Investigating the short- and long-term impacts of pain from surgical castration on affective states in piglets

doi:10.21203/rs.3.rs-2362315/v1

Surgical castration of male piglets is a routine procedure performed to improve meat quality. Pain due to castration can last for up to 4 days, negatively impacting animal welfare. The objective of this study was to assess the impact of surgical castration and practical pain alleviation methods during and after castration on piglets’ short and long-term affective states and activity. Piglets were surgically castrated (n = 22), castrated with analgesics (n = 21), or sham-handled (n = 22) at 3 days of age. Data on piglets’ activity, tails, and grimaces were collected for 1, 6, and 24 hours after castration treatments. An attention bias test was performed in week 1 (n = 31) and 12 (n = 33) to assess anxiety (an affective state), with feed (positive stimulus) and loud bangs and flashing lights (negative stimuli) presented simultaneously. Latency to eat, behavioral responses, and activity were recorded during the test. Castration treatment and sampling timepoint impacted activity levels hours after treatments. Piglets’ behavior and activity at week 1 suggest a negative impact of surgical castration on anxiety. Impacts of castration treatments observed in week 1 were no longer detected in week 12. Results confirm that surgical castration negatively impacts affective states of 1-week-old piglets, but results did not support a long-term impact detectable at 12 weeks of age. As this is the first study assessing the relationship between painful procedures and piglet affective state, more research is needed to better understand the impacts on these and other measures of pig affect.

Biological sciences/Zoology/Animal behaviour

Biological sciences/Biological techniques/Behavioural methods

Biological sciences/Biological techniques/Bioinformatics

animal welfare

attention bias

piglet castration

pain alleviation

All male piglets undergo invasive husbandry procedures within the United States (U.S.) swine industry, including surgical castration. The primary reason for castration is to prevent the development of boar taint in meat, which is a foul odor when the meat is prepared caused by the post-pubertal deposition of skatole and androsterone in body fat¹. In the U.S., carcasses that give off a pronounced boar taint are condemned for human consumption². If the odor is less pronounced, the carcass may be used for comminuted cooked meat products or for rendering. Producers in the U.S. will surgically castrate boars to prevent boar taint development and avoid potential economic consequences associated with carcass rejection or downgrading.

Surgical castration is typically performed by trained farm employees. The American Veterinary Medical Association (AVMA) recommends that after 14 days of age, boars should be castrated using analgesia, anesthesia, or both³. However, piglets in the U.S. swine industry are castrated within the first week of life so no analgesia or anesthesia is used even though the procedure is painful⁴. Pain is a combination of physical and emotional reactions to a stimulus, making it an affective experience⁵. Therefore, the pain experienced during surgical castration could possibly impact piglets’ short and long-term affective states.

Affective state can be defined as a long-lasting mood state resulting from an accumulation of subjective experiences and emotions and can range between negative and positive in valence⁶. An animal’s affective state can impact their cognitive abilities, which is called cognitive bias. Therefore, it is possible to infer an animal’s affective state by assessing their cognitive bias⁷. Three types of cognitive bias tests have been developed, including memory bias, judgment bias, and attention bias testing⁸. Attention bias testing involves the simultaneous presentation of positive and negative stimuli and the observation of the animal’s attention bias through vigilance behavior⁹. In an ambiguous situation, animals in negative affective states will show more bias towards a perceived negative stimulus than other stimuli present¹⁰ and thus will be more anxious compared to animals in a more positive affective state. The opposite can be observed in animals in a positive affective state, where they will show less bias towards the negative stimulus and respond more calmly than animals in a negative state. Attention bias tests have been used in humans¹¹ and animals, including sheep^12–14, cattle¹⁵, poultry^16–18, and pigs^9,19,20. In one study, pigs biased their attention to a threat and were more vigilant than pigs that were tested in the same arena without a threat⁹. This suggests that the threat induced an anxiety response, rather than any other aspects related to the test environment. The same researchers investigated the effects of early and later in life housing conditions on attention bias and found that pigs housed in enriched pens later in life were more anxious compared to those housed in barren pens⁹. Another study used an attention bias task and showed that Lactobacillus supplementation reduced anxiety-associated behaviors, improving affective states in piglets compared to a control diet¹⁹. Although not fully validated, these studies indicate that attention bias testing can be used as a tool to determine pigs’ affective states.

Surgical castration causes behavioral and physiological responses indicative of acute and chronic pain^21–24. However, the impact of surgical castration on short-term and long-term affective states is unclear. Therefore, the objective was to evaluate the short- and long-term impact of pain from surgical castration on pigs’ affective state. We hypothesized that surgically-castrated pigs (CAS) would bias their attention toward the threat compared to sham-handled pigs (SHAM) or pigs castrated with analgesics (CAS-A) at 1 week of age (WK1). Furthermore, we theorized that any long-term impacts of surgical castration pain would be reflected in the behavioral responses during an attention bias test at 12 weeks of age (WK12).

Body Weights

Piglet body weights on day 1, day 3, weaning, and day 84 did not differ across treatment groups (P = 0.342; Table 1).

