Effects of a topically applied anaesthetic on the behaviour, pain sensitivity and weight gain of dairy calves following thermocautery disbudding with a local anaesthetic

ABSTRACT

Aims: To compare the effect of a topically applied anaesthetic to no pain relief or meloxicam on the behavioural responses, pain sensitivity and weight gain of calves following disbudding with or without sedation.

Methods: A total of 364, 2–6 week-old calves from three commercial farms were systematically allocated to one of six treatment groups. All calves received a cornual nerve block prior to disbudding, with half restrained in a crate and half sedated with xylazine. Within these groups one third received no further treatment (control), one third were treated with meloxicam >10 minutes prior to disbudding and one third received a topical anaesthetic applied to the horn bud wounds following disbudding. The frequency of ear flicks, head shakes, head scratches and pain sensitivity of the wound were recorded on up to eight occasions over 24 hours after disbudding. Calves were weighed before, and 7 and 28 days after, disbudding to determine average daily weight gain (ADG).

Results: Compared to calves in the crate-control group, all other groups had reduced ear flicks at all times following disbudding (p < 0.01). Treatment with meloxicam and topical anaesthesia in addition to sedation reduced head scratches compared to calves in the crate-control group (p ≤ 0.013). At 22 hours after disbudding head shakes were reduced in sedated calves treated with topical anaesthetic compared to calves in the crate-control group (p < 0.001). Pain sensitivity was lower in all sedated calves than unsedated calves (p < 0.001). The ADG between Days 0–7 was 0.14 (95% CI = 0.015–0.274) kg/day greater in sedated calves treated with meloxicam than calves in the crate-control group (p = 0.03), and the ADG between Days 0–28 tended to be 0.06 (95% CI=−0.01–0.13) kg/day greater in sedated calves treated with topical anaesthetic than calves in the crate-control group (p = 0.09).

Conclusion and clinical relevance: Sedation of calves for disbudding reduced the pain experienced in the following 24 hours. There was a benefit to providing calves with topical anaesthetic following disbudding on behavioural responses and pain sensitivity, which was similar to that of treating calves with meloxicam.

KEYWORDS:

• Disbudding
• dehorning
• topical anaesthetic
• pain relief
• calves
• behaviour
• weight gain

Introduction

In New Zealand large numbers of calves are disbudded every year, usually within the first 2 months of life. While currently, disbudding is permitted without local anaesthesia, new regulations due to be introduced on 1 October 2019 require that the disbudding of all calves in New Zealand must be undertaken using an appropriately placed and effective local anaesthetic (Anonymous 2018). However, while disbudding with local anaesthesia is effective for reducing physiological and behavioural responses during disbudding (Petrie et al. 1996), once the local anaesthesia wears off (within 2–3 hours), behaviours such as ear flicking and head shaking increase, as do concentrations of cortisol, with these pain-associated changes lasting for at least 24 hours after disbudding (Faulkner and Weary 2000; Sutherland et al. 2002).

The pain associated with disbudding can be reduced by giving a systemic non-steroidal anti-inflammatory drug (NSAID; Stafford et al. 2003; Heinrich et al. 2010; Bates et al. 2015) and sedation (McMeekan et al. 1999; Bates et al. 2016), although neither of these are effective at eliminating the physiological and behavioural responses during disbudding. Furthermore, whilst many calves in New Zealand are disbudded under sedation (N. Harding¹ pers. comm.), the use of analgesics in calves disbudded, either under sedation or whilst unsedated and restrained in a crate, is limited due to the additional cost and perceived lack of economic benefit.

The welfare of dairy calves is coming under greater scrutiny as consumers increasingly consider animal welfare when purchasing food (Verbeke 2009). Therefore there is a need to reduce the impact of disbudding on calf welfare beyond the use of just local anaesthesia, and for a practical and cost effective way of delivering additional pain relief during and after disbudding. The only registered topical anaesthetic drug for large animal use is an antiseptic gel spray containing lignocaine, bupivacaine and adrenaline (Tri-Solfen; Bayer Australia Ltd.), which was originally registered in Australia for use in lambs to provide pain relief following mulesing and tail docking. Since 2016 Australian registration has included provision of pain relief during and following surgical castration of lambs and calves, and from June 2018 it has been registered for use after calf disbudding or dehorning.

This product was evaluated in 30 Holstein-Friesian, 2 month old calves, when applied after scoop dehorning, compared to no pain relief, and was found to reduce or remove responses to mechanical stimulation from 4–24 hours after application (Espinoza et al. 2013). In another study the topical anaesthetic, applied after scoop dehorning, was compared with a cornual nerve block applied prior to dehorning, and found no difference in pain sensitivity after dehorning (McCarthy et al. 2016). However this study had errors in its statistical analysis and was grossly underpowered to detect any differences. Another study, using 6–8 month-old Bos indicus calves undergoing concurrent castration and amputation dehorning, investigated the use of the topical anaesthetic and buccal meloxicam, without any local anaesthesia, on post-operative weight gain and behaviour. Very little consistent effect of treatment was found on most of the measured behaviours or weight gain over the first 6 days after castration and dehorning (Van der Saag et al. 2018b). In a similar study using 6–8 month-old Hereford calves undergoing amputation dehorning alone, there was no clear effect of topical anaesthetic or buccal meloxicam on behaviour, maximum wound temperature or wound morphology after dehorning (Van der Saag et al. 2018a). These studies have limited relevance outside Australia and similar extensive range farming systems, as they did not investigate the use of topical anaesthetic alongside local anaesthesia. Topical anaesthetics cannot be used to control pain during disbudding, as would be legally required in many jurisdictions, including New Zealand after 1 October 2019. Furthermore, all the studies used amputation or scoop dehorning of 2–8 month-old calves, rather than hot iron disbudding of 2–3 week-old calves which is much more common on New Zealand dairy farms. In addition no studies have evaluated the use of topical anaesthetic in calves sedated with xylazine prior to disbudding, which is the standard practice for veterinarian-led disbudding on dairy farms in New Zealand.

