Sulfasalazine for ankylosing spondylitis (Review) Chen J, Lin S, Liu C

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2014, Issue 11 http://www.thecochranelibrary.com

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

TABLE OF CONTENTS

HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SUMMARY OF FINDINGS FOR THE MAIN COMPARISON . . . . . . . . . . . . . . . . . . . BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.1. Comparison 1 Sulfasalazine versus placebo, Outcome 1 Spondylitis function index (Score 0 to 40, 0 to 44, 0 = the best). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.2. Comparison 1 Sulfasalazine versus placebo, Outcome 2 Spondylitis function index (2nd analysis) (score 0 to 40, 0 to 44, 0 = the best). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.3. Comparison 1 Sulfasalazine versus placebo, Outcome 3 Improvement in back pain. . . . . . . . Analysis 1.4. Comparison 1 Sulfasalazine versus placebo, Outcome 4 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.5. Comparison 1 Sulfasalazine versus placebo, Outcome 5 Back pain (2nd analysis) (VAS 100 mm, 0 = no pain, 100 = severe pain). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.6. Comparison 1 Sulfasalazine versus placebo, Outcome 6 Night pain (% no pain). . . . . . . . . Analysis 1.7. Comparison 1 Sulfasalazine versus placebo, Outcome 7 Score of sleep disturbance (end point) (0 to 4, 0 = no disturbance, 4 = severe disturbance). . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.8. Comparison 1 Sulfasalazine versus placebo, Outcome 8 Frequency of nocturnal awakening (change from baseline). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.9. Comparison 1 Sulfasalazine versus placebo, Outcome 9 Score of daily NSAIDs (change from baseline, usual dosage as 10). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.10. Comparison 1 Sulfasalazine versus placebo, Outcome 10 Reducing or stopping NSAIDs. . . . . . Analysis 1.11. Comparison 1 Sulfasalazine versus placebo, Outcome 11 Chest expansion (cm). . . . . . . . . Analysis 1.12. Comparison 1 Sulfasalazine versus placebo, Outcome 12 Chest expansion (2nd analysis) (cm). . . . Analysis 1.13. Comparison 1 Sulfasalazine versus placebo, Outcome 13 Forced vital volume (change from baseline) (L/min). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.14. Comparison 1 Sulfasalazine versus placebo, Outcome 14 (Modified) Schober’s test (cm). . . . . . Analysis 1.15. Comparison 1 Sulfasalazine versus placebo, Outcome 15 (Modified) Schober’s test (2nd analysis) (cm). Analysis 1.16. Comparison 1 Sulfasalazine versus placebo, Outcome 16 Occiput-to-wall test (cm). . . . . . . . Analysis 1.17. Comparison 1 Sulfasalazine versus placebo, Outcome 17 Occiput-to-wall test (2nd analysis) (cm). . . Analysis 1.18. Comparison 1 Sulfasalazine versus placebo, Outcome 18 Fingers-to-floor test (cm). . . . . . . . Analysis 1.19. Comparison 1 Sulfasalazine versus placebo, Outcome 19 Fingers-to-floor test (2nd analysis) (cm). . . Analysis 1.20. Comparison 1 Sulfasalazine versus placebo, Outcome 20 Chin sternum distance (change from baseline) (cm). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.21. Comparison 1 Sulfasalazine versus placebo, Outcome 21 Joint pain/tenderness score (0 to 198, the higher the score the more severe the disease) or number. . . . . . . . . . . . . . . . . . . . . . . Analysis 1.22. Comparison 1 Sulfasalazine versus placebo, Outcome 22 Joint pain/tenderness score (2nd analysis) (0 to 198, the higher the score the more severe the disease). . . . . . . . . . . . . . . . . . . . . Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

1 1 2 4 7 7 7 9 10 11 12 15 16 16 17 19 41 45 46 47 47 48 49 50 50 51 51 52 53 54 54 56 57 58 59 60 61 61 62 i

Analysis 1.23. Comparison 1 Sulfasalazine versus placebo, Outcome 23 Joint swelling score (0 to 198, the higher score the more severe the disease) or number. . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.24. Comparison 1 Sulfasalazine versus placebo, Outcome 24 Joint swelling score (2nd analysis) (0 to 198, the higher the score the more severe the disease). . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.25. Comparison 1 Sulfasalazine versus placebo, Outcome 25 Dactylitis score (0 to 3, 0 = normal, 3 = severe). Analysis 1.26. Comparison 1 Sulfasalazine versus placebo, Outcome 26 Dactylitis score (2nd analysis) (0 to 3, 0 = normal, 3 = severe). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.27. Comparison 1 Sulfasalazine versus placebo, Outcome 27 Enthesopathy index (0 to 90, 0 to 66, 0 to 90, the higher the score the more severe the disease). . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.28. Comparison 1 Sulfasalazine versus placebo, Outcome 28 Enthesopathy index (2nd analysis) (0 to 90, 0 to 66, 0 to 90, the higher score the more severe the disease). . . . . . . . . . . . . . . . . . . . Analysis 1.29. Comparison 1 Sulfasalazine versus placebo, Outcome 29 Spondylitis articular index (0 to 90, the higher score the more severe the disease). . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.30. Comparison 1 Sulfasalazine versus placebo, Outcome 30 Spondylitis articular index (2nd analysis) (0 to 90, the higher score the more severe the disease). . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.31. Comparison 1 Sulfasalazine versus placebo, Outcome 31 Improvement in patient global assessment. . Analysis 1.32. Comparison 1 Sulfasalazine versus placebo, Outcome 32 Patient assessment of disease severity (end point) (VAS 100 mm, 0 = very good, 100 = very poor). . . . . . . . . . . . . . . . . . . . . . . Analysis 1.33. Comparison 1 Sulfasalazine versus placebo, Outcome 33 General well-being (end point) (VAS 100 mm, 0 = very good, 100 = very poor). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.34. Comparison 1 Sulfasalazine versus placebo, Outcome 34 Improvement in physician global assessment. Analysis 1.35. Comparison 1 Sulfasalazine versus placebo, Outcome 35 Response to treatment (based on both patient and physician assessment). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.36. Comparison 1 Sulfasalazine versus placebo, Outcome 36 Duration of morning stiffness (hr). . . . . Analysis 1.37. Comparison 1 Sulfasalazine versus placebo, Outcome 37 Duration of morning stiffness (2nd analysis) (hr). Analysis 1.38. Comparison 1 Sulfasalazine versus placebo, Outcome 38 Morning stiffness (end point) (VAS 100 mm, 0 = no stiffness, 100 = severe stiffness). . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.39. Comparison 1 Sulfasalazine versus placebo, Outcome 39 Improvement in morning stiffness. . . . . Analysis 1.40. Comparison 1 Sulfasalazine versus placebo, Outcome 40 Erythrocyte sedimentation rate (mm/hr). . Analysis 1.41. Comparison 1 Sulfasalazine versus placebo, Outcome 41 Erythrocyte sedimentation rate (2nd analysis) (mm/hr). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.42. Comparison 1 Sulfasalazine versus placebo, Outcome 42 C-reactive protein (ug/ml). . . . . . . Analysis 1.43. Comparison 1 Sulfasalazine versus placebo, Outcome 43 C-reactive protein (2nd analysis) (ug/ml). . Analysis 1.44. Comparison 1 Sulfasalazine versus placebo, Outcome 44 Withdrawal due to side effect. . . . . . Analysis 1.45. Comparison 1 Sulfasalazine versus placebo, Outcome 45 Withdrawal due to ineffectiveness. . . . . Analysis 1.46. Comparison 1 Sulfasalazine versus placebo, Outcome 46 Drop-out for any reason. . . . . . . . Analysis 1.47. Comparison 1 Sulfasalazine versus placebo, Outcome 47 Serious adverse events. . . . . . . . . Analysis 2.1. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 1 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain). . . . . . . . . . . . . . . Analysis 2.2. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 2 Score of sleep disturbance (0 to 4, 0 = no disturbance, 4 = severe disturbance). . . . . . . . Analysis 2.3. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 3 Chest expansion (cm). . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 2.4. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 4 Schober’s test (cm). . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 2.5. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 5 Fingers-to-floor test (cm). . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 2.6. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 6 Articular index (0 to 66, the higher the score the more severe the disease). . . . . . . . . . Analysis 2.7. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 7 Degree of joint swelling (0 to 66, the higher the score the more severe the disease). . . . . . . Analysis 2.8. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 8 Patient assessment of disease severity (VAS 100 mm, 0 = very good, 100 = very poor). . . . . . Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

63 64 65 65 66 67 67 68 68 69 69 70 70 71 72 73 74 74 76 77 78 79 80 81 82 82 83 83 84 84 85 85 86 ii

Analysis 2.9. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 9 Duration of morning stiffness (hr). . . . . . . . . . . . . . . . . . . . . . . Analysis 2.10. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 10 Erythrocyte sedimentation rate (mm/hr). . . . . . . . . . . . . . . . . . Analysis 3.1. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 1 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain). . . . . . . . . . . . . . . . . . . Analysis 3.2. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 2 Score of sleep disturbance (0 to 4, 0 = no disturbance, 4 = severe disturbance). . . . . . . . . . . . Analysis 3.3. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 3 Chest expansion (cm). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.4. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 4 Schober’s test (cm). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.5. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 5 Fingers-to-floor test (cm). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.6. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 6 Patient assessment of disease severity (VAS 100 mm, 0 = very good, 100 = very poor). . . . . . . . . . Analysis 3.7. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 7 Duration of morning stiffness (hr). . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.8. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 8 Erythrocyte sedimentation rate (mm/hr). . . . . . . . . . . . . . . . . . . . . . . . . ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . . INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

86 87 87 88 88 89 89 90 90 91 91 102 110 111 111 111 111 112 112

iii

[Intervention Review]

Sulfasalazine for ankylosing spondylitis Junmin Chen1 , Shaopeng Lin1 , Chao Liu2 1 Department of Hematology and Rheumatology, The First Affiliated Hospital of Fujian Medical University, Fuzhou, China. 2 Division of Cardiology, The First Hospital of Hebei Medical University, Shijiazhuang, China

Contact address: Junmin Chen, Department of Hematology and Rheumatology, The First Affiliated Hospital of Fujian Medical University, #20 Chazhong Road, Fuzhou, Fujian Province, 350005, China. [email protected]. Editorial group: Cochrane Musculoskeletal Group. Publication status and date: New search for studies and content updated (no change to conclusions), published in Issue 11, 2014. Review content assessed as up-to-date: 28 November 2013. Citation: Chen J, Lin S, Liu C. Sulfasalazine for ankylosing spondylitis. Cochrane Database of Systematic Reviews 2014, Issue 11. Art. No.: CD004800. DOI: 10.1002/14651858.CD004800.pub3. Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

ABSTRACT Background Ankylosing spondylitis (AS) is a chronic inflammatory disease of unknown cause and affects mainly the spine, but can also affect other joints. Disease progression may result in loss of mobility and function. Sulfasalazine is a disease-modifying antirheumatic drug used in the treatment of AS. However, its efficacy remains unclear. This is an update of a Cochrane review first published in 2005. Objectives To evaluate the benefits and harms of sulfasalazine for the treatment of ankylosing spondylitis (AS). Search methods We searched for relevant randomized and quasi-randomized trials in any language, using the following sources: the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2013, Issue 11); MEDLINE (2003 to 28 November 2013); EMBASE (2003 to 27 November 2013); CINAHL (2003 to 28 November 2013); Ovid MEDLINE data, World Health Organization International Clinical Trials Registry Platform (28 November 2013); and the reference sections of retrieved articles. Selection criteria We evaluated randomized and quasi-randomized trials examining the benefits and harms of sulfasalazine on AS. Data collection and analysis Two review authors independently reviewed unblinded trial reports according to the selection criteria. Disagreements on the inclusion of the studies were resolved, when necessary, by recourse to a third review author. The same authors independently assessed the risk of bias of included trials and entered the data extracted from the included trials. We combined results using mean difference (MD) or standardised mean difference (SMD) for continuous data, and risk ratio (RR) for dichotomous data. We restructured outcome measures for this update based on recommendations from the editorial group. Major outcomes included: pain, Bath ankylosing spondylitis disease activity index (BASDAI), Bath ankylosing spondylitis function index (BASFI), Bath ankylosing spondylitis metrology index (BASMI), radiographic progression, total number of withdrawals due to adverse events, and serious adverse events. Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

1

Main results We did not add any new studies to this review following the updated search. In the original review, we included 11 studies in the analysis, involving 895 participants in total. All included studies compared sulfasalazine with placebo. We judged most of the studies as low risk of bias or unclear risk of bias in five domains (random sequence generation, allocation concealment, blinding of outcome assessment, selective reporting, and other sources of bias). However, for incomplete outcome data, we only judged one trial at low risk of bias. None of the included trials assessed BASDAI, BASFI, BASMI or radiographic progression. Different parameters were used to assess pain. The pooled MD for back pain measured on a 0 to 100 mm visual analogue scale was -2.96 (95% confidence interval (CI) -6.33 to 0.41; absolute risk difference 3%, 95% CI 1% to 6%; 6 trials). Compared to placebo, a significantly higher rate of withdrawals due to adverse effects (RR 1.50, 95% CI 1.04 to 2.15; absolute risk difference 4%, 95% CI 0.4% to 8.8%; 11 trials) was found in the sulfasalazine group. A serious adverse reaction was reported in one patient taking sulfasalazine (Peto odds ratio 7.50, 95% CI 0.15 to 378.16). Authors’ conclusions There is not enough evidence to support any benefit of sulfasalazine in reducing pain, disease activity, radiographic progression, or improving physical function and spinal mobility in the treatment of AS. A statistically significant benefit in reducing the erythrocyte sedimentation rate and easing spinal stiffness was mentioned in the previous version. However, the effect size was very small and not clinically meaningful. More withdrawals because of side effects occurred with sulfasalazine. Further studies, with larger sample sizes, longer duration, and using validated outcome measures are needed to verify the uncertainty of sulfasalazine in AS.

PLAIN LANGUAGE SUMMARY Sulfasalazine for ankylosing spondylitis We conducted a review of the effect of sulfasalazine for people with ankylosing spondylitis. After searching for all relevant studies up to November 2013, we found 11 studies involving 895 people. Our findings are summarised below. The review showed that in people with ankylosing spondylitis: - compared with fake pills, sulfasalazine probably has little or no difference in pain, disease activity, physical function, spinal mobility, patient and physician global assessment; - damage to the spine as seen on x-ray or magnetic resonance image was not measured and therefore it is not known whether sulfasalazine slows damage; - people had side effects such as stomach upsets, skin reactions/rashes and mouth sores; - more people stopped taking sulfasalazine because of the side effects than when taking fake pills; and - there is not enough evidence to be certain of the benefits and harms of sulfasalazine for ankylosing spondylitis, and more research is needed. What is ankylosing spondylitis and what is sulfasalazine? Ankylosing spondylitis is a type of arthritis, usually in the joints and ligaments of the spine. It may also affect the shoulders, hips, or other joints. Pain and stiffness occur and limit movement in the back and in other joints that are affected. Key results of this review Pain - People who took sulfasalazine rated their pain to be 3 points lower on a scale of 0 to 100 after 3 to 36 months than those who took placebo (3% absolute improvement). - People who took sulfasalazine rated their pain to be 47 on a scale of 0 to 100 after 3 to 36 months. - People who took placebo rated their pain to be 50 on a scale of 0 to 100 after 3 to 36 months. Bath ankylosing spondylitis disease activity index (BASDAI) Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

2

This outcome was not measured in the studies. Bath ankylosing spondylitis function index (BASFI) This outcome was not measured in the studies. Bath ankylosing spondylitis metrology index (BASMI) This outcome was not measured in the studies. Radiographic progress This outcome was not measured in the studies. Total number of withdrawals due to adverse events - 23 more people taking sulfasalazine withdrew due to adverse events than those taking placebo. - 13 out of 100 people taking sulfasalazine withdrew due to adverse events. - 9 out of 100 people taking fake pills withdrew due to adverse events. Serious adverse events Only one person out of 469 stopped taking sulfasalazine for serious adverse events.

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

3

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]

Sulfasalazine compared to placebo for ankylosing spondylitis Patient or population: Patients with ankylosing spondylitis Settings: Outpatients and inpatients Intervention: Sulfasalazine Comparison: Placebo Outcomes

Back pain (pooled data) 100 mm visual analogue scale, 0 = no pain, 100 = severe Follow-up: median 26 weeks

Illustrative comparative risks* (95% CI)

Assumed risk

Corresponding risk

Placebo

Sulfasalazine

Relative effect (95% CI)

The mean back pain The mean back pain (pooled data) in the con- (pooled data) in the intertrol groups was vention groups was 49.5 mm1 2.96 lower (6.33 lower to 0.41 higher)

No of Participants (studies)

Quality of the evidence (GRADE)

Comments

454 (6 studies)

⊕⊕⊕ Moderate2

Absolute risk difference 3% lower (95% CI 1% to 6%); Relative percent change = 6% (95% CI 2% to 12%); NNT4 = n/a5

Mean improve- See comment ment in Bath ankylosing spondylitis disease activity index (BASDAI) not reported

See comment

Not estimable

No data

See comment

Not measured

Mean im- See comment provement in Bath ankylosing spondylitis function index (BASFI) - not reported

See comment

Not estimable

No data

See comment

Not measured

4

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean improve- See comment ment in Bath ankylosing spondylitis metrology index (BASMI) - not reported

See comment

Not estimable

No data

See comment

Not measured

Radiographic progress - See comment not reported

See comment

Not estimable

No data

See comment

Not measured

Total number of with- 94 per 1000 drawals due to adverse events Follow-up: median 26 weeks

134 per 1000 (98 to 182)

RR 1.43 (1.04 to 1.94)

895 (11 studies)

⊕⊕⊕ Moderate2

Absolute risk difference 4% (95% CI 0.4% to 8. 8%); NNTH6 = 25 (95% CI 266 to 12)

Serious adverse events 0 per 1000 Follow-up: mean 36 weeks

1 per 1000 (0 to 0)

OR 7.5 (0.15 to 378.16)

264 (1 study)

⊕⊕⊕ Moderate3

Absolute risk difference 750% (95% CI 15% to 37816%) (W); Relative percent change = 205% (95% CI -87% to 7309%) ; NNTH = n/a

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio; OR: Odds ratio; GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: 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. Very low quality: We are very uncertain about the estimate. 1

From Clegg 1996, mean back pain at baseline in placebo = 48.9 (95% CI 3.0 to 94.8). Different baseline value (3 as endpoint value and 3 as change from baseline value). 3 Wide confidence interval. 4 NNT (Number needed to treat). NNT for dichotomous outcomes calculated using Cates NNT calculator (http://www.nntonline.net/ visualrx/). 5 n/a means result is not statistically significant. 2

5

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

6 NNTH

(Number needed to treat to harm).

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

6

BACKGROUND

Sulfasalazine has been described as an anti-inflammatory drug or immunomodulator (Smedegård 1995). We hypothesized that sulfasalazine had a similar effect on AS. However, the accurate mechanisms of sulfasalazine in AS were unclear.

Description of the condition Ankylosing spondylitis (AS) is a chronic inflammatory disease of unknown cause and belongs to a group of diseases known as spondyloarthropathies, which includes reactive arthritis, arthritis/spondylitis in inflammatory bowel disease, psoriatic arthritis/ spondylitis and undifferentiated spondyloarthropathy. The disease has a relatively early onset, presenting at a mean age of 26 years, and occurs somewhat more frequently in men than in women (Feldtkeller 2003). Depending on the geographical region, the prevalence of AS is about 0.1% to 1.4% of the population (Akkoc 2005). Disease progression may result in loss of mobility and function, and therefore patients can experience a heavy disease burden, with pain and stiffness, loss of physical function, and severe impairment in quality of life (Davis 2005; Zink 2000).

Why it is important to do this review There is uncertainty whether sulfasalazine is an effective treatment for AS. We first performed a systematic review in Chen 2005, and this is an update of that original review.

OBJECTIVES To evaluate the benefits and harms of sulfasalazine for the treatment of ankylosing spondylitis (AS).

Description of the intervention Physiotherapy and exercises are the main non-pharmacological therapies, with modest evidence showing a beneficial effect for AS (Dagfinrud 2008). Non-steroidal anti-inflammatory drugs (NSAIDs) have been the corner stone of pharmacological treatment. In 2010, the Assessment of Ankylosing Spondylitis/European League Against Rheumatism suggested that NSAIDs, including Coxibs, were recommended as first-line drug treatment for AS patients with pain and stiffness. Continuous treatment with NSAIDs is preferred for patients with persistently active, symptomatic disease (Braun 2011a). For patients refractory or intolerant to NSAIDs, disease-modifying antirheumatic drugs such as methotrexate (Chen 2013) and sulfasalazine have been used as a second-line approach. Sulfasalazine has been used in rheumatoid arthritis for decades and a clinically and statistically significant benefit in rheumatoid arthritis disease activity has been confirmed in a systematic review (Suarez-Almazor 1998). It is the best studied disease-modifying antirheumatic drug used in AS, but its efficacy remains unclear. Although AS is difficult to treat, the treatment options for AS have been broadened since the discovery of TNF-alpha inhibitors, e.g. adalimumab, etanercept, and infliximab. TNF-alpha inhibitors have been extensively studied in AS (Zochling 2005). Randomized controlled trials (RCTs) (Braun 2002; Gorman 2002; van den Bosch 2002) showed that they were highly effective in improving disease activity, spinal mobility, function, and pain. However, their expense means that few patients can access them.

How the intervention might work

METHODS

Criteria for considering studies for this review

Types of studies We evaluated randomized controlled trials (RCTs) and quasiRCTs in any language, examining the benefits and harms of sulfasalazine on ankylosing spondylitis (AS).

Types of participants Patients with AS fulfilling any one of the following criteria: 1961 Rome, 1966 New York, modified 1984 New York, Amor or ESSG (European Spondyloarthropathy Study Group) criteria (Olivieri 2002). We included studies with spondyloarthropathies/spondyloarthritis participants if there were available data assessing the outcomes specific to patients with AS. Patients with or without the impairment of peripheral joints were eligible for inclusion.

Types of interventions Sulfasalazine given orally for at least 12 weeks. We included the following comparisons. 1. Sulfasalazine versus placebo. 2. Sulfasalazine versus other medication. 3. Sulfasalazine versus no medication.

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

7

Types of outcome measures

Primary outcomes

According to the Cochrane Musculoskeletal Group (CMSG), the major outcomes included the following. 1. Pain (visual analogue scale or numerical rating scale). 2. Bath ankylosing spondylitis disease activity index (BASDAI). 3. Bath ankylosing spondylitis function index (BASFI). 4. Bath ankylosing spondylitis metrology index (BASMI). 5. Radiographic progression. 6. Total number of withdrawals due to adverse events. 7. Serious adverse events. Secondary outcomes

1. Assessment of Ankylosing Spondylitis responses (e.g. ASAS40, ASAS20, partial remission) (van der Heijde 2002; van der Heijde 2005). 2. Other disease activity index, e.g. Ankylosing Spondylitis Disease Activity Score. 3. Physical function. 4. Spinal mobility and spinal stiffness. 5. Peripheral joints/entheses (pain, swelling and tenderness). 6. Patient global assessment and physician global assessment. 7. Fatigue. 8. Level of acute phase reactants, including erythrocyte sedimentation rate and C-reactive protein.

9. Total number of withdrawals.

Search methods for identification of studies We searched for relevant randomized controlled trials (RCTs) and quasi-RCTs in any language, using the following sources: the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2013, Issue 11); MEDLINE (2003 to 28 November 2013); EMBASE (2003 to 28 November 2013); CINAHL (2003 to 28 November 2013); Ovid MEDLINE data; World Health Organization International Clinical Trials Registry Platform (28 November 2013); and the reference sections of retrieved articles. Search strategies for the original review are in Appendix 1. Search strategies for the second (current) review are in Appendix 2 and Appendix 2. Search results for the second (current) version are in Appendix 4 and Appendix 3.

Data collection and analysis Selection of studies We identified potential trials for inclusion from the search results (see Figure 1). JC and SL independently reviewed unblinded trial reports for this review update (JC and CL for the original review) according to the selection criteria. We resolved disagreements about the inclusion of studies, when necessary, by recourse to a third review author.

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

8

Figure 1. Study flow diagram.

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

9

Data extraction and management JC and SL recorded the extracted data from the included trials onto a prestructured data extraction sheet and entered the data independently into Review Manager 5 (RevMan 2014). We only included outcomes in the review that were specified in the protocol. We entered continuous data (e.g. visual analogue scales of pain, patient’s global assessment) as means and standard deviations (SDs), and dichotomous outcomes (e.g. response, improvement) as number of events.

Assessment of reporting biases We planned using a funnel plot to detect publication bias.

Data synthesis We performed meta-analysis when the participants and interventions were sufficiently homogeneous, using Review Manager 5 (RevMan 2014). We used a random-effects model to combine the results where heterogeneity was significant; otherwise, we used a fixed-effect model.

Assessment of risk of bias in included studies

Subgroup analysis and investigation of heterogeneity

For this review, we assessed risk of bias in all included studies using The Cochrane Collaboration’s tool for assessing risk of bias (Higgins 2011). There were six domains: random sequence generation, allocation concealment, blinding of outcome assessment, incomplete outcome data, selective reporting, and other sources of bias, as specified in Appendix 4. We categorized risk of bias as ’Low risk’, ’High risk’, or ’Unclear risk’. We independently addressed the included trials using a data collection form (Appendix 5).

We planned on conducting subgroup analyses based on characteristics of participants (e.g. different AS classification criteria, male or female, with or without peripheral arthritis) and intervention (e.g. different dosage, different duration).

Measures of treatment effect For continuous data, we used mean difference (MD) or standardised mean difference (SMD) (depending on comparability of scales). For dichotomous data, we used risk ratio (RR).

Sensitivity analysis We performed three sensitivity analyses according to whether: 1. the allocation to intervention or control groups was truly randomized or quasi-randomized (in order to explore the potential for selection bias); and 2. the outcome assessment was blinded (to explore the potential for assessment bias associated with knowledge of the intervention). If either of the sensitivity analyses affected the results of the review then we would make conservative conclusions.

Unit of analysis issues In this review we included only RCTs and there were no unit of analysis issues.