Table 1

Mean individual body weights (raw mean ± SD; in kg) for surgically castrated pigs (CAS; n = 22), castrated pigs that received analgesia (CAS-A; n = 21), and sham-handled pigs (SHAM; n = 22) at day 1 of age, day 3 of age, at weaning, and at day 84 of age.
Treatment	Day 1	Day 3	Weaning	Day 84
CAS	1.61 ± 0.24	1.99 ± 0.36	5.66 ± 1.68	27.60 ± 5.79
CAS-A	1.65 ± 0.22	2.02 ± 0.31	6.08 ± 1.74	26.92 ± 7.81
SHAM	1.53 ± 0.23	1.86 ± 0.36	5.57 ± 1.63	24.40 ± 7.75

[Table 1]

Home Pen Behavior

Activity Measurements

Treatment (P = 0.032), timepoint (P < 0.001), and their interaction impacted forward activity in the farrowing crate (P < 0.001), with CAS-A piglets being more active than CAS piglets at 1-hour post-treatment (P = 0.009), and CAS-A tending to be more active than CAS piglets 6 hours post-treatment (P = 0.056; Fig. 1a). Treatment (P = 0.074) tended to impact vertical activity, timepoint (P < 0.001) and the interaction between treatment and timepoint impacted vertical activity (P < 0.001), with the CAS-A piglets showing more vertical activity compared to CAS piglets at 1-hour post castration (P = 0.009; Fig. 1b).

[Figure 1]

Pain Expression and Affective States

Castration treatments did not impact total Piglet Grimace Scale scores (P = 0.352). Time post-treatment impacted grimace scores (P = 0.039). At 0-hour post-treatment, 8.2% of piglets exhibited a score of 0–1 (no to low pain experience), 61.2% exhibited a score of 2–3 (difficult to interpret pain experience), and 30.6% exhibited a score of 4–6 (moderate to high pain experience), with 4.1% of piglets exhibiting a score 6 (high pain experience). At 6-hour post-treatment, 7.1% of piglets exhibited a score of 0–1, 69.6% exhibited a score of 2–3, and 23.2% exhibited a score of 4–6, with 1.8% of exhibiting a score 6. At 24-hour post-treatment, 5.5% of piglets observed exhibited a score of 0–1, 83.6% exhibited a score of 2–3, and 10.9% exhibited a score of 4–5, with no piglets exhibiting a score 6.

Castration treatments did not impact tail motion scores indicative of affective states (P = 0.229). Time post-treatment impacted tail motion scores (P < 0.001). At 0-hour post-treatment, 63.1% of piglets showed a motionless tail, 20.0% showed a loose wag, and 7.7% an intense wag, while no piglets showed a flicking tail. Tail motions were more varied at 6-hour post-treatment, as 50.8% of piglets showed a motionless tail, 36.9% showed a loose wag, 4.6% an intense wag, and 1.5% a flicking tail. At 24-hour post-treatment, 21.5% of piglets showed a motionless tail, 43.1% showed a loose wag, 18.5% an intense wag, and 4.6% a flicking tail.

Castration treatments did not impact tail posture scores indicative of affective states (P = 0.201). Time post-treatment impacted tail postures (P = 0.034). At 0-hour post-treatment, 33.8% of piglets showed an actively hanging tail, 29.2% showed an erect tail, 16.9% showed a horizontal tail, 9.2% a passive hang and 1.5% of piglets showed a curled posture, while a tucked or jammed tail were not observed. At 6-hour post-treatment, 13.8% of piglets showed an actively hanging tail, 29.2% showed an erect tail, 13.8% showed a horizontal tail, 23.1% a passive hang, 10.8% a tucked tail, and 1.5% a jammed tail, while a curled tail was not observed. At 24-hour post-treatment, 26.2% of piglets showed an actively hanging tail, 23.1% showed an erect tail, 13.8% showed a horizontal tail, 29.2% a passive hang, 4.6% showed a curled tail, 6.2% a tucked tail, and 1.5% showed a jammed tail.

Attention Bias Test

Activity

Castration treatments tended to impact forward activity during the attention bias test (P = 0.065; Fig. 2), with more forward activity in CAS compared to CAS-A (P = 0.067) piglets, but no difference between CAS and SHAM piglets (P = 0.158), or CAS-A and SHAM piglets in WK1 (P = 0.150). In WK12, castration treatment did not impact activity during the attention bias test (P = 0.650).

[Figure 2]

Behavior

Pigs’ latency to eat did not differ among treatments in WK1 (P = 0.131) or WK12 (P = 0.479; Table 2). Castration treatment (P = 0.066) and test week (P = 0.069) tended to impact the number of pigs eating, but their interaction was non-significant (P = 0.258). Fewer CAS (n = 3/21) than SHAM piglets (n = 9/20) ate in WK1, and fewer pigs ate in WK1 (n = 6/32) compared to WK12 (n = 11/29). Castration treatment impacted the time piglets spent eating in WK1 (P = 0.020), with SHAM piglets eating longer than CAS piglets after the threat was presented (P = 0.014; Table 2), but no difference between other treatment groups. Castration treatments tended to impact vigilance behavior in WK1 (P = 0.074; Table 2), with CAS-A tending to be more vigilant than SHAM piglets (P = 0.074), but no difference between other treatment groups. Castration treatments did not impact grunting frequency in WK1 (P = 0.282) or in WK12 (P = 0.162), or high-pitched vocalizations in WK 1 (not observed) or WK12 (P = 0.403).

