Abstract #T217
Section: Graduate Student Competition
Session: ADSA Production Division Graduate Student Poster Competition, PhD
Format: Poster
Day/Time: Tuesday 7:30 AM–9:30 AM
Location: Gatlin Ballroom
Session: ADSA Production Division Graduate Student Poster Competition, PhD
Format: Poster
Day/Time: Tuesday 7:30 AM–9:30 AM
Location: Gatlin Ballroom
# T217
Validation of a hand-held meter for measuring β-hydroxybutyrate in plasma and serum of dairy cows.
Arnulfo Pineda*1, Felipe C. Cardoso1, 1University of Illinois, Urbana, IL.
Key Words: BHBA, dairy cow, Precision Xtra
Validation of a hand-held meter for measuring β-hydroxybutyrate in plasma and serum of dairy cows.
Arnulfo Pineda*1, Felipe C. Cardoso1, 1University of Illinois, Urbana, IL.
The aim of this study was to compare serum (sBHBA) and plasma (pBHBA) BHBA concentrations analyzed using either a laboratory method (Randox Laboratories Ltd., UK, Cat # RB1007 as “gold standard”; LM) or a hand-held meter (PX; Precision Xtra). Results from 187 samples taken from Holstein cows from 11 d before to 5 d after parturition were used for the analysis. Statistical analysis was performed using the MIXED, REG, and LOGISTIC procedures of SAS (v9.4). A linear mixed model with repeated measures was performed for LM and PX. A regression analysis was completed to estimate the relationship between the 2 methods. Cross-validation by randomly splitting the data in model building and validation sets was performed to estimate and validate the equation that predicted the LM results using PX. Receiver operation characteristic (ROC) curves were made to estimate the sensitivity and specificity of PX at different threshold levels. The PX yielded higher (P < 0.01) pBHBA and sBHBA than LM, 1.45 vs. 0.95 and 1.63 vs. 1.00 mmol/dL, respectively. Adjusted R2 between both methods for pBHBA was 97.1% and 97.8% for sBHBA. The equation from the model building data set that predicted pBHBA results using PX was y = −0.05353 + 0.67842x, while the equation that predicted sBHBA results was y = 0.05542 + 0.56862x, where y = predicted LM BHBA and x = PX BHBA. Mean square error (MSE) yielded by the model building data set for pBHBA and sBHBA were 0.006 and 0.005, respectively. Mean squared predictor error (MSPR) from the validation data set was 0.007 for pBHBA and 0.006 for sBHBA. Highest sensitivity and specificity for PX was achieved when threshold was set to 1.8 for pBHBA and 2.1 mmol/dL for sBHBA. The area under ROC curve (AUC) was 0.97 for pBHBA and 0.96 for sBHBA. In conclusion, the small difference between MSE and MSPR suggest that PX can be used to predict pBHBA and sBHBA by using the aforementioned equations. However, the threshold for ketosis requests to be higher than the commonly used in the field (1.2 mmol/dL) when using PX. The AUC close to 1 suggests greater ability of PX in predicting ketosis under the indicated thresholds.
Key Words: BHBA, dairy cow, Precision Xtra