How flare prevention outcomes are reported in gout studies: A systematic review and content analysis of randomized controlled trials

Journal Pre-proof

How flare prevention outcomes are reported in gout studies: A systematic review and content analysis of randomized controlled trials Sarah Stewart PhD , Amy Tallon MBChB , William J. Taylor MBChB, PhD, FAFRM (RACP), FRACP , Angelo Gaffo MD, MsPH , Nicola Dalbeth MBChB, MD, FRACP PII: DOI: Reference:

S0049-0172(19)30682-1 https://doi.org/10.1016/j.semarthrit.2019.11.002 YSARH 51558

To appear in:

Seminars in Arthritis & Rheumatism

Please cite this article as: Sarah Stewart PhD , Amy Tallon MBChB , William J. Taylor MBChB, PhD, FAFRM (RACP), FRACP , Angelo Gaffo MD, MsPH , Nicola Dalbeth MBChB, MD, FRACP , How flare prevention outcomes are reported in gout studies: A systematic review and content analysis of randomized controlled trials, Seminars in Arthritis & Rheumatism (2019), doi: https://doi.org/10.1016/j.semarthrit.2019.11.002

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Elsevier Inc. All rights reserved.

How flare prevention outcomes are reported in gout studies: A systematic review and content analysis of randomized controlled trials

Type Systematic reviews and meta-analyses

Authors Sarah Stewart*a, PhD Amy Tallona, MBChB William J. Taylorb, MBChB, PhD, FAFRM (RACP), FRACP Angelo Gaffoc, MD, MsPH Nicola Dalbetha, MBChB, MD, FRACP

a

Faculty of Medical and Health Sciences, The University of Auckland, Private Bag 92019, Auckland 1142, New Zealand b

Department of Medicine, University of Otago, PO Box 7343, Wellington South 6242, New Zealand c

School of Medicine, University of Alabama at Birmingham, 1720 2nd Ave South, Birmingham, AL 35294, United States of America

*Corresponding author: [email protected]; Faculty of Medical and Health Sciences, The University of Auckland, Private Bag 92019, Auckland 1142, New Zealand

Declaration of Interest ND has received consulting fees, speaker fees or grants from AstraZeneca, Horizon, Amgen, Kowa, Takeda, Abbvie, Pfizer, and Janssen, outside the submitted work. Angelo Gaffo has received research support from Amgen, outside of the submitted work. The other authors declare no competing interests.

1

ABSTRACT

Objectives: There are many potential ways that gout flares could be reported in clinical trials. The aim of this study was to describe the methods used to measure and report gout flare prevention outcomes in randomized controlled trials (RCTs). Methods: A systematic search of electronic databases was conducted. Articles published between 2008 and 2018 were included if they were RCTs or articles reporting on analyses of RCT data (i.e. open label extension studies) and reported the impact of an intervention on the prevention of flares in people with gout. The modified-Jadad scale was used to assess study quality. Methods used to measure and report gout flare outcomes were extracted and synthesised separately for studies of anti-inflammatory prophylaxis and urate lowering/other long term therapy. Results: A total of 38 articles were included, with 10 reporting outcomes for antiinflammatory prophylaxis and 28 for urate lowering/other long term therapies. The overall quality score of all articles was good. There was marked heterogeneity across trials in gout flare definitions, data capture methods, reporting methods and time periods used to report flares. Anti-inflammatory prophylaxis studies used multiple methods to report gout flare outcomes (mean (SD) 4.3 (2.5) methods/article), while the majority of urate lowering/other long term therapy studies used a single method to report gout flare outcomes. The most common reporting method in anti-inflammatory prophylaxis studies was the mean number of gout flares per patient (n=9 articles), and in urate lowering/other long term therapy studies was the proportion of patients with at least one gout flare (n=22 articles). Only studies of anti-inflammatory prophylaxis therapy reported flare duration or pain during flare. Conclusion: There is wide variation in methods used to measure and report gout flare prevention outcomes in long-term RCTs. These findings highlight the need for standardized methods for studies in which gout flare prevention is an outcome of interest.

Keywords: gout flare, randomized clinical trial

2

1. INTRODUCTION

A gout flare is a clinically evident episode of acute articular or peri-articular inflammation induced by monosodium urate crystals [1]. Gout flares cause severe pain and disability, and are a central concern for people with gout [2-4]. The Outcome Measures in Rheumatology (OMERACT) group has included gout flares as a core domain for use in clinical trials evaluating the chronic management of gout [5]. A definition of a gout flare has recently been developed and validated for use in clinical trials based on patient-reported criteria (> 3 of: patient-defined gout flare, pain at rest score of > 3 on a 0-10 point numeric rating scale, presence of at least one swollen joint and presence of at least one warm joint) [6, 7]. While this work has defined the state of being in flare, there is no current agreement about how flare outcomes should be measured and reported in studies of either anti-inflammatory prophylaxis or urate lowering therapies. The successful reproducibility and validity of long term clinical trials evaluating the effects of interventions on flare prevention, including anti-inflammatory prophylaxis and urate lowering therapy, requires standardised measurement and reporting methods for gout flares. The aim of this study was to describe the methods of measuring and reporting flare prevention outcomes in randomised controlled trials assessing the prevention of flares in people with gout.

2. METHODS This study was conducted according to a pre-defined protocol using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [8].

2.1 Search strategy The Cochrane Central Register of Controlled Trials (CENTRAL) database was searched in November 2018 using the search terms presented in Supplementary Table 1. The MEDLINE Ovid database was also searched in November 2018 using the Cochrane Highly Sensitive

3

Search Strategy for Identifying Randomised Trials in MEDLINE (sensitivity- and precisionmaximising version [9]) (Supplementary Table 1). Both searches were filtered to include publications between 2008 and 2018. This range in publication dates was chosen to ensure that contemporary clinical trials were included in the analysis. Reference lists from included articles and review articles were also examined for additional articles. All articles generated from the search were exported to EndNote X8. After the removal of duplicates, the titles and abstracts were screened for inclusion. Articles were included if they: were randomised trials or reported on analyses of randomised trial data (i.e. open label extension studies); reported the impact of an intervention on the prevention of gout flares in people with gout; and had a duration of follow-up of > 1 month. Articles were excluded if they: reported interventions for gout flare treatment (i.e. included participants who were currently flaring) as opposed to gout flare prevention; were not published in English; or were narrative or systematic reviews, non-intervention studies, case reports, retrospective studies, individual studies enrolling fewer than six participants, commentary letters or non-human studies. Articles which combined data from randomised trials already included in the review and provided no new data relating to gout flare measurement and reporting methods were also excluded. The full-texts of eligible articles based on title and abstract screening were then independently screened by two reviewers to confirm eligibility (SS, AT). Any disagreements in article selection were resolved by discussion of the two reviewers. If necessary, a third reviewer (ND) was involved to reach consensus.

