 21 people) of accelerated nodulosis associated with methotrexate. The authors of the review described the data as fragmented and incomplete and insufficient to draw conclusions as to prevalence and risk factors.

Stomatitis:

We found one prospective cohort study (97 people with rheumatoid arthritis) that evaluated the prevalence of stomatitis in people taking methotrexate versus those not taking methotrexate using a questionnaire developed for the study. It found a higher rate of stomatitis among people taking methotrexate, although the difference between groups did not quite reach significance (19/51 [37%] with methotrexate v 9/46 [19%] with no methotrexate; P = 0.056).

Comment

Two of the reviews we found were weak methodologically. The second review was an update of the first with an additional analysis of azathioprine. The reviews assessed tender joint counts and grip strength with a variety of disease-modifying antirheumatic drugs (DMARDs) by extracting within group data from RCTs and deriving analyses of “treatment groups”. The reviewers then performed indirect comparisons to assess the relative efficacy of the DMARDs. Clearly, these indirect analyses are justified to some extent by the fact that direct comparisons have not been carried out in large-scale RCTs. However, indirect analyses should be treated with caution as they lose the benefit of randomisation; we have therefore reported only limited data from these reviews. The second review found similar results to the first: it updated the information regarding number of people in the treatment groups, but presented all outcome data graphically.

Clinical guide:

In practice, methotrexate has become the most widely used DMARD for the treatment of people with active rheumatoid arthritis, because it is generally relatively well tolerated and effective. Methotrexate is widely used as the “gold standard” against which new DMARDs and biological agents are tested. Sulfasalazine is used in practice fairly interchangeably with methotrexate, depending on patient and clinician preference, and many rheumatologists opt for a “step up” approach, whereby they begin treatment with one of methotrexate or sulfasalazine, but add the other agent (with or without hydroxychloroquine) fairly quickly (after 3–6 months) if clinical and laboratory evidence of improvement in disease activity is not obtained. Although leflunomide has strong RCT evidence in comparison with some of the other agents, its arrival was heralded enthusiastically before biological agents and many people who had already “failed” traditional DMARDs were started on it using the rapid loading dose regimen recommended by the manufacturer. In many cases, the rapid loading of leflunomide produced problems with gastrointestinal adverse effects and, since many of these people already had severe long-standing disease, the anticipated benefits in practice did not reach the levels seen in RCTs. Leflunomide is now used more widely, often without rapid loading, and it is likely to show much better tolerability and effectiveness when used earlier in the course of the disease. Gold, azathioprine, ciclosporin, and other DMARDs are being used less as the biological agents have become more widely available.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Methotrexate plus sulfasalazine plus hydroxychloroquine (second-line treatment)

Summary

DISEASE ACTIVITY Compared with dual or monotherapy: Triple therapy with methotrexate plus sulfasalazine plus hydroxychloroquine is more effective at reducing disease activity after 2 years compared with methotrexate plus sulfasalazine or methotrexate plus hydroxychloroquine dual therapy or compared with methotrexate alone ( high-quality evidence ).

Benefits

We found no systematic review but found two RCTs.

Methotrexate plus sulfasalazine plus hydroxychloroquine versus dual treatment or monotherapy:

The first RCT (102 people aged 19–70 years with active rheumatoid arthritis recalcitrant to at least 1 previous disease-modifying antirheumatic drug compared three interventions: methotrexate alone (maximum 17.5 mg/week), dual therapy with sulfasalazine (1 g daily) plus hydroxychloroquine (400 mg/day), or triple therapy with methotrexate plus sulfasalazine plus hydroxychloroquine. Participants could take corticosteroids (up to 10 mg prednisone daily or equivalent) during the trial; dose was required to remain stable. The RCT found that significantly more people receiving the triple combination had 50% or greater improvement in measures of disease activity at 2 years compared with dual treatment or monotherapy (24/31 [77%] with methotrexate plus sulfasalazine plus hydroxychloroquine v 14/35 [40%] with sulfasalazine plus hydroxychloroquine v 12/36 [33%] with methotrexate alone; P = 0.03 for triple v dual, P < 0.001 for triple v monotherapy; intention to treat analysis). Improvement in measures of disease activity of 50% or greater was defined as fulfilling at least three of the following criteria: morning stiffness of 30 minutes or less or decreased by 50% from baseline; joint tenderness decreased by 50%, joint swelling decreased by 50%, and erythrocyte sedimentation rate (ESR) 30 mm an hour or less in women or 20 mm an hour or less in men. The second RCT (171 people aged > 18 years with active rheumatoid arthritis recalcitrant to disease-modifying antirheumatic drugs) compared three interventions: methotrexate (maximum 17.5 mg weekly) plus hydroxychloroquine 400 mg daily, methotrexate plus sulfasalazine 2 g daily, or methotrexate plus hydroxychloroquine plus sulfasalazine. Participants could take corticosteroids (up to 10 mg prednisone daily or equivalent) during the trial; corticosteroid dose was required to remain stable. The RCT found that the combination of methotrexate plus hydroxychloroquine plus sulfasalazine significantly improved clinical disease activity as measured by the American College of Rheumatology (ACR)20 after 2 years compared with dual therapy using methotrexate plus sulfasalazine (45/58 [78%] with methotrexate plus hydroxychloroquine plus sulfasalazine v 27/55 [49%] with methotrexate plus sulfasalazine; P = 0.002). The difference between triple and dual therapy using methotrexate plus hydroxychloroquine was of borderline significance (45/58 [78%] with methotrexate plus hydroxychloroquine plus sulfasalazine v 35/58 [60%] with methotrexate plus hydroxychloroquine; P = 0.05). Similar trends were seen for ACR50 response (55% with methotrexate plus hydroxychloroquine plus sulfasalazine v 40% with methotrexate plus hydroxychloroquine v 29% with methotrexate plus sulfasalazine; P = 0.005 for triple v methotrexate plus sulfasalazine, P = 0.10 for triple v methotrexate plus hydroxychloroquine; absolute numbers not reported).

Harms

Methotrexate plus sulfasalazine plus hydroxychloroquine versus dual treatment or monotherapy:

Similar rates of withdrawal owing to adverse effects were seen among those receiving the triple combination compared with those seen on methotrexate alone, sulfasalazine alone, or dual treatment. Neither RCT assessed the significance of the difference between groups.

Comment

Clinical guide:

These data suggest that the triple combination of methotrexate plus sulfasalazine plus hydroxychloroquine is relatively well tolerated and offers benefit in the control of clinical disease activity, at least in the short term. The long-term safety and efficacy of this combination is yet to be established.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Penicillamine (second-line treatment)

Summary

DISEASE ACTIVITY Compared with methotrexate: Penicillamine may be as effective as methotrexate at reducing disease activity after 26 weeks ( low-quality evidence ). Compared with sulfasalazine: Penicillamine is as effective as sulfasalazine at reducing disease activity ( moderate-quality evidence ). Compared with oral gold: Penicillamine may be more effective than oral gold at reducing disease activity (low-quality evidence). Compared with azathioprine: Penicillamine may be as effective at reducing disease activity after 24–26 weeks compared with azathioprine (low-quality evidence). Compared with ciclosporin: Penicillamine is as effective as ciclosporin at reducing disease activity after 6 months (moderate-quality evidence). ADVERSE EFFECTS Penicillamine may be associated with higher rates of adverse effects compared with methotrexate. NOTE We found no clinically relevant results about the effects of penicillamine compared with adalimumab, anakinra, antimalarial drugs, corticosteroids, cyclophosphamide, etanercept, infliximab, leflunomide, or parenteral gold.