The aims of this study were to investigate the effect of a topically applied local anaesthetic following disbudding, using either a crate for restraint or sedation, in comparison to either the NSAID meloxicam administered at disbudding or no post-disbudding pain relief, on the behavioural responses, pain sensitivity and weight gain of calves.

Materials and methods

Calf selection and allocation to groups

All animal procedures were approved by the Ruakura Animal Ethics Committee (AgResearch, Hamilton, NZ). This study was conducted between 26 March and 4 May 2018 on three conveniently located, autumn-calving, commercial farms in the Waikato region of the North Island of New Zealand.

Two weeks prior to the study start date, an electronic list of the identification, date of birth and breed of all calves born between 1st February 2018 up until the 14 days prior to the study start date were provided by the farmer. All calves between 2–6 weeks of age on the study start date (Day 0) were eligible to be examined for selection in the trial.

All eligible calves were examined on Day −1 for signs of disease (e.g. navel ill, scours, pneumonia) by a veterinarian and checked for the presence of horn buds. Calves with disease or polled were not eligible for the trial. Every healthy, eligible calf was weighed on electronic weigh scales. On Farms 1 and 2, where there were smaller numbers of calves, the age of the calves ranged from 2–6 weeks so calves were ordered by weight and systematically allocated into one of six groups. On Farm 3 there were 60 calves per visit, which were first categorised according to age and then within age category they were ordered by weight and systematically allocated into one of six groups. The mean weight and age of the six groups were determined to ensure there were no major discrepancies between groups.

A description of the treatment groups is shown in Table 1. As there were multiple outcomes measured for the trial, calves were allocated by cohort for either behavioural observation or pain sensitivity testing (Figure 1). The first cohort of eligible calves on a farm were all allocated for behavioural observation. The second cohort of calves that were disbudded either days or weeks later, depending on the farm and age of the calves, were all allocated to pain sensitivity testing. There were two cohorts per farm and behavioural observations and pain sensitivity measurements were separated by between 2 days and 3 weeks.

Figure 1. Diagrammatic representation showing the allocation within farms of calves to be disbudded in a crate or sedated, and to receive no further treatment (C), or receive a topical anaesthetic (TA) or meloxicam (M). The first cohort was observed after disbudding for behavioural measurements, and 3–21 days later the second cohort had pain sensitivity measurements recorded after disbudding.

Table 1. Description of the six different treatment groups used to investigate the effect of a topical anaesthetic (TA)^a on the behaviour, pain sensitivity and weight gain of dairy calves following cautery disbudding.

Group	Treatment
1	Crate disbudding, local anaesthetic and TA placebo	5 mL lignocaine^b applied to each horn bud, 4 mL TA placebo after disbudding
2	Crate disbudding, local anaesthetic and meloxicam and TA placebo	5 mL lignocaine^b applied to each horn bud, 1.4 mL S/C meloxicam^c, 4 mL TA placebo after disbudding
3	Crate disbudding, local anaesthetic and TA	5 mL lignocaine^b applied to each horn bud, 4 mL TA after disbudding
4	Sedation disbudding, local anaesthetic and TA placebo	0.8 mL I/M xylazine^d, 5 mL lignocaine^b applied to each horn bud, 4 mL TA placebo after disbudding
5	Sedation disbudding, local anaesthetic and meloxicam and TA placebo	0.8 mL I/M xylazine^d, 5 mL lignocaine^b applied to each horn bud, 1.4 mL S/C meloxicam^c, 4 mL TA placebo after disbudding
6	Sedation disbudding, local anaesthetic and TA	0.8 mL I/M xylazine^d, 5 mL lignocaine^b applied to each horn bud, 4 mL TA after disbudding

^aTri-Solfen (Bayer New Zealand Ltd, Auckland, NZ).

^b20 mg/mL lignocaine (Bomacaine, Bayer New Zealand).

^c20 mg/mL meloxicam (Metacam20; Boehringer Ingelheim NZ Ltd, Auckland, NZ).

^d20 mg/mL xylazine (Xylazine 2% Injection; Phoenix Pharm Distributors, Auckland, NZ).

Calves assigned to have behavioural observations recorded had their number spray marked along their sides and their backs, were separated into their designated groups and were placed in six separate pens within the same calf house on Day −1. Calves assigned to pain sensitivity testing were separated into two groups; one for sedation and one for crate disbudding.

Experimental procedure

All calves were fed the evening before Day 0 but not on the morning of Day 0. Two teams of technicians carried out the disbudding following administration of local anaesthetic nerve blocks; one team of two disbudded calves while they were fully conscious, restrained in a crate and the other team of two disbudded following sedation. The two teams worked simultaneously on the groups of calves. The same person in the group of two administered all the nerve blocks and the other team member did all the disbudding.

