Determining the clinical utility of a single pre-breeding examination for predicting subsequent reproductive performance in seasonal pasture-based dairy heifers.

Heifers that conceive early in the breeding season are key to the maintenance of a compact calving pattern in seasonal dairy herds. The purpose of this prospective observational study was to determine whether a single pre-breeding examination is useful in predicting future reproductive and calving performance in dairy heifers. This evaluation was based on a matrix of easily measurable variables which could be applied to seasonal, pasture-based dairy heifers. The matrix variables were; body weight (≥55% mature body weight at mating start date), body condition score (≥3.5), and the presence of a corpus luteum. Seasonal, pasture-based dairy heifers (n = 572) on six commercial dairy farms were examined at an average of 28 days prior to the breeding season. At this examination, they were classified as READY (meeting all three of the above criteria), INTERMEDIATE (meeting one or two of the above criteria) or PROBLEM (meeting none of the above criteria) based on the matrix and then exposed to a routine artificial insemination (AI) breeding period followed by bull breeding depending on individual farm management. Subsequent pregnancy status was determined by transrectal ultrasonography initially at 30-60 days in calf (DIC), and confirmed at over 60 DIC. After the breeding season, 519 (96%) of heifers were determined to be pregnant, of which 345 (64%) became pregnant within the first three weeks of the breeding season. Farmers collected calving data; including calving difficulty score, from the heifers the following spring. A logistic regression model was used to analyse the association between the matrix categories and the following outcomes: reproductive (21 day submission rate, 1st AI conception rate, 3 wk and 6 wk in calf rate) and calving (assistance and dystocia). Heifers classified as PROBLEM had significantly lower odds of conceiving to the first AI serve and of becoming pregnant within the first three and six weeks of the breeding season (P ≤ 0.05). In conclusion, the proposed matrix could be practically useful for classifying heifers prior to the breeding season.