Methotrexate for ankylosing spondylitis

  • Review
  • Intervention

Authors

  • Junmin Chen,

    Corresponding author
    1. The First Affiliated Hospital of Fujian Medical University, Department of Hematology and Rheumatology, Fuzhou, Fujian Province, China
    • Junmin Chen, Department of Hematology and Rheumatology, The First Affiliated Hospital of Fujian Medical University, #20 Chazhong Road, Fuzhou, Fujian Province, 350005, China. drjunminchen@hotmail.com.

    Search for more papers by this author
  • Mirella MS Veras,

    1. University of Ottawa, Centre for Global Health, Institute of Population Health, Ottawa, Ontario, Canada
    Search for more papers by this author
  • Chao Liu,

    1. The First Hospital of Hebei Medical University, Shijiazhuang, Hebei Province, China
    Search for more papers by this author
  • Junfang Lin

    1. The First Affiliated Hospital of Fujian Medical University, Department of Hematology and Rheumatology, Fuzhou, Fujian Province, China
    Search for more papers by this author

Abstract

Background

Ankylosing spondylitis (AS) is a chronic inflammatory disease of unknown cause, characterized by sacroiliitis and spondylitis. Methotrexate (MTX), a widely used disease-modifying antirheumatic drug (DMARD), is effective for rheumatoid arthritis (RA), and so might work for AS. This is an update of a Cochrane review first published in 2004, and previously updated in 2006.

Objectives

To evaluate the benefits and harms of MTX for treating AS.

Search methods

We searched CENTRAL (The Cochrane Library Issue 6, 2012), MEDLINE (2005 to June 25, 2012), EMBASE (2005 to June 25, 2012), Ovid MEDLINE Scopus, World Health Organization International Clinical Trials Registry Platform and the reference sections of retrieved articles. Trials published in any language were acceptable.

Selection criteria

Randomized controlled trials (RCTs) and quasi-randomized controlled trials (qRCTs) examining the benefits and harms of MTX versus placebo, other medication, or no medication for treatment of AS.

Data collection and analysis

Two review authors independently extracted data and assessed risk of bias. We resolved any disagreements through discussions with a third review author. In the absence of significant heterogeneity, we combined results for continuous data using mean difference or standardized mean difference values. We calculated the risk ratio for dichotomous data.

Main results

We identified three RCTs (no additional new studies), which included 116 participants. Of these three trials, one was a 12-month trial that compared naproxen plus MTX with naproxen alone. Also, there were two 24-week trials that compared different doses of MTX with placebo. We included the outcomes of response, physical function, pain, spinal mobility, peripheral joints/entheses pain, swelling and tenderness, changes in spine radiographs, and patient and physician global assessment. We judged only one trial to be at low risk of bias. Across these three trials, we did not identify any statistically significant differences favoring MTX treatment over no MTX treatment apart from one exception. The response rate in one trial showed a statistically significant absolute benefit of 36% and a number to treat for benefit (NNT) of three in the MTX group compared to the placebo group (RR 3.18, 95% CI 1.03 to 9.79). This response rate was based on a composite index that included assessments of morning stiffness, physical well-being, Bath ankylosing spondylitis disease activity index (BASDAI), Bath ankylosing spondylitis functional index (BASFI), health assessment questionnaire for spondyloarthropathies (HAQ-S), and physician and patient global assessment. We did not identify any outcome that showed a statistically significant difference between the MTX treated and no MTX treatment groups when endpoint results were compared. Furthermore, no serious side effects were reported in any of the included trials.

Authors' conclusions

There is not enough evidence to support any benefit of MTX in the treatment of AS. High-quality RCTs of larger sample sizes are needed to clarify the effect(s) of MTX on AS.

Plain language summary

Methotrexate for ankylosing spondylitis

Researchers in the Cochrane Collaboration conducted a review about the effect of the drug methotrexate in people with ankylosing spondylitis (AS). They identified three studies, which included 116 participants, that met the inclusion criteria.

The review shows that for people with AS:

- methotrexate probably improves physical function but this may have happened by chance.

- it is uncertain whether methotrexate will help to ease pain, tenderness, and swelling in the ligaments of the joints, movement of the spine, stiffness, or overall well-being because there were not enough participants studied.

- it is uncertain whether methotrexate slows damage to the joints because the studies did not look at an x-ray of the spines of the people with AS.

We often do not have precise information about side effects and complications. This is particularly true for rare but serious side effects. Possible side effects include stomach problems, mild headaches, mouth sores, changes in liver function, hair loss, or mild infections. Rare complications may include lung problems. It is important to keep medical appointments to catch any serious problems early. 

What is AS and what is methotrexate? 

AS is a type of arthritis that usually occurs in the joints and ligaments of the spine. It may also affect the shoulders, hips, or other joints. Pain and stiffness occurs and limits movement in the back and in other joints that are affected.

Methotrexate (MTX) is a disease-modifying antirheumatic drug (DMARD). It is the most commonly used DMARD in people with inflammation in their joints. It works to control inflammation in affected joints to stop the pain and stiffness. MTX is taken once per week and may be administered in pill form or as an injection.

எளியமொழிச் சுருக்கம்

தம்ப முள்ளந்தண்டழல்க்கு மெதொடிரெக்ஸேட்

காக்ரேன் ஒத்துழைப்பு ஆராய்ச்சியாளர்கள் தம்ப முள்ளந்தண்டழல் (ankylosing spondylitis (AS)) உள்ளவர்களுக்கு மெதொடிரெக்ஸேட் மருந்தின் திறன் பற்றி ஒரு ஆய்வு நடத்தினர். மொத்தம் 116 பங்கேற்பாளர்கள் கொண்ட 3 ஆய்வுகள் சேர்ப்பு விதிகளுக்கு உட்பட்டு இருந்தன.

இந்த ஆய்வு AS பாதிக்கப்பட்ட மக்களுக்கு தெரிவிப்பது:

-மெதொடிரெக்ஸேட் உடற்சார்ந்த செயல்பாட்டு திறனை அதிகரிக்க வாய்ப்பு உள்ளது .ஆனால் அது எதேச்சையாக நடக்கக் கூடும்.

-ஆய்வுகளில் போதுமான பங்கேற்பாளர்கள் இல்லாததால் மெதொடிரெக்ஸேட் (methotrexate) வலி, ரணம், மற்றும் மூட்டுகளில் தசைநார்களின் வீக்கம், முதுகெலும்பு இயக்கம், விறைப்பு மற்றும் ஒட்டுமொத்த நலனுக்காக உதவும் என்பது உறுதியற்றதாக உள்ளது.

-மெதொடிரெக்ஸேட் மூட்டுகளின் சேதத்தைத் தாமதப்படுத்தும் என்பது உறுதியற்றதாக உள்ளது ஏனெனில் இந்த ஆய்வுகளில் தம்ப முள்ளந்தண்டழல் (AS) உடைய மக்களின் முதுகு தண்டை X -கதிர் கொண்டு சோதிக்கவில்லை.

பக்கவிளைவுகளைப் பற்றி துல்லியமான தகவல்கள் இல்லை. குறிப்பாக அரிதான ஆனால் கடுமையான பக்க விளைவுகளை பற்றி தகவல்கள் இல்லை. வயிற்று பிரச்சினைகள்,லேசான தலைவலி, வாய்ப்புண், கல்லீரல் செயல்பாடு மாற்றங்கள், முடி உதிர்தல், அல்லது லேசான தொற்று நோய்கள் முதலியன சாத்தியமான பக்க விளைவுகள். அரிதான சிக்கல்களில் நுரையீரல் பிரச்சனைகள் அடங்கும். தீவிர பிரச்சனைகளைக் கண்டறிய மருத்துவ நியம சந்திப்புகளைச் சரியாக கடைப்பிடிப்பது முக்கியம்.

AS மற்றும் மெத்தோட்ரெக்ஸேட் என்றால் என்ன?

AS ஒரு விதமான வாதநோய், இவை முதுகெலும்பு மூட்டுகள் மற்றும் தசைநார்களைப் பாதிக்கும். இது தோள்கள், இடுப்பு, அல்லது மற்ற இதர மூட்டுகளையும் பாதிக்கும். முதுகு மற்றும் இதர முடுகளில் இயக்கத்தை, வலி மற்றும் விறைப்புத்தன்மை கட்டுப்படுத்தும்.

மெதொடிரெக்ஸே (MTX) ஒரு நோய் மாற்றம் அளிக்ககூடிய ஆன்டிரூமாடிக் மருந்து (DMARD) ஆகும். இது மூட்டுகளில் வீக்கம் உள்ள நோயாளிகளுக்கு DMARD பொதுவாக பயன்படுத்தப்படுகிறது. இவை பாதிக்கப்பட்ட மூட்டுகளில் உள்ள வீக்கத்தைக் கட்டுப்படுத்துவதன் மூலம் மூட்டுகளின் வலி மற்றும் விறைப்புத்தன்மையையும் குறைக்கும். MTX வாரத்திற்கு ஒருமுறை மாத்திரை வடிவிலோ அல்லது ஒரு ஊசி மூலமோ அளிக்கப்படுகின்றன.

மொழிபெயர்ப்பு குறிப்புகள்

மொழிபெயர்ப்பு: ஜெயலக்ஷ்மி மற்றும் சி.இ.பி.என்.அர். குழு

Laički sažetak

Metotreksat kao lijek za ankilozantni spondilitis

Znanstvenici Cochranea proveli su sustavni pregled literature o djelovanju lijeka metotreksata u oboljelih od ankilozantnog spondilitisa (AS). Pronašli su tri studije, koje su uključile 116 sudionika, a zadovoljile su kriterije istraživanja.

