Abstract #163

# 163
β-Hydroxybutyrate measurements in bovine milk compared between known standards, a clinical pathology analyzer, and Fossomatic high-throughput testing.
D. Wilson*1, G. Goodell2, 1Utah State University, Logan, UT, 2The Dairy Authority, Greeley, CO.

β-Hydroxybutyrate (BHB) is measured in dairy cow milk to detect ketosis, usually associated with negative energy balance. The objective was blinded comparison of spiked milk samples resulting in known concentrations of BHB to BHB measurements using a clinical pathology analyzer and Fossomatic Fourier transform infrared (FTIR) high throughput testing. Milk meter collected samples from Holstein cows with low (<0.02 mM) BHB were spiked with DL-BHB reagent to concentrations of 100 µM (0.1 mmol)/L, 200, 300, 400, 500, 600, 800, 1000, 1200, 1500, and 2000 µM/L. Two aliquots of 17 milks at each concentration were tested, blinded to operator and randomized in concentration order. Calibration was per manufacturer’s standards, as was that milk was centrifuged to milk sera for testing with the ChemWell-T 460 analyzer, and whole milk was tested using the Fossomatic. Thresholds of 200 and 300 µM/L for classifying cows as ketotic (K) vs. non-ketotic (N) were tested for test agreement and accuracy. Results were 374 runs on ChemWell and 352 runs on Fossomatic (one outlier run excluded). ChemWell results: mean correlation with standards 0.991 (range 0.970 to 0.999). 200 µM/L threshold: 97.6% agreement (K vs. N, Kappa 0.848 very good) with standards, if spiked milks considered “gold standard,” sensitivity (sens) 99.1%, specificity (spec) 82.4%. 300 µM/L threshold: 98.7% agreement (K vs. N, Kappa 0.956, very good), sens 98.4%, spec 100%. Fossomatic results: mean correlation with standards 0.920 (range 0.784 to 0.984). 200 µM/L threshold: 85.8% agreement (K vs. N, Kappa 0.355 fair) with standards, if spiked milks considered “gold standard,” sens 94.6%, spec 35.8%. 300 µM/L threshold: 91.5% agreement (K vs. N, Kappa 0.656, good), sens 96.0%, spec 66.7%. The FTIR Fossomatic BHB testing requires recalibration multiple times per day and results are not as accurate as those with the clinical pathology analyzer, but are more accurate than many other milk high throughput diagnostic tests currently used in the dairy industry. Continued improvement in BHB milk calibration standards for the Fossomatic is needed.

Key Words: β-hydroxybutyrate (BHB), milk, laboratory