Skip to main content
NCBI home page
JAMA Network logoLink to JAMA Network
. 2023 Oct 6;6(10):e2335950. doi: 10.1001/jamanetworkopen.2023.35950

Monotreatment With Conventional Antirheumatic Drugs or Glucocorticoids in Rheumatoid Arthritis

A Network Meta-Analysis

Louise S Guski 1, Gesche Jürgens 1, Hugo Pedder 2, Niels K G Levinsen 1, Stig E Andersen 1, Nicky J Welton 2, Niels Graudal 3,✉
PMCID: PMC10559183  PMID: 37801318

The meta-analysis investigates the association of conventional antirheumatic drugs and glucocorticoids with clinical outcomes among patients with rheumatoid arthritis.

Key Points

Question

Which conventional treatments are associated with the best clinical outcomes among patients with rheumatoid arthritis?

Findings

In this meta-analysis of 29 interventions investigated in 132 randomized clinical trials of 13 260 patients with rheumatoid arthritis, methotrexate was the most thoroughly investigated drug with the most reliable outcome.

Meaning

These findings suggest that methotrexate is the anchor untargeted, conventional treatment for rheumatoid arthritis but that in case of suboptimal outcomes or unacceptable adverse events, several alternative conventional treatments are available.

Abstract

Importance

This is the first network meta-analysis to assess outcomes associated with multiple conventional synthetic disease-modifying antirheumatic drugs and glucocorticoid.

Objective

To analyze clinical outcomes after treatment with conventional synthetic disease-modifying antirheumatic drugs and glucocorticoid among patients with rheumatoid arthritis.

Data Sources

With no time restraint, English language articles were searched in MEDLINE, Embase, Cochrane Central, ClinicalTrials.gov, and reference lists of relevant meta-analyses until September 15, 2022.

Study Selection

Four reviewers in pairs of 2 independently included controlled studies randomizing patients with rheumatoid arthritis to mono–conventional synthetic disease-modifying antirheumatic drugs, glucocorticoid, placebo, or nonactive treatment that recorded at least 1 outcome of tender joint count, swollen joint count, erythrocyte sedimentation rate, and C-reactive protein level. Of 1098 assessed articles, 130 articles (132 interventions) were included.

Data Extraction and Synthesis

The review followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses reporting guideline, and data quality was assessed by the Cochrane risk of bias tool RoB 2. Data were extracted by a single author and checked independently by 2 authors. Data were analyzed using a random effect model, and data analysis was conducted from June 2021 to February 2023.

Main Outcomes and Measures

A protocol with hypothesis and study plan was registered before data recording. The most complete of recorded outcomes (tender joint count) was used as primary outcome, with imputations based on other outcomes to obtain a full analysis of all studies. Absolute change adjusted for baseline disease activity was assessed.

Results

A total of 29 interventions in 275 treatment groups among 132 randomized clinical trials (mean [range], 71.0% [27.0% to 100%] females in studies; mean [range] of ages in studies, 53 [36 to 70] years) were identified, which included 13 260 patients with rheumatoid arthritis. The mean (range) duration of RA was 79 (2 to 243) months, and the mean (range) disease activity score was 6.3 (4.0 to 8.8). Compared with placebo, oral methotrexate was associated with a reduced tender joint count by 5.18 joints (95% credible interval [CrI], 4.07 to 6.28 joints). Compared with methotrexate, glucocorticoid (−2.54 joints; 95% CrI, −5.16 to 0.08 joints) and remaining drugs except cyclophosphamide (6.08 joints; 95% CrI, 0.44 to 11.66 joints) were associated with similar or lower tender joint counts.

Conclusions and Relevance

This study’s results support the present role of methotrexate as the primary reference conventional synthetic disease-modifying antirheumatic drug.

Introduction

Since the discovery of treatments for rheumatoid arthritis (RA) using gold in 1932 and glucocorticoid (GC) in 1949,1,2 many synthetic disease-modifying antirheumatic drugs (csDMARDs) have been introduced.3 The use of targeted DMARDs (tDMARDs), including biologic and targeted synthetic,4,5 has increased since 2000.

Previously, outcome measures of RA reflected joint inflammation (morning stiffness, number of tender joints [tender joint count (TJC)]) and swollen joints [swollen joint count (SJC)], and limitation of joint motion), systemic inflammation (erythrocyte sedimentation rate [ESR] and C-reactive protein [CRP] level6,7), and joint destruction (radiographic evaluation8,9). While the latter still is a separate outcome, clinical and biochemical variables have been integrated in composite disease activity scores and improvement scores,10,11,12,13 which may include TJC, SJC, ESR, and CRP.

A 2020 review14 of network meta-analyses (NMAs) of DMARDs included 37 NMAs of tDMARDs and no csDMARDs. However, 70% of patients treated for RA receive csDMARDs and 25% receive tDMARDs,15 and a combination of 2 to 3 csDMARDs may be associated with similar effectiveness as tDMARDs.16,17 Furthermore, tDMARDs are used only by specialists, whereas general practitioners may use csDMARDs. Thus, outcomes associated with csDMARDs may be of general interest. Previously, we estimated the relative effect sizes associated with 10 csDMARDs and GC using the sporadically reported radiographic destruction outcome.16 The objective in this study was to estimate relative effect sizes associated with multiple csDMARDS and GC in an integrated NMA of RCTs using frequently reported outcomes. Given that older RCTs of csDMARDs rarely used composite outcomes, we analyzed differences in absolute change of measures of TJC, SJC, ESR, and CRP adjusted for potential effect modifiers.

Methods

Protocol and Registration

A protocol for this meta-analysis was registered at Prospero on July 5, 2020 (CRD42020184585). The study was approved by the Faculty of Health and Medical Sciences, University of Copenhagen, and is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline.

Eligibility Criteria

We included trials that randomly allocated patients with RA to csDMARD, GC, placebo, or a pharmacologic non–disease-modifying comparator (no treatment; pain-relieving treatment, including nonsteroidal anti-inflammatory drugs; or very-low dose of the investigated DMARD) and that reported at least 1 of the outcome variables (eMethods 1 in Supplement 1).

Information sources (eMethods 2 in Supplement 1) included MEDLINE, Cochrane Central, Embase, Clinicaltrials.gov, and reference lists of included studies and previous meta-analyses. The search was conducted on June 13, 2018, and updated on January 12, 2022. On September 15, 2022, we performed a post hoc screening search in PubMed covering September 15, 2021, to September 15, 2022.

Search, Study Selection, and Data Collection

The electronic search string is shown in eMethods 2 in Supplement 1. Irrelevant studies were eliminated by 1 author (L.S.G.). Exclusion criteria could be not an RCT, study population including patients without RA, no report of defined outcomes, or no use of csDMARDs. Then, 2 pairs of authors (L.S.G./G.J. and S.E.A./N.G.), who were blinded to each other’s decisions, independently repeated the procedure while categorizing studies from the remaining pool according to type of intervention. Data were recorded on a standardized extraction form using Excel spreadsheet software version 16.0.166626.20134 (Microsoft). Data were collected by a single author (L.S.G.). All data were independently controlled by 2 authors (N.K.G.L. and N.G.). Any disagreements were resolved by discussion between 2 authors (L.S.G. and N.G.). See eMethods 3 in Supplement 1.

Data Items

The study population was defined as the number of patients who were randomized and received at least 1 dose of drug. We estimated factors to convert 1 outcome to another outcome to be able to impute missing outcomes. Thus, the estimated relative effect size of the missing outcome would correspond to the relative effect size of the reported outcome. Missing TJC values for the primary outcome and for the disease activity score based on 28 joints (DAS28) estimation were imputed as explained in eMethods 4 and 5 in Supplement 1. Missing SJC and ESR values for the DAS28 estimation were imputed as explained in eMethods 5 in Supplement 1. If TJC and SJC were assessed on more than 28 joints, TJC and SJC were converted to an assessment of 28 joints (TJC28 and SJC28) based on conversion factors estimated by Fuchs and Pincus18 shown in eMethods 5 in Supplement 1. Then DAS28 was calculated by the following formula: DAS28 = (0.56 × sqrt[tjc28] + 0.28 × sqrt(sjc28) + 0.70 × Ln[ESR]) × 1.08 + 0.16. To calculate SE, we used a recently calculated mean SD of 1.1.19 See eMethods 4 and 5 in Supplement 1.

Geometry of the Network

We drew the network using RStudio statistical software version 1.4.1103 (RStudio)20 by a code from the develop branch. Chloroquine and hydroxychloroquine were merged into 1 node and combined across doses. Placebo and comparators without systemic anti-inflammatory properties were merged into the placebo node. All other nodes were defined by a single intervention and combined across doses.

Statistical Analysis

Risk of Bias Within Individual Studies

Risk of bias was assessed at study level by the Cochrane risk of bias tool RoB 2.21 See eMethods 6 in Supplement 1.

Summary Measures

For our primary analysis of all 132 studies (represented by 130 publications22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151) we used the most frequently reported outcome, TJC. This included imputed TJC values for studies that did not measure TJC (eMethods 5 in Supplement 1), denominated TJCi.

For our sensitivity analysis, we used ESR, CRP, SJC, TJC without imputations, estimated TJC28 with imputations (TJC28i) (eMethods 5 in Supplement 1), and estimated DAS28 with imputations (DAS28i) (eMethods 5 in Supplement 1) as efficacy measures. The outcome measure for TJCi, TJC, SJC, ESR, CRP, DAS28i, and TJC28i was defined as the difference between the baseline mean value and the posttreatment mean value. We used absolute changes, which statistically are more efficient than percentage changes.152 Treatment outcomes of TJCi were categorized as favorable compared with placebo and methotrexate, favorable compared with placebo and not different from methotrexate, favorable compared with placebo but unfavorable compared with methotrexate, not different from placebo or methotrexate, and not different from placebo but unfavorable compared with methotrexate. Favorable means that the 2.5% credible interval (CrI) was higher than the 97.5% CrI of the compared drug. Unfavorable means that the 97.5% CrI was lower than the 2.5% CrI of the compared drug. Not different means that the CrIs were overlapping. Rankograms were plotted for all treatments. Finally, we assessed dropout rates. For each study the number of dropouts was recorded from the pool of all randomized patients. The sum of dropouts for each treatment group was divided by the sum of randomized patients for each treatment group.

Planned Methods of Analyses153,154,155,156,157,158

We conducted our network meta-analysis following the method of Lu and Ades153 using WinBUGS code taken from NICE Technical Support Document 2,154 in which handling of multigroup trials is a built-in function. We ran models using prior 0 to 5 as explained in eMethods 7 in Supplement 1. Post hoc, we reran all models with different priors and created density plots, and we ran all models with a uniform 0 to 20 prior in August 2023.

Additional Analyses

Network metaregression models were fitted using WinBUGS155 with code from NICE Technical Support Document 3,158 assuming a common regression effect size across pooled treatment effect sizes vs methotrexate. For effect modifier regression analyses, we ran univariate regression analyses for possible numerical effect modifiers: duration of study, duration of rheumatoid arthritis, publication year, mean age, percentage female, percentage with rheumatoid-factor positivity, baseline DAS28, and bias score based on the Cochrane risk of bias tool. See eMethods 8 in Supplement 1.

Subgroup analyses (eMethods 8 in Supplement 1) were used to test for the association of categorical variables (type of statistical analysis [intention to treat or completer], concomitant GC during study allowed [yes or no], and previous use of DMARDs (no, yes unspecified, or yes with inadequate response]) with outcomes. To investigate the suitability of merging chloroquine and hydroxychloroquine into 1 node, we ran the main model of TJCi separating the 2 treatments. Data were analyzed from June 2021 to February 2023 and post hoc in August 2023.