For calves that were sedated, they were injected with 3–4 mg (0.15–0.2 mL)/10 kg of 20 mg/mL xylazine (Xylazine 2% Injection; Phoenix Pharm Distributors, Auckland, NZ) I/M in the neck. For the majority of calves, this was a dose of 0.8 mL (16 mg), but varied from 0.6–1.2 mL (12–24 mg), depending on the weight of the calf. Animals were marked once injected to ensure the same animal was not injected twice.

In all calves both horns were desensitised using a cornual nerve block at least 10 minutes prior to disbudding, by injecting 5 mL (100 mg) of 20 mg/mL lignocaine (Bomacaine; Bayer New Zealand, Auckland, NZ) at depth of approximately 1 cm under the temporal ridge around each cornual nerve.

Calves that were allocated to receive meloxicam were injected with 50 mg/100 kg (1.4 mL) of 20 mg/mL meloxicam (Metacam20; Boehringer Ingelheim NZ Ltd, Auckland, NZ) S/C in the neck following the cornual nerve block and at least 10 minutes prior to disbudding.

In all calves the horn bud was prepared first by shaving the hair around the site using electric clippers. Disbudding was performed by thermocautery using a gas heated clean instrument with a 18 mm diameter head (Express LPG or Farm Hand disbudder, Shoof Hamilton, NZ). The horn bud was removed completely during the procedure. If calves showed an indication of pain after disbudding started, disbudding was stopped and the cornual nerve block was repeated.

Following disbudding the wound at each horn bud was treated with either 2 mL of topical anaesthetic solution (Tri-Solfen (Bayer New Zealand Ltd), containing 40.6 g/L lignocaine hydrochloride, 4.2 g/L bupivacaine hydrochloride, 24.8 mg/L adrenaline (as acid tartrate) and 5.0 g/L cetrimide) or 2 mL of a placebo. The placebo was identical to the topical anaesthetic but did not include any lignocaine, bupivacaine, adrenaline or cetrimide. The solutions were delivered using a hand held applicator (V-Grip 6 mL applicator; Genesis Industries, Mudgee, NSW, Australia). In some cases when it ran off the wound site excess was caught with a towel on the side of the head to avoid product going into the eyes of calves.

After disbudding, all calves that were allocated to behavioural observations remained in their separate pens until 24 hours following disbudding. All calves that were allocated to the pain sensitivity testing were in two joined pens (one for sedated calves and one for crate calves) for 24 hours. After 24 hours all calves were joined together. They were provided with clean water, a new batch of calf meal and a milk feed. The remainder of the feeding was as per usual farm practice.

Behavioural observation

Calves in the pens were video recorded, starting at 12:00 hours on Day−1 and continuing until 09:00 hours on Day 1. Four Arlo Pro 2 (Arlo Technologies Inc., San Jose, CA, USA) video cameras with night vision were mounted near the roof of the calf pens. They were positioned such that all six treatment pens were observable in at least one of the camera recordings, and were tested to ensure that they were able to capture all areas of the calf pens.

Approximately 7–10 minutes of video were downloaded at the start of each of hour, at 0, 1, 3, 5, 7, 9, 12 and 22 hours following disbudding, onto Arlo Technologies cloud system, which were then subsequently copied onto USB drives for observations. Therefore for each individual calf a total of ∼65–80 minutes of footage was recorded over 24 hours.

A total of 22 trained observers watched the video footage for ear flicking, head shaking and head scratching by individual calves, as described in Table 2. The observers were supervised by two of the authors (ELC and WAM) and worked in pairs to watch the footage together on a 50 cm computer screen. The observers were provided with a selection of specific videos and were instructed on what pen they were observing, the calf identification numbers in that pen and the time period of observation. They each watched a single calf in the pen, then another until all calves were observed once for that time. They were encouraged to re-watch the footage and ask for help from supervisors for certainty. They recorded the calf number, the description of the calf, and the count of the behaviours shown throughout the time periods. After a pair had finished watching all the calves once, another pair of observers followed the same procedure with the same videos. Therefore, each calf at each time was watched by between 2–4 separate trained observers who were unaware of any previous observation results. The calves that were positioned in the pens furthest away from the cameras were observed more frequently. The observers had no knowledge of the trial other than the behaviours that they had to record and were therefore completely blinded.

Table 2. Definitions of behaviours that were observed in dairy calves following cautery disbudding.

Behaviour	Description
Ear flicking	Calf rapidly moves one or both ears to the front and back independent of a head shake. Each time movement constitutes as an ear flick.
Head shaking	Calf rapidly shakes head from one side to the other. Recorded as one behaviour when the head moved slowly or was in a resting position.
Head scratching	Calf lifts hind leg to scratch top of head with foot or rubs head against sides of the pen.

At the completion of the observations, the data were collated so all the observations of a calf from a farm at a set time could be compared to determine discrepancies between observers. A supervisor then viewed all of the videos with discrepancies in recorded observations of >3 between observers. In cases of clear error (e.g. 2–3 observers reported 5–7 head shakes and one observer reported 18), the supervisor made a judgment call on whether to use the median result, remove a single observer or remove multiple observers for that observation, so that there was one measurement for each calf at each time.

Activity monitoring using HeyrexVet (Heyrex, Wellington, NZ) wearable monitors was also undertaken. As this activity monitor has not been validated in calves the methods and results are described in Supplementary Information 1.²

Pain sensitivity testing

It was originally intended that pain sensitivity testing would be conducted at 1, 2, 3, 4, 6, 8, 10, 12 and 24 hours after disbudding, using a pressure algometer (FPX 50, Wagner Instruments, Greenwich, CT, USA) applied at eight locations around the horn bud, as shown in Figure 2. This required placing the algometer rubber tip over the cautery wound at the edge of normal tissue. However on the first farm where pain sensitivity testing was carried out the veterinarian undertaking the testing immediately expressed concern about the level of pain that the calves were experiencing. Therefore the placement of the algometer was limited to positions 1 and 2 on each horn bud (Figure 2) and testing was carried out at 0–1, 2–5, 6–11, 12 and 24 hours after disbudding.

