Kovács L. et al./Scientific Papers: Animal Science and Biotechnologies, 2013, 46 (1)
Measuring Stress Level of Dairy Cows during Milking Using
by Geometric Indices of Heart Rate Variability
Levente Kovács*, Luca Kézér, János Tőzsér
Institute of Animal Husbandry, Faculty of Agricultural and Environmental Science, Szent István University
Gödöllő, Páter Károly str. 1, H-2103, Hungary
Abstract
Heart rate (HR) and heart rate variability (HRV) were investigated in cows (n=32, age: 3.86 years, milk production:
35±2.5 kg, DIM: 150±15) milked in a parallel milking parlour. Geometric parameters of HRV (SD1 and SD2) were
calculated using Poincare graphs. HRV indices of resting 1 h after midday milking (reference period) were compared
to those measured during the different phases of the evening milking (driving; in the holding pen; udder preparation;
milking; after milking in the milking stall). There was no difference between the reference period and the different
phases of milking in animal welfare terms. During the reference period SD2 (198.5 ms) was significantly higher
(p<0.05) than every other measured period suggesting an increasing parasympathetic tone after milking. This parasympathetic predominance decreased with time of the day (1.5 h after milking). SD2 was significantly affected by
parity, by the breeding bull (p<0.01) and by milk production (p<0.05). SD2 was notably higher (102.8 ms) in multiparous cows than in primiparous cows (p<0.017; α=0.005) during resting and milking. Results suggested that a conventional milking process is not really stressful for cows. Primiparous cows were more susceptible of milking process than multiparous ones. SD2 is a good marker of vagus activity and affected by several independent factors.
Keywords: dairy cow, geometric measures, heart rate variability, milking, stress
dairy cattle [6]. This measure is the variability in
the successive cardiac cycles (interbeat intervals,
IBI) referred to as heart rate variability (HRV)
investigating which the activity of the sympathetic
nervous system (SNS) and the parasympathetic
nervous system (PNS) can be monitored separately at the same time [7] allowing detailed interpretation of cardiac activity in terms of autonomic
nervous system (ANS) activity.
Acute stress during milking can compromise
animal welfare due to the increase in the amount
of residual milk [8]. HRV have previously been
used in welfare research to gain information about
the stress level of dairy cows during milking [6].
Some studies found higher levels of stress in
automatic milking system than in a milking
parlour [9,10], while others found no such
differences [4], or even noted lower stress level
than in a herringbone parlour [11]. In most
experiments only alterations in HR were
monitored [5,8], however, it provides little
1. Introduction

A fundamental task of applied animal science is to
evaluate and improve husbandry and management
systems in the context of animal welfare. Important aspects of this are the recognition and
evaluation of stress in farm animals. Nowadays, it
is of particular interest in intensive cattle farming
to identify those physiological characteristics
which describe animal’s responsibility in a potentially stressful environment.
Besides sympatho-adrenal [1] and non-invasive
measures of cortisol and its metabolites [2,3] as
well as simple heart rate (HR) [4,5], recently, an
additional parameter of cardiac activity have been
investigated for assessing stress and welfare in

* Corresponding author: Levente Kovács, Phone: +36
28 410 200/1633, Fax: +36 28 410 804, Mobile: +36 30
944 3990, Email: [email protected]
213
Kovács L. et. al./Scientific Papers: Animal Science and Biotechnologies, 2013, 46 (1)
For calculation of SD1 and SD2 Poincare graph
were used, which is a map of dots in an XYdiagram. Each dot represents the duration of an
IBI plotted against the duration of its preceding
IBI (Figure 1).
information on the underlying physiological
mechanisms that govern its modification in many
behavioural situations [12].
Several methods have been proposed for the
assessment of HRV reviewed in dairy cattle [6,7].
Next to the orthodox methods of HRV analysis
quantitative Poincaré measures have been also
found to provide useful information on the vagal
regulation [13].
We investigated geometric indices of HRV as
physiological measures of stress in high producing
dairy cows milked in a parallel milking parlour in
a Hungarian working farm. The aim of our study
was to find out whether a conventional milking
means stress for cows. The second aim was to
differentiate cows’ stress level in the separated
phases of the milking process. Our third interest
was with regard to compare the stress levels of
primiparous and multiparous cows.
Table 1. HRV parameters calculated in study
Parameter
Definition
HR (beats/min)
Mean heart rate
The SD of instantaneous IBI
variability measured from axis 1
SD1 (ms)
in the Poincaré plot. Index of
PNS activity.
The SD of long-term continuous
IBI variability measured from
SD2 (ms)
axis 2 in the Poincaré plot. Index of SNS activity.
ms: millisecond
2. Materials and methods
Measurements were carried out between 6:30 and
21:30 during a 2-week period in May 2012 in a
Hungarian working farm. 32 focal cows were
selected based on age (between 2 and 5 years),
milk production (35±2.5 kg) and DIM (150±15)
milked in a parallel milking parlour.
Data were collected 3-day continuously from each
animal and obtained using video observations of
the milkings and continuous HR recordings. Individuals’ HRV values were evaluated for lying
used as a baseline in the comparison with values
during milking.
