Sessions

Serum progesterone (P4) measurement is a fundamental part of cycle monitoring in breeding management practice. It enables the prediction of the time of ovulation, thereby helping to optimise the timing of mating or artificial insemination. Nowadays, these measurements are generally performed in reference laboratories using chemiluminescent immunoassay (CLIA), therefore, several hours or even a day can pass between sample collection and the results. Point-of-care analysers have been developed to overcome this delay by enabling rapid in-clinic hormone measurements. However, in most cases, it is unclear to what extent the results obtained with these devices are interchangeable with hormone concentrations measured by radioimmunoassay or CLIA, which were used to establish reference values reported in the literature. The aim of the present study was to investigate the extent to which serum P4 levels measured with Virbac’s Speed Reader™ analyser, based on laser-induced fluorescence (LIF) technology, differ from the CLIA-based IMMULITE 2000 analyser, currently considered the gold standard. Serum P4 concentrations were evaluated during 22 cycle-monitoring examinations at the Reproduction Unit of the Small Animal Clinic (University of Veterinary Medicine Budapest), using two measurement methods on serial blood samples. The primary analysis included 47 paired measurements in which both results were within the respective analytical measurement ranges. As a sensitivity analysis, all 57 paired measurements were used. A strong correlation was demonstrated between the two methods (Pearson’s r = 0.955). However, Bland–Altman analysis showed that the LIF method yielded higher values on average. The mean difference between the two methods (CLIA–LIF) was −4.08 nmol/L, with classical 95% limits of agreement (LoA) ranging from −12.50 to 4.34 nmol/L. Passing–Bablok regression gave the equation CLIA = −1.08 + 0.84 × LIF, indicating both constant and proportional bias. The sensitivity analysis resulted in a similar slope (0.831). Application of the conversion equation improved agreement between the two methods at clinically relevant P4 concentrations. When crossing the 2 ng/mL and 5 ng/mL thresholds, the mean absolute difference between CLIA and LIF decreased from 0.89 to 0.42 days and from 0.54 to 0.26 days, respectively, with improved agreement observed in 8 of 10 and 9 of 10 cycles, respectively. Overall, the LIF method could be used for monitoring the ovarian cycle, provided that method-specific differences are taken into account during interpretation. The strong correlation between the two methods alone does not imply that their results are interchangeable. Although application of the Passing–Bablok conversion equation improved agreement in most of the cycles examined, independent validation is required before clinical application.