Characterization of the lying and rising sequence in lame and non-lame sows

Abstract

This study aimed to identify possible differences in the lying and standing sequence between lame and non-lame gestating sows. A total of 85 stall-housed sows (average parity 0.9 ± 1.14; range 0-4) were scored for walking lameness on a 3-point scale (1 = normal to 3=severely lame) while moving to a separate gestation stall for recording of one lying-standing event on days 30, 60 and 90 of gestation. A video camera was positioned on the adjacent stall so sows' profiles were visible. Observations ceased when the sow laid-down and stood-up, or 2.5 h elapsed from recording commencement. From videos, postures and movements that occurred during lyingstanding sequences were identified. Time (seconds) from kneeling to shoulder rotation (KSR), shoulder rotation to lying (SRHQ), total time to lie (TLIE); latency to lie (LATENCY; minutes) and number of attempts to successfully lie were recorded. Also, time taken from first leg fold to sit (TLS), time from sit to rise (TSR), and total time to rise (TRISE) were recorded. Sows were re-classified as non-lame (score 1) and lame (scores ≥ 2). Data were analyzed using mixed model methods with gestation day, and lameness as fixed effects and sow the random effect. On average, sows took 14.3 ± 1.39 s for KSR, 7.7 ± 0.79 s for SRHQ, 21.0 ± 1.37 s for TLIE and 63.6 ± 5.97 min for LATENCY. Furthermore, sows took 8.8 ± 2.80 s for TLS, 5.95 ± 1.73 s for TSR, and 10.3 ± 2.02 s for TRISE. There were no associations between lameness status or gestation day with time required for or the likelihood of performing the different movements of the lying and standing sequences (P > 0.05). Except for lame sows tending to sit more while transitioning from lying to standing than non-lame sows (P = 0.09). Seven different lying and 4 different standing combination deviation from the normal sequences, albeit each combination was infrequent and did not allow for statistical analysis. However, all together, deviations from the normal lying and standing sequence accounted for 22.7 % and 35 % of total observations; respectively. Under the conditions of this study, lameness did not influence the time taken or the likelihood of performing different movements and/or postures during normal lying-standing sequences. However, this could be due to lameness recorded here not being severe enough to affect the sequences. The observed deviations suggest that there is variation in the way sows lie and stand although more research is necessary to understand which factors contribute to such variation.

Welfare of pigs on farm

“…lateral lying (McGlone et al., 2004). This may be reflected in deviations from the normal standing up and lying down sequence (Mumm et al., 2020), prolonged ventral lying (Marchant, 1994), standing or dog‐sitting. Such restrictions also likely disrupt the quantity and quality of sleep.…”

Posture Detection of Individual Pigs Based on Lightweight Convolution Neural Networks and Efficient Channel-Wise Attention

“…Pig postures reflect health status [1,2], welfare level [3], and environmental status [4]. Timely recognition of abnormal posture can help alleviate the spread of disease in pigs, reduce the use of veterinary antibiotics, and improve the economic benefits of commercial pig farms.…”