Results

Study Characteristics

A total of 29 interventions among 132 randomized clinical trials (RCTs) represented by 130 publications22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151 (mean [range], 71.0% [27.0%-100%] females in studies; mean [range] of ages in studies, 53 [36-70] years) were identified, which included 13 260 patients with rheumatoid arthritis. The number of studies with measured outcomes was 118 studies23,26,28,29,30,34,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,66,67,68,69,70,71,72,73,74,75,76,77,79,80,81,82,83,84,85,87,88,89,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151 for TJC, 116 studies22,23,24,25,26,27,28,29,30,32,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,80,81,82,84,85,87,88,89,90,91,92,93,94,96,97,98,99,100,101,104,105,106,107,108,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,127,128,129,135,136,137,140,141,142,143,144,146,147,148,149,150,151 for ESR, 97 studies22,26,29,30,31,32,33,37,39,40,41,44,45,46,50,51,52,54,57,58,60,62,63,65,67,68,69,70,71,72,73,74,75,76,77,78,80,81,82,83,84,86,87,88,92,93,94,96,97,98,100,101,102,103,105,107,108,109,110,111,113,114,115,116,118,119,120,122,123,124,125,127,128,129,130,131,132,133,134,135,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151 for SJC, and 47 studies40,45,56,58,64,66,67,79,85,87,95,97,98,100,104,110,111,113,114,115,116,117,122,124,125,126,127,128,129,131,132,133,134,136,137,138,140,141,142,144,146,147,148,149,150,151 for CRP. The number of study groups and participants and the dropout rate (percentage) for each intervention group appears in the Table. The dropout rate ranged from 36 of 291 participants (12.4%) for iguratimod to 91 of 169 participants (53.5%) for pyritinol. Characteristics and results of individual studies and interventions are presented in eTable 1 in Supplement 1. At baseline, the mean (range) for duration of RA was 79 (2-243) months, TJC was 18.6 (5.3-48.2) joints, SJC was 15.8 (5.0-41.6) joints, baseline ESR was 51.6 (23.8-91.6) mm/h, baseline CRP level was 4.12 (1.10-8.64) mg/dL (to convert to milligrams per liter, multiply by 10), DAS28 was 6.3 (4.0-8.8), and rheumatoid factor positivity was 77% (0%-100%).

Table. Primary Outcome and Sensitivity Outcomes With Methotrexate as Reference.

Drug Intervention groups, No. Participants Reduction in outcome (95% CrI)a
Randomized and treated, No. Dropped out, No./No. randomized (%) TJCi (n = 132) TJC (n = 118) SJC (n = 97) ESR, mm/h (n = 116) CRP level, mg/L (n = 47) DAS28i (n = 132) TJC28i (n = 132)
Methotrexate 31 2344 539/2549 (21.1) 0 [Reference] 0 [Reference] 0 [Reference] 0 [Reference] 0 [Reference] 0 [Reference] 0 [Reference]
Sulfasalazine 18 762 399/964 (41.4) −1.90 (−3.15 to −0.57)b −1.91 (−3.20 to −0.54)b −1.39 (−2.50 to −0.25)b 1.43 (−2.62 to 5.55)c −4.91 (−14.24 to 4.97)c −0.14 (−0.90 to 0.62)c −1.69 (−2.64 to −0.69)b
Leflunomide 8 1321 361/1397 (25.8) −0.26 (−1.23 to 0.89)c −0.25 (−1.22 to 0.91)c −0.92 (−1.57 to −0.18)b −7.53 (−10.50 to −4.08)b 0.90 (−5.10 to 7.57)d −0.26 (−0.76 to 0.24)c −0.35 (−1.04 to 0.51)c
Cyclosporine 11 492 147/613 (24.0) −0.79 (−2.41 to 0.88)c −0.78 (−2.43 to 0.91)c −1.14 (−2.39 to 0.12)c −9.97 (−15.64 to −4.25)d 6.52 (−7.98 to 20.64) −0.32 (−1.28 to 0.66)d −0.86 (−2.04 to 0.34)c
Injected gold 25 777 439/1126 (39.0) −0.71 (−2.10 to 0.61)c −0.72 (−2.13 to 0.62)c −0.80 (−2.07 to 0.45)c 0.29 (−4.07 to 4.65)c 5.09 (−6.25 to 16.72) −0.05 (−0.82 to 0.74)d −1.22 (−2.26 to −0.17)b
Chloroquine 19 662 133/721 (18.4) −2.51 (−4.02 to −0.93)b −2.41 (−4.03 to −0.73)b −2.20 (−3.62 to −0.81)b −7.25 (−11.86 to −2.56)b −11.19 (−23.13 to 0.36)d −0.42 (−1.32 to 0.49)d −2.15 (−3.27 to −0.97)b
D-penicillamine 28 743 321/974 (33.0) −2.21 (−3.71 to −0.67)b −2.28 (−3.82 to −0.68)b −0.90 (−2.32 to 0.54)c 2.38 (−2.70 to 7.48)c −0.65 (−16.34 to 14.81)d −0.10 (−1.03 to 0.80)d −1.46 (−2.63 to −0.27)b
Azathioprine 12 347 153/449 (34.1) −2.57 (−4.21 to −0.88)b −2.61 (−4.31 to −0.87)b −1.40 (−2.88 to 0.08)c −7.00 (−13.88 to 0.02)d −1.57 (−16.67 to 13.43)d −0.34 (−1.47 to 0.82)d −1.55 (−2.93 to −0.18)b
Oral gold 20 1200 462/1417 (32.6) −2.70 (−4.19 to −1.18)b −3.03 (−4.54 to −1.46)b −2.45 (−3.76 to −1.11)b −6.70 (−10.88 to −2.46)b −16.22 (−30.16 to −2.42)d −0.45 (−1.20 to 0.30)d −2.35 (−3.37 to −1.29)b
Cyclophosphamide 3 41 10/51 (19.6) 6.08 0.44 to 11.66)e 8.34 (2.06 to 14.61)e 8.42 (4.59 to 12.27)e −0.52 (−11.73 to 10.84)c NA 0.12 (−2.14 to 2.32)d 3.53 (−0.17 to 7.26)c
Levamisole 10 306 146/408 (35.8) −2.46 (−4.11 to −0.71)b −2.43 (−4.10 to −0.67)b −2.73 (−4.07 to −1.37)c 1.37 (−5.31 to 8.20)c 10.18 (−6.76 to 26.83) −0.15 (−1.35 to 1.03)d −1.05 (−2.48 to 0.37)c
Tacrolimus 3 275 115/287 (40.1) 1.31 (−1.46 to 4.12)c 1.39 (−1.41 to 4.22)c −0.82 (−2.81 to 1.16)c 1.73 (−6.26 to 9.74)c 0.48 (−12.18 to 13.71) −0.10 (−1.45 to 1.25)d 0.48 (−1.45 to 2.43)c
Tripterygium wilfordii 2 113 33/129 (25.6) −1.29 (−5.69 to 3.11)d −1.30 (−5.67 to 3.06)d −0.32 (−2.77 to 2.12)c −1.65 (−11.02 to 7.85)c −0.42 (−14.49 to 14.19)d −0.028 (−1.93 to 1.90)d −1.19 (−4.43 to 2.05)d
Iguratimod 3 290 36/291 (12.4) −1.13 (−2.99 to 0.71)c −1.11 (−2.98 to 0.73)c −0.75 (−2.05 to 0.56)c −0.61 (−7.85 to 6.77)c −0.01 (−9.58 to 9.49) −0.12 (−1.26 to 1.01)d −0.94 (−2.49 to 0.62)c
Timegadine 1 19 12/23 (52.2) −4.66 (−18.77 to 9.35)d −4.73 (−18.77 to 9.31)d −4.20 (−16.03 to 7.95)d 20.17 (−4.11 to 44.84)c NA 0.05 (−4.86 to 4.86)d −2.95 (−11.78 to 5.84)d
Bucillamine 1 24 3/24 (12.5) −0.85 (−12.94 to 11.03)d −0.95 (−13.10 to 11.14)d 1.29 (−6.54 to 9.20)d 1.95 (−28.56 to 32.39)d 10.34 (−27.66 to 48.19)d 0.11 (−4.49 to 4.81)d −0.50 (−9.01 to 7.99)d
Tiopronin 1 25 9/28 (32.1) −3.00 (−9.63 to 3.70)d −3.06 (−9.71 to 3.63)d NA 9.27 (−11.76 to 30.36)c NA 0.019 (−3.79 to 3.76)d −2.07 (−6.95 to 2.90)d
Pyritinol 1 139 91/169 (53.5) −2.11 (−6.22 to 2.03)d −2.43 (−6.51 to 1.74)d NA 1.23 (−9.57 to 12.03)c NA −0.23 (−2.11 to 1.62)d −1.91 (−4.80 to 1.02)d
Mycophenolate 2 286 136/291 (46.7) −3.57 (−7.09 to −0.03)f −3.50 (−7.02 to 0.04)d −3.12 (−6.11 to −0.12)f −12.26 (−20.16 to −4.37)f −13.23 (−27.55 to 1.34)d −0.63 (−2.31 to 1.03)d −2.51 (−5.78 to 0.78)d
OM-8980 2 86 23/96 (24.0) −2.20 (−5.74 to 1.34)d −2.44 (−6.02 to 1.11)d −2.89 (−4.55 to −1.16)f −6.38 (−17.94 to 5.13)d NA −0.32 (−2.37 to 1.72)d −1.77 (−4.33 to 0.78)d
Lobenzarit 1 86 36/115 (31.3) −3.55 (−7.10 to 0.0074)d NA −2.82 (−5.89 to 0.29)d −5.42 (−20.82 to 9.911)d NA −0.55 (−2.63 to 1.51)d −3.11 (−5.34 to −0.88)f
Rheumacon 2 80 13/80 (16.3) −2.82 (−5.24 to −0.32)f −2.96 (−5.39 to −0.46)f −2.84 (−7.96 to 2.20)d −21.84 (−35.67 to −7.72)d −16.39 (−35.70 to 3.29)d −0.70 (−2.73 to 1.28)d −2.36 (−4.05 to −0.63)f
Prospidin 1 23 4/27 (14.8) 0.21 (−7.19 to 7.65)d 0.17 (−7.28 to 7.68)d 3.33 (−4.47 to 11.12)d 9.36 (−3.58 to 22.25)c 7.28 (−8.41 to 23.23) 0.51 (−3.15 to 4.2)d 0.052 (−4.57 to 4.74)d
Aminopterin 1 11 2/13 (15.4) 10.37 (−12.97 to 33.61)d 5.50 (−17.38 to 28.28)d 8.34 (−13.84 to 30.55)d NA NA 0.92 (−6.44 to 8.37)d 8.36 (−8.07 to 24.75)d
Eazmov plus 1 18 6/21 (28.6) −2.84 (−15.14 to 9.72)d −2.43 (−15.08 to 10.12)d −2.64 (−10.33 to 5.01)d −11.03 (−23.63 to 1.52)d NA −0.35 (−4.80 to 3.99)d −2.26 (−14.32 to 10.24)d
Dapsone 2 44 13/57 (22.8) −1.38 (−5.24 to 2.49)c −0.90 (−5.11 to 3.28)c 0.46 (−5.31 to 6.19)d −3.66 (−19.36 to 12.19)d 4.27 (−18.95 to 27.47)d −0.17 (−2.88 to 2.49)d −1.38 (−4.05 to 1.35)c
Glucocorticoid 4 153 28/122 (23.0) −2.54 (−5.16 to 0.08)c −3.09 (−6.56 to 0.30)d −3.25 (−5.28 to −1.24)f −1.76 (−10.78 to 7.36)d −12.16 (−32.55 to 7.82)d −0.30 (−1.92 to 1.29)d −1.942 (−4.00 to 0.15)c
Injectable methotrexate 2 206 32/214 (15.0) 1.04 (−2.24 to 4.25)c 0.98 (−2.67 to 4.51)c −0.80 (−2.97 to 1.36)c −4.77 (−17.23 to 7.60)d NA 0.61 (−1.80 to 2.96)d 0.59 (−1.41 to 2.55)c
Placebo 60 2900 1042/3253 (32.0) −5.18 (−6.27 to −4.09) −4.83 (−6.08 to −3.61) −4.44 (−5.36 to −3.53) −13.35 (−16.79 to −9.87) −13.3 (−21.81 to −4.97) −0.88 (−1.51 to −0.25) −4.14 (−4.97 to −3.28)
Open in a new tab

Abbreviations: CRP, C-reactive protein; DAS28i, disease activity score based on 28 joints with imputations; ESR, erythrocyte sedimentation rate; NA, not applicable (no comparisons of this outcome within this intervention); SJC, swollen joint count; TJC, tender joint count; TJC28i, tender joint score based on 28 joints with imputations;TJCi, tender joint count imputed.

SI conversion factor: To convert CRP to milligrams per deciliter, divide by 10.

a

A negative sign indicates that the outcome reduction is less than the reduction associated with methotrexate.

b

Favorable vs placebo and unfavorable vs methotrexate.

c

Favorable vs placebo and not different vs methotrexate.

d

Not different vs placebo and not different vs methotrexate.

e

Favorable vs placebo and favorable vs methotrexate.

f

Not different vs placebo and unfavorable vs methotrexate.

Study Selection

See eResults 1 in Supplement 1. Figure 1 shows a flowchart of the identification of the pool of RCTs. We included 131 publications,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,159 which represented 133 studies. In 1 study,159 the 2 interventions (sulfapyridine and 5-aminosalicylic acid) were not connected to the network. The remaining 132 studies22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151 were included in the NMA. In the post hoc screening in PubMed on September 15, 2022, we did not identify relevant studies (eResults 1 in Supplement 1).