Dealing with missing data For continuous data (e.g. visual analogue scales of pain, patient global assessment), we analyzed only the available data. For dichotomous data of beneficial outcomes (e.g. response, improvement), we used intention-to-treat analysis, that is, we included all participants and assumed no expected event occurred in those participants who dropped out.

Assessment of heterogeneity We explored heterogeneity using the Chi2 test, with significance set at P = 0.10. We measured heterogeneity using I2 . We regarded I2 ≥ 50% as substantial heterogeneity.

Summary of findings tables We graded the evidence using GRADEpro software which was developed by the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) Working Group ( www.gradeworkinggroup.org). We included the seven main outcomes (back pain, bath ankylosing spondylitis disease activity index (BASDAI), bath ankylosing spondylitis function index (BASFI), bath ankylosing spondylitis metrology index (BASMI), radiographic progress, total number of withdrawals due to adverse events, and serious adverse events) in Summary of findings for the main comparison.

RESULTS Description of studies

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

10

Results of the search We obtained a total of 302 records in the updated search (Figure 1). In the preliminary screening of titles and abstracts, we excluded 276 records. After full-text assessment, we found 11 records that had already been included in the first version of the review (Clegg 1996; Corkill 1990; Davis 1989; Dougados 1986; Feltelius 1986; Kirwan 1993; Krajnc 1990; Nissila 1988; Schmidt 2002; Taylor 1991; Winkler 1989). We excluded 15 records (nine studies; Braun 2011 included seven records) for the reasons specified in the Characteristics of excluded studies tables. Therefore, we did not add any new studies to this update. Included studies We included eleven studies and all of these trials compared sulfasalazine with placebo. They included a total of 895 participants, 469 receiving sulfasalazine and 426 placebo. The sample size ranged from 30 to 264. The duration of treatment ranged from 12 weeks to three years. The dosage of sulfasalazine (or placebo) was 2.0 g/day or up to 3.0 g/day, depending on the efficacy and tolerance. More than 30 outcomes were assessed. Among the included participants, about 86% were male; Taylor 1991 did not present the information on the sex distribution of participants. Depending on the trial, age and duration of disease was reported as a mean or median value. The age ranged from 26.9 to 45.7 and the duration of disease ranged from 3.8 to 21.9 years. Zero to 68% of participants had peripheral arthritis. All studies

claimed that they included participants with active disease, but the definitions of active disease varied. Further details are available in the Characteristics of included studies tables.

Excluded studies Braun 2006 and Song 2011 are two trials concerning undifferentiated spondyloarthritis and early AS. We excluded them because the data specific to AS patients were unavailable. Three studies were RCTs comparing sulfasalazine with another active agent. These agents were Kunxian capsule (Chinese traditional medicine) (NCT00953979), Bushen Tongdu decoction (Chinese traditional medicine) (Xu 2008), and leflunomide (Zhao 2006). Because the efficacy of these agents was uncertain in AS and comparisons with them did not contribute to understanding the benefits of sulfasalazine, we excluded these three trials. We excluded one study, Deng 2009, because of lack of relevant outcomes specified in the protocol. Braun 2011 and Zhao 2009 were two RCTs comparing etanercept with sulfasalazine. We excluded them because there is another separate Cochrane review addressing biologics in AS (Zochling 2005).

Risk of bias in included studies The risk of bias of included trials varied greatly and is summarized in Figure 2 and Figure 3.

Figure 2. Risk of bias graph: review authors’ judgments about each risk of bias item presented as percentages across all included studies.

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

11

Figure 3. Risk of bias summary: review authors’ judgments about each risk of bias item for each included study.

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

12

Allocation Seven trials (Clegg 1996; Davis 1989; Feltelius 1986; Krajnc 1990; Nissila 1988; Schmidt 2002; Winkler 1989), were reported as randomized, but details of the methods of randomisation and concealment were not given. Four trials (Corkill 1990; Dougados 1986; Kirwan 1993; Taylor 1991), were reported as double-blind and randomized. The methods of randomisation were generated by sealed envelopes or random number table. Concealment seemed to be adequate in Dougados 1986, Kirwan 1993 and Taylor 1991.

Blinding We judged blinding at high risk of bias in Corkill 1990, as blinding was broken during the trial. We judged two trials at unclear risk of bias (Schmidt 2002; Winkler 1989). We judged all other trials at low risk of bias.

Incomplete outcome data One trial presented complete outcome data (Clegg 1996). Four studies had more than 20% (Dougados 1986; Feltelius 1986; Kirwan 1993; Schmidt 2002), and two had more than 30% of participants drop out (Kirwan 1993; Schmidt 2002). Continuous outcome results were missed in these trials and therefore we judged them at high risk of bias. We judged all other trials at unclear risk of bias.

Selective reporting Two trials were unable to provide the data of all the outcomes prespecified and we judged them at high risk of bias (Kirwan 1993; Taylor 1991). We judged all other trials at low risk of bias.

Other potential sources of bias All the included trials appeared to be free of other sources of bias and therefore we judged them at low risk of bias.

Effects of interventions See: Summary of findings for the main comparison Sulfasalazine compared to placebo for ankylosing spondylitis Because of multiple sources of heterogeneity, we could not perform meta-analyses in all outcomes. Instead, we presented a narrative summary of pertinent findings from the individual studies. The main reported findings for the individual trials are also summarised in Additional Table 1 and Table 2.

Primary outcomes Different parameters were used to assess pain in the included trials. They included improvement in back pain (Analysis 1.3), back pain measured with 100 mm visual analogue scale (Analysis 1.4; Analysis 1.5), night pain (% no pain) (Analysis 1.6), score of sleep disturbance (score 0 to 4) (Analysis 1.7), frequency of nocturnal awaking (Analysis 1.8), score of daily non-steroidal anti-inflammatory drugs (NSAIDs) (change from baseline, usual dosage as 10) (Analysis 1.9), and reducing and stopping NSAIDs (Analysis 1.10). Among these parameters, back pain was the most used one which we chose to include in Summary of findings for the main comparison. Six trials assessed back pain measured with visual analogue scale (0 mm to 100 mm) (Clegg 1996; Dougados 1986; Nissila 1988; Schmidt 2002; Taylor 1991; Winkler 1989). Two trials found significant lower back pain measured with visual analogue scale in sulfasalazine compared with the placebo group (Nissila 1988; Schmidt 2002). However, pooled MD was -2.96 (95% CI -6.33 to 0.41) and the difference was not statistically significant (Analysis 1.4). We did not find significant heterogeneity among the trials. No data were available in these trials for bath ankylosing spondylitis disease activity index (BASDAI), bath ankylosing spondylitis function index (BASFI), bath ankylosing spondylitis metrology index (BASMI), or radiographic progression. For safety outcomes, significantly higher rates of withdrawal due to side effects were found in the sulfasalazine group compared with the placebo group. Pooled RR was 1.50 (95% CI 1.04 to 2.15) (Analysis 1.44). Among 469 participants receiving sulfasalazine, a serious adverse reaction was reported in one patient who developed a generalized, erythematous, raised, pruritic eruption which was associated with nausea, anorexia and insomnia (Clegg 1996).

Secondary outcomes The included trials did not report Assessment of Ankylosing Spondylitis response, other disease activity index, or fatigue. Three trials assessed physical function index and none of them found a statistically significant difference between the sulfasalazine and placebo groups in the analysis of either end point data or change from baseline data (Clegg 1996; Dougados 1986; Schmidt 2002). Pooled analysis showed a similar result and no significant heterogeneity among these trials (Analysis 1.1; Analysis 1.2). Spinal mobility outcomes included chest expansion (Analysis 1.11; Analysis 1.12), forced vital volume (Analysis 1.13), modified Schober’s test (Analysis 1.14; Analysis 1.15), occiput-to-wall test (Analysis 1.16; Analysis 1.17), fingers-to-floor test (Analysis 1.18; Analysis 1.19), and chin sternum distance (Analysis 1.20). Two trials showed better chest expansion in the sulfasalazine compared with placebo group (Nissila 1988; Schmidt 2002) (Analysis 1.11;

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

13

Analysis 1.12). This statistically significant difference was confirmed in pooled analysis with weighted MD as 0.30 cm (Analysis 1.11) and 0.31 cm (Analysis 1.12). There was significant heterogeneity among the trials (P < 0.0001) (Analysis 1.12). Schmidt 2002, a trial with more than 30% of drop-outs, contributed more than 80% weight. When excluded from the analysis, the difference and heterogeneity were insignificant. No statistically significant difference was found between intervention groups in other outcomes of spinal mobility. However, significant heterogeneity was also found among the included studies in Schober’s test (Analysis 1.15), occiput-to-wall test (Analysis 1.16), and improvement in patient global assessment (Analysis 1.31), and physician global assessment (Analysis 1.34). Again, when the Schmidt 2002 trial was excluded, heterogeneity became insignificant. Peripheral joints/entheses were assessed in several studies (Clegg 1996; Dougados 1986; Kirwan 1993; Nissila 1988; Schmidt 2002). Outcomes included joint pain/tenderness score (0 to 198) or number (Analysis 1.21; Analysis 1.22), joint swelling score (0 to 198) or number (Analysis 1.23; Analysis 1.24), dactylitis score (0 to 3) (Analysis 1.25; Analysis 1.26), enthesopathy index (0 to 90) (Analysis 1.27; Analysis 1.28) and spondylitis articular index (0 to 90) (Analysis 1.29; Analysis 1.30). There was no statistically significant difference between intervention groups in either individual study or pooled data except for the Kirwan 1993 study, which found that occurrence of peripheral joint symptoms were lower in the sulfasalazine group (0.298 episodes/year) than in the placebo group (0.392 episodes/year) (P < 0.05). This difference was not shown in our analysis because of lack of available data. Dougados 1986 found more participants improved in terms of patient global assessment in the sulfasalazine group than placebo group, but two other trials, Clegg 1996 and Schmidt 2002, did not, and neither did our pooled analysis (Analysis 1.31). One trial, Winkler 1989, compared patient assessment of disease severity (100 mm visual analogue scale) between the sulfasalazine and placebo groups and found no difference (Analysis 1.32). One trial, Nissila 1988, assessed general wellbeing (100 mm visual analogue scale) and found the sulfasalazine group was better than the placebo group (MD -11.00, 95% CI -19.84 to -2.16) (Analysis 1.33). Two trials, Clegg 1996 and Schmidt 2002, analyzed improvement in physician global assessment and found no difference between the sulfasalazine and placebo groups, neither did our pooled result (Analysis 1.34). One trial, Clegg 1996, assessed response to treatment which was based on both patient and physician global assessment and found no difference between the sulfasalazine and placebo groups (Analysis 1.35). Five trials assessed duration of morning stiffness (Clegg 1996; Krajnc 1990; Nissila 1988; Schmidt 2002; Winkler 1989). Schmidt 2002, found a statistically significant difference between intervention groups, favoring sulfasalazine over placebo. But the difference of pooled results was statistically insignificant (Analysis 1.36; Analysis 1.37). Significant heterogeneity existed among the studies. When we excluded the Schmidt 2002 trial, heterogeneity

became insignificant. Two trials assessed morning stiffness with a 100 mm visual analogue scale (where 0 = no stiffness, 100 = severe) (Nissila 1988; Taylor 1991). One study showed a statistically significant difference favoring sulfasalazine over placebo (Nissila 1988). Pooled data showed a similar result (MD -13.89, 95% CI -22.54 to -5.24 (Analysis 1.38). For erythrocyte sedimentation rate, four trials found statistically significant differences between intervention groups, favoring sulfasalazine over placebo (Clegg 1996; Krajnc 1990; Nissila 1988; Schmidt 2002). A similar result was found in the pooled data (Analysis 1.40; Analysis 1.41). MD of change from baseline was -3.11 mm/hr, 95% CI -4.62 to -1.60 mm/hr. MD of end point was -7.07 mm/hr, 95% CI -14.39 to 0.25 (not statistically significant). Significant heterogeneity existed among the trials (P = 0.02). This could be due to the large difference of erythrocyte sedimentation rate at baseline levels among the studies. Three trials, Clegg 1996, Nissila 1988 and Schmidt 2002, assessed C-reactive protein and only one, Schmidt 2002, showed a statistically significant difference between intervention groups, favoring sulfasalazine over placebo. The pooled result showed no statistically significant difference (Analysis 1.42; Analysis 1.43). Ten trials reported the total number of withdrawals (Clegg 1996; Davis 1989; Dougados 1986; Feltelius 1986; Kirwan 1993; Krajnc 1990; Nissila 1988; Schmidt 2002; Taylor 1991; Winkler 1989). A significantly higher rate of withdrawals due to any reason was found in the sulfasalazine groups compared with the placebo groups (pooled RR 1.33, 95% CI 1.03 to 1.73) (Analysis 1.46).

Sensitivity analysis We did not perform a sensitivity analysis against concealment because we did not rank any included trials at high risk of bias. Blinding was reported in all studies, however, three participants broke the blinding during the trial of Corkill 1990. Continuous outcome data were unavailable in this trial. For the dichotomous outcome, withdrawals due to side effect, exclusion of this trial from the analysis showed a similar result (Analysis 1.44). Since continuous data included only the participants who completed the trials, we excluded the trials with more than 30% drop-outs (Kirwan 1993; Schmidt 2002), to see if a high percentage of drop-outs affected the results. After exclusion of Schmidt 2002 (no continuous data was available in Kirwan 1993), we found that the MD of chest expansion and MD of duration of morning stiffness (change from baseline) became statistically insignificant, which have been mentioned above. Other outcomes remained similar. Only the median and CI values were available in Dougados 1986. We assumed that mean was equal to median for each outcome and calculated the SD from the CI and sample size. Therefore, the mean and SD were estimated in this trial. Sensitivity analysis, however, found no obvious difference between inclusion and exclusion of this trial in the pooled result. The Davis 1989 and Feltelius 1986 trials had the least number of participants and the shortest periods. When

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

14

excluded, the pooled difference of erythrocyte sedimentation rate (end point) became statistically significant (weighted MD -6.13 mm/hr, 95% CI -11.89 to -0.37 mm/hr), while other results remained similar. Subgroup data We could not group studies of sulfasalazine versus placebo according to characteristics of interventions and participants. Only one study (Winkler 1989) presented data of subgroups (participants with and without peripheral arthritis). In participants with peripheral arthritis (N = 15), we did not find a statistically significant difference between intervention groups in back pain, score of sleep disturbance, chest expansion, Schober’s test, fingers-to-floor test, articular index, degree of joint swelling, patient assessment of disease severity, duration of morning stiffness, or erythrocyte sedimentation rate. For participants without peripheral arthritis (N = 34), we did not find a statistically significant difference in these outcomes (but articular index and degree of joint swelling were not assessed) except back pain measured with a 100 mm visual analogue scale (where 0 = no pain, 100 = severe), which was shown to significantly favor sulfasalazine over placebo (MD -9.20, 95% CI -17.81 to -0.59) (Analysis 3.1). Another study separately analyzed the results of participants with peripheral arthritis (N = 77) and found more peripheral responses in the sulfasalazine than placebo group (55.9% versus 30.2%, P = 0.023) (Clegg 1996). Here, peripheral response was a composition of four parameters, e.g. patient self assessment, physician assessment, joint pain/tenderness score, and joint swelling score. Peripheral response was defined as improvement in at least two parameters, with no worsening in other parameters. We did not analyze these parameters in Review Manager 5 because the information on treatment allocation was not given (RevMan 2014).

DISCUSSION

Summary of main results We identified 11 RCTs evaluating the benefits and harms of sulfasalazine in ankylosing spondylitis (AS). These trials enrolled 895 participants in total. All these trials compared sulfasalazine with placebo. More than 30 outcomes were assessed. Most of the pooled data in the present review included only a few trials (less than five) and few participants (less than 400). Seven major outcomes, e.g. pain, bath ankylosing spondylitis disease activity index (BASDAI), bath ankylosing spondylitis function index (BASFI), bath ankylosing spondylitis metrology index (BASMI), radiographic progression, total number of withdrawals, and total number of withdrawals due to adverse events, were included in the present review. None of the included studies assessed BASDAI, BASFI, BASMI or

radiographic progression. Different parameters were used to assess pain. We chose the most used one, e.g. back pain, to be included in Summary of findings for the main comparison. We did not find any statistically significant difference in either individual trial data or pooled analysis results, except that back pain measured with visual analogue scale (0 mm to 100 mm) was found to be statistically significantly lower in the sulfasalazine group compared with the placebo group in two trials (Nissila 1988; Schmidt 2002). However, the pooled mean difference was not statistically significant. With regard to the safety of sulfasalazine, the total number of adverse events was higher in the sulfasalazine group than in the placebo group; a serious adverse reaction occurred in one patient taking sulfasalazine. Statistically significantly higher rates of withdrawals due to side effects, and for any reason were found in the sulfasalazine group compared to the placebo group. The most frequent side effects with sulfasalazine were nausea, heartburn, epigastric distress, skin reactions, lightheadedness, and headache. In addition, the pooled data showed that sulfasalazine did significantly ease morning stiffness and reduce erythrocyte sedimentation rate, even though the slight difference does not have much clinical significance. For other outcomes, it was not possible to pool the data. Among the included trials, only two (Clegg 1996; Winkler 1989), analyzed the effect of sulfasalazine on peripheral arthritis. One trial presented data of the subgroup with peripheral arthritis (N = 15) and found no statistically significant difference between the sulfasalazine and placebo group (Winkler 1989). The other study separately analyzed the results of participants with peripheral arthritis (N = 77) and found more peripheral responses in the sulfasalazine group than in the placebo group (Clegg 1996). There is not enough evidence suggesting sulfasalazine is better for peripheral joints than for the spine.

Overall completeness and applicability of evidence The included studies did not report most of the major, validated outcomes such as BASDAI, BASFI, BASMI or radiographic progression. In addition, the study population in the included trials involved 895 AS participants who fulfilled New York criteria (Olivieri 2002) which required radiographic change. The majority of these participants were adults, and considered to have active disease, although the definition of active disease differed among the trials. The duration of disease ranged from 3.8 to 21.9 years. Therefore, the evidence was not applicable to juvenile patients and those patients with early disease onset.

Quality of the evidence The present review shows limited benefit of sulfasalazine in AS. The strength of evidence in this review should be ranked as mod-

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

15

erate level for several reasons. First, the potential benefit of sulfasalazine might be overlooked because most of the pooled data in the present review included only a few trials (less than five) and few participants (less than 400). Moreover, obvious methodological drawbacks existed as specified in the Risk of bias in included studies section. Finally, the included participants were all patients at an advanced stage of AS who might not respond well to sulfasalazine.

Potential biases in the review process We conducted a thorough search in several databases and updated the search strategies during the process of updating this review. We included all trials that met our inclusion criteria and applied no language restrictions. Two review authors independently reviewed the trial reports according to the selection criteria. When necessary, we contacted the trial authors to clarify any information regarding their trials. We resolved any disagreements regarding the inclusion of studies by discussion.

Agreements and disagreements with other studies or reviews Ferraz 1990 conducted a meta-analysis of five RCTs and concluded that sulfasalazine significantly relieved pain and morning stiffness, compared with placebo (Davis 1989; Dougados 1986; Feltelius 1986; Nissila 1988; Winkler 1989). Six RCTs have been published since then (Clegg 1996; Corkill 1990; Kirwan 1993; Krajnc 1990; Schmidt 2002; Taylor 1991). Our previous review (Chen 2005), and the current update showed limited benefit of sulfasalazine in reducing erythrocyte sedimentation rate and easing spinal stiffness, as the effect size was small and not clinically significant. We did not find any other published systematic review, meta-analysis or RCT comparing sulfasalazine with placebo in our updated search. For peripheral arthritis of AS, the present review did not support any benefit of sulfasalazine, in accordance with the experts’ recommendation of using TNF-alpha inhibitors (such as etanercept), rather than sulfasalazine, if available (van der Heijde 2011). A randomized double-blind study comparing etanercept

with sulfasalazine found that etanercept was more effective in all joint assessments, regardless of swollen joint involvement (Braun 2011). Therefore, etanercept might be a better choice for management of patients with AS and peripheral joint involvement.

AUTHORS’ CONCLUSIONS Implications for practice In this review, there was no evidence to support any benefit of sulfasalazine in reducing pain, disease activity, radiographic progression, or improving physical function and spinal mobility. Meanwhile, adverse events were common and there were more withdrawals because of side effects among participants taking sulfasalazine than placebo, although serious adverse events were rare.

Implications for research RCTs of larger sample sizes and with longer duration are needed. Assessment of Ankylosing Spondylitis response (van der Heijde 2002; van der Heijde 2005), and the core set for the evaluation of disease controlling anti-rheumatic treatment (van der Heijde 1999; van der Heijde 2002), should be used to assess outcomes. It is also important to subgroup appropriately for peripheral joint involvement, disease duration and gender.

ACKNOWLEDGEMENTS The review authors would like to thank the editorial team of the Cochrane Musculoskeletal Group (CMSG) for their helpful comments and Sally Green, Steve McDonald, Janet Piehl, Denise O’Connor, Elmer Villanueva and the staff of the Australasian Cochrane Centre for their supervision and technical support. The review authors would also like to thank M Dougados, J Kirwan, I Krajnc, J Braun, and J Heather for offering extra information about their studies; and I Krajnc, D Schuenemann and S Herter for language translation.

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

16

REFERENCES

References to studies included in this review Clegg 1996 {published data only} Clegg DO, Reda DJ, Adbellaitf M. Comparison of sulfasalazine and placebo for the treatment of axial and peripheral articular manifestations of the seronegative spondyloarthropathies: a Department of Veterans Affairs cooperative study. Arthritis and Rheumatism 1999;42(11): 2325–9. ∗ Clegg DO, Reda DJ, Weisman MH, Blackburn WD, Cush JJ, Cannon GW, et al.Comparison of sulfasalazine and placebo in the treatment of ankylosing spondylitis: a Department of Veterans Affairs Cooperative Group. Arthritis and Rheumatism 1996;39:2004–12. Reda D, Anderson R, Abdellatif M, Williams D, Clegg D. Longitudinal analysis of binary data in the V.A. Cooperative Study of sulfasalazine for the treatment of seronegative spondyloarthropathies. Controlled Clinical Trials 1995;16(3 Suppl):90s–91s. Corkill 1990 {published data only} Corkill MM, Jobanputra P, Gibson T, Macfarlane DG. A controlled trial of sulphasalazine treatment of chronic ankylosing spondylitis: failure to demonstrate a clinical effect. British Journal of Rheumatology 1990;29(1):41–5. Davis 1989 {published data only} Davis MJ, Dawes PT, Beswick E, Lewin IV, Stanworth DR. Sulphasalazine therapy in ankylosing spondylitis: its effect on disease activity, immunoglobulin A and the complex immunoglobulin A-alpha-1-antitrypsin. British Journal of Rheumatology 1989;28(5):410–3. Dougados 1986 {published data only} ∗ Dougados M, Boumier P, Amor B. Sulphasalazine in ankylosing spondylitis: A double blind controlled study in 60 patients. British Medical Journal 1986;293(6552): 911–4. Dougados M, Boumier P, Amor B. Treatment of ankylosing spondylarthritis with salazosulfapyridin. A double-blind, controlled study in 60 patients [French]. Revue du Rhumatisme et des Maladies Osteo Articulaires 1987;54(3): 255–60. Dougados M, Nguyen M, Mijiyawa M, Amor B. Sulfasalazine in Spondylarthropathies. Zeitschrift fur Rheumatologie 1990;49(Suppl 1):80. Feltelius 1986 {published data only} Feltelius N, Hallgren R. Sulphasalazine in ankylosing spondylitis. Annals of the Rheumatic Diseases 1986;45(5): 396–9. Kirwan 1993 {published data only} Kirwan J, Edwards A, Huitfeldt B, Thompson P, Currey H. The course of established ankylosing spondylitis and the effects of sulphasalazine over 3 years. British Journal of Rheumatology 1993;32(8):729–33.

Krajnc 1990 {published data only} Krajnc I. Sulphasalazine in the treatment of ankylosing spondylitis [Serbocroatian]. Lijecnicki Vjesnik 1990;112(56):171–4. Nissila 1988 {published data only} Nissila M, Lahesmaa R, Leirisalo-Repo M, Lehtinen K, Toivanen P, Granfors K. Antibodies to Klebsiella pneumoniae, Escherichia coli, and Proteus mirabilis in ankylosing spondylitis: effect of sulfasalazine treatment. Journal of Rheumatology 1994;21(11):2082–7. ∗ Nissila M, Lehtinen K, Leirisalo-Repo M, Luukkainen R, Mutru O, Yli-Kerttula U. Sulfasalazine in the treatment of ankylosing spondylitis. A twenty-six-week, placebocontrolled clinical trial. Arthritis and Rheumatism 1988;31 (9):1111–6. Schmidt 2002 {published data only} ∗ Schmidt WA, Wierth S, Milleck D, Droste U, GromnicaIhle E. Sulfasalazine in ankylosing spondylitis: a prospective, randomized, double-blind placebo-controlled study and comparison with other controlled studies [German]. Zeitschrift fur Rheumatologie 2002;61(2):159–67. Schmidt WA, Wierth S, Milleck D, Droste U, GromnicaIhle E. Sulphasalazine in ankylosing spondylitis: a prospective, randomized, double-blind, placebo-controlled study and meta-analysis of other controlled studies. Zeitschrift fur Rheumatologie 2009;59(Suppl 3):69. Taylor 1991 {published data only} Taylor HG, Beswick EJ, Dawes PT. Sulphasalazine in ankylosing spondylitis. A radiological, clinical and laboratory assessment. Clinical Rheumatology 1991;10(1): 43–8. Winkler 1989 {published data only} Winkler V. Sulphasalazine treatment in ankylosing spondylitis: A comparison of sulphasalazine with placebo. Magyar Reumatologia 1989;30(Suppl):29–37.

