Serum complement C3 correlates with insulin resistance in never treated psoriatic arthritis patients

The objective of this cross-sectional study is to investigate the relationship between inflammatory markers (C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), white blood cell (WBC) count, and complement C3) and insulin resistance (estimated with the surrogate measure homeostasis model assessment of insulin resistance (HOMA-IR)) in Psoriatic arthritis (PsA) patients. Sixty-seven patients with PsA and 138 age- and sex-matched controls were recruited from the Rheumatology Outpatient Clinic of the University of Catanzaro, Catanzaro (Italy). Plasma glucose, insulin, triglycerides, total cholesterol, LDL cholesterol, and high-density lipoprotein (HDL) cholesterol were measured after at least 12-h fasting. CRP was measured by a high-sensitivity turbidimetric immunoassay. ESR was measured by capillary photometry. Serum C3 and C4 were measured by nephelometry. Insulin resistance was estimated using the original HOMA as reported by Matthews et al. (Diabetologia 28(7):412-419, 1985) from overnight fasting insulin and glucose values, with the following formula: HOMA-IR = glucose (in milligrams per deciliter) × insulin (in micro units per milliliter)/405. Patients and controls had similar body mass index (BMI), blood pressure, and fasting glucose, but fasting serum insulin was higher in PsA patients (median (interquartile range (IQR)) 12.96 (6.35-24.65) for PsA and 10.37 (6.97-15.11) for controls; p = 0.005). Consequently, also the surrogate measure of insulin resistance HOMA-IR was significantly higher in patients (median (IQR) 3.77 (1.58-5.18) for PsA and 2.33 (1.67-3.62) for controls; p = 0.008). In univariate analysis, ln(HOMA-IR) correlated positively with BMI, systolic blood pressure (sBP), dBP, ln(TG), WBC, ln(ESR), ln(CRP) and C3, while correlated negatively with total and HDL cholesterol. In multiple linear regression analysis including age, sex and BMI, serum C3 reached the higher standardized β coefficient, while among other variables only ln(ESR) maintained a weak correlation with ln(HOMA-IR) when sBP was added to the model. When all variables were inserted in a single model, only C3 was independently correlated with ln(HOMA-IR). In a stepwise multiple regression model, serum C3 was selected as the strongest single contributing variable. The model explained 61 % of the variance in ln(HOMA-IR) (R (2) = 0.61). Insulin resistance, estimated with the surrogate measure HOMA-IR, is higher in PsA patients compared to age- and sex-matched controls. Serum C3 could represent a simple marker of insulin resistance in psoriatic arthritis patients.