Statistics from Altmetric.com
Gout is one of the most common inflammatory arthritis conditions, affecting up to 1–2% of men in Western countries1 and causing morbidity, disability and poorer quality of life.2 It is the consequence of deposition of monosodium urate (MSU) crystals in joints and other tissues, as a result of persistent hyperuricaemia. The aim of treatment is to reduce serum uric acid (SUA) levels, allowing MSU crystals to dissolve, leading to the elimination of acute episodes of inflammation, the disappearance of tophi, and, eventually, cure of the disease.3 However, suboptimal management of the condition is still reported4–6 despite the publication of a number of guidelines and recommendations,7–11 the development of new therapeutic agents, and the introduction of target-directed management strategies.12 Some evidence suggests that guidelines that are implemented improve quality of care and that interventions involving educational outreach may help the successful implementation and dissemination of guidelines.13
The 3e (Evidence, Expertise, Exchange) Initiative is a unique multinational collaboration aimed at promoting evidence-based practice in rheumatology by developing practical recommendations addressing relevant clinical problems.14–16 Unlike most existing guidelines or recommendations developed by a limited panel of experts in the field, the 3e Initiative involves a large number of practising rheumatologists from around the world. Recommendations are made in response to the identification of the 10 most important clinical questions posed by the group, rather than the more all-purpose method of generating treatment recommendations. The objective was to develop evidence-based and practical recommendations for the diagnosis and management of gout with consensus from a large number of practising rheumatologists from many countries. In addition, through the dissemination of the results of systematic literature reviews (SLRs) to such a large number of rheumatologists, an understanding of the current extent of knowledge in this field was widely shared. This educational activity may increase the uptake of the guidelines.
A total of 474 rheumatologists from 14 countries participated in the 2011 3e Initiative. Twelve scientific committees represented participating countries from Europe, South America and Australasia. The members of each of the national scientific committees formed a panel of experts who attended the multinational meetings. In addition, the bibliographic team comprised 10 multinational fellows (MA, ASRK, JM, RS, FS, MS, CvD, IvE, OV and MDW), six mentors (DA, CB, RB, LC, CJE and RBL) and the scientific chair (DMvdH). At the first international meeting, clinically relevant questions regarding gout diagnosis and management were spontaneously proposed, and 10 were selected via a modified Delphi voting process by the panel of 78 expert rheumatologists representing all 14 countries (table 1). The multinational fellows and supervising mentors then translated the questions into Population, Intervention, Comparator, Outcome (PICO) terms, agreed on the protocols, and undertook SLRs for each clinical question. A comprehensive search strategy was generated for each question aided by an experienced librarian (LF) (last date October 2011); where feasible, search terms were standardised (see online supplementary figures S1 and S2). Searches were conducted in Medline, Embase and the Cochrane Central Register of Controlled Trials (CENTRAL), and hand searches of the reference list of the selected articles and of abstracts presented at the 2010 and 2011 American College of Rheumatology (ACR) and European League Against Rheumatism (EULAR) scientific meetings were performed. Two independent reviewers screened the titles and abstracts of all citations identified by the searches, assessed potentially relevant articles in full text for inclusion according to predetermined criteria, and performed the data extraction of the selected studies (see online supplementary table S1). When discrepancies arose and no consensus could be reached, a mentor acted as arbiter. Included articles were restricted to those published in English or in a language in which at least one member of the bibliographic group was fluent (Dutch, French, German, Spanish). Standardised tools were used to assess the risk of bias of included studies (Cochrane Risk of Bias tool for intervention studies,17 Hayden tool18 for cohort studies, Newcastle-Ottawa Scale for case–control studies,19 the Consensus-based stadards for the selection of health measurement instruments (COSMIN) checklist20 for validation of measurement instruments, and the Cochrane Risk of Bias tool for diagnostic studies21). Where relevant, we considered outcomes proposed by OMERACT (Outcome Measures in Rheumatology Clinical Trials) to be used in the evaluation of interventions for acute and chronic gout.22 We planned to pool relevant data from included studies provided that they were sufficiently homogeneous. This was predefined in each SLR protocol. Details and results of the SLR for each question will be published separately, but a summary of the supporting evidence is presented under each recommendation in the Results section. After presentation of the SLR results, each of the 12 national scientific committees produced recommendations leading from the 10 clinical questions. At the final international meeting, members of each of the scientific committees merged the national recommendations into 10 final multinational recommendations through a process of discussion and a modified Delphi vote with an electronic voting system (up to three rounds with prespecified cut-off points). The participating rheumatologists quantified their agreement with each recommendation on a 1–10 scale (fully disagree to fully agree), and the potential impact of each recommendation on their clinical practice on a multiple choice question (recommendation will change my practice/is in accordance with my practice/I don't want to apply this recommendation). The level of evidence for each recommendation was appraised and graded in accordance with the Oxford Centre for Evidence-based Medicine Levels of Evidence.23 Where there was ambiguity regarding the appropriate grade or level of evidence, a lower grade or level was chosen.