References to studies excluded from this review Benitez-Del-Castillo {published data only} Benitez-Del-Castillo JM, Garcia-Sanchez J, Iradier T, Banares A. Sulfasalazine in the prevention of anterior uveitis associated with ankylosing spondylitis. Eye 2000;14(3A): 340–3. Braun 2006 {published data only} Braun J, Zochling J, Baraliakos X, Alten R, Burmester G, Grasedyck K, et al.Efficacy of sulfasalazine in patients with inflammatory back pain due to undifferentiated spondyloarthritis and early ankylosing spondylitis: a multicentre randomised controlled trial. Annals of the Rheumatic Diseases 2006;65(9):1147-53. Braun 2011 {published data only} Braun J, Pavelka K, Ramos-Remus C, Dimic A, Vlahos B, Freundlich B, et al.Clinical efficacy of etanercept versus sulfasalazine in ankylosing spondylitis subjects with

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

17

drug-free remission after 1 year of treatment with etanercept versus sulfasalazine in early axial spondyloarthritis: 2 year data of the ESTHER trial. Annals of the Rheumatic Diseases 2012;71(7):1212–5. ∗ Song IH, Hermann KG, Haibel H, Althoff CE, Listing J, Burmester GR, et al.Effects of etanercept versus sulfasalazine in early axial spondyloarthritis on active inflammatory lesions as detected by whole-body MRI (ESTHER): A 48week randomised controlled trial. Annals of the Rheumatic Diseases 2011;70(4):590–6.

peripheral joint involvement. Journal of Rheumatology 2012; 39(4):836–40. ∗ Braun J, Van Der Horst-Bruinsma IE, Huang F, BurgosVargas R, Vlahos B, Koenig AS, et al.Clinical efficacy and safety of etanercept versus sulfasalazine in patients with ankylosing spondylitis: A randomized, double-blind trial. Arthritis and Rheumatism 2011;63(6):1543–51. Freundlich B, Braun J, Huang F, Burgos-Vargas R, Van Der Horst-Bruinsma IE, Vlahos B, et al.Assessment of clinical efficacy in a randomized, double-blind study of etanercept and sulphasalazine in patients with ankylosing spondylitis. Internal Medicine Journal 2009;39:A55. Guzman R, Burgos-Vargas R, Braun J, Freundlich B, Vlahos B, Koenig AS. Etanercept is significantly more effective than sulfasalazine in patients with ankylosing spondylitis. Journal of Clinical Rheumatology 2010;16:S73. NCT00247962. Study Evaluating Etanercept and Sulphasalazine in Ankylosing Spondylitis. clinicaltrials.gov/ ct2/show/record/NCT00247962 2008. Van Der Heijde D, Braun J, Sieper J, Wishneski C, Vlahos B, Szumski A, et al.The ankylosing spondylitis disease activity score in subjects treated with etanercept (ETN) or sulfasalazine: Comparison with standard efficacy measures. Rheumatology 2010;49:i55. Van Der Heijde DM, Braun J, Dougados M, Szumski A, Pedersen R, Vlahos B, et al.Clinical improvement with Etanercept versus sulfasalazine treatment in patients with ankylosing spondylitis: Comparative performance of various efficacy measurements (ASCEND). Arthritis and rheumatism 2010;62:1927.

Xu 2008 {published data only} Xu JR, Zhang CY, Li WM. Effects of bushen tongdu decoction on serum tumour necrosis factor-alpha and transforming growth factor beta1, in patients with ankylosing spondylitis [

α TGF–β˙1 ]. Zhongguo Zhong Xi Yi Jie He Za Zhi Zhongguo Zhongxiyi Jiehe Zazhi/Chinese Journal of Integrated Traditional and Western Medicine 2008;28(12): 1093–5. Zhao 2006 {published data only} Zhao FT, Zhao H, Guan JL, Han XH. Clinical study on long-term effectiveness of leflunomide compared with sulfasalazine in treatment of ankylosing spondylitis [Chinese]. Pharmaceutical Care and Research 2006;6(6): 430–2. Zhao 2009 {published data only} Zhao FT, Zhao F, Wang YL. Efficacy of etanercept on ankylosing spondylitis

Burgos-Vargas 2002 {published data only} Burgos-Vargas R, Pacheco-Tena C, Vazquez-Mellado J. A short-term follow-up of enthesitis and arthritis in the active phase of juvenile onset spondyloarthropathies. Clinical and Experimental Rheumatology 2002;20(5):727–31. ∗ Burgos-Vargas R, Vazquez-Mellado J, Pacheco-Tena C, Hernandez-Garduno A, Goycochea-Robles MV. A 26 week randomised, double blind, placebo controlled exploratory study of sulfasalazine in juvenile onset spondyloarthropathies. Annals of the Rheumatic Diseases 2002;61(10):941–2.

[ ]. Journal of Shanghai Jiaotong University (Medical Science) 2009;29(12): 1506–8.

Additional references Akkoc 2005 Akkoc N, Khan MA. Overestimation of the prevalence of ankylosing spondylitis in the Berlin study: comment on the article by Braun et al. Arthritis and Rheumatism 2005;52 (12):4048–9. Braun 2002 Braun J, Brandt J, Listing J, Zink A, Alten R, Golder W, et al.Treatment of active ankylosing spondylitis with infliximab: a randomised controlled multicenter trial. Lancet 2002;359(9313):1187–93.

Deng 2009 {published data only} Deng X, Huang F, Zhang J. Thalidomide delays the rate of relapse in ankylosing spondylitis after discontinuing etanercept treatment [ Arthritis and Rheumatism 2009;60:1776. NCT00953979 {published data only} NCT00953979. Efficacy and safety of Kunxian capsule in treatment of patients with early ankylosing spondylitis [ ]. clinicaltrials.gov/ct2/show/record/NCT00953979. Song 2011 {published data only} Song I-H, Althoff CE, Haibel H, Hermann K-GA, Poddubnyy D, Listing J, et al.Frequency and duration of

TNF–

].

Braun 2011a Braun J, van den Berg R, Baraliakos X, Boehm H, BurgosVargas R, Collantes-Estevez E, et al.2010 update of the ASAS/EULAR recommendations for the management of ankylosing spondylitis. Annals of the Rheumatic Diseases 2011;70(6):896–904. Chen 2013 Chen J, Veras MM, Liu C, Lin J. Methotrexate for ankylosing spondylitis. Cochrane Database of Systematic Reviews 2013, Issue 2. [DOI: 10.1002/ 14651858.CD004524.pub4]

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

18

Dagfinrud 2008 Dagfinrud H, Hagen KB, Kvien TK. Physiotherapy interventions for ankylosing spondylitis. Cochrane Database of Systematic Reviews 2008, Issue 1. [DOI: 10.1002/ 14651858.CD002822.pub3] Davis 2005 Davis JC, van der Heijde D, Dougados M, Woolley JM. Reductions in health-related quality of life in patients with ankylosing spondylitis and improvements with etanercept therapy. Arthritis and Rheumatism 2005;53(4):494–501. Feldtkeller 2003 Feldtkeller E, Khan MA, van der Heijde D, van der Linden S, Braun J. Age at disease onset and diagnosis delay in HLA-B27 negative vs. positive patients with ankylosing spondylitis. Rheumatology International 2003;23(2):61–6. Ferraz 1990 Ferraz MB, Tugwell P, Goldsmith CH, Atra E. Metaanalysis of sulfasalazine in ankylosing spondylitis. The Journal of Rheumatology 1990;17(11):1482–6. Gorman 2002 Gorman JD, Sack KE, Davis JC. Treatment of ankylosing spondylitis by inhibition of tumor necrosis factor alpha. New England Journal of Medicine 2002;346(18):1349–56. Higgins 2011 Higgins JPT, Altman DG, Sterne JAC (editors). Chapter 8: Assessing risk of bias in included studies. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from www.cochrane-handbook.org. Olivieri 2002 Olivieri I, van Tubergen A, Salvarani C, van der Linden S. Seronegative spondyloarthritides. Best Practice and Research Clinical Rheumatology 2002;16(5):723–39. RevMan 2014 The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan). 5.3. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014. Smedegård 1995 Smedegård G, Björk J. Sulphasalazine: mechanism of action in rheumatoid arthritis. British Journal of Rheumatology 1995;34 Suppl 2:7–15. Suarez-Almazor 1998 Suarez-Almazor ME, Belseck E, Shea B, Tugwell P, Wells GA. Sulfasalazine for treating rheumatoid arthritis. Cochrane Database of Systematic Reviews 1998, Issue 2. [DOI: 10.1002/14651858.CD000958; : ] van den Bosch 2002 van den Bosch F, Kruithof E, Baeten D, Herssens A, De keyser F, Mielants H, et al.Randomized double-blind

comparison of chimeric monoclonal antibody to tumor necrosis factor alpha (infliximab) versus placebo in active spondylarthropathy. Arthritis and Rheumatism 2002;46(3): 755–65. van der Heijde 1999 van der Heijde D, van der Linden S, Bellamy N, Calin A, Dougados M, Khan MA. Which domains should be included in a core set for endpoints in ankylosing spondylitis? Introduction to the ankylosing spondylitis module of OMERACT IV. The Journal of Rheumatology 1999;26(4):945–7. van der Heijde 2002 van der Heijde D, Braun J, McGonagle D, Siegel J. Treatment trials in ankylosing spondylitis: current and future considerations. Annals of the Rheumatic Diseases 2002;61 Suppl 3:iii24–32. van der Heijde 2005 van der Heijde D, Dougados M, Davis J, Weisman MH, Maksymowych W, Braun J, et al.Assessment in Ankylosing Spondylitis International Working Group/Spondylitis Association of America recommendations for conducting clinical trials in ankylosing spondylitis. Arthritis and Rheumatism 2005;52(2):386–94. van der Heijde 2011 van der Heijde D, Sieper J, Maksymowych WP, Dougados M, Burgos-Vargas R, Landewé R, et al.Assessment of SpondyloArthritis International Society. 2010 Update of the international ASAS recommendations for the use of anti-TNF agents in patients with axial spondyloarthritis. Annals of the Rheumatic Diseases 2011;70(6):905. Zink 2000 Zink A, Braun J, Listing J, Wollenhaupt J. Disability and handicapin in rheumatoid arthritis and ankylosing spondylitis - results from the German rheumatological database. German Collaborative Arthritis Centers. Journal of Rheumatology 2007;27(3):613–22. Zochling 2005 Zochling J, Maxwell L, Beardmore J, Boonen A. TNF-alpha inhibitors for ankylosing spondylitis. Cochrane Database of Systematic Reviews 2005, Issue 3. [DOI: 10.1002/ 14651858.CD005468]

References to other published versions of this review Chen 2005 Chen J, Liu C. Sulfasalazine for ankylosing spondylitis. Cochrane Database of Systematic Reviews 2005, Issue 2. [DOI: 10.1002/14651858.CD004800.pub2] ∗ Indicates the major publication for the study

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

19

CHARACTERISTICS OF STUDIES

Characteristics of included studies [ordered by study ID] Clegg 1996 Methods

Multicenter Randomized allocation Double-blind allocation and assessment Parallel design Duration: 36 weeks Sample size at entry: 264 SSZ: 131 Placebo: 133 Clear description of withdrawal and drop-outs Primary outcomes were analyzed according to intention-to-treat, secondary outcomes included only those who completed the trial

Participants

Participants fulfilling the modified New York criteria for definite AS Other inclusion criteria: 1. active spondylitis, defined as morning stiffness at least 45 min, inflammatory pain, patient and physician global assessment of disease activity of “moderate” or higher, failure to respond to a trial of aspirin or another NSAID 2. maintained on a stable dose of aspirin or another NSAID for at least 4 weeks Exclusion criteria: 1. evidence of complete ankylosis of the entire spine 2. other known causes of sacroiliitis 3. positive rheumatoid factor or anti-nuclear antibody (> 1:80) 4. history of inflammatory bowel diseases or other rheumatic diseases 5. previously treated with SSZ 6. history of sensitivity to salicylates, sulfa-containing drugs or tartrazine 7. with chronic diseases (according to the investigator) Age: 44.6 +/- 12.6 Male: 95% Duration of disease: 18.5 +/- 11.6 HLA B27: 81% With peripheral arthritis: 29%

Interventions

SSZ 1.0 g orally, twice a day Placebo 1.0 g orally, twice a day

Outcomes

Primary: 1. Response to treatment (event) 2. Improvement in physician global assessment (event) 3. Improvement in patient global assessment (event) 4. Improvement in morning stiffness (event) 5. Improvement in back pain (event) Secondary: 1. Night pain (no bother, event) 2. Duration of morning stiffness (hr)

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

20

Clegg 1996

(Continued)

3. Back pain (100 mm visual analogue scale) 4. Spondylitis function index (score 0 to 40) 5. Joint pain/tenderness score (0 to 198) 6. Joint swelling score (0 to 198) 7. Dactylitis score (0 to 3) 8. Enthesopathy index (0 to 90) 9. Spondylitis articular index (0 to 30) 10. Chest expansion (cm) 11. Modified Schober’s test (cm) 12. Occiput-to- wall test (cm) 13. Fingers-to- floor test (cm) 14. ESR (mm/hr) 15. CRP (ug/mL) 16. Withdrawal due to side effect 17. Withdrawal due to ineffectiveness 17. Drop out for any reason Notes

This trial was funded by Department of Veterans Affairs and Medical Research Service All continuous outcomes were presented as both end point and change from baseline Subgroup analysis: In participants without peripheral disease (N = 187), 40.2% of SSZ group and 43.3% of placebo group showed axial response In participants with peripheral diseases (N = 77), 32.4% of SSZ group and 20.9% of placebo group showed axial response, while 55.9% of SSZ group and 30.2% of placebo group showed peripheral response No description about the number under each intervention

Risk of bias Bias

Authors’ judgement

Support for judgement

Random sequence generation (selection Unclear risk bias)

Insufficient information about the sequence generation process to permit judgment of Yes or No

Allocation concealment (selection bias)

Insufficient information to permit judgment of Yes or No

Unclear risk

Blinding (performance bias and detection Low risk bias) All outcomes

“Patients received either 500 mg of entericcoated SSZ tablets or an identical placebo”. “The following physical assessment was made by a clinician”

Incomplete outcome data (attrition bias) All outcomes

“Withdrawals from the study were fewer than was anticipated by the Planning Committee”. 36 patients from the SSZ group and 25 patients from the placebo group, the reasons for withdrawal were similar in both groups. The main study analysis based on

Low risk

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

21

Clegg 1996

(Continued)

intention-to-treat principles Selective reporting (reporting bias)

Low risk

They reported the expected outcomes

Other bias

Low risk

The study appears to be free of other sources of bias

Corkill 1990 Methods

Randomized allocation Double-blind and assessment blind Parallel design Duration: 48 weeks Sample size at entry:62 SSZ: 32 Placebo: 30 Clear description of withdrawal due to side effect Unclear about which intervention group the 6 drop-outs belonged to Intention-to-treat analysis: unclear

Participants

Participants fulfilling the New York criteria for AS and requiring daily NSAIDs or analgesics Exclusion criteria: 1. psoriasis or acknowledged inflammatory bowel diseases 2. previously treated with SSZ 3. currently treated with corticosteroid or immunosuppressive drugs Age in SSZ group: 37.4 +/- 8.5 Age in placebo group: 28.2 +/- 11.4 Male: 87% Duration of symptom in SSZ group: 12.3 +/- 8.2 yr Duration of symptom in placebo group: 16.1 +/- 11.4 yr With peripheral arthritis: 19%

Interventions

SSZ 1.0 g orally, twice a day Placebo 1.0 g orally, twice a day

Outcomes

1. Spinal pain (100 mm visual analogue scale) 2. Spinal stiffness (100 mm visual analogue scale) 3. Peripheral joint pain (100 mm visual analogue scale) 4. Modified Schober’s test (cm) 5. Chest expansion (cm) 6. Cervical flexion (degree) 7. Cervical rotation (degree) 8. ESR (mm/h) 9. Withdrawal due to side effect 10. Withdrawal due to ineffectiveness 11. Drop out for any reason

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

22

Corkill 1990

(Continued)

Notes

The active and placebo tablets were offered by Pharmacia Company and the author was supported by the Arthritis Foundation of New Zealand All continuous outcomes were presented as change from baseline Outcomes were also assessed at 4, 8, 12, 24, 36, weeks after the trial began Results were presented as means for each intervention group and 95% CI for difference between them SDs for each group were not given No significant difference was found between the intervention groups

Risk of bias Bias

Authors’ judgement

Support for judgement

Random sequence generation (selection Low risk bias)

“Patients were allocated to receive SASP or placebo using a randomisation protocol constructed from a table of random numbers in block four”

Allocation concealment (selection bias)

Insufficient information to permit judgment of Yes or No

Unclear risk

Blinding (performance bias and detection High risk bias) All outcomes

The trial is double-blind placebo-controlled. Although three participants broke the blinding, and two participants became aware of their therapy, the observers remained blinded

Incomplete outcome data (attrition bias) All outcomes

Unclear risk

Not mention how many dropped out from either active group or placebo group

Selective reporting (reporting bias)

Low risk

They reported the expected outcomes

Other bias

Low risk

The study appears to be free of other sources of bias

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

23

Davis 1989 Methods

Randomized allocation Assessment blind Parallel design Duration: 3 months Sample size at entry: 28 SSZ: 15 Placebo: 13 Clear description of drop-outs Outcomes included only those who completed the trial

Participants

Participants fulfilling the New York criteria for AS and with active disease defined as: 1. low back morning stiffness > 10 min or sleep disturbance due to pain or stiffness 2. ESR > 30 mm/hr or CRP > 20 ug/mL or IgA > 272 IU/mL Exclusion criteria: 1. history of inflammatory bowel diseases, Reiter’s disease or psoriasis 2. fused sacroiliac joints or more than three syndesmophytes in the lumber spine 3. positive rheumatoid factor Age (median and range) SSZ group: 35 (23 to 49) Placebo group: 40 (21 to 57) Male: 25/28 Duration of disease (median and range) SSZ group: 8.6 (1 to 30) Placebo group: 8.4 (1 to 25) With peripheral arthritis: 23%

Interventions

SSZ 2.0 g orally daily Placebo 2.0 g orally daily

Outcomes

1. Pain (100 mm visual analogue scale) 2. Spinal stiffness (100 mm visual analogue scale) 3. Sleep disturbance (event) 4. Occiput-to-wall test (cm) 5. Fingers-to-floor test (cm) 6. ESR (mm/hr) 7. CRP (ug/mL) 8. Withdrawal due to side effect 9. Withdrawal due to ineffectiveness 10. Drop out for any reason

Notes

The active and placebo tablets were offered by Pharmacia Company All continuous outcomes were presented as end point values We changed sleep disturbance into night pain (no bother) Participants who dropped out were counted as night pain (bother) The following were presented as median (interquartile range): Pain: SSZ group 20 (0 to 60) Placebo group 30 (20 to 70)

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

24

Davis 1989

(Continued)

Spinal stiffness: SSZ group 20 (0 to 50) Placebo group 20 (0 to 60) CRP: SSZ group 11 (6 to 23) Placebo group 28 (6 to 34) Other outcomes were presented as mean +/- 95% CI, from which we calculated SD We changed the unit from cm to mm, inch to cm and mg/L to ug/mL Risk of bias Bias

Authors’ judgement

Support for judgement

Random sequence generation (selection Unclear risk bias)

Insufficient information about the sequence generation process to permit judgment of Yes or No

Allocation concealment (selection bias)

The study did not address allocation concealment

Unclear risk

Blinding (performance bias and detection Low risk bias) All outcomes

This is placebo-controlled trial and the observer blinding was stressed

Incomplete outcome data (attrition bias) All outcomes

High risk

There are 30 participants recruited at the beginning of the trial, but only the participants who completed 3 months’ treatment were analysed

Selective reporting (reporting bias)

Low risk

They reported the expected outcomes

Other bias

Low risk

The study appears to be free of other sources of bias

Dougados 1986 Methods

Randomized allocation Double-blind and assessment blind Parallel design Duration: 6 months Sample size at the entry: 60 SSZ: 30 Placebo: 30 Clear description of withdrawal and drop out Treatment failure and drop out were analyzed according to intention-to-treat

Participants

Participants fulfilling the New York criteria for AS and requiring daily NSAIDs 8 had family history and/or minimal localised lesions of cutaneous psoriasis

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

25

Dougados 1986

(Continued)

None suffered from peripheral arthritis None had symptoms suggestive of inflammatory bowel diseases Age (median and confidence interval): SSZ group: 38.5 (13.1 to 65.3) Placebo group: 37.0 (19.0 to 59.0) Male: 77% Duration of disease (median and confidence interval): SSZ group: 10 (-8.8 to 33.2) Placebo group: 10 (-5.0 to 29.2) HLA B27: 85% With peripheral arthritis: 0 Interventions

SSZ 2.0 g orally daily Placebo 2.0 g orally daily

Outcomes

1. Treatment failure of overall patient assessment (event, including any withdrawal) 2. Score of daily NSAIDs (usual dosage as 10) 3. Pain (100 mm visual analogue scale) 4. Joint index (0 to 66) 5. Frequency of nocturnal awakening 6. Function index (0 to 40) 7. Schober’s test (cm) 8. Fingers-to-floor test (cm) 9. Chest expansion (cm) 10. ESR (mm/hr) 11. Withdrawal due to side effect 12. Withdrawal due to ineffectiveness 13. Drop out for any reason

Notes

There was no financial interest to report All continuous outcomes were presented as change from baseline and as median (95% CI) We assumed that mean was equal to median for each outcome and calculated SD from CI and sample size We changed ’treatment failure of overall patient assessment’ into ’improvement in patient global assessment’

Risk of bias Bias

Authors’ judgement

Support for judgement

Random sequence generation (selection Low risk bias)

The randomisation list was in blocks of six

Allocation concealment (selection bias)

The study supplied the active drug in numerical order

Low risk

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

26

Dougados 1986

(Continued)

Blinding (performance bias and detection Low risk bias) All outcomes

This is a double-blind (participants and investigators), placebo-controlled trial

Incomplete outcome data (attrition bias) All outcomes

High risk

For continuous outcome data, the analysis was made on those participants who completed the trials (47/60). The only one dichotomous outcome (treatment success or failure) was analyzed as intention-to-treat

Selective reporting (reporting bias)

Low risk

They reported the expected outcomes

Other bias

Low risk

The study appears to be free of other sources of bias

Feltelius 1986 Methods

Randomized allocation Blind outcome assessment: unclear Parallel design Duration: 12 weeks Sample size at entry: 37 SSZ: 18 Placebo: 19 Clear description of withdrawal and drop out Outcomes included only those who completed the trial

Participants

Participants fulfilling AS criteria of the American Rheumatism Association (1966) Other inclusion criteria: 1. HLA B27 (+) 2. ESR > 30 mm/hr or haptoglobin >= 2.0 g/L or orosomucoid >= 1.2 g/L 3. morning stiffness > 30 min or disturbed sleep due to pain or stiffness Exclusion criteria: 1. history of inflammatory bowel diseases, Reiter’s disease or psoriasis 2. with chronic infection, malignancy or other concomitant illness which might interfere the trial 3. Known allergy or intolerance to sulphonamide or salicylates 4. significant renal, hepatic, or hematological disease Age (median and actual range): SSZ group: 41.3 (25 to 57) Placebo group: 36.5 (19 to 57) Male: 76% Duration of disease (median and actual range): SSZ group: 12.1 (2 to 30) Placebo group: 10.4 (2 to 20) HLA B27: 100% With peripheral arthritis: 5%

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

27

Feltelius 1986

(Continued)

Interventions

SSZ initially 1.0 g/d orally, increased 0.5 to 1.0 g weekly until 3.0 g/d or the highest dose the patient could tolerate Placebo: the same as SSZ

Outcomes

1. Duration of morning stiffness (hr) 2. Spinal stiffness (100 mm visual analogue scale, change from baseline) 3. Pain (100 mm visual analogue scale) 4. Genral wellbeing (100 mm visual analogue scale) 5. Chest expansion (cm). Schober’s test (cm) 7. Sleep disturbance (not specify event or degree) 8. Sacroiliac pain (100 mm visual analogue scale) 9. ESR (mm/hr, end point and change from baseline) 10. Withdrawal for side effect 11. Withdrawal for ineffectiveness 12. Drop out for any reason

Notes

There was no financial interest to report Most outcomes were presented as graph which could not be transformed into figure Among continuous outcomes, only ESR was available for our analysis

Risk of bias Bias

Authors’ judgement

Support for judgement

Random sequence generation (selection Unclear risk bias)

Insufficient information about the sequence generation process to permit judgment of Yes or No

Allocation concealment (selection bias)

Insufficient information to permit judgment of Yes or No

Unclear risk

Blinding (performance bias and detection Low risk bias) All outcomes

This is double-blind, placebo-controlled trial. Although there is no information about who were blinded, it appeared to not affect the result much because only one outcome (ESR) was available for the present review

Incomplete outcome data (attrition bias) All outcomes

High risk

Only those who completed the trial were included in the analysis

Selective reporting (reporting bias)

Low risk

They reported the expected outcomes

Other bias

Low risk

The study appears to be free of other sources of bias

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

28

Kirwan 1993 Methods

Randomized allocation Double-blind Outcome assessment blind: unclear Parallel design Duration: 3 years Sample size at entry: 89 SSZ: 44 Placebo: 45 Clear description of withdrawal and drop-outs Intention-to-treat analysis

Participants

Participants fulfilling the New York criteria for AS Those who clinically had very little or no spinal mobility were excluded Age SSZ group: 44.1 +/- 13.3 Placebo group: 45.7 +/- 12.2 Male: 85% Duration of disease : SSZ group: 19.0 +/-12.0 Placebo group: 21.9 +/- 11.7 HLA B27: 98% With peripheral arthritis: 28%

Interventions

SSZ 1.0 g orally twice a day Placebo 1.0 g orally twice a day

Outcomes

Primary: 1. Modified Schober’s test (cm) 2. Chest expansion (cm) 3. Lateral cervical flexion (degree) Secondary 4. Function (Health assessment questionnaire) 5. Back pain (visual analogue scale) 6. Early morning back stiffness (visual analogue scale) 7. Consumption of anti-inflammatory drugs 8. Sleep disturbance (visual analogue scale) 9. Patient assessment of response (4 point scale) 10. Episodes of peripheral arthritis 11. Episodes of heel pain 12. Flares in general AS symptoms 13. Episodes of iritis 14. Withdrawal for side effect 15. Withdrawal for ineffectiveness 16. Drop out for any reason

Notes

There was no financial interest to report Primary outcomes were presented only as graph where no significant difference was reported between two groups For secondary outcomes, no figure was given except for episodes of peripheral arthritis

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

29

Kirwan 1993

(Continued)

which showed significant difference (0.298 episodes/yr in SSZ group vs 0.392 episodes/ yr in placebo group) (P 25 mm/hr, morning stiffness > 50 min Exclusion criteria: 1. history of intestinal tract inflammation, Reiter’s syndrome, psoriasis, malignant neoplasm 2. allergic to sulphasalazine and salicylates 3. pathologic tests of liver, kidney and hematology Age SSZ group: 38.25 +/- 6.3 Placebo group: 37.6 +/- 9.1 Male: 79% Duration of disease: not given With peripheral arthritis: 66%