2.2 Risk of bias assessment The risk of bias for included studies was assessed independently by two reviewers (SS, AT) using the 6-item modified-Jadad scale which assesses randomisation, blinding, withdrawals, dropouts, inclusion/exclusion criteria, adverse events, and the statistical analysis [10]. The scale has a maximum score of 8 points, with low quality studies yielding scores of 0 to 3 and high quality studies yielding scores of 4 to 8. Any disagreements in quality assessment were resolved by discussion of the two reviewers. If necessary, a third reviewer (ND) was involved to reach consensus. 4

2.3 Data extraction Data were extracted from all included full-text articles by two independent reviewers (SS, AT), including: study design, intervention characteristics (drug name and dose for each group, duration of intervention, and duration of follow-up), participant baseline characteristics (total sample size, gout classification criteria, mean (SD) age (years), n (%) males, mean gout disease duration (years), mean (SD) serum urate (mg/dL, converted from mmol/L, as required), and n (%) presence of subcutaneous tophi), gout flare outcome measurement and data capture methods (flare definition used, whether flares were assessed retrospectively or in real-time and the time points assessed) and gout flare outcome reporting methods (whether flare outcomes were primary and/or secondary outcomes, whether flares were an outcome of treatment efficacy and/or adverse events, the time period(s) used to report gout flare results, and details of the methods used to report gout flare outcomes). Any disagreements regarding data extraction were resolved by a third reviewer (ND).

2.4 Data synthesis For the purpose of data synthesis, data were summarised separately using descriptive statistics for studies assessing anti-inflammatory prophylaxis therapies and studies assessing urate lowering drugs/other long term therapies. Other long term therapies were considered as any therapy used in the management of gout, including dietary/ lifestyle interventions and interventions targeting bone damage. Studies comparing gout flare prevention during initiation of different urate lowering therapies were included under the anti-inflammatory prophylaxis category. Meta-analyses which were pre-planned to determine which gout flare assessment and reporting methods were most sensitive for use in clinical trials could not be undertaken due to the extent of heterogeneity across the included studies.

5

3. RESULTS

3.1 Study characteristics The search generated a total of 520 articles and following review of titles and abstracts, 87 full-text articles were assessed for eligibility. After the exclusion of 49 articles (reasons for exclusion are presented in Figure 1), a total of 38 articles were included in the review (34 randomised trials and four open label extension studies). Of the included studies, 10 investigated anti-inflammatory prophylaxis therapies and 28 investigated urate lowering/other long term interventions, which included therapies targeting bone damage (zoledronate), dietary advice, and nurse- vs. GP-led care. Characteristics of the interventions used in the studies are presented in Supplementary Table 2. The intervention durations ranged from 12 to 52 weeks for the anti-inflammatory prophylaxis studies and from four weeks to 60 months for the urate lowering/other long term therapy studies. Participant baseline characteristics are presented in Supplementary Table 3.

3.2 Risk of bias Figure 2 presents the results of the risk of bias assessment for each of the included studies. The overall quality score was classed as good for all included studies (modified-Jadad score >3). Of the 34 randomised trials, appropriate randomisation methods were described by 24 (71%) articles; 24 (71%) studies were described as double-blind, one study was single-blind, and the remaining studies were not blinded (n = 9, 37%). Of the total 38 studies, the majority provided an adequate description of: withdrawals and dropouts (n = 37, 97%), inclusion/exclusion criteria (n = 38, 100%), the method used to assess adverse effects (n = 33, 87%) and the methods of statistical analysis (n = 36, 95%).

3.3 Gout flare definitions 3.3.1 Anti-inflammatory prophylaxis therapy studies

6

Gout flare definitions for the 10 studies investigating anti-inflammatory prophylaxis therapies are shown in Table 1. Seven studies recognised flares as only those requiring treatment. Of these, three studies provided additional, more comprehensive, criteria to be met related to flare duration, pain severity and other clinical symptoms. Two of these studies reported flare outcomes using both the comprehensive criteria as well as a more conservative definition. Two studies defined flares based on patient-reported symptoms only. One study did not define gout flares. 3.3.2 Urate lowering/other long term therapy studies Of the 28 urate lowering/other long term therapy studies, gout flares were defined in 15 studies (Table 1). Fourteen of these studies recognised flares as only those requiring treatment, and of these, two studies also required participants to meet other criteria related to flare duration, pain severity and other clinical symptoms. One of these studies reported flare outcomes using the comprehensive definition as well as a more conservative definition. The remaining 13 studies did not define gout flares.

3.4 Method of capturing gout flare data 3.4.1 Anti-inflammatory prophylaxis therapy studies Table 1 also shows the methods used for gout flare data capture. Five of the 10 antiinflammatory prophylaxis studies captured gout flares in real-time using daily diaries. Four studies captured gout flares in real-time using daily telephone interactive voice response systems. One study captured gout flares retrospectively at study visits. 3.4.2 Urate lowering/other long term therapy studies Of the 28 urate lowering/other long term therapy studies, gout flare data were captured retrospectively at study visits by 12 studies, and in real-time using either daily dairies by seven studies, or by direct contact with the investigators in two studies (Table 1). The remaining nine studies did not report how gout flare data were captured.

3.5 Flare as primary or secondary outcome 7

3.5.1 Anti-inflammatory prophylaxis therapy studies Of the 10 anti-inflammatory prophylaxis studies, flares were measured as primary outcomes in nine studies, with six of these also reporting secondary gout flare outcomes (Table 2). One study reported gout flare data as a secondary outcome only. 3.5.2 Urate lowering/other long term therapy studies None of the 28 urate lowering/other long term therapy studies reported flares as a primary outcome. Flares were reported as secondary outcomes in 13 studies and as exploratory, additional or other outcomes in four studies (Table 2). The remaining 11 studies did not report the outcome level of gout flare data.

3.6 Treatment efficacy vs. adverse event 3.6.1 Anti-inflammatory prophylaxis therapy studies All 10 anti-inflammatory prophylaxis studies reported flares as a measure of treatment efficacy (Table 2). 3.6.2 Urate lowering/other long term therapy studies Twenty-four of the 28 urate lowering/other long term therapy studies reported flares as a measure of treatment efficacy (Table 2). Of these, three studies also reported flare data as an adverse event. Three studies reported flares as an adverse event only. The remaining two studies did not report whether flares were a measure of treatment efficacy or considered as an adverse event.