Figure 2. Diagram showing the position of pressure algometer placement for measurement of pain sensitivity in calves following disbudding. The black ovals indicate the position of the horn bud wound and the numbers indicate the location for placement of the pressure algometer.

Testing was conducted with two people restraining the calf. The veterinarian carrying out the testing initially placed a hand lightly on the poll of the calf until the calf was habituated to being touched and stood still. The hand was then removed and replaced with the algometer rubber tip at the two positions around each horn bud wound. The force (N) that was able to be applied against the wound before the calf moved away from the stimulus was recorded.

Half of the calves were tested on the right horn first, followed by the left horn, with the remainder tested in the opposite order. All calves returned to typical farm management (in one group) following the completion of the trial at 24 hours. All animal health events were recorded and any animals that were suffering illnesses were removed from the trial. Any calves that expressed an unusually high aversion to pain sensitivity testing (struggling, vocalisation) were removed from further assessments on welfare grounds.

In addition to the algometer measurements, the person applying the algometer also subjectively graded the calf’s pain response as low, moderate, high or extreme. The methods and results for this subjective assessment are presented in Supplementary Information 2.³

Weight gain

All calves were weighed on electronic weigh scales on Days 7 and 28. For the Day 7 visit the calibration of the scales was identified as being incorrect for 30 calves and these animals were removed from weight analysis for Day 7, but were included in the analysis for Day 28.

Statistical analyses

The sample size for total numbers of enrolled animals was estimated based on 95% confidence with 80% power of detecting a weight gain difference between groups, over 28 days, of 2 kg with a variance of 49 kg. This resulted in 284 calves being required across six groups. An extra 80 calves were enrolled due to the likelihood of losing animals to follow up after enrolment.

For the behavioural observations approximately 90 animals, 16 in each group, were required to detect an approximate 40% difference in the incidence of head shakes between groups. Approximately 132 animals were required for the pain sensitivity component of the trial, to detect a difference between groups of 7 N with a SD of 10 N, at any single time.

Behavioural observation

Data for the total number of ear flicks, head shakes and head scratches recorded at 0, 1, 3, 5, 7, 9, 12 and 22 hours after disbudding were analysed separately. The data for ear flicks and head scratches were modelled using two separate mixed effects negative binomial regression models, where calf was treated as random effects, and the natural logarithm of length of video recorded at each time period was added as an offset term. For head shakes, a mixed effects Poisson regression model with a robust standard error was adopted, which also included calf as a random effect and the same offset term. Initial explanatory variables included in the multivariable models were farm, treatment group and time, and the interaction between treatment group and time. This interaction term was included as it was possible that the treatments had a different duration of action. Starting with a full model that included the interaction term, fixed effects were dropped from the model if Akaike information criteria from the simpler model was smaller. Results are reported as incidence rate ratios (IRR), which can be interpreted as the increase, or reduction, in the rate of the behaviour per minute per calf, compared to a reference category. An IRR > 1.0 indicates an increase in the rate of a behaviour (i.e. more painful), and IRR < 1.0 indicates a reduction in the rate of a behaviour (i.e. less painful).

No differences were noted between 0, 1 and 3 hours after disbudding, so these results were re-categorised as 0–3 hours after disbudding. Further analyses of the three behavioural responses were carried out using similar models to compare results for calves sedated or not, treated with meloxicam or not, and topical anaesthetic or not.

Pain sensitivity

A small subset of calves had an extreme subjective pain response in which they head-butted the algometer, resulting in a very high algometer reading. As these values would bias the results, they were removed from the analysis. Results from algometer readings recorded at the four locations on each calf were included in the analysis, and measurements recorded at different times were re-categorised as 0–1, 2–5, 6–11 and 12–24 hours after disbudding.

A linear mixed model, with calf as a random effect, was used to analyse the algometer readings. As it was likely that sedated calves would be poorly responsive immediately after disbudding, treatment groups were initially collapsed into sedated and non-sedated, and an interaction term for time and sedation status was added to the model. A subset analysis was then performed with results from 0–1 hours removed, with the six treatment groups, farm, location on the calf and time as fixed effects. Starting with a full model, fixed effects were dropped from the model if the log-likelihood ratio test between two nested models had a significance of p > 0.05. Model diagnostics were examined to assess for outliers and influential observations.

An important assumption of linear modelling is that the variation of the residuals is constant across all ranges of the predicted outcomes and associated risk factors. With this study, the variation in algometer readings varied considerably between farms, with the algometer readings being greater in Farm 1 than Farms 2 and 3, as was the variation in the readings. This was corrected by modelling different variance weightings for each farm, thereby ensuring each farm had equal weighting when standard errors were used to compare farms.

Weight gain

Weight gain was calculated for each calf as the average daily gain (ADG) between Days 0 and 7, and between Days 0 and 28. Linear regression multivariable models were used to analyse the ADG for each time period, including the explanatory variables of farm, treatment group and weight at enrolment. An interaction term between treatment group and farm was also included in the original multivariable model, as it was biologically plausible that the treatments could have different effects on different farms. Starting with a full model that included the interaction term, fixed effects were removed from the model if the log-likelihood ratio test between two nested models had a significance of p > 0.05. Model diagnostics were examined to assess for outliers and influential observations.