Interventions

SSZ initially 1.0 g/d orally, increased 0.5 to 1.0 g weekly until 3.0 g/d or depending on the efficacy and tolerance Placebo: the same as SSZ

Outcomes

1. Duration of morning stiffness (hr) 2. Schober’s test (cm) 3. Chest expansion (cm) 4. Fingers-to-floor test (cm) 5. ESR (mm/hr) 6. Withdrawal due to side effect 7. Withdrawal due to ineffectiveness 8. Drop out for any reason

Notes

There was no financial interest to report All continuous outcomes were presented as end point values The author offered the full-text paper and English translation

Risk of bias Bias

Authors’ judgement

Random sequence generation (selection Unclear risk bias)

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Support for judgement Insufficient information about the sequence generation process to permit judgment of Yes or No 31

Krajnc 1990

(Continued)

Allocation concealment (selection bias)

Unclear risk

Insufficient information to permit judgment of Yes or No

Blinding (performance bias and detection Low risk bias) All outcomes

“Neither the physician nor the patients were acquainted about the sort of medicine they actually use”

Incomplete outcome data (attrition bias) All outcomes

High risk

“Number of patients interrupting therapy because of side effects, SSZ group 8/71 (2 skin rash, 3 nausea, 3 did not want to continue therapy), placebo group 5/24 (1 skin rash, 2 inefficiency, 2 did not want to continue)”. Did not use intention-to-treat analysis

Selective reporting (reporting bias)

Low risk

They reported the expected outcomes

Other bias

Low risk

The study appears to be free of other sources of bias

Nissila 1988 Methods

Randomized allocation Double-blind Blind outcome assessment: unclear Parallel design Duration: 26 weeks Sample size at the entry: 85 SSZ: 43 Placebo: 42 Clear description of withdrawal and drop out Intention-to-treat analysis: unclear

Participants

Participants fulfilling the New York criteria for AS Other inclusion criteria: 1. ESR >=30 mm/hr or CRP >= 20 mg/L 2. morning stiffness > 30 min 3. seronegative Exclusion criteria: 1. history or presence of intestinal disease, Reiter’s disease, psoriasis, chronic infection, malignancy and other disease which could interfere with the trial 2. allergic to sulfonamide or salicylates 3. with renal, hepatic and hematologic disease 4. advanced cases of ankylosed sacroiliac joints Age: SSZ group: 36.5 +/- 9.3 placebo group: 39.1 +/- 8.0 Male: 79%

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

32

Nissila 1988

(Continued)

Duration of disease: SSZ group: 5.4 +/- 7.3 Placebo group: 3.8 +/- 4.3 With peripheral arthritis: 68% Interventions

SSZ initially 1.0 g/d orally, increased 0.5 to 1.0 g weekly until 3.0 g/d or depending on the efficacy and tolerance Placebo: the same as SSZ

Outcomes

1. Duration of morning stiffness (hr) 2. Spinal stiffness (100 mm visual analogue scale) 3. Back pain (100 mm visual analogue scale) 4. Chest expansion (cm) 5. Schober’s test (cm) 6. Fingers-to-floor test (cm) 7. Occiput-to-wall test (cm) 8. Number of painful joints 9. Number of swollen joints 10. General wellbeing (100 mm visual analogue scale) 11. ESR (mm/hr) 12. CRP (ug/mL) 13. Withdrawal due to side effect 14. Withdrawal due to ineffectiveness 15. Drop out for any reason

Notes

There was no financial interest to report All continuous outcomes were presented as end point values We suspected that results of ’chest expansion’ were errors because they were impossible to be about 40 to 50 cm so we divided them by 10

Risk of bias Bias

Authors’ judgement

Support for judgement

Random sequence generation (selection Unclear risk bias)

Insufficient information about the sequence generation process to permit judgment of Yes or No

Allocation concealment (selection bias)

Insufficient information to permit judgment of Yes or No

Unclear risk

Blinding (performance bias and detection Low risk bias) All outcomes

“The doctors and patients were not informed of the drug codes”

Incomplete outcome data (attrition bias) All outcomes

There were 2 drop-outs in each group but no information about missing outcome results

Unclear risk

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

33

Nissila 1988

(Continued)

Selective reporting (reporting bias)

Low risk

They reported the expected outcomes

Other bias

Low risk

The study appears to be free of other sources of bias

Schmidt 2002 Methods

Randomized allocation Double-blind Outcome assessment blind: unclear Parallel design Duration: 26 weeks Sample size at the entry: 70 SSZ: 34 Placebo: 36 Clear description of withdrawal and drop out Intention-to-treat analysis: unclear

Participants

Participants fulfilling the modified New York criteria for AS Other inclusion criteria: 1. rheumatoid factor negative 2. morning stiffness >= 20 min 3. pain in the field of the axial skeleton >= 25 mm (100 mm visual analogue scale) Exclusion criteria: 1. known allergic to sulfonamide, salicylates and tartrazin 2. hematological diseases including thrombocytopenia (PLT < 140 G/L) and leukocytopenia (WBC < 4.0 G/L) 3. Severe liver diseases including GOT or GPT values > double of the upper norm limits 4. known renal diseases or increased creatinine 5. known G6PD deficiency, acute porphyria, asthma bronchial, pregnancy or urgent desire of an own baby, chronic inflammatory intestine diseases, RA, psoriasis, reactive arthritis, SLE, gout 6. corticosteroid treatment within the last month 7. severe diseases which were dangerous to participate in the trial Age: 27.5 +/- 8.3 SSZ group: 26.9 +/- 7.8 Placebo group: 28.0 +/- 8.8 Male: 87% Duration of disease: 16.7 +/- 7.2 SSZ group: 16.3 +/- 7.8 Placebo group: 17.1 +/- 6.6 With peripheral arthritis: 36%

Interventions

SSZ 1.0 g orally, 3 times a day Placebo: the same as SSZ

Outcomes

1. Back pain (100 mm visual analogue scale) 2. Nocturnal awakening (event)

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

34

Schmidt 2002

(Continued)

3. Enthesopathy index (0 to 90) 4. Duration of morning stiffness (hr) 5. Number of painful joints 6. Number of swollen joints 7. Spondylitis function index (0 to 44) 8. Effectiveness in patient assessment (event) 9. Effectiveness in physician assessment (event) 10. Schober’s test 11. Fingers-to-floor test (cm) 12. Occiput-to-wall test (cm) 13. Chin sternum distance (cm) 14. Chest expansion (cm) 15. ESR (mm/hr) 16. CRP (ug/mL) 17. Withdrawal due to side effect 18. Withdrawal due to ineffectiveness 19. Drop out for any reason Notes

There was no financial interest to report All continuous outcomes were presented as change from baseline We changed ’effectiveness in patient assessment’ to ’improvement in patient global assessment’ We changed ’effectiveness in physician assessment’ to ’improvement in physician global assessment’ Those dropping out were counted as not improved We changed ’nocturnal awakening’ to ’night pain (’no bother’ participants who dropped out were counted as night pain (bother) For ’number of painful joints’ and ’number of swollen joints’, no SD was given For ’fingers-to-floor test’ (cm), SD of SSZ group was missed German full-text was translated into English

Risk of bias Bias

Authors’ judgement

Support for judgement

Random sequence generation (selection Unclear risk bias)

Insufficient information about the sequence generation process to permit judgment of Yes or No

Allocation concealment (selection bias)

Unclear risk

Insufficient information to permit judgment of Yes or No

Blinding (performance bias and detection Unclear risk bias) All outcomes

“It treat of a prospective, randomized, double blind, placebo-controlled study with intention-to-treat analyse of the results”. But no information about which two were blinded

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

35

Schmidt 2002

(Continued)

Incomplete outcome data (attrition bias) All outcomes

High risk

There were 16/34 and 7/36 participants withdrew from the trial in SZZ and placebo, respectively. No information about the missing data

Selective reporting (reporting bias)

Low risk

They reported the expected outcomes

Other bias

Low risk

The study appears to be free of other sources of bias

Taylor 1991 Methods

Randomized allocation Double-blind Blind outcome assessment Parallel design Duration: 1 year Sample size at entry: 40 SSZ: 20 Placebo: 20 Clear description of withdrawal and drop out Intention-to-treat analysis: unclear

Participants

Participants fulfilling the New York criteria for AS Other inclusion criteria: 1. morning stiffness > 10 min or sleep disturbance due to pain 2. ESR > 30 mm/hr or PCR > 20 mg/L or IgA > 272 IU/mL 3. require regular NSAIDs or analgesics Exclusion criteria: 1. complete fusion of both sacroiliac joints or the presence of 3 bridged syndesmophytes on the lateral lumbar spine X-ray 2. Reiter’s disease, IBD or psoriasis Age: SSZ group: 34.7 +/-1.8 Placebo group: 39.4 +/- 2.2 Duration of disease: SSZ group: 11 +/- 1.6 Placebo group: 10.7 +/- 1.9 With peripheral arthritis: 15%

Interventions

SSZ orally, maximum tolerated dose or 2.0 g/d Placebo: the same as SSZ

Outcomes

1. Back pain (100 mm visual analogue scale) 2. Fingers-to-floor test (cm) 3. Chest expansion (cm) 4. Sleep disturbance (%) 5. Forced vital volume (L/min)

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

36

Taylor 1991

(Continued)

6. Occiput-to-wall test (cm) 7. Schober’s test (cm) 8. Spinal stiffness (100 mm visual analogue scale) 9. Reduction or stop NSAIDs (event) 10. Withdrawal due to side effect 11. Withdrawal due to ineffectiveness 12. Drop out for any reason Notes

This study was supported by the Haywood Rheumatism Research and Development Foundation and Pharmacia All continuous outcomes were presented as change from baseline We changed ’sleep disturbance (%)’ to ’night pain (no bother)’ We changed visual analogue scale 10 cm to visual analogue scale 100 mm and inch to cm Sacroiliac joints and lumbar spine radiograph score were assessed but no figures given

Risk of bias Bias

Authors’ judgement

Support for judgement

Random sequence generation (selection Low risk bias)

Randomization was performed within groups of 4 by Pharmacia

Allocation concealment (selection bias)

Low risk

The Department Pharmacia was not revealed to the participants or assessors

Blinding (performance bias and detection Low risk bias) All outcomes

“All observers were blinded to which therapy the patients were taking”. “The clinical measurements were performed by a single observer”. “The erosion count was confirmed for each patients by 2 observers blinded to treat”

Incomplete outcome data (attrition bias) All outcomes

Unclear risk

The study did not address the issue of missing data of those drop-outs and did not use principle of intention-to-treat

Selective reporting (reporting bias)

High risk

The study did not report the scores of patients’ general wellbeing

Other bias

Low risk

The study appears to be free of other sources of bias

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

37

Winkler 1989 Methods

Randomized allocation Single-blind Outcome assessment blind: unclear Parallel design Duration: 24 weeks Sample size at the entry: 63 SSZ: 31 Placebo: 32 Clear description of withdrawal and drop out Outcomes included only those who completed the trial

Participants

Participants fulfilling the New York criteria for AS Other inclusion criteria: 1. Duration of disease > 1 yr 2. active disease, e.g. severe pain, sleep disturbed and morning stiffness, did not respond to NSAIDs Exclusion criteria: 1. history of sulphonamide allergy or intolerance 2. psoriasis, reactive arthritis, inflammatory bowel diseases, chronic infection, malignancy, liver, renal or hematological diseases Age (median and range) SSZ group: 40.5 (25 to 66) placebo group: 40.2 (27 to 60) Male: 83% Duration of disease (median and range): SSZ group: 11.8 (1 to 30) Placebo group: 10.2 (2 to 28) With peripheral arthritis: 33%

Interventions

SSZ 2.0 g/d orally Placebo: the same as SSZ

Outcomes

1. ESR (mm/hr) 2. Duration of morning stiffness (hr) 3. Back pain (100 mm visual analogue scale) 4. Score of sleep disturbance (0 to 4) 5. Chest expansion (cm) 6. Schober’s test (cm) 7. Fingers-to-floor test (cm) 8. Patient assessment of disease severity (100 mm visual analogue scale) 9. Articular index (score 0 to 36) 10. Degree of joint swelling (0 to 36) 11. Withdrawal for side effect 12. Withdrawal for ineffectiveness 13. Drop out for any reason

Notes

There was no financial interest to report All continuous outcomes were presented as end point values We changed min to hr

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

38

Winkler 1989

(Continued)

Risk of bias Bias

Authors’ judgement

Support for judgement

Random sequence generation (selection Unclear risk bias)

Insufficient information about the sequence generation process to permit judgment of Yes or No

Allocation concealment (selection bias)

Insufficient information to permit judgment of Yes or No

Unclear risk

Blinding (performance bias and detection Unclear risk bias) All outcomes

The study did not address this item

Incomplete outcome data (attrition bias) All outcomes

High risk

The result were analyzed with intentionto-treat principle, 5 participants from active group and 9 participants from placebo group dropped out, SSZ treatment was withdrawn due to adverse side-effects, placebo treatment because of its effectiveness. Reason for missing outcome data likely to be related to true outcome, with imbalance in reasons for missing data across intervention groups

Selective reporting (reporting bias)

Low risk

They reported the expected outcomes

Other bias

Low risk

The study appears to be free of other sources of bias

AS - ankylosing spondylitis CI - confidence interval CRP - C-reactive protein ESR - erythrocyte sedimentation rate GOT - glutamic-oxaloacetic transaminase GPT - glutamic-pyruvic transaminase G6PD - glucose-6-phosphate dehydrogenease g/d - grams per day HLA B27 - human leukocyte antigen B27 Ig A - immunoglobulin A mm/hr - millimetre per hour NSAID - non-steroidal anti-inflammatory drug RA - rheumatoid arthritis SAPA - sulfasalazine SD - standard deviation SLE - systemic lupus erythematosus SSZ - sulfasalazineug - microgram Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

39

Characteristics of excluded studies [ordered by study ID]

Study

Reason for exclusion

Benitez-Del-Castillo

No relevant outcomes

Braun 2006

This was a multicenter, double-blind, placebo controlled RCT. The participants were 242 patients with inflammatory back pain due to undifferentiated spondyloarthritis and early AS, including 12 AS patients. We were unable to get the results for AS patients alone

Braun 2011

This was a multicenter, double-blind, placebo controlled RCT comparing etanercept with SSZ in patients with AS. We excluded it because there is another separate Cochrane review addressing biologics in AS

Burgos-Vargas 2002

The participants were 33 patients with juvenile onset spondyloarthropathy, including 13 AS patients. No outcome specific to AS patients was presented

Deng 2009

No relevant outcomes

NCT00953979

This is a protocol; attempts to contact the authors to access full-text have been unsuccessful so far

Song 2011

The participants were 76 patients with axial spondyloarthritis; no outcome specific to AS patients were presented

Xu 2008

This was a RCT, comparing the effects of Bushen Tongdu Decoction (Chinese traditional medicine) and SSZ on serum tumour necrosis factor-alpha and transforming growth factor beta1 in patients with AS. Because the efficacy of Bushen Tongdu Decoction is unclear in AS, we excluded this study

Zhao 2006

This was a RCT about long-term effectiveness of leflunomide compared with SSZ in treatment of AS. The efficacy of leflunomide in AS is unclear and therefore we decided to exclude this trial after a discussion

Zhao 2009

This was a RCT, comparing the effects of etanercept and SSZ. We excluded it because there was another separate Cochrane review addressing biologics in AS

AS - ankylosing spondylitis RCT - randomized controlled trials SSZ - sulfasalazine

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

40

DATA AND ANALYSES

Comparison 1. Sulfasalazine versus placebo

Outcome or subgroup title 1 Spondylitis function index (Score 0 to 40, 0 to 44, 0 = the best) 1.1 End point 1.2 Change from baseline 2 Spondylitis function index (2nd analysis) (score 0 to 40, 0 to 44, 0 = the best) 2.1 Change from baseline 3 Improvement in back pain 4 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain) 4.1 End point 4.2 Change from baseline 5 Back pain (2nd analysis) (VAS 100 mm, 0 = no pain, 100 = severe pain) 5.1 Change from baseline 5.2 End point 6 Night pain (% no pain) 7 Score of sleep disturbance (end point) (0 to 4, 0 = no disturbance, 4 = severe disturbance) 8 Frequency of nocturnal awakening (change from baseline) 9 Score of daily NSAIDs (change from baseline, usual dosage as 10) 10 Reducing or stopping NSAIDs 11 Chest expansion (cm) 11.1 End point 11.2 Change from baseline 12 Chest expansion (2nd analysis) (cm) 12.1 Change from baseline 12.2 End point 13 Forced vital volume (change from baseline) (L/min) 14 (Modified) Schober’s test (cm) 14.1 End point 14.2 Change from baseline

No. of studies

No. of participants

3

297

Mean Difference (IV, Fixed, 95% CI)

0.14 [-1.18, 1.46]

1 2 3

203 94 297

Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI)

-1.0 [-2.88, 0.88] 1.27 [-0.60, 3.13] 0.20 [-0.77, 1.18]

3 1 6

297 454

Mean Difference (IV, Fixed, 95% CI) Risk Ratio (M-H, Fixed, 95% CI) Mean Difference (IV, Random, 95% CI)

0.20 [-0.77, 1.18] Totals not selected -2.96 [-6.33, 0.41]

3 3 6

327 127 454

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

-3.47 [-10.17, 3.22] -4.29 [-12.15, 3.56] -2.38 [-5.78, 1.03]

4 2 4 1

330 124 404

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Risk Ratio (M-H, Fixed, 95% CI) Mean Difference (IV, Random, 95% CI)

-1.24 [-4.86, 2.37] -6.59 [-14.73, 1.55] 1.04 [0.75, 1.43] Totals not selected

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

Risk Ratio (M-H, Fixed, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Fixed, 95% CI)

Totals not selected 0.30 [0.17, 0.43] 0.31 [-0.03, 0.66] 0.30 [0.16, 0.44] 0.31 [0.17, 0.44]

1 7 4 3 7

536 409 127 536

Statistical method

Effect size

4 3 1

330 206

Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Random, 95% CI)

0.29 [0.15, 0.44] 0.46 [0.01, 0.90] Totals not selected

7 4 3

536 409 127

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

0.18 [-0.11, 0.46] 0.07 [-0.14, 0.29] 0.50 [0.44, 0.56]

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

41

15 (Modified) Schober’s test (2nd analysis) (cm) 15.1 Change from baseline 15.2 End point 16 Occiput-to-wall test (cm) 16.1 End point 16.2 Change from baseline 17 Occiput-to-wall test (2nd analysis) (cm) 17.1 Change from baseline 17.2 End point 18 Fingers-to-floor test (cm) 18.1 End point 18.2 Change from baseline 19 Fingers-to-floor test (2nd analysis) (cm) 19.1 Change from baseline 19.2 End point 20 Chin sternum distance (change from baseline) (cm) 21 Joint pain/tenderness score (0 to 198, the higher the score the more severe the disease) or number 21.1 End point 22 Joint pain/tenderness score (2nd analysis) (0 to 198, the higher the score the more severe the disease) 22.1 Change from baseline 22.2 End point 23 Joint swelling score (0 to 198, the higher score the more severe the disease) or number 23.1 End point 24 Joint swelling score (2nd analysis) (0 to 198, the higher the score the more severe the disease) 24.1 Change from baseline 24.2 End point 25 Dactylitis score (0 to 3, 0 = normal, 3 = severe) 26 Dactylitis score (2nd analysis) (0 to 3, 0 = normal, 3 = severe) 27 Enthesopathy index (0 to 90, 0 to 66, 0 to 90, the higher the score the more severe the disease) 27.1 End point 27.2 Change from baseline

7

536

Mean Difference (IV, Random, 95% CI)

0.12 [-0.21, 0.45]

4 3 5 3 2 5

330 206 386 306 80 386

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

0.19 [-0.35, 0.74] 0.10 [-0.15, 0.35] 0.08 [-0.90, 1.06] -0.63 [-1.33, 0.07] 0.68 [0.31, 1.05] -0.03 [-0.84, 0.79]

3 2 7 5 2 7

283 103 517 437 80 517

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

0.30 [-0.43, 1.03] -0.75 [-1.49, -0.02] -1.03 [-2.93, 0.87] -0.64 [-2.77, 1.49] -2.54 [-6.75, 1.67] -0.71 [-2.18, 0.75]

3 4 1

283 234

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Fixed, 95% CI)

-0.96 [-2.79, 0.88] -0.28 [-2.72, 2.15] Totals not selected

2

278

Std. Mean Difference (IV, Random, 95% CI)

-0.04 [-0.37, 0.29]

2 2

278 278

Std. Mean Difference (IV, Random, 95% CI) Std. Mean Difference (IV, Random, 95% CI)

-0.04 [-0.37, 0.29] -0.15 [-0.38, 0.09]

1 1 2

203 75 278

Std. Mean Difference (IV, Random, 95% CI) Std. Mean Difference (IV, Random, 95% CI) Std. Mean Difference (IV, Random, 95% CI)

-0.10 [-0.38, 0.17] -0.26 [-0.72, 0.19] 0.04 [-0.20, 0.27]

2 2

278 278

Std. Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

0.04 [-0.20, 0.27] 0.0 [-0.29, 0.29]

1 1 1

203 75

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

0.0 [-0.97, 0.97] 0.0 [-0.30, 0.30] Totals not selected

Mean Difference (IV, Random, 95% CI)

Totals not selected

1 3

297

Std. Mean Difference (IV, Random, 95% CI)

0.10 [-0.13, 0.33]

1 2

203 94

Std. Mean Difference (IV, Random, 95% CI) Std. Mean Difference (IV, Random, 95% CI)

0.06 [-0.22, 0.33] 0.19 [-0.29, 0.68]

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

42

28 Enthesopathy index (2nd analysis) (0 to 90, 0 to 66, 0 to 90, the higher score the more severe the disease) 28.1 Change from baseline 29 Spondylitis articular index (0 to 90, the higher score the more severe the disease) 30 Spondylitis articular index (2nd analysis) (0 to 90, the higher score the more severe the disease) 31 Improvement in patient global assessment 32 Patient assessment of disease severity (end point) (VAS 100 mm, 0 = very good, 100 = very poor) 33 General well-being (end point) (VAS 100 mm, 0 = very good, 100 = very poor) 34 Improvement in physician global assessment 35 Response to treatment (based on both patient and physician assessment) 36 Duration of morning stiffness (hr) 36.1 End point 36.2 Change from baseline 37 Duration of morning stiffness (2nd analysis) (hr) 37.1 Change from base line 37.2 End point 38 Morning stiffness (end point) (VAS 100 mm, 0 = no stiffness, 100 = severe stiffness) 38.1 End point

3

297

Std. Mean Difference (IV, Random, 95% CI)

0.11 [-0.12, 0.34]

3 1

297

Std. Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

0.11 [-0.12, 0.34] Totals not selected

Mean Difference (IV, Random, 95% CI)

Totals not selected

Risk Ratio (IV, Random, 95% CI)

1.52 [0.78, 2.97]

1

Mean Difference (IV, Fixed, 95% CI)

Totals not selected

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

Risk Ratio (IV, Random, 95% CI)

1.34 [0.59, 3.08]

Risk Ratio (M-H, Random, 95% CI)

Totals not selected

38.2 Change from baseline 39 Improvement in morning stiffness 40 Erythrocyte sedimentation rate (mm/hr) 40.1 End point 40.2 Change from baseline 41 Erythrocyte sedimentation rate (2nd analysis) (mm/hr) 41.1 Change from baseline 41.2 End point 42 C-reactive protein (ug/ml)

1

3

2

394

334

1

5

456

Mean Difference (IV, Random, 95% CI)

-0.20 [-0.39, -0.01]

4 1 5

409 47 456

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

-0.10 [-0.25, 0.05] -0.39 [-0.48, -0.30] -0.20 [-0.39, -0.00]

2 3 2

250 206 108

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

-0.39 [-0.48, -0.30] -0.10 [-0.25, 0.05] -13.89 [-22.54, -5. 24]

1

75

Mean Difference (IV, Random, 95% CI)

1 1

33

Mean Difference (IV, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI)

-12.00 [-23.78, -4. 22] -13.5 [-30.00, 5.00] Totals not selected

8

560

Mean Difference (IV, Random, 95% CI)

-5.66 [-10.40, -0.93]

6 2 8

466 94 560

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

-7.07 [-14.39, 0.25] -3.09 [-4.84, -1.35] -4.79 [-8.80, -0.78]

4 4 3

326 234 325

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

-3.11 [-4.62, -1.60] -9.56 [-22.03, 2.91] -1.42 [-3.76, 0.92]

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

43

42.1 End pointSub-category 42.2 Change from baseline 43 C-reactive protein (2nd analysis) (ug/ml) 43.1 Change from baseline 43.2 End point 44 Withdrawal due to side effect 45 Withdrawal due to ineffectiveness 46 Drop-out for any reason 47 Serious adverse events

2 1 3

278 47 325

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

-1.76 [-7.42, 3.90] -2.5 [-4.70, -0.30] -1.39 [-3.85, 1.07]

2 1 11 10

250 75 895 833

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Risk Ratio (M-H, Fixed, 95% CI) Risk Ratio (M-H, Fixed, 95% CI)

-1.02 [-3.37, 1.33] -7.0 [-16.80, 2.80] 1.50 [1.04, 2.15] 0.87 [0.53, 1.42]

10 1

833 264

Risk Ratio (M-H, Fixed, 95% CI) Peto Odds Ratio (Peto, Fixed, 95% CI)

1.33 [1.03, 1.73] 7.50 [0.15, 378.16]

Comparison 2. Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values)

Outcome or subgroup title 1 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain) 2 Score of sleep disturbance (0 to 4, 0 = no disturbance, 4 = severe disturbance) 3 Chest expansion (cm) 4 Schober’s test (cm) 5 Fingers-to-floor test (cm) 6 Articular index (0 to 66, the higher the score the more severe the disease) 7 Degree of joint swelling (0 to 66, the higher the score the more severe the disease) 8 Patient assessment of disease severity (VAS 100 mm, 0 = very good, 100 = very poor) 9 Duration of morning stiffness (hr) 10 Erythrocyte sedimentation rate (mm/hr)

