PSX-7 Validation of a rapid hand-held device for the quantification of glucose in bovine follicular fluid.

Abstract In the last decade, the US cattle industry has shifted to producing more in vitro than in vivo embryos. This shift has been attributed to tangible improvements in in vitro embryo production (IVEP) efficiency, as well as an increase in the number of companies offering this service. Health, nutritional and endocrine status of the donor prior to OPU can be reflected by changes in follicular fluid (FF) molecular composition. Given the myriad of molecular profiles of peripheral blood and FF of donors, personalizing the in vitro culture environment to either enhance or amend the microenvironment conditions that the oocyte was exposed in vivo may be an opportunity for improving IVEP efficiency. Our objective was to validate a handheld Glucometer (CentriVet, ACON Laboratories) for real-time FF glucose quantification. We hypothesized that FF glucose concentrations measured on a rapid hand-held device would be correlated with glucose concentrations measured with an ELISA assay. To accomplish this, we pooled FF and steer serum (SS) for glucose quantification with the glucometer and a validated glucose assay (Cell BioLabs Inc, CA). Correlations analysis was performed with the PROC-CORR function of SAS (Version 9.4). The range of detected amounts of glucose were 10 and 150mg/dL for the glucometer and 0.028 to 1.8mg/dL for the assay. The intra and inter-assay CV of the CellBio kit were 2.71% and 3.92%, respectively. To validate the precision of the device, SS and FF with known glucose concentrations was utilized. Pooled FF and SS was diluted until concentrations were within the detectable range of the assay. Steer serum and FF at a 1:100 dilution was then spiked with 100mg/dL of glucose, assayed on the device, and again diluted for analysis with the ELISA. The device and kit glucose outcomes were highly correlated, r=0.978 (P< 0.0001). The CellBio assay determined approximately 44.5% of the glucose concentration determined by the device. After adjusting the values based on the relationship between the device and kit, we estimated the accuracy of the device against the kit using our previously determined adjustment of 44.5%. The estimated kit and the actual kit had a correlation of r=0.996 (P< 0.0001). Lastly, the first set of diluted SS and FF was spiked with approximately 100 mg/dL of glucose, assayed on the device, and further diluted (1:100) for kit analysis. The correlation of all SS and FF analyses was r=0.989 (P< 0.0001). Our results support the validity of the real-time glucometer to accurately determine glucose concentrations of bovine FF and SS. Although the device detected approximately 55.5% more glucose than the kit, the overestimation is consistent between samples. Future goals are to capture real-time in vivo profiles of depleted nutrients (e.g., glucose) and promptly adjust the culture environments to improve IVEP.