J. Branco

SAT0368 Outcomes Assessed in Trials of Gout and Accordance with Omeract Recommendations

Background The OMERACT recommendations defined outcomes to be assessed in acute and chronic gout trials in 2005. Objectives With the increasing number of trials investigating new treatments for gout, the aim of our study was to assess whether the OMERACT recommendations for acute and chronic gout trials have been adopted. Methods We searched Medline, EMBASE and Cochrane databases as well as EULAR and ACR abstracts (2010-11) to identify all randomized clinical trials (RCTs) and quasi-RCTs on any intervention in adult patients with gout. We assessed risk of bias (RoB) according to Cochrane methods. We extracted all outcomes reported in all studies and categorized them as either an OMERACT recommended outcome domain for acute trials, for chronic trials, or a non-OMERACT domain. We compared the number of OMERACT outcomes included in trials according to RoB (low vs unclear and high) and according to recruitment starting before and after the publication of the OMERACT guidelines to determine their impact. Results We screened 9517 articles and 113 ACR/EULAR abstracts and 64 publications fulfilled inclusion criteria (36 acute, 28 chronic trials). Table 1 displays the most commonly reported outcomes. Acute gout trials reported an average of 3.0 (SD 1.1) OMERACT outcomes (out of 5 recommended) and chronic gout trials reported on average of 2.4 (SD 1.0) OMERACT outcomes (out of 9 recommended). Joint damage imaging, health-related quality of life (HR-QoL), participation and musculoskeletal function were not assessed in any chronic gout trial, while patient global assessment was only reported in 33% of acute and 4% of chronic trials. Function was reported in 22% of acute gout trials. There was no significant difference between low RoB studies and high or unclear RoB studies in number of OMERACT domains reported (mean (SD): acute gout trials: 3.1 (1.1) vs 2.7 (1.2), P=0.329; chronic gout trials: 2.9 (1.1) vs 2.5 (1.1), P=0.331) or between trials recruiting participants before and after publication of OMERACT guidelines (mean (SD): acute gout trials: 3.0 (1.1) vs 2.0 (1.0), P=0.144; chronic gout trials: 2.7 (1.1) vs 2.3 (1.2), P=0.507). Conclusions There is substantial heterogeneity in the measurement of outcomes in clinical trials of gout. To date there has been no appreciable impact of the published OMERACT recommendations in the assessment of outcomes in more recent gout trials. Function and patient global assessment is seldom assessed while participation and HR-QoL were not reported. Disclosure of Interest None Declared

AB0581 Two years existence of reuma.pt/vasculitis – the portuguese registry of vasculitis

Background The vasculitides are a group of relatively rare diseases with different manifestations and outcomes. New therapeutic options have led to the need for long-term registries. The Rheumatic Diseases Portuguese Register, Reuma.pt, is an electronic clinical record, created in 2008, which currently includes specific protocols for 11 diseases and >16000 patients registered from 79 national and international rheumatology centres. Since October 2014, a dedicated protocol to vasculitis has been created as part of the European Vasculitis Society initiative of having compatible European registries. Objectives To describe the structure of Reuma.pt/Vasculitis and characterize the patients registered over the last two years. Methods We developed a dedicated web-based software to enable prospective collection and central storage of anonymised data from patients with vasculitis. All Portuguese rheumatology centres were invited to participate. Data regarding demographics, diagnosis, classification criteria, imaging and laboratory tests, outcome measures of prognosis, damage, disease activity and quality of life, and treatment were collected. We performed a cross-sectional descriptive analysis of all patients registered up to January 2017. Results A total of 492 patients, with 1114 visits, from 11 centres were registered in Reuma.pt/Vasculitis. The mean age was 53±20 years at last visit; 68% were females. The diagnoses followed the 2012 Chapel Hill Consensus nomenclature (Table 1). The most common diagnoses were Behçets disease (BD) (39%) and giant cell arteritis (GCA) (20%). Patients with BD met the International Study Group 1990 criteria, the International Criteria for BD 2006 and 2013 in 84%, 95% and 95% of cases, respectively. Patients with GCA met the 1990 American College of Rheumatology criteria in 95% of cases. Data on vascular ultrasound was available in 74% of patients; 73% compatible with the diagnosis. Assessment of the Birmingham Vasculitis Activity Score (BVAS) and Vasculitis Damage Index (VDI) was available for all vasculitides and the Five Factor Score calculation of survival rate for ANCA associated vasculitis (AAV) and polyarteritis nodosa (PAN). The mean BVAS at first visit was 18±7 for AAV and 15±9 for PAN; the mean VDI at last visit was 3±2 for AAV and 2±2 for PAN. Health related quality of life assessments (SF-36, EQD5, FACIT and HADS) were also collected. Treatment registry with the disease assessment variables shown in graphics was available for all patients; only 6% were under biologic treatment. Conclusions Reuma.pt/Vasculitis is a registry adapted for routine care, allowing an efficient data repository at a national level with the potential to link with other international databases. It facilitates research, trials recruitment, service planning and benchmarking. Disclosure of Interest None declared

