The purpose of this study was to determine whether long-term modulation of inflammatory activity by tumor necrosis factor (TNF)-α inhibitors has some influence on insulin resistance (IR). 16 active rheumatoid arthritis (RA) patients without CV risk factors treated with anti-TNF-α agents were included in this study. RA activity by disease activity score 28, IR by HOMA2-IR, body composition by impedance analysis, physical activity by accelerometry, abdominal fat distribution by magnetic resonance imaging, and serum level of key adipokines by ELISA were measured at baseline and during a 1-year follow-up period. Patient body mass index increased significantly (26.94 ± 3.88 vs. 28.06 ± 4.57 kg/m2, p=0.02) after 1 year of treatment. Body composition, in terms of fat and fat-free mass, remained unchanged except for a significant elevation in body cell mass (25.50 ± 4.60 vs. 26.60 ± 3.17 kg, p=0.02). Basal levels of IR in the RA patients included in this study were significantly higher than healthy controls (1.6 ± 0.8 vs. 1.11 ± 0.56, p=0.011) but did not change during the follow-up. Nor did basal concentrations of adiponectin, visfatin, leptin, ghrelin, resistin, and apelin in response to anti-TNF-α treatment; only retinol-binding protein 4, showed a significant increase (51.7 ± 32.7 vs. 64.9 ± 28.4 μg/ml, p=0.03) at the end of the study. IR, adiposity distribution, and serum levels of most adipokines are not significantly affected by long-term inhibition of TNF-α in RA patients. Our data suggest that although systemic blockade of TNF-α exerts an anticachectic effect in RA patients, it does not seem to play a major role in IR.

Objectives It has been suggested that resistance to insulin action is a feature that accompanies rheumatoid arthritis (RA), a disease where chronic inflammation predominates and classical triggers for insulin resistance (IR) like obesity and diabetes are absents. However, data regarding characterization of RA features associated with this insulin resistance (IR) are lacking. The aim of this study was to investigate how sensitivity to insulin and markers of beta cell dysfunction (proinsulin processing metabolites) are expressed in RA. Methods 101 non-diabetic RA patients and 99 non-diabetic sex and age-matched controls were included in this study. Insulin sensitivity function through homeostatic model assessment (HOMA2), and beta cell secretion through insulin, split and intact proinsulin, and C-peptide were assessed in both groups. We performed multiple regression analysis to compare IR between groups and to explore the relation between RA features and IR. Data were adjusted for glucocorticoids intake and for IR classical risk factors. Results RA patients compared to controls show higher HOMA-IR (logHOMA-IR, beta coefficient, 0.40 [95% CI 0.20-0.59], p=0.00). When this data was adjusted for glucocorticoids intake, non-on corticosteroid patients maintained a higher IR index (beta coef. 0.14 [95% CI 0.05-0.24], p=0.00). Current prednisone treatment was not associated with higher IR in the patients’ intragroup comparison (beta coef. 0.56 [1.13-1.19], p=0.22). Insulin processing signaling in RA patients showed impaired features via an elevated intact proinsulin levels (beta coef. 3.13 [0.81-5.44] pMol/L, p=0.03 for the comparison between patients and controls). Split proinsulin levels were also higher in RA patients (beta coef. 13.7 pmol/L [3.57-9.40], p=0.00) even adjusting for prednisone intake (beta coef. 2.66 pmol/L [95%CI 1.62-5.95] for non steroids RA patients when compared to controls). In multiple regression analysis, RA features (disease duration, rheumatoid factor, erythrocyte sedimentation rate, C-reactive protein, disease activity through HAQ and DAS28 scores, and current non-biologic disease modifying antirheumatic drug), when adjusted for sex, age and body mass index, were not associated with IR or insulin propeptides. Conclusions Beta cell signaling is impaired in non-diabetic and non-corticoids RA patients. Disclosure of Interest None Declared

