Impaired beta cell function is present in nondiabetic rheumatoid arthritis patients

Introduction: To investigate how markers of beta-cell secretion (proinsulin-processing metabolites) are expressed in rheumatoid arthritis (RA) patients and their potential relation with the insulin resistance (IR) observed in these patients. Methods: The 101 RA patients and 99 nondiabetic sex-and age-matched controls were included. IR by homeostatic model assessment (HOMA2), and beta-cell secretion, as measured by insulin, split and intact proinsulin, and C-peptide levels were determined for both groups. Multiple regression analysis was performed to compare IR between groups and to explore the interrelations between RA features, proinsulin metabolites, and IR. Data were adjusted for glucocorticoids intake and for IR classic risk factors. Results: Compared with controls, RA patients showed higher HOMA-IR (beta coef., 0.40 (95% CI, 0.20 to 0.59); P = 0.00). When data were adjusted for glucocorticoids intake, noncorticosteroid patients maintained a higher IR index (beta, 0.14 (0.05 to 0.24); P = 0.00). Impaired insulin processing in RA patients was detected by the onset of elevated split proinsulin levels (beta, 0.70 pmol/L (0.38 to 1.02); P = 0.00). These data remained significant also when adjusted for prednisone intake (beta, 0.19 (0.00 to 0.36) pmol/L; P = 0.04). Split proinsulin-to-C-peptide ratios were higher in RA patients undergoing corticosteroid therapy (beta, 0.25 (0.12 to 0.38); P = 0.03) and were nearly significant in comparison between noncorticosteroids patients and controls (beta, 0.16 (-0.02 to 0.34); P = 0.08). Interestingly, the impact of HOMA-IR on the ratio of intact proinsulin to C-peptide was higher in controls compared with patients (beta, 6.23 (1.41 to 11.06) versus 0.43 (-0.86 to 1.71); P = 0.03). Conclusions: beta-Cell function is impaired in nondiabetic and in RA patients not taking corticoids by a mechanism that seems to be, at least in part, independent of IR.