Clinical evaluation of the ADVIA Centaur XPT chemiluminescent immunoassay for equine insulin measurement.
Authors: Rey-Conejo R, Toribio R E, Möller S, Müller E, Fores-Jackson P
Journal: Journal of equine veterinary science
Summary
# Editorial Summary: Equine Insulin Measurement on the ADVIA Centaur XPT Insulin dysregulation underpins equine metabolic syndrome, making reliable basal insulin measurement essential for diagnosis and monitoring, yet different laboratory analysers have historically produced inconsistent results that hamper clinical comparability. Rey-Conejo and colleagues validated the ADVIA Centaur XPT chemiluminescent immunoassay against the established Immulite 2000 XPi analyser, assessing analytical precision, linearity, and dilution recovery, then comparing results from 89 equine serum samples using Deming regression and Bland-Altman analysis. The ADVIA Centaur XPT showed good intra-assay precision (1.92–3.56 % coefficient of variation) and acceptable inter-assay variation (5.19–5.78 %), though it consistently measured lower insulin concentrations than the Immulite 2000 XPi, indicating proportional bias across the measurement range. Strong rank-order correlation (Spearman r = 0.92) between methods persisted, and critically, when adapted clinical decision thresholds were applied—essentially recalibrated cut-off values specific to the ADVIA platform—the two analysers achieved strong agreement in classifying samples (weighted kappa = 0.82). For practitioners and laboratories, this validation permits adoption of the ADVIA Centaur XPT for equine insulin testing provided that adapted reference ranges are implemented; however, samples should not be compared directly against historical Immulite results without accounting for the systematic difference in absolute values.
Read the full abstract on PubMed
Practical Takeaways
- •If your clinic switches to ADVIA Centaur XPT for insulin testing, you must use new reference ranges adapted from this study—results will be systematically lower than previous Immulite 2000 XPi measurements, but interpretation remains clinically comparable
- •The strong rank-order correlation (0.92) means the new analyzer reliably identifies which horses have dysregulated insulin relative to others, which is the critical clinical decision
- •The 0.82 kappa suggests good agreement for classifying horses into normal vs. abnormal insulin status, supporting use of this newer analyzer for metabolic syndrome screening
Key Findings
- •ADVIA Centaur XPT demonstrated acceptable precision with intra-assay coefficients of variation ranging from 1.92% to 3.56% across insulin concentration ranges
- •Proportional bias was identified with ADVIA Centaur XPT consistently measuring lower insulin concentrations than Immulite 2000 XPi method
- •Strong correlation (Spearman r=0.92, 95% CI: 0.87-0.94) was observed between the two analyzers
- •Classification agreement using adapted decision thresholds yielded weighted kappa of 0.82, indicating strong agreement for clinical interpretation