Evaluation of a Chemiluminescent Immunoassay for Measurement of Equine Insulin

Abstract

Background: Many diagnostic tests for insulin dysregulation use reference intervals established with an insulin radioimmunoassay (RIA) that is no longer available. A chemiluminescent immunoassay (CLIA) is commonly used for the measurement of serum insulin concentration in clinical practice but requires further validation, especially at clinically relevant reference intervals. Objectives: To evaluate the CLIA for measurement of equine insulin and compare it to the previously validated, but now unavailable RIA. Samples: Equine serum samples (n = 78) from clinical and experimental studies. Methods: In this experimental study, performance of the CLIA was evaluated using standard variables, including comparison with the RIA. Continuous and binary outcomes were analyzed. Results: The CLIA showed good intra-assay (coefficient of variation [CV], 1.8–2.4%) and interassay (CV, 3–7.1%) precision. Acceptable recovery on dilution (100 ± 10%) was achieved only at dilutions <1:1. Recovery on addition was acceptable. Comparison of the CLIA and RIA showed strong positive correlation (r = 0.91–0.98), with fixed and proportional bias. At 3 diagnostic cutoffs, sensitivity of CLIA compared with RIA ranged from 67 to 100% and specificity from 96 to 100%. Conclusions and Clinical Importance: The CLIA is a highly repeatable assay which is suitable for within- and between-horse comparisons. Dilution of high concentration samples should be performed with charcoal-stripped serum (CSS) and at the lowest dilution factor possible. At concentrations commonly used for diagnosis of insulin dysregulation (≤100 μIU/mL), results from the CLIA tend to be lower than from the RIA and should be interpreted accordingly. Further standardization of equine insulin assays is required.