No. of studies

No. of participants

Statistical method

Effect size

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

1 1 1 1

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

Totals not selected Totals not selected Totals not selected Totals not selected

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

44

Comparison 3. Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values)

No. of studies

Outcome or subgroup title 1 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain) 2 Score of sleep disturbance (0 to 4, 0 = no disturbance, 4 = severe disturbance) 3 Chest expansion (cm) 4 Schober’s test (cm) 5 Fingers-to-floor test (cm) 6 Patient assessment of disease severity (VAS 100 mm, 0 = very good, 100 = very poor) 7 Duration of morning stiffness (hr) 8 Erythrocyte sedimentation rate (mm/hr)

No. of participants

Statistical method

Effect size

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

1 1 1 1

Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)

Totals not selected Totals not selected Totals not selected Totals not selected

1

Mean Difference (IV, Random, 95% CI)

Totals not selected

1

Mean Difference (IV, Fixed, 95% CI)

Totals not selected

Analysis 1.1. Comparison 1 Sulfasalazine versus placebo, Outcome 1 Spondylitis function index (Score 0 to 40, 0 to 44, 0 = the best). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 1 Spondylitis function index (Score 0 to 40, 0 to 44, 0 = the best)

Study or subgroup

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

95

14.8 (6.9)

108

15.8 (6.7)

Weight

IV,Fixed,95% CI

Mean Difference IV,Fixed,95% CI

1 End point Clegg 1996

Subtotal (95% CI)

95

108

49.6 %

-1.00 [ -2.88, 0.88 ]

49.6 %

-1.00 [ -2.88, 0.88 ]

Heterogeneity: not applicable Test for overall effect: Z = 1.04 (P = 0.30) 2 Change from baseline Dougados 1986

23

-4 (32.8)

24

-2.3 (28.87)

0.6 %

-1.70 [ -19.39, 15.99 ]

Schmidt 2002

18

4.4 (3.7)

29

3.1 (2.1)

49.8 %

1.30 [ -0.57, 3.17 ]

50.4 %

1.27 [ -0.60, 3.13 ]

Subtotal (95% CI)

41

53 -10

-5

Favours SSZ

0

5

10

Favours Placebo

(Continued . . . )

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

45

(. . . Study or subgroup

SSZ N

Heterogeneity:

Chi2

Mean Difference

Placebo Mean(SD)

= 0.11, df = 1 (P = 0.74);

I2

N

Mean(SD)

Weight

IV,Fixed,95% CI

Continued)

Mean Difference IV,Fixed,95% CI

=0.0%

Test for overall effect: Z = 1.33 (P = 0.18)

Total (95% CI)

136

161

100.0 %

0.14 [ -1.18, 1.46 ]

Heterogeneity: Chi2 = 2.93, df = 2 (P = 0.23); I2 =32% Test for overall effect: Z = 0.21 (P = 0.83) Test for subgroup differences: Chi2 = 2.82, df = 1 (P = 0.09), I2 =65%

-10

-5

0

Favours SSZ

5

10

Favours Placebo

Analysis 1.2. Comparison 1 Sulfasalazine versus placebo, Outcome 2 Spondylitis function index (2nd analysis) (score 0 to 40, 0 to 44, 0 = the best). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 2 Spondylitis function index (2nd analysis) (score 0 to 40, 0 to 44, 0 = the best)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Fixed,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Fixed,95% CI

Clegg 1996

95

-0.7 (3.6)

108

-0.5 (4.7)

72.6 %

-0.20 [ -1.34, 0.94 ]

Dougados 1986

23

-4 (32.8)

24

-2.3 (28.87)

0.3 %

-1.70 [ -19.39, 15.99 ]

Schmidt 2002

18

4.4 (3.7)

29

3.1 (2.1)

27.1 %

1.30 [ -0.57, 3.17 ]

100.0 %

0.20 [ -0.77, 1.18 ]

1 Change from baseline

Total (95% CI)

136

161

Heterogeneity: Chi2 = 1.84, df = 2 (P = 0.40); I2 =0.0% Test for overall effect: Z = 0.41 (P = 0.68) Test for subgroup differences: Not applicable

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

5

10

Favours Placebo

46

Analysis 1.3. Comparison 1 Sulfasalazine versus placebo, Outcome 3 Improvement in back pain. Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 3 Improvement in back pain

Study or subgroup

Clegg 1996

SSZ

Placebo

n/N

n/N

31/131

36/133

Risk Ratio

Risk Ratio

M-H,Fixed,95% CI

M-H,Fixed,95% CI 0.87 [ 0.58, 1.32 ]

0.1 0.2

0.5

1

Favours Placebo

2

5

10

Favours SSZ

Analysis 1.4. Comparison 1 Sulfasalazine versus placebo, Outcome 4 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 4 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Nissila 1988

38

21 (24)

37

33 (25)

8.1 %

-12.00 [ -23.10, -0.90 ]

Winkler 1989

26

38.5 (11.2)

23

41.9 (13.7)

17.0 %

-3.40 [ -10.46, 3.66 ]

Clegg 1996

95

50.3 (23.4)

108

48.9 (23.4)

19.4 %

1.40 [ -5.05, 7.85 ]

44.5 %

-3.47 [ -10.17, 3.22 ]

1 End point

Subtotal (95% CI)

159

168

Heterogeneity: Tau2 = 18.46; Chi2 = 4.28, df = 2 (P = 0.12); I2 =53% Test for overall effect: Z = 1.02 (P = 0.31) 2 Change from baseline Taylor 1991

16

-24.3 (37.6)

17

-5.6 (21.3)

2.5 %

-18.70 [ -39.72, 2.32 ]

Dougados 1986

23

-9 (119.83)

24

-6.5 (185.04)

0.1 %

-2.50 [ -91.26, 86.26 ]

Schmidt 2002

18

-4.2 (3.4)

29

-1.9 (4)

52.8 %

-2.30 [ -4.44, -0.16 ]

-100

-50

Favours SSZ

0

50

100

Favours Placebo

(Continued . . . )

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

47

(. . . Study or subgroup

SSZ N

Subtotal (95% CI)

Mean Difference

Placebo Mean(SD)

57

N

Mean(SD)

Weight

IV,Random,95% CI

Continued)

Mean Difference IV,Random,95% CI

70

55.5 %

-4.29 [ -12.15, 3.56 ]

100.0 %

-2.96 [ -6.33, 0.41 ]

Heterogeneity: Tau2 = 17.27; Chi2 = 2.31, df = 2 (P = 0.31); I2 =14% Test for overall effect: Z = 1.07 (P = 0.28)

Total (95% CI)

216

238

Heterogeneity: Tau2 = 4.40; Chi2 = 6.60, df = 5 (P = 0.25); I2 =24% Test for overall effect: Z = 1.72 (P = 0.085) Test for subgroup differences: Chi2 = 0.02, df = 1 (P = 0.88), I2 =0.0%

-100

-50

0

Favours SSZ

50

100

Favours Placebo

Analysis 1.5. Comparison 1 Sulfasalazine versus placebo, Outcome 5 Back pain (2nd analysis) (VAS 100 mm, 0 = no pain, 100 = severe pain). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 5 Back pain (2nd analysis) (VAS 100 mm, 0 = no pain, 100 = severe pain)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

-6.1 (18.5)

108

-5.3 (19.9)

24.4 %

-0.80 [ -6.08, 4.48 ]

Dougados 1986

23

-9 (119.83)

24

-6.5 (185.04)

0.1 %

-2.50 [ -91.26, 86.26 ]

Schmidt 2002

18

-4.2 (3.4)

29

-1.9 (4)

48.1 %

-2.30 [ -4.44, -0.16 ]

Taylor 1991

16

24.3 (37.6)

17

5.6 (21.3)

2.5 %

18.70 [ -2.32, 39.72 ]

75.2 %

-1.24 [ -4.86, 2.37 ]

1 Change from baseline

Subtotal (95% CI)

152

178

Heterogeneity: Tau2 = 3.92; Chi2 = 3.99, df = 3 (P = 0.26); I2 =25% Test for overall effect: Z = 0.67 (P = 0.50) 2 End point Nissila 1988

38

21 (24)

37

33 (25)

8.1 %

-12.00 [ -23.10, -0.90 ]

Winkler 1989

26

38.5 (11.2)

23

41.9 (13.7)

16.7 %

-3.40 [ -10.46, 3.66 ]

-100

-50

Favours SSZ

0

50

100

Favours Placebo

(Continued . . . )

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

48

(. . . Study or subgroup

SSZ N

Subtotal (95% CI)

Mean Difference

Placebo Mean(SD)

N

64

Mean(SD)

Weight

IV,Random,95% CI

Continued)

Mean Difference IV,Random,95% CI

60

24.8 %

-6.59 [ -14.73, 1.55 ]

100.0 %

-2.38 [ -5.78, 1.03 ]

Heterogeneity: Tau2 = 14.46; Chi2 = 1.64, df = 1 (P = 0.20); I2 =39% Test for overall effect: Z = 1.59 (P = 0.11)

Total (95% CI)

216

238

Heterogeneity: Tau2 = 5.09; Chi2 = 7.17, df = 5 (P = 0.21); I2 =30% Test for overall effect: Z = 1.37 (P = 0.17) Test for subgroup differences: Chi2 = 1.38, df = 1 (P = 0.24), I2 =28%

-100

-50

0

50

Favours SSZ

100

Favours Placebo

Analysis 1.6. Comparison 1 Sulfasalazine versus placebo, Outcome 6 Night pain (% no pain). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 6 Night pain (% no pain)

Study or subgroup

SSZ

Placebo

n/N

n/N

Clegg 1996

21/131

24/133

48.0 %

0.89 [ 0.52, 1.51 ]

Davis 1989

11/15

8/15

16.1 %

1.38 [ 0.78, 2.41 ]

9/34

8/36

15.7 %

1.19 [ 0.52, 2.73 ]

10/20

10/20

20.2 %

1.00 [ 0.54, 1.86 ]

200

204

100.0 %

1.04 [ 0.75, 1.43 ]

Schmidt 2002 Taylor 1991

Total (95% CI)

Risk Ratio

Weight

M-H,Fixed,95% CI

Risk Ratio M-H,Fixed,95% CI

Total events: 51 (SSZ), 50 (Placebo) Heterogeneity: Chi2 = 1.41, df = 3 (P = 0.70); I2 =0.0% Test for overall effect: Z = 0.22 (P = 0.82) Test for subgroup differences: Not applicable

0.1 0.2

0.5

Favours Placebo

1

2

5

10

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

49

Analysis 1.7. Comparison 1 Sulfasalazine versus placebo, Outcome 7 Score of sleep disturbance (end point) (0 to 4, 0 = no disturbance, 4 = severe disturbance). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 7 Score of sleep disturbance (end point) (0 to 4, 0 = no disturbance, 4 = severe disturbance)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

26

1.38 (0.9)

23

1.74 (1)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI -0.36 [ -0.90, 0.18 ]

-4

-2

0

Favours SSZ

2

4

Favours Placebo

Analysis 1.8. Comparison 1 Sulfasalazine versus placebo, Outcome 8 Frequency of nocturnal awakening (change from baseline). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 8 Frequency of nocturnal awakening (change from baseline)

Mean Difference

Study or subgroup

SSZ

Placebo

N

Mean(SD)

N

Mean(SD)

Dougados 1986

23

0 (6.05)

24

0 (11.6)

IV,Random,95% CI

IV,Random,95% CI 0.0 [ -5.26, 5.26 ]

-100

-50

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

50

100

Favours Placebo

50

Analysis 1.9. Comparison 1 Sulfasalazine versus placebo, Outcome 9 Score of daily NSAIDs (change from baseline, usual dosage as 10). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 9 Score of daily NSAIDs (change from baseline, usual dosage as 10)

Mean Difference

Study or subgroup

SSZ

Placebo

N

Mean(SD)

N

Mean(SD)

Dougados 1986

23

-6.7 (35.82)

24

-2.9 (32.21)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI -3.80 [ -23.30, 15.70 ]

-100

-50

0

Favours SSZ

50

100

Favours Placebo

Analysis 1.10. Comparison 1 Sulfasalazine versus placebo, Outcome 10 Reducing or stopping NSAIDs. Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 10 Reducing or stopping NSAIDs

Study or subgroup

Taylor 1991

SSZ

Placebo

n/N

n/N

6/20

7/20

Risk Ratio

Risk Ratio

M-H,Fixed,95% CI

M-H,Fixed,95% CI 0.86 [ 0.35, 2.10 ]

0.1 0.2

0.5

Favours Placebo

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

1

2

5

10

Favours SSZ

51

Analysis 1.11. Comparison 1 Sulfasalazine versus placebo, Outcome 11 Chest expansion (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 11 Chest expansion (cm)

Study or subgroup

Placebo

Mean Difference

SSZ

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

3.1 (1.9)

108

3 (1.9)

6.3 %

0.10 [ -0.42, 0.62 ]

Krajnc 1990

63

3.9 (1.2)

19

3.7 (1.2)

4.6 %

0.20 [ -0.42, 0.82 ]

Nissila 1988

38

5.2 (2.5)

37

4.2 (1.3)

2.1 %

1.00 [ 0.10, 1.90 ]

Winkler 1989

26

3.59 (1.7)

23

3.13 (1.5)

2.2 %

0.46 [ -0.44, 1.36 ]

15.1 %

0.31 [ -0.03, 0.66 ]

1 End point

Subtotal (95% CI)

222

187

Heterogeneity: Tau2 = 0.00; Chi2 = 3.11, df = 3 (P = 0.37); I2 =4% Test for overall effect: Z = 1.77 (P = 0.076) 2 Change from baseline Dougados 1986

23

1 (17.66)

24

1 (19.33)

0.0 %

0.0 [ -10.58, 10.58 ]

Schmidt 2002

18

0.9 (0.2)

29

0.6 (0.3)

84.4 %

0.30 [ 0.16, 0.44 ]

Taylor 1991

16

0 (3)

17

-0.51 (2.69)

0.5 %

0.51 [ -1.44, 2.46 ]

84.9 %

0.30 [ 0.16, 0.44 ]

100.0 %

0.30 [ 0.17, 0.43 ]

Subtotal (95% CI)

57

70

Heterogeneity: Tau2 = 0.0; Chi2 = 0.05, df = 2 (P = 0.98); I2 =0.0% Test for overall effect: Z = 4.14 (P = 0.000035)

Total (95% CI)

279

257

Heterogeneity: Tau2 = 0.0; Chi2 = 3.16, df = 6 (P = 0.79); I2 =0.0% Test for overall effect: Z = 4.51 (P < 0.00001) Test for subgroup differences: Chi2 = 0.00, df = 1 (P = 0.95), I2 =0.0%

-10

-5

Favours Placebo

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

5

10

Favours SSZ

52

Analysis 1.12. Comparison 1 Sulfasalazine versus placebo, Outcome 12 Chest expansion (2nd analysis) (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 12 Chest expansion (2nd analysis) (cm)

Study or subgroup

Placebo

Mean Difference

SSZ

Weight

IV,Fixed,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Fixed,95% CI

Clegg 1996

95

-0.4 (5.9)

108

0 (3.8)

0.9 %

-0.40 [ -1.79, 0.99 ]

Dougados 1986

23

1 (17.66)

24

1 (19.33)

0.0 %

0.0 [ -10.58, 10.58 ]

Schmidt 2002

18

0.9 (0.2)

29

0.6 (0.3)

89.2 %

0.30 [ 0.16, 0.44 ]

Taylor 1991

16

0 (3)

17

-0.51 (2.69)

0.5 %

0.51 [ -1.44, 2.46 ]

90.7 %

0.29 [ 0.15, 0.44 ]

1 Change from baseline

Subtotal (95% CI)

152

178

Heterogeneity: Chi2 = 1.02, df = 3 (P = 0.80); I2 =0.0% Test for overall effect: Z = 4.06 (P = 0.000050) 2 End point Krajnc 1990

63

3.9 (1.2)

19

3.7 (1.2)

4.8 %

0.20 [ -0.42, 0.82 ]

Nissila 1988

38

5.2 (2.5)

37

4.2 (1.3)

2.3 %

1.00 [ 0.10, 1.90 ]

Winkler 1989

26

3.59 (1.7)

23

3.13 (1.5)

2.3 %

0.46 [ -0.44, 1.36 ]

79

9.3 %

0.46 [ 0.01, 0.90 ]

257

100.0 %

0.31 [ 0.17, 0.44 ]

Subtotal (95% CI)

127

Heterogeneity: Chi2 = 2.07, df = 2 (P = 0.35); I2 =4% Test for overall effect: Z = 2.03 (P = 0.043)

Total (95% CI)

279

Heterogeneity: Chi2 = 3.57, df = 6 (P = 0.74); I2 =0.0% Test for overall effect: Z = 4.48 (P < 0.00001) Test for subgroup differences: Chi2 = 0.47, df = 1 (P = 0.49), I2 =0.0%

-0.2

-0.1

Favours Placebo

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

0.1

0.2

Favours SSZ

53

Analysis 1.13. Comparison 1 Sulfasalazine versus placebo, Outcome 13 Forced vital volume (change from baseline) (L/min). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 13 Forced vital volume (change from baseline) (L/min)

Study or subgroup

Taylor 1991

Placebo

Mean Difference

SSZ

N

Mean(SD)

N

Mean(SD)

16

89 (519)

17

-123 (1013)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI 212.00 [ -332.57, 756.57 ]

-1000

-500

0

500

Favours Placebo

1000

Favours SSZ

Analysis 1.14. Comparison 1 Sulfasalazine versus placebo, Outcome 14 (Modified) Schober’s test (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 14 (Modified) Schober’s test (cm)

Study or subgroup

Placebo

Mean Difference

SSZ

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

12.4 (1.5)

108

12.4 (1.5)

18.7 %

0.0 [ -0.41, 0.41 ]

Krajnc 1990

63

3.2 (0.9)

19

3.1 (0.7)

19.7 %

0.10 [ -0.29, 0.49 ]

Nissila 1988

38

3.6 (0.8)

37

3.4 (0.8)

20.6 %

0.20 [ -0.16, 0.56 ]

Winkler 1989

26

1.88 (1.6)

23

2.3 (1.3)

8.7 %

-0.42 [ -1.23, 0.39 ]

67.8 %

0.07 [ -0.14, 0.29 ]

0.8 %

0.0 [ -3.06, 3.06 ]

1 End point

Subtotal (95% CI)

222

187

Heterogeneity: Tau2 = 0.0; Chi2 = 2.02, df = 3 (P = 0.57); I2 =0.0% Test for overall effect: Z = 0.66 (P = 0.51) 2 Change from baseline Dougados 1986

23

0.5 (5.3)

24

0.5 (5.41) -4

-2

Favours Placebo

0

2

4

Favours SSZ

(Continued . . . )

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

54

(. . . Study or subgroup

Placebo

Mean Difference

SSZ

Weight

Mean Difference

N

Mean(SD)

N

Mean(SD)

Schmidt 2002

18

0.5 (0.1)

29

0 (0.1)

30.9 %

0.50 [ 0.44, 0.56 ]

Taylor 1991

16

0.43 (5.36)

17

0.99 (5.87)

0.5 %

-0.56 [ -4.39, 3.27 ]

32.2 %

0.50 [ 0.44, 0.56 ]

100.0 %

0.18 [ -0.11, 0.46 ]

Subtotal (95% CI)

57

IV,Random,95% CI

Continued)

IV,Random,95% CI

70

Heterogeneity: Tau2 = 0.0; Chi2 = 0.40, df = 2 (P = 0.82); I2 =0.0% Test for overall effect: Z = 16.65 (P < 0.00001)

Total (95% CI)

279

257

Heterogeneity: Tau2 = 0.07; Chi2 = 16.61, df = 6 (P = 0.01); I2 =64% Test for overall effect: Z = 1.22 (P = 0.22) Test for subgroup differences: Chi2 = 14.19, df = 1 (P = 0.00), I2 =93%

-4

-2

Favours Placebo

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

2

4

Favours SSZ

55

Analysis 1.15. Comparison 1 Sulfasalazine versus placebo, Outcome 15 (Modified) Schober’s test (2nd analysis) (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 15 (Modified) Schober’s test (2nd analysis) (cm)

Study or subgroup

Placebo

Mean Difference

SSZ

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

0.1 (1)

108

0.2 (0.9)

22.6 %

-0.10 [ -0.36, 0.16 ]

Dougados 1986

23

0.5 (5.3)

24

0.5 (5.41)

1.1 %

0.0 [ -3.06, 3.06 ]

Schmidt 2002

18

0.5 (0.1)

29

0 (0.1)

26.2 %

0.50 [ 0.44, 0.56 ]

Taylor 1991

16

0.43 (5.36)

17

0.99 (5.87)

0.7 %

-0.56 [ -4.39, 3.27 ]

50.6 %

0.19 [ -0.35, 0.74 ]

1 Change from baseline

Subtotal (95% CI)

152

178

Heterogeneity: Tau2 = 0.15; Chi2 = 19.39, df = 3 (P = 0.00023); I2 =85% Test for overall effect: Z = 0.70 (P = 0.48) 2 End point Krajnc 1990

63

3.2 (0.9)

19

3.1 (0.7)

19.4 %

0.10 [ -0.29, 0.49 ]

Nissila 1988

38

3.6 (0.8)

37

3.4 (0.8)

20.0 %

0.20 [ -0.16, 0.56 ]

Winkler 1989

26

1.88 (1.6)

23

2.3 (1.3)

10.1 %

-0.42 [ -1.23, 0.39 ]

49.4 %

0.10 [ -0.15, 0.35 ]

100.0 %

0.12 [ -0.21, 0.45 ]

Subtotal (95% CI)

127

79

Heterogeneity: Tau2 = 0.0; Chi2 = 1.87, df = 2 (P = 0.39); I2 =0.0% Test for overall effect: Z = 0.77 (P = 0.44)

Total (95% CI)

279

257

Heterogeneity: Tau2 = 0.11; Chi2 = 29.30, df = 6 (P = 0.00005); I2 =80% Test for overall effect: Z = 0.72 (P = 0.47) Test for subgroup differences: Chi2 = 0.10, df = 1 (P = 0.75), I2 =0.0%

-4

-2

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

2

4

Favours Placebo

56

Analysis 1.16. Comparison 1 Sulfasalazine versus placebo, Outcome 16 Occiput-to-wall test (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 16 Occiput-to-wall test (cm)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

9.1 (8.8)

108

8.7 (6.6)

13.6 %

0.40 [ -1.76, 2.56 ]

Davis 1989

15

5.33 (6.02)

13

4.57 (5.16)

4.9 %

0.76 [ -3.38, 4.90 ]

Nissila 1988

38

1.7 (1)

37

2.5 (2.1)

32.5 %

-0.80 [ -1.55, -0.05 ]

51.0 %

-0.63 [ -1.33, 0.07 ]

1 End point

Subtotal (95% CI)

148

158

Heterogeneity: Tau2 = 0.0; Chi2 = 1.50, df = 2 (P = 0.47); I2 =0.0% Test for overall effect: Z = 1.78 (P = 0.076) 2 Change from baseline Schmidt 2002

18

1.2 (0.6)

29

0.5 (0.7)

37.9 %

0.70 [ 0.32, 1.08 ]

Taylor 1991

16

0.03 (4.52)

17

0.15 (2.44)

11.1 %

-0.12 [ -2.62, 2.38 ]

49.0 %

0.68 [ 0.31, 1.05 ]

100.0 %

0.08 [ -0.90, 1.06 ]

Subtotal (95% CI)

34

46

Heterogeneity: Tau2 = 0.0; Chi2 = 0.40, df = 1 (P = 0.52); I2 =0.0% Test for overall effect: Z = 3.59 (P = 0.00033)

Total (95% CI)

182

204

Heterogeneity: Tau2 = 0.62; Chi2 = 12.53, df = 4 (P = 0.01); I2 =68% Test for overall effect: Z = 0.17 (P = 0.87) Test for subgroup differences: Chi2 = 10.62, df = 1 (P = 0.00), I2 =91%

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

5

10

Favours Placebo

57

Analysis 1.17. Comparison 1 Sulfasalazine versus placebo, Outcome 17 Occiput-to-wall test (2nd analysis) (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 17 Occiput-to-wall test (2nd analysis) (cm)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

0 (3)

108

0.2 (2.5)

27.3 %

-0.20 [ -0.97, 0.57 ]

Schmidt 2002

18

1.2 (0.6)

29

0.5 (0.7)

33.3 %

0.70 [ 0.32, 1.08 ]

Taylor 1991

16

0.03 (4.52)

17

0.15 (2.44)

8.3 %

-0.12 [ -2.62, 2.38 ]

68.9 %

0.30 [ -0.43, 1.03 ]

1 Change from baseline

Subtotal (95% CI)

129

154

Heterogeneity: Tau2 = 0.22; Chi2 = 4.53, df = 2 (P = 0.10); I2 =56% Test for overall effect: Z = 0.80 (P = 0.42) 2 End point Davis 1989

15

5.33 (6.02)

13

4.57 (5.16)

3.5 %

0.76 [ -3.38, 4.90 ]

Nissila 1988

38

1.7 (1)

37

2.5 (2.1)

27.6 %

-0.80 [ -1.55, -0.05 ]

31.1 %

-0.75 [ -1.49, -0.02 ]

100.0 %

-0.03 [ -0.84, 0.79 ]

Subtotal (95% CI)

53

50

Heterogeneity: Tau2 = 0.0; Chi2 = 0.53, df = 1 (P = 0.47); I2 =0.0% Test for overall effect: Z = 2.00 (P = 0.046)

Total (95% CI)

182

204

Heterogeneity: Tau2 = 0.49; Chi2 = 14.45, df = 4 (P = 0.01); I2 =72% Test for overall effect: Z = 0.06 (P = 0.95) Test for subgroup differences: Chi2 = 3.93, df = 1 (P = 0.05), I2 =75%

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

5

10

Favours Placebo

58

Analysis 1.18. Comparison 1 Sulfasalazine versus placebo, Outcome 18 Fingers-to-floor test (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 18 Fingers-to-floor test (cm)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

25.8 (15.1)

108

27.6 (16.8)

18.8 %

-1.80 [ -6.19, 2.59 ]

Davis 1989

15

18.92 (18.36)

13

16.08 (16.33)

2.2 %

2.84 [ -10.01, 15.69 ]

Krajnc 1990

63

15.4 (8.2)

19

17.8 (7.8)