Članci su pokazali da u oboljelih od ankilozantnog spondilitisa:

- metotreksat vjerojatno poboljšava fizičku funkciju, ali to može biti slučajan nalaz.

- nije sigurno hoće li metotreksat pomoći kod ublažavanja bolova, osjetljivosti, oteklina ligamenata u zglobu, pokretima u kralježnici, ukočenosti, ili u sveukupnom općem stanju jer nije bilo uključeno dovoljno sudionika.

- nije sigurno hoće li metotreksat usporiti oštećenje zglobova jer istraživanja nisu uključivala rentgenske snimke kralježnice kod ljudi sa ankilozantnim spondilitisom.

Autori istraživanja često ne opisuju precizne podatke o nuspojavama i komplikacijama. To posebno vrijedi za rijetke, ali ozbiljne nuspojave. Moguće nuspojave uključuju probleme u trbuhu, blage glavobolje, rane u ustima, promjene u funkciji jetre, gubitak kose, ili blage infekcije. Rijetke komplikacije mogu uključivati plućne probleme. Važno je obavljati liječničke preglede kako bi se rano otkrili ozbiljni problemi. 

Što je ankilozantni spondilitis, a što metotreksat?

Ankilozantni spondilitis je vrsta artritisa koja se javlja u zglobovima i ligamentima kralježnice. Također može zahvatiti rame, kuk i druge zglobove. Javlja se bol i ukočenost koja ograničava pokrete u leđima i drugim zglobovima koji su zahvaćeni.

Metotreksat (MTX) je lijek koji utječe na tijek upalnih reumatskih bolesti (engl. disease-modifying antirheumatic drug, DMARD). To je najčešće korišteni DMARD lijek kod ljudi s upalom u zglobovima. Djeluje na upalu u zahvaćenim zglobovima kao bi zaustavio bol i ukočenost. Metotreksat se uzima jedanput tjedno u obliku tablete ili injekcije.

Bilješke prijevoda

Hrvatski Cochrane
Ovaj sažetak preveden je u okviru volonterskog projekta prevođenja Cochrane sažetaka. Uključite se u projekt i pomozite nam u prevođenju brojnih preostalih Cochrane sažetaka koji su još uvijek dostupni samo na engleskom jeziku. Kontakt: cochrane_croatia@mefst.hr

Summary of findings(Explanation)

Summary of findings for the main comparison. MTX treated compared to no MTX treated for AS
  1. 1 Baseline imbalance.

    2 Few patients and few events.

MTX treated compared to no MTX treated for AS
Patient or population: Patients with AS
Settings: Inpatient clinic in Turkey (Altan 2001), outpatient rheumatology clinic in Guadalajara, Mexico (Gonzalez-Lopez 2004), and United Kingdom (Roychowdhury 2002)
Intervention: MTX
Comparison: No MTX
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
No MTXMTX
ASAS 40See commentSee commentNot estimable-See commentNot reported
Partial remissionSee commentSee commentNot estimable-See commentNot reported
BASFI (score from 0 to 10, 0=excellent, 10=poor function)

Follow-up: 24 weeks

The mean change

in BASFI score in the control group was
0.4

The mean change

in BASFI score in the intervention group was
0.9 lower

(0.43 lower to 2.23 higher)

 35
(1 study)

⊕⊕⊝⊝

low 1,2

MD = 0.9 (-0.43 to 2.23)

Relative improvement 8.57%
(95% CI 29.42 to 46.57)

Absolute improvement

3% (95% CI 10.3% to 16.3%)

Not statistically significant

X-Ray of the spineSee commentSee commentNot estimable-See commentNot reported
MRISee commentSee commentNot estimable-See commentNot reported
Serious adverse eventSee commentSee commentNot estimable-See commentNot reported

Withdrawal for any reason

Follow-up: 24 weeks

56 per 100059 per 1000
(4 to 868)
RR 1.06
(0.07 to 15.62)
35
(1 study)

⊕⊕⊝⊝

low 1,2

Relative percent change 6 %
(95% CI 93% to 1462%)

Absolute risk difference 0%
(95% CI 15% to 16%)

Not statistically significant

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

Background

Description of the condition

Ankylosing spondylitis (AS) is a type of arthritis, i.e. a chronic inflammatory disease of unknown cause. Most commonly it occurs in the joints and ligaments of the spine, although it may also affect the shoulders, hips, and other joints. As a result, the condition is characterized by sacroiliitis and spondylitis, i.e. pain, stiffness, and limited movement in the back and other affected joints. AS can come and go, last for long periods, and can be quite severe. Depending on the geographical region, the prevalence of AS can range from 0.1% to 1.4% of the population (Akkoc 2005). People with AS can have a significant reduction in quality of life and an increased mortality rate. Furthermore, AS has a considerable impact on healthcare costs and also affects workforce participation (Annelies 2006).

Description of the intervention

Non-steroidal anti-inflammatory drugs (NSAIDs) are the cornerstone of pharmaceutical treatment. For patients refractory or intolerant to NSAIDs, conventional disease-modifying antirheumatic drugs (DMARDs) such as gold salts, antimalarial drugs, methotrexate, D-penicillamine, sulfasalazine, and various immunosuppressive treatments are used. However, the benefits of DMARDs are unclear except for sulfasalazine, which was found to improve peripheral joint inflammation (Chen 2005). Anti-TNF agents, including adalimumab, etanercept, and infliximab, have proven considerably effective (McLeod 2007) but the latter's expense means that few patients can access these drugs.

How the intervention might work

Methotrexate (MTX) is currently one of the most widely used DMARDs. Its benefit in patients with rheumatoid arthritis (RA) has been confirmed by several well-designed randomized controlled clinical trials (RCTs) (Suarez-Almazor 2003; Katchamart 2010).

Why it is important to do this review

There is uncertainty whether MTX is an effective drug for treatment of AS. We performed a systematic review in 2003 and this is an update of the original review.

Objectives

To evaluate the benefit and toxicity of MTX for the treatment of AS by comparing MTX versus placebo; MTX versus other medication; and MTX versus no medication.

Methods

Criteria for considering studies for this review

Types of studies

We evaluated RCTs and quasi-RCTs (qRCTs) examining the benefits and harms of MTX on AS.

Types of participants

We included patients with AS fulfilling one of the following or other criteria: 1961 Rome, 1966 New York, modified 1984 New York, Amor, or European Spondylarthropathy Study Group (ESSG) criteria (Olivieri 2002). Furthermore, we included studies with a mixture of spondyloarthropathies/spondyloarthritis patients and participants were included if data assessing the outcomes specific to patients with AS were available. Regarding inclusion of participants, we had no age or sex restrictions. Also, we included patients with or without the impairment of peripheral joints.

Types of interventions

MTX given orally or intramuscularly for at least 12 weeks. We included the following comparisons:
1. MTX versus placebo;
2. MTX versus other medication;
3. MTX versus no medication.

Types of outcome measures

Based on the core set for the evaluation of disease-controlling antirheumatic treatment (DC-ART) proposed by Assessment of Ankylosing Spondylitis (ASAS) Working Group (van der Heijde 1999; van der Heijde 2002), we included the following outcomes:

Major outcomes
  1. ASAS 40 (van der Heijde 2002; van der Heijde 2005) or ASDAS major improvement (Sieper 2009);

  2. Partial remission ( defined as a value of two or less on a scale from 0 to 10 in each of the four domains of the ASAS 20);

  3. Bath ankylosing spondylitis function index (BASFI);

  4. X-ray of the spine;

  5. MRI (spine, or SI joints, or both);

  6. Serious adverse events;

  7. Withdrawals due to adverse events.

Minor outcomes
  • Other ASAS responses (van der Heijde 2002; van der Heijde 2005);

  • Disease activity;

  • Physical function;

  • Pain;

  • Spinal mobility and spinal stiffness;

  • Peripheral joints/entheses (pain, swelling, and tenderness);

  • Changes in hip radiograph and MRI;

  • Patient global assessment and physician global assessment;

  • Fatigue;

  • Level of acute phase reactants, including erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP);

  • Any side effects.

Search methods for identification of studies

We searched for relevant RCTs and qRCTs in any language for the first version of this Cochrane Review in 2003, using CENTRAL (Cochrane Central Register of Controlled Trials), MEDLINE, EMBASE and CINAHL (see Appendix 1). For the second version review in 2006, all searches were updated on November 20, 2005 (Appendix 2). For this, the third version, we searched CENTRAL (The Cochrane Library Issue 6 of 12, 2012), MEDLINE (2005 to June 25, 2012), EMBASE (2005 to June 25, 2012), CINAHL (2005 to June 25, 2012), Scopus (2005 to June 25, 2012), World Health Organization International Clinical Trials Registry Platform (June 25, 2012) and the reference sections of retrieved articles. We included the full search strategies in Appendix 3 and the number of records retrieved in Appendix 4.

Data collection and analysis

Selection of studies

Two authors (JC and MMSV) independently reviewed the full-text trial reports for this review update according to the selection criteria. We resolved any disagreements regarding the inclusion of studies by recourse to a third review author.

Data extraction and management

We extracted the data from the included trials on a pre-structured data extraction form. Two authors (JC and MMSV) independently entered this data into Review Manager (RevMan).

In this review, we only included outcomes that were specified in the previously published protocol. We entered continuous data (e.g. visual analogue scales of pain, patient global assessment) as means and standard deviations (SD), and dichotomous outcomes (e.g. response, improvement) as number of events.