All statistical analyses were carried out using R v3.3.3 (R Core Team, 2017; R Foundation for Statistical Computing, Vienna, Austria), apart from the negative binomial modelling, that was carried out using Stata 13 (StataCorp, College Station, TX, USA)

Results

The number of calves enrolled in each treatment group on the three farm is presented in Table 3. Nine calves were excluded following enrolment; one calf that aspirated wood chips whilst under sedation and eight were found to be polled. One further calf lost weight between Day 7 and 28, it was found that this animal had been examined by a veterinarian on the morning of Day 7 and had been identified with diarrhoea and ill-thrift. It was in the sedation-meloxicam group on Farm 3 and was excluded from the analysis.

Table 3. Number of calves enrolled on three farms in six different treatment groups used to investigate the effect of a topical anaesthetic (TA) on dairy calves following cautery disbudding.

Treatment	Farm
	1	2	3	Total
Crate-control	10	6	47	63
Crate-meloxicam	10	6	46	62
Crate-TA	10	6	46	62
Sedation-control	10	6	44	60
Sedation-meloxicam	9	6	43	58
Sedation-TA	10	6	43	59
Total	59	36	269	364

Behavioural responses

Behavioural responses were recorded and analysed for 20, 30 and 44 calves, on Farms 1, 2 and 3, respectively, across eight time periods.

Ear flicks

The median number of ear flicks per minute for calves in each treatment group over time are presented in Figure 3a. Results of the final model for variables associated with the number of ear flicks per minute are shown in Table 4. Compared to calves in the crate-control group, the number of ear flicks per minute were reduced in all other treatment groups (p ≤ 0.006). The effect was similar for calves treated with meloxicam or topical anaesthetic.

Figure 3. Median number of (a) ear flicks, (b) head scratches and (c) head shakes per minute recorded in calves after disbudding in crates with no further treatment (red solid line), or were treated with meloxicam (brown dotted line), or a topical anaesthetic (green dashed line), or were sedated with no further treatment (turquoise dashed line), or were also treated with meloxicam (blue dotted line), or a topical anaesthetic (pink dashed line).

Table 4. Results of the mixed effects negative binomial regression model for variables associated with the number of ear flicks per minute recorded in dairy calves (n = 94) on three farms between 0–22 hours following cautery disbudding, that were allocated to six treatment groups (see Table 1).

Variable	IRR	95% CI	P-value^a
Intercept	0.04	0.03–0.06	<0.001
Farm
Farm 1			Ref
Farm 2	0.63	0.46–0.87	0.005
Farm 3	0.66	0.49–0.89	0.007
Time (hours)
0–3			Ref
5	1.96	1.54–2.46	<0.001
7	2.29	1.85–2.84	<0.001
9^b	3.65	2.56–5.20	<0.001
12	0.18	0.12–0.28	<0.001
22	2.16	1.75–2.66	<0.001
Treatment
Crate-control			Ref
Crate-meloxicam	0.55	0.37–080	0.002
Crate-TA^c	0.54	0.37–0.78	0.001
Sedation-control	0.58	0.40–0.86	0.006
Sedation-meloxicam	0.59	0.41–0.85	0.004
Sedation-TA	0.22	0.14–0.33	<0.001

^aSignificance of incidence rate ratio (IRR).

^bThe number of ear flicks per minute was 3.65 (95% CI = 2.56–5.20) times greater at 9 hours than at 0–3 hours.

^cThe number of ear flicks per minute was 0.54 (95% CI = 0.37–0.78) times lower in calves treated with topical anaesthetic following disbudding in a crate than calves not treated with topical anaesthetic.

IRR = Incident rate ratio; Ref = reference category; TA = topical anaesthetic.

The number of ear flicks per minute varied with time after disbudding (p < 0.001), increasing from 5–9 hours after disbudding, decreasing at 12 hours when the majority of the calves were sleeping, then increasing again at 22 hours following disbudding. There was no interaction between time and treatment group on the number of ear flicks per minute, indicating that the effect of treatment did not vary with time after disbudding.

The number of ear flicks per minute was higher in calves from Farm 1 than calves from the other two farms (p ≤ 0.007), but there was no interaction between farm and treatment group.

Head scratches

The median number of head scratches per minute for calves in each treatment group over time are presented in Figure 3b. Results of the final model for variables associated with the number of head scratches per minute is shown in Table 5. For calves that were disbudded while restrained in a crate, the addition of meloxicam or topical anaesthetic did not reduce the number of head scratches per minute (p > 0.6), however treatment with meloxicam and topical anaesthesia in addition to sedation reduced the number of head scratches per minute compared to calves in the crate-control group (p ≤ 0.013). The number of head scratches per minute was higher in calves from Farm 1 than calves from Farms 2 or 3 (p ≤ 0.002).

Table 5. Results of the mixed effects negative binomial regression model for variables associated with the number of head scratches per minute recorded in dairy calves (n = 94) on three farms between 0–22 hours following cautery disbudding, that were allocated to six treatment groups (see Table 1).