Benefits

Penicillamine versus methotrexate:

See benefits of methotrexate.

Penicillamine versus sulfasalazine:

See benefits of sulfasalazine.

Penicillamine versus oral gold:

See benefits of oral gold.

Penicillamine versus azathioprine:

We found one systematic review (search date 1990), which had weaknesses in its methods (see comment below) so we present only limited data here. We also found one subsequent RCT. The review found that penicillamine and azathioprine produced comparable improvements in measures of disease activity over a mean 26 weeks as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (37 treatment arms, 861 people aged > 18 years with active rheumatoid arthritis). The subsequent RCT (206 people aged > 18 years with active rheumatoid arthritis who had not tolerated or had failed to respond to gold) compared penicillamine 10–12 mg/kg daily versus azathioprine 1.25–1.5 mg/kg daily. Corticosteroid use was allowed during the trial; a total of 82/206 (39%) of participants took a stable dose of 10 mg prednisolone or less or equivalent daily (unclear how many from each group). In an intention-to-treat analysis at 24 weeks, the RCT found that a comparable proportion of people had “meaningful improvement” in patient and physician global assessment and in swollen and tender joint count, with no significant difference between groups (reported as not significant, P value not reported, absolute numbers presented graphically). Meaningful improvement was defined as an improvement of 2 or greater on a 5-point scale in overall assessment and a 30% or greater decrease in the number of tender or swollen joints.

Penicillamine versus ciclosporin:

We found no systematic review but found one RCT (92 people aged 18–75 years with recalcitrant rheumatoid arthritis) comparing ciclosporin 5 mg/kg daily versus penicillamine (step up regime starting with 250 mg/day) over 6 months. Corticosteroid use was not allowed during the trial. At 24 weeks, the RCT found improvements from baseline in clinical measures of disease activity (Patient and Physician Global Assessments, duration of morning stiffness, swollen joint count, Ritchie articular index, pain scores, and grip strength) in both groups, with no significant difference between groups (see table 12 ). Pain was assessed using a scale from 0 to 3 (0 = none, 3 = severe).

Table 12
Penicillamine versus ciclosporin for second-line treatment.

Penicillamine versus adalimumab, anakinra, antimalarial drugs, corticosteroids, cyclophosphamide, etanercept, infliximab, leflunomide, or parenteral gold:

We found no systematic review or RCTs.

Harms

Penicillamine versus methotrexate:

See harms of methotrexate.

Penicillamine versus sulfasalazine:

See harms of sulfasalazine.

Penicillamine versus oral gold:

See harms of oral gold.

Penicillamine versus azathioprine:

The review found that penicillamine and azathioprine were associated with similar rates of withdrawal caused by toxicity and severe toxicity (37 treatment arms, 861 people aged > 18 years with active rheumatoid arthritis; see comment below).The subsequent RCT found similar rates of withdrawals caused by adverse effects over 24 weeks with penicillamine and azathioprine (29/102 [28%] with penicillamine v 20/104 [19%] with azathioprine; significance not reported). Penicillamine was associated primarily with rash or pruritus and thrombocytopenia, and azathioprine with gastrointestinal upset.

Penicillamine versus ciclosporin:

The RCT found that ciclosporin significantly increased blood pressure compared with penicillamine (mean systolic blood pressure: 139 mm Hg with ciclosporin v 130 mm Hg with penicillamine; mean diastolic blood pressure: 83 mm Hg with ciclosporin v 79 mm Hg with penicillamine; P < 0.05 for both outcomes). More people taking ciclosporin than penicillamine withdrew because of adverse effects, primarily gastrointestinal intolerance in both groups, although ciclosporin was also associated with hypertension (9/46 [20%] with ciclosporin v 5/46 [11%] with penicillamine; significance not assessed).

Penicillamine versus adalimumab, anakinra, antimalarial drugs, corticosteroids, cyclophosphamide, etanercept, infliximab, leflunomide, or parenteral gold:

We found no systematic review or RCTs.

Comment

One of the reviews was an update of an earlier review with an additional analysis of azathioprine. The reviews assessed tender joint counts and grip strength with a variety of disease-modifying antirheumatic drugs by extracting within-group data from RCTs and deriving analyses of “treatment groups”. The reviewers then performed indirect comparisons to assess the relative efficacy of the disease-modifying antirheumatic drugs. Clearly, these indirect analyses are justified to some extent by the fact that direct comparisons have not been carried out in large-scale RCTs. However, indirect analyses in reviews should be treated with caution as they lose the benefit of the randomisation in the trials they analyse; we have therefore reported only limited data from the review. The updated review found similar results to the first: it updated the information regarding number of people in the treatment groups and added details for azathioprine, but presented all outcome data graphically.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Sulfasalazine (second-line treatment)

Summary

DISEASE ACTIVITY Compared with methotrexate: Sulfasalazine may be as effective as methotrexate at reducing disease activity after 26 weeks ( low-quality evidence ). Compared with antimalarial drugs: Sulfasalazine may be as effective as antimalarial drugs at reducing disease activity after 26–39 weeks ( very low-quality evidence ). Compared with oral gold: Sulfasalazine may be more effective at reducing disease activity compared with oral gold (low-quality evidence). Compared with parenteral gold: Sulfasalazine may be as effective as parenteral gold at reducing disease activity (very low-quality evidence). Compared with penicillamine: Sulfasalazine is as effective as penicillamine at reducing disease activity ( moderate-quality evidence ). Compared with leflunomide: Sulfasalazine is as effective as leflunomide at reducing disease activity over 6–12 months. Sulfasalazine plus methotrexate plus hydroxychloroquine versus dual or monotherapy: Triple therapy with sulfasalazine plus methotrexate plus hydroxychloroquine is more effective at reducing disease activity after 2 years compared with sulfasalazine plus methotrexate or methotrexate plus hydroxychloroquine dual therapy, or compared with methotrexate alone ( high-quality evidence ). ADVERSE EFFECTS Sulfasalazine is associated with similar rates of severe adverse effects compared with methotrexate. NOTE We found no clinically important results about the effects of sulfasalazine compared with adalimumab, anakinra, azathioprine, ciclosporin, cyclophosphamide, corticosteroids, etanercept, or infliximab.

Benefits

Sulfasalazine versus methotrexate:

See benefits of methotrexate.

Sulfasalazine versus antimalarial drugs:

We found three systematic reviews (search date 1989, search date 1990; search date 1998). The second review was an update of the first and both had the same weaknesses in their methods (see comment below) so we present only limited data here. The reviews found that sulfasalazine (minimum 2 g/day) was more effective than hydroxychloroquine in improving measures of disease activity over a mean 26 weeks as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (19 treatment groups, 547 people > 18 years with active rheumatoid arthritis). The third review (search date 1998, 2 RCTs, 120 people aged > 18 years with active rheumatoid arthritis) had much stronger methods as it assessed only RCTs directly comparing sulfasalazine versus hydroxychloroquine. It found no significant difference between sulfasalazine (mean dose 2 g/day) and hydroxychloroquine (mean dose 350 mg/day) in duration of morning stiffness, pain, swollen joint count, or articular index over a mean 39 weeks, although sulfasalazine was more effective in improving these outcomes (see table 13 ). Fewer participants receiving sulfasalazine withdrew because of lack of efficacy (3/30 [10%] with sulfasalazine v 9/30 [30%] with hydroxychloroquine; P = 0.006).