Assessment of risk of bias in included studies

For this review, we assessed risk of bias in all included studies using The Cochrane Collaboration's tool for assessing risk of bias (Higgins 2011). There were six domains: random sequence generation, allocation concealment, blinding of participants, personnel and outcome assessment, incomplete outcome data, selective reporting and other sources of bias such as baseline imbalance. We made a judgment about each risk of bias domain according to the criteria described in the Cochrane Handbook (Appendix 5). We categorized risk of bias as either 'low risk', 'high risk' or 'unclear risk'. Two review authors (JC and MMSV) independently assessed the included trials using a data collection form (Appendix 6).

Measures of treatment effect

For continuous data, we used mean difference (WMD) or standardized mean difference (SMD) values depending on the comparability of scales. For dichotomous data, we used risk ratio (RR) values.

Unit of analysis issues

In this review we included only RCTs.

Dealing with missing data

For continuous data (e.g. visual analogue scales of pain, patient global assessment), we analyzed only the available data. For dichotomous data (e.g. response, improvement), we used intention-to-treat (ITT) analysis.

Assessment of heterogeneity

We explored heterogeneity by using the Chi2 test with significance set at P < 0.10. We measured the quantity of heterogeneity by using the I 2 statistic, with I 2 > 0.50 as substantial heterogeneity.

Assessment of reporting biases

We planned to use funnel plots to detect publication bias. However, we did not pursue this any further as we identified only three trials that met the inclusion criteria.

Data synthesis

We performed meta-analysis using Review Manager (RevMan). We used a random-effects model to combine the results where heterogeneity was significant. Otherwise, we used a fixed-effect model.

Subgroup analysis and investigation of heterogeneity

We planned to perform subgroup analyses by characteristics of participants (e.g. different AS classification criteria, male or female, with or without peripheral arthritis) and intervention (e.g. different dosage, different duration). However, we did not undertake these analyses as these subgroup datasets were not available.

Sensitivity analysis

We planned two sensitivity analyses according to whether:

  1. the allocation to intervention or control groups was truly randomized or quasi-randomized, to explore the potential for selection bias;

  2. the outcome assessment was blinded, to explore the potential for assessment bias associated with knowledge of the intervention. If either of the sensitivity analyses affected the results of the review, then conservative conclusions would be drawn.

However, we did not perform these sensitivity analyses as the number of trials that met the inclusion criteria was too low.

Summary of finding table

We performed grading of the evidence using the GRADE profiler software, which was developed by the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) Working Group (www.gradeworkinggroup.org). A summary of findings table was created in Review Manager (RevMan). We presented seven outcomes that were deemed important to patients. They were ASAS40 (or ASDAS major improvement), partial improvement, BASFI, x-ray of spine, MRI (spine, or SI joints, or both), serious adverse events, and withdrawal due to adverse events.

Results

Description of studies

Results of the search

We obtained 320 records from the updated search (Figure 1). Following screening of titles and abstracts, we excluded 313 records. After assessment of full text articles, we identified two records as repeated publications of Breban 2008 and Li 2008. Two studies (Pérez-Guijo 2007; Kabasakal 2009) were not RCTs. Three trials (Breban 2008, Li 2008 and Mulleman 2011)examined the efficacy of infliximab. We excluded these trials from the present review because there is a separate Cochrane review that addresses biologics in AS (Zochling 2005), according to the Cochrane Musculoskeletal Group (CMSG). Therefore, we did not include any new studies to the previous version review (Chen 2006), which included three RCTs (Altan 2001; Roychowdhury 2002; Gonzalez-Lopez 2004).

Figure 1.

Study flow diagram.

Included studies

Among the three included trials, one trial (Altan 2001) compared naproxen plus MTX (7.5 mg/week orally) with naproxen alone over a duration of 12 months. Two trials (Gonzalez-Lopez 2004; Roychowdhury 2002) compared MTX with placebo, using MTX doses of either 10 mg/week orally or 7.5 mg/week orally, respectively, for a duration of 24 weeks. In total, 116 participants were included in the trials: 89 males and 27 females. Their mean ages ranged from 32 to 46 years and the mean duration of disease was from 5.7 to 17 years. Between 11.7% and 65% of the participants had peripheral arthritis (i.e. 6/51, 9/30, and 21/35 in the Altan 2001, Gonzalez-Lopez 2004, and Roychowdhury 2002, respectively). None of these three studies provided a definition of peripheral arthritis. A total of 55 participants received MTX and 59 did not. We have included further details in the Characteristics of included studies table.

Excluded studies

We excluded one trial (Xie 2003) because it compared the effects of different dosages of MTX and did not include the comparison specified in the inclusion criteria. Two trials (Pérez-Guijo 2007; Kabasakal 2009) were not RCTs. Another trial (Haque 1999), which treated 32 patients with seronegative spondyloarthropathy, could not be reviewed because of the lack of a full text article. To date, our attempts to contact the authors have been unsuccessful. We excluded three studies (Breban 2008; Li 2008, Mulleman 2011) because there is a separate review addressing the biologics in AS (Zochling 2005). Breban 2008 is a multicenter RCT comparing continuous infliximab group with on-demand infliximab group. The on-demand group patients were further randomized into infliximab monotherapy group and infliximab plus MTX group. Li 2008 is a single center RCT of 38 AS patients, comparing infliximab plus MTX with infliximab plus placebo. Mulleman 2011 is a two-center, open-label, prospective randomized study comparing treatment with infliximab alone and infliximab with MTX.

Risk of bias in included studies

The risk of bias of included trials varied greatly and we have summarized this in Figure 2 and Figure 3.

Figure 2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

The Altan 2001 trial was reported as a RCT, but details of the methods of randomization and concealment were not given. Indeed, since one group of patients received naproxen plus MTX and the other received naproxen alone, it is doubtful whether allocation concealment was adequate.The Roychowdhury 2002 trial was described as a RCT but the authors did not report the methods of randomization and concealment. Gonzalez-Lopez 2004 was reported as a double blind RCT. The randomization was centralized by the coordinating center and was communicated to other centers by fax. Concealment seemed to be adequate.

Blinding

All three studies were blinded. In Altan 2001, the report stated that the outcome assessment was blinded. However, we are not sure whether the patients and care providers were also blinded. Roychowdhury 2002 was described as double-blind but it was unclear who were blinded. In Gonzalez-Lopez 2004 trial, patients and researchers were blinded.

Incomplete outcome data

For Altan 2001, we contacted the trial report author who confirmed that all participants completed the trial and results included all participants. In Roychowdhury 2002, two patients were lost to follow-up but it was unclear to which intervention group they belonged. We attempted to clarify this area of uncertainty by contacting the authors, but to date we have been unsuccessful. The outcomes were presented as individual patient data from which we calculated the summary values. In Gonzalez-Lopez 2004 trial, one patient dropped out from the MTX and the placebo group. The reasons given were lack of compliance and lack of response, respectively. For those patients who missed the follow-up, the value was deduced from the previous visit. ITT analysis was used for the primary outcome (response in the composite index).

Selective reporting

The protocols were available for all the trials and reports were available for pre-specified outcomes. Therefeore, we scored all three trials at low risk of bias.

Other potential sources of bias

In the Altan 2001 trial, some baseline parameters (Table 1) were significantly different between the intervention groups, strongly indicating allocation imbalance. For example, morning stiffness was 53.75 ± 32.62 units for the MTX group and 21.31 ± 17.88 units for the control group (it was unclear whether the units were measured in minutes).

Table 1. Comparison of baseline parameters between no MTX treatment and MTX treated group in Altan 2001
InterventionFunction indexPain in the dayPain in the nightSpondylitis indexEnthesis indexPatient assessmentPhysician assessmentMorning stiffness
MTX29.12 ± 25.183.93 ± 2.065.28 ± -2.383.31 ± 3.217.21 ± 7.462.65 ± 1.003.09 ± 0.8953.75 ± 32.62
No MTX20.89 ± 18.104.00 ± 2.804.10 ± 3.071.00 ± 1.913.78 ± 5.022.73 ± 0.992.21 ± 0.9721.31 ± 17.88

In the Gonzalez-Lopez 2004 study, baseline values for the MTX group were higher than that for the placebo group in four outcomes: morning stiffness VAS (100 mm, where 0 = no stiffness and 100 = severe stiffness: 47 ± 30 versus 30 ± 28; comparison of mean and standard deviation for each group); BASDAI (on a scale from 0 to 10, where 0 = no disease activity and 10 = severe disease activity: 4.7 ± 3 versus 3.5 ± 2); BASFI (on a scale from 0 to 10, where 0 = excellent function and 10 = poor function: 4.7 ± 3 versus 3.5 ± 2); and physician global assessment VAS (100 mm, where 0 = no disease activity and 100 = severe disease activity: 49 ± 30 versus 30 ± 20), although the differences were not statistically significant (P > 0.05).

Effects of interventions

See: Summary of findings for the main comparison MTX treated compared to no MTX treated for AS

Major outcomes

The included trials did not report ASAS 40, partial remission, spine x-ray, and MRI results. BASFI was reported in one trial (Gonzalez-Lopez 2004), showing no significant difference between MTX and placebo group (Analysis 1.3).

For safety outcomes, higher rates of side effect (Analysis 1.20) and withdrawal for any reason (Analysis 1.21) were found in the MTX treated than no MTX treatment group although the differences were not statistically significant. However, serious side effects were rare.

Minor outcomes

The included trials assessed different outcomes including physical function, disease activity, pain, spinal mobility, peripheral joints/entheses, patient global assessment, physician global assessment, morning stiffness, ESR, CRP, response rate, and adverse effects.