Figure 1. Flowchart of Study Identification.

Figure 1.

Open in a new tab

DMARD indicates disease-modifying antirheumatic drug; RCT, randomized clinical trial.

Network Structure and Geometry

The network is presented in Figure 2. The network comprises 27 mono-DMARD groups (methotrexate, peroral70,71,72,78,88,90,93,98,100,101,102,103,113,114,121,123,124,127,129,131,132,133,134,135,136,141,145,147,149,150,151; sulfasalazine64,77,80,87,91,99,104,105,107,109,117,121,123,127,128,139,146; leflunomide120,128,129,132,133,134,141,149; cyclosporine76,84,85,92,96,97,111,112,125,130,135; gold, injectable26,32,33,34,39,42,44,46,49,51,52,53,59,61,63,69,80,82,83,88,100,124,125,136; chloroquine/hydroxychloroquine23,25,27,28,41,66,79,87,95,107,108,110,111,115,116,118,119,121,143; D-penicillamine30,34,35,36,38,43,44,46,49,50,52,55,56,57,58,62,64,67,68,73,75,79,89,91,94,95,96,122; azathioprine36,41,48,67,68,78,90,97,98,102,112,114; gold, peroral41,51,52,53,56,60,63,69,73,74,75,81,83,86,93,101,104,106,116,126; cyclophosphamide29,31,37; levamisole40,42,45,47,49,50,54,58,67; tacrolimus137,140,142; Tripterygium wilfordii146,150; iguratimod144,147,151; timegadine57; bucillamine122; tiopronin55; pyritinol106; mycophenolate148; OM-898081,94; lobenzarit65; rheumacon126,131; prospidin113; aminopterin103; eazmov plus143; dapsone966,110; and methotrexate, injectable82,145), 1 GC group,22,24,119,138 and 1 placebo group.22,23,24,25,26,27,28,29,30,31,32,33,35,37,38,39,40,43,45,47,48,53,54,59,60,61,62,63,65,70,71,72,74,76,77,80,84,85,86,92,99,105,108,109,115,117,118,120,128,129,130,137,138,139,140,142,143,144,148 We identified 275 treatment groups in 121 two-group studies22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,42,43,44,45,46,47,48,50,51,54,55,56,57,58,59,60,61,62,64,65,66,68,69,70,71,72,73,74,75,76,77,78,79,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,122,123,124,125,126,127,130,131,132,133,134,135,136,137,138,139,140,141,142,144,145,146,147,148,149,150,151 and 11 three-group studies.41,49,52,53,63,67,80,121,128,129,143

Figure 2. Geometry of the Network.

Figure 2.

Open in a new tab

The size of the node reflects the number of patients included in the analysis for each drug, and the thickness of the lines between 2 nodes reflects the number of studies comparing the 2 treatments. The interventions included are Mt (methotrexate; 31 study groups, 2344 patients), Su (sulfasalazine; 18 study groups, 762 patients), Le (leflunomide; 8 study groups, 1321 patients), Cs (cyclosporine; 11 study groups, 492 patients), Au (injected gold; 25 study groups, 777 patients), Cl (chloroquine/hydroxychloroquine; 19 study groups, 662 patients), Dp (D-penicillamine; 28 study groups, 743 patients), Az (azathioprine; 12 study groups, 347 patients), Af (oral gold; 20 study groups, 1200 patients), Cp (cyclophosphamide; 3 study groups, 41 patients), Lm (levamisole; 10 study groups, 306 patients), Tc (tacrolimus; 3 study groups, 275 patients), Tw (Tripterygium wilfordii; 2 study groups, 113 patients), Ig (T-614, iguratimod; 3 study groups, 290 patients), Tg (timegadine; 1 study group, 19 patients), Bc (bucillamine; 1 study group, 24 patients), Tp (tiopronin; 1 study group, 25 patients), Py (pyritinol; 1 study group, 139 patients), Mc (mycophenolate mofetil; 2 study groups, 286 patients), Om (OM-8980; 2 study groups, 86 patients), Lz (lobenzarit; 1 study group, 86 patients), Rc (reumacon; 2 study groups, 80 patients), Pd (prospidin; 1 study group, 23 patients), Ap (aminopterin; 1 study group, 11 patients), Em (eazmov plus; 1 study group, 18 patients), Ds (dapsone; 2 study groups, 44 patients), Gc (glucocorticoid; 4 study groups, 153 patients), Mx (injected methotrexate; 2 study groups, 206 patients), and Pl (placebo; 60 study groups; 2900 patients).

Risk of Bias Within Studies

The assigned risk of bias appears in eTable 1 in Supplement 1. Of 130 RCTs22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151 in the network, 122 trials22,23,24,25,26,27,28,29,30,31,32,33,35,36,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,56,57,58,59,60,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,119,120,121,122,123,124,125,126,127,128,129,130,131,132,134,135,136,137,139,140,141,142,143,144,145,146,147,148,149,150,151 were assessed as at high risk of bias, 6 trials34,37,55,61,133,138 as of some concern, and 2 trials86,118 as at low risk of bias.

Choice of Prior

Based on models using different priors, we chose prior 0 to 5 for TJCi, TJC, SJC, DAS28i, and TJC28i; prior 0 to 6 for ESR; and prior 0 to 200 for CRP level. The density plot and sensitivity analysis with a 0 to 20 prior supported this decision.

Primary TJCi Outcome, Unadjusted Analysis

We imputed 14 missing TJC values (10.6% of all primary outcomes; eMethods 5 in Supplement 1) to obtain a complete data set of 132 studies22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151 with TJC outcomes (eResults 2 in Supplement 1). Differences in estimates vs methotrexate between fixed- and random-effects models were minimal. For instance, the reduction in TJCi outcome of methotrexate vs placebo was −4.87 joints (95% CrI, −5.74 to −4.00 joints) with the fixed-effects model and −4.90 joints (95% CrI, −6.12 to −3.72 joints) with the random-effects model (eTable 2 in Supplement 1). A random-effects NMA was selected based on deviation information criterion, with a between-study SD of 1.21 (95% CrI, 0.65 to 1.83). A sensitivity analysis of TJC without imputations showed results generally not different from TJCi. For instance, the reduction in TJC outcome of methotrexate vs placebo was −4.90 joints (95% CrI, −6.12 to −3.72 joints) with imputations (TJCi) and −4.83 joints (95% CrI, −6.07 to −3.62 joints) without imputations (TJC) (eTable 3 in Supplement 1).

Effect Modifier Analysis

See eResults 3 and eFigure 1 in Supplement 1. Apart from mean baseline DAS28 with imputed values (132 studies22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151), covariates did not improve model fit significantly (eTable 4 in Supplement 1).

Primary TJCi Outcome Adjusted for Baseline DAS28 With Peroral Methotrexate and Placebo as Reference

The primary TJCi outcome adjusted for baseline DAS28 was found among 132 trials22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151 (eResults 4 in Supplement 1). Outcomes with peroral methotrexate as the reference are shown in the Table and eFigure 2 in Supplement 1; eTable 5 in Supplement 1 shows results with placebo as the reference. The methotrexate vs placebo outcome for reduction in TJCi was 5.18 joints (95% CrI, 4.07 to 6.28 joints) (Table), marginally higher than the unadjusted outcome (eTable 3 in Supplement 1). Cyclophosphamide was favorable compared with methotrexate in terms of reduction in TJCi (6.08 joints; 95% CrI, 0.44 to 11.66 joints), while remaining drugs and glucocorticoid (−2.54 joints; 95% CrI, −5.16 to 0.08 joints) were associated with similar or lower tender joint counts (Table). No drugs were unfavorable compared with placebo (eTable 5 in Supplement 1). Leflunomide, cyclosporine, injected gold, tacrolimus, iguratimod, dapsone, glucocorticoid, and injectable methotrexate were favorable compared with placebo and did not differ substantially from peroral methotrexate. Sulfasalazine, chloroquine, d-penicillamine, azathioprine, oral gold, and levamisole were favorable compared with placebo but unfavorable compared with methotrexate. Tripterygium wilfordii, timegadine, bucillamine, tiopronin, pyritinol, mycophenolate mofetil, OM-8980, lobenzarit, reumacon, prospidin, aminopterin, and eazmov plus had numerically higher but overlapping outcomes compared with placebo. Comparisons of all interventions vs each other are presented in eTable 6 in Supplement 1.

Ranking

Rankograms for all treatments are presented in eFigure 3 in Supplement 1, arranged in the order assessed by visual inspection. The most effective treatments were cyclophosphamide, tacrolimus, and oral and injected methotrexate. Placebo was poorest but not worse than lobenzarit, mycophenolate mofetil, rheumacon, timegadine, or eazmov plus.

Exploration of Inconsistency

See eResults 5 in Supplement 1. We compared estimates in the NMA model with those from an unrelated mean effect sizes model. We did not find meaningful differences in any credible interval. The dev-dev plot is shown in eFigure 4 in Supplement 1. Model fit data are shown in eTable 7 in Supplement 1.

Sensitivity Analyses for Outcomes Adjusted for DAS28

DAS28-adjusted results of TJC, SJC, ESR, CRP, TJC28i, and DAS28i are shown in the Table (peroral methotrexate as reference) and eTable 5 in Supplement 1 (placebo as reference). For TJC (118 studies,23,26,28,29,30,34,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,66,67,68,69,70,71,72,73,74,75,76,77,79,80,81,82,83,84,85,87,88,89,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151 28 treatments), unlike in the primary analysis, mycophenolate mofetil was no longer unfavorable compared with methotrexate but still not different from placebo. GC was unfavorable compared with methotrexate but still favorable compared with placebo. For SJC (97 studies,22,26,29,30,31,32,33,37,39,40,41,44,45,46,50,51,52,54,57,58,60,62,63,65,67,68,69,70,71,72,73,74,75,76,77,78,80,81,82,83,84,86,87,88,92,93,94,96,97,98,100,101,102,103,105,107,108,109,110,111,113,114,115,116,118,119,120,122,123,124,125,127,128,129,130,131,132,133,134,135,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151 27 treatments), unlike in the primary analysis, Tripterygium was favorable compared with placebo and GC was not different from placebo. For ESR (116 studies,22,23,24,25,26,27,28,29,30,32,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,80,81,82,84,85,87,88,89,90,91,92,93,94,96,97,98,99,100,101,104,105,106,107,108,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,127,128,129,135,136,137,140,141,142,143,144,146,147,148,149,150,151 28 treatments), in contrast to the primary analysis, cyclosporine and azathioprine were not different compared with placebo. Furthermore, Tripterygium, timegadine, tiopronin, pyritinol, and prospidin were favorable compared with placebo and not different compared with methotrexate. For CRP (47 studies,40,45,56,58,64,66,67,79,85,87,95,97,98,100,104,110,111,113,114,115,116,117,122,124,125,126,127,128,129,131,132,133,134,136,137,138,140,141,142,144,146,147,148,149,150,151 20 treatments), chloroquine, D-penicillamine, azathioprine, oral gold, dapsone, and GC were not different compared with placebo, whereas prospidin was favorable compared with placebo and not different compared with methotrexate. For DAS28i, injectable methotrexate was favorable compared with oral methotrexate. Mycophenolate mofetil, OM-8980, lobenzarit, reumacon, and eazmov plus were not different from placebo, whereas all other drugs were favorable compared with placebo and not different from or unfavorable compared with methotrexate. For TJC28i,unlike with TJCi, cyclophosphamide, levamisole, and mycophenolate mofetil were not different from methotrexate.

Additional Analyses

Outcomes performed with a model using prior distribution 0 to 20 are shown in eTable 8 in Supplement 1. A sensitivity analysis separating chloroquine and hydroxychloroquine in 2 nodes showed similar outcomes in the 2 groups (eResults 6 in Supplement 1).

Discussion

Results from this efficacy meta-analysis support existing guidelines.5,160 Study information on reasons for dropping out (eg, toxic effects, lack of effect, violation of study criteria, and no appearance) varied too much to be standardized. Dropout rates, which reflect tolerability and toxic effects, were instead reported.

Summary of Main Results

GC and well-defined csDMARDs (methotrexate, sulfasalazine, leflunomide, cyclosporine, gold, chloroquine, D-penicillamine, azathioprine, levamisole, and cyclophosphamide) were all favorable compared with placebo, but sulfasalazine, chloroquine, D-penicillamine, azathioprine, peroral gold, levamisole, and GC were unfavorable compared with methotrexate. Outcomes of chloroquine and hydroxychloroquine were similar. Cyclophosphamide was favorable compared with methotrexate. In addition, noncurrent practice drugs tacrolimus, iguratimod, and dapsone had outcomes comparable with those of established csDMARDs.