Table 2

Means (± SEM) and p-values of the behaviors displayed after presentation of the threat during the attention bias test at 1 or 12 weeks of age. CAS = surgically castrated; CAS-A = piglets castrated with analgesics; SHAM = sham-handed piglets. ^a−b Row means with uncommon superscripts differ at P < 0.05. ^A−B Row means with uncommon superscripts differ at P < 0.10.
	Week 1			Treatment effect (p-value)
Behavior	CAS	CAS-A	SHAM	Treatment effect (p-value)
Latency to eat (sec)	180 ± 6.7	168.5 ± 7.0	160.3 ± 6.7	0.131
Attention to threat (duration in sec)	4.1 ± 0.9	3.9 ± 0.1	2.7 ± 0.9	0.112
Vigilance (duration in sec)	34.5 ± 12.1^AB	53.7 ± 12.5^A	27.1 ± 12.5^B	0.074
Eating (duration in sec)	-0.6 ± 5.3^b	8.0 ± 5.6^ab	17.7 ± 5.5^a	0.020
Exploring (duration in sec)	138.6 ± 12.6	114.8 ± 13.5	117.6 ± 13.2	0.239
	Week 12			Treatment effect (p-value)
Latency to eat (sec)	142.8 ± 19.7	149.3 ± 19.7	115.8 ± 20.8	0.478
Attention to threat (duration in sec)	3.6 ± 0.7	3.0 ± 0.7	3.4 ± 0.7	0.112
Vigilance (duration in sec)	37.4 ± 22.5	48.9 ± 22.1	32.6 ± 22.9	0.768
Eating (duration in sec)	36.8 ± 14.1	39.1 ± 15.1	64.3 ± 15.7	0.558
Exploring (duration in sec)	113.4 ± 25.2	86.0 ± 24.7	76.5 ± 26.0	0.355

[Table 2]

This study aimed to determine the impact of a common painful management procedure (surgical castration) on piglet’s affective states shortly after the procedure (week 1 of age), and later in life (week 12 of age). Castration treatments impacted both forward and vertical activity levels within the first 24 hours after treatments were applied. While treatment did not impact grimace score or tail scores indicative of affective state, both tended to be impacted or were impacted by timepoint since treatment. Castration treatment tended to impact, or impacted forward activity, number of piglets that ate, and vigilance behavior in week 1 of life. During the attention bias test, castration treatment did not impact body weight, vertical activity levels, and other behaviors in week 1 of life. When pigs were 12 weeks old, body weight, activity levels, and behavior during the attention bias test were not impacted by castration treatments.

Pain experience can reduce piglets’ activity²⁵. In our study, CAS piglets’ forward and vertical activity in the farrowing crate during the first hour after castration were reduced compared to CAS-A piglets, and forward activity tended to be reduced 6 hours after castration compared to CAS-A piglets. This decrease in activity suggests CAS piglets experienced more acute pain from surgical castration^{22,24, 30–32} than CAS-A piglets immediately after the procedure, with pain persisting at least 6 hours afterwards. The increased forward activity in CAS-A piglets compared to CAS piglets suggests effective acute pain relief from analgesics, although whether it is due to the lidocaine, the flunixin, or both, remains unclear as previous research did not report a consistent benefit for either^26–29. In line with our findings, behavioral observations indicated reduced activity in castrated piglets in other studies^{22,24, 30–32}. Forward and vertical activity did not differ between treatment groups before the castration procedure or 24 hours afterwards, even though the activity was lowest for CAS piglets compared to the other treatment groups throughout. The lack of a difference between treatments 24 hours after the procedures suggests that the acute pain waned in the CAS piglets, with an increase in activity compared to 1 and 6 hours after the procedure. Yet, activity was still reduced compared to before the procedure, suggesting that CAS piglets’ activity, thus pain experience, did not fully return to baseline 24 hours after castration and could be indicative of pain lasting longer than 24 hours. Other studies assessing behavioral indicators of pain show that pain persists anywhere from 2 hours to 4 days post-castration^30,31. The lack of difference in forward and vertical activity between CAS and SHAM piglets 1 and 6 hours after castration treatments is unexpected and unclear. Further investigation is necessary to better understand this.

Although the Piglet Grimace Scale was previously validated to detect pain expression after castration³², and can be applied successfully by a wide range of stakeholders³³, scores did not differ between castration treatments in the current study. Grimace scores were higher at 6 hours post-treatment compared to 24 hours post-treatment. Most piglets had a total score of 2 or 3, reflecting an intermediate pain level that is difficult to interpret³². Sampling from videos rather than photos may provide more contrasts in facial grimaces associated with pain and thus may provide more insights in acute pain experiences.

