Optimisation of an existing heifer matrix and its association with reproductive performance in nulliparous...

The reproductive performance of dairy heifers can be predicted using a matrix including the percentage of mature body weight, body condition score (BCS) and presence of a corpus luteum. Given the subjectivity of condition scoring, backfat thickness determined by ultrasound could be used as an alternative measurement. The objectives of this historical retrospective observational study were to determine the relationship between BCS and backfat thickness ultrasonography (BFT), and to determine whether altering BCS thresholds or substituting BFT into this matrix was associated with subsequent reproductive outcomes. Nulliparous heifers (n = 718), from 8 seasonal, pasture-based dairy farms, were admitted for pre-breeding examination approximately one month (average 36.8 days) prior to the mating start date. Each heifer was weighed, body condition scored (on a five-point scale) and received a transrectal ultrasound examination to determine the presence of a corpus luteum by one operator. BFT was assessed by a different operator, using a separate ultrasound device placed midway between the tuber ischiae and tuber coxae on one side of the rump. Breeding was commenced on fixed calendar dates and was individually managed on each farm with subsequent pregnancy status being determined by transrectal ultrasonography. All analyses were performed in RStudio. The relationship between BCS and backfat thickness was assessed initially by Pearson's correlation (0.56). The difference between backfat thickness mean and BCS category was tested using a one-way ANOVA and was statistically significant (p ≤ 0.05). Tukey's test was used to show that the mean values of backfat thickness were significantly different (p ≤ 0.05) between each BCS category (except 3.75 and 4). The established matrix included three criteria: ≥55% mature bodyweight at mating start date, ≥3.5 BCS and presence of a corpus luteum at pre-breeding examination; with heifers achieving all three criterion being classified as READY and those achieving no criterion classified as PROBLEM. Three alternate matrices were also tested by either amending the BCS threshold (i.e., amending to ≥3.25) or substituting BCS with a backfat thickness measurement (i.e., thresholds of either ≥22.6 mm or ≥ 25 mm selected). All matrices were tested for association with reproductive outcomes (i.e., pregnant within three and six weeks post-mating start date) using logistic regression with farm treated as a random effect and heifer breed forced into each model. Goodness of fit for each model was examined by comparing the AIC and BIC of each model. Both the matrices including body condition score (≥3.25 and ≥ 3.5) and backfat thickness (≥22.6 mm) were significant associated with pregnancy by six weeks post breeding. However, the matrix using the threshold of BCS ≥3.25 was significant associated with pregnancy by three weeks post breeding and had the best statistical model fit overall. The results indicate that although BFT measurement may be used as an alternative to body condition scoring in this reproductive matrix, body condition scoring still remains preferrable.