3.7 Time periods used to report flare data 3.7.1 Anti-inflammatory prophylaxis therapy studies The time periods used to report flare outcomes varied considerably across the 10 antiinflammatory prophylaxis studies (Table 2). Follow-up periods ranged from 3 to 52 months. Reporting of flare outcomes for the total follow-up period was the most common time period used (n = 8 studies). Flare outcomes were also reported for specific time periods for the total 8

follow-up period by five studies or at the start and/or end of the follow-up period by two studies. 3.7.2 Urate lowering/other long term therapy studies The time periods used to report flare results also varied considerably across 28 urate lowering/other long term therapy studies, with follow-up periods ranging from two to 40 months (Table 2). The most common time period used to report flare results was for the total follow-up period (n = 16 studies). Ten studies reported flare outcomes for specific time periods during the total follow-up period; and six studies reported flare outcomes for periods at the start and/or end of the follow up period. One study did not report the time period used to report gout flare outcomes.

3.8 Type of gout flare reporting method 3.5.1 Anti-inflammatory prophylaxis therapy studies Of the 10 anti-inflammatory prophylaxis therapy studies, all but one study used multiple methods to report gout flare results. One study used eight methods to report gout flares, and two studies used seven methods (Table 3). The mean (SD) number of gout flare outcome reporting methods was 4.3 (2.5). The most common method used was the mean number of gout flares per patient (n = 9 studies), followed by the proportion of patients with > 1 gout flare (n = 7 studies) (Table 4). Other methods included the proportion of patients with > 2 gout flares (n = 4 studies), > 3 gout flares (n = 1 study), and exactly 1 gout flare (n = 1 study). One study also reported individual patient data showing the number of gout flares per patient. Four studies reported results relating to the median number of days to the first gout flare, mean/median gout flare duration in days per patient (n = 3 studies) and the mean number of total gout flare days (n = 4 studies). Pain-related reporting methods were also used, including mean gout flare pain in the past week on a 10 cm visual analogue scale (n = 1 study), mean number of days per patient with a pain severity score > 5 on a 10 point numeric rating scale (n = 3 studies), and pain scores during flares on a 10 point Likert scale (n = 1 study).

9

3.5.2 Urate lowering/other long term therapy studies The majority of the 28 urate lowering/other long term therapy studies used a single gout flare reporting method (n = 22, 78.6%), or two gout flare reporting methods (n = 4) (Table 3). One study used three methods and one study used four methods. The mean (SD) number of gout flare outcome reporting methods was 1.3 (0.7). The most common method used for reporting gout flares was the proportion of patients with at least 1 gout flare over a specified time period (n = 22 studies), followed by the mean number of gout flares per patient (n = 9 studies). Other methods used were the proportion of patients with > 2 gout flares and > 4 gout flares (n = 1 study), proportion who withdrew from the study due to a gout flare (n = 1 study), and the proportion requiring hospital admission due to a flare (n = 1 study). Two studies also reported the total number of flares in each intervention group. No urate lowering/long term therapy study reported gout flare time-related results or pain-related results.

4. DISCUSSION This content analysis highlights the wide variation in methods used to measure and report gout flare prevention outcomes in randomised clinical trials. Studies assessing antiinflammatory prophylaxis therapies generally measured and reported several gout flare outcome characteristics, including frequency, duration and pain. Contrastingly, uratelowering/other long term therapy studies reported limited data relating to the gout flare characteristics. The most common gout flare outcome reporting method in anti-inflammatory prophylaxis studies was the mean number of gout flares per patient, while in urate-lowering/other long term therapy studies the most common method was the proportion of patients with gout flares. These measures of flare frequency were the only reported methods used in urate lowering/other long term therapy studies. Gout flares were generally reported as a secondary outcome in studies or urate-lowering/other long term therapies, while the majority of anti-inflammatory prophylaxis studies included gout flares as primary outcomes of treatment efficacy. This may explain the additional gout flare characteristics captured and reported in the latter trials; such as time to first flare onset, flare duration and pain severity. 10

Such time-dependent reporting strategies, including gout flare duration and pain-time curves have been shown to correlate well with other measures of gout disease severity including global wellbeing, C-reactive protein and tender and swollen joint counts [11]. The reporting of gout flares has been endorsed by OMERACT as an outcome measure for gout in long term clinical trials [5]. However, there is a lack of current consensus on how to measure and report flares as an outcome [12]. A recent analysis [13], limited to RCTs of urate lowering therapy, highlighted the variability in reporting of both serum urate outcomes and gout flares. These findings are consistent with the current analysis which also showed substantial variation in gout flare measurement and data capture methods in trials of both anti-inflammatory prophylaxis and other long term therapies. The substantial heterogeneity in the reporting methods observed in this analysis meant between-study comparisons and meta-analyses could not be undertaken to determine which gout flare assessment and reporting method is most sensitive for use in clinical trials. Future work should address the development of standardized outcome measurement and reporting methods that reflect the burden of flares for clinical trials assessing interventions targeting flare prevention. In addition, further work is needed to determine which characteristics of gout flares are most important to patients as outcomes of effective therapies. To date, research has shown that one flare or less in the previous six month period is associated with patient-perceived remission [14]. Determining the importance of other characteristics, including flare duration and pain severity, would also be of value. In conclusion, there is wide variation in methods used to measure and report flare prevention outcomes in gout RCTs. Studies of anti-inflammatory prophylaxis generally report a range of gout flare characteristics, including incidence, number of flares, flare duration, and pain intensity. In contrast, studies of urate lowering/other long term therapies report limited data, mostly the proportion of participants experiencing flares. These findings highlight the need for standardized methods for studies in which gout flare prevention is an outcome of interest.

Acknowledgements Greg Gamble for his assistance in data synthesis. 11

Funding None

REFERENCES [1]

[2]

[3] [4]

[5] [6] [7]

[8]

[9] [10]

[11]

[12] [13]

[14]