Variable	IRR	95% CI	P-value^a
Intercept	0.002	0.001–0.003	<0.001
Farm
Farm 1			Ref
Farm 2	0.44	0.27–0.71	0.001
Farm 3	0.51	0.33–0.78	0.002
Time (hours)
0–3			Ref
5	1.98	1.25–3.14	0.004
7	1.60	1.06–2.42	0.026
9^b	2.85	1.58–5.14	<0.001
12	0.38	0.17–0.85	0.019
22	1.55	1.02–2.34	0.038
Treatment
Crate-control			Ref
Crate-meloxicam	1.04	0.58–1.84	0.906
Crate-TA	1.15	0.66–2.01	0.63
Sedation-control	0.62	0.35–1.11	0.105
Sedation-meloxicam	0.47	0.26–0.85	0.013
Sedation-TA^c	0.44	0.23–0.83	0.011

^aSignificance of incidence rate ratio (IRR).

^bThe number of head scratches per minute was 2.85 (95% CI = 1.58–5.14) times greater at 9 hours than at 0–3 hours.

^cThe number of head scratches per minute was 0.44 (95% CI = 0.23–0.83) times lower in sedated calves treated with topical anaesthetic following disbudding than calves not treated with topical anaesthetic following disbudding in a crate.

IRR = Incident rate ratio; Ref = reference category; TA = topical anaesthetic.

Overall, the number of head scratches per minute was reduced by 53 (95% CI = 35–67)% in sedated calves compared to non-sedated calves across the 24-hour period (p < 0.001).

Head shakes

The median number of head shakes per minute over time for the six treatment groups are presented in Figure 3c. In the final mixed effects Poisson regression model there was an interaction between treatment group and time (p < 0.01). Therefore the overall effect of treatment group could not be reported.

For all times from 0 to 12 hours, there was no statistically significant difference in the IRR for each treatment group (p = 0.29). However, at 22 hours, the number of head shakes per minute differed between treatment groups; compared to calves in the crate-control group, the IRR for calves in the crate-topical anaesthesia group was 0.43 (95% CI = 0.20–0.91), for calves in the sedation-control group was 0.47 (95% CI = 0.22–0.99) and for sedation-topical anaesthesia group was 0.16 (95% CI = 0.06–0.38). The number of head shakes per minute was similar for calves in the crate-control and crate-meloxicam (p = 0.93) and sedation-meloxicam (p = 0.08) groups. The number of head shakes per minute differed between farms (p < 0.001).

Pain sensitivity

A total of 2,469 algometer test results from 29, 36 and 59 calves on Farms 1, 2 and 3, respectively, were used in the analysis. The raw means and range of all measurements recorded at each time period, across all farms, for each of the six treatment groups are presented in Table 6. Because they had an extreme pain response, and an algometer reading > 40 N, 25 test results from nine calves on the three farms were removed from further analyses.

Table 6. Mean (min, max) algometer readings (N) for all pain sensitivity measurements in dairy calves (n = 124) on three farms between 0–24 hours following cautery disbudding, that were allocated to six treatment groups (see Table 1).

Treatment	0–1 hour	2–5 hours	6–11 hours	12–24 hours
Crate-control	25.7 (0.2, 73.2)	17.5 (0.6, 79.8)	16.1 (0.2, 66.0)	24.6 (2.4, 83.2)
Crate-meloxicam	21.9 (0, 70.2)	21.3 (1.0, 87.6)	18.6 (0.6, 86.8)	25.8 (1.6, 66.8)
Crate-TA	27.4 (0, 64.8)	22.0 (0.2, 76.4)	17.7 (0.4, 61.0)	25.2 (1.6, 66.8)
Sedation-control	38.6 (11.7, 86.0)	26.7 (3.6, 77.6)	22.2 (0.2, 82.4)	29.9 (3.4, 79)
Sedation-meloxicam	40.8 (3.6, 89.2)	31.8 (2.0, 93.8)	25.2 (0.2, 99.8)	29.9 (3.4, 89.0)
Sedation-TA	42.5 (4.2, 94.6)	36.4 (3.4, 100.4)	23.5 (1.4, 84.2)	30.0 (2.8, 84.0)

TA = Topical anaesthetic.

For the model comparing pain sensitivity between sedated and non-sedated calves there was an interaction between time and treatment (p < 0.001). At 0–1 hours, calves that had been disbudded under sedation had 15.8 (95% CI = 13.2–18.5) N greater force applied to the disbudding wound before they responded compared to calves that were disbudded and restrained in the crate (p < 0.001).

The results of the linear regression model for pain sensitivity after measurements from 0–1 hours had been excluded are presented in Table 7. There was still a significant effect of sedation, with a mean 6.5 (95% CI = 1.6–11.4) N greater force being applied to sedated calves before they showed a withdrawal response compared to calves that were disbudded and restrained in the crate (p < 0.001). Calves in all the sedation groups had a greater force applied before they responded than the calves in the crate-control group (p < 0.01), however responses were similar between calves that were given meloxicam or topical anaesthetic in either the crate or sedation groups. There was no statistically significant difference between pain sensitivity in calves in the crate-meloxicam or crate-topical anaesthesia groups in comparison to the crate-control group (p > 0.4).

Table 7. Results of the linear regression model for variables associated with algometer readings (N) for pain sensitivity recorded in dairy calves (n = 1,828 measurements from 124 calves) on three farms between 2–24 hours following cautery disbudding, that were allocated to six treatment groups (see Table 1).