Overall Completeness and Applicability

We avoided exclusions of questionable csDMARDs because such procedures may exclude relevant comparators. Cyclophosphamide can be used to treat rheumatoid vasculitis and pulmonary fibrosis, and under such conditions an associated improvement in arthritis is valuable. However, the risk of leukemia increases when the accumulated dose of cyclophosphamide exceeds 36 g.161 Thus, the general applicability of cyclophosphamide is limited. Historically, the outcome of gold treatment is interesting, although gold is no longer a current practice drug. Tacrolimus had a high dropout rate and is as expensive as tDMARDs but may be suitable for patients with RA who have had organ transplants. In patients with limited insurance coverage162 and those from resource-limited regions with limited access to expensive tDMARDs, csDMARDs are important, especially as a combination of 2 to 3 inexpensive csDMARDs may be as effective as tDMARDs.16,17,163,164,165 However, even in such patients, gold, D-penicillamine, azathioprine, and levamisole may have limited, if any, applicability given high dropout rates of 33% to 39% identified in our study.

Quality of Evidence

Generally, evidence from RCTs is of high quality. Concerning placebo and most previously established treatments (methotrexate, sulfasalazine, leflunomide, ciclosporin, gold, D-penicillamine, azathioprine, chloroquine, and GC), the number of studies and participants was adequate for a robust conclusion. However, outcomes for remaining drugs were based on 1 to 3 studies and therefore less reliable. For instance, the favorable evidence for cyclophosphamide was weak given that 41 participants were included.29,31,37 Some drugs may deserve consideration; for instance, iguratimod, which had a dropout rate of 12.4%, may be the most promising of these agents. Models using different priors gave similar results in accordance with robust outcomes.

Agreement and Disagreements With Other Reviews

Previously, in a conventional meta-analysis16 of 33 mono-csDMARD RCTs, we found that injectable gold, sulfasalazine, methotrexate, leflunomide, and cyclosporine were associated with similar radiographic outcomes and that D-penicillamine, chloroquine, oral gold, azathioprine, and cyclophosphamide were associated with inferior outcomes compared with methotrexate but superior outcomes compared with placebo. Over the last 20 years, several conventional meta-analyses have investigated outcomes associated with individual csDMARDs.166,167,168,169,170,171,172,173,174,175 All investigated drugs were superior to placebo, but in these analyses, drugs were not compared vs each other.

Limitations

This study has several limitations. Clinical heterogeneity across studies may have biased outcomes. However, several univariate sensitivity analyses disclosed only 1 factor associated with the outcome (baseline DAS28), and by adjusting the outcome for DAS28, we tried to limit this bias.

The lack of official reporting guidelines176 and protocol registers177 when most of these studies were performed may explain insufficient reporting of scientific procedures in the studies. This may have contributed to a high, potentially exaggerated risk of bias assessment and to variability in presentation of outcomes, which made imputations necessary in 14 primary outcomes (10.6%).22,24,25,27,31,32,33,35,65,78,86,90,91,107 However, the TJC full analysis with imputed values showed similar results as the TJC analysis based on measured values.

A general problem with NMA is that many comparisons are based on an assumption of transitivity, meaning that if an association between a first and second drug holds and it also holds between the second and third drug, it is assumed to necessarily hold between the first and third drug. We tested this by the unrelated mean effect sizes model, which did not reveal important inconsistency. That is, it did not disclose differences between direct and indirect comparisons, suggesting that the assumption of transitivity was acceptable.

Undoubtedly, potential drugs have been tested in unpublished studies,178 and we did not include antibiotics, such as tetracycline, associated with potential csDMARD outcomes.179 Despite these reservations, to our knowledge this analysis is the largest integrated mono-csDMARD analysis hitherto presented, and we believe that the risk of having overlooked important studies is small. Sensitivity analyses produced fewer significant differences, but numerically they were generally not in opposition to the full analysis.

Conclusions

This meta-analysis adds a comprehensive analysis comparing outcomes of multiple csDMARDs to each other in 1 integrated analysis. Currently established csDMARDs (methotrexate, leflunomide, cyclosporine, sulfasalazine, and [hydroxy]chloroquine) were found to be associated with disease-modifying outcomes in RA. The significant outcomes of former but now abandoned drugs, such as cyclophosphamide, gold, azathioprine, D-penicillamine, and levamisole, were not sufficient to justify a reintroduction considering high dropout rates. Some unapproved drugs showed outcomes potentially justifying further investigation. Outcomes confirmed the role of methotrexate as primary reference csDMARD.

Supplement 1.

eMethods 1. Eligibility Criteria

eMethods 2. Information Sources and Search

eMethods 3. Study Selection and Data Collection

eMethods 4. Data Items

eMethods 5. Estimations of Missing Values and Conversion Factors

eMethods 6. Risk of Bias Within Individual Studies

eMethods 7. Planned Method of Analysis

eMethods 8. Effect Modifier Analysis

eResults 1. Study Selection

eResults 2. Primary Outcome (Complete Analysis) With Methotrexate as Reference, Unadjusted Analysis

eResults 3. Effect Modifier Analyses

eResults 4. Primary Outcome With Methotrexate as Reference, Adjusted for Baseline Disease Activity Score 28 (DAS28)

eResults 5. Exploration of Inconsistency

eResults 6. Results of Additional Analyses

eReferences.

eTable 1. Baseline Characteristics, Primary Outcome and Bias Score of Included Studies

eTable 2. Unadjusted Tender Joint Count Imputed (TJCi) Analysis Using Fixed- and Random-Effects Models

eTable 3. Tender Joint Count Imputed (TJCi) Sensitivity Analysis Comparing Analyses With and Without Imputed Values

eTable 4. Model Fit Parameters for Tender Joint Count Imputed (TJCi) Analyses Including and Excluding Effect Modifiers in Metaregression Analyses

eTable 5. Primary Outcome and Sensitivity Outcomes With Placebo as Reference

eTable 6. Comparisons of Tender Joint Count Imputed (TJCi) Outcome of All Interventions vs All Interventions

eTable 7. Model Fit Data for Primary Outcome and Sensitivity Outcomes

eTable 8. Primary and Sensitivity Outcomes Calculated With Prior 0-20

eFigure 1. Subgroup Analyses

eFigure 2. Forest Plot of Primary Tender Joint Count Imputed (TJCi) Outcomes vs Peroral Methotrexate

eFigure 3. Rankograms of 29 Tested Interventions

eFigure 4. dev-dev Plot of Unrelated Mean Effects Mean vs Network Meta-Analyses Mean

Click here for additional data file. (1.2MB, pdf)
Supplement 2.

Data Sharing Statement

Click here for additional data file. (140.2KB, pdf)