Bursill D, Taylor WJ, Terkeltaub R, Kuwabara M, Merriman TR, Grainger R, et al. Gout, Hyperuricemia and Crystal-Associated Disease Network (G-CAN) consensus statement regarding labels and definitions for disease elements in gout. Arthritis Care Res 2019;7:42734. Tatlock S, Rudell K, Panter C, Arbuckle R, Harrold LR, Taylor WJ, et al. What outcomes are important for gout patients? In-depth qualitative research into the gout patient experience to determine optimal endpoints for evaluating therapeutic interventions. Patient 2017;10:65-79. Seow LL, Jiao N, Wang W, Holroyd E, Teng GG, He H-G. A qualitative study exploring perceptions of patients with gout. Clin Nurs Res 2018:1054773818769219. Lindsay K, Gow P, Vanderpyl J, Logo P, Dalbeth N. The experience and impact of living with gout: a study of men with chronic gout using a qualitative grounded theory approach. J Clin Rheumatol 2011;17:1-6. Schumacher Hr, Taylor W, Edwards L, Grainger R, Schlesinger N, Dalbeth N, et al. Outcome domains for studies of acute and chronic gout. J Rheumatol 2009;36:2342-5. Gaffo AL, Dalbeth N, Saag KG, Singh JA, Rahn EJ, Mudano AS, et al. Brief report:Vvalidation of a definition of flare in patients with established gout. Arthritis Rheumatol. 2018;70:462-7. Gaffo AL, Schumacher HR, Saag KG, Taylor WJ, Dinnella J, Outman R, et al. Developing a provisional definition of flare in patients with established gout. Arthritis Rheum 2012;64:1508-17. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JPA, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ. 2009;339:b2700. The Cochrane Collaboration. Cochrane handbook for systematic reviews of interventions 2011, www.handbook.cochrane.org; 2018 [accessed 26 September 2018]. Oremus M, Wolfson C, Perrault A, Demers L, Momoli F, Moride Y. Interrater reliability of the modified Jadad quality scale for systematic reviews of Alzheimer's disease drug trials. Dement Geriatr Cogn Disord 2001;12:232-6. Teoh N, Gamble GD, Horne A, Taylor WJ, Palmano K, Dalbeth N. The challenges of gout flare reporting: mapping flares during a randomized controlled trial. BMC Rheumatol 2019;3:2735 de Lautour H, Dalbeth N, Taylor WJ. Outcome measures for gout clinical trials: a summary of progress. Curr Treat Option Rheumatol 2015;1:156-66. Stamp LK, Morillon MB, Taylor WJ, Dalbeth N, Singh JA, Lassere M, et al. Variability in the reporting of serum urate and flares in gout clinical trials: need for minimum reporting requirements. J Rheumatol 2018;45:419-24. Gaffo AL, Dalbeth N, Saag KG, Singh J, Rahn EJ, Mudano AS, et al. Patient perception of gout flares as a measure of outcome: results from an international study [abstract]. Arthritis Rheumatol 2016;68:290-1.

12

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

Wortmann RL, Macdonald PA, Hunt B, Jackson RL. Effect of prophylaxis on gout flares after the initiation of urate-lowering therapy: analysis of data from three phase III trials. Clin Ther 2010;32:2386-97. Schlesinger N, Mysler E, Lin HY, De Meulemeester M, Rovensky J, Arulmani U, et al. Canakinumab reduces the risk of acute gouty arthritis flares during initiation of allopurinol treatment: results of a double-blind, randomised study. Ann Rheum Dis 2011;70:1264-71. Dalbeth N, Ames R, Gamble GD, Horne A, Wong S, Kuhn-Sherlock B, et al. Effects of skim milk powder enriched with glycomacropeptide and G600 milk fat extract on frequency of gout flares: a proof-of-concept randomised controlled trial. Ann Rheum Dis 2012;71:929-34. Schumacher HR, Jr., Sundy JS, Terkeltaub R, Knapp HR, Mellis SJ, Stahl N, et al. Rilonacept (interleukin-1 trap) in the prevention of acute gout flares during initiation of urate-lowering therapy: results of a phase II randomized, double-blind, placebo-controlled trial. Arthritis Rheum 2012;64:876-84. Schumacher HR, Jr., Evans RR, Saag KG, Clower J, Jennings W, Weinstein SP, et al. Rilonacept (interleukin-1 trap) for prevention of gout flares during initiation of uric acid-lowering therapy: results from a phase III randomized, double-blind, placebo-controlled, confirmatory efficacy study. Arthritis Care Res 2012;64:1462-70. Mitha E, Schumacher HR, Fouche L, Luo SF, Weinstein SP, Yancopoulos GD, et al. Rilonacept for gout flare prevention during initiation of uric acid-lowering therapy: results from the PRESURGE-2 international, phase 3, randomized, placebo-controlled trial. Rheumatology (Oxford) 2013;52:1285-92. Sundy JS, Schumacher HR, Kivitz A, Weinstein SP, Wu R, King-Davis S, et al. Rilonacept for gout flare prevention in patients receiving uric acid-lowering therapy: results of RESURGE, a phase III, international safety study. J Rheumatol 2014;41:1703-11. Poiley J, Steinberg AS, Choi YJ, Davis CS, Martin RL, McWherter CA, et al. A randomized, double-blind, active- and placebo-controlled efficacy and safety study of arhalofenate for reducing flare in patients with gout. Arthritis Rheumatol 2016;68:2027-34. Yu YK, Yu F, Ye C, Shen GF, Lei XM, Zhang ST, et al. Combination therapies of diacerein and febuxostat inhibit il-1beta responses and improve clinical symptoms in patients with refractory gout. Am J Ther 2017;24:e290-7. Yamanaka H, Tamaki S, Ide Y, Kim H, Inoue K, Sugimoto M, et al. Stepwise dose increase of febuxostat is comparable with colchicine prophylaxis for the prevention of gout flares during the initial phase of urate-lowering therapy: results from FORTUNE-1, a prospective, multicentre randomised study. Ann Rheum Dis 2018;77:270-6. Schumacher HR, Jr., Becker MA, Wortmann RL, Macdonald PA, Hunt B, Streit J, et al. Effects of febuxostat versus allopurinol and placebo in reducing serum urate in subjects with hyperuricemia and gout: a 28-week, phase III, randomized, double-blind, parallel-group trial. Arthritis Rheum 2008;59:1540-8. Sundy JS, Becker MA, Baraf HS, Barkhuizen A, Moreland LW, Huang W, et al. Reduction of plasma urate levels following treatment with multiple doses of pegloticase (polyethylene glycol-conjugated uricase) in patients with treatment-failure gout: results of a phase II randomized study. Arthritis Rheum 2008;58:2882-91. Becker MA, Schumacher HR, MacDonald PA, Lloyd E, Lademacher C. Clinical efficacy and safety of successful longterm urate lowering with febuxostat or allopurinol in subjects with gout. J Rheumatol 2009;36:1273-82. Reinders MK, van Roon EN, Jansen TL, Delsing J, Griep EN, Hoekstra M, et al. Efficacy and tolerability of urate-lowering drugs in gout: a randomised controlled trial of benzbromarone versus probenecid after failure of allopurinol. Ann Rheum Dis 2009;68:51-6. Schumacher HR, Jr., Becker MA, Lloyd E, MacDonald PA, Lademacher C. Febuxostat in the treatment of gout: 5-yr findings of the FOCUS efficacy and safety study. Rheumatology (Oxford) 2009;48:188-94.

