Serum resistin is associated with C-reactive protein and LDL- cholesterol in type 2 diabetes and coronary artery disease in a Saudi population

The type 2 diabetic patients for this study were randomly selected from the Diabetes Center and non-diabetic case control patients were obtained from out-patients of King Abdulaziz University Hospital at King Saud University in Riyadh, Saudi Arabia. Coronary heart disease (CHD) patients were selected from the Coronary Care units at King Khalid University Hospital and Prince Sultan Cardiac Units in Riyadh. From these, 114 Saudi Arabian patients were studied (Table 1). CHD patients (n = 22) were classified as having coronary heart disease (CHD) if they had suffered any of the following: a myocardial infarction, angina with a positive exercise tolerance test, a resting ECG consistent with CHD or angiographically proven coronary artery disease. T2DM patients (n = 35), diagnosed according to WHO criteria, but with no clinical evidence of CHD were screened with ECG, clinical symptoms and ± thallium cardiac scanning or cathertization. All patients with T2DM were treated by diet with or without oral hypoglycemic agents, mainly glibenclamide and metformin. Finally 50 Saudi individuals with no prior history of CHD or T2DM were recruited as healthy case controls from general practice. Ethical approval was obtained from the local institution's review committee.

All patients underwent a full physical examination and completed a general questionnaire. Body mass index (BMI), blood pressure, smoking habits and age were recorded. Blood samples were collected after a 12-hr fast for the determination of total cholesterol, HDL, triglycerides, insulin, apolipoprotein AI (Apo AI) and apolipoprotein AII (Apo AII), leptin and serum resistin concentrations. These studies were conducted with the approval of the King Abdulaziz Diabetic Center with approval obtained from the ethics committee and the individual consent of each patient.

Plasma samples were stored at -70°C prior to analysis. Measurements of serum cholesterol, HDL, and triglycerides, Apo AI and ApoAII were performed using routine laboratory methods. Insulin was analysed by a solid phase enzyme amplified sensitivity immunoassay (Medgenix INS-ELISA, Biosource, Belgium). Insulin resistance (HOMA IR) and β-cell function index were derived using the HOMA equation [27]. Leptin concentrations were measured by radio-immunoassay (Linco Research, St. Charles, MO). CRP was analysed by immunoturbidimetric method (Roche/ Hitachi analyser). Serum resistin was measured using the human resistin ELISA assay from Phoenix Pharmaceuticals (Belmont, California, USA; intra-assay variability <5%). The commercially available resistin ELISA had been validated as part of previous studies [15].

Data were stored and analysed using the SPSS version 10 package (SPSS, Evanston, IL, USA) for Windows. Biochemical parameters not normally distributed were analysed after being logarithmically transformed. A student's unpaired t-test or one-way ANOVA exposed the differences between the groups. Simple and partial correlation coefficients between the variables were determined and multiple regression analysis was performed to determine the relationships between the variables of interest. Data were expressed as mean and standard deviation (SD) or median and range; statistical significance was accepted at p < 0.05.

The full clinical and biochemical characteristics for the three groups are set out in Tables 1 and 2 respectively. The control subjects were noted to be younger than the T2DM patient counterparts (p < 0.01). However, there were no significant differences in age or blood pressure, Body mass index (BMI) or waist circumstance between case controls and CHD patients.

Serum resistin was significantly higher in T2DM 22.6(± 4.5)ng/mL and CHD patients 24.2(± 3.6) ng/mL compared with case controls 18.9(± 3.4) ng/mL (p-value <0.0001). CRP was significantly increased in T2DM 8.2 (0.7–27.1) μg/mL median (range) and CHD patients 7.3 (1.8–18.3) μg/mL(p = 0.007 and p = 0.002 respectively) compared with case controls 4.9(0.2–18.9) μg/mL (Table 2). Regression analysis determined that serum resistin was associated with serum CRP levels (Figure 1).

Fasting plasma glucose 15.4(5.4) serum cholesterol 15.4(5.4) mmol/L, LDL6.1(2.9) mmol/L and Apo-AI 70.8 (23.4)mg/dL were higher in T2DM patients (p < 0.0001, p < 0.05, p < 0.017, p < 0.003, respectively) compared with case control subjects, while the Apo-AII 45.9 (13.6)mg/dL was lower in the T2DM than in the controls, where p < 0.002 (Table 2). In CHD, cholesterol 7.7 (2.7), and LDL 4.9(1.9) mmol/L were significantly higher than in the case controls (p = 0.003, p < 0.01 respectively; while the Apo-AII 46.3(18.9) mg/dL in CHD patients as lower than in the case controls, where p = 0.003 (Table 2).

Univariate analysis for all groups showed strong positive associations between resistin and fasting glucose (p = 0.01, Figure 2), and negative association with hip circumference (R² 0.6, p = 0.004, p = 0.03 respectively) but in CHD patients, resistin had a strong positive association with systolic and diastolic blood pressure and Apo-AII (R² 0.82, p < 0.0017, R² = 0.45 p < 0.002 respectively). There was also a strong correlation with Apo-AII (p < 0.003), whereas, in T2DM patients, resistin was correlated with LDL (R² = 0.79, p = 0.009).