References

  • 1.Forestier J. The treatment of rheumatoid arthritis with gold salts injections. Lancet. 1932;219(5661):441-444. doi: 10.1016/S0140-6736(01)24417-1 [DOI] [Google Scholar]
  • 2.Hench PS, Kendall EC, Slocumb CH, Polley HF. The effect of a hormone of the adrenal cortex (17-hydroxy-11-dehydrocorticosterone; compound E) and of pituitary adrenocorticotropic hormone on rheumatoid arthritis. Proc Staff Meet Mayo Clin. 1949;24(8):181-197. [PubMed] [Google Scholar]
  • 3.Case JP. Old and new drugs used in rheumatoid arthritis: a historical perspective: part 1: the older drugs. Am J Ther. 2001;8(2):123-143. doi: 10.1097/00045391-200103000-00007 [DOI] [PubMed] [Google Scholar]
  • 4.Mottram PL. Past, present and future drug treatment for rheumatoid arthritis and systemic lupus erythematosus. Immunol Cell Biol. 2003;81(5):350-353. doi: 10.1046/j.1440-1711.2003.01184.x [DOI] [PubMed] [Google Scholar]
  • 5.Fraenkel L, Bathon JM, England BR, et al. 2021 American College of Rheumatology guideline for the treatment of rheumatoid arthritis. Arthritis Care Res (Hoboken). 2021;73(7):924-939. doi: 10.1002/acr.24596 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Graudal N, Tarp U, Jurik AG, et al. Inflammatory patterns in rheumatoid arthritis estimated by the number of swollen and tender joints, the erythrocyte sedimentation rate, and hemoglobin: longterm course and association to radiographic progression. J Rheumatol. 2000;27(1):47-57. [PubMed] [Google Scholar]
  • 7.Graudal N. The natural history and prognosis of rheumatoid arthritis: association of radiographic outcome with process variables, joint motion and immune proteins. Scand J Rheumatol Suppl. 2004;118:1-38. doi: 10.1080/03009740310004847 [DOI] [PubMed] [Google Scholar]
  • 8.Boers M, van Riel PL, Felson DT, Tugwell P. Assessing the activity of rheumatoid arthritis. Baillieres Clin Rheumatol. 1995;9(2):305-317. doi: 10.1016/S0950-3579(05)80192-X [DOI] [PubMed] [Google Scholar]
  • 9.Graudal NA, Jurik AG, de Carvalho A, Graudal HK. Radiographic progression in rheumatoid arthritis: a long-term prospective study of 109 patients. Arthritis Rheum. 1998;41(8):1470-1480. doi: [DOI] [PubMed] [Google Scholar]
  • 10.van Gestel AM, Haagsma CJ, van Riel PL. Validation of rheumatoid arthritis improvement criteria that include simplified joint counts. Arthritis Rheum. 1998;41(10):1845-1850. doi: [DOI] [PubMed] [Google Scholar]
  • 11.Felson DT, Anderson JJ, Boers M, et al. ; The Committee on Outcome Measures in Rheumatoid Arthritis Clinical Trials . The American College of Rheumatology preliminary core set of disease activity measures for rheumatoid arthritis clinical trials. Arthritis Rheum. 1993;36(6):729-740. doi: 10.1002/art.1780360601 [DOI] [PubMed] [Google Scholar]
  • 12.Dougados M, Aletaha D, van Riel P. Disease activity measures for rheumatoid arthritis. Clin Exp Rheumatol. 2007;25(5)(suppl 46):S22-S29. [PubMed] [Google Scholar]
  • 13.Prevoo ML, van ’t Hof MA, Kuper HH, van Leeuwen MA, van de Putte LB, van Riel PL. Modified disease activity scores that include twenty-eight-joint counts: development and validation in a prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum. 1995;38(1):44-48. doi: 10.1002/art.1780380107 [DOI] [PubMed] [Google Scholar]
  • 14.Gigliucci G, Massafra U, Frediani B, et al. A review of network meta-analysis comparing biologics in the treatment of rheumatoid arthritis. Eur Rev Med Pharmacol Sci. 2020;24(4):1624-1644. doi: 10.26355/eurrev_202002_20337 [DOI] [PubMed] [Google Scholar]
  • 15.DANBIO . National Annual Report of the Danish Rheumatology Register 2019. Landsdækkende klinisk kvalitetsdatabase for behandling af reumatologiske patienter National Årsrapport 2019. Accessed August 29, 2023. https://www.sundhed.dk/content/cms/91/4691_danbio-aarsrapport-2019_endeligversion.pdf
  • 16.Graudal N, Jürgens G. Similar effects of disease-modifying antirheumatic drugs, glucocorticoids, and biologic agents on radiographic progression in rheumatoid arthritis: meta-analysis of 70 randomized placebo-controlled or drug-controlled studies, including 112 comparisons. Arthritis Rheum. 2010;62(10):2852-2863. doi: 10.1002/art.27592 [DOI] [PubMed] [Google Scholar]
  • 17.Ma MH, Kingsley GH, Scott DL. A systematic comparison of combination DMARD therapy and tumour necrosis inhibitor therapy with methotrexate in patients with early rheumatoid arthritis. Rheumatology (Oxford). 2010;49(1):91-98. doi: 10.1093/rheumatology/kep331 [DOI] [PubMed] [Google Scholar]
  • 18.Fuchs HA, Pincus T. Reduced joint counts in controlled clinical trials in rheumatoid arthritis. Arthritis Rheum. 1994;37(4):470-475. doi: 10.1002/art.1780370406 [DOI] [PubMed] [Google Scholar]
  • 19.Leil TA, Lu Y, Bouillon-Pichault M, Wong R, Nowak M. Model-based meta-analysis compares DAS28 rheumatoid arthritis treatment effects and suggests an expedited trial design for early clinical development. Clin Pharmacol Ther. 2021;109(2):517-527. doi: 10.1002/cpt.2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.R Core Team . The R Project for Statistical Computing. Accessed August 29, 2923. https://www.r-project.org/
  • 21.Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898. doi: 10.1136/bmj.l4898 [DOI] [PubMed] [Google Scholar]
  • 22.Empire Rheumatism Council . Multi-centre controlled trial comparing cortisone acetate and acetyl salicylic acid in the long-term treatment of rheumatoid arthritis; results up to one year. Ann Rheum Dis. 1955;14(4):353-370. doi: 10.1136/ard.14.4.353 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Freedman A. Chloroquine and rheumatoid arthritis; a short-term controlled trial. Ann Rheum Dis. 1956;15(3):251-257. doi: 10.1136/ard.15.3.251 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.A comparison of prednisolone with aspirin on other analgesics in the treatment of rheumatold rthritis. Ann Rheum Dis. 1959;18(3):173-188. doi: 10.1136/ard.18.3.173 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Freedman A, Steinberg VL. Chloroquine in rheumatoid arthritis; a double blindfold trial of treatment for one year. Ann Rheum Dis. 1960;19(3):243-250. doi: 10.1136/ard.19.3.243 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.The Research Sub-Committee of the Empire Rheumatism Council . Gold therapy in rheumatoid arthritis: report of a multicentre control trial. Ann Rheum Dis. 1960;19(2):95-119. doi: 10.1136/ard.19.2.95 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Hamilton EB, Scott JT. Hydroxychloroquine sulfate (“plaguenil”) in treatment of rheumatoid arthritis. Arthritis Rheum. 1962;5:502-512. doi: 10.1002/art.1780050507 [DOI] [PubMed] [Google Scholar]
  • 28.Mainland D, Sutcliffe MI. Hydroxychloroquine sulfate in rheumatoid arthritis: a six month double-blind trial. Bull Rheum Dis. 1962;13:287-290. [PubMed] [Google Scholar]
  • 29.Cooperating Clinics Committee of the American Rheumatism Association . A controlled trial of cyclophosphamide in rheumatoid arthritis. N Engl J Med. 1970;283(17):883-889. doi: 10.1056/NEJM197010222831701 [DOI] [PubMed] [Google Scholar]
  • 30.Multicentre Trial Group . Controlled trial of D(-)penicillamine in severe rheumatoid arthritis. Lancet. 1973;1(7798):275-280. [PubMed] [Google Scholar]
  • 31.Lidsky MD, Sharp JT, Billings S. Double-blind study of cyclophosphamide in rheumatoid arthritis. Arthritis Rheum. 1973;16(2):148-153. doi: 10.1002/art.1780160203 [DOI] [PubMed] [Google Scholar]
  • 32.Cooperating Clinics Committee of the American Rheumatism Association . A controlled trial of gold salt therapy in rheumatoid arthritis. Arthritis Rheum. 1973;16(3):353-358. doi: 10.1002/art.1780160309 [DOI] [PubMed] [Google Scholar]
  • 33.Sigler JW, Bluhm GB, Duncan H, Sharp JT, Ensign DC, McCrum WR. Gold salts in the treatment of rheumatoid arthritis: a double-blind study. Ann Intern Med. 1974;80(1):21-26. doi: 10.7326/0003-4819-80-1-21 [DOI] [PubMed] [Google Scholar]
  • 34.Huskisson EC, Gibson TJ, Balme HW, et al. Trial comparing D-penicillamine and gold in rheumatoid arthritis: preliminary report. Ann Rheum Dis. 1974;33(6):532-535. doi: 10.1136/ard.33.6.532 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Dixon AJ, Davies J, Dormandy TL, et al. Synthetic D(-)penicillamine in rheumatoid arthritis: double-blind controlled study of a high and low dosage regimen. Ann Rheum Dis. 1975;34(5):416-421. doi: 10.1136/ard.34.5.416 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Berry H, Liyanage SP, Durance RA, Barnes CG, Berger LA, Evans S. Azathioprine and penicillamine in treatment of rheumatoid arthritis: a controlled trial. Br Med J. 1976;1(6017):1052-1054. doi: 10.1136/bmj.1.6017.1052 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Townes AS, Sowa JM, Shulman LE. Controlled trial of cyclophosphamide in rheumatoid arthritis. Arthritis Rheum. 1976;19(3):563-573. doi: 10.1002/art.1780190308 [DOI] [PubMed] [Google Scholar]
  • 38.Mery C, Delrieu F, Ghozlan R, et al. Controlled trial of D-penicillamine in rheumatoid arthritis: dose effect and the role of zinc. Scand J Rheumatol. 1976;5(4):241-247. doi: 10.3109/03009747609099913 [DOI] [PubMed] [Google Scholar]
  • 39.Walravens M, Dequeker J. Comparison of gold and orgotein treatment in rheumatoid arthritis. Curr Ther Res Clin Exp. 1976;20(1):62-69. [PubMed] [Google Scholar]
  • 40.Runge LA, Pinals RS, Lourie SH, Tomar RH. Treatment of rheumatoid arthritis with levamisole: a controlled trial. Arthritis Rheum. 1977;20(8):1445-1448. doi: 10.1002/art.1780200801 [DOI] [PubMed] [Google Scholar]
  • 41.Dwosh IL, Stein HB, Urowitz MB, Smythe HA, Hunter T, Ogryzlo MA. Azathioprine in early rheumatoid arthritis: comparison with gold and chloroquine. Arthritis Rheum. 1977;20(2):685-692. doi: 10.1002/art.1780200208 [DOI] [PubMed] [Google Scholar]
  • 42.El-Ghobarey AF, Mavrikakis ME, Macleod M, et al. Clinical and laboratory studies of levamisole in patients with rheumatoid arthritis. Q J Med. 1978;47(187):385-400. [PubMed] [Google Scholar]
  • 43.Berry H, Fernandes L, Ford-Hutchinson AW, Evans SJ, Hamilton EB. Alclofenac and D-penicillamine: comparative trial in rheumatoid arthritis. Ann Rheum Dis. 1978;37(1):93-97. doi: 10.1136/ard.37.1.93 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Mäkisara P, Nissilä M, Kajander A, et al. Comparison of penicillamine and gold treatment in early rheumatoid arthritis. Scand J Rheumatol. 1978;7(3):166-170. doi: 10.3109/03009747809095648 [DOI] [PubMed] [Google Scholar]
  • 45.Multicentre study group . Levamisole in rheumatoid arthritis: a randomised double-blind study comparing two dosage regimens of levamisole with placebo: multicentre study group. Lancet. 1978;2(8098):1007-1012. [PubMed] [Google Scholar]
  • 46.Haataja M, Nissilä M, Ruutsalo HM. Serum sulfhydryl levels in rheumatoid patients treated with gold thiomalate and penicillamine. Scand J Rheumatol. 1978;7(4):212-214. doi: 10.3109/03009747809095657 [DOI] [PubMed] [Google Scholar]
  • 47.Miller B, de Merieux P, Srinivasan R, et al. Double-blind placebo controlled crossover evaluation of levamisole in rheumatoid arthritis. Arthritis Rheum. 1980;23(2):172-182. doi: 10.1002/art.1780230207 [DOI] [PubMed] [Google Scholar]
  • 48.Woodland J, Chaput de Saintonge DM, Evans SJ, Sharman VL, Currey HL. Azathioprine in rheumatoid arthritis: double-blind study of full versus half doses versus placebo. Ann Rheum Dis. 1981;40(4):355-359. doi: 10.1136/ard.40.4.355 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Capell HA, Hunter JA, Rennie JA, Murdoch RM. Levamisole—a possible alternative to gold and penicillamine in the longterm treatment of rheumatoid arthritis? J Rheumatol. 1981;8(5):730-740. [PubMed] [Google Scholar]
  • 50.Mowat AG, Mowat AM. Levamisole in rheumatoid arthritis: a comparison with D-penicillamine. J Rheumatol. 1981;8(4):575-580. [PubMed] [Google Scholar]
  • 51.Smith PR, Brown GM, Meyers OL. An open comparative study of auranofin vs. gold sodium thiomalate. J Rheumatol Suppl. 1982;8:190-196. [PubMed] [Google Scholar]
  • 52.Barraclough D, Brook A, Brooks P, Boyden K, Thomas D, Tymms K. A comparative study of auranofin, gold sodium thiomalate, and D-penicillamine in rheumatoid arthritis: a progress report. J Rheumatol Suppl. 1982;8:197-200. [PubMed] [Google Scholar]
  • 53.Lewis D, Capell HA. Is auranofin preferable to gold sodium thiomalate in the management of rheumatoid arthritis? In: Capell HA, Cole DS, Manghani KK, et al. Auranofin, Proceedings of a Smith Kline & French International Symposium. Excerpta Medica; 1983:147-155. [Google Scholar]
  • 54.Multicentre Study Group . Levamisole in rheumatoid arthritis: final report on a randomised double-blind study comparing a single weekly dose of levamisole with placebo. Ann Rheum Dis. 1982;41(2):159-163. doi: 10.1136/ard.41.2.159 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Pasero G, Pellegrini P, Ambanelli U, et al. Controlled multicenter trial of tiopronin and d-penicillamine for rheumatoid arthritis. Arthritis Rheum. 1982;25(8):923-929. doi: 10.1002/art.1780250803 [DOI] [PubMed] [Google Scholar]
  • 56.Felix-Davies DD, Stewart AM, Wilkinson BR, Bateman JR, Delamere JP. A 12-month comparative trial of auranofin and D-penicillamine in rheumatoid arthritis. Am J Med. 1983;75(6A):138-141. doi: 10.1016/0002-9343(83)90487-4 [DOI] [PubMed] [Google Scholar]
  • 57.Mbuyi-Muamba JM, Dequeker J. A comparative trial of timegadine and D-penicillamine in rheumatoid arthritis. Clin Rheumatol. 1983;2(4):369-374. doi: 10.1007/BF02041557 [DOI] [PubMed] [Google Scholar]
  • 58.Clemens LE, Ansell BM, Bernstein RM, Hall MA, McCauley D. A comparison of low dose levamisole and penicillamine in rheumatoid arthritis. Aust N Z J Med. 1983;13(6):578-582. doi: 10.1111/j.1445-5994.1983.tb02607.x [DOI] [PubMed] [Google Scholar]
  • 59.Kadir N, Grayson MF, Campbell ED. A double-blind one year clinical trial comparing alclofenac with gold in rheumatoid arthritis. Ann Clin Res. 1983;15(1):4-9. [PubMed] [Google Scholar]
  • 60.Wenger ME, Alexander S, Bland JH, Blechman WJ. Auranofin versus placebo in the treatment of rheumatoid arthritis. Am J Med. 1983;75(6A):123-127. doi: 10.1016/0002-9343(83)90484-9 [DOI] [PubMed] [Google Scholar]
  • 61.Lund B, Graae A, Sonne I, Andersen RB. Long-term treatment of rheumatoid arthritis with fenclofenac or gold: a controlled, single blind trial comparing clinical and biochemical parameters. Eur J Rheumatol Inflamm. 1983;6(3):279-287. [PubMed] [Google Scholar]
  • 62.Williams HJ, Ward JR, Reading JC, et al. Low-dose D-penicillamine therapy in rheumatoid arthritis: a controlled, double-blind clinical trial. Arthritis Rheum. 1983;26(5):581-592. doi: 10.1002/art.1780260502 [DOI] [PubMed] [Google Scholar]
  • 63.Ward JR, Williams HJ, Boyce E, Egger MJ, Reading JC, Samuelson COJ Jr. Comparison of auranofin, gold sodium thiomalate, and placebo in the treatment of rheumatoid arthritis: subsets of responses. Am J Med. 1983;75(6A):133-137. doi: 10.1016/0002-9343(83)90486-2 [DOI] [PubMed] [Google Scholar]
  • 64.Neumann VC, Grindulis KA, Hubball S, McConkey B, Wright V. Comparison between penicillamine and sulphasalazine in rheumatoid arthritis: Leeds-Birmingham trial. Br Med J (Clin Res Ed). 1983;287(6399):1099-1102. doi: 10.1136/bmj.287.6399.1099 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Shiokawa Y, Horiuchi Y, Mizushima Y, et al. A multicenter double-blind controlled study of lobenzarit, a novel immunomodulator, in rheumatoid arthritis. J Rheumatol. 1984;11(5):615-623. [PubMed] [Google Scholar]
