Objective To investigate the feasibility of using human-specific insulin radioimmunoassay (Ins-RIA) kit to measure the concentrations of serum insulin in minipigs. Methods The Serum samples (219) of Bama minipigs were collected. The concentrations of serum insulin in Bama minipigs were measured by Ins-RIA kit and porcine-specific insulin enzyme linked immunosorbent assay (Ins-ELISA) kit, respectively. The linear regression analysis was undertaken in a randomly selected 175 samples (80% of total samples) to establish predictive equations between the concentrations of serum insulin measured by Ins-ELISA and the concentrations of serum insulin measured by Ins-RIA. The equations were then cross-validated in the remaining 44 samples (20% of total samples) that had not been included in the regression analysis. Results Measured insulin concentration was lower with the Ins-ELISA than that with the Ins-RIA 〔(15.32±15.50） μIU/mL vs. （32.31±21.74） μIU/mL, respectively, P<0.000 1〕; The final predictive equation for the Ins-ELISA (μIU/mL) was equal to -7.29+0.70 × Ins-RIA(μIU/mL) ( r2=0.94). The differences between the predicted values and the actual measured values were 17.18%. Conclusion Insulin values in Bama minipigs obtained from the Ins-ELISA and Ins-RIA are not equivalent and differ significantly. However, the insulin concentration by Ins-ELISA can be well estimated by Ins-RIA.
