Comparing the values of progesterone in the blood of bitches as measured with a chemiluminescence immunoassay and a radioimmunoassay

A ¹²⁵I radioimmunoassay (RIA) has long been used to determine the value of progesterone in serum or plasma of bitches but was discontinued in 2014. A chemiluminescence immunoassay (CLIA) gained prominence since 2003 to determine the value of progesterone in serum of bitches but the assay changed in 2012. This study assessed the agreement between progesterone values obtained with RIA in plasma (progRIA) and with the post‐2012 CLIA (progCLIA) in the serum of bitches. ProgCLIA was determined in 110 serum samples from 40 bitches in pro‐oestrus or early oestrus and compared to progRIA in plasma samples collected from the same bitches at the same time, where progRIA had a uniform distribution between 0.5 and 25 nmol/L. Two replicate analyses of each serum or plasma sample were simultaneously done in the same assay. For RIA and CLIA, the intra‐assay CVs were 5.85% and 6.70% and the interassay CVs 8.45% and 9.16%. For RIA and CLIA the progesterone values obtained with replicate analyses differed by as much as 11%–31% in 25% of samples. On average, the value of progCLIA was 85% of that of progRIA (95% CI 58%–112%, n = 110), with 88% of progCLIAs being lower than the progRIAs. This study shows that RIA and CLIA may yield replicate values that differ by as much as 11%–30% in about a quarter of samples analysed, necessitating replicate analyses if precise values are required. The study provides an equation by which to estimate progCLIA from progRIA.