Background Rheumatoid arthritis (RA) is characterized by an excess of cardiovascular diseases (CVD) risk, similar to type 2 diabetes (T2DM). A recent meta-analysis confirmed an elevated prevalence and incidence of T2DM in RA . In this context, the necessity to recognize early T2DM in RA is driven by the evidence that CVD risk factors tends to act synergistically, thus increasing the risk conferred by each single factor . However, it's still difficult to select patients that need to underwent extensive diabetes investigation in the rheumatology setting.
Objectives Aim of our work was to identify useful predictors of impaired glucose tolerance (IGT) and T2DM in RA patients with normal fasting glucose values in the rheumatology outpatient setting.
Methods 100 RA patients satisfying ACR/EULAR 2010 criteria  were screened with FPG. Of these, 78 had normal (<100 mg/dL) FPG values and therefore were recruited. All patients underwent a 75g oral glucose tolerance test (OGTT) following World Health Organization (WHO) recommendations. According to 2-hours postload glucose values (2h-pG), patients were classified as normal (NGT) if 2h-pG <140 mg/dL, impaired glucose tolerance (IGT) if 2h-pG 140–200 mg/dL or T2DM if 2h-pG ≥200 mg/dL. Age, sex, body mass index (BMI), waist circumference (WC) and blood pressure were recorded. After overnight fasting, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), total and HDL cholesterol, triglycerides (TG) and uric acid were measured. The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated with the formula: glucose (mg/dL) x insulin (mU/L)/405. The triglycerides and glucose index (TyG index) was calculated with the formula: ln (TG [mg/dL] x glucose [mg/dL]/2). The lipid accumulation product (LAP) was calculated with the formula [WC (cm) – 58 for women or 65 for men] x TG (mmol/L). The visceral adiposity index (VAI) was calculated with the formula proposed by Amato et al .
Results Of the 78 RA patients recruited 57 were NGT, 18 were IGT and 3 were T2DM. After correction for age, sex, BMI, and systolic blood pressure, 2h-pG correlated with diseases duration (R=0.25, p=0.04), ESR (R=0.27, p=0.02), lnTG/HDLc (R=0.24, p=0.04), HOMA-IR (R=0.29, p=0.01), LAP (R=0.28, p=0.02) but not with lnCRP, lnTG, fasting glucose, uric acid, TyG index or VAI. In addition 2h-pG category correlated only with LAP (R=0.33, p=0.005). When A1c% was set as dependent variable, it correlated with lnTG (R=0.26, p=0.03), fasting glucose (R=0.29, p=0.0), uric acid (R=0.26, p=0.03), TyG index (R=0.25, p=0.04), LAP (R=0.30, p=0.01).
Conclusions In RA patients with normal fasting glucose, LAP correlates significantly with 2hPG and predicts abnormal glucose tolerance.
Jiang, P., H. Li, and X. Li. Clin Exp Rheumatol, 2015. 33(1): p. 115–21.
Baghdadi, L.R., et al. PLoS One, 2015. 10(2): p. e0117952.
Aletaha, D., et al. Ann Rheum Dis, 2010. 69(9): p. 1580–8.
Wallace, T.M., J.C. Levy, and D.R. Matthews. Diabetes Care, 2004. 27(6): p. 1487–95.
Unger, G., et al. Endocrinol Nutr, 2014. 61(10): p. 533–40.
Kahn, H.S. BMC Cardiovasc Disord, 2005. 5: p. 26.
Amato, M.C., et al. Diabetes Care, 2010. 33(4): p. 920–2.
Disclosure of Interest None declared