Variable	Estimate	95% CI	P-value^a
Intercept	15.52
Treatment
Crate-control			Ref
Crate-meloxicam	1.09	−3.63–5.80	0.65
Crate-TA	1.89	−2.98–6.75	0.45
Sedation-control	6.5	1.64–11.35	0.01
Sedation-meloxicam	8.48	3.61–13.35	<0.001
Sedation-TA^b	9.10	4.30–13.90	<0.001
Farm
Farm 1			Ref
Farm 2	−3.5	−7.23–0.23	0.07
Farm 3	15.42	11.92–18.92	<0.001
Time (hours)
12–24			Ref
2–5	2.82	1.41–4.24	<0.001
6–11^c	−5.31	−6.37 to −4.25	<0.001
Location
Left top			Ref
Left side	−5.53	−6.80 to −4.25	<0.001
Right top	4.85	3.56–6.13	<0.001
Right side	−1.03	−2.31–0.25	0.11

^aSignificance of estimate.

^bThe force required to elicit a pain response was 9.10 (95% CI = 4.30–13.90) N greater for sedated calves treated with topical anaesthetic following disbudding than for calves not treated with topical anaesthetic following disbudding in a crate.

^cThe force required to elicit a pain response was 5.31 (95% CI = 4.25–6.37) N less for calves tested between 6–11 hours after disbudding than for calves tested at 22–24 hours.

Ref = reference category; TA = topical anaesthetic.

The force required to elicit a response was lower between 6–11 hours after disbudding compared to 22–24 hours after disbudding, indicating a reduced pain sensitivity by 24 hours.

Weight gain

The ADG of calves between Days 0–7 was analysed in 310 calves and between Days 0–28 in 315 calves.

Days 0–7

Average daily gain between Days 0–7 varied between the treatment groups (Supplementary Table 2)⁴, with the ADG of calves in the sedation-meloxicam group being 0.14 (95% CI = 0.015–0.274) kg/day greater than calves in the crate-control group. Growth rates were similar between farms, and there was no interaction between treatment group and farm. Weight at enrolment was not associated with ADG between Days 0–7.

When comparing sedated and non-sedated calves, the ADG of calves disbudded under sedation was 0.10 (95% CI = 0.02–0.17) kg/day more than calves disbudded in a crate (p = 0.009). However there was no overall effect of either topical anaesthesia (p = 0.50), or meloxicam (p = 0.75) on ADG between Days 0–7.

Days 0–28

The ADG of calves between Days 0–28 varied between farms (p < 0.05), but there was no farm by treatment interaction (p = 0.73; Supplementary Table 3).⁴ The ADG of calves in the sedation-topical anaesthetic group tended to be greater than calves in the crate-control group by 0.06 (95% CI=−0.01–0.13) kg/day (p = 0.09). There was no statistically significant difference in ADG between any of the other treatment groups compared to the crate-control group.

When comparing sedated and non-sedated calves, the ADG of sedated calves was 0.05 (95% CI = 0.01–0.09) kg/day greater than non-sedated calves (p = 0.01). There was no overall effect of either topical anaesthetic (p = 0.82), or meloxicam (p = 0.74), on ADG between Days 0–28.

Discussion

This study is one of the largest studies conducted on disbudding and showed that there was a positive effect of disbudding under sedation on all outcomes measured. There was a benefit to providing calves with topical anaesthetic or meloxicam following disbudding on behavioural responses and pain sensitivity. The benefit of topical anaesthetic was not statistically different to that of meloxicam except for a reduction in ear flicks in sedated calves.

Pain can be difficult to measure as it is a subjective state. However behaviour can be a sensitive indicator of pain, over and above that of concentrations of cortisol in serum and other physiological measures (Anil et al. 2002). Behavioural responses such as those measured in this study are an accepted indicator of pain resulting from a disbudding procedure (Faulkner and Weary 2000; Stafford et al. 2003; Heinrich et al. 2010). The behavioural observations provided an important insight into the behaviour of calves for 24 hours following disbudding. Despite the use of a cornual nerve block, the observations of ear flicks, head shakes and head scratches indicated that the pain experienced by calves in the first 24 hours after disbudding was substantial.

Both the use of meloxicam or topical anaesthetic resulted in a reduction of ear flicks, in calves that were disbudded in a crate or sedated. However, the most pronounced reduction was seen in calves treated with the combination of sedation and topical anaesthetic which resulted in a 78% reduction in ear flicks and a 56% reduction in head scratches relative to calves disbudded in a crate without sedation. The equivalent reduction in calves treated with meloxicam plus sedation was 41% and 53% for ear flicks and head scratches, respectively.

It is possible that because calves within the six treatment groups were observed for behavioural responses within the same pen, there could be a perceived lack of independence between calves. For healthy calves not experiencing pain, if a particularly active calf was present it may influence the behaviour of other calves in a pen. However in this study very specific behaviour responses were assessed that have been demonstrated to be indicators of pain (Faulkner and Weary 2000; Stafford et al. 2003; Heinrich et al. 2010). It was considered very unlikely that ear flicks, head scratches or head shakes of one calf would influence the expression of these behaviours in other calves within a pen, so they were treated as independent observations.

Practically, these results suggest that if calves are disbudded by lay disbudding operations or by farmers (who cannot use sedation), there is some benefit of using topical anaesthetic or meloxicam, but to minimise the pain-associated behaviours after disbudding, sedation is required. This study has provided some evidence (based on ear flicks and head shakes) that adding topical anaesthetic to a disbudding procedure that involved sedation reduced the expression of pain-associated behaviours compared to use of meloxicam. In addition there appeared to be an extended effect of topical anaesthetic in reducing the number of head shakes at 22 hours in calves disbudded with and without sedation. Sedation alone also reduced the number of head shakes at this time, but the use of meloxicam did not.

