Facial Expressions of Horses Using Weighted Multivariate Statistics for Assessment of Subtle Local Pain Induced by Polylactide-Based Polymers Implanted Subcutaneously.
Authors: Carvalho Júlia R G, Trindade Pedro H E, Conde Gabriel, Antonioli Marina L, Funnicelli Michelli I G, Dias Paula P, Canola Paulo A, Chinelatto Marcelo A, Ferraz Guilherme C
Journal: Animals : an open access journal from MDPI
Summary
# Editorial Summary Facial-expression coding systems are increasingly used to identify pain in horses, yet their sensitivity to mild inflammatory responses remains unclear. Researchers implanted polylactide-based polymers subcutaneously in horses whilst controls received sham surgery, then analysed facial expressions at baseline, 24 and 48 hours post-procedure using five different statistical approaches: simple scoring of seven facial features (ear movements, eyebrow tension, orbicularis oculi tension, nostril dilation, eye opening, muzzle tension, and masticatory muscle tension), summed scores, principal component analysis, weighted summed scores, and global facial scoring. Despite measurable changes in cutaneous temperature and mechanical nociceptive threshold in the implanted horses—objective indicators that inflammation was present—none of the five facial-expression analysis methods detected statistically significant differences between treatment groups, either in individual features or composite scores. These findings suggest that whilst facial-expression assessment may be valuable for detecting moderate to severe pain in horses, it may lack sensitivity for identifying low-grade inflammatory pain such as that induced by minor surgical trauma or subcutaneous implants, which has important implications for practitioners relying on these systems to assess post-operative comfort or detect early-stage conditions. Clinicians should consider combining facial assessment with objective physiological measures such as thermography or pressure algometry when evaluating subtle pain states in equine patients.
Read the full abstract on PubMed
Practical Takeaways
- •Facial expression scoring alone is insufficient for detecting mild to moderate pain in horses; practitioners should combine multiple pain assessment modalities including nociceptive threshold testing and thermal imaging
- •When using facial expressions to assess equine pain, understand that these visual indicators may only become apparent with more severe pain levels and may miss low-grade inflammatory conditions
- •Consider alternative or supplementary pain assessment methods (mechanical nociception, thermography, behavior analysis) when facial expressions appear unchanged but clinical signs suggest discomfort
Key Findings
- •Five statistical methods for analyzing facial expressions (SUM, PCoA, SUM.W, individual FE scoring, and GFS) failed to detect low-grade inflammatory pain in horses implanted with polylactide-based polymers
- •Control group showed statistically higher values for SUM, PCoA, and SUM.W at 48 hours post-implantation despite sham surgery
- •Horses with polymer implants demonstrated more obvious alterations in cutaneous temperature and mechanical nociceptive threshold compared to controls, suggesting facial expressions are insensitive to subtle pain detection
- •Facial expression coding systems may be inadequate for detecting pain below a certain inflammatory threshold in equine research