Characterization of the Lying Down Sequence in Lame and Non-Lame Sows

The objectives of this study were to characterize the different postures and movements for the lying down sequence in multiparous sows and to identify differences between lame and sound sows. Eighty-five multiparous sows (parity range 1 to 4) were used for this study. Sows were moved from their gestation stall to a stall in an empty area of the barn. A digital video camera was positioned on the adjacent stall so the sows’ profile was visible. Sows were video recorded for one lying down event on days 30, 60 and 90 of gestation. Observations ceased when the sow successfully lied down or if 2.5 hours elapsed since recording began. Prior to recording, sows were scored for lameness on a 3-point scale, (1 = normal to 3 = severely lame). From the video, postures and movements that occurred during the lying sequence were identified. Time from kneeling to shoulder rotation (KSR; seconds), time from shoulder rotation to lying (SRHQ; seconds), total time to lie down (TLIE; seconds), latency to lie down (LATENCY; minutes) and number of attempts (ATTEMPTS) to successfully lie down were recorded. Sows were re-classified as sound or lame. Parities were reclassified as 1, 2, and ≥ 3 due to small numbers of older sows. Time variables were analyzed using mixed-model methods. ATTEMPTS were classified as 1, 2 and ≥ 3 and analyzed using multinomial logistic regression. Models included gestation day, lameness status and parity. On average, sows took 13.9 seconds for KSR, 7.7 seconds for SRHQ, 20.5 seconds for TLIE and 66.1 minutes for LATENCY. Lameness was not a significant source of variation for any trait evaluated. However, lame sows tended to take longer during KSR (15.5 vs. 11.9 ± 1.59 seconds for lame and sound sows, respectively; P = 0.08), and to spend less time standing (54.1 vs. 69.8 ± 6.20 minutes for lame and sound sows, respectively; P = 0.06) compared with sound sows. Gestation day and parity were not associated with the time taken for the different movements in the lying down sequence (P > 0.05). Additionally, there were no significant associations between gestation day, lameness status or parity and ATTEMPTS. Results suggest that lameness scores do not greatly affect the lying down sequence. However, this could be due to the fact that lameness recorded in this study was not severe enough to affect the lying down sequence.