Objectives To explore the influence of body composition and abdominal adiposity on endothelial dysfunction and radiological damage in Rheumatoid Arthritis (RA) patients. Methods A total of 100 patients, 54 RA patients and 46 controls, adjusted for sex, body mass index (BMI), age and comorbidity, were included. Total body composition was measured by dual energy X-ray absorptiometry; total and regional lean mass, fat mass, fat free mass index (fat free mass/m2) and fat mass index (fat mass/m2) were established. Quantification of visceral and parietal abdominal fat area was determined using magnetic resonance imaging, as well as visceral/parietal fat index. Endothelial dysfunction was assessed by brachial artery flow-mediated dilatation (FMD) as the dilator response to 5 minutes distal cuff occlusion and after sublingual nitroglycerine administration, and the Sharp Score defined radiological damage. The 28-joint DAS (DAS-28) and disability using HAQ (Health Assessment Questionnaire) scores, erythrocyte sedimentation rate (ESR) and C-reactive protein were collected. Multivariate analysis was performed to compare body composition between controls and patients and the relationship between those and endothelial dysfunction and radiological damage, everything adjusted for demographic and comorbidity variables. Results In the univariate analysis RA patients showed less flow-mediated dilatation (5.9 vs 9.8 mm, p=0.03). Similarly, fat free mass index was higher in patients than in controls (beta coefficient 0,94 (CI95% 0,08-1,80), P=0,03) after adjusting for BMI, sex and age; on the other hand fat mass index tended to be higher in patients than in controls (beta coefficient 2,51 (CI95% -0,82-5,84), p=0.13). Parietal and visceral abdominal fat values did not show differences between controls and patients. Fat mass, lean mass and abdominal fat did not correlate with DAS-28, HAQ, ESR or CRP. Parietal abdominal fat through resonance imaging in RA patients showed a negative relation with flow-mediated dilatation (beta coefficient -0.045 mm FMD/cm2 parietal area, p=0.02) after adjusting for BMI. This relation was not found with visceral abdominal fat. Radiological damage was negatively correlated with total bone mass (beta coefficient -1,13, CI95% -2,05-0,22, p=0.02) and showed a negative trend to do it also with total lean mass (beta coefficient -1.11 CI95% 2,42-0,21, p=0,09). Conclusions Changes in body composition take place in RA patients. These changes can be related to the endothelial dysfunction and radiological damage that occurs in this disease. Disclosure of Interest None Declared

Objectives The retinol binding protein 4 (RBP4) has recently been described as a protein highly related with insulin resistance (IR) states like obesity and diabetes. The chronic inflammatory states, like in rheumatoid arthritis (RA), are linked also to insulin resistance by mechanisms that are unknown. Therefore, RA is considered as an insulin resistance model related with a chronic inflammatory state, where diabetes and obesity, classical factors for IR, are absent. The objective of this study is to estimate RBP4 expression in RA patients. Methods 101 RA patients and 115 age and sex-matched controls were included. Pancreatic beta cell function was estimated by classical insulinresistance indexes like HOMA (homeostatic model assessment 2). RBP4, C peptide and insuline levels were measured in patients and controls by a specific enzyme-linked immunosorbent assay (ELISA). Multivariate analysis was performed to compare results between patients and controls and the data were adjusted for glucocorticoids intake and for IR classical risk factors. Results RBP4 levels did not show differences between controls and patients, after adjusting for sex, age, body mass index (BMI) and waist circumference (lnRBP4 2,83 mcg/dl in patients vs. 2,70 mcg/dl, p=0,33). On going steroids patients showed higher RBP4 levels (lnRBP4 3,03 vs 2,61 mcg/dl, p=0.00), after adjusting for age, sex, BMI and waist circumference. Similarly, in the univariate analysis, RBP4 levels tended to correlate with steroid average dose (r=0,20, p=0,14). In our serie, RBP4 levels were not related with the classical IR characteristics, like abdominal waist and BMI, both in controls and patients. Nor these levels showed relation with ESR, CRP, insuline levels or disease activity indexes. Conclusions The molecular mechanisms that lead to IR in RA patients appear to be different from those that ocurr in obesity and diabetes status. RBP4 does not seem to play a role in IR in patients with RA. Disclosure of Interest None Declared

No abstract