An investigation of Dairy Cow Response to Stimulation by the cattle immobilizer
Objective
To examine plasma cortisol levels in response to stimulation of dairy cows with an
cattle immobilizer
Method
Twelve lactating dairy cows of predominantly Holstein-Friesian breed were selected. Following
morning milking the cows were grazed in a pasture for four hours and were then loaded side by
side in a herringbone race for the duration of experimental period (approximately 2 hours).
Measurements were made on two treatment days with an interval of three days between
treatment days.
Cows were assigned randomly to the treatment groups in a crossover design such that they
received two of the three possible stimulation treatments using a cattle immobilizer during the
course of the project (n=8 cows per treatment over the two treatment days). Stimulation was
applied for a sustained period of two minutes. This duration was selected as being the period
required for most examinations and minor manipulations to be carried out. The levels of
stimulation chosen were at Setting 2 and 4 on the device dial. Thus the three treatments consisted
of:

Setting 2 – probe inserted into the rectum and stimulation applied at level 2 for two
minutes (mils stimulation).

Setting 4 – probe inserted into the rectum and stimulation applied at level 4 for two
minutes (moderate stimulation).

Control treatment – probe which was unconnected to any electrical box inserted into the
rectum for two minutes.
Coccygeal blood samples were taken before the device was applied and again at 2, 15, 30 and 60
minutes after the end of the stimulation period. Blood samples were collected into heparinised
vacutainers and immediately placed into ice water. They were centrifuged at 1800g and 4°C for 12
minutes within 90 minutes of collection. Plasma was then aspirated and stored at -20°C until
analysed. Plasma cortisol concentrations were measured by radio-immune assay in a single assay
run. This assay has previously been described by Fisher et al (2001) and has a minimum detectable
level of 1,1ng/ml. Samples were analysed in duplicate. Within assay coefficients of variation were
11.7%, 9.6% and 8.2% for reference samples or 5.6, 25.7 and 52.3ng/ml, respectively.
Data were analysed using a REML variance components analysis (GenStat9.1) with raw and log
transformed data. Pre-treatment plasma cortisol concentration was initially analysed as a
covariate but found to be significant only at the 2-minute post-treatment sample.
Pre-treatment cortisol levels as a covariate were therefore omitted from further analysis. Day of
treatment was a non-significant factor in all analysis and so was treated as a fixed effect in the model.
Results
Data for plasma cortisol concentrations at each sample are given in Table 1 and are presented
graphically using values calculated as fitted splines of raw data in Figure 2. Results are reported
mean ± standard errors of differences (sed).
Setting 2 Setting 4 Control SED P Value
-2 minutes 9.6 15.3 12.3
2 minutes 13.1 18.1 15.6 2.4
15 minutes 13.8 22.8** 14.6 2.7 ** Treatment P