Pigs’ tail posture can be used as an indicator of tail biting³⁶, but also signals emotional experiences unrelated to tail biting³⁴. In the current study, tails were kept intact, as tail docking can impact tail postures, motion, and ability to express emotional states^35,36. Using the circumplex model of affectClick or tap here to enter text., tail postures and motions can provide insights into emotional valence and arousal at time of assessment³⁶. In our study, tail motions and postures were not impacted by castration treatments. Tail postures and motions did differ between sample timepoints. Tail posture and motion have not previously been investigated in the context of piglet castration. A motionless tail, the most commonly observed ‘motion’ directly after the treatments, is indicative of a low arousal and negative valence, or quadrant 3 in the model of affect, suggesting a negative affective state³⁶. Since there was no treatment effect, this may be related to the negative emotions caused by handling regardless of potential pain induced by the treatment. The transition to the loose wag as the most common tail motion after 6 and 24 hours represents a shift to a positive valance, thus a more positive affective state³⁶, suggesting that this theorized handling stress had waned. Actively hanging, erect tails, or passively hanging tails were the most common tail postures observed at 0, 6, and 24-hours post-treatments. An actively hanging tail indicates a neutral arousal level with a positive valence, indicative of a neutral affective state. Therefore, it was somewhat unexpected that most piglets showed this tail posture within an hour after the castration treatment, even though tail motion suggested negative affect. Erect tails can express high arousal and a positive or negative valence, thus are difficult to interpret in the current context. A passively hanging tail suggests high arousal and a neutral valence, mostly observed at 6 and 24-h post treatment, indicating a neutral affect state at those timepoints. More research is needed to better understand the impact of castration on tail expressions and how handling could impact those. As emotional experiences and affective states are highly individual and can be variable across time, these tail motions and postures may be reflective of individual experiences of handling and castration, thus could still be useful as a diagnostic tool for management decision making when negative affective states are expressed.

Attention bias tests can provide insights into animals’ affective state by determining anxiety towards an ambiguous threat¹⁵. The attention bias test results in the current study make it difficult to draw conclusions about increased anxiety in the CAS piglets compared to the other treatments. Forward activity during the attention bias test tended to be higher for the CAS piglets compared to the CAS-A in WK1. This hyperactivity in CAS piglets may indicate more anxiety than in the CAS-A piglets, however, this was only a tendency. This minimal difference could be in part due to the limited sample size. Latency to eat is the most common indicator of anxiety in attention bias tests^{9,17,18,37,38}. In our study, these latencies did not differ between treatments in WK1 or WK12. This could be related to the piglets’ interest in the positive stimulus provided, especially in WK1. Previous studies tested piglets after weaning or at 11 weeks of age, also using feed as a positive stimulus ^9,20. Yet no studies included piglets of pre-weaning age. As piglets solely feed from the sow at WK1 of age, the positive stimulus needed to be liquid rather than a solid feed reward. The novelty of feeding from a feeder and feeding on something other than milk, could have limited their feeding response in this test. Yet, piglets at this age did spend time ingesting the Gatorade, with SHAM piglets spending most time eating compared to the CAS piglets. Further research should focus on improving piglet responsiveness in the test, for instance by prolonging the test duration to increase the number of piglets eating, especially in WK1. Nevertheless, the results do suggest that piglets may have experienced more anxious affective states in CAS than CAS-A or SHAM, but confirmation is needed.

Castration treatments did not impact any response variables in the attention bias test when piglets were 12 weeks of age. This suggests that any impact of castration on anxiety was no longer detectable at this age. These results suggest that surgical castration did not have a long-term impact on anxiety, an affective state, in the tested piglets.

Previous research showed that surgical castration causes behavioral and physiological responses indicative of acute and chronic pain in piglets, yet the impact of the procedure and associated pain on affective states were not studied. The results in the current study suggest that, through a range of responses from surgically-castrated piglets, including reduced activity in the farrowing crate, and a tendency for increased vigilance and hyperactivity during the attention bias test, surgical castration negatively impacted piglets’ affective state in the short-term. No evidence was found of a long-term impact on affective states. Furthermore, piglets’ activity in the farrowing crate suggested that acute pain was at least partially alleviated in the piglets that received analgesics. Further research could improve the attention bias test design to increase response rates, especially in piglets at a pre-weaning age.

The use of the boars in this study was approved by Virginia Tech Institutional Animal Care and Use Committee, protocol #20–206. The study was carried out in compliance with the ARRIVE guidelines. The study was performed in accordance with relevant guidelines and regulations.

Animals And Housing

Sixty-five day-old boars (York × Landrace) originated from 11 litters born at the Virginia Tech Swine Center. Litters were farrowed in batches, with two litters in March 2021, three litters in April 2021, and six litters in August 2021. The sows were moved into conventional farrowing crates 3–6 days before the first sow in the batch was due to farrow.

On day 1 of age, boars were selected based on the number of boars in a litter (≥ 6) after which they were weighed and ear tagged. If there were more than six boars in a litter, the heaviest six males were included in the study. On day 1 and at weaning, males were administered 1 ml of CIRCUMVENT® PCV-M G2 vaccine (Merck Animal Health, Omaha, Nebraska, USA) through intermuscular injection. Males were administered 2 ml of Iron Hydrogenated Dextran (VetOne, Boise, Idaho, USA) through intermuscular injection on day 1 and were vaccinated with 2 ml of Rhini Shield TX4 (Elanco, Fort Dodge, Iowa USA) at weaning.

Sows and litters were housed in conventional farrowing crates (2.4 m × 1.5 m) until boars were weaned at (mean ± SD) 19.8 ± 1.4 days of age. Sows were fed 3 kg of a standard commercial lactation diet once per day prior to farrowing and twice per day after farrowing. While in the farrowing crate, piglets had continuous access to a water nipple and a heating mat until weaning. Piglets also had access to a commercial creep feed starting 7 days prior to weaning.