The 10 final multinational recommendations are listed in table 2 with the levels of evidence and grades of recommendation; a summary of the supporting evidence and the expert opinion on each recommendation are presented below. The level of agreement by the rheumatologists with the recommendations ranged from 8.1 to 9.2 (mean 8.7) on a 1–10 point scale where 10 represents full agreement. For every recommendation, the proportion of rheumatologists voting 7 or more was over 80%. Many rheumatologists felt that the recommendations were in full accordance with their current practice (table 3). However, for two recommendations for which there was a lower accordance with current practice (comorbidity screen of renal function and cardiovascular risk factors, and achieve tight control of SUA in patients with tophi), there was a higher willingness to change current practice.
Recommendation 1: diagnosis
Four studies used MSU crystal identification as the reference standard to evaluate the diagnostic performance of over 60 individual clinical, laboratory and imaging findings.24–27 Most clinical, laboratory and x-ray features—including podagra and hyperuricaemia—show a low diagnostic utility as stand-alone findings with the exception of response to colchicine therapy and the presence of tophi. Advanced imaging techniques, such as ultrasound (US) and dual-energy CT, performed better.
Experts showed a strong consensus that identification of MSU crystals—in a joint fluid sample or in a tophi aspirate—is required for a definite diagnosis of gout. Since life-long urate-lowering therapy (ULT) is commonly prescribed after diagnosis, this procedure should be routinely undertaken. However, as this might prove difficult in some settings, it was felt that clinical or imaging findings could support a diagnosis. The presence of hyperuricaemia on its own is insufficient to establish a diagnosis of gout. Response of acute arthritis to colchicine could support a clinical diagnosis of gout, but was felt unhelpful in differentiating types of crystal arthritis (eg, gout and acute calcium pyrophosphate arthritis). Availability, cost and the need for trained personnel and specific equipment may limit the use of advanced imaging techniques in routine clinical practice.
Recommendation 2: comorbidity screening
The focus was on those comorbidities that could be both screened for and treated. An increased incidence of end-stage renal disease was found in patients with hyperuricaemia,28 but gout was not an independent predictor for this disease.29 However, a fourfold increase in mortality due to kidney disease has been reported in patients with gout compared with non-gouty patients.30 We identified evidence that hyperuricaemia may increase the risk of developing diabetes or hypertension31 ,32; however, no prospective studies were identified that investigated the risk of these conditions in people with gout. The available data showed that hyperuricaemia does not increase the risk of developing coronary heart disease (CHD)31–36 or stroke.37–39 On the other hand, there was evidence to suggest that people with gout have an increased risk of developing CHD40–43 and slightly increased risk of CHD-related mortality.44
Experts agreed to highlight the need to screen for renal disease on the basis of the strong evidence of association and the implications for gout therapy. Experts also agreed that hyperuricaemia and gout should be considered red flags for metabolic syndrome and cardiovascular diseases.
Recommendation 3: acute gout
Twenty-six trials were included on treatment of acute gout flares (21 evaluated non-steroidal anti-inflammatory drugs (NSAIDs),45 five glucocorticoids,46 two colchicine, and one canakinumab47). The available evidence showed that low-dose colchicine (total dose 1.8 mg in 24 h) was more effective than placebo48 ,49 and as effective as high-dose colchicine (total dose 4.8 mg), but lower doses of colchicine had a significantly better safety profile.49 There was no high-quality evidence comparing NSAIDs with placebo50 and no NSAID (conventional or selective COX-2 inhibitor) has proven superior to another.51–67 Three trials concluded that systemic glucocorticoids were as effective as NSAIDs, with a similar safety profile.68–70 Despite a comprehensive search strategy, no trials assessing intra-articular glucocorticoids or paracetamol in the treatment of acute gout flares were identified.