Table 13
Sulfasalazine versus antimalarial drugs for second-line treatment.

Sulfasalazine versus oral gold:

We found two systematic reviews, which found that sulfasalazine (minimum 2 g/day) produced greater improvements in measures of disease activity as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength compared with oral gold (minimum 6 mg/day) (34 treatment groups, 1733 people aged > 18 years with active rheumatoid arthritis; see comment below).

Sulfasalazine versus parenteral gold:

We found three systematic reviews that included a meta-analysis. Two of the reviews found that sulfasalazine (minimum 2 g/day) produced greater improvements in measures of disease activity as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength than intramuscular gold 50 mg weekly (36 treatment groups, 881 people aged > 18 years with active rheumatoid arthritis; see comment below). The third review (search date 1998, 3 RCTs, 377 people aged > 18 years with active rheumatoid arthritis) had much stronger methods, as it assessed only RCTs directly comparing sulfasalazine versus parenteral gold. It found no significant difference between sulfasalazine (mean 2 g/day) and parenteral gold (1 g/week of intramuscular sodium thiomalate) in duration of morning stiffness, pain scores, swollen joint count, and tender joint count (see table 14 ). However, people receiving sulfasalazine were significantly more likely to withdraw because of lack of efficacy than people receiving parenteral gold (24/181 [13%] with sulfasalazine v 81/196 [4%] with parenteral gold; P = 0.006).

Table 14
Sulfasalazine versus parenteral gold for second-line treatment.

Sulfasalazine versus penicillamine:

We found three systematic reviews. Two of the reviews found that penicillamine (up to 500 mg/day) and sulfasalazine (up to 2 g/day) produced comparable improvements in measures of disease activity as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (30 treatment groups, 816 people aged > 18 years with active rheumatoid arthritis; see comment below). The third review (search date 1998, 3 RCTs, 111 people aged > 18 years with active rheumatoid arthritis) had much stronger methods as it assessed only RCTs directly comparing sulfasalazine versus penicillamine. It found no significant difference between sulfasalazine (2 g/day) and penicillamine (mean 667 mg/day) in duration of morning stiffness and articular index (see table 15 ). There was also no significant difference between groups in the proportion of people who withdrew because of lack of efficacy (10% with sulfasalazine v 11% with penicillamine; P = 0.58).

Table 15
Sulfasalazine versus penicillamine for second-line treatment.

Sulfasalazine versus leflunomide:

See benefits of leflunomide.

Sulfasalazine in triple combination with methotrexate and hydroxychloroquine:

See benefits of methotrexate plus sulfasalazine plus hydroxychloroquine.

Sulfasalazine versus adalimumab, anakinra, azathioprine, ciclosporin, cyclophosphamide, corticosteroids, etanercept, or infliximab:

We found no systematic review or RCTs.

Harms

Sulfasalazine versus methotrexate:

See harms of methotrexate.

Sulfasalazine versus antimalarial drugs:

Two of the reviews found that sulfasalazine was associated with a slightly higher risk of withdrawal caused by toxicity and a slightly higher rate of severe toxicity than antimalarial drugs (19 treatment groups, 547 people > 18 years with active rheumatoid arthritis; see comment below). However, the third review found similar rates of withdrawal caused by adverse effects for sulfasalazine compared with hydroxychloroquine (19% with sulfasalazine v 13% with hydroxychloroquine; P = 0.28).

Sulfasalazine versus oral gold:

Two reviews found that sulfasalazine was associated with a slightly higher risk of withdrawal caused by toxicity, and a slightly higher rate of severe toxicity than oral gold (34 treatment groups, 1733 people aged > 18 years with active rheumatoid arthritis; see comment below).

Sulfasalazine versus parenteral gold:

Two reviews found that, compared with parenteral gold, sulfasalazine had a lower risk of withdrawal caused by toxicity and a lower rate of severe toxicity (36 treatment groups, 881 people aged > 18 years with active rheumatoid arthritis; see comment below). In the third review, people taking sulfasalazine were significantly less likely to withdraw because of toxicity than people taking parenteral gold (12% with sulfasalazine v 29% with parenteral gold; P < 0.0001).

Sulfasalazine versus penicillamine:

Two reviews found that sulfasalazine and penicillamine were associated with similar rates of withdrawal caused by toxicity and severe toxicity (30 treatment groups, 816 people aged > 18 years with active rheumatoid arthritis; see comment below). The third review found similar rates of withdrawal caused by adverse effects for sulfasalazine compared with penicillamine (24% with sulfasalazine v 27% with penicillamine; P = 0.62).

Sulfasalazine versus leflunomide:

See harms of leflunomide.

Sulfasalazine in triple combination with methotrexate and hydroxychloroquine:

See harms of methotrexate plus sulfasalazine plus hydroxychloroquine.

Sulfasalazine versus adalimumab, anakinra, azathioprine, ciclosporin, cyclophosphamide, corticosteroids, etanercept, or infliximab:

We found no RCTs.

Comment

Two of the reviews we found had weak methods. The second review was an update of the first with an additional analysis of azathioprine. The reviews assessed tender joint counts and grip strength with a variety of disease-modifying antirheumatic drugs (DMARDs) by extracting within-group data from RCTs and deriving analyses of “treatment groups”. The reviewers then performed indirect comparisons to assess the relative efficacy of the DMARDs. Clearly, these indirect analyses are justified to some extent by the fact that direct comparisons have not been carried out in large-scale RCTs. However, indirect analyses should be treated with caution as they lose the benefit of randomisation; we have therefore reported only limited data from these reviews. The second review found similar results to the first: it updated the information regarding number of people in the treatment groups but presented all outcome data graphically.

Clinical guide:

In practice, methotrexate has become the most widely used DMARD for the treatment of people with active rheumatoid arthritis, because it is generally relatively well tolerated and effective. Methotrexate is widely used as the “gold standard” against which new DMARDs and biological agents are tested. Sulfasalazine is used in practice fairly interchangeably with methotrexate, depending on patient and clinician preference, and many rheumatologists opt for a “step up” approach, whereby they commence treatment with one of methotrexate or sulfasalazine, but add the other agent (with or without hydroxychloroquine) fairly quickly (after 3–6 months) if clinical and laboratory evidence of improvement in disease activity is not obtained. Although leflunomide has strong RCT evidence in comparison with some of the other agents, its arrival was heralded enthusiastically prior to biological agents, and many people who had already “failed” traditional DMARDs were commenced upon it using the rapid loading dose regimen recommended by the manufacturer. In many cases, the rapid loading of leflunomide produced problems with gastrointestinal adverse effects and, since many of these people already had severe long-standing disease, the anticipated benefits in practice did not reach the levels seen in RCTs. Leflunomide is now used more widely, often without rapid loading, and it is likely to show much better tolerability and effectiveness when used earlier in the course of the disease. Gold, azathioprine, ciclosporin, and other DMARDs are being used less as the biological agents have become more widely available.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Antimalarial drugs (second-line treatment)