13

[30]

[31]

[32]

[33]

[34]

[35] [36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

Becker MA, Schumacher HR, Espinoza LR, Wells AF, MacDonald P, Lloyd E, et al. The uratelowering efficacy and safety of febuxostat in the treatment of the hyperuricemia of gout: the CONFIRMS trial. Arthritis Res Ther 2010;12:R63. Sundy JS, Baraf HS, Yood RA, Edwards NL, Gutierrez-Urena SR, Treadwell EL, et al. Efficacy and tolerability of pegloticase for the treatment of chronic gout in patients refractory to conventional treatment: two randomized controlled trials. JAMA2011;306:711-20. Becker MA, Baraf HS, Yood RA, Dillon A, Vazquez-Mellado J, Ottery FD, et al. Long-term safety of pegloticase in chronic gout refractory to conventional treatment. Ann Rheum Dis 2013;72:1469-74. Dalbeth N, Aati O, Gamble GD, Horne A, House ME, Roger M, et al. Zoledronate for prevention of bone erosion in tophaceous gout: a randomised, double-blind, placebocontrolled trial. Ann Rheum Dis 2014;73:1044-51. Huang X, Du H, Gu J, Zhao D, Jiang L, Li X, et al. An allopurinol-controlled, multicenter, randomized, double-blind, parallel between-group, comparative study of febuxostat in Chinese patients with gout and hyperuricemia. Int J Rheum Dis 2014;17:679-86. Holland R, McGill NW. Comprehensive dietary education in treated gout patients does not further improve serum urate. Intern Med J 2015;45:189-94. Xu S, Liu X, Ming J, Chen S, Wang Y, Liu X, et al. A phase 3, multicenter, randomized, allopurinol-controlled study assessing the safety and efficacy of oral febuxostat in Chinese gout patients with hyperuricemia. Int J Rheum Dis 2015;18:669-78. Perez-Ruiz F, Sundy JS, Miner JN, Cravets M, Storgard C, Group RS. Lesinurad in combination with allopurinol: results of a phase 2, randomised, double-blind study in patients with gout with an inadequate response to allopurinol. Ann Rheum Dis 2016;75:1074-80. Yu KH, Lai JH, Hsu PN, Chen DY, Chen CJ, Lin HY. Safety and efficacy of oral febuxostat for treatment of HLA-B 5801-negative gout: a randomized, open-label, multicentre, allopurinolcontrolled study. Scand J Rheumatol 2016;45:304-11. Bardin T, Keenan RT, Khanna PP, Kopicko J, Fung M, Bhakta N, et al. Lesinurad in combination with allopurinol: a randomised, double-blind, placebo-controlled study in patients with gout with inadequate response to standard of care (the multinational CLEAR 2 study). Ann Rheum Dis 2017;76:811-20. Dalbeth N, Jones G, Terkeltaub R, Khanna D, Kopicko J, Bhakta N, et al. Lesinurad, a Selective Uric Acid Reabsorption Inhibitor, in Combination With Febuxostat in Patients With Tophaceous Gout: Findings of a Phase III Clinical Trial. Arthritis Rheumatol 2017;69:1903-13. Dalbeth N, Saag KG, Palmer WE, Choi HK, Hunt B, MacDonald PA, et al. Effects of febuxostat in early gout: a randomized, double-blind, placebo-controlled study. Arthritis Rheumatol 2017;69:2386-95. Saag KG, Fitz-Patrick D, Kopicko J, Fung M, Bhakta N, Adler S, et al. Lesinurad combined with allopurinol: a randomized, double-blind, placebo-controlled study in gout patients with an inadequate response to standard-of-care allopurinol (a US-based study). Arthritis Rheumatol 2017;69:203-12. Stamp LK, Chapman PT, Barclay ML, Horne A, Frampton C, Tan P, et al. A randomised controlled trial of the efficacy and safety of allopurinol dose escalation to achieve target serum urate in people with gout. Ann Rheum Dis 2017;76:1522-8. Stamp LK, Chapman PT, Barclay M, Horne A, Frampton C, Tan P, et al. Allopurinol dose escalation to achieve serum urate below 6 mg/dL: an open-label extension study. Ann Rheum Dis 2017;76:2065-70. Tausche AK, Alten R, Dalbeth N, Kopicko J, Fung M, Adler S, et al. Lesinurad monotherapy in gout patients intolerant to a xanthine oxidase inhibitor: a 6 month phase 3 clinical trial and extension study. Rheumatology (Oxford) 2017;56:2170-8. Doherty M, Jenkins W, Richardson H, Sarmanova A, Abhishek A, Ashton D, et al. Efficacy and cost-effectiveness of nurse-led care involving education and engagement of patients and a

14

[47]

[48]

[49]

[50] [51] [52]

treat-to-target urate-lowering strategy versus usual care for gout: a randomised controlled trial. Lancet 2018;392:1403-12. Fitz-Patrick D, Roberson K, Niwa K, Fujimura T, Mori K, Hall J, et al. Safety and efficacy of verinurad, a selective URAT1 inhibitor, for the treatment of patients with gout and/or asymptomatic hyperuricemia in the United States and Japan: Findings from two Phase II trials. Mod Rheumatol 2018;18:1-27. Gunawardhana L, Becker MA, Whelton A, Hunt B, Castillo M, Saag K. Efficacy and safety of febuxostat extended release and immediate release in patients with gout and moderate renal impairment: phase II placebo-controlled study. Arthritis Res Ther 2018;20:99-108. Saag KG, Becker MA, Whelton A, Hunt B, Castillo M, Kisfalvi K, et al. Efficacy and safety of febuxostat extended and immediate release in patients with gout and renal impairment: phase III placebo-controlled study. Arthritis Rheumatol 2019;71:143-53. Wang S. The efficacy of febuxostat and allopurinol in the treatment of gout with hyperuricemia. Pak J Pharm Sci. 2018;31:1623-7. White WB, Saag KG, Becker MA, Borer JS, Gorelick PB, Whelton A, et al. Cardiovascular safety of febuxostat or allopurinol in patients with gout. N Engl J Med 2018;378:1200-10. Yu XN, Wu HY, Deng YP, Zhuang GT, Tan BH, Huang YZ, et al. "Yellow-dragon Wonderful-seed Formula" for hyperuricemia in gout patients with dampness-heat pouring downward pattern: a pilot randomized controlled trial. Trials 2018;19:551-60.