  • 66.Fowler PD, Shadforth MF, Crook PR, Lawton A. Report on chloroquine and dapsone in the treatment of rheumatoid arthritis: a 6-month comparative study. Ann Rheum Dis. 1984;43(2):200-204. doi: 10.1136/ard.43.2.200 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67.Halberg P, Bentzon MW, Crohn O, et al. Double-blind trial of levamisole, penicillamine and azathioprine in rheumatoid arthritis: clinical, biochemical, radiological and scintigraphic studies. Dan Med Bull. 1984;31(5):403-409. [PubMed] [Google Scholar]
  • 68.Paulus HE, Williams HJ, Ward JR, et al. Azathioprine versus D-penicillamine in rheumatoid arthritis patients who have been treated unsuccessfully with gold. Arthritis Rheum. 1984;27(7):721-727. doi: 10.1002/art.1780270701 [DOI] [PubMed] [Google Scholar]
  • 69.Davis P, Menard H, Thompson J, Harth M, Beaudet F. One-year comparative study of gold sodium thiomalate and auranofin in the treatment of rheumatoid arthritis. J Rheumatol. 1985;12(1):60-67. [PubMed] [Google Scholar]
  • 70.Williams HJ, Willkens RF, Samuelson COJ Jr, et al. Comparison of low-dose oral pulse methotrexate and placebo in the treatment of rheumatoid arthritis: a controlled clinical trial. Arthritis Rheum. 1985;28(7):721-730. doi: 10.1002/art.1780280702 [DOI] [PubMed] [Google Scholar]
  • 71.Weinblatt ME, Coblyn JS, Fox DA, et al. Efficacy of low-dose methotrexate in rheumatoid arthritis. N Engl J Med. 1985;312(13):818-822. doi: 10.1056/NEJM198503283121303 [DOI] [PubMed] [Google Scholar]
  • 72.Andersen PA, West SG, O’Dell JR, Via CS, Claypool RG, Kotzin BL. Weekly pulse methotrexate in rheumatoid arthritis: clinical and immunologic effects in a randomized, double-blind study. Ann Intern Med. 1985;103(4):489-496. doi: 10.7326/0003-4819-103-4-489 [DOI] [PubMed] [Google Scholar]
  • 73.Hochberg MC. Auranofin or D-penicillamine in the treatment of rheumatoid arthritis. Ann Intern Med. 1986;105(4):528-535. doi: 10.7326/0003-4819-105-4-528 [DOI] [PubMed] [Google Scholar]
  • 74.Bombardier C, Ware J, Russell IJ, Larson M, Chalmers A, Read JL. Auranofin therapy and quality of life in patients with rheumatoid arthritis: results of a multicenter trial. Am J Med. 1986;81(4):565-578. doi: 10.1016/0002-9343(86)90539-5 [DOI] [PubMed] [Google Scholar]
  • 75.Manthorpe R, Hørbov S, Sylvest J, Vinterberg H. Auranofin versus penicillamine in rheumatoid arthritis: one-year results from a prospective clinical investigation. Scand J Rheumatol. 1986;15(1):13-22. doi: 10.3109/03009748609092663 [DOI] [PubMed] [Google Scholar]
  • 76.van Rijthoven AW, Dijkmans BA, Goei The HS, et al. Cyclosporin treatment for rheumatoid arthritis: a placebo controlled, double blind, multicentre study. Ann Rheum Dis. 1986;45(9):726-731. doi: 10.1136/ard.45.9.726 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 77.Pinals RS, Kaplan SB, Lawson JG, Hepburn B. Sulfasalazine in rheumatoid arthritis: a double-blind, placebo-controlled trial. Arthritis Rheum. 1986;29(12):1427-1434. doi: 10.1002/art.1780291202 [DOI] [PubMed] [Google Scholar]
  • 78.Hamdy H, McKendry RJ, Mierins E, Liver JA. Low-dose methotrexate compared with azathioprine in the treatment of rheumatoid arthritis: a twenty-four-week controlled clinical trial. Arthritis Rheum. 1987;30(4):361-368. doi: 10.1002/art.1780300401 [DOI] [PubMed] [Google Scholar]
  • 79.Gibson T, Emery P, Armstrong RD, Crisp AJ, Panayi GS. Combined D-penicillamine and chloroquine treatment of rheumatoid arthritis—a comparative study. Br J Rheumatol. 1987;26(4):279-284. doi: 10.1093/rheumatology/26.4.279 [DOI] [PubMed] [Google Scholar]
  • 80.Williams HJ, Ward JR, Dahl SL, et al. A controlled trial comparing sulfasalazine, gold sodium thiomalate, and placebo in rheumatoid arthritis. Arthritis Rheum. 1988;31(6):702-713. doi: 10.1002/art.1780310602 [DOI] [PubMed] [Google Scholar]
  • 81.Vischer TL. A double blind multicentre study of OM-8980 and auranofin in rheumatoid arthritis. Ann Rheum Dis. 1988;47(7):582-587. doi: 10.1136/ard.47.7.582 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 82.Suarez-Almazor ME, Fitzgerald A, Grace M, Russell AS. A randomized controlled trial of parenteral methotrexate compared with sodium aurothiomalate (Myochrysine) in the treatment of rheumatoid arthritis. J Rheumatol. 1988;15(5):753-756. [PubMed] [Google Scholar]
  • 83.Schattenkirchner M, Bröll H, Kaik B, Müller-Fassbender H, Rau R, Zeidler H. Auranofin and gold sodium thiomalate in the treatment of rheumatoid arthritis: a one-year, double-blind, comparative multicenter study. Klin Wochenschr. 1988;66(4):167-174. doi: 10.1007/BF01727786 [DOI] [PubMed] [Google Scholar]
  • 84.Yocum DE, Klippel JH, Wilder RL, et al. Cyclosporin A in severe, treatment-refractory rheumatoid arthritis: a randomized study. Ann Intern Med. 1988;109(11):863-869. doi: 10.7326/0003-4819-109-11-863 [DOI] [PubMed] [Google Scholar]
  • 85.Dougados M, Awada H, Amor B. Cyclosporin in rheumatoid arthritis: a double blind, placebo controlled study in 52 patients. Ann Rheum Dis. 1988;47(2):127-133. doi: 10.1136/ard.47.2.127 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 86.Borg G, Allander E, Lund B, et al. Auranofin improves outcome in early rheumatoid arthritis: results from a 2-year, double blind placebo controlled study. J Rheumatol. 1988;15(12):1747-1754. [PubMed] [Google Scholar]
  • 87.Nuver-Zwart IH, van Riel PL, van de Putte LB, Gribnau FW. A double blind comparative study of sulphasalazine and hydroxychloroquine in rheumatoid arthritis: evidence of an earlier effect of sulphasalazine. Ann Rheum Dis. 1989;48(5):389-395. doi: 10.1136/ard.48.5.389 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 88.Morassut P, Goldstein R, Cyr M, Karsh J, McKendry RJ. Gold sodium thiomalate compared to low dose methotrexate in the treatment of rheumatoid arthritis—a randomized, double blind 26-week trial. J Rheumatol. 1989;16(3):302-306. [PubMed] [Google Scholar]
  • 89.Carroll GJ, Will RK, Breidahl PD, Tinsley LM. Sulphasalazine versus penicillamine in the treatment of rheumatoid arthritis. Rheumatol Int. 1989;8(6):251-255. doi: 10.1007/BF00270980 [DOI] [PubMed] [Google Scholar]
  • 90.Arnold MH, O’Callaghan J, McCredie M, Beller EM, Kelly DE, Brooks PM. Comparative controlled trial of low-dose weekly methotrexate versus azathioprine in rheumatoid arthritis: 3-year prospective study. Br J Rheumatol. 1990;29(2):120-125. doi: 10.1093/rheumatology/29.2.120 [DOI] [PubMed] [Google Scholar]
  • 91.Capell HA, Marabani M, Madhok R, Torley H, Hunter JA. Degree and extent of response to sulphasalazine or penicillamine therapy for rheumatoid arthritis: results from a routine clinical environment over a two-year period. Q J Med. 1990;75(276):335-344. [PubMed] [Google Scholar]
  • 92.Tugwell P, Bombardier C, Gent M, et al. Low-dose cyclosporin versus placebo in patients with rheumatoid arthritis. Lancet. 1990;335(8697):1051-1055. doi: 10.1016/0140-6736(90)92630-Z [DOI] [PubMed] [Google Scholar]
  • 93.Weinblatt ME, Kaplan H, Germain BF, et al. Low-dose methotrexate compared with auranofin in adult rheumatoid arthritis: a thirty-six-week, double-blind trial. Arthritis Rheum. 1990;33(3):330-338. doi: 10.1002/art.1780330305 [DOI] [PubMed] [Google Scholar]
  • 94.Verstraeten A, Sileghem A, Dequeker J. OM-8980 and D-penicillamine in the treatment of rheumatoid arthritis: a 12-month double-blind randomized study. Scand J Rheumatol. 1990;19(6):422-431. doi: 10.3109/03009749009097631 [DOI] [PubMed] [Google Scholar]
  • 95.Scott DL, Greenwood A, Davies J, Maddison PJ, Maddison MC, Hall ND. Radiological progression in rheumatoid arthritis: do D-penicillamine and hydroxychloroquine have different effects? Br J Rheumatol. 1990;29(2):126-127. doi: 10.1093/rheumatology/29.2.126 [DOI] [PubMed] [Google Scholar]
  • 96.van Rijthoven AW, Dijkmans BA, Thè HS, et al. Comparison of cyclosporine and D-penicillamine for rheumatoid arthritis: a randomized, double blind, multicenter study. J Rheumatol. 1991;18(6):815-820. [PubMed] [Google Scholar]
  • 97.Ahern MJ, Harrison W, Hollingsworth P, Bradley J, Laing B, Bayliss C. A randomised double-blind trial of cyclosporin and azathioprine in refractory rheumatoid arthritis. Aust N Z J Med. 1991;21(6):844-849. doi: 10.1111/j.1445-5994.1991.tb01405.x [DOI] [PubMed] [Google Scholar]
  • 98.Jeurissen ME, Boerbooms AM, van de Putte LB, et al. Methotrexate versus azathioprine in the treatment of rheumatoid arthritis: a forty-eight-week randomized, double-blind trial. Arthritis Rheum. 1991;34(8):961-972. doi: 10.1002/art.1780340805 [DOI] [PubMed] [Google Scholar]
  • 99.Reilly PA, Burns A, Moran CJ. Sulphasalazine in early rheumatoid arthritis. Clin Exp Rheumatol. 1991;9(1):90-91. [PubMed] [Google Scholar]
  • 100.Rau R, Herborn G, Karger T, Menninger H, Elhardt D, Schmitt J. A double blind randomized parallel trial of intramuscular methotrexate and gold sodium thiomalate in early erosive rheumatoid arthritis. J Rheumatol. 1991;18(3):328-333. [PubMed] [Google Scholar]
  • 101.Williams HJ, Ward JR, Reading JC, et al. Comparison of auranofin, methotrexate, and the combination of both in the treatment of rheumatoid arthritis: a controlled clinical trial. Arthritis Rheum. 1992;35(3):259-269. doi: 10.1002/art.1780350304 [DOI] [PubMed] [Google Scholar]
  • 102.Willkens RF, Urowitz MB, Stablein DM, et al. Comparison of azathioprine, methotrexate, and the combination of both in the treatment of rheumatoid arthritis: a controlled clinical trial. Arthritis Rheum. 1992;35(8):849-856. doi: 10.1002/art.1780350802 [DOI] [PubMed] [Google Scholar]
  • 103.Alarcón GS, Castañeda O, Nair MG, Ferrándiz M, Koopman WJ, Krumdieck CL. Controlled trial of methotrexate versus 10-deazaaminopterin in the treatment of rheumatoid arthritis. Ann Rheum Dis. 1992;51(5):600-603. doi: 10.1136/ard.51.5.600 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 104.Porter D, Madhok R, Hunter JA, Capell HA. Prospective trial comparing the use of sulphasalazine and auranofin as second line drugs in patients with rheumatoid arthritis. Ann Rheum Dis. 1992;51(4):461-464. doi: 10.1136/ard.51.4.461 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 105.The Australian Multicentre Clinical Trial Group . Sulfasalazine in early rheumatoid arthritis. J Rheumatol. 1992;19(11):1672-1677. [PubMed] [Google Scholar]
  • 106.Lemmel EM; The European Multicentre Study Group . Comparison of pyritinol and auranofin in the treatment of rheumatoid arthritis. Br J Rheumatol. 1993;32(5):375-382. doi: 10.1093/rheumatology/32.5.375 [DOI] [PubMed] [Google Scholar]
  • 107.Faarvang KL, Egsmose C, Kryger P, Pødenphant J, Ingeman-Nielsen M, Hansen TM. Hydroxychloroquine and sulphasalazine alone and in combination in rheumatoid arthritis: a randomised double blind trial. Ann Rheum Dis. 1993;52(10):711-715. doi: 10.1136/ard.52.10.711 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 108.Clark P, Casas E, Tugwell P, et al. Hydroxychloroquine compared with placebo in rheumatoid arthritis: a randomized controlled trial. Ann Intern Med. 1993;119(11):1067-1071. doi: 10.7326/0003-4819-119-11-199312010-00002 [DOI] [PubMed] [Google Scholar]
  • 109.Hannonen P, Möttönen T, Hakola M, Oka M. Sulfasalazine in early rheumatoid arthritis: a 48-week double-blind, prospective, placebo-controlled study. Arthritis Rheum. 1993;36(11):1501-1509. doi: 10.1002/art.1780361104 [DOI] [PubMed] [Google Scholar]
  • 110.Haar D, Sølvkjaer M, Unger B, Rasmussen KJ, Christensen L, Hansen TM. A double-blind comparative study of hydroxychloroquine and dapsone, alone and in combination, in rheumatoid arthritis. Scand J Rheumatol. 1993;22(3):113-118. doi: 10.3109/03009749309099254 [DOI] [PubMed] [Google Scholar]
  • 111.Landewé RB, Goei Thè HS, van Rijthoven AW, Breedveld FC, Dijkmans BA. A randomized, double-blind, 24-week controlled study of low-dose cyclosporine versus chloroquine for early rheumatoid arthritis. Arthritis Rheum. 1994;37(5):637-643. doi: 10.1002/art.1780370506 [DOI] [PubMed] [Google Scholar]
  • 112.Krüger K, Schattenkirchner M. Comparison of cyclosporin A and azathioprine in the treatment of rheumatoid arthritis—results of a double-blind multicentre study. Clin Rheumatol. 1994;13(2):248-255. doi: 10.1007/BF02249021 [DOI] [PubMed] [Google Scholar]
  • 113.Benenson EV, Timina OB. Prospidine versus methotrexate pulse in highly active rheumatoid arthritis: a controlled 6-month clinical trial. Clin Rheumatol. 1994;13(1):54-59. doi: 10.1007/BF02229866 [DOI] [PubMed] [Google Scholar]
  • 114.Westedt ML, Dijkmans BA, Hermans J. Comparaison de l'azathioprine et du méthotrexate dans la polyarthrite rhumatoïde: une étude clinique ouverte randomisée. Comparison of azathioprine and methotrexate in rheumatoid arthritis: an open-randomized clinical study. Rev Rhum Ed Fr. 1994;61(9):591-597. [PubMed] [Google Scholar]
  • 115.Blackburn WD Jr, Prupas HM, Silverfield JC, et al. Tenidap in rheumatoid arthritis: a 24-week double-blind comparison with hydroxychloroquine-plus-piroxicam, and piroxicam alone. Arthritis Rheum. 1995;38(10):1447-1456. doi: 10.1002/art.1780381011 [DOI] [PubMed] [Google Scholar]
  • 116.Rao UR, Naidu MU, Kumar TR, et al. Comparison of phenytoin with auranofin and chloroquine in rheumatoid arthritis—a double blind study. J Rheumatol. 1995;22(7):1235-1240. [PubMed] [Google Scholar]
  • 117.Farr M, Waterhouse L, Johnson AE, Kitas GD, Jubb RW, Bacon PA. A double-blind controlled study comparing sulphasalazine with placebo in rheumatoid factor (RF)-negative rheumatoid arthritis. Clin Rheumatol. 1995;14(5):531-536. doi: 10.1007/BF02208150 [DOI] [PubMed] [Google Scholar]
  • 118.The HERA Study Group . A randomized trial of hydroxychloroquine in early rheumatoid arthritis: the HERA Study. Am J Med. 1995;98(2):156-168. doi: 10.1016/S0002-9343(99)80399-4 [DOI] [PubMed] [Google Scholar]
  • 119.van Schaardenburg D, Valkema R, Dijkmans BA, et al. Prednisone treatment of elderly-onset rheumatoid arthritis: disease activity and bone mass in comparison with chloroquine treatment. Arthritis Rheum. 1995;38(3):334-342. doi: 10.1002/art.1780380307 [DOI] [PubMed] [Google Scholar]
  • 120.Mladenovic V, Domljan Z, Rozman B, et al. Safety and effectiveness of leflunomide in the treatment of patients with active rheumatoid arthritis: results of a randomized, placebo-controlled, phase II study. Arthritis Rheum. 1995;38(11):1595-1603. doi: 10.1002/art.1780381111 [DOI] [PubMed] [Google Scholar]
  • 121.Salaffi F, Carotti M, Cervini C. Serum soluble interleukin-2 receptor levels in rheumatoid arthritis: effect of methotrexate, sulphasalazine and hydroxychloroquine therapy. Clin Rheumatol. 1995;14(4):458-463. doi: 10.1007/BF02207682 [DOI] [PubMed] [Google Scholar]
  • 122.Kim HA, Song YW. A comparison between bucillamine and D-penicillamine in the treatment of rheumatoid arthritis. Rheumatol Int. 1997;17(1):5-9. doi: 10.1007/PL00006849 [DOI] [PubMed] [Google Scholar]
  • 123.Haagsma CJ, van Riel PL, de Jong AJ, van de Putte LB. Combination of sulphasalazine and methotrexate versus the single components in early rheumatoid arthritis: a randomized, controlled, double-blind, 52 week clinical trial. Br J Rheumatol. 1997;36(10):1082-1088. doi: 10.1093/rheumatology/36.10.1082 [DOI] [PubMed] [Google Scholar]
  • 124.Rau R, Herborn G, Menninger H, Blechschmidt J. Comparison of intramuscular methotrexate and gold sodium thiomalate in the treatment of early erosive rheumatoid arthritis: 12 month data of a double-blind parallel study of 174 patients. Br J Rheumatol. 1997;36(3):345-352. doi: 10.1093/rheumatology/36.3.345 [DOI] [PubMed] [Google Scholar]