Summary

DISEASE ACTIVITY Compared with methotrexate: Hydroxychloroquine or chloroquine may be less effective than methotrexate at reducing disease severity after 26 weeks ( low-quality evidence ). Compared with sulfasalazine: Antimalarial drugs may be as effective as sulfasalazine at reducing disease activity after 26–39 weeks ( very low-quality evidence ). Compared with oral gold: Hydroxychloroquine may be as effective as oral gold at reducing disease activity after 1 year (very low-quality evidence). Compared with parenteral gold: Hydroxychloroquine may be less effective than parenteral gold at reducing disease activity (low-quality evidence). Hydroxychloroquine plus methotrexate plus sulfasalazine compared with dual or monotherapy: Triple therapy with hydroxychloroquine plus methotrexate plus sulfasalazine is more effective at reducing disease activity after 2 years compared with hydroxychloroquine plus methotrexate, or methotrexate plus sulfasalazine dual therapy, or compared with methotrexate alone ( high-quality evidence ). ADVERSE EFFECTS Antimalarial drugs are generally well tolerated, but their long-term effects as treatment for rheumatoid arthritis are unknown. NOTE We found no clinically important results about the effects of antimalarial drugs compared with adalimumab, anakinra, azathioprine, ciclosporin, cyclophosphamide, corticosteroids, etanercept, infliximab, or penicillamine.

Benefits

Antimalarial drugs versus methotrexate:

See benefits of methotrexate.

Antimalarial drugs versus sulfasalazine:

See benefits of sulfasalazine.

Hydroxychloroquine versus oral gold:

We found three systematic reviews (search date 1989, search date 1990; search date 2000). We also found one subsequent RCT. The second review was an update of the first, and both had the same weaknesses in their methods (see comment below), so we present only limited data here. The reviews found that hydroxychloroquine and oral gold were associated with similar improvements in measures of disease activity over a mean 26 weeks as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (37 treatment arms, 1814 people aged > 18 years with active rheumatoid arthritis). The third review assessing effects on radiographic progression did not identify RCTs that directly compared hydroxychloroquine versus oral gold; we therefore present only limited data from it here (see comment below). The review (search date 2000, 4 RCTs, 446 people) found no significant difference between hydroxychloroquine or chloroquine and placebo in radiological progression over 12 months, but found that oral gold significantly reduced radiographic evidence of erosions over 24 months compared with placebo. The subsequent RCT (40 people aged 16–75 years with active rheumatoid arthritis) compared hydroxychloroquine 400 mg daily versus oral gold 6 mg daily for 12 months. Participants could take corticosteroids (up to 10 mg prednisone daily or equivalent) during the trial; corticosteroid dose was required to remain stable. Participants who had received hydroxychloroquine in the previous 6 months, or who had received intramuscular gold in the past, were excluded from the trial. In a completer analysis of 23/40 (57%) people, the RCT found that both hydroxychloroquine and oral gold significantly improved disease activity index (index range unclear), Ritchie articular index, tender joint count, swollen joint count, grip strength, pain scores, and duration of morning stiffness from baseline at 12 months (P < 0.05 for all outcomes), with no significant difference between groups (see table 16 ). The RCT did not assess radiological progression.

Table 16
Hydroxychloroquine versus oral gold for second-line treatment.

Hydroxychloroquine versus parenteral gold:

We found three systematic reviews. Two of the reviews found that hydroxychloroquine 200 mg daily was less effective than parenteral gold 50 mg weekly in improving measures of disease activity as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (39 treatment arms, 962 people aged > 18 years with active rheumatoid arthritis; see comment below). The third review assessing effects on radiographic progression did not assess RCTs that directly compared hydroxychloroquine versus parenteral gold; we therefore present only limited data from it here (see comment below). The review found no significant difference between hydroxychloroquine or chloroquine and placebo in radiographic erosions over 12 months. It found that parenteral gold significantly reduced radiographic erosions over 24 months compared with placebo.

Hydroxychloroquine in triple combination with sulfasalazine and methotrexate:

See benefits of methotrexate plus sulfasalazine plus hydroxychloroquine.

Antimalarial drugs versus adalimumab, anakinra, azathioprine, ciclosporin, cyclophosphamide, corticosteroids, etanercept, or infliximab:

We found no systematic review or RCTs.

Harms

Antimalarial drugs versus methotrexate:

See harms of methotrexate.

Antimalarial drugs versus sulfasalazine:

See harms of sulfasalazine.

Hydroxychloroquine versus oral gold:

The reviews found that hydroxychloroquine and oral gold were associated with similar rates of withdrawal caused by toxicity and severe toxicity (37 treatment arms, 1814 people aged > 18 years with active rheumatoid arthritis; see comment below). The third review did not assess adverse effects. In the subsequent RCT, more people taking oral gold than hydroxychloroquine withdrew because of adverse effects, but the RCT did not assess the significance of the difference between groups (8/20 [40%] with oral gold v 3/20 [15%] hydroxychloroquine). Hydroxychloroquine was primarily associated with sore eyes, and oral gold with diarrhoea (absolute results tabulated for each participant).

Hydroxychloroquine versus parenteral gold:

Two reviews found that hydroxychloroquine was associated with lower rates of withdrawal caused by toxicity and severe toxicity than parenteral gold (39 treatment arms, 962 people aged > 18 years with active rheumatoid arthritis; see comment below). The third review did not assess adverse effects.

Hydroxychloroquine in triple combination with sulfasalazine and methotrexate:

See harms of methotrexate plus sulfasalazine plus hydroxychloroquine.

Antimalarial drugs versus adalimumab, anakinra, azathioprine, ciclosporin, cyclophosphamide, corticosteroids, etanercept, or infliximab:

We found no RCTs.

Comment

Two of the reviews we found had weak methods. The second review was an update of the first with an additional analysis of azathioprine. The reviews assessed tender joint counts and grip strength with a variety of disease-modifying antirheumatic drugs by extracting within group data from RCTs and deriving analyses of “treatment groups”. The reviewers then performed indirect comparisons to assess the relative efficacy of the disease-modifying antirheumatic drugs. Clearly, these indirect analyses are justified to some extent by the fact that direct comparisons have not been carried out in large-scale RCTs. However, indirect analyses in reviews should be treated with caution as they lose the benefit of the randomisation in the trials they analyse; we have therefore reported only limited data from these reviews. The second review found similar results; it updated the information regarding number of people in the treatment groups, but presented all outcome data graphically. The third review did not directly compare interventions, but assessed the effect of each drug compared with placebo. This type of analysis can only indirectly reflect the relative efficacy of each drug; analysis across different trials loses the benefit of randomisation. Therefore, results cannot be generalised to suggest that, in equivalent populations, one drug will be more (or less) effective than another.