We pooled estimates for the outcomes of BASDAI, patient global assessment, physician global assessment and CRP. Since patient global assessment and physician global assessment outcomes were reported using different scales (score 1 to 5 in the Altan 2001 trial and VAS scale in Gonzalez-Lopez 2004) and the CRP unit used differed (percent change in Altan 2001 and mg/L in Roychowdhury 2002), we used the SMD values for treatment estimates. None of these outcomes showed a statistically significant difference between the intervention groups (Analysis 1.2; Analysis 1.12; Analysis 1.13; Analysis 1.18).

In the Altan 2001 trial, the authors looked at changes from baseline values. When they compared the outcomes, they found that only the physician global assessment outcome showed a statistically significant difference between the two groups, with the group receiving MTX favoured over the group that received no MTX. In our analysis, we compared this outcome (end point) and failed to confirm this difference (Analysis 1.13). Rather, we found statistically significant differences between intervention groups in function index, spondylitis index and enthesis index, this time favouring the group that received no MTX over the MTX group. For the function index (0 to 40 scale, where 0 was the best and 40 the worst), the mean difference was 12.65, 95% confidence interval (CI) 0.73 to 24.57 (Analysis 1.4). The mean difference for the spondylitis index (0 to10 scale, where the higher the score, the more severe the disease) was 2.25 (95% CI 0.94 to 3.96) (Analysis 1.9). The mean difference for the enthesis index (0 to 90 scale, where the higher the score, the more severe the disease) was 2.18 (95% CI 0.01 to 4.35) (Analysis 1.10). Eight patients in the MTX group and none in the placebo group complained of temporary nausea (RR 16.37, 95% CI 0.99 to 269.44) (Analysis 1.20).

In Roychowdhury 2002, the change from baseline of Bath ankylosing spondylitis disease activity index (BASDAI) VAS-10 cm, Bath ankylosing spondylitis metrology index (BASMI, score from 0 to 10) and CRP (mg/L) were assessed. No statistically significant differences were found between the MTX group and the placebo group (Analysis 1.2; Analysis 1.8; Analysis 1.18). One patient withdrew because of nausea, rhinorrhoea and headache but it was unclear to which intervention group this patient belonged.

In Gonzalez-Lopez 2004 trial, the primary outcome was response, which was defined as improvement greater than or equal to 20% in at least five of following areas with no worsening in any of the scales (>20% worsening compared to baseline):

  • Severity of morning stiffness VAS;

  • Physical well-being VAS;

  • BASDAI;

  • Bath ankylosing spondylitis functional index (BASFI);

  • Health assessment questionnaire for spondyloarthropathies (HAQ-S);

  • Physician global assessment VAS;

  • Patient global assessment VAS.

According to ITT analysis, a statistically significant difference was found between the intervention groups, favouring the group on MTX over the placebo group (RR = 3.18, 95% CI 1.03 to 9.79; Analysis 1.1) and NNT was 3 (Table 2). However, no statistically significant difference was found in all the secondary outcomes, including BASDAI (Analysis 1.2), BASFI (Analysis 1.3), spinal pain VAS (Analysis 1.7), disappearance of peripheral arthritis (Analysis 1.11), patient global assessment (Analysis 1.12), and physician global assessment (Analysis 1.13), physical well-being (Analysis 1.14), and health assessment questionnaire for spondyloarthropathies (HAQ-S) (Analysis 1.17). Rates of side effects were similar between the MTX group and placebo group and these included: mild headache, oral ulcerations, upper abdominal pain, nausea, vomiting, non-severe and transitory diarrhoea, transitory liver function abnormalities, loss of hair, and mild infections. Two patients withdrew from the trial, one from the MTX group for lack of compliance and the other from the placebo group because of lack of response (Gonzalez-Lopez 2004).

Table 2. Clinical relevance table - dichotomous
  1. 1 Baseline imbalance.

    2 Few patients and few events.

Outcome# Patients (# Trials)Control event rateAbsolute RDRelative % changeNNTStatistical significanceQuality of evidence
Response to composite index at 24 weeks35 (1)16.7% (3/18)36%218% (I)3statistically significantlow 1,2

Response to composite index: defined by improvement ≥ 20% at

least 5 of following and no worsening in any of the scales (>20% worsening compared to baseline): severity of morning stiffness;

physical well-being; BASDAI; BASFI; HAQ-S; physician global assessment; and patient global assessment.
MTX = methotrexate; RR = relative risk, RD = risk difference, NNT = number needed to treat

   I = improvement   

Discussion

Summary of main results

In the present review, we identified three RCTs evaluating the benefits and harms of MTX in AS. Two trials (Roychowdhury 2002; Gonzalez-Lopez 2004) compared MTX with placebo and one (Altan 2001) compared MTX with no medication. These three trials treated a total of 116 participants. All of the included particpants were considered to have active disease, although the definition of active disease differed between the trials. Among the outcomes reported in these trials, we detected no statistically significant differences favouring MTX over no MTX group from our analyses. The only exception was for the outcome of response rates in the Gonzalez-Lopez 2004 trial, which favoured MTX over placebo. This response was judged by 'change from baseline' results of seven outcomes, i.e. morning stiffness VAS, physical well-being VAS, BASDAI, BASFI, HAQ-S, physician global assessment VAS, and patient global assessment VAS. However, none of these seven outcomes showed statistically significant differences between the MTX group and placebo group when endpoint results were compared. Therefore, the benefit of MTX for treatment of AS is questionable. We investigated the Gonzalez-Lopez 2004 trial further and found that different baseline values in some of these seven outcomes could partially explain this paradox. The baseline values of the MTX group were higher than those of the placebo group for four outcomes: morning stiffness, BASDAI, BASFI, and physician global assessment, as described in the Risk of bias in included studies section. We suppose that for patients with more severe disease it was easier for their condition to improve more after treatment when compared withpatients who were in comparatively better health before treatment who experienced less improvement after treatment.

Although there were higher incidences of side effects in the MTX treated than in the no MTX treatment group, MTX was well tolerated and serious side effects were rare.

Overall completeness and applicability of evidence

The study population from the included trials was 116 adult AS patients who fulfilled New York criteria (Olivieri 2002), including obvious sacroiliitis on X-ray. The majority of these patients were at an advanced stage of AS, with mean disease duration ranging from 5.7 to 17 years. Therefore, the evidence was not applicalbe to juvenile patients and those patients with early disease, especially those without radiological sacroiliitis.

Quality of the evidence

The evidence of the present review is not robust enough to draw a conclusion about the benefit of MTX for treatment of AS, because it included only three trials with a total of 116 patients and two of the included trials (Altan 2001; Roychowdhury 2002) had obvious methodological drawbacks. Statistically significant results could have been missed because of small sample sizes included in the present review. Dosage of MTX is another consideration: these trials used doses of 7.5 to 15 mg/week whereas the maximum dosage recommended for RA is 25 to 30 mg/week (ACR 2002).

Potential biases in the review process

We conducted an extensive search of several databases and updated the search during the review process.  We included all trials that met with our inclusion criteria. We did not limit the search to a specific language. When necessary, we attempted to contact the trial authors to clarify any information regarding their trials. Presentation of outcomes was poor in Altan 2001: both end point and change from baseline values (percent change or difference scores) were reported. However, it was unclear whether the figures given referred to 'mean ± SD' or 'mean ± standard error (SE)' data, and whether results for outcomes were 'percent change' or 'difference scores'. As a result, caution should be used in interpretation of some outcomes: for example, the figures for morning stiffness were -52.60 ± 0.70 (percent change from baseline, according to the author) for MTX and -35.09 ± 1.01 for the group without MTX, where SD (or SE) was very small. The author of this trial confirmed that this datum was a mistake but was unable to provide the correct value.

Agreements and disagreements with other studies or reviews

The benefit of MTX has been confirmed for treatment of RA (Suarez-Almazor 2003; Katchamart 2010). However, regarding use of MTX for treatment of AS, it remains uncertain. There have been several case reports and open studies evaluating MTX for the treatment of AS (Biasi 2000; Clavaguera 1998; Creemers 1995; Handler 1989; Kirnap 2000; Ostendorf 1998; Sampaio-Barros 2000). Haibel 2007 reported an open-label trial in active AS patients with predominantly axial manifestation, using higher dose of MTX (15 mg subcutaneously for four weeks and 20 mg for 12 weeks). No change from baseline in mean BASDAI was observed at 16 weeks. This result showed that even a higher dose of MTX was ineffective for treatment of AS.

In RA, combination of Infliximab and MTX was shown to cause a synergistic effect (Maini 1998) and therefore a recommendation has been made for this combination (Smolen 2010). For treatment of AS, Perez-Guijo 2007 reported an open label and prospective study of 19 patients with active disease. The results showed that infliximab in combination with MTX increased therapeutic response. Three RCTs (Breban 2008; Li 2008; Mulleman 2011) aimed to test whether concomitant use of MTX improved the efficacy of infliximab in people with AS. However, no statistically significant difference was found between MTX and no MTX group in all the assessed outcomes, including ASAS20, ASAS40 and partial remission, indicating that adding MTX to infliximab offers no demonstrable benefit.

Authors' conclusions

Implications for practice

There is not enough evidence to support any benefit of MTX in the treatment of people with AS.

Implications for research

RCTs of higher methodological quality, i.e. with larger sample sizes and of longer duration, with adequate allocation concealment and triple blinding (patients, care providers, and investigators), are needed to verify the uncertainty about the benefit and toxicity of MTX for the treatment of people with AS.

