A randomized controlled trial on the effect of incomplete milking during the first 5 days in milk on reproductive performance of dairy cows

The main objective of the current study was to measure the effect of incomplete milking on luteal activity and on pregnancy hazard. We also aimed to study the effect of early-lactation hyperketonemia (i.e., β-hydroxybutyrate blood concentration ≥1.4 mmol/L during the first 3 wk in milk) on those reproductive outcomes. Multiparous Holstein cows (n = 853) from 13 commercial herds were enrolled in a randomized controlled trial. Cows were assigned to a control or a treatment group, incompletely milked (10–14 L of milk collected/d) from 1 to 5 DIM. Blood samples were collected once a week during weeks in milk 1 to 3 for β-hydroxybutyrate blood concentration, and a threshold of 1.4 mmol/L was used to define hyperketonemia. During weeks in milk 5 and 7, cows were sampled for progesterone blood concentration, and a threshold of 1 ng/mL was used to define luteal activity. Reproductive information and culling dates were obtained through herd records. Logistic regression models and survival analyses were used to assess the effect of treatment on luteal activity and on pregnancy hazard, respectively. Analogous models were used to investigate the effect of early-lactation hyperketonemia on reproductive outcomes. The odds of luteal activity for incompletely milked cows were 1.1 (95% confidence interval: 0.72–1.7) times those of conventionally milked cows. The effect of treatment on pregnancy hazard varied as a function of time, parity, and start of the breeding period. In second-parity cows that started the breeding period <55 d in milk, the pregnancy hazard (95% confidence interval) in incompletely milked cows was 576.3 (240.0–1,383.7), 36.9 (18.9–72.1), 6.8 (3.3–13.8), 2.5 (1.0–5.9), and 0.13 (0.07–0.26) times that of conventionally milked cows at 1 to 21, 22 to 43, 44 to 65, 66 to 87, and >87 d after the voluntary waiting period, respectively. The treatment did not have an effect on pregnancy hazard in cows in third parity or greater or in those starting the breeding period ≥55 d in milk. Early-lactation hyperketonemia was not associated with any of the reproductive outcomes. In conclusion, the incomplete milking protocol had no effect on luteal activity and had a positive effect on pregnancy hazard in second-parity cows in herds with a short voluntary waiting period (<55 d). We did not observe an effect of early-lactation hyperketonemia on luteal activity or on pregnancy hazard.