The benefit of sedation may be a direct effect of the analgesic effect of xylazine, the moderation of hyperalgesia or reduced anxiety (Garcia-Villar et al. 1981; Browning et al. 1982), but there are some other reasons for the strong effect of sedation seen in this study. Firstly, to simplify study management, the sedated calves were kept separate from the unsedated calves. This meant that one technician provided the cornual nerve block to sedated calves and another to the calves disbudded in a crate. Thus it is possible that this confounding between sedation and technician could have contributed to the apparent effect of sedation if their techniques were different. However the technicians were extremely experienced and trained annually to the veterinary clinic’s standard operating procedure, and had also been retrained by the supervising veterinarian prior to study start. Furthermore the effectiveness of all the cornual nerve blocks were checked prior to disbudding. Another potential reason for the apparent benefit of sedation was that it would have been easier to administer the nerve block in calves that were sedated than restrained in a crate. Therefore the benefits observed after disbudding may have been due to more effective analgesia during the disbudding and not necessarily just the effect of xylazine. Responses of calves during disbudding were not recorded, and were only used as an indication that an animal did not have an effective block (as per the disbudding protocol), but could have had a relationship with the subsequently recorded responses.

Although there are no suitable comparable studies on the use of a topical anaesthetic following disbudding with a cornual nerve block, the results are comparable to other studies that have investigated behavioural responses following the use of NSAID. Heinrich et al. (2010), compared the use of meloxicam and a placebo following disbudding with a cornual nerve block and reported reductions in ear flick per hour of 70% and 83% with meloxicam 0 and 1 days after disbudding, respectively. There was also a 74% reduction in head shakes but no reduction in head scratches. Faulkner and Weary (2000) compared behavioural responses in calves disbudded with a cornual nerve block and sedation as well as treated with ketoprofen with calves not treated with ketoprofen. They found a significant reduction in the frequency of ear flicks, head shakes and head rubs during the 24 hours after disbudding in calves treated with ketoprofen. These results, as well as our study results, indicate that the behavioural responses of calves who received topical anaesthetic was comparable to the behavioural responses of calves who received NSAID.

The use of pressure algometry has previously been used in adult cattle to objectively assess pain associated with skin lesions of lame cattle (Dyer et al. 2007) as well as to assess the pain associated with dehorning. (Heinrich et al. 2010). The effect of topical anaesthetic and meloxicam on the pain sensitivity, as measured with the pressure algometer, was only apparent in calves that were sedated and was comparable between the two products, and were similar to those of calves that that were only sedated. These results were supported by the subjective pain assessment, where sedated calves had significantly reduced odds of showing high or extreme responses compared with unsedated calves.

The effect of providing pain relief using NSAID on weight gain after disbudding was investigated in two studies (Bates et al. 2015, 2016), which found that treatment with meloxicam or ketoprofen was associated with increased weight gain only in animals disbudded without sedation or local anaesthetic. There was no benefit if NSAID were used alongside sedation and local anaesthetic. In contrast the present study showed a significant increase in weight gain in the first 7 days after disbudding, only in sedated calves treated with meloxicam compared to the crate-control group. However, these results are not directly comparable with those of the previous studies, as in the present study all animals had a cornual nerve block which could have reduced any potential impact of meloxicam on growth rates. Nevertheless, our results are consistent with those of Bates et al. (2015, 2016) as the increased growth rates recorded in the 7 days after disbudding in sedated calves treated with meloxicam (0.144 kg/day) were only marginally greater than the increase seen in calves in the sedation-control group (0.125 kg/day). This finding supports the suggestion that the use of xylazine sedation alone may increase weight gain after disbudding (Bates et al. 2015, 2016).

One of the practical considerations of using the topical anaesthetic was that it was a liquid. In the study, the product was applied and any surplus liquid was caught on a towel before the calf was released. This may possibly have given a longer contact time for the product, but it is unknown if this would affect how well it was effective.

Management of animal welfare needs to be at the forefront of our dairy industry and this study has given a strong indication that, until we are able to predominantly breed polled calves, more can be done to alleviate pain and suffering following the disbudding procedure. The positive results shown from the use of the topical anaesthetic in combination with sedation indicate that this product produces an effective reduction in indicators of pain, similar to those resulting from use of NSAID.

In conclusion, there was a positive effect of disbudding calves under sedation on all outcomes. There was a benefit to providing calves with topical anaesthetic following disbudding on behavioural responses and pain sensitivity and this benefit was not statistically different to that of treatment with meloxicam. The effects were most pronounced when topical anaesthetic was applied in calves that were disbudded under sedation and indicate that the effect may last at least 22 hours following the disbudding procedure.

Acknowledgements

The authors would like to thank Bayer New Zealand Ltd for funding the study, the farm animal owners, managers and workers for all the time they spent drafting, helping make pens and finally the amazing VetEnt technicians and vets for their organisation, flexibility and most of all, enthusiasm.

ORCID

EL Cuttance http://orcid.org/0000-0003-0354-5295

DA Yang http://orcid.org/0000-0003-3893-9403

RA Laven http://orcid.org/0000-0002-8938-8595

Notes

1 N. Harding, DairyNZ, Hamiton, New Zealand.

2 https://doi.org/10.1080/00480169.2019.1640651.

3 https://doi.org/10.1080/00480169.2019.1640651.

4 https://doi.org/10.1080/00480169.2019.1640651.