Abstract des Autors

Determining circulating equine insulin concentrations is becoming increasingly important in equine clinical practice and research. Most available assays are optimized for human medicine, but there is strong equine cross-reactivity because of the highly conserved nature of insulin. To identify an accurate and reliable assay for equine insulin, 6 commercial immunoassays were evaluated for precision, accuracy, and specificity. Only 1 assay initially reached the requisite standard: Mercodia Equine Insulin Enzyme-linked Immunosorbent assay (ELISA). Plasma matrix interferences were identified when the provided assay buffer was used with the Siemens Count-a-Coat Insulin radioimmunoassay (RIA) but not when charcoal-stripped equine plasma was used as the diluent. This modified RIA and the Mercodia Equine Insulin ELISA were evaluated further by directly examining accuracy by comparing their results for 18 equine plasma samples with values obtained using liquid chromatography and high-resolution/high-accuracy mass spectrometry (LC-MS). Compared with LC-MS measurements, the modified Siemens Insulin RIA rendered a moderate Lin's concordance coefficient (ρc) of 0.41, whereas the Mercodia Equine Insulin ELISA rendered a very poor ρc of 0.06. This suggests that the Siemens Insulin RIA is appropriate to use for routine evaluations when LC-MS is not available. Verf.-Referat