Cows’ stress level was differentiated in the separated phases of the milking process. HR was recorded with a monitoring system with two adhesive
electrodes that stored RR-intervals for about 15 h
continuously (Polar Equine RS800 CX, Polar
Electro Oy, Helsinki, Finland). Prior to the study
cows were accustomed to wearing this equipment.
The RR data were downloaded from the HR receiver onto a computer and inspected for measurement quality and artefacts using the Kubios
HRV analysis software [14]. Because the continuous bouts from which we calculated parameters of
HRV were of varying length, we calculated all
parameters in 5-min time windows following recommendations by earlier studies [11,15]. For
HRV analysis geometric parameters (SD1 and
SD2) were calculated (Table 1).
Figure 1. Analysis of Poincaré plot
HRV indices of resting 1 h after midday milking
(reference period: 1) were compared to those
measured during the midday milking (2) and the
different phases (different behavioural categories)
of the evening milking: 1,5 h before evening milking (3); driving (4); in the holding pen (5); stepping into the milking parlour (6); udder preparation (7); milking (8); after milking, in the milking
stall (9); after letting out of the milking stall (10);
lying after milking (11). In statistical analyses, we
used generalised linear mixed-effects models in R
12.2.1. For comparisons between resting and the
different phases of milking process, individuals’
HRV values were compared with paired T-tests.
214
Kovács L. et. al./Scientific Papers: Animal Science and Biotechnologies, 2013, 46 (1)
during waiting for letting out the milking stall
after milking [16].
During the reference period SD1 (198.5 ms) was
significantly higher (p<0.05) than every other
measured period suggesting an increased vagal
tone and a decreased metabolic activity after milking. This parasympathetic predominance decreased with time of the day (1.5 h after milking)
reflecting a slow re-establishment of the sympatho-vagal balance.
Effect of parity was only calculated on SD2 but all
three parameters were significantly affected by
different phases of milking process (behavioural
categories) and breeding bulls (P=0.0001). Effect
of milk production was proved in two cases (HR:
P=0.018, SD2: P=0.044) (Table 2).
3. Results and discussion
HR was significant lower (90.32±5.69) in primiparous cows than in multiparous cows
(105.11±10.31) during the entire milking process
(p<0.05). There was no difference in the values of
HR, SD1 and SD2 between resting and the different phases of milking process (driving, being in
the holding pen, udder preparation, milking, being
in the milking stall after milking, after letting out
of the milking stall) suggesting that milking process in a parallel milking parlour is acceptable in
animal welfare terms. Namely, neither sudden
moving of the animals nor crowding in the parlours’ holding pen meant such load for cows
which was detectable in HRV. In opposite of this
an earlier study reported of higher stress level
Table 2. Tests of Model Effects (Wald Chi-Square+, Type III: P values)
Source
Dependent Variables
Mean HR (beats/min)
SD1 (ms)
Intercept
0.135
0.0001
Parity: 1,2
0.050
0.632
Behavioural categories: 11
0.0001
0.0001
Breeding bulls: 21
0.0001
0.0001
Age, year (covariate)
0.776
0.058
Milk production, kg (covariate)
0.297
0.018
SD2 value of Adept’s daughters (239 ms) was
higher in 7 cases than the daughters of other bulls.
The results of O-Man’s daughters and Zenit’s
daughters on SD2 differed not to the SD2 values
of the daughters of other bulls (Figure 2).
SD2 (ms)
0.001
0.002
0.0001
0.0001
0.346
0.044
Corresponding to others [13] SD1 was significantly higher (102.8 ms) in multiparous cows than in
primiparous cows (p<0.017; α=0.005) all the day
and during the entire milking process reflecting
higher PNS activity and lower levels of stress in
multiparous cows than primiparous ones.
Based on the literature, it can be stated that the
effects of milking systems on the cardiac activity
are well documented. Most of recent studies
evaluated CMS and AMS in respect to animal
welfare.
Till now, none of the studies evaluated cow’s
stress responses in parallel milking parlours,
several experiments were carried out on
conventional herringbone parlours, however.
PNS predominance was reported during milking
in a conventional milking system using the lowfrequency component of HRV [16] interpreted
with the release of oxytocin during udder
preparation and the pleasant sense of the udder’s
emptying, concealing the effects of technology.
In line with our results in conventional milking
systems considerable stress was not detectable
neither by HR [17] and HRV [18], in cows milked
Figure 2. SD2 by parity and behavioural categories
215
Kovács L. et. al./Scientific Papers: Animal Science and Biotechnologies, 2013, 46 (1)
Renen van, C. G., Stress responses during milking;
comparing conventional and automatic milking in primiparous dairy cows, Journal of Dairy Science, 2002,
85, 3206-3216.
5. Hagen, K., Lexer, D., Palme, R., Troxler, J., Waiblinger, S., Milking of Brown Swiss and Austrian
Simmental cows in a herringbone parlour or an automatic milking unit, Applied Animal Behaviour Science,
2004, 88, 209-225.
