Heart rate variability as a measure of dairy calf welfare
What is HRV?
The beating heart speeds up from sympathetic (S) and is slowed down by parasympathetic (P) nerves acting on the heart’s pacemaker (sinoatrial node). S and P nerves have respectively slow (seconds) and fast (milliseconds) conduction velocities. S nerves are slower from the combined effect of having to go through a chemical junction (synapse) distant from the heart following which it is not insulated (un-myelinated). The P nerve is faster because its synapse is in the heart.
HRV is a measure of how quickly the heart rate changes. When the fast P nerve influences the heart rate, it causes heart rate to change more quickly than the slow S nerve.
There are several ways to present HRV; the most informative is the route mean square of sequential inter-beat differences (RMSSD) over time, measured in milliseconds (ms). Greater inter-beat differences from one beat to the next give higher RMSSD values.
It is easy to confuse heart rate with HRV, from the counterintuitive fact that the slow S nerve speeds the heart up, while the fast P nerve slows it down (making resting heart rate the inverse to HRV). The consequence of this difference in conduction velocity between opposing sides of the autonomic nervous system, is that it allows us to determine which predominates at any one time throughout the body. High RMSSD values (heart rate changing more quickly) are the result of parasympathetic or vagal dominance, while low RMSSD values show sympathetic dominance. HRV is commonly described as a measure of the vago-sympathetic balance; with good health (physical or mental) equating to parasympathetic dominance in the resting state2.
It is important to measure HRV in the resting state because positive or negative excitement triggers sympathetic activity. In chronically stressed or ill individuals, the sympathetic side dominates also in the resting state, shown by low RMSSD values.
Why use HRV to monitor farm animal welfare?
Since the 1960’s, HRV has been utilised in human obstetrics using foetal heart monitors, which detect beat to beat variability to monitor stress in near term babies3. By the 1990’s HRV was being cited widely in scientific literature but there was a lack of consensus over its validation. To resolve this, a joint American and European task force was set up to standardise the measurement of HRV, setting out guidelines for recording and measuring HRVin 19964.
Current medical research widely reports on the importance of vagal activity on immunological, psychological and physiological health, with large epidemiological studies showing that reduced HRV is a risk factor for all-cause mortality and morbidity2.
In 2007 a comprehensive review linked HRV with welfare in farm animals. The authors concluded that stress from physical, pathological or emotional origins can be quantified in cattle using HRV, with lower RMSSD values signifying increased stress5.
Dairy calf management: a model of compromised welfare
Dairy calf management is highly stressful1 from early weaning, involving separation from the cow, change in calf nutrition and a period of enforced isolation to reduce infection spread and monitor feeding. Since most dairy breeds are horned, we must also add the stress from disbudding.
Measuring HRV in dairy calves
Calf heartbeats were recorded using the Polar Equine® girth mounted heart monitor, which incorporates two electrodes and a transmitter in an elasticated belt (Figure 1). The wristwatch recorder, that collects the data, was fastened to the belt; this data was then downloaded to a laptop for analysis. The calves did not appear averse to wearing the monitors, with their resting heart rates typically returning to normal within five minutes of them being applied. The use of electrogel ensured a good skin/electrode connection making hair removal unnecessary.
To conform to the 1996 Task Force requirements4, a five minute period of heartbeat data, after removing any errors, was selected to calculate the RMSSD value, using Artiifact® software. Typically the heart monitor only needed to be on the calf for 30 minutes to achieve a suitable resting RMSSD measurement.
HRV after separation from cow
RMSSD values in 12 Friesian dairy calves were measured on the day and three days after separation from their mothers. The age range of calves at separation was from 12 hours to five days old. There was a significant negative correlation between age at separation and the RMSSD value (r2 = -68%, p<0.05), that was still significant three days after separation (r2 = - 35%, p<0.05).
This showed that separation stress in the calf is not transient and increases the longer it has to form a bond with its mother.
HRV and dummy teats
Dummy teats were used to model another aspect of weaning stress in the calves. These were constructed from a board with a centrally drilled hole to hold a rubber teat (Figure 2). Dummy teats were fixed to each front gate of eight isolated Friesian dairy calves (<two weeks old) for five days. Concurrently nine other similar aged isolated calves were not provided with dummy teats. Mean RMSSD values increased significantly in both groups of calves with and without dummy teats. This reflected the normal increase over time of HRV at this age. But one day after dummy teat removal in the test group, the mean RMSSD value decreased significantly, showing the stress from this aspect of weaning (Figure 3).
HRV and isolation-induced social mixing stress
RMSSD values in 12 Friesian dairy calves were measured three days after being transferred to group pens following varying periods of time in isolation (range three to 28 days). There was a significant negative correlation between the time a calf spent in isolation and its RMSSD value after joining others (r2 = -60%, p<0.05). This confirmed isolating calves prevented them from developing necessary social skills, causing increased mixing stress1.
HRV and disbudding stress
RMSSD values were compared the day before and on the two days after hot iron disbudding in 18 Friesian dairy calves (mean 35 days old). To determine the effect of an analgesic on disbudding pain and stress, eight of the calves were given subcutaneously 0.5 mg/kg of the non-steroidal anti-inflammatory drug (NSAID) meloxicam (Metacam®, Boehringer Ingelheim, Germany), when the local anaesthetic was administered. And just the local anaesthetic was given in the control group. Following disbudding mean RMSSD values declined in both the meloxicam treated and untreated calves. The decline was only significant at 48hr in untreated calves, suggesting meloxicam was slowing the decline in RMSSD values by alleviating chronic pain at 48hrs (Figure 4).
To determine the longer term effect of meloxicam, a second study measured HRV in 23 Friesian dairy calves (mean nine days old) before and one week after hot iron disbudding. 11 calves were given meloxicam. One week after disbudding the meloxicam treated calves had significantly higher RMSSD values, while untreated calves did not significantly change. This result showed meloxicam preserved the normal increase in HRV of such young calves. The physiological increase in HRV was eliminated in untreated calves by the chronic pain following hot iron disbudding (figure 4).
HRV and illness
At various times during the study some calves became ill, including twins allowing valuable comparisons to be made. Typically sick calves, identified by fever and elevated heart rate, had RMSSD values below 5ms while healthy calves were greater than 10ms (Figure 5).
These results provide evidence that HRV can be used in the assessment of pain and stress after disbudding. In addition, using this measure, meloxicam (Metacam®) showed potential in alleviating chronic pain 48hrs and one week after hot iron disbudding.
REFERENCES 1. Phillips. 2002. Cattle behaviour and welfare. 2nd Ed. Blackwell Science Ltd Oxford, UK.
2. Thayer et al. 2012. Neurosci Biobehav R 36(2):747–756
3. Hon EH. 1958. Am J Obstet Gynecol. 75:1215
4. Task-Force 1996. Circulation 93: 1043–1065.
5. von Borell et al. 2007. Physiol behav 92 , 293-316
6. Kaufmann et al. 2011. Behav Res Methods, 43(4): 1161 –1170.