Bovine pain scale: A novel tool for pain assessment in cattle undergoing surgery in the hospital setting

Abstract

Pain negatively impacts animal welfare and it is still neglected in ruminants. This original study aimed to develop and validate the Bovine Pain Scale (BPS) for acute pain assessment in hospitalized cattle undergoing surgery. This was a blinded, randomized, prospective clinical study. Thirty-six animals were included in the study. The Pain Group (n = 25) included patients admitted to a veterinary teaching hospital requiring any type of soft tissue or orthopedic surgery. Videos were recorded before, 2-6 hours after surgery, 1 hour after the administration of analgesia and 24 hours after surgery. The Control Group (n = 11) included healthy animals that were video recorded twice within a 24-48h interval. The BPS was developed using content validity. A total of 118 videos of 6 minutes were randomized and analyzed by four raters who were unaware of groups, time-points and procedures in two phases with a five-week interval. Statistical analysis was performed using R software. Intra and inter-rater reliability (intra-class correlation coefficient) was very good (0.83-0.94) and ranged from good to very good, respectively (0.65-0.81). The correlation between the BPS and the Visual Analog Scale (VAS) was strong (rho = 0.77, p < 0.0001) confirming criterion validity. Item-total correlation was acceptable for 3 of 9 items (0.33-0.43) and internal consistency was below the acceptable value (0.6). The scale was responsive to pain but not the administration of analgesia. It was specific for five items, but no items showed sensitivity. The area under the curve of 0.90 demonstrated high discriminatory capacity. According to the receiver operating characteristic curve, the cut-off point for rescue analgesia was ≥ 5 of 18. The BPS is reliable and reproducible, showed content and criterion validity, and may be used in veterinary hospitals for assessing post-operative pain in cattle to guide decision-making towards rescue analgesia. Future studies should refine the instrument to guarantee construct validity and sensitivity.

Fig 1. Timeline of video analysis for the validation of the Bovine Pain Scale.

Fig 2. Plot correlating the scores of the Bovine Pain Scale (BPS) and the Visual Analog Scale (VAS) of perioperative period of cattle undergoing surgery (n

= 25; Pain Group) and in healthy cows (n = 11; Control Group).

Fig 3. Box-plot of Bovine Pain Scale (BPS) total score over time-points of perioperative period of cattle undergoing surgery (n

= 25; Pain Group) and healthy cows (n = 11; Control Group) for all raters. Time-points: P1, preoperatively, immediately prior to administration of sedatives; P2, 2 to 6 hours after the end of surgery; P3, 1 hour after the administration of analgesic intervention if required; P4, 24 hours after surgery; were assessed. C1, first video and assessment of animals of the control group; C2, the second video of same animal of the control group, recorded 24-48 hours after C1. The lower and upper bounds of the box respectively represent the first and third quartile of data; the horizontal line plus space inside the box indicates the median; the black diamond indicates the average of each time-point data separately; black circles indicate outlier. Different lowercase letters indicate statistical difference over the time-points (a > b); multiple comparisons were conducted by linear mixed model with post-test corrected by Bonferroni procedure (p < 0.05).

Fig 4. ROC curve with the diagnostic uncertainty zone for the Bovine Pain Scale (BPS).

Two-graph ROC curve, CI of 1,001 replications was used to estimate the diagnostic uncertainty zone of the cut-off point of all raters, according to the Youden index [31,32]. The diagnostic uncertainty zone was 5; ≤ 4 indicates pain-free animals (true negatives) and ≥ 6 indicates animals suffering pain (true positives). The Youden index ≥ 5 represents the cut-off point for the indication of rescue analgesia.