22.0 %

-2.40 [ -6.45, 1.65 ]

Nissila 1988

38

15.7 (7.9)

37

15.4 (8.2)

27.2 %

0.30 [ -3.35, 3.95 ]

Winkler 1989

26

25.3 (12.2)

23

23.04 (9.9)

9.4 %

2.26 [ -3.93, 8.45 ]

79.6 %

-0.64 [ -2.77, 1.49 ]

1 End point

Subtotal (95% CI)

237

200

Heterogeneity: Tau2 = 0.0; Chi2 = 2.37, df = 4 (P = 0.67); I2 =0.0% Test for overall effect: Z = 0.59 (P = 0.56) 2 Change from baseline Dougados 1986

23

-3 (56.51)

24

0 (39.17)

0.5 %

-3.00 [ -30.91, 24.91 ]

Taylor 1991

16

-6.67 (6.53)

17

-4.14 (5.92)

19.9 %

-2.53 [ -6.79, 1.73 ]

20.4 %

-2.54 [ -6.75, 1.67 ]

100.0 %

-1.03 [ -2.93, 0.87 ]

Subtotal (95% CI)

39

41

Heterogeneity: Tau2 = 0.0; Chi2 = 0.00, df = 1 (P = 0.97); I2 =0.0% Test for overall effect: Z = 1.18 (P = 0.24)

Total (95% CI)

276

241

Heterogeneity: Tau2 = 0.0; Chi2 = 3.00, df = 6 (P = 0.81); I2 =0.0% Test for overall effect: Z = 1.06 (P = 0.29) Test for subgroup differences: Chi2 = 0.62, df = 1 (P = 0.43), I2 =0.0%

-100

-50

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

50

100

Favours Placebo

59

Analysis 1.19. Comparison 1 Sulfasalazine versus placebo, Outcome 19 Fingers-to-floor test (2nd analysis) (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 19 Fingers-to-floor test (2nd analysis) (cm)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

-1.8 (7.5)

108

-1.3 (7.6)

49.6 %

-0.50 [ -2.58, 1.58 ]

Dougados 1986

23

-3 (56.51)

24

0 (39.17)

0.3 %

-3.00 [ -30.91, 24.91 ]

Taylor 1991

16

-6.67 (5.53)

17

-4.14 (5.92)

14.1 %

-2.53 [ -6.44, 1.38 ]

63.9 %

-0.96 [ -2.79, 0.88 ]

1 Change from baseline

Subtotal (95% CI)

134

149

Heterogeneity: Tau2 = 0.0; Chi2 = 0.83, df = 2 (P = 0.66); I2 =0.0% Test for overall effect: Z = 1.02 (P = 0.31) 2 End point Davis 1989

15

18.92 (18.36)

13

16.08 (16.33)

1.3 %

2.84 [ -10.01, 15.69 ]

Krajnc 1990

63

15.4 (8.2)

19

17.8 (7.8)

13.1 %

-2.40 [ -6.45, 1.65 ]

Nissila 1988

38

15.7 (7.9)

37

15.4 (8.2)

16.1 %

0.30 [ -3.35, 3.95 ]

Winkler 1989

26

25.3 (12.2)

23

23.04 (9.9)

5.6 %

2.26 [ -3.93, 8.45 ]

36.1 %

-0.28 [ -2.72, 2.15 ]

100.0 %

-0.71 [ -2.18, 0.75 ]

Subtotal (95% CI)

142

92

Heterogeneity: Tau2 = 0.0; Chi2 = 2.02, df = 3 (P = 0.57); I2 =0.0% Test for overall effect: Z = 0.23 (P = 0.82)

Total (95% CI)

276

241

Heterogeneity: Tau2 = 0.0; Chi2 = 3.04, df = 6 (P = 0.80); I2 =0.0% Test for overall effect: Z = 0.96 (P = 0.34) Test for subgroup differences: Chi2 = 0.19, df = 1 (P = 0.66), I2 =0.0%

-100

-50

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

50

100

Favours Placebo

60

Analysis 1.20. Comparison 1 Sulfasalazine versus placebo, Outcome 20 Chin sternum distance (change from baseline) (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 20 Chin sternum distance (change from baseline) (cm)

Mean Difference

Study or subgroup

SSZ

Placebo

N

Mean(SD)

N

Mean(SD)

Schmidt 2002

18

-0.5 (0.3)

29

-0.5 (0.3)

Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI 0.0 [ -0.18, 0.18 ]

-10

-5

Favours SSZ

0

5

10

Favours Placebo

Analysis 1.21. Comparison 1 Sulfasalazine versus placebo, Outcome 21 Joint pain/tenderness score (0 to 198, the higher the score the more severe the disease) or number. Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 21 Joint pain/tenderness score (0 to 198, the higher the score the more severe the disease) or number

Study or subgroup

SSZ

Std. Mean Difference

Placebo

Weight

IV,Random,95% CI

Std. Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

8.8 (17.9)

108

7.5 (12.3)

64.1 %

0.09 [ -0.19, 0.36 ]

Nissila 1988

38

1 (1.3)

37

1.4 (1.7)

35.9 %

-0.26 [ -0.72, 0.19 ]

100.0 %

-0.04 [ -0.37, 0.29 ]

1 End point

Total (95% CI)

133

145

Heterogeneity: Tau2 = 0.02; Chi2 = 1.64, df = 1 (P = 0.20); I2 =39% Test for overall effect: Z = 0.24 (P = 0.81) Test for subgroup differences: Not applicable

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

5

10

Favours Placebo

61

Analysis 1.22. Comparison 1 Sulfasalazine versus placebo, Outcome 22 Joint pain/tenderness score (2nd analysis) (0 to 198, the higher the score the more severe the disease). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 22 Joint pain/tenderness score (2nd analysis) (0 to 198, the higher the score the more severe the disease)

Study or subgroup

SSZ

Std. Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

95

-0.6 (7.6)

108

0.3 (9.5)

Weight

IV,Random,95% CI

Std. Mean Difference IV,Random,95% CI

1 Change from baseline Clegg 1996

Subtotal (95% CI)

95

73.1 %

-0.10 [ -0.38, 0.17 ]

73.1 %

-0.10 [ -0.38, 0.17 ]

26.9 %

-0.26 [ -0.72, 0.19 ]

37

26.9 %

-0.26 [ -0.72, 0.19 ]

145

100.0 %

-0.15 [ -0.38, 0.09 ]

108

Heterogeneity: not applicable Test for overall effect: Z = 0.74 (P = 0.46) 2 End point Nissila 1988

Subtotal (95% CI)

38

38

1 (1.3)

37

1.4 (1.7)

Heterogeneity: not applicable Test for overall effect: Z = 1.13 (P = 0.26)

Total (95% CI)

133

Heterogeneity: Tau2 = 0.0; Chi2 = 0.34, df = 1 (P = 0.56); I2 =0.0% Test for overall effect: Z = 1.21 (P = 0.22) Test for subgroup differences: Chi2 = 0.34, df = 1 (P = 0.56), I2 =0.0%

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

5

10

Favours Placebo

62

Analysis 1.23. Comparison 1 Sulfasalazine versus placebo, Outcome 23 Joint swelling score (0 to 198, the higher score the more severe the disease) or number. Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 23 Joint swelling score (0 to 198, the higher score the more severe the disease) or number

Study or subgroup

SSZ

Std. Mean Difference

Placebo

Weight

IV,Random,95% CI

Std. Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

2.4 (6.2)

108

2.1 (4.9)

72.9 %

0.05 [ -0.22, 0.33 ]

Nissila 1988

38

0.2 (0.5)

37

0.2 (0.8)

27.1 %

0.0 [ -0.45, 0.45 ]

100.0 %

0.04 [ -0.20, 0.27 ]

1 End point

Total (95% CI)

133

145

Heterogeneity: Tau2 = 0.0; Chi2 = 0.04, df = 1 (P = 0.84); I2 =0.0% Test for overall effect: Z = 0.33 (P = 0.74) Test for subgroup differences: Not applicable

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

5

10

Favours Placebo

63

Analysis 1.24. Comparison 1 Sulfasalazine versus placebo, Outcome 24 Joint swelling score (2nd analysis) (0 to 198, the higher the score the more severe the disease). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 24 Joint swelling score (2nd analysis) (0 to 198, the higher the score the more severe the disease)

Study or subgroup

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

95

0.2 (3.6)

108

0.2 (3.4)

Weight

IV,Random,95% CI

Mean Difference IV,Random,95% CI

1 Change from baseline Clegg 1996

Subtotal (95% CI)

95

8.9 %

0.0 [ -0.97, 0.97 ]

8.9 %

0.0 [ -0.97, 0.97 ]

91.1 %

0.0 [ -0.30, 0.30 ]

37

91.1 %

0.0 [ -0.30, 0.30 ]

145

100.0 %

0.0 [ -0.29, 0.29 ]

108

Heterogeneity: not applicable Test for overall effect: Z = 0.0 (P = 1.0) 2 End point Nissila 1988

Subtotal (95% CI)

38

38

0.2 (0.5)

37

0.2 (0.8)

Heterogeneity: not applicable Test for overall effect: Z = 0.0 (P = 1.0)

Total (95% CI)

133

Heterogeneity: Tau2 = 0.0; Chi2 = 0.0, df = 1 (P = 1.00); I2 =0.0% Test for overall effect: Z = 0.0 (P = 1.0) Test for subgroup differences: Chi2 = 0.0, df = 1 (P = 1.00), I2 =0.0%

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

5

10

Favours Placebo

64

Analysis 1.25. Comparison 1 Sulfasalazine versus placebo, Outcome 25 Dactylitis score (0 to 3, 0 = normal, 3 = severe). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 25 Dactylitis score (0 to 3, 0 = normal, 3 = severe)

Study or subgroup

Clegg 1996

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

95

0.1 (0.8)

108

0 (0.2)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI 0.10 [ -0.07, 0.27 ]

-4

-2

0

Favours SSZ

2

4

Favours Placebo

Analysis 1.26. Comparison 1 Sulfasalazine versus placebo, Outcome 26 Dactylitis score (2nd analysis) (0 to 3, 0 = normal, 3 = severe). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 26 Dactylitis score (2nd analysis) (0 to 3, 0 = normal, 3 = severe)

Study or subgroup

Clegg 1996

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

95

0.1 (0.8)

108

0 (0.2)

IV,Random,95% CI

IV,Random,95% CI 0.10 [ -0.07, 0.27 ]

-4

-2

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

2

4

Favours Placebo

65

Analysis 1.27. Comparison 1 Sulfasalazine versus placebo, Outcome 27 Enthesopathy index (0 to 90, 0 to 66, 0 to 90, the higher the score the more severe the disease). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 27 Enthesopathy index (0 to 90, 0 to 66, 0 to 90, the higher the score the more severe the disease)

Study or subgroup

SSZ

Std. Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

95

4.1 (5.3)

108

3.8 (5)

Weight

IV,Random,95% CI

Std. Mean Difference IV,Random,95% CI

1 End point Clegg 1996

Subtotal (95% CI)

95

108

69.1 %

0.06 [ -0.22, 0.33 ]

69.1 %

0.06 [ -0.22, 0.33 ]

Heterogeneity: not applicable Test for overall effect: Z = 0.41 (P = 0.68) 2 Change from baseline Dougados 1986

23

-1 (16.9)

24

0 (25)

16.1 %

-0.05 [ -0.62, 0.53 ]

Schmidt 2002

18

-1.1 (1.6)

29

-1.7 (1.1)

14.8 %

0.45 [ -0.15, 1.05 ]

30.9 %

0.19 [ -0.29, 0.68 ]

100.0 %

0.10 [ -0.13, 0.33 ]

Subtotal (95% CI)

41

53

Heterogeneity: Tau2 = 0.03; Chi2 = 1.38, df = 1 (P = 0.24); I2 =28% Test for overall effect: Z = 0.79 (P = 0.43)

Total (95% CI)

136

161

Heterogeneity: Tau2 = 0.0; Chi2 = 1.66, df = 2 (P = 0.44); I2 =0.0% Test for overall effect: Z = 0.85 (P = 0.40) Test for subgroup differences: Chi2 = 0.23, df = 1 (P = 0.63), I2 =0.0%

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

5

10

Favours Placebo

66

Analysis 1.28. Comparison 1 Sulfasalazine versus placebo, Outcome 28 Enthesopathy index (2nd analysis) (0 to 90, 0 to 66, 0 to 90, the higher score the more severe the disease). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 28 Enthesopathy index (2nd analysis) (0 to 90, 0 to 66, 0 to 90, the higher score the more severe the disease)

Study or subgroup

SSZ

Std. Mean Difference

Placebo

Weight

IV,Random,95% CI

Std. Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

-1.4 (4)

108

-1.6 (4.6)

69.1 %

0.05 [ -0.23, 0.32 ]

Dougados 1986

23

1 (16.9)

24

0 (25)

16.1 %

0.05 [ -0.53, 0.62 ]

Schmidt 2002

18

-1.1 (1.6)

29

-1.7 (1.1)

14.8 %

0.45 [ -0.15, 1.05 ]

100.0 %

0.11 [ -0.12, 0.34 ]

1 Change from baseline

Total (95% CI)

136

161

Heterogeneity: Tau2 = 0.0; Chi2 = 1.50, df = 2 (P = 0.47); I2 =0.0% Test for overall effect: Z = 0.90 (P = 0.37) Test for subgroup differences: Not applicable

-10

-5

Favours SSZ

0

5

10

Favours Placebo

Analysis 1.29. Comparison 1 Sulfasalazine versus placebo, Outcome 29 Spondylitis articular index (0 to 90, the higher score the more severe the disease). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 29 Spondylitis articular index (0 to 90, the higher score the more severe the disease)

Study or subgroup

Clegg 1996

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

95

4.8 (4.4)

108

4.8 (4.3)

IV,Random,95% CI

IV,Random,95% CI 0.0 [ -1.20, 1.20 ]

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

5

10

Favours Placebo

67

Analysis 1.30. Comparison 1 Sulfasalazine versus placebo, Outcome 30 Spondylitis articular index (2nd analysis) (0 to 90, the higher score the more severe the disease). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 30 Spondylitis articular index (2nd analysis) (0 to 90, the higher score the more severe the disease)

Study or subgroup

Clegg 1996

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

95

-1.6 (3.4)

108

-1.6 (3.6)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI 0.0 [ -0.96, 0.96 ]

-10

-5

Favours SSZ

0

5

10

Favours Placebo

Analysis 1.31. Comparison 1 Sulfasalazine versus placebo, Outcome 31 Improvement in patient global assessment. Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 31 Improvement in patient global assessment

Study or subgroup

SSZ

Placebo

n/N

n/N

53/131

56/133

44.8 %

0.96 [ 0.72, 1.28 ]

Dougados 1986

15/30

6/30

28.7 %

2.50 [ 1.12, 5.56 ]

Schmidt 2002

11/34

6/36

26.5 %

1.94 [ 0.81, 4.67 ]

Total (95% CI)

195

199

100.0 %

1.52 [ 0.78, 2.97 ]

Clegg 1996

Risk Ratio

Weight

IV,Random,95% CI

Risk Ratio IV,Random,95% CI

Total events: 79 (SSZ), 68 (Placebo) Heterogeneity: Tau2 = 0.24; Chi2 = 6.46, df = 2 (P = 0.04); I2 =69% Test for overall effect: Z = 1.24 (P = 0.22) Test for subgroup differences: Not applicable

0.5

0.7

Favours Placebo

1

1.5

2

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

68

Analysis 1.32. Comparison 1 Sulfasalazine versus placebo, Outcome 32 Patient assessment of disease severity (end point) (VAS 100 mm, 0 = very good, 100 = very poor). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 32 Patient assessment of disease severity (end point) (VAS 100 mm, 0 = very good, 100 = very poor)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

26

51 (26.3)

23

55.8 (22.3)

Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI -4.80 [ -18.41, 8.81 ]

-100

-50

0

Favours SSZ

50

100

Favours Placebo

Analysis 1.33. Comparison 1 Sulfasalazine versus placebo, Outcome 33 General well-being (end point) (VAS 100 mm, 0 = very good, 100 = very poor). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 33 General well-being (end point) (VAS 100 mm, 0 = very good, 100 = very poor)

Study or subgroup

Nissila 1988

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

38

32 (21)

37

43 (18)

IV,Random,95% CI

IV,Random,95% CI -11.00 [ -19.84, -2.16 ]

-20

-10

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

10

20

Favours Placebo

69

Analysis 1.34. Comparison 1 Sulfasalazine versus placebo, Outcome 34 Improvement in physician global assessment. Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 34 Improvement in physician global assessment

Study or subgroup

SSZ

Placebo

n/N

n/N

70/131

74/133

61.5 %

0.96 [ 0.77, 1.20 ]

Schmidt 2002

13/34

6/36

38.5 %

2.29 [ 0.98, 5.35 ]

Total (95% CI)

165

169

100.0 %

1.34 [ 0.59, 3.08 ]

Clegg 1996

Risk Ratio

Weight

IV,Random,95% CI

Risk Ratio IV,Random,95% CI

Total events: 83 (SSZ), 80 (Placebo) Heterogeneity: Tau2 = 0.28; Chi2 = 3.81, df = 1 (P = 0.05); I2 =74% Test for overall effect: Z = 0.70 (P = 0.49) Test for subgroup differences: Not applicable

0.5

0.7

1

Favours Placebo

1.5

2

Favours SSZ

Analysis 1.35. Comparison 1 Sulfasalazine versus placebo, Outcome 35 Response to treatment (based on both patient and physician assessment). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 35 Response to treatment (based on both patient and physician assessment)

Study or subgroup

Clegg 1996

SSZ

Placebo

n/N

n/N

50/131

48/133

Risk Ratio MH,Random,95% CI

Risk Ratio MH,Random,95% CI 1.06 [ 0.77, 1.45 ]

0.1 0.2

0.5

Favours Placebo

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

1

2

5

10

Favours SSZ

70

Analysis 1.36. Comparison 1 Sulfasalazine versus placebo, Outcome 36 Duration of morning stiffness (hr). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 36 Duration of morning stiffness (hr)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

3.8 (5.5)

108

4.1 (6.1)

1.4 %

-0.30 [ -1.90, 1.30 ]

Krajnc 1990

63

0.75 (0.49)

19

0.78 (0.49)

22.6 %

-0.03 [ -0.28, 0.22 ]

Nissila 1988

38

0.63 (0.82)

37

0.78 (0.65)

17.3 %

-0.15 [ -0.48, 0.18 ]

Winkler 1989

26

0.62 (0.38)

23

0.75 (0.43)

24.3 %

-0.13 [ -0.36, 0.10 ]

65.6 %

-0.10 [ -0.25, 0.05 ]

34.4 %

-0.39 [ -0.48, -0.30 ]

29

34.4 %

-0.39 [ -0.48, -0.30 ]

216

100.0 %

-0.20 [ -0.39, -0.01 ]

1 End point

Subtotal (95% CI)

222

187

Heterogeneity: Tau2 = 0.0; Chi2 = 0.51, df = 3 (P = 0.92); I2 =0.0% Test for overall effect: Z = 1.30 (P = 0.19) 2 Change from baseline Schmidt 2002

Subtotal (95% CI)

18

-0.53 (0.18)

18

29

-0.14 (0.11)

Heterogeneity: not applicable Test for overall effect: Z = 8.28 (P < 0.00001)

Total (95% CI)

240

Heterogeneity: Tau2 = 0.02; Chi2 = 10.92, df = 4 (P = 0.03); I2 =63% Test for overall effect: Z = 2.10 (P = 0.036) Test for subgroup differences: Chi2 = 10.41, df = 1 (P = 0.00), I2 =90%

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

5

10

Favours Placebo

71

Analysis 1.37. Comparison 1 Sulfasalazine versus placebo, Outcome 37 Duration of morning stiffness (2nd analysis) (hr). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 37 Duration of morning stiffness (2nd analysis) (hr)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

-0.5 (5.2)

108

-0.7 (7.3)

1.2 %

0.20 [ -1.53, 1.93 ]

Schmidt 2002

18

-0.53 (0.18)

29

-0.14 (0.11)

34.1 %

-0.39 [ -0.48, -0.30 ]

35.3 %

-0.39 [ -0.48, -0.30 ]

1 Change from base line

Subtotal (95% CI)

113

137

Heterogeneity: Tau2 = 0.0; Chi2 = 0.45, df = 1 (P = 0.50); I2 =0.0% Test for overall effect: Z = 8.26 (P < 0.00001) 2 End point Krajnc 1990

63

0.75 (0.49)

19

0.78 (0.49)

22.7 %

-0.03 [ -0.28, 0.22 ]

Nissila 1988

38

0.63 (0.82)

37

0.78 (0.65)

17.5 %

-0.15 [ -0.48, 0.18 ]

Winkler 1989

26

0.62 (0.38)

23

0.75 (0.43)

24.4 %

-0.13 [ -0.36, 0.10 ]

64.7 %

-0.10 [ -0.25, 0.05 ]

100.0 %

-0.20 [ -0.39, 0.00 ]

Subtotal (95% CI)

127

79

Heterogeneity: Tau2 = 0.0; Chi2 = 0.45, df = 2 (P = 0.80); I2 =0.0% Test for overall effect: Z = 1.27 (P = 0.20)

Total (95% CI)

240

216

Heterogeneity: Tau2 = 0.03; Chi2 = 11.25, df = 4 (P = 0.02); I2 =64% Test for overall effect: Z = 1.99 (P = 0.047) Test for subgroup differences: Chi2 = 10.35, df = 1 (P = 0.00), I2 =90%

-4

-2

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

2

4

Favours Placebo

72

Analysis 1.38. Comparison 1 Sulfasalazine versus placebo, Outcome 38 Morning stiffness (end point) (VAS 100 mm, 0 = no stiffness, 100 = severe stiffness). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 38 Morning stiffness (end point) (VAS 100 mm, 0 = no stiffness, 100 = severe stiffness)

Study or subgroup

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

38

19 (24)

37

33 (19)

Weight

IV,Random,95% CI

Mean Difference IV,Random,95% CI

1 End point Nissila 1988

Subtotal (95% CI)

38

78.1 %

-14.00 [ -23.78, -4.22 ]

78.1 %

-14.00 [ -23.78, -4.22 ]

21.9 %

-13.50 [ -32.00, 5.00 ]

17

21.9 %

-13.50 [ -32.00, 5.00 ]

54

100.0 %

-13.89 [ -22.54, -5.24 ]

37

Heterogeneity: not applicable Test for overall effect: Z = 2.80 (P = 0.0050) 2 Change from baseline Taylor 1991

Subtotal (95% CI)

16

-14.4 (31.2)

16

17

-0.9 (21.9)

Heterogeneity: not applicable Test for overall effect: Z = 1.43 (P = 0.15)

Total (95% CI)

54

Heterogeneity: Tau2 = 0.0; Chi2 = 0.00, df = 1 (P = 0.96); I2 =0.0% Test for overall effect: Z = 3.15 (P = 0.0016) Test for subgroup differences: Chi2 = 0.00, df = 1 (P = 0.96), I2 =0.0%

-100

-50

0

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

50

100

Favours Placebo

73

Analysis 1.39. Comparison 1 Sulfasalazine versus placebo, Outcome 39 Improvement in morning stiffness. Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 39 Improvement in morning stiffness

Study or subgroup

SSZ

Clegg 1996

Placebo

n/N

n/N

64/131

59/133

Risk Ratio MH,Random,95% CI

Risk Ratio MH,Random,95% CI 1.10 [ 0.85, 1.43 ]

0.1 0.2

0.5

1

Favours Placebo

2

5

10

Favours SSZ

Analysis 1.40. Comparison 1 Sulfasalazine versus placebo, Outcome 40 Erythrocyte sedimentation rate (mm/hr). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 40 Erythrocyte sedimentation rate (mm/hr)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

19.3 (19.2)

108

23.4 (22.2)

17.8 %

-4.10 [ -9.80, 1.60 ]

Davis 1989

15

17.8 (15.99)

13

25.6 (18.6)

8.6 %

-7.80 [ -20.75, 5.15 ]

Feltelius 1986

14

23.1 (17.6)

15

25.1 (13.2)

10.0 %

-2.00 [ -13.38, 9.38 ]

Krajnc 1990

63

23 (21)

19

40 (19)

11.6 %

-17.00 [ -26.99, -7.01 ]

Nissila 1988

38

23 (25)

37

42 (22)

10.8 %

-19.00 [ -29.65, -8.35 ]

Winkler 1989

26

24.4 (11.9)

23

20.3 (10.9)

16.7 %

4.10 [ -2.28, 10.48 ]

75.6 %

-7.07 [ -14.39, 0.25 ]

1 End point

Subtotal (95% CI)

251

215

Heterogeneity: Tau2 = 60.19; Chi2 = 20.51, df = 5 (P = 0.001); I2 =76% Test for overall effect: Z = 1.89 (P = 0.058)

-100

-50

Favours SSZ

0

50

100

Favours Placebo

(Continued . . . )

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

74

(. . . Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Continued)

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Dougados 1986

23

-2 (59.29)

24

-2 (85.05)

1.2 %

0.0 [ -41.77, 41.77 ]

Schmidt 2002

18

-6.9 (3.6)

29

-3.8 (1.5)

23.2 %

-3.10 [ -4.85, -1.35 ]

24.4 %

-3.09 [ -4.84, -1.35 ]

100.0 %

-5.66 [ -10.40, -0.93 ]

2 Change from baseline

Subtotal (95% CI)

41

53

Heterogeneity: Tau2 = 0.0; Chi2 = 0.02, df = 1 (P = 0.88); I2 =0.0% Test for overall effect: Z = 3.47 (P = 0.00052)

Total (95% CI)

292

268

Heterogeneity: Tau2 = 24.31; Chi2 = 21.35, df = 7 (P = 0.003); I2 =67% Test for overall effect: Z = 2.35 (P = 0.019) Test for subgroup differences: Chi2 = 1.07, df = 1 (P = 0.30), I2 =7%

-100

-50

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

50

100

Favours Placebo

75

Analysis 1.41. Comparison 1 Sulfasalazine versus placebo, Outcome 41 Erythrocyte sedimentation rate (2nd analysis) (mm/hr). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 41 Erythrocyte sedimentation rate (2nd analysis) (mm/hr)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

-5.3 (11.2)

108

-1.7 (11.8)

22.4 %

-3.60 [ -6.77, -0.43 ]

Dougados 1986

23

-2 (59.29)

