Facial Expressions of Horses Using Weighted Multivariate Statistics for Assessment of Subtle Local Pain Induced by Polylactide-Based Polymers Implanted Subcutaneously
Nenhuma Miniatura disponível
Data
2022-09-01
Autores
Carvalho, Júlia R. G. [UNESP]
Trindade, Pedro H. E. [UNESP]
Conde, Gabriel [UNESP]
Antonioli, Marina L. [UNESP]
Funnicelli, Michelli I. G. [UNESP]
Dias, Paula P.
Canola, Paulo A. [UNESP]
Chinelatto, Marcelo A.
Ferraz, Guilherme C. [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Resumo
Facial-expression-based analysis has been widely applied as a pain coding system in horses. Herein, we aimed to identify pain in horses undergoing subcutaneously polylactide-based polymer implantation. The sham group was submitted only to surgical incision. The horses were filmed before and 24 and 48 h after implantation. Five statistical methods for evaluating their facial expressions (FEs) were tested. Primarily, three levels of scores (0, 1, and 2) were applied to the seven FEs (ear movements, eyebrow tension, orbicularis tension, dilated nostrils, eye opening, muzzle tension, and masticatory muscles tension). Subsequently, the scores of the seven FEs were added (SUM). Afterwards, principal component analysis (PCoA) was performed using the scores of the seven FEs obtained using the first method. Subsequently, weights were created for each FE, based on each variable’s contribution variability obtained from the PCoA (SUM.W). Lastly, we applied a general score (GFS) to the animal’s face (0 = without pain; 1 = moderate pain; 2 = severe pain). The mechanical nociceptive threshold (MNT) and cutaneous temperature (CT) values were collected at the same moments. The results show no intra- or intergroup differences, when evaluating each FE separately or in the GFS. In the intragroup comparison and 48 h after implantation, the control group showed higher values for SUM, PCoA, and SUM.W, although the horses implanted with polymers displayed more obvious alterations in the CT and MNT. Our findings show that the five statistical strategies used to analyze the faces of the horses were not able to detect low-grade inflammatory pain.
Descrição
Palavras-chave
animal welfare, behavior, equine, principal component analysis, validation
Como citar
Animals, v. 12, n. 18, 2022.