BackgroundThe precise mechanism linking systemic inflammation with insulin resistance (IR) in rheumatoid arthritis (RA) remains elusive. In the present study, we determined whether the incretin-insulin axis and incretin effect are disrupted in patients with RA and if they are related to the IR found in these patients.MethodsWe conducted a cross-sectional study that encompassed 361 subjects without diabetes, 151 patients with RA, and 210 sex-matched control subjects. Insulin, C-peptide, glucagon-like peptide-1 (GLP-1), gastric inhibitory polypeptide (GIP), dipeptidyl peptidase 4 (DPP-4) soluble form, and IR indexes by homeostatic model assessment (HOMA2) were assessed. A multivariable analysis adjusted for IR-related factors was performed. Additionally, ten patients and ten control subjects underwent a 566-kcal meal test so that we could further study the postprandial differences of these molecules between patients and control subjects.ResultsInsulin, C-peptide, and HOMA2-IR indexes were higher in patients than in control subjects. This was also the case for GLP-1 (0.49 ± 1.28 vs. 0.71 ± 0.22 ng/ml, p = 0.000) and GIP (0.37 ± 0.40 vs. 1.78 ± 0.51 ng/ml, p = 0.000). These differences remained significant after multivariable adjustment including glucocorticoid intake. Disease Activity Score in 28 joints with erythrocyte sedimentation rate (β coefficient 46, 95% CI 6–87, p = 0.026) and Clinical Disease Activity Index (β coefficient 7.74, 95% CI 1.29–14.20, p = 0.019) were associated with DPP-4 serum levels. GLP-1 positively correlated with β-cell function (HOMA2 of β-cell production calculated with C-peptide) in patients but not in control subjects (interaction p = 0.003). The meal test in patients with RA revealed a higher total and late response AUC for glucose response, a later maximal response of C-peptide, and a flatter curve in GIP response.ConclusionsThe incretin-insulin axis, both during fasting and postprandial, is impaired in patients with RA.Electronic supplementary materialThe online version of this article (doi:10.1186/s13075-017-1431-9) contains supplementary material, which is available to authorized users.