24

-2 (85.05)

0.9 %

0.0 [ -41.77, 41.77 ]

Feltelius 1986

14

-2.2 (9.5)

15

-3.5 (17)

10.0 %

1.30 [ -8.64, 11.24 ]

Schmidt 2002

18

-6.9 (3.6)

29

-3.8 (1.5)

24.8 %

-3.10 [ -4.85, -1.35 ]

58.2 %

-3.11 [ -4.62, -1.60 ]

1 Change from baseline

Subtotal (95% CI)

150

176

Heterogeneity: Tau2 = 0.0; Chi2 = 0.87, df = 3 (P = 0.83); I2 =0.0% Test for overall effect: Z = 4.03 (P = 0.000057) 2 End point Davis 1989

15

17.8 (15.99)

13

25.6 (18.6)

7.0 %

-7.80 [ -20.75, 5.15 ]

Krajnc 1990

63

23 (21)

19

40 (19)

10.0 %

-17.00 [ -26.99, -7.01 ]

Nissila 1988

38

23 (25)

37

42 (22)

9.2 %

-19.00 [ -29.65, -8.35 ]

Winkler 1989

26

24.4 (11.9)

23

20.3 (10.9)

15.7 %

4.10 [ -2.28, 10.48 ]

41.8 %

-9.56 [ -22.03, 2.91 ]

100.0 %

-4.79 [ -8.80, -0.78 ]

Subtotal (95% CI)

142

92

Heterogeneity: Tau2 = 135.37; Chi2 = 20.05, df = 3 (P = 0.00017); I2 =85% Test for overall effect: Z = 1.50 (P = 0.13)

Total (95% CI)

292

268

Heterogeneity: Tau2 = 16.08; Chi2 = 22.11, df = 7 (P = 0.002); I2 =68% Test for overall effect: Z = 2.34 (P = 0.019) Test for subgroup differences: Chi2 = 1.01, df = 1 (P = 0.31), I2 =1%

-100

-50

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

50

100

Favours Placebo

76

Analysis 1.42. Comparison 1 Sulfasalazine versus placebo, Outcome 42 C-reactive protein (ug/ml). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 42 C-reactive protein (ug/ml)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

1.44 (1.87)

108

1.61 (2.2)

56.4 %

-0.17 [ -0.73, 0.39 ]

Nissila 1988

38

15 (16)

37

22 (26)

5.2 %

-7.00 [ -16.80, 2.80 ]

61.6 %

-1.76 [ -7.42, 3.90 ]

38.4 %

-2.50 [ -4.70, -0.30 ]

29

38.4 %

-2.50 [ -4.70, -0.30 ]

174

100.0 %

-1.42 [ -3.76, 0.92 ]

1 End pointSub-category

Subtotal (95% CI)

133

145

Heterogeneity: Tau2 = 10.78; Chi2 = 1.86, df = 1 (P = 0.17); I2 =46% Test for overall effect: Z = 0.61 (P = 0.54) 2 Change from baseline Schmidt 2002

Subtotal (95% CI)

18

18

-6 (4.5)

29

-3.5 (2)

Heterogeneity: not applicable Test for overall effect: Z = 2.22 (P = 0.026)

Total (95% CI)

151

Heterogeneity: Tau2 = 2.45; Chi2 = 5.82, df = 2 (P = 0.05); I2 =66% Test for overall effect: Z = 1.19 (P = 0.23) Test for subgroup differences: Chi2 = 0.06, df = 1 (P = 0.81), I2 =0.0%

-100

-50

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

50

100

Favours Placebo

77

Analysis 1.43. Comparison 1 Sulfasalazine versus placebo, Outcome 43 C-reactive protein (2nd analysis) (ug/ml). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 43 C-reactive protein (2nd analysis) (ug/ml)

Study or subgroup

SSZ

Mean Difference

Placebo

Weight

IV,Random,95% CI

Mean Difference

N

Mean(SD)

N

Mean(SD)

IV,Random,95% CI

Clegg 1996

95

-0.28 (1.22)

108

-0.23 (1.22)

55.6 %

-0.05 [ -0.39, 0.29 ]

Schmidt 2002

18

-6 (4.5)

29

-3.5 (2)

38.7 %

-2.50 [ -4.70, -0.30 ]

94.3 %

-1.02 [ -3.37, 1.33 ]

5.7 %

-7.00 [ -16.80, 2.80 ]

37

5.7 %

-7.00 [ -16.80, 2.80 ]

174

100.0 %

-1.39 [ -3.85, 1.07 ]

1 Change from baseline

Subtotal (95% CI)

113

137

Heterogeneity: Tau2 = 2.36; Chi2 = 4.64, df = 1 (P = 0.03); I2 =78% Test for overall effect: Z = 0.85 (P = 0.39) 2 End point Nissila 1988

Subtotal (95% CI)

38

38

15 (16)

37

22 (26)

Heterogeneity: not applicable Test for overall effect: Z = 1.40 (P = 0.16)

Total (95% CI)

151

Heterogeneity: Tau2 = 2.79; Chi2 = 6.54, df = 2 (P = 0.04); I2 =69% Test for overall effect: Z = 1.11 (P = 0.27) Test for subgroup differences: Chi2 = 1.35, df = 1 (P = 0.25), I2 =26%

-100

-50

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

0

50

100

Favours Placebo

78

Analysis 1.44. Comparison 1 Sulfasalazine versus placebo, Outcome 44 Withdrawal due to side effect. Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 44 Withdrawal due to side effect

Study or subgroup

SSZ

Placebo

n/N

n/N

Clegg 1996

11/131

6/133

14.5 %

1.86 [ 0.71, 4.89 ]

Corkill 1990

12/32

9/30

22.7 %

1.25 [ 0.62, 2.53 ]

Davis 1989

0/15

1/15

3.7 %

0.33 [ 0.01, 7.58 ]

Dougados 1986

3/30

5/30

12.2 %

0.60 [ 0.16, 2.29 ]

Feltelius 1986

4/18

3/19

7.1 %

1.41 [ 0.36, 5.43 ]

Kirwan 1993

7/44

5/45

12.1 %

1.43 [ 0.49, 4.17 ]

Krajnc 1990

5/71

1/24

3.6 %

1.69 [ 0.21, 13.75 ]

Nissila 1988

2/43

2/42

4.9 %

0.98 [ 0.14, 6.62 ]

13/34

4/36

9.5 %

3.44 [ 1.24, 9.52 ]

Taylor 1991

2/20

2/20

4.9 %

1.00 [ 0.16, 6.42 ]

Winkler 1989

4/31

2/32

4.8 %

2.06 [ 0.41, 10.47 ]

Total (95% CI)

469

426

100.0 %

1.50 [ 1.04, 2.15 ]

Schmidt 2002

Risk Ratio

Weight

M-H,Fixed,95% CI

Risk Ratio M-H,Fixed,95% CI

Total events: 63 (SSZ), 40 (Placebo) Heterogeneity: Chi2 = 6.24, df = 10 (P = 0.79); I2 =0.0% Test for overall effect: Z = 2.18 (P = 0.030) Test for subgroup differences: Not applicable

0.1 0.2

0.5

Favours SSZ

1

2

5

10

Favours Placebo

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

79

Analysis 1.45. Comparison 1 Sulfasalazine versus placebo, Outcome 45 Withdrawal due to ineffectiveness. Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 45 Withdrawal due to ineffectiveness

Study or subgroup

SSZ

Placebo

n/N

n/N

Risk Ratio

Weight

Clegg 1996

6/131

2/133

Davis 1989

0/15

0/15

Dougados 1986

2/30

3/30

10.2 %

0.67 [ 0.12, 3.71 ]

Feltelius 1986

0/18

1/19

5.0 %

0.35 [ 0.02, 8.09 ]

Kirwan 1993

12/44

9/45

30.3 %

1.36 [ 0.64, 2.91 ]

Krajnc 1990

0/71

2/24

12.6 %

0.07 [ 0.00, 1.40 ]

Nissila 1988

0/43

1/42

5.2 %

0.33 [ 0.01, 7.78 ]

Schmidt 2002

3/34

3/36

9.9 %

1.06 [ 0.23, 4.89 ]

Taylor 1991

0/20

0/20

Winkler 1989

1/31

6/32

20.1 %

0.17 [ 0.02, 1.35 ]

Total (95% CI)

437

396

100.0 %

0.87 [ 0.53, 1.42 ]

M-H,Fixed,95% CI

Risk Ratio M-H,Fixed,95% CI

6.8 %

3.05 [ 0.63, 14.82 ] Not estimable

Not estimable

Total events: 24 (SSZ), 27 (Placebo) Heterogeneity: Chi2 = 9.71, df = 7 (P = 0.21); I2 =28% Test for overall effect: Z = 0.56 (P = 0.57) Test for subgroup differences: Not applicable

0.01

0.1

Favours SSZ

1

10

100

Favours Placebo

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

80

Analysis 1.46. Comparison 1 Sulfasalazine versus placebo, Outcome 46 Drop-out for any reason. Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 46 Drop-out for any reason

Study or subgroup

Favours treatment

Placebo

n/N

n/N

Clegg 1996

36/131

25/133

32.5 %

1.46 [ 0.93, 2.29 ]

Davis 1989

0/15

2/15

3.3 %

0.20 [ 0.01, 3.85 ]

Dougados 1986

7/30

6/30

7.9 %

1.17 [ 0.44, 3.06 ]

Feltelius 1986

4/18

4/19

5.1 %

1.06 [ 0.31, 3.60 ]

Kirwan 1993

19/44

8/45

10.4 %

2.43 [ 1.19, 4.96 ]

Krajnc 1990

8/71

5/24

9.8 %

0.54 [ 0.20, 1.50 ]

Nissila 1988

5/43

5/42

6.6 %

0.98 [ 0.30, 3.13 ]

16/34

7/36

8.9 %

2.42 [ 1.14, 5.15 ]

Taylor 1991

4/20

3/20

3.9 %

1.33 [ 0.34, 5.21 ]

Winkler 1989

5/31

9/32

11.6 %

0.57 [ 0.22, 1.52 ]

Total (95% CI)

437

396

100.0 %

1.33 [ 1.03, 1.73 ]

Schmidt 2002

Risk Ratio

Weight

M-H,Fixed,95% CI

Risk Ratio M-H,Fixed,95% CI

Total events: 104 (Favours treatment), 74 (Placebo) Heterogeneity: Chi2 = 13.24, df = 9 (P = 0.15); I2 =32% Test for overall effect: Z = 2.15 (P = 0.031) Test for subgroup differences: Not applicable

0.5

0.7

Favours SSZ

1

1.5

2

Favours Placebo

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

81

Analysis 1.47. Comparison 1 Sulfasalazine versus placebo, Outcome 47 Serious adverse events. Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 1 Sulfasalazine versus placebo Outcome: 47 Serious adverse events

Study or subgroup

SSZ

Clegg 1996

Peto Odds Ratio

Placebo

Weight

Peto Odds Ratio

n/N

n/N

1/131

0/133

100.0 %

7.50 [ 0.15, 378.16 ]

131

133

100.0 %

7.50 [ 0.15, 378.16 ]

Total (95% CI)

Peto,Fixed,95% CI

Peto,Fixed,95% CI

Total events: 1 (SSZ), 0 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 1.01 (P = 0.31) Test for subgroup differences: Not applicable

0.01

0.1

Favours SSZ

1

10

100

Favours Placebo

Analysis 2.1. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 1 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values) Outcome: 1 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

9

40 (11.2)

6

42.3 (13.2)

IV,Random,95% CI

IV,Random,95% CI -2.30 [ -15.15, 10.55 ]

-100

-50

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

50

100

Favours Placebo

82

Analysis 2.2. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 2 Score of sleep disturbance (0 to 4, 0 = no disturbance, 4 = severe disturbance). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values) Outcome: 2 Score of sleep disturbance (0 to 4, 0 = no disturbance, 4 = severe disturbance)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

9

1.22 (0.97)

6

1.83 (1.17)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI -0.61 [ -1.74, 0.52 ]

-10

-5

0

Favours SSZ

5

10

Favours Placebo

Analysis 2.3. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 3 Chest expansion (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values) Outcome: 3 Chest expansion (cm)

Study or subgroup

Winkler 1989

Placebo

Mean Difference

SSZ

N

Mean(SD)

N

Mean(SD)

9

4.07 (2.47)

6

3 (1.48)

IV,Random,95% CI

IV,Random,95% CI 1.07 [ -0.93, 3.07 ]

-10

-5

Favours Placebo

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

5

10

Favours SSZ

83

Analysis 2.4. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 4 Schober’s test (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values) Outcome: 4 Schober’s test (cm)

Study or subgroup

Placebo

Winkler 1989

Mean Difference

SSZ

N

Mean(SD)

N

Mean(SD)

9

1.55 (1.54)

6

2.24 (1.32)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI -0.69 [ -2.15, 0.77 ]

-10

-5

0

Favours Placebo

5

10

Favours SSZ

Analysis 2.5. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 5 Fingers-to-floor test (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values) Outcome: 5 Fingers-to-floor test (cm)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

9

24.7 (12.1)

6

24.17 (10.5)

IV,Random,95% CI

IV,Random,95% CI 0.53 [ -11.01, 12.07 ]

-100

-50

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

50

100

Favours Placebo

84

Analysis 2.6. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 6 Articular index (0 to 66, the higher the score the more severe the disease). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values) Outcome: 6 Articular index (0 to 66, the higher the score the more severe the disease)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

9

2.67 (1.58)

6

3.83 (2.79)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI -1.16 [ -3.62, 1.30 ]

-10

-5

0

Favours SSZ

5

10

Favours Placebo

Analysis 2.7. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 7 Degree of joint swelling (0 to 66, the higher the score the more severe the disease). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values) Outcome: 7 Degree of joint swelling (0 to 66, the higher the score the more severe the disease)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

9

3 (2.12)

6

3 (2.09)

IV,Random,95% CI

IV,Random,95% CI 0.0 [ -2.17, 2.17 ]

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

5

10

Favours Placebo

85

Analysis 2.8. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 8 Patient assessment of disease severity (VAS 100 mm, 0 = very good, 100 = very poor). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values) Outcome: 8 Patient assessment of disease severity (VAS 100 mm, 0 = very good, 100 = very poor)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

9

51.1 (28.5)

6

67.5 (26.3)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI -16.40 [ -44.50, 11.70 ]

-100

-50

0

50

Favours SSZ

100

Favours Placebo

Analysis 2.9. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 9 Duration of morning stiffness (hr). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values) Outcome: 9 Duration of morning stiffness (hr)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

9

0.6 (0.48)

6

0.79 (0.51)

IV,Random,95% CI

IV,Random,95% CI -0.19 [ -0.70, 0.32 ]

-10

-5

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

5

10

Favours Placebo

86

Analysis 2.10. Comparison 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values), Outcome 10 Erythrocyte sedimentation rate (mm/hr). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 2 Sulfasalazine versus placebo (ankylosing spondylitis with peripheral arthritis, end point values) Outcome: 10 Erythrocyte sedimentation rate (mm/hr)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

9

28.8 (9.6)

6

27.3 (15.6)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI 1.50 [ -12.47, 15.47 ]

-100

-50

0

Favours SSZ

50

100

Favours Placebo

Analysis 3.1. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 1 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values) Outcome: 1 Back pain (VAS 100 mm, 0 = no pain, 100 = severe pain)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

17

34.6 (11.1)

17

43.8 (14.3)

IV,Random,95% CI

IV,Random,95% CI -9.20 [ -17.81, -0.59 ]

-100

-50

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

50

100

Favours Placebo

87

Analysis 3.2. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 2 Score of sleep disturbance (0 to 4, 0 = no disturbance, 4 = severe disturbance). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values) Outcome: 2 Score of sleep disturbance (0 to 4, 0 = no disturbance, 4 = severe disturbance)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

17

1.47 (1)

17

1.71 (0.98)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI -0.24 [ -0.91, 0.43 ]

-10

-5

0

Favours SSZ

5

10

Favours Placebo

Analysis 3.3. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 3 Chest expansion (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values) Outcome: 3 Chest expansion (cm)

Study or subgroup

Winkler 1989

Placebo

Mean Difference

SSZ

N

Mean(SD)

N

Mean(SD)

17

3.2 (1.3)

17

3.18 (1.5)

IV,Random,95% CI

IV,Random,95% CI 0.02 [ -0.92, 0.96 ]

-10

-5

Favours Placebo

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

5

10

Favours SSZ

88

Analysis 3.4. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 4 Schober’s test (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values) Outcome: 4 Schober’s test (cm)

Study or subgroup

Placebo

Winkler 1989

Mean Difference

SSZ

N

Mean(SD)

N

Mean(SD)

17

2.06 (1.64)

17

2.26 (1.4)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI -0.20 [ -1.23, 0.83 ]

-10

-5

0

Favours Placebo

5

10

Favours SSZ

Analysis 3.5. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 5 Fingers-to-floor test (cm). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values) Outcome: 5 Fingers-to-floor test (cm)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

17

25.7 (12.7)

17

22.6 (9.8)

IV,Random,95% CI

IV,Random,95% CI 3.10 [ -4.53, 10.73 ]

-100

-50

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

50

100

Favours Placebo

89

Analysis 3.6. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 6 Patient assessment of disease severity (VAS 100 mm, 0 = very good, 100 = very poor). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values) Outcome: 6 Patient assessment of disease severity (VAS 100 mm, 0 = very good, 100 = very poor)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

17

50.9 (26)

17

51.6 (20)

Mean Difference

IV,Random,95% CI

IV,Random,95% CI -0.70 [ -16.29, 14.89 ]

-100

-50

0

Favours SSZ

50

100

Favours Placebo

Analysis 3.7. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 7 Duration of morning stiffness (hr). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values) Outcome: 7 Duration of morning stiffness (hr)

Study or subgroup

Winkler 1989

SSZ

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

17

0.63 (0.33)

17

0.74 (0.41)

IV,Random,95% CI

IV,Random,95% CI -0.11 [ -0.36, 0.14 ]

-4

-2

Favours SSZ

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Mean Difference

0

2

4

Favours Placebo

90

Analysis 3.8. Comparison 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values), Outcome 8 Erythrocyte sedimentation rate (mm/hr). Review:

Sulfasalazine for ankylosing spondylitis

Comparison: 3 Sulfasalazine versus placebo (axial form ankylosing spondylitis, end point values) Outcome: 8 Erythrocyte sedimentation rate (mm/hr)

Study or subgroup

SSZ

Winkler 1989

Mean Difference

Placebo

N

Mean(SD)

N

Mean(SD)

17

22 (12.5)

17

20.3 (10.9)

Mean Difference

IV,Fixed,95% CI

IV,Fixed,95% CI 1.70 [ -6.18, 9.58 ]

-100

-50

0

Favours SSZ

50

100

Favours Placebo

ADDITIONAL TABLES Table 1. Concise comparison of included trials

Study

Methodological quality

Duration/ sample size

Disease du- Peripheral ration arthritis

Clegg 1996

Mutlicenter RCT Concealment: unclear risk Assessment: blind

36 weeks/ 18.5+/-11.6 264

29%

Baseline ESR

Intervention

Main results

Drop-out

SSZ: 24.6+/ SSZ (or ESR de- 19.3% -18.0 placebo) 2.0 clined more Placebo: 25. g/d with SSZ 2+/-22.0 than with placebo (P < 0.0001). When comparing SSZ responders with non-responders, the former had a greater decrease in ESR (P < 0. 04) Patients with peripheral arthritis

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

91

Table 1. Concise comparison of included trials

(Continued)

showed improvement that favored SSZ (P < 0. 02) No significant difference in other parameters. One patient taking SSZ had severe adverse drug reaction Corkill 1990

RCT Concealment: unclear risk Assessment: blind

48 weeks/62 SSZ: 12.3+/ 19% -8.2 Placebo: 16. 1+/-11.4

SSZ: 15+/- SSZ (or No sig- No data 16 placebo) 2.0 nificant difPlacebo: g/d ference be24+/-26 tween intervention groups

Davis 1989

RCT 3 months/ Median Conceal30 SSZ: 8.6 ment: Placebo: 8.4 unclear Risk assessment: blind

23%

SSZ: 24+/- SSZ (or Claimed ef- 6.7% 7. placebo) 2.0 fective 8(95% con- g/d on the basis fidence limof before-afits) ter compariPlacebo: 26. son 4+/-8.6

Dougados 1986

RCT 6 months/ Median 10 Conceal60 ment: low risk Assessment: blind

0%

SSZ: 13.5 SSZ (or Success 21.7% (median) placebo) 2.0 in patient Placebo: 11. g/d assessment 0 was more in SSZ than in placebo group. Function index and NSAIDs dosage were significantly improved in SSZ compared with placebo group. No

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

92

Table 1. Concise comparison of included trials

(Continued)

difference was found in other parameters Feltelius 1986

RCT Concealment: unclear risk Assessment: blind

12 weeks/37 Median 5% SSZ: 12.1 Placebo: 10. 4

Kirwan 1993

RCT 3 years/89 Concealment: low risk Assessment: blind

SSZ: 19+/- 28% 12 Placebo: 21. 9+/-11.7

SSZ: 24.3+/ -17.4 Placebo: 28. 5+/-19.5

SSZ (or placebo) up to 3.0 g/ d

No data

SSZ (or Occurence 30.3% placebo) 2.0 of peripheral g/d joint symptoms was lower in SSZ group: SSZ: 0.298 episodes/yr Placebo: 0. 392 episodes/yr, P < 0.05 No difference was found in Schober test,

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Only graphs 21.6% (no figures) were presented. Compared with placebo group, morning stiffness and sleep disturbance were significantly improved in SSZ group Analysis of SSZ group showed that the greatest improvement were those with ESR > 20 mm/ hr or haptoglobin > 3. 8 g/L

93

Table 1. Concise comparison of included trials

(Continued)

chest expansion and cervical spine lateral flexion. More dropouts in SSZ group Krajnc 1990 RCT Concealment: unclear risk Assessment: blind

24 weeks/95 No data SSZ = 71 Placebo = 24

66%

SSZ: 41+/19 Placebo: 43+/-18

SSZ (or placebo) up to 3.0 g/ d

On the basis 14.3% of before-after treatment comparison, duration of morning stiffness, number of painful and swollen joints, and ESR, there was significant improvement in SSZ group Duration of morning stiffness and ESR value were given in the paper and we found no significant difference between the intervention groups

Nissila 1988 RCT Concealment: unclear risk Assessment: blind

26 weeks/85 SSZ: 5.4+/- 68% 7.3 Placebo: 3. 8+/-4.3

SSZ: 42+/20 Placebo: 46+/-19

SSZ (or placebo) up to 3.0 g/ d

Sig12.2% nificant differences between intervention groups were observed in sever-

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

94

Table 1. Concise comparison of included trials

(Continued)

ity of morning stiffness, chest expansion and ESR. We also found severity of pain significantly improved in SSZ, compared with placebo group Schmidt 2002

RCT Concealment: unclear risk Assessment: blind?

26 weeks/70 SSZ: 16.7+/ 36% -7.2 Placebo: 16. 3+/-7.8

Taylor 1991

RCT 1 year/40 Concealment: low risk Assessment: blind

SSZ: 11+/- 15% 1.6 Placebo: 10. 7+/-1.6

SSZ: 23.1+/ SSZ (or No sig- 32.9% -3.2 placebo) 3.0 nificant difPlacebo 20. g/d ference was 4+/-2.4 found between intervention groups except IgA. There were more drop-outs in SSZ than in placebo group (18/ 34 versus 7/ 36) SSZ: 27 Placebo: 25

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

SSZ (or No sig- 17.5% placebo) 2.0 nificant difg/d ference was found between intervention groups in all parameters except pain (measured with visual analogue scale). However, the pooled re-

95

Table 1. Concise comparison of included trials

(Continued)

sult showed no statistically significant, too Winkler 1989

RCT Concealment: unclear risk Assessment: blind?

