A randomized controlled trial to study the effects of an automated premilking stimulation system on milking performance, teat tissue condition, and udder health in Holstein dairy cows.

The objectives were to examine the effect of an automated premilking stimulation (APS) by means of a high pulsation frequency (300 cycles/min) without a reduction of the vacuum in the pulsation chamber or claw piece on (1) milking performance, (2) teat tissue condition, and (3) udder health in dairy cows. In a randomized controlled field study, Holstein cows (n = 427) from 1 commercial dairy farm with a milking schedule of 3 times per day were assigned to treatment and control groups over a 90-d period. Treatments consisted of a maximum of 80 s (APS80) or 99 s (APS99) of mechanical stimulation at a pulsation rate of 300 pulses per minute and a ratio of 25:75 (no reduction of the pulsation chamber or milking vacuum). Cows in the control group (CON) received traditional premilking stimulation by means of manual forestripping for 8 s. Milking characteristics were documented with on-farm milk meters. Short- and long-term changes in teat tissue condition induced by machine milking were assessed visually on a weekly basis. Composite milk samples were analyzed once per month to determine somatic cell count. Generalized linear mixed models were used to study the effect of the treatment on the outcome variables. We observed no meaningful differences in milk yield or milking unit-on time. Least squares means and their 95% confidence intervals (95% CI) for cows in the APS80, APS99, and CON groups were 13.5 (13.1-14.0), 13.2 (12.8-13.7), and 13.2 (12.8-13.7) kg for milk yield and 222 (213-231), 219 (210-228), and 223 (214-232) s for milking unit-on time, respectively. The effect of treatment on bimodality was modified by milk yield such that the odds of bimodality increased in the treatment groups with increasing milk yield. Compared with cows in the CON group, the odds ratios (95% CI) of bimodality were 1.08 (0.62-1.89) in the APS80 group and 0.89 (0.55-1.42) in the APS99 group at a milk yield of 11 kg and 2.0 (1.24-3.22) in the APS80 group and 2.08 (1.29-3.35) in the APS99 group at a milk yield of 16 kg. We observed differences in short- and long-term changes in teat tissue condition between the treatment and control groups. Compared with cows in the CON group, the odds (95% CI) of short-term changes were 1.87 (1.35-2.58) for the APS80 group and 1.49 (1.08-2.07) for the APS99 group, and the odds of long-term changes were 1.52 (1.24-1.85) for cows in the APS80 group and 1.59 (1.31-1.94) for cows in the APS99 group. The least squares means (95% CI) for somatic cell counts (log-transformed) were 4.74 (4.68-4.81) for the APS80 group, 4.77 (4.71-4.83) for the APS99 group, and 4.79 (4.73-4.86) for the CON group. We conclude that the APS system tested here had no negative effects on milk yield or milking unit-on time. However, differences in bimodality and teat tissue condition suggest that the APS system did not provide sufficient stimulation to facilitate a gentle milk harvest and adversely affects teat tissue condition.