There was consensus that NSAIDs, colchicine and glucocorticoids (given as intra-articular, oral or intramuscular therapy) are all effective in the treatment of acute gout flares and that there was insufficient evidence to prioritise them. Individual treatment decisions should be based on consideration of an individual's characteristics and each drug's safety profile. Paracetamol, although not recommended as the primary therapy, can be useful as an adjunct analgesic.
Recommendation 4: lifestyle
There is no evidence to support the idea that intervening in lifestyle factors translates into improved outcomes in patients with gout. Despite a comprehensive search strategy,71 ,72 only one study assessing the efficacy of lifestyle interventions in the treatment of chronic gout was identified.73 The use of skimmed milk powder enriched with two dairy fractions (glycomacropeptide and G600 fat extract) did not result in a reduction in frequency of acute gout flares when compared with standard skimmed milk or lactose powder.73
Current understanding of the lifestyle factors associated with gout is largely derived from large, cross-sectional, epidemiological studies. Given the lack of evidence supporting lifestyle interventions in the treatment of gout per se, experts recommend general healthy lifestyle habits such as would be advisable for all individuals. Regarding alcohol consumption, experts agreed that there should be more emphasis on discouraging beer and spirits over wine intake. Together with general lifestyle advice, education about the need for compliance with lifelong ULT was deemed essential.
Recommendation 5: ULT
Over 40 studies were included in the evaluation of the efficacy, cost-efficacy and safety of ULT. There is high-quality evidence that allopurinol,74 febuxostat (40–240 mg daily)74 ,75 and pegloticase (8 mg intravenously every 2 or 4 weeks)76 are more effective than placebo in lowering SUA levels in patients with gout. One study showed that benzbromarone was effective in patients who failed to reach target uric acid on allopurinol.77 Febuxostat (80–240 mg daily) was more effective than potentially suboptimal doses of allopurinol (300 mg in patients with normal renal function, 100–200 mg if renal insufficiency) in lowering SUA, with a similar overall safety profile.74 ,78 ,79 Step-up therapy with allopurinol (300–600 mg) or benzbromarone (100–200 mg) are both effective in lowering SUA levels.80 Pegloticase, although highly efficacious, is associated with an increase in acute gout flares, infusion reactions and increased withdrawals due to adverse events compared with placebo.76 Available evidence for cost-efficacy was at a high risk of bias81 ,82 or outdated.83 No studies addressed the sequence of ULT.
There was a strong consensus that allopurinol constitutes first-line ULT after consideration of its safety, efficacy and cost. Low starting doses can optimise safety and minimise the risk of acute flares; doses should be gradually increased until target SUA levels are achieved (see recommendation 8). Uricosurics—where available—and low to medium doses of febuxostat (40–120 mg) are alternatives in the presence of intolerance or non-responsiveness to allopurinol. Uricase should only be considered in selected patients without other therapeutic options. Pegloticase should not be combined with other ULT, as this may mask the increase in SUA levels warning of an increased risk of infusion reactions and anaphylaxis.
Recommendation 6: flare prophylaxis
Four studies addressing flare prophylaxis when ULT is initiated were identified. In two randomised controlled trials, the use of colchicine (0.6–1.5 mg daily) for the initial 3–6 months after the start of ULT resulted in a reduction in the number of patients who developed acute gout attacks and a reduction in the severity of these flares compared with placebo.84 ,85 Despite an increase in diarrhoea in one study,84 overall adverse effects and withdrawals were similar between the colchicine and placebo groups. No evidence on the use of NSAIDs or glucocorticoids as prophylaxis was retrieved.
Experts considered that the need for acute gout flare prophylaxis when initiating ULT should be considered on an individual basis. Optimal duration is currently unclear and should be decided after assessing factors such as flare frequency, gout duration and the presence and size of tophi. There was no consensus on when ULT should be started after an acute attack. However, the majority felt that low initial doses of ULT, with slow dose increases, is an integral part of flare prevention, supporting the motto ‘start low, go slow’.