15

FIGURE LEGENDS

Figure 1. PRISMA flow chart

16

Figure 2. Modified-Jadad score for risk of bias assessment

17

TABLES Table 1. Gout flare measurement and data capture Intervention Anti-inflammatory prophylaxis studies (n = 10)

Gout flare definition

Wortmann (2010) [15]

Colchicine or naproxen while initiating urate lowering therapy

Gout flares requiring treatment

Schlesinger (2011) [16]

Canakinumab doseranging vs. colchicine

Not reported

Dalbeth (2012) [17]

Skim milk vs. lactose powder

Self-reported gout flares with pain at rest > 3 on a 0 - 10 numeric rating scale

Schumacher (2012a) [18]

Rilonacept vs. placebo

Schumacher (2012b) [19]

Rilonacept vs. placebo

Mitha (2013) [20]

Rilonacept vs. placebo

Sundy (2014) [21]

Rilonacept vs. placebo

Poiley (2016) [22]

Arhalofenate vs, allopurinol +/- colchicine vs. placebo

Yu (2017) [23]

Febuxostat +/- diacerein

Patient‐reported acute articular pain typical of a gout attack that required treatment with an anti-inflammatory agent. All of the following three criteria: (1) the presence of patient‐reported acute articular pain typical of a gout attack deemed by the patient and/or the investigator to require treatment with an anti-inflammatory therapeutic agent; (2) the presence of > 2 additional key signs/symptoms of joint swelling, redness, and tenderness; and (3) the presence of > 1 of the following additional clinical variables of rapid onset of pain, decreased range of motion, joint warmth, or other symptoms similar to a prior gout flare. A sub-analysis was done using the following definition: patient‐reported articular pain typical of a gout attack deemed to require treatment with an anti-inflammatory agent. All of the three following criteria: (1) the presence of patient-reported acute articular pain typical of a gout attack that is deemed (by patient and/or investigator) to require treatment with an anti-inflammatory therapeutic; (2) the presence of > 2 of the following three signs/symptoms: joint swelling, tenderness and redness; (3) and the presence of > 1 of the following clinical variables: rapid onset of pain, decreased range of motion, joint warmth or other symptoms similar to a prior gout flare. A sub-analysis was done using the definition: patient-reported articular pain typical of a gout attack that requires treatment with an antiinflammatory agent. Subject-reported acute articular pain typical of a gout attack that is deemed (by the subject and/or the investigator) to require treatment with an anti-inflammatory medication" All of the four following criteria: (1) an episode of patient‐reported acute articular or bursal pain that occurred at rest and was typical of past gout attacks; (2) the intensity of pain at rest had to be ≥4 on an 11‐point numerical rating scale, and the pain had to be determined by the patient and/or the investigator to require antiinflammatory/analgesic treatment; (3) presence of > 2 of 3 possible joint symptoms: swelling, warmth, or tenderness; (4) presence of > 1 of: rapid onset of pain, decreased range of joint motion, or joint redness. A pain score of > 3 on a visual analogue scale,

Method of gout flare data capture Retrospectively at study visits which occurred at week 2, week 4 and then monthly for total 12 months (FACT); at week 2, 4, 6, 8 and then monthly for next 20 weeks (APEX); and bimonthly for 6 months (CONFIRMS) In real-time using daily diaries for 24 weeks In real-time using daily diaries for 1 month prior to trial and 3 months of trial In real-time using a daily telephone interactive voice response system for 22 weeks

In real-time using a daily telephone interactive voice response system for 20 weeks

In real-time using a daily telephone interactive voice response system for 21 weeks

In real-time using daily diaries for 21 weeks

In real-time using daily electronic diaries for 14 weeks

In real-time via recording flare 18

Yamanaka (2018) [24]

Febuxostat +/- colchicine

redness, swelling, and warmth in the joints or adjacent soft tissues. Gout flares requiring analgesic treatment with NSAIDs or adrenal corticosteroid to manage symptoms (mild attacks that were tolerable without NSAID use were not defined as gout flares)

symptoms for 12 weeks In real-time on a specified patient record form when the flare occurred and verified at next study visit which occurred every 4 weeks for 24 weeks

Urate lowering therapies and other long-term intervention studies (n = 28 studies) Schumacher (2008) [25]

Febuxostat vs. placebo

Gout flares requiring treatment

Sundy (2008) [26]

Pegloticase dose-ranging

Defined at the discretion of the investigator

Becker (2009) [27]

Febuxostat vs. allopurinol

Gout flares requiring treatment

Reinders (2009) [28]

Benzbromarone vs. probenecid

Not reported

Retrospectively at study visits which occurred every 4 weeks for 28 weeks Not reported Retrospectively at study visits which occurred bimonthly for up to 40 months Not reported Retrospectively at study visits which occurred at week 2, 4, and then every 4 weeks through week 56. Visits during years 2-5 occurred every 8 weeks, except for end-ofeach-year visit which was 4 weeks after previous visit. Retrospectively at study visits which occurred bimonthly for 6 months In real-time by contacting investigators when flares occurred throughout the 25 weeks Retrospectively. Assessment time points not reported. Retrospectively at study visits which occurred at month 12 and month 24 Not reported Retrospectively at the final study visit at month 6 Retrospectively at study visits which occurred every 4 weeks for 24 weeks

Schumacher (2009) [29]

Febuxostat dose-ranging

Gout flares requiring treatment

Becker (2010) [30]

Febuxostat vs. allopurinol

Gout flares requiring treatment

Sundy (2011) [31]

Pegloticase vs. placebo

Acute joint pain and swelling requiring treatment

Becker (2013) [32]

Pegloticase dose-ranging

Not reported

Dalbeth (2014) [33]

Zoledronate vs. placebo

Not reported

Huang (2014) [34] Holland (2015) [35]

Febuxostat vs. allopurinol Intensive dietary advice vs. standard education

Gout flares requiring treatment

Xu (2015) [36]

Febuxostat vs. allopurinol

Gout flares requiring treatment

Lesinurad vs. placebo

Not reported

Not reported

Febuxostat vs. allopurinol

Not reported Clinically-relevant gout flares requiring either an increase in current medication or new medication. Permitted treatments were colchicine, analgesics and/or anti-inflammatory medications, including oral and intra-articular corticosteroids.

Not reported

Perez-Ruiz (2016) [37] Yu (2016) [38]

Not reported

Bardin (2017) [39]

Lesinurad vs. placebo

Dalbeth (2017a) [40]

Lesinurad vs. placebo

Gout flares requiring treatment

Dalbeth (2017b) [41]

Febuxostat vs. placebo

Gout flares requiring treatment

Saag (2017) [42]

Lesinurad vs. placebo

Gout flares requiring treatment. Treatments included colchicine, analgesics, and/or antiinflammatory medications, including oral and intraarticular corticosteroids.