After weaning, males from the same farrowing batch were housed together in wire-floored nursery pens (2.4 m × 1.2 m) containing a nipple drinker and a feeder, at a stocking density of 3–4 piglets/m². Pigs had ad libitum access to a commercial nursery diet. After (min-max) 25–31 days in the nursery, males were moved to grower pens. Males from the same farrowing batch were housed together in partially-slatted concrete grower pens (2.6 m × 3.0 m) with two nipple drinkers and a feeder, at a stocking density of 1–2 pigs/m². Pigs had ad libitum access to a commercial grower diet.

In the third farrowing batch, tail biting was observed at 44–46 days of age. All piglets with tail lesions were treated with 0.5 ml of ceftiofur crystalline-free acid (Excede®, Zoetis, Kalamazoo, MI, USA). Thereafter, all nursery pens received two Kong toys (Kong Company, Golden, Colorado, USA), and a herding ball (Jolly Pets, Streetsboro, OH, USA). Thereafter, all grower pens were equipped with a suspended plastic teething ring (QC Supply LLC, Schuyler, Nebraska, USA) along with the two Kong toys and ball.

Treatments

This study included three treatment groups, with males undergoing surgical castration (negative control), males receiving analgesia prior to and after surgical castration (pain relief), and males sham-handled and not castrated (positive control).

Piglets were randomly allocated to one of three treatment groups within a litter using their identification number on day 1. On day 3 of age, the six boars from a single litter were removed from the farrowing crate and placed in a holding cart. A single trained technician performed all castration treatments and handling. Piglets were arbitrarily chosen from the holding cart prior to treatment, castrated or handled, and returned to cart. Once all boars from a single litter were treated after which, piglets returned to the farrowing crate.

In the CAS treatment, individual boars (n = 22) were restrained between the technician’s legs. Piglets were surgically castrated by making two vertical incisions on the scrotum and cutting the spermatic cord with a scalpel blade. These piglets were surgically castrated without anesthesia or analgesia, as is common in the U.S. swine industry ⁴. The CAS procedure took (mean ± SD) 31 ± 4 sec. Piglets in the CAS-A treatment (n = 21) were surgically castrated as in the CAS treatment, with additional analgesics. A topical lidocaine spray (Sanifo Company, Chattanooga, Tennessee, US) was applied, and the technician waited 5 sec prior to the incisions on the scrotum. The CAS-A procedure took (mean ± SD) 32 ± 4 sec. Following castration, the CAS-A piglets were administered flunixin (VetOne, Boise, Idaho, USA) at 2.2 mg/kg once a day for three days. Piglets in the SHAM treatment (n = 22) were handled similarly as in the CAS treatment; piglets were restrained between the technician’s legs for 30 sec. The SHAM procedure took (mean ± SD) 30 ± 0 sec.

Measurements

The project timeline is provided in Supplementary Figure S1. In short, data were collected on body weight, activity, pain expression, and affective state, including anxiety.

Body Weight

Individual body weights were recorded at day 1, day 3, weaning (day 20 ± 1 of age), and the last day of the trial (day 84 ± 3 of age) for all piglets.

Activity

On day 2 of age, accelerometers (HOBO Pendant® G Data Logger, Onset Computer Corporation, Bourne, MA, USA) were placed on the piglets’ back and secured with vet wrap (Andover Healthcare, Salisbury, MA, USA) around their torso and front legs. The accelerometers were programmed to start recording the following day (day 3 of age) to allow for habituation to the vet wrap and accelerometer prior to data collection. Activity data while in the farrowing crate (home pen) were collected on the X (forward activity) and Z (vertical activity) axes in m/s² at intervals of 15 sec for 1 hour prior to castration treatments and 1, 6, and 24 hours after castration treatments, resulting in 4 hours of activity data and 960 data points per piglet. In addition, accelerometers were used to record activity during the 3-minute attention bias test at week 1 and week 12 of age. On day 6 of age, the vet wrap and accelerometers were removed. For the piglets tested at week 12 of age, vet wrap and accelerometers were reattached prior to the attention bias test.

Pain Expression and Affective States

Within 1 hour after treatment, and again after 6 and 24 hours, the grimace (indicative of pain expression³⁹), and tail posture and tail motion (indicative of affective state³⁶) were video recorded (EOS Rebel T7 DSLR Camera, Canon, Tokyo, Japan) while piglets remained in the farrowing crate with the sow. If any of the treated piglets were sleeping, the sow was gently stimulated to stand or offered some feed in order to arouse the piglets. Data were not collected from piglets that were nursing or continued to sleep at that specific time point. From the video recordings, piglets were identified and time points were noted in which grimace or tail were clearly visibly for scoring. All observers were blinded for castration treatment. A total of 165 grimaces and 176 tail postures and motions were scored (Table 3).

Table 3. Sample size for piglet face captures from video for the piglet grimace scale scoring and for piglet tail posture and motion captures for affective state assessment 0, 6, and 24 hours after treatment in surgically castrated pigs (CAS), castrated pigs that received analgesia (CAS-A), and sham-handled pigs (SHAM).