Acknowledgements

We would like to thank the editorial team of the Cochrane Musculoskeletal Group for their helpful comments and Sally Green, Steve McDonald, Janet Piehl, Denise O'Connor, Elmer V Villanueva, and the staff of Australasian Cochrane Centre for their supervision, editorial, and technical support. Also, we would like to thank Chao Liu and Junfang Lin, the original authors for their great contributions to this review.

Data and analyses

Download statistical data

Comparison 1. MTX versus No MTX
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Response in the composite index1 Risk Ratio (M-H, Random, 95% CI)Totals not selected
2 BASDAI (VAS-10 cm, 0=no disease activity, 10=severe disease activity) (end point value)263Mean Difference (IV, Random, 95% CI)0.96 [-1.22, 3.14]
3 BASFI (score 0-10, 0=excellent, 10=poor function) (change from baseline)135Mean Difference (IV, Random, 95% CI)0.9 [-0.43, 2.23]
4 Function index (score 0-40, 0=the best, 40=the worst) (end point value)1 Mean Difference (IV, Random, 95% CI)Totals not selected
5 Pain in the night (VAS-10 cm, 0=no pain, 10=severe pain) (end point value)1 Mean Difference (IV, Random, 95% CI)Totals not selected
6 Pain in the day (VAS-10 cm, 0=no pain, 10=severe pain) (end point value)151Mean Difference (IV, Random, 95% CI)0.89 [-0.44, 2.22]
7 Spinal pain (VAS 10 cm, 0=no pain, 10=severe) (end point value)135Mean Difference (IV, Fixed, 95% CI)-0.90 [-2.64, 0.84]
8 BASMI (VAS-10 cm, 0=severely limited spinal mobility, 10=overall normal) (end point value)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
9 Spondylitis index (score 0-10, the more severe the disease, the higher the score) (end point value)1 Mean Difference (IV, Random, 95% CI)Totals not selected
10 Enthesis index (score 0-90, the more severe the disease, the higher the score) (end point value)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
11 Disappearance of peripheral arthritis (event)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
12 Patient global assessment (different scales, the more severe the disease, the higher the score) (end point value)284Std. Mean Difference (IV, Fixed, 95% CI)0.12 [-0.31, 0.55]
13 Physician global assessment (different scales, the more severe the disease, the higher the score) (end point value)286Std. Mean Difference (IV, Random, 95% CI)-0.14 [-1.03, 0.75]
14 Physical well-being (VAS 10 cm, 0=feeling very well, 10=very bad)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
15 Morning stiffness duration (units not specified in trial report, we assumed minutes)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
16 Severity of morning stiffness (VAS 10 cm, 0=no stiffness, 10=severe)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
17 Health assessment questionnaire for spondyloarthropathies (HAQ-S) (0-3, 0=no difficulty, 3=unable to do)1 Mean Difference (IV, Fixed, 95% CI)Totals not selected
18 CRP (change from baseline) (percentage for Altan, ug/ml for others)279Std. Mean Difference (IV, Fixed, 95% CI)0.04 [-0.40, 0.48]
19 ESR (percentage change from baseline)151Std. Mean Difference (IV, Fixed, 95% CI)-0.04 [-0.59, 0.51]
20 Side effect151Risk Ratio (M-H, Random, 95% CI)16.37 [0.99, 269.44]
21 Withdrawal for any reason135Risk Ratio (M-H, Random, 95% CI)1.06 [0.07, 15.62]
Analysis 1.1.

Comparison 1 MTX versus No MTX, Outcome 1 Response in the composite index.

Analysis 1.2.

Comparison 1 MTX versus No MTX, Outcome 2 BASDAI (VAS-10 cm, 0=no disease activity, 10=severe disease activity) (end point value).

Analysis 1.3.

Comparison 1 MTX versus No MTX, Outcome 3 BASFI (score 0-10, 0=excellent, 10=poor function) (change from baseline).

Analysis 1.4.

Comparison 1 MTX versus No MTX, Outcome 4 Function index (score 0-40, 0=the best, 40=the worst) (end point value).

Analysis 1.5.

Comparison 1 MTX versus No MTX, Outcome 5 Pain in the night (VAS-10 cm, 0=no pain, 10=severe pain) (end point value).

Analysis 1.6.

Comparison 1 MTX versus No MTX, Outcome 6 Pain in the day (VAS-10 cm, 0=no pain, 10=severe pain) (end point value).

Analysis 1.7.

Comparison 1 MTX versus No MTX, Outcome 7 Spinal pain (VAS 10 cm, 0=no pain, 10=severe) (end point value).

Analysis 1.8.

Comparison 1 MTX versus No MTX, Outcome 8 BASMI (VAS-10 cm, 0=severely limited spinal mobility, 10=overall normal) (end point value).

Analysis 1.9.

Comparison 1 MTX versus No MTX, Outcome 9 Spondylitis index (score 0-10, the more severe the disease, the higher the score) (end point value).

Analysis 1.10.

Comparison 1 MTX versus No MTX, Outcome 10 Enthesis index (score 0-90, the more severe the disease, the higher the score) (end point value).

Analysis 1.11.

Comparison 1 MTX versus No MTX, Outcome 11 Disappearance of peripheral arthritis (event).

Analysis 1.12.

Comparison 1 MTX versus No MTX, Outcome 12 Patient global assessment (different scales, the more severe the disease, the higher the score) (end point value).

Analysis 1.13.

Comparison 1 MTX versus No MTX, Outcome 13 Physician global assessment (different scales, the more severe the disease, the higher the score) (end point value).

Analysis 1.14.

Comparison 1 MTX versus No MTX, Outcome 14 Physical well-being (VAS 10 cm, 0=feeling very well, 10=very bad).

Analysis 1.15.

Comparison 1 MTX versus No MTX, Outcome 15 Morning stiffness duration (units not specified in trial report, we assumed minutes).

Analysis 1.16.

Comparison 1 MTX versus No MTX, Outcome 16 Severity of morning stiffness (VAS 10 cm, 0=no stiffness, 10=severe).

Analysis 1.17.

Comparison 1 MTX versus No MTX, Outcome 17 Health assessment questionnaire for spondyloarthropathies (HAQ-S) (0-3, 0=no difficulty, 3=unable to do).

Analysis 1.18.

Comparison 1 MTX versus No MTX, Outcome 18 CRP (change from baseline) (percentage for Altan, ug/ml for others).

Analysis 1.19.

Comparison 1 MTX versus No MTX, Outcome 19 ESR (percentage change from baseline).

Analysis 1.20.

Comparison 1 MTX versus No MTX, Outcome 20 Side effect.

Analysis 1.21.

Comparison 1 MTX versus No MTX, Outcome 21 Withdrawal for any reason.

Appendices

Appendix 1. Search strategies for the original review

  
MEDLINE (Ovid):
1 Spondylitis, Ankylosing/
2 (bechtere$ disease$ or marie-struempell disease$ or rheumatoid spondylitis or spondylarthritis ankylopoietica or ankylo$ spondyl$ or Spin$ Ankylosis or Vertebral Ankylosis).tw.
3 1 or 2
4 methotrexate/
5 (amethopterin$ or mexate or Abitrexate or Ametopterine or Emt#exate or Methylaminopterin$ or Novatrex or Rheumatrex or methotrexate).tw.
6 methotrexate.rn.
7 or/4-6
8 clinical trial.pt.
9 randomized controlled trial.pt.
10 tu.fs.
11 dt.fs.
12 random$.tw.
13 (double adj blind$).tw.
14 placebo$.tw.
15 or/8-14
16 3 and 7
17 15 and 16
18 from 16 keep 1-62 (62)
EMBASE (Ovid):
1 Spondylitis, Ankylosing/
2 (bechtere$ disease$ or marie-struempell disease$ or rheumatoid spondylitis or spondylarthritis ankylopoietica or ankylo$ spondyl$ or Spin$ Ankylosis or Vertebral Ankylosis).tw.
3 1 or 2
4 methotrexate/
5 (amethopterin$ or mexate or Abitrexate or Ametopterine or Emt#exate or Methylaminopterin$ or Novatrex or Rheumatrex or methotrexate).tw.
6 4 or 5
7 3 and 6
8 from 7 keep 1-197

Appendix 2. Search strategy for the second version

MEDLINE

1. Spondylitis, Ankylosing/
2. (bechtere$ disease$ or marie-struempell disease$ or rheumatoid spondylitis or spondylarthritis ankylopoietica or ankylo$ spondyl$ or Spin$ Ankylosis or Vertebral Ankylosis).tw.
3. 1 or 2
4. methotrexate/
5. (amethopterin$ or mexate or Abitrexate or Ametopterine or Emt#exate or Methylaminopterin$ or Novatrex or Rheumatrex or methotrexate).tw.
6. methotrexate.rn.
7. or/4-6
8. randomized controlled trial.pt.
9. controlled clinical trial.pt.
10. randomized controlled trials.sh.
11. random allocation.sh.
12. double blind method.sh.
13. single-blind method.sh.
14. or/8-13
15. (animal$ not human).sh.
16. 14 not 15
17. clinical trial.pt.
18. exp clinical trials/
19. (clin$ adj25 trial$).ti,ab.
20. ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).tw.
21. placebo$.sh.
22. placebo$.ti,ab.
23. random$.ti,ab.
24. research design.sh.
25. or/18-24
26. 25 not 15
27. 26 not 16
28. comparative study.sh.
29. exp evaluation studies/
30. follow up studies.sh.
31. prospective studies.sh.
32. (control$ or prospectiv$ or volunteer$).ti,ab.
33. or/28-32
34. 33 not 15
35. 34 not (16 or 27)
36. and/3,7,35