Stamp (2017a) [43]

Allopurinol dose escalation vs. dose maintenance

Not reported

Stamp (2017b) [44]

Allopurinol dose escalation

Not reported

Tausche (2017) [45]

Lesinurad vs. placebo

Gout flares requiring treatment

In real-time using daily electronic diaries for 12 months

In real-time using daily electronic diaries for 15.5 months In real-time by contacting investigators when gout flare occurred for 24 months In real-time using daily electronic diaries for 12 months and 28 days Retrospectively via monthly phone calls and at study visits which occurred every 3 months for 12 months Retrospectively via monthly phone calls and at study visits which occurred every 3 months for 12 months In real-time using daily diaries for 6 months + 14 days

19

Doherty (2018) [46]

Nurse-led care vs. GP-led care

Not reported

Fitz-Patrick (2018) [47]

Verinurad vs. placebo

Not reported

Gunawardhana (2018) [48]

Febuxostat (extended and immediate release) vs. placebo

Saag (2018) [49]

Febuxostat (extended and immediate release) vs. placebo

Wang (2018) [50] White (2018) [51] Yu (2018) [52]

Febuxostat vs. allopurinol Febuxostat vs. allopurinol Yellow-dragon Wonderful-seed Formula (YWF) vs. allopurinol

All of the three following criteria: (1) patientreported acute articular pain typical of a gout attack that was deemed (by patient and/or investigator) to require treatment (and was treated) with colchicine, NSAIDs, or steroids; (2) presence of > 3 of: joint swelling, redness, tenderness, and pain; (3) and patient experienced > 1 of rapid onset of pain, decreased range of motion, joint warmth, and other symptoms similar to a prior gout flare. Patients who discontinued the double-blind study drug due to gout flares were considered to have experienced a gout flare during the 3month treatment period. All of the three following criteria: (1) patientreported acute articular pain typical of a gout attack that was deemed (by patient and/or investigator) to require treatment (and was treated) with colchicine, NSAIDs, or steroids; (2) presence of > 3 of joint swelling, redness, tenderness, and pain; (3) and patient experienced > 1 of rapid onset of pain, decreased range of motion, joint warmth, and other symptoms similar to a prior gout flare. Patients who discontinued the double-blind study drug due to gout flares were considered to have experienced a gout flare during the 3month treatment period. Not reported Not reported Not reported

In real-time using daily diaries for 24 months Retrospectively at study visits which occurred at week 1, 2, 3, 6, 8, then every 4 weeks until week 24, then at week 26

Retrospectively at study visits which occurred at week 2, month 1, month 2 and month 3

Not reported

Not reported Not reported Not reported

20

Table 2. Gout flare reporting Outcome level

Treatment efficacy or adverse event

Total duration of intervention

Time period(s) used to report the flare results

52 weeks (FACT); 28 weeks (APEX); 6 months (CONFIRMS)

Each 4-week period from week 0 to 16 and for week 48 to 52 (FACT) First 12 weeks and last 1 week of 28 week period (APEX) First 4 week period and last 4 week period (CONFIRMS)

Anti-inflammatory prophylaxis studies (n = 10)

Wortmann (2010) [15]

Primary

Treatment efficacy

Schlesinger Primary and Treatment efficacy 16 weeks (2011) [16] secondary Dalbeth (2012) Primary and Treatment efficacy 3 months [17] secondary Schumacher Primary and Treatment efficacy 16 weeks (2012a) [18] secondary Schumacher Primary and Treatment efficacy 16 weeks (2012b) [19] secondary Mitha (2013) Primary and Treatment efficacy 16 weeks [20] secondary Sundy (2014) Secondary Treatment efficacy 16 weeks [21] Poiley (2016) Primary Treatment efficacy 12 weeks [22] Yu (2017) [23] Primary Treatment efficacy 12 weeks Yamanaka (2018) Primary and Treatment efficacy 12 weeks [24] secondary Urate lowering therapies and other long-term intervention studies (n = 28 studies) Schumacher Secondary Treatment efficacy 28 weeks (2008) [25] Sundy (2008) Not reported Treatment efficacy 12-14 weeks [26] Becker (2009) Up to 40 Secondary Treatment efficacy [27] months Reinders (2009) Not reported Adverse event 2 months [28] Schumacher Additional Treatment efficacy 60 months (2009) [29] Becker (2010) Not reported Treatment efficacy 6 months [30] Sundy (2011) Treatment efficacy & Secondary 6 months [31] adverse event Becker (2013) Treatment efficacy & Up to 36 Not reported [32] adverse event months Dalbeth (2014) Treatment efficacy & Exploratory 24 months [33] adverse event Huang (2014) Secondary Treatment efficacy 28 weeks [34] Holland (2015) Secondary Treatment efficacy 6 months [35] Xu (2015) [36] Not reported Treatment efficacy 24 weeks Perez-Ruiz (2016) Not reported Adverse event 4 weeks [37] Yu (2016) [38] Not reported Treatment efficacy 12 weeks Bardin (2017) Secondary Treatment efficacy 12 months [39] Dalbeth (2017a) Secondary Treatment efficacy 12 months [40] Dalbeth (2017b) Additional Treatment efficacy 24 months [41] Saag (2017) [42] Secondary Treatment efficacy 12 months Stamp (2017a) Secondary Treatment efficacy 12 months [43] Stamp (2017b) Other Treatment efficacy 24 months [44] Tausche (2017) Secondary Treatment efficacy 6 months

Total 16 week period Each 1 month period for 3 months and total 3 month period Each 4 week period for 16 weeks, 6 week period from week 16 to 22 and, total 22 week period 16 week period from week 0 to 16 and in 4 week period from week 16 to 20 Total 16 week period First 4 week period and total 16 week period Total 12 week period Each 4 week period for 12 weeks Week 0 to 12, week 13 to 24 and for total 24 week period 8 week period from week 0 to 8 and 20 week period from week 8 to 28 Total 12 to 14 week period Each 2-month period for 40 months Total 2 month period Each 2 month period for 60 months and total 60 month period Each 2-month period for 6 months Each 3 month period for 6 months and total 6 month period First 3-month period and total 30-month period Previous 3 months for each 12 month period for 24 months and total 24 month period A 20 week period from week 9 through 28 Total 6 month period Each 4 week periods for 24 weeks Total 4 week period Total 12 week period 6 month period from month 6 to month 12 6 month period from month 6 to month 12, and the first and last 1 month periods of the total 12 months Each 6 month period for 24 months, and for total 24 months 6 month period from month 6 to month 12 Total 12 month period Previous 1 month for each 12 month period for 24 months Total 6 month period 21

[45] Doherty (2018) [46] Fitz-Patrick (2018) [47] Gunawardhana (2018) [48] Saag (2018) [49] Wang (2018) [50] White (2018) [51] Yu (2018) [52]