6. Kovács, L., Jurkovich, V., Nagy, K., Szenci, O.,
Tőzsér J., Welfare assessment in dairy cattle by heart
rate and heart rate variability – Literature review and
implications for future research, 2013, Animal. In
press.
7. Borell von, E., Langbein, J., Després, G., Hansen, S.,
Leterrier, C., Marchant-Forde, J., Marchant-Forde, R.,
Minero, M., Mohr, E., Prunier, A., Valance, D.,
Veissier, I., Heart rate variability as a measure of autonomic regulation of cardiac activity for assessing stress
and welfare in farm animals: a review, Physiology and
Behavior, 2007, 92, 293-316.
8. Rushen, J., Munksgaard, L., Marnet, P. G., Passillé,
de A. M., Human contact and the effect of acute stress
on cows at milking, Applied Animal Behaviour Science, 2001, 73, 1-14.
9. Wenzel, C., Schonreiter-Fischer, S., Unshelm, J.,
Studies on step–kick behavior and stress of cows during milking in an automatic milking system, Livestock
Production Science, 2003, 83, 237-246.
10. Gygax, L., Neuffer, I., Kaufmann, C., Hauser, R.,
Wechsler, B., Restlessness behaviour, heart rate and
heart-rate variability of dairy cows milked in two types
of automatic milking systems and auto-tandem milking
parlours, Applied Animal Behaviour Science, 2008,
109, 167-179.
11. Hagen, K., Langbein, J., Schmied, C., Lexer, D.,
Waiblinger, S., Heart rate variability in dairy cows –
influences of breed and milking system, Physiology
and Behavior, 2005, 85, 195-204.
12. Sayers, B. M., Analysis of heart rate variability,
Ergonomics, 1973, 16, 17-32.
13. Minero, M., Canali, E., Ferrante, V., Carenzi, C.,
Measurement and time domain analysis of heart rate
variability in dairy cattle, Veterinary Record, 2001,
149, 772-774.
14. Niskanen, J. P., Tarvainen, M.P., Ranta-aho, P. O.,
Karjalainen, P. A., Software for advanced HRV analysis, Computer Methods and Programs in Biomedicine,
2004, 76, 73-81.
15. Mohr, E., Langbein, J., Nürnberg, G., Heart rate
variability: A noninvasive approach to measure stress
in calves and cows, Physiology and Behavior, 2002, 75,
251-259.
16. Kovács, L., Nagy, K., Szenci, O., Tőzsér, J., Heart
rate variability during milking in dairy cows, Hungarian Veterinary Journal, 2012, 134, 653-661.
in automatic milking systems HR increased
significantly during a 5-min period prior to
entering the milking stall, however. HR rapidly
decreased in the first 5 min of milking [9]. This
slight stress diminishes through the period of
adaptation to the novel milking environment as
proved by investigating HR [4,11] and HRV [18].
This can explain that in our experiment
multiparous cows had lower levels of stress during
the entire milking process than primiparous cows.
In our opinion it was maybe a result of the longer
time spent in producing and habituation of milking
technology. More research is needed to find the
exact reasons of this difference.
4. Conclusions
Results suggested that milking process is not really stressful for cows, primiparous cows were more
susceptible of milking technology than multiparous ones in this experiment, however. SD1 seems
to be a reliable geometric measure of PNS activity
and physiological stress load. SD2 is affected by
several independent factors such as breeding bull,
parity and milk production.
Acknowledgements
This work was supported by TAMOP Project (TÁMOP
l–4.2.1.B–11/2/KMR–2011–0003) and Bolyai fellowship from the Hungarian Academy of Sciences.
The research was carried out within the framework of
the National Excellence Programme – Elaboration and
operation of a personal support provision system for
national researchers and students (grant No: TÁMOP
4.2.4 A/1-11-1-2012-0001). The project was granted
support from the European Union, co-financed by the
European Social Fund.
References
1. Mitchell, G., Hattingh, J., Ganhao, M., Stress in
cattle assessed after handling, transport and slaughter,
Veterinary Record, 1988, 123, 201-205.
2. Fukasawa, M., Tsukada, H., Kosako, T., Yamada,
A., Effect of lactation stage, season and parity on milk
cortisol concentration in Holstein cows, Livestock
Science, 2008, 113, 280-284.
3. Weston, E. I., Evaluation of cortisol in saliva relative
to serum in lactating cows, heifer calves and piglets in
response to applied stress, Thesis PhD, North Carolina
State University, Raleigh, USA, 2009.
4. Hopster, H., Bruckmaier, R. M., Werf van der J. T.
N., Korte, S. M., Macuhova, J., Korte-Bouws, G.,
216
Kovács L. et. al./Scientific Papers: Animal Science and Biotechnologies, 2013, 46 (1)
17. Rushen, J., Passillé, de A. M., Munksgaard, L., Fear
of people by cows and effects on milk yield, behavior,
and heart rate at milking, Journal of Dairy Science,
1999, 82, 720-727.
18. Neuffer, I., Influence of automatic milking systems
on behaviour and health of dairy cows, Thesis PhD,
University of Hohenheim, Hohenheim, Germany, 2006.
217