EMBASE

1. Spondylitis, Ankylosing/
2. (bechtere$ disease$ or marie-struempell disease$ or rheumatoid spondylitis or spondylarthritis ankylopoietica or ankylo$ spondyl$ or Spin$ Ankylosis or Vertebral Ankylosis).tw.
3. 1 or 2
4. methotrexate/
5. (amethopterin$ or mexate or Abitrexate or Ametopterine or Emt#exate or Methylaminopterin$ or Novatrex or Rheumatrex or methotrexate).tw.
6. 4 or 5
7. 3 and 6
8. random$.ti,ab.
9. factorial$.ti,ab.
10. (crossover$ or cross over$ or cross-over$).ti,ab.
11. placebo$.ti,ab.
12. (doubl$ adj blind$).ti,ab.
13. (singl$ adj blind$).ti,ab.
14. assign$.ti,ab.
15. allocat$.ti,ab.
16. volunteer$.ti,ab.
17. crossover procedure.sh.
18. double blind procedure.sh.
19. randomized controlled trial.sh.
20. single blind procedure.sh.
21. or/8-20
22. exp animal/ or nonhuman/ or exp animal experiment/
23. exp human/
24. 22 and 23
25. 22 not 24
26. 21 not 25
27. 7 and 26

Appendix 3. Search strategies for the third (current) version review

Ovid MEDLINE

1. Spondylitis, Ankylosing/

2. (ankylos$ or spondyl$).tw.

3. (bekhterev$ or bechterew$).tw.

4. (Marie adj struempell$).tw.

5. or/1-4

6. Methotrexate/

7. Methotrexate.tw.

8. amet?opterine.tw.

9. mexate.tw.

10. Abitrexate.tw.

11. A Met?opterine$.tw.

12. Antifolan.tw.

13. Emt?exate.tw.

14. Enthexate.tw.

15. Farmitrexate.tw.

16. Folex.tw.

17. Ledertrexate.tw.

18. Methoblastin.tw.

19. Methohexate.tw.

20. Methotrate.tw.

21. Methylaminopterin.tw.

22. Metotrexat$.tw.

23. Mtx.tw.

24. Novatrex.tw.

25. Rheumatrex.tw.

26. or/6-25

27. randomized controlled trial.pt.

28. controlled clinical trial.pt.

29. randomized.ab.

30. placebo.ab.

31. drug therapy.fs.

32. randomly.ab.

33. trial.ab.

34. groups.ab.

35. or/27-34

36. (animals not (humans and animals)).sh.

37. 35 not 36

38. and/5,26,37

 

Ovid EMBASE

1. ankylosing spondylitis/

2. (ankylos$ or spondyl$).tw.

3. (bekhterev$ or bechterew$).tw.

4. (Marie adj struempell$).tw.

5. or/1-4

6. methotrexate/

7. Methotrexate.tw.

8. mexate.tw.

9. Abitrexate.tw.

10. Amet?opterine.tw.

11. mexate.tw.

12. Abitrexate.tw.

13. A Met?opterine.tw.

14. Antifolan.tw.

15. Emt?exate.tw.

16. Enthexate.tw.

17. Farmitrexate.tw.

18. Folex.tw.

19. Ledertrexate.tw.

20. Methoblastin.tw.

21. Methohexate.tw.

22. Methotrate.tw.

23. Methylaminopterin.tw.

24. Metotrexat$.tw.

25. Mtx.tw.

26. Novatrex.tw.

27. Rheumatrex.tw.

28. or/6-27

29. (random$ or placebo$).ti,ab.

30. ((single$ or double$ or triple$ or treble$) and (blind$ or mask$)).ti,ab.

31. controlled clinical trial$.ti,ab.

32. RETRACTED ARTICLE/

33. or/29-32

34. (animal$ not human$).sh,hw.

35. 33 not 34

36. 28 and 35

37. and/5,28,36

 

The Cochrane Library via Ovid EBM Reviews

1. (ankylos$ or spondyl$).tw.

2. (bekhterev$ or bechterew$).tw.

3. (Marie adj struempell$).tw.

4. or/1-3

5. Methotrexate.tw.

6. amet?opterine.tw.

7. mexate.tw.

8. Abitrexate.tw.

9. A Met?opterine$.tw.

10. Antifolan.tw.

11. Emt?exate.tw.

12. Enthexate.tw.

13. Farmitrexate.tw.

14. Folex.tw.

15. Ledertrexate.tw.

16. Methoblastin.tw.

17. Methohexate.tw.

18. Methotrate.tw.

19. Methylaminopterin.tw.

20. Metotrexat$.tw.

21. Mtx.tw.

22. Novatrex.tw.

23. Rheumatrex.tw.

24. or/5-23

25. 4 and 24

 

CINAHL

S29 S7 and S28

S28 S8 or S9 or S10 or S11 or S12 or S13 or S14 or S15 or S16 or S17 or S18 or S19 or S20 or S21 or S22 or S23 or S24 or S25 or S26 or S27

S27 TI Rheumatrex OR AB Rheumatrex

S26 TI Novatrex OR AB Novatrex

S25 TI Mtx OR AB Mtx

S24 TI Metotrexat* OR AB Metotrexat*

S23 TI Methylaminopterin OR AB Methylaminopterin

S22 TI Methotrate OR AB Methotrate

S21 TI Methohexate OR AB Methohexate

S20 TI Methoblastin OR AB Methoblastin

S19 TI Ledertrexate OR AB Ledertrexate

S18 TI Folex OR AB Folex

S17 TI Farmitrexate OR AB Farmitrexate

S16 TI Enthexate OR AB Enthexate

S15 TI Emt?exate OR AB Emt?exate

S14 TI Antifolan OR AB Antifolan

S13 TI A Met?opterine OR AB A Met?opterine

S12 TI Abitrexate OR AB Abitrexate

S11 TI mexate OR AB mexate
S10 TI amet?opterine OR AB amet?opterine

S9 TI Methotrexate OR AB

S8 (MH "Methotrexate")

S7 S1 or S2 or S3 or S4 or S5 or S6

S6 TI Marie struempell* OR AB Marie struempell*

S5 TI bechterew* OR AB bechterew*

S4 TI bekhterev* OR AB bekhterev*

S3 TI spondyl* OR AB spondyl*

S2 TI ankylos* OR AB ankylos*

S1 (MH "Spondylitis, Ankylosing")

  

Scopus

#1 (TITLE-ABS-KEY(ankylos*) OR TITLE-ABS-KEY(spondyl*) OR TITLE-ABS-KEY(bekhterev*) OR TITLE-ABS-KEY(bechterew*) OR TITLE-ABS-KEY(marie struempell*))

#2 (TITLE-ABS-KEY(methotrexate) OR TITLE-ABS-KEY(amet?opterine) OR TITLE-ABS-KEY(mexate) OR TITLE-ABS-KEY(abitrexate) OR TITLE-ABS-KEY("A Met?opterine") OR TITLE-ABS-KEY(antifolan) OR TITLE-ABS-KEY(emt?exate) OR TITLE-ABS-KEY(enthexate) OR TITLE-ABS-KEY(farmitrexate) OR TITLE-ABS-KEY(folex) OR TITLE-ABS-KEY(ledertrexate) OR TITLE-ABS-KEY(methoblastin) OR TITLE-ABS-KEY(methohexate) OR TITLE-ABS-KEY(methotrate) OR TITLE-ABS-KEY(methylaminopterin) OR TITLE-ABS-KEY(metotrexat*) OR TITLE-ABS-KEY(mtx) OR TITLE-ABS-KEY(novatrex) OR TITLE-ABS-KEY(rheumatrex))

#3 #1 AND #2

#4 Limit to Conference Paper

 

World Health Organization International Clinical Trials Registry Platform

ankylos* OR spondyl* OR bekhterev* OR bechterew* OR marie struempell* in Condition AND methotrexate in Intervention

Appendix 4. Search result for the third version

Methotrexate for Ankylosing Spondylitis Update January 2011

All databases searched January 11, 2011

 

DatabaseDates searchedNumber of references retrievedNumber after duplicates removed

MEDLINE (Ovid)

and

 

Ovid MEDLINE(R) In-Process & Other Non-Indexed Citations

2005 to December Week 4 2010

January 10, 2011

129129
Embase (Ovid)2005 to 2011 Week 0111280

CENTRAL, part of The Cochrane Library (via Ovid EBM Reviews)

 

DARE, part of The Cochrane Library (via Ovid EBM Reviews)

 

HTA Database, part of The Cochrane Library (via Ovid EBM Reviews)

 

NHS EED, part of The Cochrane Library (via Ovid EBM Reviews)

 

4th Quarter 2010

 

4th Quarter 2010

 

1st Quarter 2011

 

 

1st Quarter 2011

6

 

 

5

 

 

0

 

 

0

0

 

 

4

 

 

0

 

 

0

CINAHL (EBSCOHost)2005 to January 20112811

Scopus

 

2005 to January 11, 20114744

World Health Organization International Clinical Trials Registry Platform

http://www.who.int/ictrp/en/

Searched January 11, 201100
All databases Totals327268

 Search:  Methotrexate for Ankylosing Spondylitis – May 2012 Update

 

Database and coverageSearch Date#Results

Results with

duplicates removed

The Cochrane Library

 

Issue 6 of 12, June

2012

www.cochranelibrary.com

Cochrane Controlled Trials Register

(CENTRAL)

June 25,

2012

33
OVID MEDLINE, Dec 2011-

June 25,

2012

2619
Ebsco CINAHL, Dec 2011-

June 25,

2012

87
OVID EMBASE  Dec 2011-

June 25,

2012

1715
Scopus Dec 2011-

June 25,

2012

1111
WHO ICTRP  2011-

June 25,

2012

11
 All databasesTotals6656

Appendix 5. Risk of bias assessment criteria

SEQUENCE GENERATION

Was the allocation sequence adequately generated? [Short form: Adequate sequence generation?]