Recommendation 7: effect of comorbidities on drug choice
Two studies, of low to moderate quality, showed that gradual dose escalation of allopurinol in patients with renal impairment resulted in a higher proportion of patients obtaining target SUA levels without a parallel increase in serious toxicity,86 ,87 when compared with the commonly used and more conservative dosing guidelines.88 Allopurinol has been compared with other ULTs in populations with renal impairment of mostly mild or moderate levels (creatinine clearance >30 mL/min). Both febuxostat (80 mg/day)74 ,79 and unadjusted benzbromarone (100–200 mg/day)89 resulted in a higher proportion of patients achieving target SUA compared with renal function-adjusted allopurinol (100–300 mg/day), with a similar safety profile. The combination of allopurinol and benzbromarone allowed a reduction in SUA levels except in cases of severe renal dysfunction.90
Recommendation 8: monitoring
The target for the treatment of any disease is either cure or control. Both of these goals may be abstract concepts and can be difficult to measure. Often a surrogate marker associated with the cure or control is used; in gout, this surrogate marker is SUA.8 ,9 ,11 ,91–93 The association of SUA with other potential outcomes22 was systematically reviewed. Six studies linked the reduction of SUA levels with a decreased rate of acute attacks,94–99 two studies with tophus regression,100 ,101 and three studies with crystal disappearance—either through US102 or synovial fluid microscopy.103 ,104 The quality of these studies was low to moderate. The most commonly used SUA cut-off point in studies was 0.36 mmol/L (6.0 mg/dL), but there is some evidence that lower SUA levels could lead to a higher speed of tophi reduction and a longer time to recurrence of acute attacks after treatment withdrawal.94 ,100 Numerous tools have been used for monitoring the different outcome domains in patients with gout, including biological markers, clinical features, patient-reported outcomes or imaging. The physical component of the SF-36 questionnaire,105 ,106 the Health Assessment Questionnaire,105 ,107 and tophus measurement by caliper108 or US109 have shown adequate clinimetric properties.
Experts considered that monitoring should include at least SUA levels, the frequency of gouty attacks, and tophi size, but recommended no specific tool. They agreed that the target should be an SUA level below 0.36 mmol/L, but recommended even lower cut-off points if tophi are present (see recommendation 9).
Recommendation 9: tophi
After a comprehensive search strategy,110 only four prospective studies assessing pharmacological agents for patients with tophaceous gout were identified: two randomised controlled trials with pegloticase,76 an open extension study with febuxostat,111 and a case series of patients with tophaceous gout on different ULTs.100 A sustained reduction in SUA led to tophi reduction and in some cases resolution, independently of which ULT was used. The only evidence for the use of surgery to treat tophi came from case reports and case series.
Experts agreed that a lower SUA level (0.3 mmol/L) should be a treatment target for patients with tophaceous gout, as the evidence suggested that lower SUA levels increase the speed of tophi reduction. Surgery should only be considered in selected cases (eg, nerve compression, mechanical impingement or infection).
Recommendation 10: asymptomatic hyperuricaemia
Defining asymptomatic hyperuricaemia was controversial; the agreed definition excluded patients with a background of arthritis or tophi, but allowed the inclusion of patients with pre-existing renal or cardiovascular disease. After an extensive search, only three studies112–114 were retrieved. Patients with asymptomatic hyperuricaemia and normal renal function112 ,113 or chronic kidney disease114 at baseline were allocated to receive allopurinol or no treatment over a 3–12 month period; no significant differences were noted in glomerular filtration rate, serum creatinine or proteinuria between the two groups. No studies dealing with prevention of gout or cardiovascular disease met the inclusion criteria.
Although there was an absence of evidence supporting the use of ULT for asymptomatic hyperuricaemia, experts agreed that lifestyle advice on diet, weight loss or exercise would apply to patients with asymptomatic hyperuricaemia, especially after considering the increased risks stated in recommendation 2.
The 3e Initiative developed 10 recommendations for the diagnosis and management of gout. These address questions relevant to the clinical setting, are informed by the currently available evidence, and are endorsed by a large international panel of rheumatologists.