OP0031 Tocilizumab is Associated with Higher CDai/Sdai Remission in Biologic-Naïve Rheumatoid Arthritis Patients – Data from Reuma.Pt

Background Tocilizumab (TCZ), an interleukin-6 receptor blocker, and anti-tumor necrosis factor (TNF) biologic agents are key therapies in the management of rheumatoid arthritis (RA). They are considered to be equally effective and very few head-to-head comparisons have been published. Objectives To compare remission rates in RA patients treated with anti-TNF agents and TCZ and assess the impact of previous biologic therapies in treatment response. Methods We included RA patients registered in the Rheumatic Diseases Portuguese Register, Reuma.pt, who started anti-TNF or TCZ after January 1, 2008, were treated for at least 6 months and had available DAS28 scores at baseline and at 6 months. Our primary outcome was the proportion of patients who achieved remission at 6 months by DAS28, CDAI, SDAI and Boolean remission criteria. Logistic regressions were performed to compare the groups and subgroup analyses of biologic-naïve patients were conducted. Results 524 RA patients were enrolled, (106 adalimumab, 202 etanercept, 43 golimumab, 78 infliximab, 95 TCZ). At baseline, the groups were similar except for proportion of biologic-naïve patients (lower in TCZ group, p<0.0001) and mean DAS28, CDAI and swollen joint count, all higher in the TCZ group (respectively: p=0.0005, p=0.037 and p<0.0001). At 6 months, more TCZ-treated patients were in DAS28 remission, with no differences for CDAI, SDAI or Boolean remission. Considering only naïve patients, DAS28, CDAI and SDAI remission were significantly higher in the TCZ group compared to anti-TNF, with similar rates of Boolean remission. This was confirmed in the multivariate logistic regression, adjusting for age, gender, number of previous biologics and baseline disease activity: DAS28 OR 10.8 (5.9-19.7), CDAI OR 2.9 (1.3-6.5), SDAI OR 4.1 (1.8-9.5), Boolean OR 1.9 (0.88-4.3). Table 1. Proportion of patients in remission according to different criteria and biologic class Anti-TNF Tocilizumab OR (95% CI) Overall Population  DAS28 (n=524) 102/429 (23.8) 55/95 (57.9) 4.4 (2.8–7.0)  CDAI (n=327) 36/260 (13.9) 14/67 (20.9) 1.6 (0.8–3.2)  SDAI (n=298) 33/239 (13.8) 14/59 (23.7) 1.9 (0.97–3.9)  Boolean (n=468) 42/358 (11.7) 11/83 (13.3) 1.1 (0.6–2.3) Biologic-naïve patients  DAS28 (n=417) 89/365 (24.4) 37/52 (71.2) 7.7 (4.0–14.5)  CDAI (n=258) 33/223 (14.8) 11/35 (31.4) 2.6 (1.2–5.8)  SDAI (n=237) 32/206 (15.5) 11/31 (35.5) 2.99 (1.33–6.76)  Boolean (n=348) 36/302 (11.9) 8/46 (17.4) 1.6 (0.7–3.5) Conclusions TCZ treatment was associated with higher rate of DAS28 remission at 6 months and previous biologic therapy significantly affected CDAI/SDAI remission. Naïve patients treated with TCZ had better DAS28, CDAI and SDAI remission rates compared to those treated with TNF inhibitors, whereas the more stringent Boolean remission was similar among all groups. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.2668

AB0451 Changes in DAS28, CDAI and SDAI are Associated with Biologic Class, Gender, Previous Biologic Therapy and ACPA/RF Status – Results from Reuma.PT