Clinical guide:

Antimalarial drugs appear modestly effective and are relatively well tolerated in the control of clinical activity of rheumatoid arthritis, but their effects in long-term treatment of rheumatoid arthritis are not established.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Azathioprine (second-line treatment)

Summary

DISEASE ACTIVITY Compared with methotrexate: Azathioprine may be as effective as methotrexate at reducing disease activity after 24–48 weeks ( very low-quality evidence ). Compared with penicillamine: Azathioprine may be as effective at reducing disease activity after 24–26 weeks compared with penicillamine ( low-quality evidence ). Compared with ciclosporin: Azathioprine may be as effective as ciclosporin at reducing disease activity after 6 months (very low-quality evidence). Compared with parenteral gold: Azathioprine may be as effective as parenteral gold at reducing disease activity after 18 months (low-quality evidence). Compared with oral cyclophosphamide: Azathioprine may be as effective as cyclophosphamide at reducing disease activity after 18 months (low-quality evidence). ADVERSE EFFECTS Azathioprine is generally less well tolerated than methotrexate. NOTE We found no clinically important results about the effects of azathioprine compared with adalimumab, anakinra, antimalarial drugs, corticosteroids, etanercept, infliximab or sulfasalazine.

Benefits

Azathioprine versus methotrexate:

See benefits of methotrexate.

Azathioprine versus penicillamine:

See benefits of penicillamine.

Azathioprine versus ciclosporin:

See benefits of ciclosporin.

Azathioprine versus parenteral gold or versus oral cyclophosphamide:

See benefits of parenteral gold.

Azathioprine versus adalimumab, anakinra, antimalarial drugs, corticosteroids, etanercept, infliximab, or sulfasalazine:

We found no systematic review or RCTs.

Harms

Azathioprine versus methotrexate:

See harms of methotrexate.

Azathioprine versus penicillamine:

See harms of penicillamine.

Azathioprine versus ciclosporin:

See harms of ciclosporin.

Azathioprine versus parenteral gold or versus oral cyclophosphamide:

See harms of parenteral gold.

Azathioprine versus adalimumab, anakinra, antimalarial drugs, corticosteroids, etanercept, infliximab, or sulfasalazine:

We found no RCTs.

Comment

Clinical guide:

Azathioprine seems to be modestly effective in the short-term control of disease activity in people with rheumatoid arthritis, but is generally not as well tolerated as methotrexate. RCTs assessing longer-term treatment and effects on radiological progression are needed.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Ciclosporin (second-line treatment)

Summary

DISEASE ACTIVITY Compared with penicillamine: Ciclosporin is as effective as penicillamine at reducing disease activity after 6 months ( moderate-quality evidence ). Compared with azathioprine: Ciclosporin may be as effective as azathioprine at reducing disease activity after 6 months ( very low-quality evidence ). ADVERSE EFFECTS Ciclosporin is associated with nephrotoxicity. Its long-term effects in the treatment of rheumatoid arthritis are unknown. NOTE We found no clinically important results about the effects of ciclosporin compared with adalimumab, anakinra, antimalarial drugs, cyclophosphamide, corticosteroids, etanercept, gold, infliximab, leflunomide, methotrexate, or sulfasalazine.

Benefits

Ciclosporin versus penicillamine:

See benefits of penicillamine.

Ciclosporin versus azathioprine:

We found no systematic review that compared azathioprine with ciclosporin but found two RCTs. The first RCT (52 people aged 30–75 years with active rheumatoid arthritis recalcitrant to at least 1 of gold and penicillamine) compared ciclosporin (maximum 5 mg/kg/day) versus azathioprine (maximum 2 mg/kg/day). Participants remained on stable doses of pre-trial corticosteroids. The RCT found that both ciclosporin and azathioprine significantly improved measures of disease activity from baseline over 6 months (duration of morning stiffness, Ritchie articular index, tender and swollen joint counts, and pain scores; P < 0.05 for change from baseline in both groups for all outcomes) with no significant differences between groups (see table 17 ). It is unclear if the analysis was by intention to treat; 33/52 (63%) people completed the trial. The second RCT (117 people aged 18–75 years with active rheumatoid arthritis recalcitrant to other treatment) compared ciclosporin (maximum 5 mg/kg/day) versus azathioprine (maximum 2 mg/kg/day) over 6 months. Participants could take corticosteroids (up to 12.5 mg prednisone daily or equivalent) during the trial; dose was required to remain stable. The RCT found similar improvements in measures of disease activity in both groups, as assessed by Ritchie articular index, duration of morning stiffness, and grip strength, with no significant difference between groups (see table 17 ). There was also no significant difference between groups in swollen joint count (reported as not significant, P value not reported, absolute results presented graphically). It is unclear if the analysis was by intention to treat; 93/117 (79%) people completed the trial.

Table 17
Ciclosporin versus azathioprine for second-line treatment.

Ciclosporin versus adalimumab, anakinra, antimalarial drugs, cyclophosphamide, corticosteroids, etanercept, gold, infliximab, leflunomide, methotrexate, or sulfasalazine:

We found no systematic review or RCTs.

Harms

Ciclosporin versus penicillamine:

See harms of penicillamine.

Ciclosporin versus azathioprine:

The first RCT found that gastrointestinal upsets (nausea, vomiting, gastric pain) were the most frequently reported adverse effect causing withdrawal in both groups (3/25 [12%] with ciclosporin v 6/27 [22%] with azathioprine; significance not reported). The second RCT found that nine people (about 15%) in each group withdrew because of adverse effects (significance not reported). More people taking ciclosporin than azathioprine had adverse effects (not leading to withdrawal), including hypertrichosis with ciclosporin and gastrointestinal upset with both drugs.

Ciclosporin versus adalimumab, anakinra, antimalarial drugs, cyclophosphamide, corticosteroids, etanercept, gold, infliximab, leflunomide, methotrexate, or sulfasalazine:

We found no RCTs.

Comment

Clinical guide:

Ciclosporin offers short-term control of rheumatoid arthritis, but is associated with nephrotoxicity. The effects of ciclosporin for long-term treatment of rheumatoid arthritis are not established.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Etanercept (second-line treatment)

Summary

DISEASE ACTIVITY Compared with methotrexate: Etanercept is as effective as methotrexate at reducing disease activity after 12 months ( high-quality evidence ). Compared with methotrexate plus etanercept: Etanercept alone is less effective at reducing disease activity after 52 weeks compared with etanercept plus methotrexate ( moderate-quality evidence ). Etanercept plus methotrexate compared with methotrexate alone: Etanercept plus methotrexate is more effective at reducing disease activity after 52 weeks compared with methotrexate alone (moderate-quality evidence). ADVERSE EFFECTS Etanercept has been associated with an increased risk of infections, including reactivation of tuberculosis. Its long-term safety is unknown. NOTE We found no clinically important results about the effects of etanercept compared with antimalarial drugs, adalimumab, anakinra, ciclosporin, corticosteroids, cyclophosphamide, infliximab, penicillamine, leflunomide, or sulfasalazine.

Benefits

Etanercept versus methotrexate:

We found one systematic review (search date 2003), which identified one RCT (632 people with active rheumatoid arthritis, disease duration of < 3 years) comparing etanercept 10 or 25 mg twice weekly versus methotrexate 7.5 mg once weekly for 12 months. Participants could take corticosteroids (up to 10 mg prednisone daily or equivalent) during the trial; corticosteroid dose was required to remain stable. The RCT found no significant difference between etanercept 25 mg twice weekly and methotrexate in the proportion of people who achieved American College of Rheumatology (ACR)20, ACR 50, or ACR70 at 12 months (see table 18 ; intention to treat analysis). However, it found that etanercept 25 mg twice daily significantly reduced radiological progression at 12 months (see table 18 ).