24 weeks/63 Median 33% SSZ: 10.8 Placebo: 11. 2

SSZ: 33.4+/ SSZ (or The advan- 22.2% -20.4 placebo) 2.0 tage of SSZ Placebo: 26. g/d over placebo 9+/-16.4 were significant only in the duration of morning stiffness and disturbance of sleep. The same results were found in the patients with axial form (N = 34) . In patients with peripheral arthritis (N = 15), articular index showed significant improvement in SSZ over placebo

ESR - erythrocyte sedimentation rate g/d - grams per day Ig A - immunoglobulin A NSAIDs - non-steroidal anti-inflammatory drugs RCT - randomized controlled trials SSZ - sulfasalazine

Table 2. Randomized controlled trials comparing sulphasalazine with placebo

Study and Du- Participants ration

Outcomes assessed

Results reported

Present analysis ESR results

Clegg 1996, 36 264 weeks (SSZ: 31) (Placebo: 133)

Primary out- Drop-out: All the results re- Absolute benefit Did not report comes included 19.3%. Both end ported have been from SSZ: -3.6 response to treat- point value confirmed mm/h

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Spinal stiffness

96

Table 2. Randomized controlled trials comparing sulphasalazine with placebo

(Continued)

29% with PA DD (year): 18.5±11.6 ESR (mm/h): 26.4±18.0 (SSZ) 25.2±22.0 (placebo)

ment, improvement in PhGA, PGA, back pain and morning stiffness. Secondary outcomes included night pain (event), duration of morning stiffness, back pain VAS, spondylitis function index, joint/tenderness score, joint swelling score, dactylitis score, enthesopathy index, spondylitis articular index, chest expansion, Schober’s test, occiputto-wall test, fingers-to-floor test, ESR and CRP

and change from baseline were presented for all continuous outcomes. No difference was found between treatment groups in all outcomes except ESR, which declined more with SSZ than placebo group (P < 0.0001) . When comparing SSZ responders with non-responders, the former had a greater decrease in ESR (P < 0.04). Subgroup analysis showed that in patients with PA, 55.9% of SSZ group and 30.2% of placebo group got peripheral response (P = 0. 023)

except subgroup analysis where no information about treatment allocation was available for analysis. MD for ESR (change from baseline) was -3. 10 mm/h, 95% CI -4.85 to -1. 35 mm/h, favoring SSZ group

Relative difference in change from baseline: 14%

Corkill 1990, 48 62 weeks (SSZ: 32) (Placebo: 30) 19% with PA DD (year): 12.3±8.2 (SSZ) 16.1±11.4 (placebo) ESR (mm/h): 15±16 (SSZ) 24±26 (placebo)

Spinal pain VAS, spinal stiffness VAS, peripheral joint pain VAS, Schober’s test, chest expansion, cervical flexion, cervical rotation, and ESR

Drop-out: 1.6%. No significant difference was found between treatment groups

Because SDs were not given for all outcomes, these results could not be analyzed

Absolute benefit from SSZ: -0.1 mm/h Relative difference in change from baseline: 1%

Absolute benefit from SSZ: -9.8 mm on 100 mm VAS Relative difference in change from baseline: 25%

Davis 1989, 3 30 months (SSZ: 15) (Placebo: 15) 23% with PA

Pain VAS, spinal stiffness VAS, sleep disturbance (event), occiput-

Drop-out: 6.7%. In SSZ group, all clinical outcomes showed

Pain VAS, spinal stiffness VAS and CRP could not be analyzed be-

Absolute benefit from SSZ: -5.4 mm/h Relatiive differ-

Absolute benefit from SSZ: -20 mm on 100 mm VAS

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

97

Table 2. Randomized controlled trials comparing sulphasalazine with placebo

DD (year): to-wall test, fin8.6 (SSZ) gers-to-floor test, 8.4 (placebo) ESR and CRP ESR (mm/h): 24±7. 8 (SSZ) 26.4±8. 6 (placebo)

(Continued)

significantly improved when initial and 3 months results are compared

cause means and ence in change SDs were not from baseline: given. No signif- 20% icant difference was found in any other outcome

Relative difference in change from baseline: 40%

Dougados 1986, 60 6 months (SSZ: 30) (Placebo: 30) None with PA DD (year, median): 10 ESR (mm/ h, median): 13.5 (SSZ) 11.0 (placebo)

PGA, score of daily NSAIDs, pain VAS, joint index, frequency of nocturnal awakening, function index, Schober’s test, fingers-to-floor test, chest expansion and ESR

Drop-out: 21. 7%. Success in PGA was more in SSZ than in placebo group (15/30 vs 6/30, P < 0.05). SSZ resulted in a significant reduction in score of daily NSAIDs (P < 0. 05) and significant improvement of function index (P was not given) compared with placebo. No significant difference was found in other outcomes

All AbsoDid not report continuous out- lute benefit from comes were pre- SSZ: 0 mm/h sented as median and 95% CI. For RevMan analysis, we assumed that mean is equal to median for each outcome and calculated SD from 95% CI and sample size. Success in PGA was more in SSZ than in placebo group (RR 2.5, 95% CI 1.12 to 5.56). No significant difference was found between treatment groups in other outcomes

Feltelius 1986, 37 12 weeks (SSZ:18) (Placebo: 19) 5% with PA DD (year, median): 12.1 (SSZ) 10.4 (placebo) ESR (mm/h): 24.3±17.4 (SSZ) 28.5±19.5 (placebo)

Duration of morning stiffness, spinal stiffness VAS, pain VAS, general wellbeing VAS, chest expansion, Schober’s test, sleep disturbance (event) , sacroiliac pain VAS, ESR

Drop-out: 21. 6%. Spinal stiffness VAS, chest expansion and sleep disturbance were significantly improved in SSZ compared with placebo group

All outcomes were presented as graphs and no data were available for analysis except ESR that showed no significant difference between treatment groups

Kirwan 1993, 3 89 years (SSZ: 44)

Primary outcomes

Absolute benefit Did not report from SSZ: 1.3 mm/h Relative difference in change from baseline: 5%

Drop-out: 30. There were sig- Did not report in- 3%. No signif- nificantly

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Did not report

98

Table 2. Randomized controlled trials comparing sulphasalazine with placebo

(Continued)

(Placebo: 45) 28% with PA DD (year): 19±12 (SSZ) 21.9±11.7 (placebo) ESR not given

cluded Schober’s test, chest expansion, and lateral cervical flexion. Secondary outcomes included function (HAQ) , back pain VAS, consumption of anti-inflammatory drugs, sleep disturbance VAS, PGA, episodes of peripheral arthritis, episodes of heel pain, flares in general AS symptoms, episodes of arthritis

icant difference was found between treatment groups in all outcomes except occurrence of peripheral joint symptoms. The episodes of PA were 0.289 episodes/year in SSZ and 0.392 episodes/year in placebo group, respectively (P < 0.05)

more drop-outs for any reason in SSZ than in placebo group. RR was 2. 43 (95% CI 1. 19 to 4.96). No data were available for analysis in any other outcome

Krajnc 1990, 24 95 weeks (SSZ: 71) (Placebo: 24) 66% with PA DD not given ESR (mm/ h): 41±19 (SSZ) 43±18 (placebo)

Duration of morning stiffness, Schober’s test, chest expansion, fingers-tofloor test, number of painful/ swollen joints and ESR

Drop-out: 14.3%. In SSZ group, duration of morning stiffness, chest expansion, number of painful/ swollen joints and ESR showed significantly improved when initial and 24 weeks results are compared

No significant difference was found between treatment groups in all outcomes except ESR (MD -17. 00 mm/h, 95% CI -26.99 to -7. 01mm/h, favoring SSZ)

Absolute benefit from SSZ: 15 mm/h Relatiive difference in change from baseline: 35%

Absoluete benefit from SSZ: -4 mm on 100 mm VAS Relative difference in change from baseline: 8%

Nissila 1988, 26 85 weeks (SSZ: 43 (Placebo: 42) 68% with PA DD (year): 5. 4±7.3 (SSZ) 3.8±4.3 (placebo) ESR (mm/ h): 42±20 (SSZ) 46±19 (placebo)

Duration of morning stiffness, spinal stiffness VAS, chest expansion, Schober’s test, fingers-to-floor test, occiput-towall test, number of painful joints, number of swollen joints,

Drop-out: 12. 2%. Significant differences between treatment groups were found in morning stiffness VAS (P = 0.02), chest expansion (P = 0. 03) and ESR (P = 0.02), favoring SSZ. No signif-

Significant differences were found in morning stiffness VAS 100 mm (0 = no stiffness, 100 = severe, MD -14. 00, 95% CI 23.78 to -4.22) , chest expansion (MD 1.00 cm, 95% CI 0.

Absolute benefit from SSZ: -15 mm/h Relatiive difference in change from baseline: 33%

Absolute benefit from SSZ: -6 mm on 100 mm VAS Relative difference in change from baseline: 15%

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

99

Table 2. Randomized controlled trials comparing sulphasalazine with placebo

Schmidt 2002, 70 26 weeks (SSZ: 34) (Placebo: 36) 36% with PA DD (year): 16. 7±7.2 (SSZ) 16.3±7.8 (placebo) ESR (mm/h): 23.1±3. 2 (SSZ) 20.4±2. 4 (placebo)

(Continued)

general well- icant difference being VAS, ESR was found in and CRP other outcomes Note: we suspected that results of chest expansion were errors because they were impossible to be about 40 to 50 cm. So we divided them by 10 for analysis

10 to 1.90 cm) , occiput-to-wall test (MD -0.80 cm, 95% CI -1. 55 to 0.05 cm) , ESR (MD -19. 00 mm/h, 95% CI -29.65 to -8. 35 mm/h), and general wellbeing VAS 100 mm (0 = the best, 100 = the worst, MD -11.00, 95% CI -19.84 to -2.16) , favoring SSZ. No significant difference was found in other outcomes

Back pain VAS, nocturnal awakening (event) , pain/tenderness score, duration of morning stiffness, number of painful joints, number of swollen joints, spondylitis function index, PGA, PhGA, Schober’s test, fingers-tofloor test, chin sternum distance, chest expansion, ESR and CRP

All continuous outcomes were analyzed as change from baseline. Significant differences were found between treatment groups in back pain VAS 100 mm (0 = no pain, 100 = severe pain, MD -2.30, 95% CI -4.44 to -0.16) , chest expansion (MD 0.30 cm, 95% CI 0. 16 to 0.44 cm) , Schober’s test (MD 0.50 cm, 95 CI 0.44 to 0. 56 cm), duration of morning stiffness (MD -0.39 h, 95% CI -0.48

Drop-out: 36%. No significant difference was reported between treatment groups. There were more dropouts in SSZ than in placebo (38% vs 11%)

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Absolute benefit Did not report from SSZ: -3.1 mm/h Relatiive difference in change from baseline: 15%

100

Table 2. Randomized controlled trials comparing sulphasalazine with placebo

(Continued)

to -0.30 h), ESR (MD -3.10 mm/ h, 95% CI -4.85 to -1.35 mm/h) and CRP (MD 2.50 µg/ml, 95% CI -4.70 to -0. 30 µg/ml), favoring SSZ group. But in occiputto-wall test, the difference (MD 0.70 cm, 95% CI 0.32 to 1. 08 cm) favored placebo over SSZ. No significant difference was found in other outcomes. There were significantly more withdrawals for side effects and dropouts for any reason in SSZ than in placebo group. RRs were 3.44 (95% CI 1. 24 to 9.52) and 2.42 (95% CI 1. 14 to 5.15), respectively Taylor 1991, 1 40 year (SSZ: 20) (Placebo: 20) 15% with PA DD (year): 11±1. 6 (SSZ) 10.7±1. 6 (placebo) ESR (mm/h, mean): 27 (SSZ) 25 (placebo)

Back pain VAS, fingers-to-floor test, chest expansion, sleep disturbance (event) , forced vital volume, occiput-towall test, Schober’s test, spinal stiffness VAS, reduction or stop of NSAIDs (event)

Drop-out: 17. 5%. No significant difference was found between treatment groups in all outcomes except pain VAS (P < 0.05, favoring SSZ)

No signif- Did not report icant difference was found between treatment groups in all outcomes including pain VAS

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Absolute benefit from SSZ: -13.5 mm on 100 mm VAS Relative difference in change from baseline: 42%

101

Table 2. Randomized controlled trials comparing sulphasalazine with placebo

Winkler 1989, 63 24 weeks (SSZ: 31) (Placebo: 32) 33% with PA DD (year, median): 10.8 (SSZ) 11.2 (placebo) ESR (mm/h): 33.4±20.4 (SSZ) 26.9±16.4 (placebo)

ESR, duration of morning stiffness, back pain VAS, score of sleep disturbance, chest expansion, Schober’s test, fingers-tofloor test, disease severity in PGA

Drop-out: 22.2%. The advantage of SSZ over placebo was significant only in the duration of morning stiffness (P < 0.05) and score of sleep disturbance (P< 0.05). In subgroup analysis, the same results were found in patients with axial form (N = 34). In patients with peripheral arthritis (N = 15), articular index showed significant improvement in SSZ over placebo (P < 0.05)

(Continued)

No significant difference was found between treatment groups in all outcomes. In subgroup analysis of patients with axial form, we found significant difference favoring SSZ over placebo in back pain VAS 100 mm (0 = no pain, 100 = severe pain). MD was -9.20 and 95% CI -17.81 to 0.59

Absolute benefit Did not report from SSZ: -2.7 mm/h Relatiive difference in change from baseline: 10%

CI - confidence interval CRP - C-reactive protein DD - duration of disease ESR - erythrocyte sedimentation rate HAQ - health assessment questionnaire MD - mean difference mm/hr - millimetre per hour NSAIDs - non-steroidal anti-inflammatory drugs PA - peripheral arthritis PGA - patient global assessment PhGA - physician global assessment RR - relative risk SD - standard deviation SSZ - sulphasalazine VAS - visual analogue scale

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

102

APPENDICES Appendix 1. Search strategies for the original review

The search strategy for MEDLINE (Ovid) was: 1 Spondylitis, Ankylosing/ 2 (bechtere$ disease$ or marie-struempell disease$ or rheumatoid spondylitis or spondylarthritis ankylopoietica or ankylo$ spondyl$ or Spin$ Ankylosis or Vertebral Ankylosis.tw. 3 1 or 2 4 (Salazosulfapyridine or sulfasalazine or Sulfosalazine or Sulfasal# zine or Salazopyridin$ or asulfidine or azulf#dine).tw. 5 sulfasalazine/ 6 Sulfasalazine.rn. 7 or/4-6 8 3 and 7 9 from 8 keep 1-132

The search strategy for EMBASE (Ovid) was: 1 Spondylitis, Ankylosing/ 2 (bechtere$ disease$ or marie-struempell disease$ or rheumatoid spondylitis or spondylarthritis ankylopoietica or ankylo$ spondyl$ or Spin$ Ankylosis or Vertebral Ankylosis).tw. 3 1 or 2 4 Salazosulfapyridine/ 5 (Salazosulfapyridine or sulfasalazine or Sulfosalazine or Sulfasal# zine or Salazopyridin$ or asulfidine or azulf#dine).tw. 6 4 or 5 7 3 and 6 8 from 7 keep 1-307

Appendix 2. Search strategies for the second (current) review (28 November, 2013) The Cochrane Library Issue 11 of 12, November 2013 1. MeSH descriptor: [Spondylarthropathies] explode all trees 2. MeSH descriptor: [Spondylitis, Ankylosing] explode all trees 3. ankylosing or spondyl.ti,ab 4. bekhterev or bechterew.ti,ab. 5. bechtere* disease* or marie-struempell disease or rheumatoid spondylitis or spondylarthritis ankylopoietica or Spin* Ankylosis or Vertebral Ankylosis 6. MeSH descriptor: [Sulfasalazine] explode all trees 7. Salazosulfapyridine or asulfidine 8. Salazosulfapyridine or sulfasalazine or Sulfosalazine or Sulfasal*zine or Salazopyridin* or asulfidine or azulf*dine 9. #1 or #2 or #3 or #4 or #5 10. #6 or #7 or #8 11. #9 and #10 EMBASE Classic + EMBASE 1947 to Nov 27 2013 1. Spondylitis, Ankylosing/ 2. (bechtere$ disease$ or marie-struempell disease$ or rheumatoid spondylitis or spondylarthritis ankylopoietica or ankylo$ spondyl$ or Spin$ Ankylosis or Vertebral Ankylosis).tw. 3. 1 or 2 4. (Salazosulfapyridine or sulfasalazine or Sulfosalazine or Sulfasal#zine or Salazopyridin$ or asulfidine or azulf#dine).tw. 5. sulfasalazine/ 6. 4 or 5 7. 3 and 6 8. random$.tw. 9. factorial$.tw 10. crossover$.tw. 11. cross over.tw. 12. cross-over.tw. 13. placebo$.tw. Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

103

14. (doubl$ adj blind$).tw. 15. (singl$ adj blind$).tw. 16. assign$.tw. 17. allocat$.tw. 18. volunteer$.tw. 19. crossover procedure/ 20. double blind procedure/ 21. randomized controlled trial/ 22. single blind procedure/ 23. or/8-22 24. 7 and 23 Database: Ovid MEDLINE(R) In-Process Other Non-Indexed Citations and Ovid MEDLINE(R) 1. Spondylitis, Ankylosing/ 2. (bechtere$ disease$ or marie-struempell disease$ or rheumatoid spondylitis or spondylarthritis ankylopoietica or ankylo$ spondyl$ or Spin$ Ankylosis or Vertebral Ankylosis).tw. 3. 1 or 2 4. (Salazosulfapyridine or sulfasalazine or Sulfosalazine or Sulfasal#zine or Salazopyridin$ or asulfidine or azulf#dine).tw. 5. sulfasalazine/ 6. Sulfasalazine.rn. 7. or/4-6 8. 3 and 7 CINAHL via Ebscohost

S16

S10 and S15

S15

S11 or S12

S14

“asulfidine”

S13

“Sulfosalazine”

S12

“Salazosulfapyridine”

S11

(MM “Sulfasalazine”)

S10

S1 or S6 or S8 or S9

S9

(MH “Spondylosis”)

S8

“Vertebral Ankylosis”

S7

“spondylarthritis ankylopoietica”

S6

(MH “Spondylarthritis”)

S5

“rheumatoid spondylitis”

S4

“”marie-struempell“”

S3

“marie-struempell”

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

104

(Continued)

S2

“bechtere disease”

S1

(MH “Spondylitis, Ankylosing”)

ClinicalTrials.gov Search Strategy: (advanced screen) Intervention: Salazosulfapyridine or sulfasalazine or asulfidine

Appendix 3. Search result for the second version (28 November, 2013)

Database, platform, issue

Results

Results after deduplicate

MEDLINE-OVID (1946 to November 28, 2013 present) 2012-2013

24

23

EMBASE-OVID (1974 to Nov November 28, 2013 26 2013) 2012-2013

21

21

(CDSR) part of The Cochrane November 28, 2013 Library 2012-2013 www.thecochranelibrary.com

4

4

CENTRAL, part of The November 28, 2013 Cochrane Library 2012-2013 www.thecochranelibrary.com

0

0

DARE, part of The Cochrane Li- November 28, 2013 brary 2012-2013 www.thecochranelibrary.com

0

0

HTA Database, part of The November 28, 2013 Cochrane Library 2012-2013 www.thecochranelibrary.com

0

0

NHS EED, part of The November 28, 2013 Cochrane Library 2012-2013 www.thecochranelibrary.com

0

0

CINAHL

5

5

101

100

TOTALS

Search date

November 28, 2013

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

105

Appendix 4. Risk of bias assessment criteria

SEQUENCE GENERATION Was the allocation sequence adequately generated? [Short form: Adequate sequence generation?] Criteria for a judgment of ’YES’ (i.e. low risk of bias)

The investigators describe a random component in the sequence generation process such as: • Referring to a random number table • Using a computer random number generator • Coin tossing • Shuffling cards or envelopes • Throwing dice • Drawing of lots • Minimization* *Minimization may be implemented without a random element, and this is considered to be equivalent to being random

Criteria for the judgment of ’NO’ (i.e. high risk of bias)

The investigators describe a non-random component in the sequence generation process. Usually, the description would involve some systematic, non-random approach, for example: • Sequence generated by odd or even date of birth • Sequence generated by some rule based on date (or day) of admission • Sequence generated by some rule based on hospital or clinic record number Other non-random approaches happen much less frequently than the systematic approaches mentioned above and tend to be obvious. They usually involve judgment or some method of nonrandom categorization of participants, for example: • Allocation by judgment of the clinician • Allocation by preference of the participant • Allocation based on the results of a laboratory test or a series of tests • Allocation by availability of the intervention

Criteria for the judgment of ’UNCLEAR’ (uncertain risk of bias) Insufficient information about the sequence generation process to permit judgment of ’Yes’ or ’No’

ALLOCATION CONCEALMENT Was allocation adequately concealed? [Short form: Allocation concealment?] Criteria for a judgment of ’YES’ (i.e. low risk of bias)

Participants and investigators enrolling participants could not foresee assignment because one of the following, or an equivalent method, was used to conceal allocation: • Central allocation (including telephone, web-based, and pharmacy-controlled, randomisation) • Sequentially numbered drug containers of identical

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

106

(Continued)

appearance • Sequentially numbered, opaque, sealed envelopes

Criteria for the judgment of ’NO’ (i.e. high risk of bias)

Participants or investigators enrolling participants could possibly foresee assignments and thus introduce selection bias, such as allocation based on: • Using an open random allocation schedule (e.g. a list of random numbers) • Assignment envelopes were used without appropriate safeguards (e.g. if envelopes were unsealed or non-opaque or not sequentially numbered) • Alternation or rotation • Date of birth • Case record number • Any other explicitly unconcealed procedure

Criteria for the judgment of ’UNCLEAR’ (uncertain risk of bias) Insufficient information to permit judgment of ’Yes’ or ’No’. This is usually the case if the method of concealment is not described or not described in sufficient detail to allow a definite judgment C for example if the use of assignment envelopes is described, but it remains unclear whether envelopes were sequentially numbered, opaque and sealed

BLINDING OF PARTICIPANTS, PERSONNEL AND OUTCOME ASSESSORS Was knowledge of the allocated interventions adequately prevented during the study? [Short form: Blinding?] Criteria for a judgment of ’YES’ (i.e. low risk of bias)

Any one of the following: • No blinding, but the review authors judge that the outcome and the outcome measurement are not likely to be influenced by lack of blinding • Blinding of participants and key study personnel ensured, and unlikely that the blinding could have been broken • Either participants or some key study personnel were not blinded, but outcome assessment was blinded and the nonblinding of others unlikely to introduce bias

Criteria for the judgment of ’NO’ (i.e. high risk of bias)

Any one of the following: • No blinding or incomplete blinding, and the outcome or outcome measurement is likely to be influenced by lack of blinding • Blinding of key study participants and personnel attempted, but likely that the blinding could have been broken • Either participants or some key study personnel were not blinded, and the non-blinding of others likely to introduce bias

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

107

(Continued)

Criteria for the judgment of ’UNCLEAR’ (uncertain risk of bias) Any one of the following: • Insufficient information to permit judgment of ’Yes’ or ’No’ • The study did not address this outcome

INCOMPLETE OUTCOME DATA Were incomplete outcome data adequately addressed? [Short form: Incomplete outcome data addressed?] Criteria for a judgment of ’YES’ (i.e. low risk of bias)

Any one of the following: • No missing outcome data • Reasons for missing outcome data unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias) • Missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups • For dichotomous outcome data, the proportion of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate • For continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes not enough to have a clinically relevant impact on observed effect size • Missing data have been imputed using appropriate methods

Criteria for the judgment of ’NO’ (i.e. high risk of bias)

Any one of the following: • Reason for missing outcome data likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups • For dichotomous outcome data, the proportion of missing outcomes compared with observed event risk enough to induce clinically relevant bias in intervention effect estimate • For continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes enough to induce clinically relevant bias in observed effect size • ’As-treated’ analysis done with substantial departure of the intervention received from that assigned at randomisation • Potentially inappropriate application of simple imputation

Criteria for the judgment of ’UNCLEAR’ (uncertain risk of bias) Any one of the following: • Insufficient reporting of attrition/exclusions to permit judgment of ’Yes’ or ’No’ (e.g. number randomized not stated, no reasons for missing data provided) • The study did not address this outcome

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

108

SELECTIVE OUTCOME REPORTING Are reports of the study free of suggestion of selective outcome reporting? [Short form: Free of selective reporting?] Criteria for a judgment of ’YES’ (i.e. low risk of bias)

Any of the following: • The study protocol is available and all of the study’s prespecified (primary and secondary) outcomes that are of interest in the review have been reported in the prespecified way • The study protocol is not available but it is clear that the published reports include all expected outcomes, including those that were prespecified (convincing text of this nature may be uncommon)

Criteria for the judgment of ’NO’ (i.e. high risk of bias)

Any one of the following: • Not all of the study’s prespecified primary outcomes have been reported • One or more primary outcomes is reported using measurements, analysis methods or subsets of the data (e.g. subscales) that were not prespecified • One or more reported primary outcomes were not prespecified (unless clear justification for their reporting is provided, such as an unexpected adverse effect) • One or more outcomes of interest in the review are reported incompletely so that they cannot be entered in a meta-analysis • The study report fails to include results for a key outcome that would be expected to have been reported for such a study

Criteria for the judgment of ’UNCLEAR’ (uncertain risk of bias) Insufficient information to permit judgment of ’Yes’ or ’No’. It is likely that the majority of studies will fall into this category

OTHER POTENTIAL THREATS TO VALIDITY Was the study apparently free of other problems that could put it at a risk of bias? [Short form: Free of other bias?] Criteria for a judgment of ’YES’ (i.e. low risk of bias)

The study appears to be free of other sources of bias

Criteria for the judgment of ’NO’ (i.e. high risk of bias)

There is at least one important risk of bias. For example, the study: • Had a potential source of bias related to the specific study design used; or • Stopped early due to some data-dependent process (including a formal-stopping rule); or • Had extreme baseline imbalance; or • Has been claimed to have been fraudulent; or • Had some other problem

Criteria for the judgment of ’UNCLEAR’ (uncertain risk of bias) There may be a risk of bias, but there is either: • Insufficient information to assess whether an important risk of bias exists; or Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

109

(Continued)

• Insufficient rationale or evidence that an identified problem will introduce bias

Appendix 5. Sample data collection form for assessment of risk of bias

DOMAIN

JUDGMENT (YES, NO, UNCLEAR)

REASON FOR JUDGMENT (copy and paste directly from text of trial)

Was the allocation sequence adequately generated? Was allocation adequately concealed? Was knowledge of the allocated interventions adequately prevented during the study? Blinding of personnel Blinding of participants Blinding of outcome assessors Were incomplete outcome data adequately addressed? Are reports of the study free of suggestion of selective outcome reporting Was the study apparently free of other problems that could put it at a risk of bias?

WHAT’S NEW Last assessed as up-to-date: 28 November 2013.

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

110

Date

Event

Description

28 November 2013

New search has been performed

New search but no new studies included

14 September 2012

New citation required but conclusions have not Change in authorship changed

HISTORY Protocol first published: Issue 2, 2004 Review first published: Issue 2, 2005

Date

Event

Description

9 November 2008

Amended

Converted to new review format. CMSG ID: C079-R

24 February 2005

Amended

Review first published

9 November 2004

Amended

Protocol first published

CONTRIBUTIONS OF AUTHORS JC: Registered the title; developed the protocol; searched for relevant studies; selected the studies and assessed their risk of bias; extracted and synthesized the data; and wrote the systematic review. SL (for the update), CL (for the original review): Selected studies, assessed their risk of bias and extracted data from them; and wrote the systematic review, independently from JC.

DECLARATIONS OF INTEREST JC - none SL - none CL - none

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

111

SOURCES OF SUPPORT Internal sources • The First Affiliated Hospital of Fujian Medical University, China.

External sources • Science development fund, Fujian Medical University, China.

DIFFERENCES BETWEEN PROTOCOL AND REVIEW In this updated review, we used The Cochrane Collaboration’s tool (Higgins 2011) for assessing risk of bias. We used GRADEpro software and included a ’Summary of findings’ table to present the main results. We modified the primary and secondary outcomes to reflect the recommendation by the CMSG editors.

INDEX TERMS Medical Subject Headings (MeSH) Antirheumatic Agents [∗ therapeutic use]; Randomized Controlled Trials as Topic; Spondylitis, Ankylosing [∗ drug therapy]; Sulfasalazine [∗ therapeutic use]

MeSH check words Humans

Sulfasalazine for ankylosing spondylitis (Review) Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

112

Sulfasalazine for ankylosing spondylitis.

Ankylosing spondylitis (AS) is a chronic inflammatory disease of unknown cause and affects mainly the spine, but can also affect other joints. Disease...
1MB Sizes 0 Downloads 15 Views