Even though gout is a potentially curable disease, its management is far from optimal5 in both primary care and rheumatology clinics. The quality of care provided to gout patients needs to improve. Guidelines that are implemented improve quality of care,115 and educational outreach has an effect on implementation14; therefore, we may suppose that multinational evidence-based recommendations developed in a way in which education—in both gout and evidence-based medicine116—and dissemination are final aims can contribute towards this goal.
Two sets of recommendations have recently been published: the first by an American group (with a USA perspective)11 and the second on behalf of the ACR.9 ,10 The first group's approach differed from ours, accepting the 2006 EULAR recommendations7 ,8 as a basis and reappraising the evidence published in the past 6 years (2005–2011). The ACR recommendations—produced following the RAND/University of California at Los Angeles (UCLA) methodology—centred on the treatment and prophylaxis of acute gout flares and the appropriate use of ULT in gout, excluding issues on gout diagnosis or asymptomatic hyperuricaemia. The recommendations are similar in some areas, but methodological differences between the 3e Initiative and the ACR guidelines—including the exclusion of benzbromarone (unavailable in the USA) and cost and cost-effectiveness appraisals—have given rise to differences in drug therapy hierarchies.
The 3e recommendations have been developed through an established process with a number of strengths. First, the formal voting process of a broad international panel representing several continents resulted in the development of 10 relevant clinical questions. Second, the available evidence was appraised and summarised following a rigorous approach, which was then combined with the experience of numerous rheumatologists. Last, the high level of agreement with the final recommendations and the multinational participation increases their utility and will hopefully facilitate their dissemination and implementation worldwide. Most participating rheumatologists either follow the recommendations or are willing to change their practice according to them, suggesting a solid potential impact of this set of recommendations.
A number of limitations of these recommendations must, however, be taken into account. Other specialties (eg, nephrology, primary care), health professionals and patients have not participated in the development of these recommendations. It is therefore unclear how applicable or relevant they are in non-rheumatological settings. Also, many recommendations are complex, including several statements with different degrees of evidence. However, experts voted on their global agreement with the entire recommendation. Finally, variability in agreement of some recommendations suggests a certain degree of dispersion; however, it must be noted that the proportion of attending rheumatologists voting 7 or over for each recommendation was over 80%, suggesting a significant degree of support for these recommendations.
In summary, 10 multinational recommendations for the diagnosis and management of patients with gout in daily clinical practice have been developed, integrating SLR and expert opinion, with the aim of improving patient care.
The authors acknowledge the work of all members of the 3e scientific committees and all participants in the national meetings.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
Handling editor Tore K Kvien
FS and MA contributed equally.
Contributors DMvdH designed and coordinated the study. FS, MA, ASRK, JM, RS, MS, CvD, IvE, OV, MDW, LC, DA, CB, RB, CJE, RBL, DMvdH and LF performed the systematic literature review. JWB, JCB, RB-V, AIC, DE, AJLF, PK, BFL, CM, UM-L, MØ and JZ coordinated the generation of national recommendations in their countries. All authors participated as the panel of rheumatologist experts. FS and MA wrote the first draft of the paper; all authors approved the final version.
Funding This paper is derived from the 3e Gout program, which was sponsored by AbbVie. Margaux Orange, Paris, France provided logistical and administrative support for the 3e Gout meetings; this work was funded by AbbVie. AbbVie employees were present during the 3e meetings, but did not influence the scientific discussions. AbbVie did not review the content or have influence on this manuscript. RB is partially funded by an Australian National Health and Medical Research Council (NHMRC) Practitioner Fellowship.
Competing interests FS received fees from Menarini for preparing educational presentations. LC received lecture fees from Menarini in 2010. AIC has received an unrestricted research grant and acted as consultant for Abbott. UML has acted as speaker and consultant for AbbVie, Berlin-Chemie, MSD. DMvdH has received consulting fees and/or research grants from AbbVie, Amgen, AstraZeneca, Augurex, BMS, Celgene, Centocor, Chugai, Daiichi, Eli-Lilly, GSK, Janssen Biologics, Merck, Novartis, Novo-Nordisk, Otsuka, Pfizer, Roche, Sanofi-Aventis, Schering-Plough, UCB and Vertex and is Director of Imaging Rheumatology bv.
Provenance and peer review Not commissioned; externally peer reviewed.
Open Access This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 3.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/3.0/
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.