Table 18
Etanercept versus methotrexate for second-line treatment.

Etanercept plus methotrexate versus either drug alone:

We found no systematic review but found one subsequent RCT. The subsequent RCT (686 people with active rheumatoid arthritis recalcitrant to at least 1 disease-modifying antirheumatic drug, disease duration 6 months to 20 years) compared three interventions: etanercept (25 mg twice weekly), methotrexate (< 20 mg weekly), or etanercept plus methotrexate. It is unclear whether participants received corticosteroids; all participants received folic acid. In an intention-to-treat analysis, the RCT found that etanercept plus methotrexate significantly increased the proportion of people achieving ACR20, ACR50, and ACR70 at week 52 compared with either drug alone (see table 19 ). After 52 weeks, etanercept plus methotrexate also reduced Disease Activity Scores and radiological progression.

Table 19
Etanercept plus methotrexate versus either drug alone for second-line treatment.

Etanercept versus antimalarial drugs, adalimumab, anakinra, ciclosporin, corticosteroids, cyclophosphamide, infliximab, penicillamine, leflunomide, or sulfasalazine:

We found no systematic review or RCTs.

Harms

Etanercept versus methotrexate:

The most common adverse effect was injection site reactions, which occurred in significantly more people taking etanercept than methotrexate (77/207 [37%] with etanercept v 16/217 [7%] with methotrexate; RR 5.04, 95% CI 3.05 to 8.35). Both etanercept and methotrexate were associated with diarrhoea and headache, with no significant difference between groups.

Etanercept plus methotrexate versus either drug alone:

The RCT found no significant difference between etanercept plus methotrexate and either drug alone in overall adverse effects (187/231 [81%] with etanercept plus methotrexate v 192/223 [86%] with etanercept alone v 185/228 [81%] with methotrexate alone; reported as not significant for etanercept plus methotrexate v either drug alone, P value not reported).

Etanercept versus antimalarial drugs, adalimumab, anakinra, ciclosporin, corticosteroids, cyclophosphamide, infliximab, penicillamine, leflunomide, or sulfasalazine:

We found no RCTs.

Drug safety alert

FDA issues a drug safety alert on the risk of opportunistic fungal infections associated with TNF-alpha blockers (tumour necrosis factor alpha-blockers). Some of these invasive fungal infections could be fatal (4 September 2008). FDA issues drug safety alert on the increased risk of lymphoma and other malignancies in children and adolescents, and the risks of leukaemia and new onset psoriasis, associated with TNF blockers (04 August 2009).

The FDA has issued a drug safety alert on the risk of opportunistic fungal infections associated with TNF-alpha blockers (tumour necrosis factor alpha-blockers). Some of these invasive fungal infections could be fatal (http://www.fda.gov). A drug safety alert has been issued on the increased risk of lymphoma and other malignancies in children and adolescents, and the risks of leukaemia and new onset psoriasis, associated with TNF blockers (http://www.fda.gov).

Comment

Clinical guide:

RCTs show equivalence between etanercept and methotrexate in improving measures of disease activity and reducing radiological progression, and show increased improvements with combined treatment compared with either drug alone. However, there are concerns with increased incidence of infections (particularly tuberculosis), and the long-term safety and efficacy of etanercept still needs to be established.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Gold (oral) (second-line treatment)

Summary

DISEASE ACTIVITY Compared with methotrexate: Oral gold may be less effective than methotrexate at reducing disease activity after 26 weeks ( low-quality evidence ). Compared with sulfasalazine: Oral gold may be less effective than sulfasalazine at reducing disease activity (low-quality evidence). Compared with antimalarial drugs: Oral gold may be as effective as hydroxychloroquine at reducing disease activity after 1 year ( very low-quality evidence ). Compared with penicillamine: Oral gold may be less effective than penicillamine at reducing disease activity (low-quality evidence). ADVERSE EFFECTS Oral gold may be better tolerated than methotrexate. NOTE We found no clinically important results about the effects of oral gold compared with azathioprine, cyclophosphamide, adalimumab, anakinra, ciclosporin, corticosteroids, etanercept, or infliximab.

Benefits

Oral gold versus methotrexate:

See benefits of methotrexate.

Oral gold versus sulfasalazine:

See benefits of sulfasalazine.

Oral gold versus antimalarial drugs:

See benefits of antimalarial drugs.

Oral gold versus penicillamine:

We found two systematic reviews (search date 1989, search date 1990). The second review was an update of the first, and both had the same weaknesses in their methods (see comment below), so we present only limited data here. The reviews found that oral gold was less effective than penicillamine in improving measures of disease activity as assessed by a composite score including tender joint count or Ritchie articular index, and grip strength (48 treatment arms, 2083 people aged > 18 years with active rheumatoid arthritis).

Oral gold versus azathioprine or cyclophosphamide:

We found no systematic review or RCTs.

Oral gold versus adalimumab, azathioprine, anakinra, ciclosporin, corticosteroids, cyclophosphamide, etanercept, or infliximab:

We found no systematic review or RCTs.

Harms

Oral gold versus methotrexate:

See harms of methotrexate.

Oral gold versus sulfasalazine:

See harms of sulfasalazine.

Oral gold versus antimalarial drugs:

See harms of antimalarial drugs.

Oral gold versus penicillamine:

Two reviews found that oral gold and penicillamine were associated with similar rates of withdrawal caused by toxicity and severe toxicity (48 treatment arms, 2083 people aged > 18 years with active rheumatoid arthritis; see comment below).

Oral gold versus adalimumab, azathioprine, anakinra, ciclosporin, corticosteroids, cyclophosphamide, etanercept, or infliximab:

We found no RCTs.

Comment

The two reviews had weak methods. The second review was an update of the first with an additional analysis of azathioprine. The reviews assessed tender joint counts and grip strength with a variety of disease-modifying antirheumatic drugs (DMARDs) by extracting within group data from RCTs and deriving analyses of “treatment groups”. The reviewers then performed indirect comparisons to assess the relative efficacy of the DMARDs. Clearly, these indirect analyses are justified to some extent by the fact that direct comparisons have not been carried out in large-scale RCTs. However, indirect analyses in reviews should be treated with caution as they lose the benefit of the randomisation in the trials they analyse; we have therefore reported only limited data from these reviews. The second review found similar results to the first review; it updated the information regarding number of people in the treatment groups but presented all outcome data graphically.

Clinical guide:

Oral gold is less effective than parenteral gold at controlling clinical activity of rheumatoid arthritis. However, parenteral gold leads to higher levels of toxicity than most of the other commonly used DMARDs. Long-term data about the effectiveness of gold salts are not currently available.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Gold (parenteral) (second-line treatment)

Summary

DISEASE ACTIVITY Compared with methotrexate: Parenteral gold may be as effective as methotrexate at reducing disease activity after 6–12 months ( low-quality evidence ). Compared with sulfasalazine: Parenteral gold may be as effective as sulfasalazine at reducing disease activity ( very low-quality evidence ). Compared with antimalarial drugs: Parenteral gold may be more effective than hydroxychloroquine at reducing disease activity (low-quality evidence). Compared with azathioprine: Parenteral gold may be as effective as azathioprine at reducing disease activity after 18 months (low-quality evidence). Compared with cyclophosphamide: Parenteral gold may be as effective as cyclophosphamide at reducing disease activity after 18 months ( low-quality evidence ). ADVERSE EFFECTS Parenteral gold is associated with higher levels of toxicity than most other commonly used disease-modifying antirheumatic drugs. NOTE We found no clinically important results about the effects of parenteral gold compared with adalimumab, anakinra, ciclosporin, corticosteroids, etanercept, infliximab, or penicillamine.