Timepoint	Piglet grimace capture (n)				Tail posture and motion capture (n)
	Treatment			Total	Treatment			Total
	CAS	CAS-A	SHAM	Total	CAS	CAS-A	SHAM	Total
0 hours	19	17	18	54	19	20	20	59
6 hours	19	18	18	55	21	18	21	60
24 hours	19	18	19	56	22	18	17	57
Total	57	53	55	165	62	56	58	176

[Table 3]

Piglet pain expression was scored by two observers using the piglet grimace scale (PGS)³⁹, which consists of a 0–6 categorical scoring system with scores of 0–1 representing a piglet experiencing “no-to-low pain” and scores of 4–6 representing “moderate-to-high pain”^32,39. Inter-rater agreement was tested for 14 piglet grimaces that were not included in this study, and the agreement was strong among observers (Cronbach’s α = 0.7531).

Affective states were assessed from tail postures and motions using a modified method described by Camerlink & Ursinus (2020)³⁶. Piglets’ tails were kept intact, thus were not docked. Seven tail postures (rather than eight in ³⁶) and five types of tail motion³⁶ were recorded and represented varying combinations of arousal and emotional valence (Supplementary Table 1; Q1: high arousal, positive valence; Q2: low arousal, positive valence; Q3: low arousal, negative valence; Q4: high arousal, negative valence³⁶). Two trained observers showed an excellent inter-rater agreement (Cronbach’s α = 0.9379) based on scores for 25 piglet tail postures not included in this study, and both observers scored all pigs. Two other trained observers showed an excellent inter-rater agreement (Cronbach’s α = 0.9028) based on scores for 20 piglet tail motions not included in this study, and both observers scored all pigs.

Anxiety

At week 1 and week 12 of age, the piglets’ (n = 61) attention bias was determined as an indicator for anxiety in a test modified from⁹. Castration treatments were balanced across the 1-week-old test group (n = 32) and the 12-week-old test group (n = 29).

A 3 m × 3 m × 1 m (length × width × height) arena with a 1 m × 0.6 m start box was constructed from plywood, placed on a concrete floor in a separate building from where the farrowing crates or pens were located (Supplementary Figure S2).

First, pigs were allowed to habituate to the arena during three sessions, three days prior to the test (days 4–6 of age for WK1 piglets and days 77–89 of age for WK12 piglets). At 4 and 5 days of age, all three WK1 littermates were placed in the arena together for 3 min. At 6 days of age, WK1 piglets were individually placed in the arena for 3 min. Similarly, WK12 pigs from the same grower pen were placed in the arena for 3 min in the first two habituation sessions, and individually in the last habituation session (also outlined online in Supplementary Fig S1). During habituation, no threat or feed were presented.

A day after the third habituation session, pigs’ attention bias was individually tested for 3 min. Testing order was randomly predetermined based on piglet identification number within a farrowing batch. At day of testing, the test arena contained a familiar metal feeding pan located in the center of the arena with an electrolyte solution (Gatorade, The Gatorade Company, Chicago, IL, USA) in week 1 and a mixture of raisins, chocolate chips and apple slices in week 12. Gatorade is a non-carbonated sports drink with water and sugar as the main ingredients and is known to be voluntarily ingested by piglets at pre-weaning ages ⁴⁰. The threat consisted of a guillotine plywood door that was repeatedly opened and closed loudly (aversive sound repeated approximately 9 times) and a strobe light shining across the arena (aversive light) for 10 sec at floor level. This threat produced flashing lights, squeaking noises, and loud bangs.

During the attention bias test, the piglet was placed in the start box, and after approximately 10 sec the door was opened and the piglet was allowed to exit the start box and enter the test arena. Once the start box door was closed, the test started. The threat was presented after the piglet looked at the guillotine door (Supplementary Fig. S2) or after 10 sec passed. During and after the threat, behavior was recorded live and from video recordings. Live observations included the latency to eat (sec) and whether the piglet ate (yes/no), frequency of urination and defecation, frequency of escape behavior, frequency of vigilance, frequency of attention to threat, and frequency of vocalizations including barks, grunts, and high-pitched vocalizations as described by ⁹ (Supplementary Table 2). One observer recorded frequency of attention to threat, frequency of urination and defecation, and latency to eat (sec). A second observer recorded frequency of escape behavior and frequency of vigilance. A third observer recorded the frequency of all vocalizations. All observers were blinded to treatments.

Video-based behavioral observations were performed to determine behavioral durations (sec) and frequencies (% of total observations; Supplementary Table 2). Two trained observers recorded these behaviors and were blinded to the treatments. The observers showed an excellent inter-rater agreement (Cronbach’s α = 0.9981).

Statistical Analysis

Statistical analyses were conducted using JMP Pro 16 (SAS Institute Inc., Cary, NC, USA). Pigs were considered the experimental units. Data residuals were assessed for normality using normal quantile plots. Residuals for exploring the duration and frequency of attention to the threat showed normal distribution. Residuals for body weights, all other behavioral durations and frequencies, all activity data, number of piglets that ate, tail posture and motion, and all vocalizations did not show a normal distribution. Even though residuals for those data were not normally distributed, the use of mixed-effects models is appropriate as these are robust to quite severe violations of model assumptions such as the residuals’ distribution⁴¹, and allow for more complex models, including random effects. Statistical models are detailed in Table 4.