Criteria for a judgement of 'YES' (i.e. low risk of bias).

The investigators describe a random component in the sequence generation process such as:

  • Referring to a random number table;

  • Using a computer random number generator;

  • Coin tossing;

  • Shuffling cards or envelopes;

  • Throwing dice;

  • Drawing of lots;

  • Minimization*.

*Minimization may be implemented without a random element, and this is considered to be equivalent to being random.

Criteria for the judgement of 'NO' (i.e. high risk of bias).

The investigators describe a non-random component in the sequence generation process. Usually, the description would involve some systematic, non-random approach, for example:

  • Sequence generated by odd or even date of birth;

  • Sequence generated by some rule based on date (or day) of admission;

  • Sequence generated by some rule based on hospital or clinic record number.

 

Other non-random approaches happen much less frequently than the systematic approaches mentioned above and tend to be obvious. They usually involve judgement or some method of non-random categorization of participants, for example:

  • Allocation by judgement of the clinician;

  • Allocation by preference of the participant;

  • Allocation based on the results of a laboratory test or a series of tests;

  • Allocation by availability of the intervention.

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

 

ALLOCATION CONCEALMENT

Was allocation adequately concealed? [Short form: Allocation concealment?]

Criteria for a judgement of 'YES' (i.e. low risk of bias).

Participants and investigators enrolling participants could not foresee assignment because of one of the following, or an equivalent method, was used to conceal allocation:

  • Central allocation (including telephone, web-based, and pharmacy-controlled, randomization);

  • Sequentially numbered drug containers of identical appearance;

  • Sequentially numbered, opaque, sealed envelopes.

 

Criteria for the judgement of 'NO' (i.e. high risk of bias).

Participants or investigators enrolling participants could possibly foresee assignments and thus introduce selection bias, such as allocation based on:

  • Using an open random allocation schedule (e.g. a list of random numbers);

  • Assignment envelopes were used without appropriate safeguards (e.g. if envelopes were unsealed or non-opaque or not sequentially numbered);

  • Alternation or rotation;

  • Date of birth;

  • Case record number;

  • Any other explicitly unconcealed procedure.

 

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

 

BLINDING OF PARTICIPANTS, PERSONNEL AND OUTCOME ASSESSORS

Was knowledge of the allocated interventions adequately prevented during the study? [Short form: Blinding?]

Criteria for a judgement of 'YES' (i.e. low risk of bias).

Any one of the following:

  • No blinding, but the review authors judge that the outcome and the outcome measurement are not likely to be influenced by lack of blinding;

  • Blinding of participants and key study personnel ensured, and unlikely that the blinding could have been broken;

  • Either participants or some key study personnel were not blinded, but outcome assessment was blinded and the non-blinding of others unlikely to introduce bias.

 

Criteria for the judgement of 'NO' (i.e. high risk of bias).

Any one of the following:

  • No blinding or incomplete blinding, and the outcome or outcome measurement is likely to be influenced by lack of blinding;

  • Blinding of key study participants and personnel attempted, but likely that the blinding could have been broken;

  • Either participants or some key study personnel were not blinded, and the non-blinding of others likely to introduce bias.

 

Criteria for the judgement of 'UNCLEAR' (uncertain risk of bias).

Any one of the following:

  • Insufficient information to permit judgement of 'Yes' or 'No';

  • The study did not address this outcome.

 

 

INCOMPLETE OUTCOME DATA

Were incomplete outcome data adequately addressed? [Short form: Incomplete outcome data addressed?]

Criteria for a judgement of 'YES'

(i.e. low risk of bias).

Any one of the following:

  • No missing outcome data;

  • Reasons for missing outcome data unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias);

  • Missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups;

  • For dichotomous outcome data, the proportion of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate;

  • For continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes not enough to have a clinically relevant impact on observed effect size;

  • Missing data have been imputed using appropriate methods.

 

Criteria for the judgement of 'NO' (i.e. high risk of bias).

Any one of the following:

  • Reason for missing outcome data likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups;

  • For dichotomous outcome data, the proportion of missing outcomes compared with observed event risk enough to induce clinically relevant bias in intervention effect estimate;

  • For continuous outcome data, plausible effect size (difference in means or standardized difference in means) among missing outcomes enough to induce clinically relevant bias in observed effect size;

  • 'As-treated' analysis done with substantial departure of the intervention received from that assigned at randomization;

  • Potentially inappropriate application of simple imputation.

 

Criteria for the judgement of 'UNCLEAR' (uncertain risk of bias).

Any one of the following:

  • Insufficient reporting of attrition/exclusions to permit judgement of 'Yes' or 'No' (e.g. number randomized not stated, no reasons for missing data provided);

  • The study did not address this outcome.

 

SELECTIVE OUTCOME REPORTING

Are reports of the study free of suggestion of selective outcome reporting? [Short form: Free of selective reporting?]

Criteria for a judgement of 'YES' (i.e. low risk of bias).

Any of the following:

  • The study protocol is available and all of the study's pre-specified (primary and secondary) outcomes that are of interest in the review have been reported in the pre-specified way;

  • The study protocol is not available but it is clear that the published reports include all expected outcomes, including those that were pre-specified (convincing text of this nature may be uncommon).

 

Criteria for the judgement of 'NO' (i.e. high risk of bias).

Any one of the following:

  • Not all of the study's pre-specified primary outcomes have been reported;

  • One or more primary outcomes is reported using measurements, analysis methods or subsets of the data (e.g. subscales) that were not pre-specified;

  • One or more reported primary outcomes were not pre-specified (unless clear justification for their reporting is provided, such as an unexpected adverse effect);

  • One or more outcomes of interest in the review are reported incompletely so that they cannot be entered in a meta-analysis;

  • The study report fails to include results for a key outcome that would be expected to have been reported for such a study.

 

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

OTHER POTENTIAL THREATS TO VALIDITY

Was the study apparently free of other problems that could put it at a risk of bias? [Short form: Free of other bias?]

Criteria for a judgement of 'YES' (i.e. low risk of bias).The study appears to be free of other sources of bias.
Criteria for the judgement of 'NO' (i.e. high risk of bias).

There is at least one important risk of bias. For example, the study:

  • Had a potential source of bias related to the specific study design used; or

  • Stopped early due to some data-dependent process (including a formal-stopping rule); or

  • Had extreme baseline imbalance; or

  • Has been claimed to have been fraudulent; or

  • Had some other problem.

 

Criteria for the judgement of 'UNCLEAR' (uncertain risk of bias).Search

There may be a risk of bias, but there is either:

  • Insufficient information to assess whether an important risk of bias exists; or

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

  

Appendix 6. Sample data collection form for assessment of risk of bias

DOMAINJUDGEMENT (YES - low risk; NO - high risk; or UNCLEAR)REASON FOR JUDGEMENT (copy and paste directly from text of trial)
Was the allocation sequence adequately generated?  
Was allocation adequately concealed?  
Was knowledge of the allocated interventions adequately prevented during the study?  
Blinding of personnel  
Blinding of participants  
Blinding of outcome assessors  
Were incomplete outcome data adequately addressed?  
Are reports of the study free of suggestion of selective outcome reporting  
Was the study apparently free of other problems that could put it at a risk of bias?  

What's new

DateEventDescription
4 July 2011New citation required but conclusions have not changedChange in authorship.
4 July 2011New search has been performedNew search but no new studies included.

History

Protocol first published: Issue 3, 2003
Review first published: Issue 3, 2004

DateEventDescription
5 November 2008New search has been performed

Converted to new review format.

CMSG ID: C050-R

Contributions of authors

JC registered the title, developed the protocol, searched for relevant studies, selected the studies and assessed their methodological quality, extracted and synthesized the data, and wrote the systematic review.

MMSV (for the update), JFL (for the second version), and CLIO (for the original review) selected studies, assessed their methodological quality, and extracted data from them independently of JC.

Declarations of interest

There are no known potential conflicts of interest for this review.

Sources of support

Internal sources

  • The First Affiliated Hospital of Fujian Medical University, China.

External sources

  • Science development fund, Fujian Medical University, China.

Differences between protocol and review

In this updated review, we used the Cochrane Collaboration's tool (Higgins 2011) for assessing risk of bias. We utilised Grade Profiler software and a summary of findings table to present the main results.