Benefits

Parenteral gold versus methotrexate:

See benefits of methotrexate.

Parenteral gold versus sulfasalazine:

See benefits of sulfasalazine.

Parenteral gold versus antimalarial drugs:

See benefits of antimalarial drugs.

Parenteral gold versus azathioprine or cyclophosphamide:

We found one RCT (121 people aged > 18 years with evidence of radiographic erosions and active rheumatoid arthritis) that compared three interventions: parenteral gold (50 mg/week to a total dose of 1 g and then 50 mg/month), oral azathioprine 2.5 mg/kg daily, or oral cyclophosphamide 1.5 mg/kg daily. Participants taking corticosteroids were included in the trial if use had been stable for past 3 months; similar numbers in each group took corticosteroids (5/38 [13%] with im gold, 6/44 [14%] with azathioprine, 5/39 [13%] with cyclophosphamide). Over 18 months, the RCT found no significant difference among groups in tender joint count, pain scores, duration of morning stiffness, and grip strength (reported as not significant, no further data reported). However, people taking gold had greater radiographic deterioration at 12–18 months than people taking azathioprine or cyclophosphamide (proportion of people with score of –2 on a score from 0 to 4 [score not specified]: 6/38 [16%] with im gold v 1/44 [2%] with azathioprine v 0/39 [0%] with cyclophosphamide; P < 0.05 for gold v azathioprine; P < 0.01 for gold v cyclophosphamide).

Parenteral gold versus adalimumab, anakinra, ciclosporin, corticosteroids, etanercept, infliximab, or penicillamine:

We found no systematic review or RCTs.

Harms

Parenteral gold versus methotrexate:

See harms of methotrexate.

Parenteral gold versus sulfasalazine:

See harms of sulfasalazine.

Parenteral gold versus antimalarial drugs:

See harms of antimalarial drugs.

Parenteral gold versus azathioprine or cyclophosphamide:

The RCT found that fewer people taking azathioprine than either oral cyclophosphamide or parenteral gold withdrew because of adverse effects. It found similar rates of withdrawal caused by adverse effects between parenteral gold and cyclophosphamide (withdrawals: 19/38 [50%] with im gold v 14/44 [32%] with azathioprine v 20/39 [51%] with cyclophosphamide; significance of difference between groups not assessed).

Parenteral gold versus adalimumab, anakinra, ciclosporin, corticosteroids, etanercept, infliximab, or penicillamine:

We found no RCTs.

Comment

Two of the reviews we found had weak methods. The second review was an update of the first with an additional analysis of azathioprine. The reviews assessed tender joint counts and grip strength with a variety of disease-modifying antirheumatic drugs (DMARDs) by extracting within-group data from RCTs and deriving analyses of “treatment groups”. The reviewers then performed indirect comparisons to assess the relative efficacy of the DMARDs. Clearly, these indirect analyses are justified to some extent by the fact that direct comparisons have not been carried out in large-scale RCTs. However, indirect analyses in reviews should be treated with caution as they lose the benefit of the randomisation in the trials they analyse; we have therefore reported only limited data from these reviews. The second review found similar results: it updated the information regarding number of people in the treatment groups but presented all outcome data graphically.

Clinical guide:

Parenteral gold is more effective than oral gold at controlling clinical activity of rheumatoid arthritis. However, parenteral gold leads to higher levels of toxicity than most of the other commonly used DMARDs. Long-term data as to the effects of gold salts are not currently available.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Adalimumab (second-line treatment)

Summary

DISEASE ACTIVITY Compared with placebo: Adalimumab is no more effective than placebo at reducing disease activity ( high-quality evidence ). ADVERSE EFFECTS Short-term toxicity with adalimumab is relatively low, but its long-term safety is unknown. NOTE We found no clinically important results about the effects of adalimumab compared with disease-modifying antirheumatic drugs or compared with corticosteroids in people who have not responded to or who are intolerant of first-line disease-modifying antirheumatic drug treatment.

Benefits

We found no systematic review or RCTs comparing adalimumab versus disease-modifying antirheumatic drugs (DMARDs; anakinra, antimalarial drugs, ciclosporin, cyclophosphamide, etanercept, infliximab, leflunomide, methotrexate, penicillamine, or sulfasalazine) or versus corticosteroids.

Harms

We found no RCTs.

Drug safety alert

MHRA issues drug safety alert on risk of hepatosplenic T-cell lymphoma associated with adalimumab (16 July 2008). FDA issues a drug safety alert on the risk of opportunistic fungal infections associated with TNF-alpha blockers (tumour necrosis factor alpha-blockers). Some of these invasive fungal infections could be fatal (4 September 2008). FDA issues drug safety alert on the increased risk of lymphoma and other malignancies in children and adolescents, and the risks of leukaemia and new onset psoriasis, associated with TNF blockers (04 August 2009).

A drug safety alert has been issued on the risk of hepatosplenic T-cell lymphoma associated with adalimumab (http://www.mhra.gov.uk). The FDA has issued a drug safety alert on the risk of opportunistic fungal infections associated with TNF-alpha blockers (tumour necrosis factor alpha-blockers). Some of these invasive fungal infections could be fatal (http://www.fda.gov). A drug safety alert has been issued on the increased risk of lymphoma and other malignancies in children and adolescents, and the risks of leukaemia and new onset psoriasis, associated with TNF blockers (http://www.fda.gov).

Comment

Clinical guide:

RCTs found that adalimumab was effective in terms of symptom control, and reduced radiographic progression compared with placebo both as monotherapy and in combination with DMARDs. Short-term toxicity was relatively low, but long-term safety less clear. We found no RCT evidence on which to base the decision about whether adalimumab is as effective as other DMARDs. At present in the UK, National Institute for Health and Clinical Excellence (NICE) guidelines restrict the use of adalimumab and other biological agents until two DMARDs (which must include methotrexate) have been tried and failed, and continued disease activity must also be demonstrated by high Disease Activity Scores on two occasions at least 4 weeks apart.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Anakinra (second-line treatment)

Summary

DISEASE ACTIVITY Compared with placebo: Anakinra reduces disease activity compared with placebo ( high-quality evidence ). ADVERSE EFFECTS Anakinra has not been associated with serious adverse effects, but its long-term safety is unknown. NOTE We found no clinically important results about the effects of anakinra compared with disease-modifying antirheumatic drugs or compared with corticosteroids in people who have not responded to or are intolerant of first-line disease-modifying antirheumatic drug treatment.

Benefits

We found no systematic review or RCTs comparing anakinra versus disease-modifying antirheumatic drugs (DMARDs; adalimumab, antimalarial drugs, ciclosporin, cyclophosphamide, etanercept, infliximab, leflunomide, methotrexate, penicillamine, or sulfasalazine) or versus corticosteroids.