Background Tocilizumab (TCZ) and anti-tumor necrosis factor (TNF) biologic agents are key therapies in the management of rheumatoid arthritis (RA). They are considered to be equally effective and very few head-to-head comparisons have been published. Objectives To compare response to therapy in RA patients treated with anti-TNF agents and TCZ according to different response measures and determine the factors influencing it. Methods We included RA patients registered in the Rheumatic Diseases Portuguese Register, Reuma.pt, who started anti-TNF or TCZ after January 1, 2008, were treated for at least 6 months and had available DAS28 scores at baseline and at 6 months. Our primary outcome was the change in DAS28, CDAI and SDAI at 6 months. We performed linear regressions to compare the groups and determined the best model predicting change in disease activity for each index. Results 524 RA patients were enrolled, (106 adalimumab, 202 etanercept, 43 golimumab, 78 infliximab, 95 TCZ). At baseline, TCZ users were less frequently naïve to biologic therapies (54.7% vs. 85%, p<0.0001), had more swollen and tender joint counts (p<0.0001 and p=0.02, respectively) and higher disease activity according to all indexes: DAS28 6.1±1.1 vs. 5.4±1.3 (n=524, p<0.0001), CDAI 33.3±13.2 vs. 28.1±13.6 (n=376, p=0.005), SDAI.35.6±13.1 vs. 29.1±30.4 (n=361, p=0.004). At 6 months, change in DAS28, CDAI, SDAI and joint counts was significantly higher in the TCZ group (Table 1). Multivariate linear regression models best predicting change in disease activity included biologic class, number of previous biologics, baseline activity, gender and ACPA/RF status (Table 2). Compared to anti-TNF, TCZ was associated with a larger difference in ΔDAS28, ΔCDAI and ΔSDAI of, respectively, 1.45, 4.25 and 5.41. Table 1. Baseline and change in disease activity according to biologic class (Mann-Whitney test) Change at 6 months Anti-TNF (n=429) Tocilizumab (n=95) p-value ΔDAS28 1.8 (1.4) 3.3 (1.6) <0.0001 ΔCDAI (n=327) 16.0 (13.6) 22.7 (15.7) 0.0003 ΔSDAI (n=298) 17.1 (14.8) 25.2 (16.5) 0.0001 ΔSJC 4.7 (4.8) 7.8 (6.6) <0.0001 ΔTJC 6.4 (7.2) 8.7 (7.7) 0.005 Table 2. Multivariate linear regression models predicting 6-months change in disease activity ΔDAS28 (n=524) ΔCDAI (n=286) ΔSDAI (n=260) Adjusted-R2 0.391 0.613 0.559 Covariables β-coefficient (p) β-coefficient (p) β-coefficient (p) Biologic class (TCZ) 1.45 (<0.0001) 4.25 (0.004) 5.41 (0.003) No. previous biologics −0.41 (<0.0001) −2.47 (0.002) −2.78 (0.004) Baseline activity 0.54 (<0.0001) 0.79 (<0.0001) 0.77 (<0.0001) Female gender −0.40 (0.02) −1.74 (0.29) −1.64 (0.41) ACPA/RF positivity −0.45 (0.004) −3.65 (0.01) −4.77 (0.008) Conclusions TCZ treatment was associated with greater change in DAS28, CDAI, SDAI and joint counts at 6 months. Biologic class, number of previous biologics, baseline activity, gender and ACPA/RF status predicted change in disease activity. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.3414

FRI0562-HPR Effectiveness of the Introduction of Proprioceptive Training in the Exercise Programs for Knee Osteoarthritis

Background Rheumatic diseases are a major social and economic problem, with increasing negative impact on public health. The prevalence of Knee Osteoarthritis (KO) among older people is increasing and it seems to be associated with proprioceptive deficits. Objectives To investigate the effectiveness of the introduction of proprioceptive training in the current program guidelines-eight week duration (Arthritis Foundation, 2009), on range of motion, pain and stiffness of the knee, as well as in physical function and quality of life of clients with knee osteoarthritis. Methods A quasi-experimental study of eight weeks was implemented, with two independent samples of subjects with knee osteoarthritis (class II-II Kellgren-Lawrence scale) aged 50 and older. It is a controlled non randomized study, with a convenience sample of 42 subjects. Subjects in the experimental group (n=21) underwent the current exercise program guidelines (Arthritis Foundation, 2009) with the additional proprioceptive training, whilst the control group (n=21) attended the current exercise program guidelines (Arthritis Foundation, 2009). Both programes were defined according to the guidleines proposed by the Arthritis Foundation (2009) for this condition. Assessment was carried out at baseline and at the end of eight weeks, on the following outcomes: range of motion of the knee (goniometer), pain, stiffness and function and quality of life [Knee and Osteoarthritis Outcome Score (KOOS)]. Results The sample consisted of 42 caucasian subjects, mostly females (83.3%) with an average age of 72.1 years (SD=7.23). At the end of the progamme there was a statistically significant improvement of all outcomes in both groups. The introduction of proprioceptive training in the current program guidelines introduces a significantly higher evolution on the knees rom (extension - p<0.05) on the sport/leisure function (p<0.0001) and on the quality of life of participants (p>0.050). Conclusions The results obtained suggest that exercise training introduces a positive effect on the most relevant symptoms, function and quality of life of subjects with knee osteoarthritis. The introduciton of of proprioceptive trainning appears to additionally improve some of the most relevant outcomes for patients with KO, according to the Arthritis Foundation, 2009: knee extension range of motion, sports and leasure and quality of life. For this reason we suggest that its introduction in the exercise plan should be considered. References Foundation, A. (2009). Arthritis Foundation Exercise Program. Instructors Manual. Atlanta: Arthritis Foundation. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.5083

