Ethics statement
Procedures were approved by The University of British Columbia Animal Care Committee under application A21-0111 and conducted in accordance with guidelines form the Canadian Council of Animal Care38. Reporting followed ARRIVE guidelines.
Animals and housing
The study was conducted at The University of British Columbia’s Dairy Education and Research Centre. To our knowledge, no study has used a similar paradigm in calves. To establish a sample size estimate, we relied on welfare studies using analogous SNC paradigms but applied to other species: rats (six subject per treatment9) and pigs (sixteen subjects per treatment group10). Considering this range and our own practical limitations, we settled on a sample size of ten subjects per treatment group. Thirty-five Holstein calves (all females) were initially enrolled in the study. Five calves were removed from the trial: three fell ill (scours and fever), one showed an extreme stress response when moved outside of her home pen, and one was not feed-restricted before a test. The thirty remaining had an average (± SD) birth weight of 38.3 ± 4.1 kg and were enrolled at 39.9 ± 4.1 d of age.
As routine farm practice, calves from all three treatments were intermingled in indoor pens (4.9 × 7.3 m, bedded with sawdust, and each containing eight to ten calves). Calves were provided ad libitum access to water and hay (RIC; Insentec B.V., Netherlands), and time-restricted access to 12 L of whole milk through a nipple feeder (CF 1000 CS Combi; DeLaval Inc., Sweden). To avoid long delay during trials, small replicates (average number of subjects per replicate = 3.5) were conducted.
Apparatus
The experimental apparatus was located in the same barn as the calves’ home pen, approximately 10–30 m away. The apparatus was a 1.8 × 1.2 m start-box leading to a 3.6 × 2.4 m pen through a vertical gate (Fig. 2A). Directly across from the start-box was a bottle and rubber teat mounted on rails, with an algometer (FPX 25, Wagner, Greenwich, USA) installed behind the bottle allowing measures of the maximum pressure applied to the bottle (Fig. 2B).
Training
The trial was divided in three phases over seven days: training (three days), treatment (one day) and testing (three days). During training, calves were feed-restricted overnight (from 22:00 h) to ensure a high motivation for milk rewards over repeated trials. At approximately 10:00 h calves were individually brought into the apparatus, with no set order, and then placed in the start-box. The vertical gate was lifted and calves could approach and drink a 0.5 L milk reward from the bottle (this amount was based on previous studies on motivation trade-offs studies in calves37,39). Latency to contact the bottle (with mouth or tongue), latency to finish the reward, number of vocalisations and maximum pressure applied to the bottle were recorded live. The calf was then brought back to the start-box, the bottle refilled, and two more trials were conducted (i.e., for a total of three trials/d). After these trials were completed, the calf was returned to her home pen with full access to her daily milk allowance of (12 L/d). Training took place over three consecutive days, for a total of nine training trials. During the first day of training (for all three trials), no cues were given to the calf for the first minute after opening the start-box gate. After one minute, auditory (calls/whistle) and tactile (finger suckling) cues were given from the experimenter from outside the test-pen to get the calf’s attention towards the bottle. If these cues had failed after an additional minute, the experimenter would go inside the test pen and lead the calf to the bottle.
During the second and third day of training, no cues were given. If the calf had not approached the bottle within two minutes, the trial was recorded as a no-approach (and a pressure of zero applied to the bottle). Once a calf had approached the bottle, she had three additional minutes to finish the reward.
Treatments
Calves were pseudo-randomly assigned to one of three treatments (Disbudding, Disbudding + Analgesia, or Sham; ten calves each). Treatment assignment was balanced for age and birthweight (Disbudding: 40.7 ± 4.3 d, 38.7 ± 3.9 kg; Disbudding + Analgesia: 39.2 ± 7.0 d, 38.0 ± 5.6 kg; Sham: 39.7 ± 6.0 d, 38.3 ± 2.4 kg). On treatment day, calves were not feed-restricted and went through their treatment in their group pen at approximately 10:00 h. Regardless of treatment, calves were weighted and administered a multimodal pain mitigation strategy of sedative, local anesthesia and analgesia. The sedative was used to facilitate following injections and disbudding (xylazine 0.2 mg/kg Subcutaneous, Rompun 20 mg/mL, Bayer, Leverkusen, Germany). After sedation was reached (recumbency and eye rotation, approximately 10 min), a local anesthetic was injected as a cornual nerve block to mitigate the acute pain of the procedure (5 mL per side, lidocaine 2%, epinephrine 1:100,000, Lido-2, Rafter8, Calgary, AB, Canada), an NSAID was provided to minimize inflammation (meloxicam 0.5 mg/kg Subcutaneous, Metacam 20 mg/mL, Boehringer Ingelheim, Burlington, ON, Canada), and the horn bud area was shaved with an electric trimmer. Ten minutes after lidocaine injection, a pinprick test was done on the horn buds to test for pain reflex. For calves in the Disbudding and Disbudding + Analgesia treatments, a pre-heated electric dehorner (X30, 1.3 cm tip, Rhinehart, Spencerville, IN, USA) was applied to both horn buds until a consistent dark ring formed around each bud (requiring approximately 10 to 15 s). Calves from the Sham group were treated identically but instead of being disbudded, only pressure on the horn buds was applied with the plastic handle of the dehorner. After the procedure was completed, calves were positioned in sternal recumbency and left to recover in the pen. As the magnitude and duration of NSAID effects following disbudding remain unclear 18, calves from the Disbudding + Analgesia group received an additional NSAID injection (ketoprofen, 3 mg/kg, Subcutaneous, Anafen, 100 mg/mL, Boehringer Ingelheim, Ontario, Canada) 1 h before each of the three test sessions to provide supplemental pain control at the time of testing. Based on a previous study on the efficacy of ketoprofen after disbudding29, we expected ketoprofen to provide analgesic effects for up to 2 h following treatment.
Tests
In the three days following treatment, calves were tested for sensitivity to reward loss. Tests were similar to training: calves were brought individually to the apparatus after overnight feed restriction, and allowed access to a milk reward three times in a row (for a total of nine trials), but during testing the reward was reduced to 0.1 L. The time allowed for calves to approach and drink the reward was matched with their performance during training. Maximum pressure applied to the bottle, number of vocalisations and latency to approach were recorded. Calves from the Disbudding + Analgesia group received an additional NSAID injection (ketoprofen, 3 mg/kg, Subcutaneous, Anafen, 100 mg/mL, Boehringer Ingelheim, Ontario, Canada) 1 h before each of the three test sessions. After each session calves were returned to their home pen and again provided access to their full milk allowance (12 L). After the three test days calves were returned to routine farm care.
Statistical analysis
A mixed model was conducted on each outcome (maximum pressure, vocalisations and approach latency) on test phases (post treatment) using R’s lme4 package40. For pressure and latency, data were log transformed to fit model assumptions of linearity, normality and homoscedasticity. For vocalisation counts, we used a Poisson mixed model. Fixed factors were treatment (2 df), test day (1 df), daily trial (1 df) and their interaction (3 df). Daily trial, nested within day and Calf ID, was included as a random factor. Significance and tendency thresholds were set at P ≤ 0.05 and P ≤ 0.10, respectively. Data (Supplementary Information 1) and R code (Supplementary Information 2) are available in supplementary materials.