Secondary

Treatment efficacy

24 months

Each 12 month period for 24 months

Not reported

Treatment efficacy

24 weeks

Total 24 week period

Secondary

Treatment efficacy

3 months

Total 3 month period

Secondary Not reported

Treatment efficacy Not reported

Total 3 month period Total 6 month period

Not reported

Treatment efficacy

Not reported

Not reported

3 months 6 months 24 months (median) 4 weeks

Not reported Total 4 week period

22

Table 3. Gout flare reporting methods used for each study Number of flare reporting methods used Anti-inflammatory prophylaxis therapies (n = 10 studies) Wortmann (2010) [15] 1 Schlesinger (2011) [16]

4

Dalbeth (2012) [17]

2

Schumacher (2012a) [18]

7

Schumacher (2012b) [19]

8

Mitha (2013) [20]

7

Sundy (2014) [21]

5

Poiley (2016) [22]

5

Yu (2017) [23]

2

Yamanaka (2018) [24]

2

Reporting method a Proportion of patients with > 1 gout flare for each intervention group Mean number of gout flares per patient for each intervention group Proportion of patients with > 1 gout flare for each intervention group Time to first flare per patient for each intervention group (hazard ratio) Mean duration of gout flares (days) per patient for each intervention group Mean number of gout flares per patient for each intervention group Mean pain scores (on a 10 point Likert scale) during gout flares for total group and for each intervention group Mean number of gout flares per patient for each intervention group Total number of gout flares for each intervention group Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with 1 gout flare for each intervention group Proportion of patients with > 2 gout flares for each intervention group Mean number of gout flare days per patient for each intervention group Mean number of days with a pain severity score of > 5 (on a 0-10 NRS) per patient for each intervention group Total number of gout flares for each intervention group Mean number of gout flares per patient for each intervention group Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with > 2 gout flares for each intervention group Proportion of patients with 1 gout flare for each intervention group Mean number of gout flare days per patient for each intervention group Mean duration of gout flares (days) per patient for each intervention group Mean number of days with a pain severity score of > 5 (on a 0-10 NRS) per patient for each intervention group Total number of gout flares for each intervention group Mean number of gout flares per patient for each intervention group Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with > 2 gout flares for each intervention group Mean number of gout flare days per patient for each intervention group Mean number of days with a pain severity score of > 5 (on a 0-10 NRS) per patient for each intervention group Median number of days to onset of first flare per patient for each intervention group Mean number of gout flares per patient for each intervention group Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with > 2 gout flares for each intervention group Mean number of gout flare days per patient for each intervention group Median number of days to onset of first flare per patient for each intervention group Mean number of gout flares per patient for each intervention group Proportion of patients with no gout flares for each intervention group Proportion of patients with > 3 flares for each intervention group Median duration of gout flares (days) per patient for each intervention group Median number of days to onset of first flare per patient for each intervention group Mean number of gout flares per patient for each intervention group Mean pain due to gout flares in the past week on a 10 cm VAS per patient for each intervention group Proportion of patients with > 1 gout flare for each intervention group Mean number of gout flares per patient for each intervention group

Urate lowering therapies and other long-term intervention studies (n = 28 studies) Schumacher (2008) [25] 1 Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with > 1 gout flare for total group and for each Sundy (2008) [26] 1 intervention group Becker (2009) [27] 1 Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with > 1 gout flare for each intervention group Reinders (2009) [28] 2 Proportion of patients who withdrew from the study because of a gout flare Proportion of patients with > 1 gout flare for total group and for each Schumacher (2009) [29] 1 intervention group Becker (2010) [30] 1 Proportion of patients with > 1 gout flare for each intervention group Mean number of gout flares per patient for each intervention group Sundy (2011) [31] 2 Proportion of patients with > 1 gout flare for each intervention group 23

Becker (2013) [32]

2

Dalbeth (2014) [33]

4

Huang (2014) [34] Holland (2015) [35] Xu (2015) [36] Perez-Ruiz (2016) [37] Yu (2016) [38] Bardin (2017) [39]

1 1 1 1 1 1

Dalbeth (2017a) [40]

2

Dalbeth (2017b) [41] Saag (2017) [42] Stamp (2017a) [43] Stamp (2017b) [44] Tausche (2017) [45]

1 1 1 1 1

Doherty (2018) [46]

3

Proportion of patients with > 1 gout flare for total group and for non-responders and responders in RCT Mean number of gout flares per patient for total group Mean number of gout flares per patient for each intervention group Proportion of patients with gout flares resulting in hospital admission for each intervention group Proportion of patients with > 1 gout flare for each intervention group Total number of gout flares for each intervention group Proportion of patients with > 1 gout flare for each intervention group Mean number of gout flares per patient for each intervention group Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with > 1 gout flare for each intervention group Mean number of gout flares per patient for each intervention group Mean number of gout flares per patient for each intervention group Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with > 1 gout flare for each intervention group Mean number of gout flares per patient for each intervention group Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with > 1 gout flare for each intervention group Proportion of patients with > 2 gout flares for each intervention group Proportion of patients with > 4 gout flares for each intervention group Mean number of gout flares per patient for each intervention group Proportion of patients with > 1 gout flare for each intervention group

Fitz-Patrick (2018) [47] 1 Gunawardhana (2018) 1 Proportion of patients with > 1 gout flare for each intervention group [48] Saag (2018) [49] 1 Proportion of patients with > 1 gout flare for each intervention group Wang (2018) [50] 1 Proportion of patients with > 1 gout flare for each intervention group White (2018) [51] 1 Mean number of gout flares per patient for each intervention group Yu (2018) [52] 1 Total number of gout flares for each intervention group a Proportion indicates percentage and/or number of patients

Table 4. Number of studies using each gout flare reporting method

Proportion of patients with gout flares

Method

Anti-inflammatory prophylaxis studies (n = 10)

Urate lowering / other long term therapy studies (n = 28)

Proportion of patients with > 1 gout flare

7

22

Proportion of patients with > 2 gout flares

4

1

Proportion of patients with > 3 gout flares

1

0

Proportion of patients with > 4 gout flares

0

1

Proportion of patients with 1 gout flare

2

0

Proportion of patients with no gout flares

1

0

0

1

0

1

Mean number of gout flares

9

9

Number of gout flares (individual patient data shown)

1

0

Total number of gout flares in each group

3

2

Mean number of gout flare days

4

0

Median number of days to onset of first flare

4

0

Mean duration of gout flares (days)

2

0

Median duration of gout flares (days)

1

0

Mean pain score due to gout flares in the past week (on a 10 cm VAS)

1

0

Proportion of patients who withdrew from the study due to a gout flare Proportion of patients requiring hospital admission for a flare Number of gout flares per patient Number of gout flares per group Total number of days in flare per patient Time to first flare per patient Gout flare duration per patient Gout flare pain per patient

24

Mean number of days with a pain severity score > 5 (on a 0-10 NRS) Mean pain scores (on a 10-point Likert scale) during gout flares

3

0

1

0

25