SAT0372 The Burden of Gout in Patients Included in Clinical Trials: A Systematic Review

Background The impact of gout on patients’ lives is still largely unknown. Objectives We performed a systematic review of the baseline characteristics of participants of acute and chronic gout trials as an indirect measure of the burden of gout. Methods We searched Medline, EMBASE and Cochrane databases as well as EULAR and ACR abstracts (2010-11) to identify all randomized clinical trials (RCTs) and quasi-RCTs on any intervention in adult patients with gout. We assessed risk of bias (RoB) according to Cochrane methods. To assess burden, we extracted all reported baseline characteristics, and how each domain was measured. For each included trial we quantified the number of OMERACT defined domains that had been included and reported at baseline. We compared the number of OMERACT baseline domains included in trials according to RoB (low vs unclear and high), and according to recruitment starting before and after the publication of the OMERACT guidelines. Results We screened 9517 articles and 113 ACR/EULAR abstracts. Sixty-four publications fulfilled inclusion criteria (36 acute gout trials (AT), 28 chronic gout trials (CT)). Burden was incompletely captured across trials; there was no uniformity in what baseline characteristics were reported; many different measures were used to capture the same domain, as presented in more detail in Table 1. Certain OMERACT domains were seldom reported, such as patient global assessment of disease (AT: 3 (7.9%); CT: 0), disability (AT: 3 (7.9%); CT: 3 (7.1%)) and health-related quality of life (AT: 2 (5.3%), CT: 0). There was a statistically significant trend for CT with low RoB (LRoB) to report more baseline characteristics in comparison to studies judged to be at unclear or high RoB (UHRoB) (LRoB: 3.6 (1.4) vs. UHRoB: 2.0 (1.4), p=0.017). This tendency came close to statistical significance in AT (LRoB: 1.9 (1.2) vs. UHRoB: 1.2 (1.0), p=0.078). Similarly, there was a non significant trend for trials recruiting participants after publication of the OMERACT guidelines to report more baseline characteristics (AT: 2.0 (1.7) vs. 1.4 (1.1), p=0.410; CT: 3.5 (1.9) vs. 2.2 (1.4), p=0.129). Image/graph Conclusions Burden of gout in both acute and chronic trials is incompletely captured and using heterogeneous measures. While there was a trend towards more complete assessment of core domains of burden after the publication of OMERACT recommendations, many domains were still absent or poorly reported. Disclosure of Interest None Declared

AB0631 Efficacy and safety of urine alkalinization for patients with uric acid nephrolithiasis with or without gout arthritis:. a systematic review

Background Gout is a complex metabolic and inflammatory disease with varying clinical presentations including gouty arthritis, uric acid nephrolithiasis and renal impairment. Urine alkalinization may be a useful adjunct in the management of gout. As part of the 3e initiative for generating recommendations for the diagnosis and management of gout, we performed a systematic review using Cochrane methods to determine the efficacy and safety of urine alkalinization in patients with uric acid nephrolithiasis with or without gouty arthritis. Objectives Determine the efficacy and safety of urine alkalinization in patients with uric acid nephrolithiasis with or without gouty arthritis. Methods We searched Medline, EMBASE and Cochrane databases to March 2012, and 2010-11 ACR/EULAR abstracts to identify all randomized controlled trials (RCTs) and quasi-RCTs that compared urine alkalinization to placebo or another therapy in people with uric acid nephrolithiasis with or without gouty arthritis. Primary outcomes were uric acid stone regression and withdrawals due to adverse effects. Two review authors independently selected studies for inclusion, assessed risk of bias (RoB) and extracted data using Cochrane methods. Results A total of 7103 articles were identified, of which 76 articles were selected for detailed review and two fulfilled inclusion criteria. One trial (60 participants) was judged to be at low risk of bias (RoB) and compared potassium citrate to phytotherapy. At 12 weeks, 14/30 (47%) people who received potassium citrate achieved uric acid stone remission compared to 9/30 (30%) in the phytotherapy group (P=0.05). The other trial (191 participants) was judged to be at unclear RoB and compared four treatment arms: potassium citrate and tamsulosin versus placebo or potassium citrate or tamsulosin. At 4 weeks, 27/46 (59%) who received potassium citrate achieved stone remission compared to 12/46 (26%) in the placebo group (P=0.003). There were no withdrawals in either trial due to adverse events. Conclusions There is limited evidence from two trials (1 at low RoB and 1 at unclear RoB) that urine alkalinization may be an efficacious and safe treatment for patients with uric acid nephrolithiasis with or without gouty arthritis. Disclosure of Interest None Declared