Table 4

Statistical models applied for each response variable with fixed and random factors included in the analysis.
Response variable	Model	Fixed	Random
Body weights	Mixed	Treatment, age, interaction	Piglet identification
Activity during the attention bias test	Mixed	Treatment (by week)	Piglet identification
Attention bias behavior (duration/latency)	Mixed	Treatment (by week)	Piglet identification, farrowing batch
Attention bias behavior (frequency/number of piglets)	Ordinal logistic	Treatment, test week, interaction
Grimace scale scores, tail postures, tail motions	Ordinal logistic	treatment, time since castration, interaction

[Table 4]

Interaction effects at P < 0.10 were removed from the models. Post-hoc comparisons were performed with Tukey HSD corrections for effects with P < 0.10. Escape, laying down, tail circling, WK1 high-pitched vocalizations, and barks were rarely observed and, therefore, not analyzed (frequency of 7 or lower). Data are presented as LSmeans ± SEM unless otherwise noted.

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request.

Acknowledgements

We thank the Virginia Pork Council for the financial support of this study. We would like to thank Garrett Walker (Virginia Tech) for assistance with animal work, Alyshia Talmage, Ema Hollinger, Jane Perkins, Cody Howard, Kara Crudup, Hailey Roman, and Elise Dow (Virginia Tech) for assisting with data collection and Sadie Grant, Jake Tiss, and Kayleah Corral (Virginia Tech) for assistance with animal care and data collection.

Author contributions statement

J.M.N.: methodology, conceptualization, funding acquisition, investigation, data curation, writing—original draft, writing—review and editing. L.J.: methodology, conceptualization, funding acquisition, data curation, project administration, writing—original draft, writing—review and editing. A.B.A.A.: methodology related to activity data, writing—review and editing. All authors read, and approved the submitted version.

Additional information

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding

This work was financially supported by the Virginia Pork Council.