Harms

We found no RCTs.

Comment

Clinical guide:

RCTs found that anakinra alone or in combination with methotrexate improved disease activity and slowed radiological progression compared with placebo. Injection site reactions were the most common adverse effect. Safety studies assessing anakinra in combination with another DMARD suggest that it is not associated with serious adverse effects. However, we found no RCT evidence on which to base the decision about whether anakinra is as effective as other DMARDs. In practice in the UK, anakinra is started as a fourth-line treatment, after failure of response, or intolerance of, at least two DMARDs, one of which must be methotrexate, and at least one of the other biological agents.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Corticosteroids (second-line treatment)

Summary

DISEASE ACTIVITY High-dose corticosteroids compared with lower-dose corticosteroids: The effects of higher-dose parenteral corticosteroids (1 g) compared with lower-dose parenteral corticosteroids (40 mg, 100 mg, 250 mg or 500 mg) are unclear ( very low-quality evidence ). ADVERSE EFFECTS We don't know how the long-term risks of corticosteroid treatment of rheumatoid arthritis compare with their possible benefits. NOTE We found no clinically important results about the effects of corticosteroids compared with adalimumab, anakinra, antimalarial drugs, ciclosporin, cyclophosphamide, etanercept, infliximab, leflunomide, methotrexate, penicillamine, or sulfasalazine.

Benefits

Corticosteroids versus adalimumab, anakinra, antimalarial drugs, ciclosporin, cyclophosphamide, etanercept, infliximab, leflunomide, methotrexate, penicillamine, or sulfasalazine:

We found one systematic review (search date 1994), which compared corticosteroids versus other “second-line treatments commonly used for the treatment of rheumatoid arthritis”. However, the review had weak methods (see comment below), and did not perform separate analyses for corticosteroids versus each disease-modifying antirheumatic drug (DMARD). We therefore report no data from it here.

Dose finding studies: intravenous methylprednisolone:

We found no systematic review but found three RCTs comparing different doses of intravenous methylprednisolone. Two RCTs (73 people aged > 18 years with active rheumatoid arthritis) compared intravenous methylprednisolone (100 mg single infusion in 1 RCT, 250 mg single infusion in the other), versus intravenous methylprednisolone 1 g single infusion, and found no significant difference in Patient Global Assessment, Ritchie articular index, tender or swollen joint counts, pain scores, or duration of morning stiffness at 6 or 12 weeks between groups (see table 20 ). The third RCT (71 people aged 22–69 years with active rheumatoid arthritis) compared three doses of intravenous methylprednisolone: 40 mg, 500 mg, and 1 g each administered as a single pulse during the course of a 24-hour inpatient admission. It found that significantly more people taking methylprednisolone 1 g than other doses had improved at 6 weeks as assessed by Patient Global Assessment and Physician Global Assessment (Patient Global Assessment [proportion who felt treatment was worthwhile]: 10/24 [42%] with 40 mg v 13/22 [59%] with 500 mg v 20/23 [87%] with 1 g; P < 0.001 for 1 g v other doses; Physician Global Assessment [proportion in whom clinician felt treatment was beneficial]: 8/24 [33%] with 40 mg v 6/22 [27%] with 500 mg v 17/23 [74%] with 1 g; P < 0.005 for 1 g v other doses). Results for effects on specific symptoms were not reported. These results should be interpreted with caution as there were high withdrawal rates from the trial at 3 weeks; significantly more people receiving methylprednisolone 40 mg and 500 mg withdrew because of any cause (12/24 [50%] with 40 mg v 11/22 [50%] with 500 mg v 3/22 [14%] with 1 g; P < 0.05 for 1 g v other doses).

Table 20
Methylprednisolone 100 or 250 mg versus 1 g for second-line treatment.

Harms

Corticosteroids versus adalimumab, anakinra, antimalarial drugs, ciclosporin, cyclophosphamide, etanercept, infliximab, leflunomide, methotrexate, penicillamine, or sulfasalazine:

We found no RCTs.

Dose-finding studies: intravenous methylprednisolone:

The first and third RCTs found similar rates of adverse effects with different doses of intravenous pulsed methylprednisolone, including transient significant elevation of white blood cell count, and mild facial flushing. The second RCT reported “no clinically evident adverse effects” with either dose of methylprednisolone either at 3 days or 3 months.

Comment

The review assessed clinical outcomes such as tender joint counts and grip strength for corticosteroids and a variety of DMARDs by extracting within-group data from RCTs and deriving analyses of “treatment groups”. The reviewers then performed indirect comparisons to assess the relative efficacy of corticosteroids and a variety of DMARDs. Indirect analyses in reviews should be treated with caution as they lose the benefit of the randomisation in the trials they analyse.

Clinical guide:

Corticosteroids (oral or parenteral) offer short-term relief of clinical disease activity in people with rheumatoid arthritis, but the longer-term risk compared with benefit analyses in rheumatoid arthritis have not been evaluated.

Substantive changes

2007; 2007: 1124.
Published online 2007 August 1.

Cyclophosphamide (second-line treatment)

Summary

DISEASE ACTIVITY Compared with parenteral gold: Cyclophosphamide may be as effective as parenteral gold at reducing disease activity after 18 months ( low-quality evidence ). Compared with azathioprine: Cyclophosphamide may be as effective as azathioprine at reducing disease activity after 18 months (low-quality evidence). ADVERSE EFFECTS Cyclophosphamide has been associated with similar rates of adverse effects as parenteral gold, and higher rates of adverse effects compared with azathioprine. NOTE We found no clinically important results about the effects of cyclophosphamide compared with adalimumab, anakinra, antimalarial drugs, ciclosporin, corticosteroids, etanercept, infliximab, leflunomide, methotrexate, penicillamine, or sulfasalazine.

Benefits

Cyclophosphamide versus parenteral gold or oral azathioprine:

See benefits of parenteral gold.

Cyclophosphamide versus adalimumab, anakinra, antimalarial drugs, ciclosporin, corticosteroids, etanercept, infliximab, leflunomide, methotrexate, penicillamine, or sulfasalazine:

We found no systematic review or RCTs.

Harms

Cyclophosphamide versus parenteral gold or oral azathioprine:

See harms of parenteral gold.

Cyclophosphamide versus adalimumab, anakinra, antimalarial drugs, ciclosporin, corticosteroids, etanercept, infliximab, leflunomide, methotrexate, penicillamine, or sulfasalazine:

We found no RCTs.

Comment

Clinical guide:

In clinical practice, cyclophosphamide has traditionally been reserved for the treatment of the severe, life-threatening complications of rheumatoid arthritis, such as vasculitis or interstitial lung disease. Given the relative rarity of these complications, and their acute presentation, RCTs are difficult to perform. Cyclophosphamide has shown benefit in the treatment of these complications in small RCTs of people with primary vasculitis or interstitial lung disease, and for these reasons is currently widely used in clinical practice. It remains to be seen whether more aggressive early disease-modifying antirheumatic drugs and biological treatments will reduce the long-term risks of these severe life-threatening complications.

Substantive changes


Articles from BMJ Clinical Evidence are provided here courtesy of BMJ Publishing Group