  • 125.Zeidler HK, Kvien TK, Hannonen P, et al. Progression of joint damage in early active severe rheumatoid arthritis during 18 months of treatment: comparison of low-dose cyclosporin and parenteral gold. Br J Rheumatol. 1998;37(8):874-882. doi: 10.1093/rheumatology/37.8.874 [DOI] [PubMed] [Google Scholar]
  • 126.Bjørneboe O, Moen F, Nygaard H, Haavik TK, Svensson B. CPH-82 (Reumacon) versus auranofin (Ridaura): a 36-week study of their respective onset of action rates in RA. Scand J Rheumatol. 1998;27(1):26-31. doi: 10.1080/030097498441137 [DOI] [PubMed] [Google Scholar]
  • 127.Dougados M, Combe B, Cantagrel A, et al. Combination therapy in early rheumatoid arthritis: a randomised, controlled, double blind 52 week clinical trial of sulphasalazine and methotrexate compared with the single components. Ann Rheum Dis. 1999;58(4):220-225. doi: 10.1136/ard.58.4.220 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 128.Smolen JS, Kalden JR, Scott DL, et al. ; European Leflunomide Study Group . Efficacy and safety of leflunomide compared with placebo and sulphasalazine in active rheumatoid arthritis: a double-blind, randomised, multicentre trial. Lancet. 1999;353(9149):259-266. doi: 10.1016/S0140-6736(98)09403-3 [DOI] [PubMed] [Google Scholar]
  • 129.Strand V, Cohen S, Schiff M, et al. ; Leflunomide Rheumatoid Arthritis Investigators Group . Treatment of active rheumatoid arthritis with leflunomide compared with placebo and methotrexate. Arch Intern Med. 1999;159(21):2542-2550. doi: 10.1001/archinte.159.21.2542 [DOI] [PubMed] [Google Scholar]
  • 130.Altman RDA, Schiff M, Kopp EJ. Cyclosporine A in rheumatoid arthritis: randomized, placebo controlled dose finding study. J Rheumatol. 1999;26(10):2102-2109. [PubMed] [Google Scholar]
  • 131.Lerndal T, Svensson B. A clinical study of CPH 82 vs methotrexate in early rheumatoid arthritis. Rheumatology (Oxford). 2000;39(3):316-320. doi: 10.1093/rheumatology/39.3.316 [DOI] [PubMed] [Google Scholar]
  • 132.Emery P, Breedveld FC, Lemmel EM, et al. A comparison of the efficacy and safety of leflunomide and methotrexate for the treatment of rheumatoid arthritis. Rheumatology (Oxford). 2000;39(6):655-665. doi: 10.1093/rheumatology/39.6.655 [DOI] [PubMed] [Google Scholar]
  • 133.Kraan MC, de Koster BM, Elferink JG, Post WJ, Breedveld FC, Tak PP; MC K . Inhibition of neutrophil migration soon after initiation of treatment with leflunomide or methotrexate in patients with rheumatoid arthritis: findings in a prospective, randomized, double-blind clinical trial in fifteen patients. Arthritis Rheum. 2000;43(7):1488-1495. doi: [DOI] [PubMed] [Google Scholar]
  • 134.Kraan MC, Reece RJ, Barg EC, et al. Modulation of inflammation and metalloproteinase expression in synovial tissue by leflunomide and methotrexate in patients with active rheumatoid arthritis: findings in a prospective, randomized, double-blind, parallel-design clinical trial in thirty-nine patients at two centers. Arthritis Rheum. 2000;43(8):1820-1830. doi: [DOI] [PubMed] [Google Scholar]
  • 135.Singh CP, Brar GK, Singh G, Singh J. The open randomised trial of cyclosporine vs methotrexate in refractory rheumatoid arthritis. Journal of Internal Medicine of India. 2000;3(1):19-25. [Google Scholar]
  • 136.Hamilton J, McInnes IB, Thomson EA, et al. Comparative study of intramuscular gold and methotrexate in a rheumatoid arthritis population from a socially deprived area. Ann Rheum Dis. 2001;60(6):566-572. doi: 10.1136/ard.60.6.566 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 137.Furst DE, Saag K, Fleischmann MR, et al. Efficacy of tacrolimus in rheumatoid arthritis patients who have been treated unsuccessfully with methotrexate: a six-month, double-blind, randomized, dose-ranging study. Arthritis Rheum. 2002;46(8):2020-2028. doi: 10.1002/art.10427 [DOI] [PubMed] [Google Scholar]
  • 138.van Everdingen AA, Jacobs JW, Siewertsz Van Reesema DR, Bijlsma JW; AA E . Low-dose prednisone therapy for patients with early active rheumatoid arthritis: clinical efficacy, disease-modifying properties, and side effects: a randomized, double-blind, placebo-controlled clinical trial. Ann Intern Med. 2002;136(1):1-12. doi: 10.7326/0003-4819-136-1-200201010-00006 [DOI] [PubMed] [Google Scholar]
  • 139.Choy EHS, Scott DL, Kingsley GH, et al. Treating rheumatoid arthritis early with disease modifying drugs reduces joint damage: a randomised double blind trial of sulphasalazine vs diclofenac sodium. Clin Exp Rheumatol. 2002;20(3):351-358. [PubMed] [Google Scholar]
  • 140.Yocum DE, Furst DE, Kaine JL, et al. ; Tcrolimus Rheumatoid Arthritis Study Group . Efficacy and safety of tacrolimus in patients with rheumatoid arthritis: a double-blind trial. Arthritis Rheum. 2003;48(12):3328-3337. doi: 10.1002/art.11363 [DOI] [PubMed] [Google Scholar]
  • 141.Bao C, Chen S, Gu Y, et al. Leflunomide, a new disease-modifying drug for treating active rheumatoid arthritis in methotrexate-controlled phase II clinical trial. Chin Med J (Engl). 2003;116(8):1228-1234. [PubMed] [Google Scholar]
  • 142.Kondo H, Abe T, Hashimoto H, et al. Efficacy and safety of tacrolimus (FK506) in treatment of rheumatoid arthritis: a randomized, double blind, placebo controlled dose-finding study. J Rheumatol. 2004;31(2):243-251. [PubMed] [Google Scholar]
  • 143.Rani PU, Mur N. Clinical efficacy and safety evaluation of eazmov plus in rheumatoid arthritis. Phytomedica. 2006;7(1):1-7. [Google Scholar]
  • 144.Lü LJ, Teng JL, Bao CD, et al. Safety and efficacy of T-614 in the treatment of patients with active rheumatoid arthritis: a double blind, randomized, placebo-controlled and multicenter trial. Chin Med J (Engl). 2008;121(7):615-619. doi: 10.1097/00029330-200804010-00008 [DOI] [PubMed] [Google Scholar]
  • 145.Braun J, Kästner P, Flaxenberg P, et al. ; MC-MTX.6/RH Study Group . Comparison of the clinical efficacy and safety of subcutaneous versus oral administration of methotrexate in patients with active rheumatoid arthritis: results of a six-month, multicenter, randomized, double-blind, controlled, phase IV trial. Arthritis Rheum. 2008;58(1):73-81. doi: 10.1002/art.23144 [DOI] [PubMed] [Google Scholar]
  • 146.Goldbach-Mansky R, Wilson M, Fleischmann R, et al. Comparison of Tripterygium wilfordii Hook F versus sulfasalazine in the treatment of rheumatoid arthritis: a randomized trial. Ann Intern Med. 2009;151(4):229-240, W49-51. doi: 10.7326/0003-4819-151-4-200908180-00005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 147.Lu L-J, Bao CD, Dai M, et al. Multicenter, randomized, double-blind, controlled trial of treatment of active rheumatoid arthritis with T-614 compared with methotrexate. Arthritis Rheum. 2009;61(7):979-987. doi: 10.1002/art.24643 [DOI] [PubMed] [Google Scholar]
  • 148.Schiff M, Beaulieu A, Scott DL, Rashford M. Mycophenolate mofetil in the treatment of adults with advanced rheumatoid arthritis: three 24-week, randomized, double-blind, placebo- or ciclosporin-controlled trials. Clin Drug Investig. 2010;30(9):613-624. doi: 10.2165/11537460-000000000-00000 [DOI] [PubMed] [Google Scholar]
  • 149.Ishaq M, Muhammad JS, Hameed K, Mirza AI. Leflunomide or methotrexate: comparison of clinical efficacy and safety in low socio-economic rheumatoid arthritis patients. Mod Rheumatol. 2011;21(4):375-380. doi: 10.3109/s10165-010-0405-4 [DOI] [PubMed] [Google Scholar]
  • 150.Lv QW, Zhang W, Shi Q, et al. Comparison of Tripterygium wilfordii Hook F with methotrexate in the treatment of active rheumatoid arthritis (TRIFRA): a randomised, controlled clinical trial. Ann Rheum Dis. 2015;74(6):1078-1086. doi: 10.1136/annrheumdis-2013-204807 [DOI] [PubMed] [Google Scholar]
  • 151.Zhao L, Jiang Z, Zhang Y, Ma H, Cai C. Analysis of efficacy and safety of treatment of active rheumatoid arthritis with iguratimod and methotrexate. Biomed Res (Aligarh). 2017;28(5):2353-2359. [Google Scholar]
  • 152.Vickers AJ. The use of percentage change from baseline as an outcome in a controlled trial is statistically inefficient: a simulation study. BMC Med Res Methodol. 2001;1:6. doi: 10.1186/1471-2288-1-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 153.Lu G, Ades AE. Combination of direct and indirect evidence in mixed treatment comparisons. Stat Med. 2004;23(20):3105-3124. doi: 10.1002/sim.1875 [DOI] [PubMed] [Google Scholar]
  • 154.Dias S, Sutton AJ, Ades AE, Welton NJ. Evidence synthesis for decision making 2: a generalized linear modeling framework for pairwise and network meta-analysis of randomized controlled trials. Med Decis Making. 2013;33(5):607-617. doi: 10.1177/0272989X12458724 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 155.Lunn DJ, Thomas A, Best N, Spiegelhalter D. WinBUGS—a bayesian modelling framework: concepts, structure, and extensibility. Stat Comput. 2000;10:325-337. doi: 10.1023/A:1008929526011 [DOI] [Google Scholar]
  • 156.Brooks SP, Gelman A. General methods for monitoring convergence of iterative simulations. J Comput Graph Stat. 1998;7(4):434-455. doi: 10.1080/10618600.1998.10474787 [DOI] [Google Scholar]
  • 157.Daly C, Downing BC, Welton NJ. A practical guide to inconsistency checks in bayesian network meta-analysis. Accessed August 29, 2023. https://www.bristol.ac.uk/media-library/sites/social-community-medicine/documents/mpes/Guide%20to%20Checking%20for%20Inconsistency%20in%20NMA_TSU.pdf
  • 158.Dias S, Sutton AJ, Welton NJ, Ades AE. Evidence synthesis for decision making 3: heterogeneity–subgroups, meta-regression, bias, and bias-adjustment. Med Decis Making. 2013;33(5):618-640. doi: 10.1177/0272989X13485157 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 159.Taggart AJ, Neumann VC, Hill J, Astbury C, Le Gallez P, Dixon JS. 5-Aminosalicylic acid or sulphapyridine: which is the active moiety of sulphasalazine in rheumatoid arthritis? Drugs. 1986;32(suppl 1):27-34. doi: 10.2165/00003495-198600321-00006 [DOI] [PubMed] [Google Scholar]
  • 160.Smolen JS, Landewé RBM, Bijlsma JWJ, et al. EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2019 update. Ann Rheum Dis. 2020;79(6):685-699. doi: 10.1136/annrheumdis-2019-216655 [DOI] [PubMed] [Google Scholar]
  • 161.Faurschou M, Mellemkjaer L, Voss A, Keller KK, Hansen IT, Baslund B. Prolonged risk of specific malignancies following cyclophosphamide therapy among patients with granulomatosis with polyangiitis. Rheumatology (Oxford). 2015;54(8):1345-1350. doi: 10.1093/rheumatology/keu372 [DOI] [PubMed] [Google Scholar]
  • 162.Tatangelo M, Tomlinson G, Paterson JM, et al. Association of patient, prescriber, and region with the initiation of first prescription of biologic disease-modifying antirheumatic drug among older patients with rheumatoid arthritis and identical health insurance coverage. JAMA Netw Open. 2019;2(12):e1917053. doi: 10.1001/jamanetworkopen.2019.17053 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 163.Graudal N, Hubeck-Graudal T, Tarp S, Christensen R, Jürgens G. Effect of combination therapy on joint destruction in rheumatoid arthritis: a network meta-analysis of randomized controlled trials. PLoS One. 2014;9(9):e106408. doi: 10.1371/journal.pone.0106408 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 164.Scott DL, Ibrahim F, Farewell V, et al. Tumour necrosis factor inhibitors versus combination intensive therapy with conventional disease modifying anti-rheumatic drugs in established rheumatoid arthritis: TACIT non-inferiority randomised controlled trial. BMJ. 2015;350:h1046. doi: 10.1136/bmj.h1046 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 165.Graudal N, Hubeck-Graudal T, Faurschou M, Baslund B, Jürgens G. Combination therapy with and without tumor necrosis factor inhibitors in rheumatoid arthritis: a meta-analysis of randomized trials. Arthritis Care Res (Hoboken). 2015;67(11):1487-1495. doi: 10.1002/acr.22618 [DOI] [PubMed] [Google Scholar]
  • 166.Suarez-Almazor ME, Belseck E, Shea B, Wells G, Tugwell P. Sulfasalazine for rheumatoid arthritis. Cochrane Database Syst Rev. 2000;1998(2):CD000958. doi: 10.1002/14651858.CD000958 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 167.Clark P, Tugwell P, Bennet K, et al. Injectable gold for rheumatoid arthritis. Cochrane Database Syst Rev. 2000;1997(2):CD000520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 168.Suarez-Almazor ME, Spooner CH, Belseck E, Shea B. Auranofin versus placebo in rheumatoid arthritis. Cochrane Database Syst Rev. 2000;2000(2):CD002048. doi: 10.1002/14651858.CD002048 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 169.Suarez-Almazor ME, Belseck E, Shea B, Wells G, Tugwell P. Cyclophosphamide for rheumatoid arthritis. Cochrane Database Syst Rev. 2000;2:CD001157. doi: 10.1002/14651858.CD001157 [DOI] [PubMed] [Google Scholar]
  • 170.Suarez-Almazor ME, Spooner C, Belseck E. Azathioprine for rheumatoid arthritis. Cochrane Database Syst Rev. 2000;2:CD001461. doi: 10.1002/14651858.CD001461 [DOI] [PubMed] [Google Scholar]
  • 171.Suarez-Almazor ME, Spooner C, Belseck E. Penicillamine for rheumatoid arthritis. Cochrane Database Syst Rev. 2000;2:CD001460. doi: 10.1002/14651858.CD001460 [DOI] [PubMed] [Google Scholar]
  • 172.Osiri M, Shea B, Robinson V, et al. Leflunomide for treating rheumatoid arthritis. Cochrane Database Syst Rev. 2003;2002(1):CD002047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 173.Golicki D, Newada M, Lis J, Pol K, Hermanowski T, Tłustochowicz M. Leflunomide in monotherapy of rheumatoid arthritis: meta-analysis of randomized trials. Pol Arch Med Wewn. 2012;122(1-2):22-32. doi: 10.20452/pamw.1131 [DOI] [PubMed] [Google Scholar]
  • 174.Hazlewood GS, Barnabe C, Tomlinson G, Marshall D, Devoe DJ, Bombardier C. Methotrexate monotherapy and methotrexate combination therapy with traditional and biologic disease modifying anti-rheumatic drugs for rheumatoid arthritis: a network meta-analysis. Cochrane Database Syst Rev. 2016;2016(8):CD010227. doi: 10.1002/14651858.CD010227.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 175.Alfaro-Lara R, Espinosa-Ortega HF, Arce-Salinas CA; PRECIS study group, all physicians belong to Division of Internal Medicine. Hospital Central Sur de Pemex . Systematic review and meta-analysis of the efficacy and safety of leflunomide and methotrexate in the treatment of rheumatoid arthritis. Reumatol Clin (Engl Ed). 2019;15(3):133-139. doi: 10.1016/j.reuma.2017.07.020 [DOI] [PubMed] [Google Scholar]
  • 176.Begg C, Cho M, Eastwood S, et al. Improving the quality of reporting of randomized controlled trials. The CONSORT statement. JAMA. 1996;276(8):637-639. doi: 10.1001/jama.1996.03540080059030 [DOI] [PubMed] [Google Scholar]
  • 177.National Library of Medicine . About ClinicalTrials.gov. Accessed August 29, 2023. https://clinicaltrials.gov/ct2/about-site/background
  • 178.Khan NA, Singh M, Spencer HJ, Torralba KD. Randomized controlled trials of rheumatoid arthritis registered at ClinicalTrials.gov: what gets published and when. Arthritis Rheumatol. 2014;66(10):2664-2674. doi: 10.1002/art.38784 [DOI] [PubMed] [Google Scholar]
  • 179.Stone M, Fortin PR, Pacheco-Tena C, Inman RD. Should tetracycline treatment be used more extensively for rheumatoid arthritis: metaanalysis demonstrates clinical benefit with reduction in disease activity. J Rheumatol. 2003;30(10):2112-2122. [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