Characteristics of studies

Characteristics of included studies [ordered by year of study]

Altan 2001

MethodsRandomized allocation: randomization method not given.
Outcome assessment blinded.
Treatment allocation possibly open to patients and care providers.
Parallel groups.
Duration: 12 months.
Sample size at entry: 51.
MTX and naproxen group: 26.
Naproxen group: 25.
Personal communication with the trial report author confirmed that all participants completed the trial, and results include all the participants.
ParticipantsPatients with AS fulfilling New York criteria.
Other inclusion criteria:
morning stiffness > 15 min; sleep disturbance caused by pain; limitation of daily functional activity; CRP > 0.5 mg/dl.
Exclusion criteria:
total ankylosis involving the whole spine; failure of hepatic and renal tests; severe anaemia, leucopenia or thrombocytopenia.
Age: 46.78 years ± 11.76 (mean ± SD).
M:F ratio: 76%: 24%.
Mean disease duration: 9.52 (4 to 30) years.
HLA B27 positive: 88.2%.
With peripheral arthritis: 12%.
InterventionsMTX + naproxen group: MTX orally, 7.5 mg/week, plus naproxen 1000 mg/day.
Naproxen group: naproxen 1000 mg/day.
Outcomes1. Morning stiffness (units of scale not specified, we assumed the scale was in minutes).
2. Pain in the day (VAS 0 to 10 cm, where 0 = no pain and 10 = severe pain).
3. Pain in the night (VAS 0 to 10 cm, where 0 = no pain and 10 = severe pain).
4. Patient global assessment (scale of 1 to 5, where 1 = no disease activity and 5 = severe disease activity).
5. Physician global assessment (scale of 1 to 5, where 1 = no disease activity and 5 = severe disease activity).
6. Enthesis index (0 to 90, where the more severe the disease, the higher the score).
7. Function index (score 0 to 40, where 0 = the best and 40 = the worst).
8. Spondylitis index (score 0 to 10, where the more severe the disease, the higher the score).
9. ESR (percent change from baseline, according to information from author).
10. CRP (change from baseline, according to information from author).
NotesOutcomes were poorly presented, for example:
Morning stiffness was -35.09 ± 1.01 (naproxen alone) and -52.60 ± 0.70 (naproxen plus MTX), respectively. The unit was unknown and likely to be minute. We are not sure whether this figure was mean ± SD or mean ± SE. Nor are we sure whether this figure was score or percent change from baseline. According to author's correspondence, data were presented as percent change from baseline. After discussion, we entered end point mean ± SD (we assumed as SD) values.
Some baseline parameters were significantly different between intervention groups (summarised in Table 1).
Our attempts to contact the authors to clarify the uncertainties have been unsuccessful so far.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information about the sequence generation process to permit judgment of low risk or high risk.
Allocation concealment (selection bias)Unclear riskInsufficient information about the sequence generation process to permit judgment of low risk or high risk.
Blinding (performance bias and detection bias)
All outcomes
High riskOnly outcome assessment was reported to be blind. Because they did not use placebo, blinding of personnel and participants was impossible.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing outcome data.
Selective reporting (reporting bias)Low riskThe study protocol is available and all of the study's pre-specified (primary and secondary) outcomes that are of interest in the review have been reported in the pre-specified way.
Other biasHigh riskSome baseline parameters were significantly different between intervention groups. Furthermore, presentation of outcomes was poor.

Roychowdhury 2002

MethodsRandomized allocation: randomization and concealment methods not specified.
Outcome assessment blinded.
Patients and care providers blinded.
Parallel groups.
Duration: 24 weeks.
Sample size at entry: 30.
MTX group: 12 completed;
Placebo group: 16 completed.
Two participants withdrew, but it was unclear to which intervention group(s) they belonged.
Outcomes of each patient were provided.
ParticipantsPatients with AS fulfilling modified New York criteria and active disease defined as at least two of the following symptoms despite treatment with NSAIDs and/or other DMARDs:
morning stiffness > 30 min; sleep disturbance due to pain and stiffness; peripheral arthritis; ESR > 30 mm/h or CRP > 20 mg/L; spinal pain; pain in both buttocks during the day or night.
Sulphasalazine or other DMARDs discontinued at least four weeks before study entry.
Exclusion criteria:
serious systemic disease, malignancies, impaired organ function or serious infections; pregnancy and breast feeding; mental disorders; alcohol or drug abuse; history of intestinal disease.
Age: 44 ± 10 years (mean ± SD).
M:F ratio: 87%: 13%.
Mean disease duration: 17 (3 to 26) years.
With peripheral arthritis: 30%.
InterventionsMTX group: MTX, orally, 10 mg/week.
Placebo group: placebo orally, 10 mg/week.
Outcomes1. Bath ankylosing spondylitis disease activity index (BASDAI) (change from baseline) (VAS-10cm, where 0 = no disease activity and 10 = severe disease activity).
2. Bath ankylosing spondylitis metrology (BASMI) (change from baseline) (score from 0 to 10, where 0 = severely limited spinal mobility and 10 = overall normal).
3. CRP (change from baseline) (mg/L).
Notes30 patients entered the trial and 28 completed.
It is not known to which intervention group the two drop-outs belonged.
Our attempts to contact the authors have been unsuccessful to date.
Each patient's data were given, from which we calculated the summary values.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information about the sequence generation process to permit judgement of low risk or high risk.
Allocation concealment (selection bias)Unclear riskInsufficient information about the sequence generation process to permit judgement of low risk or high risk.
Blinding (performance bias and detection bias)
All outcomes
Low riskReported as double blind, placebo-controlled.
Incomplete outcome data (attrition bias)
All outcomes
Low riskFor dichotomous outcome data, the proportion of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate.
Selective reporting (reporting bias)Low riskThe study protocol is available and all of the study's pre-specified (primary and secondary) outcomes that are of interest in the review have been reported in the pre-specified way.
Other biasLow riskThe study appears to be free of other sources of bias.

Gonzalez-Lopez 2004

MethodsRandomized allocation: computer generated randomization numbers.
Patients, care-providers and outcome investigators blinded.
Parallel groups.
Duration: 24 weeks.
Sample size at entry: 35
MTX - 17
Placebo - 18
Withdrawals and drop-outs clearly specified. Outcomes analysis: ITT.
ParticipantsPatients with AS fulfilling New York criteria.
Other inclusion criteria:
minimum duration since diagnosis one year; having active disease at the time of the baseline evaluation, defined by the combination of three conditions: (1) physician assessment VAS at least 30 mm (0 to 100), (2) presence of inflammatory back pain (stiffness and pain worsening with rest and improving with exercise), (3) morning stiffness lasting 45 mins and/or peripheral arthritis; aged 18 years or more; stable dose of anti-inflammatory drugs; DMARDs ceased eight weeks before entry.
Exclusion criteria:
previous MTX treatment; history of hepatitis, pneumonitis; anaemia (Hb lower than 11 g/L); active infection, alcohol or drug abuse, mental or psychiatric disorders; those requiring corticosteroid or immunosuppressive treatment(s).
Age: 32 years ± 10 (MTX group), 38 years ± 10 (placebo group) (mean ± SD).
M:F ratio: 69%: 31%.
Mean disease duration since first symptoms: 9.5 ± 8 years (MTX group), 5.7 ± 5 years (placebo group).
With peripheral arthritis: 60%.
InterventionsMTX group: MTX 7.5 mg/week orally.
Placebo group: placebo orally,
OutcomesPrimary outcomes: response in the composite index (*see Notes section for this trial for details).
Secondary outcomes:
1. Morning stiffness (VAS 100 mm, 0 = no stiffness, 100 = severe).
2. Physical well-being (VAS 100 mm, 0 = feeling very well, 100 = feeling very bad).
3. Bath ankylosing spondylitis disease activity index (BASDAI) (from 0 to 10, 0 = no disease activity and 10 = severe disease activity).
4. Bath ankylosing spondylitis functional index (BASFI) (from 0 to 10, 0 = excellent, 10 = poor function).
5. Physician global assessment (VAS 100 mm, 0 = no disease activity and 100 = severe disease activity).
6. Patient global assessment (VAS 100 mm, 0 = no disease activity and 100 = severe disease activity).
7. Health assessment questionnaire for spondyloarthropathies (HAQ-S) (from 0 to 3, 0 = no difficulty and 3 = unable to do).
8. Spinal pain (VAS 10 cm, 0 = no pain, 10 = severe).
9. Disappearance of peripheral arthritis (event).
Notes

* Composite index defined by improvement of 20% in at least 5 of following:
1. Severity of morning stiffness VAS 100 mm;
2. Physical well-being VAS 100 mm;
3. BASDAI;
4. BASFI;
5. HAQ-S;
6. Physician global assessment VAS 100 mm;
7. Patient global assessment VAS 100 mm.
as well as no worsening in any of the scales (defined as >20% worsening compared to baseline).

8. Disappearance of peripheral arthritis

9. Spinal pain (from 0 to 10)
All outcomes reported as mean ± SD of endpoint value except BASFI. Mean change from baseline ± SD reported for BASFI.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer generated random list.
Allocation concealment (selection bias)Low riskSequentially numbered drug containers of identical appearance.
Blinding (performance bias and detection bias)
All outcomes
Low riskThe patients and researchers involved in administration of the interventions, clinical evaluations, and statistical analyses were blinded to the treatment assignment for the duration of the trial.
Incomplete outcome data (attrition bias)
All outcomes
Low riskFor dichotomous outcome data, the proportion of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate.
Selective reporting (reporting bias)Low riskThe study protocol is available and all of the study's pre-specified (primary and secondary) outcomes that are of interest in the review have been reported in the pre-specified way.
Other biasHigh riskSome baseline parameters were significantly different between intervention groups.

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
Breban 2008There is another separate Cochrane review addressing biologics in AS.
Kabasakal 2009Not a RCT.
Li 2008There is another separate Cochrane review addressing biologics in AS.
Mulleman 2011There is another separate Cochrane review addressing biologics in AS.
Pérez-Guijo 2007Not a RCT.
Xie 2003This trial compared the effects of different dosage of MTX and did not include the comparison specified in the inclusion criteria.

Characteristics of studies awaiting assessment [ordered by study ID]

Haque 1999

MethodsRCT
ParticipantsPatients with seronegative spondyloarthropathies. Data of AS patients were not available in the article abstract.
InterventionsMethotrexate. Comparison is unknown.
OutcomesTherapeutic response.
NotesThis study could not be reviewed because of lack of full text.

Ancillary