Keenan, D. F. Pork Meat Quality, Production and Processing on. in Encyclopedia of Food and Health 419–431 (Elsevier Ltd., 2016). doi:10.1016/B978-0-12-384947-2.00551-1.
9 CFR 311.20 - Sexual odor of swine. 143 (2012).
AVMA. Swine policy revised to recognize immunocastration. https://www.avma.org/javma-news/2013-08-01/swine-policy-revised-recognize-immunocastration (2013).
Rault, J. L., Lay, D. C. & Marchant-Forde, J. N. Castration induced pain in pigs and other livestock. Appl Anim Behav Sci 135, 214–225 (2011).
Steagall, P. v., Bustamante, H., Johnson, C. B. & Turner, P. v. Pain management in farm animals: Focus on cattle, sheep and pigs. Animals 11, (2021).
Mellor, D. J. Enhancing animal welfare by creating opportunities for positive affective engagement. N Z Vet J 63, 3–8 (2014).
Roelofs, S., Boleij, H., Nordquist, R. E. & van der Staay, F. J. Making decisions under ambiguity: Judgment bias tasks for assessing emotional state in animals. Front Behav Neurosci 10, (2016).
Paul, E. S., Harding, E. J. & Mendl, M. Measuring emotional processes in animals: the utility of a cognitive approach. Neurosci Biobehav Rev 29, 469–491 (2005).
Luo, L., Reimert, I., de Haas, E. N., Kemp, B. & Bolhuis, J. E. Effects of early and later life environmental enrichment and personality on attention bias in pigs (Sus scrofa domesticus). Anim Cogn 22, 959–972 (2019).
Lee, C. et al. Anxiety influences attention bias but not flight speed and crush score in beef cattle. Appl Anim Behav Sci 205, 210–215 (2018).
Tanda, T., Toyomori, K. & Kawahara, J. I. Attentional biases toward real images and drawings of negative faces. Acta Psychol (Amst) 229, (2022).
Rodger, J. et al. Attention Bias Test Differentiates Anxiety and Depression in Sheep. Frontiers in Behavioral Neuroscience 12, (2018).
Lee, C., Verbeek, E., Doyle, R. & Bateson, M. Animal behaviour Attention bias to threat indicates anxiety differences in sheep. Biol Lett 12, (2016).
Monk, J. E., Lee, C., Dickson, E. & Campbell, D. L. M. Attention Bias Test Measures Negative But Not Positive Affect in Sheep: A Replication Study. Animals 10, (2020).
Crump, A., Arnott, G. & Bethell, E. J. Affect-Driven Attention Biases as Animal Welfare Indicators: Review and Methods. Animals 8, 1–24 (2018).
Campbell, D. L. M. et al. An attention bias test to assess anxiety states in laying hens. PeerJ 2019, e7303 (2019).
Anderson, M. G. et al. Effect of environmental complexity and stocking density on fear and anxiety in broiler chickens. Animals 11, (2021).
Campbell, A. M., Johnson, A. M., Persia, M. E. & Jacobs, L. Effects of Housing System on Anxiety, Chronic Stress, Fear, and Immune Function in Bovan Brown Laying Hens. Aniamls 12, (2022).
Verbeek, E., Dicksved, J. & Keeling, L. Supplementation of Lactobacillus early in life alters attention bias to threat in piglets. Nature 11, 1–10 (2021).
Ipema, A. F. et al. Assessing the Effectiveness of Providing Live Black Soldier Fly Larvae (Hermetia illucens) to Ease the Weaning Transition of Piglets. Front Vet Sci 9, (2022).
Prunier, A., Mounier, A. M. & Hay, M. Effects of castration, tooth resection, or tail docking on plasma metabolites and stress hormones in young pigs. J. Anim. Sci 83, 216–222 (2005).
Sutherland, M. A., Davis, B. L., Brooks, T. A. & McGlone, J. J. Physiology and behavior of pigs before and after castration: Effects of two topical anesthetics. Animal 4, 2071–2079 (2010).
Lonardi, C., Scollo, A., Normando, S., Brscic, M. & Gottardo, F. Can novel methods be useful for pain assessment of castrated piglets? Animal 9, 871–877 (2015).
Gottardo, F. et al. Pain alleviation during castration of piglets: a comparative study of different farm options. J Anim Sci 94, 5077–5088 (2016).
Ison, S. H., Eddie Clutton, R., di Giminiani, P. & Rutherford, K. M. D. A Review of Pain Assessment in Pigs. Front Vet Sci 3, (2016).
Burkemper, M. C., Pairis-Garcia, M. D., Moraes, L. E., Park, R. M. & Moeller, S. J. Effects of Oral Meloxicam and Topical Lidocaine on Pain associated Behaviors of Piglets Undergoing Surgical Castration. Journal of Applied Animal Welfare Science 23, 209–218 (2020).
Kluivers-Poodt, M., Zonderland, J. J., Verbraak, J., Lambooij, E. & Hellebrekers, L. J. Pain behaviour after castration of piglets; Effect of pain relief with lidocaine and/or meloxicam. Animal 7, 1158–1162 (2013).
Merenda, V. R. et al. Impact of transdermal flunixin administration on serum prostaglandin E2 and cortisol concentrations in piglets following castration. Am J Vet Res 83, (2022).
Langhoff, R. et al. Investigation about the use of analgesics for the reduction of castration-induced pain in suckling piglets. Berl Munch Tierarztl Wochenschr 122, 325–332 (2009).
Hay, M., Vulin, A., Génin, S., Sales, P. & Prunier, A. Assessment of pain induced by castration in piglets: Behavioral and physiological responses over the subsequent 5 days. Appl Anim Behav Sci 82, 201–218 (2003).
Taylor, A. A., Weary, D. M., Lessard, M. & Braithwaite, L. Behavioural responses of piglets to castration: the effect of piglet age. Applied Animal Bhaviour Science 73, 35–43 (2001).
Viscardi, A. V., Hunniford, M., Lawlis, P., Leach, M. & Turner, P. V. Development of a piglet grimace scale to evaluate piglet pain using facial expressions following castration and tail docking: A pilot study. Front Vet Sci 4, (2017).
Neary, J. M., Porter, N. D., Viscardi, A. v. & Jacobs, L. Recognizing Post-Castration Pain in Piglets: A Survey of Swine Industry Stakeholders and the General Public. Frontiers in Animal Science (2022) doi:10.3389/FANIM.2022.937020.
Camerlink, I., Coulange, E., Farish, M., Baxter, E. M. & Turner, S. P. Facial expression as a potential measure of both intent and emotion. Sci Rep 8, (2018).
Herskin, M. S., di Giminiani, P. & Thodberg, K. Effects of administration of a local anaesthetic and/or an NSAID and of docking length on the behaviour of piglets during 5 h after tail docking. Res Vet Sci 108, 60–67 (2016).
Camerlink, I. & Ursinus, W. W. Tail postures and tail motion in pigs: A review. Appl Anim Behav Sci 230, 105079 (2020).
Campbell, D. L. M., Dickson, E. J. & Lee, C. Application of open field, tonic immobility, and attention bias tests to hens with different ranging patterns. PeerJ 2019, e8122 (2019).
Verbeek, E., Colditz, I., Blache, D. & Lee, C. Chronic stress influences attentional and judgement bias and the activity of the HPA axis in sheep. PLoS One 14, e0211363 (2019).
Viscardi, A. v. & Turner, P. v. Use of Meloxicam or Ketoprofen for piglet pain control following surgical castration. Front Vet Sci 5, 299 (2018).
Lin, L. Q. Development of a Novel Piglet Model to Simulate the Physiology of a Child with Hypoplastic Left Heart Syndrome during the First Interstage for Investigating Changes in Tricuspid Valve as it Adapts to a High-pressure and High-volume Stress Environment. (University of Alberta, 2020). doi:10.7939/R3-9TYJ-KK96.
Schielzeth, H. et al. Robustness of linear mixed-effects models to violations of distributional assumptions. Methods Ecol Evol 11, 1141–1152 (2020).

No competing interests reported.

NearyimpactsofcastrationonaffectivestatesinpigletsSM.docx

Investigating the short- and long-term impacts of pain from surgical castration on affective states in piglets

Status:

Version 1

Abstract

Figures

Introduction

Results

Body Weights

Home Pen Behavior

Activity Measurements

Pain Expression and Affective States

Attention Bias Test

Activity

Behavior

Discussion

Conclusion

Methods

Animals And Housing

Treatments

Measurements

Body Weight

Activity

Pain Expression and Affective States

Anxiety

Statistical Analysis

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1