Development of a Web App to Convert Blood Insulin Concentrations among Various Immunoassays Used in Horses.
Authors: Delarocque Julien, Feige Karsten, Carslake Harry B, Durham Andy E, Fey Kerstin, Warnken Tobias
Journal: Animals : an open access journal from MDPI
Summary
# Editorial Summary: Standardising Insulin Measurements Across Laboratory Assays Insulin dysregulation diagnosis in horses relies on blood insulin concentration and comparison against diagnostic cut-offs, yet the seven commonly used assays produce substantially different results, creating confusion when comparing values across laboratories or referencing literature standards. The authors compiled original and published insulin data from CLIA, RIA and ELISA platforms into a web-based conversion tool, fitting linear regression models with non-decreasing splines to 13 assay pairs to mathematically translate readings between methods. Models demonstrated strong overall performance (median r² = 0.94), though notably, multi-laboratory datasets showed reduced accuracy, revealing previously undocumented technical variation between facilities that affects the reliability of converted values. Whilst this variation warrants caution when diagnostic cut-offs are provided by different laboratories, the tool's high performance supports its practical utility for interpreting insulin results when your preferred reference assay is unavailable and for contextualising values obtained from different sources. For practitioners working with endocrine cases, this resource addresses a genuine clinical problem: the ability to meaningfully compare insulin measurements now exists, though awareness of inter-laboratory technical differences remains essential when making diagnostic decisions.
Read the full abstract on PubMed
Practical Takeaways
- •Insulin results from different laboratories and assay types can now be converted and compared using this web app, improving diagnostic interpretation of insulin dysregulation
- •Be aware that the same assay may produce slightly different values between laboratories due to technical variation, even when converting between standardized assays
- •This tool helps standardize insulin dysregulation diagnosis across different testing facilities, reducing confusion when reference labs differ from your preferred diagnostic centre
Key Findings
- •Seven different insulin immunoassays were compiled with conversion models for 13 assay pairs integrated into a web app tool
- •Models demonstrated high overall performance with median r² = 0.94 (range 0.57–1.00) for single-study assay comparisons
- •Technical variation between laboratories was identified when assay comparisons included data from multiple studies, indicating inter-laboratory differences not previously described
- •The conversion tool enables interpretation of diagnostic insulin measurements and comparison of values across different assays when reference assays are unavailable