eMethods 1. Eligibility Criteria

eMethods 2. Information Sources and Search

eMethods 3. Study Selection and Data Collection

eMethods 4. Data Items

eMethods 5. Estimations of Missing Values and Conversion Factors

eMethods 6. Risk of Bias Within Individual Studies

eMethods 7. Planned Method of Analysis

eMethods 8. Effect Modifier Analysis

eResults 1. Study Selection

eResults 2. Primary Outcome (Complete Analysis) With Methotrexate as Reference, Unadjusted Analysis

eResults 3. Effect Modifier Analyses

eResults 4. Primary Outcome With Methotrexate as Reference, Adjusted for Baseline Disease Activity Score 28 (DAS28)

eResults 5. Exploration of Inconsistency

eResults 6. Results of Additional Analyses

eReferences.

eTable 1. Baseline Characteristics, Primary Outcome and Bias Score of Included Studies

eTable 2. Unadjusted Tender Joint Count Imputed (TJCi) Analysis Using Fixed- and Random-Effects Models

eTable 3. Tender Joint Count Imputed (TJCi) Sensitivity Analysis Comparing Analyses With and Without Imputed Values

eTable 4. Model Fit Parameters for Tender Joint Count Imputed (TJCi) Analyses Including and Excluding Effect Modifiers in Metaregression Analyses

eTable 5. Primary Outcome and Sensitivity Outcomes With Placebo as Reference

eTable 6. Comparisons of Tender Joint Count Imputed (TJCi) Outcome of All Interventions vs All Interventions

eTable 7. Model Fit Data for Primary Outcome and Sensitivity Outcomes

eTable 8. Primary and Sensitivity Outcomes Calculated With Prior 0-20

eFigure 1. Subgroup Analyses

eFigure 2. Forest Plot of Primary Tender Joint Count Imputed (TJCi) Outcomes vs Peroral Methotrexate

eFigure 3. Rankograms of 29 Tested Interventions

eFigure 4. dev-dev Plot of Unrelated Mean Effects Mean vs Network Meta-Analyses Mean

Click here for additional data file. (1.2MB, pdf)
Supplement 2.

Data Sharing Statement

Click here for additional data file. (140.2KB, pdf)

Articles from JAMA Network Open are provided here courtesy of American Medical Association

RESOURCES